t
Win
.
-
.
;
r
-
INTRtiUCTItN T# PHYSICAL METALLURGY
Sec«nd E ititn
SliNEY H. AVNER Professor
iview York City Community College " .
City University of New York
V;
.
I
.
v-
' .
-
i
I J
.
i
it
I
"
W
M
McGRAW-HILL
BQQK COMPANY
New York
Kuala Lumpur
Panama
St. Louis
London
San Francisco
Mexico
Rip de Janeiro Singapore
Dtisseldorf
Montreal
Sydney
Johannesburg
New Delhi
Toronto:
.
'
;
r
a
-
66% Ofi .
'
a
.
If
in Publication Dat
t
«
CONTENTS Preface Introduction 1
Tools of the Metallurgist Metal Structure and Crystallization
3
Plastic Deformation
.
.
65 ....
,
107
.
W? Annealing and Hot Working
yi?. Constitution of Alloys
i i
; reserved.
The Iron-lronCarbide Eauilibrium Diagramv- iflvfi fi. The Heat Treatment of Steel IP IiSR 7
publica-
.
r transnical
r
V f
,
.
Steels
|349*|,
Tool Steels
387
11.
Cast Iron
42i3
12.
Nonferrous Metals and Alloys
461 * gg|
13.
Metals at.High and Low Temperatures
547
14.
Wear of Metals
567
15.
Corrosion of Metals
583
«uction
16.
Powder Metallurgy
605
1a by
17.
Failure Analysis
633
I
| V
-Alloy
10.
ir
rna
r
,
1
1 1
I
Appendix: Temperature-conversion Table Glossary Index
'
:
mm
.
.
666 : 667 689
- <; '
.
,
4
; :w
PREFACE
I 1 The emphasis of the seco
cepts and applications of The level of this editioi
considered appropriate fc who are not majors in me introductory course. It ha grams in industry. The fu
I
plified form yet as accure is an elementary course ir During the past decad( quite effective, and manj students and faculty mer suggestions from users h; The following is a surr
i
To Judy Kenny, and Jeffrey, ,
in whose hands
the future lies
second edition:
In Chapter 1, which co\ field of metallurgy, the s panded to include eddy c code has been used in the
\
\
Some changes have be( tied explanation of atom brief explanation of x-ra
'
3
added.
Chapters 3 and 4 whicl and the effect of heat on c
except for an expanded c Chapter 6, on binary f detailed explanation of th diagrams as illustrations. Chapter 7 which consii
i '
i
some detail now also d
quantities of other eleme The section on case h£
to include a more detail'
induction hardening. A s added.
1: "
In Chapter 9, the por sections on precipitatior ausforming.
In Chapter 10 the secti and a new section on ce
J
Chapter 11 now cover 4
|;.; -
:. 0
tional diagrams.
T!5
PREFACE
.
.:
m
il
a
.
The emphasis of the second edition of this text remains on the basic concepts and applications of physical metallurgy. The level of this edition is also essentially unchanged. The text is still considered appropriate for the teaching of physical metallurgy to students who are not majors in metallurgy as well as to engineering students as an introductory course. It has also proved useful for technician training pro-
grams in industry. The fundamental concepts are still presented in a simplified form yet as accurately as possible. The only background required
is an elementary course in physics.
1
During the past decade, the first edition of this text was found to be quite effective, and many favorable comments were received from both "
students and faculty members. However, advances in certain areas and
m s
.
suggestions from us6rs have necessitated a revision of the first edition. The following is a summary of the most notable improvements in the second edition:
In Chapter 1, which covers some of the important topis and tests in the field of metallurgy, the section on nondestructive testing has been ex-
panded to include eddy current testing and holography.The latest ASTM code has been used in the section on hardness testing. \ Some changes have been made in Chapter 2 in order to make the simpliV
fied explanation of atomic and metal structure more, understandable. A brief explanation of x-ray diffraction and grain size measurement was added.
r
.
Chapters 3 and 4 which cover the fundamentals of plastic deformation and the effect of heat on cold-worked materials remain essentially the same
except for an expanded discussion of dislocations and {fracture. Chapter 6, on binary phase diagrams, now includes diffusion, a more detailed explanation of the theory of age hardening, and more actual phase diagrams as illustrations. i
Chapter 7 which considers the iron-iron carbide equilibrium diagram in some detail now also discusses wrought iron and the effect of small
,
quantities of other elements on the properties of steel.. The section on case hardening of steel in Chapter 8 has been expanded to include a more detailed explanation of nitriding flame hardening and induction hardening. A section on hardenable carbon steels has also been ,
added.
In Chapter 9 the portion on stainless steels now encompasses new sections on precipitation-hardening stainless steels maraging steels, and ausforming. ,
,
In Chapter 10 the section on cemented carbide tools has been expanded and a new section on ceramic tools has been added
.
Chapter 11 now covers only cast iron and has been enhanced by additional diagrams.
mil
4
INTRODUCTION
PREFACE
vt
Metallurgy is the science
1 > -;l 1
7-
m
The numerous additional photomicrographs added to Chapter i2 il-
of this text to cover the
lustrate various nonferrous microstructures. An entire section on titanium
certain highlights will be
and titanium alloys has been included because of their increased commercial importanolt The chapter on wear of metals has been moved next to the one on cor-
mythology. Metallurgy a
The worker of metals i Ancient man knew and l
rosion of metals to improve the continuity of subject matter. Chapter 15 now discusses the corrosion of metals in greater detail. A brief discussion of the powder metallurgy processing techniques has
i
1
ments, plates, and utens ing, and shaping metals
4
the Chinese. The ancier
been added to Chapter 16, There are two major changes in this edition as compared to the first
ore and that steel had tl
edition.
tendency to rust, and t
before 1000 B.C. Iron wa brass, and bronze.
The replacement of Chapter 17 on extractive metallurgy by an entirely new chapter on failure analysis. It was felt that extractive metallurgy was not really part of physical metallurgy and that a chapter on failure analysis would be of greater in\ terest and value to technicians and engineers. \2 The addition of a glossary of terms related to physical metallurgy. 1
.
Knowledge of dealing master to apprentice in stition surrounding man lurgical processes until followed by Agricola's "I much knowledge was a(
-
.
a
\v There is very little on the details of operation of heat-treating and testing equipment since they are covered in the laboratory course which is taken
-
m
composition and etched
in conjunction with the theory course. Numerous photomicrograptis have been used to illustrate typical structures. Many tables have been included to present representative data on
i
Until the beginning ol
investigations of metal si ficial. The science of th(
commercial alloys.
The aid received from the following people in reading portions of the manuscript or in preparations of photomicrographs for the first edition is
situation was ripe for tl
.
f
ground was more scienti for the period of rapid d<
"
gratefully iacknowledged: J.E. Krauss, G. Cavaliere, A. Dimond, A. Smith, Cendrowski, J. Sadofsky, C. Pospisil, T. Ingraham, J. Kelch, and O. Kammerer. Many companies have contributed generously from their pub-
A
-
Sorby was an amatei
.
lications and credit is given where possible.
meteorites and then wer
m:-
I make no particular claim for originality of material. The information of other authors and industrial companies has been drawn upon. The only
if
justification for this book, then, lies in the_particular topics covered, their sequence, and the way in which they are presented. I would like to express my appreciation to Miss Barbara Worth for typing
f
number of microscopic;
This paper marks the be the use of the microscc
f
while many people appr were done, none of the
most of the first edition manuscript, to Mrs. Helen Braff and Mrs. Lillian .
independently, and met Additional work by W in metallurgical problerr
Schwartz for typing the second edition material, and finally to my wife, without whose patience and understanding this book could never have been written. .
Sidney H. Avner
In September 1864, S for the Advancement o
I
Steel Institute which s' attention was now ge
lurgists in other countr
if ..
Sauveur convinced Am .
practical tool to aid in l
;7 .
3
About 1922, more kn
was added by the appli
a
f
INTRODUCTION Metallurgy is the science and technology of metals. It is beyond the scope of this text to cover the development of metallurgy as a science. Only certain highlights will be mentioned here for the purpose of orientation.
iraphs added to Chapter 12 iljs. An entire section on titanium
.
The worker of metals is mentioned irfthe Bible and in Greek and Norse
because of their increased com-
mythology. Metallurgy as an art has been practiced since ancient times. i moved next to the one on cor-
Ancient man knew and used many native metals. Gold was used for orna- jV
j of subject matter.
ments, plates, and utensils as early as 3500 b.c. The art of smelting, refining, and shaping metals was highly developed by both the Egyptians and the Chinese. The ancient Egyptians knew how to separate iron from its ore and that steel had the ability to harden, but iron was not used widely
ii of metals in greater detail.
lurgy processing techniques has I
Edition as compared to the first
before 1000 b.c. Iron was not popular with ancient people because of its
tendency to rust, and they preferred working with gold, silver, copper; Knowledge of dealing with metals was generally passed directly from master to apprentice in the Middle Ages, leading to an aura of superstition surrounding many of the processes. Very little was written on metal-
Was not really part of
ure analysis would be of greater in-
b physical metallurgy. ation of heat-treating and testing laboratory course which is taken
lurgical processes until Biringuccio published his "Pirotechnia" in 1540, followed by Agricola's "D« Re Metallurgica" in 1556. In succeeding years, much knowledge was added to the field by people trying to duplicate the
icM'n used to illustrate typical struc-
I
composition and etched structure of Damascus steel. Until the beginning of the last quarter of the nineteenth century, most
fc
ficial. The science of the structure of metals was almost nonexistent. The
) present representative data on
ople in reading portions of the icrographs for the first edition is
investigations of metal structure had been macroscopic;(by eye) and supersituation was ripe for the detailed attention of individuals whose back-
;
'
I
Cavaliere, A. Dimond, A. Smith, T Ingraham, J. Kelch, and O. )uted generously from their pub-
y of material. The information of has been drawn upon. The only
e particular topics covered, their presented. to Miss Barbara Worth for typing Mrs. Helen Braff and Mrs. Lillian
material, and finally to my wife,
iling this book could never have
V
meteorites and then went on to study metals.
In September 1864, Sorby presented a paper to the British Association for the Advancement of Science in which he exhibited and described a
number of microscopical photographs of various kinds of iron and steel. This paper marks the beginning of metallography, the field concerned with the use of the microscope to study the structure of metals. It seems that while many people appreciated the value of Sorby's studies at the time they
were done, none of them had sufficient interest to develop the technique independently, and metallography lay dormant for almost twenty years Additional work by Martens in Germany (1878) revived Sorby's interest in metallurgical problems and in 1887 he presented a paper to the Iron and .
,
.
attention was now generated by both scientists and industrial metal-
lurgists in other countries. In the early part of the twentieth century Albert Sauveur convinced American steel companies that the microscope was a ,
I
f
for the period of rapid development that followed was Henry Clifton Sorb'yife Sorby was an amateur English scientist who started with a study of
Steel Institute which summarized all his work in the field Considerable
1
.
ground was more scientific than practical. The individual most responsible V-
.
le.
,
brass, and bronze.
3 metallurgy by an entirely new chapive metallurgy
"
practical tool to aid in the manufacture and heat treatment of steel. About 1922 more knowledge of the structure and properties of metals ,
was added by the application of x-ray diffraction and wave mechanics
.
Metallurgy is really not an independent science since many of its fundamental concepts are derived from physics, chemistry, and crystallography. The metallurgist has become increasingly important in modern technology. Years ago, the great majority of steel parts were made of cheap lowcarbon steel that would machine and fabricate easily. Heat treatment was reserved largely for tools. Designers were unable to account for structural inhomogeneity, surface defects, etc., and it was considered good practice to use large factors of safety. Consequently, machines were much heavier than they should have been, and the weight was considered a mark of '
quality. This attitude has persisted, to some extent, to the present time but has been discouraged under the leadership of fhe aircraft and automotive
TOC MET
industries. They have emphasized the importance of the strength-weight ratio in good design, and this has led to the development of new high-
strength, lightweight alloys. New technical applications and operating requirements pushed to higher levels have created a continued need for the development of new alloys. For example, an exciting development has been the Wankel rotary enginean internal combustion engine of unusual design that is more compact, lighter, and mechanically far simpler than the ordinary reciprocating piston motor of equivalent horsepower. A particularly bothersome problem has been the seals between the rotor and tfie metal wall. Originally, the seals
The purpose of this chapter of the common tools and te M Temperature Scales In scier
the standard temperature-m
were made of carbon and seldom lasted more than 20,000 miles. Research
ever, in American industria
developed a new sintered titanium-carbide alloy seal which has given lifetimes of up to 100,000 miles. The metallurgical field may be divided into two large groups: Process or extractive metallurgy-the science of obtaining metals from their ores, including mining, concentration, extraction, and refining metals and alloys. 2 Physical metallurgy-the science concerned with the physical and mechanical characteristics of metals and alloys. This field studies the properties of metals and alloys as affected by three variables: 1
exclusively. Therefore all r ,
in terms of the Fahrenheit si countered by the industrial
.
S
other may be made by the fol
.
a
.
Chemical composition-the chemical constituents of the alloy
Mechanical treatment-any operation that causes a change in shape such as rolling, drawing, stamping, forming, or machining Thermal or heat treatment-the effect of temperature and rate of heating c b
.
The accuracy with which will determine the successfi
such as casting smelting, r ,
.
,
and cooling
/
a profound effect on the st
REFERENCES TEMPERATURE MEASUREMENT
Hoover and Hoover: "Georgius Agricola's De Re Metallurgica," Dover Publications, New York, 1912.
Howe, H. M.: The Metallurgy of Steel, The Engineering etnd Mining Journal, 1st ed., New York, 1890. 4
it is necessary to have some Pyrometry deals with the
Rickard, Thomas: "Man and Metalg," McGraw-Hill Book Company, New York, 1932.
erally above 950oF, and Inst
Sauveur, Albert: "The Metallography and Heat Treatmenl of Iron and Steel," 4th ed.,
pyrometers.
McGraw-Hill Book Company, New York, 1935.
Smith and Gnudi: "Pirotechnia of Vannoccia Biringuc0'0-" American Institute of Mining and Metallurgical Engineers, New York, 1943.
Smijh, Cyril Stanley: "A History of Metallography," University of Chicago Press, I960..
Sullivan, F.: "The Story of Metals," American Society for Metals, Metals Park, Ohio, Vlll
In order to understand the (
1951.
\
!
Thermometry deals with th and instruments for this pur[ 1-2 Temperature Measurement by mating the temperature of a There is an apparent correla
T i
.
t
science since many of its funda-
s, chemistry, and crystallography. ingly important in modern techeel parts were made of cheap lowvv-pricate easily. Heat treatment was ;
.
" .-;
B unable to account for structural i it was considered good practice
;
itly, machines were much heavier /eight was considered a mark of me extent, to the present time but
hip of the aircraft and automotive nportance of the strength-weight
TOOLS OFTHE METALLURGIST
o the development of new high"
:
'
" -
ng requirements pushed to higher ir the development of new alloys. s been the Wankel rotary engine-
.
jal design that is more compact, 1 the ordinary reciprocating piston
The purpose of this chapter is to give the student an understanding of some of the common tools and tests that are used in the metallurgical field 11 Temperature Scales In scientific research and in most foreign countries the standard temperature-measuring scale is the centigrade scale How-
i ; iicularly bothersome problem has
.
.
ie metal wall. Originally, the seals
'
'
more than 20,000 miles. Research
. .
,
r
.
ever, in American industrial plants, the Fahrenheit scale is used almost exclusively. Therefore, all references to temperature in this book will be
# |de alloy seal which has given lifei
...u:_u
i
j into two large groups:
in terms of the Fahrenheit scale since this is the one most likely to be en-
iience of obtaining metals from their ion, and refining metals and alloys.
other may be made by the following equations:
countered by the industrial technician. Conversion from one scale to the
led with the physical and mechanical d studies the properties of metals and
°
C = V,rF - 32) -
(1.1)
The accuracy with which temperatures are measured and controlled
constituents of the alloy
that causes a change in shape such
will determine the successful operation of some metallurgical processes
machining
such as casting, smelting refining, and heat treatment. It will also have ,
of temperature and rate of heating
a profound effect on the strength properties of man metals and alloys
.
i
I i
TEMPERATURE MEASUREMENT
1 Re Metallurgica,
"
Dover Publications,
gineering and Mining Journal, 1st ed.,
In order to understand the effect of thermal treatment on the properties it is necessary to have some knowledge of how temperature is measured Pyrometry deals with the measurement of elevated-temperatures gen-
{; v-HIII Book Company, New York, 1932.
erally above 950oF and instruments used for this purpose are known as
.
,
.
,
,
"
it Treatment of Iron and Steel, 4th ed., .
.
Via Biringuccio," American Institute of
" fork, 1943.
'
graphy," University of Chicago Press,
.
pyrometers.
pT
Thermometry deals with the measurement of temperatures below 950oF
,
N
and instruments for this purpose are known as thermometers 1-2 Temperature Measurement by Color One of the simplest methods of esti.
.
-
mating the temperature of a metal is by noting the color of the hot body k ,
Society for Metals, Metals Park, Ohio,
m
There is an apparent correlation between the temperature of a metal and %
I
.
.
i
Ml - . .-
It.
2
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 1 1
Variation
of
Color
Spii
with Temperature
exp
COLOR
N
TEMP., 0F
Faint red
I
950
Dark red
1150
Dark cherry Cherry red Bright cherry Dark orange Orange
1175
1475
Yellow
1800
Filled bulb
Y
Armor-'
1300
covered
capillary
1650 1750
Fig. 1 -2
Simple thermal system for indusl
measurement. (By permission from P. J. O
Instrumentation/' McGraw-Hill Book Comf 1966.)
its color, as sHown by Table 1-1. Except when applied by an experienced observer, this method will give only rough temperature estimates. The principal difficulty is that judgment of color varies with the individual.
Most bimetallic strips hi cient of expansion, and yi tures or a nickel alloy for range of -100 to 1000oF
Other sources of error are that the color may not be uniform and may vary somewhat with different materials.
,
If a continuous indication or recording of temperature is required, then the instruments in use may be divided into two general classifications: (1) mechanical systems that deal essentially with the expansion of a metal, i
tenance. Their main disad\
ing the element in a prote than that of other instrumi
t
a liquid, a gas or vapor; and (2) electrical systems which deal with resistance, thermocouple, radiation, and optical pyrometers. fl-3 Metal-expansion Thermometers Most metals expand when heated, and
1-4
vapor-pressure, consist ol
fH-;:'-; the amount of expansion will'depend upon the temperature and the co-
sured and an expansible indicating pointer or a re connected by capillary tut
efficient of expansion. This principle is incorporated in the bimetallic strip '
A-k;
-
which is used in the common thermostat. The bimetallic strip is made by
bonding a high-expansion metal on one side with a low-expansion metal
The,liquid-expansion tt suitable organic liquid or r liquid to expand or contr expand or contract. Temi case also cause some exp form of compensation is tf pensated liquid-expansion arranged so that motions the temperatures covered
on the other. As a result of small temperature changes, the strip will curve and therefore make or break an electrical circuit which will control the
heating of a house. When it is used as an industrial temperature indicator, the bimetallic strip is usually bent into a coil, one end of which is fixed so that on expan-
sion a rotary motion is automatically obtained (Fig. 1-1). :
' -
.
.!..
Mercury
1
mi
Liquid-expansion Thermome temperature-measuring in
Alcohol Pentane
Temp
Creosote
1-5 *
Flg. 14 Industrial temperature indicator with a helical
bimetallic element. (By permission from P. J. O'Higgins, -
ft '
if
I
35 to + 950°
110to+160ol
+330 to + 85°l
+20 to +400°l
Gas-or Vapor-pressure Ther
;
a volatile liquid partially fil cause corresponding press
'
Sp asic Instrumentation," McGraw-Hill Book Company, iNew York. 1966.)
-
-
liquid surface in the bulb.
TOOLS OF THE METALLURGIST
Spiral expansion element
1
3
Pointer
7
Filled bulb
n
Scale
Armor-' couered
capillary
Fig. 1 2 Simple thermal system for industrial temperature measurement. (By permission from P. J. O'Higgins, "Basic
m
Instrumentation," McGraw-Hill Book Company, New York, 1966.)
!
yhen applied by an experienced gh temperature estimates. The olor varies with the individual
i
1
Most bimetallic strips have Invar as one metal, because of its low coeffi-
I:
cient of expansion, ahd yellow brass as the other metal for low tempera-;-; tures or a nickel alloy for higher temperatures. They can be used in the
.
lay not be uniform and may vary
range of -100 to 1000°F, are very rugged, and require virtually no main-
5 >g of temperature is required,
tenance. Their main disadvantage is that, owing to the necessity for enclosing the element in a protecting tube, the speed of response may be lower
J into two general classifications:
|pl" ysystems with thewhich expansion of a metal, deal with resis-
:
1-4
al pyrometers.
als expand when heated and ,
on the temperature and the co-
corporated in the bimetallic strip The bimetallic strip is made by side with a low-expansion metal
ture changes, the strip will curve al circuit which will control the
erature indicator, the
bimetallic
f which is fixed so that on expan-
lined (Fig. 1-1).
1
than that of other instruments.
I
r
The remainder of the mechanical system temperature-measuring instruments, whether liquid-expansion or gas- or vapor-pressure, consist of a bulb exposed to the temperature to be measured and an expansible device, usually a Bourdon tube, operating an
Liquid-expansion Thermometers
-
indicating pointer or a recording pen. The bulb and Bourdon tube are' connected by capillary tubing and filled with a suitable medium (Fig. 1-2). The liquid-expansion thermometer has the entire system filled with a suitable organic liquid or mercury. Changes In bulb temperature cause the liquid to expand or contract, which in turn causes the Bourdon tube to expand or contract. Temperature changes along the capillary and at the case also cause some expansion and contraction of the liquid, and some form of compensation is therefore required. Figure 1 -3 shows a fully compensated liquid-expansion thermometer using a duplicate system, less bulb, arranged so that motions are subtracted. Some of the liquids used and the temperatures covered by them are; ,
Mercury Alcohol Pentane
Creosote
35 to+950°F
-
110 to+160°F
-
+ 330 to+ 850F
+20 to+400oF
1 -5 Gas- or Vapor-pressure Thermometers In the vapor-pressure thermometer a volatile liquid partially fills the bulb. Different temperatures of the bulb cause corresponding pressure variations in the saturated vapor above the ,
liquid surface in the bulb These pressure variations are transmitted to the .
1
m
; 4 INTRODUCTION TO PHYSICAL METALLURGY
1 -7 Thermoelectric Pyrometer
Bourdon \
lurgical temperature mes
tube
Auxiliary compensating
up to about 3000°F
.
.
/ capillary
The simple thermoelec following units: 1
The thermocouple comp
2 3
The junction block just c The extension leads
,
,
4 The indicating instrumen
Compensating system
.
The operation of this p;
«
3/
Peltier Effect If two diss contact an emf will exist ,
the emf developed will be wires and the temperaturt
Bulb )
Capillary
Thomson Effect
Fig. 1 -3 A fully compensated liquid-expansion thermometer.
If there
single homogeneous wire The magnitude of the em
(From "Teniperature Measurement," American Society for Metals, 1956.)
tion, the chemical uniforn
m
Bourdon tube, the pressure indications acting as a measure of the temperature in the bulb. By suitable choice of volatile liquid, almost any temperature from -60 to +500oF can be measured. Some liquids used are methyl
Seebeck effect is thereto
chloride, ether, ethyl alcohol, and toluene.
of the wires.
The total emf in a th ,
'
emf s at the hot and cole
.
The gas-pressure thermometer is similar to the vapor-pressure ther-
-
The cold junction or ref perature. This is usually ( usually done by means of a ,
mometer except that the system is filled with a gas, usually nitrogen. The range of temperature measured by the gas-pressure thermometer is from
1
200 to+800oF.
its resistance with fluctuat
Filled-system thermometers are used primarily for low-temperature
instrument at 320F.
cooling water and oil temperatures, and subzero temperatures in the cold treatment of metals. These instruments are relatively inexpensive but are not
applications such as plating and cleaning baths, degreasers,
If the
constant temperature thei ,
be a definite function of ti
used where quick repair or exceptionally high accuracy is required. 1 6 Resistance Thermometer The principle of the resistance thermometer Extension leads
depends upon the increase oi electrical resistance with increasing tem
-
perature of a conductor. If the temperature-resistance variations of a metal are calibrated, it is possible to determine the temperature by measuring its electrical resistance. The resistance coil is mounted in the closed end
L
"
of a protecting tube and the leads are extended to a suitable resistancemeasuring instrument, usually a Wheatstone bridge. v
.
.
;i
Resistance coils are usually made of copper, nickel, or platinum. Nickel and copper are most satisfactory for temperatures between 150 and SOO-F, whereas platinum may be used between-350 and+1100oF The resis-
I
K
Indicating -. Instrument
.
tance thermometer is very accurate and is of great importance in the >
m Cold junction
laboratory.-However, its industrial use is limited because it is fragile and requires many precautions in use.
ill J i
i
1
.
Fig. 1-4
A simple thermoelectric pyrometer
I
-
TOOLS OF THE METALLURGIST
if m
iluxiliory "
bmpensQting
Thermoelectric Pyrometer This is the most widely used method for metallurgical temperature measurement and control; it will perform satisfactorily up to about 3000oF. The simple thermoelectric pyrometer, shown in Fig: 1-4 consists of the
t
iopillary
1-7
S
,
following units:
D
1 2
The thermocouple, composed of two different metals or alloys The junction block, just outside the furnace
3
The extension leads
4
The indicating instrument or recorder
btinq syslem
The operation of this pyrometer is based upon two principles: Peltier Effect If two dissimilar metallic wires are brought into electrical contact an emf will exist across the point of contact. The magnitude of the emf developed will be determined by the chemical composition of the
m
,
wires and the temperature of the junction point. Thomson Effect If there is a temperature difference between the ends of a single homogeneous wire an emf will exist between the ends of the wire.
;
,
The magnitude of the emf developed will bo determined by the composition, the chemical uniformity of the wire, and the temperature difference. The total emf in a thermoelectric pyrometer sometimes called the
m iting as a measure of the temperaolatile liquid, almost any temper-
/
:
,
Seebeck effect, is therefore the algebraic sum of four emf's, two Peltier emf s at the hot and cold junctions and two Thomson emf s along each
3d. Some liquids used are methyl a
'
of the wires. .
lilar to the vapor pressure therwith a gas, usually nitrogen. The
The cold junction or reference junction, must be kept at a constant temperature. This is usually 0°C, or 32°F. At the indicating instrument this is usually done by means of a cold-junction compensating coil which changes its resistance with fluctuations in ambient temperature always keeping the instrument at 32°F. If the cold junction or reference junction, is kept at constant temperature, then the measured emf in the pyrometer circuit will be a definite function of the temperature of the hot junction. By suitable
-
jas-pressure thermometer is from
,
4
,
d primarily for low-temperature
ning baths, degreasers, cooling
o temperatures in the cold treat
-
i:;:,
'
,
4 T
relatively inexpensive but are not f high accuracy is required. of the resistance thermometer
Proteclmg tube Extension leads
il resistance with increasing tem-
Thenriocouple
jre-resistance variations of a metal ne the temperature by measuring
u
coil is mounted in the closed end
Junction
block
,
extended '
to a suitable resistance-
tone bridge.
Nickel Stopper, nickel, or platinum. between 150 and SOCF, y peratures
Indicating
en -350 and -MIOOT. The resis-
'
nd is of great importance in the -
is limited because it is fragile and
-
f
=
-* Instrument
Cold junction
Fig. 1 -4 A simple thermoelectric pyrometer
.
Hot
junction
-
.
1
-
1
6 INTRODUCTION TO PHYSICAL METALLURGY
TABLE 1 2 Temperature vs. Electromotive Force
The purpose of the e to a point where the te
*
°
Emf In millivolt*; cold |uncllon 32 F
TEMP, "F
i
CHROMEL
IRON
COPPER
usually not long enougl
VS. PLATINUM
VS.
VS.
VS.
instrument.
ALUMEL
CONSTANTAN
CONSTANTAN
as the thermocouple wii vidual covering color-c
00
00
00
00
100
0 221
1 52
1 94
1 517
200
0 595
3 82
4 91
3 967
300
1 017
6 09
7 94
6 647
400
1 474
8 31
11.03
9 525
500
1 956
10.57
14.12
12.575 15.773
19.100
32
> :
PT+10%RH
.
.
.
.
may be used in some ca
.
.
.
.
.
.
.
.
.
.
.
block instead of the ins
.
.
.
.
The extern
.
1-8
constant temperatures. Thermocouple Materials will develop an emf whe junction points. Industri
600
2 458
12.86
17.18
700
2 977
15.18
20.25
3 506
17.53
23.32
4 046
19.89
26.40
looo
4 596
22.26
29.52
to be measured. The firs
1100
5 156
24.63
32.72
to the positive terminal.
1200
5 726
26.98
36.01
Chromel-Alumel
1300
6 307
29.32
39.43
1400
6 897
31.65
42,96
1500
7 498
33.93
46.53
Alumel (94 percent nick( percent silicon) is one c It has a fairly linear calib
1600
8 110
36.19
50.05
is most useful in the ran(
1700
8 732
38.43
Iron-Constantan
% 1800
9 365
40.62
1900
10.009
42.78
f 2000
10.662
44.91
c
2100
11. 323
47.00
2200
11.989
49.05
percent copper and 46 p the range from 300 to 1 tively low cost, high the atmospheres. Copper-Constantan Tht
2300
12.657
51.05
from that used with Iron
2400
13.325
53.01
80(3
.
.
1
.
.
used for thermocouples electric potential reasc ,
temperature-emf curve
,
i
; '
fi
f; -
;
\
;
.
.
.
.
.
.
.
.
.
2500
13.991
2600
14.656
2700
15.319
2800
15.979
2900
16.637
3000
17.292
54.92
.
m
m
down to -420 . The up[ ;
The above combination
I
Platinum, 10 percent Rh< mocouple. It is used for base-metal thermocouple
"
.By permission from P. H. Dike.
Const:
and iron. This comblnatic
I
m
Chron
"
Thermoelectric Thermometry,
p. 82, Leeds and Northrup Company, 1954.
are not satisfactory. It is i to 3000oF but deteriorates
Thermocouples are me
calibration, it is possible to determine an exact relationship between the developed emf and the true temperature of the hot junction (Table 1-2).
Another1 useful thermoelectric law states that, if a third metal is introduced intb the circuit, the total emf of the circuit will not be affected if the temperature of this third metal is uniform over its entire length. 1
the two wires; the ends ar
or sometimes butted togei head (Fig. 1-5a). The thermocouple wire junction, since contact at
Si
TOOLS pF THE METALLURGIST
1 517
The purpose of the extension leads is to move the reference junction to a point where the temperature will not vary. Therrnocouple wires are usually not long enough nor well enough insulated to run directly to the instrument. The extension leads are usually made of the same material as the thermocouple wires and are placed in a duplex cable with the individual covering color-coded for identification. Copper extension leads may be used in some cases, but then the cold junctions are at the junction
3 967
block instead of the instrument and may be more difficult to maintain at
V
COPPER
DN
VS.
bNSTANTAN .
00
0
.
CONSTANTAN .
94 91
.
.
.
.
94 03
6 647
constant temperatures.
.
9 525
1-8
.
.
12
12.575
.
18
15.773
.
) 25
19.100
.
$ 32 .
3 40 .
Thermocouple Materials Theoretically any two dissimilar metallic wires will develop an emf when there is a temperature difference between.their junction points. Industrially, however, only a few combinations are actually used for thermocouples. These were chosen primarily for their thermoelectric potential, reasonable cost, grain-size stability, linearity of the 1:emperature-emf curve, and melting points higher than the temperature ,
to be measured. The first material in the combination is always connected
9 52 .
to the positive terminal. Chromel-Alumel Chromel (90 percent nickel, 10 percent chromium) vs. Alumel (94 percent nickel, 3 percent manganese, 2 percent aluminum, 1 percent silicon) Is ohe of the most widely used industrial combinations.
12.72 6 01 .
9 43 .
2 96 .
It has a fairly linear calibration curve and good resistance to oxidation. It is most useful in the range from 1200 to 2200oF. Ifon-Constantan Constantan is an alloy containing approximately 54 percent copper and 46 percent nickel. This combination may be used in
6 53 .
0 05 .
-
the range from 300 to 1400°F. The primary advantages are its comparatively low cost, high thermoelectric power, and adaptability to different atmospheres.
Copper-Constantan
The constantan alloy used with copper differs slightly
from that used with iron and may contain small amounts of manganese and iron. This combination is most useful for measuring low temperatures,
down to -420°F. The upper limit is approximately 60b°F. The above combinations are known as base-metal thermocouples. 10 percent Rhodium-Platinum This is a "noble-metal" ther-
Platinum
i ;
p Leeds and Northrup Company.
1954.
,
,
mocouple. It is used for measuring temperatures which are too high for base-metal thermocouples and where radiation or optical pyrometers are not satisfactory. It is suitable for continuous use in the range from 32 to 3000°F but deteriorates rapidly in a reducing atmosphere .
Thermocouples are manufactured by cutting off suitable lengths of e an exact relationship between the jre of the hot junction (Table 1-2).
.
states that, if a third metal is mtrof the circuit will not be affected if the orm over its entire length.
!
the two wires; the ends are carefully twisted together for about two turns
,
or sometimes butted together and welded to form a smooth well-rounded head (Fig. 1-Sa). ,
The thermocouple wirefe should be in electrical contact only at the hot junction, since contact at any other point will usuallyi result in too low a
8
INTRODUCTION TO PHYSICAL METALLURGY
1 -9 Measurement of Emf
The tc
measuring the emf genera most accurate Instrument
4
tially the emf developed b emf and is measured in ter .
'
3 :
calibrated in millivolts or c
;
mi
strument should be used o
calibrated. \
This informat
instrument.
-
A simple direct-indicati Current from the dry cell i slide-wire and an adjustab resistance wire which may of divisions. With the poh potential along the slide-w
[a) I
upon the current flowing t
is of uniform resistance, t division.
In order to stand
to the fixed markings on tl fixed voltage Is connectei standard cell (S.C.) posliic is such that the current fl<
the dry cell. The resistanc
equal, with the net result indicated by zero deflectio dardized so that the poter
corresponds to a definite ;
i ( )
R
.
Fig. 1-5 (a) Examples of properly welded thermocouples. {Hi Different types of porcelain separators. (Leeds & Northfup Company.)
-
Slide wire
'
emf. The two wires are insulated from each other by porcelain beads or ceramic tubes (Fig. 1-56). measured
In most cases, thermocouples are enclosed in protecting tubes. The protecting tubes may be either ceramic or metallic materials. I
:
r
A
The tube
guards the thermocouple against mechanical injury and prevents contamination of the thermocouple materials by the fyrnace atmosphere. A variety of metallic protecting tubes are available such as wrought iron or cast iron (up to 130(rF); 14 percent chrome iron (up to 150p°F); 28 percent chrome iron, or Nichrome (up to 2000oF). Above 2000oF porcelain or
Galvanometer (7) T C .
,
Thermocouple
0 1p
_
-
.
SC .
.
£
,
silicon carbide protecting tubes are used-
Air, .
P
jL
Fig. 1.6 A simple direct-indicating potent
Northrup Company.)
TOOLS QF THE METALLURGIST
1-9
I
,
9
Measurement of Emf The temperature of the hot junction is determined by measuring the emf generated in the circuit. A potentiometer is one of the most accurate instruments available for measuring small emfs. Essen-
tially the emf developed by a thermocouple is balanced against a known !
emf and is measured in terms of this standard. The slide-wire scale may be
IT r
I 1
calibrated in millivolts or directly in temperature. In the latter case, the instrument should be used only with the type of thermocouple for which it is calibrated. This Information is usually stamped on the dial face of the instrument.
A simple direct-indicating potentiometer circuit is shown in Fig 1-6. Current from the dry cell is passed through a main circuit consisting of a slide-wire and an adjustable resistance R. The slide-wire AB is a uniform resistance wire which may be considered as divided into an equal number of divisions. With the polarity of the dry cell as shown there is a drop of potential along the slide-wire from A to 6, the magnitude of which depends upon the current flowing through it from the dry cell. Since the slide-wire is of uniform resistance, there are equal drops of potential across each division. In order to standardize the drop between A and S to correspond to the fixed markings on the indicating dial, a standard cell of known and fixed voltage is connected into the circuit by moving the switch to the standard cell (S.C.) position. Notice that the polarity of the standard cell is such that the current flowing from it opposes the current flowing from the dry cell. The resistance R is adjusted so that these currents are made equal, with the net result that no current flows through the circuit-as indicated by zero deflection of the galvanometer. The circuit is now standardized so that the potential drop across each division of the slide-wire corresponds to a definite amount of millivolts. ,
I:
3»
'
;r i .
'
.
....
i
J>w
-
I:
iaasssaaaaasaaaaaaaaswaaaaaasa*
I R +'
-
Dry cell Slide wire
isulated from each other by porcelain
\
Je enclosed in protecting tubes. The amic or metallic materials. The tube
;
A
l&schanical injury and prevents contamis by the furnace atmosphere.
" .
Galuanometer {/) Standard
A variety
T C .
ailable, such as wrought iron or
cast
*
S.C,
!
cell
Thermocouple
rome iron (up to ISOOT); 28 percent 2000oF). Above 2000oF, porcelain or \ used.
m
.
+
Fig. 1.6 A simple direct-indicating potentiometer (Leeds & Northrup Company ) .
.
i
10 INTRODUCTION TO PHYSICAL METALLURGY
When the emf of the thermocouple is to be measured it replaces the standard cell in the circuit by moving the switch to the thermocouple
where W = rate at which em
(T.C.) position. The thermocouple must be properly connected so that the
7= absolute tempen
,
current flowing from it opposes the flow of current from the dry cell. The circuit is! balanced not by adjustirtg the resistance R but by adjusting the
The apparent temperature
resistance of that portion of the slide-wire which is contained in
the ther-
of the material, which is de
mocouple circuit. This adjustment is made by turning the indicator dial
energy is emitted from the
until the galvanometer reads zero
emitted from a blackbody ai
,
r
K = proportionality c
always be lower than the tr
,
.
At this point the drop of potential ,
through the-slide-wire up to the point of contact is equal to the emf of the thermocouple and the millivolts may be read directly on the slide-wire scale. Reference to a suitable calibration table such as Table 1 -2 for the ,
,
or
,
particular, thermocouple being used will allow the conversion of millivolts or the temperature may be read directly if the dial is so
to temperature
,
where Ta = apparent absolu
calibrated.
pyrometer
Since the cold junction at the instrument is usually higher than the stan
-
et = total emissivity c Therefore, knowing the total eter temperature may be ea
dard cold junction (320F) it is necessary to compensate for this variation ,
.
s
The compensation may be made manually or automatically by a temperature-sensitive resistor called a cold-junction copipensator In contrast ,
.
to most mjaterials
,
that would be read by the p
the cold-junction compensator has a negative tempera-
ture resistance coefficient
.
Figure 1-7 shows a cross
This means that its resistance decreases with
Radiation from the target
increasing temperature
It will, therefore, maintain the cold junction at a constant temperature by balancing any change in resistance as the instru.
ment temperature varies 1-10
focused to form an image
phragmJ. This image is the
.
Recording and Controlling Pyrometer
In most industrial installations
couples called a thermopile
the
,
instrument is required to do more than simply indicate temperature. The pointer of the potentiometer may be replaced by a pen that moves over a traveling chart to obtain a complete record of the temperature This is
determined whether the ims
hole and whether the pyrorr
'
!
.
called a recording pyrometer The instrument through the use of electric .
,
circuits
of the thermopile is approxi energy impinges on it, and 1 practice, however, not all
I I
rnay also be used to control the flow of gas to the burners or electricity to the heating elements and thereby maintain a constant pre,
since some will be absorb instrument. Therefore, th
,
determined furnace temperature This is called a controlling pyrometer It is possible to design the instrument to record and control the temperature from one or more thermocouples .
.
.
1 11 .
closely, and the relation b« and the emf of the thermoc
I
Radiation Pyrometer The basic principles of the operation of the radiation pyrometer, involve a standard radiating source known as a blackbody A .
blackbody is a hypothetical body that absorbs all the radiation that falls
The constants K and b mu
upon it. Such a body radiates energy at a higher rate than any other body at the same temperature Radiation pyrometers are generally calibrated to
at two standardization poir The radiation pyrometer
.
indicate blackbody or true temperatures The Stefan-Boltzmann law which is the basis for the temperature scale of radiation pyrometers, shows that
body, and therefore the upf of the pyrometer itself to \ stops in the optical system temperature limit is appro;
.
,
the rate of radiant energy from a blackbody is proportional to the
fourth
power of its absolute temperature. i
i: ' .
V
t & ( t
f
\
W=KT*
(1-3)
1-12
Optical Pyrometer
The ins
.
1 .
TOOLS OF THE METALLURGIST
i
is to be measured it replaces the
where W= rate at which energy is emitted by a blackbody
,
.
>
/ig the switch to the thermocouple st be properly connected so that the
..
K=
proportionality constant absolute temperature of blafckbody The apparent temperature measured from non-blackbody materials will always be lower than the true temperature. This is due to the emissivity T=
bw of current from the dry cell. The ie resistance R, but by adjusting the [wire which is contained in the ther-
of the material, which is defined as the ratio of the rate at which radiant
made by turning the indicator dial t this point, the drop of potential of contact is equal to the emf of the
energy is emitted from the non-blackbody material to the rate of that emitted from a blackbody at the same temperature. Hence
r ,
i
W=Ke,Tt
|r be read directly on the slide-wire .
V
.;
:
iion table, such as Table 1-2
"
for the
,
/ill allow the conversion of millivolts
pay be read directly if the dial is so
.
(1.5) where Ta
= apparent absolute temperature of non-blackbody measured by pyrometer
e, = total emissivity of non-blackbody knowing the total emissivity of the material, the indicated pyrometer temperature may be easily corrected to the true absolute temperature that would be read by the pyrometer under blackbody iconditions. Figure 1-7 shows a cross section of a mirror-type radiation pyrometer. Radiation from the target passes through window A to mirror 6 and is focused to form an image of the target in the plane of the internal diaTherefore
r
v ; ns that its resistance decreases with
fore, maintain the cold junction at a ly change in resistance as the instru-
,
.
phragm J. This image is then focused by mirror D upon a group of thermocouples called a thermopile E. By viewing hole C through lens H it can be determined whether the image of the target is sufficiently large to cover the hole and whether the pyrometer is properly aimed. The rise in temperature of the thermopile is approximately proportional to the r,ate at which radiant energy impinges on it, and the emf is therefore proportional to T4. In actual practice, however, not all the radiant energy reaches the thermocouple since some will be absorbed by the atmosphere and optical parts of the
In most industrial installations, the
an simply indicate temperature. The eplaced by a pen that moves over a record of the temperature. This is strument, through the use of electric )l the flow of gas to the burners or '
hd thereby maintain a constant pre-
instrument.
s is called a controlling pyrometer. It p record and control the temperature
Therefore, the Stefan-Boitzmann law is not followed very
closely, and the relation between the temperature of the radiating source
and the emf of the thermocouple may be expressed empirically as
pies of the operation of the radiation ig source known as a blackbody. A
E = KV
it absorbs ail the radiation that falls
pyrometers are generally calibrated to ; .
es. The Stefan-Boitzmann law which
'
body, and therefore the upper temperature range is not limited by the ability of the pyrometer itself to withstand high temperature By using suitable
,
% of radiation pyrometers, shows that ickbody
(1-6)
The constants K and b must be determined experimentally by calibration at two standardization points. The radiation pyrometer does not require direct contact with the hot
i at a higher rate than any other body "
(1-4)
i
,:
i
= KTai
or
i
ment is usually higher than the stanary to compensate for this variation. ; j \;anually, or automatically by a temunction compensator. In contrast compensator has a negative tempera-
.
is proportional to the fourth
stops in the optical system there is no upper temperature limit. The lower temperature limit is approximately lOOOT 1-12 Optical Pyrometer The instrument described in the previous section which ,
.
T"
(1-3)
!
si L
i
11
.
P
--
..
""""""~'
V
12
INTRODUCTION JO PHYSICAL METALLURGY
in Fig. 1-8£). These includf the lamp, a rheostat R to ad
G
wire, with associated stand a,
ment current accurately. T
until the filament matches tl
mm
upon and the filament seerr then obtained by rotating / attached to the potentiomet The temperature range o to about 2400 . This upp< of the filament at higher ter the eye of the brightness at may be extended upward b jective lens and the filame
F
H
D
J
A
if
1 I c
E
secured at lower filament ti
brated for the higher tempt peratures. Thus, by using v
I
optical pyrometer can be e> Red glass
r
Eyepiece
Fig. 1-7 A mirror-type radiation pyrometer (Leeds & Northrup Company.) .
e
responded to all wavelengths of radiation is known as a total-radiation pyrometer. While the general principles on which the optical pyrometer is based are the same as for the radiation pyrometer, they differ in that the optical pyrometer makes use of a single wavelength or a narrow band of wavelengths in the visible part of the spectrum. The optical pyrometer measures temperature by comparing the brightness of light ernitted by the source with that of a standard source. To make the color comparison
Hood
i
[TjLamp
Tempera
rrrr
AWW
The type most widely used in industry is the disappearing-filament type.
-
i
telescope (Fig. 1-8a) contains a red-glass filter mounted in front of the
J
.
;
;J
'
:
.
:
r
W
Galv.
This pyrometer consists of two parts, a telescope and a control box. The
-
.
-W
Filamer rh
easier, ajred filter is used which restricts the visible radiation to only the wavelength of red radiation. >
1
-
eyepiece ;and a lamp with a calibrated filament upon which the objective
Switch on
liens focuses an image of the body whose temperature is being measured. It also contains a switch for closing the electric circuit of the lamp and an absorbing screen for changing the range of the pyrometer.
telescope
0-®" Std cell
it)) Circuit
,
Fig. 1 -8 The disappearing-filament type of i eter. (a) Telescope; (b) circuit diagram; (c) fi ance. (Leeds & Northrup Company.)
The cqntrol box contains the main parts of the measuring circuit shown 6
mm
TOOLS OF THE METALLURGIST
13
in Fig. 1-86. These include dry cells to provide the cufrent to illuminate the lamp, a rheostat R to adjust filament current and a potentiometer slide,
wire, with associated standard cell and galvanometer, to measure the fila-
ment current accurately. This current is manually adjusted by rotating R-, until the filament matches the brightness of the image ofitheobject sighted upon and the filament seems to disappear (Fig. 1 -Sc). Accurate balance is then obtained by rotating P, until the galvanometer reads zero. A scale
i F
attached to the potentiometer contact P indicates the temperature directly. The temperature range of the optical pyrometer described is from 1400 to about 2400oF. This upper limit is due partly to danger of deterioration of the filament at higher temperatures and partly to the dazzling effect on the eye of the brightness at elevated temperatures. The temperature range may be extended upward by use of an absorbing screen between the objective lens and the filament, thus permitting brightness matches to be secured at lower filament temperatures. The pyrometer can then be calibrated for the higher temperature range by using the lower filament tem-
D
J
i
i c
1
sv. A 'i.Q
E
:
1
i
mJ
peratures. Thus, by using various absorbing screens, the upper limit of the optical pyrometer can be extended to 10,000oF or higher.
ft. V
Screen
Red glass
Objective
Eyepiece
E
5 1
/ J
a
\ ation is known as a total-radiation
Screen
3
Hood
shifting
les on which the optical pyrometer
on pyrometer, they differ in that the jle wavelength or a riarrow band of
'
i spectrum. The optical pyrometer the brightness of light emitted by irce. To make the color comparison
device
Lamp Switch
i I \
icts the visible radiation to only the
[a] Telescope ff
1
llllll,-i-JWvWWVr-| Filament current
\ Knob moves contacts
\ /f an(j p] simultaneously. I
rheostat
I
I Filament too dark n
..
J Lamp Temperature scale ri.l.i.u.i.i.u.ii
:
ry is the disappearing-filament type.
'
i
v>-
-
Xv a telescope and a control box. The "
.
v
.
Galv.
WWWW.
0I
'
1
u
/
Filament too bright
--
==
---
-
I -
'
JWWV
-
glass filter mounted in front of the d filament upon which the objective
OJ
\CJ
cell
Switch on
V . lose temperature is being measured.
telescope
le electric circuit of the lamp, and an nge of the pyrometer.
i
parts of the measuring circuit shown
i
\iy Std
Knob moves potentiometijr contact Z3, only-is pressed to close switch 5
"
Filament correct
id] Circuit diagram
(c) Filament appearance
Fig. 1 -8 The disappearing-filament type of optical pyrometer. (a) Telescope; (6) circuit diagram; (c) filament appearance. (Leeds & Northrup Company ) .
m I
-
r -
14
INTRODUCTION TO PHYSICAL METALLURGY
Some; advantages of the optical and radiation pyrometers are:
i
1
a specimen is comparative! developed only after constat
Measurement of high temperature.
duce a flat, scratch-free, min
2
Measiirement of inaccessible bodies
3
Measurement of moving or small bodies
4
No part of the instrument is exposed to the destructive effects of heating
Li .,
i
.
a metallographic specimen p
.
f;M4 Sampling
.
portant. If a failure is to be
The principal-disadvantages are:
I
1
as close as possible to the ar
Errors! introduced because the photometric match is a matter of individual
taken from the normal sectic If the material is soft, such
judgments
2
The choice of a sa
Errors introduced by smoke or gases between the observer and the source
.
treated steels, the section r
3 Uncertainty as to the amount of departure from blackbody conditions
.
the material is hard, the secti
off wheel. This wheel is a th METALLOGRAPHY 1-13 Introduction
highspeed. The specimen sh>
Metallography or microscopy consists of the microscopic
1-15
Whenever p( is convenient to handle. Asi
Rough Grinding
'
study of the
structural characteristics of a metal or an alloy The microscope is by far the most important tool of the metallurgist from both the .
it up and back across the sl specimen may be rough-groi cool by frequent dropping ii grinding and polishing opei pendicular to the existing sc stage when the deeper sera
1 scientific and and the size
technical standpoints. It is possible to determine grain size shape, and distribution of various phases and inclusions which have a great effect on the mechanical properties of the metal The ,
.
microstrlicture will reveal the mechanical and thermal treatment of the
' '
V -
i
" -
-
'
metal, arjd it may be possible to predict its expected behavior under a given
-
.
-
characteristic of the finer ab the surface is flat and free of hacksaw or cutoff wheel are
set of conditions.
Experience has indicated that success in microscopic study depends
I
largely upon the care taken in the preparation of the specimen
.
The most
grinding is shown in Fig. 1-1
expensive microscope will not reveal the structure of a specimen that has
been pobrly prepared. The procedure to be followed in the preparation of
j;
1-16
Mounting
Specimens that a
mounted to facilitate intermi
sheet metal specimens, thin
i
in a suitable material or rigic
Synthetic plastic materials 1
mounts of a uniform conve 0
,
,
s
m I
1
mm
a
I
-
. .
((7) Fig. 1 -9 1
reached, the specimen moi
,
(Q Specimen mounted in Lucite, enlarged 2X. (c) Specimen held in metal clamp, enlarged 2X.
I i
m
W
(a) Specimen mounted in Bakellte enlarged 2X.
diameter) for handling in sut when properly made, are ve ordinarily used. The most Bakellte, Fig. 1-9a. Bakellte colors, which simplifies th( specimen and the correct art are placed in the cylinder of ally raised to ISO , and a i simultaneously. Since Bakt
f i
it is still hot. Lucite is the most commc
_
i TOOLS OF THE METALLURGIST
tS . . .
;
iadiation pyrometers are: -
a specimen is comparatively simple and involves a technique which is developed only after constant practice. The ultimate objective is to pro-
. 1
duce a flat
scratch-free, mirrorlike surface. The steps required to prepare a metallographic specimen properly are covered in Sees. 1 -14 to 1 -19. ,
I;
le destructive effects of heating.
1-14
i
etric match is a matter of individual
taken from the normal section.
If the material is soft
such as nonferrous metals or allocs and non-heattreated steels, the section may be obtained by manualjihacksawing, If
|/veen the observer and the source.
,
5 from blackbody conditions.
.
. . .
Sampling The choice of a sample for microscopic study may be very important. If a failure is to be investigated, the sample should be chosen as close as possible to the area of failure and should be compared with one
the material is hard, the section may be obtained by use of an abrasive cutoff wheel. This wheel is a thin disk of suitable cutting abrasive, rotating at high speed. Thespecimen should be kept cool during the cutting operation. 115 Rough Grinding Whenever possible, the specimen should'be of a size that
opy consists of the microscopic )f a metal or an alloy. The micro1 of the metallurgist from both the is possible to determine grain size of various phases and inclusions
is convenient to handle, A soft sample may be made flat by slowly moving it up and back across the surface of a flat smooth file. The soft or hard
specimen may be rough-ground on a belt sandor, with thife specimen kept cool by frequent dropping in water during the grinding operation. In all
grinding and polishing operations the specimen should! be moved perpendicular to the existing scratches. This will facilitate recognition of the
inical properties of the metal. The
;
cal and thermal treatment of the
ts expected behavior under a given
stage when the deeper scratches have been replaced by shallower ones characteristic of the finer abrasive. The rough grinding is continued until
sss In microscopic study depends aration of the specimen. The most e structure of a specimen that has o be followed in the preparation of
the surface is flat and free of nicks, burrs, etc., and all scratches due to the
1-16 -
hacksaw or cutoff wheel are no longer visible. (The surface after rough grinding is shown in Fig. 1-10a.) Mounting Specimens that are small or awkwardly shaped should be mounted to facilitate intermediate and final polishing. Wires, small rods,
sheet metal specimens, thin sections, etc., must be appropriately mounted in a suitable material or rigidly clamped in a mechanical mount. Synthetic plastic materials applied in a special mounting press will yield mounts of a uniform convenient size (usually 1 in. 1.25 in., or 1.5 in. in ,
i
diameter) for handling in subsequent polishing operations. These mounts, when properly made, are very resistant to attack by the etching reagents ordinarily used. The most common thermosetting resin for mounting is Bakelite, Fig. 1-Qa. Bakelite molding powders are available in a variety ol colors, which simplifies the identification of mounted specimens. The specimen and the correct amount of Bakelite powder or a Bakelite preform, are placed in the cylinder of the mounting press. The temperature is gradually raised to 150oC, and a molding pressure of about 4 000 psi is applied simultaneously. Since Bakelite is set and cured when thi's temperature is reached, the specimen mount may be ejected from the molding die while
-
,
,
ic)
it is still hot.
Ij
--
Lucite is the most common thermoplastic resin for mounting. Lucite is I
i
.
1
:
n
INTRODUCTION TO PHYSICAL METALLURGY
completely transparent when properly molded, as shown In Fig. 1 -db. This transparency is useful when it is necessary to observe the exact section that is being polished or when it is desirable for any other reason to see the entire spepimen in the mount. Unlike the thermosetting plastics, the thermoplasticresins do not undergo curing at the molding temperature; rather
they set on cooling. The specimen and a proper amount of Lucite powder are placed in the mounting press and are subjected to the same temperature and pressure as for Bakelite (150oC and 4,000 psi).
|,
After this tem-
perature has been reached, the heating coil is removed, arid cooling fins are placed around the cylinder to cool the mount to below 750C in about
J
'
:
7 min while the molding pressure is maintained. Then the mount may be ejected from the mold. Ejecting the mount while still hot or allowing it to 1 cool slowly in the molding cylinder to ordinary temperature before ejection will cause the mount to be opaque. Small specimens may be conveniently mounted for metallographic preparation in a laboratory-made clamping device as shown in Fig. 1-9c. Thin sheet specimens, when mounted in such a clamping device, are usually alternated with metal filler" sheets which have approximately the same hardness as the specimens. The use of filler sheets will preserve surface irregularities of the specimen and will prevent, to some extent, the edges of the specimen from becoming rounded during polishing. Intermediate Polishing After mounting the specimen is polished on a series of fernery papers containing successively finer abrasives. The first paper is usually No. 1, then 1/0, 2/0, 3/0, and finally 4/0. "
'
1
-17
ii
,
The surface after intermediate polishing on 4/0 paper is shown in Fig. The intermediate polishing operations using emery paper are usually done dry; however, in certain cases such as the preparation of soft materials, silicon carbide abrasive may be used. As compared to emery 1-10/j
1-18
.
paper, silicon carbide has a greater removal rate and, as it is resin-bonded,
Fig. 1-10
can be used with a lubricant. Using a lubricant prevents overheating the
100X. (t>) Surface after intermediate polishing
sample, minimizes smearing of soft metals, and also provides rinsing action to flush away surface removal products so the paper will not become clogged. ;
magnification 100X. (c) Scratch-free surface al polishing, magnification SOX. Black spots are
Fine Polishing The time consumed and the success of fine polishing depend iargjely- upon the care that was exercised during the previous polishing steps. The final approximation to a flgt scratch-free surface is obtained by use of a wet rotating wheel covered vyith 3 special cloth that is charged with carefully sized abrasive particles. A wide range of abrasives is available for final polishing. While many will do a satisfactory job, there appears to be a preference for the gamma form Of aluminum oxide for ferrous and copper-based materials, and cerium oxide for aluminum, magnesium, and their alloys. Other final polishing abrasives often used are diamond paste, chromiunj oxide, and magnesium oxide.
I 1
(a) Surface after rough grinding, ms
impurities.
The choice of a proper pol
terial being polished and the cloths are available of varyinj as silk, to those of intermec '
and canvas duck and finally ,
ishing cloths are also availal two, under the trade names
used. A properly polished sar and will be scratchf ree (Fig. 1
i
"
£
:v
.
--
.
TOOLS OF JHt METALLURGIST .
tiolded, as shown in Fig. 1-9£). This
I
'
ssary to observe the exact section "'i able for any other reason to see the
j '
17
-
i
'
.
-
tie thermosetting plastics, the therat the molding temperature; rather I a proper amount of Lucite powder ire subjected to the same temperaPC and 4 000 psi). After this tem-
P I
,
g coil is removed, and cooling fins
?ilri
the mount to below 75°C in about
.
.
iiSaintained. Then the mount may be aunt while still hot or allowing it to
&
11'"' J
f 'J
irtfm
f'A
mm \
I
'
rdinary temperature before ejection &
y mounted for metallographic prep'
device as shown in Fig. 1-9c. Thin M&ch a clamping device, are usually -
.
.
.
-
:
'
:
' -
" -
|hich have approximately the same pf filler sheets will preserve surface the edges
?-]prevent, to some extent, ted during polishing.
'
'
C
-
| -f »
i . p &
l the specimen is polished on a ;| jcessively finer abrasives. The first f"
'0
,
and finally 4/0.
.
.
.
hing on 4/0 paper is shown in Fig.
-
.
*
*
rations using emery paper are usu-
es such as the preparation of soft
y be used. As compared to emery
i
icval rate and, as it is resin-bonded,
1
lubricant prevents overheating the
{
netals, and also provides rinsing oducts so the paper will not become
terial being polished and the purpose of the metallographic study Many cloths are available of varying nap or pile from those having no pile, such
i flat scratch-free surface is obtained
.
iviiv) with a special cloth that is charged
,
as silk, to those of intermediate pile such as broadcloth, billiard cloth,
! A wide range of abrasives is availxiyM do a satisfactory job, there appears
,
and canvas duck, and finally to a deep pile such as velvet. Synthetic pol-
-
-v;n of aluminum oxide for ferrous and
v
-
xide for aluminum, magnesium, and
jsives often used are diamond paste, le. !
final
The choice of a proper polishing cloth depends upon {the particular ma-
|
'
'
magnification lOOX. (c) Scratch-free surface after
polishing, magnification SOX. Black spots fire oxide impurities.
d the success of fine polishing de-
.
Fig. 1-10 (a) Surface after rough grinding, magnification 100X. (b) Surface after intermediate polishing on 4/0 paper,
,
1
ishing cloths are also available for general polishing plurposes of which two, under the trade names of Gamal and Microclothi1 are most widely .
,
used. A properly polished sample will show only the nonrtietallic inclusions
and willbescratchfree(Fig. 1-10c).
|
i S
18
INTRODUCTION TO PHYSICAL METALLURGY
1-19
In alloys composed of tw
Etching Tihe purpose of etching is to make visible the many structural characteristics of the metal or alloy. The process must be such that the various |Sarts of the microstructure may be clearly differentiated. This is
during etching by a prefere by the reagent, because i phases (Fig. 1-Ha). In unit is obtained and grain boum
accbmpliished by use of an appropriate reagent which subjects the polished surface tp chemical action.
the rate at which various c I
This difference in the rate .
V
r
the different grain sections chemical attack by the etch valleys in the polished surf
Wm m
11 1, !
.
1
V
mm
.
j
of these valleys will be refl boundaries appear as dark
1>.
I
V
1-11c
.
The selection of the apt: metal or alloy and the spe
TO' ;-
lists some of the common <
1 20 Metallurgical Microscopes the principles of the metall logical type, the metallurgi the specimen is illuminatet light, the sample must be i
\
if-
.
i
v
m
1-12
,
M s
V
'
mm Microscope
a horizontal beam o'
means of a plane-glass ref jective onto the surface of fleeted from the specimen j lower lens system, the ob plane-glass reflector and bi eyepiece. The initial magr is usually engraved on the objective and eyepiece is u fication is equal to the pro the eyepiece. Figure 1-13£ It is possible to mount a table-type microscope for metallograph illustrated in both visual examination am
/
? : i =V .
/surface
which limits the resolution
Grain boundary
'
.
tures by photographic metl The maximum magnifici about 2,000x. The princip
/ Polished
.
; \ .
.
Fig. 1-11 (a) Photomicrograph of a mixture revealed by etching, (b) Photomicrograph of pure iron. (The International Nickel Company.) (c) Schematic illustration of the
(c)
4
microscopic appearance of grain boundaries as dark lines.
Jfai ilnfl
i ;
.
#1
I I
The magnification may be t length radiation, such as l technique is more involved The greatest advance in
TOOLS OF THE METALLURGIST
19
;
In alloys composed of two or more phases, the components are revealed ,
pake visible the many structural ;
during etching by a preferential attack oj one or more of ;these constituents
he process must be such that the
by the reagent, because of difference in chemical composition of the.' phases (Fig. 1 -11 a). In uniform single-phase alloys or pure metals, contrast is obtained and grain boundaries are made visible because of differences in the rate at which various grains are attacked by the reagent (Fig. 1-116).
be clearly differentiated. This is lished
eagent which subjects the po
This difference in the rate of attack is mainly associate with the angle of the different grain sections to the plane of the polished surface. Because of chemical attack by the etching reagent, the grain boundaries will appear as valleys in the polished surface. Light from the microscope hitting the side of these valleys will be reflected out of the microscope; making the grain boundaries appear as dark lines. This is illustrated schematically in Fig.
1 t
IHc.
The selection of the appropriate etching reagent is determined by the metal or alloy and the specific structure desired for viewing. Table 1-3
i
lists some of the common etching reagents. 1 20
Metallurgical Microscopes At this point it is appropriate to discuss briefly the principles of the metallurgical microscope. In comparison with a bio-
logical type, the metallurgical microscope differs in thef manner by which
i
the specimen is illuminated. Since a metallographic sample is opaque to
4
light, the sample must be illuminated by reflected light/ As shown in Fig. 1-12
,
en
«5 I'
a horizontal beam of light from some light sour'pe is reflected, by
means of a plane-glass reflector, downward through the microscope objective onto the surface of the specimen. Some of this incident light re-
flected from the specimen surface will be magnified in plassing through the Ufc.
lower lens system the objective, and will continue upward through the plane-glass reflector and be magnified again by the upper lens system, the eyepiece. The initial magnifying power of the objective and the eyepiece ,
is usually engraved on the lens mount. When a particular combination of objective and eyepiece is used at the proper tube length the total magni-
1
,
fication is equal to the product of the magnifications o\ the objective and
Microscope
the eyepiece. Figure 1'13a shows a table-type metallurgical microscope. It is possible to mount a camera bellows above the eyepiece and use the table-type microscope for photomicrography. However the bench-type ,
metallograph illustrated in Fig. 1-13b which is specifically designed for ,
both visual examination and permanent recording of metallographic structures by photographic methods will give superior photomicrographs.
Polished
,
The maximum magnification obtained with the optical microscope is about 2 000x. The principal limitation is the wavelength of visible light which limits the resolution of fine detail in the metallographic specimen The magnification may be extended somewhat by the use of shorter-wavelength radiation such as ultraviolet radiation, but the sample preparation
j< surface
m
,
Grain boundary yii 4
,
: .
,
.
technique is more'involved.
,
The greatest advance in resolving power was obtained by the electron
vv
I
i
'
20 INTRODUCTION :TO PHYSICAL METALLURGY s
Retina
Retinal image (Erect and unreversed)
Human eye ris
I
Cornea
Entrance pupil af eye Eye lens
/
I
I /
Eyejiiece diaphragm-
mmm ) Huygenian eyepiece
Tl
First focal point of $ye lens
Primary real image
of object formed by objective and field
// //
Field lens
lens at principal focal point of eye lens, or
/ /
within focus distance
// \
//
1
as illustrated (Image inverted and reversed)
\
//
\
//
\
// //
//
\
/
II //
i I
From light
Plane gloss
/I II
V
reflector \
reflector
source
4
Ij Back focal
//
point of -
\
> Objective
w
i
// w
Front focal point of objective
Object area of
Image. (By permission from G. L. Kehl, "Principles of Metallographic Laboratory Practice," 3d ed., McGraw-Hill Book Company, New Yorfc, 1949.)
the shadowed replica is fragil
unreversed)
ll
i
(Inverted and reversed)
K
optical system fromlthe object field to the final virtual
( I
J
i
I .
i
.
\\
Final virtual image
Fig. 1 -12 Illustrating the principle of the metallurgical compound microscope and the trace of rays through the
:? >.
I
of the coils, and the image strength of the coils while the microscope the image is bro
Since metallographic spec is necessary to prepare, by sp to be studied. The specim( metallographic practice. It is of suitable plastic on the etc plastic begins to flow and pr between the plastic and the fully peeled off. To improve sten is evaporated onto the t
specimen (Erect and
Metallographic specimen
i
,
The lenses of the electron
/
i
microscope. Under certain c like light of very short wavele it a wavelength nearly 100 00 light, thus increasing the rej croscope is shown in Fig. 1-' Although in principle the e croscope (Fig. 1-14£)), its ap larger because of the highly produce and control the elei pumped to a high vacuum si electrons.
ob|ective
/ /
Fig. 1-13 (a) Metallurgical microscope, (fa) Bi metallograph. (Bausch & Lomb, Inc.)
TOOLS OF THE METALLURGIST
' "
21
Retinal Image
)
(Erect and unreversed)
;
Cornea
Entrance pupil of eye
I
> Huygenion eyepiece Primary real image
of object formed by
\\
objective and field
\\\
lens at principal focal
w
point of eye lens, or
[o)
within focus distance
( )
as illustrated (Image
.
inverted and Reversed)
\ \
\
Fig. 1-13 (a) Metallurgical microscope, (b) Bench-type metallograph. (Bausch & Lomb, Inc.)
microscope. Under certain circumstances, high-velocity electrons behave like light of very short wavelength. The electron beam has associated with it a wavelength nearly 100,000 times smaller than the wavelength of visible light, thus increasing the resolving power tremendously. An electron microscope is shown in Fig. 1 -143. Although In principle the electron microscope is similar to the light microscope (Fig. 1-14to), its appearance is very much different. It is much larger because of the highly regulated power supplies that are needed to produce and control the electron beam. The entire system must be kept pumped to a high vacuum since air would interfere with the motion of the
I \
\
\ \
\ \ \
n f
From light source
> Objective
J
\
electrons.
i \\
\V J
.
fle
\
j 1
e '
'
. .
I
etal-
: r
The lenses of the electron microscope are the powerful magnetic fields of the coils, and the image is brought into focus by Changing the field strength of the coils while the coils remain in a fixed position. In the optical microscope the image is brought into focus by changing the lens spacing. Since metallographic specimens are opaque to an electron beam, it is necessary to prepare, by special techniques, a thin replica of the surface . to be studied. The specimen is polished and etched following normal metallographic practice. It is then placed on a hot plate with a small pellet of suitable plastic on the etched surface. As the temperature rises, the plastic begins to flow and pressure is applied to ensure intimate contact between the plastic and the surface. After cooling, the replica is carefully peeled off. To improve contrast, a thin coating of carbon or tungsten is evaporated onto the replica at an angle and from one side. Since the shadowed replica is fragile, it is supported on a disk of very fine copper'
Object area of
specimen (Erect and unreversed)
-
!
[
i
' .
.
!,.: :, .:
TABLE 1-3
Etching Reagents for Microscopic Examination*
ETCHING REAGENT
COMPOSITION
Nitric acid
White nitric acid
(nital)
Ethyl or methyl alcohol (95% or absolute)
USES
REMARKS
1 -5 ml
In carbon steels: (1) to darken
Etching rate is increase'd selectivity
100 ml
pearlite and give contrast between pearlite colonies, (2) to reveal ferrite boundaries, (3) to dif-
decreased, with increasing percentages of HNO3. Reagent 2 (picric acid) usually superior
ferentiate ferrite from martensite
Etching time a few seconds to 1 min
H 31
(also amyl alcohol)
..
Picric acid
(picral)
Ethyl or methyl alcohol (95% or absolute)
Ferric chloride
Ferric chloride and
100 ml
annealed, normalized, quenched,
useful. Does not reveal ferrite grain
and tempered, spheroidized, austempered. For all low-alloy steels attacked by this reagent
boundaries as readily as nital Etching time a few seconds to 1 min or more
O c
o H
o z H
o I
<
o > v
S m
and stainless steels
50 ml
Water
More dilute solutions occasionally
Structure of austenitic nickel
5 g
Hydrochloric acid
hydrochloric acid
For all grades of carbon steels;
4g
Picric acid
,
O
> r;
100 ml
c
Ammonium hydroxide and hydrogen peroxide
Ammonium hydroxide
5 parts
Water
5 parts
Hydrogen peroxide Ammonium persulfate
2-5 parts
Ammonium persulfate
10 g
Copper, brass, bronze, nickel
Water
90 ml
silver, aluminum bronze
Chromic oxide Sodium sulfate
Palmerlon reagent
200 g 15 g
Water
Ammonium molybdate
Generally used for copper and many of its alloys
Peroxide content varies directly with copper content of alloy to be etched Immersion or swabbing for about 1 min. Fresh peroxide for good results
Use either cold or boiling; immersion
General reagent for zinc and its alloys
Immersion with gentle agitation
Rapid etch for lead and its alloys; very suitable for removing thick layer of worked metal
Alternately swab specimen and wash in running water
General microscopic for aluminum and its alloys
Swab with soft cotton for 15 s
1,000 ml
Molybdic acid (85%) Ammonium hydroxide (sp gr 0.9)
100 g
Water
210 ml
140 ml
Filter and add to nitric
acid (sp gr 1.32) Hydrofluoric acid
. From
"
60 ml
Hydrofluoric acid (cone) HzO
0.5 ml 99.5 ml
Metals Handbook," 1948 ed., American Society for Metals. Metals Park, Ohio.
a m 2
-
.
o
i
rn
5 fD
.
q
3
CO CO
CD
/
o
3 3 o
o
i
a
.Z!
1 id
a .
J=3
n
,
2
.
4
jD
o <
-
im-
imwi-
i4:
TOOLS OF THE METALLURGIST 23
1
o
5
0)
"
o
X
t5
lO
E o
o 0
w 0)
o
01
0)
r
o
o
;
II
4
E
0) 2
CO
5
G
9 o 0
-
c
a
o
E
N
o
o
:
s
,
o g 10
O
E
3
S
x:
J3
o
o
ro
5
E
'
B = .s
!
1
2 g E
I ST i
Si 5l
-
1
a>
=j
CD
c
6
6
Electron microscope Source of illumination
Anode-1E
E
E
E
§ §
O CD
LO
LO
E
«
Light microscope
Cathode I j (Electrons)
n
Lamp
[Light]
'
? g g CM
s s O
d oi
OJ
Oi
0)
S
Condenser lens
w ra
s in
Ml Q
o
QJ
I
5
(Magnetic)
S
[Glass) S
o
S 1 O CO
Specimen stage -
M
.
x 2=
S E
"
a. |
ni o
Objective lens E <
<5 oco5 5 < a 5 ll. « x iii
(Magnetic)
iGlass)
d
- Projector lens 1 Magnetic)
(Glass
o *-
5 .
;/
I
.
!
'
j E
il
a
OI
o
(0
E
I
a
o
.
o
CO
Image
«
o
|
o
ro
(Viewing Screen)
s
(Eye piece)
t a)
E
o
E E <
[t]
E .
o
o
Fig. 1 -14 (a) The electron microscope (b) Similarity of light ,
and electron microscopes (Radio Corporation of America )
!
.
.
* .
i
'
-
yr-..
24
INTRODUCTION TO PHYSICAL METALLURGY
wire mesh. The disk is then placed over the opening in the specimen holder
,
which is inserted in the column of the instrument
.
i
The electrons emitted by a hot tungsten-filament cathode are accelerated, to form a high-velocity beam by the anode. This beam is concentrated on the replica by the condensing lens Depending upon the density and thickness of the replica at each point some of the electrons are absorbed or scattered while the remainder pass through The magnetic field of the objective lens focuses and enlarges the electron beam that has passed through the replica. Some of the electrons in this image are brought into a second focus on a fluorescent screen by the projector lens The electron microscope shown in Fig 1-14a has a basic magnification
4/
S3
,
m
.
s
*
,
I
.
mm
lllli
.
.
r
range of 1,400 to 32 000x, which may be extended to 200 000>< with accessory lenses. ,
,
9 _
*
/
.
A
Iji VESTS FOR MECHANICAL PROPERTIES %
1-21 Hardness The property of "hardness" is difficult to define except in reia-
fi
tion to the particular test used to determine its value. It should be observed that a hardness number or value cannot be utilized directly in design, as
mm
t
1!,' ,
can a tensile strength value, since hardness numbers have no intrinsic significance. Hardness is not a fundamental property of a material but is related to the elastic and plastic properties. The hardness value obtained in a particular test serves only as a comparison between materials or treatments. The test procedure and sample preparation are usually simple, and the results may be used in estimating other mechanical properties. Hardness testing is
'
;
"
'
;
Fig. 1-15
1
ment & Manufacturing Company.)
widely used for inspection and control. Heat treatment or working usually results in a change in hardness. When the hardness resulting from treating a given material by a given process is established, it affords a rapid and simple means of inspection and control for the particular material and
J
;
y
'"
:
of potential energy. When it energy until it strikes the su
now absorbed in forming tl
hammer for its rebound. The
process.
I;
Scleroscope hardness tester. (The
an arbitrary scale such that and the harder the test piece
The various hardness tests may be divided into three categories: Elastic hdrdness
of a material, that is, the enc
1 -23 Resistance to Cutting or Abr;
Resistance to cutting or abrasion
Scratch Test
Resistance to Indentation
This test was
sists of 10 different standard 1-22
Elastic Hardness This type of hardness is measured by a scleroscope (Fig. 1 -15). which is a device for measuring the height of rebound of a small diamond-tipped hammer after it falls by its own weight from a definite height onto the surface of the test piece. The instrument usually has a selfindicating dial so that the height of rebound is automatically indicated. When the hammer is raised to the starting position, it has a certain amount
(
)
ness. Talc is No. 1, gypsurr V
mond.
If an unknown matt
.
by No. 5, the hardness value used to any great extent in i primary disadvantage is thai hardness of the minerals i;
i
'
TOOLS OF THE METALLURGIST
25
the opening in the specimen holder, .
instrument.
gsten-filament cathode are accel"
by the anode. This beam is conjing lens. Depending upon the den;h point, some of the electrons are inder pass through. The magnetic
3 -
0
Enlarges the electron beam that has of the electrons in this image are
escent screen by the projector lens. ,
Jg. 1-14a has a basic magnification
'
fc
. .
.
s
.
#Me extended to 200,000x with accesm
s
:
Is difficult to define except in relanine its value. It should be observed v;.?:jiot be utilized directly in design, as hardness numbers have no intrinsic '
irty of a material but is related to the dness value obtained in a particular jen materials or treatments. The test
usually simple, and the results may
J
-
1 Ik
1f
il
'
PiQ-I IS
Scleroscope hardness tester (The Shore Instru.
if .<. rn8nt & Manufacturing Company.)
cal properties. Hardness testing is Heat treatment or working usually n the hardness resulting from treat-
I
of potential energy. When it is released this energy is converted to kinetic energy until it strikes the surface of the test piece Some of the energy is now absorbed in forming the impression and the rest is returned to the
.
,
4 is established, it affords a rapid and
itrol for the particular material and
.
,
I
hammer for its rebound. The height of rebound is indicated by a number on an arbitrary scale such that the higher the rebound the larger the number and the harder the test piece This test is really a measure of the resilience of a material, that is the energy it can absorb in the elastic range 123 Resistance to Cutting or Abrasion Scratch Test This test was developed by Friedrich Mohs The scale con,
ivided into three categories: r
.
,
.
.
sists of 10 different standard minerals arranged in order of increasing hard-
viess is measured by a scleroscope
uring the height of rebound of a small ;
rijs by its own weight from a definite : ice. The instrument usually has a self-
j rebound is automatically indicated. rting position, it has a certain amount
ness. «
Talc Is No. 1 gypsum No. 2, etc., up to 9 for corundum, 10 for dia,
mond. If an unknown material is scratched noticeably by No
.
6 and not
by No. 5 the hardness value is between 5 and 6. This test has never been ,
used to any great extent in metallurgy but is still used in mineralogy
.
The
primary disadvantage is that the hardness scale is nonuniform. When the
hardness of the minerals is checked by anothe- hardness-test method
>
i
,
26
INTRODUCTION TO PHYSICAL METALLURGY
it is found that the values are compressed between 1 and 9, and there is a
large gap in hardness between 9 and 10. File Test
The test piece is subjected to the cutting action of a file of
known hardness to determine whether a visible cut is produced. Compara-
tive tests with a file depend upon the size, shape, and hardness of the file; the speed, pressure, and angle of filing during the test; and the composition and he at treatment of the material under test. The test is generally used industrially as one of acceptance or rejection.
1
'
:
In rnany cases, particularly with tool steels, when the steel is properly
heat-trbated it will be hard enough so that if a file is run across the surface it will hot cut the surface. It is not unusual to find heat-treating specifica-
tions which simply say "heat-treat until the material is file-hard." By run- | ning a file across the surface an inspector may rapidly check a large number of heat-treated parts to determine whether the treatment has been successful. 1-24
Resistance to Indentation This test is usually performed by impressing into the specimen, which is resting on a rigid platform, an indenter of fixed and known geometry, under a known static load applied either directly k or by means of a lever system. Depending on the type of test, the hardness is expressed by a number that is either Inversely proportional to the depth of indentation for a specified load and indenter, or proportional to a mean load over the area of indentation. The common methods of making Indentation hardness tests are described below.
The Brinell hardneM lestejL-USiLaiiy_consLsts of
Brinell Hardness Test
[0)
Fig 1-16 (a) Brinell hardness tester. (Amel Equipment Systems, East Moline, III.) (fc) Rc lester. (Wilson Mechanical Instrument Divis
_
a hand-operajed verticaMhydraulic presg. designed to force a ball indenter \
Chain & Cable Company.)
"
into the test specimen (Fig. 1-16ak Standard procedure requires that the f: test be made with a ball of 10 mm diameter under a load of 3,000 kg for ;r ferrous metals, or 500 kg for nonferrous metals. For ferrous metals the : loaded ball is pressed into the test specimen for at least 10 s; for nonferrous metals the time is 30 s. The diameter of the impression produced is measured by means of a microscope containing an ocular scale, usually -
graduated in tenths of a millimeter, permitting estimates to the nearest 0 05 mm.
ing the observed diametei (see Table 1 -4). The Brinell hardness m suffix numbers denotes s
diameter and a load of 3,0( the hardness number and
.
The Brinell hardness number (HB) is the ratio of the load in kilograms to the impressed area in square millimeters, and is calculated from the follow- ; ing formula: HB
L
inD/2){D - VD2
-
(F)
(1 -7)
I I
M
I i;
hardness of 75 measured
kg applied for 30 s. The Brinell hardness nu
imately 500 HB. As the me rate. The upper limit of th<
'
i
and duration of loading. F
for the indenter itself to st;
where il = test load, kg P = diameter of ball, mm d = diameter of impression, mm Calculation is usually unnecessary because tables are available for convert-
eating the test conditions
I
bide ball rather than a har
to approximately 650 HB.
MM*
TOOLS OF THE METALLURGIST
27
I
essed between 1 and 9, and there is a -
4 10.
i v|ed to the cutting action of a file of ar a visible cut is produced. Compara5 size, shape, and hardness of the file; g during the test; and the composition )nder test. The test is generally used rejection. doI steels, when the steel is properly
i
) that if a file is run across the surface
i
3:,v nusual to find heat-treating specifica-
llliitll the material is file-hard." By run-
P
i
)ector may rapidly check a large numI
.
ine whether the treatment has been
is usually performed by impressing n a rigid platform, an indenter of fixed vS/Wn static load applied either directly iding on the type of test, the hardness ler inversely proportional to the depth
t :
p
.
id indenter, or proportional to a mean e common methods of making inden-
rtOCKWEUL
r
jwdness testeujsJLtaJIv consists of
'
tous metals.
For ferrous metals the
'
specimen for at least 10 s; for non-
-
Wlldiameter of the impression produced pe containing an ocular scale, usually
:
f :
;
:
\
ters, and is calculated from the follow-
i (1 7)
nm
(cause tables are available for convert-
WILSON J
tester. (Wilson Mechanical Instrument Division, American Chain & Cable Company.)
I
permitting estimates to the nearest is the ratio of the load in kilograms to
r
Fig. 1-16 (a) Brineil hardness tester. (Atnetek/Testing Equipment Systems, East Molina, III.) (b) Rockwell hardness
:
,
"
I
)elow.
ress. designed to force a ball indenter Standard procedure requires that the liameter under a load of 3,000 kg for
1
i
I
ing the observed diameter of impression to the Brineil hardness number (see Table 1 -4). The Brineil hardness number followed by the symbol HB without any suffix numbers denotes standard test conditions using a ball of 10 mm diameter and a load of 3,000 kg appliedfor10to15s. For other conditions, the hardness number and symbol HB are supplemented by numbers indicating the test conditions in the following order: diameter of ball, load, and duration of loading. For example, 75 HB 10/500/30 indicates a Brineil hardness of 75 measured with a ball of 10 mm diameter and a load of 500
kg applied for 30 s. The Brineil hardness number using the standard ball is limited to approximately 500 HB. As the material tested becomes harder, there is a tendency for the indenter itself to start deforming, and the readings will not be accurate. The upper limit of the scale may be extended by u ing a tungsten carbide ball rather than a hardened steel ball. In that case it is possible to go to approximately 650 HB. ,
i
!
.
I .
i
ii I
I
:
.
28
INTRODUCTION TO PHYSICAL METALLURGY
I
!
>
TABLE 14
Approximate Hardness Relations for Steel' TABLE 14
ROCKWELL, USING BRALE
BRINELL, 3,000 KG
BRINELL,
Q
i
-
-
z
O _
j
DC UJ LU
S < 5
i
-
I
j
5 5
.
m
Q DC
< Q Z
<
-I
I-
<
t/5
CO
1 I
'
'
'
'
d m tr <
o
cc f
C3
cc UJ
o
<
Q-
S
o 00
CO
c/)
O
CC
<
>
O
LL
CO
682
737
61.7
72.0
82.2
79.0
84
2 40
653
697
60.0
70.7
81.2
77.5
81
.
'
o
UJ
CO
o
o
z s
5
tu
60
39
60
96
4 15
212
38
59
95
4 20
207
37
59
94
4 25
202
37
58
93
4 30
197
36
58
92
4 35
192
35
57
91
259
4 40
187
34
57
90
253
4 45
183
34
56
89
4 50
179
33
56
88
80.5
76.3
79
68.7
79.8
75.1
77
309
578
615
56.0
67.7
79.1
73,9
75
297
323
285
2 60
555
591
54.7
66.7
78.4
72.7
73
2 65
534
569
53.5
65.8
77.8
71.6
71
274
2 70
514
547
52.1
64.7
76.9
70.3
70
263
.
2 75 .
; i .,*>.
2 80 .
2 85 .
2 90 .
495
| | | |
;495
477 ;477
461 :
461
444 444
539
51.6
64.3
76.7
69.9
528
51.0
63.8
76.3
69.4
516
50.3
63.2
75.9
68.7
508
49.6
62.7
75.6
68.2
495
48.8
61.9
75.1
67.4
491
48.5
61.7
74.9
67.2
474
47.2
61.0
74.3
66.0
472
47.1
60.8
74.2
65.8
7 5 .
68
t
.
.
.
.
.
r
-
.
5
.
.
.
247 66
s = en ;
40
69.7
.
C
s
217
58.7
.
<
223
57.3
,
C o
CD
4 10
667
,
-I
C3
4 05
640
80
0 c C3
< d <
Z
o
627
.
H CO
CL
601
.
5 5
of
]
_
2 50 2 55
UJ
c DC < Q
_
CO
CO X
tU
2 45 .
BRALE
LU
DC
O
2 35 .
'
<
5
tu
,
z
UJ
2 <
F
3 000 KG
o
.
<
Approximate Hardness Relatl
.
97
243
4 55
174
33
55
87
237
4 60
170
32
55
86
.
.
235
4 65
166
32
54
85
226
4 70
163
31
53
84
63
225
4 75
159
31
53
83
217
4 80
156
30
52
4 85
153
81
65 70
.
.
.
82
2 95
429
429
455
45.7
59.7
73.4
64.6
61
3 00
415
415
440
44.5
58.8
72.8
63.5
59
210
3 05
401
401
425
43.1
57.8
72.0
62.3
58
202
4 90
149
80
3 10
388
388 :
410
41.8
56.8
71.4
61.1
56
195
4 95
146
79
3 15
375
375
396
40.4
55.7
70.6
59.9
54
188
5 00
143
78
182
5 05
140
76
137
75
134
74
.
.
.
.
.
3 20 .
363
363
383
39.1
54,6
70.0
58.7
6 5 .
52
.
.
V
.
.
.
.
,
3 25
352
352
372
37.9
53.8
69.3
57.6
51
176
5 10
3 30
341
341
360
36.6
52.8
68.7
56.4
50
170
5 15
3 35
331
331
350
35.5
51.9
68.1
55.4
48
166
5 20
131
73
34.3
51.0
67.5
54.3
47
160
5 25
128
71
.
.
.
J 40
321
321
339
3 45
311
311
328
33.1
50.0
66.9
53.3
46
3 50
302
302
319
32.1
49.3
66.3
52.2
45
3 55
293
293
309
30.9
48.3
65.7
51.2
3 60
285
285
301
29.9
47.6
65.3
50,3
3 65
277
277
292
28.8
46.7
64.6
49,3
.
1
.
.
.
.
.
3 70 .
269
269
284
3 75
262
262
276
3 80
255
255
269
3 85
248
248
261
.
.
.
3 90 .
241
241
,
253
27.6 i
45.9
64.1
48.3
5 30
126
70
5 35
124
69
43
145
5 40
121
68
42
141
5 45
118
67
41
137
5 50
116
65
133
5 55
114
64
129
5 60
112
63
63.6
47.3
39
44.2
63.0
46.2
38
24.2
43.2
62.5
45.1
37
43.9
60 .
40
45.0
61.8
36
55
.
.
.
.
5 65
109
62
107
60
118
5 75
105
58
5 80
103
57
235
247
21.7
41.4
61.4
42,9
35
115
229
229
241
20.5
40,5
60.8
41,9
34
111
. Adapted from H E. Davis. G. E. Troxell, and C. T. Wiskocii. "The Testing and Inspection of Engineering Materials." 2d ed., .
McGraw-Hill Book Company, New York, 1955; based on "Metals Handbook." 1948 ed., American Society lor Metals, Metals
V
.
5 70
235
Park, Ohio. See ASTM E 140 for additional relations.
.
126
4 00 .
.
122
3 95 .
.
.
155
25.4
42.0
.
150
26.6
22.8
.
.
.
.
,
TOOLS OF: THE METALLURGIST
29
a
.
TABLE 14
USING BRALE
Approximate Hardness Relations for Steel (Continued)
BRINELL, 3 000 KG ,
BRALE
z
LU
.
ROCKWELL
SUPERFICIAL
LU
a
z
.
O
<
a
.
CO UJ
O cc
a to
z
LU
CO
O 5
.
3 CO
-
Q
CO
CD
LU
CO X
<
73 r
<
-
1
Z
-I
<
-I
82.2
79.0
84
77.5
81
80.5
76.3
79
323
.
75,1
77
309
79.1
73.9
75
297
78.4
72.7
73
79.8
71.6
76.9
70.3
76.7
69.9
76.3
69.4
75.9
68.7
75.6
68.2
75.1
67.4
74.9
67.2
74.3
66.0
74.2
65.8
274
71
259
70 .
63
O
41
80.5
33
108
96
40
80.0
32
105
4 15
212
38
59
95
39
79.0
31
02
4 20
207
37
59
94
38
78.5
31
100
4 25
202
37
58
93
110
37
78.0
30
4 30
197
36
58
92
110
36
77.5
29
4 35
192
35
57
91
109
35
77.0
28
4 40
187
34
57
90
09
34
76.0
28
92
4 45
183
34
56
89
109
33
75.5
27
90
.
.
i
.
98 96
50 .
94
179
33
56
88
108
32
75.0
27
88
4 55
174
33
55
87
108
31
74.5
26
86
237
4 60
170
32
55
86
107
30
74.0
26
84
235
4 65
166
32
54
85
107
30
73.5
25
82
226
4 70
163
31
53
84
106
29
73.0
25
81
225
4 75
159
31
53
83
106
28
72.8
24
79
27
71.5
24
77
r
.
.
.
.
.
.
61
217
4 80
156
82
105
59
210
4 85
153
81
105
71.0
23
4 90
149
80
104
70.0
23
.
.
30
52
76 4 5
75
62.3
58
202
61.1
56
195
4 95
146
79
104
69.5
22
188
5 00
143
78
103
69.0
22
72
5 05
140
76
103
68-0
21
71
5 10
137
75
102
67.0
21
70
170
5 15
134
74
102
66.0
21
68
48
166
5 20
131
73
101
65.0
20
66
5 25
128
71
100
64.0
65
59.9
5
54
70.0
58.7
52
182
69.3
57.6
51
176
68.7
56.4 55.4
50
.
i
.
.
.
V
.
.
.
'
6
.
74
67.5
54.3
47
160
66.9
53.3
46
155
5 30
126
70
100
63.5
64
66.3
52.2
45
150
5 35
124
69
99
62.5
63
121
68
98
62
62
.
.
.
51.2
43
145
5 40
65.3
50.3
42
141
5 45
118
67
97
61
61
64.6
49.3
41
137
5 50
116
65
96
60
60
114
64
95
59
59
112
63
95
58
58 56
65.7
64.1
48.3
.
r
.
40
133 129
5 60
63.6
47.3
39
63.0
46.2
38
62.5
45.1
37
61.8 61.4 60.8
.
.
126
5 65
109
62
94
58
122
5 70
107
60
93
57
55
43.9
36
118
5 75
105
58
92
55
54
42.9
35
115
5 80
103
57
91
54
53
41.9
34
5
111 "
2ded.,
iidbook, 1948 ed., American Society for Metals, Metals "
.
5 55
tsting and Inspection of Engineering Materials,
.
o
LU
97
63.5
68.1
o
z
60
64.6
70.6
cn
60
72.8
71.4
I
O CO
73.4 72.0
.i
CO
39
.
243
65
CD
4 50
247 66
CO
CO
-
253
68
5
LU
40
.
263
70
CD
CC
217
.
77.8
CD
r-
-
CO
O
223
,
285
5
<
2 <
t
CO
4 10 .
i
33
4 05 .
80
<
CD
s
81.2
i
1D < LU
LU
z
<
o
.
.
J
.
.
30
INTRODUCTION TO' PHYSICAL METALLURGY
I
Rockwell Hardness Test This hardness test uses a direct-reading instrument based on the principle of differential depth measurement (Fig.
Major loads are usually 60, 10 30, and 45 kg on the superficial The most commonly used Rc and 100-kg load) and the C (di
1 -ISb). The test is carried out by slowly raising the specimen against the indenter until a fixed minor load has been applied. This is indicated on the dial gauge. Then the major load is applied through a loaded lever system. After the dial pointer comes to rest, the major load is removed and, with the minor load still acting, the Rockwell hardness number is read on the dial gauge. Since the order of the numbers is reversed on the dial gauge, a shallow impression on a hard material will result in a high number while a deep impression on a soft material will result in a low number. There are two Rockwell machines the normal tester for relatively thick
tained with the normal tester.
hardness number must be spe the letter designating the seal For example, 82 HRB means a scale (Via-in. ball and 100-kg some typical applications are c The performance of the m; standard test blocks supplied should be returned gently to 1 remove the major load may cat
,
sections, and the superficial tester for thin sections.
The minor load is
10 kg on the normal tester and 3 kg on the superficial tester. A variety of indenters and loads may be used, and each combination determines a particular Rockwell scale. Indenters include hard steel balls Via, Vb, V4, and V2 in. in diameter and a 120° conical diamond (brale) point.
cation. Care must be taken to;
cal movement at these points
on the gauge and, therefore, a ~
'
: .
~
Vickers Hardness Test
r :VI
TABLE 1-5
SCALE
MAJOR LOAD,
TYPE OF INDENTER
"
A
diamond-pyramid indenter wit site faces (see Fig. 1-17). The
The Rockwell Hardness Scales*
KG
Diamond cone
60
TYPICAL MATERIALS TESTED
i
I
The Vickers hardness tester o .
Brinell tester, the numbers bei Extremely hard materials, tungsten
the impression. As a result of surface of the specimen will t
carbides, etc. B
Vu" ball
100
In thi
Medium hard materials, low- and medium-carbon steels, brass,
the square is measured througl
bronze, etc.
eter that contains movable knif
Hardened steels, hardened and
is indicated on a counter calib
C
150
Diamond cone
D
100
Diamond cone
Case-hardened steel
E
100
Vb" ball
Cast iron, aluminum and magnesium alloys
tempered alloys
F
60
Vu" ball
Annealed brass and copper
G
150
Vu" ball
Beryllium copper, phosphor
H
60
Vb" ball
Aluminum sheet
K
150
Vb" ball
Cast iron, aluminum alloys
L
60
VV ball
Plastics and soft metals such as lead
vj Operating
1
position
bronze, etc.
M
100
VV ball
Same as L scale
P
150
V/ ball
Same as L scale
R
60
V2" ball
Same as L scale
S
100
V2" ball
Same as L scale
V
150
V2" ball
Same as L scale
. Ametek Testing Equipment Systems, East Moiine, III.
h
:
i
J
136 ,
VI
136
E
<3> Fig. 1 -17 The Vickers diamond-pyramid indent
TOOLS OF THE METALLURGIST
ess test uses a direct-reading ins
Major loads are usually 60 100, and 150 kg on the normal tester and 15, 30, and 45 kg on the superficial tester. The most commonly used Rockwell scales are the B (Vu-in. ball indenter ,
fferential depth measurement (Fig.
.
.
vly raising the specimen against the jen applied. This is indicated on the plied through a loaded lever system.
. -
and 100-kg load) and the C (diamond indenter and 150-kg load)
,
,
.
jrs is reversed on the dial gauge, a result in a low number.
rA:; he normal tester for relatively thick -
r thin sections.
The minor load is
scale (Vu-in; ball and 100-kg load). The Rockwell hardness scales and some typical applications are given in Table 1-5.
i
s
The performance of the machine should be checked frequently with standard test blocks supplied by the manufacturer. The operating crank should be returned gently to its starting position; snapping the crank to remove the major load may cause an error of several points in the dial indication. Care must be taken to seat the anvil and indenter firmly. Any verti-
I
.
i the superfici&l tester. be used, and each combination deIndenters include hard steel balls 1120
°
both ob-
tained with the normal tester. Because of the many Rockwell scales the hardness number must be specified by using the symbol HR followed by the letter designating the scale and preceded by the hardness numbers For example, 82 HRB means a Rockwell hardness of 82 measured on the B
|lardness major load is removed and, with the number is read on the dial I will result in a high number while a
31.
conical diamond (brale) point.
cal movement at these points results in additional depth being registered on the gauge and, therefore, a false hardness reading. Vickers Hardness Test In this test the instrument uses a square-based diamond-pyramid indenter with an included angle of 136° between oppo,
'
'
'
z
' -
r
site faces (see Fig. 1-17). The load range is usually between 1 and 120 kg. The Vickers hardness tester operates on the same basic principle as the Brinell tester, the numbers being expressed in terms <:)f load and area of
TYPICAL MATERIALS TESTED
Extremely hard materials, tungsten
the impression. As a result of the indenter's shape, the impression on the surface of the specimen will be a square. The length of the diagonal of the square is measured through a microscope fitted with an ocular microm-
carbides, etc.
Medium hard materials, low- and medium-carbon steels, brass, bronze, etc.
Hardened steels, hardened and
tempered alloys
f
eter that contains movable knife-edges. The distance between knife-edges is indicated on a counter calibrated in thousandths of a millimeter. Tables
Case-hardened steel
Cast Iron, aluminum and magnesium alloys Annealed brass and copper
*
J Operating position
Beryllium copper, phosphor bronze, etc.
Aluminum sheet
am
Cast iron, aluminum alloys Plastics and soft metals such as lead
;
136
.
/
/ r
Same as L scale Same as L scale .
. .
-
;
.
.
<3>
7j
Same as L scale Same as L scale Same as L scale
Fig. 1-17
i j
The Vickers diamond-pyramid indenter
.
i
'
32
INTRODUCTION TO PHYSICAL METALLURGY
where L = applied load k( ,
d = length of long (
Operating position
4
The Tukon microhardne:
plications of Indentation ha 125 Accuracy of Any Indentatioi influence the accuracy of;
V
Condition of the indenter in errors in the hardness r 130°
> .
r Fig. 1-18
1
4
The Knoop diamond-pyramid indenter.
are usually available to convert the measured diagonal to Vickers pyramid hardness number (HV) or the following formula may be used: ,
HV =
1 854/.
(1-8)
d2
6
where L = applied load kg d = diagonal length of square impression mm As a result of the latitude in applied loads, the Vickers tester is applicable to measuring the hardness of very thin sheets as well as heavy sections. ,
51
,
Microhardness Test This term, unfortunately, is misleading, as it could refer to the testing of small hardness values when it actually means the use of small indentations. Test loads are between 1 and 1,000 gm. Two types of Indenters are used for microhardness testing: the 136° square-base Vickers diamona pyramid described previously, and the elongated Knoop
-
:
sM:
diamond indenter. in
The Knoop indenter (Fig. 1-18) is ground to pyramidal form that produces a diamond-shaped indentation having long and short diagonals of approximate ratio of 7:1. The pyramid shape employed has included longitudinal angles of 172c30 and transverse angles of 130°. The depth of indentation is '
about Vao of its length. As in the Vickers test, the long diagonal of the impression is measured optically with a filar micrometer eyepiece. The Knoop hardness number is the load divided by the area of the impression. Tables are usually available to convert the measured diagonal to Knoop hardness number (HK), or the following formula may be used: HK =
14.229L d2
it
it
K Fig. l -19
(1-9)
Company.)
I
i
The Tukon microhardness tester
anical Instrument Division, American Chai
TOOLS OF THE METALLURGIST
where L = applied load kg d= length of long diagonal
33
,
,
;
mm
The Tukon microhardness tester is shown in Fig 1 19. Some typica. applications of indentation hardness tests are given in Table 1 -6. 125 Accuracy of Any Indentation Hardness Test Some of the factors that influence the accuracy of any indentation hardness test are: Condition of the Indenter Flattening of a steel-ball indenter will result in errors in the hardness number. The ball should be checked frequently
>v--y
.
-
t
m
i .
5;
imm i '
neasured diagonal to Vickers pyramid ing formula may be used:
.
mm
4J
m
m
'
;;
(1-8)
V
i
i m
I
0
V: t
V
impression, mm i loads, the Vickers tester is applicable hin sheets as well as heavy sections.
II
%
fortunately, is misleading, as it could ss values when it actually means the Js are between 1 and 1,000 gm. Two nardness testing: the 136 square-base °
5S<
i
2
r
Si
previously, and the elongated Knoop
.as
ound to pyramidal form that produces g long and short diagonals of approxie employed has included longitudinal
;
les of 130°. The depth of indentation is /ickers test, the long diagonal of the vith a filar micrometer eyepiece. The
m
.
'
idivided by the area of the impression
-
.
::
-
.
;>ert
the measured diagonal to Knoop
JS*?mg formula may be used: .
"
" ""
'
J Fig. 1-19
!
The Tukon microhardness tester. (Wilson Mech-
.
.
229L
Id2
III
(1-9)
anical Instrument Division, American Chain & Cable
Company.)
.
34
:.V.a..:.V-v
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 1-6
Typical Applications of Indentation Hardness Tests*
BRINELL
ROCKWELL
ROCKWELL SUPERFICIAL
VICKERS
for permanent deformation and
Structural steel
Finished parts,
Same as
and other rolled
such as bear-
standard
Rockwell and
sections
ings, bearing
Rockwell except
Rockwell
Coatings, such as
races, valves, nuts, bolts,
where shallower
lacquer, varnish, or paint
gears, pulleys, rolls, pins, pivots,
necessary, as in:
Superficial except where higher accuracy
stops, etc.
Thin case-
Most castings, Including steel, cast iron, and aluminum
Most forgings
penetration Is
hardened parts,
Cutting tools,
Same as
Thin materials
scissors
down to .006 in.
Accuracy of Load Applied The range with negligible error. Loa
I
should not be used for accurate
Foils and very thin
penetration is
materials down to
to .010 in.
such as saws, knives, chisels,
Plated surfaces
or shallower
necessary, as in:
Diamond indenters should be cl
MICROHARDNESS
.
Impact Loading Besides causi loading may damage diamond ir pot will ensure smooth, steady c Surface Condition of the Specir which the hardness reading is to of sound material. Any pits, sea ing or polishing. Thickness of Specimen The sf
0001 in.
To establish case
Thin case-
gradients
hardened parts, 005 to .010 in.
.
no bulge appears on the surfai Bimetals and
recommended thickness of the
laminated
Forming tools
Small castings and forgings
Cemented
Thin materials
carbides
down to .005 in.
Powdered metals
Sheet metal
Highly finished parts to avoid a removal
operation
Large-diameter wire
materials
of the impression. Shape of the Specimen
Very small parts or
surface is flat and normal to the1
areas, such as
men should be properly suppor! should be prepared, if possible,
watch gears, cutting tool edges, thread crests, pivot points, etc.
such as tubing Electrical contacts
Weak structures
Plastic sheet
Plating thickness
V-notch anvil should be used to
i
are ground, in which case a flat < I
Thin sections,
test is made on a round specime
-
Very brittle or fragile materials (Knoop indenter),
germanium, glass,
ing a flat, the observed reading n tion factor (Table 1 -7). Location of Impressions Impre from the edge of the specimen
tooth enamel
for ball tests.
Opaque, clear, or
Uniformity of Material if there the material, the larger the impre hardness reading. It is necessai
such as silicon,
or parts Case-hardened
parts
translucent
materials
area is small to obtain a true avt
Cemented carbides
1 26
Powdered metals
To investigate individual constituents of
a material
To determine grain or grain boundary
Advantages and Disadvantages c of a hardness test is usually detei of accuracy desired. Since the E sion, it is limited to heavy sectio This is an advantage, however, neous. The surface of the test f have to be so smooth as that for
*
hardness '
The gr
Ametek/Testing Equipment Systems. East Moline, lit.
croscope to measure the diame as reading a dial gauge. Becj Brinell test is generally inaccur; tended to about 650 HB with a t
h
The Rockwell test is rapid anc
1
I !
i.
~"
111! 11 I liillMMMIfllMWIilwiMlllii
"
TOOLS OF THE METALLURGIST
ts Tests*
35
for permanent deformation and discarded when such deformation occurs Diamond indenters should be checked for any sign of chipping.
.
VICKERS
MICROHARDNESS
Same as
Plated surfaces
.
Accuracy of Load Applied Rockwell and Rockwell
Coatings, such as
Superficial except where higher accuracy
lacquer, varnish, or paint
or shallower
Foils and very thin materials down to
penetration Is necessary, as in:
.
,
0001 in.
To establish case
Thickness of Specimen The specimen should be thick enough so that no bulge appears on the surface opposite that of the impression. The recommended thickness of the specimen is at least ten times the depth of the impression. Shape of the Specimen The greatest accuracy is obtained when the tost
gradients
Thin case-
hardened parts, .
Bimetals and
005 to .010 in.
laminated
Thin materials down to .005 in.
materials ,
Very small parts or
surface is flat and normal to the vertical axis of the indenter. A long speci-
Highly finished parts to avoid
areas, such as
a removal
cutting tool edges, thread crests, pivot
men should be properly supported so that it does not tip. A flat surface should be prepared, if possible, on a cylindrical-shaped specimen, and a V-notch anvil should be used to support the specimen unless parallel flats are ground, in which case a flat anvil may be used. If a Rockwell hardness test is made on a round specimen less than 1 in. in diameter without grinding a flat, the observed reading must be adjusted by an appropriate correc-
operation
watch gears,
points, etc. Thin sections,
such as tubing Weak structures
Plating thickness
Pi-
Very brittle or fragile materials (Knoop Indenter),
tion factor (Table 1 -7).
such as silicon,
Location of Impressions
germanium, glass,
from the edge of the specimen and should be at least 5 diameters apart
tooth enamel
for ball tests,
Impressions should be at least 2V2 diameters ,
Uniformity of Material
If there are structural and chemical variations in
the material, the larger the impression area the more accurate the averagehardness reading. It is necessary to take many readings if the impression area is small to obtain a true average hardness for the material.
Opaque, clear, or translucent materials
1 26
Powdered metals
To investigate individual
Advantages and Disadvantages of Different Types of Tests The selection of a hardness test is usually determined by ease of performance and degree of accuracy desired. Since the Brinell test leaves a relatively large impression, it is limited to heavy sections. This is an advantage, however when the material tested is not homogeneous. The surface of the test piece when running a Brinell test does not have to be so smooth as that for smaller impressions; however using a microscope to measure the diameter of the impression is not so convenient
constituents of
,
a material
m
The tester should apply loads in the stated
range with negligible error. Loads greater than the recommended amount should not be used for accurate testing. Impact Loading Besides causing inaccurate hardness readings, impact loading may damage diamond indenters. The use of a controlled oil dashpot will ensure smooth, steady operation of the loading mechanism. Surface Condition of the Specimen The surface of the specimen on which the hardness reading is to be taken should be flat and representative of sound material. Any pits, scale, or grease should be removed by grinding or polishing.
To determine grain or grain boundary
,
hardness
as reading a dial gauge. Because of deformation of the steel ball the Brinell test is generally inaccurate above 500 HB. The range may be extended to about 650 HB with a tungsten carbide ball. The Rockwell test is rapid and simple in operation. Since the loads and ,
\
'
~
36
INTRODUCTION TO PHYSICAL METALLURGY
The tube must be perpendici erly supported and clampec
TABLE 17 Wilson Cylindrical Correction Cnart* Cylindrical work correictions (approximate only) to be added to observed Rockwell number
than for most other testing
DIAMOND BRALE 1NDENTER
chipped or cracked.
DIAMETER OF SPECIMEN IN.
C D, A SCALES
,
1 27 Hardness Conversion
,
3/a 80
05
05
05
0
70
10
10
05
05
05
.
.
.
.
,
.
0 .
.
0
0
0
0
15
10
10
05
05
05
50
25
20
15
10
10 .
05
05
40
35
25
20
15
10
10
10
30
50
35
25
20
15 .
15
10
20
60
45
35
25
20
15
15
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
tests on carbon and alloy stt
05 .
1 28 Stress and Strain
to change its size or shape tt resistance of the body is kno\ ,
.
.
dimensions of the body are c
.
is the total internal resistance
usually determined is the intt
DIAMETER OF SPECIMEN IN. ,
V4
3/8
V2
5/a
3/a
Va
1
100
35
25
15
15
10 .
10 .
05
90
40
30
20 .
15
1 5 .
15
10
80
50
35
25
20
15 .
15
15
70
60
40
30
25
20
20
15
60
70
50
35
30
25
20
20
50
80
55
40
35
30
25
20
40
90
60
45
40
30
25
25
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
as the stress per unit area. pounds per square inch (psi] it is calculated as the load pi
I
The total deformation ort(
of a dimension of the body
.
unit strain is the deformatior
.
1-29 The Tensile Test
.
Courtesy of Wilson Mechanical Instrument Division, American Chain 4 Cable Co.
Next to the
quently performed to determ cally prepared sample is plac
.
V
'
When an e
.
/i6-IN. BALL INDENTER
F, G SCALES
The a
various hardness-test mach
generally applicable to steel
60
'
B
.
1
7/8
\ i
indenter's are smaller than those used in the Brinell test, the Rockwell test
may be used on thinner specimens, and the hardest as well as the softest materials can be tested.
The Vickers tester is the most sensitive of the production hardness 1
1
1
:
testers. It has a single continuous scale for all materials, and the hardness number is virtually independent of load. Because of the possibility of using
light loads, it can tost thinner sections than any other production lest, and the square indentation is the easiest to measure accurately. The microhardness test is basically a laboratory test. The use of very light loads permits testing of very small parts and very thin sections.. It can be used to determine the hardness of individual constituents of the microstructure. Since the smaller the indentation, the better the surface
finish must be, a great deal more care is required to prepare the surface
for microhardness testing. The surface is usually prepared by the technique of metallographic polishing described in Sees. 1 -14 to 1 -19. The principal advantages of the scleroscope are the small impressions that remain, the rapidity of testing, and portability of the instrument. However, results tend to be inaccurate unless proper precaustions are taken.
MM
It
!
i I
Fig. 1 20
A tensiilo sample with an extensomi
1
TOOLS OF THE METALLURGIST
37
The tube must be perpendicular to the test piece thin pieces must be properly supported and clamped the surface to be tested must be smoother than for most other testing methods and the diamond tip should not be ,
Bbserved Rockwell number
-
,
... .
: 'ENTER
,
OF SPECIMEN, IN.
1-27
0
0
0
0
05
05
0
05
05 .
05
05
10
10
05
05
15
10
10
10
20
15 .
15 .
10
25
20
15
15
.
The approximate hardness conversion between the various hardness-test machines is shown in Table -4. These data are generally applicable to steel and have been derived by extensive hardness
1
3/4 0
chipped or cracked. Hardness Conversion
.
tests on carbon and alloy steels mainly in the heat-treated condition. ,
.
.
.
.
.
.
.
.
.
.
.
.
.
1-28
.
.
.
.
Iter
,
OF SPECIMEN, IN. Vb
3U .
10
05
it is calculated as the load per unit area.
15
15
.1 0 1
The total deformation or total strain in any direction is the total change of a dimension of the body in that direction, and the unit deformation or unit strain is the deformation or strain per unit of length in that direction.
.
.
'
/
-
.
.
.
:
.
.
6
20
15
15
25
20
20
15
30
25
20 .
20
35
30
25
20
40
30
25
25
.
* .
1
10 .
.
.
.
.
Stress and Strain When an external force is applied to a body which tends to change its size or shape, the body resists this external force. The internal resistance of the body is known as stress and the accompanying changes in dimensions of the body are called deformations or strains. The total stress is the total internal resistance acting on a section of the body. The quantity usually determined is the intensity of stress or unit stress which is defined as the stress per unit area. The unit stress is usually expressed in units of pounds per square inch (psi), and for an axial tensile or compressive load
.
.
.
.
.
,
.
.
.
.
.
.
.
1-29 The Tensile Test
.
Next to the hardness test, the tensile test is most fre-
quently performed to determine certain mechanical properties. A specifically prepared sample is placed in the heads of the testing machine and an :
.
;hain & Cable Co.
HI
d in the Brinell test, the Rockwell test
and the hardest as well as the softest
i l
M . .
.
,
.
,
ensitive of the production hardness
.,
ale for all materials/and the hardness
1 | lO'l
ad. Because of the possibility of using
is than any other production test, and
i
4
to measure accurately.
ily a laboratory test. The use of very mall parts and very thin sections. It
i i
Less of individual constituents of the
;NMie -
::;
;
--
.
.
indentation, the better the surface
are is required to prepare the surface iace is usually prepared by the tech-
|§|scribed in Sees. 1 -14 to 1 -19. '
0
cleroscope are the small impressions
rid portability of the instrument. Howmless proper precaustions are taken.
\ i
:
Fig.
20
A tensile sample with an extensometer attached
.
v
.-:;
3 '
i
38
INTRODUCTION TO PHYSICAL METALLURGY
axial load is placed on the sample through a hydraulic or mechanical lever loading system. The force is indicated on a calibrated dial. If the original cross-sectional area of the specimen is known, the stress developed at any load may be calculated. The deformation or strain is measured at a fixed length, usually 2 in., by a dial gauge called an extensometer (see Fig. 1 -20). The unit strain may then be determined by dividing the measured elongation by the gauge length used. In some cases, an electrical strain gauge may be used to measure the total strain.
W
B
Y
P
The relation between unit stress s and unit strain e, found experimentally, is represented by the stress-strain graph in Fig. 1-21 for a ductile material and by the graph in Fjg. 1 -22 for a brittle material. 1-30 Tensile Properties The properties which may be determined by a tension
f
test follow.
I
-
-. .
J
-
Proportional Limit It is found for many structural materials that the early part of the stress-strain graph may be approximated by a straight line OP jn Figs. 1'.21 and 1-22. In this range, the stress and strain are proportional to each other, so that any increase in stress will result in a proportionate increase in strain. The stress at the limit of proportionality point P is known as the proportional limit. Elastic Liinit If a small load on the test piece is removed, the extensometer needle will return to zero, indicating that the strain, caused by the load, is elastic. If the load is continually increased, then released after each increment and the extensometer checked, a point will be reached at
I
which the extensometer needle will not return to zero. This indicates that
Plastic range
03
CP
o
W= specified offset Unit strain
Fig. 1 -22
i
:V :: r
the material now has a perm; fore be defined as the mini
first occurs.
For most struc
same numerical value as the Yield Point
As the load in
limit, a stress is reached at w an increase in load
.
The stre
point. This phenomenon o stress may actually decrease
Y in
Stress-strain diagram for a brittle i
P
i
r\-:
-
i
yield point.
Since the yield
permanent deformation is srr
in the design of many machir by considerable permanent that exhibit a well-defined yi' Yield Strength Most nonfe do not possess a well-define
4
c
0 Unit strain
mum useful strength is the y at which a material exhibits <
tionality of stress to strain
.
Fig. 1 -21
Stress-strain diagram for a ductile steel.
I
i
set method." In Fig. 1 -22
,
the
TOOLS OF THE METALLURGIST
rough a hydraulic or mechanical lever
39
B
ii;
ed on a calibrated dial. If the original l i Is known, the stress developed at any ,
r
Y
iation or strain is measured at a fixed
called an extensometer (see Fig. 1 -20). led by dividing the measured elongaome cases, an electrical strain gauge
P
ain.
ind unit strain e, found experimentally, :;
in
aph in Fig. 1-21 for a ductile materia!
Mjittle material.
Milch may be determined by a tension any structural materials that the early ie approximated by a straight line OP iihe stress and strain are proportional
stress will result ih a proportionate in-
: .
V
»f|ilt of proportionality point P is known vO
' -
0
OX-specified offset
test piece is removed, the extensomiating that the strain, caused by the
Unit strain
Fig
Lually increased, then released after
22
Stress-strain diagram for a brittle material
,
!r checked, a point will be reached at lot return to zero. This indicates that
the material now has a permanent deformation. The elastic limit may there-
:
fore be defined as the minimum stress at which permanent deformation first occurs. For most structural materials the elastic limit has nearly the same numerical value as the proportional limit.
Yield Point As the load in the test piece is increased beyond the elastic
i B
limit a stress is reached at which the material continues to deform without ,
an increase in load. The stress at point V in Fig 1 -21 is known as the y;e/d point. This phenomenon occurs only in certain ductile materials. The stress may actually decrease momentarily resulting in an upper and lower .
I
,
yield point. Since the yield point is relatively easy to determine and the permanent deformation is small up to yield point, it is a very important value in the design of many machine members whose usefulness will be impaired by considerable permanent deformation
that exhibit a well-defined yield point
This is true only for materials
.
.
Yield Strength Most nonferrous materials and the high-strength steels do not possess a well-defined yield point For these materials, the maxi.
:
mum useful strength is the yield strength The yield strength is the stress .
1 n
c
at which a material exhibits a specified limiting deviation from the proportionality of stress to strain. This value is usually determined by the "offset method." In Fig. 1 -22 the specified offset OX is laid off along the strain ,
r
i
40
INTRODUCTION TO PHYSICAL METALLURGY
strength. For a brittle material, tl
axis. Then XIV is drawn parallel to OP and thus V, the intersection of XW with the stress-strain diagram is located. The value of the stress at point ,
coincide.
,
/gives the yield strength. The value of the offset is generally between 0.10
Ductility
and 0.20 percent of the gauge length. Ultimate Strength As the load on the test piece is increased still further the stress and strain increase as indicated by the portion of the curve YM (Fig. 1-21) for a ductile material, until the maximum stress is reached at
tion that is possible until fractur
point M. The ultimate strength or the tensile strength is therefore the maximum stress developed by the material based on the original cross-sectional area. A brittle material breaks when stressed to the ultimate strength (point
marks.
The ductility of a mate
two measurements:
,
Elongation
,
This is determined b
of the specimen and measuring
Elongation (per
S in Fig. 1-22), whereas a ductile material will continue to stretch Breaking Strength For a ductile material, up to the ultimate strength, the deformation is uniform along the length of the bar. At the maximum stress .
,
localized deformation or necking occurs in the specimen and the load falls off as the area decreases. This necking elongation is a nonuniform deformation and occurs rapidly to the point of failure (Fig. 1-23). The breaking1 strength (point S, Fig. 1 '21), which is determined by dividing the breaking load by the original cross-sectional area, is always less than the ultimate ,
:
,
\
\
where L,= final gauge length L0 = original gauge length In reporting percent elongatior fled since the percent elongatior Reduction in Area
This is also d{
tensile specimen by measuring using the following formula: Reduction in area (
where A0 = original cross-sectioi Af= final cross-sectional Modulus of Elasticity, or Young'
portion of the stress-strain curv mx + b, where y is the vertical ; zontal axis, in this case strain.
1
and in this case it is zero since tl
of the line is m.
When the equa
y/x. Therefore, the slope of the right triangle and finding the ta equal to y/x or stress/strain. The ality between stress and strain bf
-
.
:
the modulus of elasticity or Youi The modulus of elasticity, whk terial, is measured in pounds per elasticity of steel is approximate
10 million psi.
Therefore, stee
aluminum. The modulus of elas 4$
ft '
and will appear in formulas deal!
mi
where stiffness is important. m
EM
Fig. 1-23
1 31 True Stress-strain
as
Tension sample before and after failure.
<
.
. v -
1
The conventh
able information up to the point
TOOLS OF THE METALLURGIST
i
41
a r
strength. For a brittle material, the ultimate strength and breaking strength
and thus V, the intersection of
-
coincide.
v d. The value of the stress at point
..
|f he offset is generally between 0.10
Ductility The ductility of a material is indicated by the amount of deformation that is possible until fracture. This is determined in a tension test by
test piece is increased still further ted by the portion of the curve WW
two measurements:
,
:
Elongation
This is determined by fitting together after fracture, the parts of the specimen and measuring the distance between the original gauge
he maximum stress is reached at
isile strength is therefore the maxised on the original cross-sectional
,
marks.
Elongation (percent) -
ssed to the ultimate strength (point
(1-10)
x 100
jal will continue to stretch. al, up to the ultimate strength, the
where L,= final gauge length
of the bar. At the maximum stress,
L0= original gauge length, usually 2 in. In reporting percent elongation, the original gauge length must be specified since the percent elongation will vary with gauge length.
j in the specimen, and the load falls |g elongation is a nonuniform de-
jit of failure (Fig. 1 -23). The break-
'
' .
ft
Reduction in Area
This is also determined from the broken halves of the
tensile specimen by measuring the minimum cross-secjtional area and
determined by dividing the breakea, is always less than the ultimate
using the following formula: Reduction In area (percent);
l (1-11)
x 100
where A0 = original cross-sectional area Af= final cross-sectional area
Modulus of Elasticity, or Young's Modulus Consider the straight-line portion of the stress-strain curve. The equation of a straight line is y = mx + b, where y is the vertical axis, in this case stress, and x is the horizontal axis, in this case strain. The intercept of the line on the y axis is b-, . and in this case it is zero since the line goes through the origin. The slope
of the line is m. When the equation is solved for m, the slope is equal to y/x. Therefore, the slope of the line may be determined by drawing any right triangle and finding the tangent of the angle 0 (Fig. 1-22), which is equal to y/x or stress/strain. The slope is really the constant of proportionality between stress and strain below the proportional limit and is known as the modulus of elasticity or Young's modulus. The modulus of elasticity, which is an indication of the stiffness of a material, is measured in pounds per square inch. For example, the modulus of elasticity of steel is approximately 30 million psi, while that of aluminum is,
V
10 million psi. Therefore, steel is approximately three times as stiff as aluminum. The modulus of elasticity is a very useful engineering property and will appear in formulas dealing with the design of beams and colgmns where stiffness is important. 1-31
True Stress-strain
The conventional tensile test described will give valu-
able information up to the point of yielding. Beyond this point, the stress
i
'
J
'
v
.
42
INTRODUCTION TO PHYSICAL METALLURGY
The ordinary impact machin "
True stress-strain curve
'
' " ' '
' .
'
which is raised to a standard h
1
V
I 1
si
lum has a definite amount of p leased, this energy is converte men. The Charpy specimen w
specimen, placed with the V n
Ordinary
!
tested (see Fig. 1 -26). At that h
stress-strain
the V notch.
In either case, s
curve
used to rupture the specimen lower than the initial height (
weight of the pendulum times energy, usually in foot-pounds impact strength. From the description of thi
Strain
Fig. 1 -24
impact test does not yield the
True stress-strain and conventional stress-strain
diagrams for mild teel. '
v
r
. .
;
behavior with a particular note
parative purposes. The notch
" .
-
.
values Eire fictitious since the actual cross-sectional area will be consider- |
-
motive industries, which
ha\
,
ably reduced.
The true stress is determined by the load divided by the oross-sectional area at that moment of loading. The true strain is deter- '
mined by the change in length divided by the immediately preceding length. ;: The true stress-strain diagram (Fig. 1 -24) yields useful information regard- * ing plastic flow and fracture of metals.
1 '32 Resilience and Toughness
lOrr
(0.3!
It is possible to divide the stress-strain diagram f
55mm
into tvyo parts as shown in Fig. 1 -21. The part to the left of the elastic limit s
(2.165")
may be called the elastic range and that to the right of the elastic limit the I plastic range. The area under the curve in the elastic range (area OPR) is a ?
Simple beam V-no
measurfe of the energy per unit volume which can be absorbed by the mate- h
!5
rial without permanent deformation. This value is known as the modulus
10
'
of resilience. The energy per unit volume that can be absorbed by a mate-
;
0 _
rial (the area under the entire stress-strain diagram) up to the point of frac- l ture is known as toughness. This is mainly a property of the plastic range, ;;
(0.3 _
55 mm
_
(2.165")
since only a small part of the total energy absorbed is elastic energy that J recovered when the stress is released. j 1-33 impact Test Although the toughness of a material may be obtained by
Simple beam key-
'
can be
the area under the stress-strain diagram, the impact test will give an indi-
cation of the relative toughness.
28mm
(1.092")
((
0
,
[
Generally, notch-type specimens are used for impact tests. Two general ;
75mm
types of notches are used in bending impact tests, the keyhole notch and \
(2.952")
the V notch. Two types of specimens are used, the Charpy and the Izod, j!
Conlilever beam I;
shown in Fig. 1 -25. The Charpy specimen is placed in the vise so that it is a |
Fig. 1-25 Notched-bar impact test specimens.
vise so that one end is free and is therefore a cantilever beam.
Testing of Metallic Materials, ASTM Designati'
simple beam supported at the ends. The Izod specimen is placed in the |
1
i
mission from
"
Tentative Methods for Notched"
r
-
.
~
.
TOOLS OF THE METALLURGIST
43
.
I
The ordinary impact machine has a swinging pendulum of fixed weight which is raised to a standard height depending upon the type of specimen tested (see Fig. 1 -26). At that height with reference to the vise, the pendulum has a definite amount of potential energy. When the pendulum is released this energy is converted to kinetic energy until it strikes the speci,
,
men. The Charpy specimen will be hit behind the V notch while the Izod specimen, placed with the V notch facing the pendulum will be hit above the V notch. In either case, some of the energy of the pendulum will be used to rupture the specimen so that the pendulum will rise to a height lower than the initial height on the opposite side of the machine. The weight of the pendulum times the difference in heights will indicate the energy, usually in foot-pounds, absorbed by the specimen, or the notched impact strength. From the description of the test, it is apparent that the notched-bar impact test does not yield the true toughness of a material but rather its behavior with a particular notch. The results are useful, however, for corn- . parative purposes. The notched-bar test is used by the aircraft and auto,
,
! i
ain
cross
motive industries, which have found by experience that high impact
sectional area will be consider-
-
termined by the load divided by the
?ViV; of loading. The true strain is deter-
r
by the immediately preceding length. .24) yields useful information regard-
0 25mm
8mm
p
(0.315")
J
.
.
/(0.0I0 )R "
Jl
10mm
s
.
(0.3941
ble to divide the stress-strain diagram ;
T
55mm
jThe part to the left of the elastic limit
(0.394") }
Simple beam V-notched Charpy type 5 mm
e which can be absorbed by the mate-
(0.197")
1
This value is known as the modulus
JL 55 mm
_
(0.3941
! }
imen is placed in the vise so that it is a
T
srefore a cantilever beam.
10mm
_
-
*
(0.394")
28mm
,.
10mm
8mm
(0.394")
(0.315")
t
u Mr
T
75 mm
Cantilever beam Izod lype
Fig. 1-25 Notched-bar Impact test specimens (By permission from "Tentative Methods for Notched-Bar Impact Testing of Metallic Materials ASTM Designation E23-56T.) "
,
i
0
V-45
f
/ .
A
(2.952")
« lb
Simple beam key-hole notched Chorpy type
(1.092")
SIqw cut
-
li6mm ( -) or less
"
.
The Izod specimen is placed in the
1
(2.165")
released.
of a material may be obtained by am, the impact test will give an indi-
(6.079")
n
10 mm
'
nergy absorbed is elastic energy that
re used for impact tests. Two general 5 impact tests, the keyhole notch and is are used, the Charpy and the Izod
lOmm
(2.165")
hat to the right of the elastic limit the ve in the elastic range (area OPR) is a
ume that can be absorbed by a matetrain diagram) up to the point of fracnainly a property of the plastic range,
T
I0mm
Q 25mm
(0.394")
(p.010")R
.
44
INTRODUCTION TO PHYSICAL METALLURGY
the stress corresponding to sonT plot for alloy steel heat-treated,
num-copper alloy, and gray cast Fatigue tests are widely used 1 for type and range of fluctuating surface conditions, temperature, 1-35 Creep Tests The creep test dete formation of a material at elevat tin
yield strength. The results are i which are exposed to elevated t cussed in greater detail in Chap.
m m mm
NONDESTRUCTIVE TESTING mmmm
1-36 Introduction
0}
A nondestructive te
manner which will not impair the in most cases nondestructive te
of mechanical properties, they an that could impair the performan Fig. 126 The impact testing machine (Ametek/Testing
service.
Equipment Systems, East Moline III.)
machined into component parts sembly, to measure the thicknes; level of liquid or solid contents
Such a test is used to d
.
,
i
strength by test generally will give satisfactory service where shock loads are encountered.
1-34
.
r
materials, and to discover defect;
;
ing or use.
Fatigue Tests The fatigue test is a dynamic type of test which determines the relative behavior of materials when subjected to repeated or fluctuat-
Nondestructive tests are used
ing loads. It attempts to simulate stress conditions developed in machine parts by vibration of cycling loads. The magnitude of the stress may be changed on the machine, and the type of stress (tension, compression,
economical. Increased reliability
bending, or torsion) is determined by the machine and the type of specimen tested. The stress placed on the specimen during test continually alternates between two values, the maximum of which is usually lower than the yield strength of the material. The cycles of stress are applied until failure of the specimen or until a limiting number of cycles has been reached. Those results are then plotted on a semilogarithmic scale with the stress' S as the ordinate and the number of cycles N, to cause failure, as the abscissa. The endurance limit of any material is defined as the limiting stress below which the metal will withstand an indefinitely large number of "
ioo
Vj, SO
j '
t
rreo
60
40
Co Pper
20
"
becomes parallel to the abscissa. For steel this will occur at approximately 107 cycles of stress. For some nonferrous alloys, however, the curve does not become horizontal, and the term endurance limit is often applied to
\
01
9
-
Gr
I
10
i
3
y cast iron
1J-|J_1
0
cycles of stfess without fracture. At that point on the S-N curve, the curve
I
Parts may also be
moval before failure occurs.
100
L_LJ 1
Thousand
Cycles of Reversed F
[
Fig. 1-27
Typical S-N (stress-cycle) diagrams. (Frc
Metals Handbook," 1948 ed., American Society fo Metals Park, Ohio.) "
L
TOOLS OF THE METALLURGIST
45
ft
the stress corresponding to some specific number of cycles. A typical S-N plot for alloy steel heat-treated, medium carbon steel heat-treated, alumi-
num-copper alloy, and gray cast iron is shown in Fig. 1 -27. Fatigue tests are widely used to study the behavior of materials not only for type and range of fluctuating loads but also for the effect of corrosion ,
surface conditions, temperature, size and stress concentration. 1-35 Creep Tests The creep test determines the continuing change in the deformation of a material at elevated temperature when stressed below the yield strength. The results are important in the design of machine parts which are exposed to elevated temperatures. Creep behavior will be disr cussed in greater detail in Chap. 13. ,
'
it.:
I NONDESTRUCTIVE TESTING 1-36
N
*-
-.
Introduction A nondestructive test is an examination of an object in any manner which will hot impair the future usefulness of the object. Although in most cases nondestructive tests do not provide a direct measurement
.
of mechanical properties, they are very valuable in locating material defects that could impair the performance of a machine member when placed in service. Such a test is used to detect faulty material before it is formed or machined into component parts, to detect faulty components before assembly, to measure the thickness of metal or other materials, to determine level of liquid or solid contents in opaque containers, to identify and sort materials, and to discover defects that may have developed during processing or use. Parts may also be examined in service, permitting their re-
t Cv -v-
jtisfactory service where shock loads /namic type of test which determines
moval before failure occurs.
ien subjected to repeated or fluctuatess conditions developed in machine The magnitude of the stress may be /pe of stress (tension, compression, the machine and the type of specimen lecimen during test continually alterum of which is usually lower than the /cles of stress are applied until failure
|mber of cycles has been reached. semilogarithmic scale with the stress cycles A/, to cause failure, as the ab,
ly material is defined as the limiting istand an indefinitely large number of hat point on the S-N curve, the curve > steel this will occur at approximately
'
;
jrrous alloys however, the curve does n endurance limit is often applied to ,
Nondestructive tests are used to make products more reliable, safe, and economical. Increased reliability improves the public image of the manui
100
\\
80
\|
rreotec
60
ot Ireated
9
i %:
J)
10
i t
T Co oper .
20
n
Y°y cast iron 1 11 I
i
1
10 (00 Thousand
l._.L-L
I
10
100
1000
Million
Cycles of Reversed Flexure
Fig. 1 -27
Typical S-N (stress-cycle) diagrams (From .
"
Metals Handbook," 1948 ed American Society for Metals, ,
.
i
.
Metals Park, Ohio.)
i
i
46
INTRODUCTION TO PHYSICAL METALLURGY
;
facturer, which leads to greater sales and profits. In addition, manufacturers use these tests to improve and control manufacturing processes. i Before Wold War II, nondestructive testing was not urgent because of
5
the large safety factors that were engineered into almost every product. Service failures did take place, but the role of material Imperfections In such failures was not then fully recognized and, therefore, little concentrated effort was made to find them. During and just after. World War II the significance of imperfections to the useful life of a product assumed greater importance. In aircraft design, in nuclear technology, and in space exploration, high hazards and costs have made maximum reliability essential. At the same time, there has been extensive growth of all Inspection methods in industrial and scientific applications. There are five basic elements in any nondestructive test. 1 Source A source which provides some probing medium, namely, a medium that can be used to inspect the item under test. 2 Modification This probing medium must change or be modified as a result of the variations or discontinuities within the object being tested. 3 Detection A detector capable of determining the changes In the prob-
x-ra
,
i i
y tube
,
ing medium. 4 Indication detector.
Target
x-rays
r I
5:
Filament
Facal spot
Fig. 1-28 Schematic representation of the use of x-r, (or examination of a welded plate. (From "Basic Meta v
A means of indicating or recording the signals from the
5
vol. 2, American Society for Metals, Metals Park, Ohic 1964.)
,
5 Interpretation A method of interpreting these indications. While there are a large number of proven nondestructive tests in use,
part of their kinetic energy is com
this section will concentrate on the most common methods and on one re-
The essential conditions for the (
cent development. The most common methods of nondestructive testing
(cathode) to provide the source o
get, (2) a target (anode) located in
or inspection are:
ference between the cathode and
Radiographyi
of the electrons striking the target
Magnetic-particle Inspection
rays produced, and (4) a means o
v
.
Fluorescent-penetrant inspection Ultrasonic inspection
Eddy current inspection
1 -37 Radiography of Metals The radiography of metals may be carried out by | using x-rays or gamma rays-short-wavelength electromagnetic rays ca-
pable of going through relatively large thicknesses of metal. Gamma raya may be obtained from a naturally radioactive material such as radium or a radioactive isotope such as cobalt-60. Gamma radiation is more penetrat-
ing than that of x-rays, but the inferior sensitivity limits its application. There is no way that the source may be regulated for contrast or variable thickness, :;and it usually requires much longer exposure times than the x
-ray method X-rays
.
are produced when matter is bombarded by a rapidly moving
stream of electrons.
-t-
When electrons are suddenly stopped by matter, a
'
number of electrons striking the usually incorporated in an x-ray tut of a welded plate Is shown schema dangerous, and adequate safeguai ing personnel. A radiograph is a shadow pictur to radiation. The x-rays darken tt which readily permit penetration af
with regions of higher density whi a hole or crack appears as a darl aluminum alloy appear as lighter a While the radiography of metals tion of castings and welded produ thickness of materials. Fig. 1 -30 si"
TOOLS OF THE METALLURGIST
-
.
.
.
)es and profits. In addition, manufacsd control manufacturing processes.
Safety box
ve testing was not urgent because of mgineered into almost every product. the role of material imperfections in
:
Leat shields
pgnized, and, therefore, little conceni
y tube
x-ra
During, and just after. World War 11 the useful life of a product assumed |n, in nuclear technology, and in space s have made maximum reliability es-
: "
K-rays
.
Lead
Target
backing
Filament
x
W?)een extensive growth of all inspection ' "
47
Focal spot
Applications.
-ray film
Weld
ny nondestructive test.
js some probing medium, namely, a \ the item under test. .
jm must change or be modified as a cities within the object being tested.
;
. .. .
determining the changes in the prob-
Fig. 1 -28 Schematic representation of the use of x-rays for examination of a welded plate (From "Basic Metallurgy," .
MMq or recording the signals from the
< p: vol. 2, American Society for Metals
,
ireting these indications. )1 proven nondestructive tests in use, host common methods and on one reOon methods of nondestructive testing
.v
:
I
:
:
iphy of metals may be carried out by wavelength electromagnetic rays cage thicknesses of metal. Gamma rays -
dioactive material such as radium or a
Metals Park, Ohio,
] I I954"'
5
part of their kinetic energy is converted to energy of radiation, or x-rays. The essential conditions for the generation of x-rays are (1) a filament (cathode) to provide the source of electrons proceeding toward the target, (2) a target (anode) located in the path of electrons, (3) a voltage difference between the cathode and anode which will regulate the velocity of the electrons striking the target and thus regulate the wavelength of xrays produced, and (4) a means of regulating tube current to control the number of electrons striking the target. The first two requirements are usually incorporated in an x-ray tube. The use of x-rays fpr the examination of a welded plate is shown schematically in Fig. 1 -28. X-rays are potentially dangerous and adequate safeguards must be employed to protect operat,
ing personnel.
A radiograph is a shadow picture of a material more or less transparent to radiation. The x-rays darken the film so that regions of lower density
Gamma radiation is more penetrat-
which readily permit penetration appear dark on the negative as compared
ferior sensitivity limits its application.
with regions of higher density which absorb more of the radiation Thus a hole or crack appears as a darker area whereas copiper inclusions in
0
.
y be regulated for contrast or variable puch longer exposure times than the r is bombarded by a rapidly moving is are suddenly stopped by matter, a
.
,
aluminum alloy appear as lighter areas (see Fig 1 '29). While the radiography of metals has been used primarily for the inspection of castings and welded products it may also be used to measure the thickness of materials. Fig. 1 -30 shows a simple radiation thickness gauge .
,
.
t
48
INTRODUCTION TO PHYSICAL METALLURGY
I i
-
!
'
|
Indicator
1
Detector
:
Radiation
I
r
source
Fig. 1 30
iff:
A simple radiation thickness gauge
.
,
The radiation from the sourc As the thickness increases t ,
decreases. If the response o nesses, the detector reading inspected material. With a si used to control the thickness
%0)
r
ii;1
138 Magnetic-particle Inspection (I the presence of cracks, laps, tinuities in ferromagnetic mat detect surface discontinuities also detect discontinuities w
r
,
i i
f Fig. 1-29
(a) Radiograph of a stainless steel casting; dark
I
spots are shrinkage voids, (b) Radiograph of a brass sand
casting; numerous black spots indicate extensive porosity
r
.
t
I 151
.
\
applicable to nonmagnetic m Magnetic-particle inspectic piece to be inspected may be netic particles (iron powder). the magnetization and appli neously. This is known as the may be held in suspension in piece may be immersed in thi tions, the particles, in the for face of the workpiece (dry rr shown by the formation and e
of the workpiece over the disc and assumes the approximat continuity. The Magnaglo m tion is a variation of the Mag magnetized workpiece contai piece is then viewed under bl out more clearly. When the discontinuity is c out to the surface and forms
TOOLS OFjTHE METALLURGIST
49
Indicotor
Detector ;::::;
:
.
.
.
i:i:i;:'M-i;-;L-;y
-
.
V.V.V.V.
-
-
Material
being inspected
V.V.V.'i
.
Radiation source
Fig. 1-30 A simple radiation thickness gauge
.
The radiation from the source is influenced by the material being tested As the thickness increases, the radiation intensity reaching the detector .
v
1-38
decreases. If the response of the detector is calibrated for known thicknesses, the detector reading can be used to indicate the thickness of the inspected material. With a suitable feedback circuit, the detector may be used to control the thickness between predetermined limits. Magnetic-particle Inspection (Magnaflux) This is a method of detecting
the presence of cracks, laps, tears, seams, inclusions, and similar disconv
tinuities in ferromagnetic materials such as iron and steel. The method will detect surface discontinuities too fine to be seen by the naked eye and will also detect discontinuities which lie slightly below the surface. It is not applicable to nonmagnetic materials.
Magnetic-particle inspection may be carried out in several ways. The piece to be inspected may be magnetized and then covered with fine magnetic particles (iron powder). This is known as the residual method. Or, the magnetization and application of the particles may occur simultaneously. This is known as the continuous method. The magnetic particles may be held in suspension in a liquid that is flushed over the piece, or the piece may be immersed in the suspension (wet method), in some applications, the particles, in the form of a fine powder, are dusted over the surface of the workpiece (dry method). The presence of a discontinuity is shown by the formation and adherence of a particle pattern on the surface
of the workpiece over the discontinuity. This pattern is balled an indication and assumes the approximate shape of the surface projection of the discontinuity. The Magnaglo method developed by the Magnaflux Corporation is a variation of the Magnaflux test. The suspension flowed over the ; dark sand
rV-V-'-osity.
magnetized workpiece contains fluorescent magnetic particles. The work-
piece is then viewed under black light, which makes the indicatiohs stand out more clearly.
When the discontinuity is open to the surface, the magnetic field leaks out to the surface and forms small north and south poles that attract the
1
'
i
'
ST
50
INTRODUCTION TO PHYSICAL METALLURGY
Particles cling to the
tors, including strength of tht with the suspension, time allow magnetizing current, and strenc amples of cracks detectable by
defect like lacks to a
I
simple magnet
N
Defect
Speci men
-
S
1-33
.
All machine parts that have bi through a demagnetizing opera without demagnetizing, they wil steel particles which may cause
S
Magnetic field
s N
5 N
s
Detection of parts which have
N
ii
By inducing a magnetic field within the part to be tested
,
plished by keeping a compass c
and applying a coating of magnetic particles
,
surface cracks are made visible, the cracks in effect
5
forming neyi magnetic poles
Fig. 1-31
1 39
Fluorescent-penetrant Inspectior
five method of detecting minute and porosity that are open to tl plied to both magnetic and non
Principle of the Magnaflux test. (Magnaflux
Corporation, Chicago, III.)
i
magnetic particles (see Fig. 1 -31). When fine discontinuities are under the
Electric
ft
current
surface, some part of the field may still be deflected to the surface, but the leakage; is less and fewer particles are attracted, so that the indication obtained is much weaker. If the discontinuity is far below the surface, no leak-
age of he field will be obtained and consequently no indication. Proper use of magnetizing methods is necessary to ensure that the magnetic field set up will be perpendicular to the discontinuity and give the clearest I; indication.
As shown in Fig. 1-32 for longitudinal magnetization, the magnetic field
may bei: produced in a direction parallel to the long axis of the workpiece
by placilng the piece in a coil excited by an electric current so that the long ;;
i
axis of the piece is parallel to the axis of the coil. The metal part then becomes the core of an electromagnet and is magnetized by induction from
the magnetic field created in the coil. Very long parts are magnetized in steps by moving the coil along the length. In the case of circular magnetization, also shown in Fig. 1 -32, a magnetic field transverse to the long axis of the workpiece is readily produced by passing the magnetizing current
3
,
||
r
through the piece along this axis. Direct current, alternating current, and rectified alternating current are all used for magnetizing purposes. Direct current is more sensitive than
alternating current for detecting discontinuities that are not open to the surface; Alternating current will detect discontinuities open to the surface
4 Electric
and is (lised when the detection of this type of discontinuity is the only in-
I current
terest. When alternating current is rectified, it provides a more penetrating
magnetjc field. The Sensitivity of magnetic-particle inspection is affected by many fac-
t
1
Fig
32
Illustrating two kinds of magnetization;
tudinal magnetization; (b) circular magnetization. flux Corporation, Chicago, III.)
(MM
'
1 TOOLS OF THE METALLURGIST
Parficles cling to the defect like tacks to a
tors, including strength of the indicating suspension, time in contact with the suspension, time allowed for indications to form, time subject to magnetizing current, and strength of the magnetizing current. Some ex-
4
simple magnet
51
amples of cracks detectable by Magnaflux or Magnaglo are shown in Fig. Defect
5
1-33
.
AH machine parts that have been magnetized for inspection must-be put through a demagnetizing operation. If these parts are placed in service without demagnetizing, they will attract filings, grindings, chips and other steel particles which may cause scoring of bearings and other engine parts. Detection of parts which have not been demagnetized; is usually accom-plished by keeping a compass on the assembly bench.;
5 N s
,
S N
S N
1:39 Fluorescent-penetrant Inspection (Zyglo) This is a sensitive nondestructive method of detecting minute discontinuities such as cracks, shrinkage, and porosity that are open to the surface. While this method may be ap-
plied to both magnetic and nonmagnetic materials, its primary application Electric
When fine discontinuities are under the
.
Coil
Crack
magnetic
current
field S
still be deflected to the surface, but the are attracted, so that the indication ob-
t
/
itinuity is far below the surface, no leak-
id consequently no indication. Proper
I
jssary to ensure that the magnetic field
\e discontinuity and give the clearest
f
)
iinal magnetization, the magnetic field
j allel to the long axis of the workpiece -
I by an electric current so that the long ds of the coil. The metal part then bet and is magnetized by induction from )!!. Very long parts are magnetized in sngth. In the case of circular magnetiignetic field transverse to the long axis d by passing the magnetizing current
3
3>
,
:
»
t
,
and rectified alternating current are
Iron powder
Direct current is more sensitive than
Magnetic field
scontinuities that are not open to the ect discontinuities open to the surface /lis type of discontinuity is the only in'
I Electric I current
i
actified, it provides a more penetrating
-
Fig. 1-32
,
Illustrating two kinds of magnetization: (a) Longi-
i I: tudinal magnetization; (/>) circular magnetization. (Magna,
.
e inspection is affected by many fac-
flux Corporation Chicago, III.) ,
I
JL, ' -
"
"
: :
:
'
r
.
i
i
v
52
INTRODUCTION TO PHYSICAL METALLURGY
f
-;
is for nonmagnetic materials. specting any homogeneous ma '
glass, plastic, and some ceramn Parts to be tested are first tre;
ally light, oil-like liquids which a ing, or in some other convenier into cracks and other discontini
penetrant has had time to seep i
removed by wiping or washing. connected discontinuities. The l
or a suspension of powder in a lii sponge drawing the penetrant f the area of penetrant indication. completed, the penetrant must b
[0 [6]
der.
One method is to use con
veioper.
A combination of whi
common.
Another method is to use a flL
fluorescent penetrant inspection exactly the same as described pre contains a material that emits vis
4t mm.
radiation. Lamps that emit uitrav visible light they might normally e pear black or dark purple. When black light, the defect appears as background. Figure 1 -35 shows i being tested by fluorescent penel Fluorescent penetrant inspectic in castings, cracks in thefabricatii and pits in welded structures, cr P r.etrate
Wash
/
1 Fluorescent penetrant is
Fig. 1-33 Typical defects revealed by Magnaflux and
drawn into crack by capillary action
Water spray removes penetrant from surfa' but not from crack
Magnaglo. (a) Grinding cracks; (b) fatigue crack in an aircraft crankshaft; (c);casting cracks in a lawnmower casting;
(rf) cracks in critical-jet engine blades. (Magnaflux Corporation, Chicago, III.)
\ <
i
| Fig. 1-34 Major steps in fluorescent-penetrant insp
i; TOOLS OF THE METALLURGIST
53
I
J
is for nonmagnetic materials. Penetrant techniques ban be used for in'
specting any homogeneous material that is not poroMS, such as metals, glass, plastic, and some ceramic materials. Parts to be tested are first treated with a penetrant. Penetrants are usu-
ally light, oil-like liquids which are applied by dipping, spraying, or brushing, or in some other convenient manner. The liquid penetrant is drawn into cracks and other discontinuities by strong capillary action. After the penetrant has had time to seep in, the portion remaining on the surface is removed by wiping or washing. This leaves the penetrant in all surfaceconnected discontinuities. The test part is now treated with a dry powder or a suspension of powder in a liquid. This powder or developer acts like a sponge drawing the penetrant from the defect and enlarging the size of the area of penetrant indication. In order for the inspection process to be completed, the penetrant must be easily observed in the developing powder. One method is to use contrasting colors for the penetrant and developer. A combination of white developer and red penetrant is very common.
Another method is to use a fluorescent penetrant. The major steps in fluorescent penetrant inspection are shown in Fig. 1 r34. The steps are
exactly the same as described previously except that the penetrating liquid contains a material that emits visible light when it is exposed to ultraviolet radiation. Lamps that emit ultraviolet are called black lights, because the visible light they might normally emit is stopped by a filter, making them appear black or dark purple. When the part to be inspected is viewed under black light, the defect appears as a bright fluorescing mark against a black
background. Figure 1 -35 shows a nonmagnetic stainless steel valve body being tested by fluorescent penetrant.
Fluorescent penetrant inspection is used to locate cracks and shrinkage in castings, cracks in the fabrication and regrinding of carbide tools, cracks and pits in welded structures, cracks in steam- and gas-turbine blading,, Penetrate
Wosh
Inspect
Develop
iilif
/ - / .f r w
I
-
.
.
P £2 (
7i
Fluorescent penetrant is Water spray removes drawn into crack by penetrant from surface but not from crack capillary action
i
Developer acts like a blotter ta diaw
Black light causes penetrant to glow in
penetrant out of crack
dark
an air-
casting; iorpo-
1
1-34
Major steps in fluorescent-penetrant inspection. 1
V
54
INTRODUCTION ; TO PHYSICAL METALLURGY .
i Signal reflected
m
1
i
4
/
I
-
.
.
:
.
i
Transmitting
Electrical
ultrasonic transducer
pulse
1
generator
Tesl specimen
Fig, 1-36 The through-transmission and pulse-ec methods of ultrasonic inspection ,
1
i
frequency of the alternating elei
mechanical vibration (sound w widely used ultrasonic transduc one form of energy to another Two common ultrasonic test
\
.
the pulse-echo methods are ill ,
Fig. 1 -35
mission method uses an ultrasc being inspected If an electrical
Nonmagnetic stainless steel valve body being
inspected by fluorescent penetrant. (Magnaflux Corporation, Chicago, ML)
.
to the transmitting crystal
,
and cracks in ceramic insulators for spark plugs and electronic appli-
opposite side receives the vibra
cations. 1-40 Ultrasonic
tin
through the specimen to the otl
signal that can be amplified anc
'
Inspection The use of sound waves to determine defects is a very ancient method. If a piece of metal is struck by a hammer, it will radiate certain audible notes, of which the pitch and damping may be influenced by the presence of internal flaws. However, this technique of
oscilloscope a meter or some travels through the specimen v received is relatively large If tl,
,
.
hammeripg and listening is useful only for the determination of large
defects. |
A more refined method consists of utilizing sound waves above the audi-
Reflectio,
ble range with a frequency of 1 to 5 million Hz (cycles per second)-hence
from floi
the term; ultrasonic. Ultrasonics is a fast, reliable nondestructive testingf
method |which employs electronically produced high-frequency sound voltage
waves that will penetrate metals, liquids, and many other materials- all speeds of several thousand feet per second.
destructive testing are usually produced by piezoelectric materials. These! material! undergo a change in physical dimension when subjected to an
Imtiol pulse from transmitter
electric ;field. This conversion of electrical energy to mechanical energyl i
fime
is known as thep/ezoe/ecfr/c effect. If an alternating electric field is appliedl to a piezoelectric crystal, the crystal will expand during the first half oi the cycle and contract when the electric field is reversed. By varying the !
|\/vwvw
Ultrasonic waves for non
Fig. l -37
j
Oscilloscope pattern for the pulse-echo
of ultrasonic inspection
,
TOOLS OF THE METALLURGIST
55
J m :
-
Transmitting
Electrical ncol
ul1
pulse generator
-
-nr
,
Receiving
Indicating 'ndlcot'n5 dev,ce
ultrasonic u|t[OSonic
ic
ransdUCe, '
,,onsducer Test specimen
Fig. 1-36 The through-transmission and pulse-echo methods ol ulliasonic inspection
.
frequency of the alternating electric field we can vary the frequency of the ,
mechanical vibration (sound wave) produced in the crystal Quartz is a widely used ultrasonic transducer. A transducer is a device for converting .
one form of energy to another. Two common ultrasonic test methods the through-transmission and the pulse-echo methods are illustrated in Fig. 1-36. The through-trans,
,
Sipeing
mission method uses an ultrasonic transducer on each side of the object being inspected. If an electrical pulse of the desired frequency is applied
1
rporafion,
to the transmitting crystal the ultrasonic waves produced will travel ,
through the specimen to the other side. The receiving transducer on the .
s for spark plugs and electronic appli-1
opposite side receives the vibrations and converts them into an electrical
signal that can be amplified and observed on the cathode-ray tube of an 1 sound waves to determine defects is a
.
oscilloscope, a meter or some other indicator. ,
of metal is struck by a hammer, it will 3 I
:
[ which the pitch and damping may be .
f-
If the ultrasonic wave
travels through the specimen without encountering finy flaw
the signal received is relatively large. If there is a flaw in the path of the ultrasonic
[;
ternal flaws. However, this technique of ;:.:ful only for the determination of large
5 of utilizing sound waves above the audi-1
Retlection from flow
) 5 million Hz (cycles per second)-hence \
is a fast, reliable nondestructive testing |
+
nically produced high-frequency sound mt s,
Voltage
liquids, and many other materials at J-.
per second. Ultrasonic waves for nonoduced by piezoelectric materials. These .
Retlection from back surface
Initial pulse f ram transmit ter
physical dimension when subjected to an f jf electrical energy to mechanical energy
Time
Ut
Imil
i
bt. If an alternating electric field is applied Jft
stal will expand during the first half of fjjk
electric field is reversed
.
By varying the
Fig. 1 -37
Oscilloscope pattern for the pulse-echo method
oruitrasonic inspection
.
.
i
it
,
I wave.ipart of the energy will be reflected and the signal received by the
receiv|ng transducer will be reduced. Thelpulse-echo method uses only one transducer which serves as botR transn*jitter and receiver. The pattern on an oscilloscope for the pulse-echo method would look similar to that shown in Fig. 1-37. As the sound wave
enters the material being tested, part of it is reflected back to the crystal wherejit is converted back to an electrical impulse. This impulse is ampli-
fied arid rendered visible as an indication or pip on the screen of the oscilloscope. When the sound wave reaches the other side of the material, it is reflected back and shows as another pip on the screen farther to the right of the first pip. If there is a flaw between the front and back surfaces of the material, it will show as a third pip on the screen between the two indica5
tions for the front and back surfaces.
Since the indications on the oscil-
r
-
loscope screen measure the elapsed time between reflection of the pulse Fig. 1-39 An eddy current tester and two enc (Magnetic Analysis Corporation, Mount Verno
I
from the front and back sur measure of the thickness of
therefore be accurately deter V
In general, smooth surface
testing pulse and thereby per mission of the ultrasonic wav
test results.
{
w
s f K
For large parts
,
the crystal searching unit an
in a tank of water, oil, or g sound waves through the m (Fig. 1-38). Close examinati shows the presence of three piece, the right pip the back indication of a flaw.
Ultrasonic Inspection is usi age cavities, internal bursts elusions.
\ i
f
\ i
Fig. 1-38 Ultrasonic inspection by immersion tank. (Fansteel Metallurgical Corporation)
I 1 T
in a water
!
Wall thickness ca
where such measurement ca
141 Eddy Current inspection Edc electrically conducting mate and variations in compositioi field is produced if a source When this field is placed ne electric current, eddy curren currents, in turn, will produci
unit will measure this new m;
TOOLS OF THE METALLURGIST
57
sflected and the signal received by the id. '
:
'
i:
ly one transducer which serves as both n on an oscilloscopeforthe pulse-echo shown in Fig. 1-37. As the sound wave art of it is reflected back to the crystal -
bctrical impulse. This impulse is amplipation or pip on the screen of the oscilches the other side of the material, it is j-
r pip on the screen farther to the right of /een the front and back surfaces of the
;;
jySilon the screen between the two indicales Since thejndications on the oscil.
ed time between reflection of the pulse
Fig. 1-39 An eddy current tester and two encircling coils. (Magnetic Analysis Corporation, Mount Ve.-non, N.Y.)
from the front and back surfaces, the distance between indications is a
measure of the thickness of the material.
The location of a defect may
therefore be accurately determined from the indication on the screen. In general, smooth surfaces are more suitable for the higher-frequency testing pulse and thereby permit detection of smaller defects. Proper transmission of the ultrasonic wave has a great influence on the reliability of the test results. For large parts, a film of oil ensures proper contact between the crystal searching unit and the test piece. Smaller parts may be placed in a tank of water, oil, or glycerin. The crystal searching unit transmits sound waves through the medium and into the material being examined (Fig. 1-38). Close examination of the oscilloscope screen in this picture shows the presence of three pips. The left pip indicates the front of the piece, the right pip the back of the piece, and the smaller center pip is an
I
indication of a flaw.
i
Ultrasonic inspection is used to detect and locate such defects as shrinkage cavities, internal bursts or cracks, porosity, and large nonmetallic inclusions.
ill
Wall thickness can be measured in closed vessels or in cases
where such measurement cannot otherwise be made. 1-41
Eddy Current inspection
Eddy current techniques are used to inspect
electrically conducting materials for defects irregularities in structure, and variations in composition. In eddy current testing a varying magnetic field is produced if a source of alternating current is connected to a coil. When this field is placed near a test specimen capable of conducting an ,
,
i
electric current, eddy currents will be induced in the specimen, the eddy in turn, will produce a magnetic field of their own, The detection unit will measure this new magnetic field and convert the signal into a volt-
currents er
f
,
!;
58
INTRODUCTION TO PHYSICAL METALLURGY
1
TABLE 1 8 jMajor Nondestructive Testing Methods INSPECTION
METHOD f
I
Edrly current
i!
'
A
WHEN TO USE
WHERE TO USE
ADVANTAGES
Measuring variations in
Tubing and bar stock, parts of uniform geometry,
High speed, noncon-
Fal
tact, automatic.
fro
wall thickness of thin met-
als or coatings; detecting longitudinal seams or cracks in tubing; determining heat treatments and metal compositions for sorting
flat stock, or sheets and
Detecting internal flaws and defects; finding weld-
Assemblies of electronic
works well on thin
sections; high sensitivity; fluoroscopy techniques
lack of fusion; measuring
parts, casting, welded vessels; field testing of welds; corrosion surveys; components of nonmetallic
variations In thickness.
materials.
goe
wire.
ma
of |
Provides permanent
.
-
Radiographjiy: x-rays
i
record on film;
ing flaws, cracks seams, porosity, holes. Inclusions, ,
Garrjma
Detecting Internal flaws,
Forgings, castings, tubing,
x-rays
cracks, seams, holes. In-
welded vessels; field test-
clusions, weld defects;
ing welded pipe; corrosion
measuring thickness vari-
surveys.
i
LIN
available; adjustable
Hig sou
haz
nici
energy level.
ations.
Detects variety of flaws; provides a permanent record; portable; low
Ons
Initial cost; source is
sou
small (good for inside shots); makes panoramic
cor
sou
trai
exposures.
cracks, porosity, non-
Only for ferromagnetic materials; parts of any size, shape, composition,
metallic inclusions, and
or heat treatment.
Magnetic
Detecting surface or shal -
Darticle
low subsurface flaws,
Economical, simple in principle, easy to per-
<
weld defects. Penetrant
;:
j
Locating surface cracks,
porosity, laps, cold shuts, lack of weld bond, fatigue, and grinding cracks.
All metals, glass, and ceramics, castings, forgings, machined parts, and cutting tools; field inspections. All metals and hard non-
Ultrasonic
Finding internal defects,
pulse echo
cracks, lack of bond, lami-
metallic materials; sheets,
nations, inclusions, porosity; determining grain
tubing, rods, forgings, castings; field and production testing; inservlce part testing; brazed and adhesive-
structure and thicknesses.
! 1 i
\
I
ism
I it
tion testing.
clei
ne€
Simple to apply, portable,
Lin
fast, low in cost; results
sur
I
Fast, dependable, easy to operate; lends itself to
Ret sio
automation, results of
of t
test immediately known;
trai
relatively portable, highly accurate, sensitive. .
Metals Progress Data Sheet, August 1968. American Society lor Metals, Metals Park, Ohio.
is r
orate setup required.
L
'
form; portable (for field testing); fast for produc-
easy to Interpret; no elab-
bonded joints.
v- - J
.
Ma
der
'
TOOLS OF THE METALLURGIST' 59
Methods i
-
WHERE TO USE
ADVANTAGES
Tubing and bar stock, parts of uniform geometry.
High speed, noncon-
False indications result
lin met-
tact, automatic.
iecting
flat stock, or sheets and
from many variables; only good for conductive materials; limited depth of penetration.
: >:-:
-
.
.
.
v
-
.
is in
wire.
bter-
LIMITATIONS
snts :ions
Provides permanent record on film; :*: ;::
'
"
Assemblies of electronic
works well on thin
sections; high sensitivity; fluoroscopy techniques
usions,
parts, casting, welded vessels; field testing of welds; corrosion surveys;
suring
components of nonmetallic
energy level.
3SS.
materials.
1aws weldjams,
'
aws.
.
.
.
:>->:s in.
:ts;
s es
vari-
available; adjustable
High initial cost; power source required; radiation hazard; trained tech-
nicians needea.
Forgings, castings, tubing, welded vessels; field test-
ing welded pipe; corrosion surveys.
Detects variety of flaws; provides a permanent record; portable; low
One energy level per
initial cost; source is
source loses strength
small (good for inside shots); makes panoramic
continuously.
source; radiation hazard; trained technicians needed;
exposures.
r shals, n
-
Only for ferromagnetic materials; parts of any size, shape, composition,
and
or heat treatment.
acks,
All metals, glass, and ceramics, castings, forgings, machined parts, and cutting tools; field inspections.
shuts.
atigue,
ects,
All metals and hard non-
J lami-
metallic materials; sheets,
porosin
tubing, rods, forgings, castings; field and production testing; inservice part testing; brazed and adhesivebonded joints.
,
esses.
is required; power source needed; parts must be
tion testing.
cleaned before finishing.
Material must be magnetic;
Simple to apply, portable,
Limited to surface defects.;
fast, low in cost; results
surfaces must be clean.
automation, results of
test immediately known;
I Requires contact or immejsion of part; Interpretation of readings requires training.
relatively portable, highly
V .
I
;
.
form; portable (for field testing); fast for produc-
Fast, dependable, easy to operate; lends itself to
i
.
demagnetizing after testing
easy to interpret; no elaborate setup required.
:
ian Society for Metals, Metals Park, Ohio.
..
Economical, simple in principle, easy to per-
accurate, sensitive. -
r
-
1 60
INTRODUQTION TO PHYSICAL METALLURGY
:
age that can be read on a meter or a cathode-ray tube. Properties such as % hardness lloy composition, chemical purity, and heat treat condition |: ,
influince the magnetic field and may be measured directly by a single coil. | An:limportant use for eddy current testing is sorting material for heat|
I
treat/variations or composition mix-ups. This application requires the use 1
of twib coils (see Fig. 1 -39). A standard piece is placed in one coil and the f
test f|iece in fhe other coil. Acceptance or rejection of the test piece may be |
deter|nined by comparing the two patterns on the oscilloscope screen. |;
Eddy current testing may be used to detect surface and sub-surface 1
defeqts, plate or tubing thickness, and coating thickness. A summary of the major nondestructive testing methods is given in j:
11
Tab,e'1-8. 1-42
Recent-Developments
j The most interesting of the recent developments
;
-
in nondestructive testing methods is the use of holography,1 a unique | method of recording on film visual data about a three-dimensional object f
and recreating a three-dimensional image of the object. Whereas conven- \ tional photography shows the image of an object on film, the holographic j; process records interference patterns which are used to reconstruct the f image.
A simplified setup for making and viewing holograms is shown in Fig. '
i
J
f '
1-40. The narrow, single-wavelength light beam from the laser passes .
through a lens-pinhole assembly to increase its area of coverage. The |.
diverged beam strikes a mirror and the object. The reference beam re-1 fleeted from the mirror and the light reflected from the object both strike j?
a phcitographic plate. Since no lens is used to focus the light from the f;
3T-,;
object, the film records not an image but the Interference pattern resulting | i
f
Much of the following desenpuan was taken Irom publications ot GC Optronics, Inc., Ann Arbor, Mich. '
s
Relcrerce mirror
Loser
-
U
/ I
/ Reco'ding
-
Lens - pinhole ossembly
I
Fig. 1 -41
A real-time hologram of an aluminui
ndwich structure with a T-extrusion. Two de revealed after mild stressing by heat. (GC Optr
j.a
Ann Aroor, Mich.)
J*L Honeycon'b
~
t -Na/ -sandwich
J
;
-
-
.
L®-
structure
'"
'
from a mixture of these two I
(object)
'
-
.
and the hologram is ready fc
Photographic plate
to the reference beam used d
Reference mirror .
Laser
,
f -t -"I
Virluo1 image ot the object
f
Viewing
Lens-pinhole assembly Observer or
Developed hologram
camera
t; ,
Fig. 1-40 A simplified setup for making and viewing holograms. (GC Optronics, Inc., Ann Arbor, Mich.)
An observer, looking througl
Reconstrucied
V
sees a reconstructed, three
apparently in the exact posit In the real-time method, tf
constructed virtual image is then stressed. Any stress-ir
as fringe patterns on the pi areas, and inclusions. Fig. 1
I
:
-
f
.
t
4 TOOLS OF THE METALLURGIST
:
y
61
a cathode-ray tube. Properties such as mical purity, and heat treat condition
ay be measured directly by a single coil
ent testing is sorting material for heat |
.
<-ups
Flaws
'
This application requires the use
.
idard piece is placed in one coil and the ance or rejection of the test piece may be patterns on the oscilloscope screen.
:
sed to detect surface and sub-surface
and coating thickness.
T
-
. ..
.
.
destructive testing methods is given in nteresting of the recent developments s is the use of holography,1 a unique I data about a three-dimensional object J image of the object. Whereas conven-
;
.
section
3J
i
-
<
> ge of an object on film, the holographic
.
.
v
ri-varns which are used to reconstruct the
: .
nd viewing holograms is shown in Fig. igth light beam from the laser passes to increase its area of coverage. The id the object. The reference beam re-
i
jht reflected from the object both strike bns is used to focus the light from the
be but the interference pattern resulting IHP '
'
'
-
-
.
IS of
Fig. 1 -41
A real-time hologram of an aluminum honeycomb-
sandwich structure with a T-extrusion. Two debonds are
revealed after mild stressing by heat (GC Optronics, Inc., .
Reference mirror
Ann Arbor, Mich ) .
J\ Honeycomb
'
< J*; ; y y
struciurc
'
(oojeel)
.
Photographic plote Reference mirror ,
Ai
Q
-
Reconstructed
from a mixture of these two beams. After exposure, the film is developed and the hologram is ready for viewing. In viewing the laser light, similar to the reference beam used during the exposure illuminates the hologram. An observer, looking through the hologram as though it were a window sees a reconstructed, three-dimensional image of the original object apparently in the exact position occupied by the object during exposure In the real-time method, the original hologram is viewed so that the reconstructed virtual image is superimposed on the object The "object is then stressed. Any stress-induced deformation of the object will appear as fringe patterns on the picture and will reveal voias flaws, unbonded areas, and inclusions. Fig. 1 -41 shows a hologram of an aluminum honey,
,
,
,
.
"
'
::
'
f:::-
Developed hologram
.
1 holo-
,
62 i
..
INTRODUCTION TO PHYSICAL METALLURGY
.
i 1
comb-sandwich structure with a T-extrusion bonded to the surface. The
1-31
structure was mildly stressed by heat, and the real-time hologram reveals.
1-32
Explain the difference betw of ultrasonic inspection. 1 -33 What is a transducer? 1-34 Name and explain three tra
two debonds under the T-section. l
1-35
QUESTIONS 1-1
1-2 1-3
What are the limitations of
Which nondestructive test
thickness at the bottom of a st«
How are thermocouples calibrated? What factors may lead to errors in a thermoelectric circuit? Aside from being able to measure high temperatures what is another advan-
1-36
Which nondestructive test
mixed steel?
,
tage ofithe optical pyrometer? 1-4
Assuming that bare not fused wires of copper and constanlan are put into a liquid metal below the melting point of copper, will the thermocouple measure the temperature of the liquid metal? 1 -5 Assume that the thermocouple wires are reversed when connected to the potentiometer; how may this be detected?
'
If
i
,
"
How is
1-7
Differentiate between resilience and toughness
"
"
.
-8 Which property in a tension test is an indication of the stiffness of a material?
1 -9 1-10
Which properties in a tension test indicate the ductility of a material?
5:
Hfow will the speed of testing affect the yield strength and ultimate strength?
-11 Cj n a stress-strain graph for a load beyond the yield strength that is suddenly removela, show the elastic strain and the plastic strain.
1
Why is the yield strength usually determined rather than the elastic limit?
1'13
what is the difference between the proportional limit and the elastic limit?
'
-14 v|hy are impact specimens notched?
New York, 1960.
f
1-15
Discuss the effect of the type of notch and velocity of the hammer on the | results bf the impact test.
ii
I
?
1-20
Vyhy is a correction factor necessary for Rockwell readings on a specimen '% Is the correction factor in Question 1 -20 to be added or subtracted from the
Book Gompany, New York, 194 Lysaght, V. E.: "Indentation Hare
What is the minimum thickness of the specimen if a reading is to be taken
l
If the specimen in Question 1 22 is to be checked with the Brinell test what |
what should its minimum thickness be?
„
|
How may one determine whether the specimen was too thin to be checked with %
a particular Rockwell scale?
|-
liist three factors that contribute to the inaccuracy of a scieroscope reading
1-27
What factors may be varied in taking a radiograph with x-rays? 1-28 In a radiograph what will be the difference in appearance of cracks, and impurities? 1-29 What are the limitations of magnetic-particle inspection? ,
.
|
gas cavities, |
What are the limitations of fluorescent-panetrant inspection?
|
I
m i
New York, 1949. New York, 1959.
should|ts minimum thickness be? | 1 -24 If the specimen in Question 1 -22 is to be checked on the Rockwell 15 N scale. |
1-30
1957.
Williams, S.
in the r nge of Rockwell C 60? (Refer to "Metals Handbook," 1948 edition.)
1-26
III., 1940.
McMaster, R. G.. "Nondestructive
observed readings? Explain.
1 -25
phia. Pa., 1954. Doane, F. B.: "Principles of Magna
'
less thSin 1 in. in diameter?
1 -23
Dike, P. H.- "Thermoelectric Ther
Kohl, G. L. "Principles of Metallc
Is there a unit associated with the Rockwell hardness number? Explain.
1-22
"
Why is it possible to obtain the approximate tensile strength of steel by 500 f
1 -19
1 -21
neering Materials, 2d ed., McG
Eastman Kodak Company: "Rac
times the Brinell hardness number?
i
Davis, H. E., G. E. Troxell, and G.
r p-
hat limits the range of hardness in the Brinell machine? 1 -17 What are the units for the Brinell hardness number?
1-16
1-18
Garlin, B.: "Ultrasonics," McGraw
Goxon, W. F.: "Temperature Mea
,
1-12
1
American Society for Metals: "Me : Nondestructive Testing f : Temperature Measureme American Society for Testing anc Philadelphia, Pa.. 1971. Betz, G. E.; "Principles of Penetr
true stress" calculated?
1-6
1
REFERENCES
R.
"Hardness and
Metals, Metals Park, Ohio, 1942
-
r
1
:
TOOLS OF THE METALLURGIST
63
K
1
sxtrusion bonded to the surface.
The
131 1 -32
M at, and the real-time hologram reveals
What are the limitations of ultrasonic inspection? Explain the difference between through-transmission and pulse-echo methods
of ultrasonic inspection. 1-33
What is a transducer?
1 -34
Name and explain three transducers
.
1-35
Which nondestructive testing method is best suited to determine the wall thickness at the bottom of a steel tank?
1 -36 ;a thermoelectric circuit?
Which nondestructive testing method should be used to sort out bars of
mixed steel?
ihigh temperatures what is another advan,
:
.
s of copper and conslantan are put into a
;
SSV copper, will the thermocouple measure the
REFERENCES
Amnrionn Socioty lor Molrtls: "Melnls Handbook : Nondestructive Testing for Management
"
.
p are rovorsQd when connoclod to the polen-
"
IfMS od.. Metals Park, Ohio.
"
Metals Park, Ohio, 1963. Temperature Measurement," Metals Park, Ohio 1956. American Society for Testing and Materials: "Annual Book Of Standards Part 31, Philadelphia, Pa., 1971. ,
:
"
,
"
,
d toughness. ;
an indication of the stiffness of a material?
/
"
.
1
Betz, C. E.: "Principles of Penetrants Magnaflux Corporation, Chicago, III., 1963. McGraw-Hill Book Company, New York, 1960. Coxon, W. F.; "Temperature Measurement and Control The Macmillan Company, "
,
indicate the ductility of a material?
'
.
.
"
'
'
v
.
Carlin, B.: "Ultrasonics
.
"
ct the yield strength and ultimate strength? id beyond the yield strength that is suddenly
,
"
,
New York 1960. ,
plastic strain. ' .
v-
' -
.
"
Davis, H. E., G. E. Troxell, and C. T. Wiskocil: "The Testing and Inspection of Eingineering Materials, 2d ed., McGraw-Hill Book Company, New York, 1955. Dike, P. H.: "Thermoelectric Thermometry Leeds & Northrup Company, Philadelphia, Pa., 1954. Doane F. B.: "Principles of Magnaflux Inspection," Magnaflux Corporation, Chicago;
determined rather than the elastic limit?
"
e proportional limit and the elastic limit?
"
,
3d?
notch and velocity of the hammer on the
,
III., 1940. ;
in the Brinell machine?
'hardness
Eastman Kodak Company: "Radiography in Modern Industry
number?
" ,
Rochester, N.Y.,
1957.
ipproximate tensile strength of steel by 500
Kehl. G. L.: "Principles of Metallographic Laboratory Practice 3d ed., McGraw-Hill Book Company, New York, 1949. Lysaght, V. E.; "Indentation Hardness Testing Van Nostrand Reinhold Company, "
,
i Rockwell hardness number? Explain. ;sary for Rockwell readings on a specimen
"
,
New York 1949.
.
,
McMaster, R. C: "Nondestructive Testing Handbook." The Ronald Press Company
pn 1 -20 to be added or subtracted from the
Williams, S. R,:
[of the specimen if a reading is to be taken
"Hardness and Hardness Measurement
"
American Society for
,
Metals Metals Park, Ohio, 1942. ,
to "Metals Handbook," 1948 edition.) is to be checked with the Brinell test what ,
Is to be checked on the Rockwell 15 N scale,
*
.
;he
'
.
specimen wys too thin to be chocked with
\
:
to the inaccuracy of a scleroscope reading.
'
ie difference in appearance of gas cavities
,
i Jetic-particle
inspection?
jscent-penetrant inspection?
v
.
,
New York, 1959.
.
i
.
1
f
'
MET/
i
STRL I
AND CRYS
. .
i
i .
2-1 Introduction
All matter is conside
known as chemical elements. Th --
\ \
-
.
guishable on the basis of their c The elements are composed of at acteristic of each element. 2-2 Atomic Structure
While each cher
the difference in atomic structur
t
properties. It is useful to think of elementary particles: (1) electro (2) protons particles of positiv particles, called neutrons. Almo
>
'
,
I
trated in the nucleus, which cont
.
Si
of the proton is approximately slightly heavier, approximately 1.
tron is much lighter, approximiti is approximately 1/1800 the mass
is of the order of lO-12 cm and is \
1
eter, which is approximately 10"' tively charged nucleus made up a sufficient number of electrons neutral.
Since the electron and
charge, the neutral atom must c protons.
m i:
r i '" -
T
.
I i
m
MM WW""1
i
The discovery of the atomic Crooks, an Englishman, while * inside a partially evacuated tub rays could turn a small paddle v\ shadow of any solid object plao the rays were material in nature
I'
km
METAL STRUCTURE AND
i
CRYSTALLIZATIO i
2-1
All matter is considered to be composed of unit substances known as c/?em/ca/ i/emef?fs. These are the~smallest units that are distin-
introduction
~
guishable on the basis of their chemrcal a ctivity and physical properties. The elements are composed of atoms which have a distinct structure char'
acteristic of each element. 2-2
Atomic Structure
While each chemical element is composed of atoms it is ,
the difference in atomic structure that gives the element its characteristic properties. It is useful to think of the free atom as being composed of three elementary particles: (1) electrons, tiny particles of negative electricity, (2) protons, particles of positive electricity, and (3) electrically neutral particles, called neutrons. Almost the entire mass of the atom is concentrated in the nucleus, which contains the protons and neutrons. The mass of the proton is approximately 1.673 x 10 1 g and the neutron is just slightly heavier, approximately 1.675 x IQ-24 g, whiie the mass of the elec-
V
_
J_
.
tron is much lighter, approximateiy 9.11 x lO-2* g. Therefore, the electron is approximately 1/1800 the mass of a proton. The diameter of the nucleus
is of the order of ID-12 cm and is very small compared with the atomic diameter, which is approximately lO-8 cm. The atom consists of a minute positively charged nucleus made up of protons and neutrons surrounded by a sufficient number of electrons to keep the atom as a whole electrically neutral. Since the electron and proton have equal but opposite electrical charge, the neutral atom must contain an equal number of electrons and
i
protons. .
The discovery of the atomic particles started in 1874 when William
1
,
Crooks
,
an Englishman, while experimenting with electrical discharges
inside a partially evacuated tube discovered that the so-called cathode rays could turn a small paddle wheel placed in the tube and would cast a
.
,
:
:
.
tj.%
.
;5
shadow of any solid object placed in their path. This convinced him that the rays were material in nature. From the direction that they were de,
,
Ib66 V
-:.
INTRODUCTION TO PHYSICAL METALLURGY
I: \
5
1 fleeted ih a magnetie field he knew that the particles possessed negative |'
;
;
.
charges.; It was suggested by his colleagues that these rays consisted o( jp
negatively charged particles. In 1897 his countryman, J. J. Thomsonj k' made another discovery. From the extent to which the rays were deflectect £ by a positively charged object and by a magnet, Thompson determined that; |; ,
.
1
the particles of which the rays were composed were traveling at high speeds ind that each one had the mass of about 1/1800 of the hydrogeir atom. Further investigations showed that these little particles had identical ,
1
fe>-
11 +
I2N
I
properties, regardless of the kind of material used as the electrodes. Thus;
it was evident that these small negatively charged particles called elec-1 ,
I
trons, were parts of all atoms. In 1911, fourteen years after Thomson's ; discovery of electrons, Ernest Rutherford bombarded light elements such» Fi9- 2,1 ,
,
.
u- u
,
i i
.
.
.
,
f
Iff)
[6)
Changing concepts of atomic stmcture-t
sodium atom, (a) Lewis-Langmuir model; (6) Bohr t
as aluminum, with high-speed alpha particles. The targets emitted a new; % {c)„ave mftchanics model. kind of particle, positive in charge, and of a mass almost identical witlr i,
that of the hydrogen atom.
These particles were named protons. The
bombardment of light elements, notably beryllium, with alpha particles|
16 neutrons or a total of 16 of th
in 1932, named this particle the neutron. It is sometimes difficult for the mind to conceive of such tiny particles. If the simplest atom, hydrogen, were magnified so that its diameter would
is correct?
be V2 mile, its nucleus would be the size of a baseball, and its electron,
V4 of a mile distant, would be the size of a softball. The first conception
atomic weight, the number of nuc In other words, hydrogen, our ligl
of atomic structure that met with wide acceptance was that advanced byj the American scientists Gilbert Lewis and Irving Langmuir about 1916. In their theory, the protons and neutrons constituted a central dense
The next heavier element, heliun Uranium, the 92nd element, has atomic number of 8, and its atorr
nucleus. Each electron was supposed to occupy a fixed position outsidel
of the atom is composed of eigf the atomic number of an elemei
ranged in concentric cubes. The inner cube could hold eight electrons,
since the atom has to be electr
cube could hold eight electrons, and even more in cases of ele-|
equal to the number of electrons
'
ments of large atomic weight.
Successive cubes showed an increasing
complexiity of electronic structure. Figure 2.1(a) shows the Lewis-Lang-I
The electrons, spinning on the cleus, are arranged in definite sf
twelve neutrons, is placed in the center, and the electrons, represented by
that can fit in each shell is 2nJ, w maximum number of electrons th
dots, are placed at the corners of hypothetical stationary concentric cubes) Shortly thereafter a dynamic model of the atom was developed by the!
further subdivided into energy :
muir mqdel of the sodium atom. The nucleus, with eleven protons and-
Danish scientist Niels Bohr. It possessed fewer defects than the Lewis-| Langmuir model and is substantially the working model now in use. The]
Bohr concept of the sodium atom is shown in Figure 2.1 (fa). According to| this concept, the nucleus of an atom contains all the neutrons and pre*
tons. Hfence the mass or weight of an atom is centered in its nucleus. The] electrons revolve in circular and elliptical orbits about the nucleus some-.] what asithe planets and comets of our solar system revotve about the sun,
Each ijinit of atomic weight is supplied by a proton or a neutron. Thus anjj i
I
Mi
To determine the answer to t
of atomic number. If the element
the nucleus. These positions, beginning with the third electron, were arthe next
: :
atom of oxygen with an atomic v
resulted ;in the emission of a third kind of atomic building unit. Chadwick.l
eight, the third eighteeen, the fc Exclusion Principle, no more tha level, and if two are to fit on the
The number of energy levels incr electrons tend to occupy the lo shells tend to fill up first, but this elements. It is possible for the hi more energy than the lowest enc trated later.
w
.
METAL STRUCTURE AND CRYSTALLIZATION
67
ew that the particles possessed negative colleagues that these rays consisted of
,
:
&sA 1897, his countryman
,
J: J. Thomson
,
f
e extent to which the rays were deflected
i by a magnet
,
Thompson determined that |
were composed were traveling at high J| e mass of about 1/1800 of the hydrogen /ed that these little particles had identical
JJIi:
of material used as the electrodes Thus .
egatively charged particles
,
called elec[a]
gMi 1911, fourteen years after Thomson's "
'
. .
herford bombarded light elements
,
Fig. 21
such ;
.
ha particles. The targets emitted a new
je, and of a mass almost identical with se particles were named protons. The notably beryllium with alpha particles kind of atomic building unit. Chadwick
16 neutrons or a total of 16 of the two. Which of these three possibilities
,
>| utron.
is correct? r
mind to conceive of such tiny particles.
To determine the answer to this question, you must use the concept of atomic number. If the elements are arranged in order of their increasing
sre magnified so that its diameter would the size of a baseball, and its electron,
atomic weight, the number of nuclear protons increases in the same order. In other words, hydrogen, our lightest atom, has one proton in the nucleus.
size of a softball. The first conception wide acceptance was that advanced by
The next heavier element, helium, has two.
Wis and Irving Langmuir about 1916
of the atom is composed of eight protons and eight neutrons. Therefore
osed to occupy a fixed position outside ginning with the third electron, were ar-
the atomic number of an element is equal to the number of protons, or since the atom has to be electrically neutral, the atomic number is also equal to the number of electrons.
inner cube could hold eight electrons,
pctrons, and even more in cases of eleJSjuccessive-cubes showed an increasing .
.
j
-
.
|
.
The electrons, spinning on their own axes as they rotate around the nucleus, are arranged in definite shells. The maximum number of electrons
Figure 2.1(a) shows the Lewis-Lang-
that can fit in each shell is 2n2. where n is the shell number. Therefore the maximum number of electrons that will fit in the first shell is two, the second
! The nucleus with eleven protons and ,
enter, and the electrons, represented by lypothetical stationary concentric cubes.
eight, the third eighteeen, the fourth thirty-two, and so on. Each shell is further subdivided into energy states or levels. According to the Pauli
jodel of the atom was developed by the assessed fewer defects than the Lewis::
.
:?illy
the working model now in use. The
"" '
v
is shown in Figure 2.1 (Jb). According to ytom contains all the neutrons and pro-
'
' -
.
.
f$?i an atom is centered in its nucleus. The
|lliptical orbits about the nucleus someour solar system revolve about the sun. Dplied by a proton or a neutron Thus an
,
-
,
Exclusion Principle, no more than two electrons can fit on any one energy level, and if two are to fit on the same level, they must be of opposite spin. The number of energy levels increases with distance from the nucleus, and electrons tend to occupy the lowest energy levels. Therefore, the i.iner shells tend to fill up first, but this is not always true for some of the heavier elements. It is possible for the highest energy level of an inner shell to have
more energy than the lowest energy level of the next shell, as will be illustrated later.
.
-
The third, lithium, has three.
Uranium, the 92nd element, has 92 protons. Now gince oxygen has an atomic number of 8, and its atomic weight is 16, we know that the nucleus
d neutrons constituted a central dense
.
Changing concepts of atomic structure-the
sodium atom, (a) Lewis-Langmuir model: [ti] Bohr concept; (c) wave mechanics model.
atom of oxygen with an atomic weight of 16 could contain 16 protons or
,
v
(0
i
ill:
TABLE 2-1
i
Atomic Number Atomic Weight, and Chemical Symbols of the Elements* ,
ELEMENT
SYMBOL
"
#
1
Ac A!
13
Am
95
243
Sb
51
121.76
Mi
illI 11
I
if.
m
i
I
The modified Bohr theory is
Mercury Molybdenum Neodymium
Mo
42
Nd
60
Neon
Ne
10
144.2| 20.li
Hg
80
200.6t
Neptunium
Np
93
74.91
237 I
Astatine
N
At
28
85
211
Barium
Nickel Niobium
Nb
Ba
41
56
137.36
58.6 92.9fi
Berkelium
Bk
97
247
Beryllium
Be
4
Bi
Boron
B
Bromine Cadmium Calcium
Br
83 5 35
9 013 .
209.00 10.82
Cd
48
79.916 112.41
Ca
20
40.08
Cf
98
(Columbium) Nitrogen
(Cb) 7
14.00
Nobelium Osmium
No
102
Os
254 4
76
Oxygen
O
190.24 16.001
Palladium
Pd
46
196.71
15
30.971
N
8
chanical model of the atom, b
95.9
39.944
Phosphorus
P
Platinum
Pt
78
Plutonium Polonium
195.23|
Pu
94
242 |
agreed that the electron has t the early 1920s, through the wc
V
the development of wave mec best expressed by mathemati . visualized as concentric halos ties concentrated as in the Bo .
5;
;
;
It is not possible to determi its position is determined by t region of the atom. This probe
i tain wave function and the s ,
Californium Carbon Cerium
Ce
Po
84
58
140.13
K
210 |
Cesium
Potassium
quantities known as the quam the principal quantum numbe in a particular state and may h
Cs
19
55
132.91
39.100
Pr
CI
Praseodymium
turn number / is a measure of
CMIorine Chromium Cobalt
59
17
35.457
61
24
54.01
Promethium Protactinium
Pm
Cr
140.92' 145
Pa
Co
91
27
58.94
Copper
Radium
Cu
Ra
88
29
63.54
Radon
Rn
Cm
86
96
Re
75
231 | 226.051 222 1 mail
Rh
45
102.91
Rb
37
C
6
251
12.011
247
87
233
Rhenium Rhodium Rubidium Ruthenium Samarium Scandium Selenium Silicon
Gd
64
156.9
Silver
Ga Ge
Ag
47
31
107.88(
69.72
Na
11
32
22.991
72.60
Sr
38
87.63J
.
Dysprosium
Dy
66
162.46
Einsteinium
E
99
245
Erbium
Er
68
167.2 152.0
Europium
Eu
63
Fermium
Fm
100
Fluorine Francium Gadolinium Gallium Germanium
F
9
Fr
253 19.00
Ru
44
101.1 ,
Sm
62
150.43
Sc
21
44.96:
Be
34
78.96
Si
14
28.09:
79
197.0
S
16
Hf
72
178.6
Tantalum
Ta
73
He
2
Holmium
43
67
Technetium Tellurium Terbium
re
Ho
Te
52
127.61 J 158.931 204.39232.05
4 003 .
Hydrogen
H
1
Indium
in
49
Iodine Iridium
I
53
126.91
Ir
77
192.2
Iron
Fe
26
Krypton
Kr
Lanthanum Lawrencium ; Lead
Lithium
Li
i
1 008
98
65 81
Th
90
Tm
69
55.85
t68.94:
Tin
Sn
50
36
118.70
83.80
Titanium
Ti
La
57
138.92
Lw
103'
Pb
82
238.07 |
U
92
V
23
Xenon
Xe
54
131.3
Ytterbium Yttrium Zinc
Yb
70
173.04
Y
39
88.92
Zn
30
65.38
Zirconium
Zr
40
91.22
Lu
71
174.99
12
24.32
Mn
25
54.94
Mendelevium
Mv
101
256
mements o 104 3 "00 (Ha) ' l961 6 011, Arnerican Society for Metals nl Hahnium t+ El and 105 Kurchatorijm (Ku) have also beenMetals foundPark, Ohio. .
.
.
t
47.90: 183.92-
Uranium
Mg
w
22 74
Vanadium
.
1
2
0
-
il
-
H
-
CP
207.21 6 940
-
il
'
257
Lutecium i Magnesium Manganese :
3
-
:
Tb
3
momentum in a specified dir eluding zero. The fourth qu? electron on its own axis, ma; direction of the spin. Figure;
mi
32.066
Tl
Tungsten
6?
Is called the (2p) level, e
iS/-,. The quantum number m, i
180.95
Thallium Thorium ThuJium
.
114.76
Mv!";
'
Au
164.94
fling to n = 1 and / = 0 is called is: ? ||/>», 1 .
Hafnium Helium
I
may have values from 0 to (n introduced to signify / = 0,1,2
85.48*
Sodium Strontium Sulfur
'
mi
AT.
18
Gold
-r
26.98
NO.
33
Curium
iiii
227
AT,
SYMBOL
A i
if
'
89
ELEMENT
As
Arsenic
Bismuth
m
AT. WT.
Aluminum Amerlcium
Argon ri'l' /'j
NO.
Actinium
Antimony n
AT.
-
ff
H
-
50.95
fij. 2-2 Schematic representation of possib "
v
:
-
alfetes in an atom. (From M. M. Eisenstadt, I Mechanical Properties of Materials, The Ma "
<5oinpany, New York, 1971.)
1
. .
.
;
.
.
v.7.
--
-
.
-
.
:
i
in,
METAL STRUCTURE AMD CRYSTALLIZATION
69
ihemical Symbols of the Elements* ELEMENT
SYMBOL
Mercury Molybdenum
Hg
1 Neodymium
.
NO. 80
AT. WT
Mo
42
95.95
Nd
60
144.27
Ne
10
Np
93
Nickel
Ni
28
58.69
Nb
41
92.91
jNiobium (Columbium)
The modified Bohr theory is our nearest approach to a satisfactory mechanical model of the atom, but it contains certain defects. It is generally agreed that the electron has the nature of both a particle and a wave. In
200.61
Neptunium
Neon '
AT.
the early 1920s, through the work of Heisenberg and Schroedinger and with
20.183
the development of wave mechanics, the modern concept of the atom is best expressed by mathematical equations. The electron shells may be visualized as concentric halos of negative electricity with its greatest densi-
237
(Cb) 7
Nitrogen
N
Nobelium
No
102
Os
76
ties concentrated as in the Bohr shells, Fig. 2-1 (c).
4 008 .
Rn
86
222
Rhenium
Re
75
186.31
Rhodium
Rh
45
102.91
Rubidium
Rb
37
Ruthenium
Ru
44
101.1
Samarium
Sm
62
150.43
It is not possible to determine the exact orbit of an electron, but rather its position is determined by the probability that it will be found iri a given region of the atom. This probability is represented mathematically by a certain wave function, and the solution of the wave equation leads to four quantities known as the quantum numbers n, I, m,, and ms. Of these, n is the principal quantum number related to the total energy of the electron in a particular state and may have values n = 1, 2, 3, etc. The second quantum number / is a measure of the angular momentum of the electron and may have values from 0 to (n - 1). The letters s, p, d, f, g, and h have been introduced to signify / = 0, 1, 2, 3, 4, 5, so that the energy level corresponding to n = 1 and / = 0 is called the (Is) level, that corresponding to n = 2 and / = 1 is called the (2p) level, etc. The quantum number m, is related to the component of the angular momentum in a specified direction and may have values of+1 to -1, including zero. The fourth quantum number m.., related to the spin of the electron on its own axis, may have the value of ± Vs depending upon the
Scandium
Sc
21
44.96
direction of the spin. Figure 2 2 schematically .shows some of the possible
Selenium
Se
34
78.96
Silicon
Si
14
28.09
Silver
Ag
47
107.880
Sodium
Na
11
22.991
Strontium
Sr
38
87.63
Sulfur
S
16
32.066
254
"
iOsmium .
,
.
f
.
:
16.000
Oxygen
0
Palladium
Pd
46
Phosphorus
P
15
Platinum
Pt
78
195.23
Plutonium
Pu
94
242
Polonium
Po
84
210
Potassium
K
19
'
..
8
190.2
Praseodymium
196.7
30.975
39.100
59
140.92 145
Promethium
Pm
61
Protactinium
Pa
91
231
Radium
Ra
88
226.05
';/v; Radon
.
85.48
Tantalum
Ta
73
Technetium
Te
43
tellurium
Te
52
127.61
Terbium
Tb
65
58.93
Thallium
Tl
81
204.39
Thorium
Th
90
232.05
Thulium
Tm
69
168.94
Tin
Sn
50
8 70
titanium
Ti
22
47.90
Tungsten
W
74
183.92 238,07
.
mi
3
-
2
-
-
1
.
"
.
. . '
:
.
4/
4d
180.95 5j
98
Id
H-45
4- -H--M-3,
.
Uranium
U
92
Vanadium
V
23
!Xenon
Xe
54
131.3
Yb
70
173.04
-
H-H- -H- 2 0 -
.
jYtterbium
jYttrium
50.95
Y
39
88.92
[Zinc
Zn
30
65.38
'Zirconium
Zr
40
91.22
r Metals, Metals Park, Ohio. also been found.
3
5p
-
...
2
0
-
:
i
Fig. 2-2
H- t
.
5
Schematic representation of possible electron
states in an atom. (From M. M. Eisenstadt, "Introduction to Mechanical Properties of Materials," The Macmillan Company, New York, 1971.)
liii 1
70
INTRODUCT,ON TO PHYS,CAL METALLURGY
[ ;r
to
electron states in an atom. Each horizontal line may be occupied by
.
.
? < <*?
CM
Ul
ft 3
electrons of opposite spin, which are shown in the first few levels as vei
CO
cal arrows. Notice the overlap of energy levels at larger values of n. TP (4s) level is below the (3c0 level, the (5s) level is below the (4c0 level, eti
4. '
This will have an interesting effect on the periodic table which will beejji plainied in the next section. The state of an electron is completely specifi|
V
CD
J
oo O o
>
to
CO
3 <
the same four quantum numbers. The electron configuration of an atom may be indicated by a numbi
.
CD
<
by the four quantum numbers, and no two electrons in an atom may har
f
LO CO
>
CO
<
1
>
2
CO
« D
2 °
CO
Z
representing the principal shell number n followed by the letter reprt* senting the subshell (value of /) and finally the number of electrons in tl
c\j
to O q
>
CO
CNJ
lO CD
containing eight electrons, as (Is)2 (2s)2 (2p)4. I The atomic weight of an element is the weight of the particular atom rel; five to the atomic weight of oxygen, which is taken to be 16.000. On t
- <
<
CO
CO
C
CO
5 = O
scale, the atomic weight of hydrogen is 1.008. The atomic number of an] element is equal to the number of electrons or the number of protons. Thj]
CM
9
r,
,
-
J
CO C\J
f
-
<
a Russian scientist, Mendeleev, of what is called the periodic table periodic classification of the elements. The periodic table is the best clai
n
5
>
C\J
3
sification device for the elements and is one of the most fundament;
concepts in science.
UJ
o
fiL
-
ties, and they tried various groupings. This led to the tabulation in 1869,bj
(D C\J
<
,
Elements of any given family which show a similarity in chemical properj ties were arranged in the same column or group. For Example, Group Vi|; fluorine, chlorine, bromine, and iodine are all similar in properties. elements in each horizontal row or period increase from left to right li atomic weight, with few exceptions. Using all the elements that wen known in 1869 naturally left many blank spaces in the first table. So satis factory was this classification of the elements that Mendeleev was able t(
i
CO -
o
i
atomic number and atomic weight of the elements are given in Table 2'1, 2-3 The Periodic Table Several scientists in the first half of the 1800s recognize that certain elements showed a similarity in physical and chemical proper?
1-
00
three electrons, would be represented as (Is)2 (2s)'; the element oxyge(
CO
CM
subsliell as a superscript. For example, the element lithium, which contaii|
\
o CO
O) .
5 >
-
>
cn c\j
i
-
CD >
co
E
make certain predictions and formulate generalizations that have beei
exactly verified by subsequent research. Each blank space in the tabljj predicted the existence of an element and predicted its approximate atomil
i
1
Cc
Lb
>
'
weight and physical and chemical properties.
In the years that have elapsed since Mendeleev formulated his table, alt] of his missing elements have been discovered, and twelve elements beyom CO
uranium have been prepared in the course of atomic-nuclear research. Thi
. i
complete periodic table is shown in Table 2-2. The modified form of th| periodic table is shown in Table 2-3. The lines running Irom top to botto connect elements of similar chemical properties. In each horizontal re
5
\
i
CM
<
CD R
-
.
f
i
CM Oj CO
5 C\J
UJ
X
<
o
r- X
0
OJ
CO
CM
CD
CM
4
-
CTJ
CO
f
'
7
'
1 i'
horizontal line may be occupied by two
LU in
o
are shown in the first few levels as verti-
f
C\J
CO
CO 0) o
CD »- S
x p CM
energy levels at larger values of n. The
1
3
CO
CO
CD
ne (5s) level is below the (4of) level, etc. on the periodic table which will be exite of an electron is completely specified 1 no two electrons in an atom may have
to
>
m CO
O
< CO
>
Si5?
CD CD
to p
O P CD
i/3 -rr
OX)-
Tt O 0 "Q. w
S
oo <
CD
CM 0)
co co od
cd
55 S
CO
CO -
I
< in
LU
n atom may be indicated by a number umber n followed by the letter repred finally the number of electrons in the
CD
o
< >
a cn
z q
z
m <
o
C\j
S UJ i '
CM
CD
oo
o t m - U
CM
-
O z < >
CD
u P
o
CM
(U CD
CO
"
O CM
pie the element lithium, which contains
.
m oo
CM l
S m CO
(D x
1
CM
,
nted as (Is)2 (2s)1; the element oxygen,
2
!(2s)2 (2pr
10 111
c\j
CO
CO
m - cn
d
-
< m
C\J
CM
CD >
CO
ca "
»i- *
O ai
c\j
.
(DQg
CD
So ? CM
iisthe weight of the particular atom relan,
which is taken to be 16.000.
$
On this
CD
5|S
i: er\ is 1.008. The atomic number of an p&jlectrons or the number of protons. The
aO
00
O C CO CO M
CO
o cnP
"
CO
--
CM
ooxg
(D
m
P
CD
- CO in
CM
3) Ol CO
:
of the elements are given in Table 2-1. ,5 in the first half of the 1800s recognized lilarity in physical and chemical propergs. This led to the tabulation in 1869, by )f what is called the periodic table or
CO CO
E
.
-
0
5
_
o
3
CN
5
LU
>
515
CM LL in
CO Q-i cn z co
Q
_
in
CM
CM
in LU X
3 CO
CD UJ f n in O
CM
OD 0) CO
1
c: as
cm 5
>
CO
* 1
O Ol ai
m
O
"
D
CM
CD Z *
-
S
od CO CM
in
5
5
>
C\J
CD
in
CM
CM
CO
O Ol
LD
S) Q- o
CO
a:
5aS
LU LD
LD
CD
le elements that Mendeleev was able to
>
b
co CM
ai a: .
T
O
- £1
5>
CO
CO
* z C\J
CO
O
-
s CO
ID
mulate generalizations that have been ;earch. Each blank space in the table nt and predicted its approximate atomic properties.
in o 9
as
CM I"
O
»- CM
CM
|
rsi
S H pi co CNJ
CM
CNJ
CM
>
1
-
I 00
(0
ID
I CO CO
:
CD
? nee Mendeleev formulated his table, all
-
discovered, and twelve elements beyond
5
Is
CM
as
ai
co > aj CO
v>
HJ TO O
oo
1 C (/)
-
-
5
CO _
-
i - - cn
ID
<
.
cm E
"
cm
cm O od
blank spaces in the first table. So satis-
;
ID £ CO
m
o S o as
Q
5
>
W or period increase from left to right in jns. Using all the elements that were
vv in Table 2-2.
2mS
ID OS
CO in
LU
ii
5
Sz
J
<
11
jdine are all similar in properties. The
course of atomic-nuclear research. The
o
9 O
LU
.
at; in
oo oo
a
.:
""
HI
and is one of the most fundamental
.
- <
CD
snts. The periodic table is the best clas-
ch show a similarity in chemical properumn or group. For Example, Group VII:
CD CM
o
CM O CO
The modified form of the
J
<
2
CO
CM
CQ p
5
O
CO
CO
CO o
00 CD CO «
cm o d CM
CO
in CQ
CO
LU
3. The lines running from top to bottom
(D
CD CO
.
CO
LU
1 .s
LU
00 CO P
CO
0O cc
LU
CM CM
9 z
o
CO
jcal properties. In each horizontal row
r
CO
-
I
<
CO
'
™ a,
CD
CM CM
> CM
2
-
<
1
*
co DC in
CO
ID
saoid3d
in
w
m O cm
1
»- CO
0O LL CM CM
CO
CO
r
--
LU LU
CO LU
9 5
I z
3 CD
CO
<
i 72
INTRODUCTION TO PHYSICAL METALLURGY
MET
? t«r. - ~
5::
the brackets designated (Is)
,
(2s), (2p) and so on denote the filling of sub-i
Modified Periodic Table*
mBlE 2-3
shells of electrons. The period number refers to values of the principal
H
shell, or the first quantum number n. Table 2-4 shows the placement of th electrons in the principal shell and subshells of the first 26 elements Some understanding of the arrangement of the elements may now be ob.
tained. The subsequent discussion may be followed by referring to Tables]
<
2-2, 2-3, and particularly 2-4. Since the first principal shell can contain th| maximum of only two electrons, the first period contains only two elements! hydrogen, written as (Is)1 and helium, (Is)2. In the next element lithium
3
4
6
Li
He
c
N
0
13
14
lb
16
A
Si
P
S
J
it
,
the third electron enters the second principal shell and the electron coni figuration is written therefore (Is)2 (2s)1. The sum of the superscripts indij ,
.
cates the total number of electrons, which in the case of lithium is 3. Bot|
11
hydrogen and lithium have similar electron configurations that is, one elecl tron in the (s); subshell, and therefore are listed In the same group. The; ,
12
Mg
3S
electrons in the unfilled shells are known as valence electrons and arei
largely responsible for the chemical behavior of the elements. Therefore! the\e|ements in the same group will have the same chemical valence.
In going frojm lithium to neon (atomic number 10)
,
the second shell \i
19
filled to its maximum of eight electrons. In this process the (2s) subsheHj is filled first with beryllium. Then from boron to neon the electrons fill thai
20
21
22
23
Ca
Sc
Ti
V
40
4 1
2A
25
26
27
Mn
he
Co
44
J 4S
(2p) subshell to its maximum of 6 electrons. The outer group of 8 electrons5
when the (s) nd (p) subshells are filled (or 2, in the case of helium), isa| very stable group, and whenever this occurs, the element shows very littl
J
-
.
chemical activity. Helium, neon, and the rest of the elements in the samel
37
38
Hb
3r
group are known as inert gases.
T
Zi
42
43
Mo
Ic
9
70
71
72
Vb
Lu
HI
Rh
5S
In the next element, sodium (atomic number 11), the last electron hasto;! enter the third principal shell.
The electron configuration for sodiunif
therefore, is (Is)2 (2s)2 (2p)6 (3s)1. Notice that the electron configuration; is similar to that of hydrogen and lithium. Therefore sodium is placed inl
69 Ce
the same group. From sodium to argon the process is the same as that|
60
61
Nd
Pm Sm
S2
53
54
Eu
3d
OS lb
HO
Fi
Tm
J
occurring between lithium and neon. The (3s) and (3p) subshells are now!
u
filled, and a very stable grouping is obtained. Therefore argon is an inert gas.
According to the 2n2 rule, the third shell can contain a maximum of 18] electrons, and you would think that the next electron should enter the (3c/)l ;;: :
90
Pa
T
I (
i
\ k
J
Nip
ij
96
97
9o
Am Cm
Bk
CI
95
bs
100
101
102
103 101
Fm
Md
No
Lw
5/
'
,
:>
J
ft
next electron goes into the (4s) subshell, and we have the element potas-
shown schematically in Figs. 2-2 and 2-3.
U
J
subshell. However, the (4s) subshell has less energy, that is, it is closer to| the nucleus than the (3c0 subshell, and therefore it is filled first, so thef sium (atomic.number 19), with an electron configuration similar to that of hydrogen, lithium, and sodium, and it is placed in the same group. This is:
94
91 . 92
From Glenn T. Seaborg and Justin L. Bloom, The Syn
:
After the (4s) subshell is filled with third subshell (3d), expanding the thir
,
-I METAL STRUCTURE AND CRYSTALLIZATION
(2p) and so on denote the filling of subnumber refers to values of the principal
TABLE 2-3
,
Modified Periodic Table* H
WM n- Table 2-4 shows the placement of the d subshells of the first 26 elements.
1
ngement of the elements may now be obn may be followed by referring to Tables
V
HG
_
::
-
73
r
-
,
e the first principal shell can contain the 3 first period contains only two elements: ium, (1s)2. In the next element, lithium, id principal shell, and the electron con(2s)1. The .sum of the superscripts indi-
He
3
0
J
.
SsSp, which in the case of lithium is 3. Both jlectron configurations, that is, one elec-
11
:
fore are listed in the same group. The e known as valence electrons and are
behavior of the elements.
3 .
S
IS
lb
I /
P
s
c
A;
29
E
I
'
;al
12
Ha Mg
Therefore
ill have the same chemical valence.
f;:fsatomic number 10), the second shell is ; :";ftrons In this process the (2s) subshell
19
'
?0 Ca
?5
?3
Sc
Ti
V
Cr
26
27
?e
Fe
Co
V
30
31
32
33
34
35
In
Ga
Ge
As
Ss
3r
36
.
J
Jfrom boron to neon the electrons fill the jlectrons. The outer group of 8 electrons,
4s
i filled (or 2, in the case of helium), is a
IP
r
pis occurs, the element shows very little nd the rest of the elements in the same
37
36
Rb
Si
Y
40
41
42
43
44
45
46
47
46
49
50
51
52
Zr
Nb
Mo
Tc
Ru
Rh
Pd
Ag
Cd
In- Sn
Sb
Te
54
53
Xe
4d
5s
3P
mic number 11), the last electron has to "
he electron configuration for sodium, Notice that the electron configuration lithium. Therefore sodium is placed in
argon the process is the same as that n The (3s) and (3p) subshells are now S obtained. Therefore argon is an inert .
58
J 1
53
yj
La
Pm
Nd
Sm
64
Eu
58
t
lb
Dy
Ho
Ei
71'
Tm
Vb
I
,
u
.
J
J
be
41
HI
T
74
75
7 7
W
Re
ii
a
1
PI
Au
30
31
8
83
64
35
Pb
Si
Po
Al
5d
:
ilrd shell can contain a maximum of 18
t the next electron should enter the (3d)
3ll has less energy, that is, it is closer to
911
91 ,
Ih
Pa
3
93
Np
95 Pi.
96
Am Cm
97
98
Bk
J J .
s
100
Es
7
101
102
103
Em Md
No
Lw
104
105
106
09 110 111
112 113 114 115 116
J
J
5/
6 .
117
I and therefore it is filled first, so the ,
bshell, and we have the element potas-
'
From Glenn T. Seaborg and Justin L. Bloom. The Synthetic Elements. Sc/em/Y/'c/lmerican April 1969. ,
ielectron configuration similar to that of
rid it is placed in the same group. This is Vid 2-3.
After the (4s) subshell is filled with calcium the nexi electrons enter the third subshell (3c/) expanding the third shell from a group of 8 electrons to ,
,
74
INTRODUCTION TO PHYSICAL METALLURGY i
1
is -
TABLE 2 4
- id - As
Electron Conflgur
In the First Four Periods* F
- ip - Is
2p-
ATOMIC NUMBER AND ELEMENT
n
I
2s -
ii
- Zp - 25
15 -
15
Fig. 2-3
Relative electronic energies in two free atoms:
(a) magnesium, (b) iron. (From A. G. Guy, "Elements of Phyfiical Metallurgy," 2d bd, Addison-Wesley Publishing Company, Inc., Reading,'Mass., 1959.)
a maximum of 18.
These elements are known as the transition elements
and are all pf variable valence. After copper (atomic number 29), the electrons fill the (4p) subshell in a normal manner, reaching a stable grouping of eight electrons in the fourth shell with krypton, atomic number 26, an
i
i
inert gas.
.
.
The electron configuration of the elements in the first four
periods is given in Table 2-4. Transition processes also occur in the later periods, and a similar process gives rise to the rare-earth metals, atomic numbers 58 to 71. The periodic table is constantly referred to in the development of new alloys for specific purposes. Since elements in the same group have similar electron configurations, they often may adequately replace each other in alloys. Tungsten is added to tool steels to improve their softening resistance at elevated temperatures, and molybdenum or chromium is sometimes used as a substitute. Sulfur is used to improve the machinability
;-.:- :c
of steel; selenium and tellurium are added to stainless steels for the same purpose. 2-4
-
: .v
j
Isotopes It is possible for the nucleus of an element to have more or less than the normal number of neutrons. Since the number of protons or elec,
trons has npt changed, the atomic number will remain the same. However, the atomic weight will be different. The atoms of varying atomic weight
are called the isotopes of the element. Most of the elements are mixtures of two;pr more naturally occurring isotopes. Therefore the atomic weights
which are treasured are usually not whole numbers. Hydrogen, for example, is a mixture of three isotopes. Most of its atoms contain one proton,
i
f
t
1
I 'I .
1
2 He
2
3 Li
2
4 Be
2
5 B
2
6 C
2
7 N
2
8 O
2
9 F
2
10 Ne
2
11 Na
2
[b)
[a]
1 H
12 Mg
2
13 Al
2
14 SI
2
15 P
2
16 S
2
17 CI
2
18 A
2
19 K
2
20 Ca
2
21 22 23 24 25 26 27 28
2
Set Tit Vt Crf Mnt Fet Cot Nit
2 2
2 2 2 2 2
29 Cu
2
30 Zn
2
31 Ga
2
32 Ge
2
33 As
2
34 Se
2
35 Br
2
36 Kr
2
.
From W Hume-Rothery, "Atomic of Metals, London, 1955. .
t Transition elements.
-
METAL STRUCTURE AND CRYSTALLIZATION 75
TABLE 2-4
Electron Configuration ot the Elements
In the First Four Periods*
m
PRINCIPAL SHELL AND SUBSHELLS ATOMIC NUMBER AND ELEMENT
;
s
s
3
P
1
2 He
2
3 LI
2
1
4 Be
2
2
5 B
2
2
1
6C
2
2
2
7 N
2
2
3
8 O
2
2
4
2
2
5
2
2
6
i of
11 Na
2
2
6
hing
12 Mg
2
6
2
ter copper (atomic number 29), the elec; mal manner, reaching a stable grouping lell with krypton, atomic number 26, an
r t
ration of the elements in the first four
in the later periods, and a similar process
[ atomic numbers 58 to 71. referred to in the development of new elements in the same group have similar j
P
of
s p d f
.
9 F
xv-Jts are known as the transition elements
4
s
10 Ne
1
13 A!
2
2
6
2
1
14 SI
2
2
6
2
2
15 P
2
2
6
2
3
16 S
2
2
6
2
4
17 CI
2
2
6
2
5
18 A
2
2
6
2
6
19 K
2
2
6
2
6
1
20 Ca
2
2
6
2
6
2
21 Set
2
2
6
2
6
1
2
22 Tif 23 Vt 24 Crt
2
2
6
2
6
2
2 2
r
;
2
1 H
ims:
-
n . - 1
25 Mnt 26 Fet
2
2
6
2
6
3
2
2
6
2
6
5
1
2
2
6
2
6
5
2
2
2
6
2
6
6
2
27 Cot 28 Nit
2
2
6
2
6
7
2
2
2
6
2
6
8
2
ed molybdenum or chromium is some-
29 Cu
2
2
6
2
>6
10
1
30 Zn
2
2
6
2
6
10
2
ir is used to improve the machinability
31 Ga
2
2
6
2
,6
10
2 1
e added to stainless steels for the same
32 Ge
2
2
6
2
6
10
2 2
33 As
2
2
6
2
S
10
2 3
34 Se
2
2
6
2
'6
10
2 4
35 Br
2
2
6
2
i6
10
2 5
36 Kr
2
2
6
2
S
10
2 6
H:.ji may adequately replace each other in
.
steels to improve their softening resist-
:
jus of an element to have more or less
is. Since the number of protons or elecnumber will remain the same. However,
*
It. The atoms of varying atomic weight
,
ol Metals, London 1955. t Transition elements. ,
Most of the elements are mixtures
> isotopes. Therefore the atomic weights "
:
:
iiot whole numbers. Hydrogen, for exis. Most of its atoms contain one proton,
J Jl'
i
From W. Hume-Rothery, Atomic Theory for Students of Metallurgy
"
:
1
The Institute
I
76 INTRODUCTION TO PHYSICAL METALLURGY
"'
I
--
I"
but a few contain one proton and one neutron :
,
and still fewer contain one j
proton and two neutrons. The double-weight form of hydrogen is known as
together, there is a transfer o
deuterium, the prime constituent of heavy water. The triple-weight form is|
atoms, resulting in a strong ( sodium ions and the negative i
known as tritium. In ordinary hydrogen these masses of one, two, and three
'
"
-
:i
'
.:>
:
are mixed in such proportions aslo give an average atomic weight of 1.008.|
pound sodium chloride, whic
Nickel has isotopes of atomic weights 58, 60, 61, and 62, producing an average atomic weight of 58.71. Many artificially prepared isotopes, such
compound has its own propi chlorine, demonstrates that tt
as cobalt 60, are radioactive and are used for industrial and medical
fortunate, since sodium is a hi
applications. 2-5 Classification of Elements The chemical elements may be roughly classified into three groups, metals, metalloids, and nonmetals. Elements con-
This explains the strong attra(
sidered to ,
(1) in the solid state they exist in the form of crystals; (2) they have relatively-!
by six negative chlorine ions i ,
in all directions Z8 Covalent Bond
.
i
shell. To attain a stable structui
Generally they have some conductivity but little or no plasticity. Examplesij
than do sodium and chlorine
of metalloids are carbon, boron, and silicon.
three hydrogen atoms and in three hydrogen atoms to form ions are not formed; instead
.
The remaining elements are known as nonmetals. This includes the inert gases, the elements in Group VIIA, and N, 0, P, and S. 2-6
Atom Binding
,
the shared electrons by the poi united to the nitrogen atom b nishing one electron of each pi
Itls characteristic of the solid state that all true solids exhibit*!
a crystal structure which is a definite geometric arrangement of atoms or
molecules. Some materials, such as glass or tar, that are rigid at room tem-
perature do not have a regular arrangement of molecules but rather the.; random- distribution that is typical of the liquid state. These materials are not true solids but rather supercooled liquids. The question now arises as to what holds the atoms or molecules of a
1
J
2
Metallic Bond
The lack of oppos
and the lack of sufficient valen
necessitate the sharing of valen of a negative electron "cloud
jC 1 lonicbond
J
polar bond. The covalent bone 2-9
of the atoms of the metal contri
solid together. There are four possible types of bonds:
-
Atoms of some i
Metalloids resemble metals in some respects and nonmetals In others. '
r
.
ture by sharing one or more e Fig. 2-5, nitrogen (atomic num needs 3 more to complete that
formed plastically; (4) they have relatively high reflectivity of light (metallic j luster). The metals are on the left side of the periodic table and constituted about three-fourths of the elements (see Table 2-2).
-
,
or liquid state. In the solid sta
be metals are distinguished by several characteristic properties: j
high thermal and electrical conductivity; (3) they have the ability to be de-|
f
ous gas; yet table salt is som
" .
particular ion but are free to
Covalent or homopolar bond Metallic bond
L i 4 Van der Waals forces 2-7
Ionic Bond As was pointed out earlier the electron structure of atoms isrelatively stable when the outer shells contain eight electrons (or two in ,
the case of the first shell). An element like sodium with one excess elec-
tron will readily give it up so that it has a completely filled outer shell. It 3 will then have more protons than electrons and become a positive ion i
Nq
(chargefd atom) with a +1 charge. An atom of chlorine, on the other hand, with seven electrons in its outer shell, would like to accept one electron, j When it does, it will have one more electron than protons and become a negative ion with a -1 charge. When sodium and chlorine atoms are placed
Fig. 2-4 Electron transfer in NaCI formation (Frc Van Vlack, "Elements of Materials Science," Addi .
|;-;Publishing
i
c
Company, Inc., Reading, Mass., 1959.)
METAL STRUCTURE AND CRYSTALLIZATION 77
|d one neutron, and still fewer contain one.,|
together, there is a transfer of electrons from the sodium to the chlorine
)uble-weight form of hydrogen is known as
lemical elements may be roughly classi-
atoms, resulting in a strong electrostatic attraction between the positive sodium ions and the negative chlorine ions (Fig. 2 4) and forming the compound sodium chloride, which is ordinary table salt. The fact that this compound has its own properties, not necessarily related to sodium or chlorine, demonstrates that the ionic bond is a very strong bond. This is fortunate, since sodium is a highly reactive metal and chlorine is a poisonous gas; yet table salt is something we use every day without any harm. This explains the strong attraction between paired ions typical of the gas
etalloids, and nonmetals.
or liquid state. In the solid state, however, each sodium ion is surrounded
of heavy water. The triple-weight form is| jrogen these masses of one, two, and three i to give an average atomic weight of 1 008.3 weights 58, 60, 61, and 62, producing an .
Many artificially prepared isotopes, such| rid are used for industrial and medical vj
.
Elements con-
by six negative chlorine ions, and vice versa, so that the attraction is equal
vV.shed by several characteristic properties:
.
in all directions.
he form of crystals; (2) they have relatively jctivity; (3) they have the ability to be deelatively high reflectivity of light (metallic
Covalent Bond Atoms of some elements may attain a stable electron structure by sharing one or more electrons with adjacent atoms. As shown in
Fig. 2-5, nitrogen (atomic number 7) has 5 electrons in the outer shell and.1
t side of the periodic table and constitute'!
needs 3 more to complete that shell. Hydrogen has 1 electron in the outer shell. To attain a stable structure, nitrogen and hydrogen behave differently than do sodium and chlorine. A nitrogen atom shares the electrons of
s (see Table 2-2). some respects and nonmetals in others. :
'
itivity but little or no plasticity. Examples . |
jnd silicon.
'
three hydrogen atoms and in turn shares three of its electrons with the three hydrogen atoms to form the compound ammonia (NH3). In this case ions are not formed; instead, the strong bond is due to the attraction of
|
; »wn as nonmetals. This includes the inert I
and N, O, P, and S.
„
.
the shared electrons by the positive nuclei. The three hydrogen atoms are . united to the nitrogen atom by three pairs of electrons, each atom fur-
the solid state that all true solids exhibit
nite geometric arrangement of atoms or as glass or tar, that are rigid at room temrrangement of molecules but rather the of the liquid state. These materials are
.
nishing one electron of each pair. This is known as the covalent or homopolar bond. The covalent bond is typical of most gas molecules. 2-9
Metallic Bond
oled liquids.
necessitate the sharing of valence electrons by more than two atoms. Each
what holds the atoms or molecules of a :
,
The lack of oppositely charged ions in the metallic structure
and the lack of sufficient valence electrons to form a true covalent bond of the atoms of the metal contributes its valence electrons to the formation
r. isible types of bonds:
of a negative electron
"
cloud.
"
These electrons are not associated with a
particular ion but are free to move among the positive metallic ions in i
lier, the electron structure of atoms is
1
®
hells contain eight electrons (or two in yv.-iment like sodium with one excess elec-
,
;V;h
it has a completely filled outer shell. It electrons and become a positive ion
'
would like to accept one electron. re electron than protons and become a in sodium and chlorine atoms are placed
. hell
ci
No
An atom of chlorine, on the other hand, "
: r.1
Fig. 2-4
Electron transfer in NaCI formation (From L. H. Addison-Wesley Publishing Company, Inc. Reading, Mass., 1959.) .
Van Vlack, "Elements of Materials Science ,
i
f
.
1
NcT
©
,
l
-
®
"
,
CI
"
78
INTRODUCTION TO PHYSICAL METALLURGY
©
H
Total internal energy Repulsion
©
©
a
e-i
Dist
to H N
Fig. 2-5
betv ato
H
NH3
Attraction
Covalent bond in the formation of ammonia.
definite energy levels. The metallic ions are held together by virtue of their
I
Fig. 2-7
mutual attraction for the negative electron cloud This is illustrated sche- i|
,
Internal energy in relation to distance b
atoms.
.
matically in Fig. 2-6. The metallic bond may be thought of as an extension of the covalent bond to a large number of atoms
tive nuclei and also between the
.
'
:
2
-10
dn der Waals Forces This type of bond arises in neutral atoms such
the inert gases. When the atoms are brought close together there is a sep-
the internal energy and the sect tance these two forces will jus
aration of the centers of positive and negative charges, and a weak attractive force results. It is of importance only at low temperatures when the weak attractive force can overcome the thermal agitation of the atoms.
energy £0 will be a minimum, < (Fig. 2-7). The equilibrium dist determined by measuring the
as
the solid state. If the atoms are
librium, then the distance betw
METAL STRUCTURE 2-11
the approximate atomic diamei
Atomic Diameter When atoms of a metal approach each other two opposing forces influence the internal energy, an attractive force between the electrons and both positive nuclei, and a repulsive force between the posi-
number of occupied shells in
,
ft
number of valence electrons in TABLE 2 5
Atomic Diameti
the Periodic Table*
Negative electron cloud ELEMENT
m
m
ATOM I
Lithium
3
Sodium
11
Potassium
19
Rubidium
37
Cesium
55
Beryllium Magnesium
12
4
Calcium
20
Strontium
38
Barium
56
Positive metal ion *
Fig. 2-6 Schematic? illustration of the metallic bond.
;
I'
l
i
From "Metals Handbook," 194S
t One angstrom = 10"* cm.
METAL STRUCTURE AND CRYSTALLIZATION
0
79
Total internal energy Repulsion 0
a
Distance
' .
between
atoms Attraction
p ions are held'together by virtue of their
Fig. 27
electron cloud.
atoms.
This is illustrated sche-
Internal energy in relation to distance between
Dond may be thought of as an extension mber of atoms. .
V-
tive nuclei and also between the electrons. 1 he first force tends to decrease
if bond arises in'neutral atoms such as
>e brought close together there is a sepnd negative charges, and a weak attrac-
: .
Jnce only at low temperatures when the "
e the thermal agitation of the atoms.
the internal energy and the second force tends to increase it. At some distance these two forces will just balance each other and the total internal energy E0 will be a minimum, corresponding to an equilibrium condition (Fig. 2-7). The equilibrium distance r0 is different for each element and is determined by measuring the distance of closest approach of atoms in
i 1 1
the solid state. If the atoms are visualized as spheres just touching at equi1
librium, then the distance between centers of the spheres may be taken as
1
the approximate atomic diameter. The atomic diamelter increases as the metal approach each other, two opposnergy, an attractive force between the and a repulsive force between the posi-
number of occupied shells increases (Table 2-5) and decreases as the number of valence electrons increases (Table 2-6). TABLE 2-5
Atomic Diameter of the Elements in Groups' IA and MA of
the Periodic Table* ELEMENT
ATOMIC NUMBER
ATOMIC DIAMETER,
ANGSTROMSt
GROUP IA
n
Lithium
3
3 03
Sodium
11
3 71
.
Potassium
19
4 62
Rubidium
37
4 87
Cesium
55
5 24
t
i
Beryllium Magnesium
.
.
4
2 22
12
3 19
.
.
Calcium
20
3 93
Strontium
38
4 30
Barium
56
4 34
*
1
.
GROUP IIA
1
.1
.
.
.
.
.
From "Metals Handbook," 1948 ed., American Society for Metals, Metals Park, Ohio.
t One angstrom = 10" cm,
80
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 2-6
Mi
Atomic Diameter and Valence of Some
z
Elements in the Fourth Period*
'
.
;
ATOMIC DIAMETER,
ELEMENT
VALENCE
Potassium
1
4 62
Calcium
2
3 93
ANGSTROMS .
.
Scandium
3
3 20
Titanium
4
2 91
Vanadium
5
2 63
Chromium
Variable
2 49
Manganese
Variable
2 37
.
.
b
.
0/
.
Iron
Variable
2 48
Cobalt
Variable
2 50
Nickel
Variable
2 49
.
x
.
.
Fig. 2-8 . From "Metals Handbook Park, Ohio.
" .
Y
.
Space lattice illustrating lattice parameters
1948 ed.. American Society for Metals, Metals
t
Examples of metals that crystalli; tungsten, alpha (a) iron, delta sodium.
2-12
;1
Crystal Structure
Since atoms tend to assume relatively fixed positions
2-14 Face-centered Cubic
,
this gives rise to the formation of crystals in the solid state. The atoms oscillate about fixed locations and are in dynamic equilibrium rather than J statically fixed. The three-dimensional network of imaginary lines connect-
In addition 1
there is one in the center of each r
ing the atoms is called the space lattice, while the smallest unit having the full symmetry of the crystal is called the unit cell. The specific unit cell for each metal is defined by its parameters (Fig. 2-8), which are the edges of the unit cell a, b, c and the angles a (between b and c), B (between a and Msc), and y (between a and b). There are only 14 possible types of space lattices, and they fall into seven crystal systems listed in Table 2-7. Fortunately, most of the important metals crystallize in either the cubic or hexagonal systems, and only three types of space lattices are commonly encountered: the b.c.c. (body-centered cubic), the f.c.c. (face-centered cubic), and the c.p.h. (close-packed hexagonal). Unit cells of these are shown schematically in Figs. 2-9 to 2-11. In each case the atom is represented as a; point (left) and more accurately as a sphere (right). 2-13 Body-centered Cubic If the atoms are represented as spheres the center:
Each face atom touches its neare;
TABLE 27
The Crystal Systems*
In this table
means "not necessarily equal '
1 Triclinic .
I
2
.
Monoclinic I
3
.
4
.
5
.
Orthorhomblo
Rhombohedral (trigonal) Hexagonal
,
f
atom touches each corner atom but these do not touch each other. Since
each corner atom is shared by eight adjoining cubes and the atom in the
ili
center cannot be shared by any other cube (see Fig. 2-12a), the unit cell
i
1 center atom
i
Total i
.
.
Tetragonal
7
Cubic
.
8 atoms at the corners x Va:
v
6
of the b.c.c. structure contains:
1 -
.s
1 atom
.
1 atom 2 atoms
*
From C S. BarrettN'Structure of Metals,'
11
r<","";: ;iV: ;
-
I Valence of Some
ATOMIC DIAMETER
,
r
i METAL STRUCTURE AND CRYSTALLIZATION
i
81
f
ANGSTROMS
i
4 62 .
3 93 .
j !
3.20 2.91
m
2 63 .
m
2 49 .
2 37 .
'
v\'.: | '
-
. .
-
2 48 .
j
2.50 2 49 .
Pig. Z'8 '
nerican Society for Metals, Metals
Space lattice illustrating lattice parameters,
4
Examples of metals that crystallize in the b.c.c. structure are chromium, tungsten, alpha (a) iron, delta (5) iron, molybdenum, Vanadium, and
SMt
sodium.
assume relatively fixed positions,
2-14 Face-cehtered Cubic
iv itals in the solid state. The atoms
fc
n dynamic equilibrium rather than Imms.
In addition to an atom at each corner of the cube,
there is one in the center of each face, but none in the center of the cube. Each face atom touches its nearest corner atom. Since each corner atom
letwork of imaginary lines connect- Jpltf while the smallest unit having the le unit cell. The specific unit cell
1
,
ers (Fig. 2-8), which are the edges
.
'
S
TABLE 2 7
The Crystal Systems*
In this table
means "not necessarily equal to"
1
between b and c), p (between a and
.
Triclinic
Three unequal axes, no two of which are perpendicular
a
ace lattices, and they fall into seven '
2
.
Monoclinic
V etals crystallize in either the cubic oes of space
.
d cubic), the f.c.c. (face-centered
|
sxagonal). Unit cells of these are
|
In each case the atom is reprejtely as a sphere (right).
3
.
Orthorhombic
4
,
5
.
. .
. .
y. v
./j
b b
.
c c
Tetragonal
7
j atom = 1 atom Total = 2 atoms
.
c c
Cubic
~ '= a
o = P -
,
on" y = 90 '
a = /3 = 90°
y = 120°
a = /i = 7 = 90o
u
Three equal axes, mutually perpendicular
a = b=c *
a = 7 = 90o#/3
Three perpendicular axes, only two equal a = b
p x V8 = 1 atom
90°
Three equal coplanar axes at 120° and a fourth unequal axis perpendicular to their plane
.! '
,
Hexagonal
a=b 6
y
Three equal axes, not at right angles a-b=c a --- p = y 90°
ise do not touch each other. Since
iljoining cubes and the atom in the pube (see Fig. 2-12a), the unit ceil
p
Rhombohedral (trigonal)
.
epresented as spheres, the center
a
Three unequal axes, all perpendicular a
1
c
Three unequal axes, one of which Is perpendicular to the other two ,
a
lattices are commonly
b
ct = /3 = 7
I
1
1
1
90o
From C, S. Barrettfs"Struclure of Metals." McGraw-Hill Book Company, Inc., New Vork. 1952.
T v
i
,
1
82
INTRODUCTION TO PHYSICAL METALLURGY
Mi
--
3l
Lefl
Fig. 2-9
yy.-
Unit cell of the b.c.c. structure represented by
RB- S'll
points (left) and as spheres (right).
120°
Lefl
The c.p.h. structure-as points, with the unit
Still shown in heavy lines (left), and as spheres (right)
.
is shared by eight adjoining cubes and each face atom is shared by only|
compare it to.that of a b.c.c cell. Sin
one adjacent cube (see Fig. 2-t2b)t the unit cell contsfrns: -
.
and each atom is a sphere of radius r
t;
i
8 atoms at the corners x Vs
1 atom
6 face-centered atoms x V2
3 atoms
Total
4 atoms
and
This indicates that the f.c.c. structure is more densely packed than the| c c structure. Another way to show the difference In packing is to calcu| late the fraction of the volume of a f.c.c. cell that is occupied by atoms andl
b
.
.
,
.
9%
where a is the lattice parameter. It is n in terms of ra. Consider a cube face a
9
0
S
(I o
7
O
i it
IS
-
55
a\/2 = 4r
mm
I
4
i
V
Packing factor fi
mm Left
Fig, 2-10 Unit cell of the f.c.c. structure represented liy points (left) and as spheres (right).
J i
7:
toms
ce„
It is left as an exercise for the student the b.c.c. structure turns out to be ttV
Examples of metals that crystallize nickel, copper, gold, silver, lead, platin
1
METAL STRUCTURE AND CRYSTALLIZATION
83
'
I
5
i
6
3s
1
-.
5
;
H20O
I eM .
by
S
Fig 2-11
The c p h structure-as points, with the unit .
.
.
I cell shown in heavy lines (left), and as spheres (right). and each face atom is shared by only -
'
..
V
--
.-
%
), the unit cell contdfns:
. .
.
7
1
,
.
compare it to. that of a b.c.c. cell. Since there are four atoms per unit cell and each atom is a sphere of radius ra, then
orners x Vb = 1 atom 16
atoms x 72 = 3 atoms
,0
3
Total = 4 atoms
and
jcture is more densely packed than the x :: ow the difference in packing is to calcu-
oil
-
-
f cc .
.
.
cell that is occupied by atoms and i
1
a3
where a is the lattice parameter. It is now necessary to find the cell volume in terms of ra. Consider a cube face as shown,
is
o
J2
S 4/i
r
I
::::
a
\'2
1 i
:
4
2\ 2r„
or a
16 3
ft
i5
Packing factor == -
n
77
(aN- z-J3
-- = 0,74
3\/2
ft
'
it;-'
It is left as an exercise for the student to show that the packing factor for .
the b.c.c. structure turns out to be it V3/8 or 0.68. ,
by i
Examples of metals that crystallize in the f.c.c. lattice are aluminum nickel, copper, gold, silver lead, platinum, and gamma (y) iron.
,
,
'
84
I
INTRODUCTION TO PHYSICAL METALLURGY
°
0
II
i
r
"
""*
7
[o)
i
'
Fig. 2-12 (a) Unit cell of the b.c.c. structure; (b) unit cell of the f.c.c. structure. (From "Basic Metallurgy," American Society for Metals, Metals Park, Ohio, 1967.)
2-15
Close-packed Hexagonal The usual picture of the close-packed hexagonal lattice shows two basal planes in the form of regular hexagons with an atom1
at each corner of the hexagon and one atom at the center. In addi-
tion, there ar& three atoms in the form of a triangle midway between the two basal planes. If the basal plane is divided into six equilateral triangles the additional three atoms are nestled in the center of alternate equilateral ) triangles (Fig. 2-11). The parallel repetition of this hexagonal prism will not build up the entire, lattice. The true unit cell of the hexagonal lattice is in fact only the portion ,
;
-
i
shown by heavy lines in Fig. 2-11. The hexagonal prism, therefore, contains two whole unit cells and two halves.
It may not be readily apparent from the unit cell why the structure is| called hexagonal. It remains to show that, if a number of these unit cellsj Fig. 2-13
are packed together with axes parallel to one another, as in a space lattice, a hexagonal prism may be carved out of them. Figure 2-13 shows many unit cells, with the open circles representing
axial ratio of a c.p.h. structure forn reality, metals of this structure have i
atoms in the plane of the paper and the filled circles representing atomsj
halfway above and below. The hexagonal lattice derived from the unit cellsj
lium to 1.88 for cadmium.
is shown by means of either the filled or the open circles. In each case, there are seven atoms representing the basal plane and three atoms in thea form of a triangle in the center of alternate equilateral triangles.
,
Since each atom at the corner of the unit cell is shared by eight adjoining]
Polymorphism and Allotropy
'
"
TTie
"
'
umrcj3l l of the cubic system may be specified by a single lattice;
.
*
c/a, which is sometimes given. It may be shown mathematically that the]
1
1 lib
:
i
" "
.
v v:-:. .
.
i
Polym
to exist in more than one type of i change in structure is reversible, the allotropy. At least fifteen metals sh
,|
parameter a,;but the hexagonal cell requires the width of the hexagon and the distance between basal planes c. These determine the axial ratio!
Therefo
be in contact, they must be spheroi The types of crystal structure I: property data of the common metal:
cells and one atom inside the cell cannot be shared, the c.p.h. unit celt: contains two atoms. Examples of metals that crystaNize in this type ofi
structure are magnesium, beryllium, zinc, cadmium, andjaafmum.
Derivation of the hexagonal lattice from mar
unit cells.
example is iron. When iron crystalli: the structure changes to f.c.c. (y Fe) (a Fe).
METAL STRUCTURE AND CRYSTALLIZATION '85
r
5
4
[6) cell of an '
: r
-
-
i
t
-
ill picture of the close-packed hexagonal the form of regular hexagons with an jn and one atonvat the center. In addiinform of a triangle midway between the > is divided into six equilateral triangles led in the center of alternate equilateral ,
'
jagonal prism will not build up the entire i agonal .
lattice is in fact only the portion
! The hexagonal prism therefore, con,
halves. :
rom the unit cell why the structure is ow that, if a number of these unit cells ,
lei to one another, as in a space lattice,
Fig. 2-13
Derivation of the hexagonal lattice from many
unit cells.
j)ut of them. lis with the open circles representing ,
axial ratio of a c.p.h. structure formed of spheres in contact is 1 633. .
reality, metals of this structure have axial ratios that vary from 1.58 for beryl-
igonal lattice derived from the unit cells
lium to 1.88 for cadmium.
led or the open circles. In each case, the basal plane and three atoms in the Iternate equilateral triangles.
he unit cell is shared by eight adjoining cannot be shared, the c.p.h. unit cell
Therefore if the atoms are still considered to ,
be in contact, they must be spheroidal in shape rather than spherical The types of crystal structure lattice parameters, and other physical property data of the common metals are given in Table 2-8. 2-16 Polymorphism and Allotropy Polymorphism is the property of a material to exist in more than one type of space lattice in the solid state If the change in structure is reversible then the polymorphic change is known as allotropy. At least fifteen metals show this property and the best-known example is iron. When iron crystallizes at 2800oF it is b c c (S Fe), a\ 25540F the structure changes to f.c.c. (y Fe) and at 1670oF it again becomes b.c.c. (a Fe). .
,
.
metajsJJiat crystalljzejn this type of
zinc, cadmium, and baimum. i may be specified by a single lattice I requires the width of the hexagon a _
,
,
.
,
nes c.
These determine the axial ratio
nay be shown mathematically that the
-
>
In
d the filled circles representing atoms
.
.
r
i '
ft
4
INTRODUCTION TO PHYSICAL METALLURGY
'
x
217
o <
I
Crystallographic Planes
H
o
,
0
:
I- S
-
.
>
*
+-
-
indices. One corner of the unit c
< w 2
2 0
coordinates, and any set of pla
UJ V I-
intersections with these coordir
I-
W q EC
,
w 5 w -
CVJ ID O CO CT)
o o <
Cvi CM CM
o < o j ll 2:
tCM CM
Tti-
oo
CM
cotfjCMCMOtf)
iDi-in-i-eoco
CM
o in oo oooo>
cm o
t-
CM
O
CO CM
in to t-; CM CM CM CO CO
r cm
in co i-
cm rn. co co o CM CM CVJ CM CM CM CO
'
lattice parameter of the crystal, I it at infinity. in Fig. 2-14, or the cubic syst
co co co h- CO CD '
CM
CM CM CM
in
co
o>
too "
o co in co
co
cc «
sects the V axis at one unit fron
m 2 I-
o m uj cc O 2 Ic < w
F EC o < < z _
) £L <
in
OO CM in Tf oo oo
co i-
en
I itco o
co
cm in
co oo
in o>
o
o
Ttcn
2,inr>lco,J'tncococ0c,'l
S S cBo ii ii S
-
t
co co o
i
S co
o
axes or intersects them at infinit
cod> -
a>
c>,tom,-00fooo5CM r
§ 5 S 5 i i 5 ? S 8 5? S ii !? 8 ii ii Qsocaocvico cvi 'caooococococo 'caocaoeoeocBu
Intersection CD
2
CO
o
I
Reciprocal
i
Miller indices
_
a
cc _
.
'
0)
I : I
J 3
< K i
'
V -V-
I- O W 3 >- DC
Cv;."
CC hO W (3 z
LL
J
o
n
d R-
sJCCo
O I-"
CM .
m o.
n
,o
i .=
.f
;= « OCo
,
o 05
CO O iCM in o cm m
.
I X
o
O co
t-
co
.
E
|
o
oi :
.
-
i|l o
it
a)
co co
o cd cm
d o>
CM cm
CM O in co
T-
E
CO CO i- co ,'
T- CM
x
Y
oc
1
1
1
0C
1
0
1
Z
1
o
«.
6 6 v. 6*.
0. ddOdqo
ja js-i
c c
3 « d d -g 6 s>o a
6
o
O n
o
in
o 2 o in o o o
oj o oo
co co
CM
o 2 in oo co i- cm
oo
t- O in co CM
o
co o Is*
co in
O
- O) t-
00 O
co o t co
tj-
coor
t-
j-
o no a 6
The illustrated plane has Millet on the negative side of the origin, by placing a minus sign above th indices of the plane ADEF which be determined without changing cube may be selected as the ori'
p s s
o> m
T-
CD
o
co
i-
o
o
» ~
t-'
co
IT) f-
CM
odoi
m
O CO Tt 05 05
05
CM O CO CM
r-
co tj r- cd cMNcocMinr-Tt
o
i-
CD
CO
CM
i-
CM
CO
co
co
o
o
co
co
r05 CD
cm CM
in CM
o
d d d
o
T-
t-
t-
0
««: =
5ti6didi:
(3 z
(0
The lay
atoms are arranged are known < relation of a set of planes to the ;
t -
- CM
CO
CM
t-
co
in
co
I
o
E E
3 O
o
o o
E
s
E cr>
ra
.
in
oi O
n
Q >.
O co
I 05
co co CM O
odd
CO
co t-
5
co
co
o
co cm o> co O CO CM CO CO CM *0
d d
d
d d d d o d
oo
0)CM
1
OOCMin -OlCO f COCMf CMCOCOt
OOI
CO
OCOCM
d d d d d d d
J
(112):
t Q o
.
O m -
Ml
o
>
-
M
< co m
o
6 o
< u. q. S
£ i
.
cp CM
J
I1 CL -
<
= e
p
< < CQ ' m U
O
£
I
I i z a: 55
A
< co
P
$1
§ >
N
a>
3
X
.w
U) T3 Ui 03
I
Fig. 2-14
LU
2
D
c E
-
m
OOC3i:_j2
Y
S5ZQ.C0C0I-
I-
l->N
Determination of Miller indices: the (01C
Is-- the (111) plane,and the (112) plane.
'
I
!
,
METAL STRUCTURE AND CRYSTALLIZATION ,
87
2-17 Crystallographic
Planes The layers of atoms or the planes along which atoms are arranged are known as atomic or crystallographic planes. The
I
relation of a set of planes to the axes of the unit cell is designated by Miller
m
indices. One corner of the unit cell is assumed to be the origin of the space coordinates, and any set of planes is identified by the reciprocals of its intersections with these coordinates.
The unit of the coordinates is the
CO
.*
CM
-*
CO
CO
CD
CO
CD
CM
CM
CT>
CO
cO cJ oj csi
CO
CM
Cxi
CO
CM
lattice parameter of the crystal. If a plane is parallel to an axis it intersects it at infinity. In Fig. 2-14, or the cubic system, the crosshatched plane BCHG intersects the V axis at one unit from the origin and is parallel to the X and Z axes or intersects them at infinity. Therefore,
.
,
1 co o
NT
v:-
o o
.IS i it
O
co CO
CO
oo
03
o
cn
o
o a
t co
t--
o m
co cn
co i"
co co-r-cocoincoiNTitt c\j
co
O)
t-
m
05
CO
CO
CO
«
c\j
co
CO 05
II
o .
O
13
CM
S II I! CO
CO
TO
Intersection
E
4
-
CM
-
O
CO
CNJ
T
S ddid 2 =9dd!2d9
o 2 o in o O S 00
o> H. o> t- co co o
ti-
co
o o
i-
= q-
o6£ 9 « 9-
o
S o co
a> m
- io co
O
5
Miller indices
»
-'
co
1
0&
1
X
0
1
0
indices of the plane ADEF which goes through the origin (poinM) cannot \
co i-
be determined without changing the location of the origin. Any point in the cube may be selected as the origin. For convenience, take point B. The
Z
S CM
X
1
by placing a minus sign above the index, as (hkl). for example, the Miller I
P ID CO
i
c S
1
1
The illustrated plane has Miller indices of (010). If a plane cuts any axis on the negative side of the origin, the index will be negative and is indicated
s
:
X
.
cn
o
6 9-
z
Reciprocal
CL
6
y
X
o
co o i-- i - o d oi r~ co tj- t
i~CM
co t i-i-
in
o
o
S
LO
CO
CO
oo 1
5
s 8 6
-
/
,
5
oo>cjinTi-cnco h-cocMt coi--co CMCOCOI OCOCM
zi o
o o d o d o d
t co
l
~
-
05 CO
o
O
v
in CM
CM
c
N
si i
e
r
si ;3
-
D
.
c
2 S Z D. W C/5 K
E 2 TO
mo X
s W
"D
d co .E
\f Fig
.
I- > N
2-14
Delermination of Miller indices: the (010) plane,
the (111) plane and the (112) plane .
if Sis
-
C
CO
il-ic
giE 8 5 to o o « - .5
1010
CO
d d d
S £ - O) c 2 5 z d. w < w .
£
5;
00 3>cm oco .o
.
Y
'
t ;
..
88
,NTRODUCT,ON TO PHYS,CAL METALLURGY
\
plane ADEF is parallel to the X axis (BC) and the Z axis (SG) but intersects the / axis at -1. The plane has Miller indices of (010) As another illustration, the Miller indices of the plane BDJ (Fig 2-14) .
.
may be determined as follows; x
Intersection
Y
Z
1
1
1
1
1
1
1
V2
1
2
,
O
Reciprocal ,
Miller indices
1
0
(1201 This plane has Miller indices of (112). If the Miller indices of a plane result in fractions, these fractions must be cleared. For example, consider a plane that intersects the X axis at 1, the V axis at 3, and the Z axis at 1. Taking reciprocals gives indices of 1, 'A and 1. Multiplying through by 3 to clear fractions results in Miller indices'
of (313) for the plane.
"
10
Fig. 2-15
|
BCC
Projection of the lattice on a plane perpe
lo the Z axis to illustrate interplanar spacing. Filled are In the plane of the paper.
All parallel planes have the same indices. Parentheses (hW) around Miller ]
x '
(M01
indices signify a specific plane or set of parallel planes. Braces signify a| family of planes of the same "form" (which are equivalent in the crystal),-; such as the cube faces of a cubic crystal: {100} = (100) + (010) + (001)
An infinite number of planes me but most are just geometrical cc portance. Remembering that eac
(Too) + (oro) + (ooT). Reciprocals are not used to determine the indices of a direction In order!
account for all the atoms, the mo;
.
to arrive at a point on a given direction, consider that starting at the orjgin it|
atomic population and largest inte these are the {110} planes, and ii planes (see Figs. 2-15 and 2-16). 2-18 X-ray Diffraction One may wondei
is necessary to move a distance u times the unit distance a along the X axis 1; ,
v times the unit distance b along the V axis, and w times the unit distanod c along the Z axis. If u, v, and w are the smallest integers to accomplish!
the desired motion, they are the indices of the direction and are enclosed]
sure lattice dimensions since atom
in square brackets [uvw]. A group of similar directions are enclosed in atv
angstrom units. The most useful ti
gular brackets (uvw). For example, in Fig. 2-14, to determine the direction AC, starting at the origin (point A), it is necessary to move one i along the X axis to point O and one unit in the direction of the Yaxis to read point C. The direction AC would have indices of [110]. In a cubic crystal, direction has the same indices as the plane to which it is perpendicular.
diffraction, and a brief introductio 1»
Since x-rays have a wavelength
An approximate idea of the packing of atoms on a particular plane ma; be obtained by visualizing a single unit ce|l of the b.c.c. and f.c.c. structure;! Considering the atoms as the lattice points, the number of atoms on a p ticular plane would be:
y ; .
.
I
PLANE
,
b.c.c.
f
.
c c .
.
(100)
u
(110)
5
6
(111) (120) (221)
3
6
[a] BCC
2
3
|Fig. 2-16 Interplanar spacing of (111) planes in (a) t
1
1
and (b) f.c.c. structures. (By permission from L. H. Ve
5
Elements of Materials Science," Addison-Wesley Pl
''
Company, Inc., Reading, Mass., 1959.)
'
r
{
is
METAL STRUCTURE AND CRYSTALLIZATION
89
.
: .
:
x
.
x
O
o
o
o
o
"
(220)
o
o "(110)
o
o
o a
"
. -
:
.
V::l 12).
o
o
o
o
1120) o
(IIO)
'-
-
o
o
Bsult in fractions, these fractions must be
"
j plane that intersects the X axis at 1, the Taking reciprocals gives indices of 1 V3,
.
-
,
.
'
r
;
{b) F.C.C.
[o] B.C.C.
,
o clear fractions results in Miller indices
Fig. 2-15 F'rojectlon of the lattice on a plane perpendicular to the Z axis to illustrate interplanar spacing. Filled circles arg in the plane of the paper.
3 Indices. Parentheses (hkl) around Miller r set of parallel planes. Braces signify a ;m" (which are equivalent In the crystal),
An Infinite number of planes may be taken through the crystal structure, but most are just geometrical constructions and have no practical im-
||jc crystal: {100} = (100) + (010) + (001) + |
portance. Remembering that each complete set of parallel planes must account for all the atoms, the most Important planes are the ones of high atomic population and largest interplanar distance. In the b.c.c. structure
ermine the indices of a direction. In order
ption, consider that starting at the origin It | Itlmes the unit distance a along the X axis, the V axis, and w times the unft distance
are the smallest integers to accomplish
2-18
ndlces of the direction and are enclosed of similar directions are enclosed in an-
le, in Fig. 2-14, to determine the direc)int A), It is necessary to move one unit
these are the {110} planes, and in the f.c.c. structure these are the {111} planes (see Figs. 2.15 and 2-16). X ray Diffraction One may wonder at this point now it is possible to measure lattice dimensions since atomic spacings are In the order of only a few angstrom units. The most useful tool for studying crystal structure is x-ray -
diffraction, and a brief introduction will be given here. Since x-rays have a wavelength about equal to the distance separating
unit in the direction of the /axis to reach
lave indices of [110]. In a cubic crystal a % ?, ,
the plane to which it is perpendicular.
z
I
king of atoms on a particular plane may ,
:
unit cell of the b.c.c. and f.c.c. structure.
e points, the number of atoms on a pari
K
FCC
[a] BCC
.
Fig. 2 16
.ii ii-i
.
-
Interplanar spacing of (111) planes in (a) b c c .
.
.
a
and (b) l.c c structures. (By permission Irom L. H. Van Vlack.
;
Company. Inc Reading, Mass., 1959.)
.
"
Elements of Materials Science .
,
" .
Addison-Wesley Publishing
;
90
INTRODUCTION TO PHYSICAL METALLURGY
the atoms ih solids
,
At some lower temperature, tf creased so that the attractive forc(
when x-rays are directed at a crystalline material they]
will be diffracted by the planes of atoms in the crystal If a beam of x-ray| strikes a set of crystal planes at some arbitrary angle there will usually|
\
.
the atoms together in a liquid. No main in the vapor above the liquic atoms between the vapor and liqui
,
be no reflected beam because the rays reflected from the crystal planes must travel different lengths and will tend to be "out of phase" of to cancel] each other put. However at a particular angle, known as the Bragg angled .
vessel, at a definite temperature, t
,
librium and there will be a const
the reflected rays will be in phase because the distance traveled will be an integral number of wavelengths Consider the parallel planes of atomsl
6
,
above the liquid.
.
The waves may be reflected!
in Fig. 2-17,from which a wave is diffracted
.
from an atom at H or H' and remain in phase at K; however the reflectedl rays from atoms in subsurface planes such as H", must also be in phase ,
,
at K. In order for this to occur
,
the distance MH"P must equal one or moref
k>
integral wavelengths. If d is the spacing between planes and e is the angle is equal to 2d sin 0 or n\ = 2d sin e
>
2,3, etc. This is known as the Bragg equation. Without going into the details of equipment for a given wavelength \, B ,
vibrating around fixed points, giv crystal structures discussed previo
,
can be measured and d calculated.
2 20 Mechanism of Crystallization Crys
The States of Matter Three states of matter are distinguishable: gas liquid, and solid. In the gaseous state the metal atoms occupy a great deal of space because of their rapid motion. Their motion is entirely random and ,
liquid to the solid state and occurs
,
.
:
The combination of all the collisions with the wail is the pressure of the gas on the wall. The atoms move independently and are usually widely sep- i arated so that the attractive forces between atoms are negligible. The arrangement of atoms in a gas is one of complete disorder.
.
it, resulting in evaporation. Evider a liquid may be demonstrated by be easily compressed into a smalle compress a liquid. There is, howe to allow the atoms to move about As the temperature is decreasec attractive forces pull the atoms cl Most materials contract upon solic atoms in the solid state. The atom
where n can have values of 1
as they travel they collide with each other and the walls of the container
.
V
of incidence, the distance MH" is equal to cf sin e and the distance MH"P
2-1P
If the vapor is
reached and more atoms will leavi
1
Nuclei formation
2
Crystal growth
Although the atoms in the liquid i ment, it is possible that some atot J
'
.
exactly corresponding to the spa (Fig. 2-18). These chance aggregat tinually break up and reform at ot! mined by the temperature and size ture, the greater the kinetic enerc of the group. Small groups are ver a small number of atoms and the
group. When the temperature of XK
ment decreases, lengthening the ii
present at the same time. Atoms in a material have botl-
H
energy is related to the speed at (hkl) planes <
1
d(hkl)
;:
-
M
Fig. 2-17
"
H
X-ray reflection. (From L. H. Van Vlack, "Elements of Materials Science, Addison-Wesley Publishing Company, "
Inc., Reading, Mass., 1959.) <
j
!
function of temperature. The higl the atoms and the greater is their other hand, is related to the distar
age distance between atoms, the Now consider a pure metal at
METAL STRUCTURE AND CRYSTALLIZATION
91 >
j
re directed at a crystalline material they | %3atoms in the crystal. If a beam of x-rays
5
v some arbitrary angle there will usuallyM
.
,
3 rays reflected from the crystal planes
II tend to be "out of phase" or to cancel f icular angle, known as the Bragg angle
.
e because the distance traveled will be
Consider the parallel planes of atoms diffracted. The waves may be reflected n in phase at K; however the reflected ,
gjies, such as H", must also be in phase jjistance MH"P must equal one or more icing between planes and d is the angle '
;qual to d sin 0 and the distance MH"P
.
As the temperature Is decreased, the motions are less vigorous and the attractive forces pull the atoms closer together until the liquid solidifies. Most materials contract upon solidification, indicating a closer packing of
2d sin 6
y
.
atoms between the vapor and liquid across the liquid surface In a confined vessel, at a definite temperature the interchange of atoms will reach equilibrium and there will be a constant value of vapor pressure of the gas above the liquid. If the vapor is free to escape, equilibrium will not be reached and more atoms will leave the liquid surface than are captured by it, resulting in evaporation. Evidence of attractive forces between atoms in a liquid may be demonstrated by the application of pressure. A gas may be easily compressed into a smaller volume, but it takes a high pressure to compress a liquid. There is, however, still enough free space ih the liquid to allow the atoms to move about irregularly. ,
,
.
,
At some lower temperature, the kinetic energy of tihe atoms has decreased so that the attractive forces become large enough to bring most of the atoms together in a liquid. Not all the atoms are in the liquid Atoms remain in the vapor above the liquid, and there is a continual interchange of
This is known as the Bragg equation.
'
quipment, for a given wavelength \, e
1 matter are distinguishable: gas, liquid, 9 metal atoms occupy a great deal of i Their motion is entirely random, and .
i other and the walls of the container
.
with the wall is the pressure of the gas )endently and are usually widely sepetween atoms are negligible. The arDf complete disorder.
atoms in the solid state. The atoms in the solid are not stationary but are i
vibrating around fixed points, giving rise to the orderly arrangement of crystal structures discussed previously.
.
/-2
20 Mechanism
of Crystallization
Crystallization is the transition from the
liquid to the solid state and occurs in two stages; 1
Nuclei formation
2
Crystal growth
Although the atoms in the liquid state do not have any definite arrangement, it is possible that some atoms at any given instant are in positions exactly corresponding to the space lattice they assume when solidified (Fig. 2-18). These chance aggregates or groups are not permanent but continually break up and reform at other points. How long they last is determined by the temperature and size of the group. The higher the temperature, the greater the kinetic energy of the atoms and the shorter the life of the group. Small groups are very unstable since they: are formed of only a small number of atoms and the loss of only one atom may destroy the group. When the temperature of the liquid is decreased, the atom movement decreases, lengthening the life of the group, and more groups will be present at the same time. Atoms in a material have both kinetic and potential energy. Kinetic energy is related to the speed at which the atoms move and is strictly a function of temperature. The higher the temperature, the more a6tive are the atoms and the greater is their kinetic energy. Potential energy, on the other hand, is related to the distance between atoms. Th!e greater the aver'
K
f d(hkl) P
ems
any,
age distance between atoms, the greater is their potential energy. Now consider a pure metal at its freezing point where both the liquid !
3 V
p
:
92 INTRODUCTION TO PHYSICAL METALLURGY 1
A
C
' ': .
D
B o
Undercooling -
3®
x
temperotui
,
"
Melting or freezing
c
A
I
E
-
0 Time
Fig. 2-19 Cooling curve for a pure metal; ABOE ABODE actual.
three dimensions, the atoms e
directions, ysually along the axe Fig. 2-18 ;:
.
:
Schematic diagram of structures of (a) crystal and
5
(/>) liquid, krea ABODE in liquid is identical in arrangement "
Physical Metallurgy. Wiley & Sons, Inc., New York, 1959.)
as in crystal. (From Chalmers,
nucleus is formed by chance, tl the dendrites growing from the crystal. Finally, as the amount
John
and solid states are at the same temperature. The kinetic energy of the atoms in th6 liquid and the solid must be the same, but there is a significant difference in potential energy. The atoms in the solid are much closer together, so that solidification occurs with a release of energy. This differ-j ence in potential energy between the liquid and solid states is known as the latent heat of fusion. However, energy is required to establish a surface between the solid and liquid. In pure materials, at the freezing point, insufficient energy is released by the heat of fusion to create a stable bound- J ary, and some undercooling is always necessary to form stable nuclei. Subsequent release of the heat of fusion will raise the temperature to the ' freezing point (Fig. 2-19). The amount of undercooling required may be reduced by the presence of solid impurities which reduce the amount of .
>. . .
:
,
.
teristic treelike structure which
arms of the dendrite will be fill
surface energy required.
When the temperature of the liquid metal has dropped sufficiently below its freezing point, stable aggregates or nuclei appear spontaneously at various points in the liquid. These nuclei, vyhjch have now solidified act ,
as centers for further crystallization. As cooling continues, more atoms tend to freeze, and they may attach themselves to already existing nuclei or form new nuclei of their own. Each nucleus grows by the attraction of atoms from the liquid into its space lattice. Crystal growth continues in
i I
m r
Fig. 2-20 Magnesium dendrites growing from liq
mmmmmstaass
METAL STRUCTURE A D CRYSTALLIZATION 93
A
B o
Undercooling
1
-
;
Melting or freezing temperature
c
E
m Time
Fig. 2-19
»-
Cooling curve for a pure metal;
ideal,
ABODE actual.
three dimensions, the atoms attaching themselves in certain preferred
directions, usually along the axes of the crystal. This gives rise to a characteristic treelike structure which is called a dendrite (Fig. 2-20). Since each nucleus is formed by chance, the cyrstal axes are pointed at random and the dendrites growing from them will grow in different directions in each
ystal and lement '
t
-
John
i
crystal. Finally, as the amount of liquid decreases, the gaps between the arms of the dendrite Will be filled and the growth of the dendrite will be
:emperature.
The kinetic energy of the ust be the same, but there is a significant he atoms in the solid are much closer
jrs with a release of energy. This differie liquid and solid states is known as the
4
#
nergy is required to establish a surface ;| )ure materials, at the freezing point, inheat of fusion to create a stable bound-
ways necessary to form stable nuclei. fusion will raise the temperature to the
I w -n
iount of undercooling required may be mpurities which reduce the amount of
id metal has dropped sufficiently below es or nuclei appear spontaneously at
;;
;
vvr? nuclei which have now solidified, act ,
)n. As cooling continues, more atoms h themselves to already existing nuclei ach nucleus grows by the attraction of ;e lattice. Crystal growth continues in
* Fig. 2-20
Magnesium
dendrites growing from liquid.
;
94
INTRODUCTION TO PHYSICAL METALLURGY
mutually obstructed by that of its neighbors. This leads to a very irregular external shape. The crystals found in all commercial metals are commonly called grains because of this variation in external shape. The area along
which crystafs meet, known as the grain boundary, is a region of mismatch (Fig. 2-21). This leads to a noncrystalline (amorphous) structure at the grain boundary with the atoms irregularly spaced. Since the last liquid to solidify is generally along the grain boundaries, there tends to be a higher .a concentration of impurity atoms in that area. Figure 2-22 shows schemat-
ically the proisess of crystallization from nuclei to the final grains. 2-21
S:S:S;o-:;,
Crystal Imperfections It is apparent from the preceding section that most materials when solidified consist of many crystals or grains. It is possible under carefully controlled conditions to manufacture a single crystal. The "
so-called metal
"
whiskers,
which in some cases are made directly from the vapor, are nearly perfect single crystals. Figure 2-23 shows tin whiskers growing from a copper substrate made with the scanning electron microscope at a magnification of 20,000 times. Single crystals may also be made by withdrawing a crystal fragment or
:: vrK-:::
{0]
i
m
seed at a carefully controlled speed from a melt which is held at just above
the freezing point. In any case, single crystals approach a nearly perfect
if
.i
i 5 c)
Fig. 2-22 Schematic representation of the process crystallization by nucleation and dendritic growth. (I F. Mondolfo and O. Zmeskal, "Engineering Metal McGraw-Hill Book Company, New York, 1955.) L
.
lattice structure.
It is possible t
metal by the force required to sep This turns out to be several mil
preached by the strength of sine the ordinary strength of metals is a difference? The answer is foun Fig. 2-21
Schematic representation of a grain boundary
between two crystals. The cross-hatched atoms are those
which constitute the boundary material. (From L. F. Mondolfo and 0. Zmeskal,
"
Engineering Metallurgy," McGraw-Hill
Book Company, New York,. 1955.) '
< -
mi
i
I sip
I i !
I
tal structure.
It is interesting to realize the ai surface of a crystal during growtl
per day, requires the deposition <
--hiAwM* ,
mi
wwawiwww-wb
METAL STRUCTURE AND CRYSTALLIZATION
eighbors. This leads to a very irregular .
..
95
-
in all commercial metals are commonly Ion in external shape. The area along rain boundary, is a region of mismatch /stalline (amorphous) structure at the jularly spaced. Since the last liquid to
a
boundaries, there tends to be a higherhat area. Figure 2-22 shows schematrom nuclei to the final grains. from the preceding section that most
:
.
'
F
Ms to manufacture a single crystal. The
-
some cases are made directly from the stals. Figure 2-23 shows tin whiskers
mft
P1
c cijmany crystals or grains. It is possible Q
.
_
Uj]
bide with the scanning electron micro-
p1
imes.
by withdrawing a crystal fragment or . from a melt which is held at just above gle crystals approach a nearly perfect
i c
i
:
ic)
id)
Fig. 2-22 Schematic representation of the process of crystallization by nucleation and dendritic growth. {From L F. Mondolfo and O. Zmeskal. "Engineering Metallurgy," McGraw-Hill Book Company, New York, 1955.) ,
lattice structure. It is possible to calculate the theoretical strength of a metal by the force required to separate the bond between adjoining atoms. This turns out to be several million pounds per square inch and is approached by the strength of single crystals or metal whiskers. However, the ordinary strength of metals is 100 to 1,000 times less. Why is there such a difference? The answer is found in the occurrence of defects in the crystal structure.
la
%
It is interesting to realize the amount of activity that is occurring on the surface of a crystal during growth. A very slow growth rate, such as 1 mm
dolfo
per day, requires the deposition of about one hundred layers of atoms per i
INTPODUCTION TO PHYSICAL METALLURGY
'
.
VV .
o
'
.
\)0
o
Vacant-
lattice
i:
O
O
O
site
Interst atom
O (j) O Q) O io)
ffig. 2-24 Vacancy and interstitial crystal defects.
be produced by raising the tempi nuclear particles. Interstitial ato distortion during plastic deforma A dislocation may be defined stantially perfect parts of a crysl illustrated schematically in Fig. 2 extra half plane of atoms in the ci because of the spiral surface fc
: 4
"
'
screw-dislocation line.
Fig. 2-23 Tin whiskers growing from a copper substrate! Made with the scanning electron microscope at a magnification of 20,000xi
The disk
below the dislocation and tensile
in the lattice structure. Where X\
is not too great, the grain boun parallel edge dislocations (Fig.
the surface. All these atoms must be laid down in exactly the right sort of order for the crystal to be perfect. It is therefore not surprising that few crystals are perfect and that imperfections exist on an
second on
interaction between dislocations
,
properties of metals. The student should refer to the
atomic scale.
a more complete description of t
The most important crystal imperfections are vacancies, interstitials,
and dislocations. Vacancies are simply empty atom sites (Fig. 2-24a). It may be shown by thermodynamic reasoning that lattice vacancies are a stable feature of metals at all temperatures above absolute zero. By successive jumps of atoms, just like playing Chinese checkers, it is possible for a vacancy to move in the lattice structure and therefore play an important part in diffusion of atoms through the lattice. Notice that the atoms surrounding: a vacancy tend to be closer together, distorting the lattice planes. 1 It is possible, particularly in lattice structures that are not close-packed and in alloys between metals that have atoms widely different in atomic diameters, that some atoms may fall into interstitial positions or in the spaces of the lattice structure (Fig. 2-24/3). Interstitials tend to push the surrounding atoms farther apart and also produce distortion of the lattice planes. Vacancies are not only present as a result of solidification but can
'
.>
i m
[0]
Fig. 2-25 Dislocations, (a) Edge dislocation; (b) s( dislocation. (From L. F. Mondolfo and 0. Zmeskal, "
Engineering Metallurgy," McGraw-Hill Book Com
New York, 1955.)
:
-
'
'
.
"
:
..
'
"
;V : :::
'
v ; .
: >: . -
METAL STRUCTURE AND CRYSTALLIZATION
o \ 0(j)0 Vocont lattice site
o
(T cp o
OCpO
\
o Interstitial
o
o
rate.
atoms must be laid down in exactly the to be perfect. It is therefore not sur;ct and that imperfections exist on an
it)
Vacancy and interstitial crystal defects
.
properties of metals.
The student should refer to the references at the end of this chapter for
a more complete description of the types and theory of dislocations.
erfections are vacancies, interstitlals,
simply empty atom sites (Fig. 2-24a). :;
reasoning that lattice vacancies are a
eratures above absolute zero. By suclaying Chinese checkers it is possible ,
i structure and therefore play an imporiugh the lattice Notice that the atoms i closer together distorting the lattice .
,
;
e structures that are not close-packed have atoms widely different in atomic
;
lall into interstitial positions or in the 2.246) Interstitials tend to push the
[0
Fig. 2-25 Dislocations (a) Edge dislocation; (6) screw dislocation. (From L F. Mondolfo and O. Zmeskal, ,
.
d also produce distortion of the lattice
.
.
sent as a result of solidification but can
if
o
be produced by raising the temperature or by irradiation with fast-moving nuclear particles. Interstitial atoms may be produced by the severe local distortion during plastic deformation as well as by irradiation. A dislocation may be defined as a disturbed region between two substantially perfect parts of a crystal. Two simple kinds of dislocation are illustrated schematically in Fig. 2-25. The edge dislocation consists of an extra half plane of atoms in the crystal. The screw dislocation is so named because of the spiral surface formed by the atomic planes around the screw-dislocation line. The dislocation line produces compressive stress below the dislocation and tensile stresses above it and a disturbed region in the lattice structure. Where the mismatch between neighboring grains is not too great, the grain boundary may be represented by an array of parallel edge dislocations (Fig. 2-26). The creation, multiplication, and interaction between dislocations are very useful in explaining many of the
A -
o
o
atom
[a]
Fig. 2-24
4
97
[;
"Engineering Metallurgy," McGraw-Hill Book Company, New York, 1955.)
J
'
i !
98 introduction! to physical metallurgy
4
%
rttrn
3
WW. Fig. 2-26
A small angle boundary composed of edge dislocations', indicated at the T symbols (By permission
i
.
.
from C. S. Barrett, "Structure of Metals," 2d ed
McGraw-Hill
,
.
Book Company, New York 1952.) ,
j
2-22
<
Macrodefects in Castings
The preceding section discussed defects on an Other defects that may result
atomic scale that arise from solidification
.
from solidification are large enough to be visible to the naked eye
.
These
are known as macrodefects. The most common macrodefects are shrink-
Fig. 2-27 Shrinkage cavity in a niobium (columl
v
'
:V :V:
age cavltips and porosity.
billet, (Fansteel Metallurgical Corporation.)
Liquid metals, with few exceptions undergo a contraction in volume due ? ,
by reaction of the liquid metal
to solidification. This decrease in volume may be as much as 6 percent In a properly designed mold, with provision for liquid supply to the portion that solidifies last, the contraction in volume presents no serious problem If, however, the entire exterior of the casting should solidify first, the decrease in volume of the interior during solidification will result in a large shrinkage cavity at the mid-section, as shown in Fig. 2-27. In the solidification of steel ingots, the shrinkage cavity, called pipe, is usually concentrated in tifie top central portion of the ingot. This portion is cut off and dis.
in the mold. Porosity may be gn
and by not unduly compacting
.
Hottears.
re cracks due to I
casting just after solidifirritinn -
mold to collapse and allow thi result from the same nonuniforr
carded before working. .
The ide l solidification would be that in which the metal first freezes at
the bottorli of the mold and continues upward to a riser at the top; however,
.
/V'
'
heat is dis sipated more rapidly from the top of the mold. To minimize the
formation pf shrinkage cavities, abrupt changes in thickness and combinations of heavy and light sections should be avoided. If the casting does have heavy sections, they should be designed with risers at the top to supply liquid metal during solidification. Heavy sections should be cast uppermost in the mold, and chills may be used in the sand adjacent to the
i
i
mm.
4
slow-coolijhg parts.
Porosityjsor blowholes occur whenever gases are trapped in the casting.
. .
They are Usually more numerous and smaller than shrinkage cavities and
may be distinguished by their rounded form (Fig. 2-28). Air may be entrapped inhhe casting by the suciden rusih of metal during pouring. Since
.
gases are penerally more soluble in liquici metal than the solid, dissolved gases ma ibe liberated during solidification. Gases may also be produced
I li
'
J
1}
i ;
.
.
.
.
. u
.
.
.
< .'
Fig. 2-28 Gas pockets and porosity in a tool st£ enlarged 2x. ft
i
.
£3
.
1
1
ir..
.
-
METAL STRUCTURE AND CRYSTALLIZATION
99
3 s
3
ait
"" - :mi
ed of edge ly permission McGraw-Hill
wm seeding section discussed defects on an '
immm
lification. Other defects that may result gh to be visible to the naked eye. These ; most common macrodefects are shrink-
.
FiO 2 27
Shimkage cavity in a niobium (columbium) alloy
billet, (Fansteel Metallurgical Corporation.)
ms, undergo a contraction in volume due 1 volume may be as much as 6 percent
.
by reaction of the liquid metal with volatile substances, such as moi&ture,
.
in the mold. Porosity may be greatly reduced by proper venting of the mold, and by not unduly compacting the sand.
provision for liquid supply to the portion i In volume presents no serious problem, the casting should solidify first, the deuring solidification will result in a large i as shown in Fig, 2-27. In the solidificae cavity, called pipe, is usually concenthe ingot. This portion is cut off and dis-
Hotjears are cracks_due to heavy shrinkage strains set up in the solid casting just ailsL3oMiil 3lioa A common cause is the failure of the sand
mold to collapse and allow the casting to contract. Hot tears may also
,
result from the same nonuniform cooling conditions that give rise to shrink-
e that in which the metal first freezes at
ajbs upward to a riser at the top; however, m the top of the mold. To minimize the rupt changes in thickness and combina-
:
should be avoided. If the casting does 1 be designed with risers at the top to J
cation. Heavy sections should be cast I may be used in the sand adjacent to the anever gases are trapped in the casting,
.
" .
.
.
-
and smaller than shrinkage cavities and mded form (Fig. 2-28) Air may be en:en rush of metal during pouring. Since
i
-
.
'
in liquid metal than the solid dissolved ,
dification. Gases may also be produced
Air-
Fig. 2-28 Gas pockets and porosity in a tool steel casting, enlarged 2x,
INTRODUCTION TO PHYSICAL METALLURGY
i . t
age cavities. Proper design of the casting will minimize the danger of hot tears. Fig. 2-29 shows a hot tear in a stainless steel casting
'
.
.
2-23 I
Grain Size The size of grains in a casting is determined by the relation between the rate of growth G and the rate of nucleation N. If the number of nuclei formed is high, a fine-grained material will be produced, and if only a few nuclei are formed, a coarse-grained material will be produced. The rate of cooling is the most important factor in determining the rate of nucleFation and therefore the grain size. Rapid cooling (chill cast) will result in 5 a large number of nuclei formed and fine grain size, whereas in slow cool-
jl ing (sand cast or hot mold) only a few nuclei are formed and they will have ia chance to grow, depleting the liquid before more nuclei can form.
Other faqtors that increase the rate of nucleation, thus promoting the formation df fine grain, are: 1
Insoluble Impurities such as aluminum and titanium that form insoluble oxides
in steel.
2
J la)
Stirring the melt during solidification which tends to break up the crystals before
ttiej have a qhance to grow very large.
Fig. 2'30 (a) Brittleness of a coarse-gr shock; (b) frequent change of directior of a fine-grained metal. (By permission Mahla, "Principles of Physical Metallur Book Company, New York, 1941.)
The rate of growth relative to the rate of nucleation is greatest at or just
under the freezing point. If the liquid is kept accurately at the freezing temperature anjd the surface is touched by a tiny crystal (seed), the crystal will grow downward into the liquid. If it is withdrawn slowly, a single crystal can be produced.
will exist in the liquid. The out ter and therefore starts to solid
In genera , fine-grained materials exhibit better toughness or resistance
mold wall and begin to grow ii
to shock (Nig. 2-30). They are harder and stronger than coarse-grained material.
)
of the mold and each other,
'
j In industrial casting processes, where a hot liquid is in contact with an originally cool mold, a temperature gradient (difference in temperature)
growth is toward the center. 1 ones, perpendicular to the surf purity lead, as cast, in Fig. 2-3 rate is fast, the grains are smal
rate is much slower, the grain; If the mold has sharp edges, corner because both gaseous
along this plane. Such castinc or forging operations. To a vol i vide the mold with rounded cc 2 24
Grain Size Measurement Thai recommended by the ASTM ar 1 2 3
-
Comparison method Intercept (or Heyn) method Planlmetric (or Jeffries) methoi
Comparison Method The sp
to the metallographic procedi microstructure projected at a i Fig. 3-29
Hot tear (arrow) in a stainless steel casting.
WW
I
i
of the structure at the same
'
1
1
,
4 METAL STRUCTURE AND CRYSTALLIZATION
.
101
3 casting will minimize the danger of hot in a stainless steel casting
v
.
casting is determined by the relation beIf the number of I
le rate of nucleation N
.
ed material will be produced and if only grained material will be produced. The it factor in determining the rate of nude,
Rapid cooling (chill cast) will result in nd fine grain size whereas in slow cool,
:
ew nuclei are formed and they will have -.juid before more nuclei can form. jrate of nucleation thus promoting the
;
,
i
urn and titanium that form insoluble oxides
D
i which tends to break up the crystals before
Fig. 2-30 (a) Brlttleness of a coarse-grained metal under shock; {b) frequent change of direction of force In rupture cf a fine-grained metal. (By permission from Doan and
'
.
id)
rate of nucleation is greatest at or just d is kept accurately at the freezing tem3d by a tiny crystal (seed), the crystal If it is withdrawn slowly a single crystal
Mahia, "Principles of Physical Metallurgy," McGraw-Hill Book Company, New York, 1941.)
will exist in the liquid. The outside is at a lower temperature than the cen-
,
ter and therefore starts to solidify first. Thus many nuclei are formed at the
exhibit better toughness or resistance
mold wall and begin to grow In all directions. They soon run into the side
der and stronger than coarsergrained
of the mold and each other, so that the only unrestricted direction for
growth is toward the center. The resulting grains are elongated columnar ones, perpendicular to the surface of the mold. This is illustrated for highpurity lead, as cast, in Fig. 2-31. Next to the mold wall, where the cooling rate is fast, the grains are small, while toward the center, where the cooling
/here a hot liquid is in contact with an
3 gradient (difference in temperature)
rate is much slower, the grains are larger and elongated.
If the mold has sharp edges, a plane of weakness will develop from this |
corner because both gaseous and solid impurities tend to concentrate j along this plane. Such castings may cause internal rupture during rolling or forging operations. To avoid this plane, it is good casting design to pro-
1
vide the mold with rounded corners. 2-24
r
Grain Size Measurement The three basic methods for grain size estimation recommended by the ASTM are: 1 2 3
Comparison method Intercept (or Heyn) method Planimetrlc (or Jeffries) method
Comparison Method The specimen is prepared and etched according to the metallographic procedure described in Chap. 1. The image of the microstructure projected at a magnification of 10Ox, or a photomicrograph of the structure at the same magnification, is compared with a series of
1
I
102
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 2-9
ASTM grain-size Ranges t ,
GRAIN SIZE NO.
GRAINS PE MEAN
-
n -
1
:
2
N
1
:
2
2
3
4
4
8
5
16
6
32
7
64
8
128
9
256
10
512
least three fields to assure a n
in millimeters divided by the a gives the average intercept le method is recommended partic Planimetric (or Jeffries) Metho (usually 5,000 sq mm) is inscrib' glass of the metallograph. A rr give at least 50 grains in the fie included completely within the grains intersected by the circun or equivalent whole grains with the specimen, the number of gn multiplying the equivalent num magnification factor (Jeffries rr
r
Fig. 2-31
High-purity lead, as cast. Magnified 2x, (National
Lead Co. Research Laborktoiv.)
graded standard grain-size charts (ASTM E112-63). By trial and error a match is secured, and the grain size of the metal is then designated by a
'
number corresponding to the index number of the matching chart. Metals showing a mixed grain size are rated in a similar manner, and it is customary in such cases to report the grain size in terms of two numbers de-
TABLE
noting the approximate percentage of each size present. The comparison method is most convenient and sufficiently accurate for specimens con-
sisting of equiaxed grains. The ASTM grain-size number n maybe obtained
2-10
Relationship
be-
tween Magnification Used and Jeffries' Multiplier f for an Area of 5.000 sq. mm
h
MAGNIFICATION USED
as follows: N = 2"- 1
1
0 002
25
0 125
50
05
.
.
,
where N is the number of grains observed per square inch at 100x magnification (see;Table 2 9). Intercept (or Heyn) Method The grain size is estimated by counting on a ground-gldss screen, or photomicrograph, or on the specimen itself, the
300
18.0
number of grains intersected by one or more straight lines. Grains touched
500
50.0
by the end of the line count only as half grains. Counts are made on at
1000
200.0
I
f 4
5
I
75
.
1 125 .
100
20
200
80
.
.
"
5 :
METAL STRUCTURE AND CRYSTALLIZATION 103
TABLE 2-9 ASTM grain-size Ranges
ii
i
N = 2n-'
,
GRAINS PER SQ IN. AT lOOx
GRAIN SIZE NO
.
,
MEAN 1
n
N
:
RANGE 1
r
2
2
1 5-3
3
4
3-6
A
8
6-12
5
16
12-24
6
32
24-48
64
48-96
.
.i
8
128
96-192
9
256
192-384
10
512
384-768
"
TV Hi*:
1,
.
least three fields to assure a reasonable average. The length of the line in millimeters divided by the average number of grains intersected by it "
gives the average intercept length or grain diameter." The intercept method is recommended particularly for grains that are not equiaxed. Planimetric (or Jeffries) Method A circle or a rectangle of known area (usually 5,000 sq mm) is inscribed on a photomicrograph or on the ground glass of the metallograph. A magnification should be selected which will
give at least 50 grains in the field to be counted. The sum of all the grains
3TM E112-63). By trial and error a of the metal is then designated by a jmber of the matching chart. Metals
:
included completely within the known area plus one-half the number of grains intersected by the circumference of the area gives the total number or equivalent whole grains within the area. Knowing the magnification of the specimen the number of grains per square millimeter is determined by multiplying the equivalent number of whole grains by the corresponding magnification factor (Jeffries multiplier) / given in Table 2-10. ,
'
\i in a similar manner, and it is cuslin size in terms of two numbers de-
each size present. The comparison iently accurate for specimens congrain-size number/? may be obtained ;
TABLE 2-10 Relationship between Magnification Used and Jeffries' Multiplier f for an Area of 5,000 sq. mm MAGNIFICATION USED
rved per square inch at 100x magni.
yn size is estimated by counting on graph, or on the specimen itself, the
" --
1
0 002
25
0 125
50
05
75
1 125
.
.
.
.
100
20
200
80
300
18.0
.
.
j nnore straight lines. Grains touched
500
50.0
ialf grains. Counts are made on at
1000
200.0
I
'
f
i
i 104
INTRODUCTION
0 PHYSICAL METALLURGY
Thus, if the equivalent number of whole grains is found to be 75 at a mag-
2-22
List three factors that tend to
nification of lOOx, the number of grains per square millimeter is equal to |
2-23
Why is it important to avoid s
75 x 2.0, of 150.
2-24
Describe a method of obtainir
thick and thin section.
In case df dispute, the pianimetric method is preferred over the compariI
2-25
son method for equiaxed grains. it Is impd/tant to realize that in using any method to determine grain size, -i
In the cubic system, the inter
the planes is given by the following
the estimafSon is not a precise measurement. A metal structure is a mixture of three-dimensional crystals of varying sizes and shapes. Even if all the crystals were identical in size and shape, the cross sections of the grains
where {hkl) are the Miller indices an planar spacing for the (110), (111), I of the above planes has the greatest
on the polished surface would show varying areas depending upon where the plane ctits each individual crystal. Therefore, no two fields of observa- i
tion can bejexactly the same.
2 26 In the cubic system, the ang
"
TftX/J may be found by the followin cos $
QUESTIONS
Calculate the angle between the (10C 2-1
Differentiate between atomic number and atomic weight
2-2
What Is meant by an isotope? Explain the arrangement of the elements in the periodic table What rrlay be said about all the elements in the same group of the periodic
2-3 2-4
.
.
the (111) plane; the (110) plane and 27 An x ray diffraction analysis c
2
-
length of 0.58 A. A reflection is ob planar spacing?
table? 2-5
Why are some elements known as transition elements ? Using the system explained in Sec 2-2, write the electron configuration of the elements in the fourth period. 2-7 Define a solid Glass is not considered a true solid. Why? '
2-6
.
2-28
Density calculations may be
analysis. Density
.
2-8
How does the metallic bond differ from the ionic and covalent bonds?
2-9
Describe an experiment to show that the atoms of a solid are in motion Explairi the existence of attractive and repulsive forces between atoms 2-11 What is atomic diameter and how may its value be approximated? 2-12 How does the atomic diameter change within one group of the periodic table? .
2-10
Why? 2-13
H How does the atomic diameter change within a period of the periodic table?
Why? 2-14
Give three examples of allotropic metals. For each example, give the tempera-
tures and the* changes In crystal structure. 2-15
What are the Miller indices of a plane that intersects the X axis at 2 and the
Y axis at Vj aVid is parallel to the Z axis? The structure Is cubic. 2-16
where weight/unit cell = (no. of at(
.
of atom is the atomic weight/Avogai
(a) Copper is f.c.c. and has a latt density in gm/cm3, and check wit! (b) Do the same with iron (b.c.c. i 2-29 If copper is f.c.c. and has an i parameter. 2 30 If iron is b.c.c. and has a lal diameter. 2-31
The equation for the equilibrii
What dre the Miller indices of a plane in the cubic structure that goes through
y = 1/2, Z = 1J and is parallel to the X axis? Draw the cubic structure and Crosshatch this plane. 2-17
What
1
re the Miller indices of a plane in the cubic structure that intersects
where AH = molar heat of reaction R = gas content
the X axis at Va, the Y axis at 1, and is perpendicular-to the XV plane?
When a gas liquefies, energy is released as the heat of vaporization. What is this energy due to? 2-19 Is ther§ any difference in the kinetic energy of the atoms in the liquid and the gas at the bojling point? Explain. 2-18
'
.V
2-20 2-21
Differe|itiate between a crystal, a dendrite, and a grain. Isithe grain boundary irregular? . W: '
Why
*
.
l!
i
! 3 5
;r
'
T=
absolute temperature ii
Using AH = 20,000 cal/mol for alun and R = 2 cal/mol-deg calculate the fraction of lattice site:
Using the pianimetric method lent grains at 200x was 62. Calcul
2-32
What is the equivalent ASTM grain :
.A
"
I
METAL STRUCTURE AND CRYSTALLIZATION
; yVAwhole .
grains is found to be 75 at a magVr-'- rains per square millimeter is equal to
2-22 2-23 2-24
List three factors that tend to promote fine grain in a casting. Why is it important to avoid sharp corners in castings? Describe a method of obtaining a uniform grain size in a casting which has a
thick and thin section.
3 method is preferred over the compari-
jing any method to determine grain size
In the cubic system, the interplanar spacing dnki measured at right angles to the planes is given by the following formula:
2-25
,
a
Wement. A metal structure is a mixture
\/h' + k! 4 I'
rying sizes and shapes. Even if all the
shape, the cross sections of the grains ;
105
where (hkl) are the Miller indices and a is the lattice parameter. Calculate the interplanar spacing for the (110), (111), (120), (221), and (123) planes of copper. Which of the above planes has the greatest interplanar spacing?
s |yv varying areas depending upon where aI. Therefore, no two fields of observa-
-
gg In the cubic system, the angle (/> between the plane (/),/<,/,) and the plane 1frJ<2l2) may be found by the following formula: h,h2 + k,k, + 1,1.2 cos 4>
\/(h,2 + /(1! + /,2)(V + k* +12')
Calculate the angle between the (100) plane and the (110) plane; the (100) plane and :
-
-
;: ber -
.
and atomic weight.
the (111) plane; the (110) plane and the (111) plane. An x ray diffraction analysis of a crystal Is made, with x-rays having a wavelength of 0.58 A. A reflection is observed at an angle of 6.45°. What is the inter-
2-27 .
:
.
jments in the periodic table.
V ;V)iements '
:
in the same group of the periodic
planar spacing?
1 i
-
transition elements? ;
.
-
Ans. 2.'575 A Density calculations may be used to check the validity of x-ray diffraction analysis.
2-28
2-2, write the electron configuration of the
weight/unit cell
Jered a true solid. Why?
Density
from the ionic and covalent bonds?
volume/unit cell
lat the atoms of a solid are in motion
.
9 and repulsive forces between atoms
where weight/unit cell = (no. of atoms/unit cell) (weight of atom) and the-weight
/ may its value be approximated?
of atom is the atomic weight/Avogadro s number (6.02 x 1023 atoms).
.
'
(a) Copper is f.c.c. and has a lattice parameter of 3.61 A(10-8cm). Calculate its
iange within one group of the periodic table?
density in gm/cm3, and check with the density value obtained in a handbook. '
:
(b) Do the same with iron (b.c.c. and lattice parameter of 2.86 A).
Jiange within a period of the periodic table?
if copper is f c c and has an atomic diameter of 2.556 A, calculate the lattice parameter. 2-30 If iron is b c c and has a lattice parameter of 2,86 A, calculate the atomic
2-29
metals. For each example give the tempera,
te.
.
plane that intersects the X axis at 2 and the ? The structure is cubic
.
4
2-31
perpendicular to the XY plane?
-
.
:
:
,:
t
.
.
m
The equation for the equilibrium fraction of atom sites that are vacant is:
where AH = molar heat of reaction accompanying the formation of vacancies R = gas content T= absolute temperature in K Using AH= 20,000 cal/mol for aluminum and R = 2 cal/mol-deg *
?tic energy of the atoms in the liquid and the
calculate the fraction of lattice sites that are vacant at 300, 500, 1000, and 1300 K
1
2-32
dendrite, and a grain
.
What is
I
lar?
"
.
°
..
v;->::. Jleased as the heat of vaporization c'
.
/\/ = g-iH/nr
is? Draw the cubic structure and Crosshatch
y: plane in the cubic structure that intersects
.
.
diameter.
lane in the cubic structure that goes through
.//
.
L
.
°
.
Using the planimetric method for measuring grain size the number of equivalent grains at 200x was 62. Calculate the number of grains per square millimeter. ,
What is the equivalent ASTM grain size? (Refer to Table li, ASTM El 12-63).
i
106
INTRODUCTIOIil TO PHYSICAL METALLURGY i
REFERENCES
American Society for Metals: "Atom Movements Mietals Park, Ohio, 1951. Barrett, C. S.: "Structures of Metals 2d ed., McGraw-Hill Book Company New York, "
,
"
,
,
1952.
Cottrell, A. H.: "Dislocations and Plastic Flow in Crystals
"
,
1/
Oxford University Press,
Fair Lawin, N J 1956. .,
.
Cullity, B, D.: "Elements of X-ray Diffraction Inc., Reading Mass., 1956.
,
Addison-Wesley Publishing Company
,
J|
,
Eisenstadt, M. M.: "Introduction to Mechanical Properties of Materials
" ,
PLAJ DER
The Mac-
millan Company New York, 1971. ,
Guy, A. G.: "Elements of Physical Metallurgy
2d ed., Addison-Wesley Publishing
"
,
Company, Inc. Reading, Mass., 1959. ,
Hume-Rothery, W.: "Atomic Theory for Students of Metallurgy Metals, London, 1955.
"
,
andG
.
The Institute of
V. Raynor: "The Structure of Metals and Alloys," The Institute of Metals
,
London 1969. ,
Mason, C. W.' "Introductory Physical Metallurgy
,
'
American Society for Metals
,
Metals Park, Ohio 1947.
Mondolfo, tL. F. and O. Zmeskal; "Engineering Metallurgy," McGraw-Hill Book Com,
pany, Nevy York, 1955.
Rogers, B.jA.: "The Nature of Metals
"
,
American Society for Metals, Metals Park
,
"
,
pany, Inc., Reading, Mass., 1958.
:
"
Materials Science for Engineers
"
,
Reading, Mass., 1970. Weertman, U., and J. R. Weertman::
mi Man Coirnpany, New York 1964. ,
'
\
Introduction
When a material is
"
Addison-Wesley Publishing Company,
Elementary Dislocation Theory
"
,
ual return of the object ot its
stressed beyond its elastic linr
Ohio, 195j1. Van Vlack, i. H.: "Elements of Materials Science Addison-Wesley Publishing Com'
3 1
deformation or strain is tempor
,
The Mac-
place, and it will not return to i alone. The ability of a metal to most outstanding characteristi
shaping operations such as stc drawing, and extruding involv machining operations such as involve plastic deformation. Th tion and the mechanism by wh
fecting the working operation.
ij
Much information regarding
be obtained by studying the b later applying this knowledge 1 Plastic deformation may tak of both methods.
i
by Slip If a single yond its elastic limit, it elong
3-2 Deformation
indicating relative djsplaeetp
i .
;
to the rest, and the elongation ment on another parallel plam
boring thin sections of the cry cards on a deck. Each succe;
\
results in the appearance of ; tion of a slip plane with the si the load eventually causes the Investigations showed that
-
r-
t
I i
-
-
Vlovements
"
Metals Park, Ohio, 1951.
.
2d ed., McGraw-Hill Book Company tic Flow in Crystals
"
,
iction,
"
,
New York,
Oxford University Press,
Addison-Wesley Publishing Company
,
.
PLASTIC DEFORMATION
pchanical Properties of Materials," The Mactallurgy." 2d ed., Addison-Wesley Publishing
:
,v
'
pr Students of Metallurgy," The Institute of iof Metals and Alloys," The Institute of Metals Metallurgy
" ,
,
American Society for Metals
noering Metallurgy
"
,
,
f
3-1
McGraw-Hill Book Com-
-
American Society for Metals, Metals Park
,
Science," Addison-Wesley Publishing Com-
; ,
srs,
"
.' I
most outstanding characteristic in comparison with other materials. All shaping operations such as stamping, pressing, spinning, rolling, forging, drawing, and extruding involve plastic deformation of metals. Various machining operations such as milling, turning, sawing, and punching also involve plastic deformation. The behavior of a metal under plastic deforma-
Addison-Wesley Publishing Company,
Elementary Dislocation Theory
"
,
When a material is stressed below its elastic limit, the resulting
deformation or strain is temporary. Removal of the stress results in a gradual return of the object ot its original dimensions. When a material is stressed beyond its elastic limit, plastic or permanent deformation takes place, and it will not return to its original shape by the application of force alone. The ability of a metal to undergo plastic deformation is probably its
i '
Introduction
The Mac-
tion and the mechanism by which it occurs are of essential interest in perfecting the working operation. Much information regarding the mechanism of plastic deformation may be obtained by studying the behavior of a single crystal under stress and later applying this knowledge to a polycrystalline material. Plastic deformation may take place by slip, twinning, or a combination
1
of both methods. 3 2
Deformation by Slip
If a single crystal of a metal Is stressed in tension be-
yond its elastic limit, it elongates slightly, a step appears on the surface
indicating relative displacement of one- pftrt of the crystal with respect and the elongation stops. Increasing the load will cause move-
to the rest
,
ment on another parallel plane resulting in another step. It is as if neigh,
7
boring thin sections of the crystal had slipped past one another like sliding cards on a deck. Each successive elongation requires a higher stress and results in the appearance of another step which is actually the intersection of a slip plane with the surface of the crystal Progressive increase of ,
.
i
the load eventually causes the material to fracture
.
Investigations showed that sliding occurred in certain planes of atoms
'
"
"
; ; ;
'
:
-
,
r
108
INTRODUCTIONjTO PHYSICAL METALLURGY
in the crystal and along certain directions in these planes
The mechanism
.
by which a metal is plastically deformed was thus shown to be a new type |
m
[a)
of flow, vastly different from the flow of liquids or gases It is a flow that depends upon the perfectly repetitive structure of the crystal which allows the atoms; in one face of a slip plane to shear away from their original neighbors in the other face/HcJ slide in an organized-way along this face carrying their own half of the crystal with them and finally to join up again with a new set of neighbors as nearly perfect as before It was pointed out in the preceding chapter that parallel planes of high .
,
.
atomic density and corresponding large interplanar spacing exist in the ] crystal structure. Any movement in the crystal takes place either along these planes or parallel to them.
Investigation of the orientation of the slip plane with respect to the applied stress indicates that slip takes place as a result of simple shearing
stress. Resolution of the a ial tensile load F in Fig. 3-1 gives two loads
.
One
|s a shear ;load (Fs = F cos e) along the slip plane and the other a normal
fig. 3 2
Plastic flow occurs when planes of atoms
one another. Close-piacked planes do this more eas
(a) than planes aligned in another direction (6),
The resulting stresses are F cos
tensile load (F = F sin e) perpendicular to the plane The area of the slip n
.
Shear stress Ss:
plane is /A/sin 0, where A is the cross-sectional area perpendicular to F.
A/s\n F sin
Normal stress S„
i
A/sin
From Eq. (3-1), it is evident th maximum when e = 45
° .
A more important factor in de of shear on the slip plane. Slipo most closely packect. since this
\
was pointed out in the previous c apart from each other than rows can slip past each other with I since the atoms are not bonded
s
Slip; plane
gether by the free electron gas could slide past each other parth ring to Fig. 3-2, the atoms in a n are farther apart vertically than tl for a given horizontal displacemi bars between the atoms. In addi
the atoms into unstable positioi into stable ones, when these s
Fig. 3-2(a).
Figure 3'3 shows the packing directions in which the atoms £ Fitj. 3-1
Components of force cn a slip plane.
i I mimm
"
T i
\
mm i
\ i
.
easy slip directions. The shear
t,
GY
PLASTIC DEFORMATION
The mechanisml
directions in these planes
.
Vjeformed was thus shown to be a new type
7
109
sssssm
9 flow of liquids or gases It is a flow thal etitive structure of the crystal which allows'|! d plane to shear away from their originals .
islide in an organized way along this
stai with them and finally to join up again | ,
iearly perfect as before
.
{eding chapter that parallel planes of high
.
ng large interplanar spacing exist in the '
: v:::,v .
.it in the crystal takes place either along
i of the slip plane with respect to the ap
Tig. 3-2 Plastic flow occurs when planes of atoms slip past iiOne another Close-packed planes do this more easily
ikes place as a result of simple shearing !
i ja) than pianos aligned in another direction (b).
-
sile load F in Fig
.
3.1 gives two ioads
.
.
One 1
Fhe resulting stresses
mg the slip plane and the other a normal dicular to the plane The area of the slip
are F cos 6
.
'
Shear stress Ss =
cross-sectional area perpendicular to F
.
A/s\n e
F sin e
Normal stress S„
/A/sin e
F
F -
cose sin e-
- sin 26
(3-1)
2A
A F = - sin2 e A
(3-2)
From Eq. (3-1), it is evident that the shear stress on a slip plane will be maximum when (9 = 45°.
A more important factor in determining slip movement is the direction of shear on the slip plane. Slip occurs in directions in which the atoms are most closely packed, since this requires the least amount of energy. If was pointed out in the previous chapter that close-packed rows are farther apart from each other than rows that are not close-packed, therefore they can slip past each other with, less interference. We might also expect, since the atoms are not bonded directly together but are merely held together by the free electron gas, that these close-packed rows of atoms could slide past each other particularly easily without coming apart. Referring to Fig. 3-2, the atoms in a row in (a) are closer together and the rows are farther apart vertically than they are in (£)), so that less force is required
"
,
:
-
5
for a given horizontal displacement, as suggested by the slope of the black bars between the atoms. In addition, less displacement is required to move the atoms into unstable positions from which they will be pulled forward into stable ones, when these stable positions are closer together as in Fig. 3 2(a). Figure 3-3 shows the packing of atoms on a slip plane. There are three
V
directions in which the atoms are close-packed, and these would be the easy slip directions. The shear stress S on the slip plane, which was des
'
. I
"
' -
110
INTRODUCTION TO PHYSICAL METALLURGY
I
5
i Hi
kr-Sr.
Eosy slip directions
i i
i
5
X
Fig. S-'S
Resolution of shear stress into a slip direction. \
. .
,
rived earlier (Eq. 3-1), may not coincide with one of these easy slip direc-|
\tions and has to ;be resolved into the nearest slip direction to determine the resolved shekr stress S As the diagram shows, the stresses are| related by the cosine of the angle x, and the resolved shear stress is rs.
a]
i
F S„ = - sin 20 cos \ 2A
"
(3-3) .
Investigation has shown that differently oriented crystals of a given metal will begin to slip when different axial stresses are applied but that the
,
critical resolved shear stress, that is, the stress required to initiate slip.f is always the same.
If the slip planes are either parallel or perpendicular to the direction of applied stress, slip cannot occur, and either the material deforms by twin-
ning or it fractures. As deformation proceeds and the tensile load remainsl axial, both the plane of slip and the direction of slip tend to rotate into the;
Fig. 3:4 Schematic representation of slip in tensi \ Before straining; (b) with ends not constrained; (c Sljbnstrained. (From B. D. Cuillty, "Elements of X-r.
te-Diflraotion,'' Addison-Wesley Publishing Company Ifteading, Mass., 1956.)
packed [TlO] direction, a dista of that dimension. The series ol
the microscope as a group of i 3-8) In Fig. 3-7, a single vertic .
axis of tension (Fig. 3-4). 3 3
Mechanism pf Slip
Portions of the crystal on either side of a specific slip
plane move in opposite directions and come to rest with the atoms in nec|rly| equilibrium positions, so that there is very little change in the lattice orientation. Thus the external shape of the crystal is changed without destroying it. Sensitive x-ray methods show that some bending or twisting of the lattice planes has occurred and that the atoms are not in exactly normal posi-
tions after deformation. Slip is illustrated schematically in Figs. 3-5 and 3-6| in an f.c.c. (face-centered cubic) lattice.
The (HIi) plane (Fig. 3-5), which is the plane of densest atomic population, intersects the (001) plane in the line ac. When the (001) plane is assumed to be the plane of the paper and many unit cells are taKen together
(Fig. 3-6), slip is seen as a movement along the (111) planes in the close-
m
o
[Tio] Direction I Fig. 3-5 Slip plane and slip direction in an f.c.c.
PLASTIC DEFORMATION
111
Rotation
Bending ' .
"
e with one of these easy slip direcnearest slip direction to determine
yl diagram shows, the stresses are
'
v
id the resolved shear stress is [c]
1 ' / cos A
Fig. 3-4 Schematic representation of slip in tension, (a) Before straining; (b) with ends not constrained; (c) ends
(3-3)
constrained. (From B. D. Culllty.
ly oriented crystals of a given metal stresses are applied but that the the stress required to initiate slip,
:
jither the material deforms by twiniceeds and the tensile load remains ?":pction of slip tend to rotate Into the
Elements of X-ray
Reading, Mass., 1956.)
packed [110] direction, a distance of one lattice dimension or multiple of that dimension. The series of steps formed will generally appear under the microscope as a group of approximately parallel lines (Figs. 3-7 and 3-8) In Fig. 3-7, a single vertical line was scribed on the surface before
ir perpendicular to the direction of ;
"
Diffraction," Addison-Wesley Publishing Company, Inc..
.
i
1
on either side of a specific slip ome to rest with the atoms In nearly )ry little change in the lattice orienystal Is changed without destroying
I
Am) Plane
ome bending or twisting of the lat-
|- Dms are not in exactly normal posid schematically in Figs. 3-5 and 3-6
c
plane of densest atomic popula-
(OOI) Plane
te ac. When the (001) plane is asjmany unit cells are taken together long the (111) planes in the close-
%
0
Fig. 3-5
i
MTlO] Direction
Slip plane and slip direction in an f.c.c. lattice.
A
i
i
!
112
INTRODUCTION TO PHYSICAL METALLURGY
/[WO] direction 3
t:
F
11;
Plane of paper (OOll llll) plane
Pr/rffo
Fig. 3-6 Schematic diagram of slip in an t c c crystal. (By .
.
.
permission from Doan and Mahla, "Principles of Physical
I
I?
Metallurgy," McGraw-Hill Book Company New York, 1941.) ,
straining. After straining, the slip lines appear as parallel lines or steps at an angle of approximately 45°. The vertical line is no longer straight and
= :Fig 3 8 Slip lines in copper. Specimen polished, Pi and then strained. lOOx.
each step amounts to a movement of about 700 to 800 atoms
if'
.
H
,
.
Figure 3-8 shows slip lines in two grains of copper Notice that the slip lines are parallel inside each grain but because of the different orientation
"
'
i
.
across each other. This, however have the same value over all point; atoms and the difficulties of apply
of the unit cells in each grain, they have to change direction at the grain boundary. The grain boundary starts in the lower left corner and runs up-
this condition unattainable.
ward and to the right.
From the schematic picture of slip in Fig. 3-6, one may assume, at first that the motion consists of a simultaneous movement of planes of atoms ,
:v
A m
atoms slip consecutively, starting a plane, and then move outward ove Sir Neville Mott has likened slip t a floor. If you try to slide the entir
great. What you can do instead istc the whole thing a little at a time b larging the slipped region behind i
„.
in front of it. A similar analogy to t the earthworm (see Fig. 3-9). Exar cation of the shear force, an extra |
been formed above the slip plane.
plane and leaves a step when it i |
Each time the dislocation moves i
one atom spacing. Since the aton
tions after the passage of the disb
1
.
location across the same slip plan Fig. 3-7 Single crystal of brass strained in tension, 200x. (By permission from R. M. Brick and A. Phillips, "Structure and Properties of Alloys, 2d ed., McGraw-Hill Book Company, New York, 1949.)
tually, this resistance or distortion to lock the dislocation In the crys Further deformation will require nru
"
9 ~
-
i,T
:
5 i
PLASTIC DEFORMATION
113
v
t
11
-
.
m
1 P'cne
.
,
t
.;
41
4
WW
10 )
.Fig 3-8 Slip lines in copper. md then strained. 100x.
iM appear as parallel lines or steps at Kical line is no longer straight, and
.
i
Specimen polished, etched,
"
"
5 I
i ibout 700 to 800 atoms. kins of copper. Notice that the slip
across each other. This, however, requires that the shearing force must have the same value over all points of the slip plane. The vibrations of the atoms and the difficulties of applying a uniformly distributed force make this condition unattainable. A more reasonable assumption is that the
because of the different orientation
ve to change direction at the grain i the lower left corner and runs up- .
Fig. 3-6, one may assume, at first, Sfeous movement of planes of atoms
1
atoms slip consecutively, starting at one place or at a few places in the slip plane, and then move outward over the rest of the plane. Sir Neville Mott has likened slip to the sliding of a large heavy rug across a floor. If you try to slide the entire rug as one piece, the resistanceis-too great. Whatyou can do instead is to make a wrinkle in-the rug and tnen slide the whole thing a little at a time by pushing the wnnkle along, thereby enlarging the slipped region behind it at the expense of the unslipped region in front of it. A similar analogy to the wrinkle in the rug is the movement of the earthworm (see Fig. 3-9). Examination of Fig. 3-9 shows that, by appli-
cation of the shear force, an extra plane of atoms (called a dislocation) has been formed above the slip plane. This dislocation tnoves across the slip plane and leaves a step when it comes out at the surface of the crystal. Each time the dislocation moves across the slip plane the crystal moves ,
I
one atom spacing. Since the atoms do not end up in exactly normal positions after the passage of the dislocation subsequent movemecjt of the dislocation across the same slip plane encounters greater resistance Eventually, this resistance or distortion of the slip plane becomes great enough to lock the dislocation in the crystal structure and the movement stops. Further deformation will require movement on another slip plane Although ,
.
,
.
1 it
1
| ..
.
1
i
is
1
I i I i
l
«:i14 INTRODUCTION TO PHYSICAL METALLURGY
-
Dislocation line
-
1
metal.
i
Slip plane
the distortion is greatest on the a the lattice structure, and the appli ment on another slip plane. The stress required to initiate required to move one atom ove This result, however, is 1
mentally observed critical resolvt The much lower observed critic;
since it was pointed out in Sec. ; crystal structure as a result of sc
!;
to create a dislocation, but simp slip plane. This theory suggests t in metals would be to manufactur
3-4
perfections. The strength of me crystals, has approached the ttr the dislocation theory. Slip in Different Lattice Structures
slip direction is known as a s//p sy of densest atomic packing in the s in the slip system. In f.c.c. materials there are foi
packed (110) directions (Fig. 3-3) ble slip systems. These slip systerr it is almost impossible to strain a {111} plane in a favorable positk solved shear stress for slip would type of lattice structure (silver, golc c) TABLE 3-1
o'l
Fig. 3-9
Analogy between the movement of a dislocation
through a crystal and |ie movement of an earthworm as it
arches its back while gping forward. (Courtesy Westinghouse Electric Corp.)
i
if:
Critical Resolved Shear Stress
METAL
STRUCTURE
Silver
f
c c
Copper
f
c c
Aluminum
f
c c
Magnesium
c.p.h
Cobalt
c.p.h
Titanium
c.p.h
Iron
b
c c
Columbium
b
c c
Molybdenum
b
c c
.
.
.
.
.
.
.
P
.
.
.
.
S
.
{( {( {
.
.
.
Values wpre tabulated from various sources by Macmillan Company, New York, 1966, p. 106.
*
t
,
PLASTIC DEFORMATION
16-
115
the distortion is greatest on the active slip plane its effect is felt throughout ,
the lattice structure and the applied load mjjst be increased to cause move,
ment on another slip plane.
i
The stress required to initiate slip in a perfect crystal that is, the stress required to move one atom over another may be calculated for a given ,
"
T i
,
metal.
This result
however, is 100 to 1.000 times larger than the experimentally observed critical resolved shear stress tor slip in single crystals The much lower observed critical resolved shear stress is not surprising since it was pointed out in Sec. 2-21 that dislocations already exist in the crystal structure as a result of solidification. It is therefore not necessary to create a dislocation but simply to start an existing one moving on the slip plane. This theory suggests that one method of attaining high strength
!
,
.
S:.?.ri
,
-
PI
in metals would be to manufacture more nearly perfect crystals without imperfections. The strength of metal whiskers, which are nearly perfect crystals, has approached the theoretical strength and lends support to the dislocation theory. '
\
mi
3-4
'
Slip in Different Lattice Structures The combination of a slip plane and a slip direction is known as a s//p sysfem. The slip direction is always the one of densest atomic packing in the slip plane and is the most important factor in the slip system. In f.c.c. materials there are four sets of (111) planes and three closepacked (110) directions (Fig. 3-3) in each plane, making a total of 12 possible slip systems. These slip systems are well distributed in space; therefore, it is almost impossible to strain an f.c.c. crystal and not have at least one {111} plane in a favorable position for slip. As expected, the critical resolved shear stress for slip would be low (Table 3-1), and metals with this
I
type of lattice structure (silver, gold, copper, aluminum) are easily deformed.
TABLE 3 1
Critical Resolved Shear Stress for Several Metals at Room Temperature*
l
CRITICAL
METAL
STRUCTURE
SLIP
SLIP
PLANE
DIRECTION
RESOLVED
'
1 Silver
f
Copper
f
.
Aluminum
f
.
c c .
.
c c .
.
c c
.
.
.
Magnesium
c.p.h
Cobalt
c.p.h
Titanium
c.p.h
Iron
b
Columbium
b
Molybdenum
b
.
.
.
c c
.
.
.
c c .
.
c c
.
.
.
.
{111} {111} {111} {0001} {0001} {1010} {110}
{110} {110}
(110) (110) (110) (1120) (1120) (1120) (111) (111) (111)
SHEAR STRESS
(PSI) 54 1
71
1
114 64
960 1 990 ,
3 980 ,
4 840
\
,
10,400
Values were tabulated from various sources by W. J. M. Tegart. "Elemeits of Mechanical Metallurgy." The Macmillan Company. New York, 1966. p. 106.
*
I
I
i
Ma 116
INTRODUCTION TO PHYSICAL METALLURGY
The c.p.h. (close-packed hexagonal) metals (cadmium, magnesium,!
A
cobalt, titanium) have only one plane of high atomic population, the (0001)| plane (or basal plane) and three close-packed (1120) in that plane (see]
C
Fig. 3-10). ThHs structure does not iiave so many slip systems as the f.c.cj lattice, and the critical resolved shear stress is higher than for f.c.c. ma-| terials (Table 3-1). While the number of slip systems is limited, deformation by twinning helps to bring more slip systems into proper position, thereby! approaching the plasticity of the f.c.c. structure and surpassing that of] b
c c
.
.
.
E
[112] Twin direction
metals.
Since b.c.c. (body-centered cubic) metals have fewer atoms per unitl cell, they do not have a well-defined slip system and do not have a truly|
.v
ET
close-packed plane. The slip direction is the close-packed (111) direction
'
E
Slip lines jin b.c.c. metals are wavy and irregular, often making identifica-| tioh of a slip plane extremely difficult. The {110}, {112}, and {123} planei
have all b6en identified as slip planes in b.c.c. crystals. Studies have indi-1 cated that an
Fig. S'H
y plane that contains a close-packed (111) direction can aot|
across the twin plane. Twinninc lattice in Figs. 3-11 and 312. In Fig. 3-11, the (111) twinning f
'
as a slip plane. In further agreement with the lack of a close-packed plane is the relatively high critical resolved shear stress for slip (Table 8-1)J Therefore, b.c.c. metals [molybdenum, alpha (a) iron, tungsten] do not
line AB', which is the twin directic
1
show a high degree of plasticity. 3-5
Diagram of a twin plane and twin directio
lic.c. lattice.
in Fig. 3-12. The plane of the pap
Deformation.by Twinning In certain materials, particularly c.p.h. metals| twinning is a major means of deformation. This may accomplish an ex-
?
Twinned
.
tensive change in shape or may bring potential slip planes into a morel
region
favorable position for slip. Twinning is a movement of planes of atoms i the lattice parallel to a specific (twinning) plane so that the lattice is divided,
c
t
G
into two symmetrical parts which are differently oriented. The amount oi movement of each plane of atoms in the twinned region is proportional ti its distance from the twinning plane, so that a mirror image is formei
E A
'
B
B
Plane of paper (llO)
'9
3'12
Schematic diagram of twinning in an
fcc .
.
"
Fig. 3-10
Close-packed directions in the basal plane of the
hexagonal lattice.
ly permission from G. E. Doan, Principles of Phys Wallurgy," 3d ed.. McGraw-Hill Book Company, N( 953.)
[ I
PLASTIC DEFORMATION
agonal) metals (cadmium, magnesium,f ane of high atomic population, the (0001)'
: .
:
A
close-packed (1120) in that plane (seel it have so many slip systems as the f.c.c.!}|
'
:
117
-
ihear stress is higher than for f.c.c. ma-: 5er of slip systems is limited, deformation
t
lip systems into proper position, thereby f
.
c c
.
.
.
Twin direction
bic) metals have fewer atoms per un it I
,
(III) Twinning plane
[112]
structure and surpassing that of :
-
jjphed slip system and do not have a truly .J '
'"
'
:
3
(110) Plone
'
tion is the close-packed (111) direction.
;
E
y and irregular, often making identificacult. The {110}, {112}, and {123} planes mes in b.c.c. crystals. Studies have indii a close-packed (111) direction can act :; -
Fig. 3-11
Diagram of a twin plane and twin direction in an
t c c lattice. .
.
.
across the twin plane. Twinning is illustrated schematically in an f.c.c.
\nX with the lack of a close-packed plane
lattice in Figs. 3-11 and 3-12.
blved shear stress for slip (Table 8-1)..J|
In Fig. 3-11, the (111) twinning plane intersects the (TlO) plane along the
snum, alpha (a) iron, tungsten] do not
line /AS', which is the twin direction. The mechanism of twinning is shown if
in Fig. 3-12. The plane of the paper is the (TlO) plane, and many unit cells
t materials, particularly c.p.h. metals, ormation. This may accomplish an ex-
Twinned
region
bring potential slip planes into a more qs; ng is a movement of planes of atoms in
t
C
G
nning) plane so that the lattice is divided are differently oriented. The amount of
\
in the twinned region is proportional to '-A
E .
ane, so that a mirror image is formed
4
B r
I)
4 (1
"
.
Iv""
"
"
-
B
]
0
Plane of poper (llO)
r *
.
V
.J
(III) Twin plane'
H
\(>12]
Twin direction
j
Fig. 3-12 Schematic diagram of twinning in an f c c lattice. (By permission from G. E. Doan, "Principles of Physical Metallurgy," 3d ed.. McGraw-Hill Book Company, New York. 1953.) .
of the
.
.
% 118
INTRODUCTION TO PHYSICAL METALLURGY
are taken together.
W
.l
Each (111) plane in the twin region moves in shear! M
in the [112] direction. The first one, CD, moves one-third of an inter|| atomic distance; the second one, EF, moves two-thirds of an interatomic! distance; and the third one, GH, moves an entire spacing.
;f
If a line |is drawn perpendicular to the twin plane (AET) from atom A' l ,
notice that we have another atom, C, exactly the same distance away froni|
the twinned plane but on the other side. The same thing is true for all the! atoms in the twinned region, so what we really have here is a mirror imageM
in the twinned region of the untwinned portion of the crystal. Since theif
atoms end up in interatomic spaces, the orientation of the atoms, or the!
distance between atoms, has been changed. Generally, the twinned regionf involves the movement of a large number of atoms and usually it appearsl ,
microscopically as a broad line or band, as shown in Fig. 3-13. This picturei
shows twin bands in zinc; notice how the bands change direction at the i grain boundary. v
f
The twinning plane and direction are not necessarily the same as thosef
for slip. In f.c.c. metals the twin plane is the (111) plane, and he twin direc-|Bp
:
tion Is the [112] direction; in b.c.c. it is the (112) plane and the [111]|| direction.
Two kinds of twins are of interest to the metallurgist: i
1 Deformation or mechanical twins, most prevalent in c.p.h. metals (magnesium,| zinc, etc.) and b.c.c. metals (tungsten, a iron, etc.).
2 Annealing twins, most prevalent in f.c.c. metals (aluminum, copper, brass, etc.).! These metals have been previously worked and then reheated. The twins are formed because of a change in the normal growth mechanism.
3-6
Slip vs Twinning .
Slip and twinning differ in:
1 Amount of movement: in slip, atoms move a whole number of interatomic spac--ings, while in twinning the atoms move fractional amounts depending on their
distance from the twinning plane.
'
Fig. 3-13
Twin bands in zinc. 2X.
|
2 Microscopic appearance: slip appears as thin lines, while twinning appears asj the (100) plane in iron] In polyc surface shows a granular appearar in orientation of the cleavage plan(
broad lines or bands.
3
.
Lattice orientation: in slip there is very little change in lattice orientation, and,
the steps are visible only on the surface of the crystal. If the steps are removed byii
polishing, there will be no evidence that slip has taken place. In twinning, however,!
As with plastic deformation the ture strength and the actual fractur
since there is a different lattice orientation in the twinned region, removal of the|
,
steps by surface polishing will not destroy the evidence of twinning. Proper etching
solutions, sensitive to the differences in orientation, will reveal the twinned regionfl 3-7
Fracture
ties. Freshly drawn glass fibers h values, but anything that can give
Fracture is the separation of a body under stress into two or morQ;J|
nicks or cracks weakens them.
T
,
*
parts. The|failure is characterized as either brittle or c/ucf/7e. Brittle fra'
was given by A. A. Griffith in 1921
ture generally involves rapid propagation of a crack with minimal energyl
rials was caused by many fine el metal. The sharpnesss at the tip c stress concentratipn which may e:
.
absorptioti and plastic deformation. In single crystals, brittle fracture!
occurs byijcleavage along a particular crystaiiographic plane [for example
I I
i
i f "
Si
i;
.A
? i ,;.m,w»j,a * m www mmmimm
-
'
7
* '"
A
PLASTIC DEFORMATION
119
Diane In the twin region moves in shear W one, CD, moves one-third of an inter? .
'
A
"
"
i EF moves two-thirds of an interatomic ,
noves an entire spacing. r to the twin plane (AB ) from atom A'/ '
'
exactly the same distance away from r side. The same thing is true for all the lat we really have here is a mirror image inned portion of the crystal. Since the
|C
,
mm.5i;
,
es, the orientation of the atoms, or the
.
h
.
fM. changed. Generally, the twinned region ""
dumber of atoms, and usually it appears
band, as shown in Fig. 3-13. This picture how the bands change direction at the .
is
V
1
T are not necessarily the same as those :)vsine
is the (111) plane, and the twin direc-
<::: kc.
it is the (112) plane and the [111]
'
i'
-
MM to the metallurgist:
post prevalent in c.p.h. metals (magnesium, I Iron, etc.). Lc.c. metals (aluminum, copper, brass, etc.). [ed and then reheated. The twins are formed
1
vth mechanism.
ifferin:
move a whole number of Interatomic spac'
'
i - ;
Fig 3-13
Ve fractional amounts-depending on their
Twin bands in zinc 2x. .
rs as thin lines, while twinning appears as
ery little change in lattice orientation, and I of the crystal. If the steps are removed by slip has taken place. In twinning, however,
:?
'
tion in the twinned region, removal of the :;|
W the evidence of twinning, Proper etching ;.|
| orientation, will reveal the twinned region. !| i
a body under stress into two or more
is either br/ff/e or ductile. Brittle fracjation of a crack with minimal energy i In single crystals, brittle fracture .
ir crystallographic plane [for example,
.
the (100) plane in iron]. In polycrystalline materials, the brittle-fracture surface shows a granular appearance (Fig. 3-14b) because of the changes in orientation of the cleavage planes from grain to grain As with plastic deformation, the difference between the theoretical frac.
ture strength and the actual fracture strength is due to structural irregularities. Freshly drawn glass fibers have strengths approaching theoretical values, but anything that can give rise to surface irregularities such as nicks or cracks weakens them. The first explanation of this discrepancy ,
,
was given by A. A. Griffith in 1921 He theorized that failure in brittle materials was caused by many fine elliptical submicroscopic cracks in the .
,
metal. The sharpnesss at the tip of such cracks will result in a very high stress concentration which may exceed the theoretical fracture strength
IP
m 120
INTRODUCTION TO PHYSICAL METALLURGY
Mr-
!
The final stage leaves a circul£
fy»?
on the surface of the other ha shallow cup, and the other h
gr| igiving rise to the term cup-anc .8 Slip Twinning, and Fracture ,
Thi
before fracture is determined b
for slip, twinning, and cleavag for slip which is increased by i deformation.
There is a critic!
is also increased by prior defon for cleavage on a particular pl£
la)
ifc-
tion and temperature. When a takes place depends upon wh
v--
critical resolved shear stress fo
will slip or twin and show sonstress is reached first, the crystf little or no plastic deformation.
m
.9 Polycrystalline Material The pn formation in single crystals. Co up of polycrystalline grains, w
/ Fig. 3-14 (a) Ductile cup-and-cone fracture in low-carbon steel; (b) brittle frabture in high-carbon steel.
When a polycrystalline materh
those grains in which the slipsi to the applied stress. Since coi
at this localized area and cause the crack to propagate even when the bod] of the material is under fairly low applied tensile stress. I
It is possible that microcracks may exist in the metal due to the previous history bf solidification or working. However, even an initially sound matgs
rial may develop cracks on an atomic scale. As shown in Fig. 3-15a, tli| pile-up of dislocations at a barrier which might be of grain boundary
Dislocation
included particle may result in a microcrack. Another method (Fig. 3-15t|
Jislocations
source
i
.
is for thjree unit dislocations to combine into a single dislocation. From thi
above explanation, it is apparent that any method that will increase thj
,
mobility of dislocations will tend to reduce the possibility of brittle fractutt
'
Ductile fracture occurs after considerable plastic deformation prior
wrack
Ban
failure. J The failure of most polycrystalline ductile materials occurs with] cup-and-cone fracture associated with the formation of a neck in a tensil specimen. See Fig. 3-14a. The fracture begins by the formation of caviti in the benter of the necked region. In most commercial metals, these i
V
ternal cavities probably form at nonmetallic inclusions. This belief is su| ported by the fact that extremely pure metals are much more ductile thi (1
those of slightly lower purity. Under continued applied stress, the cavitie coalesqe to form a crack in the center of the sample. The crack proceei outward toward the surface of the sample in a direction perpendicular,
the applied stress. Completion of the fracture occurs very rapidly along! surface that makes an angle of approximately 45° with the tensile axisl
i
.0 '
:
-
:V
4 a
g 3-15 .
A mechanism of crack formation, (a) I
(locations piling up at a barrier, (b) three piled itoeations form an incipient crack within the c Froni A. G. Guy, "Elements of Physical Metallur Son-Wesley Publishing Company, Inc., Reac
P».)
..
PLASTIC DEFORMATION
iY
121
The final stage leaves a circular lip on one half of the sample and a bevel on the surface of the other half. Thus one half has the appearance of a shallow cup, and the other half resembles a cone with a flattened top, giving rise to the term cup-and-cone fracture.
| 3-8 Slip, Twinning, and Fracture The amount of deformation that can occur before fracture is determined by the relative values of the stresses required
for slip, twinning, and cleavage. There is a critical resolved shear stress for slip which is increased by alloying, decreasing temperature, and prior deformation. There is a critical resolved shear stress for twinning which
is also increased by prior deformation. There is also a critical normal stress
for cleavage on a particular plane which is not sensitive to prior deformation and temperature. When a stress is applied to a crystal, which process
takes place depends upon which critical stress is exceeded first. If the critical resolved shear stress for slip or twinning is reached first, the crystal will slip or twin and show some ductility. If, however, the critical normal stress is reached first, the crystal will cleave along the plane concerned with little or no plastic deformation.
\
3-9 i.
arbon
;
Polycrystalline Material The preceding discussion described plastic deformation in single crystals. Commercial material, however, is always made up of polycrystalline grains, whose crystal axes are oriented at random. When a polycrystalline material is subjected to stress, slip starts first in those grains in whieh the slip system is most favorably situated with respect
"
crack to propagate even when the body
to the applied stress. Since contact at the grain boundaries must be main-
applied tensile stress.
ay exist in the metal due to the previous However, even an initially sound mate-
*
mic scale. As shown in Fig 3-15a, the r which might be of grain boundary or ilcrocrack. Another method (Fig 3-15b) nbine into a single dislocation. From the .
Dislocation
:
.
.
Dislocations-
source
.
-(>
1
-
J
.
.
J. J.
that any method that will increase the
'
reduce the possibility of brittle fracture nsiderable plastic deformation prior to .
1 i-: Crack
/stalline ductile materials occurs with a with the formation of a neck in a tensile
Barrier
ture begins by the formation of cavities .
.
.
.
..
..
i. In most commercial metals these in,
metallic inclusions. This belief is sup-
I i
jure metals are much more ductile than | l- pr continued applied stress the cavities ;'-hter of the sample. The crack proceeds ,
,
sample in a direction perpendicular to
the fracture occurs very rapidly along a pproximately 45° with the tensile axis.
if)
Fig. 3-15 A mechanism of crack formation, (a) Edge dislocations piling up at a barrier, (b) three piled-up dislocations form an incipient crack within the crystal lattice. (From A. G. Guy, "Elements of Physical Metallurgy," 2d ed., Addison-Wesley Publishing Company, Inc., Reading, Mass., 1959.)
122 INTRODUCTION lb PHYSICAL. METALLURGY
| tained, it may be necessary for more than one slip system to operate. The rotation into the axis of tension brings other grains originally less favor-i ably orienteti, into a position where they can now deform. As deformation; ,
1
and rotation proceed, the individual grains tend to elongate in the direc-1 tion of flow'(Fig. 3-16). After a certain amount of deformation most grainsi ,
will have a?particular crystal plane in the direction of deformation. TheS material noto shows preferred orientation which will result in somewhati different prbperties depending upon the direction of measurement. A fine-grained metal in which the grains are oriented at random wilill ,
,
possess idelhtical properties in all directions, but a metal with preferred! orientation of grains will have directional properties. This may be troubie-i
some-for e'xample, in the deep drawing of sheet metal. Preferred orienta-f
5*
tion is also lof prime importance in the manufacture of steel for electrical
Fig. 3-17 Polycrystalline brass, polished, etched, and
instruments}:because the magnetic properties will be different depending! upon the direction of working. If the deformation is severe the grains may!
deformed slightly in a vise. 100x. (By permission from Brick and A. Phillips, "Structure and Properties of All(
jded., McGraw-Hill Book Company, New York, 1949.)
,
be fragmented or broken. '
Not all thp work done in deformation is dissipated in heat; part of it is] zontal when they cross the twinnir
stored in the crystal as an increase in internal energy. Since the crystals
axes of adjacent grains are randomly oriented, the slip planes and twinning||
direction on the other side. Since t
planes must change direction in going from grain to grain (Figs. 3-8 ar 3-13) This means that more work is done at the grain boundaries and morel internal energy will exist at those points.
either side of the twinning band, tl curring in the same grain.
.
I
When a crystal deforms, there is This distortion is greatest on the s creases with increasing deformati in resistance to further deformati
Figure 3-17 shows the microstructure of polycrystalline brass after being -
i
deformed slightly in a vise. Notice that the thin, parallel slip lines change! '
direction at, the grain boundaries. The grain in the lower portion of thei picture is ihteresting in that it illustrates both slip and twinning in the; same grain.!; The slip lines, running vertically and to the right, become hori-
-
hardening or work hardening. Or deformation is that the stress req -
-
required to continue deformation i
tortion of the lattice structure, the
\ 0
(such as grain boundaries and for
.
Kit -
i
.
m
5 1
tions on intersecting slip planes formation.
-
In reality, crystals usually conta dislocation lines, as well as othe lattice. When the dislocations b«
Hi
IP !i
other parts of the network, or to c
i <
5
(a) * .
--
* .
v;-i
Fig. 3-16
The microstructure of a nickel-copper alloy,
Monel, showing: (a) th equiaxed grains in the unworked condition, hardness BHN 125, etched in cyanide persulfate,
lOOx, and (b) cold-dravSfn condition showing the grains
elongated in the directiion of drawing, hardness BHN 225, '
etched in Carapellas rdagent, 100 x. (The International Nickel
Company.)
,
i|
I p
!
X
chored, the active slip planes can fact, the dislocations in the plane ease with which a dislocation mo'
of the ductility of the material, i
harder by putting various obstacl dislocations pile up at grain bou hardened by reducing the size of Alloying introduces foreign ato
i
' .
PLASTIC DEFORMATION
123
imt'
are than one slip system to operate. The irings other grains, originally less favor,
i/./e they can now deform. As deformation iial grains tend to elongate in the directain amount of deformation, most grains e in the direction of deformation.
.
mm,
The
[entation, which will result in somewhat
il
on the direction of measurement.
„
7!
:
the grains are oriented at random will
directions, but a metal with preferred
'ctional properties. This may be trouble.vy{ awing of sheet metal Preferred orienta.
,
irt
.
n the manufacture of steel for electrical
i
properties will be different depending ie deformation is severe the grains may
;
,
lation is dissipated in heat; part of it is : V>e in internal energy. Since the crystal ily oriented, the slip planes and twinning
.
:
.
icing from grain to grain (Figs. 3'8 and s done at the grain boundaries and more
=
)oints.
ture of polycrystalline brass after being that the thin, parallel slip lines change The grain in the lower portion of the jstrates both slip and twinning in the vertically and to the right, become hori-
r ; -
11
Fig. 3-17 Polycrystalline brass, polished etched, and then deformed slightly in a vise. 100x. (By permission from R. M. Brick and A. Phillips, "Structure and Properties of Alloys," 2d ed., McGraw-Hill Book Company, New York, 1949.) ,
zontal when they cross the twinning band and then resume their original direction on the other side. Since the slip lines have the same direction on
'
either side of the twinning band, this indicates that the deformation is occurring in the same grain. When a crystal deforms, there is some distortion of the lattice structure.
This distortion is greatest on the slip planes and'grain boundaries and increases with increasing deformation. This is manifested by an increase
in resistance to further deformation.
The material is undergoing strain
hardening or work hardening. One of the remarkable features of plastic deformation is that the stress required to initiate slip is lower than that required to continue deformation on subsequent planes. Aside from distortion of the lattice structure, the pile-up of dislocations against obstacles (such as grain boundaries and foreign atoms) and the locking of dislocations on intersecting slip planes increase the resistance to further deformation.
In reality, crystals usually contain complex networks of interconnected
if ;o
V J : :od -
-
-
,
.
:
,
v
;
;Hfate. 25,
Nickel
dislocation lines, as well as other defects and impurities in the crystal lattice. When the dislocations begin to move, their ends remain tied to m
other parts of the network, or to other defects. Because the ends are anchored, the active slip planes can never get rid of their slip dislocations; in
fact, the dislocations in the plane multiply when the plane slips. Since the ease with which a dislocation moves across the slip plane is an indication of the ductility of the material it suggests that materials may be made harder by putting various obstacles in the way of the dislocations. Since dislocations pile up at grain boundaries, metals can, to some extent, be hardened by reducing the size of the grains. Alloying introduces foreign atoms that distort the crystal locally around ,
124
INTRODUCTION TO PHYSICAL METALLURGY
\
1
themse'ves, and these 'oca' d'stort'ons offer res'stance to the movement o|
I
a nearby d,s,ocat,on. ,f the a,,oy atoms are gathered together ,n c,umpsj their effect is enhanced, and this can be accorpplished, as will be explainei later in the section on age-hardening, by heat treatment. In the hardenini
Tens,,e strength
that is produced by various processes of plastic working, such as hammer|
ing or rctlling, the obstacles are paradoxically the dislocations themselves|
I
Yield strength (0.2%)
to
When the number of dislocations in the worked metal becomes largi I
-
enoughjhose moving along intersecting slip planes obstruct one another's]
-
movement, an effect readily appreciated by anyone who has been held up au a road junction in dense traffic.
Hi-1'
J-
It is important to remember that whenever there is distortion of the lattice
0
structural, whether it is a result of plastic deformation, heat treatment,
Percent cold working -*-
alloying,!'there will be an increase in the strength and hardness of the;
,
IpFifl. 3-18 Effect of cold working on tensile and yii
material. 3-10
Strength of copper.
Effect of Cold Working on Properties
A material is considered to be cold-*
worked if its grains are in a distorted condition after plastic deformation is|
the first 10 percent reduction an
v pompletesd. All the properties of a metal that are dependent on the lattica structure are affected by plastic deformation or cold working. Tensile! strength, yield strength, and hardness are increased, while ductility, as]
decreases electrical conductivity
Distortion of the lattice structi
is appreciable in alloys (Fig. ZAi
represented by percent elongation, is decreased (Table 3-2). Although both| strength and hardness increase, the rate of change is not the same. Hard-ii ness generally increases most rapidly in the first 10 percent reduction, TABLE 3-2
\
REDUCTION BY COLD
TENSILE STRENGTH,
ELONGATION,
HARDNESS
ROLLING, PERCENT
PSI
% IN 2 IN.
ROCKWELL Xt;
43,000
70
12
48,000
52
62
53,000
35
83
30
60,000
20
89
40
70,900
12
94
50
80,000
8
97
60
90,000
6
100
f
0
10
I
20
0
Effect of Plastic Deformation on the Tensile Properties of 70:30 Brass*
10
\
20 Q
M. Brick and A. Phillips, "Structure and Properties of Alloys," McGraw-Hill Book Company, New York, 1942. t Rockwell x = Vu-in, ball indenter, 75 kg load.
. From R
\
_
.
whereas-the tensile strength increases more or less linearly. The yieldj strength iincreases more rapidly than the tensile strength, so that, as th e amount dif plastic deformation is increased, the gap between the yield and tensile strengths decreases (Fig. 3-18). This is important in certain formingl operations where appreciable deformation is required. In drawing, forex' "
.:-v
ample, tlje load must be above the yield point to obtain appreciable de-J
.
.
Iff.-:-.
formation but below the tensile strength to avoid failure. narrow, very close control of the load is required.
If the gap is
Ductility follows a path opposite to that of hardness, a large decrease inl \ i
\ I
\ i
i
I
30 0
20
I
40 Percent cold
,
Pg. 3-19 Effect of cold working on the electrical c
1 1 pure aluminum; pure copper; Cu + 30 percent 1 percent Si; Cu + 5 percent and 7.5 percent Al; Cu
l ercent, 20 percent, and 30 percent Zn. (By permi: "
.
.
Brick and A. Phillips, "Structure and Proper
Alloys," 2d ed., McGraw-Hill Book Company, New
: 1949.1
3
1
PLASTIC DEFORMATION
125
ions offer resistance to the movement of |
: :
itoms are gathered together in clumps, Jpn be accomplished, as will be explained '
"
-
-
1
Tensile strength
ng, by heat treatment. In the hardening-'f ses of plastic working, such as hammer adoxically the dislocations
a>
Yield strength
themselves.
(0.2%)
in the worked metal becomes large
cting slip planes obstruct one another's v L
,
.
ated by anyone who has been held upat MBK
1>
henever there is distortion of the lattice J Ppolastic deformation, heat treatment, or
0
Percent cold working -
in the Strength-and hardness Of thejjBp Fig. 3-18
Effect of cold working on tensile and yield
.
i
U strength of copper,
A material is considered to be cold-;J
I
d condition after plastic deformation is 4
netai that are dependent on the lattice - aK-
the first 10 percent reduction and then a decrease at a slower rate. Distortion of the lattice structure hinders the passage of electrons and decreases electrical conductivity. This effect is slight in pure metals but is appreciable in alloys (Fig. 3-19).
-
'
.
Reformation or cold working. Tensile
.
less are increased, while ductility, as .- p 5 decreased (Table 3-2). Although both-JBpV . rate of change is not the same. Hard- B "
jdly in the first 10 percent reduction/.-mo '
e Properties of 70:30 H
Brass*
Pure A
fe
-
HARDNESS
'
\
-
:
;y
: :
"
:
s
.
70
12
52
62
35
83
20
89
12
94
-
8
97
6
100
30% Ni
\
'
% IN 2 IN.
Cu
\
fafe ROCKWELL Xt R;
ELONGATION,
,
0
10% Zn \
X3
3% Si
10 \
'
m Ml;
in
20% Zn
-
5% At
-
i
S 20
\
30% Zn
Hoys," McGraw-Hill Book Company, New York. 1942. X
1
ses more or less linearly. The yield i the tensile strength so that, as the eased the gap between the yield and ,
S
11 it
s
7 5% AI ,
30 0
,
40
20
60
Percent cold reduction This is important in certain forming tation is required. In drawing for ex- W£- FI9.3-19 Effect of cold working on the electrical conductivity
4
.
.
.
.
...
Ll
,
. , i
1 / of pure aluminum; pure copper; Cu + 30 percent NI; Cu + 3
yield point to obtain appreciable de- ... : per;ent Si. Cu + 5 percent and 7 5 percent A,; cu + io .
mgth to avoid failure. If the gap is
I is required.
that of hardness, a large decrease in
i V percent
20 percent, and 30 percent Zn. (By permission from Structure and Properties of Alloys," 2d ed., McGraw-Hill Book Company, New York,
* |: R' M- Brick an
1949.)
1
-
. -: T -
PI
,
"
80
100
; .
'
.
if
f
,
1
.
12(>
INTRODUCTION TO PHYSICAL METALLURGY
The increase in internal energy, particularly at the grain boundaries |
Mondolfo, L. F., and O. Zmeskal:
,
'
makes the :
-
.
V::;'-
material more susceptible to intergranular corrosion (see Fig f
Company, New York, 1955.
.
12-4), thereby reducing its corrosion resistance.
Known as stress cor-'t
rosion, this'is an acceleration of corrosion in certain environments duetoi
pany. Inc., Reading, Mass., 1959.
heat treatment after cold working and before placing the material in|
Inc., Reading, Mass., 1970. Weertman, J., and J. R. Weertman: Company, New York, 1964.
:
I
J
-1 Describe an experiment to determine the increase in internal energy of a crystal
as a result of deformation. | 3-2 The movement in slip is sometimes described as analogous to simple gliding ! '
.
Why Is this a poor analogy?
Is it possible to have both slip and twinning occur In the same grain? Explain
Jf |
.
3 4 How may one distinguish between slip and twinning if the width of the twin | | band Is of thje same order as a slip line? ,
3
-5 What is the effect of the rate of deformation on the mechanical properties? f How would a difference in grain size affect the change in mechanical properties;!
3-6
due to deformation?
3
-7 Which properties would be affected by preferred orientation and why?
3
J
In Fig 3-19, why does the addition of 30 percent nickel to copper have lessl
3-8
.
effect on electrical conductivity than the addition of 5 percent aluminum to copper?| 3-9 Using Eqs (3-1) and (3-2), plot a curve showing the variation of the shear stress? .
i
and the normal stress as 0 varies from 0 to 90°. Assume F is constant. 3-10
|
The critical resolved shear stress for slip in copper is 142 psi
Using Eq. (3-3)| and assumiijig k to remain constant at 10 plot a curve showing the change in the;! .
°
,
axial load F with change in 0. 3-11
Explain the reason for the increase in ductility of most metals as the tempera-
ture is raised. 3-12 Why is there a greater tendency for brittle failure straining? 3-13
to occur at high rates ot|
What is the difference between brittle fracture and ductile fracture?
REFERENCES
I
.
:
j
Barrett, C. S.: "Structure of Metals," 2d ed., McGraw-Hill Book Company, New York| .
1952.
Boas, W.: "An Introduction to the Physics of Metals and Alloys," John Wiley & Sons,
Inc., New Vork, 1947.
|
Brick, R. M.!, and A. Phillips: "Structure and Properties of Alloys," 2d ed., McGraw-| Hill Book 'Company, New York, 1949. V * '.",*"
"
Cottrell, A. H.: "Dislocations and Plastic Flow in Crystals," Oxford University Press.S air Lawn1
,
N.J., 1956.
uy, A. G.: Elements of Physical Metallurgy," Company, Inc., Reading, Mass., 1959. "
i
-
.
Materials Science For Engim "
El«
Wulff, J.: "The Structure and Propertie
Wulpi, D. J.: "How Components Fail,
3
3-3
"
John Wiley & Sons, Inc., New York, 1
QUESTIONS
-
Ohio, 1951.
Van Vlack, L. H.: "Elements of Material;
m
I
Rogers, B. A.: "The Nature of Metals,
residual stresses resulting from cold working. One of the ways to avoid stress corrosion cracking is to relieve the internal stresses by suitable! service.
E
"
1
. .
\ i
I
-
11 ,
I i
f 2d ed., Addison-Wesley Publishing
'
Ohio, 1967.
PLASTIC DEFORMATION
particularly at the grain boundaries,
'
,
Mondolfo, L. F., and O. Zmeskal: "Engineering Metallurgy,
. ,
yple to intergranular corrosion (see Fig,j|
"
127
McGraw-Hill Book
Company, New York, 1955.
Rogers, B. A.: "The Nature of Metals, American Society for Metals, Metals Park, "
s Sjon resistance. Known as stress cor-||
Ohio, 1951.
irrosion in certain environments due to f »ld working. One of the ways to avoid li
Van Vlack, L. H.; "Elements of Materials Science, Addison-Wesiey Publishing Com"
pany, Inc., Reading, Mass., 1959.
ilieve the internal stresses by suitable |
_:
3 and before placing the material in'il
Materials
Science For Engineers, Addison-Wesiey Publishing Company, "
Inc., Reading, Mass., 1970. Weertman, J., and J. R. Weertman; Elementary Dislocation Theory," The Macmillan Company, New York, 1964. Wulff, J.; "The Structure and Properties of Materials. vol. 3: Mechanical Behavior, "
"
John Wiley & Sons, Inc., New York, 1965. Wulpi, D. J.: "How Components Fail, American Society for Metals, Metals Park, "
;
ie the increase in internal energy of a crystal JL
Ohio, 1967.
I
'
s described as analogous to simple gliding. twinning occur in the same grain? Explain. i slip and twinning if the width of the twin
jormation on the mechanical properties? 3 affect the change in mechanical properties '
'
J by preferred orientation and why?
*
i of 30 percent nickel to copper have less
mb-
yg}
addition of 5 percent aluminum to copper?
ve showing the variation of the shear stress :M£, to 90°. Assume F is constant.
:r
or slip in copper is 142 psi. Using Eq. (3-3) 0°
,
plot a curve showing the change in the
i in ductility of most metals as the temperaor brittle failure to occur at high rates of
,
.
_
it |
I t | |;
f
ttle fracture and ductile fracture?
d
.,
McGraw-Hill Book Company, New York,
s of Metals and Alloys," John Wiley & Sons,
g: ind Properties of Alloys
.
-
,
2d ed., McGraw-
low in Crystals," Oxford University Press
. . .
" ,
-
,
v "
urgy,
2d ed., Addison-Wesiey Publishing
;
si
-
1
.
!
Jv
/
:( :
i
\
ANN HOT
f
f
FULL ANNEALING
1
Introduction In the previous chapt by slip and twinning and the effec metal were studied.
As a resul
strength, and electrical resistance There was also a large increase i planes in the crystal structure we that while most of the energy us< V
in heat, a finite amount was st
energy associated with the lattl The stored energy of cold work i: mately 10 percent of the energj
5
cold-worked state, which is retail
' t
4'2
i
relationship between stored enen purity copper. Full annealing is the process b structure is changed back to on cation of heat. This process is ci usually followed by slow cooling ture. The annealing process ma
recrystallization, and grain growt Recovery This is primarily a lowchanges produced do not cause { The principal effect of recovery 5 due to cold working. This is she the rate of decrease in residual st
and drops off at longer times, fi stress that occurs in a practical tir 1
i
m <
-:
'
i
: :
1 I
'
.
7
-
-
" .
" -"
mat'.mttimmmmm
i
-
/ v.-
ANNEALING AND HOT WORKING
m
FULL ANNEALING
fe
'" .
In the previous chapter, the mechanism of plastic deformation by slip and twinning and the effect of cold working on the properties of the metal were studied. As a result of cold working, the hardness, tensile
'
,
strength, and electrical resistance increased, while the ductility decreased.
There was also a large increase in the number of dislocations, and certain
J : .-
usually followed by slow cooling in the furnace from the desired temperai
i
ture. The annealing process may be divided into three stages: recovery recrystallization, and grain growth.
,
4-2 Recovery
This is primarily a low-temperature process, and the property changes produced do not cause appreciable change in the microstructure The principal effect of recovery seems to be the relief of internal stresses due to cold working. This is shown in Fig. 4-2. At a given temperature the rate of decrease in residual strain hardening is fastest at the beginning and drops off at longer times. Also the amount of reduction in residual stress that occurs in a practical time increases with increasing temperature. .
,
,
1
.
planes in the crystal structure were severely distorted. It was emphasized . that while most of the energy used to cold-work the metal was dissipated :. in heat, a finite amount was stored in the crystal structure as internal , energy associated with the lattice defects created by the deformation. The stored energy of cold work is that fraction, usually from 1 to approximately 10 percent of the energy put into a material while producing a cold-worked state, which is retained in the material. Figure 4-1 shows the relationship between stored energy and the amount of deformation in highpurity copper. Full annealing is the process by which the distorted cold-worked lattice structure is changed back to one which is strain-free through the application of heat. This process is carried out entirely in the solid state and Is .
1
'
/
t
INTRODUCTION TO PHYSICAL METALLURGY
130
s
Recrystallizatlon As the upper te iy. minute new crystals appear in tl the same composition and latl grains and are not elongated bi (equiaxed). The new crystals c formed portions of the grain, u;
3
a)
15
V
r: -;
Fraction of energy
i
.
.
stored
IM '
10
o
s
Hit'
.
,
Stored energy
5
;The
£f 75 o
o
*-
1
Recrystallization takes
strain-frge- rains and the growl
_
10
0
20
30
40
worked material.
Percent elongation
,
cluster of atoms from whic
cleus.
9
.
.;. Figure 4-3 shows a typical re
Fig. 4-1 Stored energy of cold work and fraction of the total work of deformation remaining as stored energy for
percent of the material recrysta stent temperature for a fixed coi
11 fe f high-purity copper plotted as functions of tensile elongation. 1 tesyXFrom data of P. Gordon, Trans. AIME, vol. 203, p. 1043,
curve is typical of any process 1 tially, there is an incubation pe
|n order to start the process goi
When the load which has caused plastic deformation in a polycrystallihi
allow the strain-free nuclei to re
material is released, all the elastic deformation does not disappear. Ttttj
mm
a
is due to the different orientation of the crystals, which will not allow so '
ithem to move back
when the load is released. As the temperature];
i
HAcation from the liquid would s
increased, there is some springback of these elastically displaced ato ~
i which relieves most of the internal stress. In some cases there may be
Ipqritical size to form a stable c than critical size, would redissc
.
slight amount of plastic flow, which may result in a slight increase in hai ness and strength. Electrical conductivity is also increased apprecial
i
&
during the recovery stage. Since the mechanical properties of the metal are essentially unchange
no simple way to recreate the
crystallization embryo cannc
jTierely wait for additional ener
"
i
the principal application of heating in the recovery range is in strei 1
jtant to realize that the growth During the study of crystallizatk
;
|structure. Eventually the critic
||ation begins. The incubatio
relieving cold-worked alloys to prevent streps-corrosion cracking or
growth of the embryos. Exactly how recrystallization
minimize the distortion produced by residual stresses. Commercial this low-temperature treatment in the recovery range is known as sfres.
however, some idea of the pre
re//e annealing. too
m
I
if
5
o
a
v
-
.
Incubation
.
0
Recovery time -*
Fig. 4-2 Residual strain hardening vs. recovery time at three constant annealing temperatures.
Time of annealing -
ilf-S:1 Atypical recrystallization curve at cons fere. '
i
1!
:
i
i
I
period
ANNEALING AND HOT WORKING
131 v
4*3
Recrystallization As the upper temperature of the recovery range is reached 4 minute new crystals appear in the microstructure. These new crystals have \ l the same composition and lattice structure as the original underformed grains and are not elongated but are approximately uniform in dimensions (equiaxed). The new crystals generally appear at the most drastically de,
'
St
formed portions of the grain, usually the grain boundaries and slip planes.
The cluster of atoms from which the new grains are formed is called a nu- \ cleus.
Recrystallization takes place by a combination of nucleation of 1
strain-free grains and the growth of these nuclei to absorb the entire cold- \ worked material.
I
#1
F,gure 4-3 shows a typ,ca, recrysta,,,zat,on curve. Th,s ,s a p,ot of the percent of the mater,a, recrysta,,,zed versus the t,me of annea,,ng at constant temperature for a fixed composition and degree of cold working. This
,
tion.
curve is typical of any process that occurs by nucleation and growth. Initially, there is an incubation period in which enough energy Is developed
in order to start the process going. In this case, the incubation period is to allow the strain-free nuclei to reach a visible microscopic size.vlt is important to realize that the growth of recrystallized embryos is irreversible. During the study of crystallization in Chap. 2, it was pointed out that solidi-
astic deformation in a polycrystailine
:
Reformation does not disappear. This ''-Z he crystals which will not allow some ,
i
ld is released. As the temperature is
fication from the liquid would start when a group of atoms had reached a
: of these elastically displaced atoms tress. In some cases there may be a
:
critical size to form a stable cluster. Embryos, that is, clusters of less than critical size, would redissolve or disappear. However, since there is
nay result in a slight increase in hard-
no simple way to recreate the distorted, dislocation-filled structure, the recrystallization embryo cannot redissolve. Therefore, these embryos
ictivity is also increased appreciably
merely wait for additional energy to attract more atoms into their lattice structure. Eventually the critical size is exceeded, and visible recrystallization begins. The incubation period corresponds to the irreversible
the metal are essentially unchanged in the recovery range is in stress[ent stress-corrosion cracking or to jy residual stresses. Commercially, ,
'
growth of the embryos.
Exactly how recrystallization takes place is not yet clearly understood; however, some idea of the process may be obtained by examining it in
'
.;Z}h
recovery range is known as sfress-
.i too
.-J
In
i a
Incubation 0
period
Time of annealing ree
(Jig. 4-3
A typical recrystallization curve at constant
temperature.
s
'
:
Z
: ~
..
S51i 132
%
INTRODUCTION TO PHYSICAL METALLURGY
11
it was Emphasized that the slip planes and grain boundaries were localizr
workedjriaterial completely fernperature f several metals
points of high internal energy as a result of the pile-up of dislocationl Because of the nature of strain hardening, it is not possible for the disloci tion or the atoms to move back to form a strain-free lattice from the di$«
Notice that very pure metal; tures as compared with impui recrystallization temperatures
torted: lattice.
these metals cannot be cold-
terms of the energy of the lattice. In the discussion of pfastic deformatioif '
some atoms
,
A simplified analogy is shown in Fig. 4-4.
Consider \h
crystallize spontaneously, refc The greater the amount of f
at the grain boundaries or slip planes, have been pushed ups
an energy hill to a value of E, above the internal energy of atoms in the un|
for the start of recrystallizatic
deformed lattice. The energy required to overcome the rigidity of the disl
i
torted lattice is equal to £2. The atoms cannot reach the energy of th| strain-free crystal by the same path they went up the hill; instead the|
tortion and more internal enei
must get over the top, from which they are able to roll down easily. This| difference in energy, £2 - £,, is supplied by heat. When the temperature! is reached at which these localized areas have an energy content equal toj
ture. The recrystallization pro
Increasing the annealing tii
,
perature than to variations in ti time and temperature on the t( is shown in Fig. 4-6. Recrysi tensile strength. A tensile strf
E2, they give up part of their energy as heat of recrystallization and fornil
nuclei of new strain-free grains. Part of this heat of recrystallization ls|
390oF.
structure of the strain-free grains, initiating grain growth. The numberarii energy content of these high-energy points depend to a large extent off
For equal amounts of cold-\A into initially fine-grained met
the amount of prior deformation, the number increasing with increasini
Therefore, the finer the initis
f
c
ing for 12 hr at 300oF, 6 hr at:
absorbed by surrounding atoms so that they have sufficient energy tp overcome the rigidity of the distorted lattice and be attracted into the lattt
deformation. 4-4
Recrystallization Temperature The term recrystallization temperature do»
TABLE 4-1 Appro; for several metals ai
not refer to a definite temperature below which recrystallization will no
occur but refers to the approximate temperature at which a highly cold|
it
MATERIAL
Energy required to overcome rigidity
/
Copper (99.999%)
oi the distorted lattice
O
.
Copper, 5% zinc
s
Copper, 5% alumi Copper, 2% beryll Aluminum (99.99E Aluminum (99.0% Aluminum alloys Nickel (99.99%)
energy supplied by heat £2
m
m '
m :k
ft
internal energy due to deformation
/
Monel metal
Iron (electrolytic) .
m mm
Low-carbon steel
Magnesium (99.9
Magnesium alloy Zinc Tin
P:
I
Lead
.
1
Energy of a strain-free crystal
""
.
*
By permission from /
2d ed., Addison-Wesle
Fig. 4-4
Schematic representation of recrystallization.
1 1
m
turns*
I I
.i
1959.
1 ANNEALING AND HOT WORKING
the discussion of plastic deformation;
result of the pile-up of dislocations ning, it is not possible for the disloca| s
'
i)rm a strain-free lattice from the dis-)
recrystallization temperatures below room temperature. This means that
!
is shown in Fig. 4-4. Consider that:
these metals cannot be cold-worked at room temperature since they re-
or slip planes, have been pushed up te internal energy of atoms in the un-
b to overcome the rigidity of the dis-
crystallize spontaneously, reforming a strain-free lattice structure. The greater the amount of prior deformation, the lower|the temperature for the start of recrystallization (Fig. 4-5), since there will be greater dis-
)ms cannot reach the energy of the.
tortion and more internal energy left.
lied by heat. When the temperature
i
If:
that they have sufficient energy to iting grain growth. The number and
ing for 12 hr at 300"F, 6 hr at 320"F, 2 nr at 340 , 1 hr at 370oF,«r Va hr at J ;: :/i '
into initially fine-grained metal than into initially coarse-grained metal.
Therefore, the finer the initial grain size the lower the: recrystallization
recrystallization temperature does low which recrystallization will not amperature at which a highly cold-
J
% '
"
TABLE 4-1
for several metals and alloys*
1
pK
1
TEMP, "F
Copper (99.999%) Copper, 5% zinc Copper, 5% aluminum Copper, 2% beryllium Aluminum (99.999%) Aluminum (99.0%+) Aluminum alloys Nickel (99.99%)
250 600 550 700 175 550 600 700
\
.
3
'
4
RECRYSTALLIZATION
ly
""
Approximate recrystallization temperatures
MATERIAL
f2
'
For equal amounts of cold-working, more strain hardening is introduced ,; |
5
.
;
tensile strength. A tensile strength of 40,000 psi may be ojbtalned by heat- ||
39o°f.
tttice and be attracted Into the lattice § 4points depend to a large extent on number increasing with increasing
Increasing the annealing time decreases the recrystallization temperature. The recrystallization process is far more sensitive toi changes in temperature than to variations in time at constant temperature.; The inf luence of
"
t of this heat of recrystallization Is .
i
is shown in Fig. 4-6. Recrystallization is indicated by the sharp drop in ...
is heat of recrystallization and form V
1
time and temperature on the tensile strength of highly cold-worked copper
eas have an energy content equal to il
r
j
Notice that very pure metals seem to have low recrystallization tempera-
ay are able to roll down easily. This
.
1
tures as compared with impure metals and alloys. Zinc, tin, and lead have ,;?|
yjthey went up the hill; instead, they '
i
worked material completely recrystallizes in 1 hr. The recrystallization temperature of several metals and alloys js listed in Table 4-1.
and grain boundaries were localized
"
I
133
Monel metal
1100
Iron (electrolytic)
750
Low-carbon steel
1000
Magnesium (99.99%) Magnesium alloys
150 450
Zinc Tin
50 25
Lead
25
1
\
(I "
By permission from A. G Guy, "Elements of Phyrical Metallurgy, 2d ed., Addison-Wesley Publishing Company, Inc., Reading, Mass.,
*
.
.
1959.
ft-
1
134
ft
INTRODUCTION TO PHYSICAL METALLURGY
700 24 U
.
20
600
i
"
O '
S 16
s
I
500
2
0}
-
Critical amount of cold - work
5 400 4
.2 300
0 0 .
_
10
20
30
40
50
L
_
60
7(
Percent degree of deformation
a
9-ZOO
4-7 Effect of cold working on grain size dev low-carbon steel after annealing at 1740° F. (Fn Wendbook, 1948 ed., American Society for Metal k, Ohio,) "
100 0
20
.
,
j
100
Fig. 4-5 Effect of prior deformation on the temperature for
'
}!
40 60 80 Percent prior deformdlion
'
the startvof recrystallization of copper
necessary lor recrystal I ization. approximately 7 percent is requ
.
I?
[PPerature. By the same reasoning, the lower the temperature of col*
I ..
workftig, the greater the amount of strain introduced effectively decreases
r
,
X ing the recrystallization temperature for a given annealing time. A certain minimum amount of cold working (usually 2 to 8 percent) ij
1
This is known as the critical
smaller than this, the number small.
»5 Grain Growth
Large grains have
is associated with the reductioi 70
I
4 9 C O £
60
fore, under ideal conditions, th
I
One-half hour One hour Two hours Six hours Twelve hours
as a single crystal. This is the this force is the rigidity of the rigidity of the lattice decreases At any given temperature there effects are in equilibrium (Fig.
F Twenty-iour hours
o
S
It is therefore theoretically pi a specimen for a long time higl
50
8
4
40
1 30 200
30C
400
500
600
Temperoture °F ,
Fig. 4-6 Effect of time and temperature on annealing (From .
"
Metals Handbook," 1948 ed American Society for Metals ,
.
Metals Park, dbio ) r .
,
Temperature -»-
£ fig 4-8 Effect of temperature on recrystallizec
\
1 ANNEALING AND HOT WORKING
135
i
j
V
20-
I
s 16
i
c
>:
!0
=0
i:
e-ce * degree :! ee'c-5"s-
m
Fig. 4-7
"0 .
Effect of cold working on grain size developed in i
'
a low-carbon steel after annealing at 1740' F, (From "Metals Handbook.' 1948 ed American Society for Metals. Metals ..
i
Park, Ohio ) .
Ii
r
!
the lower the temperature of cold Jin introduced, effectively decreasV a given annealing time.
r
necessary for recrystallization. In Fig. 4-7, It is seen that a deformation of approximately 7 percent is required before any change in grain size occurs. This is known as the critical deformation. At degrees of deformation smaller than this, the number of recrystallization nuclei becomes very
.
'
,
V Working (usually 2 to 8 percent) is
small.
r
4-5
Grain Growth
Large grains have lower free energy than small grains
.
This
is associated with the reduction of the amount of grain boundary. There-
i
i fore, under ideal conditions, the lowest energy state for a metal would be ; % as a single crystal. This is the driving force for grain growth. Opposing this force is the rigidity of the lattice. As the temperature increases, the
rigidity of the lattice decreases and the rate of grain growth is more rapid. \
'
At any given temperature there is a maximum grain size at which these two effects are in equilibrium (Fig. 4-8). It is therefore theoretically possible to grow very large grains by holding a specimen for a long time high in the grain-growth region. The very large
!
I
Temperature-*-
Fig. 4-8
Effect of temperature on recrystall'ized grain size
3
r
; .
i;
$ 1
t
m
I
i
! 136
INTRODUCTION TO PHYSICAL METALLURGY :
grains shown in Fig. 4-9 were obtained by this method The specimen wasl
along the slip lines and at the
.
held at a temperature just under the melting point of this alloy Notice thai
internal energy. After 8 min a
.
W Mr
k fi-ii
i i
some melting has occurred in the lower left corner because of temperaturS fluctuation in the furnace 4-6 Grain Size Since annealing
.| tion is just about complete,
Jl
.
_
involves nucleation and grain growth factorll ,
structure.
'
m .;!
I
in coarse-grained material
.
The factors that govern the final recrystailizedl
due to previous straining.
fe
Degree of Prior Deformation This is the most important factor
.
Annealing Temperature The
Increas-|
ing the amount of prior deformation favors nucleation and decreases the| final grain size (see Fig 4-7). It is interesting to note that, at the cr deformation the grains will grow to a very large size upon annealing Thef
lization temperature, the finer Heating Time The shorter tin the finer the final grain size.
:
MJ':
.
...t
grain growth and resulting in
,
.
formation of large grains during recrystallization with the minimum formation is due to the very few recrystallization nuclei that are formed!
ill
...
Beyond this point,
graphs d to g) merely serves to of annealing twin bands, which They arise during annealing b;
that favor rapid nucleation and slow growth will result in fine-grained ma|| terial, 'and those which favor slow nucleation and rapid growth will result| grain size are:
tf
H- '
Insoluble Impurities The gre
tion of insoluble impurities tl-
during;the time available for recrystallization If the deformation is care-l
increase nucleation but act £
fully controlled at the critical amount
shown in Table 4-1 that addin
.
subsequent annealing will result inl very large grains or single crystals This is the basis for the strain-anneail ,
.
method of producing single crystals tion
' ,
.
With increasing degrees of defortna||
an increasing number of points of high stress or high energy
present, leading to recrystallization from a greater number of nuclei and| finally to a greater number of grains and thus to a continually smallerl grain size. Ji ,
,
j|p;'-;- or soluble impurities raised t WiP j impurities, such as Cu20 in cc .
'
\
lization temperature but decre
fetj"' £
effect is used commercially tc
metals.
.
fe '
The amount of strain harder
ilfc tion increases as the grain siz<
Time at Temperature Increasing the time at any temperature above th|| W recrystallization temperature favors grain growth and increases the finai| |pj! material are giveruthe same grain size. The progress of recrystallization of cold-worked brass is shownl
|6$ behavior will be very similar. ml;. The rate of cooling from t
in the group of photomicrographs of Fig 4-10. The first one (a) shows the! .
|i effecton final grain size. Thh
alloy after the rolling operation in which the reduction was 33 percent|
m
M "Has been heated far in the gi
There are numerous slip lines and several dark twinning bands The maining samples were reheated at 1075DF in a lead bath for increasinc periods of time. The second one (b) shows new grains beginning to
m
.
II;' slow cooling the material m It-growth, and some coarsening
JEffect on Properties Since full free lattice structure, it is esse '
7v produced by plastic deforms
flpvery nearly to its original prop r
.
:
.
,
; ness and strength decrease,
jj irt properties is shown scher
.
Table 4-2.
Working
r
Hot working is usually descri lization temperature. The al
Wk' Rg- 4-9 Large grains in a titanium-vanadium alloy formed fe- by holding for a long time high in the grain-growth region. .ST
*
Magnification 2x,;.
Wtj | ,
m
ii i
.
i
i
5 !
.account the rate of working.
i
ANNEALING AND HOT WORKING
ed by this method The specimen was)
along the slip lines and at the grain boundaries, which are points of high internal energy. After 8 min at 1075°F (photomicrograph c) recrystalliza-
.
melting point of this alloy Notice that jwer left corner because of temperature .
,
tion is just about comfslete, there being no evidence of the old distorted structure. Beyond this point, increasing the time at 1075 (photomicrographs d to g) merely serves to increase the grain size. Notice the presence of annealing twin bands, which are found in wrought and annealed brasses. They arise during annealing by a change in the normal growth mechanism due to previous straining. Annealing Temperature The lower the temperature above the recrystal lization temperature, the finer the final grain size (see Fig. 4-8).
nucleation and grain growth factors growth will result in fine-grained ma,
|ucleation and rapid growth will result t>rs that govern the final recrystallized s the most important factor Increas.
d:.- avors nucleation and decreases the
S#iteresting to note that
,
Heating Time
at the critical
4 very large size upon annealing
.
137
The shorter the time heating to the annealing temperature,
the finer the final grain size. Slow heating will form few nuclei, favoring
The
grain growth and resulting in coarse grain.
crystallization with the minimum de-
Insoluble Impurities
[rystallization nuclei that are formed
The greater the amount and the finer the distribu-
1
tion of insoluble impurities the finer the final grain size. They not only increase nucleation but act as barriers to the growth ;of grains. It was
illization. If the deformation is care1. subsequent annealing will result in
.
With increasing degrees of deforfriaof high stress or high energy are
shown in Table 4-1 that adding alloying elements (such as zinQ in copper) :;f or soluble impurities raised the recrystallization temperature. Insoluble 1 impurities, such as Cu20 in copper, do not noticeably affect the recrystallization temperature but decrease the recrystallized grain size. This latter
om a greater number of nuclei and
effect is used commercially to obtain fine-grained structures in annealed
.
:|| his is the basis for the strain-anneal "
-
,
s,
and thus to a continually smaller
metals. '
The amount of strain hardening Introduced by a given amount of elonga-' time at any temperature above the jrain growth and increases the final
tion increases as the grain size decreases. If both coarse- and fine-grained material are given-the same amount of strain hardening, their annealing
;
ation of cold-worked brass is shown
behavior will be very similar.
:
fig. 4-10. The first one (a) shows the
The rate of cooling from the annealing temperaturd has a negligible
' jhich the reduction was 33 percent.
effect on final grainjize. This factor will be of interest only iMKermaterial
veral dark twinning bands. The re-
has been heated far in the grain-growth range and slow-cooled. During slow cooling the material may have enough energy to continue grain
075oF in a lead bath' for increasing ;fhows new grains beginning to form
growth, and some coarsening may result. 4-7
Effect on Properties Since full annealing restores the material to a strain free lattice structure, it is essentially a softening process. Property changes produced by plastic deformation are removed, and the material returns very nearly to its original properties. Therefore, during annealing, the hardness and strength decrease, whereas the ductility increases. The change in properties is shown schematically in Fig. 4-11 and for 70-30 brass in -
Table 4-2.
,
HOT WORKING i
Hot working is usually described as working a material above its recrystal-
lization temperature.
The above definition, however, does not take into
account the rate of working.
.
-
I
i 11
138
,NTRODUCT,ON TO PHYS,CAL METALLURGY
14:
mm 1
1
E
i
II...
J .
5
i 1
-
5S
My
'
s
5
i
Ml mm V
if
1
i in
i
as i
(a)
e)
to) Si
i i
Pi
1
5
,
(c)
f
(d)
a
ANNEALING AND HOT WORKING
II
S
139
3
J
111
mi
i
i %1
si Mi
s
ii
it
1
m
m
i;
I
6
(e)
(6)
i
i
ft
"
i Fig. 4-10 Series of photomicrographs illustrating the progress of recrystallization and grain growth of cold-worked brass after annealing at a constant temperature of 1075"F. (a) cold-worked; (6) start of recrystallization; (c) after 8 min.; (d), (e), (f). (g) increasing time at 1075°F. Magnification 40X. (J. E. Burke, General Electric Company.)
rt >
1
3 4
1
.
1
-
v.
i 1 (g)
51 J
i
.
.
.
. ..
.
;
:
140
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 4-2
Annealing of 70-30 brass after 50 percent cold reduction with time constanl
E
30 min* =3
ANNEALING TEMP,
HARDNESS,
TENSILE STRENGTH,
ELONGATI
ROCKWELL Xf
PSI
% IN 2 IN. 1
2
CP o
a a>
None (cold work)
97
80,000
8
300
98
81,000
8
392
100
82,000
8
482
101
82,000
8
572 '
662
\
842 1112
5
-
98
76,000
12
80
60,000
28
i
58
46,000
51
lj
34
44,000
66
14
42,000
70
1292
2
%
M. Brick and A. Phillips, "Structure and Properties of Alloys," 2d ed., McGraw-Hill Book Comps New York, 1949
>. From R
5
.
eS
t Roskwall X = l/is-in. ball penetrator, 75-kg load.
f
j
4 3 Dividing Line between Hot and Cold Working
%
When a material is plastical
deformed, it tends to become harder, but the rate of work hardening dl creases asilthe temperature is increased. When a material is plastical! deformed at an elevated temperature, two opposing effects take place the same time-a hardening effect due to plastic deformation and a softei Ing effect d,ue to recrystallization. For a given rate of working, there mui ii
be some temperature at which these two effects will just balance, if t i material isifworked above this temperature, it is known as hot workini below this {temperature it is known as cold wor/f/nsfQ The hardening | copper unqer slow deformation in a tension test at various temperatim
ft-
is shown irif Fig. 4-12. At about TSCTF, the rate of softening will equal tnj rate of hardening and the material can be continuously deformed withoi .
an increase; in load. If the rate of deformation is increased considerabl| such as in jammer forging, the temperature will have to be increased |i about 1475?F before the two rates will be equal. The effect of working te
perature ori-hardness with varying rate of working is shown schematicall;
in Fig. 4-131 a
The terms hot and cold as applied to working do not have the same sigj nificance that they ordinarily have. For example, lead and tin, whosi recrystallization temperature is below room temperature, may be hot-
worked at rjpom temperature; but steel, with a high recrystallization perature, may be cold-worked at lOOOT. 4-9
Hot Working vb Cold Working .
Most of the metal shapes are produced froiftl
cast ingotsj To manufacture sheet, plate, rod, wire, etc., from this ing 3
the most economical method is by hot working. However, in the case
steel, the hOt-worked material reacts with oxygen as it cools down to room| temperature and forms a characteristic dark oxide coating called scale.
\ i
1
'
A
sssupJDH - qi6u3JtS
i:
2
1 I s5' Icn
0 percent cold reduction with time constant at '
M
,
TENSILE STRENGTH,
ELONGATION
PS!
% IN 2 IN.
*
a
5
cri
o
3
5
80,000
8
81,000
8
82,000
8
82,000
8
76,000
12
60,000
28
46,000
51
44,000
66
42,000
J:
Cl.
s
o
o
C3i CT>
= o
D
o
70 O "
ertios of Alloys, 2d ed.. McGraw-Hill Book Company,
.
1
a
a
£5 .
It
,
rking
Sr-jbut
, '
:
.
When a material is plastically
the rate of work hardening de- j
ied. When a material is plastically two opposing effects take place at - to
plastic deformation and asoften-
5! 2 o '
A
j
a>
a given rate of working, there must wo effects will just balance. If the
>
2 -a o
ature, it is known as hot working;
c
i cold working. \ The hardening of :
snsion test at various temperatures the rate of softening will equal the be continuously deformed without ormation is increased considerably,
I
!
vrature will have to be increased to
1
O m
? 2
8! 11
e equal. The effect of working temof working is shown schematically
si" £ to -
5
in
o
a)
r
9i £2 £ 2
.
1
O working do not have the same sig:
11
o
or example, lead and tin, whose f|
ll
<
room temperature may be hot,
with a high recrystallization tern-
f
O iE O)
o t r
ll
\e metal shapes are produced from te, rod, wire, etc., from this ingot, working. However, in the case of h oxygen as it cools down to room
dark oxide coating called scaie.
;
I
353UpJr)||
f
Hlbusiis
1
c
J
If! l°5 0)
c a
.
.
.
142 INTRODUCTION TO PHYSICAL METALLURGY
70 0F
Melting begins
F
212°F 60
400 0F
Hot working or gram
growth range
45 o
575 0F
3 €30
si
CP
1 in
Q
-
'
;: ::V:1
Recrystallization
QJ
range
155
-
Nc
Cold working a v 0
20 '
;
-
40
60
range
BO
Percent reduction in area -
.S;
Coarse
Fig. 4-12 i Effect of the amount of cold working on the
grain
strength of copper determined by tension tests at various temperatures.
Room temperature
Occasionally, this scale may give difficulty during machining or formi
6
operations.
4-14
Effect of finishing temperature in forging on
n size.
It is nof possible to manufacture hot-worked material to exact size bi cause of dimensional changes that take place during cooling. Cold-worki material, on the other hand, may be held to close tolerances. It is free
3W: n
*
surface-s ale but requires more power for deformation and is thereto) more expensive to produce. Commercially, the initial reductions are a
3
r:
v .
Rapid rate of working
MM
.
1 -
a)
Slow rate of
working
Unworked
Cold
worked
Hot worked
Cold worked
Hot worked
c)
Working lemperalurc
. .,
Fig. 4'13 Sohematicjillustration of the effect of working lomperature on hardhess with varying rate of working.
ii
I
|.15
Effect of grain size on the surface appearar
lirawn 70-30 brass sheet. (H. L. Burghoff, Chase Qopper Company.)
I
ANNEALING AND HOT WORKING
Melting begins
m
143
Forging starts
Hot working or gtoin
!5l
giowlh tanqe Forging ends
11 5 a
Secrystallization
Conlinq
langp
r
Mote: Width ol column
i
represents grain
Cold working
diameter
range Coarse
gram
-
ine
I
grain :
ir Room temperature
lulty during machining or forming
Fig. 414
Effect of finishing temperature in forging on
grain size.
worked material to exact size be-
'
1
r.
;
r
Dlace during cooling. Cold-worked Id to close tolerances.
at
It is free of
fe
„ t
for deformation and is therefore ally, the initial reductions are car-
'
"
1
li-
mm
\
i
if
m
s
i?5
a
Hot
.
'
V/: worked
worked
c)
l
-
.
1
*
Fig. 415 Effect of grain size on the surface appearance of cold-drawn 70-30 brass sheet. (H, L, Burghoff, Chase Brass and Copper Company.)
(of)
m
m
m
'
MO!;
144
INTRODUCTION'TO PHYSICAL METALLURGY
r TABLB;4-3 Recommended Grain Sizes in Brass for Cold-forming Operations*
GRAIf SIZE, MM 0 015
r
expressed, In general, as Rate
TYPE OF COLD-FORMING OPERATION -
-
r
0 025
Slight forming operations Shallow drawing
0 035
For best average surface combined with drawing jJ
.
.
f 0 050 .
I
.
. From "Metals
Ohio.
4-10
Handbook," 1948 ed., table II, p. 879, American Society for Metals
,
ried out with the material at an elevated temperature
,
i
s
and the final redui
American Society for Metals; Metals Hi Brick, R. M., and A. Phillips: Structure Hill Book Company, New York, 1949. "
"
Initially to promote uniformity In the material, and the resulting large grai allow more economical reduction during the early working operation. the material cools and working continues, the grain size will decrease, bi
Byrne, J. E.: "Recovery, Recrystallizatic
.
pany, New York, 1965.
Guy, A. G.: "Elements of Physical Metalli
coming very fine just above the recrystallization temperature. This Is illui trated schematically in Fig. 4-14. Proper control of annealing temperature will approximate the final gn size required for subsequent cold working. Although coarse-grained m;
Inc., Reading, Mass., 1959. Mason, C. W.: "Introductory Physical Metals Park, Ohio, 1947.
Reed-Hill, R. E.: "Physical Metallurgy Pi New York, 1964.
terial has better ductility, the nonunlformlty of deformation from grain
1
Rogers, B. A.: "The Nature of Metals,
Figure 4-15 shows tl
Ohio, 1951.
orange-pjeel surface on coarse-grained material that is subjected
Smith, C. O.: "The Science of Engineer!
"
Cliffs, N.J., 1969.
severe deformation. The choice of grain size is therefore a compromi;
determined by the particular cold-forming operation (Table 4-3).
QUESTIONS '
4-1
i
*
Explam the importance of heating in the recovery range
for some industrii
application .
4-2
Give tyi/o methods of lowering the recrystallization temperature of a given met!
4-3
Assurrje that a tapered piece of copper has been stressed in tension beyond I'
yield pointland then annealed. Explain how the grain size will change along thi
taper. 4-4
-
>
-
-
.
[
Suppose a bullet hole is made in a plate of aluminum How will the grain .
from the hole outward vary if the plate Is annealed? 4 5 Describe two methods of making a single crystal. .
4-6 4-7
Why cfoes the recrystallization temperature vary with different metals? Why 4oes the addition of alloying elements change the recrystallization
tei
perature? \ 4-8
Cite ah industrial application which requires periodic annealing between cold-
working operations. 4-9 Many ;|heat-treating processes, including annealing, involve nucleation and growth. Tfie relation between the rate of the process and the temperature may b«i ,
;
! f
.
j r
[
j ::v ::r .
i
Extrapolate the line obtained in Q at which copper will be 50 percent recr>
4-11
|erences
M
The finishing temperature in hot working will determine the grain si: that is available for further cold working. Higher temperatures are usi
"
Plot the results obtained in Quest
log scale and temperature as the ordina
Metals Park,
tions are done cold to take advantage of both processes.
grain creates a problem in surface appearance.
where A and B are constants, and T is t
For 50 percent recrystallization of coppe for 50 percent recrystallization may bet£ the time for 50 percent recrystallization
Deep drawing Heavy drawing on thick sheet
1
.
0 100
j.
5.
'
WWWS
i
.
mi ANNEALING AND HOT WORKING
145
;l
expressed, in general, as
ass for Cold-forming Operations* FORMING OPERATION
Rate = Ae-BIT
-
-
:: :< -
j
perations
where A and e are constants, and 7 is the absolute temperature in degrees Kelvin.
I surface combined with drawing
For 50 percent recrystallization of copper, A = -\0:2 min"1 and 8 = 15,000. The time for 50 percent recrystallization may be taken as the reciprocal of the rate. Calculate the time for 50 percent recrystallization at 100, 150, 200, 250, and 275°F.
in thick sheet
4-10
Plot the results obtained in Question 4-9 on semilog paper with time on
the
log scale and temperature as the ordinate. This plot should be a straight line. 4-11 Extrapolate the line obtained in Question 4-10 and determine t ie temperature
American Society (or Metals Metals Park, .
at which copper will be 50 percent recrystallized after 15 years.
temperature, and the final reduc-f iboth processes ;;|
REFERENCES
.
ing will determine the grain size I |. Higher temperatures are used !
American Society for Metals: "Metals Handbook," 1948 ed., Metals Park, Ohio. Brick, R. M., and A. Phillips: "Structure and Properties of Alloys," 2d ed., McGraw-
;
rial
,
Hill Book Company, New York, 1949.
and the resulting large grains
Byrne, J. E.: "Recovery, Recrystallization and Grain Growth," The Macmlllan Com-
the early working operation. As
:
r&j, the grain size will decrease, be-;|| v f zation temperature This is illus-":'
.
;e will approximate the iinal grain '
g.
::|
f?
Inc., Reading, Mass., 1959, i Mason, C. W.: "Introductory Physical Metallurgy," American Society for Metals,
.
;
pany. New York, 1965.
Guy, A. G.: "Elements of Physical Metallurgy," Addison-Wesley Publishing Company, Metals Park, Ohio, 1947.
Reed-Hill, R. E.: "Physical Metallurgy Principles," Van Nostrand Reinhold Company,
Although coarse-grained ma-
New York, 1964.
ity of deformation from grain to | larance. Figure 4-15 shows the d material that is subjected to size is therefore a compromise ,
;
|
Rogers, B. A.: "The Nature of Metals," American Society for Metals, Metals Park;,
.
'
Ohio, 1951.
,
-
-
-.
M
;m
Smith, C. O.: "The Science of Engineering Materials," Prentice-Hall, Inc., Englewood Cliffs, N.J., 1969.
I operation (Table 4-3).
recovery range for some industrial
lization temperature of a given metal ! been stressed in tension beyond Its i .
he grain size will change along the f aluminum. How will the grain size led?
jfi: crystal. > vary with different metals? .
ts change the recrystallization tem-
ill
'
''
m Vf:- j9s -
periodic annealing between cold-
i
il
1 annealing, involve nucleation and I
irocess and the temperature may be '
i .
1
i.
18 i
i
f
!
m
CON c OFAL
.
1
I
\
5-1 Introduction
1
metal.
I 1 i '
r.
\
I
%
-
An alloy system contains all the elements combined in all possible of two elements, it is called a binary a//oy sysfem; etc Taking only 45 ( bination of two gives 990 binary sys 14,000 ternary systems. However, ferent alloys are possible. If the cc binary system will yield 100 differen contain many elements, it is appare .
almost infinite.
Alloys may be classified accordin
systems may be classified accordi
\
§ , phase diagram. The basic types of pi
1
i
An alloy is a substanc
ccJmposed of two or more chemic
Classification of Alloys
Alloys may b
sClf the alloy is homogeneous it will
.
mixture it will be a combination o
which is homogeneous and physic?
phase is not determined on an ato each unit lattice cell, but rather o -V
.
which is visible as physically distin
1 1
a phase. For most pure elements th
.
There is, therefore, for pure elemei Some metals are allotropic in the s
S
'
i;
phases. When the metal undergoes goes a phase change since each
' .
i
distinct. V
I I
"
> '
'
(
i. >.
A
'
.
r .
CONSTITUTION .
OF ALLOYS
.1
5 1
Introduction An alloy Is a substance that has metallic properties and is coVnposed of two or more chemical elements, of which at least one is a
\
metal.
'"'
An alloy system contains all the alloys that can be formed by several elements combined in all possible proportions. If the system is made up \
of two elements, it is called a binary alloy system; three elements, a ternary -
i i
,
iif
alloy system; etc. Taking only 45 of the most common metals, any combination of two gives 990 binary systems. Combinations of three give over
14,000 ternary systems. However, in each system, a large number of different alloys are possible. If the composition Is varied by 1 percent, each binary system will yield 100 different alloys. Since commercial alloys often contain many elements, it is apparent that the number of possible alloys is almost Infinite.
Alloys may be classified according to their structure, and complete alloy systems may be classified according to the type of their equilibrium or phase diagram. The basic types of phase diagrams will be studied in Chap. 6. 5-2 Classification of Alloys Alloys may be homogeneous (uniform) or mixtures If the alloy is homogeneous it will consist of a single phase, and if it is a mixture it will be a combination of several phases. A phase is anything which is homogeneous and physically distinct. The uniformity of an alloy .
i
i
m
i i
i
phase is not determined on an atomic scale, such as the composition of each unit lattice cell
but rather on a much larger scale. Any structure which is visible as physically distinct microscopically may be considered a phase. For most pure elements the term phase is synonymous with state. There is, therefore for pure elements, a gaseous liquid, and solid phase. ,
,
,
Some metals are allotropic in the solid state and will have different*6lid
phases. When the metal undergoes a change in crystal structure, it undergoes a phase change since each type of crystal structure is physically distinct.
is m
1 i
1
.
148
INTRODUCTION TQ PHYSICAL METALLURGY
In the soilid state there are three possible phases: (1) pure metal (2) ,
termediate;.alloy phase or compound
,
and (3) solid solution.
Jf an alloV is homogeneous (composed of a single phase) in the
atom of oxygen), and table salt, NaC ,
,
P one atom of chlorine). The atoms tte
|
p Whlch is the smallest unit that has th(
state, it can be only a solid solution pr a compound If the alloy is a n
together in a. definite bond. Variou:
r
.
ture, ifis then composed of any combination of the phases possible itr, solid state/ It may be a mixture of two pure metals or two solid solutioj
discussed in Chap. 2. The bond is g
easily separated. Most students are
,
Chemistry demonstration of the ej at rggsp. current through water it is possihle.
or two compounds or a pure metal and a solid solution and so on. Tl ,
,
mixture may also vary in degree of fineness Pure Metal The characteristics of a pure metal have been discussed in de in an earlier chapter. However one property is worth repeating Unci .
5 3
,
equilibrium
;
atoms. _
conditions, all metals exhibit a definite melting or freezfj
point. The term under equilibrium conditions implies conditions of cl tremely sloW heating and cooling In other words if any change is to occiij sufficient tiime must be allowed for it to take place If a cooling curve )
t
q compnund is fnnTieriJti arid characteristic properties to a \i salt (NaCI). Sodium (Na) is a very ai When
.
.
m
usually stored under kerosene. Chli
,
atom of each combines to give the table salt. Water (H20) is compose at .room temperature, yet the comp What exists then is not the individu or compound. The comBauacLwiJ.1 mechanical, and chemical propertie Most compounds, like pure meta
.
.
plotted for & pure metal, it will show a horizontal line at the melting or freej
.
ing point (Fig. 5-1).
,
5 4
iQterrnediate Alloy phase or Compound
1
Because the reason for referring
tothis type of solid phase as an intermediate alloy phase will be more af
parent during the study of phase diagrams it will be simpler at this poir ,
to call it a Compound.
i'
It is now necessary to obtain some understanding of compounds li general. Most ordinary chemical compounds are combinations of positiw and negative valence elements The various kinds of atoms are combine! in a definite proportion which is expressed by a chemical formula Soi typical examples are water H20 (two atoms of hydrogen combined with o .
,
.
'
_
m
~
"
within narrow limits of temperatur
compoundjs similar to that for a pui
(
.
||||:; . to as a congruent melting phase. Ir intermediate alloy phases are pha
6?
.
intermediate between the two pure
,
tures different from those of the pu
The three most common interme "Tptermetallic
Compounds or Valen yrorlneSTjSfween' chemically dissim
1 Liquid antimony
'
antimony farmed in-the liquid antimony
Liquid phase
t 1170
m Dendrites of solid
Freezing
antimony growing in liquid antimony
paint o
Solid
'
phase
'
ing The rules of chemical valence. j ing (ionic or covalent), their prope usually show poor ductility and po a complex crystal structure. Exar Mg2Pb, Mg2Sn, and Cu2Se.
Nuclei of solid
Ipterstitiai Compounds These cc tiBrTrfi taTs sLrch as scandium (Sc '
(W), and iron (Fe), with hydrogen "
-
rv-
.
Three grains
l
of solid antimony
i Fig. 5-1
Time
solidification of a smalj crucible of liquid antimony i
I
\
have relatively small ktoms that fi of the metairrfiese same five elerr which will be described shortly.
may have a narrow range of coi extremely hard. Examples are T
Time-temperature cooling curve for the
;
The word interstitial means, betvyee
.-
.
:
1
4 CONSTITUTION OF ALLOYS
149
0
sible phases: (1) pure metal (2) ,
.
one atom of chlorine). The atoms that are combined to form the molecule, which is the smallest unit that has the properties of the compound, are held together in a definite bond. Various types of atomic bonding have been
.
P)ed of of a single phase) in
the solid
a compound. If the alloy is a mix
-
lation of the phases possible in the
discussed in Chap. 2. The bond is generally strong, and the atoms are not easily separated. Most students are familiar with the classical high school
pure metals, or two solid solutions d a solid solution and so on. The
,
chemistry demonstration of the ejg trolysis of water. By passing an electdc cijjTRnt th rmoh water LtAa_pr)SRihlft tn sftparate Jtie hydrogen and oxygen
,
less.
„
_
inetal have been discussed in detail
foperty is worth repeating
.
.
.
.
When
Under
_
lose their individual identity .,
largT xtent
it, Labte-
usually stored under kerosene. Chlorine (CI) is a poisonous gas. Yet on
,
;
,
-
,
atom of each combines to give the harmless and important compound,V;y table salt. Water (H20) is composed of elements that are normally gases at room temperature, yet the compound is a liquid at room temperature.
.
-
'
cQmjicujndJ
_
i
salt (NaCI). Sodium (Na) is a very active metal that oxidizes rapidly and is
-
s
a
and characteristic projaerlLesJo
.
er words, if any change is to occur I take place If a cooling curve is : izontal line at the melting or f reez '
4
atoms.
bit a definite melting or freezing M iditions implies conditions of ex :
1
atom of oxygen), and table salt, NaCI $0ine atom of sodium combined with
in
ind (3) solid solution
i
.
Jecause the reason for referring
What exists then is not the individual elements but rather the combination
/diate alloy phase will be more ap
or compound. The comB
-
id-jatUUmve4t6
wn- characteristic physical
mechanical, and chemical properties. Vios\ compoundS like pure metals, also exhibit a definite melting point
ms, it will be'simpler at this point
ir
.
j
limits of temperature. Therefore, the cooling curve for a compound is similar to thatfor a pure metal (see l ig 5-1). It is"then referred
I
understanding of compounds in
; .
within narrow
.
-
mds are combinations of positive
.
jus kinds of atoms are combined ed by a chemical formula Some s of hydrogen combined with one
to as a congruent melting phase. In reference to equilibrium diagrams, the intermediate alloy phases are phases whose chemical compositions are intermediate between the two pure metals and generally have crystal struc- ., tures different from those of the pure metals. The three most common intermediate alloy phases are:
I
.
V
i
Ihtermetaliic Compounds or Valency Compounds These are generally -
"
formed between cfiemKafiy dissimilar metals and are combined by fgHow-
1 -
f
'
ing the rules of chemical valence. Since they gener&lly havp strong bond_
'
I
ing (ionic or covalent), their properties are essentially nonmetallic. They usually show poor ductility and poor electrical conductivity; and may have a complex crystal structure. Examples of valency compounds are CaSe, Mg2Pb, Mg2Sn and Cu2Se. ,
Interstitial Compounds These compounds formed between the transiii oh metals such as scandium (Sc), titanium (Ti), tantalum-(Ta), tungsten '
(W), and iron (Fe), with hydrogen, oxygen, carbon, boron, and nitrogen.
The word interstitial means between the spaces and the latter five eleffients ,
5-
have relatively small atoms that fit into the spaces of the lattice structure of the metal. Th
~
"
same five elements also form interstitial solid solutions, which will be described shortly. The interstitial compounds are metallic, may have a narrow range of composition, high melting points, and are ese
extremely hard. Examples are TiC, T C, Fe4N, FejC, W2C, CrN, and TiH.
m
1
150
INTRODUCTION TO PHYSICAL METALLURGY
s s
%->
-
.
_
v
Many of thiese compounds are useful in hardening steel and in cementeci
than it could dissolve at a given ten
carbide tools.
unsaturated. If it is dissolving the lii If it is dissolving more of the solute
Electron Compounds
-
. .
A study of the equilibrium diagrams of the alley
of copper, gold, silver, iron, and nickel with the metals cadmium
,
ditions, the solution is supersatura
ma
nesium, tin, zinc, and aluminum shows striking similarities. A number TABLE 5:.1
r
ft: 1
condition is an unstable one, and g solution tends to become stable or
ELECTRON-ATOM
RATIO 3:2 (B.C.C.ISTRUCTURE)
RATIO 21:13
RATIO 7:4
(COMPLEX CUBIC)
(C.P.H. STRUCTURE)
i
Ag5Cd8 Cu9AI4 CU3,SN8 AusZn,, Fe5Zn21 NisZnj,,
FeAl
Cu5Sn
.
equilibrium conditions by rapidly co
Examples of Electron Compounds
ELECTRON-ATOM
AgCd AgZn CU3AI AuMg
complished by doing work on the s
ELECTRON-ATOM
the excess solute.
A solid solution is simply
v
AgCda Ag5Al3 AuZria CUaSi FeZn7
SQiu
solution starts, the temperature me
point of the pure solvent. Most solh perature. Figure 5-2 shows the coo taining 50 percent Sb (antimony) a
.
found to exist at or near compositions in each system that have a definii
this cooling curve with the one sh begins to solidify at a temperature antimony (1170oF) and higher tha
ratip of valence electrons to atoms and are therefore called electron col Some examples are given in Table 5-1. For example in the coi pound AgZn, the atom of silver has one valence electror while that of zli has two valence electrons so that the two atoms of the compound
m
_
liquid state than in the solid state.
Ag3Sn
hntermediate phases are formed in these systems with similar lattice struj tiires. Hume-Rothery first pointed out that these intermediate phases aj pounds.
a _
two kinds of atoms combined in on( a considerable difference in the s( solid states of the solution. The s
,
have three valence electrons, or an electron-to-atom ratio of 3:2.
(520oF). The process of solidifica 1;
solution alloy and the liquid solul
In tl
.
compound CugALj, each atom of copper has one valence electron and ea( atom of aluminum three valence electrons, so that the 13 atoms that rn
up the compound have 21 valence electrons, or an electron-tc-atom raff
11
of 21:13. For the purpose of calculation, the atoms of iron and nickel
assumed to have zero valence.
,|
Many electron compounds have properties resembling those of sol] s
i/j
&
solutions, including a wide range of composition, high ductility, and loj hardness. 5/6 Solid Solutions
Liquid [50% 5b 50 /o Bi)
'
Solid nuclei rich in Sb formed m
liquid rich in Bi Liquid solulion
- -- Beginning ol solidilicahonrq Freezing
Any solution is composed of two parts: a solute and a sl
.
3
range
5/660
vent. The solute is the minor part of the solution or the material whichj dissolved, while the solvent constitutes the major portion of the solutioj
"""
Liquidtsolid solulion
Dendrites 01
Solid solulion
It is possible to have solutions involving gases, liquids, or solids as eitH| }
i
%
The arriount of solute that may be dissolved by the solvent is general a function of temperature (with pressure constant) and usually increa
rated, and supersaturated. If the solvent is dissolving less of the solufi
i
.
Three grt
alloy (50
-
1
'
with increasing temperature. There are three possible conditions for a solution: unsaturated, sati
mm
growing 11 rich in 81
i
the solutdior the solvent. The most common solutions involve water astl
solvent, spch as sugar or salt dissolved in water.
[SI
End of solidification
Time
ip.Tlme-temperature cooling curve for the
lllication of a small crucible of 50 percent antirru «rcent bismuth alloy.
CONSTITUTION OF ALLOYS
fful in hardening steel and in cemented
than it could dissolve at a given temperature and pressure, it is said to be unsaturated. If it is dissolving the limiting amount of solute, it is saturated. If It is dissolving more of the solute than it should, under equilibrium con-
the equilibrium diagrams of the alloys "
Inickel with the metals cadmium mag- m hows striking similarities A number ol W
ditions, the solution is supersaturated.
,
'
pounds SON-ATOM
*EX
ELECTRON-ATOM RATIO 7:4
CUBIC)
solution tends to become stable or saturated by rejecting or precipitating the excess solute.
(C.P.H. STRUCTURE)
I 1
A solid solution is simply a salMon-in the-soJid state and conorets-oftwoTTnds of atoms combined in onejypejyf space lattice. There is usually;
AgCd3 Ag5AI3 AuZna CUjSi
'
considerable difference in the solubility of the solute in the liquid and solid states of the solution. The solute is generally more soluble in the; "
a
FeZn,
liquid state than in the solid state. Moreover, when Solidification of the 5
Ag3Sn
solution starts, the temporature may be higher or lower than the freezing!
point of the pure solvent. Most solid solutions solidify over a range in tem-j
hese systems with similar lattice struc-
5ut that these intermediate phases are '
i
ns in each system that have a definite
:
The latter condition may be ac- .
complished by doing work on the solution, such as stirring, or preventing equilibrium conditions by rapidly cooling the solution. The supersaturated condition is an unstable one, and given enough time or a little energy, the
.
£l:13
begins to solidify at a temperature lower than the freezing point pf pure antimony (1170oF) and higher than the freezing point of pure bismuth (520oF). The process of solidification and the composition of the solid solution alloy and the liquid solution during freezing will be explained
Wm Table 5-1. For example, in the comjne valence electro while that of zinc the two atoms of the compound will electron-to-atom ratio of 3:2
In the
.
perature. Figure 5-2 shows the cooling curve for a solid solution alloy containing 50 percent Sb (antimony) and 50 percent Bi (bismuth). Compare
this cooling curve with the one shown in Fig. 5-1. Nbtice that this alloy
and are therefore called electron com-
)er has one valence electron and each :trons,
so that the 1 S atoms that make
lectrons or an electron-to-atom ratio ,
tion, the atoms of iron and nickel are
1 bond nudei ncn
Liquid -
properties resembling those of solid composition, high ductility, and low
n ;ib lormed m
. ..
liquid i icn in /
1
sed of two parts: a solute and a solthe solution or the material which is es the major portion of the solution ing gases liquids, or solids as either
i 5 660
,
J
Liqurfl soluhon Beginninc
Freezing (onge
o* soudtUcuhon
iquid [-solid solution - fnd of tolidificunofi
_
V
Solid sdunon y
qiowmg m liquid rich in Hi
-
'
4
M
)mmon solutions involve water as the sd in water.
I hi pi* qiutnsnf snlicj illey (r.OlV : | 07.Hi)
dissolved by the solvent is generally jure constant) and usually increases
;
Dcnflnt s i iih mSb
:
.
-
..
.
' . . |
.
\
Time
.
151
s for a solution: unsaturated satu,
vent is dissolving less of the solute
Fig. 5-2 Time-temperature cooling curve for the solidification of a small crucible of 50 percent antimony, 50 percent bismuth alloy. :
*
152
%
in the next chapter. There are two types of solid solutions substitution!
By considering the above four fa bility of one metal in another can t
,
and interstitial. 5-6
Substitutional Solid Solution "
I
In this type of solution, the atoms of the solul
an unfavorable relative-size fac to a low value. If the relative-size fi factors should be considered in de
that
SUPStHUta i'd 'afoms'oflKe solvent in the lattice structure of the solvel For example silver atoms may substitute for gold atoms without losing'tl ,
f
Cc: face-centered cubicfstructure of g6rd7and oid atoms may subs)
solubility! While the Hume-Rother solubility, there are exceptions to tl
-
tute for silver atoms in the f.c.c. lattice "structure of silver All alloys in t|i silver-gold system consist of an f.c.c lattice witfcLSiiyer and gold atoiw distributed at random through the lattice structure This entire system cou, 1 sists of a continuous series of solid solutions Several ifactors are now known largely through the work of Hurrr .
IE
.
The lattice structure of a solid s
with slight changes in lattice param atorrTis larger than the solvent ato
.
.
.
is smaller.
,
Rothery, that control the range of solubility in alloy systems | Crystal-structure Factor Complete solid solubility of two elemeni .
v
is never attained unless the elements have the same type of crystal latti
fi-7 Interstitial Solid Solutions These art '
Ijk'hd radii fit"into the spaces or interstic solvent atoms.
structure.
The size factor is favorable for solid-solution forma-
tion when, the difference in atomic radii is less than about 15 percent
.
the alloy system usually shows a minimum, if the relative-size factor! greater thfrn 15 Bgixent, solid-solution formation is very limited. For e: °
ample, silfer and lead are both f.c.c and the raiatiwo-c o f tor 1° abouj 20 percent. The solubility of lead in solid silver is about 1.5 percent, anj ,
_
the solubility of silver in solid lead is about 0 1 .
percent. Antimony and bis[
muth are completely soluble in each other in all proportions They havethi .
same type of crystal structure (rhombohedral) and differ in atomic radii bf about 7 percent. However, the solubility of antimony in f.c.c. aluminum
of solute is added to the solvent ar
enough, an interstitial solid sqlutioi
c6nsiderab)e mobjliiy
'
atQm§ b ave ..„
-
of tliejattice structure. More solu until the solution becomes saturat
composition starts to form. The it range of composition, is expresse
toward compound formation. Generally, the farther apart the elemeni are in the periodic table, the greater is their chemical affinity.
Relative-valence Factor If the solute metal has a different valence frois
..
that of the solvent metal, the number of valence electrons per atom, call6f
stifial solution, being of variable i a chemical formula. The lattice str
this type of solution is formed. Interstitial solid solutions norrm
erally are of little importance. Ca
the electron ratio, will be changed. Crystal structures are more sensitiv|
formsTTieTjasis for hardening ste
to a decrease in the electron ratio than to an increase. In other wordsj[ metal of lower valence tends to dissolve more of a metafof higher valerr
Carbon dissolves in iron interstitii
tnah vice versa.
For example, in the aluminum-nickel alloy system bol metals arie face-centered cubic. Th relative-size factor is approximatelj
carbon in a iron (b.c.c.) is only 0.C
14 percerit. However, nickel is lower in valencelhan aluminum and accord with the relative-valence factor solid nickel dissolves 5 percerii
lattice structure will exist in the re will interfere with the movemept
aluminum, but the higher valence aluminum dissolves only 0.04 percerj
therefore increase the strength ol
nickel.
the strengtheningj3LajTietaLby_al
,
-
,
i
;V
than the amount that may be dissol
Chemical-affinity Factor The greater the chemical affinity of two metals]
the more restricted is their solid solubility and the greater is the tendenc|
.
amount of smaller/atoms requiredj
is only about |
.
percent.
.
.
(0.77), nitrogen (0.71), and oxygen This type of solution differs frc
amount of solute atoms beyond 1 of these atoms in a particular are;
less than :0.1 percent, although the relative-size factor
.
interstitial solid solutions. These at
B
the relative size factor is greater than 8 percent but less than 15 percenf
I
Since the spaces c
size, only atoms with atomic radii
Relative-size Factor
:3
-
Me
INTRODUCTION TO PHYSICAL METALLURGY
'
.
I
! c
i
'
f
I
i
-
'"
in y iron (f.c.c.) is 2 percent at 2 '
Both types of solid solutions art
1
' " wucKSSSSS iSSjHSWSI
CONSTITUTION OF ALLOYS 153 ' .
itypes of solid solutions
,
|
i
substitutionltj
By considering the above four factors, some estimate of the solid solubility of one metal in another can be determined. It is important to note:-
pe of solution, the atoms of the solu
"'
'
that an unfavorable relative-size factor alone js sufficient to limit solubility to a low value. If the relative-size factor is favorable, then the other three:i factors should be considered in deciding on the probable degree of solid.
In the lattice structure of the solven tute for gold atoms withq ut losing thej _
_
.
of goiaTand gold atpms ay substi-| be structure of silver. All alloys in the lattice witti
solubility. While the Hume-Rothery rules are a very good guide to solid |§i| "
solubility, there are.exceptions to these rules.
silyer and gold atoms |
jlqicel.utipns. structure. This entire system con-
'
'
is smaller. -H' M 57 Interstitial Solid Solutions These are formed when atoms of small atomicfeS
argely through the work of Hume-
)>iubility in alloy systems
.
v
solid solubility of two elements
The jattjce strurture of a solid solution is basicallyJJiaLof the solvent v with slight changesjn lattice parameter. An expansion results if the solute atom is larger than the solvent atom and a contraction if the solute atom . . 1
J
M
.
have the same type of crystal lattice
v
radii fit into the spaces or interstices of the lattice structure of the largefpS
solvent atoms. Since the spaces of the lattice structure are restricted
infe|l size, only atoms with atomic radii less than 1 angstrom are likely to form Si" '
is favorable for solid-solution forma-
J
interstitial solid solutions. These are hydrogen (0.46), boron (0.97), carbo i
dii is less than about 15 percent If 8 percent but less than 15 percent
(0.77), nitrogen (0.71), and oxygen (0.60).
.
,
--
..
lirnum. If the relative-size factor is i formation is very limited For ex-
amount of smaller atoms required to form the compound is always greater
.
?;v£and tho ro|qti"'=t-fii7P fa trir js about
'
olid silver is about 1 5 percent and .
<, .> ...
This type of solution differs from interstitial compounds in that the>c?|
than the amount that may be dissolved interstitially. When a small amount s of solutejs added to the solvent and the difference in atomic radii is great
.
.
.
,
enough, an interstitial solid solution is formed. In this condition, the solute : ; ! atp.ms .have.considerable mobility and may move in the Inte fltfol pac' r jif of the lattice structure. More solute atoms may be dissolved Interstitlai1y,f.: until the solution becomes saturated at that temperature. Increasing the amount of solute atoms beyond this limit severely restricts the mobility of these atoms in a particular area and the interstitial compound of fixed 1: composition starts to form. The interstitial compound, showing a narrow . range of composition, is expressed by a chemical formula, but the interstitial solution, being of variable composition, cannot be represented by .
)out 0.1 percent. Antimony and bisler in all proportions They have the .
.
'
hedral) and differ in atomic radii by ty of antimony in f co. aluminum is .
elative-size factor s only about
i!
the chemical affinity of two metals
,
::W
-
.
lity and the greater is the tendency
illy, the farther apart the elements their chemical affinity metal has a different valence from .
valence electrons per atom
,
'
:..- Mriore of a metal of higher v lfence
£&uminum-nickel alloy system
,
:
forms ffie'basfe
| .
1
for hardening steel, which will be discussed in Chap. 8. i
in y iron (f.c.c.) is 2 percent at 2065°F, while the maximum solubility of s
carbon in 1
a iron (b.c.c.) is only 0.025 percent at 13330F
.
Both types of solid solutions are illustrated in Fig. 5-3. Distortion of the
lattice structure will exist in the region of the solute atom. [This distortion 1
.
i
I
m
"
Carbon dissolves in iron interstitially. The maximum solubility of carbon?
l§|n valence than aluminum, and in
4 solid nickel dissolves 5 percent jiinum dissolves only 0 04 percent
i
erally are of littlejmportance. Carbon in iron is a notable exception and ;
both
ilative-size factor is approximately
a chemical formula. The lattice structure always shows an expansion when this type of solution is formed.
Interstitial solid solutions normally have very limited solubility and gen-|
called
,
ystal structures are more sensitive i to an increase. In other words a
:
I
will interfere with the movem ent pj„dislocati.ons..on. slip planes and-WjU. .
,
therefore increase the strength of the alloy- This .is the erimaryJbMis for. the strengthening oLa .
._
m .
etal by alloying J
I V
\ I
1
'
Kc.'..".:.:.*.-::
' .
I
v ,
154 INTRODUCTION TO PHYSICAL METALLURGY
11 .
Si ~
4 L
1
-/
x
-
'
\
A i
r
I
-
-
I
J
r
.
I
m ,
;
(A)
Fig. 5-3- Schematic representation of both types of solid
PHASE
m
solutions, (a) Substitutional; (b) interstitial
.
In the previous chapter it possibilities for the structure of an all depend to a large extent on the typ
Introduction
NXv r con compounds solid solutioi m general are easier to separate melt over a range in temperature ha properties, that are influenced by those of the solvent and solute a usually show a wide range of composition so that they are not express ,
,
by a chemical formula
'
phases present, and can be changec sential to know (1) the conditions ur the conditions under which a change A great deal of information concer
.
A summary of the possible alloy structures is shown in Fig
.
5-4.
systems has been accumulated, and1 is in the form of phase diagrams, al
Alloy structure
S
Homogeneous
Solid solution
/
Mixlure
Intermediote
\
/
TPure metal
I \
I Solid solution
Intermetallic Interstitial Electron
.
Fig. 5-4
constitutional diagrams. In order to specify completely the
necessary to specify three independ are externally controllable, are tem With pressure assumed to be const;
Any combination of solid phases
alloy phase (compound)
Substitutional .Interstitial
.
rium diagram indicates the structura
}
UWiwdiate alloy
5
Possible alloy structures
ture and cnmposilino-The diagram tion of an alloy system. [ "
.
f
ldeally>the phase diagram will
equilibrium conditions, that is, unde change with time. Equilibrium cond
i v
.
i
slow heating and cooling, so that if time is allowed. In actual practice, \
higher or lower temperaturesjjjefig is heated or cooled.
m
.t
Hapid variati
phase changes that would normall will distort and sometimes limit the
i
si
a
It is beyond the scope of this tex1 equilibrium between phases in bina
:1 ;
1
r
1
""
'
'
.
PHASE DIAGRAMS r
( 1
I
6-1
titial compounds, solid solutions iver a range in temperature have Av'tS of the solvent and solute, and an so that thiey are not expressed ,
Introduction In the previous chapter it was indicated that there were many possibilities for the structure of an alloy. Since the properties of a material depend to a large extent on the type, number, amount, and form of the phases present, and can be changed by altering these quantities, it is essential to know (1) the conditions under which these phases exist and (2)
1
1
the conditions under which a change in phase will occur.
res is shown in Fig. 5-4.
Mixlure
11111 Any combino tion of solid phases TPure metal < Solid solution
[intermediate alloy
A great deal of information concerning the phase changes in many alloy systems has been accumulated, and the best method of recording the data is in the form of phase diagrams, also known as equilibrium diagrams or constitutional diagrams. , In order to specify completely the state of a system in equilibrium, it is necessary to specify three independent variables. These variables, which are externally controllable, are temperature, pressure, and composition. With pressure assumed to be constant at atmospheric value, the equilibrium diagram indicates the stmr.tuml changes due to variation of tempera-
}
ture and _
cpmposltiQrL~The diagram is essentially a graphical representa-
tion of an alloy system, j ldeally>the phase diagram will show the phase relatiqnships under equilibrium conditions, that is, under conditions in which there will be no change with time. Equilibrium conditions may be approached by extremely slow heating and cooling, so that if a phase change is to occur, sufficient time is allowed. In actual practice, phase changes tend to occur at slightly
I
I
"
1
J
i if
ii k
ependinqjjpon the rate at which the alloy
1
is heated or cooled. Rapid variation in temperature, which may prevent phase changes that would normally occur under equilibrium conditions,
i 1
higher or lower temperatures
will distort and sometimes limit the appKcation of these diagrams.
\
It is beyond the scope of this text to cover all the possible ponditions of equilibrium between phases in binary alloys. Only the most irfiportant ones
ili m
I
156
INTRODUCTION TO PHYSICAL METALLURGY
will be considered and they may be classified as follows: ,
1
Metallographic Methods This n
Components completely soluble in the liquid state a Completely soluble in the solid state (Type I)
an alloy to different temperatures, and then quickly cooling to retai
b Insoluble in the solid state: the eutectic reaction (Type II) c Partly soluble in the solid state: the eutectic reaction (Type III) d Formation of a congruent-melting intermediate phase (Type IV) 2
samples are then examined micro This method is difficult to app\)
e The'peritectic reaction (Type V)
the rapidly cooled samples do r
Components partly soluble in thejiquid state: the monotectic reaction (TypeVlj
structure, and considerable skill i microstructure correctly. This m diagram.
3 Components insoluble in the liquid state and insoluble in the solid state (TypeVII)j
4
Transformations in the solid slate
a
Allotropic change
b
Order-disorder
X-ray
c The|eutectoid reaction d
The iperitectoid reaction
-
A stud of these diagrams will illustrate basic principles which may] b appl'ed to understand and interpret more complex alloy systems y K
.
f ( Zj ordinates of Phase Diagrams Phase diagrams are usually plotted with,, temperature, in degrees centigrade or Fahrenheit as the ordinate and the| alloy composition in weight percentage as the abscissa It is sometimes| more convenient for certain types of scientific work to express the alloyl -
.
. .
-
'
yZ&nce the two metals are compl
1;
type of solid phase formed will b metals will generally have the s a
,
"
.
cotnpositibn in atomic percent
.
'
i
atomic radii by less than 8 perce The result of running a series o
.
.
The conversion from weight percentage!
to atomic percentage may be made by the following formulas: Atomic percent of A Atomic percent of S
or alloys between metals A and B A 0 percent S to 0 percent A 100
100X
mi
X + y(/W/A/)
to see the relationship between t
on a single set of axes. However ing curve has its own coordinati a separate experiment starting fr
iooy(M//v)
where M =f atomic weight of metal A N === atomic weight of metal 6
pure metals A and B show only a end of solidification take place s
X=f
weight percentage of metal A weight percentage of metal B Regardless of the scale chosen for temperature or composition there
intermediate compositions form i two breaks or changes in slope. first break is at temperature T,,
,
will be no difference in the form of the resulting phase diagram 6-3
\
.
Experimental Methods
cation, and the lower break at
.
The data for the construction of equilibrium dia-;
/ grams are determined experimentally by a variety of methods
,
intermediate alloy compositions The sense of the phase diagram,
the moSti
common being:
by drawing a line connecting £
Thermal Analysis This is by far the most widely used experimental|
solidification, the upper dotted li
method. As was shown in Chap. 5, when a plot is made of temperature!
all the points that show the end
vs. time, at constant composition the resulting cooling curve will show aj
line in Fig. 6-1.
,
a change in slope when a phase change occurs because of the evolution of|
P
.
solely in the solid state generally involve only small heat changes methods pive more accurate results.
I m
It is now possible to determine
heat by the phase change. This method seems to be best for determining the initial;and final temperature of solidification. Phase changes occurring!
I
!
Since this met
mension or by the appearance ol simple, precise, and very useful i bility with temperature. 6-4 Tyotf/I -Two Metals Completely J
'
.
diffraction
indicate the appearance of a new
,
and other
perature vs. composition. The a of cooling curves and plotted or Y
since the left axis represents thi
Similarly, TB is plotted. Since i
i
9
.
-
PHASE DIAGRAMS
SI
157
;
'
I
lassified as follows:
Metallographic Methods This method consists in heating samples of an alloy to different temperatures, waiting for equilibrium to be established and then quickly cooling to retain their high-temperature structure. The samples are then examined microscopically. This method is difficult to apply to metals at high temperatures because the rapidly cooled samples do not always retain their high-temperature structure, and considerable skill Is then required to interpret the observed
jquid state
,
Mi (Type I) itic reaction (Type II) utectic reaction (Type III) prmediate phase (Type IV) !
ptate: the monotectlc reaction (Type VI) and insoluble in the solid state (Type VII)
microstructure correctly. This method is best suited for verification of a diagram. X-ray diffraction
trate basic principles which may .
'
lagrams are usually plotted withal ahrenheit
,
.
as the rdinate and the
; as the abscissa.
"
,
.
nore complex alloy systems
:
Since this method measures lattice dimensions it will
indicate the appearance of a new phase either by the change in lattice dimension or by the appearance of a new crystal structure. This method is simple, precise, and very useful in determining the changes in solid solability with temperature. 6-4 Type J -Two Metals Completely Soluble in the Liquid and Solid States Ince the two metals are completely soluble in the solid state, the only type of solid phase formed will be a substitutional solid solution.xThe two metals will generally have the same type of crystal structure and differ in-
It is sometimes .
ientific work'to express the alloy nversion from weight percentage
»i5
:
.
.
atomicj adii by less than 8 percent.
e following formulas: 100X
+ Y{M/N)
(6-1)
00Y(M/N) + Y(M/N)
(6-2)
'
::
two breaks or changes in slope. For an alloy containing 80-4 and OS
,
.
"
,
;cursbecauseof the evolution of .
ems to be best for determining
..
-
Ration. Phase changes occurring jy small heat changes, and other .
I j I
|
intermediate alloy compositions will show a similar type of cooling curve. The sense of the phase diagram or some idea of its form may be obtained by drawing a line connecting all the points that show the beginning of solidification, the upper dotted line in Fig. 6-1 and another line connecting ,
f
11
,
,
all the points that show the end of solidification which is the lower dotted line in Fig. 6-1. ,
It is now possible to determine the actual phase diagram by plotting temperature vs. composition. The appropriate points are taken from theseriefe of cooling curves and plotted on the new diagram. For example in Fig. 6-2, since the left axis represents the pure metal A TA is plotted along this line. ,
,
I
~' '
the
first break is at temperature T,, which indicates the beginning of solidification, and the lower break at T2 indicates the end of solidificati6ri7 Sll
suiting phase diagram onstruction of equilibrium diaa variety of methods the most
.
,
on a single set of axes. However, the student should realize that each cooling curve has its own coordinates. In other words, each cooling curve is a separate experiment starting from zero time. The cooling curves for the pure metals A and B show only a horizontal line because the beginning and end of solidification take place at a constant temperature. However, since intermediate compositions form a solid solution, these cooling curves show
nperature or composition there
„
.
to see the relationship between the cooling curves, they have 6een plotted
:v.>;
lost widely used experimental a a plot is made of temperature Suiting cooling curve will show a
..#%v y,¥
The result of running a series of cooling curves for various combinations or alloys between metals A and 6, varying in composition from 100 percent A 0 percent 6 to 0 percent A 100 percent 6, is shown in Fig. 6-1. In order
.
Similarly, TB is plotted. Since all intermediate compositions are percent-
m
M v
1S8
INTRODUCTION TO PHYSICAL METALLURGY
100 0
80 20
50 40
40 50
20 60
C 100
us label the solid solution alpha
Percent /I Percent S
will be used to represent the pure lines there exists a two-phase rec
-
of a mixture of a liquid solution £
!
Specification of temperature ar region indicates that the alloy cot not give any information regardii
v: is
N
to know the actual chemical cor * "
;
'
two phases that are present. In necessary to apply two rules.
6 5 Rule ]-Chemical Composition of P composition of the phases of an
r /
.
peratureTin a two-phase region, <
-
:
a tie line+lo the boundaries of t
X
droppecHo the base line, and thj -
:
:
.
I
In Fig. 6-3, consider the alloy c< The alloy is in a two-phase regk
' -
to the boundaries of the field. P
the solidus line, when dropped I
f
-
the phase that exists at that boi solution a of composition 90A-1C base line, will give the composith ture, in this case the liquid solutic ;
Fig. 6-1
Time -»-
6-6
Series of co6ling curves tor different alloys in a
Rule Jl-Relative Amounts of Each I of the two phases in equilibriut
completely soluble system. The dotted lines indicate the form of the phase diagram.
Liquid solution
ages of A and B, for simplicity the percent sign will be omitted. A vertical line representing the alloy 8(M-20e is drawn and 7, and T2 from Fig. 6 are plottejd along this line. The same procedure is used for the other ,
h
Liquidus
,
compositions.
iquid +
.
The phase diagram consists of two points two lines, and three areas
solid solution a
,
The two points TA and TB represent the freezing points of the two pure| metals.
a
Solidus 5
The upper line, obtained by connecting the points showing the beginning| of solidification, is called the liquidus line; and the lower line determined ,
Solid solution
by connecting the points showing the eriti of solidification, is called the solidus Me. The area above the liquidus line is a single-phase region and any alloy in that region will consist of a homogeneous liquid solution. Similarly, the area below the solidus line is a single-phase region and any alloy ,
,
,.
..i
;
in this region will consist of a homogeneous solid solution. It is common
practice, in the labeling of equilibrium diagrams, to represent salkLsalutions and sometimes intermediate alloys by Greek letters. In this case let ,
.
r
.
.
.
j
§0\ I
A
20
40
60
80
Composition, weight percent B
Fig. 6-2 Phase diagram of two metals completely in the liquid and solid states.
!
-
.
1
i
PHASE DIAGRAMS
20 80
3 100
Percent /I Percent B
1S9
us label the solid solution alpha (a). Uppercase letters such as A and B will be used to represent the pure metals. Between the liquidus and solidus lines there exists a two-phase region. Any alloy in this region will consist of a mixture of a liquid solution and a solid solution. Specification of temperature and composition of an alloy in a two-phase region indicates that the alloy consists of a mixture of two phases but does not give any information regarding this mixture. It is sometimes desirable to know the actual chemical composition and the relative amounts of the two phases that are present. In order to determine this ir.formation it is necessary to apply two rules. 6'5 Rule J-Chemical Composition of Phases To determine the gctual chemical composition of the phases of an alloy, in equilibrium at any specified temperature in a two-phase region, draw a horizontal temperature line, called a tie line, to the boundaries of the field. These points of intersection are dropped to the base line, and the composition Is read directly. In Fig. 6-3, consider the alloy compos ed of 80A-20S at the temperature T.
i
n 1 1 .
'
A
if
,
"
i
.
The alloy is in a two-phase region. (Applying Rule I, drawjjip tie 'i"0 m" to the boundaries of the field.
m
Point m, the intersection of the tie line with
"
-
.
the
i
.J
ill
-
r
solidus line, when dropped to the base line, gives the composition of the phase that exists at that boundary. In this case, the phase is a solid solution a of composition 90>4-106. Similarly, point o, when dropped to the base line, will give the composition of the other phase constituting the mix-
i 1
ture, in this case the liquid solution of composition 74A-266. 6-6
RuleJI-Relative Amounts of Each Phase
To determine the relative amounts
i
of the two phases in equilibrium at any specified temperature in a two-
nt sign will be omitted A vertical .
rawn, and T, and 7 from Fig. 6-1
.
.
Liquid solution
5
,
Liquidus
procedure is used for the other
;
lOints
,
two lines, and three areas.
1
freezing points of the two pure
if
1
Liquid + . a>
solid solution
3
Solidus }
the points showing the beginning J e; and the lower line
p$nfcl of solidification, is called the line is a single-phase region
,
" -
Imogeneous liquid solution
.
.
.
r "ingle-phase '
it'
'
determined l "
,
and
Solid solution
1
Simi-
region, and any alloy
pus solid solution, jt is common |agrams, to represent SQlkLsalu-
jy Greek letters. In this case, let
_
£
20
40
1
_
60
1
_
i'i
_
80
B
Composition weight percent B ,
'
life
-
.
-
XT'?.
'
-
. ..
-
r ; .
.
-
.
S
.
.
1 '
Fig. 6-2 Phase diagram of two metals completely soluble in the liquid and solid states
i
ii h!
!
1-
-'
<
V,
M
160
INTRODUCTION TO PHYSICAL METALLURGY
Liquid (percent) = Liquid a
T
aj (percent) =
Tie line
>
0121-2
T2L2
>
0 rn
a
! a
.
Amount of liquid
Amount of a '
W a
10/9
6
10
><3
Liquid
26/?
205
Fig. 6-4 The tie line mo removed from Fig. 6-3 to illi application of the lever rule. 10
20
26 30
To summarize both rules, the a
Composition, weight percent 5 Fig. 6-3
perature 7 consists of a mixture (
Diagram showing the tie line mo drawn in the
two-phase region at temperature T
of composition 74/\-266 constituti
.
ent and the other a solid solution
phase region, draw a vertjcaMine representing the alloy and a horizontal temperature line to the boundaries of the field. The vertical line will divide! '
/ ...
the horizontal line into two parts whose lengths are inversely proportional I to the amount of the phases present. This is also known as the /ever nj/e >/The point where the vertical line intersects the horizontal line may be con.
sidered as the fulcrum of a lever system. The relative lengths of the lever i arms multiplied by the amounts of the phases present must balance. [v In Fig. 6-3
,
the vertical line, representing the alloy 20B
,
divides the hori-
zontal tie line into two parts, mn and no. If the entire length of the tie line mo is taken to represent 100 percent or the total weight of the two phases
WiS: >
. .
.:.
.
.
material very rich in A from the lie after the start of solidification, tl
mated as 69/A-31S (Fig. 6-5£)). \
mo
,
no
tion become richer in 6 but also t
If the tie line is removed from the phase diagram and the numerical values are inserted, it will appear as shown in Fig. 6-4. Applying the above equations, Liquid (percent) = ~ x 100 = 62.5 percent
.
6
a (percent) =- x 100 = 37.5 percent 16
-am
:
formed surrounding the
comp(
of az (Fig. 6-6). In order for equi solid phase must be a compositio the /A-rich core not only from the i
~
io
I
at L2. The only solid solution in e solid solution forming at 72 is a2. Hence, as the temperature is deci
mo
1
-
When the lower temperature 7
a (percent) = - x 100
.
|
as
mn
*
,
present at temperature 7, the lever rule may be expressed mathematically Liquid (percent) =-x 100
1
percent of all the material present Equilibrium Cooling of a Solid-Soiuti . equilibrium conditions of a parti to observe the phase changes tha perature T0 is a homogeneous sin remains so until temperature 7, is freezing or solidification now bee form, a,, will be very rich in the I
composed of 95/A-5e (Rule 1). S
,
,
6-7
immm
This is possible only if the coolin keep pace with crystal growth (Fi . t 72, the relative amounts of tl mined by applying Rule II:
i
-
v PHASE DIAGRAMS
161
/'
I: Liquid (percent) =
1
aj2 0:2 i-j
20
x 100 = 35
100 = 57 percent
fl
15
a2 (percent) =
0:2/-2
x 100 = - x 100 = 43 percent 35
v
.
Amount of liquid
4
1 Amount of a
fit! a
6
10
105
-I Liquid 2&B
t
I i
20 B 1
teRg
6-4
The tie line mo removed from Fig 6-3 to illustrate .
|£application of the lever rule. To summarize both rules, the alloy of composition 80>A-206 at the ternperature 7 consists of a mixture of two phases. One is a liquid solutionVf
S3
of composition 744-266 donstituting 62.5 percent of all the material prea-|$||
f
>enting the alloy and a horizontal
ent and the other a solid solution of composition 90-4-1 OS making up 37.5 ;V i
percent of all the material present.
,
y . V field. The vertical line will divide
6-7
lengths are Inversely proportional iis is also known as the iever rule.
its the horizontal line may be conI The relative lengths of the lever lases present must balance. ig the alloy 20B divides the hori- ,
.
,
if the entire length of the tie line
he total weight of the two phases lay be expressed mathematically
i.
"
'
.
equilibrium conditions, of a particular alloy 70-4-30S will now be studied to observe the phase changes that occur (see Fig. 6.5). This alloy at temperature T0 is a homogeneous single-phase liquid solution (Fig. 6.5a) and remains so until temperature 7, is reached. Since 7, is on the liquidus line, freezing or solidification now begins. The first nuclei of solid solution to form, a,, will be very rich in the higher-melting-point metal A and will be composed of 95-4-5S (Rule 1). Since the solid solution in forming takes material very rich in A from the liquid, the liquid must get richer in B. Just after the start of solidification, the composition of the liquid is approxi-
mated as 69-4-31B (Fig. 6-56).
-
- x 100 T70
- X 100
r
When the lower temperature 72 is reached, the liquid composition is
vn
no
i
o
[
at L2. The only solid solution in equilibrium with /L2 and therefore the only solid solution forming at 72 is a2. Applying Rule I (v2 is composed of 10B. ,
Hence
,
as the temperature is decreased, not only does the liquid composi-
se diagram and the numerical
crystals of a2 are formed surrounding the a, composition cores and also separate dendrites
WM1 in F|9- 6-4. Applying the above
of at (Fig. 6-6). In order for equilibrium to be established at ;72, the entire
tion become richer in S but also the solid solution. At 72,
ai
'
:
-
v:-:.
'
v
'
solid phase must be a composition a2. This requires diffusion bf B atoms to the A-rich core not only from the solid just formed but also from the liquid This is possible only if the cooling is extremely slow so that diffusion may .
= 62 5 percent .
-
keep pace with crystal growth (Fig 6-5c). .
:
37.5 percent
At 72, the relative amounts of the liquid and solid solution rrtay be determined by applying Rule II:
mm.
>''||
Equilibrium Cooling of a Solid-Solution Alloy The very slow cooling, under $1
ill
ri
"
162
INTRODUCTION T;0 PHYSICAL METALLURGY
Liquid solution (70/1-30 5)
Nuclei-solid solution a, (95/4-55)
Pnmory dendrites ff] (55)
Liquid solution (694-311
Liquid Lz
Dendrites of a2(90/Mo|
(455)
-Liquid Ljf
Iff)
Liquid
1 Liquid us
I
a2(10 5)
155/)-il5| 2
Liquid Ljl
a2 +
3
Dendrites ot 03 (80,4-205)
1
a3
/4
Fig. 6-6 Schematic picture of the alloy 30S at temp
'4
J
3
T, before diffusion.
Wl
5
\
Liquid L4(35/J-65l
Solidus
Groins of a4 \ ;-| (almost 704-A \ :'j
\ Groins of a
6-8 Diffusion
V (704-305)
a
Through the mechanism of diffi
structure disappeared, and the gr of this section is to explain briefly Diffusion is essentially statistic!
l« '
'
-
r* .'i
It was pointed out in thi
movement of atoms, in the soli(
305
if
movements of individual atoms.
be zigzag and unpredictable, wl '
A
10
20
30
40
«
50
58 60
65
-V7
V
70
Composition weight percent 5 ,
AO
Fig. 6-5 The slow cooling of a 70A-30S alloy, showing this microstructure at various points during solidification.
As the temperature falls, the solid solution continues to grow at the ex-
pense of the liquid. The composition of the solid solution follows the soli- M
> a/ 7
dus line while the composition of liquid follows the liquidus line and both M ,
'
phases are becoming richer in 6. At 7, (Fig. 6-5d), the solid solution wili make up approximately three-fourths of all the material present. The stu- iM
,
<
'h y :
dent should apply the lever rule at T3 and determine the relative quantities al of aj and'Lj. Finally, the solidus line is reached at 74 and the last liquid L4, very rich in 6, solidifies primarily at the grain boundaries (Fig. 6-5e). w
t>
However, diffusion will take place and all the solid solution will be of uni- B
.
form composition «(7(M-30S), which is the overall composition of the
alloy (Fig. 6-5/), Figure 6-7 shows the microstructure f a slow-cooled
;
solid solution alloy. There are only grains and grain boundaries. There is -
no evidence of any difference in chemical composition ihsidejhe grains, »
/
indicating that diffusion has made the grain homogeneous.
.; ! .
J.W.ii,vsJ..'
I'
I
*
\
Fig. 6>7 Microstructure of a solid solution iron al magnification 100X. (Courtesy of the Research Lai U S Steel.) .
.
PHASE DIAGRAMS
Nuclei-solid solution 0 (954-55) ,
Primary dendrites
Liquid solution (694-3lff);l| '
Dendrites of azOO -105)
i
(5fl)
-
iii
163
Liquid Lz
J f
i
Liquid L2 .. I t '
-
Liquid
I*)
(554-455) i ?
"
j s:
:
Liquidus
2
3
(f)
Liquid Lj
.
(42 4-58/91
i
Dendrites of aj (804-206')
mm
If
a2 (10 51
i
1 Fig. 6-6
i
Schematic picture of the alloy 30fl at temperature
T, before diffusion.
I
-
' ;
Liquid L4(354-655)
Groins of 04
\
6-8
(almost 70 4-\ \
Diffusion
iOB]
It was pointed out in the previous section that .diffusion
,
or the
movement of atoms, in the solid state was an important phenomenon.
Through the mechanism of diffusion under slow cooling the dendritic structure disappeared, and the grain became homogeneous. The purpose
(e)
1
of this section is to explain briefly how diffusion in solids may occltr. Diffusion is essentially statistical in nature, resulting from many random
-
movements of individual atoms. While the path of an individual atom may
be zigzag and unpredictable, when large numbers of atoms make such
50
58 60
65
.
1
ie solid solution follows the soli-
d I
.
eached at Tt and the last liquid he grain boundaries (Fig. 6-5e). ;the solid solution will be of unlv the overall composition of the
'
i
i
1 V?..
and grain boundaries. There is composition inside he grains in homogeneous.
,
5> fn >
f
fiicrostructurejaf a slow-cooled
Fig. 6-7 Microstructure of a solid solution iron alloy; magnification 100X. (Courtesy of the Research Laboratory, U S Steel.) .
.
.
.
i
I
5 ii.
>llows the liquidus line, and both Fig. 6-5d), the solid solution will II the material present. The studetermine the relative quantities
i I I
'
m
on continues to grow at the ex-
;
i
1
a
;
m
iii
'
.
;
1 Ii
1
\y\
r
I
ill
re
e
:
1f i
il
i i
164
INTRODUCTION TO PHYSICAL METALLURGY
the diffusion coefficient is a tunc
movements they can produce a systematic flow.
tant is temperature.
There are three methods by which diffusion in substitutional solid so-|
As a geners
coefficient doubles for every 20-c
lutions may take place; the vacancy mechanism interstitial mechanism, and atom interchange mechanism. These are illustrated schematically In ,
is not surprising, since all atoms librium positions in the lattice,
Fig. 6-8.
creases with increasing tempera thermal vibrations, often referred
In Chap. 2, under crystallization, it was pointed out that vacancies and
interstitial sites were a normal feature of a crystal structure. These imper- 3
cause an atom to jump out of its Therefore, temperature is obvio
fections greatly facilitate diffusion, or the jumping of adjoining atoms 1 Figure 6-8a shows how a solute atom might move one atomic spacing to the left by jumping into a vacancy, it is equally probable, of course,' .
that any one of the other atoms neighboring the vacancy could have made the jump. The vacancy has moved to the right to occupy the position of the previous atom and is now ready for another random interchange. The interstitial mechanism is illustrated in Fig. 6-8£>, where an atom in normal
whether jumping or diffusion is free energy when it is in a homoi force for diffusion. 6-9
Nonequilibrium Cooling-Origin of difficult to cool under equilibriur
position moves into an interstitial space, and the vacated spot is taken by the interstitial atom. As shown in the same figure diffusion may occur by an interstitial atom wandering through the crystal, but this method is more
state takes place at a very slow r ing rates there will be some diffei equilibrium diagram. Referring
likely in interstitial solid solutions.
cation starts at Tu forming a sc
It is possible for movement to take place by a direct interchange between two adjacent atoms, as shown in Fig. 6-8c or by a four-atom ring interchange, as in Fig. 6-8d. However, these would probably occur only under special conditions, since the physical problem of squeezing between
liquid is at L2 and the solid solut Fig. 6-6). Since diffusion is too i
,
,
closely packed neighboring atoms would increase the barrier for diffusion
,
Experimental evidence has indicated that the use of vacancies is the pri-
mary method of diffusion in metals. The rate of diffusion is much greater | in a rapidly icooled alloy than in the same alloy slow-cooled. The difference is due to the larger number of vacancies retained in the alloy by fast cooling. Vacancy migration also has a lower activation energy when compared .
with the other methods.
The rate of diffusion of one metal In another is specified by the diffusion coefficient, which is given in units of square centimeters per second. While
enough time will be allowed to ac
age composition will be betwee drops, the average composition ( from equilibrium conditions. It solution is following a
"
noneq
dotted in Fig. 6-9. The liquid, on
sition given by the liquidus line, At T3 the average solid solution Under equilibrium cooling, solit ever, since the average compt
reached the composition of the
plying the lever rule at T4 gives InlersfitiaK
/'Vacancy o
ofoooo
o\oy
o
o
mH
olom
o o
ooopo
ooooo
ooooo
o o oio o
o o,OwO o
o o/rb>o
o
o o xy o o
o o Qjy o
ffo o
"
t (percent) =
L,
(percent) =
o o o o o
o
cm5 o o
OOOOO
ooooo
Solidification will therefore con
o o o o o
o o o o o
o o o o o
ooooo
ture the composition of the soli( position, and solidification is c
(£) Interstitial
ic) Two-atom
mechanism
it/) Four-atom ring
interchange
richer in B than the last liquid 1
a) Vacancy mechanism
Fig. 6-8
'
a
\
Schematic dittusion mechanisms.
i
.
f
.
interchange
is apparent from a study of Fig.
v
'
m M
PHASE DIAGRAMS
3tiC flOW.
{:
the diffusion coefficient is a function of many variablesjthe most important is temperature. As a general rule, it-may be stated that the diffusion coefficient doubles for every 20-deg (centigrade) rise in temperature. This
1
diffusion in substitutional solid so interstitial mechanism pse are illustrated schematically in
lechanism
,
,
is not surprising, since all atoms are constantly vibratingiabout their equi-
as pointed out that vacancies and f a crystal structure These imper-'
,
librium positions in the lattice, and since the amplitude of vibration increases with increasing temperature. The energy associated with these
4
the jumping of adjoining atoms might move one atomic spacing
if
'
'
:
tL
m
cause an atom to jump out of its lattice position under suitable conditionsTherefore, temperature is obviously an important factor in determining whether jumping or diffusion is likely to occur. An alloy has the lowest
.
It is equally probable, of course,'
Spring the vacancy could have made pe right to occupy the position of
/
thermal vibrations, often referred to as the thermal energy, is sufficient to
.
:
m
free energy when it is in a homogeneous condition, and this is the driving force for diffusion.
Si
: .
Another random interchange. The
I
6-9
:t
Nonequilibrium Cooling-Origin of Coring In actual practice it is extremely
difficult to cool under equilibrium conditions. Since diffusion in the solid state takes place at a very slow rate, it is expected that with ordinary cool- j ing rates there will be some difference in the conditions as.indicated'by tfiiie !
ig. 6-8b, where an atom in normal
,
'
,
and the vacated spot is taken by me figure, diffusjon may occur by
t
Xvie crystal, but this method is more
.
equilibrium diagram. Referring again to the alloy 30B (Fig. 6-9), solidlf(*? j cation starts at Tu forming a solid solution of composition
;
.8c
or by a four-atom ring interwould probably occur only under problem of squeezing between
55 1
-
enough time will be allowed to achieve uniformity in the solid, and the average composition will be between ai and 02, say a a As the temperature. drops, the average composition of the solid solution will depart still further from equilibrium conditions. It seems that the composition of the solid '
-
.
t the use of vacancies is the prirate of diffusion is much greater
solution is following a
lloy slow-cooled. The difference
"
"
nonequilibrium
solidus line
to a'5, shown
,
dotted in Fig. 6-9. The liquid, on the other hand, has essentially the composition given by the liquidus line, since diffusion is relatively rapid in liquid. At T3 the average solid solution will be of composition a'3 instead of .03. Under equilibrium cooling, solidification should be complete at 74; how- .
retained in the alloy by fast coolictivation energy when compared other is specified by the diffusion
ever, since the average composition of the solid solution au has not
re centimeters per second While .
reached the composition of the alloy, some liquid must still remain. Applying the lever rule at T4 gives
.
'
1
o o o o
o o o o o
o,ao o
o o/5 b.o
oxyo o
o o qjy o
oooo
o o o o o
a*
(percent):
i
interchange
o o o o o
Id) Four-atom ring interchange
1
'
a,
x 100 ~ 75 percent '
'
r) Two-atom
fc .
Lt
nA
~
o o o o
!;!
Fig. 6'6). Since diffusion is too slow to keep pace with crystal growth, not
,
increase the barrier for diffusion
At T2,,th6- i|
liquid is at L2 and the solid solution now forming is of composition 02 (see
e by a direct interchange between
Of
4
L, (percent)
x 100 = 25 percent
i
a 4 -4
i
Solidification will therefore continue until Ts is reached. At this temperature the composition of the solid solution a'5 coincides with the alloy composition, and solidification is complete. The last liquid to solidify, L5, is richer in B than the last liquid to solidify under equilibrium conditions. It is apparent from a study of Fig. 6-9 that the more rapidly thfe alloy is cooled
i
) 1
.
I j
Ml
..
li
St-'- '"
166
INTRODUCTION TO PHYSICAL METALLURGY
act as a plane of weakness. It will
I
"
Liquid
corrosive medium.
2
?2
in mechanical and physical propel ceptibility to Intergranular corrosi
a2
3
f
l3
Therefore for ,
objectionable.
" 3
4
a5
r
There are two methods for solvi
vent its formation by slow freezinc
grain size and requires a very long is to achieve equalization of comf
+
structure by diffusion in the solid At room temperature for most rr ,
if the alloyjs reheated to a tempt
a
will be more rapid and homogef time.
Figure 6-11 shows the actual ea system, and the alloy 85Cu-15Ni I homogenization on the cored stru
by the series of photomicrographs quence shows the microstructure 10
20
30
10
50
60
70
Composition, weight percent 8 Fig. 6-9
Nonequilibrium cooling; the origin of coring.
i
.
the greater will be the composition range In the solidified alloy. Since the rate of chemical attack varies with composition, proper etching will reveal the dendritic structure microscopically (Fig. 6-10). The final solid consists of a
"
cored
"
structure with a higher-melting central portion surrounded by M
the lower-melting, last-to-solidify shell. -
t6
The above condition is referred f
5
_
'
as con
ng or dendritic segregation.
|
To summarize, nonequilibrium cooling results In an Increased tempera-
ture range over which liquid and solid are present; final solidification | occurs at a lower temperature than predicted by the phase diagram; the | .
final liquid to solidify will be richer in the lower-melting-point metal; and \ since diffusion has not kept pace with crystal growth, there will be a differ- i "
ence in chemical composition from the center to the outside of the grains, The faster the rate of cooling, the greater will be the above effects. 6-10
Homogenization Cored structures are most common in as-cast metals From the atpove discussion of the origin of a cored structure, it is apparent .
that the last solid formed along the grain boundaries and in the interden-f dritic spaces is very rich in the lower-melting-point met k Depending upon
the properties of this lower-melting-point metal, the grain boundaries may ;| V >
I
I
V
/
I hi
|T Fig. 6-10 Fine-cored dendrites in a copper-lead alio) |; waflnlflcation 100X. (Research Laboratories, Nationa' Jf- Ltad Company.)
1
\
1
:
J;
PHASE DIAGRAMS
'
A
act as a plane of weakness. It will also result in a serious lack of uniformity in mechanical and physical properties and, m some cases, increased susceptibility to intergranular corrosion because of preferential attack by a corrosive medium. Therefore, for some applications, a cored structure is
Liquid
-
167
2
objectionable.
'3
v
There are two methods for solving the problem of coring. One is to prevent its formation by slow freezing from the liquid, but this results in large grain size and requires a very long time. The preferred method industrially is to achieve equalization of composition or homogenization of the cored structure by diffusion in the solid state.
5 L
+
2
At room temperature, for most metals, the diffusior rate is very slow; but
if
m
i i|
if the alloy is reheated to a temperature below the solidus line, diffusion will be more rapid and homogenization will occur in a relatively short time.
i
j
50
60
Figure 6-11 shows the actual equilibrium diagram for the copper-nickel system, and the alloy 85Cu-15Ni is shown as a dotted line. The effect of homogenization on the cored structure of an 85Cu-15Ni alloy is illustrated
by the series of photomicrographs in Fig. 6-12. The first picture of this sequence shows the microstructure of the alloy as chill-cast. As the equi-
_
70
V
i
\
3
*
A."
4
in the solidified alloy. Since the sition, proper etching will reveal 3. 6-10)
.
9
The final solid consists
central portion surrounded by ; $:
0'
"
he above condition is referred
esults in an increased temperaire present; final solidification
Hi
*
1 v
ted by the phase diagram; the lower-melting-point metal; and
;
'
.
V
:;
v. '
A
tal growth, there will be a differ1:
?f£arter to the outside of the grains. l(vilt be the above effects.
-
v
..
::
5i
"':
'
St common in as-cast metals
i
.-
»
.
:a cored structure, it is apparent
3m
*
boundaries and in the interdeng-point met k Depending upon
jietal, the grain boundaries may
f
1
Fig. 6-10
Fine-cored dendrites In a copper-lead alloy;
magnification 100X. (Research Laboratories Lead Company.)
,
National
iii
-
i
168
"
11
INTRODUCTION TO PHYSICAL METALLURGY
am®-.
w
m
1500
S3 2600
r
1400 ,
I
Liquid :
oT 1300
I
2400 s
i + a
a 2200 £
£ 1200 a
1100
55
I
PL
m as
mm
m
J
Ni
10
20
30
40
50
60
70
80
90
i
Fig. 6-11 , Copper-nickel equilibrium diagram. {From Metals Handbook," 1948 ed„ p. 1198, American Society
38
m
'
Weighl percent copper
\3
asp
m
Cu
.
.
3
w
1000 .
.
t
2000
m
p
;
la]
"
t
for Metals, Metals Park, Ohio.)
ft
Iibrium diagram predicts, the first solid to be formed in the central axes i
of the dendrites is rich in nickel. Because of rapid cooling, there is a great
difference in nickel content between the central axes of the dendrites and 1 the interdendritic spaces. This difference is revealed by suitable etching ft? ,
The hext figure shows the same sample after heating at 13820F for 3 h K Counterdiffusion of nickel and copper atoms between the nickel-rich cores m-' .
,
If; -7 ] h r A 'f"y~-r':. . '
and the copper-rich fillings has reduced the composition differences some- fll what. The microstructure of the same sample heated to 17420F for 9 h is J -
shown In the third figure. The composition is completely equalized and «* the dendrites have disappeared. The grain boundaries are clearly evident
m
,
.
r
.
Black particles are copper oxide or nickel oxide inclusions.
The fourth
,
figure illustrates the same alloy slowly cooled by casting in a hot mold The dendritic structure is coarser than that of the chill-cast alloy. The last W .
.
figure shows this same sample heated 15 h at 1742°F. The structure is now Hi completely homogenized. Despite the smaller initial composition differ- I ences across the coarse dendrites as compared with the fine dendrites, it m took a longer time for equalization because of the greater distance through which the copper and nickel atoms had to diffuse in the coarse structure. Extreme care must be exercised in this treatment not to cross the solidus
;
line; otherwise liquation of the grain boundaries will occur, impairing the shape and physical properties of the casting (Fig. 6-13). 6-11 Rroperties of Solid-solution Alloys In general, in an alloy system forming a continuous, series of solid solutions, most of the property changes are caused by distortion of the crystal lattice of the solvent metal by additions
of the solu.te metal. The effect of composition on some physical and me--i
7
chanical properties of annealed alloys in the copper-nickel system is given ftin Table 6-1. structure.
Electrical resistivity depends upon distortion of the lattice 1
Since the distortion increases with the amount of solute metal
m a -
-
!
(e)
i
immmmm V*
PHASE DIAGRAMS
169
(V.
Mi
2600 an*
iquid
2400
?!
V
wm
m
1i
3
2200 f
is
i
*1
m
S3
2000
mm
V
70
80
90
.
K5
Cu
m
Si
S2
m
1
.
(/>)
(7)
m1M m
I to be formed in the central axes :;
;
-
.
SIM
e of rapid cooling, there is a great
5 central axes of the dendrites and :||
fie is revealed by suitable etching. || | \je after heating at 13820F for 3 h. 1
:
i
IB
m
v
1
ft
mmA
m r
|
,
ition is completely equalized and 1 ,
kin boundaries are clearly evident. The fourth
.
'
the composition differences some'sample heated to 1742°F for 9 h is
ikel oxide inclusions.
I
SSI
[oms between the nickel-rich cores
:
'
0ii
i
I
II
.
cooled by casting in a hot mold. hat of the chill-cast alloy. The last :
:
>
,
;
?
-
1 :
at 17420F The structure is now .
smaller initial composition differmpared with the fine dendrites it ,
ise of the greater distance through to diffuse in the coarse structure. :reatment
not to cross the solidus
undaries will occur impairing the ,
iting (Fig. 6-13).
>neral, in an alloy system forming a P
bst of the property changes are ,
|)sition on some physical and me-
A,
Fig. 6-12 Photomicrographs of an 85 Cu-15 Ni alloy, (a) chill-cast, 50x; (b) chill-cast,
? of the solvent metal by additions
i
ithe copper-nickel system is given
reheated 3 h at 1382
, 50x; (c) chill-cast,
\
50x; (d\ cast in a hot mold, 50x; (e) cast in a hot mold, reheated reheated 9 h at 17420F
,
15 h at 1742' , 50y. (By permission from Brick, Gordon, and Phillips, "Structure and
fids upon distortion of the lattice s with the amount of solute metal
Properties of Alloys," 3d ed., McGraw-Hill Book Company, New York, 1965.) J
4
(
.
170
INTRODUCTION TO PHYSICAL METALLURGY
6 ) Properties of Annealed Copper-Nickel Alloys*
J
TENSILE STRENGTH,
SITION
I
EL
ELONGATION, %
Bl
IN 2 IN.
1C
PSI
S
It
30,000
53
36
35,000
47
51
39,000
43
se
44,000
40
67
48,000
39
7C
50,000
a
73
53,000
41
7A
53,000
42
7c
50,000
43
6E
48,000
45
61
43,000
48
5A
,
-
4
fmmuionlmm H. M. Brick and A. Phillips, -Structure and Properties
r
9
Fig. 6-13 Photomicrograph of an aluminum alloy in which some melting has occurred at the grain boundaries during heating. After cooling, these portions of the grain boundaries appear as dark.broad lilies; magnification, 1,0OOX. (Alcoa Research Laboratories, Aluminum Company of America.)
the eutectic type to be discussed shoi doeutectic alloy. Examples of alloy Cu-Au and Ni-Pd. Those showing a known metallic systems of this type,
613 Type II-Two Metals Completely Solul added, and since either metal can be considered as the solvent, the maxi-J
Insoluble in the Solid State
mum electrical resistivity should occur in the center of the composition
insoluble in each other.
Techr
range. This is verified by the values given in the table. As copper is added| to nickel, the strength of the alloy increases, and as nickel is added to|
stricted that for practical purposes t
However,
copper, the strength of that alloy also increases. Therefore, between pure; copper and pure nickel there must be an alloy which shows the maximum strength. It turns out to be at approximately twcHhirdsjiickel and one-j third
copper. This is a very useful commercial alloy known as Monel. it
shows good strength and good ductility along with high corrosion resistance.
ft x
The same behavior is also true of hardness in that there will be ai
alloy that shows the maximum hardness, although the maximum tensile strength and hardness do not necessarily come at the same composition. 6-12
m
Variations of Type I Every alloy in the Type I system covered has a melting point between the melting points of A and B. It is possible to have asysteni|
if a
In which the liquidus and solidus lines go through a minimum or a maxi-| mum (Fig. 6-14a, b). The alloy composition x in Fig. 6-14a behaves just like
a pure metal. There is no difference in the liquid n
5
-
ji composition.!
Jt begins and ends solidification at a constant temperature with no changed in composition, and its cooling curve will show a horizontal line. Such
alloys are known as congruent-meliing alloys. Because alloy x has
A
Composition
(<7)
jfig. 6'14 (a) Solid-solution system showing a minimur
lowest melting point in the series, and the equilibrium diagram resembles|«||(6) Solid-solution system showing a maximum.
PHASE DIAGRAMS 171 e-i
m
|
|
Properties of Annealed Copper-Nickel Alloys*
ELECTRICAL :!
3
TENSILE
OSITION KEL
RESISTIVITY,
?
PARAMETER,
MICROHMS
l
10-8 CM
PER CU CM
i
BHN,
IN 2 IN.
10 MM, 500 KG
30,000
53
36
3 6073
35,000
47
51
3 5975
14
39,000
43
58
3 5871
27
44,000
40
67
3 5770
38
48,000
39
70
3 5679
46
50,000
41
73
3 5593
51
53,000
41
74
3 5510
50
53,000
42
73
3 5432
40
50,000
43
68
3 5350
30
48,000
45
61
3 5265
19
43,000
48
54
3 5170
PS1
permission Irom R. M. Brick and A. Phillips.
r
LATTICE
ELONGATION, %
STRENGTH,
17 .
.
.
.
.
.
.
.
.
.
.
*
'
!
68 .
.
Structure and Properties of Alloys.
i
2d ed,. McGraw-Hill BookCompany, New York, f 949. |
4
the eutectic type to be discussed shortly. It Is sometimes known as a pseu doeutectic alloy. Examples of alloy systems that show a minimum are Cu-Au and Ni-Pd. Those showing a maximum are rare, apd there are no -
known metallic systems of this type.
t
i
6-13
Type II-Two Metals Completely Soluble in the Liquid State and Completely
bonsidered as the solvent the maxi,
i
pr in the center of the composition
insoluble in each other. However, in some cases the solubility is so restricted that for practical purposes they may be considered insoluble.
j.yen in the table. As copper is added
.
.
Insoluble in the Solid State Technically, no two metals are completely
creases, and as nickel is added to
increases. Therefore, between pure
.
V
an alloy which shows the maximum
.
:
?;; fTimercial alloy known as Monel
.
It
ty along with high corrosion resistof hardness in that there will be an ss, although the maximum tensile
rily come at the same composition ype I system covered has a melting nd 6. It is possible to have a system
\ \
i j
; .
I
L
5
9
:
V
x
,
3
a CL
.
;
s5:i 9° through a minimum or a maxi-
a
L+ a
ft
i
a
\ion x in Fig. 6-14a behaves just like ;
..
v
'
.
' '
n the liquid and solid composition. wtant temperature with no change
vwtll show a horizontal
A
line. Such
V
-
1
I I
B
Iff)
-
? a//oys. Because alloy x has the she equilibrium diagram resembles
x
Composition
'
Fig. 6-14
B
A
Composition 1*1
(a) Solid-solution system showing a minimum.
(b) Solid-solution system showing a maximum. V
1
" t
172
i
INTRODUCTION TO PHYSICAL METALLURGY
Raoult's law states that the freezing point of a pure substance will bel lowered by the addition of a second substance provided the latter is soIu-ME ble in the pure substance when liquid and insoluble when solidified The| amount of lowering of the freezing point is proportional to the molecular
lowered. Therefore, since each meta the line connecting the points show
_
liquidus line, must show a minimum line in Fig. 6-15, showing a minimi point, for a composition of 40/1-606. positions, a portion of the cooling
.
weight of the solute.
This phase diagram can be developed from a series of cooling curves in a manner analogous to that used for the solid solution diagram described previously, but in this case, the experimental curves show a different kind
of behavior. The series of cooling curves for the pure metals and various
i
'
tion occurs at a fixe d temperature.
dotted in Fig. 6-15, is known as the eutectic composition 4n/4-fiQfl con temperature, the eutectic temper tu
_
alloys, and the room-temperature microstructures are shown in Fig. 6 15 1 The cooling curves for the pure metals A and B show a single horizontal line at their freezing points as expected. As S is added to A the temperature for the beginning of solidification is lowered As A is added to S the ] temperature for the beginning of solidification for those alloys is also ,
,
,
tic composition thus resembles tha melting alloy since we will shortly s
,
'
of two phases. The actual phase diagram may n breaks on the cooling curves to £
.
,
Si
100 0
80 20
'
50 40
50 50
40 50
30 70
20 B0
0 100
as shown in Fig. 6-16. The meltim M and N, are plotted on the vertic; For an alloy containing 8(M-206 th
Percent 4 Percent B
end of solidification 7f are plotted -
-
lowed for the remaining alloys. Th
- , -1
necting the two melting points, M beginning of solidification. The pc the minimum point E, is known as 1 tic temperature and 40/A-606 the is always a continuous line conr
O) =3
metals, so that the complete solidi
This phase diagram consists of 1 '
line is a single-phase homo geneo are soluble in the liquid state. Tl areas. Every two-phase area on a a horizontal line by single phase;
-
t
it
first, then the two-phase areas m
in Fig. 6-16, to determine the phas a horizontal tie line OL is drawn which means that the liquid is om
Time
area and intersects the left axis at
phase, the pure metal A, which be the two phases existing in the are reasoning is applied to determim These are liquid and solid S. The
i
:
-
:c.;3
Fig. 6-15
Cooling curves and room-temperature micro-
structures for a series of alloys of two metals that are insoluble in the solid state. The upper dotted line indicates the form of the liquidus and the lower dotted line the form
diagram and will be. useful to the diagrams.
of the solidus line.
Since the two metals are assurr
i
.
r
11
7 1 PHASE DIAGRAMS
3 point of a pure substance will ts|ibstance provided the latter is sol|| W and insoluble when solidified, tl
lowered. Therefore, since each metal lowers the freezing point of the other,
.
;;;
the line connecting the points showing the beginning of solidification, the liquidus line, must show a minimum. This is illustrated by the upper dotted
int is proportional to the molecul
line in Fig. 6-15, showing a minimum at pnlnt F
.
1
Hental curves show a different kind ,
eutectic compositionjtM-fiOfi compile solidifi ti p n
are shown in Fig 6-15; |
Wm A and B show a single horizontal ,
t
V
0
nrc at a single
of two phases. The actual phase diagram may now be constructed by transferring the breaks on the cooling curvfes to a plot of temperature vs. composition, as shown in Fig. 6-16. The melting points of the two pure metals, points
,
dification for those alloys is also
20 80
'
tic composition thus resembles that of a pure metal, it is not a congruent melting alloy since we will shortly see that the resulting solid is composed
As S is added to A the temperas lowered. As A is added to S the J
0
41
temperature the eutectic temperature. Although the freezing of the eutec-
.
'
_
dotted in Fig. 6-15, is known as the eutectic temperature. In one alloy, the _
>es for the pure metals and variou$"i structures
I
nnwn asjhe sulectic
point, for a composition of 404-605. Notice that over a wide range of compositions, a portion of the cooling curve that shows the end of solidification occurs at a fixe"d temperature. This lower horizontal line at TE, shown
d from a series of cooling curves in le solid solution diagram describee! ,
173
M and N, are plotted on the vertical lines that represent the pure metals.
0 -"-Percent d 100-«-Percent
For an alloy containing 80>A-20e the beginning of solidification 7, and the end of solidification TE are plotted as shown. The same procedure is followed for the remaining alloys. The upper line on the phase diagram connecting the two melting points, MEW, is the liquidus line and shows the
J
:
beginning of solidification. The point at which the liquidus lines intersect, the minimum point £, is known as the eutectic point. TE is called the eutectic temperature and 40/4-606 the eutectic composition. The solidus line
is always a continuous line connecting the melting points of the pure metals, so that the complete solidus line is MFGN.
This phase diagram consists of four areas. The area above the liquidus line is a single-phase homogeneous liquid solution, since the two metals are soluble in the liquid state. The remaining three areas are two-phase areas. Every two-phase area on a phase diagram must be bounded along a horizontal line by single phases. If the single-phase areas are labeled first, then the two-phase areas may be easily determined. For example, in Fig. 6-16, to determine the phases that exist in the two-phase area MFE, a horizontal tie line OL is drawn. This line intersects the liquidus at L, which means that the liquid is one of the phases existing in the two-phase
i
m
1
area and intersects the left axis at point O. The left axis represents a single phase, the pure metal A, which below its meltfh g point is solid.1 Therefore, the two phases existing in the area MFE are liquid and solid A. The same reasoning is applied to determine the two phases that exist in area A/EG. These are liquid and solid fi. The above ideas may be applied to any phase diagram and will be. useful to the student for the labeling of more complex diagrams. Since the two metals are assumed to be completely insoluble in the solid '
m
,
1 1
I f
1 m
I
i
I
174
INTRODUCTION TO PHYSICAL METALLURGY
will solidify. If slightly too much £ have shifted to the left, requiring A fore, at constant temperature, the
Melting point of
pure 6, resulting in an extremely fi microscope. This is known as the of this liquid of composition £ in
Liquid- solution
0
4
-
Melting point of a
-
N
I
Liquid
known as the eutectic reaction an
uquidus coolir
Liquid
Liquid
3
heatlr
_
Solid -4
+ rv
Solid B >
Since solidification of the eutect
.
Eutectic
its cooling curve would be the si gruent-melting alloy. The eutecti
point Sohdus
since there is a difference in com Solids
+
Solid 5 i .
vidual solid phases.
!
alloy 2
A
40
20
60
0 100 B
BO
Composition, weight percent B
Fig. 6-16
Liquid
Eutectic-type phase diagram.
1
»
state, it should be apparent that when freezing starts the only solid that can form is a pure metal. Also, every alloy when completely solidified must I be a mixture of the two pure metals. It is common practice to consider
alloys to the left of the eutectic composition as hypoeutectic alloys and those to the right as hypereutectic alloys. The way in which solidification. takes place is of interest and will now be studied by following the slow j; cooling of several alloys.
Jquid + Solid A
F
.
Alloy 1 in Fig. 6-17 is the eutectic composition 40A-60S. As it is cooled (
from temperature T0, it remains a uniform liquid solution until point £
Solid ,
the eutectic-temperature line, is reached. Since this is the intersection of
'
the liquidus and solidus lines, the liquid must now start to solidify and the temperature cannot drop until the alloy is completely solid. The liquid will ,
Solid A
+
Eutectic mixturi
solidify into a mixture of two phases. These phases are always the ones that \ appear at either end of the horizontal eutectic-temperature line in this case point F, which is the pure metal A, and point G, the pure metal B. Let us assume that a small amount of pure metal A is solidified. This leaves the ,
j remaining liquid richer in S; the liquid composition has shifted slightly ;
to the right. To restore the liquid composition to its equilibrium value
i L
m
.
,
B
10
20
30
40
Hypoeutectic alloys Composil
Fig. 6-17 Eutectic-type phase diagram,
r 3
PHASE DIAGRAMS
GY
175
will solidify. If slightly too much Q is solidified, the liquid composition will have shifted to the left, requiring A to solidify to restore equilibrium. There fore, at constant temperature, the liquid solidifies alternately pure A and pure B, resulting in an extremely fine mixture usually visible only under the microscope. This is known as the eutectic mixture (Fig. 6-18). The change of this liquid of composition E into two solids at constant temperature is -
Liquid solution
Melting point of £
N
known as the eutectic reaction and may be written as
Liquidu;
Liquid .
Liquid
"""
I"9
. solid+solid fl
heating
+
eutectic nnixture
i
Solid b
Since solidification of the eutectic alloy occurs at constant temperature,
V
6
tect c
1
its cooling curve would be the same as that for a pure metal or any con
-
lint
gruent-melting alloy. The eutectic solidification, however, is incongruent 1/
4
since there is a difference in composition between the liquid and the indi-
I
Solid /5
vidual solid phases.
i ff
1
Alloy
Alloy
Alloy
3
2
60
8C
C
IOC
veight percent 6
Liquid
1
when freezing starts the only solid that
ery alloy when completely solidified must l; itals It is common practice to consider ft
'
.
composition as hypoeutectic alloys and
,
...
vp
w,
Kt\ now be studied by following the slow Jic composition 40>4-606 As it is cooled .
i uniform liquid solution until point £
,
sached. Since this is the intersection of
liquid must now start to solidify
,
\ G
E
F
i
Solid A
and the Solid A
i alloy is completely solid The liquid will
+
+
solid B
Eutectic mixture + Solid 5
Eutectic mixture
l
.
.
Liquid + Solid b
Liquid + Solid A
a//oys. The way in which solidification J
vivs. These phases are always the ones that '
tal eutectic-temperature line in this case and point G, the pure metal S Let us ,
.
,
.
re metal A is solidified
.
L
.
This leaves the M
liquid composition has shifted slightly composition to its equilibrium value S ,
10
20
Hypoeutectic alloys x
1
Fig. 6-17 Eutectic-typ« phase diagram
1 t
m --r '
i
40
30
60
7C
80
90
Hypereutectic alloys
Compositiorv, weight percent B
B
.
176
INTRODUCTION TO PHYSICAL METALLURGY
a traction of a degree above Te are; Phases
Liquid
Composition
404-608
T£x£ Relative amount
x 100 = 33%
1
I
The microstructure would appear
»>
?1
(33 percent), having reached the fine intimate mixture of A and 6 as
tied, the alloy will consist of 67 perc
m
m
<3
A (which formed between 7, and 7£ percent eutectic (A + 6) mixture (F eutectic point £, when solidified,
m
and the eutectic mixture. The close
it
composition, the more eutectic nr alloy (see microstructures in Fig. 6
i
:: :-y
*:
SMI WW. Fig. 6-18
Alloy 3, a hypereutectic alloy con cooling process as alloy 2 except the liquid deposits crystals of pur( decreased, more and more 6 will
Lead-bismuth eutectic mixture 100X. (Research ,
Laboratories, National Lead Company )
The amount of liquid gradually de
.
moves down and to the left along \
Alloy 2, a hypoeutectic alloy composed of 804-206 remains a uniform liquid solution until the liquidus line, temperature 7,, is reached. At this
at the eutectic temperature. The i
j
,
eutectic (A + S) mixture. After sc percent grains of primary B or pro 6) mixture. The student should ve
point the liquid L, is saturated in A and as the temperature is dropped slightly, the excess A must solidify. The liquid, by depositing crystals of pure A, must become richer in S. Applying Rule I at temperature 72 shows the solid phase to be pure A and the liquid composition L2 as 70A-306. The
structure at room temperature.
point, when solidified, will consist < tic mixture. The only difference wi structures in Fig. 6-15). The relati
amount which has solidified up to this temperature would be found by applying Rule II: A (peicent)
L} (percent)
<
x?L7
10 x 100
7
30
r2x2
x 100 = 33 percent
Primary A
20
T2L2
X 100
30
CI
i
x 100 = 67 percent
The microstructure would appear as in Fig. 6-19a. As solidification con-
\ tinues, the. amount of pure solid A increases gradually by continued pre\ cipitation from the liquid. The liquid composition, becoming richer in fl
o .
,
'
/
>
.
.
'
-
i
l
\ m
Q
is slowly traveling downward and to the right along the liquidus curve,
while the aimount of liquid is gradually decreasing. When the alloy reaches xe, the eutectic line, the liquid is at point £. The conditions existing just 1
.
I
[
(o)
Fig 6 19
Stages in the slow cooling of an 80/4-206 ;
PHASE DIAGRAMS 177
a fraction of a degree above TE are: Phases Composition
Liquid 40/4-606 TgXc
Relative amount
Solid A 100>4 XbE
- x 100 = 33%
- x 100 = 67%
T
jE
TEE
The microstructure would appear as in Fig. 6-19b. The remaining liquid (33 percent), having reached the eutectic point, now solidifies into the
as?'.
v5
fine intimate mixture of A and B as described under alloy 1. When solidi-
m
fied, the alloy will consist of 67 percent of grains of primary A or proeutectic A (which formed between 7, and TE or before the eutectic reaction) and 33 percent eutectic (A + 6) mixture (Fig. 6-19c). Every alloy to the left of the eutectic point £, when solidified, will consist of grains of proeutectic A and the eutectic mixture. The closer the alloy composition is to the eutectic composition, the more eutectic mixture will be present in the solidified alloy (see microstructures in Fig. 6-15). Alloy 3, a hypereutectic alloy composed of 10A-90S, undergoes the same cooling process as alloy 2 except that when the liquidus line is reached the liquid deposits crystals of pure B instead of A. As the temperature is decreased, more and more 6 will solidify, leaving the liquid richer in A.
;
m .i i
The amount of liquid gradually decreases, and its composition gradually moves down and to the left along the liquidus line until point £ is reached at the eutectic temperature. The remaining liquid now solidifies into the eutectic (A + S) mixture. After solidification, the alloy will consist of 75 percent grains of primary B or proeutectic B and 25 percent eutectic (A + 6) mixture. The student should verify these figures and sketch the microstructure at room temperature. Every alloy to the right of the eutectic point, when solidified, will consist of grains of proeutectic S and the eutectic mixture. The only difference will be in the relative amounts (see microstructures in Fig. 6-15). The relationship between alloy composition and
i of 80A-208, remains a uniform | iperature T,, is reached. At this V? ,
I as the temperature is dropped
liquid, by depositing crystals of |l| "
ig Rule I at temperature 7 2 shows I composition/.2 as 70/A-30S. The temperature would be found by
100 = 33 percent
Li
Liquid of composition E
Primary A
I
m i
i i
Eutectic mixture
§
x 100 = 67 percent
?3
0
M g. 6-19a. As solidification con-
,
11
CP CD S
ses gradually by continued preiposition, becoming richer in 6
Primary /I
.
:
right along the liquidus curve, reasing. When the alloy reaches E
.
Q
} '
1
(a)
-
.
The conditions existing just
Rg, 6-19 :
i
Stages
.1
or
proeutectic /I
n i
[/>)
in the slow cooling of an 80A-20fl alloy.
..
A
J
..
J
.
178
;
.
INTRODUCTION TO PHYSICAL METALLURGY
I:
microstructure may be shown by using the eutectic composition as Wmm-
11
imaginary boundary line. The area below the solidus line and to the leftlMli.
1400
of the eutectic composition is labeled solid A + eutectic mixture, and thaf|P f to the right, solid 6 + eutectic mixture (Fig. 6-17). Figure 6-20 shows the Aprelation between alloy composition and relative amounts. jmjk From the previous discussion it is apparent that, regardless of alloyipRS
\zoo
composition, the same reaction takes place whenever the euiectic-temi perature line is reached, namely, cooling
Liquid
Liquid
1000
solid A + solid S
heating eutectic mixture
The above reaction applies specifically to this diagram; however, the'B*
.
,
800 - Liquid
eutectic reaction may be written in general as Liquid
cooling heating
+
Isolid Al
E
solid, -r solid2
s
v
eutectic mixture
600
the only requirement being that the eutectic mixture consist of two dif-J-jE ferent solid phases. This mixture may be two pure metals, two solid solu-' tions, two intermediate phases, or any combination of the above. 400
/The simplified aluminum-silicon phase diagram is shown in Fig. 6-21,
QJ
Si + eutec
neglecting the slight solubility of silicon in aluminum. The numbers at
the bottom of this diagram refer to the photomicrographs in Fig. 6 22. Be-
5
ginning with alloy 1 at the left of Fig. 6-21, the microstructure of pure alumi- § num is shown in Fig. 6-22a. Alloy 2 (Fig. 6-225), containing 8 percent silicon, consists of dendrites of primary or proeutectic aluminum surrounded by the eutectic mixture of aluminum and silicon. Notice the fine alternate light and dark structure of the eutectic. Since the eutectic is formed from I oo
200
Al
12
Fig. 6-21 Primary ff or
proeutectic /
proeutectic B
20
30
40
50
Composition, weight pe
3
Primary A or
10
3
4
5
Simplified aluminum-silicon phase diagrar
numbers on the bottom correspond to the photomicr graphs in Fig. 6-22. O
QJ
i
50
50
Eutectic mixture I
I a)
0 20
40
60
Compositicn weight percent 5 ,
Fig. 6-20 Diagram showing the linear relationship between the parts of the microstructure and alloy composition tor the eutectic system of Fig. 617.
80
100 0
the last liquid to solidify, it fills th drites. AlloyS (Fig. 6-22c) is the ei and consists entirely of the eutect microstructure will consist of prim ture, the amount of primary sili content as shown in Fig. 6-22cy an structure of pure silicon. It is the
librium diagram, with reasonable ; which will exist in an alloy after si
3? PHASE DIAGRAMS
179
"
it
3 the eutectic composition as an )w the solidus line and to the left '
2600 1400
)lid A + eutectic mixture, and that
3
Fig. 6-17). Figure 6-20 shows the
:
2400
relative amounts.
pparent that, regardless of alloy Si '
220C
200
)lace whenever the eutectic-tem-
Liquid 2000
I000
I A + solid 5
I800
,
iSrilectic mixture
o
Liquid + solid Si
ly to this diagram; however the J
1600: OT
,
"
a
al as
S
800
Liquid +
a>
1400 e
solid Al I
-
lid, + solid 1200 .
ectic mixture
M
500
Miotic mixture consist of two diftwo pure metals, two solid solu-
.
'
5
[ diagram is shown in Fig. 6-21, 1 in aluminum.
.
800
-
400 Si + eutectic
The numbers at
otomicrographs in Fig. 6-22. Bethe microstructure of pure alumi6-22b) containing 8 percent sill-
.
I
J
mbination of the above.
'
f
:
1000
600
1
400
200
i
,
roeutectic aluminum surrounded 10
Al
jsilicon. Notice the fine alternate
20
Since the eutectic is formed from 0
I
1*3
Fig. 6-21
Primary B or proeutectic B
30
40
50
60
70
80
90
Si
Composition, weighl percent silicon
2
3
4
5
6
Simplified aluminum-silicon phase diagram. The
numbers on the bottom correspond to the photomicrographs in Fig. 6-22. o
50
i
the last liquid to solidify, it fills the spaces between the arms of the dendrites. Alloy 3 (Fig. 6-22c) is the eutectic composition of 12 percent silicon
and consists entirely of the eutectic mixture. As we move to the right, the 80 ;<::lKn\ B
,
100 B
I:--)
'
microstructure will consist of primary silicon (black) and the eutectic mix-
3i
ture, the amount of primary silicon increasing with increasing silicon content as shown in Fig. 6-22d and e. Finally, Fig. 6-22 shows the microstructure of pure silicon. It is therefore possible to predict from an equir librium diagram with reasonable accuracy, the proportions of.each phase1 which will exist in an alloy after slow cooling to room temperature. ,
'
I1 !
.
i
.
180
V-!
INTRODUCTION TO PHYSICAL METALLURGY
6*14
r«tpy
Type III-Two Metals Completel Partly soluble in the Solid State for each other in the solid state,
I
7
7
fore, the most important alloy sy
From the discussion of the gi student is familiar with the metf _
a series of cooling curves. The 1 directly. The phase diagram of this type of the two pure metals are indie; ' "
.
liquidus line is TAETB, and the si
J
areas should be labeled first. Ab
Alloy 1. 99 95% Al
a
phase liquid solution. At the mel lines meet, the diagram resemt (complete solid solubility), and s
Alloy 2. 8% SI
.
solid state, a solid solution mus
solidify crystals of pure A or pun of solid solutions,- The single-p !
if Ta
\c
Alloy
12% Si
(1/) Alloy 4. 20% Si Liquid + a
(2 phoses) 5
F
«
m
II phase)
z1 4 1
J
i
Solvus '
v
i
H
A
Alloy 5 50% Si
j
-
Solidus
a
10
40
20
[f] Alloy 6. Pure Si Composition
Fig. 6-22 Photomicrographs of aluminum-silicon alloys as numbered at the bottom of Fig. 6-21. (Research Laboratories, Aluminum Company of America.)
Fig 6-23
Phase diagram illustrating partial
bility.
\
r
1
solid
r
PHASE DIAGRAMS : 181
SMi fype III-Two Metals Completely Soluble in the Liquid State but Only
Partly solubie in the Soiid State Since most metals show some solubilit|| for each other In the solid state, this type is the most common and, there- . fore, the most important alloy system. From the discussion of the previous two types, it is assumed that the student is familiar with the method of determining a phase diagram from a series of cooling curves. The remaining types will be drawn and studied directly. The phase diagram of this type is shown in Fig. 6-23. The melting points of the two pure metals are indicated at points TA and TB, respectively. The
5
liquidus line is T ETg, and the solidus line is TAFEGTB. The single-phase areas should be labeled first. Above the liquidus iine, there is only a singlephase liquid solution. At the melting points, where the liquidus and solidus lines meet, the diagram resembles the cigar-shaped diagram of Type I (complete solid solubility), and since these metals are partly soluble in the solid state, a solid solution must be formed. Alloys in this system never
(/>) Alley 2. 8% Si
I:-
k
f m
1
i
solidify crystals of pure A or pure B but always a solid solution or mixture of solid solutions - The single-phase a (alpha) and (3 (beta) solid-solution
4
X
l|
Liquid
11 phase)
i (
KIIIIM I (f
'
"' A
J
'
' a
uqutd '
a
1
1 ' i'Ii-imv;)
.
a)
7 Solidus
( 1 a
A
4
.
G
(1 phase]
Eutsctic
poin l
a + 13 (2 phases) 1
..to
i /
Solvus
4
Soiuus
H
10
If) Alloy 6. Pure Si
n 20
40
SO
Composition weight percent B ,
Fig. 6'23 bility.
Phase diagram illustrating partial solid solu-
30
95
8
182
INTRODUCTION TO PHYSICAL METALLURGY
areas are nbw labeled.
Since these solid solutions are next to the axes l
Liquid
.
Cooling
they are known as terminal solid solutions. The remaining three two-phasef
curve
r
areas may now be labeled as liquid + a liquid + p, and a+ (3. At TE, the a *
solution
Alloy
(55)
(20/9)
solid solution dissolves a maximum of 20 percent B as shown by point f| and the /3 sblid solution a maximum of 10 percent A as shown by point G
Liquid solution
,
.
With decreasing temperature the maximum amount of solute that can be 1 dissolved decreases, as indicated by lines FH and GJ. ,
.
These lines are | I
called solvus lines and indicate the maximum solubility (saturated solution) * of 6 in /A (a solution) or A in 6 (/3 solution) as a function of temperature
3
Alpho solid solution VIB
.
Point E, where the liquidus lines meet at a minimum as in Type II, is known ,
as the eutect/c po/n/. The slow cooling of severafalloys will now be studied Alloy 1 (Fig. 6-24), composed of 95A-5Bl when slow-cooled will follow
v
-
a
.
.
a process exactly the same as any allby in Type I: When the liquidus line Alpho solid solution (bB)
is crossed at 7,, it will begin to solidify by forrriing crystals of a solid solution extremely rich in A. This process continues with the liquid getting richer in B and gradually moving down along the liquidus line. The a ,
Ti
fig 6-25
Alloy .
'
'
.
' .
.
20
The cooling
curve and microstructure at ve
temperatures during solidification of a 95/1-58
'
.
i0
55
.
.
me
Alloy
Alloy
Alloy
3
2
alloy
.
solid solution, also getting richer
\.
line. When the solidus line is final
ing pace with crystal growth, the et solution and will remain that way c
of solidification and the cooling ci Alloy 2, 30/A-70e, is the eutectii
Liquid
l:
3
1
the eutectic temperature is reache< line, the liquid now undergoes the
ture, forming a very fine mixture of
I
5
L
the eutectic mixture are given by ture line, a of composition F and j3 may be written as
+
Tf M F
E
6
0
tect ic mixture
-
:
8
eulec he
,
cool
Liquid ;=
mixture
a
heal
N
This reaction is the same as th(
+
;
.iV
'
:
"
"
'
'
.
.
.
H
J
HL A
3
10
15
20
30
10
50
70
Composition, weight percent B
SO
90
relative amounts of a and /3 in the
applying Rule II (lever rule):
J
K
gram, except for the substitution
8
EG
a (percent) =- x 1( rG
EF
Fig. 6-24 Ity.
Phase diagram illustrating partial solid solubil-
rG
\
(
m
!
jS (percent) = -- x I1
9*
I
PHASE DIAGRAMS
olid solutions are next to the axes
Liquid
,
:
'
"
.
.
183
ins. The remaining three two-phase
Cooling :urve
liquid + p, and a + p At TE, the a 20 percent B as shown by point F 10 percent A as shown by point G
Alloy
solution
(55)
.
Liquid solution
mm
(20/9)
.
Ttium amount of solute that can be
f
.
l
jines FH and GJ.. These lines are
;
4
mum solubility (saturated solution) tion) as a function of temperature : a minimum, as in Type II, is known f severafalloys will now be studied.
+
S: :
.
i
Alpho solid
.
solution (25)
? f
.
'
.
rSSj when slow-cooled will follow
Tin Type!: When the liquidus line Dy forcing crystals of « solid solucontinues
,
n along the liquidus line.
.
i
Alpha solid solution (55)
with the liquid getting
Time
The <*
10
20
Fig. 6-25 The cooling curve and microstructure at various temperatures during solidification of a 95 -58 alloy. Allov 2
solid solution, also getting richer in B, is moving down along the solidus line. When the solidus line is finally crossed at T4 and with diffusion keep'
A .
til
ing pace with crystal growth, the entire solid will be a homogeneous a solid
1
solution and will remain that way down to room temperature. The process of solidification and the cooling cijrve for this alloy are shown in Fig. 6-25. Alloy 2, 30A-70B, is the eutectic composition and remains liquid until
1 ' '
the eutectic temperature is reached at point E. Since this is also the solidus line, the liquid now undergoes the eutectic reaction, at constant temperature, forming a very fine mixture of two solids. The two solids that make up the eutectic mixture stfe given by the extremities of the eutectic-temper&ture line, a of composition F and /3 of composition G. The eutectic reaction
/3
i
may be written as
if
is
/3
/S+ eutectic mixture
Mil'
Liquid
cooling
1
a + /3
heating eutectic mixture
1
.
This reaction is the same as the one which occurred in the Type II dia-
gram, except for the substitution of solid solutions for pure metals, .'f he . '
-
70 cent 5
80
90
3
I
relative amounts of a and p in the eutecticjnixture may be determined by
i
applying Ru[e II (lever rule): It £G
20
a (percent) = - x100 = - x100 = 28.6 percent FQ
70
EF
50
/3 (percent) = - x 100 = - x 100 = 71.4 percent FG
70
1 .
v
,
-
,:-
-
184
INTRODUCTION TO PHYSICAL METALLURGY
a
-
/Because of, the change in solubility of B in A line FH, and of A in fi, line GJ, there will be a slight change in the relative amounts of a and p as the alloy is cooled to room temperature. The relative amounts of a and j3at
Liquid
Cooling
,
Liqu
solution (405)
curve
Alloy 3
room temperature ars KJ
25
a (percent): = - x 100 = - x 100 = 29.4 percent :
HJ
85
HK
60
B (percent) = - x 100 = - x 100 = HJ
85
Liquid
5
+
70.6 percent
Q
The eutectic mixture is shown in Fig. 6-29c. Notice the similarity between this picture and the eutectic mixture formed in Type II (Fig. 6-18). It is not possible to tell microscopically whether the eutectic mixture is made up of two splid solutions or two pure metals.
0 + /?
Alloy 3, 60A-40B, remains liquid until the liquidus line is reached at Tj The liquid starts to solidify crystals of primary or proeutectic a solid solution very rich in A. As the temperature decreases the liquid becomes richer
'
Time
.
and richer in S, gradually moving down and to the right along the liquidus line until it reaches point E. Examining the conditions which exist just above the eutectic temperature TB there are two phases present: ,
N9 6 27
A
30
40
50
The cooling curve and microstructure at vai
temperatures during solidification of a 604-408 alloy.
Notice the similarity in microstrucl As the alloy cools to room tempera
indicated by the solvus line FH, sc
Phases
Liquid
Primary a
Chemical composition
30A-70B
80 -203
Relative amounts
40%
60%
solution. The process of solidificat are shown in Fig. 6-27.
Alloy 4, 854-15/3, follows the sam The student should verify the above numbers by applying Rules I and II '
at the eutectic temperature.v Since the remainingJjguid (401 percent) is at point E, the right temperature and composition to form the eutectic mix-
ture, it now solidifies by forming alternately crystals of a and p of the composition appearing at the ends of the eutectic temperature line (points >
F and G). "The temperature does not drop gntiLsolidification is complete;! and when copiplete, the microstructure appears as shown in Fig 6-26
microstructure at various tempera
alloy are shown in Fig. 6-28. Solid TG, the resultant solid being a he solution. At point M the solution explained previously, shows the d<
creasing temperature. s the alloy N
.
The a solution is now saturate
.
conditions of slow cooling, the exci A is soluble in B, the precipitate d Primary 3 or proeutecvic a 60%
Eutectin mixture
(tf+/9)
40%
but rather the j3 solid solution. At largely of a with a small amount boundaries (Fig. 6-28). The studet cess /? by applying the lever rule al
If the £ phase is relatively brittle, tile. The strength of an alloy to a I ..
that is continuous through the alio constitutes only about 5 percent oi
>
Fig. 6-26
Schematic picture of the microstructure, after
solidification, of alloy .3 in Fig. 6-24.
work along the grain boundaries. along these boundaries. This alio
PHASE DIAGRAMS
?f B in A, line FH, and of A in 8 line te relative amounts of a and as the ,
r The
relative amounts of « and /Sat
solution (405)
i
/fffjjK Alloy
r
iquid solution 1705) Alpha solid solution (205)
25
- x
t
Jo
100 = 29.4 percent
.
30
35
2?
Liquid
x 100 = 70.6 percent
p-29c
+ a
Eutectic mixture (705)
a) -
Notice the similarity between germed in Type II (Fig. 6-18). It is not her the eutectic mixture is made
+-
(alpha + Ijeta solid solution)
a
.
a + /3 4 -
netals.
il the liquidus line is reached at T3 primary or proeutectic a solid soludecreases the liquid becomes richer
Time
.
: ,
I
Liquid solution (SO ) /.Alpha solid solution (105)
Liquid
Cooling curve
185
10
20
30
40
50
GO
Alpha solid solution (205)
r
70
Fig. 6-27 The cooling curve and microstriicture at various temperatures during solidification of a 60/1-406 dilloy.
I ll
n and to the right along the liquidus
.
,
ng the conditions which exist just
Notice the similarity in microstructure between this alloy and Fig. 6-19c.
e are two phases present: .
As the alloy cools to room temperature because of the change in solubility indicated by the solvus line FH, some excess /? is precipitated from the
»
Primary a 0A-20B
solution. The process of solidification and the cooling curve for this alloy
0%
are shown in Fig. 6-27.
microstructure at various temperatures and the cooling curve for this alloy are shown in Fig. 6-28. Solidification starts at T2 and is complete at Ts, the resultant solid being a homogeneous single phase, the a solid solution. At point M the solution is unsaturated. The solvus line FH, as
remaining liquid (40 percentl is at
pqg[tion. tojorm the eutectic mix-
nately crystals of a and
of the
e eutectic temperature line (points
f f
AJloy 4, 85/A-156, follows the same process as described for alloy 1. The lumbers by applying Rules I and II
I
explained previously, shows tht fiecj asej
i
i
de-
creasing temperature. s the alloy cools, the solvus line is reached at point
giqp untiLsolidification is complete, e appears as shown in Fig 6-26.
N
.
.
fhe a solution is now saturated in B. Below this temperature, under
conditions of slow cooling, the excess 6 must come out of solution. Since A is soluble in 6, the precipitate does not come out as the pure metal 6,
but rather the /3 solid solution. At room temperature, the alloy will consist
largely of a with a small amount of excess [3, primarily along the grain
boundaries (Fig. 6-28). The student should determine the kmount of excess p by applying the lever rule at the line HJ (Fig. 6-24). ' If the /3 phase is relatively brittle the alloy will not be very strong or duc,
I
tile. The strength of an alloy to a large extent is determined by the phase that is continuous through the alloy. In this case, although the f3 solution constitutes only about 5 percent of the alloy, it exists as a continuous network along the grain boundaries. Therefore, the alloy will tend to rupture
along these boundaries. This alloy, however, may be madd to undergo a i
V
r
i 186
Cooimg;
V
f
'
INTRODUCTION TO PHYSICAL METALLURGY
a
-
Liquid solution (15/7)
curve
700
Alloy 600
Liquid
Liquid solution(40/9) 500
Alpha solid solution (5/9)
-
400
Liquid + a
I Liquid + a
Alpha solid solutiondS/?)
Liq
-
300
5
?00
CD
Eutectit -
Alpha solid solut ion (105)
M
100 eutectic
Beta solid solufion(955)
a + 13
0
0
Ht 10
Time
LiJ
30
20
10
100 Pb
20
30
40
I
l
1
2 3
Percent B
Fig. 6-28
The cooling curve and microstructure at various
temperatures for an 85/A-15S alloy. ' -
.
'
"
: ; :V
'
significant, change in strength and hardness after being properly heat- E treabd.
m
The lead-antimony equilibrium diagram and photomicrographs of var- «
ious alloys;in this system re shown in Fig. 6-29. Alloy 1 (Fig. 6-29ti)
,
taining 6.5 percent antimony
,
con- m
illustrates a typical hypoeutectic structure Jr
of primary a dendrites (black) and the eutectic mixture filling the spaces W between thp dendrites. Alloy 2 (Fig. 6-29c), containing 11.5 percent anti- temony, consists entirely of the eutectic mixture of a and B solid solutions. K
To the right of the eutectic composition, the alloys consist of primary/j * (white) surrounded by the eutectic mixture (Fig. 6-29d) and differ only in
f (b) Alloy 1
the relative amounts of the phases present. The amount of the eutectic mixture decreases as the alloy composition moves away from the eutectic composition.
The lead-tin equilibrium diagram and photomicrographs of various alloys in this system are shown in Fig. 6-30. Alloy 1 (Fig. 6-30b), containing 70 percent tin, is to the right of the eutectic composition. The microstructure consists of primary (3 dendrites (white) surrounded by the eutectic mixture.
„
a
.
& '
Alloy 2 (Fig. 6-30c) is the eutectic composition and consists entirely of a
very fine mixture of a and p solid solutions. Alloys 3 and 4 (Fig. 6-30cy and , | e), containing 60 and 50 percent tin, respectively, consist of dendrites of the lead-rich primary a solid solution (black) surrounded by the eutectic mixture, the amount of a increasing as the alloy composition moves to the left. 1 Notice the similarity of the photomicrographs shown in Figs. 6-22, 6-29, and 6-30.
; \d) Alloy 3 i
H
40
50
Weight percent an
l
I PHASE DIAGRAMS
187
f
V ; 1
.
I
70C
1200
s I
600
Liquid
Liquid solution (40 )
/- 1000
If'
500
Alpha solid solution (55)
300
400
3
Liquid + a
.Alpha solid solution(l55)
Liquid + j3
a
600
300
-
400
200
Eutedic + /3
y
.
2
Alpha solid solution (\0B]
;iPt*r-Btfa solid solutionOS
100 -+ %
1
-
200
eutectic iuj
1)
I 0 '
-
0 lOOPb
5
4
30
20
10
40
50
60
70
80
90
Weight percent antimony
100 0
i i 8 f
2 3
[a]
-S
i
j-dness
after being properly heat-
f
ft
1
II
m and photomicrographs of varrig 6-29. Alloy 1 (Fig. 6-29b), con-
* „
| *
.
3 a typical hypoeutectic structure jg f '
5Utectic mixture filling the spaces
.
"
y
?9c), containing 11.5.percent anti-
nixture of a and ft solid solutions. n,
'
.'
l:; f-
m
the alloys consist of primary ft . )
ure (Fig. 6-29d) and differ only in o:- :3ent.
The amount of the eutectic
m
(C) Alloy 2
(Z>) Alloy 1
M
ion moves away from the eutectic
1 1
US
hotomicrographs of various alloys Uloy 1 (Fig. 6-306), containing 70 composition. The microstructure rrounded by the eutectic mixture.
ill
i
Josition and consists entirely of a: ;
mhs. Alloys 3 and 4 (Fig. S-30d and., | sctively, consist of dendrites of the surrounded by the eutectic mixJoy composition moves to the left. Yaphs shown in Figs. 6-22, 6-29,
i
.
;
Fig. 6-29 (a) The lead-antimony equilibriunf>diagram. (b) 6.5 percent antimony alloy, 75X. (c) Eutectic alloy, 11.5 percent antimony, 250X. (cO 12.25 percent antimony alloy, 250X. (American Smelting and Refining Company.) {
1 i i
a
si
188
INTRODUCTION TO PHYSICAL METALLURGY
350
250
v is a linear relationship between the structure and the alloy compositio seem to indicate that the physical a system should also show a linear \ this Ideal behavior is rarely found. depend upon the individual charac
500
Q
i
(15 Properties in Eutectic Alloy Systems
oOO
300
a + L
.
I 200
400
9
.
19,2
91
97.5\
51.9
phases are distributed in the micro
150 JOO
eutectic alloy systems. Strength, h< size, number, distribution, and pro In many commercially important eu tively weak and plastic while the ot
a + /3 100
3 Pb
10
20
30
40
50
60
70
I
ill
4
3 2
80
90
100
1
Weight percent tin
As the eutectic composition is api there will be an increase in the stn
crease in strength beyond the eute(
1» .
- :
the am ounf of the smalTeutecffc '
"
i
.
amount of the proeutectic brittle pi the eutectic composition will gene illustrated in Fig. 6-31, which shov elongation for cast aluminum-silic silicon. The tensile strength show composition. The aluminum-silicc Another important point is that th nearly resemble those of the phase that forms the background or matr
X . -
5 .
x
>
1
r
.
mm. I
(*) Alloy 1
are imbedded. The eutectic mixture
(
continuous, since it is the last liqui
PC
iff
grains. It is generally true that the portion in the eutectic mixture will
iT5. 5.
28
2
-
24,000
Tensile strength -
H 20,000 £ 16,000
if. fa
£
24 o
20
S
iG
12 000 ,
w
Elongotion
12
S
8
£
4
(rf) Alloy 3
(e) Alloy 4 (Al) 2
Fig. 6-30 (a) The lead-tin equilibrium diagram (6) 70 per-
4
6
8
10 12 14
.
Percent silicon
cent tin alloy, (c) Eutectic alloy [d) 60 percent tin alloy. ,
(e) 50 percent tin alloy. All photomicrographs at 200X (From H Manko, "Solders and Soldering," McGraw-Hill Book .
.
Company, New York 1964.) .
:
i!
i
Fig. 6-31 Variation of typical properties of cast alum
t
Silicon alloys up to 14 percent silicon. (From Guy, "E ments of Physical Metallurgy, Addison-Wesley Publi Co., Reading, Mass., 1959.)
'
"
HISS
4 PHASE DIAGRAMS
1
6-15
600
189
Properties in Eutectic Alloy Systems It was shown in Fig. 6-20 that there V is a linear relationship between the constituents appearing in the microstructure and the alloy composition for a eutectic system. This would
seem to indicate that the physical and mechanical properties ofji eutectic
500 k1-
system sjiould also, show a.linear variation. In actual practice, however, this ideal behavior is rarely found. The properties of any multiphase alloy depend "upon the individual characteristics of the phases and how these phases are distributed in the microstructure. This is particularly true for eutectic alloy systems. Strength, hardness, and ductility are related to the size, number, distribution, and properties of the crystals of both phases. In many commercially important eutectic alloy systems one phase is relatively weak and plastic while the other phase is relatively hard and brittle. As the eutectic composition is approached from Ihe plastic-phase side
v
.
400
a
.
61,9
3
97.5
300
s
m
eo '
70 i
i
3 2 nl fin
I
80
90
100 Sn
'
,
,
there will be an increase in the strength of the alloy. There will be a de crease in strength beyond the eutectic composition due to the decrease in
the amourirbf the small eufecTilTparticies and the increase in size and
amount of the proeutectic brittle phase. Therefore, in this kTnd oTsystem '
'
"
the eutectic composition wllT generally show maximum strength. This is illustrated in Fig. 6-31, which shows the variation in tensile strength and
Siia
s;.-v -tfX.:'* /' 7a (c) Alloy 2
I 4
elongation for cast aluminum-silicon alloys containing up to 14 percent silicon. The tensile strength shows a maximum at very near the eutectic composition. The aluminum-silicon phase diagram is given in Fig. 6-21. Another important point is that the resulting properties of a mixture most nearly resemble those of the phase which is continuous-that is, the phase that forms the background or matrix in which particles of the other phase are imbedded. The eutectic mixture is always the microconstituent which is
continuous, since it is the last liquid to solidify and surrounds the primary grains. It is generally true that the phase which makes up the greater proportion in the eutectic mixture will be the continuous phase. If this phase
28
24,000
Tensile strength -
f 20,000 1 16,000 i
24 |
20 I 16
12 000
"r-J
12
,
8
Elongation
a,
a
4
[e] Alloy 4 (A!) 2
4
6
8
10 12 14
Percent silicon
Fig. 6-31 Variation of typical properties of cast aluminumsilicon alloys up to 14 percent silicon. (From Guy, Elements of Physical Metallurgy, Addison-Wesley Publishing "
"
Co., Reading, Mass., 1959.)
!
I
\,
'. -
-
i
-
vs. v, -
190
INTRODUCTION TO PHYSICAL METALLURGY
is plastic, the entire series of alloys will show some plasticity If this phase is brittle, the entire series of alloys will be relatively brittle. .
a so |-
In additbn to the above factors, an increase in cooling rate during freez- Jl
D
ing may result in a finer eutectic mixture a greater amount of eutectic mix-
m 80
ture, and smaller primary grains which in turn will influence the mechanical properties considerably.
I 75
,
,
>
7
,
f 70
Age Hardening There are only two principal methods for increa§\ng the *
strength and hardness of a given alloy: coldjvorking or heat treatment
Tj (100°
f 85
.
Tj (2l0of i
fi5 0 1
.
1
The most; important heat-treating process for nonferrous alloys is age "
hardening, or precipitation hardening
10
100
100
Time, in hours
In order to apply this heat treat-
.
320F
Fig. 6 33 Effect of temperature on the aging curve: precipitation hardening. Curves are for a 0.06 perce
.
id4 ment, the-equilibrium diagram must show partial solid solubility ancnhe ,
bon steel. (After Davenport and Bain.)
slope of the solvus line must be such that there is greater solubility at a > higher temperature than at a lower temperature These conditions are .
satisfied by Fig
.
up stresses which often result i intricately designed. In such ins
6-24. Let us consider the left side of the diagram involving j .
the a soli'd solution. NKlloy compositions that can be age-hardened are i' usually chosen between point F containing 20 percent B and point H contkining 10 percent 6. Commercially most age-hardenable alloys are
quench medium to minimize dist will be ductile immediately after torted parts to be straightened ea be carried out as soon as possil is shown schematically in Fig. 6v
I
,
t
chosen of compositions slightly to the left of point F, although the maxi- jp mum hardening effect would be obtained by an alloy containing 20 per- 1 cent B. The phase which is dissolved may be a terminal solid solution, as in this case, or an intermediate alloy phase. Two stages are generally required in heat treatment to produce age ft hardening: solution treatment and aging. Solution Treatment If alloy 4 (Fig. 6-24, microstructure shown in Fig. 6-32a) is reheated to point M all the excess p will be dissolved and the JJP structure will be a homogeneous a solid solution. The alloy is then cooled .
Aging Process .
of solution, vfhe speed at which ture.
rapidly (quenched) to room temperature. supersaturated solution results, l
that no appreciable precipitation due to rapid-diffusion, but soften
with the excess
in a lower maximum hardness. T
trapped in solution. The quench is usually carried out %
at which maximum hardening oc
in a cold-water bath orTay a water spray. Drastic quenching tends to set*!*-. 8 particles [actually submicroscopic)
0 a
a
Those alloys in which precipita so that they obtain their full stren -are known as natural-aging allc
'
a
groins
temperaturesJo dev€ alloys. However, these alloys als
to elevated
a
_
_
3 a
a
tureTthe rate and extent of the s
a
a
*
a a
v.
0
.
(see Table 6-2). Refrigeration ret; beginning of the aging process i
a
a
-
3
a
(-50 to -100oF) retards aging foi fact in the aircraft industry whe
a
a
0 [a) Fig. 6-32
Figure 6-33 shows the efl
curves of an iron alloy. At the lov\
,
.
The alloy, as qi
tion and is in an unstable state.
[t>)
age at room temperature, arejg
ic)
are driven. The rivets have previo
Microstructure of an 85/4-15B alloy, (a) After slow
phase they are very ductile. After
cooling; (b) after reheating and rapid cooling to room temperature; (c) after aging.
temperature, with a resulting incn A
9
i
.
iiiLWinuii ;
WW
mm
-
A
)
r
.
v
.
PHASE DIAGRAMS
I
191
:
k II show some plasticity If this phase .
.
11 be relatively brittle
.
90
i
.
6
v;:-hcrease in cooling rate during freez-
13
:
'1 -
nre, a greater amount of eutectic mix-
f 0
m 80
h in turn will influence the mechani75
fincipal methods for increg&mg the
\y: cold working or heaUreltment Vcess for nbnferrous alloys s age
65 3 1
.
,
Fig. 6-33
:
ie left side of the diagram involving
,
as
I
Effect of temperature on the aging curves during
%
Is
„
,
.
up stresses which often result in distortion, especially if the parts are intricately designed. In such instances, boiling water may be used as a quench medium to minimize distortion. If a is a ductile phase, the alloy will be ductile immediately after quenching. This allows warped or dis-
,
ling 20 percent B and point H conmost age-hardenable alloys are left of point F, although the maxi)ed by an alloy containing 20 perlay be a terminal solid solution
1000
.
sns that can be age-hardened are
-
100
precipitation hardening. Curves are for a 0.06 percent carbon steel. (After Davenport and Bain )
,
that there is greater solubility at a These conditions are
r.i
10
Time, in hours
' In order to apply this heat treat:¥;iow partial solid solubility ancK h e imperature.
(32°F)
* 70
i
torted parts to be straightened easily. The straightening operations should
be carried out as soon as possible after quenching. The microstructure '..g g
\
is shown schematically in Fig. 6-32£».
n heat treatment to produce age
Aging Process The alloy as guienched, is a supersaturated solid solu.: The excess solute will tend to come out ' , tion and is in an unstable state. stat
24, microstructure shown in Fig
ture.
ase
.
.
_
_
_
.
-
i
.:
-
.
llii'B
"-
-
-
._
_
_
.
.
_
of solution. "-The speed ait which precijjjtation occursj/arlesjadlb tempera-.
.
"
jsolution. The alloy is then
ji .
that no appreciable precipitation occurs. At T3, hardening occurs quickly, .- Ml
cooled
supersaturated solution results
due to rapid diffusion, but softening effects are also accelerated, resulting
,
in a lower maximum hardness. The optimum temperature seems to be T21 : ||
The quench is usually carried out Drastic quenching tends to set -
at which maximum hardening occurs within a reasonable length of ftme. ;: p
Those alloys in which precipitation takes place at room temperatureso that they obtain their full strength after 4 or 5 days at room temperature
S particles (actually submicroscopicl
Figure 6-33 shows the effect of three temperatures on the aging
curves of an iron alloy. At the low temperature the diffusion rate is so slow / 'ifw
xcess j8 will be dissolved and the
-are known as natural-aging alloys
.
a grains
Those alloys wh[ch require reheating
i
to elevated temperatures to develop their full strength are artificial-aging -
alloys. However, these alloys also age a limited amount at room temperatureTthe rate and extent of the strengthening depending upon the alloys (see Table 6-2). Refrigeration retards the rate of natural aging. At 32°F, the beginning of the aging process is delayed for several hours while dry ice
a
m
:|p||
,
(-50 to -100
F) retards aging for an extended period.(jJse is made of this which nornially
-til |||
°
fact in the aircraft industry when aluminum-alloy rivets
,
age at room temperature, are kept in deep-freeze refrigerators until they . ?(p
are driven. The rivets have previously been solution-treated and as a single ,
phase they are very ductile. After being driven, aging will take place at room
C
'
temperature, with a resulting increase in strength and hardness In the early .
1
.
I
'
:
-
.
.
192
%
INTRODUCTION TO PHYSICAL METALLURGY
i TABLE 6-2
Effect of Aging on Properties of Aluminum Alloy 2014 (3.5 to 4 5 Percent Copper)
ALLOY AND CONDITION Annealed I
ULTIMATE
YIELD
STRENGTH,
STRENGTH,
PSI
PSI
27,000
14,000
ELONGATION, % IN 2 IN. 18
BHN, 500 KG, 10 MM 45
I I
. Solute atom
|
ooo»ooooooo
o
o
c
ooooooooo*
o
o
c
o
c
I
.
SHEAR
B
strength!
psi
i
18,000
o
I
Solution-treated,
naturally aged
i
62,000
42,000
105
20
38,000
Solution-treated,
artificially aged
70,000
60,000
13
135
42,000
. oooooooooo
o
ooooo»oo»oo
ooo
ooooooooooo
0
o«ooooooooo
ooo
o
oooooo»ooo
O O 0
o
o
oo»oooooo
o
o
o
o
o
theory of the aging process, it was thought that the excess phase comes M
o
o
ooo
o
,
o
o
c
o
o
o
.
o
o
o
c
o
o
o
o
o
o
o
o
.
<
o
o
.
o
o
o
o
o
o
(<7)
Equilibrium precipitate o«ooooooc
thus
ooooooooc
increasing strength and hardness.
o
o
Subsequent studies have led to a more complete understanding of the age-hardening process. The strengthening of a heat-treatable alloy by *
O
O 1 .
o
.
o
.
o
.
c
o
o
. I o 0 I . 10 ,
o . o .
. o . o
o . o .
c < <
o
»
O
aging isjnot due merely to the presence of a precipitate. It is due to both the uniform distribution of a finely dispersed submicroscopic precipitate
o
o
"0
_
__
o oL»
_
o
o
__
_
'
p
__
o
and the distortion of the lattice structure by those particles before they
oooooo
o
o
reach a visible size.
o
tion, with the solute atoms distributed at random in the lattice structure
,
Fig. 6-34a.
During an incubation period, the excess solute atoms tend to
migrate to certain crystallographic planes, forming clusters or embryos of the precipitate. During aging, these clusters form an intermediate crystal structure; or transitional lattice, maintaining registry (coherency) with the
o
c _
o
o
o
o
< jp
o
It is not possible to state definitely in what manner precipitate particles harden the matrix or solvent lattice. While there are several theories of precipitation hardening, the most useful is the coherent lattice theory. After solution treatment and quenching, the alloy is in a supersaturated condi-
m
0
o
_
v
0
o
out of solution as fine submicroscopic particles, many of which fall on the p slip planes (Fig. 6-32c). These particles were considered to have a keying action, thereby interfering with movement along planes of ready slip
Transit
o Solvent atom
c
o
o
o
o
o
o
(
ooo
o
o
o
o
0
1
o
ic)
fig. 6 34 The stages in the formation of an equil piecipitate. (a) Supersaturated
solid solution. (6)
lattice coherent with the solid solution, (c) Equilit
precipitate essentially independent
of the solid sc A(
(From Guy, Elements of Physical Metallurgy "
Wesley Publishing Company, Inc., Reading,
"
Mass
then it increases when precipit;
on some properties is shown in
lattice structure of the matrix. The excess phase will have different lattice
Aging does not have the san
parameters from those of the solvent, and as a result of the atom matching (coherency), there will be considerable distortion of the matrix. Fig. 6-34b.
some alloys the change in hare
The distortion of the matrix extends over a larger volume than would be the case if the excess phase were a discrete particle. It is this distortion that
cess of the saturation limit, bi
interferes with the movement of dislocations and accounts for the rapid increase jn hardness and strength during aging (see Fig. 6-35). Eventually the equilibrium excess phase is formed with its own lattice structure (Fig. 6-34c) This causes a foss of coherency with the matrix and less distortion.
eutectic temperature but only
,
.
.
Hardnes and strength will decrease, and the alloy is "over-aged." There will now'be a boundary between the excess phase and the matrix so that the precipitated particle will be visible under the microscope. Electri'
cal conductivity decreases during aging because of lattice distortion;
m i
1
the changes may be large. Thi: distortion.
For example, magr
propert heat treatment, precip occurs. This is due to the ab
localized distortion durfng pre ciplTation involves extensive ch tortion occurs with wide chan the amount of metal dissolved the distortion produced and th
PHASE DIAGRAMS
193
314 (3.5 to 4 5 Percent Copper) .
. Solute atom
JGATION, '
2 IN.
BHN, 500 KG,
STRENGTH, i
10 MM
PSI
SHEAR
45
18 000 ,
105
38,000
135
.
42,000
yXS: that the excess phase comes Teles, many of which fall on the
o o o o . o o\o o
oooooooooo*
O
O
O
0
o
o
o
o
o
.
o
o
o
o
o
o
o
o
oooooooooo
OOOOOO
ooooooooooo
oo o[« o
09000000000
o o o lo
*
o
*
ooooooo«ooo
b o o
o
*
o
o
o
long planes of ready slip
,
thus
-
-
o
o
o
o
o o
o
Oj o o o o
ooooo»oo»oo
.jo o 0 Oj O 0 0
.
jO o o
ooooooooooo
o o o o
o
o
o
o
o
000000000*0
o a o o
o
o
o
o
o
o
*
000*0000000
o
.
o
o
o
o
o
o
o
o
o
00
o
o
0
o
o
o
o
o
o
o
o
0
000*0000
(<7)
"
e considered fo have a keying
Transition lattice
o Solvent atom
ooo#ooooooo
m
4 Equilibrium precipitate
m
0*000000000
OOOO
i
0 0
0
OOOOOOO
OtO
060000
"
Dmplete understanding of the of a heat-treatable alloy by
o o[.
\ precipitate. It is due to both
o
01*
o
o
o
o oI. 1 o
d submicroscopic precipitate
}y those particles before they i
manner precipitate particles
0
*
*1 o ol
o
o *
O
*l o
o
o
*
oi « „ *| o o
*
0
0 o
Sj o o
000000
00000
000000
o
o
o
o
o
000000
o
o
.
o
o
OOOOOO
00000
10
*
o
o
1
i;
I
[c]
f
ere are several theories of prezoiierent lattice theory After
Fig. 6-34 The stages in the formation of an equilibrium precipitate, (a) Supersaturated solid solution, (b) Transition lattice coherent with the solid solution, (c) Equilibrium precipitate essentially independent of the solid solution. (From Guy, "Elements of Physical Metallurgy," AddisonWesley Publishing Company, Inc., Reading, Mass., 1959.)
.
is in a.supersaturated condi1ndom in the lattice structure
,
excess solute atoms tend to :
,
orming clusters or embryos yvv3 form an intermediate crystal
'
s
Aging does not have the same effect on the properties of all alloys. In some alloys the change in hardness and strength may be small; in others the changes may be large. This is not due to the amount dissolved in ex-
jer volume than would be the icie. It is this distortion that
cess of the saturation limit, but to the effect of the precipitate on lattice
. and accounts for the rapid
|;g (see Fig. 6-35). Eventually,
eutectic temperature but only 2 percent lead at room temperature. With propert heat treatment, precipitation takes place but no age hardening
te own lattice structure (Fig
occurs. This is due to the absence of a transition lattice and very little
,
.
.
;ne
matrix and less distortion There )hase and the matrix so that .
lalloy is "over-aged
"
.
.
ir the microscope. Electribause of lattice distortion;
7
on some properties is shown in Fig. 6-35.
i result of the atom matching
:;
mi.-
then it increases when precipitation takes place. The effect of aging time
ase .will have different lattice
Fig. 6-34i).
'
"
-registry (coherency) with the
,
'
'
/
tion of the matrix
1
<
distortion. For example, maignesium can dissolve 46 percent lead at the 5
localized distortiorTduring precipitation. On the other hand, if the precipitation involves extensive changes in the lattice, a large amount of distortion occurs with wide changes in properties. In this case, the greater the amount of metal dissolved in excess of the saturation limit, the greater
the distortion produced and the greater will be the effect on hardness and
m
:
,
,
5' ''
194
INTRODUCTION TO PHYSICAL METALLURGY
Over
Aging
6 17 Type IV-The Congruent-meltin changes into another phase is without any change in chemici phase change or congruent tra
aging
-
Hardness
Incubation
period Tensile strength
gruently. We have previously se
i
as a variation of a Type I phase c a liquid phase to a single solid change in composition, and it i Q
termediate phases are so namec between the terminal phases on formation of an intermediate p
.
Electncal
conductivity
Percent
will cover the incongruent melt phase may be treated as anoth
elongation
intermediate phase has a narro\
compounds and interstitial con Yield strength
gram as a vertical line and labe
IO-2
I02 Time,
lO4
106
pound.
108
usually an electron compound i years some authors tend to us( In Fig. 6-36, the intermedial
hours (logorithnic scale)
Fig. 6-35 Effect of aging time on properties. (By permission from L. F. Mondolfo and 0. Zmeskal, "Engineering Metallurgy," McGraw-Hill Book Company, New York 1955.)
Since it is a compound, it is in
,
scripts which indicate the num For example, magnesium and the chemical formula MgjSn. In lent to m; tin, Sn, is equivalent
strength. This is illustrated for some copper-beryllium alloys in Table 6-3.| Figure 12-16 shows the copper-rich portion of the copper-beryllium alloy;; system.
Notice that copper dissolves up to a maximum of 2.2 percent
beryllium at 1600oF.
If the intermediate pha
At room temperature the solubility of beryllium m'
It is apparent from Fig. 6-36 i
copper is approximately 0.2 percent. Referring to Table 6-3, while thereis|
two independent parts, one to s
no difference in strength and hardness of both alloys after solution anneal-j
ing (solution treatment), there is a difference after aging. The alloy containj
Melting point
of 4mS„
Ing 1.90-2.15 percent beryllium is closer to the maximum solubility oj|
Melting point
beryllium in copper, and upon aging more of the y (gamma) phase will b8|
XoM
precipitated, resulting in a greater hardness and strength as shown. TABLE 6-3
Alloys
V
Melting po
Effect of Aging and Composition on the Mechanical Properties of Some Copper-Berylllum|
of 0 2
ALLOY AND CONDITION
TENSILE STRENGTH,
ELONGATION,
ROCKWELUl
PSI
% IN 2 IN.
hardness!
CL
1 90-2.15% Be: .
Solution-annealed
After aging 1 60-1.80% Be: .
35-50
B 45-65
5-8
C 36-40
60,000-80,000
35-50
B 45-65
150,000-165 000
5-8
G 33-37
,
4
i
Solution-annealed
After aging
60,000-80,000 165 000-180,000
i
Composition
,
Fig. 6-36 Composition and melting point of pure |
;
and
a compound AmBn.
PHASE DIAGRAMS
Over - aging
6-17
Si
195
Type IV-The Congruent-melting Intermediate Phase When one phase changes into another phase isothermally (at constant temperature) and without any change in chemical composition it is said to be congruent ,
phase change or congruent transformation. All pure metals solidify congruently. We have previously seen an example of a congruent-melting alloy as a variation of a Type I phase diagram. The alloy x in Fig. 6-14a goes from a liquid phase to a single solid phase at constant temperature without a change in composition, and it is therefore a congruent-melting alloy. Intermediate phases are so named because they are single phases that occur between the terminal phases on a phase diagram. Type IV will consider the formation of an intermediate phase by congruent melting while Type V will cover the incongruent melting intermediate phase. Any intermediate phase may be treated as another component on a phase diagram. If the intermediate phase has a narrow range of composition as do intermetallic compounds and interstitial compounds, it is then represented on the dia-
t
M
,
,
gram as a vertical line and labeled with the chemical formula of the comJ
-
pound. If the intermediate phase exists over a range of composition, it is
_
108
.
beryllium alloys in Table 6-3
usually an electron compound arid is labeled with a Greek letter. In recent years some authors tend to use Greek letters for all intermediate phases. In Fig. 6-36, the intermediate alloy phase is shown as a vertical line. Since it is a compound it is indicated as AmBn, where m and n are sub-.
i ill If
,
:
scripts which indicate the number of atoms combined in the compound.. For example, magnesium and tin form an intermediate phase which has the chemical formula Mg2Sn. In this case Mg is equivalent to A; 2 is equiva-
.
)f the copper-beryllium alloy
,
a maximum of 2.2 percent
Vie
lent to m; tin, Sn, is equivalent to 6; and n is equal to 1. It is apparent from Fig. 6.36 that the A-B system may be separated into two independent parts, one to show all the alloys between A and the com-
solubility of beryllium in
ig to Table 6-3
,
while there is
alloys after solution anneal-
ter aging. The alloy contain'
he y (gamma) phase will be id strength as shown
.
[4
Melting point
f the maximum solubility of
of AmBn
\
-Melting point
I
i AU
Melting point
operties of Some Copper-Beryllium
n
O
LONGATION IN 2 IN.
,
1
ROCKWELL HARDNESS
.
A r5 50
B 45-65
i-8
C 36-40
-
50
B 45-65
8
C 33-37
-
-
-us
B
Composition
Fig. 6-36
Composition and melting point of pure A, pure S,
and a compound AmBn.
if
1
,
i
P
1
r
196
INTRODUCTION TO PHYSICAL METALLURGY
(21
pound AmBn and the other to show those between AmBn and 6. The portion I
of the diagram between A and AmBn may be any of the types studied in this |
5
chapter; similarly for the portion between AmBn and 6. If the compound shows no solubility for either pure metal and the pure metals show some
,
solubility for each other) the equilibrium diagram will be as shown in Fig. sj 6-37
.
equations rrjay be written as follows.
i
i
-
L + /I
This diagram shows two different eutectic mixtures. The eutectic
At T,:
Liquid
cooling
F
ir
a + AmBn
1
cooling
Liquid
1
D
G H
F
I
heating
At r2:
h
If.
L
heating ]
.
The study of many actual systems that show the formation of several congruent-melting intermediate phases may be simplified by the above
r
>
approach. [ G-IS ype V-The Peritectic Reaction In the peritectic reaction a liquid and \
'
solid react isothermally to form a new solid on cooling. The reaction is expressed in general as
a
Liquid + solid, .
cooling
/M,
20
10
A
new solid2
-
i 1 '
heating
Phase diagram showing the formation o
6-38
coHgruent melting intermediate
Melting point of AmBn
Liquid
40 Compositi
phase by a peritect
reaction.
the new solid formed is usua
'
in some cases it may be a termir Consideration of Fig. 6-38 sh when heated to the peritectic tei
L + Antti +
phases, liquid and solid A. Th
rt
a
217° F 5200 L
+ L
2800 3
+
3
+
e 2400
12%
2I650F
P
Q
+
/3
2000
a + G
1600 P
A Fig. 6-37
\
Composition
Equilibrium diagram illustrating an intermediate
alloy which is an intermetallic compound. : -
m
...
i ....
.
AmB„
J. i
tr-<
B
Fig 6-39
20
45%
30 40 50 6C Composilion -percenl
Silver-platinum alloy system showing tt
tion ol a terminal solid solution by a peritectic
re;
. -
uiwrnm*."
PHASE DIAGRAMS
(2)
(1) .
197
tween AmBn and B. The portion any of the types studied in this
AmSn
.
5
and S. [Hhe compound
idjhe pure metais sh ow some igram will be as shown in Fig .
Liquid
.
.
+ /I
jtectic mixtures. The eutectic
1
%
0
|+ AmBn
3
V
1 m
L+ 8 "
:f
-
J
low the formation of several
E
K
.
'
/ be simplified by the above AmB„ + B
tactic reaction a liquid and d on cooling. The reaction is
.
'
~
' -
v : '
' .
.
A
AmBn
20
to
3W solid2
40
'
iJ
50
60
70
30
9C
B
Composition weight percent B ,
Fig. 6-38 Phase diagram showing the formation of an incongruent melting intermediate phase by a peritectic
ig point
:
reaction. t
the new solid formed is usually aninterme ate phase (Fig. 6-38), but
iSi
'
in some cases it may be a terminal solid solution (Fig. 6'39). Consideration of Fig. 6-38 shows that the compound AmBm 70-4-30S, when heated to the peritectic temperature, point G, decomposes into two phases, liquid and solid A Therefore, this is an example of an incon'
1
I m
32170F
.:|
J200
1
iquid
2800 4
CP
+
II
S 2400
1
12%
2165°F
2000 +
13
P
59%
I760°F
+
46%
1600 Pt
B
10
20
30
40
50
60
70
80
90
Ag
i
Composition-percent silver -
Fig. 6-39 Silver-platinum alloy system showing the formation of a terminal solid solution by a peritectic reaction.
t '
yy->
.
j
I
i 198
gruent-mglling intermediate alloy.
1
r
INTRODUCTION TO PHYSICAL METALLURGY
The student should realize that tho 1
peritectic reaction is just the reverse of the eutectic reaction, where a -A
single pnase rormed two new phases on cooling.
The liquidus line is |
B
TADETB and the solidus line is TATpGJTETB. The peritectic-reaction line is |
A
TpD. Notice that only part of this line the length TpG, coincides with the solidus line. The slow cooling of several alloys will now be studied.
A
Alloy 1, 90/4-106, remains liquid until the liquidus line is reached at Tv Solidification now takes place by forming crystals of the pure metal A As the temperature falls, the liquid is decreasing in amount, and its composition is moving down along the liquidus line. Let us examine the conditions that exist just above the peritectic temperature T?:
Liquid
,
1
.
Phases Composition .
Relative1 amount
Liquid
Solid A
60/A-40S TPF
100 A
-x100 =
!
Envelope of kmB„ increases in thickness by diffusic atoms outward and S atoms inward.
the compound/!\men. Figure 6-41
%I
FD 25%
TpD
;
Fig. 6-40 Schematic picture of the peritectic react
x 100:
75%
TpD
r
V
The conditions that exist just below the peritectic temperature are: Phases
A„Bn
Solid A
Composition
70A-Z0B
100 A
T pF
FG
Relative amount
TpG
excess A remaining after the peri
alIoy composition is to the comp A will remain.
Alloy 2, 65A-35e, solidifies pu
T2, and as solidification continue point H is reached, the liquid co
x 100 =
-- x 100 = 33%
-
peratures in the slow cooling of any alloy to the left of point G. Tl
67%
rule for this alloy, there is
TPG
A first glance at these two areas seems to indicate that the liquid has disappeared gt the horizontal line and in its place is the compound AmBn. Consideration of the chemical compositions shows that this is not possible. %
disappear in reacting with some Liquid solution (10 B) Alloy (I)
enough in A to form the compound by itself.Hfhe liquid must therefore re- i act with just the right amount of solid /A, in this case 8 percent, to bring its «|
compositicin to that of the compound AmBn. The following reaction must JP 60/A
Composition:
\
25%
Relative amount:
'
8%
Liquid
cooling
Liquid solut
*l
TK Pnmorv G
70/\
100/A
Liquid + solid A -
Equation:
'
solid A. Since the line GD is nol remain after the reaction takes pi
The liquid.contains 60/4, while AmBn contains 70A The liquid is not rich ,'| have taken, place at the peritectic temperature:
3S/aq x
D
AmBn (3(
solid AA 33%
W+AmBn
The reaction takes place all around the surface of each grain of solid A
where the liquid touches it. When the correct composition is reached, tha
Primary/
layer solidifies into AmBn material surrounding every grain of A. Further reaction is slow since it must wait for the diffusion of atoms through the
peritectic wall of AmBn in order to continue (see Fig. 6-40).
mi -
When diffusion
is completed, all the liquid will have been consumed, and since only 8 percent of pure Aw as required for the reaction, there will be 67 percent oi A leftTTFe final microstructure will show grains of primary/A surrounded by ,
_
1
.
"-
v;;:;::-;-i
r
.
A
IC
20
30
40
AmBn Percent B
Pig. 6-41 Slow cooling of a 9(M-10S alloy showir microstructure at various temperatures.
'
Ait
PHASE DIAGRAMS
'
199
m student should realize that the :
the eutectic reaction, where a cooling. The liquidus line is The peritectic-reaction line is
;H;n '
V/
:
,
.
M 4
e length TpG coincides with the
5
/I
.
,
alloys will now be studied
.
ie liquidus line is reached at Tv crystals of the pure metal A As
Liquid
I
.
'
ing in amount, and its composi-
\
i
Let us examine the conditions
.
jiture f
i-
'
-
the compound AmBn. Figure 6-41 shows the microstructure at various tem-
OA
i
.
)
X 100
0
atoms out-ward and B atoms inward.
Tp-.
-
Fig, 6-40 Schematic picture of the peritectic reaction. Envelope of AmB„ increases in thickness by diffusion of A
:
75%
5
peratures in the slow cooling of this alloy. The story will be the same for any alloy to the left of point G. The only difference will be in the amount of excess A remaining after the peritectic reaction is complete he closer the the less primary alloy composition is to thej qm£ositiqn qnhe _
I
ritectic temperature are:
A will remain.
Alloy 2, 65A-35B, solidifies pure A when the liquidus line is crossed at
)
72, and as solidification continues, the liquid becomes richer in B.
1
A
point H is reached, the liquid composition is 60A-40B.
J '
>>: : ; X 100 = 67%
i
'
IWhen
Applying the lever
rule for this alloy, there is 35/ao x 100 or 87.5 percent liquid and 12.5 percent
,
L
idicate that the liquid has dise is the compound iA 6 Conows that this is not possible m
solid A. Since the line GD is not art qHhe sqlidusJine, some liquid Must
i
remain after the reaction takes, place. It is therefore the solid A which must 1i
'
9
-
I
n.
--
disappear in reacting with some of the liquid to form the compound AmBn. j
t
.
V: ,
:
is 70A The liquid is not rich he liquid must therefore relis case 8 percent, to bring its
:
The following reaction must
ir .
-
V>
Liquid sojution(lOB) 1 1
Alloy (I)
i
1
iquid
Liquid solution(40B)-25%
_
4 -
4
70A
G
0
Plid AmBn
.AmBn
r
33%
(3081-33%
f+AmB.
ace of each grain of solid A i
rimary>«000/4)-67%
,
ig every grain of A. '
Further
fusion of atoms through the
>e Fig. 6-40).--When diffusion v . jsumed, and since onjy 8 per:;
"
there will be 67 percent of A
\ of primary A surrounded by
'
1
1
01
-
t composition is reached the
1
Primory/5(10CMl-75%
i
A
10
20
30
40
AmBn
I
Percent B
Fig. 6-41
Slow cooling of a 90A-10B alloy showing the
microstructure at various temperatures.
m
1
m
200
INTRODUCTION TO PHYSICAL METALLURGY
The same reaction takes place again: Composition:
60/4
Reaction:
The student should study the slow
100A
peritectic point P in the equilibrium
70-4 cooling
Liquid + solid
of a terminal solid solution by a per
AmBn
vthe amount of liquid entering into the above reaction may by applying the leyer rule below the reaction temperature GH
10 -x 100 20
Liquid (percent) = - x 1 00 DG
:
i
The peritectic reaction was descr actual practice, however, this condi phase forms an envelope around t
E
drance to the diffusion which is e
be determined
.
50 percent
Fig. 6-40). As the layer of the new distance increases, so that the reac
Since there was 87 5 percent liquid before the reaction and 50 percent liq-
ample, according to the phase diagi
.
) uid after the reaction, it is apparent that 37.5 percent of the liquid reacted | with 12.5 percent of solid A to give 50 percent of the compound A B at the j Reritectic temperature. As cooling continues the liquid now separates m
Fig. 6-39, a 60 percent Ag alloy sh( perature. The actual cast structui phase. The light areas are primary
n
,
crystals of A B m
n.
The liquid becomes richer in 6 and its composition grad,
ually moves down and to the right along the liquidus line until it reaches point E, the eutectic temperature. At this temperature there is only Vso
i
,
x 100 or 10 percent liquid left Since the liquid has reached the eutectic point, it now solidifies into the eutectic mixture of A B + S. This alloy at .
'
m
i
room temperature
n
,
will consist of 90 percent primary or proeutectic A
,
m
Bn i 0
surrounded by 10 percent of the eutectic {A Bn + B) mixture Figure 6-42 shows the cooling curve and the changes in microstructure at various m
.
points in the slow cooling of this alloy.
Liquid solution (35 B)
Liquid solution (40 5)-87 5% .
Primory /)(100/3) - 12 5%
Cooling
.
curve
Alloy \
Liquid
2
\
Liquid solution (40 B]- 50 % AmBn (30 51-50%
-
AmBn (30 5)-907o
>4
iquid solution (8051-10%
i
CO
i + AmRn
J
E
j-Euteclic mixture
AmBn + B
(80 B) - 10 %
J
-AmBn(30 51-90%
I
AmBn + B
Time
A
10
20
30
40
50
Am Bn
Percent B
Fig. 6-42
The cooling curve and the microstructure at
various temperatures during the slow cooling of a 65/1-350 alloy.
V I
!
I
60
70
80
Fig 6 43 40 percent Platinum + 60 percent silver c. alloy. Light areas are primary a; dark"two-toned are-
are 0. Magnification 100X. (By permission Phase
Diagrams in Metallurgy, pany, New York, 1956.)
"
from F. N McGraw-Hill Book
i
j
PHASE DIAGRAMS.' 201
The student should study the slow cooling pf alloys on either side of the peritectic point P in the equilibrium diagram that illustrates the formatibn of a terminal solid solution by a peritectic reaction (Fig. 6-39). The p eri tectic„. re.action was described under e ujllbrimp condj fli . In actuaj practice, however, his condition Is rarely attained. Since the new phase forms an envelope around the primary phase, it will act as a hindrance to the diffusion which is essential to continue the reaction (see Fig. 6-40). As the layer of the new phase becomes thicker the diffusion
'OA
.
)ve reaction may be determined
.
pa.tejnperature.
.
.
-
r
..
_
rcent
. .
,
distance increases
the reaction and 50 percent liq 5 percent of the liquid reacted
,
-
: :.
,
..
.
;
so that the reaction is frequently incomplete. For ex-
"
ample, according to the phase diagram of the silver-platinum alloy system,
.
Fig. 6-39, a 60 percent Ag alloy should be a single phase 13 at room tem-
y yi.nt of the compound A B at the ues, the liquj jioyy separates m
perature. The actual cast structure. Fig. 643, however, is not a single
n
phase. The light areas are primary a grains surrounded by thie dark two-
[in B, and its composition gradrie liquidus line until It reaches there is only 5/55
4
J temperature
-
,
iquid has reached the eutectic
|fure of AmBn + B. fhis alloy, at :
it primary or proeutectic A B B„ T S) mixture. Figure 6-42 m
m
6
n
m fm
m
,
in microstructute at various
iquid solution(40 5)-8y 5% .
-
IriiTary .4(100 )- 12 5%
«
-
,
I
iquid solution (40 51-50% -
I
AmBn (30 51-50% 1
.
AmBn (30 51 90%
I
-
V
iquid solution(8051-10%
x
vrry (-Eutectic mixtuie AmBn+ 5
f l (80 51-10%
I
mBn (30 51-90%
1
'
I :1 '
80
i
4
1
i
Fig. 6-43 40 percent Platinum +60 percent silver cast alloy, Light areas are primary a; dark two-toned areas are p. Magnification 100X. (By permission from F. N. Fthrnes, "
Phase Diagrams in Metallurgy," McGraw-Hill Book Com-
pany, New York, 1956.)
202
INTRODUCTION TO PHYSICAL METALLURGY
much /A. Therefore, what must occu solid A is precipitated f omj., to bn -
toned areas of [3, indicating that the peritectic reaction was not complete.;! Since peritectic alloys are usually two-phase mixtures, their mechanical"!
properties follow the same principles stated for eutectic alloys with twof differences: (1) the individual phases are more likely to be different thanl
predicted for equilibrium condition|, and (2) cast grain size
form L2. m
is usuallyl usually
coarse.
'
ap;:
6 19 Type VI-Two Liquids Partly Soluble in the Liquid State: The Monotecticf Reaction
S;
To prove this, we can apply Rule II line TM. "Abovejhe line we have 50 percent L,. Below the line, we hav percent solid A. Therefore, at the ho have formed 17 percent L2 and 33 pe
In all the types discussed previously, it was assumed that there
howeveif|fe
50 percent already existing, gives a
that over a certain composition range two liquid solutions are formed that ap!'
mined by the calculations. The reac
was complete soTubility in the liquid state. It is quite possible
,
are not soluble in each other. Another term for solubility is miscibility
.
Substances thai are not soluble in each other, such as oil and water
,
areiLi
Composition:
BOA 80A
40A cooling
said to be immiscible. Substances that are partly soluble in each other arep|;
Equation:
The equilibrium diagram for this type is shown in Fig. 6-44. The liquidus »} line is TACFETB, and the solidus line is TATgJTB. Alloys having compositions between point C and point F at a temperature just above TM will consist M
as a monotectic reaction; the genera
said to show a miscibility gap, and this is Type VI.
of two liquid solutions,
C
/When one liquid forms another liqi
and L2. The lines CD and FG show the composi- If
tion of the liquid phases in equilibrium, with each other at higher temperatures. In most cases, these lines are shown dotted because experimental
17%
50%
Relative amount:
Alloy
Alloy >
G
0
r
difficulties at high temperatures usually prevent an accurate determination of their position. Since these lines tend to approach each other, it is pos- i sible that at higher temperatures the area will be closed and a single homo-
geneous liquid solution will exist.
This area should be treated like any
L
' -
2
other two-phase area, and the same rules may be applied to determine the
chemical composition of L, and
perature.
jsaliqi
4+ 1 I
and their relative amounts at any tem-
c
solutionof B dissolved in A, whereas L2 \sa liquid
*2
solution of /A dissolved In B.
Let us study the slow cooling of several alloys. Alloy x containing 10
percent 6 is a single-phase liquid solution L,, and it remains that way until the liquidus line is crossed at x,. Solidification starts by forming crystals
re
and to the right along the liquidus line,
/3
5
of the pure metal A. The liquid becomes richer in S, gradually moving down
hen the alloy has reached the
monotectic temperature Jin J £t poJnt x2, the liquid composition is given _
by point CVwhich is 80A and 206. Thg horizontal line on any phase diagram
/I
indicates that a reaction must take place. What is the reaction in this case? Below the ifne, the two phases present are solid A and L2. Offhand, it seems
f that L, has disappeared and that in its place we have L2, but a more careful study of the diagram indicates that this could not have happened. The composition of L2 is given by point F, which is 40A and 60S, whereas L, has a i composition of 80A and 2QB; so although L, has disappeared, by itself it
\ could not have formed L2. Remember that L, is a liquid solution rich in A, whereas L2 is a liquid solution rich in 8. The problem is that L, has too
::v
-
;
:
m
i
'
L
A
10
20
30
40
5(
Composition, wei( fig 6-44 Hypothetical equilibrium diagram of two n paTtly soluble in the liquid state; the monotectic reac
m
PHASE DIAGRAMS
203
i
much /A.therefore, what must occur at the horizontal line is that enough
sritectic reaction was not complete. ; ;;.o-phase mixtures, their mechanical
.
solid A is precipitated from
%ated for eutectic alloys with two
to bring its composition to the right one to
form L2.
ire more likely to be different than
To prove this, we can apply Rule II both above and below the horizontal line TM. Abovejhe line we have 50 percent solid A (10/20 x 100) and 50
and (2) cast grain size is usually
\
percent L,. Belowjhejine, we have 17 percent L2 (10/60 x 100) and 83 the Liquid State: The Monotectic
percent solid A. Therefore, at the horizontal line, the 50 percent of must \ have formed 17 percent L2 and 33 percent solid A. The 33 percent, plus the 1 50 percent already existing, gives a total of 83 percent solid A as determined by the calculations. The reaction is summarized as follows:
eyiously, it was assumed that there ,
cooling
Equation:
pre partly soluble in each other are :
is shown in Fig. 6-44. The liquidus TpJTg. Alloys having compositions
100 A
Lz + solid A 17%
50%
Relative amount:
is Type VI. ;
4(M
80A
Composition:
H other, such as oil and water, are
|es CD and FG show the composiwith each other at higher tempera-
l/Vhen one liquid forms another liquid, plus a solid, oncooling, it js known as a monotectic reaction; the general equation for the monotectic reaction .
1
..
,
%)ral alloys. Alloy x containing 10
|?vin Lu and it remains that way until .
x
G
D
.
+
iI '
4+
, F
C /2
I
' A +
i
'
.
J
/3
(3
1
I
/henjhe alloy has reached the
1
Xj, the liquid composition is given jzontal line on any phase diagram
I
.
A
What is the reaction in this case?
ice we have L2, but a more careful
I
Tb
h
starts by forming crystals icher in B, gradually moving down
'
1
1
r
Tication
Ifllj solicM and Lj. Offhand, it seems
1
ftlloy
Alloy
'
to approach each other, it is posi will be closed and a single homo; area should be treated like any s may be applied to determine the heir relative amounts at any temsolved in A, whereas L2 is a liquid
:
-
.
jrevent an accurate determination
;,;;
I
'
:
own dotted because experimental
-
j
33%
erafure just above T„ will consist
f
i
'
rMif term for solubflity is miscibliity.
.
I
1 !
ate. It is quite possible however, no liquid solutions are formed that '
\
1
+
r
v juld not have happened The com-
.
.
"
v;x; js
40A and 60S, whereas
has a
h Li has disappeared by itself it at L, is a liquid solution rich in A ,
,
The problem is that L1 has too
A
10
20
30
40
50
60
Composition, weight percent B
Fig. 6-44 Hypothetical equilibrium diagram of two metals partly soluble in the liquid state: the monotectic reaction.
70
80
90
B
1 I 1
¥;1
204
INTRODUCTION TO PHYSICAL METALLURGY
may be written as.
mixture which is composed of L, r cooling
equation to form more of i-j + soll
/-2 + solid
heating
A is formed from L2, its compositio
Point C is known as the monotectic point. Hfshould be apparent that the
temperature is approached at poir L2 (37.5 percent) undergoes the ei
monotectic reaction resembles the eutectic reaction, the only difference
being that oneTof the products is a liquid phase instead of a solid phase.
fine mixture oM + /3. An example of an alloy system s
It turns out that all known
located nearer the composition of the solid phase, so that the solid phase
-
.
predominates in the reaction. In this case, 33 percent solid A was formed compared with 17 percent L2. As in the case of the eutectic, alloys to the left of point C, such as alloy x, are known as hypomonotectic alloys whereas: alloys to the right of point C up to point F are known as hypermonotectic alloys. We now continue with the discussion of the slow cooling of alloy x.
V;..:..i
f :
,
,
tween copper and lea_d given in Fi L2 region is closed. Also, althouc and 3, the solubility is actually sc metals, copper and lead.
6 20 Type VII-Two Metals Insoluble In complete the study of basic phai
/After the monotectic reaction is complete and as the temperature is
the liquid and solid states. If poin! site directions, they will eventually
dropped, more solid >A will be formed from L2. 'When the eutectic tempera-
in Fig. 6-46. There are many com
ture is reached at TE, the alloy is at X3 and L2 will be at point E. This is the right temperature and right composition to form the eutectic mixture. The}
insoluble in each other.
,
eutectic reaction now takes place, with L2 forming
a _
_
vein/ fine mixture of
solid A, plus solid S. fhe final microstructure wili consist of 87.5 percent
'
grains oTprimary A surrounded by 12.5 percent of the eutectic (A + /3) mixture. The student should verify these percentages.
When c(
at their individual freezing points of contact and almost no diffusioi An alloy system which comes vi minum and lead shown in Fig. 6A i extends almost entirely across th '
The occurrence of two liquids in the hypermonotectic alloys that is, alloys of composilions between C and F, above the monotectic tempera,
ture, introduces structural considerations which have not been discussed
0c
up to this point. Given enough time, the two liquids will separate into two
1100
layers according to density, with the lighter layer on top
.
1083°
i
It is quite possi-
1000 -q+Z,
ble, however, to have two liquids existing as an emulsion wherein tiny droplets of one liquid remain suspended in the other liquid. Unfortunately knowledge of this behavior with respect to metals is very limited at the
41
953° 900
,
present time. Consider the
slow cooling of a hypermonotectic alloy y containing 70A300. At the elevated temperature, this alloy will be composed of a singlejj homogeneous liquid phase L,. Upon cooling, the limit of liquid immiscK bility is; crossed at y,, and the second liquid L2 will make its appearance
800
O
700
600 500
,
probably at the surface of the confining vessel, and possibly also at various]
points through the liquid bath. The composition of L2 may be obtained byfc
400 326°
300 a
-
drawing a tie line in the two-phase region and applying Rule I. As the tern-«
perature decreases, the quantity of L2 increases, so that just above theC monotectic temperature, at point y2 the amount of L2 present would be equal to 10/40 x 100, or 25 percent. Conditions being favorable this liquid J
200
'
j
_
Cu
10
20
30
40
50
60
7
Weight percentage leai
,
composition
,
will exist as a separate layer in the crucible or mold, fhat portion of the
Fig. 6-45 The copper-lead equilibrium diagram. (Fr Metals Handbook, 1948 ed„ p. 1200, American Socii Metals, Metals Park, Ohio.)
4
-
r
1
'
v
i
*
.
5 as
PHASE DIAGRAMS
205
mixture which is composed of /., now reacts according to the monotectic ,
equation to form more of L2 + solid A. With continued cooling, more solid [
id
A is formed from L?, its composition becoming richer in 6, until the eutectic temperature is approached at point y3. At that temperature, the remaining L2 (37.5 percent) undergoes the eutectic reaction and solidifies into a very fine mixture of >4 + /}. An example of an alloy system showing a monotectic reaction is that be-
fshould be apparent that the reaction, the only difference ase instead of a solid phase. _
points in metal systems are base, so that the solid phase
tween copper and lead given in Fig. 6-45. Notice that in this case the + L2 region is closed. Also, although the terminal solids are indicated as a and p, the solubility is actually so small that they are practically the pure metals, copper and lead. 6 20 Type VII-Two Metals Insoluble In the Liquid and Solid States This will complete the study of basic phase or equilibrium diagrams that involve the liquid and solid states. If points C and F in Fig. 6-44 are moved in opposite directions, they will eventually hit the axes to give the diagram shown in Fig. 6-46. There are many combinations of metals which are practically insoluble in each other. When cooled, the two metals appear to solidify at their individual freezing points into two distinct layers with a sharp line
& percent solid A was formed
.J of the eutectic, alloys to the
. '
"
"
h as hypomonotectic alloys, point F are kriown as hyperthe slow cooling of alloy x and as the temperature is .
Vi/hen the eutectic temperam\\ be at point E. This is the '
'
:: '
-
rm the eutectic mixture The .
"
rming a very
.
..
nejriixture of
of contact and almost no diffusion.
s will consist of 87.5 percent
ii
3
if
i
:
.
An alloy system which comes very close to this type is that between aluminum and lead shown in Fig. 6-47. Notice that the two-phase liquid region extends almost entirely across the. diagram. This condition corresponds
it of the eutectic {A + p) mixages.
ermonotectic alloys that is, ,
;)ve
the monotectic tempera-
ich have not been discussed
0C
liquids will separate into two lyer on top It is quite possi-
1083°
1100
2000
.
1000 '
) emulsion wherein tiny drop-
-
y:v-
953°
mother liquid. Unfortunately
2
900
w
1600
,
netals is very limited at the
1800
92 6 V
41
800 1400
ectic alloy y containing 70A/vill be composed of a single g, the limit of liquid immisciL2 will make its appearance and possibly also at various
|
700
2 a
-
ion of L2 may be obtained by
1000 ;)0o -
400
127
;. V '
.
ses, so that just above the unt of L2 present would be
a!
-
is being favorable this liquid pr mold. That portion of the
800
0
600
326
300
"a
-
.
i applying Rule I. As the tem*
in %
600
,
.
'
1200 .
8
-*
200
400 Cu
10
20
30
40
50
60
70
80
3b
Weight percentage lead composition
,
Fig. 6-45 The copper-lead equilibrium diagram (From y Metals Handbook 1948 ed., p. 1200, American Society for .
,
Metals, Metals Park Ohio.) ,
i
;
-
,
J
r 206
.
INTRODUCTION TO PHYSICAL METALLURGY
between the basic types to make the s
pier. The first three types differ onl
Liquid A + liquid ff
Starting with a completely insoluble
points at either end of the eutectic li Solid
other, that is, toward greater solubilit
+ liquid B
'
diagram of Type III, partly soluble in 1 moved until they coincide with the eu
soluble system results (Fig. 6-48c). 1
i
Solif /I + solid B
intermediate phase. A
5
Composition -
_
' .
1
If this phase d
melting), the diagram will show a p( IS
Fig. 6-46 Hypothetical equilibrium diagram for two metals Insoluble In the liquid and solid states.
to a limiting case of the monotectic reaction and the eutectic reaction
.
The upper of the two horizontal lines represents a monotectic reaction inE
which the monotectic point is very close to the composition and melting*
u
point of pure aluminum. The lower horizontal line represents a eutectic reaction in which the eutectic point is practically coincident with the composition and melting point of pure lead. 6-21
"
a
Interrelation of Basic Types The various types of equilibrium diagrams that! have been discussed may be combined in many ways to make up actuals
diagrams. It is important for the student to understand the interrelation! °
c
F
noo
C V
((?) Insoluble in the
2000
I0OO
1800
900
IfaOO
800
1400
ID 13
700
e
560°
658 5° 1200
152
600
1 500 400
. 1000
E
F
n + i-
G
800 -
326.2°
3274V oCC
300
i
200
400
At
..10
20
30
40
50
60
Weight percentage '
;
.
70
80
lead
compobilion
Fig. 6-47 The aluminum-lead alloy system. (From Metals Handbook, 1948 ed.,;p. 1165, American Society for Metals, Metals Park, Ohio.)
I -
1
90
Pb
Composition
(A) Partly soluble in the solid state
Fig. 6 48 Interrelation of equilibrium diagrams as the solubiliiy in the solid state is varied.
5
PHASE DIAGRAMS
207
m >*
between the basic types to make the study of complex diagrams much simpler. The first three types differ only by the.solubility in the solid state. Starting with a completely insoluble system of Type II (Fig. 6-48a), if the points at either end of the eutectic line (F and G) are moved toward each other, that is, toward greater solubility in the solid state, this will result in a diagram of Type III, partly soluble in the solid state (Fig. 6-48fa). If they are moved until they coincide with the eutectic composition at E, a completely
1
soluble system results (Fig. B-48c). Types IV and V are determined by the intermediate phase. If this phase decomposes on heating (incongruent melting), the diagram will show a peritectic reaction.
;tion
If the intermediate
m
4 :j m m
4
C|
and the eutectic reaction
.
f 1
esents a monotectic reaction in
o the composition and melting ;vjontal line represents a eutectic "
a
licaiiy coincident with the com-
a)
A
jes of equilibrium diagrams that many ways to make up actual to understand the interrelation
!
H
I
:
1 4
°F -
Composition
n 2000
[a] Insoluble in the solid state
ii
1800 1600 1400
-
-
1200
a)
if
3
o
1000
E -*-G
cL
800 600
i
400 Pb
1 Composition 16] Partly soluble in the solid state Fig. 6-48 Interrelation of equilibrium diagrams as the solubility in the solid state is varied.
Composition
(c) Completely soluble in the solit state i
i
:
AS V
208
INTRODUCTION TO PHYSICAL METALLURGY
phase shows a true melting point (congruent melting)
,
show a eutectic reaction
the diagram
maj
.
' ,
60C
(539° ,1512°
3 4
TRANSFORMATIONS IN THE SOLID STATE
45
S
There are several equilibrium changes and reactions which take plai ,
L
62
1500
140C
1400'
entirely in the solid state
.
v6-22 Allotropy During the discussion of metals and crystal structure in Chap 2 ,! it was pointed out that several metals may exist in more than one t y /
1300
.
ype
_
_
li
100C
crystajl structure depending upon temperature fron tin, manganese, a cobalt are examples of metals which exhibit this property known as a//u .
9(0C
,
,
ropy. On an equilibrium diagram
,
1
80C
this allotropic change is indicated byl
770
point or points on the vertical line which represents the pure metal Thii 6-49. In this diagram, the gamma solid-solution field .
f.OC
is illustrated in Fig
.
is "looped." The pure metal A and alloys rich in A undergo two transform mations. M
a + y
\ 40C c
_
200 10
Fe
20
30
40
50
,
called the'gamma loop
Weight percentage
.
In some alloy systems involving iron the gamma loop is not closed Thitj. ,
.
is illustrated by the iron-nickel equiibrium diagram shown in Fig 6-50. Thi| .
diagram shows the freezing point of pure iron at 15390C(2795°F) the 6 solid solution
,
,
formin
whiolijsjwdy-centered cubic, fhe y sojid solution i .
.
,
$-50
.
The
Metals Handbook, 1948 ed., p. 1211, American Society Ifctals, Metals Park, Ohio.)
,
forme by a p i ctic eactiojiat 1512°C(27570F) Notice that forpure iroij the allotropic change from 6 (b c c.) crystal structure to the -y (f.c.c.) for occurs at 1400 C(2554oF) but for the alloy this change begins at a high .
a
iron-nickel equilibrium diagram. (From
temperature. The last allotropic forming the a. (bx.c.)crystal struct
8 3 9*'c'e''-disorder Transformation On tional type of solid solution the sc
position but are distributed at rand The alloy is said to be in a
Liquid
4
disordi
solid solutions, if cooled slowly, i where the solute atoms move int(
Melting point of /I L
"
a
structure is now known as an ore
5?
6-51) Ordering is most common the solid state, and usually the m .
5
V\
Allotropic Allc change
simple atomic ratio of the two elen
a
is sometimes given a chemical fc
ill
the
gold-copper alloy system. On
tions are frequently designated a; which they are found is usually I A
Composition
B
Fig. 6-49 Hypothetical equilibrium diagram showing metal A undergoing two allotropic changes
.
i
equilibrium diagram for the Au
-
C
When the ordered phase has th
phase, the effect of ordering on rr
m
i
.i
'
mm
W -
PHASlE DIAGRAMS
209
M -
ngruent melting)
,
the diagram may
f
1600
I539M5120
34
L
62
H 2800
tsoo
y+l68
~
45
S
is and reactions which take place
srature.- fron
1400 r
H 2400 t300 <
1000
tin, manganese and Hihibit this property, known ss allotKiillotropic change is indicated by a ih represents t'he pure metal This ,
H 2600
1436 I400«
ils and crystal structure in Chap. 2, hiay exist in more than one type of
1455
910°
,
H 1600 800
~
r
770°
H 1200
.
600
im, the gamma solid-solution field /s rich in A undergo two transfor.
I
a + y
4
800
400
:ams involving iron such as Fe-Si,
353
_
:
d-solution field Since the type of
200
Js gamma iron, the field is usually
:
400
Fc
10
20
30
40
50
60
70
80
90
Ni
Weight percentage nickel
Wrvje '
gamma loop is not closed This i diagram shown in Fig. 6-50 This e iron at 1539°C(27 950F) forming
Fig. 6'50
.
The iron-nickel equilibrium diagram. (From
1
Metals Handbook, 1948 ed., p. 1211, American Society for Metals, Metals Park, Ohio.)
.
,
red cubic. The 7 solid solution Is 27 570F). Notice that for pure iron ital structure to the (f.c c.) form by this change begins at a higher ;
.
.
temperature.
t
The last allotropic change takes place at 910oC(1666oF),
n
forming the « (b.c.c.) crystal structure. 6:23 Order-disorder Transformation Ordinarily in the formation of a substituv
tional type of solid solution the solute atoms do not occupy any specific position but are distributed at random in the lattice structure of the solvent. The alloy is said to be in a "disordered" condition. Some of these random solid solutions, if cooled slowly, undergo a rearrangement of the atoms where the solute atoms move into definite positions in the lattice. This structure is now known as an ordered solid solution or superlattice (Fig. 6-51) Ordering is most common in metals that are completely soluble in the solid state, and usually the maximum amount of ordering occurs at a simple atomic ratio of the two elements. For this reason, the ordered phase is sometimes given a chemical formula, such as AuCu and AuCu3 in the .
i
gold-copper alloy system. On the equilibrium diagram, the ordered solutions are frequently designated as a' p', etc. or a', a", etc., ind the area in which they are found is usually bounded by a dot-dash line. The actual equilibrium diagram for the Au-Cu system is shown in Fig. 6-52. When the ordered phase has the same lattice structure as thedisprdered ,
i v
I
phase, the effect of ordering on mechanical properties is negligible. Hard-
11
1
210 INTRODUCTION TO PHYSICAL METALLURGY
1
Quenched
(disordered)
Annealed
Disordered
-
Fig 6-51
[ordered!
Orderec
4 atoms
oloms L
Atomic arrangements in a disordered and ordered
.
50
25
0 Cu
solid solution
100 Au
75
Atomic percent Au
1
Fig. 6'5J Electrical resistivity vs. composition for the gold-copper system. (By permission from C. S. Barrett,
!S
:
!he. erin9 .pr0Cess is ™* Pronounced in thos*
rTrrthe shape of the
*
;
"
g 3
less of the structure formed as a result of ordering an important prope*! change prod d
J 6.53). Notice the sharp decrease It
' c raw
C°-
,
even in the absence of
,
duct.on m electncal resistance (Fig
m
.
electrical resistivity at the composit
a
phases AuCua and AuCu. In the gold no two-phase region between the d In some cases there will be a two-f
Atomic percent gold 10
HOC
20
30
40
5C
60 70 8090
2000
disordered solid solutions.
1800
formation of a crystal structure wl ordered phase from which it forrr
1600
palladium alloy system shown in ordered phases: «', a", and /3. Co(
100C a
+ L
a+l
900
Most
800
I40C 7or,
Li_
(7
600
1200
0C 1G00
i
03
140C I554c
1000
500
1200 Q + L
40C
80C 4
30C
\
a
sac
/
600 o
I
20C
100C
eoc
'
a"
' /
40C
7
n
+
100 Cu
10
20
30
40
50
60
80
90
Pd
Au
+ \I 0
8
200 70
Hr
o4-/3--aVA-
40C
10
Weight percent gold
20
30
40
50
60
70
1
a
81
Weight percentage copper
Fig. 6-52 The gold-copper equilibrium diagram (From .
Metals Handbook
,
1948., p. 1171, American Society for
Metals Metals Park ,
,
Ohio.)
fig 6-54 ,
| Metals Park, Ohio.) .
m
The copper-palladium alloy system (Frorr .
Handbook, 1948 ed., p. 1201, American Society for N
1 PHASE DIAGRAMS
211
I
I 1 M
Quenched
i
(disordered)
:
g
Annealed
(ordered)
Ordered
I
oms
50
25
0
100 Au
75
Atomic percent Au
Cu
M
Fig. 6-53 Electrical resistivity vs. composition for the gold-copper system. (By permission from C. S. Barrett, Structure of Metals," 2d ed., McGraw-Hill Book Company, New York, 1952.) "
i
,
iss is most pronounced in those s changed by ordering Regard.
ordering, an important property
electrical resistivity at the compositions which correspond to the ordered phases AUCU3 and AuCu. In the gold-copper equilibrium diagram there was no two-phase region between the disordered and ordered solid solutions. In some cases there will be a two-phase region between the ordered and disordered solid solutions. Most frequently this is associated with the formation of a crystal structure which is different from that of the disordered phase from which it forms. This is illustrated by the copperpalladium alloy system shown in Fig. 6-54. This diagram shows three ordered phases: a', a", and (3. Copper and palladium are both face-cen-
Wfy hardening, is a significant re|
Notice the sharp decrease in
.
50
60 70 8090 -
-
I
[
i
|-i 2000.
1800
a+ L
1600
1400
% i 'til
'
u
F
_
1200 |
If
1600
5
-
1400
cx
10001
2800
L
1554°
1083°
-
2400
1200
2000
a+L
800
1000
\
a \
-
1600
-
1200
800
\
A M
m
600
\
600
m
.
\
400
400
a
_
Ll 90
-
+
1
80
'
200 Pd
Au
0 10
20
30
40
A
'
hi 50
a
60
a
70
80
90
Weight percentage copper
Fig. 6-54
The copper-palladium alloy system (From Metals p. 1201, American Society for Metals. Metals Park, Ohio.) .
Handbook, 1948 ed
,
.
800
m
400 Cu
1
"
i
m
M .
-
.
111
212 INTRODUCTION TO PHYSICAL METALLURGY
\
tared cubic and the ordered solutions «' and «" are face-center ed cubic;| -fi i - etlth side. s body-centered cubic and shows a two-phase region onj
however
The liquidus line is TAETB and th
,
mixture is composed of the phases
,
/ 6-24 $\e Eutectoid Reaction This is a common reaction in the solid stat | ( 'very similar to the eutectic reaction but does not involve the liquid ItiInJisl e
y
v The significance of the solvus line '
.
this case a solid phase transforms on cooling into two new solid
.
/ increased in B, the temperature at \ 1: is decreased, reaching a minimum
,
The general equation may be written as Solid,
phases.1
i the decrease in solubility of S in y
cooling
gglidg + solldj,
heating
I N is known as the eutectoid point. position, and the line OP is the eutt tic diagram, it is common practice
eutectoid mixture mux lure
The resultant eutectoid mixture is extremely
just like the eutectic1«| ndeTthe microscope both mixtures generally appear the same » [and it is not possible to determine microscopically whether the mixture| resulted from a eutectic re _
mh
re
fine
,
:
-
,
action or a eutectoid reaction An equilibrium
diagram illustrating the eutectoid reaction Alioy
temperature line, namely, y solid ; (point G). Pomt M indicates an a
a.
Alloy
Alloy
2
3
.
is shown in Fig 6-55. .
4
J9
toid composition hypoeutectoid a
-
hypereutectoid alloys. When the hypoeutectoid alloy 1 i when the liquidus line is crossed the solidus line is crossed at x2. It i solvus line is crossed at X3. The pi
allotropic change, forming the a solution dissolves much less of S
X
Ml
the B atoms that are dissolved in
Liquid
change must now diffuse out of t atoms has taken place, the remai the new crystal structure, forming
"2
dissolve in the remaining y solutic perature falls. The composition
B
L
y
down and to the right along thest eutectoid temperature x4, the rer
r
F
E
G
M
point N. The significance of the
a
the end of the crysbl structurecf y must now transform by the eute
3
as
q3
a and /3 in an extremely fine mix /V
!
P C(
a + eutectoid mixture
1
Ahe microstructure at room temf toid a which was formed betwe
13 + eutectoid mixture
i
for a given alloy from the phase
A
Hypoeutectoid alloys
Hypereutectoid alloys Composition Phase diagram illustrating the eutectoid re<
Fig. 6-55
mixture of a + 3. This is shown
action.
i
.
B
whenever a line is crossed on t
_
sponding break in the cooling cl on the phase diagram, indicatir curve as a horizontal line. The
PHASE DIAGRAMS
' ' .
213
The liquidus line is TAETB and the solidus line is TAFGTB. The eutectic
and a" are face-centered cubic; shows a two-phase region on
mixture is composed of the phases that occur at both ends of the eutectic temperature line, namely y solid solution (point F) and p soWd solution '
,
q
n reaction in the solid state It is : does notjnvolve the liquid In
(point_ G). Point M indicates an allotropic change for the pure
metal A.l
"
I
vf The significance of the solvus line MN is that, as the alloy composition is
.
.
;
)oling into two new solid phases
1
solidj
,
nlxture
the decrease in solubility of 6 in y as the temperature is decreased. Point N is known as the eutectoid point. Its composition is the eutectoid composition, and the line OP is the eutectoid temperature line. Like the eutec-
tures generally appear the same
toid composition hypoeutectoid alloys and those to the right of point N
oscopically whether the mixture
hypereutectoid alloys. When the hypoeutectoid alloy 1 is slow-cooled y solid solution is formed when the liquidus line is crossed at x, More and more y is formed until the solidus line is crossed at x2. It remains a uniform solid solution until the solvus line is crossed at X3. The pure metal A must now start to undergo an
'
,
tectoid reaction
is shown in Fig
.
An equilibrium
.
I
tic diagram, it is common practice to call all alloys to the left of the eutec-
ymely fine, just like the eutectic "
increased in B, the temperature at which the allotropic change takes place
(j is decreased, reaching a mihlmum at point N. The solvus line FN shows
.
I
1
,
6'55.
1
.
allotropic change Jeiining the a solid solution,
I
Notice that the a solid 1
solution dissolves much less of B than does the y solid solution. Some of 1 _
the B atoms that are dissolved in the area that will undergo the allotropic
iquid
i
change must now diffuse Out of that area. When sufficient diffusion of 6 \ «
atoms has taken place, the remaining A atoms rearrange themselves into
i
the new crystal structure, forming the a solid solution. The excess B atoms dissolve in the remaining y solution, which becomes richer in 6 as the tem-
L + B
-
-.
'--
-
perature falls.
- .
I T
.t
t
-
_
The composition of the remaining y is gradually moving
down and to the right along the solvus line MN. When the alloy reaches the G
eutectoid temperature x4, the remaining y has now reached the eutectoid
/3
point N.
The significance of the eutectoid line is that this temperature is
al structure change thaytortedjatXa, and the remaining y must now transform by the eutectoid reaction, forming alternate layers of a and p in an extremely fine mixture. The reaction may be written as the end of the cr
X+/3
P
_
i»
.
cooling y:
a+ p
heating eutectoid mixture
v/fhe microstructure at room temperature consists of primary a or proeutec.ixture
1
toid a which was formed between X3 and x4 surrounded by the eutectoid
1
,
mixture of a + /3. This is shown in Fig. 6'56. In drawing the cooling curve for a given alloy from the phase diagram it is important to remember that whenever a line is crossed on the phase diagram, there must be a corresponding break in the cooling curve. Also, when a horizontal line is crossed on the phase diagram, indicating a reaction, this will show on the cooling
:
n
,
B .ctoid
j
curve as a horizontal line, the cooling curve for alloy 1 is shown in Fig.
M
11
1
214
INTRODUCTION TO PHYSICAL METALLURGY
Primery n or Droei i ctoicl q
Liquid
L ,
Eulectoid
L
+
+
(q + /3) mixture a
3 + 6
Fig. 6-56 Microstructure of a slow-cooled hypoeutectoid alloy, alloy 1 of Fig. 6-55.
I + /3
a +y
i -
,
6-57
.
The description of the slow cooling of the hypereutectoid alloys 2
fnd 3 is left as an exercise for the student. 6-25
6
A -
Composition
he Peritectoid Reaction This is a fairly common reaction in the solid f 58 phase diagram showing the formation state and appears in many alloy systems. The peritectoid reaction may be f ite,modiate phase y.by a peritectoid reaction
'
of
ir
written in general as Solid! + solid2
In Fig. 6-58, two solid phases a £ line £F to form an intermediate p
new solids healing
a +
The new solid phase is usually an intermediate alloy, but it may also be a solid solution. The peritectoid reaction has the same relationship to the peritectic reaction as the eutectoid has to the eutectic. Essentially, it is the replacement of a liquid by a solid. Two hypothetical phase diagrams to illustrate the peritectoid reaction are shown in Figs. 6-58 and 6-59.
.
In Fig. 6-59, two solid phases, react at the peritectoid-temperat the terminal solid solution y. Th( A +
1
Liquid '2
.1
. J
-
L + A
I
A + a
E 1
D
'3 \ -
y + /3
si
. . .
.
4 + 7
..
Composition Time-*~
Fig. 6-57
:
Cooling curve for alloy 1 of Fig. 6-55.
Fig 6-59 Phase diagram showing
the formation
terminal solid solution y.by a peritectoid reaction
1
PHASE DIAGRAMS
215
Liquid
L +
+
a
/5
a
a
a
3Q
1
m
y ng of the hypereutectoid alloys 2
:
nt.
Composition
y common reaction in the solid Fig. 6-58
Phase diagram showing the formation of an intermediate phase y by a peritectoid reaction.
The peritectoid reaction may be
.
0
.
In Fig. 6-58, two solid phases a and /3 react at the peritectoid-temperature line EF to form an intermediate phase y. The equation may be written as
cvif± new solid
f
:
viediate alloy but it may also be a
cooling
a+p-
,
has the same relationship to the to the eutectic. Essentially it is wo hypothetical phase diagrams
In Fig. 6-59, two solid phases, the pure metal A and (3 solid solution, react at the peritectoid-temperature line CD to form a new solid phase, the terminal solid solution y. The equation may be written as
,
1
y
heating
own in Figs. 6-58 and 6-59
.
,
A+P. n
cooling
j I |
-y
heating
I
I
Liquid 1
i 1
i
I
4 + y y \
Ft
Composition Fig. 6-59 Phase diagram showing the formation of the terminal solid solution 7 by a peritectoid reaction .
Ii 1
41
I i
I
ew
f
-
216
INTRODUCTION TO PHYSICAL METALLURGY
f
It was pointed out in the discussion of the peritectic reaction thattl microstructure of a peritectic alloy rarely shows complete transformatit (Fig. 6-43). This is because diffusion through the new phase is requir( z in order to reach equilibrium. Since the peritectoid reaction occurs entire! .
I
in the solid state and usually at tower, temperatures than the peritectii
reaction, the diffusion rate will be slower and there is less likelihood thaf equilibrium structures will be reached. Figure 6-60 shows a portion of the!
silver-aluminum phase diagram containing a peritectoid reaction. If aTipfc percent aluminum alloy is rapidly cooled from the two-phase area justl above the peritectoid temperature, the two phases will be retained, andthep microstructure will show a matrix of y with just a few particles of a (Fig.f 6-61 ,
a)
.
7
d
If the same alloy is slow-cooled to just below the peritectoid tem-|
r
perature and held there for 20 min before rapid cooling, some transforma-l tion will take place. The microstructure, Fig. 6-616, shows that some y hasW transformed to the new phase p, while much of the original a remains. The phase diagram indicates that there should be only a single-phase fi, Obviously equilibrium has not been achieved. Even after holding for 2 hrat «
just below the peritectoid temperature, a single-phase structure is still S
not produced (Fig. 6-61 c).
||
The similarity, in both the general equation and the appearance on an
equilibrium diagram, for the monotectic, eutectic, and eutectoid reactions and the peritectic and peritectoid reactions is apparent from a study of w Table 6-4. These reactions are by no means the only ones that may occur on equilibrium diagrams. However, they are by far the most common ones, m
and the student should become familiar with them. 6-26
Complex Diagrams
ft
Some of the equilibrium diagrams discussed under the w
i
.
J
MM
simple types are the same as actual ones. Many alloy systems have dia- if grams which show more than one type of reaction and are generally more complex than the simple types.
However, even the most complex dia-
grams show mainly the reactions that have been covered. The student
d -
mm
: 1000 0 u
900
m
.
| 800
7
o
.
.
-
700
a + a 600 "
Fjg. 6-60
Ag
0+ v
2 4 6 8 10 Weight percent aluminum
A portion of the silver-aluminum phase diagram.
m j
m
PHASE DIAGRAMS
:
?n of the peritectic reaction that the
217
0 i
' -
m
irely shows complete transformation '
through the new phase is required e peritectoid reaction occurs entirely er temperatures than the peritectic wer and there is less likelihood that
Figure 6-60 shows a portion of the lining a peritectoid reaction. If a 7 oled from the two-phase area just .
two phases will be retained, and the
with just a few particles of a (Fig i:?d to just below the peritectoid temxe rapid coolipg some transforma5 Fig. 6-61 b, shows that some y has nuch of the original a remains. The ould be only a single-phase (3, so ieved. Even after holding for 2 hr at
f i
.
*
,
I
-
,
Jle, a single-phase structure is
;:
i
still
|uation and the appearance on an
:
i
eutectic, and eutectoid reactions
-
tions is apparent from a study of eans the only ones that may occur are by far the most common ones,
'
. with them.
Lium
diagrams discussed under the
es. Many alloy systems have diaof reaction and are generally more
16 *
pver, even the most complex diahave been covered.
i
The student
1 4
-
Fig. 6-61 (a) A peritectoid alloy, Ag + 7 percent Al, stabilized by long heat treatment above the peritectoid tem-
5
perature and then quenched. Islands" are the y phase embedded in a matrix of y. Magnification 150X. (fa) Same as (a), cooled to a temperature slightly "
.
below that of the peritectoid and held
20 min before quenching. Much of the
light-colored y has transformed to the dark-colored p without greatly affecting the islands of y. Magnification 150X. (c) Same as (a), cooled to a tempera-
11
ture slightly below the peritectoid and held for 2 hr before quenching. Dark
matrix is /3; light area is residual a that has not yet been dissolved by the /3. Magnification 150X. (By permission from F. N. Rhines,
Phase Diagrams in
"
Metallurgy," McGraw-Hill Book Company, New York, 1956.)
i
m
18
INTRODUCTION TO PHYSICAL METALLURGY
dent should recognize that this is a peri
Equilibrium-diagram Reactions
iL 6-4
VIE OF REACTION
GENERAL EQUATION
has been labeled, the area to the right m
APPEARANCE ON DIAG
tne left of the S region above 14650C ma) between 1100° and 1465° may also be lat recognize the eutectic point at 45 percen liquid solution and below the line two sc
'
cooling
notectic
V
L, + solid
neating
L, + solid L
Liquid J
ectic
S
horizontal line and therefore, another
solid, + solid,
healing
solids 8 and p. Below the line a new ph phase, which is labeled y. This is a perit
Solid, + solid, Solid,
tectoid
Solid
cooling
has been labeled, the area to the right I
solid, + solid,
neatmg
Solid?
left
i
solid3
Liquid + solid,
Liquid + solid,
ntectic
cooling
new solid2
healing
+ y. The point A on the vertical lii
aliotropic change. Below point A there of tungsten in cobalt. The solid solutioi
the diagram, but it is labeled a as indica
ew soli
labeled as a + Solid, + solid ntectoid
Solid, + solid
cooling
new solids
heating
.
tion, with the eutectoid point at 3 percer solid solution, while below the line are I
ew solid3
should.be able to label a phase diagram completely; understand the sig-1 |
gram is now completely labeled. Two in
that occur at the horizontal lines; and describe the slow cooling and micro-S:
on the diagram, y and 8. Since their ran they are most likely intermetallic compo respond to the formula Co7W2 and the S |
a binary equilibrium diagram. The application of ; |s
significant fact that may be determined
nificance of every point, line, and area; determine the various reactions; jf? structure of any alloy on
.
some of the principles in this chapter will now be illustrated for a complex | equilibrium diagram such as the cobalt-tungsten alloy system shown in ; ,
Fig. 6-62.
freezing point of tungsten, 3410oC
,
difference in solubility of tungsten in c
If
f
The freezing point of cobalt is shown on the left axis as 14950C. The
L-
is where the two dotted lines on ths-' f'
right would meet and is above the range of temperatures shown. Since! I two-phase areas must be bounded by single phases on either side oil a horizontal line, it is necessary to first label the single-phase areas. Thei upper line on the diagram is the liquidus line, so above the liquidus linej there is a single homogeneous liquid solution indicated by L. On the left
2600 L
-
2200
1800 I500
1495°
=-i-Ll465°-I35
1400
,
from the freezing point of cobalt, there is a very thin, cigar-shaped area!
which resembles Type I. Therefore, there must be a solid solution formed| below this area. The solid solution is labeled j8. Once the p area has been labeled, the portion between the very thin lines is labeled as /3 + L. Nexttoj
this area is a portion of the diagram which looks like a variation of TypelJ and shows a maximum at 1500oG. The area between the points 35, 45, and
1
1100°
-
6 5
1000
.
=00 350°
200 Co
T 10
""
20
30
40
50
60
the dotted line indicatesa| Weight percentage tungs small solid solution area which is labeled as epsilon, e. Now the area above f 6-62 The cobalt-tungsten alloy system. (From Metals 1690oC may be labeled as L + e. The first horizontal line is at 1690oG. Abovej inp dbook, 1948 ed., p. 1193, American Society for Metals thelinetherejis L+e. Below the line a new solid solution S appears. Thestu-l iMetals Park, Ohio.) 1500oC would also be p + L. Along the right axis
a
B+8
/3
'
I
although it is not labelc
zontal line at 350oC indicates another re
-
,
I
PHASE DIAGRAMS
219
dent should recognize that this is a peritectic reaction. After the S region APPEARANCE ON DIAGRAM
has been labeled
the area to the right may Ije labeled as S + e. The area to the left of the S region above 1465°C may now be labeled as L + S. The area between 1100° and 1465° may also be labeled as jti + 8. The students should recognize the eutectic point at 45 percent W and 1465°C. Above the line is a ,
/-2 + solid
liquid solution and below the line two solids (3 and S. At 1100°C is another horizontal line and therefore another reaction.
lld2
Above the line are two
,
>
Solid, + solid2 .
solids S and £!. Below the line a new phase appears an intermediate alloy phase, which is labeled y. This is a peritectoid reaction. Once the-y region has been labeled the area to the right becomes 7 + S and the area to the left (3 + y- The point A on the vertical line representing cobalt must be an allotropic change. Belovv point A there is a very tiny amount of solubility ,
Solid
,
-1
Solid2 + solid3 -
Liquid + solid,
of tungsten in cobalt. The solid solution area is too small to be shown on the diagram, but it is labeled a as indicated. The triangle AB3 may now be
solid2 2
labeled as a + /3, although it is not labeled in the diagram. The dotted hori-
Solid, + solid
zontal line at 350°C indicates another reaction. This is the eutectoid reac-
| solid:3 '
tion, with the eutectoid point at 3 percent W and 350°C. Above the line is
ew solid,
solid solution, while below the line are two solid solutions a + y. The diagram is now completely labeled. Two intermediate alloy phases are shown
1 completely; understand the sig-
on the diagram, y and s. Since their range of composition is not very great,
.
determine the various reactions ;
they are most likely intermetallic compounds. The y phase appears to correspond to the formuja Co7W2 and the S phase to the formula CoW. Another significant fact that may be determined from the diagram is the very large
cribe the slow cooling and micro-
rium diagram The application of .
I now be illustrated for a complex -
tungsten alloy system shown in on the left axis as 1495°C
.
difference in solubility of tungsten in cobalt depending upon the type of
The
°
c 7
L
+ 440C
S
2200
ibel the single-phase areas The 3 line, so above the liquidus line Jtion indicated by L On the left is a very thin cigar-shaped area
r
f- 5600
/
.
1800
must be a solid solution formed
ih looks like a variation of Type I ,
;,
,
ba between the points 35
45, and
,
ht axis, the dotted line indicates a asepsilon e Now the area above
:
.
.
iorizontal line is at 1690oC Above .
solid solution S appears
.
Thestu-
'
1500°
1495°
,
Jib/
1400
99.7 f- ? 80C
72
.
1465°
1
35
/3+8
/5 !
I690c
S2
,
3led /3. Once the y3 area has been lines is labeled as 3 + L Next to
I
0F
2600
.
.
I
;y
here the two dotted lines on the 9 of temperatures shown Since single phases on either side of
'
I
100c
A
32
/3+y
I- 2000
65
-
.
8W
y+S
y
60C A
iiooc
1 r -
-
120c
Co
i I
jfi'
350°
B
200
I
a
d400 10
20
30
40
50
60
70
90
W
I
Weight percentage tungsten Fig. 6-62 The cobalt-tungsten alloy system (From Metals Handbook, 1948 ed., p. 1193, American Society for Metals Metals Park, Ohio.) .
,
i, !
-
:
m
I
i-
220
INTRODUCTION TO PHYSICAL METALLURGY
crystal structure. In the /3 solid solution, where cobalt is face-centen cubic, the maximum solubility of tungsten in cobalt is given as 35 pera at 14650C. However, once the allotropic change has occurred and cobi
r
2 For an alloy containing 70 percent solidification; (b) give the temperature
composition and relative amounts of
tl
below (a); (d) sketch the microstructur
becomes close-packed hexagonal below point A, the solubility of tungstee]
curve.
in cobalt is almost negligible. The reactions and specific equations thaSj
3
Same as part 2 but for an alloy cor
Lead melts at 620oF and tin melts a percent tin at 360oF The maximum soli
occur at each horizontal line on this diagram are given below.
6-6
TEMPERATURE
is 19 percent; of lead in tin, 3 percent. As
.
REACTIQN
EQUATION
ture is 1 percent.
.
y
1690oC
Peritectic
L + e
14650C
Eutectic
L
cooling
:
-.
1100oC
Peritectoid
cooling
/S + S
1 5
2
/S + S 7
the phases present.
i
350oC
Eutectoid
P
Describe the solidification of a 40
at room temperature, giving the che
cooling
cooling
Draw the equilibrium diagram to s
points, lines, and areas.
3
a + y
Draw the cooling curve for the ab(
4 Repeat 2 and 3 for an alloy contai
Calcium (melting point 1560oF) an
6-7
compound CaMg2 which contains 45 ! compound forms a eutectic with pure n
QUESTIONS £.1
calcium. The solubility of the compoi
What information may be obtained from an equilibrium diagram?
eutectic temperature and decreases to
/6-2 Explain the importance of equilibrium diagrams in the development of ne*W The melting point of platinum is 32250F and that of gold is 1945UF. An alloy containing 40 percent gold starts to solidify at 2910oF by separating crystals of 15 perceni
,
and calcium at 830oF containing 78 per
between the compound and pure calcii
,
1 2
W:
b
phases present.
m.
3
fok an alloy containing 70 percent gold (1) give the temperature of iniliSM' '
'
Draw the cooling curve.
4 Write the specific equation of th
solidification/ ) give the temperature of final solidification; (3) give the chemical jb
composition and relative amounts of the phases present at 2440 F; (4) draw ths cooling curve. 6'4 Bismuth and antimony are completely soluble in both the liquid and solid states a Check the crystal-structure factor and calculate the relative-size factor forthese
Describe the slow cooling of an ;
the microstructures at room temper
*
a Draw the equilibrium diagram to scale on a piece of graph paper and label a? points, lines, and areas.
Draw the equilibrium diagram to
points, lines, and areas.
gold. An alloy containing 70 percent gold starts to solidify at 2550 'F by separating! crystals of 37 percent gold.
a
is not soluble in the compound. A seco
alloys. ft 6-3 Platinum and gold are completely soluble in both the liquid and solid staieiB
temperature.
'
: .
6-8
1
Label Fig. 6-63 completely.
2 Write the specific equation of tht
.
line.
3
metals.
b
Sketch the microstructure of alio
Bismuth melts at 520"F and antimony melts at 1170oF. An alloy containing 69.W
percent bismuth starts to solidify at 940oF by separating crystals of 90 percent
x
antimony. An alloy containing 80 percent bismuth starts to solidify at 750oF bf;» separating crystals of 75 percent antimony.
1
m
Draw the equilibrium diagram to scale on a piece of graph paper labelingal«!
lines, points, and areas.
2
For an alloy containing 40 percent antimony, (a) give the temperature of initi
s
solidification; (b) give the temperature of final solidification; (c) give the chemical composition and relative amounts of the phases present at 800oF; (d) draw cooling curve. 6-5
BisVnuth (melting point 520oF) and cadmium (melting point 610oF) are assumi
to be completely soluble in the liquid state and completely insoluble in the soli l state. They form a eutectic at 290°F containing 40 percent cadmium.
4
Composition
1 Draw the equilibrium diagram to scale on a piece of graph paper labeling ! points, lines, and areas.
Fig. 6-63
St.
'
mmamstSKSSSSSSSSSSi
m
PHASE DIAGRAMS
ion
,
»
f iten in cobalt is given as 35 percent 3 sic change has occurred and cobalt 9w point A, the solubility of tungsten i ictions and specifip equations that j
; ;: .
.
where cobalt is face-centered
2 For an alloy containing 70 percent cadmium (a) give the temperature of Initial solidification; (b) give the temperature of final solidification; (c) give the chemical
.;.
.
agram are given below
.j
.
221
composition and relative amounts of the phases present at a temperature of 100oF below (a); (d) sketch the microstructure at room temperature; (e) draw the cooling curve.
3 Same as part 2 but for an alloy containing 10 percent cadmium. Lead melts at 620oF and tin nrielts at 450oF They form a eutectic containing 62 6-6 .
percent tin at 360oF The maximum solid solubility of tin in lead at this temperature .
is 19 percent; of lead in tin, 3 percent. Assume the solubility of each at room tempera-
MION
ture is 1 percent. cooling
1
8
Draw the equilibrium diagram to scale on a piece cf graph paper labeling all
points, lines, and areas.
2 :r
"
'
-
r
cooling
Describe the solidification of a 40 percent tin alloy. Sketch its microstructure
at room temperature, giving the chemical composition and relative amounts of
.
y
the phases present.
3
a + y
Draw the cooling curve for the above alloy.
4 Repeat 2 and 3 for an alloy containing 90 percent tin. 6-7 Calcium (melting point 1560"F) and magnesium (melting point 1200°F) form a
compound CaMg2 which contains 45 percent calcium and melts at 1320oF
.
This
compound forms a eutectic with pure magnesium at 960 and contains 16 percent calcium. The solubility of the compound in magnesium is about 2 percent at the eutectic temperature and decreases to almost zero at room temperature. Magnesium '
vvv/sian equilibrium diagram? diagrams in the development of new :
;
A'le in both the liquid and solid states that of gold is 19450F
.
I I
is not soluble in the compound. A second eutectic is formed between the compound and calcium at 830oF containing 78 percent calcium, and there is no solid solubility
.
between the compound and pure calcium.
An alloy con-
1 Draw the equilibrium diagram to scale on a piece of graph paper labeling all
i0F by separating crystals of 15 percent rts to solidify at 2550oF by separating
points, lines, and areas.
ri
2 Describe the slow cooling of an alloy containing 30 percent calcium. Sketch the microstructures at room temperature and give the relative amounts of the
i a piece of graph paper and label all
phases present. 3
d (1) give the temperature of initial al solidification; (3) give the chemical
Draw the cooling curve.
4 Write the specific equation of the reaction that takes place at each eutectic temperature.
lases present at 2440nF; (4) draw the
6-8
jible in both the liquid and solid states. bulate the relative-size factor for these
1
Label Fig. 6-63 completely.
2 Write the specific equation of the reaction that takes place at each horizontal line.
3
Sketch the microstructure of alloy x when slow-cooled to room temperature.
Its at 1170oF
An alloy containing 50 ay separating crystals of 90 percent .
)ismuth starts to solidify at 750oF by (1 a piece of graph paper labeling all
i>ny, (a) give the temperature of initial
*
1
3
a)
2
'
solidification; (c) give the chemical ases present at SOO-F; (d) draw the
QJ
'
.
sm (melting point 6100F) are assumed id completely insoluble in the solid
:
140 percent cadmium.
,
i a piece of graph paper labeling all
Composition Fig. 6-63
m
9i
:
:
I, < r
222
INTRODUCTION
,
TO PHYSICAL METALLURGY
1
3
7
3
2
h
3
i a)
'5
Fe
FeSi
Fig. 6-64 The modified iron-silicon
Si
system.
CugSb
Cu
Composition
2
x rs sTisr"could be p,o,ted from Plot this diagram to scale on
fig. 6-66 The copper-antimony system.
a piece of graph paper and label all the areas
3 Write the reaction that takes place at each
horizontal line
4 Draw the cooling curve of a 40 pe 5 Describe the slow cooling of this perature, and give the relative amoui Label completely the phase c 6-10 1 2 Give the name and write the spec
.
at each horizontal line.
3
Discuss the significance of each
L
REFERENCES
American Society for Metals: "Metals
4
:
"
Metals Handbook," 8th ed., 19
Gordon, P.: "Principles of Phase Di Book Company, New York, 1968.
Fe2Sn
5
Guy, A. G.: "Elements of Physical Meta Inc., Reading, Mass., 1959. Hansen, M., and K. Anderko: "Consti
6
Book Company, New York, 1958.
Hume-Rothery, W., J. W. Christian, a Diagrams," The Institute of Physics Fe
I;
FeSn
-
FeSn 2
,
Sn
Fig. 6-65
R. E. Smallman, and C. W. H;
Institute of Metals, London, 1969.
Composition
Marsh, J. S.: "Principles of Phase I
The iron-tin system.
York, 1935.
4i
_
. .
~
I
' .
PHASE DIAGRAMS
223
1 I
m
.
f
> T
:
1
5
)
2 3
Si
S
CugSb
Sb
Composition Fig. 6-66
em so that It could be plotted from tho p. 1225).
graph paper and label all the areas
The copper-antimony system.
%
.
ch horizontal line
4 5
.
Draw the cooling curve of a 40 percent magnesium alloy. Describe the slow cooling of this alloy, sketch its microstructure at room tem-
perature, and give the relative amounts of the phases present. 6-10
2
1 Label completely the phase diagrams given in Figs. 6-64 to 6-66. Give the name and write the specific equation of the reaction that takes place
at each horizontal line. s
3
Discuss the significance of each line on the diagrams.
REFERENCES
American Society for Metals: "Metals Handbook," 1948 ed.. Metals Park, Ohio. ;
"
Metals Handbook," 8th ed., 1973, Metals Park, Ohio.
Gordon, P.; "Principles of Phase Diagrams in Materials Systems," McGraw-Hill Book Company, New York, 1968. Guy, A. G.: "Elements of Physical Metallurgy," Addison-Wesley Publishing Company, Inc., Reading, Mass., 1959. Hansen, M., and K. Anderko; "Constitution of Binary Alloys," 2d ed., McGraw-Hill
h
Book Company, New York, 1958.
FeSn 2
'
Hume-Rothery, W., J. W. Christian, and W. B. Pearson: "Metallurgical Equilibrium Diagrams," The Institute of Physics, London, 1952. j \ R. E. Smallman, and C. W. Haworth: "The Structure of Metals and Alloys,"
Sn
,
Institute of Metals, London, 1969.
Marsh, J. S.: "Principles of Phase Diagrams," McGraw-Hill Boqk Company, New York, 1935.
m
1
I I
I I I
I
I
f I
224
INTRODUCTION TO PHYSICAL METALLURGY
New Yo iS
63'3' "E
"
Rhines
,
F. H.: "Phase Diagrams
York, 1956
m Metallurgy
.
M " .
cGraw-Hill BookCoJ
McGraw-Hill Book Company
,
NeJB|-
THE IF
Reading Mass., 1959, ,
CARB EQUIL DIAGF
t
The metal iron is a pr important engineering alloys. In ai it is used for drainage culverts, ro ceiain enamel in refrigerator cabi
7-1 Introduction
typical analysis for ingot iron is: i
carbon
0012 percent
manganese
0017
phosphorus
0 005
sulfur
0 025
silicon
trace
.
.
Typical mechanical properties of 40,000 psi
Tensile strength Elongation in 2 in.
40 percen
Rockwell B hardness
30
Iron is an allotropic metal, whic "
ilctule dependir
afjattice
type
_
pure iron is shown in Fig. 7-1. When iron first solidifies at 2
cubic) S (delta) form. Upon furtf curs and the atoms rearrange the fc .
.
c
.
(face-centered cubic) anc
reaches 16660F another phase ,
iron to b.c.c. nonmagnetic a (aif comes magnetic without a chanc netic a iron was called /3 iron i change in lattice structure at H "
does not affect the heat treatrr
.
I
J
3
.
1
J!
J 1
ng Metallurgy McGraw-Hill Book Com"
,
4
'
imW McGraw-Hill Book Company
,
Science
" ,
New
THE IRON-IRON CARBIDE
Addison-Wesley Publishing
EQUILIBRIUM DIAGRAM
.
>
i
7-1
Introduction
The metal iron is a primary constituent of some of the most
important engineering alloys. In an almost pure form, known as ingot iron,
I
-
f »
it is used for drainage culverts, roofing, and ducts, and as a base for porcelain enamel in refrigerator cabinets, stoves, washing machines, etc. A typical analysis for ingot iron is: carbon
OOl? percent
manganese
0 017
phosphorus
0 005
sulfur
0 025
silicon
trace
i
.
.
.
Typical mechanical properties of ingot iron are as follows: Tensile strength
40,000 psi
Elongation in 2 in.
40 percent
Rockwell B hardness
30
Iron is an allotropic metal, which means that it caDjexij '
type ofjattice stri rrtTIrfi daDaildinq upon temperature A cooling curve for pure iron is shown in Fig. 7-1. When iron first solidifies at 2800oF, it is in the b.c.c. (body-centered cubic) S (delta) form. Upon further cooling at 2554°F, a phase change oc,
curs and the atoms rearrange themselves into the y (gamma) form which is f c c (face-centered cubic) and nonmagnetic. When the temperature reaches 16660F, another phase change occurs from f.c.c. nonmagnetic y iron to b.c.c. nonmagnetic a (alpha) iron. Finally at 1414°F, the «iron becomes magnetic without a change in lattice structure. Originally nonmagnetic « iron was called p iron until subsequent x-ray studies showed no change in lattice structure at 14140F. Since this magnetic1 transformation does not affect the heat treatment of iron-carbon alloys it will be disre,
.
.
.
*
,
,
,
i
I /
Li-
.
1
i 226 INTRODUCTION TO PHYSICAL METALLURGY
'
/ SS
2800
I AO )
2554
5
_
I
5'
,
I
Liquid
8 (delta! Fe B.CC
a bessemer converter and then machine. An exact iron silicate and certain siliceous materials i ladle and moved directly below The next step is the key ope
disintegration and slag incorpoi
y (gamma) Fe F. C.C.
ture of about 2800oF is poured
nonmagnetic
/
taining the molten slag, which
LJ_
got
THE
uniform distribution of the rel machine automatically oscillate
I 1666 a (alpha) Fe B.CC nonmagnetic
I 1414
Since the slag is maintained at
.
freezing point of the iron, the ire liquid iron contains large qua metal solidifies, the gases are r
tion liberates the gases in the I
a (alpha) Fe B.CC.
force to shatter the metal int(
magnetic
\
*
tom of the slag ladle. Because tion is called shotting. Since because of the fluxing action
Time-
Fig. 7-1
Cooling curve for pure iron,
garded in our discussion. Ail the allotropic changes give off heat (exo- }
I
thermic) when iron is cooled and absorb heat (endothermic) when iron is heated. 7-2
Manufacture of Wrought
Iron Wrought iron is essentially a two-component j
v
,
metal consisting of high-purity iron and slag The slag is composed mainly? .
of jxon
_
silLQala. The small and uniformly distributed particles of slag exist "i
physically separate in the iron. There is no fusion or chemical relationship . between the slag and the iron
j
.
Wrought iron was originally produced by the hand-puddling process, later by mechanical puddling and since 1930 by the Byers or Aston process. Regardless of the process there are three essential steps in the man- ] ufacture of wrought iron: first to melt and refine the base metal; second to produce and keep molten a proper slag; and third to granulate ordis-jj ,
,
'
,
,
,
integrate
,
,
the base metal and mechanically incorporate with it the desired!
amount of slag.
i
/
.
In the Byers process each step is separated and carried out in individual!
pieces of equipment. The raw materials of pig iron iron oxide, and silica are melted in cupolas The pig iron is purified to a highly refined state in ,
.
i
Fig. 7-2 The key operation in the manufactur iron. (A. M. Byers Company.)
l.- .v.*-l...wfc*l._-*
f»
,
,
W ..
. -:
0 -I "SjS = F THE IRON-IRON CARBIDE EQUILIBRIUM DIAGRAM
a bessemer converter and then transferred to the ladle of the processing machine. An exact iron silicate slag made by melting together iron oxide and certain siliceous materials in an open-hearth furnace is poured into a ladle and moved directly below the processing machine. The next step is the key operation of the process-that of base-metal disintegration and slag incorporation. The liquid refined iron at a tempera-
1
ir-
J 1 .
C C .
227
.
I
ture of about 2800oF is poured at a predetermined rate into the ladle con'
taining the molten slag, which is at about 2300oF (Fig. 7-2) To ensure a uniform distribution of the refined metal into the slag, the processing machine automatically oscillates as well as moves toward and backward. Since the slag is maintained at a temperature considerably lower than the a freezing point of the iron the iron is continuously and rapidly solidified. The liquid iron contains large Quantities of gases in solution, but when the metal solidifies, the gases are no longer soluble in it. This rapid solidifica- :
I
.
f 1 t
,
i
ia) Fe B.C
tion liberates the gases in the form of many small explosions of sufficient;J force to shatter the metal into small fragments which settle to the bot
C
.
.
stic
tom of the slag ladle. Because of the noise of the explosions, this operai
!
tion is called shotting. Since the iron is at a welding temperature, and.J
i
because of the fluxing action of the siliceous slag, these fragments stick .<
r
'
x r
-
opic changes give off heat (exo b heat (endothermic) when iron
-
;
1
if
n is essentially a two-component lag. The slag is composed mamly ;
}
i
.
[
-
5istributed particles of slag exist
I
o fusion or chemical relationship
by the hand-puddling
process,
930 by the Byers or Aston : ; :
three essential steps in the manp refine the base metal; second, 3; and third, to granulate or dis-
I
I
proc.
1
1 J
,
y incorporate with it the desired V
ted and carried out in individual if pig iron iron oxide and silica '
,
,
ified to a highly refined state
Fig. 7 2 The key operation In the manufacture of wrought
in
iron. (A M. Byers Company.)
\
I
.
I
.
.A
THE IRON >28
INTRODUCTION TO PHYSICAL METALLURGY
i5
together to form a spongelike ball of iron globules coated with silicate slag. The excess slag is poured off and the sponge ball, weighing between 000 and 8,000 lb, is placed in a press. The press squeezes out the surplus slag and welds the cellular mass of slag-coated particles of plastic
m
6
,
»
-
iron into a bloom. The bloom is reduced in cross section to a billet, which 7-3
is reheated and rolled into plate, bars, rods, tubing, etc. Properties nd Applications of Wrought Iron Quality wrought iron is distinguished by its low carbon and manganese contents. The carbon content is generally below 0.08 percent and the manganese content below 0.06 percent. The phosphorous content is usually higher than that of steel and ordinarily ranges from 0.05 to 0.160 percent. The sulfur content is kept low, and the silicon content of between 0.10 and 0.20 percent is concentrated alrfiost entirely in the slag. The slag content usually varies from about 1 to 3 percent by weight. A typical chemical analysis of wrought iron is as follows:
IS
ii m m
7r
V
V-i
m
mm
PERCENT
Carbon
0 06
Manganese
0 045
Silicon
0 101
Phosphorus
0 068
Sulfur
0 009
Slag, by weight
1 97
.
.
.
.
.
Slag in fig. 7-4 The microstructure of wrought iron. itudinal territe matrix, (a) Transverse section; (b) long lion. Etched in 2 percent nital; 100X.
.
Since wrought iron is a composite material, there are many methods of distinguishing between wrought iron and steel. Figure 7-3 shows the typical fibrous fracture of wrought iron; steel, on the other hand, shows a crystalline or granular break. The uniform distribution of the slag throughout the ferrite matrix is clearly shown by microscopic examination of a transverse section (Fig.
7-4a)
.
The threadlike appearance
examination of a longitudinal se
direction of rolling (Fig. 7-4£>). The mechanical properties of
'
iron. Because of the nature of
tfility are greater strength and duct
than in the direction transverse t
of wrought iron in the longitudin
Table 7-1. Improvement of rolling zation of the tensile strength and
It is possible to improve the sti most popular alloy wrought irons TABLE 7 1 Tensile Properties PROPERTY
Tensile strength, psi Yield point, psi
Ellongatlon, % in 8 in. Fig, 7-3
(Left) Fibrous fracture of wrought Iron; (right)
crystalline fracture of steel.
I
Reduction in area, %
THE IRON-IRON CARBIDE EQUILIBRIUM DIAGRAM
i globules coated with silicate slag e sponge ball weighing between The press squeezes out the sur)f slag-coated particles of plastic in cross section to a billet which
if
.
r K
wm
1
,
;;;
,
ds, tubing etc. ,
n Quality wrought iron is dis jse contents
.
2
M
mm
i 1
.V
-
r,
i
and ,
ent usually va ries ,
from about 1
i
1
mm r
i
.
'
35s
Hi
it. The sulfur content is kept low
SJnd 0 20 percent is concentrated
Mi
mm
.
.
.
m
V
manganese content below 0 06 .
J
Hi
»-
The carbon content
ially higher than that of steel
229
.2
ft
m
E
ron is as follows:
s-
1 i 1 J.
.
. .
(a
V
i
a
.
Fig, 7-4 The microstructure of wrought iron. Slag in a ferrite matrix, (a) Transverse section; (b) longitudinal section. Etched in 2 percent nital; 100X.
al, there are many methods of
.
7-4a)
eel. Figure 7-3 shows the typi
-
.
The threadlike appearance of the slag is evident from microscopic
on the other hand shows a
examination of a longitudinal section, that is, a section parallel to the direction of rolling (Fig. 7-4b).
oughout the ferrite matrix is
The mechanical properties of wrought iron are largely those of pure iron. Because of the nature of the slag distribution, however, the tensile strength and ductility are greater in the longitudinal or rolling direction than in the direction transverse to rolling. Typical mechanical properties of wrought iron in the longitudinal and transverse directions are given in Table 7-1. Improvement of rolling procedure has made possible the equali-
,
of a transverse section (Fig
.
i
zation of the tensile strength and ductility in both directions. It is possible to improve the strength of wrought iron by alloying. The most popular alloy wrought irons are those containing between 1.5 and 3.5 I
/
TABLE 7 1
Tensile Properties of Wrought Iron
PROPERTY
LONGITUDINAL
TRANSVERSE
Tensile strength, psi Yield point, psi Elongation, % in 8 in.
48,000-50,000
36,000-38,000
27,000-30,000
27,000-30,000
18-25
2- 5
Reduction in area, %
35-45
3- 6
v
,
\
i
i
a
i
230
THE 1RON-
ft.
INTRODUCTION TO PHYSICAL METALLURGY
r t .
percent nickel. The comparative mechanical properties of unalloyed and nickel wrought iron are given in Table 7-2. Charpy impact tests reveal that nickel-alloy wrought iron retains its'
m
Impact st-ength to a high degree at subzero temperatures
2800
LiQ
.
Y +
One of the principal virtues of wroOght iron is its ability to resist corrosion. When exposed to corrosive media, it is quickly coated with an oxide film. As corrosion continues, the slag fibers begin to function as rust resistors. The dense, uniform, initial oxide film is securely fastened to the surface of the metal by the pinning effect of the slag fibers and protects
r + L
c
2065 F
,
the surfaces from further oxidation.
/ 7
1666
Wrought iron is used for standard pipe, nails, barbed wire, rivets, and welding fittings. It is available in plates, sheets, tubular forms, and structural shapes. Wrought iron has many applications in the railroad, shipbuilding, and oil industries, as well as for architectural purposes and for farm implements.
2
1333 0F
H I
-
Eutectoid
0 025% C .
-4 The Iron-Iron Carbide Diagram The temperature at which the allotropic
\Jj V-/
changes take place in iron is influenced by alloying elements, the most r
important of which is carbon. The portion ofJhe iron-carbon alloy system which is of interest is shown in Fig. 7-5. This is the part between pure iron and an interstitial compound, iron carbide, FeaC containing 6.67 percent carbon by weight. Therefore, we will call this portion the iron-iron carbide
a + FejC
equilibrium diagram. Before going into a study of this diagram, it is important for the student to understand that this is not a true equilibrium diagram, since equilibrium implies no change of phase with time. It is a fact,
0
of relatively slow heating and cooling.
;
special names to most of the structures that appear on the diagram. The |
3
rig 7.5 The iron- iron carbide equilibrium diagram
I
ks general terms.
y solid .
The diagram shows three horizontal lines which indicate isothermal reactions. Figure 7-5 has been labeled in general terms with Greek letters to represent the solid solutions. However, it is common practice to give
2
.
Percent carb
however, that the compound iron carbide will decompose into iron and car-
bon (graphite). This decomposition will take a very long time at room temperature, and even at 1300oF it takes several years to form graphite. Iron carbide is called a metastable phase. Therefore, the iron-iron carbide diagram, even though it technically represents metastable conditions, can be considered as representing equilibrium changes, under conditions
08 1
solution is called austenite.
(e Wiand corner is expanded in Fi because of the 5 solid solution.
zontal line at 2720oF as being a |
peritectic reaction may be written Liquid + 5
TABLE 7 2
Tensile Properties of Unalloyed and Nickel Wrought Iron UNALLOYED
NICKEL
WROUGHT IRON
WROUGHT IRON
Tensile strength, psi Yield point, psi Elongation, % in 8 in.
48,000 30,000 25
60,000 45,000
Reduction In area, %
45
40
PROPERTY
! I
22
The maximum solubility of carb
while in f.c.c. y Fe the solubility ii
influences the 8 y allotropic ch; perature of the allotropic change
centC. Consider the significance
231
THE iroN-IRON CARBIDE EQUILIBRIUM DIAGRAM
m ileal properties of unalloyed and
1
gs -alloy wrought iron retains its =to temperatures
i 1
iron is its ability to resist cort is quickly coated with an oxide jrs begin to function as rust refilm is securely fastened to the
pf the slag fibers and protects nails, barbed wire, rivets, and
-a
.
.
'
i
/I t
-
2600
Liquid
s
o r 2554
I i 1 I
L + Fe C
r + L
I
0
E
C
f
2065 °F
Eutectic
it
m r
i
r
1666 a
J eets, tubular forms, and struc-
2
ilications in the railroad ship-
jrchitectural purposes and for
K
1333 °F
H
,
J Eutectoid
0 025% C .
.
ature at which the allotropic / alloying elements, the most
I
I r
|;if 4he iron:carbon allo sy tem
la + FejC
} is the part between pure iron
-
J
I
jF C, containing 6.67 percent portion the iron-iron carbide
udy of Jhis diagram, it is Ims is not a true equilibrium diaf phase with time. It is a fact,
jdecompose intpjrpjn jnd car'
. .
i very long time at room temyears to form graphite. Iron
f 4
.
0
2
0.8 1
1 1
4
3
43 .
5
6
:
6.67 .
1 ,
-
/
Percent carbon by weight
ir
Fig. 7-5 The ironviron carbide equilibrium diagram labeled in general terms.
efore, the iron-iron carbide
5nts metastable conditions,
The portion of the diagram in the upper g/pn l t-hand corner is expanded inTig. 7-6. This is known as the de/faherehoriThe student should recognize t because of the 8 sojid solution. tion of the
i changes, under conditions
Yjoli solution_is ca[led austenite. _
hich indicate isothermal re-
ral terms with Greek letters
zontal line at
is common practice to give
2729°F as being a peritectic reaction. The equa
peritectic reaction may
Ippear on the diagram. The
Liquid + 8
Jickel Wrought Iron
austenite
heating
he solubility is much grea r. Ihe ™
WROUGHT IRON
I
whi e in f.cc. y Fe t
60,000 45,000 22
40
22
=
The maximum solubility of carbon in b.c.c. 8 Fe is O.IOpercent (point M).
NICKEL '
SI
be written as
t
influences the 8 7 allotropic change. As J 2720 F at 0.10 per perature of the allotropic change increases from 2554 to the portion cent C Consider the significance of the line NMPB. On cooling,
8 /1a
m
232
THE IRQ
INTRODUCTION TO PHYSICAL METALLURGY
The eutectic mixture is not_usuaJJy st ite is not stable at room temperatu duringcooliDg. There is a small solid solution an
A 2800
Liquid + S 0 10 %c p .
Li_
,
16660F represents the change in ci
0 50% C .
y to b;c.c. a. That area
2720
8
8
3
#/ 0.18 %C i
i
is a solid so
solved in b.c.c. a Fe and is called f zontal line HJK, which represents V
point, J, is at 0.80 percent C and
+ Austenite
-
.
transform into the very fine eutec
called pearlite. The equation may
Austenite
2554
A/
cool
a 4
.
-
Fig. 7-6
m
The delta region of the iron-iron carbide diagram
iquld heating
Below the eutectoid temperature I of ferrite and cementite as indicate On the basis of carbon contenLU
.
NM represents the beginning of the crystal structure change from b.c c 8 Fe'; .
.
to f.c.c, y Fe for alloys containing less than 0.10 percent C. The portion MP|
percent carbon argJsnown as sfg
represents the beginning ofthis crystal structure change by means of a peritectic reaction for alloys between 0.10 and 0.18 percent C. For alloys con-: taining less than 0.18 percent C, on cooling, the end of the crystal structure change is given by the line NP. The portion PB represents the beginning and the end of the crystal structure change by means of the peritectic reaction. In other words, for alloys between 0.18 and 0.50 percent C, the allo-
percent carbon jire known as cas r/M 2800
.
ml--2554
tropic change begins and ends at a constant temperature. Notice that any|
L + Austenite
alloy containing more than 0.50 percent C will cutthe diagram to the right of '
point B and will solidify austenite directly. The delta solid solution and the ; mm
'
treatment is done in the delta region, there will be no reason to refer to this portion of the diagram again.
2065oF
Austenite
allotropic change will be completely bypassed. Since no commercial heat
S!
Aus.+ Fer,
'
Austenite -
o
j
| 1666
G
The diagram in Fig. 7-7, which has the common names inserted, shows
I3330F
H
a eutectic reaction at 2065oF. The eutectic point, E, is at 4.3 percent C|
£utectoid ; pearlite)
Ferrite
and 2065oF. Since the horizontal line CED represents the eutectic reaction, whenever an alloy crosses this line the reaction must take place.
Ferrite +JCementite
Any liquid that is present when this line is reached must now solidify intothe very fine intimate mixture of the two phases that are at either end of the horizontal line, namely austenite and iron carbide (called cementite). This eutectic mixture has been given the name ledeburite, and the equation
0
08
Hypo-
| Hyper
eutectoid| eutectoid
may be written as
Hypoeutectic Ca;
-
Steels-
-
Percent carbon by we
Liquid
cooling :
heating
austenite + cementite -
v
,
-
eutectic mixture-ledeburite
w-.
.i,.!.?.
v.-:-
T
I E
'
Fig 7-7 The iron-iron carbide equilibrium dlagran With the common names for the structures.
m
THE IRON-IRON CARBIDE EQUILIBRIUM DIAGRAM
1
233
The eutectic mixture is not usuailv seen in the microstrudture, since austen-
stable at room temperature and must undergb another reaction
ite te not
_
_
durmg coolipg.
] Thereisa small solid solution area to the left of line GW. We know that _
i 1
0 50% C
,
.
1
+ Austenife
f
16660F represents the change in crystal structure of pure iron from f.c.c. y to b.c.c. a. That ar6a is a solid solution of a small amount of carbon dissolved in b.c.c. a Fe and is called ferrite. The diagram shows a third horizontal line HJK, which represents a eutectoid reaction. The eutectoid point, J, is at 0.80 percent C and 13330F Any austenite present must now .
transform into the very fine eutectoid mixture of ferrite ayd cementite,
1
-
called pearlite. The equation may be written as
ferrite + cementite
\
,
I
/
9
heating
eutectoid mixture-pearlite
Below the eutectoid temperature line every alloy will consist of a mixture of ferrite and cementite as indicated.
On thejjasis of carbon conteaLLLis common practice to divide the iron-
structure change from b c c 8 Fe an 0.10 percent C The portion MP tructure change by means of a pernd 0.18 percent C For alloys con.
.
irori carbide diagram intojwo-pacts. Tbgse alloys containing less than 2
.
percent carbon
.
are
knowh as _
_
stsels, and those containing more than-2
_
percent carbon are known as casf irons. The steel range
,
.
ng, the end of the crystal structure /Jtt 2600
tion PS represents the beginning ige by means of the peritectic re-
in 0.18 and 0.50 percent C
,
IfiH 2554
the allo-
itant temperature Notice that any
F
Austenite
|:fi?iere will be no reason to refer to
Aus.+ Fer.
common names
inserted, shows
1333°F
Eutectoid
,
(pearlite!
Ferrite
a
-
Ferrite +,Cementite
.
. .
.
jhases that are at either end of the
.
carbide (called cementite) This me ledeburite, and the equation .
.
1
.
:e
+ cementite
Mxture-ledeburite
3
Austenite + Cementite
G
ctic point, £ is at 4.3 percent C ED represents the eutectic reac-
e the reaction must take place s reached must n6w solidify into
D
Eutectic
(ledeburite!
1666
:
Cementite
i will cut the diagram to the right of
The delta solid solution and the ;assed. Since no commercial heat
.
I 4
L \ Austenite
.
.
3 0
«
Hypereutectoid
6 67
4 3
2
08 Hypoeutectoid
.
.
FexC
Hypoeutectic
\
Hypcreutectic
Cast irons
Steels
Percent carbon by weight ,
Fig. 7-7 The iron-iron carbide equilibrium diagram labeled with the common names for the structures.
)
is further sub-
1' \ 234
divided by the tectoid ar laaO ConLent (0.8 percent C). SteejsjCQDtainn less_than .0.8- pereent © are-cailedJiypQeuiecte/d-steeys, while those col tajnin bie rx0J and.2I ,
.
-
0
4
_
cast iron range may also be subdivided by eutectic car.b6n content'(4. percent C). Cast irons that contain less than 4.3 percent C are known hypuButectjc cast irons, whereas those that-G04 m-iwjceJhanA3 Jjercei C arg called hypereutectic casLimns. ' Definition of Structures The names which, for descriptive or commemora* tive reasons, have been assigned to the structures appearing on this dia-'
ill
.
5'
'
rs
-
J
gram will now be defined.
Cementite or iron carbide, chemical formula Fe
MM
ii
THE IRON-
INTRODUCTION TO PHYSICAL METALLURGY
contains 6 67 fl£teent1 ,
,
ft
'
C by weight. It is a typical hard and brittle interstitial compound of low ,
5
If
tensile strength (approx. 5,000 psi) but high compressive strength the hardest structure that appears on the diagram. Its crystal structure is orthorhombic.
Austenite is the name
ft
given to the y solid solution. It is an interstitial _
,
is 2 pec£grit C-at 2065oF (point C). Average properties are: tensile strength 150,000 psi; elongation, 10 percent in 2 in.; hardness, Rockwell C 40 ap* prox.; and toughness, high. It is normally not stable at room temperature Under certain conditions it is possible to obtain austenite at room temperature, and its microstructure is shown in Fig. 7-8a.
(a)
,
_
m
m
solid solution of carbon dissolvedln y (i.c.c.) iron.. Maximum solubility .
IS
<;X"':
Si,'.if'
,
.
Ledeburite is the eutectic mixture of austenite and cementite '"
tains 4.3 percent C and is formed at 2065T.
if
It coik
"
0 _
Ferrite is the name given to the a solid solution. It is an interstitial solid solution of a small amount of carbon dissolved in a (b.c.c.) iron (Fig. 7-86). The maximum solubility is 0.025 percent C at 13330F (point H), and it dissolves only 0.008 percent C at room temperature. It is the softest structure that appears on the diagram. Average properties are: tensile strength, 40,000 psi; elongation, 40 percent in 2 in.; hardness, less than Rockwell CO
0
.7 5
7
Cementit e
g
or less than Rockwell B 90. Pear//fe
_
Xpoint J) is the eutectoid mixture containing 0.80 percent
and is formed at 13330F on very slow cooling. It is a very fine platelike orl lamellar mixture of ferrite and cementite.
The fine fingerprint mixture
i 1
cementite. The same structure, magnified 17,000 times with the electron |
n
microscope, is shown in Fig. 7-dd. Average properties are: tensile strength, well B 95,-100, or BHN 250-300. 6
Carbon Solubility in Iron Austenite, being f.c.c. with four atoms per unit| cell, represents a much denser packing of atoms than ferrite, which is b.c.c.| with two atoms per unit cell. This is shown by the expansion that takesj .
1
A
called pearlite is shown in Fig. 7-8c. The white ferritic background or ma-' trix which makes up most of the eutectoid mixture contains thin plates of <
120,000 psi; elongation, 20 percent in 2 in.; hardness, Rockwell C 20, Rock!
m
Ferrite
il , ! :i!l. 7-8
'
i
The microstructure of (a) austenite, 500X; (b)
ie. 100X; (c) pearlite, 2,500X; (cf) pearlite, electron i ih, 17,0OOX: enlarged 3X in printing, (a, b, and c, R sii Laboratory, U.S. Steel Corporation.)
www
wwmhiiiiihiii
mr
n
THE IRON-IRON CARBIDE EOUIUBRIUM DIAGRAM 235
it (0.8 percent C). Steel jCflataining
. -I
i '
-
!'
f
vjautectoid-steBJs, while those confMre calledi>w3ersutecjo/d steelsTThe ed by eutectic carJaon content' (4.3
m mi
4
ss than 4.3 percent C are known as ! that contain morfi than 4J3_perc ent "
1
m
};h, for descriptive or commemora-
10
,
;e structures appearing on this dia-
i
Qrmula Fe,C. contains 6.67 percent
|
0m
i
mm
m
m
J arittle interstitial compound of low |
8
X
I high compressive strength, It is the diagram. solid solution.
Its crystal structure
TJ
1
i
It is an interstitial
r
ff.c.c.) iron.. Maximum solubility
ge properties are: tensile strength, ;in.; hardness, Rockwell C 40, ap-
'
;V
(a)
35
:::jly not-stable at room temperature.
V; to obtain austenite at room tem-
i in Fig. 7 6a.
1
$
austenite and cetfientite. It con.
I solution. It is an interstitial solid
I '9
i Ferrite i .
nil
iolved in a (b.c.c.) iron (Fig. 7-Qb).
Cemenfite!
C at 13330F (point H), and it disiperature. It is the softest struce properties are; tensile strength,
;
I
i
hardness, less than Rockwell CO
\1
: : .
jd 17,000 times with the electron
:
.
.
..
(c)
iVX'p properties re: tensile strength, .
.
i; hardness, Rockwell C 20, Rock-
Fig. 7-8 The mlcrostructure of (a) austenite, 500X; (6) *
:
m < ferrite, 100X; (c) pearlite, 2,500X; (d) pearlite, electron micro-
-
!
|f.c.c. with four atoms per unit iwn by the expansion that takes
3
'
>
I
graph, 17,000X; enlarged 3X in printing, (a, b. and c. Research Laboratory, U.S. Steel Corporation.)
3toms than ferrite, which is b.c.c.
d)
:4V
236
|
INTRODUCTION TO PHYSICAL METALLURGY
place when austenite changes to ferrite on slow cooling
.
THE IRON-IR<
If the iron at
are assumed to be spheres it is possible, from the lattice dimensions assuming the distance of closest approach to be equal to the atom di eter, to calculate the amount of empty space in both crystal structures
ii
AllOV
A
A
,
Austenite
A
A
.
calculation shows that the percentage of unfi[led
.
A
J:
space in the f.c c la
1666
.
is 25 percent and in the b.c.c. lattice 32 percent. In both austenite
Fefriie
,
.
H-Au; Austenite
ferrite, the carbon atoms are dissolved interstitially that is, in the until spaces of the lattice structure In view of the above calculations it. seem strange that the solubility of carbon in austenite is so much grei
Prooutla doit
,
,
than it is in ferrite a study of Fig 7-9. .
.
a:
territe\
.
c
V
10
This seemingly unusual behavior may be explained The largest hole in b.c.c. ferrite is halfway between
A
4,
J
1333
Ferrite
/[p*|2 Pearlite! Ferrite I
I
Peorlite !
Peatiite
Peorl + Cem
Ferrite + Cemei
center of the face and the space between the two corner atoms. Two of four possible positions for a carbon atom on the front face of a bo centered cube are shown in Fig 7-9. The largest interstitial sphere U would just fit has a radius of 0 36(10"8) cm. The largest hole in f c c a.
i
0 7
.
.
0.2
0.8
I Hypoeutectoidl steel
.
.
.
.
Percent carbon t
tenite is midway along the edge between two corner atoms. One possiL
position for a carbon atom on the front face of a face-centered cube
7-10 Schematic representation of the changes in
M
.
shown in Fig. 7-9 The largest interstitial sphere that would just fit ha| radius of 0.52(10_8) cm Therefore, austenite will have a greater solubil .
.
foe carbon than ferrite Since the carbon atom has a radius of abqa 0 70(10-8) cm the iron atoms in austenite are spread apart by the solutL of carbon so that at the maximum solubility of 2 percent only about 1 .
.
Structure during the slow cooling of 0.20 percent Bdnste l. (a) Austenite; (b) formation of ferrite grains
iteniro gi in boundaries; (c) growth of ferrite grains-
(position Li austenite is now 0.8
percent carbon; (d) °
lenite transforms to pearlite at 1333
F
.
,
,
,
percent of the holes are filled. The distortion of the ferrite lattice bytl carbon atom is much greater than in the case of austenite; therefore,!
slow cooling, from the austenite rang
Alloys containing more than 2 percent Alloy 1 (Fig. 7-10) is a hypoeute
carbon solubility is much more restricted 7-7 Slow Co oling of Steel The steel portion of the iron-iron carbide diagram, of greatest interest and the various changes that take place during thevei .
carbon. In the austenite range, this solid solution. Each grain contains
,
I
spaces of the f.c.c. iron lattice struc nothing happens until the line GJ is c as the upper-critical-temperature lit
labeled A3. The allotropic change ft 16660F for pure iron and decreases i content, as shown by the Aj line. The at the austenite grain boundaries (F
Fig. 7-9
.
.
.
.
lattice is indicated by the black atom
tenite, so that, as cooling progress?
can
the remaining austenite becomes r;
with
two of the four possible positions on one face shown here as filled The f.c.c. lattice has far fewer holes but, as shown by the black sphere the hole is much larger (By permission
gradually moving down and to the
.
,
,
from Brick
,
HJ is reached at point x2. This line ture line on the hypoeutectoid sid
.
Gordon, and Phillips, "Structure and Properties
of Alloys," 3d ed
.
,
McGraw-Hill Book Company 1965.)
'
eutectoid-temperature line and is
,
i y
m k i -
-
.
.
must come out of solution before .
.
The maximum-diameter foreign sphere (black) that enter the b.c c
th thi
The carbon which com s out of soIl
Interstices of the b c c (left) and f c c (right). .
very little carbon, in those areas
i
i
tJ
i
THE IRON-IRON CARBIDE EQUILIBRIUM DIAGRAM 237
3rrite on slow cooling If the iron atomsQi issible from the lattice dimensions and: .pproach to be equal to the atom diam.
/
-
.
*V:r
)ty space in both crystal structures
.
.
A
1*1
The
Ll.
A
1666
6
i
/
i1
.
W9%
.
(A)
,
yiew of the above calculations, it may arbon in austenite is so much greater jnusual behavior may be explained by !in b.c.c. ferrite is halfway between the
Pearlite
Ferrite n
[C]
Cementite
J
At
i=J333
ferrite
m
Austenite
3
Fefnte
Austenite H-Alis
Proeutectoid
t
:
-
Austenite
A
ige of unfilled space in the f c c lattice ice 32 percent In both austenite and ved interstitially that is, in the unfilled. .
Alloy
A
A
,
Pearlite + Cementite
Ferrite +
Pearlite
Pearlite
.
:
Ferrite + Cementite
iiWtteen the two corner atoms Two of the n atom on the front face of a body-
i
.
9
-
0 .
The largest interstitial sphere that
.
i siA i -
!
.
0,2
0 8
I Hypoeutectoidi steel
8) cm The largest hole in f.c.c aus.
.
/een two corner atoms One possible
Percent carbon by weighl-
.
r -ront face of a face-centered cube is i
-ivititial sphere that would just fit has a
v;viustenite
will .have a greater solubility
t Flg. 7-10 Schematic representation of the changes in microstructufe during the slow cooling of 0.20 percent
1
. carbon atom has a radius of about :
;
carbon steel, (a) Austenite; (b) formation of ferrite grains at
m *
ignite are spread apart by the solution
austenite giVin boundaries; (c) growth of ferrite grainscomposition austenite is now 0.8 percent carbon; (d)
austenite trans'orms to pearlite at 1333°F.
.
olubility of 2 percent only about 10 ,
listortion of the ferrite lattice by the the case of austenite; therefore the ,
ted. :
-
of the iron-iron
carbide diagram is anges that take place during the very
i
i
slow cooling, from the austenite range, of several steels will be discussed.
Alloys containing more than 2 percent carbon will be discussed in Chap. 11. Alloy 1 (Fig. 7-10) is A hypoeutectoid steel containing 0.20 percent . carbon. In the austenite range, this alloy consists of a uniform interstitiaj solid solution. Each grain contains 0.20 percent carbon dissolved in the spaces of the f.c.c. iron lattice structure (Fig. 7 10a). Upon slow cooling -
nothing happens until the line GJ is crossed at point x,.
This line is known
as the upper-critical-temperature line on the hypoeutectoid side and is
labeled A3. The allotropic change from f.c.c. to b.c.c. iron takes place at 1666°F for pure iron and decreases in temperature with increasing carbon content, as shown by the A3 line. Therefore, at x,, ferrite must begin to form
at the austenite grain boundaries (Fig. 7-1 Ob). Since ferrite can dissolve very little carbon, in those areas that are changing to Jerrite the carbon must come out ot solution before the atoms rearrange themselves to b.c.c/
The carbon which corn s Out of solution is dissolved in the remaining austenite, so that, as cooling progresses and the amount of ferrite increases,
the remaining austenite becomes richer in carbon. Its carbon content is
gradually moving down and to the right along the A3 line. Finally, theMine HJ is reached at point Xj. This line is known as the lower-ctitical-tempera-
n
ture line on the hypoeutectoid side and is labeled A,. The A, line is thie eutectoid-temperature line and is the lowest temperature at which f.c.c. s
m
i ;
J
.
238
I
THE IRON
II INTRODUCTION TO PHYSICAL METALLURGY
iron can exist Lnder equilibrium conditions. Just above the /A, line
,
a
microstructure consists of Approximately 25 percent austenite and 75
cent ferrite (Fig. 7-10c). TheVemaining austenite, about 25 percent ofi
total material (Rule II) and containing 0.8 percent carbon, nowexperiena the eutectoid reaction
1
Austenite
cooling
I
Jerrite + cernentite,
heating
V
pearlite
Note that it is only austenite which is changing at the A line. Therefi when the reaction is complete the microstructure will show approximate 25 percent pearlite and 75 percent ferrite (Fig. 7-1 Od).
!
.
Let us consider the eutectoid reaction in a little more detail
i
.
Austen
j!
/S
S *i V'' i-iT ' JCV--
i
is to change to ferrite. Austenite is an interstitial solid solution with eaj remaining grain dissolving 0.8 percent C in f.c.c. Fe. Ferrite, however ! b c c Fe and dissolves very little carbon, so the change in crystal struct ,
.
.
.
cannot occur until the carbon atoms come out of solution. Therefore
,
first step'is the precipitation of the carbon atoms to form plates of cemeiiK'
ite (iron carbide). In the area immediately adjacent to the cementlte plai I the iron is depleted of carbon, and the atoms may now rearrange thei» I--
selves to fdrm b.c.c. ferrite. Thin layers of ferrite are formed on eachsii % of the cernentite plate. The process continues by the formation of alter K
nate layers of cernentite and ferrite to give the fine fingerprint mixtun b
known as pearlite. The reaction usually starts at the austenite grain bountf |-
ary, with the pearlite growing along the boundary and into the grain, s« % Fig, 7-11.
m
Since ferrite and pearlite are stable structures, the microstructure remains substantially the same down to room temperature and consists
approximately 75 percent proeutectoid ferrite (formed between the
and /A, lines) and 25 percent pearlite (formed from austenite at the
line). Figure 7-12a shows the microstructure of a 0.2 percent C stee!(i slow-cooled. As predicted, it consists of 75 percent proeutectoid ferrite] (light areas) and 25 percent pearlite (dark areas). The dark areas in thft
/
"
f
1 i
A
.
mi
micro certainly do not look like a mixture, which pearlite is supposedtti|
mm a
J Austenite
-Cernentite
Fig 7.12 Photomicrographs of (a) 0.20 percent carb steel, slow-cooled, 100X; (b) same as (a), but at 500X boundary
Fig. 7-11 pearlite.
ic) 0 40 percent carbon steel, slow-cooled, 100X; (of)
toid (0 80 percent carbon) steel, slow-cooled, 500X. f
.
samples etched with 2 percent nital. Dark areas are p
Schematic picture of the formation and growth of
He; light areas are ferrite.
"
T
.
.
.
.
J
THE IRON-IRON CARBIDE EQUILIBRIUM DIAGRAM
239
n
J
onditions. ..
Just above the A. line the
I
,
nately 25 percent austenite and 75 per-
..
,
Jping austenite about 25 percent of the f | g 0.8 percent carbon, now experiences ; | % ,
Mm
:" :
.
-
.
11
: "
Jerrite + cementite
i
,
pearlite
1
m
i!
4
js changing at the A, line. Therefore, 1 icrostructure will show approximately i I :
1
ii
mrrite (Fig. 7-1 Od).
?S:|tion in a little more detail
.
Austenite
:
interstitial s'olid solution with each,,- f
nt C in f.c.c. Fe.
Ferrite however, is ,
:
on, so the change in crystal structure i the V/'bon atoms to form plates of cementcome out of solution. Therefore
'
,
ijately adjacent to the cementite plate $ >he atoms may now rearrange them-
XVs of ferrite are formed on each side continues by the formation of alter-
{
,
to give the fine fingerprint mixture
y starts at the austenite grain bound-
Bp
tm&K
\
I'
ie boundary and into the grain, see *
3
» .V*
structures, the microstructure re-
;
room temperature and consists of id ferrite (formed between the A, (formed from austenite at the A,
'
a
i <
1
Substructure of a 0.2 percent C steel of 75 percent proeutectoid ferrite lark areas). The dark areas in this ure, which pearlite is supposed to
f
I .
ft V
i
4 Ml) Fig. 7-12 Photomicrographs of (a) 0.20 percent carbon steel, slow-cooled, 100X; (t>) same as (a), but at 600X; i ;
if
(c) 0.40 percent carbon steel, slow-cooled, 100X; (d) eutectoid (0.80 percent carbon) steel, slow-cooled, 500X. All samples etched with 2 percent nital. Dark areas are pearlite; light areas are ferrite.
l
240 INTRODUCTION TO PHYSICAL METALLURGY
THE IRON-I
be. Higher.magnification (Fig 7-12b), however reveals the fine fingerprii| mixture of pearlite The changes just described would be the same for any hypoeutectoii steel. The only difference would be in the relative amount of ferrite ai
along the Acrn line toward point J. The from x3 to X4, the excess carbon abc austenite is precipitated as cementite
The closer the carbon content to the eutectoid composition (0 (j
line is called the lower-critical-temf. side and is labeled A3A. Just above thi
.
,
.
pearlite
.
.
percent C), the more pearlite will be present in the microstructure Th| microstructure of a 0 4 percent C steel'slow-cooled (Fig 7-12c} shows apf1 .
,
.
.
proximately 50 percent pearlite, while the eutectoid composition (0 8 pen '
.
cent C) shows 100 percent pearlite (Fig 712d).
\
.
Alloy 2 (Fig
.
7-13) is a hypereutectoid steel containing 1 percent carbonj
In the austenite range this alloy consists of a uniform f c c solid solution? with each grain containing 1 percent carbon dissolved interstitially (Fig |
m
,
.
.
.
(Fig. 7-13b,c). Finally, the eutectoid-t€ largely of austenite, with the excess | surrounding the austenite grains (Fig. ite on the right side of the line, the at 10 .
% cementite = 6 67 .
.
7-13a)
.
Upon slow cooling nothing happens until the line CJ is crossed al|
and the amount of austenite would b
point X3. This line is known as the upper-critical-temperature line on Vntl
hypereutectoid side and is labeled A amount of carbon that can be dissolved in austenite as a function of tem-| cm.
The Acm line shows the maximum|
perature. Above the Acm line, austenite is an unsaturated solid solution!,
As the tempi
At the Acm line, point x3 the austenite is saturated in carbon perature is decreased, the carbon content of the austenite that is, the maxk mum amount of carbon that can be dissolved in austenite moves down! ,
.
,
,
6 67 .
% austenite = 6 67 .
The A,,, line for hypereutectoid stee end of the allotropic change from f.
same process described earlier, the
percent carbon) transforms to the e
i
At room temperature the microstruc
(formed from austenite at the A,,, lii
c
Alloy
eutectoid cementite (formed betwee
a
A
2
cementite
A
(a:
a
a
Austenite
acn-
1666
Austenitv
J
Excess
a
a
network
ntite
'
Cementite
i
a
+ =1
a
1
,
1
ic)
Pearlite
Cementite
+
network
Cementite
.
I
Percent corbon by weight
microstructure during the slow cooling of a 1 0 percent carbon steel, (a) Austenite; (b) formation of excess cementite at austenite grain boundaries; (c) growth of excess cement ite to form a network-austenite composition Is now 0 8 percent carbon; {d) Austenite transforms to pearlite at .
-
.
1
V '
1
I
i i
.
i
lines, the A3 and the Acrn. The forr
structure. As was pointed out in low tensile strength, while cementr The combination of these two phas
Fig. 7-13 Schematic representation of the changes in
1333°F.
-
iar (platelike) structure of pearlite.
phases and the way in which thes
2
I Hypereutectoid steels ,
7
whereas the latter involves only a c The mechanical properties of an i
Ferrite + Cementite
08
work generally increases. Figure
Note the difference in significa
Pearlite
0
the same for any hypereutectoid stei of the alloy increases, the thickne
percent carbon steel. Both photomi
i
J
g- 1333
at Fig. 7-14a, particularly where the white proeutectoid cementite netwo
produces an alloy of much great phase. Since the amount of pearli content for hypoeutectoid steels, also increase up to the eutectoid c ductility, as expressed by percent
r
""-:
.
:JlVJ1...w.
r
,H.
[(cW(V,..
*
m
.
f
1 .
,; _
THE IRON-IRON CARBIDE EQUILIBRIUM DIAGRAM
iwever
,
reveals the fine fingerprint
along the Acm line toward point J. Therefore, as the temperature decreases from X3 to xA, the excess carbon above the.amount required to saturate austenite is precipitated as cementite primarily along the grain boundaries
7
the same for any hypoeutectoid he relative amount of ferrite
;
and
.
,
.
r
.
eel containing 1 percent carbon W:Of a uniform f c c solid solution
largely of austenite, with the excess proeutectoid cementite as a network surrounding the austenite grains (Fig. 7-13c). Applying Rule II with cementite on the right side of the line, the amount of cementite would be
.
.
'
'
.
1 0-0 8 .
.
x 100 = 3.4%
% cementite :
.
bon dissolved interstitially (Fig.
i n
n
,
"
i
line is called the lower-critical-temperature line on the hypereutectoid side and is labeled A,* Just above the Aj , line the microstructure consists
ow- cooled (Fig 7-12c) shows ap} eutectoid composition (0.8 per.12c0
I
(Fig. 7'13£>,c). Finally, the eutectoid-temperature line is reached at x4. This
to the eutectoid composition (0.8 jsent in the microstructure. The
'
241
.1
6 67-0,8 .
ns until the lirte CJ is crossed at critical-temperature line on the he Acm line shows the maximum
and the amount of austenite would be
-
> austenite as a function of temv 3 an unsaturated solid solution. v aturated in carbon As the tem)f the austenite that is, the maxi-
'
v
.
.
.
The A3i{ line for hypereutectoid steels represents the beginning and the
.
-
.
,
olved in austenite, moves down
(formed from austenite at the A3A line) and a network of 3.4 percent proeutectoid cementite (formed between the Acm and A3 , lines). Look closely at Fig. 7-14a, particularly where the pearlite areas meet, to see the thin, white proeutectoid cementite network. The story just described would be the same for any hypereutectoid steel, slow-cooled. As the carbon content ,
e
t k
end of the allotropic change from f.c.c. austenite to b.c.c. ferrite. By the same process described earlier, the remaining austenite (containing 0.8 percent carbon) transforms to the eutectoid mixture, pearlite (Fig. 7'10cy). At room temperature the microstructure consists of 96.6 percent pearlite
.
;
6 67 - 1.0 .x 100 = 96 6% 6 67 - 0.8 .
% austenite :
t '
J
I
of the alloy increases, the thickness of the proeutectoid cementite network generally increases. Figure 7-14£) shows the microstructure of a 1.2 percent carbon steel. Both photomicrographs show very clearly the lamellar (platelike) structure of pearlite. Note the difference in significance of the upper-critical-temperature lines, the A3 and the Acm. The former line involves an allotropic change, whereas the latter involves only a change in carbon solubility. The mechanical properties of an alloy depend upon the properties of the
phases and the way in which these phases are arranged to make up the structure.
As was pointed out in Sec. 7-5, ferrite is relatively soft with .
low tensile strength, while cementite is hard with very low tensjle strength. The combination of these two phases in the form of the eutectoid (pearlite) produces an alloy of much greater tensile strength than that of either phase. Since the amount of pearlite increases with an increase in carbon content for hypoeutectoid steels, the strength and Brinell hardness will also increase up to the eutectoid composition of 0.80 percent carbon. The ductility, as expressed by percent elongation and reduction In area, and
5
-
1
1
i
I
.:
242 INTRODUCTION TO PHYSICAL
METALLURGY
THE I
V
m
3 '
i
a
390
mm
_
S350
1|330
Brmell ho
270
i
Chorpy impocl
210 J
'
15
*
1
ISO
« 9 -
120
5
19,
-
si e
A
\
r
Reduction Df area
90
60 :
.
0
Ion ganon in
-
r
(a)
in
0 0 0.1 0.2 0.J 0.4 0.5 0.6 0.7 C.8 0.9
5
.
.
5
Carbon (percen'
/
-
"
rig. / 15 Effect of carbon on mechanical properl hot-worked steel. (By permission from S/sco, Allc Iron and Carbon, McGraw-Hill Book Company, N 1937.) "
"
m
Cementite network
variation in properties is shown produce significant changes in t r in detail in Chap. 8. 7 ftflThe Critical-temperature Lines Ii v temperature lines are shown as
Peorlite
and are sometimes indicated a;
actually determined, it is found ture. r
.
.4
-
,
The critical line on heatir
cooling. To distinguish critical cooling, the former are callec which means heating) and the
'
5?
dissement, which means cooli:
(b)
hypoeutectoid steel on heating Fig. 7-14 Photomicrographs ot (a) 1 percent carbon steel slow-cooled 500X; (b) 1,2 percent carbon steel slow-
cooling Ar3. The rate of heating and cool
,
,
cooled, 300X Pearlite areas surrounded by a white pro eutectoid cementite network Note the increase in thick ness of the cementite network with the increase in carbon .
-
gap between these lines. The
.
-
nearer will the two lines apprc
content
.
li
n impact strength decrease with increasing carbon content Beyond the eu- S
tectoid composition
.
the strength levels off and may even show a decrease !
due to the brittle cementite network The Brinell hardness ,
however
tinues to increase due to the greater proportion of hard cementite coiThe!kI .
,
.
i
1*9*-
t
11
heating and cooling they wou perature. The results of thermal analy
erage heating and cooling rate
gram shows the effect of this of the critical lines. Also show
,
the magnetic change in iron a
THE IRON-IRON CARBIDE EQUILIBRIUM DIAGRAM
t:,
243
390
I
£
.
"
3
r40
!S 360
_
§330 Brmell hardness
300
100
-
1' r
Charpy impact
J
--
80
'
V
°
.
30
90
?70
70
-
20.
2
3
.
180
s
150
K
7
50
Yield strength
30
.
-
I
20
;A
UJ
0
1.1 1.2 1.3 1.4 1.5
Carbon (percent)
1
3
30
10
Elongation in 2 in
0 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0,9 1.0
/i
10 -iS
-
0
40
90 ,2
a
50
Reductmn of area
-
>« 120
ii
60
Tensile strength
Fig. 7-15 Effect of carbon on mechanical properties of hot-worked steel. (By permission from S/'sco, "Alloys of Iron and Carbon," McGraw-Hill Book Company, New York, 1937.)
4 J
6mm
variation in properties is shown in Fig. 7-15. The heat treatment of steel to produce significant changes in the mechanical properties will be discussed
V
/
. in detail in Chap
.
8.
7;8phe Critical-temperature Lines In Fig. 710, the upper- and l£wej-criti cal4
v 7sS St
5 (i
f
f
J;
temperature lines are shown as single lines under equilibrium conditions and are sometimes indicated as Aev Ae,, etc. When the critical lines are actually determined, it is found that they do not occur at the same temperature. The critical line on heating is always higher than the critical line on cooling. To distinguish critical lines on heating from those occurring on cooling, the former are called Ac (c from the French word chauffage, which means heating) and the latter Ar (r from the French word refroidissement, which means cooling). Therefore, the upper critical line of a hypoeutectoid steel on heating would be labeled Ac3 and the same line on cooling Ar3.
The rate of heating and cooling has a definite effect on the temperature gap between these lines. The slower the rate of heating and cooling the nearer will the two lines approach each other, so that with infinitely slow heating and cooling they would probably occur at exactly the same temperature.
i carbon content Beyond the eu.
jff and may even show a decrease f Brinell hardness, however )portion of hard cementite .
,
T
con-
The results of thermal analysis of a series of carbon steels with an average heating and cooling rate of 110F/min are shown in Fig 7-16. The dia.
gram shows the effect of this rate of heating and cooling on the position of the critical lines. Also shown are the /\c2 and /4r2 lines, which are due to
The
the magnetic change in iron at 14140F
.
1 i
«
IS 244
THE IROI
INTRODUCTION TO PHYSICAL METALLURGY
steel. SAE specifications now emplo nations as the AISI specifications, wi
1700
n 1600
X
The basic numbers for the four-di
bon and alloy steels with approxima
Austenite
are:
Ac
s .
n
Ar3
N
Ferrite + austenite
s
10xx
Basic open-hearth and acid bess
11 XX
Basic open-hearth and acid bes; phorus
N
Ac2
,
7
V:
Af2
12xx
Basic open-hearth carbon steels
13xx
Manganese 1.75
23xx
Nickel 3.50 (series deleted in 19!
Ac,
1300 h _
t
Ferrite
/t
An
-_
__
25xx
Nickel 5.00 (series deleted in 19;
31 xx
Nickel 1.25, chromium 0.60 (seri
33xx
Nickel 3.50, chromium 1.50 (seri
Ferrite + cementite 1220
-
'
Fig. 7-16
L
L
1
_
0
0.10
0.20
0 30 .
-
0.40 0 50 0.60 Percent carbon content .
0 70 .
.
0.80
0 90 .
Hypoeutectoid portion of the iron- iron carbide
diagram. (By permission from Dowdell General Metallography," John Wiley & Sons Inc.. New York, 1943 )
40xx
Molybdenum 0.20 or 0.25
41 xx
Chromium 0.50, 0.80, or 0.95, m
43xx
Nickel 1.83, chromium 0.50 or 0
44xx
Molybdenum 0.53
4fixx
Nickel 0.85 or 1.83, molybdenur
47XX
Nickel 1.05, chromium 0.45, mo
48xx
Nickel 3.50, molybdenum 0.25
50xx
Chromium 0.40
51 xx
Chromium 0.80, 0.88, 0.93, 0,95
5xxxx
Carbon 1.04, chromium 1.03 or
61 xx
Chromium 0.60 or 0.95, vanadii
86xx
Nickel 0.55, chromium 0.50, mc
87xx
Nickel 0.55, chromium 0.50, mc
"
,
,
.
i
7-9
Classification
of Steels Several methods may be used to classify steels:
Method of Manufacture
This gives rise to bessemer steel
,
_
open-hearth %
steel, electric-furnace steel crucible steel, etc. ,
Use
This is generally the final use for the steel such as machine steel ,
m
,
spring steel, boiler steel structural steel, or tool steel Chemical Composition This method indicates by means
K of a numbering K system, the approximate content of the important elements in the steel, fr ,
.
,
This is the most popular method of classification and will be discussed
in
greater detail.
The steel specifications represent the results of the cooperative effort
88xx
Nickel 0.55, chromium 0.50, mc
of the American Iron and Steel Institute (AISI) and the Society of Automotive Engineers (SAE) in a simplification program aimed at greater efficiency in meeting the steel needs of American industry The first digit of the four- or five-numeral designation indicates the type
92xx
Silicon 2.00
93xx
Nickel 3.25, chromium 1.20, mc
98xx
Nickel 1.00, chromium 0.80, mc
.
to which the steel belongs Thus 1 indicates a carbon steel 2 a nickel steel
94Bxx Nickel 0.45, chromium 0.40, mc '
.
,
Series deleted" does not mean
,
3 a nickel-chromium steel etc.
"
In the case of simple alloy steels the second digit indicates the approximate percentage of the predominant alloying element The last two or three digits usually indicate the mean carbon content divided by 100 Thus the symbol 2520 indicates a nickel steel ,
J
,
simply means that the tonnage pre included in the listing of standarc
.
..I
ically by AISI. Some reioresentative standardcarbon and free-machining stee
.
of approximately 5 percent nickel and 0 20 percent carbon In addition to the numerals AISI specifications may include a letter .
.
,
prefix to Indicate the manufacturing process employed in producing the
If
'
Table 7-4.
THE IRON-IRON CAF-tBIDE EQUILIBRIUM DIAGRAM
245
steel. SAE specifications now employ the same four-digit numerical designations as the AISI specifications, with the elimination of all letter prefixes.
4
The basic numbers for the four-digit series of the various grades of carbon and alloy steels with approximate percentages of identifying elements
Austenite
are:
lOxx i
.
11 xx
Basic open-hearth and acid bessemer carbon steels
Basic open-hearth and acid bessemer carbon steels, high sulfur, low phosphorus
12xx
9
-
nentite
0 60
0 70
.
.
1
0 80
0 90
.
.
7:m
may be used to classify steels: to bessemer steel open-hearth
|
40xx
Molybdenum 0.20 or 0.25
41xx
Chromium 0.50, 0.80, or 0.95, molybdenum 0.12, 0.20, or 0.30
43xx
Nickel 1.83, chromium 0.50 or 0.80, molybdenum 0.25
44xx
Molybdenum 0.53
46xx
Nickel 0.85 or 1.83, molybdenum 0.20 or 0.25
47xx
Nickel 1.05, chromium 0.45, molybdenum 0.20 or 0.35
«
"
.
„
.
Chromium 0.80, 0.88, 0.93, 0,95, or 1.00 Carbon 1.04, chromium 1.03 or 1.45
61xx
Chromium 0.60 or 0.95, vanadium 0.13 or 0.15 min.
jfication and will be discussed in
86xx
Nickel 0.55, chromium 0.50, molybdenum 0.20
87xx
Nickel 0.55, chromium 0.50, molybdenum 0.25
results of the cooperative effort
88xx
Nickel 0.55, chromium 0.50, molybdenum 0.35
(AISI) and the Society of Auto-
92xx
Silicon 2.00
i program aimed at greater effiican industry al designation indicates the type
93xx
Nickel 3.25, chromium 1.20, molybdenum 0.12 (series deleted in 1959)
98xx
Nickel 1.00, chromium 0.80, molybdenum 0.25 (series deleted in 1964)
94Bxx
Nickel 0.45, chromium 0.40, molybdenum 0.12, boron 0.0005 min.
icates
,
,
i
m
.
by means of a numbering
important elements in the steel.
.
5S a carbon steel 2 a nickel steel ,
;;; '
'
Nickel 3.50, chromium 1.50 (series deleted in 1964)
5xxxx
,
;
Nickel 1.25, chromium 0.60 (series deleted in 1964)
33xx
51xx
he steel such as machine steel
vv
Nickel 5.00 (series deleted in 1959)
Nickel 3.50, molybdenum 0.25
or tool steel
.
25xx 31xx
Chromium 0.40
,
.
Manganese 1.75
Nickel 3.50 (series deleted in 1959)
50xx
I etc.
i
:
13xx
23xx
48xx
,
,
Basic open-hearth carbon steels, high sulfur, high phosphorus
ase of simple alloy steels
,
,
the
oercentage of the predominant its usually indicate the mean car
-
'bol 2520 indicates a nickel steel
) percent carbon
.
jifications may include a letter ess employed in producing the i
i 'J
:
1
if
"
Series deleted" does not mean that these steels are no longer made. It simply means that the tonnage produced is below a certain minimum to be included in the listing of standard grades. This listing is revised period-
$
.
ically by AISI. N Some representative standard-steel specifications are given for plain- : carbon and free-machining steels in Table 7-3 and for alloy steels in Table 7-4.
246
INTRODUCTION TO PHYSICAL METALLURGY
THE IRON-IF
TABLE 7 4 Some Representative Alloy-steel Specificati TABLE 7-3
Some Represenlative Standard steel Specifications -
AISI NO.a:
% C
% Mn
% S max
SAE NO
.
% Mn
11330 «340
0 28-0.33
1 60-1.90
0 38-0.43
1 60-1.90
| 2317 I;030
0
15-0.20 28-0.33
0.40-0.60 0.60-0.80
3.25-3.75 3.25-3.75
"
PLAIN-CARBON STEELS C1010
0 08-0.13
0 30-0.60
C1015
0 13-0.18
0 30-0.60
,
.
'
.
.
0.04
0 05
1010
0 04
0 05
1015
.
.
.
C1020
0 18-0.23
0 30-0.60
0 04
0 05
C1025
1020
0 22-0.28
0 30-0.60
0 04
0 05
1025
.
.
CI 030
0 28-0.34 .
;
C1035
0 32-0.38 .
C1040
,
CI 045
0 37-0.44 .
0 43-0 50 .
C1050
.
0 48-0.55 .
C 1055
0 50-0.60
,
.
.
.
.
.
.
0 60-0.90 .
.
0 04
0 05
1030
0 04
0 05
1035
.
0 60-0.90 .
.
0 60-0.90
0 04
.
.
0 60-0.90
0 04
.
.
0 60-0.90 .
.
.
0 05 0 05
1045
.
0 05
1050
0 04
0 05
1055
.
.
.
.
.
0
.
.
0 55-0.65 .
CI 065
0 60-0.70 .
0 60-0.90 .
.
0 65-0.75 0 70-0 80
0 50-0.80
C1080
0 75-0.88
0 60-0.90
.
.
.
.
0 05
1060
0 04
0 05
1065
0 04
0 05
1070
.
0 60-0.90
C1070 CI 074
0 04 .
0 60-0.90 .
.
0 04
.
.
0 04
.
.
.
.
.
0 05 0 05
0 45-0.60
4 75-5.25
0 40-0.60
4 75-5.25
3115
0
13-0.18 28-0.33 0 38-0.43 0 08-0.13
0,55-0,
0
0.40-0.60 0.60-0.80 0.70-0.90 0.45-0.60
1 10-1,40
S130
1 10-1,40
0.55-0.
1 10-1.40
0.55-0.
3 65-3,75
1,40-1.
0 20-0.25
0 70-0,90
0 35-0.40
0 70-0.90
1 4419
0 18-0.23
0 45-0.65
4118
0 18-0.23
0 70-0.90
0 40-0
»t30
0 28-0,33
0 40-0.60
0 80-1
4140
0 38-0,43
0 75-1.00
0 80-1
4150
0 48-0,53
0 75-1.00
0 80-1
43W
0 17-0.22
0.45-0.60 0.60-0.80 0.50-0.70
1 65-2.00
0.40-C
38-0.43 0 17-0.22
65-2.00 0 90-1.20
0.70-C 0.35-C
4820
17-0.22 24-0.29 0 18-0.23
0.45-0.60 0.45-0.65 0.50-0.70
1.65-2.00 0.70-1.00 3.25-3.75
Mi 5120
0 17-0.22
$130
38-0.43 0 48-0.53 0 95-1.10
0.70-0.90 0.70-0.90 0.70-0.90 0.70-0.90 0.25-0.45
0 70-1
0 28-0.33
6118
0 16-0.21
0 50-0.70
0 50-
6150
0 48-0.53
0 70-0.90
0 80-
8820 8630 8640
0
18-0.23 0 28-0.33 0 38-0.43
0.70-0.90 0.70-0.90
0.40-0.70 0.40-0.70
0.400.40-
0.75-1.00
0.40-0.70
0.40-
8720
0 18-0.23
0 70-0.90
0 40-0.70
0 40-
8740
0 38-0.43
0 75-1.00
0 40-0.70
0 40-
88?2
0 20-0.25
0 75-1.00
0 40-0.70
0 40-
.
|
K 314C m;.E33io 4023
1080
.
C1085
0 80-0.93
0 70-1.00
0 04
0 05
1085
C10901
0 85-0.98
0 60-0.90
0 04
0 05
1090
0 04
0 05
1095
.
.
C1095
0 90-1.03 .
.
.
.
.
0 30-0.50 .
.
.
.
.
v..
E«40
.
.
.
.
.
.
0 13 max
0 70-1.00
0 07-0.12
0 16-0.23
1112
B1113
0 13 max
0 70-1.00
0 07-0.12
0 24-0.33
1113
C1110
.
.
-
C1113
0 08-0.13
0 10-0.16
1 00-1.30
0 04
0 24-0.33
0 18-0.23 .
0 32-0.39 .
0 37-0,45 .
0 13 max .
.
.
0 60-0.90 .
0 70-1.00
.
.
0 04 .
.
0 08-0.13 .
1115
0 08-0.13
1120
1 35-1.65
0 04
0 08-0.13
1137
1 35-1.65
0 04
0 08-0.13
1141
.
.
0 70-1.00 .
.
.
.
0 07-0.12 .
,
.
.
0 16-0 23
1112
.
.
0 70-1.00
0 07-0.12
0 24-0.33
1113
0 15 max
0 80-1.20
0 04-0.09
0 25-0.35
12L14
,
.
.
.
.
.
.
. Prefix AISI letters: B = acid bessemer carbon steel; C = basic open-hearth carbon steel
,
,
,
.
0
.
4620
0
.
0
.
'
il' 5140 I 5,50 '
0 04
.
.
KW26
.
0 13 max ,
C12L14t
.
0 04
.
C1137 C1141 CI 212 C1213
.
0 30-0.60
0 13-0.18
C1120
.
.
0.08-0.13 .
C1115
.
.
.
.
FREE-MACHINING CARBON STEELS B1112
.
0 12-0.17
1074
.
.
0 09-0.14
.
C1060
% Cr
»15
1040
.
0 04 .
0 60-0.90
.
% Ni
% C
«Sl NO
% P max
: £52100
.
.
.
0
.
.
.
.
.
.
.
.
,
,
.
,
,
.
.
.
,
.
.
.
.
.
.
.
1
.
.
.
0 80.
0 70.
0 70.
1 30.
.
.
.
.
.
.
t Lead, 0.15 to 0.35 percent.
Steels are sometimes classified by the broad range of carbon content such as:
.
.
.
.
.
.
.
.
.
.
.
.
Si
.260
0
.
56-0.64
0.75-1.00
1 80-2.20 .
Ni
>
j Low-carbon steel: up to 0.25 percent carbon
j/ Medium-carbon steel: 0 25 to .
0.55 percent carbon
I High-carbon steel: above 0.55 percent carbon
i
I
E9310
0 08-0.13
0 45-0.65
3 00-3.50
1 00
9«40
0 38-0.43
0 70-0.90
0 85-1.15
0 70
9$50
0 48-0.53
0 70-0.90
0 85-1.15
0 70
94B30
0
48-0.53
0.70-0.90
0.85-1.15
0.70
.
.
.
.
.
.
.
.
.
.
.
.
.
. E = basic electric-furnace process. All others are normally
m
.
:.-..-L "_ .
:. -
'
'
:
T'yy/: r:
'
~ -
C
.
'
'
' .
. Z
'
-
-
.
,
r
.
i
. „,
-a-jtw,-*,,*
.
MWIm
W
:
THE IRON-IRON CARBIDE EQUILIBRIUM DIAGRAM
247
f
SMmlf TABLE 7 4 Some Representative Alloy-steel Specifications
el Specifications
imk- AISI NO. %C
% P max
% S max
SAE NO
.
V M STEELS .
< 1340 J
.
.
.
2317
0 15-0.20
0 40-0.60
3 25-3.75
2330
0 28-0.33
0 60-0.80
3 25-3.75
E2512*
0 09-0.14
0 45-0.60
4 75-5.25
2515
0 12-0.17
0 40-0.60
4 75-5.25
0 04
0 05
1020
0 05
1025 1030
.
.
.
0 04
0 05
.
.
0.04
0 05
104
0 05
1035 1040
.
.
V. p.04
0 05
SiSp 04
1045
0 05
) 04
1050
0 05
1055
.
.
.
.
.
04
0 05
04
0 05
04
1065
0 05
1070
0 05 '
1074
i
.
1060
.
.
.
.
1 60-1.90
.
1010
.
-
0 38-0.43
1015
0 04
.
04
.
04
0 05
1080
.
% V : SAE NO.
1340
0 05 .
% Mo
1330 I
0 05 .
:
%Cr
1 60-1.90
0 04 .
-
%Ni
0 28-0.33
0 04 .
,
% Mn
1330
.
.
.
.
.
.
.
.
2315 |
.
.
3115
0 13-0.18
0 40-0.60
1 10-1.40
3130
0 28-0.33
0 60-0.80
1 10-1
3140
0 38-0.43
E3310
515 }
5% Ni
: 2515
0 70-0.90
0 55-0.75 0.55-0.75 0.55-0.75
0 08-0.13
0 45-0.60
3 65-3.75
1.40-1.75
4023
0 20-0.25
0 70-0.90
0 20-0.30
4023
4037
0 35-0.40
0 70-0.90
0 20-0.30
4037
4419
0 18-0.23
0 45-0.65
0 45-0.60
4419
4118
0 18-0.23
0.70-0.90
0 40-0.60
0.08-0.16
4118
4130
0 28-0.33
0.40-0.60
0 80-1.10
0.15-0.25
4130
Cr-Mo
4140
0 38-0.43
0 75-1.00
0 80-1.10
0.15-0.25
4140
steels
4150
0 48-0.53
0 75-1.00
0 80-1.10
0.15-0.25
4150
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3115
.
.
'
.
.
.
.
.
.
.
.
.
.
.
I
3130
3140
3310
I
0 05
1085
4320
0 17-0.22
0 45-0.60
1 65-2.00
0.40-0.60
0.20-0.30
4320
0 05
1090
4340
0 38-0.43
0 60-0.80
1 65-2.00
0.70-0.90
0.20-0.30
4340
1095
4720
0 17-0.22
0 50-0.70
0 90-1.20
0.35-0.55
0.15-0.25
4720
4620
0 17-0.22
0 45-0.60
1 65-2.00
0 20-0.30
4620
4626
0 24-0.29
0 45-0.65
0 70-1.00
0 15-0.25
4626
4820
0 18-0.23
0 50-0.70
3 25-3.75
0 20-0.30
4820
5120
0 17-0.22
0 70-0.90
0 70-0.90
5120
5130
0 28-0.33
0 70-0.90
0 80-1.10
5130
5140
0 38-0.43
0 70-0.90
0 70-0.90
5140
5150
0 48-0.53
0 70-0.90
0 70-0.90
5150
E52100*
0 95-1.10
0 25-0.45
1 30-1.60
52100
)4
0 05 .
.
.
.
.
.
.
.
.
.
30N STEELS 17-0 12 7-0 12 .
.
0 16-0 23
1112
0
1113
.
.
24-0.33 08-0.13 0 24-0.33 0 08-0.13 .
4
0
4
4
.
steels
40 1 10-1.40
.
34
.
3% Ni steels
:o 4
.
Mn steels
2330 J
.
.
TYPE
.
.
.
.
.
.
.
.
.
.
.
.
Ni-Cr steels
Mo
Steels
Ni-Cr-Mo steels
Ni-Mo steels
-
.
.
.
.
4
.
i
1141
6118
0 16-0.21
0.50-0.70
0 50-0.70
0 12
6118
6150
0 48-0.53
0.70-0.90
0 80-0.10
0 15
6150
0 24-0 33 0 25-0 35
1113
.
.
.
.
0 12
-
.
.
.
0 09
-
.
.
.
1112
.
.
.
.
.
0 16-0 23
.
-0 12
.
.
.
0 08-0 13 0 08-0.13 0 08-0 13 .
7
1115
.
.
.
1120
.
.
.
Cr steels
1137 .
.
.
,
.
Cr-V
steels
'
"
12L14
ipen-hearth carbon steel
.
8620
0 18-0.23
0 70-0.90
0 40-0.70
0 40-0.60
0.15-0.25
8620
8630
0 28-0.33
0 70-0.90
0 40-0.70
0 40-0.60
0.15-0.25
8630
8640
0 38-0.43
0 75-1.00
0 40-0.70
0 40-0.60
0.15-0.25
8640
.
.
.
.
.
.
.
.
.
.
.
.
Low
Ni-Cr-Mo
?
,:
road range of carbon content
8720
0 18-0.23
0.70-0.90
0.40-0.70
0.40-0.60
0.20-0.30
8720
8740
0 38-0.43
0 75-1.00
0 40-0.70
0 40-0.60
0 20-0.30
8740
8822
0 20-0.25
0 75-1.00
0 40-0.70
0 40-0.60
0 20-0.40
8822
.
.
.
.
.
.
.
.
.
.
.
steels
Si 9260
0 56-0.64 .
0.75-1.00
1 80-2.20
Si steel
9260
.
Ni
!
.r
E9310*
0 08-0.13
0 45-0.65
3 00-3.50
1 00-1.40
0 08-0.15
.
.
.
.
9310
.
9840
0 38-0.43
0 70-0.90
0 85-1.15
0 70-0.90
0 20-0.30
9840
9850
0 48-0.53
0 70-0.90
0 85-1.15
0 70-0.90
0.20-0.30
9850
94B30
0.48-0.53
0.70-0.90
0.85-1.15
0.70-0.90
0.20-0.30
.
.
.
.
.
.
.
.
.
...
I
94B30
Higher Ni-Cr-Mo steels
Boron
,
steel "
I: , ,
. E = basic electric-furnace process. All others are normally manufactured by the basic open-hearth process.
,
248
INTRODUCTION TO PHYSICAL METALLURGY
7-10
Effect of Small Quantities of Other Elements The previous discussion the iron-iron carbide equilibrium diagram assumed that only iron andii carbide were present. However, examination of Table 7-3 indicates commercial plain carbon steels contain small quantities of other elenn
besides iron and carbon as part of the normal composition. Sulfur Sulfur in commercial steels is generally kept below 0.05 parcel
THE H TREA1 OF ST
Sulfur combines with iron to form iron sulfide (FeS). Iron sulfide forms!! low-melting-point eutectic alloy with iron which tends to concentrate
the grain boundaries. When the steel is forged or rolled at elevated tet peratures, the steel becomes brittle, or hot-short, due to the melting the iron sulfide eutectic which destroys the cohesion between grail allowing cracks to develop. In the presence of manganese, sulfur tends form manganese sulfide (MnS) rather than iron sulfide. The manganese! sulfide may pass out in the slag or remain as well-distributed inclusion
.
11
throughout the structure. It is recommended that the amount of mangal nese be 2 to 8 times the amount of sulfur. In the free-machining steels, thai v
to a metal or alloy.in the solid state ir
sulfur content is increased to between 0.08 and 0.35 percent. The improve-
erties.
ment in machinability is due to the presence of more numerous sulfides inclusions which break up the chips, thus reducing tool wear.
"
All basic heat-treating proci
tion or decomposition of austenite. transformation products determine ties of any given steel. The first step in the heat treatm*
}
Manganese Manganese is present in all commercial plain carbon steels, in the range of 0.03 to 1.00 percent. The function of manganese in counteracting the ill effects of sulfur was just pointed out. When there is more
some temperature in or above the ci
manganese present than the amount required to form MnS, the excess;]
In most cases, the rate of heating to tant than other factors in the heat-tr produced by cold work should be
combines with some of the carbon to form the compound MnjC which is!
found associated with the iron carbide, FejC, in cementite. Manganese1]
materials to avoid distortion.
also promotes the soundness of steel casting through its deoxidizing
The
action on liquid steel. The effect of larger quantities of manganese will bej
thick and thin sections
discussed in Chap. 9 on alloy steels.
sidered, and whenever possible, pn
Phosphorus
heating of the thinner sections to i Usually less overall damage will be
The phosphorus content is generally kept below 0.04 per-.J
cent. This small quantity tends to dissolve in ferrite, increasing the strength: and hardness slightly.
In some steels 0.07 to 0.12 percent phosphorus ?
seems to improve cutting properties. In larger quantities, phosphoriiff i' reduces ductility, thereby increasing the tendency of the steel to crack when cold worked, making it cold-short. Silicon Most commercial steels contain between 0.05 and 0.3 percent silicon. Silicon dissolves in ferrite, increasing the strength of the steel
,
without greatly decreasing the ductility. It promotes the deoxidation oi molten steel through the formation of silicon dioxide, SiOj, thus tendingi
to make for greater soundness in casting. Silicon is an important element! i
Introduction The definition of heat tre is; "A combination of heating and c
in cast iron, and its effect will be discussed in Chap. 11.
t ? .
of articles o
heating rate as is practical.
Full Annealing This process consist; perature and then cooling erably in the furnace
slowly t
or in any gc
cooling is generally continued to k
The purpose of annealing may I improve electrical and magnetic pr machinability.
Since the entire mass of the
fui
the material, annealing is a very sk closest to following the iron-iron Assume that we have a coarse-c
1
! 5
.
ments
The previous discussion of
l fram assumed that only iron and iron "
mination of Table 7-3 indicates that in small quantities of other elements normal composition
;
THE HEAT TREATMENT OF STEEL
.
generally kept below 0.05 percent
.
.
i sulfide (FeS) Iron sulfide forms a .
iron which tends to concentrate at is forged or rolled at elevated tem -
: :; .
.
.
or hot-short, due to the melting of
Spoys the cohesion between grains
,
sence of manganese sulfur tends to ,
than iron sulfide
The manganese rnain as well-distributed inclusions .
J
.
nended that the amount of mangav-jir. In the free-machining steels the
B ).08 and 0.35 percent. The Improve'
: ..
"
erties.
esence of more numerous sulfide commercial plain carbon steels,
The first step in the heat treatment of steel is to heat the material to C
pointed out. When there is more equired to form MnS the excess orm the compound MnjC which is !, FesC, in cementite Manganese -
.
.
jl casting through its deoxidizing Ur quantities of manganese will be
The difference in temperature rise within thick and thin sections of articles of variable cross section should be considered, and whenever possible, provision should be made for slowing the materials to avoid distortion.
per-
heating of the thinner sections to minimize thermal stress and distortion.
\e in ferrite, increasing the strength 0
.
Usually less overall damage will be done to the steel by utilizing as slow a .
07 to 0.12 percent phosphorus
heating rate as is practical.
In larger quantities phosphorus ie tendency of the steel to crack ,
8-2
lin between 0 05 and 0 3 percent .
m
It promotes the deoxidation of
.
. .
.
.
silicon dioxide
a|g
.
Si02, thus tending Silicon is an important element ,
sed in Chap. 11
.
Full Annealing
This process consists in heating the steel to the proper tem-
perature and then cooling slowly through the transformation range, preferably in the furnace or in any good heat-insulating material. The slow cooling is generally continued to low temperatures.
.
greasing the strength of the steel ,
some temperature in or above the critical range in order to form austenite.
In most cases, the rate of heating to the desired temperature is less important than other factors in the heat-treating cycle. Highly stressed materials v: produced by cold work should be heated more slowly than stress-free
,
.
proper- .
ties of any given steel.
function of manganese in counter-
-1 is generally kept below 0 04
All basic heat-treating processes for steel involve the transforma-
transformation products determine the physical and mechanical
.
W-ia\\
3
m
The purpose of annealing may be to refine.the grain, induce softness, improve electrical and magnetic properties, and, in some cases, to improve machinability.
Since the entire mass of the furnace must be cooled down alpng with the material, annealing is a very slow cooling process and therefore comes closest to following the iron-iron carbide equilibrium diagram. Assume that we have a coarse-grained 0.20 percent carbon (hypoeutec-
i
-l
produce desired prop-
tion or decomposition of austenite. The nature and appearance of these
us reducing tool vyear
:
The definition of heat treatment given in the Metals Handbook is: "A combination of heating and cooling operations, timed and applied
to a metal or alloy.in the solid state in a way that will
,
:: :
'
8-1 rntroductiori
.
J
250 INTRODUCTION TO PHYSICAL METALLURGY
toid) steel and that it is desired to refine the grain size by anneali ng, m microstructure is shown in Fig 8-1 a. When this steel is heated will occur until the /A (lower-critical) line is crossed At that temperatmf the pearlite areas will transform to small grains of austenite by meansflf the eutectoid reaction but the original large ferrite grains will remain unto changed (Fig 8-1 b). Cooling from this temperature will not refine the .
,
,
100
no chang|
.
,
Continued
graiiwl heating between the A and lines will allow the large territeJH grains to transform to small grains of austenite so that above the A
Proeutectoid lerrite
1 C
SO
.
Pearlite
,
critical) line the entire microstructur 3 (upper* e will show only small grains of austei»| ite, Fig. 8-1c Subsequent furnace cooling will result in small grains di; proeutectoid ferrite and small areas of coarse lamellar pearlite (Fig Therefore S-ld);
n
,
.
c
08
0
.
Hypoeutectoid steels
.
the proper annealing temperature for hypoeutectoid steels if :| Refinement of the grain si 3 line.
-
,
t!
Percent car
approximately 50oF above the A
gjfto 8-2 Proportion of the constituentsfunction present ofin carboi the
Imicrostructure of annealed steels as a
ze of hypereutectoid steel will occur about'
50"F above the lower-critical-temper
temperature will coarsen the auste transform to large pearlitic areas. T
1666 Ausfemte
I
eutectoid steel will consist of coars V:'
Austenite + Ferrite
annealing should never be a final he;
The presence of a thick, hard, grain
Id /J,
m Austenite
Cooling to room temp -
errite +
by a network of proeutectoid cemer cess cementite network is brittle at
3
6]
Ferrite
.
Ferrite
Deorlite
chinability.
A careful microscopic study of tl or pearlite and cementite present ir determine the approximate carbon ( The approximate tensile strengl may be determined by the proportic
Dearlife
Approx tensile strength [a] C
40,00C
3earlite
02
For example, an annealed 0.20 p mately 25 percent pearlite and 7!
08
.
.
Percent carbon by weight
formula,
Fig, 8-1 Schematic representation ol the changes in microstructure during the annealing of a 0.20 percent
Approx tensile strength = 40,000(0.75)
carbon steel (a) Original structure coarse-grained ferrite and pearlite (£)) Just above the A, line; pearlite has ,
,
.
transformed to small grains of austenite, ferrite unchanged
(c) Above the A, line; only fine-grained austenite
.
(d) After
cooling to room temperature; fine grained ferrite and small -
pearlite areas..
.
This same idea cannot be appl strength is determined by the cerr ous phase. The presence of the t
I
¥1 mm
THE HEAT TREATMENT OF STEEL 25'l> grain size by annealing
.
.
The
n this steel is heated no change
m
ure
ii
grains of austenite by means of e ferrite grains will remain un
t
. .
m
100
Proeutectoid cementite
,
ViVls crossed. At that temperat
Proeutectoid ferrite cz
-
o
)erature will hot refine the grain
a
50h
.
Pearlite
lines will allow the large ferrite ite, so that above the A, (upper-
CD O
m
jow only smajl grains of austen g will result in small grains of -
/
.
se lamellar pearlite (Fig.
.
-
>
'"'
.
jre for hypoeutectoid
0
8-1d)
.
0
08 .
Hypoeutectoid steels
.
-
steels is
i
.P
Hypereutectoid steels-
.
m
Percent carbon
Fig. 8-2
tectoid steel will occur about
*4-<
fell
Proportion of the constituents present in the
microstructure of annealed steels as a function of carbon content.
1
3;
/
]
-
50oF above the lower-criticai-temperature (Aj ) line. ,
,
m
-
it
I
1
i
ferrite
jPearlife
annealing should never be a final heat treatment for hypereutectoid steels.
_
The presence of a thick, hard, grain boundary will also result in poor machinability. A careful microscopic study of the proportions of ferrite and pearlite or pearlite and cementite present in an annealed steel can enable one to determine the approximate carbon content of the steel (Fig. 8-2). The approximate tensile strength of annealed hypoeutectoid steels may be determined by the proportion of ferrite and pearlite present: Approx tensile strength
40,000(percent ferrite) + 120,000(percent pearlite) 100
i
I i i
For example, an annealed 0.20 percent carbon steel contains approximately 25 percent pearlite and 75 percent ferrite. Applying the above formula,
1 Approx tensile strength = 40,000(0.75) + 120,000(0.25) = 60,000 psi
i
This same idea cannot be applied to hypereutectoid steels since their strength is determined by the cementite network which forms the continuous phase. The presence of the brittle network results in a drop in tensile ,
.1 m .
-
'
I
Heating above this
temperature will coarsen the austenitic grains, which, on cooling, will transform to large pearlitic areas. The microstructure of annealed hypereutectoid steel will consist of coarse lamellar pearlite areas.surrounded by a network of proeutecteid cementite (see Fig. 7-14). Because this excess cementite network is brittle and tends to be a plane of weakness,
I
Jl
-
Ir
262
INTRODUCTION TO PHYSICAL METALLURGY
strength above 0.8 percent carbon (see Table 8-1). The proper annealir* range for hypoeutectoid and hypereutectoid steels Is shown in Fig. 8-3. ,
Normalizing range
8-3 Spheroidizing As was pointed out earlier, an annealed hypereutectoid steeij with a microstructure of pearlite and a cementite network will generally
give poor machinability. Since cementite is hard and brittle, the cuttinjl tool cannot cut through these plates. Instead
,
1600
the plates have to be brokeiij |
Ac 3
Therefore, the tool is subjected to continual shock load by the cementite: % plates, and a ragged surface finish results. A heat-treating process which; I will improve the machinability is known as spheroidize annealing
1500
ThSl I
process will produce a spheroidal or globular form of carbide in a ferrilit; p matrix, as shown in Fig. 84. One of the following methods may be used: 1
Prolonged holding at a temperature just below the lower critical line.
2
Heating and cooling alternately between temperatures that are just above and
5 1400
I:
just below the lower critical line.
3
Heating to a temperature above the lower critical line and then either cooling vetj
'
Prolonged time at the elevated temperature
Ac
1300
|
.
slowly in the furnace or holding at a temperature just below the lower critical line
Full annealing and hardening range
5
.
will completely break up theft
1200
0 60
0 40
0 20
.
.
.
Perc
Table 8-1
Mechanical Properties of Normalized and Annealed Steels*
f.g 8 3 Annealing, normalizing, and hardening rangefor plain-carbon sleeis,
.
CARBON, %
YIELD POINT
,
1 000 PS I ,
TENSILE STRENGTH, 1 000 RSI
, ELONGATION,
REDUCTION
% IN 2 IN.
IN AREA, %
„_„.,„,.,„.,
I
,
Normalized
,
k
(hot-rolled steel):
,
3
5m
i
a
.
a?
0 01
26
45
45
71
0 20
45
64
35
60
0 40
51
85
27
43
0 60
60
109
19
28
0 80
70
134
13
18
260 i r
1 00
100
152
7
11
295-
1 20
100
153
3
6
315
1 40
96
148
3
300
0 01
18
41
47
71
90;
0 20
36
59
37
64
115 j
0 40
44
75
30
48
145-'
.
.
.
.
.
.
.
.
90
:
.
.
T
.
.
165 t
2201
0 60
49
96
23
33
0 80
52
115
15
22
220.
1 00
52
108
22
26
1 20
51
102
24
39
195 200
1 40
50
99
19
25
215:
.
.
.
.
.
1 .
.
9
mm
L
A
-
By permission from Brick, Gordon, and Phillips, "The Structure and Properties of Alloys," 3d ed. McGraw-Hill ,
Company, New York, 1965.
Bo*
8-4
I' showing
i v
4Z-
i Fig *
.
120;; py
Annealed:
.
-
i3
iwm
.
A 1 percent carbon steel spheroidize-annea
spheroidized cementite in a ferrite matrix. E
in 2 percent nital, 750X.
THE HEAT TREATMENT OF STEEL
253
-
3e Table 8-1). The proper annealing ectoid steels is shown in Fig. 8-3.
1700
vi>?r, an annealed hypereutectoid steel a cementite network will generally itite is hard and brittle the cutting istead, the plates have to be broken, itinual shock load by the cementite jits. A heat-treating process which
4
1500
Full annealing and
lobular form of carbide in.aierritic
hardening range
o
ie following methods may be used:
.
,
4c 3
in as spheroidize annealn i g. This -
v
1600
,
v
Normalizing range
1400
4ielow the lower critical line
.
temperatures that are just above and
1
critical line and then either cooling very
Ac 3
4c
1300
:
1
,
.
,
ature will completely break up the
1200
0 20
0 ilcd Steels*
Fig. 8-3
ELONGATION,
REDUCTION
/o IN 2 IN.
IN AREA, %
I
'
.
1
ature just below the lower critical line. ; C.
i
..
.
0 60
0 40
20
1 40 .
,
.
1
.
.
Annealing, normalizing, and
,1
1 00
0,80
,
Percent carbon
hardening range
i
for plain-carbon steels. BHN
:
l *
.
5
71
90
IS
60
120
!7
43
165
9
28
220
18
260
.
7
11
295
3
6
315
1
3
300
71
90
64
115
48
145
33
190
22
220
26
195
'
I M r
m
0
s 4 .
200
25
215
m
.
i
.
3
..
Properties of Alloys,-1 3d ed., McGraw-Hill Book
i
6
.
-
1
S5-
f<5
39
i
1
.
;
/
.J
a
.
SB
\
Fia 8-4 A 1 percent carbon steel spheroidize-annealed, showing spheroidized cementite in a ferrite matrix. Etched
j
in 2 percent nital, 750X.
! 1
n
i
254
INTRODUCTION TO PHYSICAL METALLURGY
pearlitic structure and cementite network. The cementite will beco spheres, which is the geometric shape in greatest equilibrium with surroundings. The cementite particles and the entire structure may
called sgheroidite (see Fig. 8-4). Contrast this microstructure with one shown in Fig. 7-14. Notice tttat in both cases the microstructure made up of ferrite and cementite. The difference is in the form cementite, and this greatly affects the properties of the materials. Tl spheroidized structure is desirable when minimum hardness, maxi™
ductility, or (in high-carbon steels) maximum machinability is important
Low-carbon steels are seldom spheroidized for machining, because inth| "
"
spheroidized condition they are excessively soft and gummy. The ci ting tool will tend to push the material rather than cut it, causing excess!' heat and wear on the cutting tip. Medium-carbon steels are some! spheroidize-annealed to obtain maximum ductility for certain _wQrkini operations. If the steel is kept too long at the spheroidize-annealing te _
v
roeute sequently there will be less p proeutectoid c toid steels and less
compared with annealed ones. Fig normalized 0.50 percent carbon stet would have approximately 62 percer ferrite. Due to air cooling, this samp toid ferrite, which is the white netwc For hypereutectoid steels, normali;
proeutectoid cementite network, an
entirely. Since it was the presence c the strength of annealed hypereute Th show an increase in strength.
given in Table 81, particularly for £ carbon.
Aside from influencing
the amou
perature, the cementite particles will coalesce and become elongate!
form, the faster cooling rate in norr
thus reducing machinability.
of austenite
Stress-relief Annealing
This process, sometimes called subcritical anneal
transformation and tithe faster the cooling rate, the lov
ing, is useful in removing residual stresses due to heavy machining orothi
formation and the finer the pearlit
cold-working processes. It is usually carried out at temperatures below th
mentite plates
lower critical line (1000 to 1200oF).
\ i,
Process Annealing This heat treatment is used in the sheet and wire iiwl
in the pearlite betwe
in Fig. 8-6. Ferrite k With the cementite plates closer to;
schematically
'
dustries and is carried out by heating the steel to a temperature below the| lower critical line (1000 to 1250°F). It is applied after cold working an(£
softens the steel, by recrystallization, for further working. It is very similar! to stress-relief annealing. 86 e
Normalizing The normalizing of steel is carried out by heating approximately! 100°F above the upper-criticai-temperature (A, or Acm) litie followed
cooling in still air to room temperature. The temperature range for normal-
izing is shown in Fig. 8-3. The pjjrjsogejjf normaNzMT£jsJ steel thanlulLannealing, sojhat for some applications! norm ing may b ajinaj heat treatment. Therefore, for hypereuteotoid|
hardeiiand stronfler
_
_
'
s
teels, it is necessary to heat above the Acm line in order to dissolve ther|
cementite network,
Normalizing may also be used to improve machin-j
ability, modify and refine cast dendritic structures, and refine the grainj
and homogenize the microstructure in order to improve the response ml hardening operations. The increase in cooling rate due to air cooling as
SI
compared with furnace cooling affects the transformation of austenitej
and the resultant microstructure in several ways. Since we are no longerf cooling under equilibrium conditions, the iron-iron carbide diagram can-
not be used to predict the proportions of proeutectoid ferrite and pearlitel
teel, h,
or proeutectoid cementite and pearlite that will exist at room temperature,
percent rbon sfernte Flg 8.5 Normalizedd;0,5*, and air-coole 100X, Proeutectoid
There is Iqiss time for the formation of the proeutectoid constituent; con-
surrounding pearlite
areas.
'
THE HEAT TREATMENT OF STEEL: ;255
work. The cementlte will become 1 pe in greatest equilibrium with its mm :
-
'
jpHs and the entire structure may be itrast this microstructure with the 0 both cases thejTT[ci sta ture is ie difference is in the form of the
I'M
1
_
;
of the materials The |en minimum hardness maximum
I
.
.
.
1
ly about 10 percent proeutecferrite. ferrite. Due to air cooling, this sample has on toid ferrite, which is the white network surrounding the dark peariite areas.
li|
V|,
,
show an increase in strength.
.
,
'
-
This is illustrated by the strength values gJ
carbon.
_
Aside from influencing the amount of proeutectoid constituent that_wjll - m form, the faster cooling rate in normalizing will also affect the temperaturel ;: || of austenite transformation and the fineness of the peariite. In genera
at the spheroidize-annealing tem-
'
loalesce and become elongated
,
,
the faster the cooling rate, the lower the temperature of austenite transformation and the finer the peariite. The difference in spacing of the cementite plates in the peariite between annealing and normalizing is shown
Retimes called subcrltical anneal
-
5 due to heavy machining or other
ied out at temperatures below the
schematically in Fig. 8 6 -
.
-
.
applied after cold working and
.
' -
V
.
.
line followed by
4 '
e temperature range for normal of normalizing Is to produce a
-
>
r-
if
4
Ag, sojhat for some applications
'
.
Therefore, for hypereutectoid
lcm
line in order to dissolve the o be used to improve machin
-
tructures
and refine the grain ier to improve the response in ,
- t- -
A
V
pling rate due to air cooling as Wye transformation of austenite -
.
.
'
f .
I ways. Since we are no longer
.
carblde diagram canA 'jroeutectoid ferrite and peariite '
;
on-iron
_
will exist at room temperature
Fig. 8-5 Normalized 0(6$) percent carbon stefilite, heated to ISOOT and air-cooled; 100X. Proeutectoid ferr
.
,
proeutectoid constituent;
I
Ferrite is very soft, while cementite is very hard.
urther working It is very similar
re (A, or Acm)
'
With the cementite plates closer together in the case of normalized medium
used in the sheet and wire in steel to a temperature below the
Wd out by heating approximately
m
given in Table 8-1, particularly for steels containing more than 0.8 percent
ium-carbon steels are sometimes
urn ductility for cedaui working
:
Wt
'
the strength of annealed hypereutectoid steels, normalized steels should
jvely soft and "gummy The cutvS\|ther than cut it causing excessive
:
"i|j
V
.
hich reduced entirely. Since it was the presence of the cementite network w
"
'
;|1
have approximately 62 percent peariite and 38 percent proeutectoid
in some cases it may be suppressed proeutectoid uoiiicniii cementite nc... networ..,, and proeuieuiuiu ...
.
,,
:. f
bon steel. normalized lized 0.50 percent car caroon sxeei. In m the me annealed aimcaicu condition this stee
k
,
:
,
compared with annealed ones. Figure 8-S shows the microstructure of al
For hypereutectoid steels, normalizing will reduce the continuity of the
,
ximum machinability is important. ized for machining because in the ,
.
-
_
j properties
sequently there will be less proeutectoid ferrite in normalized hypoeutec-\:;||| eeirajs,. j: told steels and less proeutectoid cementite in hypereutectoi
con-
surrounding peariite areas.
.
I 1
mi
'
1
1
m
256
INTRODUCTION TO PHYSICAL METALLURGY
r
called martensite, is a supersatura
r errify
body-centered tetragonal structui
Lementite
equal, but the third is slightly exi The axial ratio c/a increases witt
1
(see Fig. 8-7). This highly distorte( Fe-C-martensite 3 04
Annealed
Niormalized
Coarse lamellar pearlile
Medium lamellar pearlite
.
3 00 .
a)
Fig. 8-6 Schematic picture of the difference in pearlitic structure due to annealing and normalizing.
.-.v.-.
2 96 .
2
2 92
a
,
O
-
2 88 .
I
2 84
pearlite, they tend to stiffen the ferrite so it will not yield as easily, thus increasing hardness. If the annealed coarse pearlite has a hardness of about Rockwell C 10, then the normalized medium pearlite will be aboui Rockwell C 20. Nonequilibrium cooling also shifts the eutectoid POintW
.
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1. Weight percent carbon Fia 8-7 Variation of lattice constants a and c of m
SenS,ie with carbon content in plain-carbon steelsj St toward lower carbon .content in hypoeutectoid steels and toward hiqherlp permission from c. Barrett andT. b. Massaiski, Ne "
3d ed., McGraw-Hill Book Company,
'
"
of Metals.
produces a finer and more abundant pearlite structure than is obtained by J| annealing, which results in a harder and stronger steel.
i
the high hardness of martensite. densely packed than in austenitf
While annealing, spheroidizing, and normalizing may be employed to improve machinability, the process to be used will depend upon carbon
content. Based on many studies, the optimum microstructures for roa«lp chining steels of different carbon contents are usually as follows: CARBON, %
OPTIMUM MICROSTRUCTURE
0 06 to 0.20
As cold-rolled
0 20 to 0.30
Under 3-in. dia., normalized;
.
.
as a white needlelike or aciculai
0 40 to 0.60
Annealed, to give coarse pearlite
tained austenite, the acicular stn (Fig. 8-9).
There are several important formation:
or coarse spheroidlte .
8-7
Hardening
1 The transformation is diffuslonles tion. Small volumes of austenite si
100% spheroidlte, coarse to fine
tion of two shearing actions.
Under slow or moderate cooling rates, the carbon atoms are's
2 The transformation proceeds
able to diffuse out of the austenite structure. The ifon atoms then moyjj transformation takes place by a process Qf nucleation and growth andisL time-dependent. With a still further increase in cooling rate insufficient
in contrast to one that will occur at c The amount of martensite formed
.
,
number of martensite needles prod
time is allowed for the carbon to diffuse out of solution and although some* movement of the iron atoms takes place the structure cannot becomel
and finally, near the end it decreasi of martensite formation is known a tensile formation as the M, tempera
,
,
,
c c
.
.
.
while the carbon is trapped in solution.
The resultant structure ! ,
Si
ii
on
rupted. Therefore, the transformatii ture and is independent of time. A t
slightly to become b.c.c. (body-centered cubic). This gamma-to-alphaj
b
steels, the ma
ofjstraw. In most
resolvable (Fig. 8 8). In high-c;
Annealed, to give coarse pearlite
0 60 to 1.00
high localized stresses which re After drastic cooling (quenchir
Over 3-in. dia., as cold-rolled
.
formation. This expansion durii
-
0 30 to 0.40 .
1966 *
mm,
.
THE HEAT TREATMENT OF STEEL
1
'
1
t
The axial ratio c/a increases with carbon content to a maximum of 1.08
1
(see Fig. 8-7). This highly distorted lattice structure is the prime reason for
3 08
J i
.I
1
i
e-C-martensite
.
Cellar pearlite
I
called martensite, is a supersaturated solid solution of carbon trapped in a body-centered tetragonal structure. Two dimensions of the unit cell are equal, but the third is slightly expanded because of the trapped carbon.
o
itialized
257
"
c
3 04 .
<3
3 00
CD
.
n
2 96
5
2 92
.
1
.
2 88 .
it will not yield as easily, thus
.
irse pearlite has a hardness of
V medium pearlite will
0
also shifts the eutectoid
1
> :toid steels and toward higher
1
le net effect is that normalizing e structure than is onger steel
1.4 1.6 1.8 2.0
obtained by
I
Fig. 8-7 Variation of lattice constants a and c of martensite with carbon content in plain-carbon steels. (By permission from C. Barrett and T. B. Massalski, Structure of Metals," 3d ed., McGraw-Hill Book Company, New York, 1966.) "
.
the high hardness of martensite. Since the atoms of martensite are less :;Jp
malizing may be employed to ised will depend upon carbon mum microstructures for ma .
.
Weight percent carbon
point
.
0 2 0.4 0.6 0.8 1.0 1.2
be about
.
;
3
Lj
-
2 84
densely packed than In austenlte, an expansion occurs during the transformation. This expansion during the formation of martensite produces
-
ire usually as follows: .
Hi
high localized stresses which result in plastic deformation of the matri J p
After drastic cooling (quenching), martensite appears microscopically
as a white needlelike or acicular structure sometimes described as a pile rj'g'; of straw. In most steels, the martertsitic structure appears vague and.un- aK '
1
;
resolvable (Fig. 8-8). In high-carbon alloys where the background is re-
-{W
tained austenlte, the acicular structure of martensite is more clearly defined . 4j:
(Fig. 8-9).
{
i|p t
There are several important characteristics of the martensite trans-
e
formation:
j ates, the carbon atoms .. . i The iyon atoms then
are
move
liable). This gamma-to-alpha iucleation and
growth and is
in cooling rate insufficient ,
solution
,
and although some
e structure cannot become The resultant structure,
n
.
I|i;
1 The Tansformation is diffusioniess, and there Is no change in chemical compositlon. Small volumes of austenite suddenly change crystal structure by a comblna-
-
1
"
,
j|i
tion of two shearing actions. 2 The transformation proceeds only during cooling and ceases if cooling is inter- ,
||!
;
'
'HI
rupted. Therefore the transformation depends only upon the decrease In tempera,
'
ture and is independent of time. A transformation of this type is said to be athermal
in contrast to one that will occur at constant-temperature (/sofherma/transformation)
'
,
I
_i
„
l
:x_
ii
i
i
_x
i
.
it.
ii
.i.
_
i
.
, "
"
:
.
The amount of martensite formed with decreasing temperature Is not linear The number of martensite needles produced at first is small; then the number increases
'
'
.
.
,
Ijlj
L
" i!;'f
,
'
and finally, near the end, it decreases again (Fig. 8-10). The temperature o f the start of martensite formation is known as the Ms temperature and that of the end of mar-
tensite formation as the M, temperature. If the steel is held at any temperature below
i
258 INTRODUCTION TO PHYSICAL METALLURGY
« 3
,
'
7
0
100 Percent martensite
Fig. 8 10
Schematic representation of percentage of
martensite formed as a function of temperature. {From Metals Handbook," 1948 ed., American Society for
"
"
! »
Metals, Metals Park, Ohio.)
of the formation of martensite is charactt
by increasing the cooling rate. The Ms U ical composition only, and several formu calculated. One such formula (R. A. Gn June, 1946) is
( F) = 1,000 - (650 x % C) - (70 x % "
M5
s: i
rzx10 ma,",nsi,e .""s,op a"d
The influence of carbon on the Ms and A
temperature line is shown dotted bece retically, the austenite to martensite ti rate
.
The temperature range
amounts of retained austenite will rer formation of the last traces of austeni amount of austenite decreases.
This i
100 percent martensite in Fig. 8-10. Cu on visual estimations of the structures c
of retained austenite are very difficult 1 martensite needles. Therefore, the M, ture at which the transformation is co
OOOrr
'
i
.
300 h
\ \
S 600
\ \
181
r
1
i
\
v
\
400
\
\
m
\
200
\ \
0
0
02 .
04 .
0.6
OS
1.0
12 .
Percent carbon
Fig. 8-11 austenite matrix
m
Influence of carbon on the martensite n
(From Metals Handbook," 1948 ed., American So( lor Metals, Metals Park, Ohio.) "
.
.
.
a
"
THE HEAT TREATMENT OF STEEL 259/ V ill! | :
'
I
m m
a
4%. PI:
11
.
1
-
ii
,
o
100 Percent martensite
Fig. 8-10
l
Schematic representation of percentage of
martensite formed as a function of temperature. (From
e|
"
Metals Handbook," 1948 ed., American Society for Metals, Metals Park Ohio.) ,
1
\ .
of the formation of martensite is characteristic of a given alloy and cannot be lowered by increasing the cooling rate. The Ws temperature seems tc be a function of chemical composition only and several formulas have been developed by which it may be calculated. One such formula (R. A. Grange and H. M. Stewart, Metals Technology, June, 1946) is
* I
f
,
Ms
(0F) = 1,000 - (650 X % C) - (70 X % Mn) (35 x % Ni) - (70 x % Cr) - (50 x % Mo)
V
) .
.
> -
-
vill not proceed again unless
The influence of carbon on the Ms
mot be suppressed nor can ,
and M, temperatures is shown in Fig. 8-11. The M, temperature line is shown dotted because it is usually not clearly defined. / Theoretically, the austenite to martensite transformation is never complete and small amounts of retained austenite will remain even at low temperatures. The trans-
V
formation of the last traces of austenite becomes more and more difficult as the
t
,
jrate. The temperature range mi
amount of austenite decreases. This Is apparent from the slope of the curve near 100 percent martensite in Fig. 8-10. Curves such as shown in these figures are based on visual estimations of the structures of metallographic samples and small amounts ,
of retained austenite are very difficult to measure when there are manyoverlapping martensite needles. Therefore, the M, temperature is usually taken as the temperature at which the transformation is complete as far as one is able to determine by
By
1000
800
fife t 600 i
a CD
E 400 1
A
200
0
02 .
04 .
0.6 0.8 1.0 Percent carbon
12 ,
14 .
Fig. 8-11 Influence of carbon on the martensite range. (From "Metals Handbook," 1948 ed., American Society for Metals, Metals Park, Ohio.)
1
.
V
1
:
-
1!
260
INTRODUCTION TO PHYSICAL METALLURGY
visual means.
4
j oo
Martensite is probably never in a condition of real equilibrium although it m,
65
,
persist indefinitely at or near room temperature. The structure can be considered
300
a transition between the unstable austenlte phase and the final equilibrium conditio 1 The most significant property of martensite is its potential of very great hardnesi'
of a mixture of ferrite and cementite
60
700
.
5
Although martensite is always harder than the austenite from which it forms 8-12) extreme hardnesses are possible only in steels that contain sufficient carbon The hardness of martensite increases rapidly at first with increase in carbon content reaching about 60 Rockwell C at 0.40 percent carbon. Beyond that point the curvi levels off,-and at the eutectoid composition (0.80 percent carbon) the hardness!
V Martensite
GOO
in
(Boin and Paxton)
.
50 "e
500 (X
45 " 40
S 400
,
is about Rockwell C 65. The leveling off is due to the greater tendency to retail
4 35
-
,
I I
55
ron-carbon steels
'
,
austenite in higher-carbon steels. The high hardness of martensite Is believed to be; a result of the severe lattice distortions produced by its formation since the amountl
/
,
of carbon present is many times more than can be held in solid solution The maximum hardness obtainable from a steel in the martensitic condition seems to be a function of carbon content only.
50 25 20
-Austenite in
iron-nickel-carbon
.
200 h
alloys (Krauss)
100
v '. The martensite transformation
,
for steel.
However
,
for many years, was believed to be uniquej;
0
in recent years, this martensite type of transformation!
0.4
0.6
0,8
1.0
Weight percent carbon the hardness of
has been found in a number of other alloy systems such as iron-nickel ,
0.2
Fig. 8-12 The effect of carbon on
,
i austenite and on the hardness of martensite. The s represents the effect of ret: area of the upper curve austenite. (By permission from Brick, Gordon, and of Alloys," 3d ed., McGrE Structure and Properties
copper-zirVc, and copper-aluminum. The transformation is therefore recog4 nized as a basic type of reaction in the solid state, and the term martensiMi is no longer confined only to the metallurgy of steel.
'
Book Company, New York, 1965.)
Thejiasie-ptii puye uf lwdening isJ:o pr_D.d]jce a fuMyjriartensltic strua ._
_
ture.jand the minimum coolinfl-iate (°F per second) that will avojcj- the
i 1
'
forrnatiori
of any oyhe
_
critical cooling rate
softer products of transformatiojrjsj nown
asjfia|
JTrejgritical cooling rate, determined by chemical:!
composition and austenitic grain size, is an Important Drbpertvjjf aleelj Einne it indicates how fast a steel must
'
be cooled in order to form only|
mattensiie-
carbon is
Tl
thej5im£lesLQnfi.iQ_s.U
uent pre iiLiiiJhfi-JmcrostRtetu The steps usually followed tc
8/8 The Isothermal-transformation Diagram It is apparent from the previousl discussion that the iron-iron carbide equilibrium diagram is of little value in the study of steels cooled under nonequilibrium conditions. Many metal-
diagram are:
lurgists realized that time and temperature of austenite transformation had|
method of handling the small s of a wire threaded through a hoi cross section has to be small in
.
Step 1 Prepare a large numbe
a profound influence on the transformation products and the subsequent| properties of the steel. However, this was not given scientific basis until E
.
'
S. Davenport and E. C. Bain published their classic paper {Trans. AIME, \
perature.
in a
vol. 90, p, 117, 1930) on the study of the transformation of austenite at|
Step 2 Place the samples
constant subcritical temperature. Since austenite is unstable below the lower critical temperature Ae,, it is necessary to know at a particular subcritical temperature how long it will take for the austenite to start to trans- « form, how long it will take to be completely transformed, and what will be the nature of the transformation product. The best way to understand the isothermal-transformation diagrams
Step 3 Place the samples in a
.
v
vim
is to study their derivation.
austenitizing temperature. For approximately 14250F. They sh enough to become completely
subcritical temperature (a temj 1300oF.
' .
v
i '
9
THE HEAT TREATMENT OF STEEL
*»«
261
1
.
.
if real equilibrium
.
,
900
mi
although it may
65
Mpie structure can be considered as
i
BOO
and the final equilibrium 6onditlon
ts potential of very great hardness
B
ustenite from which it forms (Fig
r
600
.
iels that contain sufficient carbon st with increase in carbon content
.
i
bon. Beyond that point the curve 10 percent carbon) the hardness to the greater tendency to retain
m
bss of martensite is believed to be
if
-
v
.
.
pssAy its formation, since the amount held in solid sojution The maxi.
tensitic
Martensite in iron-carbon steels
55 w
(Bain and Paxton)
50 I
| 45 5
-
I 400
t
-
35 1 H30a: /-Austenite in
20
iron-nickel-carbon 200
condition seems to be a
I
40 1
o
5 300
Ah
11
O
Q
i
1
a>
500
,
,
60
700
'
.
alloys (Krauss)
m IOC
mt
rs, was believed to be unique ;;\;.;|ensite type of transformation Systems, such as iron-nickel, r formation is therefore recog-
1 i
I
ice a fujjy martensitic strucsecondl that wULauBm the
known
isformation is .
'
carbor is the simplest one to study sincithftrfi is ncLpmeutectoid const J uent pn?seriLjjill ic«>st
of a wire threaded through a hole in the sample, as shown in Fig. 8 13. The cross section has to be small in order to react quickiy to changes in tem-
perature. J Step 2 Place the samples in a furnace or molten salt bath at the proper
Ignite is unstable below the
l-transformati
on diagrams
| f
-
,
and what will be
-
Step 1 Prepare a large number of sampies cut from the same bar. One | ;| method of handling the small samples during heat treatment is by means |. ;|
classic paper (Trans. AIME iisformation of austenite at
trans-
'
'
diagram are:
..
b know at a particular sub-
'
M' The steps usualiy foiiowed to determine an isothermal-transformation :,;|| |
given scientific basis until
,
4
is to study their derivation. The_ eutectoid composition of 0.8 percent. / flf-, it-
jjm diagram is of little value
Vv sformed
Phillips,
the
:
rS austenite to start to
,
"
oducts and the subsequent
.
10
Structure and Properties of Alloys, 3d ed., McGraw-Hill Book Company, New York, 1965.)
"
jstenite transformation had
.
0.8
austenite. (By permission from Brick, Gordon, and
um conditions. Many metal-
,
0.6
Fig. 8-12 The effect of carbon on the hardness of
gwipparent from the previous :
_
-
0.4
.
Weight percent carbon
te. determined by chemical nportantjDropertv of a steel doled in order to form nn|y
'
02
austenite and on the hardness of martensite. The shaded area of the upper curve represents the Effect of retained
tate, and the term martensite if steel.
Ii
'
'
l
1
-
0
.
austenitizing temperature. For a 1080 (eutectoid) steel, this temperature is
r.
approximately 14250F. They should be left at the given temperature long
enough to become completely austenite.
\
Step 3 Place the samples in a molten salt bath which is held at a constant subcritical temperature (a temperature below the Ae, line),, for example, 1300oF..,
.
,
.
(
1
1
"\ }
'
'
/
'?.
:| )
v
262
INTRODUCTION TO PHYSICAL METALLURGY
6
1
Molten solt bath at
30E
2
100
BOOT
95/) +5 75')+ 25
F'g. 8-13 A typ'ca' samp'e wh'ch 's used to determ'ne an ,-T d,agram.
mi
.
-
50,1+ 50
254 + 75/' mp
-
8-14
Step 4 After varying time intervals in the salt bath, each sample IS quenched in cold water or iced brine. Step 5 After cooling, each sample is checked for hardness and studied microscopically. Step 6 The above steps are repeated at different subcritical temperatures until sufficient points are determined to plot the curves on the diagram. We are really interested in knowing what is happening to the austenite at 1300°F, but the samples cannot be studied at that temperature. There-
The progress of austenite transformati
coarse pearlite at 1300T as related to the structu emperature; A is austenite, M is martensite, P is -
transformation curve at 1300oF structures are shown in Fig. 8 that the transformation from £ the rate of transformation is ve it slows down toward the end.
fore, we must somehow be able to relate the room-temperature micro
scopic examination to what is occurring at the elevated temperature. Two facts should be kept in mind: 1
if
Martensite is formed only from austenite almost instantaneously at low tempera-
tures.
,
2 If austenite transforms at a higher temperature to a structure which is stable at t:j room temperature, rapid cooling will not change the transformation product. In M other words, if peanite is formed at 1300oF the pearlite will be exactly the same at ::; room temperature no matter how drastically it is quenched, since there is no reason ||for the pearlite to change. ,
3
.
100 r
75 h
Steps 3, 4, and 5 are shown schematically in Fig. 8-14. Sample 1, after 30 s at T300oF and quenched, showed only martensite at room temperature. Since martensite is formed only from austenite at low temperatures, it
50
ere was only austenite present mB|
means that at the end of 30 s at 1300oF th
5
25
and the transformation had not yet started. Sample 2, after 6 h at 1300oF | 2
and quenched, showed about 95 percent martensite and 5 percent coarse
pearlite at room temperature (see Fig. S-l 5). It means that at the end of 6 h 1|
Ti
at 1300°F there was 95 percent austenite and 5 percent coarse pearlite. Ju.
:i
Fin 8-15 Typical isothermal-transformation c austenite to pearlite for a 1080(eutectoid) stee
The transformation of austenite at 1300°F has already started, and thelHp
m
,
1
n
ri
s corresporu
transformation product is coarse pearlite. Using the same reasoning as|B| martens n Fig. s u. Magnificat above, the student should be able to follow the progress of austenite tran&mmk: os courtesy of Research Laboratory, u.s '
formation by studying samples 3, 4, 5, and 6. The typical isothermak||H| corporation.)
s
g
THE HEAT TREATMENT OF STEEL 263
'
I
!
-
66h-
i
i 21h-
-
- 18h6he n -
,
Molten salt bath
i
-
t
i
at
1300 °F
30 s
1
elMHRNM y y y v vI i IOO/J1004
p
ULlJLlI i tTT lOO/!/-J
4
5
water
6
I
95/!/+ 5/°-J
11
754+25/°
m
50/
504+ 50
V
I
-
\00P
;
1
+ 50/°
-
-25// !
°
.254+75-
(
100/1/
954+5/°
i " .
,
/'
J
-\00P
Fig 8-14 The progress of austenite transformation to
salt bath, each sample is
coarse pearlite at 1300oF as related to the structure at
temperature; A is austenite. M is martensite, .
Cold
ii Li.
llllll
room
P is pearlite.
.d for hardness and studied
transformation curve at 1300 F and several of the room-temperature micro-
t
"
snt subcritical temperatures e curves on the diagram. happening to the austenite it that temperature. Theres room-temperature microthe elevated temperature.
:
structures are shown in Fig. 8-15. The light areas are martehsite. Notice that the transformation from austenite to pearlite is not linear. Initially, the rate of transformation is very slow, then it increases rapidly, and finally
s i
it slows down toward the end. I
4
JM Istantaneously at low temperaa structure which is stable at
e transformation product.- In Ite will be exactly the same at
<
1
-
5
1
ched, since there is no reason ICQ
Fig. 8-14. Sample 1 after nsite at room temperature. e at low temperatures, it i/as only austenite present
75
,
*
;;;nple 2, after 6 h at 1300oF "
a)
50 QJ
MM
25i
6
isite and 5 percent coarse
oo
10
.
eans that at the end of 6 h
m percent coarse pearlite.
Fig. 8-15 Typical isothermal-transformation curve of
°
F;
already started and the
austenite to pearlite for a 1080(eutectoid) steel at 1300
g the same reasoning as
martensite is the light area. Micros correspond to the sample numbers given in Fig. 8 14. Magnification 500X.
,
-
ogress of austenite trans; The typical isothermal-
m
1
Time-hour, log scale
(Micros courtesy of Research Laboratory. U.S. Steel Corporation.)
j
i
j
i \
si
264 INTRODUCTION TO PHYSICAL METALLURGY
As a result of this experiment two points may be plotted at 130W the time for the beginning and the time for the end of transform tion. It is also common practice to plot the time for 50 percent transformedi The entire experiment is repeated at different subcritical temperatures unl sufficient points are determined to draw one curve showing the beginnin
Austenite (stable)
,
,
namely,
I
1000 H
of transformation another curve showing the end of transformation and dotted curve in between showing 50 percent transformed (Fig ,
QJ
Bainite
800
n
8-16). Ths]
o
principal curves on the l-T diagram are drawn as broad lines to emphasize
I 600
that their exact location on the time scale is not highly precise Portiori of these lines are often shown as dashed lines to indicate a much hi degree of uncertainty ghek Time is plotted on a logarithmic scale so that times! of 1 -min or less as well as times of 1 day or week can be fitted into a reason*! able space and yet permit an open scale in the region of short times
(unstable) | -
.
0
Fine pearlite R/C 40
\\ Upper or feathery bainile R/C 40
Lu
,
.
Coorse pearlite R/C 15 Medium pearlite R/C 30
Nose
I200h
w
Ms
100
Lower or acicular
Beginning
bainite I
50%
-%)
*
.
200
--Mso
,
.
Martensite R/C 64
Thej
diagram is known as an l-T (isothermal-transformation) diagram Other| names for the same curves are TTT (transformation temperature, timejw
0 05 .
10
102
I03
10"
.
curves or S curves. Construction of a reasonably accurate diagram requiresM;,
100
0
li 75
-
Typical Isolherma) Ironsformation
1 50 -
Ending
25
curve ot 700 0F
50%
50 75
Beginning r
,
10C
diagram for a 17 lIsothermal-transformation sothermai '
eu iccioid)
steel.
the heat treatment and metallographi
1 r
individual samples.
The l-T diagram for a 1080 eutectoi( the Ae, austenite is stable. The area formation consists of unstable austen of-transformation line is the product
constant temperature. The area
bet
B cor
J
1200
Time-second log scale
-
transformation labeled A + rite, and carbide, or austenite plus th
I
The point on the beginning of the tr
1000
\
is known as the nose of the diagrarr
\ U
-
eutectoid steel, there is an additional
1 80C
c
r
line to the left indicates the beginnir eutectoid ferrite in hypoeutectoid s
.
mentite in hypereutectoid steels. Ending
60C
ning of austenite
T
transformation to p(
is labeled A + f (austenite plus pro
50
Beginninc
plus proeutectoid cementite).
40C
The 1
region.
The Ms temperature is indicated to the temperature scale indicate th
m
ioz
10
Time
,
Fig
.
seconds
8-16 Diagram showing how measurements of
isothermal transformation are summarized by the l-T
i
diagram (From j'Atlas of Isothermal Diagrams U.S. Steel Corporation ) ,
"
,
.
m m
i
. .
f
t
if
i
Transformation
I03
10"
cent of the total austenite will, on
tensite. In some diagrams, the data
by direct meiasurement using a me grams, the temperatures were cal determine the progress of marter
THE HEAT TREATMENTfOF STEEL
4
'
1
265
i
AO
nts may be plotted at 1300oF time for the end of transforma-
4e
,
Austenite (stable)
:;;
:
.
.
.
ime for 50 percent transformed
S'?
it subcrltical temperatures until
je curve showing the beginning
!000 a
.
m
,
A
,
800
t transformed (Fig 8-16). The
Medium pearlite R/C 30.
(( Fine pearlite R/C 40 Upper or feathery bainite R/C 40
"
e end of transformation and a
Coarse pearlite R/C 15
Nose
1200
.
Bamite
+
3
.
/n as broad lines to emphasize
"
s not highly precise. Portions
Beginning \ \
les to indicate a much higher '7
\ bainite R/C 60
400
oP ogarithmic scale so that times :
s N < Endin5 (unstable) J N X N. Lower or acicuiar "
Austenite >v
600
|*- 50
:
ek, can be fitted into a reasonhe region of short times The isformation) diagram Other ormation temperature time)
50%
90
200
Martensite R/C 64
.
0 Ob I
.
,
1
,
ly accurate diagram requires
i
to 102 I03 Time-second log scale
.
10"
fell
Fig. 8-17 Isothermal-transformation diagram for a 1080(eutectoid) steel
-
.
;
>i
0
25
£
the heat treatment and metalldgraphic study of more than bne hundr id SM '
1
individual samples.
50 75
J 100
f
.
W
The l-T diagram for a 1080 eutectoid steel is shown in Fig.;8-17. Above '''M$k
i
the Ae, austenite is stable. The area to the left of the beginning of trans
pl
"
i
formation consists of unstable austenite. The area to the right of the end |M|| of-transformation line is the product/to which austenite will transform at >mi|
constant temperature. The area between the beginning and the end ;of:4;M | '
transformation labeled
A + f+ qconsists of three phases, austenite, fer- ' |
rite, and carbide, or austenite plu the product to which it is transforming. The point is the is known as the nose nose of of the the diagram. diagram. In In all all diagrams, diagrams, except except the the one for:
.
. 4 on the beginning of the transformation line farthest to the left \ji; c|| v
'
Ml
eutectoid steel, there is an additional line above the nose region. The first' line to the left indicates the beginning of austenite transformation to pro-
eutectoid ferrite in hypoeutectoip steels (Fig. 8-29) or proeutectoid cementite in hypereutectoid steels.
'9
The second line indicates the begin-
ning of austenite transformation to pearlite. The area between the two lines
is labeled A + F (austenite plus proeutectoid ferrite) or A + C (austenite plus proeutectoid cementite). The two lines generally merge at the nose ,
region.
if
The Ms temperature is indicated as a horizontal line. Arrows pointing to the temperature scale indicate the temperature at which 5)0 and 90 per10"
cent of the total austenite will
,
on quenching, have transformed to mar-
tensite. In some diagrams the data on martensite formation Were obtained '
,
by direct measurement using a metallographic technique, in other diagrams, the temperatures were calculated by an empirical formula. To determine the progress of martensite formation metallographically, let
11 i
i
Tl
266
IMTRODUCTION TO PHYSICAL METALLURGY
us drastically quench a sample to a temperature below the Ms line
,
s i
350oF. Approximately 20 percent of the austenite will have transform?!
V
to martensite. If this sample is now reheated for a short time to a tempei ture below the lower critical line, say 800oF, the martensite just formed become dark. The austenite will remain unchanged. Upon quenching tl sample will show 20 percent dark martensite and 80 percent fresh orwhiti
s.
,
S
martensite, and the amount formed at 350oF may therefore be determinei:
m
microscopically. A typical example of the transformation of austenite to martensite determined by the heat-treating procedure described above is shown in Fig. 8-18. 8-9 Transformation to Pearlite and Bainite Returning to Fig 817, the trans-;
1i
.
formation product above the nose region is pearlite. The pearlite micro-'
"
I
structure is the characteristic lamellar structure of alternate layers of fers
.
"
1
mm
i -
I
11
450 r
1 400
5.
350 5 JZ
u
Hf.
50-
3
-
300
to
5
'<:
o
a
-
D
250 -
90
I
E
200
1
(c). 40
20
0
60
Percent martensite observed i
Fig. 8-18
Typical example of the transformation of
austenite to martensite. Microstructures at 500X. (From
Atlas of Isothermal Transformation Diagrams," U.S. Steel
"
Corporation.)
MS
.v.
'
150
\
:
80
100
I, 8.9
Pearlites formed by the isothermal transformatic
Zus.oni.0 at various subcritica, = [cx 1150
°
F
,
f
(d) 1075°F. Magnified 1,500X.
Zl Note
£f s m the fineness of pearlite With decreasing ormation
temperature, (Courtesy 0, Research
taboratory. U.S. Steel
Corporation.)
s 1
THE HEAT TREATMENT OF STEEL »T
perature below the Ms line, say
i .
austenite will have transformed
5s
:
the martensite just formed will ichanged. Upon quenching this
s
\
,
,
jte and 80 percent fresh or white
.
F may therefore be determined
i
I 5 at
i transformaiion of austenite to
\
1 procedure described above is
i
\
drning to Fig.\8-17, the trans-
P:s
pearlite. Th|e pearlite micro-
cture of alterrtate layers of fer-
\
t
i
lb. .
.s
.V5
A
i A* n
1
IP
w
ft 4
4
.
i1
5 '
5
J
.
5 '
4
SB
'
1'?
1
r
s
it
v
1
5 00
(C);
Fig. 8-19 Pearlites formed by the
subcritical Ihe increase in the fineness
isothermal '
'ormation
pe?tUres (a, 1300 F,
of austenite at various Note lb) 1225'F, (0) 1150 F (d) 1075»F. Magnified 1.500X of pearlite with decreasmg transformation temperature, (Courtesy of Research °
.
Laboratory, U.S. Steel Corporation.)
ri
I
\
i iN for a short time to a tempera-
;
.
. .
i
i
'
268 INTRODUCTION TO PHYSICAL METALLURGY
m
i;
-
'
v.
Si 1
V
.
i
ii -
!
11
9k
'
>
i
i
.
la)
Fig. 8-21
4 Fig
8-20 (a) Feathery bainite and fine pearlite In a martehsitic (white) matrix I.OOOX. (b) The microstructure
of bainite transformed at SOOT, taken with the el
microscope, 15,000X, (Courtesy of
.
,
,
,
,
.
.
US
|
of bainite transformed at 850"F taken with the electron microscope 15,000X, (Courtesy of Research Laboratory US
.
Steel Corporation.)
.
transformation temperature decreases the characteristic lamellar struc ;
,
,
ture Is maintained
but the spacing between the ferrite and carbide layersi
,
°
becomes increasingly smaller until the separate layers cannot be resolvedwith the light microscope (see Fig 8-19). As the temperature of transformation and the fineness'of the pearlite decreases it Is apparent that the hardness will Increase. As explained under normalizing this hardness increase is the result of decreasing the spacing between plates of the hard constit.
is nucleated by a carbide crystal, bai
,
and this results in a different growth The hardness of bainite varies fr bainite to about Rockwell C 60 for lo
,
jjent, cementite, within the soft ferrite matrix, r y '-j
Between the nose region of approximately 950oF and the Ms tempera-
dark-etching aggregate of ferrite and cementite appears
.
This structure named after E. C. Bain, is called bainite At upper tempera- , tures of the transformation range it resembles pearlite and is known as upper or feathery bainite {Fig 8-20a). At low temperatures it appears as a ,
.
,
.
black needlelike structure (Fig 8-21a) resembling martensite and is known .
as lower or acicular bainite structure of bainite and the .
I
,
These photomicrographs show the gross electron microscope is required to resolve
I
which make up the matrix. As the tra the ferrite needles become thinner smaller and more closely spaced. The at an angle of about 60 to the long a parallel to this direction (see Fig. 8-21 ite transformed at 500oF and magnif
,
\ i
Research Lab
Steel Corporation.)
generally oriented parallel with the
.
,
.
the details. Figure 8-20/3 shows the 850oF and magnified 15,000 times.
rite and cementite described in Chap 7. Just below the Ae line, coarse lamellar pearlite is formed with a hardness of about Rockwell C 15 As the
ture, a new
(a) Acicular or lower bainite, black net
a martensitic (white) matrix, 2,500X. (b) The micr
f
as with pearlite, is a reflection of th( carbide platelets as the transformati chahicarpfoperties of a completely t in this chapter under austempering. 6 10, Cooling Curves and the l-T Diagram m
perimentally by placing a thermocc sample and then measuring the vari; the coordinates of the l-T diagram
0k
i
THE HEAT TREATMENT OF STEEL
269
m arc
r
4
.
i
tm i
1
id)
Fig 8-21 (a) Aoicular or lower bainite, black needles in a martensitic (white) matrix, 2,500X. (b) The microstructure of bainite transformed at SOOT, taken with the electron microscope, 15,0OOX. (Courtesy of Research Laboratory, US .
.
Steel Corporation.)
m
the detfils. Figure 8-20b shows the structure of bainite transformed at
SSCF and magnified 15,000 times. generally oriented parallel with
; : .
.
hacac!£nstLcJ mfiJJai tm
he ferrite and carbide layers
Vufite layers cannot be resolved
crystal, and this results in a different growth pattern, as illustrated in Fig. 8-22.
The hardness of bainite varies from about Rockwell C 40 for upper bainite to about Rockwell C 60 for lower bainite. This hardness increase, as with pearlite, is a reflection of the decrease in size and spacing of the carbide platelets as the transformation temperature decreases. I he mechanical properties of a completely bainitic structure will be covered later in this chapter under austempering.
950oF and the /Ws tempera.
v¥
?ite and cementite appears
.
) bainite. At upper temperaV
:;
.
5S pearlite and is known as
. .
§: "
:
mperatures it appears as a
Ha
ope is required to resolve
i
8
is determined ex-10/cooling Curves and the l-T Diagram A cooling curve finite location in a steel
Kj
Vng martenslte and is known prographs show the gross ;
-f '
is nucleated by a carbide crystal, bainite is nucleated by a ferrite
it is apparent that theJiard:
izing, this hardness increase n plates of the hard constit-
'
-
,
°
r
b temperature of transforma,
] '\ '
the ferrite needles, become thinner and the carbide platelets J>.ec.Qme smaller and more closely spaced., the carbide platelets are usually oriented V at an angle of about 60 to the long axis of the ferrite needles, rather than parallel to this direction (see Fig. 8-21 b, which shows the structure of bainite transformed at 500oF and magnified 15,000 times). Whereas pearlite
,
,..
||
the long direction of the ferrite needles
which make up the matrix. As the transformation temperature decreases,-
3t below the Ae, line coarse about Rockwell C15. As the
S
It consists of tiny carbide plateleter
'
i
perimentally by placing a thermocouple at a de
sample and then measuring the variation of temperature with time. Since the coordinates of the l-T diagram are the same as those for a cooling
,
270
INTRODUCTION TO PHYSICAL METALLURGY
curve, it is possible to superimpose various cooling curves on the l-T diagram. This was done in Fig. 8-23. Cooling curve 1 shows a very slow cooling rate typical of conventional annealing. The diagram indicates that the material will remain austenitic for a relatively long period of time. Transformation will start when the cooling curve crosses the beginning of transformation at point x,. The transformation product at that temperature will be very coarse pearlite. Transformation will continue until point x',. Since there is a slight difference in temperature between the beginning and end of transformation, there will be a slight difference in the fineness of pearlite formed at the beginning and at the end. The overall product will be coarse pearlite with low hardness. Below the temperature of x', the rate of cooling will have no effect on the microstructure or properties. The material may now be cooled rapidly without any change occurring. This is of great value to companies doing commercial annealing, since the diagram indicates that it is not necessary to cool in the furnace to room temperature but that the material may be
Austenile (sloblel
'
1
7
4\
\
\
,
Uf
\
\ \
\CCR
\
\ \ V
A
\
\
\
\
\
Auslenite \
.
\ (unstable!
\
\
\
in air. ,
tween X3 and x'3 than there is between x, and x',, the normalized micro-
N
\
\ \
E
250/
\
CD
CD
Fine pec
V5
removed at a relatively high temperature after transformation and cooled
of medium/pearlite. Since there is a greater temperature difference be-
\
V
\ \
.
3
\
\
'
Cooling curve 2 illustrates "isothermal" or "cycle annealing" and was developed directly from the l-T diagram. The process is carried out by cooling the material rapidly from above the critical range to a predetermined temperature in the upper portion of the l-T diagram and holding for the time indicated to produce complete transformation. In contrast to conventional annealing, this treatment produces a more uniform microstructure and hardness, in many cases with a shorter time cycle. Cooling curve 3 is a faster cooling rate than annealing and may be considered typical of normalizing. The diagram indicates that transformation will start at X3 with the formation of coarse pearlite, in a much shorter time than annealing. Transformation will be complete at x 3 with the! formation
3
4
\
\
6
Ms
i
1
i
\ \
m
.
,.2* coiw ""."ST.iV. \
structure will show a gr
;
smaller proportion of co Cooling curve 4 typic; described, and the micr
Pearlite
,
1
.
±1 1
li
Fig. 8-22 Growth of pearlite, nucleated by a carbide crystal, and of bainite, nucleated by a ferrite crystal with carbide rejected as discontinuous small platelets. (After
pearlite.
Cooling curve 5 typic? form (at x5) to fine pear ,
to fine pearlite will cont centage transformed, s ing cooling curve is go
Since pearlite cannot f
stop at x 5 The micros '
.
Hultgren.)
J
THE HEAT TREATMENT OF STEEL
It;
s
271
"
> P
)
1
us cooling curves on the i-T diaAusienile (stoble)
jailing rate typical of conventional e material will remain austeniticj rmation will start when the cool- ;i ormatlon at point x, The trans- :"
CooTse peo'lile R/C 15
3
4
.
be very coarse pearlite Transince there is a slight difference iend of transformation there will
1
Ae._
\'
Xy
.
-
5
K Medium peorlile R/C JO
.
>
,
1
Vse pearlite with low hardness
sS oling will have no effect on th
\ \Fine peorlite R/C 40
*
,
lite formed at the beginning and
\ 0
"
.
.
e
=3
al may now be cooled rapidly idicates that it is not necessary e but that the material may be
\ \CCR
1
jreat value to companies doing
\
11
i 9
\* Austenite AUSltJIIMC
1
\
\
em
L
or "cycle annealing" and was rO process is carried.out by cooljcal range to a predetermined
_
Ms
\
A+F+C
\
\
\V
NN
\
-
_W
\
\
V
\
\
Morlensite R/C64
\
cycle.
in annealing and may be conearlite
in a much shorter time plete at x 3 with the formation )r temperature difference he,
ld x', the normalized micro,
It
mi
Lower bainite R/C 60
\ (unstable! \
agram and holding for the time m. In contrast to conventi onal i uniform microstructure and
indicates that transformation
P-
\
\
\
v fter transformation and cooled
.
Upper boinite R/C 40
-
Time, log scale
Fig 8-23 Cooling curves superimposed 1-T
on a
i
hypothetical
diagram for a eutectoid steel. Cross-hatched portion
of the cooling curve indicates transformation. \
!
f pearlite and a structure will show a greater variation in the fineness o smaller proportion of coarse pearlite than the annealed microstructure.
Cooling curve 4, typical of a slow oil quench is similar to the one just ,
described, and the microstructure will be a mixture of medium and fine pearlite.
Cooling curve 5, typical of an intermediate cooling rate, will start to transform (at X5) to fine pearlite in a relatively short time. The transformation 1
r
to fine pearlite will continue until the curve becomes tangent to some pe centage transformed, say 25 percent, at x'5. Below this temperature, the d -
cooling curve is going in a direction of decreasing percent transforme
1
Since pearlite cannot form austenite on cooling, the transformation must t of fine, stop at x The microstructure at this point will consist 25 percen '
5
.
..
3
.
.
'
'i
272
.
.
INTRODUCTION TO PHYSICAL METALLURGY
I
nodular pearlite largely surrou Mum ndingy the existing» i r>r\r\r4Ul ..... ,
I
austenitic grains.
will remain in this condition until the Ms line is crossed at x's
it
,
.1
transformation in the nose rc
„|
ing austenite now transforms to martensite. The final mi The remain- % crostructure at room temperature will consist of 75 percent martensite (whit 25 percent fine nodular pearlite largel e areas) and along the original austenite grain boundaries (Fig. 8-24) yIf concentrated only a small a
reached at x6.
.
Ms
a
Transformati and /Wf lines. The final
high hardness. It is apparent that to obtai
is present the black etching nodules of pearlite in th mount of pearlit e white martensiti matrix nicely reveal the former a .
to avoid transformation in th(
e
,
is tangent to hejLDse, would
c
tion to pearlite took place
ustenitic grain size when the'tran sforma-
for tTTis sTeel. Any cooling r
.
Cooling curve 6, typical of a drastic quench is rapid enough to avoid
curve above the nose and f(
,
m
cooling rate faster than the ferent steels may be compa Notice that it is possible
3*
martensite by continuous cc
r
bainite. A complete bain'H rapidly enough to miss the i perature range at which bail
f
This is illustrated by coolin
J
continuously cooled steel sj ite, and this is probably the until the isothermal study.
811 Transformation on Continuoi
should not be superimpose! section. The l-T diagram austenite transformation o »
most heat treatments invc r
is possible to derive from tf the transformation under ( C-T C-T
diagram (cooling-trar diagram for a eutecto
which it was derived.
Co
location of lines of the C-T
3
downward and to the righl
tangent to the nose of the what slower than the rate of isothermal
MS
"
"
nose
tim
to some error; however,
slightly faster cooling rat( ite.
Notice the absence
gram. In this steel the pearlite nose, and bainite '
R. A. Grange and J. M. Kiefer, Translormatio
Constant Temperature, Trans. A.S.M.. vol. 29, i
it
i THE HEAT TREATMENT OF STEEL
c grains! !
the existing austeniti ,ine is crossed at x" The remetft)
transformation in the nose region. It remains austenitic until the Ms line Is reached at x6. Transformation to martensite will take place between the Ms and M, lines. The final microstructure will be entirely martensite of high hardness. It is apparent that to obtain a fully martensitic structure it is necessary to avoid transformation in the nose region. Therefore, cooling rate 7, which is tangent to thejaose, vwuid bfiJJie-appfoximate critiCat-coolmg-tate CCB)-
s
.
-
Sansite. The final microstructur|| rcent martensite (white
areas)
273
r
'
y concentrated along the original
nt of pearlii|
If only a small amou f pearlite in the white martensltii ,c grain size when theVansformai
I
-
_
.
for tfiis steeTAny cooling rate slower than the one indicated will cut the 1
-
curve above the nose and form some softer transformation product. Any: J
quench, is rapid enough to avoid
0
]
,
cooling rate faster than the one illustrated will form only martensite. Different steels may be compared on the basis of their critical cooling rates. Notice that it is possible to form 100 percent pearlite or 100 percent
martensite by continuous cooling, but it is not possible to form 100 percent bainite. A complete bainitic structure may be formed onlyjby cooling
rapidly enough to miss the nose of the curve and then holding iDth em-;; perature range at which bainite is formed until transformation Is complete.. This ir. illustrated by cooling rate 6 then 8 in Fig. 8-23. It is apparent that
4| continuously cooled steel samples will contain only small amounts of bain- >| ,
ite, and this is probably the reason why this structure was not recognized until the isothermal study.
BH
Transformation on Continuous Cooling Theoretically, cooling-rate curves should not be superimposed on the l-T diagram as was done in the previous section.
4 '
'
..v. '
The l-T diagram shows the time-temperature relationship for
S
g|
austenite transformation only as it occurs at constant temperature, but
r
most heat treatments involve transformation on continuous cooling.
It
is possible to derive from the l-T diagram another diagram which will show the transformation under continuous cooling.* This is referred to as the C-T diagram (cooling-transformation diagram) Figure 8-25 shows the .
C-T
'
diagram for ai eutectoid steel superimposed on the l-T diagram from which it was derived. Consideration of the l-T diagram in relation to the location of lines of the C-T diagram shows that the "nose" has been moved downward and to the right by continuous cooling. The critical cooling rate tangent to the nose of the C-T diagram is shown as 250°F/s. This is somewhat slower than the rate indicated by the l-T diagram. Therefore, the use of isothermal nose times to determine required cooling rates will lead to some error; however the error will be on the safe side in indicating a slightly faster cooling rate than is actually necessary to form only martensite. Notice the absence of an austenite-to-bainite region in the C-T diagram. In this steel the bainite range is "sheltered" by the overhanging pearlite nose, and bainite is not formed in any appreciable quantity on ordi-
ft
"
"
,
I
. R A. Grange and J. M. Kiefer, Transformation of Austenite on Continuous Cooling and Its Relation to transformation at .
Constant Temperature, Trans. A.S.M., vol. 29, no 85, 1941. .
J
274 INTRODUCTION TO PHYSICAL METALLURGY
greatly aided the classificatic during continuous cooling, ai
800
1400
emper alurel " _
_
-
~
s
-
'4
e
ize at what stage of the cooli
-H
Austenife peom begm
l-T
?oc
why steel responds as it doe:
I20C
be used directly to predict a
\
1
\
\
Austenite
V
\
pearlite complete
continuous cooling. It should be noted that a f
600
\
\
1000 u
diagram is useful in plai
V
\
_
one group of samples; samp
500 "
tions in the same heat, are
1 800
=
\
m
Transformation stops
\
When used with its limitatio
400 g
600
i
300
1600 -Austenilizmg
400
;
Austenite
m
l
-martensilic structure I I I
200
-
Final structure
1
i
Isothermal diagram - Continuous transformation diagram Constant rate cooling curves
temperature
200 I400
2100 °F/I .
-J
100
I I I Martensite - Mar'er,site_ - Pearlite (softer, coarser-H1 pearlitel
\
.
0
1200 \
I
.
10
I02
\
s
0 0 1
I04
10'
\ \,54000
Transformation time seconds ,
,
"F/h
1000
Fig. 8-25 Continuous cooling-transformation (C-T) diagram
derived from the isothermal-transformation diagram for
U
\ -
a plain-carbon eutectoid steel (From "Atlas of Isothermal Transformation Diagrams U.S. Steel Corporation )
\
\
.
'
"
,
800
.
1)
nary continuous cooling
.
600
This situation is generally different for alloy steels
.
Figure 8-26 shows the C-T diagram for a triple-alloy steel. This is a hypo-
eutectoid steel so there is an additional area austenite-to-ferrite which was not present in the eutectoid steel In this alloy steel the pearlite zone
Austenite -
,
,
,
400
2
.
,
lies relatively far to the right and does not shelter the bainite region Therefore with cooling rates between 2,100 and 54 000oF/h it is possible to obtain appreciable amounts of bainite in the microstructure Noyice that the .
,
,
200
.
cooling rate tangent to the "upper nose" (2 100oF/h) ,
is not the critical
cooling rate. The cooling rate tangent to the "lower nose" or "knee" of the diagram (54 000'F/h) would have to be exceeded to form only mar-
i
l I Li
_
j
The derivation of a C-T diagram is a tedious task and for many purposes not essential provided that the fundamental relationship of the C-T diagram to the corresponding l-T diagram is understood. Isothermal studies have ,
Fig. 8-26 C-T diagram of a triple-alloy ste
,
percent carbon, 0.78 percent
manganese,
(From "U.S.S. Cariiloy Steels,
"
nickel, 0.80 percent chromium, 0.33 perce;
Li
m
100
0
,
tensite.
Mar
Manensile
U.S. Steel (
THE HEAT TREATMENT OF STEEL
275
is
greatly aided the classification of the microstructure of steel transformed during continuous cooling and with the l-T diagram it is possible to visualize at what stage of the cooling cycle different, structures are formed. The
Hsoo
,
l-T
700
:
-
peorlite complete
why steel responds as it does to a particular heat treatment but it cannot be used directly to predict accurately the course of transformation under continuous cooling.
600
It should be noted that a particular l-T diagram exactly represents only one group of samples; samples from other heats, or even from other loca-
500 o
tions in the same heat
are likely to have slightly different l-T diagrams.
i i
,
400 S
I
When used with its limitations in mind the l-T diagram is useful in inter-
,
.-)nsformation stops
:
1
diagram is useful in planning heat treatments and in understanding ,
,
l i
300
-Austemtizmg
:
-
:: : -
.
.
lermal diogram > nuous transformation
i
rj.ii (softer
,
Lrser-H
c
lemperature
200
1400 Mfj
ont rate cooling curves J -
i
1600
T
=rr.r=b
_
r.=
r:
,-
±zn=:
2100 0F/h
4ei
100
1200
-
i 103
J
.
5S N
\\
Austemte
\
\
I04
\
i
\2l00oF/h
\54000 "F/h
1000
\ 40 "F/h
\ 150"F/h-r
\
\
LL.
pearlite
\
2 S
800
i
a
.
Auslemle
baimte
I
;
1
600
enerally different for alloy steels iple-alloy steel This is a hypo-
1/
.
.
irea,
austenite-to-ferrite which
Austemte
,
his alloy steel the pearlite zone
rtensite
22
400
i
,
lelter the bainite region There54,000oF/h it is possible to ob microstructure Notice that the (2,100oF/h) is not the critical
m
.
-
200
.
.:. :the -
.
.
"lower nose" or "knee
'
Martensite
of UU
e exceeded to form only mar-
us task and
,
for many purposes
n
!
10
Fig. 8-26
,
relationship of the C-T diagram stood. Isothermal studies have
it-
.
11: It
100
Martensite bainite
I I 1111 1000
Martensite -
ferrite balnite I I Mil
10 000 Transformation time, seconds
C-T diagram of a triple-alloy steel (4340): 0.42
percent carbon, 0.78 percent manganese, 1.79 percent nickel, 0.80 percent chromium, 0.33 percent molybdenum.
(From "U.S.S. Carilloy Steels," U.S. Steel Corporation.)
,
Martensite!
Lferrite ' Tpearlile , bpinite lOQCOO
Ferrite
pearlite J
i
i
i 11 n
1 000,000 ,
m 276 INTRODUCTION TO PHYSICAL METALLURGY
preting and correlating observed transformation phenomena on a rational i
basis, even though austenite transforms during than at constant temperature
m
continuous cooling rather |
n
.
8-12
Position of the l T Curves
There arp only two factors that will change
-
,
the position of the curves of the l-T diagram namely, chemical composition ,
and austenitic grain size With few exceptions an increase in carbon or alloy content or in grain size of the austenite always retards transforma.
,
tion (moves the curves to the right) at least at temperatures at or above the nose region. This in turn slows up the critical cooling rate making ,
,
it easier to form martensite
This retardation is also reflected in the greater hardenabiiity or depth of penetration of hardness of steel with higher alloy .
,
,
content or larger austenitic grain size The effect of increasing carbon content may be seen by referring to Figs. 8-27 and 8 29 Figure 8-27 shows the l-T diagram of a 1035 steel The /W .
.
.
.
s
Fig 8 28 Microstructure of a low-carbon siee
queached, showing a white ferrite network sut Ihe gray low-carbon martensite areas, (s) lOOx Etched in 2 percent nital.
800
"
Austenite
-Ae3
temperature is approximate!'
It
-
notice the presence of the au;
1400
Austenite and ferrite
Ae
is not visible, indicating that i
700
7
1200
-
I?
600 PS
Ferrite and corbid€ u
o °
.
- 500 g
,
n
i
| 800
CD
CL
25
Austenite ferrite and carbide n
O
D
I
iooo
hV
400
27
N
L-Llli: '
-
50% mortensite on quenching lo this temperature 99% mortensite on quenching lo this temperature
reduced to 620oF
.
Theoret
Notice that the A + F region
46
the vicinity of the nose. T water-quenched. Fig. 8-30, outline some of the previou
47
400
200
the right to make the nose j
35 |
43 I
600
300
low-carbon martensite areas
steel. The increase in carbo
necessary to cool fast enoi L!-_ -L
l l- /y -
to obtain only martensite. Th quenched, Fig. 8-28, shows
ery bainite, and substantial! in the low-carbon steel.
Although alloy additions t tion and to increase the tim
I00
200
magnitude and nature of th mm
oL
I
I I III!
i
L
i
_
10
m -
15 30 I min min hour i
i
i 11111
100
1000
Time, seconds
Fig, 8-27 l-T diagram of a 1035 steel: 0 35 percent carbon 0 37 percent manganese Grain size: 75 percent 2 to 3; 25 percent, 7 to 8 Austenitlzed at 1550"F, (From "Atlas of ,
,
10
I - 52
hour day
11
jJi
_
10 000 ,
Ii lOQOOO
of a 1335 manganese steel. that the addition of 1.50 pt to the right. The nose of tht to completely harden this steel. Notice that the end-
,
.
,
,
Isothermal Transformation Diagrams Corporation.)
T
~
"
,
U.S. Steel
Where there is a pronounc
.
v-
i THE HEAT TREATMENT OF STEEL
(formation phenomena on a rational ;
.
.
.
4
*-
is during continuous cooling rather
; :
277
.
.
5
pnly two factors that will change Jram, namely, chemical compogitjpj] exceptions an increase in carbon ,
.
iustenite always retards transformai least at temperatures at or above ip the critical cooling rate making ation is also reflected in the greater ,
Martensite
| :f hardness, of steel with higher alloy int may be seen by referring to Figs ,
herrlte
.
-T diagram of a 1035 steel. The Ms
Fig. 8-28 Microslructure of a low-carbon steel, waterquenched, showing a white ferrite network surrounding the gray low-carbon martensite areas, (a) 100x; (b) 500x. Etched in 2 percent nltal.
e
7
2
-
25
25 27 35
i
Lit
5
:
43 5
iq to this temperature
46
ng to this temperature
47
temperature is approximately 750oF. Since this is a hypoeutectoid steel, notice the presence of the austenite-to-ferrite region. The nose of the curve is not visible, indicating that it is very difficult to cool this steel fast enough to obtain only martensite. The microstructure of a low-carbon steel waterquenched, Fig. 8-28, shows a white ferrite network surrounding the gray low-carbon martensite areas. Figure 8-29 shows the l-T diagram for a 1050 steel. The increase in carbon content has shifted the curve far enough to the right to make the nose just visible, and the Ms temperature has been reduced to 620°F. Theoretically, in order to form only martensite, it is necessary to cool fast enough to get by 1000°F in approximately 0.7 s. Notice that the A + F region has become much narrower and disappears in the vicinity of the nose. The microstructure of a medium-carbon steel water-quenched. Fig. 8-30, shows dark areas of fine pearlite that seem to outline some of the previous austenite grain boundaries, some dark feathery bainite, and substantially more martensite as the matrix than appeared in the low-carbon steel.
Although alloy additions tend in general to delay the start of transformation and to increase the time for its completion they differ greatly in both ,
15 30 I min min hour
0
|-52
hour . day I I' '
1000
10 000 ,
conds
10Q000
1
magnitude and nature of their effects. Figure 8-31 shows the l-T diagram of a 1335 manganese steel. Comparison of this diagram to Fig. 8-27 shows that the addition of 1.50 percent manganese has shifted the entire curve to the right. The nose of the curve is now visible and it should be possible ,
to completely harden this steel much more easily than the plain carbon
steel. Notice that the end-of-transformation line now shows an S shape. Where there is a pronounced minimum time in the ending line at relatively
=1
278
INTRODUCTION TO PHYSICAL METALLURGY
°
r
0F
c
in
800
4
800
1400
a
1400
C3
3 -
Ae
-
700
1200
18
1000
28
600
Ae, -
700 1200
23 F+C
500 -
/
600
32
1000 800
ii
a
.
500 42
4
,2
600
300
48
Austen,te, fer
QJ
S 800 400
52
200 -400
6C»
1 -T diagram 100
Estimated temperature
200
-
'
r
i
! hour
mm
-
.
0 2
.
5
10
MY
102
103
99% martei
Jjuj I06
toJ
10"
50% marler
I week 62
nnm
mi
05 1
%
300
'
C
7
.
50*
.
/
36
50%
400
3
as
4- -
4
200
400
Time-seconds 100
Fig. 8-29 l-T diagram of a 1050 steel; 0.50 percent carbon, 0 91 percent manganese. Grain size: 7 to 8; austenitized at 1670°F. (From Atlas of Isothermal Transformation Diagrams," U.S. Steel Corporation.)
200
.
"
0 -
I
I Mill
10
high temperature, it is possible to take advantage of this minimum to design a short annealing cycle. In this diagram the upper minimum (point x) is approximately 5 min at HOOT for the end of transformation, whereas at 1200oF the end of transformation Is approximately 1 h. Therefore, isother-
Fig. 8-31 l-T diagram of a 1335 steel. 0.35 p 1 85 percent manganese. Grain size: 70 perc percent, 2. Austenitized at 1550 F. {From "At
.
1>
.
"
thermal Transformation Diagrams,
U.S. Ste
r
mal annealing at 11 GOT wc
ft
ure 8-32 shows the effect of
V
»
Mortensite matrix
has not only shifted the c 4
particularly in the region o
V
cumulative.
This is iliustr
(Fig. 8-26). The critical co mately 54,000oF/h
,
or abo
rate of 250oF/s for the eu
that this eutectoid steel h
Feathery bainite
iOFine '
X
f
.
enability. The triple-alloy
«
peorlite, ,
hardening, and illustrates addition of alloying eleme
1 r
.
.
!
i
Fig. 8-30 Microstructure of a medium-carbon steel, waterquenched, showing dark areas of fine pearlite that seem to outline, some of the previous austenite grain boundaries, some dark feathery bainite, and a matrix of martensite. (a) lOOx; (6) 750x. Etched in 2 percent nilal.
ing steel. In general, the on the movement of the
(strongest), tungsten, mt nickel (weakest).
i
t
h
>
mmm
-
f
THE HEAT TREATMENT OF STEEL
279
3
800 Ae X .
u
.
3
1400
3
Austemte
.
Ae, -
700
18
Austemte and fernte
23
1200
28
/
60C
r
-
18
1000
36
L
15
x
/
32
500
12
Austemte
QJ
fernte and carbide
21 rernte and corbide
48 a
52
S 800
-
\
--/
33
.
a)
T3
400 41
600
1
% -1
300
-
46
50% martensite on quenching to this temperature 48
.
.
Lday_
our
-
i Til i ii
j i -
104
I week 62
105
i
106
200 -
400
100
200
99% martensite on quenching to this temperature 48
I -
..
::-}
15 0
vantage of this minimum to design m the upper minimum (point x) is
Martensite
matrix. a
.'
i
10
10
I
LUJ
i i
100
1000
1
i
1
hour day lull
-
10 000 ,
i
59
i Ii l
100,000
Time, seconds
Fig. 8-31 l-T diagram of a 1335 steel. 0.35 percent carbon, 1 85 percent manganese. Grain size: 70 percent, 7; 30 percent, 2. Austenitized at 1550T. (From Atlas of Isothermal Transformation Diagrams," U.S. Steel Corporation.)
,
,
ii i
i
lend of transformation whereas at
proximately 1 h. Therefore
i
I
30
min min hour
min
.
"
isother-
mal annealing at HOO F would be complete in a relatively short time. Figure 8-32 shows the effect of chromium. The addition of 2 percent chromium has not only shifted the curve to the right but has changed its shape, particularly in the region of 900 to 1200oF. The effect of alloy additions is cumulative. This is illustrated in the C-T diagram for a triple-alloy steel
1
-
3f
(Fig. 8-26). The critical cooling rate for this steel is shown to be approximately 54,000°F/h, or about 15°F/s as compared with the critical cooling
5
A
rate of 250oF/s for the eutectoid steel shown in Fig. 8-25. It is apparent that this eutectoid steel has a very fast critical cooling rate and low hardenability. The triple-alloy steel has a slow critical cooling rate, is deephardening, and illustrates how much easier it is to form martensite by the
addition of alloying elements. This is one of the principal reasons for alloying steel. In general the relative effect of the common alloying elements on the movement of the nose of the l-T diagram to the right is: vanadium (strongest), tungsten, molybdenum, chromium, manganese, silicon, and nickel (weakest). ,
v i -
-
.
-
MM,
200
INTRODUCTION TO PHYSICAL METALLURGY
quenched to obtain full hardr
m
attain the same hardness anc 800
.
- Ae 3
when cooled more slowly by
400 700
Ferrile and carbide
Austemte
i 1200 .i
.
rate reduces the danger of di; While coarsening the austeni
4
adding alloyng elements, the
5
26
600
ness of the steel.
28
Austenite and ferrile 1000 500
27
OJ
31
ftusfemte ferrite and carbide
to
,
|2
S 800 400 L i
35 4
-
300
600
10C
20C
soft spots and lower hardnes For plain-carbon hypereut(
48
99% mcrtensite an quenching to this temperature 400
eutectoid ferrite present whl(
48
50 /o mcrtensile an quenching to this temperclure
200
rate, this may be done best b than by coarsening the austei 8 13 Hardening or Austenitizing Tei temperature for hypoeutectoi cal-temperature (A,) line. Thi temperature. At any tempera
44
Fernte and carbide
Ms
Therefore,
800
1400
15 39
mir
C
1
1 1 mi
UJ
0
10
min min hour 11
100
100c
70C
1
hour day
58
1200
1 li 1 ml
10000
V
600 h
I0000C
Time seconds "
l-T diagram of a 0 33 percent carbon, 0 45 .
.
1000
,
l:ig. 8-32
/
.
500 2?
percent manganese, 1.97 percent chromium steel. Grain size: 6 to 7. Austenltlzed at 1600oF (From "Atlas of Isother mal Transformation Diagrams U.S. Steel Corporation.)
1 8oc
.
fe 400
"
.
1 30C
60C
1
The effect on the retardation of the critical cooling rate by coarsening the grain size of the austenite is shown in Fig 8-33. The numbers that indicate grain size are related to the number of grains in a unit area so that the higher the number the finer the grain size Notice the consider-
20C
r
.
Ms*
40C
M50*
-
,
-
!0C
20C
Man*
-
.
.Eslimaled
able shift of the curves to the right by the higher austenitizing temperature 0
and the resultant coarse austenitic grain size To summarize there are only two factors that will decrease the critical cooling rate or move the l-T diagram to the right One is increasing the
temperature 1 1 1 mill-1
.
m
1C
,
Fig. ff-33 l-T diagram of a 0.87 percent carb percent manganese, 0.27 percent vanadium :
.
amount of carbon and alloying elements added and the other is coarsen,
ing the austenitic grain size. While the addition of alloying elements does not affect the maximum hardness obtainable from the steel that property ,
;
being controlled by carbon content only they make it much easier to com,
pletely harden the steel. A plain carbon steel may have to be water-
1
m
'
Grain si e: 2 to 3, austenitized at 1925 Grain size: 11, austenitized at 1500oF,
r
of the l-T curve to the right by coarsening th
grain. {From US .
.
"
Atlas of Isothermal Transformi
Steel Corporation.)
I
THE HEAT TREATMENT OF STEEL
quenched to obtain full hardness, while the same steel when alloyed may attain the same hardness and may be hardened to a greater depth even
1
when cooled more slowly by oil quenchjng. The use of a slower cooling rate reduces the danger of distortion and cracking during heat treatment. While coarsening the austenitic grain size has an effect similar to that of adding alloying elements, the coarser grain will tend to reduce the tough-
4
Fernte and carbide
26
ness of the steel, therefore, if it is desired to reduce the critical cooling
28
rate, this may be done best by changing the chemical composition rather than by coarsening the austenitic grain. 813 Hardening or Austenitizing Temperature The recommended austenitizing temperature for hypoeutectoid steels is about SCPF above the upper-critical-temperature (A}) line. This is the same as the recommended annealir g temperature. At any temperature below the A3 line there will be some proeutectoid ferrite present which will remain after quenching, giving rise to soft spots and lower hardness. For plain-carbon hypereutectoid steels the recommended austenitizing
27
.
..
3)
iPJte, ferrile and carbide
m
r
35 44
Fernte and carbide 48
ung to this temperature
48
ung to this temperature
281
>
A
800 1400 Ae
15 30 I min min hour
10 I hour da
32
700 58
1200
[ i mill Llll
i ii 11 in
1000
10000
42
600
100000
seconds
38
F+C
1000
5:>
500
°
39 a>
a>
to
1 800
o
400 "
e
50
2f
.
r
600
300
SI
critical cooling rate by coarsening vn in Fig. 8-33. The numbers that lumber of grains in a unit area so le grain size. Notice the consider-
200
400
Ms*-
100
200
= 90*-
1-T
,
he higher austenitizing temperature [in size.
. Estimated
t
to the right. One is increasing the .
its added and the other is coarsen,
1
1
i
iiiiimI
10
LLU
I03
102
Grain size: 2 to 3 austenitized at 1925"F. ,
Grain size: 11 austenitized at 1500°F, shows the shift ,
of the l-T curve to the right by coarsening the austenitic grain. (From "Atlas of Isothermal Transformation Diagrams," U S Steel Corporation.) .
.
day
hour
mm
Fig. 8-33 l-T diagram of a 0.87 percent carbon, 0.30 percent manganese, 0.27 percent vanadium steels.
"'
)y, they make it much easier to comrbon steel may have to be water-
diagram
Time, seconds
Ma addition of alloying elements does
finable from the steel, that property
X
week
66
temperature
0
vectors that will decrease the critical ..
\
50%\
104
J
1
1 1 mil
10=
106
1
4 282
INTRODUCTION TO PHYSICAL METALLURGY
temperature is usually between the and A lines (Fig. 8-3); therefore undissolved carbides would t £ end to be present in the microstructure at 1 room temperature Figure 8-34 shows the microstructure of a hy
fast critical cooling rate, tenc
3i [ -
,
m
while those richer in carbon, I
.
toid steel, oil-quenched from between the A pereutec-| and Aj lines. Notice the! small white undissolved carbide spheroids clearly visible in the dark flnefl pearlite areas. They are much more difficult to see in the light marten cm
,
matrix.
hardness. This condition may formity is established by cart excessive time required by tl practical. A more suitable me ing temperature. At this temp formity will be established in
sitic
*
The Acm line rises so steeply that an excessively high temperature m aybe required to dissolve all the proeutectoid cementite in the austenite This tends to develop undesirable coarse austenitic grain si ze, with danger of cracking on cooling 8 -14 Homogeneity of Austenite This refers to the uniformity in carbon co ntent of the austenite grains If a hypoeutectoid steel is heated for h ardening e , line is crossed the austenite grains formed from pearlite will contain 0.8 percent carbon With ,vmi lAj.umuea continued neat heating mg, the austenite grains fnrmpH from nr- Qnt t' -' ormed f rom proeutectoid ferrite will contain very little carbon so that when the A3 Ime is crossed the austenite grains will not be uniform in carbon content. Upon quenching, the austenite grains leaner in carbon having a
I
'1
.
.
.
when th
form martensite. This results i
,
.
,
,
;
recommended that the materi
i< :|
for each inch of thickness or 8-15 Mechanism of Heat Removal C
S
and strength resulting from the actual cooling rating obtccooling rate exceeds the cril
.
,
If the actual cooling rate Is le not completely harden. The
,
,
Fine
Mortensite
Undissolved
pearlite
matrix
carbide
ing rates the softer will be tl hardness. At this point, it Is heat removal during quenchi To illustrate, a typical cool in warm water is shown in Fii
rate throughout the quench in mind the difference betwe
ing curve shows the variatio A cooling rate, however, sho
.5
The cooling rate at any ter curve by drawing a tangent
ing the slope of the tangen slower is the cooling rate.
9* I
on the cooling curve in Fl< constantly changing during Stage A-Vapor-blanket C( ture of the metal is so high surface of the metal and a tl
.At .
f i
Fig. 8-34
Cooling is by conduction ar vapor films are poor heat through this stage. Stage B-Vapor-transport
. L.
Microstructure of a high-carbon steel, oil-
quenched from between the A
c„
and A,
,
,
metal has cooled to a terr
lines. Fine
pearlite (dark) in a martensitic (light) matrix Etched in 2 percent nital 500X. Notice the small white undissolved .
,
carbide spheroids in the dark pearlitic areas.
:
i
stable. Wetting of the mete boiling occur. Heat is ren heat of vaporization. This
7
;
m and A3 ,
.
f
1 the >4c
m
and A, , lines.
,
form martensite. This results in a nonurwform microstructure with variable
Notice the
,
283
,
>|the microstructure of a hypereutec.
THE HEAT TREATMENT OF STEEL
fast critical cooling rate tend to transform to nonmartensitic structures, while those richer in carbon having a slower critical cooling rate, tend to
,
e present in the microstructure at
. .
>.V
lines (Fig. 8-3); therefore
.
hardness. This condition may be avoided by very slow heating so that uniformity is established by carbon diffusion during heating. However the
olds clearly visible in the dark fine
,
ifficult to see in the light martensitic Excessively high temperature may be
excessive time required by this method does not make it commercially practical. A more suitable method is to soak the material at the austeniti2ing temperature. At this temperature diffusion of carbon is rapid, and uni-
lid cementite in the austenite This
formity will be established in a short time.
jaustenitic grain size, with danger
recommended that the material be held at the austenitizing temperature
;
for each inch of thickness or diameter.
.
'
y
,
h
lite grains formed from pearlite will heating the austenite grains
Mechanism of Heat Removal During Quenching The structure hardness, and strength resulting from a heat-treating operation are determined by the actual cooling rating obtained by the quenching process. If the actual cooling rate exceeds the critical cooling rate, only martensite will result.
ntain very little carbon so that when ains will not be uniform in carbon
If the actual cooling rate is less than the critical cooling rate, the part will not completely harden. The greater the difference between the two cool-
8-15
ptoid steel is heated for hardening, ;inued
,
,
'
W te grains leaner in carbon
s
,
'
v&sjto the uniformity in carbon content
.
To be on the safe side it is
.-j
isite
,
having a
,
ing rates the softer will be the transformation products and the lower the hardness. At this point, it is necessary to understand the mechanism of heat removal during quenching.
Undissolved corbide
To illustrate a typical cooling curve for a small steel cylinder quenched in warm water is shown in Fig. 8-35. Instead of showing! a constant cooling rate throughout the quench, the cooling curve shows three stages. Keep in mind the difference between a cooling curve and a cooling rate. A cooling curve shows the variation of temperature with time during quenching. A cooling rate, however, shows the rate of change of temperature with time. The cooling rate at any temperature may be obtained from the cooling curve by drawing a tangent to the curve at that temperature and determining the slope of the tangent. The more nearly horizontal the tangent, the slower is the cooling rate. Visualizing tangents drawn at various points on the cooling curve in Fig. 8-35, it is apparent that the cooling rate is ,
Hii *
r
5
mm
constantly changing during cooling.
Stage A-Vapor-blanket Cooling State In this first stage, the temperature of the metal is so high that the quenching medium is vaporized at the surface of the metal and a thin stable film of vapor surrounds the hot metal.
Cooling is by conduction and radiation through the gaseous film, and since vapor films are poor heat conductors the cooling rate is relatively slow ,
ff
"
through this stage.
Stage B-Vapor-transport Cooling Stage This stage starts when the metal has cooled to a temperature at which the vapor film is no longer stable. Wetting of the metal surface by the quenching medium and violent boiling occur. Heat is removed from the metal very rapidly as the latent heat of vaporization. This is the fastest stage of cooling.
.
.
-i. -
!'
;..
284 INTRODUCTION TO PHYSICAL METALLURGY
900
i
H600
diameter stainless-steel bar the left is a 10 percent brin< medium has a very short v
800h
5s
Transition
700
between stages '
quickly into the boiling stage goes into the third stage at tap water at 750F, notice th
1200
-
60C
sfsoo i
g 400
800
brine. It drops into the boili
-
.
ai
E
ing this stage, while very ra third stage is reached after £
300
II
200
40C
100 4 0 v
fused salt. This is usually heated until it is liquid; the this case, the fused salt is i
I I c
J
o
1
2
3
4
5
6
r e 9 Time, seconds
10
II
12
13
14
15
16
short vapor stage, approxim ing rate during the boiling
Fig. 8-35 Typical cooling curve for a small cylinder
quenched in warm water. (Gulf Oil Corporation )
water, and it reaches the thii with oil, the dotted line be solid line slow oil at 1250F the difference being that Gi about 7 s, whereas the slo\
.
.
Stage C-Liquid Cooling Stage This stage starts
when the surface tem-
perature of the metal reaches the boiling point of the quenching
liquid.
Vapor no longer forms so cooling is by conduction and convection
The third stage is reached b
,
i
through the liquid
.
The rate of cooling is slowest in this stage
.
Many f actors determine the actual cooling rate. The most important are the type of quenching medium the temperature of the quenching medium ,
the surface condition of the part and the size and mass of the part
,
,
8-16
Quenching Medium
In view of the mechanism of heat removal
,
I
-
1600
1
I
.
the ideal
quenching medium would show a high initial cooling rate to avoid trans-
1400 «
.
formation in the nose region of the l-T diagram and then a slow cooling rate throughout the lower-temperature range to minimize distortion. Unfortunately there is no quenching medium that exhibits these ideal pro p erties. Water and water solutions of inorganic salts have high initial cool-
1200
-
,
1000
-
-
ing rates through the A and B stages but these high cooling rates persist
I I
Con-
I
,
to low temperatures where distortion and cracking tend to occur
.
ventional quenching oils have a longer A or vapor-blanket stage and a ,
shorter B stage with a slower rate of cooling
-
1 o
.
,
\\
1 \V
QJ
V
800
\ 600
.
\
The following industrial quenching media are listed in order of decreasing quenching severity:
i
1 Water solution of 10 percent sodium chloride (brine)
2
Tap water
3 4
Fused or liquid salts Soluble oil and water solutions
5
Oil
6
Air
\
400
\ X\
200
0 3
i
10
15
Fig. B-36 Center-cooling curves for stain The cooling curves obtained by different media in the center of a V2-in
specimens, V? in. diameter by 2,/2 in. long -
(Gull Oil Corporation.)
.
i
I !
i
THE HEAT TREATMENT OF STEEL
J
H600
285
diameter stainless-steel bar are shown in Fig, 8-36. The curve furthest to the left is a 10 percent brine solution at 750F. Notice that this quenching medium has a very short vapor stage lasting about 1 s and then drops quickly into the boiling stage, where the cooling rate is very rapid. It finally goes into the third stage at about 10 s. Looking at the cooling curve for tap water at 750F, ndtice that the vapor stage is slightly longer than for brine. It drops into the boiling stage after about 3 s. The cooling rate dur'
es -
f
1200
.
800
-
ing this stage while very rapid, is not quite so fast as that for brine. The third stage is reached after about 15 s. Now examine the cooling curve for fused salt. This is usually an inorganic low-melting-point salt which is heated until it is liquid; the liquid is then used as a quenching medium. In this case the fused salt is at 400oF. Notice that the fused salt has a very short vapor stage, approximately equal to that of brine. However, the cooling rate during the boiling stage is not so rapid as that for brine or tap water/and it reaches the third stage at about 10 s. The next two curves deal with oil, the dotted line being Gulf Super-Quench oil at 1250F and the solid line slow oil at 1250F. They both show a relatively long vapor stage, the difference being that Gulf Super-Quench enters the boiling stage after about 7 s, whereas the slow oil enters the boiling stage after about 13 s. The third stage is reached by Gulf Super-Quench after about 15 s and about ,
h400
-
c 10
,
2
13'
14
15
16
:i
1
-jstage starts when the surface temling point of the quenching liquid. is by conduction and convection is slowest in this stage .
)oling rate. The most important are perature of the quenching medium he size and mass of the part
I 1600
,
.
:
hanism of heat removal the ideal
1400
,
i initial cooling rate to avoid trans '
-
diagram and then a slow cooling range to minimize distortion
lium that exhibits these ideal
iKjbrganic salts have high initial W these high cooling rates
.
1200
1\
1
\
-
Un-
10% brine at 75 F
propcool-
1000
Ll.
persist
1 \\
300
3
a)
.
QJ
,
600
F °
F
Fused salt at 400 "F
\
"
Slaw ail at 125 F
\
joling.
°
Gull super-quench at 125 N
ZD
and cracking tend to occur Conr A, or vapor-blanket stage and a
Tap water at 75
Still air at 82 °F
\ \
ledia are listed in order of decreas-
V \ 400
p;)ride (brine)
200
0 0
5
10
15
20
25 Time,secands
Fig. 8-36
rent media in the center of a Vs-in
Center-cooling curves for stainless-steel
specimens Vj in. diameter by 2,h in. long. No agitation. ,
-
,
(Gulf Oil Corporation.)
5
-
i
30
35
40
45
50
28C
INTRODUCTION TO PHYSICAL METALLURGY
V
22 s for the slow oil. The final cooling curve for still air at 82UF never gets j
1600
out of the vapor stage and therefore shows a very slow cooling rate over the !i
entire range.
\
1
1400
The usual methods of comparing the quenching speed of different media J are by determining the rate of cooling at some fixed temperature or the
1200
average rate between two temperatures. Cooling rates of several media are given in Table 8-2. 8-17
iOOO
Temperature of Quenching Medium
Generally as the temperature of the ,
u
,
.
\
medium rises, the cooling rate decreases (Fig. 8-37; Table 8-2). This is due §
.
to the increase in persistence of the vapor-blanket stage. Since the medium is closer to its boiling point, less heat is required to form the vapor film. This is particularly true of water and brine. The figures for Gulf SuperQuench in Table 8-2, however, show an increase in- cooling rate with a rise in temperature of the medium, which seems contrary to the previous statement. In the case of oil, there are two opposing factors to be con-
800 p
\ .
600
m. M
\ \
400
sidered. As the temperature of the oil rises there is a tendency for the cool-
ing rate to decrease due to the persistence of the vapor film. However, as the temperature of the oil rises it also becomes more fluid, which increases
0
the rate of heat conduction through the liquid. What happens to the cool-
lor sP™nS
curves fig 8 37 Center thcooling temperatures of 75 tapv eratba -
m
ing rate is determined by which factor has the greatest influence. If the
20
IS
10
0
and 125
3n„at,on. (OuK Oil Corporalion.)
increase in the rate of heat conduction is greater than the decrease due to the persistence of Ihe vapor film, the net result will be an increase in the actual cooling rate as in the case of Gulf Super-Quench. However, if the reverse is true then the net result will be a decrease in actual cooling rate, as shown by the figures for slow oil. The optimum rates of cooling are obtained with conventional quenching oils at bath temperatures between 120 and 150oF. To prevent a temperature rise in the medium during quenching it is always necossary to provide sufficient volume of medium. In some ,
cases, cooling coils are inse
,
temperature of the medium. The cooling rate may
be irr
kept constant by circulation o
effectively wipes off the vap length of the vapor blanket s1 The quenching severity, relat for various conditions of que
,
-
TABLE 8-2
Cooling Rates at Center of Vi-in.-dlameter by aVj-in.-long Stainless-steel Speci-
men When Quenched from 1500 F in Various Media* '
RATE AT 1300oF,
RATE AT 1200
,
"
BATH
Brine (10%) Tap water Gulf Super-Quench Slow oil
,
1/
F/s
-
i
75"
125
75 o
125 o
75"
125
382
296
382
325
383
287
211
46
223
117
220
176
80
85
170
180
135
137
36
32
30
26
39
44
36
30
36
30
34
28
90% water
Fused salt (at 400oF)
3
4
5
162
in the quen
of circulation, to obtain a wi(
TABLE 8-3 Quenching Severity Ri Quench*
METHOD OF COOLING No circulation of liquid or agital Mild circulation or agitation Moderate circulation
10% soluble oil, Still air
times overlooked
AVERAGE RATE 1250-900
F/s
F/s
°
130
66
Good circulation
Strong circulation Violent circulation A. Grossmann. Principles of h "
Courtesy of Gulf Oil Corp.
m
m
'
From M
,
I
1
'
7
) curve for still air at 820F never gets
THE HEAT TREATMENT OF STEEL,
287
1600
M PWS a very slow coo|'ng rate over the
'
\ 1400
I quenching speed of different media
\
g at some fixed temperature or the s Cooling rates of several media are
1200
.
enerally, as the temperature of the 68 (Fig. 8-37; Table 8-2). This is due , ior-blanket stage Since the medium VV. is required to form the vapor film brine. The figures for Gulf Superan increase in- cooling rate with a tich seems contrary to the previous re two opposing factors to be conises there is a tendency for the cool-
1000
Tap water at 75 \
fc 800
"
F
Tap water at 125
.
:
O
.
.
600
f
400
200
' ence of the vapot- film. However, as '
'
.
0
lecomes more fluid which increases
i
5
,
,
e liquid. What happens to the coolhas the greatest influence. If the is greater than the decrease due to
10
15
- JL
-
-
1-
20 25 30 Time, seconds
_
35
L
_
40
50
45
Fig. 8-37 Center-cooling curves for specimens quenched in tap water at bath temperatures of 75 and 1250F agitation.
net result will be an increase in the
jiulf Super-Quench. However if the
.
No
(Gulf Oil Corporation.)
,
ae a decrease in actual cooling rate
cases, cooling coils are inserted in the quenching tank to control the
,
The optimum rates of cooling are l oils at bath temperatures between
temperature of the medium.
.
Je rise in the medium during quenchyifficient volume of medium. In some
I r jameter by 2V2-in.-long Stainless-steel Specijdia*
RATE AT 1200°F
,
T/s
for various conditions of quench. Circulation is a factor which is sometimes overlodked in the quenching process. It is possible, by proper choice of circulation, to obtain a wide variety of cooling rates with an oil quench.
AVERAGE RATE 1250-900
,
0F/s 75°
125 O
75'»
125°
382
325
383
287
223
117
220
176
170
180
137
30
26
135 39
The cooling rate may be improved and the temperature of the medium kept constant by circulation of the medium and agitation of the piece. This effectively wipes off the vapor film as quickly as it forms, reducing the length of the vapor-blanket stage, and results in faster cooling (Fig. 8-38). The quenching severity, relative to still water as 1.0, is shown in Table 8-3
TABLE 8-3 Quenching Severity Relative to Still Water As 1.0 for Various Conditions of Quench*
36
30
4
44
34
28
3
130
66
OIL
WATER
No circulation of liquid or agitation of piece Mild circulation or agitation
0 25-0.30
0 9-1 0
2
0 30-0.35
1 0-1 1
2-2 2
Moderate circulation
0 35-0.40
1 2-1 3
Good circulation
0 40-0.50
1 4-1 5
Strong circulation
0 50-0.80
1 6-2 0
Violent circulation
0 80-1.10
. From M
ti.,,
i
BRINE
METHOD OF COOLING
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
\
.
.
4
5
A. Grossmann, ' Principles of Heat Treatment," American Society for Metals, Metals Park, Ohio, 1953.
288 INTRODUCTION TO PHYSICAL METALLURGY
1600
Protective Atmosphere
1400 \ \
from the partial or complete as methane and propane in s|
\ 120C L \ \
i
ioooH
Nc ogltution Violent ogitotion
Liquid-salt Pots The part to salt furnace that is neutral wit
\
pletely surrounded by the ne
\
3
800h
scale.
\
Cast-iron Chips The part is Any oxygen entering the furn;
\ \
600 h
\
the steel.
\
8-19 Size and
400h
Mass
Since it is or
with the quenching medium, tant factor in determining th of the geometric shape of t Thin plates and small-diame
200K
C w
An a
may be introduced under pre purpose are hydrogen, dissoci
0
5
10
15
20
25
30
35
40
45
mass and therefore rapid cc the surface area of the ends circumference times the len the cross-sectional area tim(
50
Time seconds ,
Fig. 8-38 Effect of agitation on center-cooling curves of a stainless'-steel specimen quenched in conventional quenching oil. Oil temperature 125°F (Gulf Oil Corporation.) .
The ratio is Surfa
-
8-18
Surface Condition When steel is exposed to an oxidizing atmosphere b ecause of the presence of water vapor or oxygen in the furnace a layer of iron oxide called scale is formed Experiments have shown that a thin layer of scale has very little effect on the actual cooling rate but that a thick layer of scale (0 005 in. deep) retards the actual cooling rate (see Fig. 8-39). There is also the tendency for parts of the scale to peel off the surface when the piece is transferred from the furnace to the quench tank thus giving
,
where p = density.
.
,
.
.
loor
H(ifdness(R/C)
,
'
to a variation in cooling rate at different points on the surface The presence of scale need be considered only if the actual cooling rate is rise
surfoce
140C
center
.
very close to the critical cooling rate. More important since scale is softer than the hardened steel there will be a tendency for the scale to clo
12
g up
the grinding wheel during the finishing operations. In any case, the formation of scale is usually avoided in commercial heat treatment Many methods are used industrially to minimize the formation of scale . I The choice of method depends upon the part being heat-treated type of W furnace used availability of equipment and cost. Copper Plating A flash coating of only a few ten-thousandths of an inch of copper will protect the surface of a steel against the formation of scale This method is economical when copper-plating tanks are available in
time (sec.) Time in0-930(
1000
,
,
Half temperature
C
2
z
800
1.
I 60C i -
.
4 0C
.
,
20C
,
,
the plant.
5
0
-
I
15 20 25 30 3! Time, seconds
Fig. 8-39 Effect of scale on center coolir -
1095 steel specimens quenched in Gulf S temperature 125 F violent agitation. (Gull °
,
j-I A
10
THE HEAT TREATMENT OF STEEL
Protective Atmosphere
289
An atmosphere that is inert with respect to steel
may be introduced under pressure into the furnace. Gases used for this purpose are hydrogen, dissociated ammonia, and combusted gas resulting from the partial or complete combustion of hydrocarbon-fuel gases such as methane and propane in special generators. :5
No agitation
Liquid-salt Pots Th6 part to be heat-treated may be immersed in a liquidsalt furnace that is neutral with respect to the steel. The piece, being com-
-
Violent agitation
pletely surrounded by the neutral liquid salt, cannot be oxidized to form scale.
Cast-iron Chips The part is buried in a container having cast-iron chips. Any oxygen entering the furnace reacts with the cast iron before it reaches the steel. 8'19
Size and Mass Since it is only the surface of a part which is in contact with the quenching medium the ratio of surface area to mass is an important factor in determining the actual cooling rate. This ratio is a function of the geometric shape of the part and is smallest for a spherical part. Thin plates and small-diameter wires have a large ratio of surface area to mass and therefore rapid cooling rates. Consider a long cylinder so that the surface area of the ends is negligible. The surface area is equal to the ,
30 '
35
40
45
50
r
circumference times the length of the cylinder, and the mass is equal to the cross-sectional area times the length times the density of the material. The ratio is
)sed to an oxidizing atmosphere T or oxygen in the furnace a layer of ,
-
,
nDL
4
Mass
(77-/4)D2Lp
Dp
s where p = density.
meriments have shown that a thin
8 actual cooling rate
Surface area
but that a thick
1
je actual cooling rate (see Fig. 8-39). '
.
ie scale to peel off the surface when
1600
.
ce to the quench tank thus giving fferent points on the surface. The
1400
1 only if the actual cooling rate is
l?00
Half temperature
1000
Time IH0-930lsec.)
,
lore important since scale is softer
[
time (sec.) o
a tendency for the scale to clog up
800
operations. In any case the forma-
a
,
.
:
8 90 1 75 .
.
44.0
39.5 1 M0
2 98 .
\
o minimize the formation of scale
400 .
he part being heat-treated type of ,
/ and cost
scale
Heavy scale Light scale
.
600
nercial heat treatment.
Heauy
scale
53.5 49,0
center
,
....
Hardness (R/C) surface
Liqhl
.
a few ten-thousandths of an inch
teel against the formation of scale per-plating tanks are available in
f
200 0 0
5
10
15
20 25
30 35 40
45 50
Time, seconds
.
Fig. 8-39 Effect of scale on center-cooling curves of SAE 1095 steel specimens quenched in Gulf Super-Quench. Oil temperature 125°F, violent agitation. (Gulf Oil Corporation.)
.
mi 290
INTRODUCTION TO PHYSICAL METALLURGY
1600
From the above data we
I
14O0K -
r
\ 1200
i
Specimen 3A in diameter .
Specimen 1/2 in diameter
\
Specimen 'A in. diameter
\
\
\
800
ceived slow cooling, with a resul
I
\
1000
m
.
can con
surface of the Vs- and 1-in. piec this steel, so that a fully martensi hardness. The surface of the 2 cooling, and the structure is profc and a small amount of ferrite. T
r
\ 160C
S
500
1400 40C
Center 1200
200
Midway 1000 L
0 0
m
-
5
,
10
-
15
20
25 30 Time secands
1
-
35
-
L
-
40
-
L
45
50
Surface
300
,
Fig. 8-40 Effect of mass on center-cooling curves of staiiiless-steel specimens quenched in conventional quenching oil. Oil temperature 1250F (Gulf Oil Corporation.)
i
600
1
.
The calculation shows that the ratio is inversely proportional to diameter.
400
I:
200
If the diameter is increased, the ratio of surface area to mass decreases,
and the cooling rate decreases. In other words, with a fixed quenching medium a large piece will be cooled more slowly than a small piece (Fig. 8-40) As the diameter increases, the duration of the vapor-blanket stage increases. The vapor-transport stage is less distinct, and the transition from the vapor-transport stage to the last stage becomes more gradual. The rate of cooling in all three stages decreases sharply (Fig. 8-41). Let us now perform an experiment on a medium-carbon steel of about
3
H 1600
,
1400
.
m
45 percent carbon. A series of pieces ranging from 1/2 to 5 in. diameter are heated to the proper austenitizing temperature and quenched in water.
0
Center 1000
1 300
.
When the surface hardness was determined on these pieces, the following results were obtained: DIAMETER OF PIECE
SURFACE HARDNESS,
WATER-QUENCHED, IN.
ROCKWELL C
05
59
1
58
2
41
.
3
1200
i
I
35
4
30
5
24
Midway
I:
500 -
i i 1
400h Surface .
200
0
40 T
c
S 4i r 8
u s
20
0
nd
Suiting the Heat Tre water-quenched (From "
Joh," U.S. Steel Corporation.)
in
THE HEAT TREATMENT OF STEEL
291
From the above data we can conclude that the actual cooling rate at the
surface of the Vs- and 1-in. pieces exceeded the critical cooling rate for this steel, so that a fully martensitic structure was obtained with maximum hardness.
- Specimen 3/4 in. diameter -
-
Specimen '/2 in diameter
The surface of the 2- and 3-in. pieces received intermediate
cooling, and the structure is probably a mixture of martensite, fine pearliie
.J
and a small amount of ferrite. The surface of the 4- and 5-in. pieces re-
Specimen'A In.diameter
ceived slow cooling, with a resulting structure of pearlite and ferrite. As a
1600
1400 Center
1200
Midway
1000
&0
round water quenched
2
35
40
45
50
V
Surface
800
J
600
400
inversely proportional to diameter
u
200 .
.
f surface area to mass decreases
,
:
ier words, with a fixed quenching ore slowly than a small piece (Fig duration of the vapor-blanket stage
a)
f
_
a
.
:
is less distinct and the transition ,
L
J
_
i 1600 .
i2 1400 "
5
2
i
round water quenched
1200
aast stage becomes more gradual. .
Cente-
ecreases sharply (Fig. 8-41) h a medium-carbon steel of about 5 ranging from Vj to 5 in. diameter .
mperature and quenched in water
1000
800
Midway
.
ned on these pieces the following
600
,
400
Surface
RDNESS,
.
200
-
:
SV:
1
1
0
_
Fig. 8-41
L
_
20
0
40
60 80 Time, seconds linear scale
Cooling curves at the surface, midway on the
radius, and at the center of two different-sized bars when
water-quenched. (From Suiting the Heat Treatment to the Job," U.S. Steel Corporation.) "
7
>,
-
.
.
.
-
i
_
100
120
'
i
292
.S
r
INTRODUCTION TO PHYSICAL METALLURGY
\
-i
matter of fact, for the 5-in. piece, the amount of heat to be removed is| so large compared with the surface area available that the water quench is|
i
ineffective, and very nearly the same hardness would have been obtained
if that piece had been cooled In the furnace. The approximate relation of ] the actual cooling curves of the surface to the l-T diagram is shown inl
Fig. 842.
)
1470
Up to this point, the discussion has concerned itself only with the surface -: hardness. The surface being in actual contact with the quenching medium, was cooled most rapidly in quenching. The heat in the interior of the piece must be removed by conduction, through the body of the piece ;|| eventually reaching the surface and the quenching medium. Therefore the cooling rate in the interior is less than that at the surface. Figure 8-43 shows the time-temperature cooling curves at different positions in a 1-in diameter bar during a drastic quench. If such a variation in cooling rates exists across the radius of a bar during cooling, it is to be anticipated
G
F
,
3
2
E D
,
c
,
770 B A
-
.
that variations in hardness would be evident when the bars are cut and 70
4
Z
0
5
Tir
erature cooling curves at e
Fla 8-43 Time-temp Ae -
rAmerican Society tor Metals.Me.a.
i \
\
Ohio, 1955.)
A+F. 5
R/C 24
coarse pearlite +
A+F+C
femte
\
\
4
R/C 30
\
F+C
\
medium peorme
made on th J. . hardness surveys
+ ferrite
„
\
center during quenching. If i
\ Q
\
-
\
\
\
\
3
curves, thie surface will havt the center is still at 1470oF .
"
of about 870 between the si temperature difference will which may result in distortio °
\
V
\
-V-\R/C 59
_
have a considerable temperat
R/C 58
Marlensite
V R/C 41
I
will be discussed in greater (
R/C 35
in this chapter.
Marlensite+ peorlite
The results of a hardness
+ ferrite
face hardness was previous
m
may be called
a hardnes: at a glance since it shows
quenching. Hardenability hardness, and in no case he
Time
Fig. 8-42 Surtace-cooling curves, the final structure and hardness of different-sized rounds related to the l-T diagram of a 0.45 percent carbon steel.
steel 1
ii
Si. I
is said to have low ha
.
1
THE HEAT TREATMENT OF STEEL
293
g amount of heat to be removed ii
,
.jj ea available that the water quench iS Sr. hardness would have been obtained urnace. The approximate relation of
face to the l-T diagram is shown
X
in
AlBCDEF 147C 6
concerned itself only with the surface §
u
.
\
ual contact with the quenching me=3
snchiQg. The heat in the interior of Dtion
,
S Mthe quenching medium. Therefore '
'"
?
"
"
F
through the body of the piece, ,
lthan that at the surface. Figure 8-43
:
770
l
: curves at different positions in a lench. If such a variation in cooling during cooling it is to be anticipated ,
evident when the bars are cut and 7C
'
0
2
4
6
8
10
I?
14
16
Time, seconds Ae
E i
--
,
Fig. 8-43 Time-temperature cooling burves at different positions in a 1-in.-diameter bar quenched drastically in water. (From M. A. Grossmann, Principles of Heat Treatment," American Society for Metals, Metals Park, "
Ohio, 1955.)
R/C 24 coarse pearlite +
ferrite TR/C
30 medium pearlite + ferrite
hardness surveys made on the cross section. Notice that it is possible to have a considerable temperature difference between the surface and the
!
'
center during quenching. If a vertical line is drawn at 2 s intersecting the curves, the surface will have come down to approximately 600oF while the center is still at 1470CF. Therefore there is a temperature difference of about 870
°
between the surface and the center at the end of 2 s. This
temperature difference will give rise to stresses during heat treatment i
which may result in distortion and cracking of the piece. These stresses will be discussed in greater detail in the section on residual stresses later in this chapter.
The results of a hardness survey on the different diameters, whose sur:
:
3m
i
face hardness was previously mentioned, are shown in Fig. 8-44a. This may be called a hardness-penetration or hardness-traverse diagram, since it shows at a glance to what extent the steel has hardened during quenching. Hardenability is related to the depth of penetration of the hardness, and in no case has the hardness penetrated deeply, so that this steel is said to have low hardenability. As anticipated, it is seen that the
294
INTRODUCTION TO PHYSICAL METALLURGY
6C
size, the actual cooling rate n
Op
if
actual rate at the center of a p steel, the hardness at the cer but the actual cooling rate at t
2
u 5G
in Fig. 8-44b for a VHn. bar.
c2
IT
_
I
Increase in the hardenabili
71
2 1
may be accomplished by eith< 1
1 1
size.
c
Ik
m
2
a
-
c
8
Since increasing cooling ra
1
r
ing, the addition of alloying creasing hardenability. Fic diagrams, after water quenct
'
2 3
"
"
4
'
5 o
With the l-T curve fixed, inc
quenching medium or increasing
7
o
J 30
With the actual cooling rates f
curve to the right) by adding aid
40
vanadium steel of about the
c
20 2
m
:
'
3"
-
4'
SAE 1045
"
5
-
SAE 104S
50
SAE 614C L
10
DiQme,er
Diomeler
Fig. 8-44 Hardness-penetration or hardness-traverse curves for various sizes quenched in water (a) SAE 1045 steel; (b) SAE 6140 chromB-vanadium steel (From M A. Grossmann ,
.
.
o 40
,
"
Principles of Heat Treatment American Society for Metals, Metals Park 1 Ohio 1955.) "
,
,
,
"
1
£ 30
: I
hardness of the quenched piece is less as its
size increases and also that each piece is lower in hardness at the center than at the surface ,
.
A study of the curves of Fig 8-_44a shows an interesting situation
-
hardness of Rockwell C 30 was obtained (1) at the surface of a 4-in (2) at about Va in. under the surface of a 3-in
.
center of a 2-in
.
.
The
round round and (3) at almost the .
,
,
f
20
I
round. These three points are equivalent and have reached
the same hardness because the actual cooling rate was the same at each location. This leads to a very important conclusion: that for a steel of fixed composition and austenitic grain size regardless of the shape or size of the piece and the quenching conditions wherever the actual cooling rate is the same the hardness must be the same. The student should realize that the converse of this statement is not necessarily true. Wherever the hardness is the same in a steel of fixed composition and austenitic grain
5 5
*
10
,
,
,
MM
?" 2 -"
-
.
t
,,
Diameter
Fig. 8-45 Hardness-penetration or
hardm
nurves for various sizes quenched in oil. ( steel; (b) SAE 6140 chrome-vanadium ste Grossmann, Principles of Heat Treatmen "
,
m
Society for Metals, Metals Park, Ohio.
19E
,
THE HEAT TREATMENT OF STEEL
295
size, the actual cooling rate may or may not have been the same. If the actual rate at the center of a piece exceeds the critical cooling rate for the steel, the hardness at the center will be the same as that at the surface,
2
but the actual cooling rate at both locations will be different. This is shown in Fig. 8-44b for a Vj-in. bar. Increase in the hardenability or depth of penetration of the hardness may be accomplished by either of two methods;
i With the actual coollhg rates fixed, slow up the critical cooling rate (shift the l-T curve to the right) by addihg alloying elements or coarsening the austenitic grain size.
2 With the l-T curve fixed, increase the actual cooling rates by using a faster quenching medium or increasing circulation.
Since increasing cooling rates increase the danger of distortion or cracking, the addition of alloying elements is the more popular method of increasing hardenabitity. Figure 8-44/5 shows the hardness-penetration diagrams, after wat6r quenching, for different-size rounds of a chromium-
"
2 3 4 5
"
"
"
vanadium steel of about the same carbon content as Fig. 8-44a. The hard
-
1
SAE 1045 50
SAE 6140 Diameter
r
-
-
-I
"
I 40
1
2
c
"
-
I
-
1"
as its size increases and also that
J
,
center than at the surface
,
-
.
hows an interesting situation The j (1) at the surface of a 4-in round a 3-in. round and (3) at almost the
crfra-
3
.
" "
4
"
V- 5
.
,
,
its are equivalent and have reached
cooling rate was the same at each conclusion: that for a steel of fixed
SAE 6140
I
10
regardless of the shape or size of
Diameter
igs, wherever the actual cooling rate
Fig. 8-45
Hardness-penetration or hardness-traverse
curves for various sizes quenched In oil. (a) SAE 1045
Jtyisame. The student should realize
steel; (6) SAE 6140 chrome-vanadium steel. (From M. A.
not necessarily true. Wherever the
Grossmann, "Principles of Heat Treatment," American
d composition and austenitic grain
Society for Metals, Metals Park, Ohio, 1955.)
4
9is
Diameter
m
296
INTRODUCTION TO PHYSICAL METALLURGY
,
size has almost achieved a fully martensitic structure across its diameter. :9%
If the actual cooling rate is reduced by using an oil quench, the hardness .1 level of all pieces made of the plain-carbon steel will drop as shown by Fig. 8-45a. Even the chromium-vanadium steel (Fig. 8-45b) shows a drop in hardness level in an oil quench as compared with the water quench ,
-J
,
specimen (Fig.
During the entire discussion of cooling rates it was assumed that the thermal conductivity of all steels is the same. This is not technically true but the variation in thermal conductivity between different steels is so small compared with the other variables in the quenching process that it may be ,
specimen is removed
and two
at 'Ai-in. intervals from the qui curve of hardness values vs. c
hardenability curve is shown
ir
procedure may be obtained b
End Quench Test for Hardenal
,
.
Each location on the Jominy
,
;;.'.--v
,
nally to a depth of 0.015 in. Ro
,
:
.
ture receives the same rate of
considered constant with little error.
'Hardenability The usual method of purchasing steel is on the basis of chemical composition. This allows a considerable variation in the carbon and alloy content of the steel. For example, AISI 4340 steel has the following composition range: 0.38 to 0.43 percent C, 0.60 to 0.80 percent Mn 0.20 to 0.35 percent Si, 1.65 to 2.00 percent Ni, 0.70 to 0.90 percent Cr and 0 20 to 0.30 percent Mo. Let us determine the percent variation of each element within the stated limits. For example in the case of carbon, the difference tpetween 0.43 and 0.38 is 0.05. If we divide 0.05 by the average between the limits or 0.40 and express this as a percentage, it turns out to be 12.5 percent. Following" the same procedure for the other elements the percent variation turns out to be even greaterr28.7 percent for Mn, 53.8 percent for Si, 19.1 percent for Ni, 25 percent for Cr, and 40 percent for Mo. These figures illustrate that it is possible when expressed on a percentage basis to have a considerable variation in chemical composition. This variation in chemical composition within a particular grade will cause a variation in the critical cooling rate and in turn a variation in the response of the steel to heat treatment. Figure 8-46 is a schematic representation of what might happen in an actual case. Assume the l-T curve on the left shows the
The size
8-47)
to the bottom of the specimen, water are all standardized, so
,
8-20
hardenability test, or the doming The test has been standardize ing this test, a 1-in.-round spec proper austenitizing temperatur placed on a fixture where a jet c
but note that the Win. round still attained full hardness.
m
have a test that will most widely used method of de
necessary to
ness level of all sizes has been raised appreciably and notice that the 1-in.
represents a certain cooling all steels is assumed to be the!
position on the
,
test piece reg
which the test piece is made. S
of cooling rates
varying con
i "
Cooling curue JAJ 1
Elements 01 the low sidi
Elements or
the high sir
beginning of transformation of a steel with all the elements on the low side and the l T curve on the right shows the beginning of transformation with -
.
3e tr
all the elements on the high side. Superimposed on the diagram is a cooling curve for a steel part quenched under certain conditions.
If the ele-
2
ments are on the high side, this cooling curve will miss the nose of the l-T diagram, and the steel will attain full hardness.
However, if the elements
are on the low side, the cooling curve will hit the l-T curve above the nose, and the steel will not attain full hardness. Therefore, buying a steel according to chemical composition is no assurance that full hardness will be at-
tained under certain quenching conditions. Since strength is the prime factor in design, unless special properties gre desired, it would seem more economical to base the
"
nnaterial specification on the response to heat
treatment (hardenability) rather than chemical composition. It is therefore
{ -
.
v : -
.
.
.
iV-i-'j
s
Time
Fin 8-46 Schematic representation of the 6 variation in chemical composition for a g,ve
Cooling curve superimposed Icrmation curve.
on the begmm
THE HEAT TREATMENT OF STEEL
appreciably and notice that the 1-in. o i:.ensitic structure across its diameter
necessary to have a test that will predict the hardenability of the steel. The most widely used method of determining hardenability is the end-quench
0ip using an oil quench, the hardness carbon steel will drop as shown by
hardenability test, or the Jominy test. The test has been standardized by the ASTM, SAE, and AISI. In conduct-
dium steel (Fig. 8-45b) shows a drop
ing this test, a 1-in.-round specimen 4 in. long is heated uniformly to the
3 compared with the water quench
proper austenitizing temperature. It is then removed from the furnace and
,
, '
;
.
,
'
,
led full hardness.
oling rates it was assumed that the ,
e same. This is not technically true
;
i
,
y between different steels is so small ;he
quenching process that it may be
r
jurchasing steel is on the basis of
to the bottom of the specimen, and the temperature and circulation of the water are all standardized, so that every specimen quenched in this fixture receives the same rate of cooling. After 10 min on the fixture, the
at Vu-in. intervals from the quenched end. The results are expressed as a curve of hardness values vs. distance from the quenched end. A typical hardenability curve Is shown In Fig. 8-48. Details pertaining to the testing
considerable variation in the carbon mple, AISI 4340 steel has the followrcent C 0.60 to 0.80 percent Mn 0 20 k£?nt Ni' 0-70 t0 0-90 Percent Cr, and : !; ine the percent variation of each ele,
placed on a fixture where a jet of water impinges on the bottom face of the specimen (Fig, 8-47). The size of the orifice, the distance from the orifice
specimen is removed, and two parallel flat surfaces are: ground longitudinally to a depth of 0.015 in. Rockwell C scale hardness readings are taken
i
.
,
procedure may be obtained by referring to ASTM Designation
mple, in the case of carbon the difi?:-J'5. If we divide 0.05 by the average
Each location on the Jominy test piece, quenched in a standard manner,
,
> this as a percentage
,
represents a certain cooling rate, and since the thermal conductivity
of is the same for a given all steels is assumed to be the same, this cooling rate
it turns out to
position on the test piece regardless of the composition of the steel from which the test piece is made. Each specimen Is thus subjected to a series of cooling rates varying continuously from very rapid at the quenched
rocedure for the other elements the ,
reater: 28.7 percent for Mn 53.8 perand 40 percent for Mo. ,
xent for Cr
,
)le when expressed on a percentage n chemical composition. Thisvariai particular grade will cause a variajrn a variation in the response of the a schematic representation of what : ne the l-T curve on the left shows the it
/Coolmq curv
ilements on the low side
r ilements on
with all the elements on the low side
the
high side
ne beginning of transformation with
erimposed on the diagram is a cool-
Beginning of transformation
nder certain conditions. If the eleig curve will miss the nose of the l-T hardness. However if the elements
e
,
iJwill hit the l-T curve above the nose ss Therefore, buying a steel accord,
.
jurance that full hardness will be at-
jjons. Since strength is the prime f ac.
A255-48T
End Quench Test for Hardenability of Steel.
:
'
:
;197
Time
ss are desired it would seem more
Fig. 8-46 Schematic representation of the effect of
jcification on the response to heat
variation in chemical composition for a given grade of steel.
,
Cooling curve superimposed on the beginning of trans-
;hemical composition It is therefore
formation curve.
.
L
i
1
'
298 INTRODUCTION TO PHYSICAL METALLURGY
4;:
graph paper (Fig. 8-48). Noti
S
scale parallel to the distance actual cooling rate at 1300oF 8-4 Therefore, the end-quen .
ness varies with different act
Figure 8-49 shows the cor an alloy steel of the 8630 type at selected locations along t
f
.A
posed.
f
This serves to clarii
'
v
A S T M. END QUENCH TEST FOR HARDENABILITY .
.
.
OF STEEL (A 255-48T)
r
HEAT NO. | G5R| N
TYPE
35862
a
REMARKS' V;
Fig. 8-47 End-quench hardenability specimen b eing quenched. (Bethlehem Steel Company.) APPROXIMATE
CC
iol ml 9| °l pI 9l " M oM pK-!
end to very slow at the air-cooled end (Table 8 4). Since each dista alon nce
(Tirol- u3csjio
3-
-
J'fvJ
70r
g the quenched bar is equivalent to a certain actual cooling rate could just as
well plot Rockwell C hardness vs actual cooling rateyou as Rockwell C hardness vs distance. This is exactly what is done on the ASTM
65r
,
.
60 =
.
TABLE 8-4 Cooling Rateis at Distances from the Water-cooled Hardenability Test Bar DISTANCE FROM
COOLING RATE
QUENCHED END IN.
0F/s AT
,
V,
,
DISTANCE FROM QUENCHED END IN.
1300oF
,
End-quench
50E CD
COOLING RATE
<40P
0F/s AT
1300oF
35-
19.5
30F
49C
3/,6
30£
3A
195
13/
125
.
End of the Standard
I
56.0 4?
33 26
5/B
21.4
-
'5/
1
1V4
5
12.4 16
20h
11.0 15b
10.0
10b
70 .
1V2
t
14.C
7/„
77.0 -
U
16.3
5 1 .
I3/.
2
4 0
4
6
10
12
.
2
DISTANCE FROM OUEI
35 .
Fig. 8-48 1
m
. .
l
A typical end-quench hardenabil
THE HEAT TREATMENT OF STEEL
299
H 1
.
graph paper (Fig. 8-48). Notice that the upper part of the graph contains a scale parallel to the distance scale which has readings of the approximate actual cooling rate at 1300oF that coprespond to the values given in Table 8-4 Therefore, the end-quench hardenability curve really shows how hardness varies with different actual cooling rates for a particular steel Figure 8-49 shows the continuous cooling-transformation diagram for an alloy steel of the 8630 type on which cooling curves representing those at selected locations along the end-quench test bar have been superimposed. This serves to clarify the relationship between the end-quench
r
.
.
,
,
A S T M .
.
.
.
DATE : LABORATORY TYPE SPECIMEN. TEST NO.
END QUENCH TEST
FOR HARDENABILITY
OF STEEL (A 255-4BT) TYPE
HEAT NO.
GRAIN
C
SIZE
8
r
.
P
Mn 55
IS
.
Si
s
016
.
.
018 27 .
Cr
Ni 1
79
.
.
47
Mo .
23
Cu .
NORMAL
TEMP°F
14
QUENCH
TEMPT 1700
REMARKS:
APPROXIMATE
COOLING RATE
"F PER SECOND AT I300F
q
15
rO
end (Table 8-4). Since each distance
70
ent to a certain actual cooling rate, you C hardness vs. actual cooling rate as
65
60
This is exactly what is done on the ASTM _
i 55-
3 CO
s Water-cooled End of the Standard End-quench
QUENCHED END, IN.
COOLING RATE, °
F/s AT
1300
n/l6
19.5
3/4
16.3
'
14.0
Ve
12.4
3/l6
-
S £U5E
v
DISTANCE FROM
= 50
%
40 C J
If
_
35E
uj 30 = o 25 20
11.0
15b
10.0 -
70
10 =
5 1
5
1 1V4 1V2 13/4
40
2
35
.
.
2
4
6
8
10
12
14
18
16
20
22
24
26
28
30
32
34
36
.
DISTANCE FROM QUENCHED END OF SPECIMEN IN SIXTEENTHS OF INCH
.
Fig. 8-48
A typical end-quench hardenability curve
.
38
40
:
J:: >. ; '
*
;; r
300 INTRODUCTION TO PHYSICAL METALLURGY
End-quench hardenabilify fesl
5oL®
Legend -
tures of ferrite, bainite, and m
-- Cooling Ironsf diagram Isothermal transf diagram Cooling curve .
.
3
SOW! tn
n"" Tronsf
.
2
130
during cooling
f
-
1®
20 3
1500
35
I
.
.
,
1
'S A
3
®
©
1
1200 \
1100 U
\
\
\
_
1000
\
\
tained for a considerable d
\
\
3
900
steels 4140 and 5140 the hare
\
Austenite
800 h
bainite
By the analysis of data col
\
of steel, the AISI has establ
\
700
cent carbon but of different h
steels develop the same ma: water-quenched end, since tt only. However, in the high-hf
\
\
and bainite regions, thus pern rates of cooling. The end-quench curves of
-
end, inches
1400
1300 h
Although hardenability is us it is the changes in miprost which are of importance in ti
general, increase hardenabili
2-5
Jislance from quenched
ing with the decrease in cool cooling rate is necessary to ol
curves known as hardenabi 500
Austenite
shown in Fig. 8-51. The suffi basis of a hardenability spe size, etc., being of secondary Two points are usually des the following methods:
marte nsite
500 \
400 H
®
®
\
\
\
I
©
A
The minimum and maximum
tance selected should be that ' 4', Gb
0 Marlensite
Martensite
Morfensite
ferrile a bainite
ferrile a bainite
1
i
31
-
i
Martensite
55
ferrile a boinlle
I
50
'
0
.
100 fime
,
seconds
a:
1000
10,000
45
40
re
30 25 20
end
.
exceeds the c ft Cal
coo
n
V ?86?
rate neareSt the
15
0
2
4
6
8 10 12 1 Distance from
Fig. 8-50 End-quench hardenability curves samples of 4340, 4140, and 5140 alloy steel;
Mi
-
:
?
-
:
-
Yhe heat treatment of steel
1
.
y test
Cooling transf. diagram Isothermal transf. diagram Cooling
3 I
tures of ferrite, bainite, and martensite, the amount of martensite decreas-
Legend
l
Miiiniiiiiiii
Transf
.
ing with the decrease in cooling rate. It should be noted that a very slow
cooling rate is necessary to obtain pearlrte in this steel. ,
curve
Although hardenabillty is usually expressed in terms of hardness changes,
during cooling
3
jend,
30t
inches
:
7
.
it is the changes in miprostructure reflected by those hardness values which are of importance in the properties of steel. Alloying elements, in general, increase hardenabillty by delaying transformation in the pearlite and bainite regions, thus permitting the formation of martensite with slower rates of cooling. The end-quench curves of three alloy steels, each containing 0.40 percent carbon but of different hardenabillty, are shown in Fig. 8-50. All three steels develop the same maximum hardness of Rockwell C 52.5 at the water-quenched end since this is primarily a function of carbon content only. However, in the high-hardenability steel 4340, this hardness is maintained for a considerable distance whereas in the lower-hardenability steels 4140 and 5140 the hardness drops off almost immediately. By the analysis of data collected from hundreds of heats of each grade of steel, the AISI has established minimum and maximum hardenabillty curves known as hardenabillty bands. A typical hardenabillty band is shown in Fig. 8-51. The suffix H denotes steels that may be bought on the basis of a hardenabillty specification, with chemical composition, grain size, etc., being of secondary importance. ,
\
\
,
Samite
m
,
e
,
\
Two points are usually designated in specifying hardenabillty by one of the following methods: ®
A The minimum and maximum hardness values at any desired distance. The distance selected should be that distance on the end-quench tpst bar which correi
.
i
65
60
Martensite
J
i
100
i
55
Martensite
ferrite fi bainite
L 1000
ferrite 9 bainite I i i
S 50
__
10,000
15
r
e, seconds
4340 "
!
40
I 35 or
30
4140
25
on behavior previously discussed
s on cooling at these various rates
.
20 -
,
15
ents the rate nearest the quenched ind will result in transformation to It in transformation to Various mix-
0
2
4
6
8
10
12
14
16
18
20
22
24
Distance from quenched end, 4 of on inch Fig. 8-50
End-quench hardenabillty curves for individual
samples of 4340, 4140 and 5140 alloy steels. ,
26
28
30
5140
"
32
302
INTRODUCTION TO PHYSICAL METALLURGY
i '
"
j
c
0
distance
V'S"
-
Mjn
50
53
50 60 59
53 52 51
6
59 58
7 B
58 57
51 50 48
9
57
4 4
0
56
42
56
40 39
2 4
5
12
55
13 14
55 54
whether that steel will meet t
Mo
37A0 44
-
Max
steel whose hardenability cun,
4140 H
Hardenobility bono
Hardness limits for
specification purposes
0 65 .
1 10 .
.
a2%035
0 75 .
A
.
0
the problem it is first necessai
0 25 .
the center of a 2-in. diameter conditions, or the distance a -
that has the same cooling
4 7
cooled end of the end-quench rate as the center of a 2-in. n
4
60 _
Referring to Fig. 8-50, a vertic for 4140 steel as shown. The
38 37 36
1 5
54
16
63
35
18
52
34
20 22 24
5 49 48
33 33 32
26 28
47
32
46
3
30
45
31
32
44
30
50
Since the required hardness that requirement under these
a;
4
40
ing medium were changed
1
recommended by SAE >
20 4
For forged or ralled specimens only
::r:
to int
Returning to Fig. 8-52, po from the water-cooled end. I tersecting the 4140 curve she
30
Heat treating temperatures .Normalize 1600 °F Austcnitize 1550 °F
rate
eter and H = 0.35, point X is Ic
70
8
12
16
20
24
28
C 45. Therefore, going to a v
32
Distance from quenched end, 1/16"
Fig. 8-51 Test data and standard hardenability band for 4140H steel. (American Iron and Steel Institute.) r
6 0h .
spends to the section used by the purchaser. For example, in Fig. 8-51, the specification could be J50/58 = 6/i6 in.
-
50 ,
B The minimum and maximum distances at which any desired hardness value occurs. This method of specification, in Fig. 8-51, could be J50 = 6/i6 to 5Vi6 in. 8-21
3/4
« 30 .
5
20 ,
§
point on the end-quench test bar that has the same cooling rate. The rela-
1
10
tion between the end-quench test bar and the center and mid-radius locations of various sizes of rounds quenched under different conditions
m
1/2
£4.0
different locations under various quenching conditions or the equivalent
is shown in Figs. 8-52 and 8-53. The severity of quench is designated by
i 1-
01
Use of Hardenability Data To select a steel to meet a minimum hardness at a given location in a part quenched under given conditions, the cooling rate at the given location must first be known and the reference point on the end-quench test bar having the same cooling rate must be determined. If the part is simple in cross section, such as round, flat, or square, numerous charts are available in the literature which give the cooling rate at
Table 8-5. Let us consider a practical application of the end-quench hardenabilityi test. Assume that a company is required to manufacture a steel shaft 2 in. in diameter to a specified minimum hardness at the center after hardening of Rockwell C 42. They plan to use a good oil quench and moderate agitation (H = 0.35, Table 8-5). They would like to use a bar of 4140
r-
,
§
Distance frc
Fiq 8 52 Locaiion on the end-quench hare * ba( corresponding to the center of round US vanous quenching conditions. (From .
Steels," U S .
.
Steel Corporation.)
..
-
'
Hardenability band 4140 H ;
i ,o
-
,
SI 0 20/ 0 35 /o.
Cr
a75X1 20
.
.
1
1
Mo .
/O. 0 25
THE HEAT TREATMENT OF STEEL
303
steel whose hardenability curve is Fig. 8-50. The problem is to determine whether that steel will meet the above specifications. In order to solve
.
the problem it is first necessary to know'what the actual cooling rate is at the center of a 2-in.-diameter round when it is quenched under the given conditions, or the distance along the end-quench hardenability test bar
that has the same cooling rate. Referring to Fig. 8-52, for a 2-in. bar diameter and H = 0.35, point X Is located. Therefore, 3/a or "/u from the watercooled end of the end-quench hardenability test bar has the same cooling rate as the center of a 2-in. round quenched under the given conditions.
Referring to Fig. 8-50, a vertical line is drawn at 12/i6 intersecting the curve for 4140 steel as shown. The hardness, read to the left, is Rockwell C 37.'
Since the required hardness was Rockwell 42, this steel will not satisfy that requirement under these quenching conditions. Suppose the quenching medium were chahged to water with no agitation (H = 1.0, Table 8.5), Returning to Fig. g-52, point Y is located, which gives a distance of 7/i6 from the water-cooled end. Drawing a vertical line at 7/i6 in Fig. 8-50 in-
a
\ 8
-
12
16
20
24
28
>
32
Distance from quenched end, i/16"
tersecting the 4140 curve shows that the hardness now will be Rockwell
C 45. Therefore, going to a water quench will meet the hardness require-
00 .
r
50 20 1 5 IO 0 70
60
.
.
.
.
iser For example, in Fig. 8-51, the speci-
.
.
y*-ff->
50 .
iss at which any desired hardness value : ig. 8-51, could be J50 = to 21/16 in.
0 50 .
0 35
o
.
.
E4 0
a steel to meet a minimum hardness
.
Jd under given conditions, the cooling De known and the reference point on jme cooling rate must be determined.
0 20 .
« 30 .
3
CD
1,
such as round, flat, or square, nu-
srature which give the cooling rate at
pnching conditions or the equivalent 1 has the same cooling rate. The relair and the center and mid-radius loca1
1
o
20 .
y
.
I
Round ba rs
io .
o
juenched under different conditions ..
e severity of quench is designated by ;il application of the end-quench hardiny is required to manufacture a steel minimum hardness at the center after
an to use a good oil quench and modI They would like to use a bar of 4140
2 Distance from water-cooled end inches ,
Fig. 8-52 Location on the end-quench hardenability test bar corresponding to the center of round bars under various quenching conditions. (From "U S S Carilloy Steels," U.S. Steel Corporation ) .
.
"
1
.
.
S
m 304 INTRODUCTION TO PHYSICAL METALLURGY
90 .
Examination of published
=0 5 .
.
70 ,
50 20 1 5
the selection of several steels
1
show the greatest overall ecoi The typical U curves (Fig. 8-
.
selection will be based on c
,
.
60
.
0
,
0 70 .
quenched under given condi hardenability band (Fig. 8'5'i)
05
g 5.0
.
0 35
8-52 and 8-53
.
5 4.0
.
The calculate
mum hardness variation acres
i
It is possible to show the
0? ,
hardness at the center locatic
m 3.0 1
quenched under the same C'
QJ
ability band (Fig. 8-51) and th at other locations by curves j
U
20
s g
.
Round
Dors
The approximate cooling
10 .
O)
any location in an irregularly ability curve for that steel is a
(SI
5 s CO
.-r
C 40 was obtained at a pan
Oistonce from woter-cooled end Fig. 8-53 Location on the end-quench hardenability test bar corresponding ,o the mid-radius positio bars under vanous n of round quenching conditions (From "U S S
2
'
Canlloy Steels
,
whose hardenability curve is
inches
at a distance of ,0/i6 on the
cooling rate t that location When steel is purchased
.
"
,
U.S. Steel Corporation ) .
the purchassr is certain that . .
J ,
'
8-22
ties after heat treatment.
Tl
/ and greater economy. Tempering In the as-quench for most applications.
The
sidual stresses in the steel.
TABLE 8 5 H or Se
verity Quenching Conditions*
H VALUE 0 20 .
0 35 .
0 50 .
0 70 .
1 00 .
m
1 50 .
2 00 .
5 00 .
CO
'
of Quench Values for Various 0 8
QUENCHING
CONDITION
Poor oil quench-no agitation Good oil quench-moderate agitation Very good oil quench-good agitation Strong oil quench-violent agitation Poor water quench-no agitation Very good water quench-strong agitation
Brine quench-no agitation Brine quench-violent agitation Ideal quench
From .US.S. Carilloy Steels." U.S Steel Corporati .
by tempering or drawing.jy perature below the lower ing is to relieve residual stre ~
ofThe stiel. This increase the hardness or strength. In general, over the brof decreases and toughness creased.
This is true if toi
tensile test.
However, this
Izod or Charpy is used as show a decrease in notchi
and 800oF, even though tl
strength. The reason fortt" The variation of hardness i
on.
perature shown in Fig. 8-5
m
a i
THE HEAT TREATMENT OF STEEL
305
r
Examination of published minimum hardenability limits will result in the selection of several steels to meet the original requirements. The final selection will be based on other manufacturing requirements and will show the greatest overall economy. The typical U curves (Fig. 8-44) or hardness-penetration curves of rounds quenched under given conditions may be calculated from the standard
50 .
W 7 K!
20 5 .
:
10 .
0 70 .
05
hardenability band (Fig. 8-51) and a series of curves such as those in Figs.
.
8-52 and 8-53
0 55
.
.
The calculated results will show the maximum and mini-
mum hardness variation across the cross section of various sizes of rounds.
It is possible to show the relation between the minimum as-quenched 02
hardness at the center location and the diameter of different-sized rounds
.
quenched under the same conditions by means of the standard harden-
ability band (Fig. 8-51) and the curves of Fig. 8-52, The same may be done at other locations by curves similar to Fig. 8-53. The approximate cooling rate, under fixed quenching conditions, at any location in an irregularly shaped part may be determined if the hardenability curve for that steel is available. Assume that a hardness of Rockwell C 40 was obtained at a particular location in a part made of 4140 steel whose hardenability curve is shown in Fig. 8-50. Rockwell C 40 is obtained at a distance of 10/i5 on the end-quench test bar, and Table 8-4 gives the cooling rate at that location as 21.40F/s. When steel is purchased on the basis of a hardenability specification, the purchaser is certain that he will obtain the desired mechanical proper-
CD
°
dots
c
O 1
t 9 -
2 end,inches :
9St
S
.
ties after heat treatment. This results in fewer rejections or retreatments
/ and greater economy. 8-22 Tempering In the as-quenched martensitic condition, the steel is too brittle
ter quench cannot be used. Then there teel of higher hardenability. Reference /vill certainly meet the original require-
for most applications.
The formation of martensite also leaves high re-
sidual stresses in the steel. Therefore, hardening is almost always fnllowftd.
ill C 52 is probably too high.
by tempering or dewingtjftdlix±LXjQQslslsJii±
thQ steol to some4 m--
'
p
erature below the lower criticaltemgerature. The purpose of temper-
ing is to relieyeTesidual stresseslind to improve the ductility and tounhness
Quench Values for Various
of the steel. This increase in ductility is usually attained at the sacrifice of the hardness strength. ' ""
""
I G CONDITION
In general over the broad range of tempering temperatures, hardness decreases and toughness increases as the tempering temperature is increased. This is true if toughness is measured by reduction of area in a tensile test. However this is not entirely true if the notched bar such as Izod or Charpy is used as a measure of toughness. Most steels actually show a decrease in notched-bar toughness when tempered between 400 and SOOT even though the piece at the same time loses hardness and strength. The reason for this decrease in toughness is not fully understood. The variation of hardness and notched-bar toughness With tempering temperature shown in Fig. 8-54 is typical of plain-carbon and low-alloy steels. ,
lench-no agitation uench-moderate agitation K::::
:
oil quench-good agitation quench-violent agitation
| quench-no agitation
'
;:;
v]water quench-strong agitation
V?";.|ch-no agitation
ich-violent agitation Ich 3
'
> >;
-
i
.
Steel Corporation.
V
,
5
,
1
306
r
INTRODUCTION TO PHYSICAL METALLURGY
.
AISI100
'
Rd. size treoted
AC,
1395 °F
Ar
"
Rd. size tested
AC3
I450°F
Ar,
0 530
80
.
Heot tested
Hardness
60
Brinell
311
EC
40
27
44
Shore
B89
33
Rockwell
72
3
531
578 78
72
C57
C53
I zed
280000
Toughness
1
t3 3
3
.
0 505
5
s
(s
i
270000
20
260000 Oi 0
m
200 !
400
600
800 1000 1200 1400
250000
Tempering temperature, T 240000
1
Fig. 8-54 Hardness and notched-bar toughness of 4140 steel after tempering 1 h at various temperatures. (From Suiting the Heat Treatment to the Job," U.S. Steel
230000
"
A
Corporation.)
220000 210000 200000
The tempering range of 400 to 800l,F is a dividing line between applications that require high hardness and those requiring high toughness. If
190000 180000
the principal desired property is hardness or wear resistance, the part is
o
. .
170000
tempered below 400oF; if the primary requirement is toughness, the part
is tempered above 800oF. If the part does not have any "stress raisers" or notches, the change in ductility may be a better indication of toughness than the notched-bar test, and tempering in the range of 400 to 800oF may not be detrimental. The effect of tempering temperature on the mechanical properties of a low-alloy steel 4140 is shown in Fig. 8-55. Residual stresses are relieved to a large extent when the tempering temperature reaches 400oF, and by 900oF they are almost completely gone. Certain alloy steels exhibit a phenomenon known as temper brittleness, which is a loss of notched-bar toughness when tempered in the range of 1000 to 1250oF followed by relatively slow cooling. Toughness is maintained, however, if the part is quenched in water from the tempering temperature. The precise mechanism which causes temper brittleness has not been established, although the behavior suggests some phase which precipitates along the grain boundaries during slow cooling. High manganese, phosphorus, and chromium appear to promote susceptibility, while molybdenum seems to have a definite retarding effect. Martensite, as defined previously, is a supersaturated solid solution of
P
.
CD
160000 -
150000 140000 130000 120000 110000
i
100000
90000
,
carbon'trapped in a body-centered tetragonal structure. This is a metastable condition, and as energy is applied by tempering, the carbon will be
precipitated as carbide and thejj nwJIIbeeo
b.c.c. There will be diffu-
sion and coalescence of the carbide as the tempering temperature is raised. r
v;.i
80000 70000
60000
50000 A Annealed
1500 "FX.
Draw
Normalized leOO-A.C,
400
500
Normalized
Fig 8-55 Mechanical properties of 4140 st€ quenching and tempering at various temper
(Bethlehem Steel Company.)
YHE HSAT TREATMENT OF STEEL
307
AISI-4140 properlies chart
(single heal results)
5 0
.
530" Rd. size treated "
0 506 ,
Rd. size tested
Ac,
C
1395 °F
AC3 1450 "F
Mn
0 41 .
27
44
B89
C33
0.85 429
578
Shore Rockwell
78
72
§1
S3
59
C57
C53
C50
C47
C45
495
461
Ni
Or
Mo
Grain size
0.0J4 0031 0.2 388 54
0 12
2i
.
311
341
1.01
0.24
277
6-8
235
48
44
39
34
C36
C33
C29
C22
48
69
83
108
.
Izcd
9
Si
.
Brinell
534
S
.
Heal tested
311
P
Arj 1330 'F OZB/Ofy Minn. Ma« 020/ 0.80/ 0.15/ 1280"F 43 4.00 0.04 0.04 -6,35 Nil aiO -0 25
Ar,
21
Ml
51
280000
4140
270000
As quenched1 Brinell hardness 601 260000
250000
P
240000
230000 220000 V
210000
Tensile strength
200000
;F
is a dividing line between applicathose requiring high toughness. If Iness or wear resistance, the part is requirement is toughness, the part t does not have any "stress raisers"
190000 180000 o
'
170000 160000
jy be a better indication of toughness
150000
oering In the range of 400 to 800oF if tempering temperature on the me-
Yield point I
-
,
140000
3l 4140 is shown In Fig. 8-55. a large extent when the tempering 0oF they are almost completely gone.
130000
50%
110000
Jmenon known as temper brittleness,
Reduction
mess when tempered in the range of slow cooling. Toughness is mained in water from the tempering temhich causes temper brittleness has ehavior suggests some phase which '
Q
100000
50 %
90000
40%
80000
30 %
-
20 %
Elongation
70000
3s during slow cooling. High mangapear to promote susceptibility, while "
70%
120000
10 %
60000
s retarding effect.
"'"
Annealed
I500°F.C.|
is a supersaturated solid solution of
Normalized
gp|tragonal structure. This is a metastaled by tempering the carbon will be
J
vill become b.c.c. There will be diffu-
i
I600°A,C
,
50000 Draw
400
500
600
700
800
900
1000
1100
Normolized ot 1500 °F, reheated to IS'iO0
quenched in agitated oil
Fig. 8-55 Mechanical properties of 4140 steel after oil quenching and tempering at various temperatures. (Bethlehem Steel Company.)
s the tempering temperature is raised.
1200
Tempering temperature, °F
t :
i
!.
1300
308 INTRODUCTION TO PHYSICAL METALLURGY
has high strength, high hardnes 1 04
many of the residual stresses
.
Heating In the range from 450 orthorhombic cementite (FesC), 1
02
5
I OC
"
.
are
0
200
400
600
800
Tempormq temperalure
ferrite, and any retained austen carbides are too small to be res
1000
entire structure etches rapidly
"F
,
Fig. 8-56 The axial ratio c/a of martensite as a function of tempering temperature When c/a = 1.00 the martensite has decomposed into ferrite and a carbide phase (From
(Fig. 8-57). If the sample is magr scope, the carbide precipitate
.
,
i
.
Brick, Gordon Alloys," 3d ed 1965.)
m
come out along the original mai the tensile strength has droppei
and Phillips, "Structure and Properties of
,
.
,
McGraw-Hill Book Company New York. ,
ductility has increased slightly, I: decreased to between Rockwell
When plain carbon steel is heated In the range of
100 to 400oF
,
temperature.
thestruc-W
ture etches dark and is sometimes known as black martensite The originalm
Tempering in the range of 7
.
as-quenched martensite is beginning to lose its tetragonal crystal structure* by the formation of a hexagonal close packed transition carbide (epsilon
cementite particles. This coale of the ferrite matrix to be seen, lower-temperature product. In the carbide is just about resol in the electron micrograph (Fig are: tensile strength 125,000 2 in., hardness Rockwell C 20-
-
carbide) and low-carbon martensite X-ray studies Fig. 8-56, show the!' decrease in c/a ratio as carbon is precipitated from martensite forming epsilon carbide The precipitation of the transition carbide may cause a» slight increase in hardness particularly in high-carbon steels The steel .
,
.
.
-.
v
..
]
.
'
.
-
,
.
toughness, as shown by Fig. 8
Unfempered
Tempered
mortenslte
mortensite
ifaf
wmmi
v
.
t
St
i
mm '
-
'
T
m ft
/r
i
X
:
4 4
r
,
6
i
,
r A
Fig. 8-57 1045 steel water-quenched and tempered at for 1 h. Tempered martensite (dark) and untempered martensite (light) Etched in 2 percent nital 500X, .
,
!
.:
3
600oF
..
Fig. 8-58 Same sample as Fig. 8 57, -
microscope, 9.000X.
\ ft.
it
r
taken w
1
J
THE HEAT TREATMENT OF STEEL 309 has high strength, high hardness, low ductility, and low toughness, and many of the residual stresses are relieved.
Heating in the range from 450 to 750oF'changes the epsilon carbide to orthorhombic cementite (FeaC) the low-carbon martensite becomes b.c.c. ferrite, and any retained austenite is transformed to lower bainite. The carbides are too small to be resolved by the optical microscope, and the entire structure etches rapidly to a black mass formerly called troostite (Fig. 8-57). If the sample is magnified 9,000 times using the electron microscope, the carbide precipitate is clearly seen. Some of the carbide has ,
in
isite
come out along the original martensitic plate directions (Fig. 8-58). While
Of
It
the tensile strength has dropped, it is stjll very high, over 200,000 psi. The ductility has increased slightly but the toughness is still low. Hardness has decreased to between Rockwell C40 and 60 depending upon the tempering.
"
i
,
in the range of 100 to 400
the struc- % ; 3 to lose its tetragonal crystal structure
temperature.
,
lown as Jb/ac/( martensite The original
Tempering in the range of 750 to 1200oF continues the growth of the
jse-packed transnion carbide (epsilon
cementite particles. This coalescence of the carbide particjes allows more of the ferrite matrix to be seen, causing the sample to etch lighter than the
.
;
X-ray studies, Fig. 8-56
"
"
-
show the j f precipitated from martensite forming *
lower-temperature product. In this structure, formerly known as sorbite, the carbide is just about resolvable at 500x (Fig. 8-59) and is clearly seen in the electron micrograph (Fig. 8-60). Mechanical properties in this range are: tensile strength 125,000-200,000 psi, elongation 10-20 percent in 2 in. hardness Rockwell C 20-40. Most significant is the rapid increase in toughness, as shown by Fig. 8-54.
,
fpi the transition carbide may cause a
'
-
:
arly in high-carbon steels
.
The steel |
,
Jntempered
Tempered
nortensite
martensite
mm
r
3 mm it 6
3
m
,
v
.
tic
0
,
f
is b
m .
.. jr
on
\
.
r.
00°F
Fig. 8-58
Same sample as Fig. 8-57, taken with the electron
microscope, 9,000X.
i
I
310 INTRODUCTION TO PHYSICAL METALLURGY
For many years, metallurgists m
stages. The microstructure apf troostite and sorbite. However.
ual that it is more realistic to a
ture simply tempered martensit and martensite are summarize In the above discussion, time
stant. Since tempering is a p temperature are factors. The
shorter time at a higher temp temperature. Figure 8-62 sho peratures for a eutectoid ste 1
occurs in the first few minute;
results from increasing the tir It is important to realize the in order to compare different same hardness or strength le If a medium tensile strengtl first to form a fully martensit substantially in tempering, v
Fig. 8-59 1045 steel wafer-quenched and tempered at 1150oF for 1 h Precipitated carbide particles in a ferrite matrix Etched in 2 percent nital 500X, .
.
,
Heating in the range from 1200 to 13330F produces
mentite particle .
large
spherbidized cementite
,
structure obtained directly from austenit e
spheroidized anneal (see Fig
,
8-4).
tained, with less difficulty in
globular ce- {
This structure is very soft and tough a nd is similar to th
s
e1
by a |
coo
Ajstemte or
7 'o°/iS'
Coorse pearlite /
R/c 15 s
.4
'
Medium pearlite
It
R/c 30
11
v
ff / S
5
"
s-
s
o/
0
£/
o
_
in CO
§
o
..
5
v
Fine pearlite
o
R/c 40 ai Q>
n
O
7?
| = 900-400°F § ho,d
Boin,te
H/c 40-60
'
'
O
I5 i
S Z T OOxmP,e " Fi9S
i
cr
pe
e
o
With the eleC,ri
ii
1
Marten siti R/c 64
Fig. 8-61
Transformation products of aust
tensite for a eutectoid steel.
!
3 THE HEAT TREATMENT OF STEEL
r
5
23
.
m
.
m
m
"
3
0 .
5? ft
311
For many years, metallurgists divided the tempering process into definite stages. The microstructure appearing in these stages was given names like troostite and sorbite. However, the changes in microstructure are so gradual that it is more realistic to call the product of tempering at any temperature simply tempered martensite. The transformation products of austenite and martensite are summarized in Fig. 8-61. In the above discussion, time of tempering has been assumed to be con-
stant. Since tempering is a process that involves energy both time and temperature are factors. The same effect can be achieved by using a shorter time at a higher temperature as by using a longer time at a lower temperature. Figure 8.62 shows the effect of time at four tempering tern-
V;
,
,
4
peratures for a eutectoid steel.
.
Note that most of the softening action
occurs in the first few minutes and that little further reduction in hardness-
results from increasing the time of tempering from say, 1 to 5 h. It is important to realize that, when toughness measurements are made in order to compare different steels the comparisons must be made at the same hardness or strength levels and at the same temperature of testing. If a medium tensile strength is desired, one may ask why it is necessary first to form a fully martensitic structure and then to reduce the strength substantially in tempering, when the same tensile strength may be obtained, with less difficulty in quenching, from mixtures of martensite and
3S
,
i
V:/
i
m
,
i
,
i13330F produces large globular ce-
1
.
:
,
y soft and tough and is similar to the
tained directly from austenite by a Austenite
cooling rote 30-500F/h or
v > ,
hold
.»-Spheroidized cementite
I200-1300°F
Larqe, rounded cementite particles R/c
5-10
|
.. .
s
l200-i300°F
Sorbite
Coarse pearlite
Small, round, resolved c?rnentite particles,
R/c 15 Z5
matrix 'errite
c>
R/c 20-40
Medium pearlite R/c 30
[750-1200°F
in CM
1 <'MK
i
Tempered
Fine pearlite Cr
R/c 40
matrix territe retained
Tiartensite
,
austenite changed to
1
lower bainite
s
2
Bainite
nod
R/c 40-60
900-400 F §
R/c 40-60
|
400-750°F
Black martensite / / Formation of epsilbn
Martensite R/c 64
up to 400 r
carbide and low carbon -
marlensite .
on
Fig. 8-61
Transformation products of austenite and mar-
tensite for a eutectoid steel.
1
/ ;
Troostite / Cementite particles too small to be resolved,
a XL
R/c 60-64
11
i
/
312
INTRODUCTION TO PHYSICAL METALLURGY
i
70
Logarithmic scale
65
60k\\
to ferrite and pearlite and thei pearlite and martensite. Thetl' tensile strength of 125,000 psi tensitic before tempering had tility, the highest fatigue stren
400 °F
8 55k \
500 "F
5 50
800 0F
45
shows the notched-bar tough ing temperatures. The curves
3
40 q:
I000 0F
35
-
.
the tempered martensite is ies than pearlite.
1
30
As a further aid in the sele
25 !
10
10
Seconds
W 1 2
Minutes
5
Hours
possible to extend the usefL by subjecting additional end peratures (Fig. 8-64).
25
I ime interval at temperature
2
'-tempering te erpercent in Steel" a (Fr0m Bain and PaX,0n A'loying E lem atures upon the softening of a quenched 0 82
OhK r 103" S00iety f0r M "
v
8-23
'
'
et
.
Austempering This is a heatgram to obtain a structure wl-
Petals Park
by first heating the part of the cooling rapidly in a salt bath and 800oF). The piece is left is complete. The steel is ca
m
(2)
.ra47o™: s
r ;rr"ched:0 mar,ensi,8
and at no time is it in the fu
i
tempering is a complete hea tempering. Figure 8-65 illu
n
120r 500 200 100 I00
I
-
80
Tempered mortensite
65
Tempered mortensite
60
+ boinite
50 30 L
1
J_l
-
_
20
LJ
L
As quenched
55
Q
5C
60
Tempered mortensite
Tempered BOO'F
+ peorlite =
P 45 5
40
Tempered 1000°F
40
f 35
-
11
§4
'
20
6
lempered 1200"F
rr
30 C 80
25
40
-
C
-
40
80
I20
Temperature of test "c
160
200 20
,
Fig. 8-63 Variation temperature for threeof notched-bar toughness with testina tensiie strength of t25structures tempered to the same 000 psi ,
.
Book Company New York 1949 ) ,
,
,
( mBHck anTplTps
5 16
1
3
1
S
16
1
5
i 2
0
Distonce from
Fig. 8-64
End-quench test results of 434C
as-quenched condition and after temperin temperatures. (Joseph T. Ryerson & Son, I
m
m mi I
THE HEAT TREATMENT OF STEEL
313
0
to ferrite and pearlite and then quenched, resulting in a mixture of largely pearlite and martensite. The three samples were then tempered to the same
tensile strength of 125,000 psi and tested. The sample that was fully martensitic before tempering had the highest yield strength, the highest ductility, the highest fatigue strength, and the greatest toughness. Figure 8-63 shows the notched-bar toughness of the three structures at different testing temperatures. The curves also indicate that the presence of bainite in
V
the tempered martehsite is less detrimental at room temperature and above than pearlite. As a further aid in the selection of a steel for a given application, it is
possible to extend the usefulness of the end-quench hardenability test by subjecting additional end-quench samples to various tempering temV 8-23
>er-
ts
I
9
is complete. The steel is caused to go directly from austenite to bainite, and at no time is it in the fully hardened martensitic state. Actually, aus-
Samples of a medium-carbon alloy ent ways: (1) quenched to martensite
peratures (Fig. 8-64). Austempering This is a heat-treating process developed from the l-T diagram to obtain a structure which is 100 percent bainitei It is accomplished by first heating the part of the proper austenitizing temperature followed by cooling rapidly in a salt bath held in the bainite range (usually between 400 and 800oF). The piece is left in the bath until the transformation to bainite
tempering is a complete heat treatment and no reheating is involved as in tempering. Figure 8-65 illustrates austempering schematically, showing
,
,
f to bainite and quenched to form a
3) partially transformed isothermally
Cooling rote, "F per second 50 30
500 200 100
Tempered martensite -
j
65
60
Tempered martensite + bainite
15
20
10 J
.
9 8
7
G
5
4
3 5
|3
z
.
L
As quenched
55 50
-
Tempered martensite + pearlite
1/1
Tempered 800°F
45
Tempered 1000oF
40
35
Tempered 1200°F 30 25
160
200
20
g
15
i 6
1
I 16
1
8
2
3
I
4
8
8
'
J
It
li2
5
is
Distance from water cooled end of standard bar, inches
l 1
.
Fig. 8-64 End-quench test results of 4340H steel In the as-quenched condition and after tempering at the indicated temperatures. (Joseph T. Ryerson & Son. Inc.)
r
A
.
i
314
4tf 3 Center
cooling curve
i
TABLE 8 6
Summary of Tensile and Im
by Quench and Temper Method and by
Mr
PROPERTY MEASURED
Rockwell C hardness
Msolhermal IWnsformatlon
Ultimate tensile strength, psi Elongation, % in 2 in.
S diogrcm
Surface
Reduction In area, %
cooling
impact ft-lb (unnotched ,
curve
round specimen)
m
Free-bend test
Analysis: 0.78% C; 0.58% Mn; 0.14f
Heat treatment resulting in grain si
f
Producl
Quench and Temper Pb bath 1450T, 5 min
Bamile
Oil quench Tempered 650 F
la
,
Une, log scole 1
Research Laboratory. U.S Sleel Corporalioi
-
v
30 min
"
-
.
-
;
the difference between auste Center
temper method. The comparison of mechar and the quench and temper n or strength (Table 8-6). The s
cooling curves Surface
cooling curve />. '
properties as reduction of ar
Isothermol ronsformation
m
slow-bend test (Fig. 8-66). A austempered shovel is show the impact strength of auster
Tempered to desired hardness
ness range of Rockwell C 45 Aside from thejadvantage f i
M f
of the conventLQnatmethod.
m rronsformotion
.
i
Time, log scale Frg 8-65 (a) Schematic transformation diagram for austempering; (t>) schematic transformation diagram for c
.
conventional quench and temper method (From "U S S
Canlloy Steels
,
.
"
high hardness as a result of ; danger of quenching cracks
U.S. Steel Corporation )
Product
The primary limitation of ; being heat-treated. Only s
Tempered
avoid transformation to pear
martensite
are suitable. Therefore, mo?
less than '/j in. thick (Fig. 8-6 by the use of alloy steels, bu1 to bainite may become exce 8-24 Surface Heat Treatment or C
tions require a hard wear-re
I
:
C
\
THE HEAT TREATMENT OF STEEL
315
TABLE 8-6 Summary of Tensile and Impact Properties of 0.180-in. Round Rods Heat-treated by Quench and Temper Method and by Austempering* QUENCH AND TEMPER
PROPERTY MEASURED
Ultimate tensile strength, psi Elongation % in 2 In.
49.8
50.0 259,000
Reduction in area, %
.
Impact, ft-lb (unnotched round specimen)
if
3 75
50
26.1
46.4
.
.
,
5
36.6
14.0
1
Greater than 1 0°
Ruptured at 45°
Free-bend test
AUSTEMPERING
259,000
Rockwell C hardness
'
.
METHOD
without rupture
Analysis: 0.78% G; 0.58% Mn; 0.146% Si; 0.042% P; 0.040% S Heat treatment resulting iri grain size (1450oF) 5 to 6 with 6 predominating:
5
Austempering
Quench and Temper
Product Bcinile
Pb bath 1450oF, 5 min
Pb bath 1450oF, 5 min
Oil quench Tempered 650°F, 30 min
Transformed in Pb-Bi bath at 600oF, 20 min
*
i
Research Laboratory, U.S. Steel Corporation,
the difference between austempering and the conventional quench and temper method.
The comparison of mechanical properties developed by austempering and the quench and temper method is usually made at the same hardness or strength (Table 8-6). The superiority of austempering shows up in such properties as reduction of area in tension, resistance to impact, and the slow-bend test (Fig. 8-66). A striking demonstration of the resiliency of an
austempered shovel is shown in Fig. 8-67. The marked improvement in the impact strength of austempered parts is most pronounced in the hardness range of Rockwell C 45 to 55 (Fig. 8-68).
Tempered to desired hardness
Aside fromjh jadvantage of greater ductility and toughness along with high hardnessj aj esult of austempering, there is also less distortion and .
danger of quenching clacks
,
because the quench..isjmt ajdmstic..as that
.
of the conventlOJiaLmethod.
The primary limitation of austempering is the effect of mass of the part being heat-treated. Only sections which can be cooled fast enough to avoid transformation to pearlite in the termperature range of 900 to 1200oF are suitable. Therefore, most industrial applications have been in sections less than V2 in. thick (Fig. 8-69). This thickness may be increased somewhat by the use of alloy steels but then the time for completion of transformation to bainite may become excessive.
Product
T
Tempered
,
mortensite
,
,
i
8 24
Surface Heat Treatment or Case Hardening
Numerous industrial applica-
tions require a hard wear-resistant surface called the case and a relatively ,
{
\
-
v
,
"
.
ij
316 INTRODUCTION TO PHYSICAL METALLURGY
soft, tough inside called the cor
HARDNESS ROCKWELL C 50
hardening:
QUENCHED
AUSTEMPERED
AND TEMPERED
REDUCTION OF AREA IN TENSION 34.5 PERCENT
.
IMPACT
Flame hardening Induction hardening
The first three methods chat
by the addition of carbon, nitric ing by the addition of both car
2 9 FOOT POUNDS .
.
Nltriding
4 5
0 7 PERCENT
,
Carburizing
2
3 Cyaniding or carbonitriding
35.3 FOOT POUNDS l
1
not change the chemical cor shallow-hardening methods. SLOW BEND m- 8-25
must be capable of being hard< about 0.30 percent or higher. Carburizing This is the oldest hardening. A low-carbon steel
is placed in an atmosphere tt monoxide. The usual carburiz Fig. 8-66 Improved toughness and ductility of austempered rods compared with quenched and tempered rods of the same hardness
(Courtesy of Research Laboratory
.
,
BO
Rod diameter 0.18 Carbon 0.7'
U.S.
Steel Corporation )
'
Monqanese
.
0.3
Siiicon
sulfur
o.o;
Phosphorus
0.0'
Eoch plotted pi
represents the overoge o' sev
g40
tests
5 a
H 50
-
.
if,
O
CJ
13
1>J ID
i
a
1
i
20h o
-
CD
5
to 3
3
10
J
i
Fig. 8-67 A striking demonstration of an austempered shovel. The picture on the left shows the extent to which the shovel can be bent without failure and that on the
-
r
r
.
a
o
45
40
,
50
55
Rockwell C hardness
right shows how the bent shovel after removal of the bending force returns to its original position without ,
rin Fig, 8-68 Comparison between austempe Suitln
,
permanent deformation. (Courtesy of Research Laboratory U S Steel Corporation.) .
.
i
1
m
,
quenched and tempered steel. (From
"
Treatment to the Job, U.S. Steel Corporatio "
1 GY
'
50
317
soft, tough inside called the core. There are five principal methods of case hardening:
.ENCHED '
THE HEAT TREATMENT OF STEEL
TEMPERED
1 2 3 4 5
ENSION
ERCENT
Carburizing Nitriding Cyaniding or carbonitriding Flame hardening Induction hardening
The first three methods change the chemical composition carburizing ,
by the addition of carbon nitriding by the addition of nitrogen, and cyanid,
ing by the addition of both carbon and nitrogen. The last two methods do not change the chemical composition of the steel and are essentially shallow-hardening methods. In flame and induction hardening the steel must be capable of being hardened; therefore, the carbon content must be about 0.30 percent or higher.
POUNDS .
.
.
8-25
Carburizing This is the oldest and one of the cheapest methods of case hardening. A low-carbon steel, usually about 0.20 percent carbon or lower, is placed in an atmosphere that contains substantial amounts of carbon monoxide. The usual carburizing temperature is 1700oF. At this tempera-
t
jtcmpcrod 60
of the
Rod diameter 0.180
US .
.
i5 50-
Carbon
0.74
Manganese
0.37
Silicon Sulfur
0.145 0.039
Phosphorus
0.044
i
Each plotted point represents the average of several
y 40-
V
4
tests
| 30 CD
03
-
20-
CD
CP
I sred
St
iwhich the he ut
joratory
.
-
.
.
v
.
.
m
0
i 40
45 50 55 Rockwell C hardness
60
Fig. 8-68 Comparison between austempering and quenched and tempered steel. (From Suiting the Heat Treatment to the Job," U.S. Steel Corporation.) "
,
IL
.
o
0
i
318 INTRODUCTION TO PHYSICAL METALLURGY '
.
i
tration, the form of the carbc cm
m -
n2
accuracy, as a function of els the required amount of tirr the part is removed from th
o
1
cooled and examined micrc
bie in the gradual change c eutectoid zone consisting o lowed by the eutectoid zone zone of pearlite and ferrite, i
i
m
core is reached.
This is ilU
measured microscopically w ent and the case depth mav part in a lathe and machinir
1
of 0.005 in. until the core is r
o
1
i
is made, and the results can
lation of time and tempera Table 8-7.
The carburizing equation reversible and may proceec (
layer it the steel is heated (COz).f This is called deca
!
actions are
r Fe
i
Decarburization is a probl steels. The surface, deplet'
subsequent hardening, ar many tool applications, th(
i
ture
,
the following reaction takes place:
ice are maximum at or nea
Fe + 2CO -» Fe(c) + CO,
where Fe
(c) represents carbon dissolved in austenite The maximum amount of carbon that can be dissolved in austenite at 1700oF is indi on the iron iron carbide equilibrium diagram at the A line. Therefore cated very quickly a surface layer of high carbon (about 1 2 percent) is built up. Since the core is of low carbon content the carbon atoms tr reach equilibrium will begin to diffuse inward The rate of diffusio ying to enite at a given temperature, is dependent upon th sion coefficient and the carbon-concentration gradient Under known e diffui standard operating conditio and ns, with the surface at a fixed carbon concen .
-
cm
,
,
.
il
,
carbon in aust
n of
.
,
.
-
m
i
Figure 8-73 shows decarb Decarburization may be p sphere in the furnace to carbon dioxide, and water
pared by reacting relative! ally natural gas) in an ex nickel catalyst. The gas p
cent hydrogen, and 20 pe Commercial carburizini
burizing, gas carburizing, work Is surrounded by a c
i
.THE HEAT TREATMENT OF STEEL
319
tration, the form of the carbon gradient may be predicted, with reasonable accuracy, as a function of elapsed time. After diffusion has taken place for the required amount of time depending upon the case depth desired, the part is removed from the furnace and cooled. If the part is furnacecooled and examined microscopically the carbon gradient will be visi,
ble in the gradual change of the structure.
I
At the surface is the hyper-
eutectoid zone consisting of pearlite with a white cemefitite network, fol'
lowed by the eutectoid zone of only pearlite and finally t he hypoeutectoid
1
zona of pearlite and ferrite, with the amount of ferrite increasing until the core is reached. This is illustrated in Fig. 8-70. The case depth may be measured microscopically with a micrometer eyepiece. The carbon gradi-. ent and the case depth may be determined experimentally by placing the
n o
part in a lathe and machining samples for chemical analysis at increments of 0.005 in. until the core is reached. Analysis to determine carbon content
£3
is made, and the results can be plotted graphically, as in,Fig. 8-71. The relation of time and temperature to case depth is shown in Fig. 8-72 and
if
Table 8-7. -
i.
V
-
The carburizing equation given previously, Fe + 2CO -* Fe(0) + CO2, is reversible and may proceed to the left, removing carbon from the surface
layer if the steel is heated in an atmosphere containing carbon dioxide '
(C02). This is called decarburization. Other possibleJdecarburizing reactions are
&a
Fe(c) + H20 -» Fe + CO + H2 -
Fe(c, + Oj
J
Fe + C02
Decarburization is a problem primarily with high-carbon steels and tool steels. The surface, depleted of carbon, will not transform to martensite on
sred.
subsequent hardening, and the steel will be left with a soft skin. For many tool applications, the stresses to which the part is subjected in servce:
. Fe(c)
ice are maximum at or near the surface, so that decarburization is harmful.
Figure 8-73 shows decarburization on the surface of a high-carbon steel.
+ C02
lolved in austenite
Decarburization may be prevented by using an endothermic gas atmo.
The maximum
sphere in the furnace to protect the surface of the Steel from oxygen, carbon dioxide, and water vapor. An endothermic gas atmosphere is prepared by reacting relatively rich mixtures of air and hydrocarbon gas (usu-
ed in austenite at 1700oF is indicated i diagram at the A
cm
line. Therefore
,
? carbon (about 1.2 percent) is built
ally natural gas) in an externally heated generator in the presence of a nickel catalyst. The gas produced consists of 40 percent nitrogen, 40 percent hydrogen, and 20 percent carbon monoxide. i Commercial carburizing may be accomplished by means of pack carburizing gas carburizing, and liquid carburizing. In pack carburizing, the work is surrounded by a carburizing compound in a closed container. The
content, the carbon atoms trying to Jse inward The rate of diffusion of .
jerature, is dependent upon the diffuentration gradient Under known and .
[he surface at a fixed carbon conc
,
en-
i! x y-
.
I 1
m
:)
320 INTRODUCTION TO PHYSICAL METALLURGY
TABLE 8 7
ft
Case Depth In Inches
Si
TIME, H I
*3
Hypoeufectoid zone
Eu tec toid zone
Hypereutectoic
1
0 012
0 01
0 017
0 02
3
0 021
0 02
4
0 024
0 02
5
0 027
COS
6
0 030
0 0G
7
0 032
o o:
8
0 034
0 0-:
9
0 036
00
10
0 038
00
11
0 040
0 0'
12
0 042
0 0!
13
0 043
0 0!
14
0 045
0 01
15
0 047
00
16
0 048
00
17
0 050
00
18
0 051
00
19
0 053
00
20
0 054
oc
21
0 055
oc
22
0 056
oc
23
0 058
oc
24
0 059
oc
one
Total case
™t7? 0 20 Peri ent Carb0n Steel Pack-carburized at -
nrt
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
'
L°r 6 h and furnace-cooled. Etched In 2 percent
nital, SOX
.
container is heated to the proper temperature for the required amount of
.
155(
2
4
i
1500
lend itself to high production. Commercial carburizin ' consist of hardwood g compounds usuallv charcoal
.M:!
,
coke, and about 20 percent o barfum car-
1 40
.
.
.
.
,
.
.
.
.
.
.
.
.
.
.
.
.
.
.
l 20h .
.
.
.
.
.
.
.
.
.
.
.
.
1 00 .
*
v
-
-rjr.
Courtesy of Republic Steel Corp.
-
. . .
bonate as an energizer. The particles or lumps, so that, ficient air will be trapped ir advantages of pack carburi; pared atmosphere and tha processing of small lots o1 vantages are that it is not v
= 0 80 .
a
.
3 60 H .
0 40 H .
cases that must be contro
close control of surface ce '
9k
Base carbon
0 20
parts cannot be dlrect-que
,
C
0 020
0 040 Cose depth inches
.
.
,
Fig. 8-71
Carbon-concentration gradient in a carburized
steel with 0 080 in. total case .
.
0 060 .
0 080 .
excessive time is consume of the inherent variation ir
pack carburizing is not us 0 030 in., and tolerances a .
t
t
St r
I
4
I
pi THE HEAT TREATMENT OF STEEL
TABLE 8 7 TIME
TEMP, 'F
H
1500
1550
1600
1650
1700
1750
if
0 012
0 015
0 018
0 021
Q 025
0 029
i
0 017
0 021
0 025
0 030
Q 03 5
0 041
0 021
0 025
0031
C 03"
0 0<3
0 051
0 024
0 029
0 035
042
3 050
0 059
5
0 027
0 033
0 040
0 047
0 056
0 066
6
0 030
0 036
0 043
0 052
0 061
0 072
7
0 032
0 039
0 047
0 056
|0.066
0 078
8
0 034
0 041
0 050
0 060
i 0 071
0 083
9
0 036
0 044
0 053
0 063
0 075
0 088
10
0 038
0 046
0 056
0 067
0 079
0 093
11
0 040
0 048
0 059
0 070
0 083
0 097
12
0 042
0 051
0 061
0 073
0 087
0 102
13
0 043
0 053
0 064
0 076
0 090
0 106
14
0 045
0 055
0 066
0 079
f0.094
0 110
15
0 047
0 057
0 068
0 082
0 097
0 114
16
0 048
0 059
0 071
0 084
0 100
0 117
17
0 050
0 060
0 073
0 087
0 103
0 121
18
0 051
0 062
0 075
0 090
: o.i 06
0 125
19
0 053
0 064
0 077
0 092
0 109
0 128
20
0 054
0 066
0 079
0 094
0 112
0 131
21
0 055
0 067
0 081
0 097
0 114
0 134
22
0 056
0 069
0 083
0 099
0 117
0 138
23
0 058
0 070
0 085
0 101
0 120
0 141
24
0 059
0 072
0 086
0 103
0 122
0 144
,
.
.
3
.
!
t
.
4 Eutectoid zone h<- Hypereutectoid zone case
1
iperature for the required amount of
/ qentially a batch method and does not y v:?rcial carburizing compounds usually "
:
j
Case Depth in Inches by Carburizing*
2
.
321
i
ft:
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
I
.
.
.
.
.
.
I
.
.
.
.
.
.
.
.
.
.
.
.
'
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
I
.
,
.
I
.
.
.
.
.
.
.
.
.
.
.
.
-
and about 20 percent of barium car-
I
'
t
*
JK
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Courtesy of Republic Steel Corp.
bonate as an energizer. The carburizing compound is in She form of coarse :
:
particles or lumps, so that, when the cover is sealed on the container, suf-
ficient air will be trapped inside to form carbon monoxide. The principal advantages of pack carburizing are that it does not require the use of a pre-
pared atmosphere and that it is efficient and econorriical for individual V V:"::;:'J
:
I
Nil .
: : -
4
0-060
0 080 .
processing of small lots of parts or of large, massive |)arts. The disad-
vantages are that It is not well suited to the productionfof thin carburlzed
cases that must be controlled to close tolerances; it Cannot provide the close control of surface carbon that can be obtained by gas carburizing; parts cannot be direct-quenched from the carburizing temperature; and excessive time is consumed in heating and cooling the charge. Because of the inherent variation in case depth and the cost of (jacking materials pack carburizing is not used on work requiring a casetiepth of less than ,
,
burized
0 030 in., and tolerances are at least 0.010 in. .
!_
i
c
.
t '
I
322 INTRODUCTION TO PHYSICAL METfl
URGY
carbon. By using a diffusioi
0 140 .
the temperature maintainec
1750 F
be reduced to any desired v much cleaner work by dissi
0 120.
when no gas is flowing. Ga 0 100
quencjiing, lower cost, cle greater flexibility of operati
1700 "F
.
Liquid carburizing is a rr
a bath of molten cyanide sc
metal and produce a case carburizing: Liquid carbut
5 0 080
"
E
.
1650 °F
a>
2 0 060
1600 0F
.
'
550 0F
0 040 .
I
0 020 .
r
0
0
4
12
8
16
20
24
:
-
'
.
i
28
Carburizing time,hours Fig. 8-72
Relation of time and temperature to case depth. (
Gas carburizing may be either batch or continuous and lends itself better to production heat treatment. The steel is heated in contact with carbon monoxide and/or a hydrocarbon which is readily cecomposed at v
.
'
.
. .
.
.
>-
the carburizing temperature. The hydrocarbon may be methane, propane, natural gas, or vaporized fluid hydrocarbon. Commercial practice is to use
if
t ! £
,
a carrier gas, such as obtained from an endothermic generator, and enrich , it with one of the hydrocarbon gases. -i t It was mentioned previously that carburized parts will 'usually have a thin outer layer of high carbon. There are two reasons why it may be desirable to avoid this hypereutectoid layer. First, if the piece is cooled slowly from the carburizing temperature, a proeutectoid cementite network will form at the grain boundaries. On subsequent hardening, particularly if the steel is heated to below the Acm line, some grain-boundary cementite will remain
in the finished piece and is a frequent cause for failure. Second
,
i
: ' 0
the hyper-
i3
eutectoid surface-carbon content will increase the amount of retained austenite.
Therefore
Fig, 8-73 Decarburized layer of ferrite or a high-carbon annealed steel. Etched in ;
if the steel is highly alloyed, the. carbon content of thecase should be no greater than the eutectoid content of 0.80 percent
Si
m
,
200X.
I
-
i
1
.
-
THE HEAT TREATMENT OF STEEL
325
-
v
carbon. By using a diffusion period, during which the gas is turned off bu
1750 0F
the temperature maintained, gas carburizing allows[:the surface carbon tc
be reduced to any desired value. Use of the diffusio| period also produce; much cieaner work by dissipation of carbon deposi|;(soot) during thetinv when no gas is flowing. Gas carburizing allows quicker handling by direc quenching, lower cost, cleaner surroundings, closjbr quality control, an' greater flexibility of operation in comparison with pack carburizing.
700 F
Liquid carburizing is a method of case-hardening steel by placing it i a bath of molten cyanide so that carbon will diffuse from the bath into th metal and produce a case comparable to one resulting from pack or ga 16b0°F
carburizing.
1600 "F
Liquid carburizing may be distinguished from cyaniding b
" '
i
1550 0F
3
v
.
i :
c .
16 '
20
24
28 v
izing time hours ,
depth. 1
r
batch or continuous and lends itself
3nt.
The steel is heated in contact with
Pi
:
icarbon which is readily decomposed at hydrocarbon may be methane propane, ocarbon. Commercial practice is to use m an endothermic generator and enrich ,
,
s
ses.
t carburized parts will usually have a thin are two reasons why it may be desirable
I
First, if the piece is cooled slowly from
i
jv 'oeutectoid cementite network will form
5:
'
squent hardening particularly if the steel ,
me grain-boundary cementite will remain
: .
xBjient cause for failure.
Second
,
the hyper-
will increase the amount of retained aus-
ilghly alloyed the carbon content of the
I \
.
,
the eutectoid content of 0 80 percent
4
Fig. 8-73 Decarburized layer of ferrite on the surface of a high-carbon annealed steel. Etched in 2 percent nital, 200X.
.
/: y
'
JL. t
I '
1
i
m 324
INTRODUCTION TO PHYSICAL METALLURGY
I
the character and composition of the case produced The cyanide case
to 0.120 in., although it is f
is higher in nitrogen and lower in carbon; the reverse is true of liquidcarburlzed cases Cyanide cases are seldom to a depth greater than 0 010 in.; liquid-carburizing baths permit cases as deep as 0.250 in. Low-temperature salt baths (light-case) usually contain a cyanide content of 20 percent and operate between 1550 and 1650"F These are best suited for case depths up to 0.030 in High-temperature salt baths (deep-case) usually
liquid carburizing is best J large parts are difficult to p
.
.
.
carburizing are: (1) freedo uniform case depth and car
(4) the fact that the bath p
.
ducing the time required to Disadvantages include; (1) ment to prevent rusting; (2 composition is necessary t
.
have a cyanide content of 10 percent and operate between
1650 and 1750oF
.
High-temperature salt baths are used for producing case depths of 0.030 I
Carburizinq temperature -
C
L
F
cannot be handled becaus(
F
r
out of salt; and (4) cyanid(
riticaMemjjerature of core
tion to safety.
2
8 26 Heat Treatment after Carbu CI
h
-
I
!<_,
Critical/
Is?
temperature
region, direct quenching both the case and core if t
at case
ing rate. Direct quenchinc ness and distortion, so th;
Carbon content
grained steels. Alloy steels
Time
because of the large amo TREATMENT
CASE
A-best
-
Figure 8-74 shows a diac CORE
adapted to fine-grained steels
Refined; excess carbide
Unrefined; soft and
not dissolved
machinable
S-best
adapted to fine-grained steels
Slightly coarsened; some
C-best
Somewhat coarsened; solu-
solution of excess
treatments for carburized When a carburized pat martensite zone followed
Partially refined; stronger and tougher than A
carbide
adapted to fine-grained steels
tion of excess carbide
Refined; maximum core strength and hardness;
favored; austenite reten-
better combination of
tion promoted in highly
strength and ductility
alloyed steels
than B
D-best treatment for
Refined; solution of excess
Refined; soft and machin-
coarse-grained steels
carbide favored; austenite retention minimized
able; maximum tough-
/
.
ness and resistance to
impact £-adapted
to fine
-
grained steels only
Unrefined with excess carbide dissolved; austenite
Unrefined but hardened
retained; distortion minimized
m
adapted to fine-
grained steels only
Refined; solution of excess
Fig. 8-74 Various heat treatments for carburized (From "Metals Handb ook, .
for Metals
,
"
steels.
1948 ed., American Society
Metals Park, Ohio.)
r
Unrefined; fair toughness
carbide favored; austenite retention minimized
Fig. 8-75 A properly carburized, harder gear. Etched in 2 percent nital 7X. ,
tea,
!
>
3 THE HEAT TREATMENT OF STEEL
325
.!
H
case produced. The cyanide case ,
although it is possible to go as high as 0.250 in. In general, liquid carburizing is best suited to small and medium-size parts, since
to 0.120 in.
/bon; the reverse is true of liquld-
,
,
|pbldom to a depth greater than 0.010
? i
,
large parts are difficult to process in saft baths. The advantages of liquid carburizing are: (1) freedom from oxidation and sooting problems, (2) uniform case depth and carbon content, (3) a rapid rate of penetration, and (4) the fact that the bath provides high thermal conductivity, thereby reducing the time required for the steel to reach the carburizing temperature. Disadvantages include- (1) parts must be thoroughly washed after treat-
as as deep as 0.250 in Low-temper.
ntain a cyanide content of 20 per50°F. These are best suited for case
ure salt baths (deep-case) usually d operate between 1650 and 1750oF
.
for producing case depths of 0 030
i
ment to prevent rusting; (2) regular checking and adjustment of the bath composition is necessary to obtain uniform case depth; (3) some shapes cannot be handled because they either float or will cause excessive drag-
.
out of salt; and (4) cyanide salts are poisonous and require careful attention to safety. 8-26
,
Heat Treatment after Carburizing region
,
Since steel is carburized in the austenite
direct quenching from the carburizing tempbrature will harden
both the case and core if the cooling rate is greater than the critical cooling rate. Direct quenching ol coarse-grained steels often leads to brittleness and distortion so that this treatment should be applied only to fine,
Time .
grained steels. Alloy steels are rarely used in the direct-quenched condition
because of the large amount of retained austenite in the hardened case. Figure 8-74 shows a diagrammatic representation of various hardening
CORE
rbide
treatments for carburized steels together with case and core properties.
Unrefined; soft and
When a carburized part is hardened, the case will appear as a light
machinable some
martensite zone followed by a darker transition zone (Fig. 8-75). The hard
Partially refined; stronger and tougher than A
led; soluide
Refined; maximum core strength and hardness; better combination of
ighly
i
strength and ductility than B
f excess
Refined; soft and machin-
isten-
able; maximum tough-
ized
ness and resistance to
sss car-
Unrefined but hardened
m
Hp
impact tenite -
v
i
mm
A1
-
i
v:s -f excess
|sten-
Unrefined; fair toughness
v. »
ized
Fig. 8-75 A properly carburized hardened, and tempered gear. Etched in 2 percent nltal, 7X. ,
f
'
Si
'
I*
i;
11
326
INTRODUCTION TO PHYSICAL METALLURGY
are absorbed simultaneously. 1 cause it implies a modified nitrii
case or effective case is measured from the outer edge to the middle of the dark zone. From the nature of the carbon gradient the hard case contains the portion of the case above 0.40 percent carbon and is approximately
modification of carburizing, ani more descriptive. The processJj
,
equal to two-thirds of the total case.
Hardness-traverse measurements
ing, and nicarbing. The atmo comprise aTrTixture of carrier g£
may also be used to determine the depth of the effective case since the middle of the transition zone is at approximately Rockwell C 50. 8-27 Cyaniding and Carbonitriding Cases that contain both carbon and nitrogen are produced in liquid salt baths (cyaniding) or by use of gas atmospheres (carbonitriding). The temperatures used are generally lower than ,
those used in carburizing being between 1400 and 16pooF.
-
,
for a shorter time
rier gas is usually a mixture of i produced in an endothermic ge
gas is supplied to the
furnace
filtration and acts as a diluent ammonia), thus making the pre is usually propane or natural ge added to the surface. At the
Exposure is
and thinner cases are produced, up to 0.010 in. for cyaniding and up to 0.030 in. for carbonitriding. In cyaniding the proportion of nitrogen and carbon in the case produced by a cyanide bath depends on both composition and temperature of the bath, the latter being the most important. Nitrogen content is higher ,
,
(NH3) breaks up or dissociates t
steel. Figure 876 shows a carl
iq baths operating at the lower end of the temperature range than in those operating at the upper end of the range. Generally carbon content of the case is lower than that produced by carburizing ranging from about 0.5 to 0.8 percent. The case also contains up to about 0.5 percent nitrogen; therefore, file-hard cases can be obtained on quenching in spite of the rela,
1m
,
JJ**
.
tively low carbon content. Several mixtures of cyanides are available for the bath. Although baths of higher sodium cyanide concentrations are employed, the most commonly used mixture is made up of 30 percent sodium cyanide, 40 percent sodium carbonate, and 30 percent sodium chloride. This mixture has a melting point of 1140oF and remains quite stable under continuous operating conditions. The active hardening agents of cyaniding baths, carbon and nitrogen, are not produced directly from sodium cyanide (NaCN). Molten cyanide decomposes in the presence of air at the surface, of the bath to produce sodium cyanate (NaNCO), which in turn decomposes as follows;
\
2NaCN + 02-
-
v
-
.
12
i.
2NaNCO m
4NaNCO -> Na2C03 + 2NaCN + CO + 2N
3
.
The carbon content of the case developed in the cyanide bath increases with an Increase in the cyanide concentration of the bath thus providing considerable flexibility. A bath operating at 1550oF and containing about 3 percent cyanide may be used to restore carbon to decarburized steels while a 30 percent cyanide bath at the same temperature will develop a 0 005-ln. case on the surface of a 0.65 percent carbon steel in 45 min. This process is particularly useful for parts requiring a very thin hard case, such as screws small gears, nuts, and bolts. The principal disadvantages of cyaniding are the same as those mentioned under liquid carburizing. ,
.6
,
if
.
,
Carbonitriding is a case-hardening process in which a steel is heated in a gaseous atmosphere of such composition that carbon and nitrogen
i
\
Fiq 8-76 Carbonltrided case on AISI
C1213 !
"
ia-propane at l550oF for 20 min In an ammon
then oil-quenched. Etched in 2 percent nital,
depth approximately 0.0025
in.
.,
1
i
1
THE HEAT TREATMENT OF STEEL, 327
m the outer edge to the middle of the || Spijbon gradient the hard case contains l|
.
V
,
'
''
percent carbon and is approximately ?' f Hardness-traverse measurements 'H .
epth of the effective case, since the
rier gas is usually a mixture of nitrogen, hydrogen, and carbon monoxide
[hat contain both carbon and nitro(cyaniding) or by use of gas atmo-
%tures used are generally lower than
"
d
)roximately Rockwell C 50. .
are absorbed simultaneously. The term carbonitriding is misleading because it implies a modified nitriding process Actually carbonitriding is a modification of carburizing, and the nafne "nitrocarburizing would be more descriptive. The process is also known asjinLcyanidina. oas cvaniding, and nicarbing. The atmospheres used in carbonitriding generally comprise a mixture of carrier gas, enriching gas, and ammonia. The car-
produced in an endothermic generator, as in gas carburizing. The carrier gas is supplied to the furnace under positive pressure, to prevent air in-
1 1
filtration and acts as a diluent for the active gases (hydrocarbons and ammonia), thus making the process easier to control. The enriching gas
veen 1400 and 16pO°F. Exposure is 3re produced, up to 0.010 in. for cyatriding. rogen and carbon in the case proboth composition and temperature nportant. Nitrogen content is higher the temperature range than in those
is usually propane or natural gas and is the primary source for the carbon
added to the surface. At the furnace temperature, the added ammonia (NH3) breaks up or dissociates to provide the nitrogen to the surface of the steel. Figure 8-76 shows a carbonitrided case obtained by heating C1213
e. Generally, carbon content of the
:
6
1
carburizing, ranging from about 0.5 is up to about 0.5 percent nitrogen; "
: .
ned on quenching in spite of the relaixtures of cyanides are available for isodium cyanide concentrations are nixture is made up of 30 percent soponate, and 30 percent sodium chlo!t of 1140oF and remains quite stable ns. The active hardening agents of are not produced directly from sodecomposes in the presence of air at um cyanate (NaNCO), which in turn
t v
-
.
i
,
3.
*
.
A-
» 2NaNCO
r
-
2NaCN + CO + 2N
I'
A mm
i
I -
oped in the cyanide bath increases
I:
pntration of the bath, thus providing
I
jting at 1550oF and containing about
|store carbon to decarburized steels,
y
ne same temperature will develop a
.
percent carbon steel in 45 min
This
requiring a very thin hard case
such
.
,
ts. The principal disadvantages of ioned under liquid carburizing.
I
..
Fig. 8-76 Carbonitrided case on AISI C1213 steel. Heated 1:
F for 20 min in an ammonia-propane atmosphere,
°
at 1550
process in which a steel is heated
then oil-quenched. Etched in 2 percent nital, 125X. Case
nposition that carbon and nitrogen
depth approximately 0.0025 in.
326
INTRODUCTION TO PHYSICAL METALLURGY
steel in an ammonia-propane atmosphere at 1550oF for 20 min followed by K Each division of the micrometer eyepiece is 0.001 in. andrJp
oil quenching.
,
the effective case depth measured to the middle of the dark transition zone; is approximately 0.0025 in.
The presence of nitrogen in the austenite accounts for the major differ-
|j
.
ences between carbonitriding and carburizrng. Carbon-nitrogen austenite is stable at lower temperatures than plain-carbon austenite and transforms more slowly on cooling. Carbonitriding therefore can be carried out at lower temperatures and permits slower cooling rates than carburizing in
i
the hardening operation. Because of the lower temperature treatment and :|
"
m
1
oil quenching rather than water quenching, distortion is reduced, and there is less danger of cracking. Since nitrogen increases the hardenability carbonitriding the less expensive carbon steels for many applications will provide properties equivalent to those obtained in gas-carburized alloy ,
'
steels.
It has also been found that the resistance of a carbonitrided sur-
face to softening during tempering is markedly superior to that of a carburized surface. 8-28
Nitriding
This is a process for case hardening of alloy steel in an atmo
-
.
sphere consisting of a mixture in suitable proportions of ammonia gas and dissociated ammonia. The effectiveness of the process depends on the formation of nitrides in the steel by reaction of nitrogen with certain alloying elements. Although at suitable temperatures and with the proper atmo-
sphere all steels are capable of forming iron nitrides, the best results are obtained in those steels that contain one or more of the major nitride-forming
alloying elements. These are aluminum, chromium, and molybdenum. The nitrogen must be supplied in the atomic or nascent form; molecular nitro((7)100X
gen will not react.
The parts to be niti ided are placed in an airtight container through which the nitriding atmosphere is supplied continuously while the temperature is raised and held between 925 and 1050oF. The nitriding cycle is quite
long, depending upon the case depth desired. As shown in Fjg. 8-77, a 60-h cycle will give a case depth of approximately 0.024 in. at 9750F. A nitrided case consists of two distinct zones.
"
is commonly known as the
0 016
a nital etch. In the zone bene
S 0.012
"
v
been precipitated. A typical m the white layer and the underh
0 008
S- 0 004 .
=1
20
40
60
80
Time in hoursot temperature(9750F1 "
In the outer zone the
This region, which varies in th
.
.
Fig. 8-77
dissociation. Diai
shapes are Knoop hardness impressions. (Court. Nitralloy Corporation.)
0 020
.
a>
975-F and 30 percent ammonia
0 024
.
=
a nitrided c
nitride-forming elements, inch
S 0.032 5 0 028 E
Fig 8-78 Microstructure illustrating
produced by a single-stage process. Nitrided for
Depth of nitrided case vs. time at 975 , (From
Heat Treatment of Steels," Republic Steel Corporation.)
illustrated in Fig. 8-78a, the lie nitride case. The depth of nitr of nitrogen from the white la'
'
THE HEAT TREATMENT OF STEEL
;J29-
i
IT
sphere at 1550oF for 20 min followed by :
j micrometer eyepiece is 0 001 .
in.
,
White loyer
andf
jo the middle of the dark transition zone >«:
*
ustenite accounts for the major differ-arburizing. Carbon-nitrogen austenite:
..
T
'
it
.
plain-carbon austenite and transforms
iding therefore can be carried out at
)wer cooling rates than carburizing in
J
»f the lower temperature treatment and : iching distortion is reduced and there
V
,
:
,
,
nitrogen increases the hardenability rbon steels for many applications will lose obtained in gas-carburized alloy ,
| 1
-
the resistance of a carbonitrided surf
...
Is markedly superior to that of a car-
hardening of alloy steel in an atmoitable proportions of ammonia gas and :i /.less of the process depends on the foriction of nitrogen with certain alloying
iperatures and with the proper atmo- j \ ng iron nitrides the best results are ob- j | le or more of the major nitride-forming J ?
It
,
um, chromium and molybdenum. The ,
)mic or nascent form; molecular nitro(J)100X
in an airtight container through which d continuously while the temperature i 1050oF. The nitriding cycle is quite )th desired. As shown in Fig 8-77, a .
Fig. 8-78
( );500X
Microstructure illustrating a nitrided case
produced by a single-stage process. Nitrided for 48 h at 9750F and 30 percent ammonia dissociation. Diamond
shapes are Knoop hardness impressions. (Courtesy of The Nitralloy Corporation.)
approximately 0.024 in at 9750F. listinct zones. In the outer zone the .
nitride-forming elements, including iron, have been converted to nitrides. This region, which varies in thickness up to a maximum of about 0.002 in., is commonly known as the "white layer" because of its appearance after a nital etch. In the zone beneath this white layer, alloy nitrides only have
hi
.0
i
been precipitated. A typical microstructure, illustrated in Fig. 8-78b, shows the white layer and the underlying nitride case. At the lower magnificat on, illustrated in Fig. 8-78a, the lighter core structure can be seen beneath the nitride case. The depth of nitride case is determined by the rate of diffusion of nitrogen from the white layer to the region beneath. The nitriding me-
330
:
INTRODUCTION TO PHYSICAL METALLURGY
dium, therefore, needs to contain only sufficient active nitrogen to maintain the white layer. Any increase beyond this point serves to increase the depth of white layer and does not affect the thickness of the inner layer. The concentration of active nitrogen on the surface of the s;teel, which determines the depth of the white layer, is fixed by the degree of dissociation of the ammonia. In the single-stage nitriding process this dissociation is held between 15 and 30 percent by adjusting the rate of flow. A temperature in the 925 to 9750F range is employed. The double-stage process, also known as the Floe process, has the advantage of reducing the thickness of the white nitride layer. In the first stage of the double-stage process, the ammonia dissociation is held at 20 percent for a period of 5 to 10 h at 9750F. During this period the white layer is established, and the useful nitride starts to form by diffusion of nitrogen out of it. In the second stage, the ammonia dissociation is increased to 83 to 86 percent, and the temperature is usually raised to 1025 to 1050oF. During this second stage the gas composition is such that it maintains only a thin white layer on the
,
5
.
'
* :
finished part. A typical structure of the case produced by this method is shown in Fig. 8-79. The white layer is brittle and tends to chip or spall from the surface if it has a thickness in excess of 0.0005 in. Thicker white layers produced by the single-stage process must be removed by grinding or lapping after
.v
.
nitriding. Ordinarily an allowance of at least 0.002 in. on a side is made in the finish machining dimensions if grinding is necessary after nitriding. If the double-stage process is used, however, grinding or other finishing operations may be omitted except insofar as they are required in order to meet dimensional tolerances. The very thin white layer obtained by this method, usually from 0.0002 to 0.0004 in. in depth, does not chip or pit, and the frictional characteristics of the surface are excellent. This layer
{a] 100 X f-ic,
also has good wear-in properties and may be expected to improve cor-
.
h at 975 F and 20 percent ammonia dissoc
ciaiion Diamond
shapes are Knoop hardness in Notice the much smaller white layer compared w .r, Firj.
8-78. (Courtesy of The Nitralloy Corporate
is an advantage of nitriding ov cannot be case hardened satis without difficulty. Wear resisti -
nitrided case and is responsibl
hardness of a nitrided case is l
the original nitriding temperat
does occur due to the increase in volume of the case. However, this growth
i
.
monia iiy 42 h at 1050JF and 83 to 86 percent am
Hardest cases, approximately R/C 70, are obtained with aluminum alloy steels known as Nitralloys. These are medium-carbon steels containing also chromium and molybdenum. For some applications where lower hardness is acceptable, medium-carbon standard steels containing chromium and molybdenum (AISI 4100, 4300 series) are used. Nitriding has also been applied to stainless steels and tool steels for certain applications. The steel is usually hardened and tempered between 1100 and 1300oFto produce a sorbitic structure of maximum core toughness and then nitrided. Since nitriding is performed at relatively low temperatures and no quenching is required, distortion is reduced to a minimum, although some growth
parts may be machined very close to final dimensions before nitriding. This
c;
Nitri prciJuced by the double-stage Floe process. iatio
rosion resistance.
is constant and predictable for a given part and cycle, so that in most cases
8-79 Microstructure illustrating a nitrided
1
least 1150°F, in marked contras its hardness at relatively low t
important advantage. Tool m
THE HEAT TREATMENT OF STEEL
V
.
)nly sufficient active nitrogen to main-l48 )eyond this point serves to increase the|
-
.
,
,
:
3
White layer f
\Sv enffect the thickness of the inner layen:t| on the surface of the steel
T
which
,
331
ayer, is fixed by the degree of dissocia-
m m
tage nitriding process this dissociation H y adjusting the rate of flow. A temperamployed. The double-stage process s the advantage of reducing the thickle first stage of the double-stage proceld at 20 percent for a period of 5 to ,
,
;
V
mm
mm
white layer is established and the use,
|n of nitrogen out of it.
In the second
increased to 83 to 86 percent and the 5 to 1050oF. During this second stage naintains only a thin white layer on the the case produced by this method is ,
*
I
:
to chip or Spall from the surface if it
y j in. Thicker white layers produced by
:
' '
'
emoved by grinding or lapping after
-
)f at least 0.002 in. on a side is made
if grinding is necessary after nitriding however, grinding or other finishing isofar as they are required in order to very thin white layer obtained by this 004 in. in depth, does not chip or pit the surface are excellent. This layer nd may be expected to improve corm .
,
,
70, are obtained with aluminum alloy
are medium-carbon steels containing For some applications where lower
bon standard steels containing chro4300 series) are used Nitriding has and tool steels for certain applications .
.
:tempered
between 1100 and 1300oF to num core toughness and then nitrided .
yely low temperatures and no quench\;::to a minimum, although some growth ume of the case. However this growth ,
,
n part and cycle '
,
so that in most cases
final dimensions before nitriding
.
i
This
4
.
to
(<7)100X
mm mm (b)b00)(
Fig. 8-79 Microstruolure illustrating a nitrided case produced by the double-stage Floe process. Nitrided (or 8 h at 975T and 20 percent ammonia dissociation followed by 42 h at 1050"F and 83 to 86 percent ammonia dissociation. Diamond shapes are Knoop hardness impressions. Notice the much smaller white layer compared with that in Fig. 8-78. (Courtesy of The Nilralloy Corporation.)
is an advantage of nitriding over carburizing. Some complex parts which cannot be case-hardened satisfactorily by carburizing have been nitrided without difficulty. Wear resistance is an outstanding characteristic of the nitrided case and is responsible for its selection in most applications. The hardness of a nitrided case is unaffected by heating to temperatures below the original nitriding temperature. Substantial hardness is retained to at least 11 SOT, in marked contrast with a carburized case, which begins to lose
its hardness at relatively low temperatures. Fatigue resistance is also an important advantage. Tool marks and surface scratches have little effect
t
.
V
-
.
m 332
INTRODUCTION TO PHYSICAL METALLURGY
on the fatigue properties of nitrided steels Although it is sometimes indi ;: cated that nitriding improves the corrosion resistance of a steel this is t true only if the white layer is not removed Corrosion resistance of stainless .
gressive; (3) spinning, (4) progressi
"
-
and work are stationary.
This metlr
,
,
,
.
steels is reduced considerably by nitriding a factor which must be taken f ,
into account when nitrided stainless steels are used in corrosive atmo spheres. Disadvantages of nitriding include the long cycles usually re-
I
to the treatment of teeth of large geai
is to be obtained cost of ammonia atmosphere and the technical control required. Nitriding is used extensively for aircraft engine parts such as if
harden parts of circular cross sectior similar components. The progressivi
quired, the brittle case, use of special alloy steels if maximum
hardness
,
,
cylinder liners, valve stems shafts, and piston rods. Flame Hardening The remaining two methods cams
m
method, the torch moves over a st;
hardening of large parts, such as the
-
R
,
,
8-29
small parts such as valve stems and oi
,
duction hardening
flame hardening and in-If
do not change the chemical composition of the steel *
,
the torch is stationary while the W(
moves over a rotating workpiece, is
parts such as shafts and rolls. In all procedures, provision must'
.
'
They are essentially shallow hardening methods Selected areas of thesur-% .
face of a steel are heated into the austenite range and then quenched to # form martensite. Therefore it is necessary to start with a steel which is f ,
capable of being hardened
.
Generally, this is in the range of 0 30 to 0.60% .
percent carbon.
»
In flame hardening heat may be applied by a single oxyacetylene torch ; ; as shown in Fig 8-80, or it may be part of an elaborate apparatus which ,
,
.
automatically heats
,
quenches, and indexes parts Depth of the hardened; .
zone may be controlled by an adjustment of the flame intensity heating time, or speed of travel Skill is required in adjusting and handling manuall y operated equipment to avoid overheating the work because of high flame ,
.
temperature cessive
.
Overheating can result in cracking after quenching and ex
;
-
grain growth in the region just below the hardened zone Fourf methods are in general use for flame hardening: (1) stationary; (2) pro- fc ,
.
'
surface has been heated to the requ
plished by the use of water sprays, or oil, or even by air-coo/ing for s should be stress-relieved by heatin cooled. Such a treatment does not The hardened zone is generally m
<
burizing, ranging from Vs to V4 in. i Vu in. can be obtained by increasir Among the advantages of flame bility. The equipment can be take the area which requires hardening. can be handled easily and quickly \
ability to treat components after scaling, decarburization, ordistorti bility of overheating and thus dama
ing hardened zones less than Vi« 8-30 Induction
Hardening Induction hf localized heating produced by c
rapidly changing magnetic
field.
in which the primary or work coi
tubing that are water-cooled, and ondary of a high-frequency indi
Flame head
work coils for use with high-fre veloped by each are shown in Fig. .
(a) a simple solenoid for externa
. .
"
for heating bores, (c) a pie-plat current densities in a narrow bar
Hardness
pattern 1
-Workpiece
the hardness pattern developed (From "Metals Handbook."
turn-coil for scanning a rotating turn that will aid in heating the
vol. 2, American Society for Metals Metals Park, Ohio
heating.
,
Fig. 8-80 Progressive method of flame hardening, showing .
,
1964.)
When high-frequency alternat
p
.,
THE HEAT TREATMENT OF STEEL
4
333
s
)d steels. Although it is sometimes indi
gressive; (3) spinning; (4) progressive-spinning.
In the first, both torch and work are stationary. This method is used for the spot hardening of small parts such as valve stems and open-end wrenches. In the progressive
corrosion resistance of a steel this & ,
"
noved. Corrosion resistance of stainle«|
nltriding, a factor which must be taken! less steels are used in corrosive atmol
method, the torch moves over a stationary work piece; this is used for
jing include the long cycles usually re-1
to the treatment of teeth of large gears (Fig. 8-80). In the spinning method, the torch is stationary while the work rotates. This method is used to harden parts of circular cross section, such as precision gears, pulleys, and
hardening of large parts, such as the ways of a lathe, but is also adaptable
ecial alloy steels If maximum hardness
p atmosphere, and the technical control! lively for aircraft engine parts such as N shafts, and piston rods v;;;;;:wo methods flame hardening and in5 the chemical composition of the steel fling methods. Selected areas of the sur-: i austenite range and then quenched tb ,
;
"
similar components. The progressive-spinning method, in which the torch moves over a rotating workpiece, is used to surface-harden long circular
.
,
,
parts such as shafts and rolls.
.
Ih all procedures, provision must be made for rapid quenching after the
.
surface has been heated to the required temperature. This may be accomplished by the use of water sprays, by quenching the entire piece in water
or oil, or even by air-cooling for some steels. After quenching, the part should be stress-relieved by heating in the range of 350 to 400oF and aircooled. Such a treatment does not appreciably reduce surface hardness. The hardened zone is generally much deeper than that obtained by carburizing, ranging from Vs to 1/4 in. in depth. Thinner cases of the order of Vi6 in. can be obtained by increasing the speed of heating and quenching. Among the advantages of flame hardening are adaptability and portability. The equipment can be taken to the job and adjusted to treat only the area which requires hardening. Parts too large to be placed in a furnace can be handled easily and quickly with the torch. Another advantage is the
necessary to start with a steel which is
srally, this is in the range of 0 30
to 0.60
.
j applied by a single oxyacetylene torch part of an elaborate apparatus which I ,
|e
?:]dd indexes indexes
pa parts. Depth of the hardened
justment of the flame intensity heating Mired in adju; djusting andjiandling manually; ,
{heating the work because of high flame suit in cracking after quenching and exn just below the hardened zone
.
'
Four
ame hardening: (1) stationary; (2) pro-
r
,
:
r
ability to treat components after surface finishing, since there is little
scaling, decarburlzation, or distortion. Disadvantages include (1) the possibility of overheating and thus damaging the part and (2) difficulty in producing hardened zones less than Vu in. in depth. 8-30 Induction Hardening Induction hardening depends for its operation on localized heating produced by currents induced in a metal placed in a rapidly changing magnetic field. The operation resembles a transformer in which the primary or work coil is composed of several turns of copper tubing that are water-cooled, and the part to be hardened is made the secondary of a high-frequency induction apparatus. Five basic designs of work coils for use with high-frequency units and the heat patterns developed by each are shown in Fig. 8'81a through e. These basic shapes are;
(a) a simple solenoid for external heating, (b) a coil to be used internally for heating bores, (c) a "pie-plate" type of coil designed to provide high current densities in a narrow band for scanning applications {d) a single,
howing Jbook." o,
i:. ZC'
turn-coil for scanning a rotating surface provided with a contoured half,
turn that will aid in heating the fillet, and (e) a "pancake" coil for spot heating. When high-frequency alternating current passes through the work coil,
v.
-.
-.
334 INTRODUCTION TO PHYSICAL METALLURGY r Z5 TABLE 8-8 Effect of Frequency on Depth of C
THEORETICAL DEP1
Heoting pattern Healing pattern
,
FREQUENCY,
OF PENETRATION
HZ
OF ELECTRICAL EN IN. 1 000
0 059
3 000
0 035
10,000
0 020
120,000
0 006
500,000
0 003
1 000,000
0 002
Healing pattern
10]
,
,
m
,
:
10
Heating pattern
.
.
.
.
.
.
From Metals Handbook, vol 2. p. 180, American So
work through the coils is adjusted to ol cause these methods of temperature ( ducing the required case depth are ger
[e]
radiation pyrometer may be used to me alure of the work and improve uniform!
! i S5
Heating porlern
I
!19J the and t8\heatTyPiCal WOrkSdeveloped COils forbhiy geach. h-fluency units patterns (From ' Metals
Handbook
1
Park,
vol. 2 American Ohio 1964,) , ,
,
Society for Metals Metals
ing, provision must be made for rapii reached the desired temperature. The case obtained by induction hare flame hardening, and thinner cases arf lar to those used for flame hardening. bon content are used for most applica ot thin cases. The carbon dissolves c
,
,
quired to heat the steel to the quenchi be induction-hardened and are needi
high-frequency magnetic field is set up This magnetic field induces high-frequency eddy currents and hysteresis currents in th e metal. Heating results from the resistance of the metal to passa high-frequency induced currents ge of these The tend to travel at the surfacecurrents. of the metal This is known as skin effect Therefore, it is possible to heat a shallow layer of the steel without heating the interior. However heat applied to the surface tends to flow toward the center by conduction and so time of heating is an important factor in controlling the depth of the hardened zone -Th surface layer is heated practically Instantaneously to a depth which ie inversely proportional to the square root of the frequency s frequencies commonl The range y used is between 10 000 and 500 000 Hz Table 8 8 shows the effect of frequency on depth of case hardn d ess. Greater.case
alloy steels are readily surface-harden steels are more sluggish and may n
a
.
achieve the desired structure for sa
.
.
,
,
i
.
.
1
,
,
epths may be obtained at each frequency by
t
.
increasing the
of
cause of the rapid heating, the alloy i from 100 to 200oF higher by induct methods without danger of excessivi been surface-hardened by induction than the same parts quenched from steel before induction hardening is ir to be used. Structures obtained aftei
carbides are small and uniformly dij
accordingly, minimum case depths face hardness while using very raj structures typical of normalized, hotto 0.50 percent carbon also resf
0 40 .
lT processes " COntro,,ed timingTe timing the ZTe cycle TIn continuous '
.
'
n r
,
-
1
omaticaHy |
the speed of passage of
the
ing. Another advantage of induction ment directly into the production I since the operation is practically aul
i THE HEAT TREATMENT OF STEEL
335
TABLE 8 8 Effect of Frequency on Depth of Case Hardness*
THEORETICAL DEPTH OF PENETRATION OF ELECTRICAL ENERGY,
'
Heofmg pattern
FREQUENCY.
HZ
IN.
Heating pattern
OF CASE HARDNESS, IN.
1,000
0 059
0 1S0 to 0.350
0 035
0 150 to 0.200
10,000
0 020
0 100 to 0.150
i o.ooo
0 006
0 060 to 0.100
500,000
0 003
0 040 to 0.080
1 000,000
0 002
0 010 to 0.030
'
6]
. From
.
.
.
.
.
.
.
.
.
.
,
Heating pattern
PRACTICAL DEPTH
3 000
.
,
id
'
.
.
Melals Handbook, vol. 2. p. 180. American Society for Metals, Metals Park. Ohio, 1964.
work through the coils is adjusted to obtain the required temperature. Because these methods of temperature control are indirect conditions producing the required case depth are generally determined by experiment. A ,
|; radiation pyrometer may be used to measure and control the actual temperature of the work and improve uniformity of hardening. As in flame hardening, provision must be made for rapid quenching of the part after it has reached the desired temperature. The case obtained by induction hardening is similar to that obtained by flame hardening and thinner cases are possible. The steels used are similar to those used for flame hardening. Plain-carbon steels of medium carbon content are used for most applications, particularly for the production of thin cases. The carbon dissolves completely even in the short time required to heat the steel to the quenching temperature. Alloy steels can also be induction-hardened and &re needed particularly for deep cases. Lowalloy steels are readily surface-hardened by this method, but highly alloyed steels are more sluggish and may require an increase in temperature to achieve the desired structure for satisfactory hardening. However, because of the rapid heating, the alloy steels can be heated to temperatures from 100 to 200oF higher by induction hardening than by conventional methods without danger of excessive grain growth. Steel parts that have ,
als s
3 set up. This magnetic field induces iysteresis currents in the metal Heating .
petal to passage of these currents. The
ind to travel at the surface of the
I
metal.
ore, it is possible to heat a shallow layer
been surface-hardened by induction generally exhibit less total distortion
jrior. However heat applied to the surr by conduction and so time of heating 3 the depth of the hardened zone The ,
than the same parts quenched from a furnace. The microstructure of the steel before induction hardening is important in selecting the heating cycle
,
to be used. Structures obtained after quenching and tempering so that the carbides are small and uniformly dispersed are most readily austenitized; accordingly, minimum case depths can be developed with maximum sur-
.
> /7
instantaneously to a depth which is :'
'
»>>
.
ween 10,000 and 500,000 Hz Table 8-8
:
.
ix
'
ijdepth of case hardness
Greater case frequency by increasing the time of
,
i
iis generally controlled by recesses
,
face hardness while using very rapid rates of heating Pearlite-ferrite structures typical of normalized hot-rolled, and annealed steels containing 0 40 to 0.50 percent carbon also respond successfully to induction hardening. Another advantage of induction hardening is the ability to fit the equipment directly into the production line and use relatively unskilled labor .
.
m mi
automatically fp .
the speed of passage of the *
.
since the operation is practically automatic Among disadvantages are the .
i f
336
t
INTRODUCTION TO PHYSICAL METALLURGY
This total stress of 156,000 psi m and outside layers, and the aver area available to support this str( stitute one-fourth of the crosson the outside would be equal average compressive stress on t
cost of the equipment, the fact that small quantities or irregular-shaped parts cannot be handled economically, and high maintenance costs. Typical parts that have been induction-hardened are piston rods, pump shafts, spur gears, and cams. 8-31 Residual Stresses These are stresses that remain in the part after the force has disappeared. Residual stresses always arise from a nonuniform plastic deformation. In the case of heat treatment, this nonuniform plastic deformation may be caused by the temperature gradient or the phase change or
psi. This stress distribution
is p
in tension must balance the are to be in equilibrium across the c
usually a combination of both factors during cooling. Residual stresses are a very serious problem in heat treatment, since they often result in dis-
sharp drop in stress at the juncti sharp drop in temperature frorr
tortion or cracking and in some cases in premature failure of the part in service. Actually the problem of residual stresses is quite complex, but it is hoped that the following discussion, although simplified, will give the student some insight into and appreciation of the factors that give rise to
does not drop sharply but chan shown by the curves of Fig. 8-43 bution is shown by the dotted shows that the tensile stress or
these stresses.
If this stress exceeds the ultim occur. This is what usually he
Consider first the effect of temperature gradient alone. It was shown earlier, under the effect of size and mass, that during quenching the surface is cooled more rapidly than the inside. This results in a temperature gradient across the cross section of the piece or a temperature difference
temperature difference. In th< alone very rarely lead to crackii
of the steel, the stress will be b
between the surface and the center. For example, let us examine the cool- §:
reached room temperature,
ing curves of a 2-in. round water-quenched (Fig. 8-41). At the end of 10 s
AT
will be zero, there will be strength, the surface layer wi no
the surface has cooled to about 700oF, while the center is at about 1500oF.
Almost all solids expand as they are heated and contract as they are cooled.
elongated. At room temperatur
This means that at the end of 10 s the surface, since it is at a much lower
temperature, should have contracted much more than the inside. However, since the outside and inside are attached to each other, the inside, being
150
longer, will prevent the outside from contracting as much as it should. It will therefore elongate the outside layers, putting them in tension while the inside in turn will be in compression. The approximate magnitude of this thermal stress may be calculated from the following formula: s
a
100 \
\ \
E AT
thermal stress, psi coefficient of linear expansion, in./(in.) (0F) E modulus of elasticity, psi A7 = difference in.temperature, 0F
50
II
I
where s
a
Assuming an average value for the coefficient of expansion for steel as 6 5 x 10"6 in./(in.)(0F) and £ = 30 x 106 psi, insertion of these values in the above equation with AT = 800 (1500 to 700bF) gives
\
f
I
-
0
\
Diometer -
A
/
\ \
.
7 /
\
/
.
s
-
= 6
5 x 10 - 6 x 30 x 106 x 800= 156,000 psi
.
This is the approximate value of the thermal stress existing between the outside and inside layers because of the temperature difference of 800°F.
-
I
100
Fig 0-82 Schematic representation of the stre
tnbution across the diameter due to a temperal gradient. Dotted curve indicates a truer represe the stress distribution.
.
I THE HEAT TREATMENT OF STEEL
This total stress of 156,000 psi must now be distributed between the inside and outside layers, and the average stress is inversely proportional to the area available to support this stress. Assuming that the outside layers con-
at small quantities or irregular-shaped -J
;:;;:
lly, and high maintenance costs Tvpi.
y;V.;:-)iardened are piston rods, pump shafts
I
,
337
stitute one-fourth of the cross-sectional area, the average tensile stress
es that remain in the part after the force always arise from a nonuniform plastic satment, this nonuniform plastic defop
on the outside would be equal to 3/a x 156,000, or 117,000 psi, while the average compressive stress on the inside would be V4 x 156,000, or 39,000 psi. This stress distribution is plotted schematically in Fig. 8-82. The area
irature gradient or the phase change or
in tension must balance the area in compression in order for the stresses
eatment, since they often result In dis-
to be in equilibrium across the cross section. The plot in Fig. 8-82 shows a sharp drop in stress at the junction of the inside and outside layers due to a
brs during cooling. Residual stresses
.
ses in premature failure of the part in
sidual stresses is quite complex
'
,
sharp drop in temperature from 1500 to 700 F. Actually the temperature does not drop sharply but changes gradually across the cross section, as shown by the curves of Fig. 8-43. A truer representation of the stress distribution is shown by the dotted curve in Fig. 8-82. The above discussion shows that the tensile stress on the surface may reach a very high value. If this stress exceeds the ultimate strength of the material, cracking will occur. This is what usually happens when glass is subjected to a large temperature difference. In the case of steel, however, thermal stresses
but it
5ion, although-simplified, will give the eciation of the factors that give rise to 5rature gradient alone
.
It was shown
mass, that during quenching the suri inside. This results in a temperature the piece or a temperature difference
'
alone very rarely lead to cracking. If the stress is below the yield strength
For example, let us examine the coolenched (Fig. 8-41). At the end of 10 s
of the steel, the stress will be borne elastically. When the entire piece has
0F while the center is at about 1500°F.
will be zero, there will be no distortion.
Seated and contractus they are cooled
strength, the surface layer will be plastically deformed or permanently elongated. At room temperature the surface will have residual compressive
reached room temperature, Ar=0, and therefore, since the thermal stress
,
.
ne surface, since it is at a much lower
i much more than the inside. However
,
iched to each other, the inside being ,
150
m contracting as much as it should ! layers, putting them in tension while
:
.
:
-
|sion. The approximate magnitude of
r
icoh \\
'
d from the following formula:
i
I I
\
o
I
\ \
E AT
50
ision, in./(in.) (0F)
I I
1
I
I
-
Diometer -
-
0
\
\ v
-
\
7 /
0F
/
o tn in
coefficient of expansion for steel as '
| 50
I06 psi, insertion of these values in the .
- to 700oF)
o
gives 100
lO6 x 800 = 156 000 psi
v
,
? thermal stress existing between the .jf the temperature difference of SOOT.
I i
Fig. 8-82
Schematic representation of the stress dis-
tribution across the diameter due to a temperature
gradient. Dotted curve indicates a truer representation of the stress distribution.
If the stress exceeds the yield
..
. .
.
.
J
-
-
:
338
INTRODUCTION TO PHYSICAL METALLURGY
stress and the inside, residual tensile stress. cylindrical, it will now be barrel-shaped.
If the piece was originally
i
Ae
Austenite, being f.c.c. (face-centered cubic), is a denser structure than any of its transformation products
.
Center cooling curve
Therefore, when austenite changes to
ferrite, p'earlite, bainite, or martensite, an expansion occurs. The austeniteto-martensite expansion is the largest and amounts to a volume increase ol about 4.6 percent. The martensite expansion will be greater the lower the ir Ms
:
A + F+C
temperature. Figure 8-83 shows the changes in length, during cooling, jf
of a small-diameter cylinder as measured in a dilatometer. The piece is austenitic at the elevated temperature, and normal contraction of the austenite takes place until the /Ws temperature is reached. Between the M,
-
a
and the M, the transformation of austenite to martensite causes an expan-
EU
Surfoce
cooling
sion in length. After the M, temperature, the martensite undergoes normal
curve
contraction.
Let us now consider the combined effect of temperature gradient and
phase change for two possibilities: (1) through-hardened steel and (2) shallow-hardened steel.
Figure 8-84 shows the surface- and center-cooling curves superimposed on the l-T diagram for the through-hardened steel. Since the centercooling rate exceeds the critical cooling rate, the part will be fully martensitic across its diameter. During the first stage, to time /„ the stresses present are due to the temperature gradient. The surface, prevented from contracting as much as it should by the center, will be in tension while the center will be in compression. During the second stage, between times f,
1
and f2, the surface, having reached the iW5 temperature, transforms to mar-
31
ISI
stage
2nd
3rd
stage
'
2
Tr
Fig, 8 84 Center- and surface-cooling curves on the l-T diagram to illustrate the through-ha condition.
tensite and expands. The ce tion due to cooling. The cei
expanding as much as it shot sion while the center will tei reached room temperature ai
During the third stage, the c begins to expand, forming n
QJ
to pull the surface along witl condition in the three stages Martensite
austenite +
austenite
martensite
#1
STAGE
I
First (temperature gradient) Second (A -> M of surface) Third (A -* M of center)
Temperature -
Fig. 8-83
Schematic dilation curve lor marlensite lor-
mation.
.
i
mm
1
THE HEAT TREATMENT OF STEEL
Isile stress. .
.
;
.
.
.
.
339
If the piece was originally
v,iaped.
.
|4>Aered cubic), is a denser structure than
:
Center
Therefore, when austenite changes to ite, an expansion occurs. The austenite-
cooling
curve
.
jest and amounts to a volume increase of V expansion will be greater the lower the
A+F+C
F
C
:
s the changes in length, during cooling,
leasured in a dilatometer. The piece is :ure,
fi -
and normal contraction of the aus-
Tiperature is reached. Between the Ms F ustenite to martensite causes an expanature, the mapterisite undergoes normal
a
.
Surface
cooling curve
ned effect of temperature gradient and =s: (1) through-hardened steel and (2) :
>V.]nd center-cooling curves superimposed
1
ijgh-hardened steel Since the center>;)cooling rate, the part will be fully mar-
Mf
.
9 the first stage, to time f,, the stresses
isi
'
e gradient. The surface, prevented from
stage
2nd
stage
3rd stage
ay the cer\ter, will be In tension while the Time, log scale
ring the second stage, between times f,
Fig. 8-84 Center- and surface-cooling curves superimposed on the l-T dlngram lo illustrate the through-hardened
1 the Ms temperature, transforms to mar-
condition.
tensite and expands. The center, however, is undergoing normal contraction due to cooling. The center contracting will prevent the surface from expanding as much as it should, and the surface will tend to be in compres-
sion while the center will tend to be in tension.
.V/V'
Austenite
After f2, the surface has
reached room temperature and will be a hard, brittle, martensitic structure. During the third stage, the center finally reaches the /Ws temperature and begins to expand, forming martensite. The center, as it expands, will try to pull the surface along with it, putting the surface in tension. The stress condition in the three stages is summarized below. STRESS CONDITION STAGE
SURFACE
First (temperature gradient) Second (/\ /W of surface) Third (yA -> /W of center)
CENTER
Tension
Compression
Compression
Tension
Tension
Compression
-
;:
.
340
..
INTRODUCTION TO PHYSICAL METALLURGY
w li v '
To initiate and propagate a crack it is necessary for tensile stress to be
present. Let us examine the three stages with regard to the danger of | cracking. In the first stage, the surface is in tension; however, it is austen-
I
itic, and if the stress is high enough, rather than cracking, it will deform i
4e
plastically, relieving the stress. In the second stage, the center is in tension |
5
Center
cooling curve
and is austenitic, so that the tendency is to produce plastic deformation %|
rather than cracking. In the last stage, the surface is again in tension. Now however, the surface is hard, unyielding martensite. As the center expands there is little likelihood of plastic deformation. It is during this stage that ,
,
j
the greatest danger of cracking exists. Depending upon the difference in :| time between the transformation of the surface and center the cracking | may occur soon after the quench or sometimes many hours later. Figure |
|
8-85
shows schematically the type of failure that may occur
.
Isotherm Irajnsformationw.
a)
,
diagram\ $:
The crack will
take place in the tension layers and will be widest at the surface.
Surface
it win 9
gradually disappear when it gets to the compression layers on the inside. % Very rarely does one end up with two pieces, because the crack cannot be |
cooling curve
5
propagated through the compression layers. By a microscopic study of the crack in the cross section, it is possible to determine how much of the cross section was in tension and how much was in compression. One heattreating rule which minimizes the danger of cracking is that parts should be tempered immediately after hardening. Tempering will give the surface martensite some ductility before the center transforms. Another very effective method of minimizing distortion and cracking is
f
Time,
illu Fig. 8-86 Schematic transformation diagram S C martempering or marquenching. (From U S . "
by martempering or marquenching, illustrated in Fig. 8-86. It is carried out by heating to the proper austenitizing temperature, quenching rapidly in a liquid-salt bath held just above the Ms temperature, and holding for a period of time. This allows the surface and the center to reach the same temperature; air cooling to room temperature then follows. Since air cooling from just above the martensite-formation range introduces very little temperature gradient, the martensite will be formed at nearly the same time throughout the piece. Thus martempering minimizes residual stresses and greatly reduces the danger of distortion and cracking. The heat treatment is completed by tempering the martensite to the desired hardness.
.
Figure 8-87 shows the surfac on the l-T diagram for the sha up to time f,, the stresses prest and as in the through-harden while the center will be in com i;
1
Crack
r Tension loyers
if,:
.
Steels," U.S. Steel Corporation.)
-
times f, and t%, both the surfac transform to martensite while like pearlite. The entire piece
ing from the formation of mar formation of pearlite, the sur This tends to put the center in
After f2, the center will contrac ature to room temperature. reached room temperature m
tracting as much as it should l
Fig, 8-85
Sketch of possible fracture in a through-
hardened steel.
m m
\ \
the center.
page 342.
The stress com
THE HEAT TREATMENT OF STEEL
is necessary for tensile stress to be
341
ft.
stages with regard to the danger of J See is in tension; however, it is austen,
rather than cracking, It will deform*|
Ae
second stage, the center is in tension cy is to produce plastic deformation the surface is again in tension. Now,
Cenler coolina curve
f
t
,
ig martensite. As the center expands
3
r ,
ormation. It is during this stage that is. Depending upon the difference in
5
Slsothermol
the surface and center, the cracking
.
.
transformation
r sometimes many hours later
will be widest at the surface.
desired hardness
Surface
cooling
It will
he compression layers on the inside
Tempered to
diag ram
Figure failure that may occur. The crack will .
curve
.
pieces, because the crack cannot be
H layers. By a microscopic study of the v4ible to determine how much of the
M f
Product
.
much was in compression. One heatger of cracking is that parts should be ig. Tempering will give the surface :enter
'
\
Tempered martensite
transforms.
I
i
ilnimizing distortion and cracking is ustrated in Fig. 8-86. It is carried out g temperature, quenching rapidly in 5 Ms temperature, and holding for a
-
Fig. 8-86 Schematic transformation diagram illustrating martemperini) or marquenching. (Ffofn
"
USS .
.
.
Carilloy
Steels," U.S. Steel Corporation.)
jce and the center to reach the same jmperature then follows.
Time, log scale
,
Figure 8-87 shows the surface- and center-cooling curves superimposed on the l-T diagram for the shallow-hardened steel. During the first stage, up to time f,, the stresses present are due only to the temperature gradient, and as in the through-hardened condition, the surface will be in tension while the center will be in compression. During the second stage, between times f, and f2, both the surface and center will transform. The surface will
Since air
>ite-formation range introduces very isite will be formed at nearly the same smpering minimizes residual stresses
tortion and cracking. The heat treatnartensite to the desired hardness
:
.
transform to martensite while the center will transform to a softer product,
like pearlite. The entire piece is expanding, but since the expansion result-
>
l
i V :
VvV
.e
ing from the formation of martensite is greater than that resulting from the formation of pearlite, the Surface tends to expand more than the center. This tends to put the center in tension while the surface is in compression. After f2, the center will contract on cooling from the transformation temperature to room temperature. The surface, being martensitic and having reached room temperature much earlier, will prevent the center from contracting as much as it should. This will result in higher tensile stresses in the center. The stress condition in the three stages is summarized on page 342.
:
-
J
34:J
INTRODUCTION TO PHYSICAL METALLURGY
Cornpressio layers
STRESS CONDITION
-
STAGE
SURFACE
CENTER
First (temperature gradient)
Tension
Compression
Second [A -> M of surface A -> P of center)
Compression
Tension
Greater compression
Greater tension
C'acki
,
Tension
layers
Third (cooling of center to room temperature)
Fiy. 8-80
Sketch of possible fracture in a shallow-
hardenpd steel.
Let us examine the three stages with regard to the danger of cracking, In the first stage the surface is in tension, but being austenitic if the stress is high enough it will yield rather than crack. In the second stage the center is in tension However, since both the surface and center are ex,
,
,
,
: i'
.
panding, the tensile stress will be small. During the third stage
,
.
1
the surface
,
as a hard, rigid shell of martensite will prevent the center from contract-
ing as it cools to room temperatu reach a high value, and since th( strength, it is during this stage tt Figure 8-86 shows schematically shallow-hardened steel. The ten
,
the surface because of the comp
„e
they are internal, these cracks ? some cases Magnaflux inspectic sures. Very often these parts at the internal quenching cracks. ;
nler cooling
:urve
Surface
sile stress in the surface due t
cooling
through and the part will fail. In many applications, the tens
curve
is maximum at or near the sl
F-VC
hardened or case-hardened par
stresses are usually compressivi the residual compressive stresse
tively increases the strength of t greatly increased life have been face compressive stresses were
a)
were placed in service. 8-32 Hardenable
Carbon Steels
Cart
and have wider use than any o low cost. The low-carbon stee'
subjected to process annealing carbon steels have low harden h
1st stoge-«-
/,
'nc
3rd stage
5taqe
/2 Time, log scale
Fig. 8-87 Center- and surface-cooling curves superimposed on the l-T diagram to illustrate the shallow-hardened
transformation temperature (A recrystallization and some gra
condition.
I
\ \
quenching, the improvement in hardly worth the cost, and this annealing of these steels is ap mills to prepare it for forming c the work strains and permit fur at temperatures between the r
4"
a
THE HEAT TREATMENT OF STEEL
343
Compression
STRESS CONDITION
layers .
URFACE
CENTER
snsion
Compression
ompression
Tension
reater compression
Greater tension
Crack-
Tension
layers
Fig. 8-88
Sketch of possible fracture in a shallow-
hardened steel
with regard to the danger of cracking. / jsion, but being austenitic, if the stress than crack. In the second stage, the
I
.
-
ing as it cools to room temperature. The tensile stresses in the center may reach a high value, and since the center is pearlite of relatively low tensile strength, it is during this stage.that the greatest danger of cracking exists. Figure 8-88 shows schematically the type of failure that may occur in a
'
e both the surface and center are ex-
lall. During the third stage, the surface, will prevent the center from contract-
shallow-hardened steel.
The tensile crack is internal and cannot come to
the surface because of the compressive stress in the surface layers. Since they are internal, these cracks are difficult to detect. X-ray testing or in some cases Magnaflux inspection may show the presence of internal fis-
sures. Very often these parts are placed in service without knowledge of the internal quenching cracks. As soon as there is the slightest bit of tensile stress in the surface due to the external load, th
crack will come
through and the part will fail. In many applications, the tensile stress developed by the external force is maximum at or near the surface. For these applications, shallow-
c
hardened or case-hardened parts are preferred, since the surface residual stresses are usually compressive. In order for the surface to be in tension, the residual compressive stresses must first be brought to zero. This effectively increases the strength of the surface. The same beneficial effect ar.d greatly increased life have been found for leaf springs Where residual surface compressive stresses were induced by shot peening before the springs 8-32
}
were placed in service. Hardenable Carbon Steels
Carbon steels are produced in greater tonnage
and have wider use than any other metal because of their versatility and low cost. The low-carbon steels (0.10 to 0.25 percent carbon) are usually subjected to process annealing or case-hardening treatments. Since lowcarbon steels have low hardenability and form little or no martensite on quenching, the improvement in mechanical properties is so small that it is
hardly worth the cost, and this heat treatment is rarely applied. Process annealing of these steels is applied principally to sheet and strip at the mills to prepare it for forming or drawing or between operations to relieve the work strains and permit further working This operation is carried out ,
pie posed
.
at temperatures between the recrystallization temperature and the lower transformation temperature (A line). The effect is to soften the steel by recrystallization and some grain growth of the ferrite ;A stress-relieving ,
.
344
INTRODUCTION TO PHYSICAL METALLURGY
the Stillson type), and similar h<
treatment at 1000oF is applied to low-carbon cold-headed bolts, Th
quenching followed by tempering
lower temperature relieves much of the stress induced by cold workingl yet retains most of the strength and provides ample toughness.
when no reduction in the as-quenc
Case-
at 300 to 3750F is employed to hi
hardening of low-carbon steels was described in detail earlier in thi| chapter. The medium-carbon steels (0.25 to 0.55 percent carbon)
because op their higher carbon content, are generally used in the hardened and temlf pered condition. By varying the quenching medium and tempering temper ature a wide range of mechanical properties can be produced. They are thelte ,
most versatile of the three groups of carbon steels and are most commonly Jll
m
the Stillson-type wrenches, the ja usually quenched in water or brir 50 to.60. Either the jaws may be I
used for crankshafts, couplings, tie rods, and many other machinery partelp;
,
where the required hardness values are in the range of R/C 20 to 48 The K
may be heated all over and the brine. The entire part is then que .
remainder. Hammers require high what lower hardness on the claws
tempered on each end. Final h
.
medium-carbon steels are usually either normalized or annealed before hardening in order to obtain the best mechanical properties after harden-
ing and tempering. Cold-headed products are commonly made from these p steels, especially the ones containing less than 0.40 percent carbon Pro- W .
cess treatment before cold working is usually necessary because the 8 .
-
higher carbon decreases the workability. Frequently, a spheroidizing treat- J?
I
.
ment is used depending upon the application. Water is the quenching
Rockwell C 50 to 58; on the claw hatchets, and mower blades mus
toughness in their cutting edge, a For hardening, the cutting edges baths to the lowest temperature £
then quenched in brine. The final C 55 to 60.
medium most commonly used because it is the cheapest and easiest to
install. In some cases, where a faster quench is desired, brine or caustic C
QUESTIONS
soda solution may be used. When the section is thin or the properties re- w
quired after heat treatment are not high, oil quenching is used. This nearly £
8-1 0
Describe
completely the chang«
5 percent carbon steel from the au;
.
always solves the breakage problem and is very effective in reducing distortion; Many of the common hand tools, such as pliers, open-end wrenches
8-2
and screwdrivers are made from medium-carbon steels. They are usually
percent carbon, (d) 1.2 percent
quenched in water, either locally or completely, and then suitably tempered. High-carbon steels (above 0.55 percent carbon) are more restricted in their application since they are more costly to fabricate and have decreased machinability, formability, and weldability compared with medium-carbon steels. They are also more brittle in the heat-treated condition. Highercarbon steels such as 1070 to 1095 are especially suitable for springs, where resistance to fatigue and permanent set are required. Most of the
8-3 8-4
8-5
and martempering are often successfully applied to take advantage of the -
;
;J
-
.
Is it
possible to determine the
8 6 In Table 8-1, why do annealed ; 0 80 percent carbon? 8-7
In Table 8-1, why do normaliz
up to 1.2 percent and then a t
decrea
Define critical cooling rate. What factors influence the crit' 8-9 DeUne actual cooling rate. 8-10 What factors influence the ac 8-11 How is the actual cooling rat( 8-12
8-8
8 -13 Calculate the surface-area-toand compare it with a sphere of th£
at high hardness. It is important to remember that even with the use of a
8 14 Explain why the cooling rati
drastic quench, these steels are essentially shallow-hardening compared
temperature. Explain two ways in which tl 8-15
with alloy steels and that there is therefore a definite limitation to the size of section that can be hardened
\
carbi
What are the limitations on the What is the effect of increasin
considerably reduced distortion and in some cases the greater toughness ,
\
.
steel from microscopic study? Exp!
ing. Water quenching is used for heavy sections of the lower-carbon steels §:?
0 30 pel
transformation, (b) fineness of pea
parts made from steels in this group are hardened by conventional quench- || and for cutting edges, while oil quenching is for generalise. Austempering
Calculate the relative amounts o
cooled steels containing (a)
,
.
Screwdrivers, pliers, wrenches (except
show a straight horizontal
line,
a
THE HEAT TREATMENT OF STEEL
-.-
d low-carbon cold-headed bolts.
M vOf -
the Stillson type), and similar hand tools are usually hardened by oil
The
quenching followed by tempering to the required hardness range.
the stress Induced by cold working
Wefand provides ample toughness
.
Even
when no reduction in the as-quenched hardness is desired, stress relieving at 300 to 3750F is employed to help prevent sudden service failures. In the Stillson-type wrenches, the jaw teeth are really cutting edges and are usually quenched in water or brine to produce a hardness of Rockwell C 50 to 60. Either the jaws may be locally heated and quenched or the parts
Case
was described in detail earlier in this ;5 to 0.55 percent carbon), because of
ienerally used in the hardened and tem-
may be heated all over and the jaws locally time-quenched in water or brine. The entire part is then quenched in oil for partial hardening of the remainder. Hammers require high hardness on the striking face and somewhat lower hardness on the claws. They are usually locally hardened and
jenching medium and tempering temper-
jroperties can be produced. They are the
.
345
Df carbon steels and are most commonly xnhs rods, and many other machinery parts
.
:%::¥
tempered on each end. Final hardness on the striking face is usually Rockwell C 50 to 58; on the claws, 40 to 47. Cutting tools such as axes, hatchets, and mower blades must have high hardness and high relative
>s are in the range of R/C 20 to 48. The / either normalized or annealed before
est mechanical properties after harden-
toughness in their cutting edge, as well as the ability to hold a sharp edge. For hardening, the cutting edges of such tools are usually heated in liquid baths to the lowest temperature at which the piece can be hardened, and then quenched in brine. The final hardness at the cutting edge is Rockwell
iroducts are commonly made from these
ing less than 0.40 percent carbon. Pro-
ving is usually necessary because the ability. Frequently, a spheroidizing treat- * o annliratinn e application. Watpr Water ic is thp the nupnrhinn quenching i !
.
C 55 to 60.
'
cause it is the cheapest and easiest to
ster quench is desired, brine or caustic
the section is thin or the properties re- | fe
high, oil quenching is used. This nearly
!m and is very effective in reducing dis- -
|
5
QUESTIONS
'
.
8-1
0
.
Describe completely the changes that take place during the slow cooling of a 5 percent carbon steel from the austenite range.
8-2
ools, such as pliers, open-end wrenches. nedium-carbon steels. They are usually or completely, and then suitably tem-
Calculate the relative amounts of the structural constituents present in furnace-
cooled steels containing (a) 0.30 percent carbon, (b) 0.60 percent carbon, (c) 0.80
percent carbon, (d) 1.2 percent carbon, (e) 1.7 percent carbon. 8-3
What are the limitations on the use of the iron-iron carbide diagram?
8-4
. . . . . .
..
'
What is the effect of increasing cooling rate on (a) temperature of austenite transformation, (b) fineness of pearlite, (c) amount of proeutectoid constituent? 8-5 Is it possible to determine the approximate carbon content of a normalized
percent carbon) are more restricted in costly to fabricate and have decreased
? 5;?'e
dabillty compared with medium-carbon ' in the heat-treated condition. Higher)95 are especially suitable for springs, Brmanent set are required. Most of the p are hardened by conventional quench-
eavy sections of the lower-carbon steels inching is for general use. Austempering ssfully applied to take advantage of the id In some cases the greater toughness
.
o remember that, even with the use of a
ssentially shallow-hardening compared therefore a definite limitation to the size ;
Screwdrivers, pliers, wrenches (except
steel from microscopic study? Explain.
s t
why do annealed steels show a decrease in tensile strength above
8-6
In Table 8-1
0 80
percent carbon?
.
,
In Table 8-1 why do normalized steels show an Increase in tensile strength up to 1.2 percent and then a decrease? 8-7 8-8
,
Define critical cooling rate
.
8-9
What factors influence the critical cooling rate? Explain 8-10 Define actual cooling rate 8-11 What factors influence the actual cooling rate? Explain. 8-12 How is the actual cooling rate determined? 8'13 Calculate the surface area-to-mass ratio of a 2-in.-diameter 10-ft-long cylinder, and compare it with a sphere of the same mass. 8-14 Explain why the cooling rate of oil may be increased by increasing the oil .
.
-
temperature. 8-15 Explain two ways in which the hardness traverse curve of a given steel may show a straight horizontal line. -
}
346
INTRODUCTION TO PHYSICAL METALLURGY
8'16
REFERENCES
Explain why the surface hardness of quenched high-carbon steel may be lesi
I
than the hardness under the surface. 8'17
What are the advantages of specifying steel on the basis of hardenability?» 8-18 How will the microstructure differ for three samples of a 0.20 percent C steelM after the following heat treatments? '(a) Heated to 1700°F and furnace-cooled; heated to 1800oF and furnace-cooled; (c) heated to 1700oF and air-cooled .
8-19
How will the microstructure differ in four samples
steel after the following heat treatments?
of a 0.40 percent carbon l|
(a) Heated to 1500"F and air-cooled; f
(b) heated to 1500 F and oil-quenched; (c) heated to 1500oF and water-quenched; (d) heated to 1350'F and water-quenched. : 8-20
Sketch the l-T diagram of a 1080 steel (Fig 8-17) and (a) show a cooling curve * .
that will result in a structure of 50 percent martensite and 50 percent pearlite; (b)
,
"
"
Felbeck, D. K.:
|;
What will be the approximate hardness at the quenched end of (a) a 1050
hardenability test specimen?
"
Quenching of S
Ohio, 1959. "
Introduction to
Englewood Cliffs, N.J., 1968.
JF
L Describe how an l-T diagram is determined experimentally 8-26 What are the limitations on the use of the l-T diagram? K 8-27 What will be the hardness at the center and mid-radius position of (a) 2-in.diameter 4140 steel with a poor oil quench, strong oil quench, brine quench-no .
Grossmann, M. A.: Elements of I"
Park, Ohio, 1952. Principle and E. C. Bain: Metals Park, Ohio, 1964. "
-
Guy, A. G.: "Elements of Physic Company, Inc., Reading, Mass
agitation; (b) same as (a) for 5140 steel; (c) same as (a) and (b) for a 2V2-in.-diameter
:
bar; (tf) same as (a) and (b) for a 3-in.-diameter bar?
"Ph
ysical Metallurgy for
Inc., Reading, Mass., 1962. '
8'28 1
Steel and Its Heat
Clark, D. S., and W. R. Varney: Reinhold Company, New York, Crafts, W., and J. L. Lament: Hi
ability test. What was the approximate carbon content?
i hardenability test specimen; (b) a 6150 hardenability test specimen; (c) a 4150 .,
r
"
New York, 1948-1949.
ing Corporation, New York, 194
-
8-25
K
Brick, R. M., R. B. Gordon, and McGraw-Hill Book Company, N(
DuMond, 7. C:
8'24
"
Pa., 1959.
show a cooling curve that will result in a uniform pearlitic structure of Rockwell C40 8-23 A steel showed a hardness of Rockwell C 40 at the quenched end of a harden m .
'
:
Bethlehem Steel Corporation:
.
.
1
"
"Metals Handbook, vol. 2 "
Bullens, D. K.:
If the samples in Question 8-19 are 2 in in diameter, sketch the approximate hardness-traverse curves after the given heat treatments. 8-21 Give two different methods of obtaining a spheioidized cementite structure K 8-22
'
American Iron and Steel Institute: Hot Rolled and Cold Finished B American Society for Metals: Me-
Plot the change in hardness and Izod impact strength as a function of tempering from the data in Fig. 8-55. 8-29 What are the principal advantages of austempering compared with the con-
Hultgren, R.:
.
o
"
1966. '
The Strengthenu New York, 1964.
Peckner, D.:
,
"
Reed-Hill, R. E.: Physical Metal "
New York, 1964.
,
1900 and 2000oF?
Fundamental s o-
wood Cliffs, N.J., 1952 Hume-Rothery, W.: Structure
ventional quench and temper method? 8-30 What are the limitations of austempering? From the data in Table 8-7 plot case depih vs. time at 1500, 1600, and 1700CF. 8'31
What conclusions may be drawn from the shape of these curves? From the data in Table 8-7 plot case depth vs. temperature at 4, 10, and 20 h. 8'32 What conclusions may be drawn from these curves? 8-33 What are the limitations on the use of high carburizing temperatures such as
"
Republic Steel Corporation: il
"
Rogers, B. A.:
Hi
"
The Nature of
Ohio, 1951.
The Science of
8-34
What are the advantages of gas carburizing compared with p ack carburizing? I
Smith, C. O.:
8-35
Define hard case or effective case
8'36
Describe an application and the heat treatment used so that advantage may i
wood Cliffs, N.J., 1969 Smith, M. C: Alloy Series in Pt
be taken of the residual stresses resulting from heat treatment. 8-37
j
8-38
What will be the nature of the residual stresses after carburizing? What are the limitations of martempering?
8-39
You have been given a gear with a broken tooth that has failed prematurely
.
"
.
.
Explain.
in service. The normal heat treatment is carburlze, harden, and temper. Describe
completely how this gear would be studied to determine possible metallurgical cause
"
porated, New York,
.
i
.
"
"Suiting the Heat Treat Williams,, R. S., and V O. Honn Hill Book Company, New Yor :
-
.
for failure. 8-40
1956.
Corporation: Atlas (
U S Steel 1963.
Assume that a cold chisel is to be made of plain-carbon steel Analyze the application for properties required, select the hardness range desired, select the .
carbon content, and specify the heat treatment.
t
THE HEAT TREATMENT OF STEEL
347
REFERENCES
3S of quenched high-carbon steel may be less i
-
v-.
American Iron and Steel Institute: "Steel Products Manual Alloy Steel-Semifinished; Hot Rolled and Cold Finished Bars New York, 1970.
r' ficifylng steel on the basis of hardenability? :
,
"
sr for three samples of a 0.20 percent C steel
,
American Society for Metals: "Metals Handbook
la) Heated to 1700°F and furnace-cooled; (6) =|
:
,
(a) Heated to 1500oF and air-cooled;
Brick
-
4:;|n
.
,
heat treatments.
,
Clark, D. S., and W. R. Varney: "Physical Metallurgy for Engineers," Van Nostrand Reinhold Company, New York, 1962. Crafts, W., and J. L. Lamont: "Hardenability and Steel Selection," Pitman Publishing Corporation, New York, 1949. DuMond T. C: "Quenching of Steels," American Society for Metals, Metals Park,
steel (Fig. 8-17) and (a) show a cooling curve
rcent martensite arid 50 percent peariite; {b) a uniform pearlitic structure of Rockwell C 40.
,
Ickwell C 40 at the quenched end of a harden-
Ohio, 1959.
3 carbon content?
Feibeck, D. K.: "Introduction to Strengthening Mechanisms," Prentice-Hail, inc., Englewood Cliffs, N.J., 1968. Grossmann, M. A.: "Elements of Hardenability," American Society for Metals, Metals
lardness at the quenched end of (a) a 1050 50 hardenability test specimen; (c) a 4150
i
Park, Ohio, 1952.
determined experimentally. .
'
;
i
use of the l-T diagram?
and E
.
].
|
uench, strong oil quench, brine quench-no
|;
(c) same as (a) and (b) for a 2V2-in.-diameter
f
.diameter bar?
f
Izod impact strength as a function of temper-
wood Cliffs, N.J., 1952. 1966.
Peckner, D.: "The Strengthening of Metals," Van Nostrand Reinhold Company, .
New York, 1964.
the shape of these curves?
Reed-Hill, R. E.: "Physical Metallurgy Principles," Van Nostrand Reinhold Company,
iiXri case depth vs. temperature at 4, 10, and 20 h. V
'
-
these curves?
t
use of high carburizing temperatures such as
i
carburizing compared with pack carburizing? 3se.
|: I
Smith, C. O.: "The Science of Engineering Materials," Prentice-Hall, Inc., Engle-
|
Smith, M. C: "Alloy Series in Physical Metallurgy," Harper & Row, Publishers, Incor-
e heat treatment used so that advantage may Iting from heat treatment.
residual stresses after carburizing? Explain.
tempering?
lith a broken tooth that has failed prematurely
nt is carburize, harden, and temper. Describe rjdied to determine possible metallurgical cause
'
-
New York, 1964.
Republic Steel Corporation: "Heat Treatment of Steels," Cleveland, Ohio. Rogers, B. A.: "The Nature of Metals," American Society for Metals, Metals Park,
,
"
"
Hume-Rothery, W.: "Structure of Alloys and Iron," Pergamon Press Inc., New York,
?
.
Guy, A. G.: "Elements of Physical Metallurgy," 2d ed., Addison-Wesley Publishing Company, Inc., Reading, Mass., 1959. ; Physical Metallurgy for Engineers," Addison-Wesley Publishing Company,
Hultgren, R.: "Fundamentals of Physical Metallurgy," Prentice-Hall, Inc., Engle-
empering?
;
_
Inc., Reading, Mass., 1962.
f
jes of austempering compared with the con-
case depTh vs. time at 1500,1600, and 170boF
C. Bain: "Principles of Heat Treatment," American Society for Metals,
Metals Park, Ohio, 1964.
vie center and mid-radius position of (a) 2-in.-
::
;
,
New York 1948-1949.
,
btaining a spheroidized cementite structure.
X:
.
'
f
|
Ohio, 1951.
wood Cliffs, N.J., 1969.
porated, New York, 1956. US .
r
P
.
Steel Corporation: "Atlas of Isothermal Transformation Diagrams," Pittsburgh,
1963. :
| "
Suiting the Heat Treatment to the Job," Pittsburgh, Pa. 1967. Williams, R. S., and V. O. Homerberg: "Principles of Metallurgy 5th ed., ,
"
,
Hill Book Company
,
.
o be made of plain-carbon steel. Analyze the select the hardness range desired-, select the treatment.
'
1 i '
.
i
" ,
,
are 2 in. in diameter, sketch the approximate .
R. M., R. B. Gordon, and A. Phillips: "Structure and Properties of Alloys
,
McGraw-Hill Book Company New York, 1965. Bullens D. K.: "Steel and Its Heat Treatment," vols. 1 to 3, John Wiley & Sons Inc.,
led.
>
,
,
Pa., 1959.
J; (c) heated to 1500oF and water-quenched;
'"
1948 ed.. Metals Park Ohio.
,
"
in four samples of a 0.40 percent carbon
mts?
Metals Handbook
"
vol. 2, 1964, and vol. 7, 1972, Metals Park Ohio. Bethlehem Steel Corporation: "Modern Steels and Their Properties Bethlehem,
;(c) heated to 1700oF and air-cooled.
|er
"
,
"
New York, 1948.
McGraw-
I
m
ALL 9 1 Introduction
,
I
f
Plain-carbon stee
other requirements are not to ordinary temperatures and in but their relatively low harde tained except in fairly thin se
pered to reduce internal stress that plain-carbon steels sho\ pering temperature. This bet that require hardness above r plain-carbon steels may be o\ An alloy steel may be define due to some element other th
contain moderate amounts o'
silicon (up to about 0.30 pe because the principal functio deoxidizers. They combine v effect of those elements.
9 2 Purpose of Alloying Alloying e poses. Someof the most imp' 1
Increase hardenability
2 3 4
Improve strength at ordinary Improve mechanical propertic Improve toughness at any mi
5 6
Increase wear resistance Increase corrosion resistancs
-J
7
Improve magnetic properties
J
distributed in the two main <
i
Alloying elements may be '
.
I
Group 1 Group 2
Elements which dissc Elements which com!
if '
if
i: i
3
i
: i
.
ALLOY STEELS
3
t
if 9-1 Introduction
Plain-carbon steels are very satisfactory where strength and J
other requirements are not too severe. They are also used successfully at; .
ordinary temperatUrfeS and in atmospheres that are not highly corrosive|i||
but their relatively low hardenability limits the strength that can be ob ||
tained except in fairly thin sections. Almost all hardened steels are tem-:?g pered to reduce internal stresses. It was pointed out in the previous chapter f
m
'
that plain-carbon steels show a marked softening with increasing tem-i pering temperature. This behavior will lessen their applicability for parts that require hardness above room temperature. Most of the limitations of:*! plain-carbon steels may be overcome by the use of alloying elements.
An a//oy steel may be defined as one whose characteristic properties are4® due to some element other than carbon. Although all plain-carbon steels contain moderate amounts of manganese (up to about 0.90 percent) and silicon (up to about 0.30 percent), they are not considered alloy steels because the principal function of the manganese and silicon is to act as deoxidizers. They combine with oxygen and sulfur to reduce the harmful i
effect of those elements.
1
9-2
Purpose of Alloying
Alloying elements are added to steels for many pur-
poses. Some of the most important are: 2
Increase hardenability Improve strength at ordinary temperatures
1 r
1
3
Improve mechanical properties at either high or low temperatures
4
Improve toughness at any minimum hardness or strength
5
Increase wear resistance
6
Increase corrosion resistance
7
Improve magnetic properties
Alloying elements may be classified according to the way they may be distributed in the two main constituents of annealed steel
.
Group 1 Group 2 .
5SV
Elements which dissolve in ferrite
Elements which combine with carbon to form simple or complex carbides
:
m
350
INTRODUCTION TO PHYSICAL METALLURGY
9-3
Effect of Alloying Elements upon Ferrite Technically, there is probably some solubility of all the elements in ferrite, but some elements are not found extensively in the carbide phase. Thus nickel, aluminum, silicon,
??o Silicon Man
200
copper, and cobalt are all found largely dissolved in ferrite. In the absence
of carbon, considerable proportions of the group 2 elements will be found Wk dissolved in ferrite.
Therefore, the carbide-forming tendency is apparent only when there is a significant amount of carbon present. The behavior WL
180
of the individual elements is shown in Table 9-1, and the relative tendency ijf of certain elements to exist in both groups is shown by the size of the IB60
arrowhead. jK Any element dissolved in ferrite increases its hardness and strength in |
t
'
n
accordance with the general principles of solid solution hardening. The
'
X3
'
*
40
order of increasing effectiveness in strengthening iron, based upon equal additions by weight, appears to be about as follows: chromium, tungsten, vanadium, molybdenum, nickel, manganese, and silicon (Fig. 9-1). The
/ cn
m
I
/
120
hardening effect of the dissolved elements is actually small and illustrates how relatively little is the contribution of the strengthening of the ferrite
to the overall strength of the steel. This is shown in Fig. 9-2 for low-carbon .
chromium alloys.
100
The upper curve indicates the influence of chromium
/
to change the tensile strength by changing the structure, while the lower curve Indicates the minor influence of chromium in essentially constant
t
80
it 94 Effects of Alloying Elements upon Carbide structures.
.i!
The influence of the amount of and the*form and dispersion of the carbide on the properties of steel has been discussed in Chap. 8. Since all carbides found in steel are hard and brittle, their effect on the room-temperature tensile properties isyi§imilar regardless of the specific composition. carbide
,
60
o
2
!
4 Percenf alloying e
Fiq 9.1
Probable hardening
effect of the vario
as dissolved in alpha iron. (From E, C. Bain anc American Paxton, Alloying Elements in Steel, "
*
"
i TABLE 9-1
Behavior of the Individual Elements in Annealed Steel*
ALLOYING ELEMENT
.
I
i
m
.
The presence of elements 1
GROUP 1
GROUP 2
DISSOLVED
COMBINED IN CARBIDE
temperature and soaking time
Nickel
Ni
and tend to remain out of so carbon and alloy content of ;
Silicon
Si
Undissolved carbides also act
Aluminum
Al
reduce hardenability.
Copper Manganese
Cu
elements are very powerful
IN FERRITE
Mn
4
Mn
Chromium
Or
Or
Tungsten Molybdenum
W
W
Vanadium
V
Titanium
Ti
Mo -
-
. Adaptej)
V Ti
from E. C. Sain and H. W. Paxton, "Alloying Elements in Steel
for Metals. Metals Park, Ohio, 1961.
d
mium and vanadium carbide
Mo <
When
While all the carbides four
-
I
tor Metals, Metals Park, Ohio, 1961.) -
'
2d ed.. American Society
sistance. The hardness and ' are in a large measure deter these hard particles. These composition, method of mar
m -.
-
11
.
ALLOY STEELS
351
I i
'
I-:-:
there
Technically
,
is
220
proba
Sfi errite, but some elements are n ot
1
Silicon
Thus nickel, aluminum, silicon |l '
.
,
200
dissolved in ferrite. In the absence '
Mongonese
m
the group 2 elements will be found
j)ide-forming tendency is apparent
1
180
t of carbon present. The behavior
able 9-1, and the relative tendency oups is shown by the size of the s
Molybdenum
s
jases its hardness and strength in '
n
of solid solution hardening. The 140
mgthening iron, based upon equal jt as follows: chromium tungsten, anese, and silicon (Fig. 9-1). The
: S
vanodium
,
'
.
?
CD
nts is actually snhall and illustrates of the strengthening of the ferrite is shown in Fig. 9-2 for low-carbon
120
Jicates the influence of chromium
100
} "
....
f
I
Chromium
I
ging the structure, while the lower chromium in essentially constant 80
e
The influence of the amount of
)f the carbide on the properties of
I
'
SO
ince all carbides found in steel are
|om-temperature tensile properties
0
4
2
6
8
10
12
Percent alloying element, in alpha iron
Fig. 9-1 Probable hardening effect ot the various elements as dissolved in alpha iron. (From E. C. Bain and H. W. Paxton, "Alloying Elements in Steel," American Society .
nposition.
I
for Metals, Metals Park, Ohio, 1961.)
I
in Annealed Steel*
The presence of elements that form carbides influences the hardening
iROUP 2
temperature and soaking time. Complex carbides are sluggish to dissolve
lOMBINED IN CARBIDE
and tend to remain out of solution in austenite.
This serves to lower the
carbon and alloy content of austenite below that of the steel as a whole. -
Undissolved carbides also act to reduce grain growth. Both effects tend to reduce hardenability. When dissolved in austenite, the carbide-forming i
'
; ; :r -
.
elements are very powerful deep-hardening elements. While all the carbides found in steel are hard brittle compounds, chromium and vanadium carbides are outstanding in hardness and wear resistance. The hardness and wear resistance of alloy steels rich in carbides are in a large measure determined by the amount, size, and distribution of
Mn
UiCr '
:
.
w Mo
IV
fTi lements in SteBl," 2d ed
.
these hard particles. These factors, in turn, are controlled by chemical .
American Society
.
t
Tungsten
: .
m
Nickel
160 a)
.
.
f J
v|
composition, method of manufacture, and heat treatment.
m-MM
J
1 352
INTRODUCTION TO PHYSICAL METALLURGY
carbon content of the eutectoid
220
the eutectoid temperature (Fig. 200
manganese may lower the critic. transformation of austenite on si
180h
stabilizing elements. Therefore, i
Air cooled
ature. This situation occurs in tf
160
Certain alloying elements, not;
07
titanium, in increasing amounts,
g M0
and enlarge the field in which ; change is shown in Fig. 9-4, whe
120
of the austenitic field with incr 100
-
so
Alloy compositions to the right with increasing amounts of carb
hurnoce cooled
austenite with more or less ferr
9 6 Effect of Alloying Elements in Ten
6C
of plain-carbon steels, it was si reheating. As the tempering ten continuously. The general effec ening rate, so that alloy steels v -
0
Fig. 9-2
1
2 3 4 5 Percent chromium content
6
The minor effect of chromium in annealed
steels
compared with its powerful effect as a strengthener through its Influence on structure in air-cooled steels (From E. C Bain and H W. Paxton, "Alloyiny Elements in Steel American Society for Metals Metals Park, Ohio 1961.) .
.
"
.
,
,
,
T
Si
2l00r
9-5
Influence of Alloying
Elements on the Iron-Iron Carbide Diagram Although
no mention was made in Chap 7 of the possible modification of the iron.
iron carbide diagram by the presence of elements other than carbon
determining the effects of alloying elements this possibility must be
,
in
fK
given j|
due consideration When a third element is added to steel the binary iron,
iron carbide diagram no longer represents equilibrium conditions Although the construction and interpretation of ternary equilibrium diagrams S .
are outside the scope of this book the presence of alloying elements will H ,
change the critical range the position of the eutectoid point ,
,
and the loca-
tion of the alpha and gamma fields indicated by the binary iron-iron car-
a
-
I 1
1900 -
1700
1500 h
1300 -
bide diagram.
Nickel and manganese tend to lower the critical temperature on heating
,
wnile molybdenum aluminum, silicon, tungsten and vanadium tend to ,
raise it. The change in critical temperature produced by the presence of alloying elements is important in the heat treatment of alloy steels since it
will either raise or lower the proper hardening temperature as compared with the corresponding plain-carbon steel
.
The eutectoid point is shifted from the position it normally has in the iron-iron carbide diagram All the alloying elements tend to reduce the .
m
m
4.
hoc
,
C
?
1
()
Wciijhl porconl (illoyiiKj elomo
Fiy 9 3 Eutectoid composition and eutectoid I as influenced by several alloying elements. (Fro Bain and H. W. Paxton, Alloying Elements in S American Society lor Metals. Metals Park, Ohio "
ALLOY STEELS
353
carbon content of the eutectoid, but only nickel and manganese reduce the eutectoid temperature (Fig. 9-3). lncreasing ar m ofj kel and _
_
_
manganese may lower the critical tetriperature sufficiently to prevent the transformation of austenite on siovv cooling; they are knoMLag austenitestabilizing elements. Therefore, auster e wijib .
ature. This situation occurs in the austenitic stainless steels.
Certain alloying elements, notablyjriolybdenum. chromium, silicon, and titanium, in increasing amounts, tend to contract the pure aust itjc reaion and enlarge the field in whichja!|3haja) or delta (8) iron is found. This change is shown in Fig. 9-4, where the solid lines represent the contraction of the austenitic field with increasing amounts of the alloying element. Alloy compositions to the right of the "triangles" will be largely austenite with increasing arhburtts of carbide while to the left of the austenite areas, ,
9-6
6
d steels ;
-.
;!-r
"
x;
.
through
austenite with more or less ferrite (solutions in a or S iron) will be found. Effect of Alloying Elements In Tempering In the discussion of tempering of plain-carbon steels, it was shown that hardened steels are softened by reheating. As the tempering temperature is increased, the hardness drops continuously. The general effect of alloying elements is to retard the softening rate, so that alloy steels will require a higher tempering temperature
T! e. 61
)
.
Mo
Ti
Si
2 IOC
lie Iron-Iron Carbide Diagram
Although
T of the possible modification of the iron-
A
Isence of elements other than carbon, in
LL
ig elements this possibility must be given element is added to steel, the binary iron
-
represents equilibrium conditions. Al
-
>rpretation of ternary equilibrium diagrams ok, the presence of alloying elements will sition of the eutectoid point, and the loca ilds indicated by the binary iron iron car-
1900
S 1700 -
|
0 80
I
.
1
Cr
,
t 1500 -/
1
0 6C .
II
<
_
>
Cr 1300
-
01c
Mn
.
Mr
i lower the critical temperature on heating,
Si
0 2C
Ni
1100 -
.
silicon, tungsten, and vanadium tend to
.
temperature produced by the presence of
i
c
'
n the heat treatment of alloy steels since It
Voper hardening temperature as compared irbon steel.
c
2
4
6
8
IC
Weight percent olloying element
Fig. 9-3
Eutectoid composition and eutectoid temperature
as influenced by several alloying elements (From E. C, Bain and H. W. Paxton "Alloying Elements in Steel,
"
,
imerican Society for Metals Metals Park, Ohio, 1961.) ,
0
2
4
6
8
Weight percent olloying element
.
d from the position it normally has in the i the alloying elements tend to reduce the
vio
II
10
354
INTRODUCTION TO PHYSICAL METALLURGY
.
g
and their application will be give
lJ|f-7 Nickel Steels (2xxx Series) Nickel
2600h 5
2400h
steel-alloying elements. It has l is highly soluble in ferrite, contr
2
\7
2200
this phase. Nickel lowers the ( temperature range for successfL
12
19%
tion of austenite, and does noff
5
2000h
157, Cr
:
1800
to dissolve during austenittemg. of the eutectoid; therefore, the s tains a higher percentage of pe steels. Since the pearlite form
12% Cr
15
s
57. Cr
12
1600 p-
a brief consideration of the spec
m
2800
tougher than the pearlite in un
\
ii
5
' '
/
0%
attainment of given strength
Cr
1400 h 0
:
1200
'
0
.0.2
0.4
0.6
ing toughness, plasticity, and fa
/ .
0.8
leve
suited for high-strength structu 1.0
1.2
14 .
16 .
1 8 .
Fig. 9-4 Range of austenite in chromium steels. (From Metals Handbook," 1948 ed, American Society for Metals, Metals Park, Ohio.)
50
"
50
to obtain a given hardness. The elements that remain dissolved in ferrite ; such as nickel, silicon, and to some extent manganese, have very little ,
:
effect on the hardness of tempered steel. The complex carbide-forming elements such as chromium tungsten, molybdenum, and vanadium, however, have a very noticeable effect on the retardation of softening. Not only do they raise the tempering temperature but when they are present in higher percentages, the softening curves for these steels will show a range in which the hardness may actually increase with increase in tempering temperature. This characteristic behavior of alloy steels containing carbide-forming elements is known as secondary ,
,
7D
30 S
hardness and is believed to be due to delayed precipitation of fine alloy carbides. The effect of increasing chromium content is illustrated in Fig. 9-5. The specific effects of the alloying elements in steel are summarized In Table 9-2. 10 200
Classification of alloy steels according to chemical composition as in the AISl series has been covered in Chap. 7, and compositions of some representative alloy steels are given in Table 8-2.
The preceding sections were devoted to a general discussion of the alloying elements in steel. Since a large number of alloy steels are manufactured it is not feasible to discuss the individual alloy steels; however ,
m
m
,
500
too
I?
! Tempering
Fig. 9-5 The softening, with increasing
iure.
a
temper
of quenched 0.35 percent carbon steels £
C Bain and H. ' American Society
by chromium content. (From E. "
Ailoying Elements in Steel,
-
Metals Park, Ohio, 1961.)
.
$1% Y
ALLOY STEELS
355
a brief consideration of the specific effects of the common alloy elements and their application will be given. 9-7
Nickel Steels (2xxx Series)
Nickel is one of the oldest, most fundamental
steel-alloying elements. It has unlimited solubility in gamma (y) iron and is highly soluble in ferrite, contributing to the strength and toughness of this phase. Nickel lowers the critical temperatures of steel, widens the temperature range for successful heat treatment, retards the decomposition of austenite, and does not form any carbides which might be difficult to dissolve during austenitizing. Nickel also reduces the carbon content
5
of the eutectoid; therefore, the structure of unhardened nickel steels con-
tains a higher percentage of pearlite than similarly treated plain-carbon steels. Since the pearlite forms at a lower temperature, it is finer and
s
tougher than the pearlite in unalloyed steels. These factors permit the attainment of given strength levels at lower carbon contents, thus increasing toughness, plasticity, and fatigue resistance. Nickel steels are highly suited for high-strength structural steels which are used in the as-rolled
no
S
0% Cr
1L 16
L .
o
.
12 .
14
.
.
.
1,
.
18 .
From
60
)x Metals,
.
1
:
c
50
! elements that remain dissolved in ferrite,
some extent manganese, have very little = 27, Cr
red steel.
1 elements such as chromium, tungsten,
v>
wever, have a very noticeable effect on the ly do they raise the tempering temperature, jher percentages, the softening curves for which the hardness may actually increase perature. This characteristic behavior of forming elements is known as secondary
47, Cr
40
2
Cr
0 57 Cr .
0% Cr
1 30 en
due to delayed precipitation of fine alloy | asing chromium content is illustrated in
20
lloying elements in steel are summarized MMi
\
4
according to chemical composition as In | ed in Chap. 7 and compositions of some ,
..
jiven in Table 8-2.
devoted to a general discussion of the alloya large number of alloy steels are manuscuss the individual alloy steels; however J ,
.
. .
i? -
,
10
200
400
600
800
1000
Tempering temperature, "F Pg. 9-5 The sottening, with increasing tempering temperature, of quenched 0.35 percent carbon steels as influenced
[by Chromium content. (From E. C. Bain and H. W, Paxton, |"Alloying Elements in Steel," American Society for Metals, Metals Par<, Ohio, 1961.)
1200
1400
\
11
111 TABLE 9-2 Specific Effects of Alloying Elements in Steel* INFLUENCE EXERTED THROUGH SOLID SOLUBILITY ELEMENT
Aluminum
IN ALPHA
IRON
IRON
1 1%
36%
CARBIDE-
(HARDENABILITY)
Hardens consid-
Increases hard-
(increased
erably by solid
by C)
solution
enability mildly, if
.
O)
PRINCIPAL FUNCTIONS
AUSTENITE
FERRITE IN GAMMA
u
CARBIDE
INFLUENCE ON
INFLUENCE ON
ACTION 33
FORMING
DURING
TENDENCY
TEMPERING
O
Negative (graphitizes)
O C 1
o
Deoxides efficiently Restricts grain growth (by forming dispersed oxides or
.
2
.
dissolved in
o z I
nitrides)
austenite
3
-
o
Alloying element in nitriding
.
steel <
Chromium
12.8%
Unlimited
(20% with 5% C)
0
Hardens slightly;
Increases hard-
Greater than
Milc'y . exists
increases corro-
enabllity moderately
Mn; less
softening
than W
sion resistance
.
1
Increases resistance to cor-
.
o >
rosion and oxidation .
3
-
r
Increases hardenability Adds some strength at high
2
.
S m
temperatures 4
>
Resists abrasion and wear
,
r;
(with high carbon) Cobalt
Manganese
Unlimited
Unlimited
75%
3%
Similar to Fe
Decreases hard-
Hardens consid-
Sustains
Contributes to red-hardness
1
.
enability as
solution
dissolved
Hardens
Increases hard-
Greater than
Very little
markedly;
enability
Fe; less
in usual
moderately
than Cr
percentages
2
Opposes softening, by secondary + hardening
1 Raises grain-coarseiVing tem-
by hardening ferrite
solid solution ,
1
3%±
37.5% (less
(8% with 0 3% C)
with lowered
.
temperature)
Provides agehardening system in high Mo-Fe alloys
Increases hard-
Strong;
enability strongly
greater than Cr
(Mo > Cr)
i
Counteracts brittleness Irom
.
.
the sulfur
Increases hardenability inexpenslve(y
.
.
ticity somewhat Molybdenum
f
Q <
erably by solid
reduces plas-
hardness by
I
c 33
perature of austenfte 2
Deepens harden/ng Counteracts tendency toward
.
3
.
4
.
temper brittleriess
Raises hot arid creep strength
.
i:
red-hardness 5
Enhances corrosion resistance
.
in stainless steel
Foms abrasion-resisting par-
6
.
ticles Nickel
Unlimited
10%
increases hard-
Negative
(irrespective
Strengthens and toughens
enability
(graphirizes)|
ol carbon
by solid
mildlyt but
-
|
iMPt-OCNCE ON BnFUTE
ELEMENT
-
v
1
Phosphorus
0 5% .
Toughvna poarmic-fsrrltlo
WflilK O* on
Strengthens unquenched or
PfWNCifAl. rUNCTlONS
AUSTEN.TE
(HARDENABIMTYl IRON
annealed steels
pftrcontaooa
wm
m
IN GAMMA
Very little in small
IN ALPHA ,
|
RON
2 8%
Hardens
Increases hard-
strongly by solid solution
enability
CAPBICE-
ACTION
FORMiNG
DURING
TENDENCY
TEMPERING 1
Mi-
.
(irrespective of carbon
2
Strengthens low-carbon steel
.
Increases resistance to corro-
.
sion 3
Improves machinability in free-cutting steels
.
content)
Silicon
Hardens with
ncreases hard-
2%±
18.5% (not
(9% with
much
oss in plasticity
enability
changed
(Mn < Si < P)
moderately
0
.
35% C)
Negative
Sustains
Used as general-purpose de-
[graphitizes)
hardness by
oxidizer
solid solution
2
.
Alloying element for electrical and magnetic sheet
by carbon)
3
.
4
.
Improve oxidation resistance Increases hardenability of
steels carrying nongraphitizing elements 5
Strengthens low-alloy steels
1
Fixes carbon in inert particles
.
Titaniurr
0 75% .
6% =: (less
Provides age-
Probably in-
Greatest
Persistent
twrtenina
creases harden-
known
carbides
a
.
Reduces marlensitic hard-
!
It Cobalt
Manganese
Moiybdsnum
Unlimited
Unlimited
75%
3%
3%±
37.5% (less
(8% with 0 3% C)
with lowered
.
temperature)
.
hi
Hardens consid-
Decreases hard-
erably by solid
enability as
hardness by
solution
dissolved
solid solution
Hardens
Increases hard-
Greater than
Very little.
markedly; reduces plasticity somewhat
enability moderately
Fe; less
in usual
than Cr
percentages
2
Provides agehardening system in high Mo-Fe alloys
Increases hard-
Strong; greater
strongly
than Cr
Opposes softening, by secondary hardening
1
enability
Similar to Fe
(Mo > Cr)
Sustains
vX .
.
1
3)
Contributes to red-hardness
.
by hardening ferrite Counteracts brittleness from
1
the sulfur
\,
Increases hardenability inexpensively
.
Raises grain-coarsening temperature of austenite
2
.
3
.
Deepens hardening
Counteracts tendency toward temper brittleness
4
Raises hot and creep strength
.
,
red-hardness 5
.
6
.
Enhances corrosion resistance in stainless steel
Forms abrasion-resisting particles
Nickel
Unlimited
10%
Strengthens and toughens
Increases hard-
(irrespective of cartion
by solid
mildly, but
enability
otution
Negative (graphitizes)
Very little in
1
Strengthens unquenched or
.
annesle
small «r<
steels
nla«M
una* to .
.
m f
ELEMENT
ihid tMStat .K INFLUENCE ON
FERRITE
INFLUS.VCE
Phosphorus
PBINOPAfc
AUSTENITE
(HARDENABILITY)
CARBIDE-
ACTION
IN GAMMA
IN ALPHA
FORMING
DURING
IRON
IRON
TENDENCY
TEMPERING
0 5%
2 8%
Hardens
Increases hard-
(irrespective
strongly by
enability
Df carbon
solid solution
,
.
Nil
sion
18.5% (not
Hardens with
Increases hard-
Negative
Sustains
(9% with 0 35% C)
much
loss in plasticity (Mn < Si < P)
enability moderately
(graphitizes)
hardness by
changed
Increases resistance to corro-
.
2%±
.
Strengthens low-carbon steel
.
content)
Silicon
1
2
solid soiLftion
3
.
1
.
2
.
Improves machinability in free-cutting steels Used as general-purpose deoxidizer
by carbon) 3
.
4
.
5
.
1
.
Alloying element for electrical and magnetic sheet Improve oxidation resistance Increases hardenability of steels carrying nongraphitizing elements
Titanium
Strengthens low-alloy steels Fixes carbon in inert particles
0 75%
6% = (less
Provides age-
Probably in-
Greatest
Persisterit
(1% ±
with lowered
hardening
creases harden-
known
carbides
with
temperature)
system in high Ti-Fe alloys
ability very strongly as dis-
(2% Ti ren-
probably
ness and hardenability in
ders 0.50%
unaffected.
medium-chromium steels
solved.
carbon steel
Some
unharden-
secondary hardening
.
0
20% C)
.
The
carbide effects
reduce hard-
able)
a
.
b
.
Reduces martensitic hartJ-
Prevents formation )f austenite in high-chromium
> .
c
enability
_
.
ii;
Prevents localized depletion of chromium in stainless
steel during long heating
! Tungsten
Strong
Opposes
33% (less
Provides age-
Increases hard-
(11% with
with lowered
26% C)
temperature)
hardening system in high W-Fe alloys
enability strongly in
secondary
small amounts
hardenfng
6%
0
.
1
.
Forms hard, abrasion-resistant
particles in tool steels
softening by 2
.
Promotes hardness and
strength at elevated temperature
Vanadium
1%
(4% with 0 20% C) .
Unlimited
Hardens moder-
Increases hard-
Very strong
Maximum
ately by solid
enability very strongly, as
(V < Ti or Cb)
for
solution
dissolved
secondary hardening
.
steels
1
2
3
.
.
.
Elevates coarsening temperature of austenite (promotes fine grain) Increases hardenability (when dissolved) Resists tempering and cs .'ses marked secondary hardening
> r;
O <
-
CO -i m
m
ai CO en
358
INTRODUCTION TO PHYSICAL METALLURGY
i:; condition or for large forgings which are not adapted to quenching Tti |,
ability and wear resistance. It is
.
3
.
5 percent nickel steels (23xx series) with low carbon are used extensive| |
effect of two or more alloying el than the sum of the effects of th
'
;
for carburizing of drive gears
,
connecting-rod bolts, studs, and kingpitnl p'
The 5 percent nickel steels (25xx series) provide increased toughness an| W
i
The low-carbon nickel-chromi
are used for heavy-duty applications such as bus and truck gears camsj | and crankshafts Nickel has only a mild effect on hardenability but is out-' t1 standing in its ability to improve toughness particularly at low tem» !
mium supplies the wear resistanc improve the toughness of the cc
,
.
,
cent chromium (31 xx series) th etc. For heavy-duty application:
;
peratures.
While nickel steels of the 2xxx series have been deleted from the AISI i SAE classification of standard alloy steels it does not mean that they are ' not manufactured. Removal from the classification simply means thatth» t
the nickel content is increased t(
-
1
,
tonnage produced is below a certain minimum '
9-8
The steels in this series p have been largely replaced in many applications by the lower-cost triple-
alloy steels of the 86xx series Chromium Steels (Sxxx Series)
placed by the triple-alloy steels lower cost.
C3j Cr4C) or complex car-r
These carbides have high hardness and good wear*:
resistance. Chromium is soluble up to about 13 percent in y iron and has;f unlimited solubility in a ferrite In low-carbon steels chromium tends to go into solution thus increasing the strength and toughness of the ferrite! When chromium is present in amounts in excess of 5 percent the.high-
also reduces the tendency tow;
,
from the presence of sulfur, the
,
When manganese is absent or
,
temperature properties and corrosion resistance of the steel are greatly improved.
which forms a eutectic with ire
The plain-chromium steels of the 51 xx series contain between 0 15 and 0 64 percent carbon and between 0.70 and 1 15 percent chromium. The'! low-carbon alloy steels in this series are usually carburized The presence1 of chromium increases the wear resistance of the case but the toughness in the core is not so high as the nickel steels; With medium carbon these f .
.
.
.
,
,
steels are oil-hardening and are used for springs engine bolts, studs, axles, etc. A high -carbon (1 percent) high-chromium (1 5 percent) alloyL steel (52100) is characterized by high hardness and wear resistance Thl$« steel is used extensively for ball and roller bearings and for crushing machinery. A special type of chromium steel containing 1 percent carbon and 2 to 4 percent chromium has excellent magnetic properties and is used for ,
.
.
permanent magnets.
The high-chromium steels containing over 10 percent chromium-are noted for their high resistance to corrosion and will be discussed in
Sec. 9-15.
.
.
I
Jj- 9-9 Nickel-chromium Steels (3xxx Series) In these steels the ratio of nickel f to chro:mium is approximately 2V2 parts nickel to 1 part chromium A combination of alloying elements usually imparts some of the characteristic properties of each one. The effect of nickel in increasing toughness and .
ductility is combined with the effect of chromium in improving harden | -
\
\
m i i
The very high nickel-chromiur
V Manganese Steels (31 xx Series)
9.l
alloying elements and is preser
.
,
deleted from the classification.
Chromium is a less expensive alloying ele- f 7
m
5 percent (33xx series). The rr As in the case of the nickel ste
.
.
ment than nickel and forms simple carbides (Cr
bides [(FeCr)3C].
.
used in the manufacture of aut(
uous films around the primary These films are liquid at the r condition of hot-shortness whi
bounda-ies during working.
I
combine with sulfur, and man
point than the iron sulfide eute at the rolling temperature and I properties of steel. It is only when the mangane the steel ipay be classed as an f
to strength and hardness, but effective in the higher-carbon s and has a moderate effect on h the critical range and decrees'
Fine-grained manganese stf These steels are often used foi With a moderate amount of v
for large forgings that must be yield properties equivalent to a full hardening and temperin When the manganese cont
ALLOY STEELS
359
J .
s which are not adapted to quenching.
ability and wear resistance. It is important to remember that the combined
series) with low carbon are used extensivi
effect of two or more alloying elements on hardenability is usually greater than the sum of the effects of the same alloying elements used separately.
connecting-rod bolts, studs, and kingpift 5xx series) provide increased toughness
:
The low-carbon nickel-chromium alloy steels are carburized. The chromium supplies the wear resistance to the case, while both alloying elements improve the toughness of the core. With 1.5 percent nickel and 0.60 percent chromium (3ixx series) they are used for worm gears piston pins, etc. For heavy-duty applications, such as aircraft gears, shafts and cams, the nickel content is increased to 3.5 percent and the chromium content to
cations such as bus and truck gears, cai
nly a mild effect on hardenability but is cull iprove toughness, particularly at low terrn
,
,
xxx series have been deleted from the AlSl-ii d alloy steels, it does not mean that they arej:
1
rom the classification simply means thattn
used in the manufacture of automotive connecting rods and drive shafts.
certain minimum.
The steels in this serieat
As in the case of the nickel steels, the steels in this series have also been
many applications by the lower-cost tripli
deleted from the Classification. In many cases, these steels have been replaced by the triple-alloy steels of the 87xx and SSxx series because of
.
4
Chromium is a less expensive alloying ele't'i
r
lower cost.
mple carbides (OA, Cr4C) or complex ".S -V 'p and good wgir|
bides have high hardness
9
,
'
h\e up to about 13 percent in y iron and has l In low-carbon steels, chromium tends to goi| I the strength and toughness of Jhe ferritft*
"
.
'
5 percent (SSxx series). The medium-carbon nickel-chromium steels are
.
y
alloying elements and is present in all steels as a deoxidizer. Manganese also reduces the tendency toward hot-shortness (red-shortness) resulting from the presence of sulfur, thereby enabling the metal to be hot-worked. When manganese is absent or very low, the predominant sulfide is FeS,
.
_
1 amounts in excess bflTpercent, the higteji orrosion resistance of the steel are greatly! 3f the 51xx series contain between 0.15
t<
The very high nickel-chromium alloy steels will be discussed in Sec. 9-15.
-1 Manganese Steels (31xx Series) Manganese is one of the least expensive
which forms a eutectic with iron and has a tendency to form thin contin-
uous films around the primary crystals during solidification of the steel. These films are liquid at the rolling temperature of steel and produce a condition of hot-shortness which js a tendency to crack through the grain boundaries during working. Manganese is outstanding in its power to combine with sulfur, and manganese sulfide has a much higher melting
an d
_
leen 0.70 and 1.15 percent chromium. The-
series are usually carburized. The presenc| ar resistance o| the case, .
but the toughness;,,
ie nickel steelsf With medium carbon, thes |
point than the iron sulfide eutectic. The manganese sulfide remains solid at the rolling temperature and has a less adverse effect onjhe hot-working
are used for springs, engine bolts, studi percent) high-chromium (1.5 percent) alloy;
.
properties of steel...
&y high hardness and wear resistance. This
It is only when the manganese content exceeds about 0.80 percent that the steel may be classed as an alloy steel. Manganese contributes markedly
all and roller bearings and for crushing ma-l mium steel containing 1 percent carbon anrfj
to strength,and.b ardne.5sJ ,
xcellent magnetic properties and is used foi| containing over 10 percent chromium
f
t
ce to corrosion and will be
jries)
>
_
arei
discussed inf
1
In these steels the ratio of nickel ; '
.
vVS'/z parts nickel to 1 part chromium. A com-
usually imparts some of the characteristic
3ffect of nickel in increasing toughness and P
3 effect of chromium in improving harden-|L
i
2;
but Jo ajesser degree than carbon, and is most
effective in the higher-carbon steels. This element is a weak carbide former
,
and has a moderate effect on hardenability. Like nickel, manganese lowers the critical range and decreases the carbon content of the eutectoid. Fine-grained manganese steels attain unusual toughness and strength.
These steels are often used for gears, spline shafts, axles, and rifle barrels. With a moderate amount of vanadium added, manganese steels ar,e used for large forgings that must be air-cooled. After normalizing, this steel will yield properties equivalent to those obtained in a plain-carbon steel after a full hardening and tempering operation.
When the manganese content exceeds about 10 percent, the steel will
360
INTRODUCTION TO PHYSICAL METALLURGY
triple-alloy nickel-chromium
-
be austenitic after slow cooling. A special steel, known as Hadfield manga-
'
,
have the advantages of the nic hardenability imparted by molyb
nese steel, usually contains 12 percent manganese. After a properly controlled heat treatment, this steel is characterized by high strength, high?
m
aircraft industry for the structu
ductility, and excellent resistance to wear.}; It is an outstanding material for resisting severe service that combines abrasion and wear as found in i power-shovel buckets and teeth, grinding and crushing machinery, and; railway-track work. If this alloy is slow-cooled from 1750"F, the structure will consist of large brittle carbides surrounding austenite grains. This structure has low strength and ductility. In this condition the tensile '
m
m
1
feet, it is not used in general ( marily in tool steels.
and to the conversion of some austenite to martensite.
0- 9 11 Molybdenum Steels (4xxx Series)
I
Molybdenum is a relatively expensive alloying element, has a limited solubility in y and a iron, and is a strong
V i
though larger quantities are req
1850°F, the structure will be fully austenitic with a tensile strength of about; 120,000 psi, elongation of 45 percent, and a HN of 180. The alloy now has much greater strength and ductility as compared with the annealed condition. The steel is usually reheated below 500°F to reduce quenching 1 stresses. In the austenitic condition following rapid cooling, the steel is
hardness is due to the ability of manganese steels to work-harden rapidly jj -
and landing gear. Tungsten Steel Tungsten has a carbide former, and retards the general, the effect of tungsten irr
sten is equivalent to 1 percent
peated impact, the hardness increases to about 500 BHN. This increase in
.
C 9-12
strength is about 70,000 psi, with elongation values down to 1 percent. If the same alloy, after allowing the carbides to dissolve, is quenched fronv
riot very hard; however, when it is placed in service and subjected to re-
i
mo
carbide former. Molybdenum has a strong effect on hardenability and, like chromium, increases the high-temperature hardness and strength of steels
Steels containing molybdenum are less susceptible to temper brittleness I than.other alloy steels. This element is most often used in combination with nickel or chromium or both nickel and chromium. For carburizing applications it improves the wear resistance of the case and the toughness of the core. The plain-molybdenum steels (40xx and 44xx series) with low carbon content are generally carburized and are used for spline shafts, transmission gears, and similar applications where service conditions are not too severe. With higher carbon they have been used for automotive coil and leaf springs. The chromium-molybdenum steels (41 xx series) are relatively cheap and possess good deep-hardening characteristics, ductility, and weldability. They have been used extensively for pressure vessels, aircraft structural parts, automobile axles and similar applications. The nickel-molybdenum steels (46xx and 48xx series) have the advantage of the high strength and ductility from nickel combined with deep-hardening
expensive and large quantities 913 Vanadium Steels Vanadium elements. It is a
is t
powerful deo)
inhibits grain growth. Vanadiui
a sound, uniform, fine-grain c marked effect on hardenability, cooling. Therefore, carbon-van and machinery forgings that ar The low-carbon chromium-v
the case-hardened condition shafts. The medium-carbon cff ness and strength and are us< grade with high hardness and tools. 9-14 Silicon
Steels (92xx Series) Si
steels as a cheap deoxidizer. V silicon, it is classed as a silico former but rather dissolves in A steel containing 1 to 2 perc structural applications requirii with less than 0.01 percent c cellent magnetic properties f machinery.
A properly balanced combir steel with unusually high stre1
,
,
and improved machinability imparted by molybdenum. They have good toughness combined with high fatigue strength and wear resistance. They are used for transmission gears chain pins, shafts, and bearings. The ,
This silicon-manganese steel and also for chisels and punc 9-15 Stainless
Steels Stainless ste resisting applications. A thref
tify stainless steels. The last t
:
I 1
ALLOY STEELS
A special steel known as Hadfield mangi'
triple-alloy nickel-chromium-molybdenum steels (43xx and 47xx series) have the advantages of the nickel-chromium steels ialong with the high
,
N
.
>]Brcent manganese. After a properly co
.;;
,
M|j is characterized by high strength, bij e to wearj It is an outstanding mater!
hardenability imparted by molybdenum. They are used extensively in the aircraft industry for the structural parts of the wing assembly fuselage, ,
combines abrasion and wear as found i i
,
361
and landing gear.
grinding and crushing machinery,
J
Tungsten Steel Tungsten has a marked effect on hardenability is a strong carbide former and retards the softening of martensite on tempering. In ,
is slow-cooled from 1750°F the structurei ,
,
ides surrounding austenite grains Thlii
general, the effect of tungsten in steel is similar to that of molybdenum, al-
.
ductility. In this condition the tensile!
i elongation values down to 1 percent
.
3 carbides to dissolve
,
.
though larger quantities are required. Approximately 2 to 3 percent tung-
If
-
r
is quenched from :
--austenitic with a tensile strength of about
sten is equivalent to 1 percent molybdenum. Since tungsten is relatively expensive and large quantities are necessary to obtain an appreciable effect, it is not used in general engineering steels. TungstenJs usecLprir .
ent, and a BHN of 180. The alloy now has lity as compared with the annealed con-
9-13
eated below SOOT to reduce quenching ?f ition following rapid cooling, the steel IS 1
v
.
.
Vanadium Steels Vanadium is the most expensive of the common alloying elements. It is a powerful deoxidizer and a strong carbideJarmer which inhibits grain growth. Vanadium additions of about 0.05 percent produce
a sound, uniform, fine-grain casting. When dissolved, vanadium has a marked effect on hardenability, yielding high mechanical properties on air cooling. Therefore, carbon-vanadium steels are used for heavy locomotive and machinery forgings that are normalized. The low-carbon chromium-vanadium steels (61 xx series) are usedjn
is placed in service and subjected to re-
Sases to about 500 BHN. This increase in
.r
hanganese steels to work-harden rapidly "Tl
vXv: tenite to martensite. Molybdenum is a relatively expensive olubility in y and a iron, and is a strong
the case-hardened condition in the manufacture of pins and crank-
shafts. The medium-carbon chromium-vanadium steels have high tougfP ness and strength and are used for axles and springs. The high-carbon grade with high hardness and wear resistance is used for bearings and
a strong effect on hardenability and, like iperature hardness and strength of steels. ire less susceptible to tgm per_b.ri.ttleriess .
ment is most often used in combination
ti nickel and chromium.
_
marily in tool steels.
tools.
914 Silicon Steels {92xx Series) Silicon, like manganese, is present in all steels as a cheap deoxidizer. When a steel contains more than 0.60 percent
For carburizing
resistance of the case and the toughness
silicon, it is classed as a silicon steel. Like nickel, silicon is not a carbide [
(40xx and 44xx series) with low carbon
A steel containing 1 to 2 percent silicon known as; navy steel is used for structural applications requiring a high yield point, Hadfield silicon steei with less than 0.01 percent carbon and about 3 percent silicon has excellent magnetic properties for use in the cores and poles of electrica1
d and are used for spline shafts, transcations where service conditions are not
they have been used for automotive coil molybdenum steels (41 xx series) are rela-
deep-hardening characteristics, ductility, n used extensively for pressure vessels, X;bile axles, and similar applications. The ;
..
"
::
1
machinery. A properly balanced combination of manganese and silicon produces i steel with unusually high strength and with good ductility and toughness
"
and 48xx series) have the advantage of the
;. i nickel, combined withjteep-hardening r: parted by molybdenum. They have good
;
.
former but rather dissolves in ferrite, increasing strength and toughness.
,
"
:r;
atigue strength and wear resistance. They s
,
VvV.
'
I
chain pins, shafts, and bearings. The
This silicon-manganese steel (9260) is widely used for coil and leaf spring; and also for chisels and punches. 9-15 Stainless Steels
Stainless steels are used for both corrosion- and heat
resisting applications. A three-numeral numbering system is used to iden tify stainless steels. The last two numerals have no particular significance
362 INTRODUCTION TO PHYSICAL METALLURGY
:
but the first numeral indicates the group as follows:
m
SERIES DESIGNATION
GROUPS
2xx
Chromiurrbnickel-manganese;
hardenable
,
3xx
non-
austenitlc, nonmagnetic
Chromium-nickel: nonhardenable
,
austenitic
nonmagnetic Chromium; hardenable martensitic
4xx
,
,
,
magnetic
4xx
Chromium; nonhardenable ferritic, ,
m
magnetic
5xx
Chromium; low chromium heat,
resisting
m I +
1400 13
°
Z + y
2400
I
1200
-:
Fig. 9 7 Microstructure of a 12 percent chrom (a) Annealed; small carbide particles in a ferritt i0) Quenched from 1850 F tempered .at 600"F; .
Y ;
I
Z +y f S
8
.
i
2800
Liquid
t + r + C/v
is tempered martensite. Etched in picric-hydroi acid, 500x. (Research Laboratory, Universal-Cy Steel Corporation.)
Since stainless steels cor
12000
"
the iron-chromium-carbon a
10001r
3
oT
5 9
r -
a+r + Cm
BOO
| 1600 |
3
.
i
,2
I
H 1200
600
I
a + Cm
and 9-8 represent plane sec these plane figures are not t study of phase changes and Figure 9-6 shows a diagn varying carbon. In compari presence of this amount of < and reduced the austenite
400 -
1
800
carbon, these steels may be
the plain-carbon steels.
m
i
ZOOj1
i
The microstructure of a 1;
L 10
-
88% Fe 12%Cr
0 5 ,
15
.
,
20
25 ,
400 30 .
Percenf carbon
0%C
35 ,
I
40 .
Fig. 9-6 Cross-section diagram for steels containing 12 percent chromium (From E, E, Thum Book of Stainless
Figure 9-8 shows a diagre
"
,
,
umo \£tf -' American S°cie,y 1935.) ,
dition is shown in Fig. 9-7a. This same steel after quenc1 sists of tempered martensitf
Metals, Metals Park,
bon.
Consideration of this (
the steel is low austenite w ,
.1
J
i
m
1
"
ALLOY STEELS
363
-
group as follows:
i-nickel-manganese; e,
St
non
3 -
austenltic, nonmagnetic
5
-
-nickel; nonhardenable,
i
9te
<
-
el
4
nonmagnetic i; hardenable, martensitic, V
i; nonhardenable, ferritic, 5 :
>Mh; low chromium, heat-
i
is
5»
.
is due to a thin, adherent, stable chro-
1
j
lat effectively protects the steel against erty is not evident in the low-chromium
y
ed and is apparent only when the chro-
percent. 1
t
Liquid
2400
ran>
acid, 500x. (Research Laboratory, Universal-Cyclops Steel Corporation.)
Since stainless steels contain relatively large amounts of chromiur the iron-chromium-carbon alloys belong to a ternary system. Figures 9 and 9-8 represent plane sections through such a ternary system. Whi these plane figures are not true equilibrium diagrams, they are useful in study of phase changes and in interpreting structures. Figure 9-6 shows a diagram for steels with 12 percent chromium ar varying carbon. In comparison with the iron-iron carbide diagram, tl presence of this amount of chromium has raised the critical temperatun and reduced the austenite area. However, with the proper amount carbon these steels may be heat-treated to a martensitic structure, as we the plain-carbon steels.
2000
1600 f
1200
n
J
800
0
25 .
30 .
,
The microstructure of a 12 percent chromium steel in the annealed co dition is shown in Fig. 9-7a. It consists of ferrite and small carbide particlt This same steel after quenching from 18500F and tempering at SOCTF co sists of tempered martensite and bainite (Fig. 9-7b). Figure 9-8 shows a diagram with 18 percent chromium and varying c< bon. Consideration of this diagram indicates that if the carbon content the steel is low, austenite will not be formed on heating. These steels £
400
_
35 .
40 .
it carbon v .-X
.;
; ig 12
'
.
Microstructure of a 12 percent chromium steel.
(a) Annealed; small carbide particles in a ferrite matrix; (b) Quenched from 1850oF, tempered at 600°F; structure is tempered martensite. Etched in picric-hydrochloric
I +Y + Cm
-
-
*
Fig. 9'7 2800
m
Jinless
,
s Park,
. . ..
.
364
INTRODUCTION TO PHYSICAL METALLURGY
tent. The austenite formed i stable phase reluctant to tran
2800
Liquid 1400
ing. Figure 9-11a shows the fi chromium, 8 percent nickel ; this steel after cold working i; The response of stainless depends upon their compos
8
Y+8 + y Y
8
1200
2400
+ y
2000
r
groups. 1000
a + r + C/n
a
1600 I
3
5
800
I 1200
1
-
,
600 a + Cm
H 800
400
200,
L 10
-
05 .
82% Fe I S'/oCr '
.
.
L
L 1 5
-
-
-
2,0
.
L
2.5
_
30 .
L
35 .
d 400 40 .
Percent carbon
0%C
Fig. 9-8 Cross-section diagram for iron-carbon alloys containing 18 percent chromium. (From E. E. Thum, Book of Stainless Steels, 2d ed,, American Society for Metals, Metals Park, Ohio, 1935.) "
"
(a)
nonhardenable, since subsequent quenching will only form ferrite of low hardness. Figure 9-9a illustrates the ferritic microstructure obtained by quenching a 0.03 percent carbon, 18 percent chromium steel from the delta region. If the carbon content is increased so that on heating the steel will be in the 8 plus y field, some hardness will result on quenching because of
the transformation of y iron. This is illustrated in Fig. 9-9b for a 0.075 percent carbon, 18 percent chromium steel water-quenched from 185G0F. The microstructure consists of ferrite (light area) and transformation product (dark area). If the carbon is still further increased so that the steel is in the austenite y + Cm field on heating, subsequent quenching will pro-
.5
'T
3v
If
S3
n
duce full hardness. This is shown in Fig. 9-9c for a high-carbon, 18 percent chromium steel water-quenched from 1850oF and tempered at 600oF. The microstructure consists of tempered martensite plus some undissolved
8
7
ft
carbides.
M
The addition of nickel to the chromium steel will produce further modifications in the diagram. Figure 9-10 indicates the trend in the changes of steel with 18 percent chromium 8 percent nickel, and varying carbon con,
mi
i
(c)
1
-
-
Cm -
ALLOY STEELS
tent.
2800
36
The austenite formed at the elevated temperature Is a particularl
2400
stable phase reluctant to transform and tends to be retained after anneal ing. Figure 9-11a shows the fully auslenitic microstructure of an 18 percen chromium, 8 percent nickel steel after annealing. The microstructure o
2000
this steel after cold forking is shown in Fig. 9-11 b. The response of stainless and heat-resisting steels to heat treatmer depends upon their composition. They are divided into three geners Li_
1
groups.
.
a)
f Cm
4
1600
-
1200
J
t
Cm
r
300
400 25 .
;
:
3.0
3.5
4.0
\V:\:arbon
s
dm
Book als,
4$
/ (a)
quenching will only form ferrite of low the ferritic microstructure obtained by 8 percent chromium steel from the delta creased so that on heating the steel will ess will result on quenching because of is illustrated in Fig. 9-96 for a 0.075 perm steel water-quenched from 1850oF.
23
m Fig. 9-9 Microstructures of an 18 percent chromium steel with varying carbon content (a) 0.03 percent carbon, water-quenched troi
ite (light area) and transformation prod-
.
till further increased so that the steel is
4
2100oF, etched in 20 percent HCI; structure is ferrite. (b) 0.075 percent carbon, waterquenched from ISSCFF and tempered at lOOO1! etched in picric-HCI; ferrite (light area) and
eating, subsequent quenching will proV:
-
n Fig. 9-9c for a high-carbon, 18 percent
p om 1850°F and tempered at 600oF. The
transformation product (darker area), (c) 0.61 percent carbon, water-quenched from 1850 and tempered at tOOOT, etched in picric-HCI
red martensite plus some undissolved
; 4 mium steel will produce further modifi.
|l0 indicates the trend in the changes of percent nickel, and varying carbon con-
(c)
undissolved carbides in a tempered-martensi matrix. All magnifications 500x. (Research
Laboratory, Universal-Cyclops Steel Corporation.) !
m 366
INTRODUCTION TO PHYSICAL METALLURGY
i
i
Liquic 2800
i MOO
X
L+cm+r
8+ y
120C
2400
i
Austemte 2000
1
IGOOh r +ca
I
1600 |
I 80C I
a
a
.
1200
1
I
as for plain-carbon or low-allc ness depend chiefly on carbc the high alloy content of the to be so sluggish, and the hai
is produced by air cooling. T ing them above the transforr then cooling in air or oil. Ti prevent decarburization or ex in this group should not be dc drop in impact properties. Te higher tempering temperaturt with a subsequent reduction carbon content on the low s
600
800
400
The heat treatment process
I
resistance is not too severe.
i i
Stainless steels as a group a carbon steels. The use of a s
nium in type 416Se improves 20C
Ol .
|
t jl !!
1 i i
400
02 .
0 3 .
04
0,5
.
74%Fe l8%Cr
0.6
07 .
08 .
09 ,
IO
i
.
Percent carbon
selenium has less effect in re
Stainless steels of type 440, w cent and 16 to 18 percent ch
e
I
8% Ni 0%C
1
Fig. 9-10 Tentative cross-section diagram showing trend of reactions in steels alloyed with 18 percent chromium and 8 percent nickel. (From E E. Thum, "Book of Stainless .
Steels," 2d ed., American Society tor Metals
,
4 ;
Metals
Park, Ohio. 1935.)
.\ -
f ,
Martensitic Stainless Steels These steels are primarily straight chromium steels containing between 11 5 and 18 percent chromium Some examples of this group are types 403 410, 416, 420, 440A, 501, and 502. Some of the properties and applications of the martensitic stainless steels
I t
.
.
,
i
are given in Fig. 9-12. Types 410 and 416 are the most popular alloys in this group and are used for turbine blades and corrosion-resistant castings The chemical composition and typical mechanical properties are given in .
SI
Table 9-3. The martensitic types of stainless steels are magnetic, can be cold-worked without difficulty especially with low carbon content, can be machined satisfactorily have good toughness, show good corrosion resistance to weather and to some chemicals and are easily hot-worked They attain the best corrosion resistance when hardened from the recom-
1
\
\
t .
1
I
\
I /
,
(a)
;
/
,
,
,
mended temperature but are not as good as the austenitic or terrific stain-
Fig. 9-11 Microstructures of an 18 percent t 8 percent nickel steel, (a) After annealing, all lb) after cold working. Etched In glyceregia,
less steels.
International Nickel Company.)
ALLOY STEELS
The heat treatment process for martensitic steels is essentially the same as for plain-carbon or low-alloy steels, where maximum strength and hardness depend chiefly on carbon content. The principal difference is that the high alloy content of the stainless grades causes the transformation to be so sluggish, and the hardenability so high, that maximum hardness is produced by air cooling. These steels are normally hardened by heating them above the transformation range to temperatures near 1850oF, then cooling in air or oil. Time at temperature is held to a minimum to prevent decarburization or excessive grain growth. Tempering of steels
i
viquid
2800
L+Cm
L + Cm + r 2400
I
-
.
H2000
in this group should not be done in the range of 750 to 950oF because of a drop in impact properties. Tempering is usually done above HOOT. The
Ll.
+ cm
1600 i o
higher tempering temperatures will cause some precipitation of carbides with a subsequent reduction in corrosion resistance. However, with the carbon content on the low side of the range, the lowering of corrosion
I
1200
367
h
"
resistance is not too severe.
Stainless steels as a group are much more difficult to machine than plain-
800
carbon steels. The use of a small amount of sulfur in types 416 and sele-
\
nium in type 416Se improves the machinability considerably. The use of selenium has less effect in reducing the corrosion resistance than sulfur. Stainless steels of type 440, with carbon content between 0.60 and 1.20 percent and 16 to 18 percent chromium, will have high corrosion resistance,
H400 L
_
06 .
07 .
_
08 .
l
_
_
09 .
10 .
irbon
-
.
2 trend
/
jainless
.
3Sm \ -
ST
m
i
sse steels are primarily straight chro11.5 and 18 percent chromium. Some
mm
403, 410, 416, 420, 440A, 501, and 502. nations of the martensitic stainless steels ind 416 are the most popular alloys In this Dlades and corrosion-resistant castings,
pical mechanical properties are given
m
DISSS' "-7
be
,
(a)
-
sistance when hardened from the
recom-
as good as the austenitic or ferritic stain-
1
:
mm 3
in
of stainless steels are magnetic, can pecially with low carbon content can be jod toughness, show good corrosion re 5e chemicals, and are easily hot-worked.
T
Jig. 9-11
Microstructures of an 18 percent chromium,
percent nickel steel, (a) After annealing, all austenite; ?(b) after cold working. Etched in glyceregia, 100x. (The International Nickel Company.)
(b) -~
368
INTRODUCTION TO PHYSICAL METALLURGY
403
Ferritic Stainless Steels Tl steels contain approximately 1 405, 430, and 446 (see Fig. 9
410
Turbine- quality
Base alloy for this
grade; similar to 4t0,
group,- general-pur-
used for steam ti/r-
higher in chromium than thi
pose, heat-treatable
bine blading and other highly stressed
type; used for machine parts pump
parts.
shafts.
hardened by heat treatment ;
working. They are magnetic , they develop their maximum :
,
JZ 414
41$
Higher Ni content increases hardenabi-
lity and corrosion resistancej used for springs tempered
420
Free-machining modi-
High-carbon modifi-
fication of 410
Highest carbon con-
cation of 410; has
(contains S); for heavy cuts.
higher hardness and
tent (0.95 to 1 20%) of the stainless steels; used for balls bearings races.
wear resistance
,
,
,
machine parts
steels is approximately 50 per they are superior to the martei and machinability. Since thf
.
-
used for cutlery surgical instruments,
,
rolls
in the annealed condition. Int
440C
,
they are used extensively for di
,
and food industries and for ar
values.
431
\
416Se
440A
Higher chromium content improves
Free-machining modi-
corrosion resistance; has high mechanical properties; used for
cuts and where hot
can have higher
working or cold head-
hardness than 420;
air-craft fittings heater bars paper machinery parts bolts ,
fication of 410(con-
tains Se); for light ing is involved
The only heat treatment app treatment serves primarily to i
440B
Slightly lower carbon
Slightly lower carban
content than 440B
content than 440C
for greater toughness;
improve; roughness; used lor line cutlery valve ports
important form of brittleness from prolonged exposure to, c from about 750 to 950oF. Note
,
Although the precise cause o effects increase rapidly with type 446. Certain heat treatr ductility, must be controlled usually annealed at temperat1
.
good corrosion resistance; used for cutlery valve parts
.
,
,
Fig. 9-12 The martensitic stainless steels (From Machine Design Metals Reference Issue, The Penton Publishing .
,
Co., Cleveland
,
Ohio, 1967.)
strength and wear resistance. parts, and bearings. ,
These alloys are used for cutlery
,
valve f
The addition of about 2 percent nickel to the 16 to 18 percent chromium low-carbon alloys (type 431) extends the austenite region and thus renders them heat-treatable. They are usually air-cooled and the heat treatment requires careful control of composition and quenching temperature because of the possible presence of delta ferrite at the austenitizing temperature. Type 431 has been used for aircraft fittings paper machinery parts,
430F
Free-machining modification of 430) (conioms S); for
,
heavy cuts and screw machined parts -
,
J
405
,
pumps, and bolts.
Addition of Al
improves weldability
The relatively low-chromium alloy steels containing 4 to 6 percent chro-
of this otherwise
mium (types 501 end 502) have excellent resistance to oxidation and much better corrosion resistance than ordinary steel These steels may be hardened by oil quenching or in some cases air cooling The properties attainea
i martensitic ohoy, making it nonhardenable-, used where
.
air-hardening types (410 or 403) are objectionable
.
are really intermediate between the series 5xxx alloy steels and the type 400 martensitic stainless steels Therefore they are suitable for mild corro.
,
sion conditions or at temperatures below 1000oF
.
extensively for petroleum refining equipment such as heat exchangers, valve bodies pump rings, and other fittings. ,
i
They have been used |
Fig. 9-13 The ferritic stainless steels. (From Design, Metals Reference Issue, The Penton Co., Cleveland, Ohio, 1967.)
mm.
ALLOY STEELS
1
Ferritic Stainless Steels
This group of straight-chromium stainless
steels contain approximately 14 to 27 percent chromium and include types
410
ur-
369
405, 430, and 446 (see Fig. 9-13). Low in carbon content
but generally higher in chromium than the martensitic grades these steels are not
Base alloy for this groupi general-purpose, heat-treatable type; used for
,
,
hardened by heat treatment and are only moderately hardened by cold
machine parts,pump
wbrking. They are magnetic and can be cold-worked or hot-worked but ,
shafts.
they develop their maximum softness, ductility and corrosion resistance in the annealed condition. In the annealed condition the strength of these steels is approximately 50 percent higher than that of carbon steelr. and they are superior to the martensitic stainless steels iri corrosion resistance and machinability. Since the ferritic steels may be cold-formed easily ,
di-
420
440C
High-carbon modifi-
Highest carbon con-
,
tent (0.95 to 1.20%)
cation of 410! has
of the stainless
higher hardness and wear resistance; used for cutlery, sur
,
-
,
steels; used for balls, bearings, races.
they are used extensively for deep-drawn parts such as vessels for chemical
gical instruments,
and food Industries and for architectural and automotive trim.
valves.
440A
(Odi-
-
. .
'
.
lot . iad-
Slightly lower carbon
The only heat treatment applied to truly ferritic steels is annealing. This treatment serves primarily to relieve welding or cold-working stresses. An important form of brittleness peculiar to the ferritic grades can develop
440B
Slightly lower carbon
content than 440B
content than 440C
for greater toughness-,
improves toughness;
can have higher
used for fine cutlery,
hardness than 420;
valve parts.
from prolonged exposure to or slow cooling within, the temperature range ,
from about 750 to 950 . Notch-impact strength is most adversely affected. Although the precise cause of the brittleness has not been determined, its effects increase rapidly with chromium content, reaching a maximum in type 446. Certain heat treatments, such as furnace cooling for maximum
m
good corrosion resistance-, used for
i
cutlery, valve parts.
1
ductility, must be controlled to avoid embrittlement. Ferritic steels are usually annealed at temperatures above the range for 850oF embrittlement
Machine
shing
430F
These alloys are used for cutlery, valve |
modification of 430
nt nickel to the 16 to 18 percent chromium, / . ends the austenite region and thus renders usually air-cooled, and the heat treatment nposition and quenching temperature be-
430Se
430
Free-machining (contains S); for
heavy cuts and screw machined parts -
Basic alloy for this group-o nonhardening chromium type used for decorativ» trim nitric ocio tanks annealing ocy ,
,
Free-machining modiflcatian of 430
(contains Se); for light cuts and where
hot-working or cold heading may be involved.
kets.
of delta ferrite at the austenitizing tempera-
for aircraft fittings, paper machinery parts,
alloy steels containing 4 to 6 percent
405
chro-
Myjn ordinary steel. These steels may be har
I
.
d
High chromium con -
Higher chromium
improves weldability
tent tor increased corrosion and scol-
content than 442 increases corrosion
martensitic alloy making it nonharden-
ing resistance; useiJ
abte; used where
nozzles combustion chambers
and scaling resistance at high tempera tures used especially for intermittant
of this otherwise
excellent resistance to oxidation and much
,
-
446
442
Addition of A
for furnace ports ,
,
,
he cases air cooling. The properties attained
air- hardening types
in the series 5xxx alloy steels and the type
(410 or 403) ore
service, often in
objectionable.
sulfur-bearing atmosphere.
;
.
Therefore, they are suitable for mild corro-
itures below 1000oF. They have been used !ning equipment such as heat exchangers, iother fittings.
. .
;
Fig. 9-13 The ferritic stainless steels (From Machine Design Metals Reference Issue, The Penton Publishing Co., Cleveland, Ohio 1367.) .
,
,
;
-
1
370
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 9-3
m
Chemical Composition and Typical Mechanical Properties of Some Stainless Steels*
GROUP
MARTENSITIC GROUP
AUSTENITIC GROUP
TYPE NUMBER
201
202
301
302
300
316
*10
416
420
440A
501
12.0-14.0
12.0-14.0
16.0-18.0
4 0-6 C
Analysis %: ,
Chromium
16.0-18.0
17.0-19,0
16.0-18.0
3 5-5 5
4 0-6 0
6 0-8 0
Nickel
.
.
.
.
.
1 7,0-19,0
22,0-24,0
16 0-18.0
r!l1.5-13.£
8 0-10 0
12.0-15.0
lO.O-U.O
050 max
.
.
.
Other elements
N, 0.25 max
N, 0.25 max
Carbon
0 15 max
0 15 max
0 15 max
0 15 max
0 20 max
0 08 max
7 5-10 0
2 0 max
2 0 max
2 0 max
1 0 max
1 0 max
Manganese
5 5-7 5 .
.
Silicon
1
00 max
.
.
.
.
1 0 max .
.
.
,
Mo 0.75 max
Vlo 0.'
0 15 max
0 15 max
0 60-0.75
J 10 rr
2 0 max
1 0 max
1 25 max
1 0 max
1 0 max
1 0 rr
1 0 max
1 0 max
1 0 max
1 0 max
1 0 max
1 0 max
1 0 rr
.
,
,
.
,
015 max
Mo 2,0-3 0
i
,
.
,
,
0 50 max
.
.
.
,
.
,
.
.
.
,
.
.
,
.
.
.
.
Temperature, aF:
Forging-start
2300
2300
2200
2200
2150
2200
JIM
2150
2000
2100
2150
Annealing-ranges
1850-200
1850-2000
1950-2050
1850-2050
2050-2150
1975-2150
(500.1650
1500-1650
1550-1650
1550-1650
1525-
Annealing-coolingt
WQ (AC)
WQ (AC)
WQ (AC)
WQ (AC)
WQ (AC)
WQ (AC)
SFC
FC
FC
FC
Hardening-ranges
t
t
t
t
t
Quenching
1700-1850
1700-1850
1800-1900
1850-1900
1600-
0 or A
O or A
0 or A
O or A
0
Over 1100
Jp to 1200
Below 700
Over 1100
Over
Under 700
Below 700
Below 700
Under 700
Undei
r-c
F-C
F-C
32
40-50
50-60
55
30
60-80
90-100
S5
70
30-20
25-20
20,0
28.0
60-50
50-40
40,0
6b,0
29,0
29.0
30,0
29,0
145-185
200-230
240 max
160
B 79-90
B 93-98
B 100 max
Tempering-for intermediate hardness
Drawing-for relieving stresses ;
-
:
Mechanical properties -annealed:
Structure annealed
A
Yield strength 1,000 psi min
A
40
40
Ultimate strength, 1,000 psi min
115
100
Elongation, % in 2 in. min
40.0
40.0
,
Reduction in area. % min.
Modulus of elasticity in tension, 10" psi
29,0
Hardness, Brinell
210 max
A
35 100 50.0
A
A
30
30
30
8C
75
75
50.0
40.0
10,0
SO
60.0
60,C
50,0
50,0
29.0
29.0
29,0
29.0
29.0
210 max
180 max
180 max
200 max
200 max
m 29"
Hardness, Rockwell
B 95 max
B 95 max
B 90 max
B 90 max
B 95 max
B 95 max
Impact values, Izod, ft-lb
85 min
85 min
85 min
85 mm
80 min
70 min
A .
'
300 max
lr E 95 max o mm
t
1;
A
F-C
Low
50-30
Mechanical properties -heat-treated;
Yield strength 1,000 psi ,
Ultimate strength, 1 000 psi ,
§
§
§
35-180
30-130
120-220
55-240
90-1
jO-200
M-160
150-250
96-275
I1&.
Elongation, % in 2 in.
25-?,
Hardness, Brinell
' .
Hardness, Rockwell
i
20.400
8 70-0 45
20-10
12-2
20-
20-2
180-300
275-500
200-555
B 88-107
C 30-52
B 95-C 55
240-
:
Abbreviations; AC = air cool FC = furnace cool, SFC = slow furnace cool WQ = water quench O AC = air cool F i= ferrite, C = carbide, A = austenite. ,
,
,
oil quench,
1
,
*
t i
§ H
From "Stainless Steel Handbook," Allegheny Ludlum Steel Corp. Thin sections of 300 series, marked WQ (AC) are usually air-cooled, heavy sections waler-quenchod. Hardenable only by cold working. Ultimate strengths up to 350,000 psi lor wire and 250,000 psi lor strip can be obtained by cold working. Generally used in the annealed condition only,
When heat treated
,
,
Is negligible
m
,
and because of possible embrittlement in the 850'5F range
Th
and chromium-nickel-manganes
are austenitic, are essentially nc do not harden by heat treatment is at least 23 percent. They cat worked when proper allowance
and below temperatures at which austenite might form above the A, line to obtain maximum ductility these steels are cooled slowly. They are not tempered since the amount of martensite formed .
Austenitic Stainless Steels
.
Cold working develops a wide ra in this condition may become sli resistant and difficult to machir
(types 303 and 303Se).
Thes
strength and resistance to seal
"
RGY
ALLOY STEELS
371
A
i
ileal Properties of Some Stainless Steels* AUSTENITIC GROUP 32
301
302
MARTENSITIC GROUP 309
316
416
420
12.0-14.0
12,0-14.0
FERRITIC GROUP 405
502
440A
501
16,0-18,0
4 0-6 0
430
446
11.5-14.5
14,0-18,0
23,0-27,0
0 50 max
0 50 max
0 50 max
J: ii
7 0-19 0 .
.
; o-6 .
17.0-19.0
22.0-24.0
16.0-18.0
13.5
6 0-8 0
8 0-10 0
12.0-15.0
10.0-14.0
max
b
.
.
.
.
.
0.25 max
,
0
.
,
4 0-6 0 ,
,
50 max
.
Mo 2.0-3, jii
.
Mo 0.75 max
Mo 0,4-0,65
Mo 0 4-0,65
A! 0.10-0,30
0 10 min
0 10 max
0 08 max'
.
.
N, 0,25 max
15 max
0 15 max
0 15 max
0 20 max
o oa max
IP m*
0 15 max
0 15 max
0 60-0.75
0 12 max
0 20 max
5-10 0
2 0 max
2 0 max
2 0 max
2 0 max
may
1 25 max
1 0 max
1 0 max
1 0 max
1 0 max
1 0 max
1 0 max
1 5 max
1 0 max
1 0 max
1 0 max
IfAmax
1 0 max
1 0 max
1 0 max
1 0 max
1 0 max
1 0 max
1 0 max
1 0 max
.
.
.
.
0 max
1 0 max .
00
.
16.0-18.0
.
,
.
.
,
.
2200
2200
2150
2200
)50-2000
1950-2050
1850-2050
2050-2150
1975-2150
Q (AC)
WQ (AC )
WQ (AC)
WQ (AC)
WO (AC)
t
t
1650
,
.
.
.
.
.
.
,
,
,
,
.
.
,
.
,
,
.
.
.
.
.
2150
2000
2100
2150
2150
2100
2100
1500-1650
1550-1650
1550-1650
1525-1600
1525-1600
1350-1500
1400-1500
1450-1600
FC
FC
FC
FC
FC
AC
FC
WQ
H
1700-1850
1800-1900
1850-1900
1600-1700
O or A
O or A
O or A
O
1100
Up to 1200
Below 700
Over 1100
Over 1000
r7fl0
Below 700
Below 700
Under 700
Under 700
t
-
A 40
3f
2150
,
30
35 max min
ii.
A
A
A
F-C
F-C
F-C
F-C
F-C
F-C
F-C
F-C
30
30
36
40-5C
50-6C
55
30
25
32
35
45
80
75
75
60-80
90-100
95
70
65
50
60
75
50. C
40,C
40.C
30-2f
25-2C
20.0
28.0
30.0
20.0
20.0
20.0
00,0
fi0,0
50,0
50,0
60-50
50-40
40 0
65.0
76.0
50.0
40.0
40.0
29.0
29.0
29,0
29,0
29.0
29.0
300
29.0
29.0
29 0
29.0
29.0
180 max
180 max"
200 max
200 max
145-185
200-230
240 max
160
150
180 max
200 max
200 max
B 79-90
B 93-98
.
10 max
Monhardenable
50.C
100
o c
B 90 max
B 90 max
a 95 max
B 95 max
85 min
35 min
80 mm
70 mm
0
max
to
max
ilow furnace cool
,
B 100 max
B 75
8 90 max
B 95 max
B 95 max
85 min
25 min
3-85
Low
50-30
180
60-130
120-220
55-240
90-135
200
! 10-160
150-250
95-275
115-175
12-2
.
Low
mm
20-10
20-2
00
180-300
275-500
200-555
70-0 45
B 88-107
C 30-52
B 95-C 55
20-15 240-370
WQ = water quench, O = oil quench
,
Austenitic Stainless Steels These are the chromium-nickel (type 3xx) and chromium-nickel-manganese stainless steels (type 2xx). These types are austenitic, are essentially nonmagnetic in the annealed condition, and do not harden by heat treatment. The total content of nickel and chromium is at least 23 percent. They can be hot-worked readily and can be coldworked when proper allowance is made for their rapid work hardening
Steel Corp. ly air-cooled heavy sections water-quenched.
j
.
'
Ska
00
.
,
! :|00 psi for strip can be obtained by cold working
.
.
C
h austenite might form When heat treated" .
laximum ductility these steels are cooled:; ,
d,
since the amount of martensife formed
ossible embrittlement in the 850°F range.
.
M
Cold working develops a wide range of mechanical properties, and the steel in this condition may become slightly magnetic. They are extremely shockresistant and difficult to machine unless they contain::sulfur and selenium (types 303 and 303Se). These steels have the best high-temperature strength and resistance to scaling of the stainless steels The corrosion .
r
372
INTRODUCTION TO PHYSICAL METALLURGY
resistance of the austenitic stainless steels is usually better than thatd "
-
-
-
'
the martensitic or ferritic steels.
-
202
,,
I
Generol-purpose low-nickel equivo-
Type 302, the basic alloy of the austenitic stainless steels, has been mod-, ?;
ified into a family of 22 related alloys (see Fig. 9-14). For example, lowers | ing the carbon content to 0.08 percent maximum led to type 304 with im» '
Bo
gr,
lent of 302 , Ni
fo
parfiatiy replaced by Mn,
Co
eq sp tL
proved weldability and decreased tendency toward carbide precipitation, t To avoid carbide precipitation during welding, a lower-carbon version,
;
[
_
type 304L, was developed which contains only 0.03 percent carbon max- j imum. Although type 304L suppresses carbide precipitation during cooling
304L
:
Ej.ttolow-corbon modi ficof ion of 304 for further reslriction of
through the range of 1500 to 800oF after welding, potentially more serious precipitation problems are encountered in multiple-pass welding or serv- ;
j corbide precipitalion
bide precipitotion during welding;
ft
ex a:
a(
and food-processing
eguipmeni/ecording
grades, type 321 with Ti added, and type 347 with Cb or Ta added, are % recommended. In both alloys, a carbide other than chromium carbide /
wire "
I
precipitates, thus chromium is retained in solution and the alloy maintains ;
.
Si
used for chennca!
during welding.
ice in the 800 to 1500oF range. To meet these requirements, the stabilized ;
Low - cotbon modihcotion of 302 for resl nction of cor-
321
303
its corrosion resistance. A stabilizing heat treatment consists of holding |
f ree-mochining
Ti content prevents chromiumcdrbide
Si
modif icotion
either annealed or welded types at 1600 to 1650oF for 2 to 4 h, followed by
(contoins S) of 302 i for heovy
precipitotion during welding,
ir
cuts; used for screw machine
for severe corrosive
n
conditions ond
o
rapid cooling in air or water. The purpose is to precipitate all carbon as a carbide of titanium or columbium in order to prevent subsequent precipita- .
-;
v -vZ.
-
'
tion of chromium carbide.
i
| valves.
to 1600 F, used for oircrolt
exhdust rnomfolds, ooiler shellstproce5s equiprnenr.
Although all stainless steels can be hardened somewhat by cold work, the response becomes pronounced in the austenitic alloys, 303Se
reaching a maximum in types 201 and 301. Table 9-4 compares the work- |
icontoins loins Se) Ss)
REDUCTION,
CONDITION OF METAL
%
TENSILE
ELONGATION
ROCKWELL
YIELD
BUCKLING
STRENGTH,!
STRENGTH,
IN 2 IN.,
HARDNESS
STRENGTH,t
STRENGTH
PS I
PSI
%
NUMBER
PSI
!!!
j
s
3021 for light
i
cuts ond where
r
involved,
L
COMPRESSION
YIELD
of
hot working or cold heddmq mdy be
Effect of Cold Rolling in the Longitudinal Direction on Types 301 and 302 Austenitic Stainless Steels(*) TENSION
e
modiflicolicn icoticr
which the austenitic stainless steels have been used, and Table 9'3 gives
COLD
S
rnochsning Free mochinmg
hardening behavior of the 17-7 Cr-Ni type 301 with the more stable 18-8 type-302. Figure 9-14 shows the interrelation and wide variety of applications for
TABLE 9 4
1
service from 800
j products, shafts.
i
e
j Worn- hordening
Low- nickel
1 rate increased by
equivalent of 301;
1 lower Cr ond Ni
Ni portially replaced by Mn;
I I content j used for ;
I
201
301
Ingh-strenqth hiqhducliHty applications
has high workhardening rate.
such as railroad
I cors, trailer bodies,
TYPE 301 STAINLESS STEEL
I oircraft structural Annealed
33,000
117,800
68
B 85
40 000
57,800
10
Cold-rolled
67 000
147.600
47
C 32
54,000
89 400
0
,
,
] members.
,
25
Cold-rolled
127 000
165,200
24
C 38
96 000
151 400
35
Cold-rolled
164,000
196 000
15
C 43
139 000
184,500
'
45
Cold-rolled
200.000
225,000
7
C 46
163 000
218,000
! Design Metals Reference Issue,
,
,
,
,
,
,
36,000
94,000
61
B 80
36,000
50 250
20
Cold-rolled
121 000
139,300
22
C29
74.000
120,100
35
Cold-rolled
131,000
155,300
15
C 36
95 000
151 800
50
Cold-rolled
151,000
177,400
6
C 38
99,000
155 200
0
Annealed
,
,
Adapted from "Metals Handbook. vol, 1, American Society for Metals. Metals Park, Ohio, 1961, t 0.2 percent offset. "
:
.
The austenitic stainless steels. (From N
Co,, Cleveland, Ohio, 1967.)
TYPE 302 STAINLESS STEEL - .:
F19 9 14
,
,
.
The Penton Publ
JRGY
ALLOY STEELS
373
;
-
tainless steels is usually better than-
202
the austenitic stainless steels, has beeB]
lent of 302; Ni
d alloys (see Fig. 9-14). For example; 8 percent maximum led to type 304 witi ased tendency toward carbide precipn
partially replaced
by Mn.
'
I
ip4L__] r
.
.
04 for further
isslnc'icr of
i
303 f'ee-'nachininq
modification
es at 1600 to 1650oF for 2 to 4 h, followedij.4
{contains S) of
302for heavy
The purpose Is to precipitate all carbon ast
-
djts; used for screw machine
ium in order to prevent subsequent precipita
products, shaf ts,
'
. .
valves.
jels can be hardened somewhat by coli les pronounced
in the austenitic afl-
> 201 and 301.
4 compares the wori with the more stable 18-1
Table 9
-7 Cr-Ni type 301
7
E
IGTH
J01
COMPRESSION
i (Ni and Cr) is higher;
for severe corrosive conditions and service horn BOO
nuclear energ-y applications.
( corrosion and
sjxepi alloycontent I
has excellent
Low-carbon modification of
309 for improved weldability,
j1 scaling resistance, used in aircraft
to 1600 F,
j heoters eoHreot-
used for aircrof t exhaust manifolds, boiler shells, process
j mg equipment, turnace ports.
I 314
3I0S
310
Similar la 310, except higher
Similar to 309,
silicon content
(NiandCr) is tiigheri
increases scaling resistance at high
used for heat
temperalure.
nace ports combusl ion chambers, woidinq liller metols.
,
s
except olloy content
Low-corbon modification of
310 for improved weldability.
enchongers, fur,
lower Cr and Ni
Ni portiolly repioced by Mn; has high workhordemng rote.
IN 2 IN.
HARDNESS
STRENGTHS
STRENGTH '1
high-strength high-
NUMBER
PSi
PSI
duclilify opclicotions
content
used for I
68
B 85
40,000
67,800
47
C 32
54,000
89,400
0
equivalent of 301;
Higher Mo content than 316improves resistance to corrosion and creep.
24
C38
96,000
151,400
15
C 43
139,000
184 500
fig. 9-14
0
7
C 46
163,000
218.000
Oesign Metals Reference Issue, The Penton Publishing Co., Cleveland, Ohio, 1967.)
61
.
B 80
36,000
22
C 29
74,000
120 400
16
C 36
95,000
151.800
6
C 38
99,000
155,200
1 Society for Metals, Metals Park, Ohio, 1961
50,260 ,
Low-carbon
resistance than 30? or 304
modification of 316,for welded construction.
.
because of Mo
content j has high creep resistoiicej phatoqrophic,ond food equipment.
0
102 STAINLESS STEEL
316L
Higher corrosion
used for chemical, pulp- handling,
cors, trailer bodies, aircraft structurol members.
,
316
317
Low-nickel
%
,
201
Work- hardening rate increased by
BUCKLING"
'
m
309S
moximum limit on To; used for
involved.
YIELD
.
i
.
precipitation during welding;
hot working or cold heading moy bo
ROCKWELL
0
.Jo
SO'S O* .3'-
S'milor to 308
ercept fc a
1 and severe drc«ing j ( operctions. i
-
303
3 8
| 'Jied fofS(xnfo'mi |
j 1:65 'i *- = 'C ir,g
Sirmla' to 3 7
cuts and where
ELONGATION
0
.:.
I
Ti coitfif prevents chrofr.ij'mccf bide
modilication
01 STAINLESS STEEL
.
"W C
Free-machining
such as railroad
"
;
521
"
r -ito's
' ,
'
,
Direction on Types 301 and 302 Austenitic Stainless SteeisC) TENSION
..
j
iT'd,-l:
*
ork -
j nar er.ir.q re!*;
Si
;
'
[
:.
c:"3' c-r';:: '-; '
to iower
:"-:ira'ij
*
V
'
.
!
ri tz' revs*;-
equipment.
!
; :S;I ; i
ini
.
'
303Se
rrelation and wide variety of applications fcrti| I :
305
High- nickel content
-
s steels have been used, and Table 9-3 giveft |
; :
308
Higher alloy (Ni and -
(contains Se) of 302; for light
:
heating -'
V.
1
wire
tabilizing heat treatment consists of hold™
i
Similar to 321 i Ck 0' To is
:
€3'JlpTr*,VeC:r-!r,q
ys, a carbide other than chromium car s retained in solution and the alloy maintailf
..
elements.
liners,
347
ca*'C- ot 102 *Cf
,
ed, and type 347 with Cb or Ta added,
'
tural products,
scaling than 302 because of SI content; used for f ui noce portSjStill
p*
corbide precipitation i rtar g melding. i
the stabili
i04
Low - COrOon nodifi-1
Extrolow-carbon
wwiificotior. of
icountered in multiple-pass welding or; To meet these requirements,
More resistant l-o
X
ich contains only 0.03 percent carbon ! spresses carbide precipitation during ;ii 800"F after welding, potentially moresei i
502B
Bose olloy for t is group; used for trim, food- handling equipmen oircraft cowling, ontennos, springs rchlteccookware,
)n during welding, a lower-carbon v/
ij
502
General-purpose low-nickel equiva-
The austenitic stainless steels (From Machine .
i
374
INTRODUCTION TO PHYSICAL METALLURGY
the chemical composition and typical mechanical properties of somestaifr.
less steels.
4
if
*
The shortage of nickel in times of national emergency has presented a serious problem to stainless-sjeel producers and consumers. Stai
55
MM
S75
during World War II and continued through the Korean emergency, d| velopment work covering the substitution of manganese for nickel in staii
less steel led to the production of types 201 and 202, the chromium-nick9l|
:!
manganese stainless steels. Type 201 with a nominal composition of l| >5
percent Cr, 4.5 percent Ni, and 6.5 percent Mn is a satisfactory substitute
for type 301 (17 percent Cr, 7 percent Ni) where machinability and sevefi forming characteristics are not essential. Where those characteristicsaiW
m
. ,
„„„
,
.
.
,
„
.
„
.
°
9'15 17~4PH alloy solution Ftreated at 190o f and air-cooled.
fe- air-cooied, then aged 4 h at 925
essential, type 202 With a nominal composition of 18 percent Cr, 5 Percei|»
,empered
°
'
martensite. Etched in Fry s reagent, u
Ni, and 8 percent Mn is more desirable because the higher manganesMr (From Metals Handbook, vol. 7, "Atlas of Microsta
reduces the rate of work hardening. Although types 201 and 202 hawp American Society for Metals, 1972.)
somewhat less resistance to chemical corrosion than 301 and 302, their
'
resistance to atmospheric corrosion is entirely comparable. 9-16
Precipitation-hardening Stainless Steels World War
structure of the same alloy after t
As a result of research duringf
a new group of stainless steels with precipitation-hardenmg;i p
characteristics were developed. The first of these nonstandard gradeso! \ stainless steels, 17-7PH was made available in 1948. The nominal chemicaT ;; ,
composition of some representative precipitation-hardening stainless ; steels is given in Table 9-5. These steels are usually solution-annealed at the mill and supplied in that condition. After forming they are aged to
at a higher temperature, 1100oF
that is more refined and has gn
structure shown in Fig. 9-15. Tai erties of some precipitation-har should not be put into service in ;
1
attain the increase in hardness and strength. In general they have lower I
nickel content, thus reducing the stability of the austenite. These steels* may also have elements such as copper and aluminum that tend to form coherent alloy precipitates.
dition because its ductility can b
corrosion cracking is poor. Asid<
aging also improves both tougl
The 17-7PH and PH15-7MO a
lowed by air cooling. This treatr about 5 to 20 percent delta ferrit(
The 17-4PH alloy is solution-treated at 1900oF followed by air cooling j* with the resultant transformation of austenite to martensite. Aging is carried out by reheating in the range from 900 to 1150°F to cause a precipi- \ tation effect. The lower temperature results in the highest strength anfl hardness. Figure 9-15 shows the microstructure of a 17-4PH alloy solution- \ treated at 1900;F and air-cooled, then aged for 4 h at 925CF and air-cooled , :
f
,
,
Transformation to martensite is essentially complete, and the martensile has been tempered by the aging treatment. Figure 9-16 shows the microTABLE 9-5
Nominal Composition of Precipitation-hardening Wrought Stainless Steels C
Mn
Si
Cr
Ni
Mo
OTHERS
%
%
%
%
%
%
%
0 04
0 40
0 50
16,50
4 25
0 25 Cb 3.60 Cu
aged ai 1100
17-7 PH
0 07
0 70
0 40
17.00
7 00
1 15 Al
PH 15-7 Mo
0 07
0 70
0 40
15.00
7 00
1 15 Al
and has greater ductility than the structure show Fig 9 15. Etched in Fry s reagent, 100X. (From IV
0 28 P
Handbook, vol. 7,
GRADE
-
..
,
.
1
17-4 PH
r
.
.
.
.
.
.
.
.
.
.
fig. 916 Same alloy and treatment as Fig. 9'15 ,
'
F
.
The tempered martensite is more '
17-10 P
,
0 12 .
,
0 75 .
.
0 50 .
17.00
.
10.50
2 25 .
.
.
Society for Metals, 1972.)
t
"
Atlas of Microstructures, Arr
"
'
mm
r . .
„;,
..
JRGY
ALLOY STEELS
375
1
typical mechanical properties of some
t
mes of national emergency has presei Sti ;s-steel producers and consumers
If
s
1
.
Itinued through the Korean emergency||
<5
mm 5
substitution of manganese for nickel ifl
i of types 201 and 202, the chromiunwif
Type 201 with a nominal composition p|
MS
d 6.5 percent Mn is a satisfactory substitj
i
m
mm
percent Ni) where machinability and sev
Jhg. 9-15 I7-4PH alloy solution treated at 1900°F and
W£>\ essential. Where those characteristic linal composition of 18 percent Cr, 5 pen
|iir-cooled, then aged 4 h at 925°F and air-cooled. Structure w tampered martensite. Etched in Fry's reagent, 100X. (From Metals Handbook, vol. 7, "Atlas of Microstructures,"
e desirable because the higher mangane
American Society for Metals, 1972.)
rdening. Although types 201 and 202 hi chemical corrosion than 301 and 302, th(
/rosion is entirely comparable.
Structure of the same alloy after the same treatment except that it was aged at a higher temperature 1100oF. The structure is tempered martensite
.
,
?ss Steels As a result of research durl stainless steels with precipitation-hardenlr
"
' -
. .
,
that is more refined and has greater ductility but lower strength than the structure shown in Fig. 9-15. Table 9-6 gives the nominal mechanical properties of some precipitation-hardening stainless steels. The 17-4PH alloy should not be put into service in any application in the solution-treated condition because its ductility can be relatively low and its resistance to stresscorrosion cracking is poor. Asidefrom the increase in strength and ductility aging also improves both toughness and resistance to stress-corrosion.
The first of these nonstandard grades / nade available in 1948. The nominal chemi ,3d
.
jsentative precipitation-hardening stainli These steels are usually solution-anneali lat condition. After forming they are
m
.
iss and strength. In general they have lo ta the stability of the austenite. These st as copper and aluminum that tend to form o
,
-
The 17-7PH and PHl5-7Mo alloys are solution-annealed at 1950oF followed by air cooling. This treatment produces a structure of austenite with about 5 to 20 percent delta ferrite (see Fig. 9-17a) In this condition the alloy .
pn-treated at 1900oF followed by air coolinil,
,
:
:
. ,
.
.
iation of austenite to martensite.
Aging
'
?
e range from 900 to 1150 F to cause a precipf|
'
°
hperature results in the highest strength antf
5
m .
the microstructure of a 17-4PH alloy, solutioft
.
ii
led, then aged for 4 h at 925°F and air-cooled| e is essentially complete, and the martensltdl ging treatment. Figure 9-16 shows the microns
IMS
1
.
Ni
Mo
OTHERS
%
%
%
%
Fig. 9-16 Same alloy and treatment as Fig 9-15, except .
.
50
16.50
4 25
40
17.00
7 00
40
15.00
7 00
50
17.00
10.50
aged at 1100°F The tempered martensite is more refined
1 15 Al
and has greater ductility than the structure shown in
.
.
.
0 25 Cb, 3.60 Cu .
.
2 25 .
.
1 15 Al
Fig. 9-15 Etched in Fry's reagent, 100X. (From Metals
0 28 P
Handbook
.
.
Wmk
»
f v/. ecipitation-hardening Wrought Stainless Steels Cr
11m
.
vol. 7, "Atlas of Microstructures," American Society for Metals 1972.) ,
,
i
m
376
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 9-6
CONDITION
GRADE
17-4 PH
I
17-7 PH
PH 15-7 Mo
17-10 P
9 17 Maraging Steels A series of i strengths up to 300,000 psi in
Nominal Mechanical Properties of Precipilalion-hardoning Wrought Stainless Steels' TENSILE
0 2% YIELD
ELONGA-
REDUCTION
STRENGTH,
STRENGTH,
TION IN 2 IN.,
OF AREA.
PSI
PSI
%
%
.
'
ROCKWELL
HARDNESS
'>:
ness was made available early
s
taining 18 to 25 percent nick(
Solution-annealed
150,000
1-10,000
10
45
C 33
H 900-Ft
200,000
178,000
12
48
C 44
H 1025oFt
170,000
165,006
15
56
C 38
H 1075=Ft H1150Ft
165,000 145,000
150,000 125 000
16
58
C 36
19
60
C33
,
Solution-annealed
130 000
40.000
35
B 85
TH 1050-F} RH 950°F§
200 000 235 000
9
C 43
,
185 000 220 000 ,
6
C 48
CH 950°Ffl
265 000
260 000
2
Solution-annealed
130 000
55 000
35
B 88
TH 1050°Ft RH 950-T§
210 000 240 000 ,
200 000 ,
7
C 44
,
225 000
6
C 48
Solution-annealed
89 000 143 000 .
37 000
70
76
B 82
.
98 000
20
32
C 32
Hardened, 1300' . 2-1 h
,
.
,
,
,
,
,
,
,
.
,
(
and are called maraging (mart be martensitic as annealed am
0
*
Adapted from "Metals Handbook," vol. 1, 1961, American Society lor Metals t Solution-annealed, then reheated al indicated temperature.
,
*
o /
V'-.l p
>
Metals Park. Ohio.
X Solution-annealed, reheated at 1400:F for 1 Vj h, air-cooled to room temperature, then reheated at 1050 for 1 Vi h. § Solution-annealed, reheated at 1750 for 10 min, air-cooled, then held at -100oF for 8 h, followed by reheating at950f for 1 h. i
if
Cold-rolled 60% then reheated at 950oF for 1 h. ,
'
[o]
is soft and may be easily formed. Several hardening sequences are possible. In the TH {temper, hard) sequence, the austenite is conditioned by reheating to 1400oF. This will precipitate chromium carbides thus reducing the carbon and chromium content of the austenite and allowing transfor,
mation on cooling. Cooling is continued to below 60oF but above 320F in
p
11 1
strength and ductility (see Fig. 9-17jb). The TH sequence gives better ductility but lower strength than other sequences (see Table 9-6). In the RH sequence, the austenite is conditioned at 1750oF. This higher temperature results,in more carbon in solution in the austenite and therefore a lower
Ms temperature. The transformation to martensite is obtained by a subzero treatment at -100oF with subsequent aging at 950oF (Fig. 9-17c). The 17-7PH alloy may also be supplied in the cold-rolled condition. Here, transformation is achieved by cold rolling, and heat treatment is reduced ton single aging at 950oF. Although strength is greatly increased, ductility is reduced and formability is limited. PH15-7Mo is a high-strength modification of 17-7PH and requires identical heat-treating procedures. The 17-1 OP alloy is a nickel stainless vyith 0.28 percent phosphorus. This alloy is solution-annealed at 2100oF followed by rapid cooling to produce a supersaturated austenitic matrix with excellent ductility and forming characteristics. Subsequent aging In the. range of 1200 to 1400oF will produce lower tensile strength than other precipitation-hardening alloys but much higher ductility (Table 9-6).
i
i
V
1
MS?
\
b
order to obtain the amount of martensite necessary for the strength levels desired. Aging is usually carried out at 1050oF for the best combination of a;
.
.
.
t
ft
Wml
;
'
i
Fig. 9-17 The structure of a 17-7PH alloy, (a)
treatment 1 h at 1950oF air-cooled. Ferrite islai austenite matrix, etched in Fry s reagent, 100X. treated as in (a) and reheated to 1400 F for IV-, ,
'
°
tc 60JF and held for /? h. aged for 1V5 h at 105 cooled (condition TH1050). Ferrite islands and chromium carbides in a martensite matrix; etcf '
(©agent, l,000X. (c) Solution-treated
as in (a) b
to 1750'F for 10 min and air-cooled, held 8 h a aged for 1 h at 950 F air-cooled (condition RH Stringers in a martensite matrix with less carbii °
,
solution compared with (5); etched in Vilelia's 1
000X. (From Metals Hanbook, vol. 7,
Atlas o
"
,
"
structures,
American Society for Metals, 1972
t
ALLOY STEELS
377
V
stlon-hardonlng Wrought Slalnlot* Sleda* 0 2% YIELD .
M STRENGTH
ELONGATION IN 2 IN.,
OF AREA,
%
%
110,000
10
45
C33
178.000
12
48
C 44
,
HAM
,
165 00(5
15
56
C 38
150 000
16
58
C36
125,000
19
60
C33
40,000
35
3 85
185 000
9
C 43 C 48
.
strengths up to 300 000 psi in combmation with excellent fracture toughness was made available early in 1960. These steels are low-carbon con-
RO
PSI
,
|7 Maraging Steels A series of iron-base alloys capable of attaining yield
-
REDUCTION
taining 18 to 25 percent nickel together with other hardening elements
,
,
220 000
6
260,000
2
55.000
35
0 88
200.000
7
C 44
,
6,
225 000 ,
C 48
37,000
70
76
3 82
98,000
20
32
C 32
an Society (or Metals, Metals Park, Ohio.
mm
lure.
pled to room temperature, then reheated at 1050'? for I'/j h. ooled. then held at -10CFF lor 8 h, followed by reheating at 950T
si
I
0
7
sd. Several hardening sequences are possiequence, the austenite is conditioned by re- ;, ecipitate chromium carbides, thus reducing ,
itent of the austenite and allowing transfercontinued to below 60oF but above 320F in
4
0]
I
'
j
jmartensite necessary for the strength levels
:
s
ed out at 1050oF for the best combination of v |..
j 9-17£)). The TH sequence gives better duo-
;
.
m
|
pther sequences (see Table 9-6). In the RH
.
I
ditioned at 1750oF. This higher temperature j |,
lition in the austenite and therefore a lower
f
liation
|;
to martensite is obtained by a subzero
3
' Ic,
3k
9-17c). The [ lied in the COld-rolled condition. Here, trans- || f\a 9M
psequent aging at 950oF (Fig
*
a:
.
.
rolling, and heat treatment is reduced to a igh strength is greatly increased, ductility is
The structure of a 17-7PH alloy, (a) After solution
treatment 1 h at 1950°F, air-cooled. Ferrite Islands in an
austenite matrix, etched in Fry's reagent 100X. (b) Solutiontreated as in (a) and reheated to 1400CF for 1V2 h air-cooled ,
,
ited. PHI 5-7Mo is a high-strength modifica- [ r ,0 60t.F and he|d for Vj h aged for 1 ,/a h at 1050op
'
:
-
/intical heat-treating procedures. stainless With 0.28 percent phosphorus. This „
, ,
..
.
2100 F followed by rapid cooling to produce ' '
f chromium carbides in a martensite matrix; etched in Vilella s . I
1
matrix with excellent ductility and forming 1
aging in the range Of 1200 to 1400°F will pro- l„
reagent, ),000X. (c) Solution-treated as in (a) but reheated to 1750'
for 10 min and air-cooled held 8 h at -100oF, ,
aged for 1 h at 950°F, air-cooled (condition RH950). Ferrite stringers in a martensite matrix with less carbide out of solution compared with (b); etched in Vilella s reagent, '
. . -i . , .... han other precipitation-hardening alloys but M , ooox (From Metals Ha 00k vo| 7 „At|as of M« ,
9-6)
.
S structures," American Society for Metals, 1972.) M
1
air.
! cooled (condition THIOSO). Ferrite islands and small
'
.
,
nd are called maraging (martensitic plus aging). They are considered to be martensitic as annealed and attain ultrahigh strength on being aged in
ro.
.
.I
378
.
INTRODUCTION TO PHYSICAL METALLURGY
V
o
o
in
m
the annealed, or martensitic, c
tough rather than the hard, I
d d I
I
o
o
steels.
d d
cold-worked to a high degree. i
o CO
o CO
T-
m T-
in T-
<
I I uo in TO
I m O
18 percent nickel grades use
d d d d d
small amounts of titanium am
to
co
i co
i co
co
l m O
Thus far, the commercial s
groups which differ in the har
d d d d d i in T-
This ductile martensite
f
percent nickel grades use titar
m
d d d :
t-
i m
i co
i m
:
d d T o
cn
in
H
5
in
lo
: -sf
Soak lo anneal auslenite and dissolve hardening
1300
elements Ti and Al
I100
(may be bypassed depending an hoi work finishing lemp)
co
O
2
1500
co IH
900
(allernatel
in
f 700
o
d co d i
t
in
o
I
also
Air-cool
V
I 500
I
o o
co K co
o
s,re
i
! in
2
300 o CO
o
o
o
o
o
o>
o>
o>
CM
i-
1-
i-
I
I
I
I
I
q o q o q
tri d cd K CM
T-
T-
T-
T-
<
o
o
o
o
o
5
d d d d d
X
.
X
01 0)
35
<
w 5
T~
1-
1-
l-
T-
o
o
o
o
o
d d d d d
i
100
i
100
-
i
I I
x
2 ra
S
<
<7 )
o -
i
o
o
o
o
-
i-
i-
i-
i
d d d d d
«
X
c <
2 5 *-
o -
T
o
o
o o
T-
T-
T-
T-
d d d d d
I
{
cold-work as desired
Soak lo anneal auslenite
and dissolve hardening 1300
elements Co and Mo
1100
(may be Dypassed depending on hot work finishing temp)
£ o>
I
T
Maroged(52 R/C)
1 H
1500
c '
J |
Martensite (26-35 Rc)
"
"
.
Age lo hcr
900
S 700
stress ieh occurs
Air-cool
CVJ
c
o '
S
x
CO
<
CO
CO
CO
q o o o d d d d
O 2
o
E o
o
z
o
p 1
<
m
CD
m
<
1
\
\ m
w
LU
Q
§
I
2 0 2 o lo o
I
Z Z 5 Z 2
1
in
o oo co co
C\J
CM
T-
T-
T-
? 500 300 100
Martensite 100
-
Maraged(52F
(28-32 R/C) cold-work as desired
Fig. 9-18 Heat-treating cycles for the maraglni {The International Nickel Company.)
r
i ALLOY STEELS
!RGY
-
o
the annealed or martensitic, condition. The martensite formed is soft and
o
,
tough rather than the hard brittle martensite of conventional low-alloy steels. This ductile martensite has a tow work-hardening rate and can be cold-worked to a high degree.
d d
Ml?
o CO
'
:
' - .
. -
.
.
I
3
,
o CO
d d A* o CO
o CO
id 1
1
Thus far, the commercial steels developed fall into two distinct alloy groups which differ in the hardening elements used (see Table 9-7). The
id 1
d d d d d i
i
i
i
i
LO I-
LO i-
LO o
LO o
LO o
18 percent nickel grades use mainly cobalt-molybdenum additions with small amounts of titanium and aluminum. The 20 percent nickel and 25 percent nickel grades use titanium-aluminum-colurnbium additions.
d d d d d
<
CO t
-
CO
LO
LO CM
d d d
:
l
: i
CD
l
l
I
i
CO CO LO CO ID d d T ! r-
25% Ni
-
4
;
o
CVJ
CVJ
LO
LO
LO
CO
IH
2
tf
1 H
1500
Soak to anneal dustenite and dissolve hardening
1300
elements Ti and Al
1100
(may be bypassed depending an hot work finishing temp) Age to harden,4H
5
ciii
o
co
IH
_
in
. .
!
in
| 700
o
I
I
o
LO o
o
o
od r-
od
| 500
c\j
cvj t-
I
I
o
o
I
T-
T-
I
o o
I o
lo d oo r- r-
I
a
.
<
o o o o o
2
d d d d d
<
o o o o o
2
d d d d d
X
o o o o o
<
65 2
t-
"r-
-y-
-
i
d d d d d
I
-
c <
2 2
o
o
o
o
o
d d d d d
i
I
1
co
<
co co
co
d d d d
1i
§
Z 2 2 Z H lo
o oo co oo
i-
t-
i-
Air-cool
Ms
300 -
100
Martensite (26-35 Rc)
ioo
Maraged(52 R/C)
cold-work as desired
00-
-
r .
1 l
Ms
I dora:)
t
Refrigerate
Cold-work
{
at least 25%
1 H
Soak to anneal austenite
18 % Ni
and dissolve hardening 1300
elements Co and Mo
1I00 -
(may be bypassed depending on hot work finishing temp) 3H
900
Age to harden; stress relief also
a>
S 700
occurs
Air-cool
I' 500 Ms
III i
o
4H
(alternate)
"
o o o o
O 2
1 H
r 900 -
J
Maraged
(52 R/C)
(optional for
ausaged, high Ti heats;needed for low Ti heats)
i cm
x
to 160 F
1100
(alternate)
I
1
Ausoge to precipitate elements to raise Ms
"
.
x
4H
Ausfenite
100
1500 X
dissolve hardening elements Ti and Al
I 500
I X
1500 -
700
300 o o o o o
stress relief also occurs
Air-cool
d od d I
20% Ni
Soakto anneal austenite and
1300 -
r*! (alle rnole)
900
.
379
300
100 Martensite 100
-
Maraged (52 R/C)
(28-32 R/C) cold-work as desired
im '9- '18 Heat-treating cycles for the maraging steels. « (Tho International Nickel Company.)
:
380
INTRODUCTION TO PHYSICAL METALLURGY
The greatest interest has been directed to the 18 percent nickel grades designed primarily for ultrahigh strength at room temperature. An important reason for the interest is their superior fracture toughness com- : pared with quenched and tempered medium-carbon ultrahigh strength ? steels. All that is required is a simple heat treatment carried out at mod- ; erate temperature to develop full strength in these steels (see Fig. 9-18). :;, Section size and heating and cooling rates are not important factors in the f hardening process. Since these steels are extra low in carbon content 1 ;
/
.
i
* . ' ; .
s
" .
it
.i;
,
decarburization is not a problem, and a protective atmosphere is not re- '. Fig. 9-19 The microstructure of an is percent nici
quired. The use of low aging temperatures reduces distortion to a minimum. \
Jll
This means that little machining or forming is required after hardening to ' jn moijified Fry s reagent, 250X. (The international '
produce parts v ith high dimensional accuracy. These steels are fully weldable and have good machinability in the annealed condition. The mechani-
. Company.)
cal properties of five maraging steels are given in Table 9-8. Figure 9-19 i
I
in annealed hardness and con;
martensitic matrix. The precise role of each of the elements in the harden- 1
as shown in Fig. 9-20, when moly cent cobalt, an increase in hard
shows the microstructure of an 18 percent nickel 250 grade maraging steel at 900oF for 3 h. The structure is made up of very fine precipitates in a
ing process is not completely understood, but some idea may be obtained 1; by considering the effect of alloying elements on the hardness of the mar- 1tensitic base. The hardness of the binary iron-nickel martensite is Rockwell 1
both elements is obtained.
C 25. A weak response to maraging is found after adding 7 percent cobalt § to the alloy base. The addition of molybdenum alone gives a slight increase
52 I
w
7 Co + MO 48
TABLE 9-8
Mechanical Properties of Maraging Steels
Mo
20%Ni
18%Ni
1 'I
25%N
Annealed: 40
'
Yield Strength, (0.2% off-
m
110
115
40
140
152
132
set), 1,000 psi
Soluhon anne
Tensile Strength, 1,000 psi
Elongation (in 1 in.), % Reduction in area, %
17
30
75
72
26-35
28-32
Hardness, R/C
Moroged v J
18Ni{200)t
10-15
18Ni(300)t
18Ni(250)t
Yield Strength, 1,000 psi
295-303
240-268
190-210
246
249-256
Tensile Strength, 1000 psi
297-306
250-275
200-220
256
265-270
§
After Maraging:
-
.
.
Elongation, %
12
10-12
14-16
11
12
Reduction in area %
60
45-58
65-70
45
53
1
1
,
*
The International Nickel Company. t Maraged 900T, 3 h. t Maraged 900°F, 1 h. § Conditioned 1300CF, 4 h, refrigerated, maraged 800-850:F, 1 h.
a
i
24
0
2
4
5
8
Percent molybdenum or percent cobo
Fig. 9 20 Cobalt additions alone have little effec
maraged hardness. Molybdenum additions alone Hie hardness of the iron-nickel martensite after £ SOO F. A strong additive effect of both cobalt anc molybdenum can be seen. (The International Nic Company.)
RGY
;
;; ;w
;
.
,
ALLOY STEELS
381
en directed to the 18 percent nickel gradef
l Sisgh strength at room temperature. An irn| it is their superior fracture toughness con$ Inpered medium-carbon ultrahigh streng a simple heat treatment carried out at modfull strength in these steels (see Fig. 9-18),-|
Ml
looling rates are not important factors in thtf|
:
5se steels are extra low in carbon content,
em, and a protective atmosphere is not re-:
Fig. 9-19 The microstructure of an 18 percent nickel ?250-grade maraging steel after aging at 900°F for 3 h.
; -V: mperatures reduces distortion to a minimum.
:
igcS'i
,
.
.
.
,,
.
.
.. .
.-Slructure is fine precipitates in a martensite matrix. Etched
ng or forming is required after hardening to «;|rt modifjed Fry,s reagent 250X (The ,nternational Nickel sional accuracy. These steels are fully v;eld- .» Company.)
ility in the annealed condition. The mechani- :W
|g steels are given in Table 9-8. Figure 9-19 |p n 18 percent nickel 250 grade maraging steel B-
vX jire is made up of very fine precipitates in a MSje role of each of the elements in the harden-
in annealed hardness and considerable maraging response. However as shown in Fig. 9-20 when molybdenum is added in the presence of 7 percent cobalt, an increase in hardness greater than the combined effect of
*
,
.
,
understood, but some idea may be obtained : |
both elements is obtained.
Jv: jiloying elements on the hardness of the mar- , i the binary iron-nickel martensite is Rockwell
raging is found after adding 7 percent cobalt
\
,
52
i of molybdenum alone gives a slight increase
7 Co + Mo
I :
48
ng Steels '
20%Ni
i
25%Ni
115
40
152
132
8
30
26-35
10-15
a
6
Moroged
32 Co
§
28
240-268
190-210
246
249-256
250-275
200-220
256
265-270
0
2
4
5
8
10
.
.
18Ni(200)t
Solution annealed
5G
72
18Ni(250)t
Mo
,1
.
.
*
0
10-12
14-16
11
12
45-58
65-70
45
53
Percent molybdenum or percent cobalt
f fig. 9-20 Cobalt additions alone have little effect on '
I maraged hardness. Molybdenum additions alone increase
the hardness of the iron-nickel martensite after aging at ift 900'F. A strong additive effect of both cobalt and 3-850oF 1 h. ,
molybdenum can be seen. (The International Nickel Company.)
382
INTRODUCTION TO PHYSICAL METALLURGY
Compared with the 18 percent nickel grades the 20 percent grade has
420
,
the advantage of lower alloy content and freedom from cobalt and molyb- | denum, which may be desirable for some applications and environments However, it is lower in toughness in resistance to stress-corrosion crack.
,
|
,
P
t
t
380 360
ing, and in dimensional stability during heat treatment. In comparison with t
340
the 25 percent nickel grade this steel does not require a conditioning
320
,
treatment at an intermediate temperature to become martensitic (see Fig 9-18) Also, because of its lower nickel content, it has an Ms temperature
Deformotio
400
30
r
.
.
above room temperature. However to ensure complete transformation to
to
380 -
Oeforrnationl 94
,
o
martensite, refrigeration at -ICF is recommended before aging. In contrast to the other grades of maraging steels the 25 percent nickel grade is largely austenitic after annealing. To reach high strength levels, this steel after forming must be completely transformed to martensite be-
'
60
' ,
,
40
2 20 v
.
300 30
260 >
for maraging. This may be accomplished in either of two ways: (1) ausaging
-
700
and (2) cold work and refrigeration (Fig. 9-18). In ausaging the steel is given a Conditioning treatment at 1300oF for 4 h after forming. This treatment reduces the stability of the austenitic structure by causing nickel-
800
900
I0O0 Deformatior
Fig. 9-22 Effect of temperature and deformation i and tensile strength of ausformed H11 hot-work tc
[Data from Vanadium-Alloys Steel Company.)
titanium compounds to precipitate from the austenitic solid solution and as a result it raises the Ms temperature so that austenite will transform to ; ,
Applications for maraging s
martensite on cooling to room temperature. In the second method the austenite is cold-worked at least 25 percent to start the transformation to martensite, which is completed by refrigeration at -100:'F.
motor cases for missiles, and lo
also been suggested for hot e and rifle tubing, and pressure v<
9 IB Femperat ure of deformation
420
900 "F 4
iS.380
900 "F
8 360 2 340
usforming A new type of exti been developed as a result of ai nique consists of deforming uns at a temperature below the A, pearlite and bainite reactions. prevent the formation of nonm;
,
en
sultant microstructure consists
320
persion of which are determir
30C
[
80 iioM
.
irenqth 40
1
300,000 psi (by conventional satisfactory ductility, by ausfoi completed fairly quickly to av Figure 9-21 shows that the tei
}5 .
s at
|
30
10 25 Hon
to
5
0
20
40
60
a
80
.
100
may be increased to the vicini
Percent deformation al indicaled temperature
Fig. 9-21
of ductility, by a 94 percent de
Effect on mechanical properties of deforming
austenite of H11 hot-work tool steel, austenitized at 1900T
amount of plastic deformation. steels, it has been used on Ai; in Table.10.1). The tensile stn
,
indicate that the temperature c
°
cooled to 900 F
deformed, cooled to room temperature, and tempered twice (double-tempered) at 950oF (Data from Vanadium-Alloys Steel Company.) ,
.
Wilful «i
4
-'
-
.
"'
r
structural parts for aircraft ai attractive applications for aust
# ALLOY STEELS
nt nickel grades, the 20 percent grade has
4?()
££|)ntent and freedom from cobalt and molyb'
'"" "
8
e for some applications and environments;;
CP
ss, in resistance to stress-corrosion crackfi r during heat treatment. In comparison with
Delormntion (percent)
100
380
,
5V (HID
bose tensile strength 310,000 psi
50
,
ihis steel does not require a conditioning1
.
0
Tempered al950F (5100
360
:
0 tOC-5.0CR-l.3MO
91
.
383
"
.
0
10
320
imperature to become martensitic (see Fig.: jer nickel content, it has an M5 temperature; Delorrnalion (percent)
/ever, to ensure complete transformation to
o
| fP is recommended before aging. gWtes of maraging steels, the 25 percent nickflj; '
w
;
-
-.
.
'
.
3
300
He completely transformed to martensite be-
5
280 800
900
1000
1100
1200
1300
1400
1500
Deformation temperature(F)
fig. 9-22 Effect of temperature and deformation on yield Snd tensile strength of ausforfhed H11 hot-work tool steel.
V
.
aitate from the austenitic solid solution, ancj
'
30
700
t at 1300oF for 4 h after forming. This treatr
Tempered ot 950 F(5t0C) lnn.e yield ' itronglh MO OOOpsi .
jr annealing. To reach high strength levels,
V|the austenitic structure by causing hickei
.
.
310
.
jmplished in either of two ways: (1) ausagingj ration (Fig. 918). In ausaging the steel is :;
0 1OC-5.0CR-I.3M0O bV (HII)
94
350
(Data from Vanadium-Alloys Steel Company.)
f
e nperature so that austenite will transform to Applications for mafaging steels are hulls for hydrospace vehicles and low-temperature structural parts. They have
temperature. In the second method the aus?|
:
motor cases for rihissiles
t 25 percent to start the transformation td| d by refrigeration at
,
also been suggested for hot extrusion dies, cold-headed bolts, mortal and rifle tubing, and pressure vessels.
100"F.
9 IS .
usforming A new type of extremely high-strength steels have recently been developed as a result of ausforming or austenitic forming. The technique consists of deforming unstable austenite of moderately alloyed steels
peral ure o( (or met ion
900 "F
"
at a temperature below the A, line in the bay" that exists between tht pearlite and bainite reactions, followed immediately by oil quenching tc prevent the formation of nonmartensitic transformation products. The re sultant microstrucfure consists of fine martensitic plates, the size and dispersion of which are determined by prior austenitic grain size and thf amount of plastic deformation. While the technique is applicable to man
900 "F
steels, it has been used on AISi 4340 and H11 hot-work tool steel (listec in Table 10.1). The tensile strength of 4340 steel may be increased frorr 300,000 psi (by conventional heat treatment) to over 400,000 psi, witf satisfactory ductility, by ausforming. The hot-working operation must be . completed fairly quickly to avoid the transformation to softer products Figure 9.21 shows that the tensile properties for H11 steel in sheet torn may be increased to the vicinity of 400 000 psi, maintaining a useful leve of ductility, by a 94 percent deformation at 900°F. The curves of Fig 9.2; indicate that the temperature of deformation is not critical Highly stresse( structural parts for aircraft and automotive leaf springs are particular!; attractive applications for ausforming.
30 s s -
> 125 g 100
lO mperature
,
J&i deforming
i
.
"
ized at 1900oF,
,
|mperature,
.
ITF. (Data
.
5
.
:
4
-
.
384
INTRODUCTION TO PHYSICAL METALLURGY
QUESTIONS 9-1
:
V;';~.*::;
9-2
of a »9-3 9-4
Bilby, Glover, and Wakeman: "Mod
Explain how alloying elements that dissolve in ferrite increase its strength What effect would the addition of 1 percent chromium have on the properties low-carbon steel? A high-carbop steel? What factors determine the wear-resisting properties of a steel? Devise and explain a practical test to compare the wear resistance of different
steels. 9-5
i
'
i
ii
of Metallurgists, London, 1967. Special Steels,"
Burnham, T. H.:
.
i
;
1933. "
I
If the primary consideration is hardenability which alloy steel should be se-
Hall, A. M., and C. J. Slunder: T 18 Per Cent Nickel Maraging Stt
Space Administration, Washingto
Hull, A. M.: "Nickel in Iron and Stec International Nickel Co.: Nickel Allc "
,
lected? 9-6
,
Monypenny, J. H. G.: "Stainless Iro
Look up the chemical composition of 4340 steel (Table 7-2) This steel is to be
Londor, 1951-1954.
.
used as a structural member in an aircraft landing-gear assembly. What mechanical properties would be desirable for this application? Describe a heat treatment that would be applied to obtain these properties. 9'7 Same as Question 9-6 for 6150 steel to be used as a front coil spring in an auto-
Parr, J. G., and A. Hanson: "An Int for Metals, Metals Park, Ohio, 19( Thum, E. E.: "Book of Stainless Ste Park, Ohio, 1935.
mobile.
US .
Look up the chemical composition of 4620 steel On the basis of chemical composition, what mechanical properties would you expect this steel to have? 9-8
a
. 9-9
"
1949.
Why was manganese chosen as a substitute for nickel in the development of the
9-10 Which stainless steel is best suited for surgical instruments? Explain . 9-11 Why do some pots and pans have a copper bottom and stainless steel inside? What type of stainless steel is best for this application? Why? 9-12 What is the difference in chemical composition between the standard stainless steels and the precipitation-hardenable stainless steels? 9-13 Describe three ways in which precipitation-hardenable stainless steels maybe .
hardened.
Why does aging 17-4PH at 1150oF give lower mechanical properties than aging
at 10500F? 9-15
What effect does reduction of the carbon and chromium content of the aus-
tenite have on the Ms temperature? 9-16 Why is refrigeration or subzero treatment sometimes used for precipitationhardening stainless steels? 9-17 Explain the term maraging .
9-18
What is the difference In heat treatment between the 18 percent Nl and the
25 percent Ni maraging steels? 9-19 Give at least four advantages of maraging steels as compared to regular stainless steels.
REFERENCES
Allegheny Ludlum Steel Corp.: "Stainless Steel Handbook," Pittsburgh American Iron and Steel Institute: "Stainless and Heat-Resisting Steels
,
"
,
1951. Steel Prod-
ucts Manual, New York, 1963.
Americash Society for Metals: "Metals Handbook," 7th ed 1948; 1954 Supplement, .
,
8th ed., vol. 1, 1961, vol. 2, 1964 and vol. 7, 1972, Metals Park, Ohio
.
Archer, R. S., J. F. Briggs, and C. M. Loeb: "Molybdenum; Steels, Irons Alloys," Climax Molybdenum Co., NeW York, 1948. Bain, E. C, and H. W. Paxton: "Alloying Elements in Steel 2d ed., American Society ,
"
,
for Metals, Metals Park, Ohio, 1961.
0
1
Steel Corp.: "The Making, She
Zapffe, C. A.: "Stainless Steels,
2xx series of stainless steels?
9-14
.
1957.
.
Specify a heat treatment for this steel and give some possible applications. n
"
.
ALLOY STEELS
?85
Bilby, Glover and Wakeman: "Modern Theory in the Design of Alloys," The Institute of Metallurgists London, 1967. ,
,
Si at dissolve in ferrite increase its strength. p cv i percent chromium have on the properties |
Bumham, T. H.: "Special Steels," 2d ed., Sir Isaac Pitman & Sons, Ltd., London,
"
v :
,
.
1933.
steel?
Hall, A. M., and C. J. Slunder: "The Metallurgy, Behavior and Application of the 18 Per Cent Nickel Maraging Steels," NASA SP-5051, National Aeronautics and
resisting properties of a steel?
jt to compare the wear resistance of different
Space Administration, Washington, D.C., 1968. Hull, A. M.: "Nickel in Iron and Steel," John Wiley & Sons, Inc., New York, 1954. International Nickel Co.: "Nickel Alloy Steels," 2d ed., New York, 1949.
prdenability, which alloy steel should be s4-
Monypenny, J. H. G.: "Stainless Iron and Steel," 2 vi s., 3d ed.. Chapman and Hall,
[n of 4340 steel (Table 7-2). This steel is to be
London, 1951-1954.
Saft landing-gear assembly. What mechanical application? Describe a heat treatment that
.
Parr, J. G., and A. Hahson: "An Introduction to Stainless Steel," American Society for Metals, Metals Park, Ohio. 1965.
-6 rties.
.
Thum, E. E.: "Book of Stainless Steels," 2d ed., American Society for Metals, Metals
lei to be used as a front coil spring in an autobn of 4620 steel.
On the basis of chemical
rties would you expect this steel to have? and give some possible applications.
substitute for nickel in the development of the ted for surgical instruments? Explain.
e a copper bottom and stainless steel inside? this application? Why?
cal composition between the standard stainnable stainless steels?
scipitation-hardenable stainless steels may be P give lower mechanical properties than aging ie carbon and chromium content of the aus-
treatment sometimes used for precipitation-
; ;
treatment between the 18 percent Ni and the
of maraging steels as compared to regular
sss Steel Handbook,
"
Pittsburgh, 1951.
g&Anless and Heat-Resisting Steels," Steel ProdHandbook," 7th ed., 1948; 1954 Supplement; vol. 7, 1972, Metals Park, Ohio.
fM Loeb: "Molybdenum; Steels. Irons. Alloys." '
; ' 1948. i) Elements in Steel, 2d ed., American Society
'
,
"
.
Park, Ohio, 1935. US .
t
.
Steel Corp.: "The Making, Shaping and Treating of Steel," 7th ed., Pittsburgh,
1957.
Zapffe, C. A.: "Stainless Steels," American Society for Metals, Metals Park, Ohio, 1949.
.
sr. J
-
;
I
'
-
..H
m i
10 1 Classification of Tool Steels classed as a tool steel.
How
quality special steels used fot There are several methods
according to the quenching i oil-hardening steels, and air-
means of classification, such
and medium-alloy tool steels. application of the tool stee
steels, high-speed steels, and The method of identificatior
by the AISI (American Iron ; quenching, applications, spt industries. The commonly us
major headings, and each grc betical letter as follows: GROUP
SYMBOL
Water-hardening
W
Shock-resisting
S
Cold-work
0
Hot-work
Oil-hs
A
Medit
D
High-i
H
(Hl-H H20-I H40-I
High-speed !!
Mold
T
Tung;
M
Molyt
P
Mold carbo
Special-purpose
TJ -rW":! .
\
i
.
L
Low-;
F
Carbc
'
\ v. -y*
\
5
TOOL STEELS' 1
101
Classification of Tool Steels
Any steel used as a tool may be technically
classed as a tool steel. However, the term is usuaJy restricted to highquality special steels used for cutting or forming purposes. \
53
There are several methods of classifying tool steels. One method is according to the quenching media used, such as water-hardening steels, oil-hardening steels, and air-hardening steels. Alloy content is another means of classification, such as carbon tool steels, low-alloy tool steels, and medium-alloy tool steels. A final method of grouping is based on the application of the tool steel, such as hot-work steels, shock-resisting steels, high-speed steels, and cold-work steels.
The method of identification and type classification of tool steels adopted by the AISI (American Iron and Steel Institute) includes the method of quenching, applications, special characteristics, and steels for special industries. The commonly used tool steels have been grouped into seven major headings, and each group or subgroup has been assigned an alphabetical letter as follows: GROUP
SYMBOL AND TYPE
Water-hardening
W
Shock-resisting
S
Cold-work
o A
D
Oil-hardening Medium-alloy air-hardening High-carbon high-chromium
Hot-work
H
{H1-H19, incl., chromium-base; H20-H39, incl., tungsten-base; H40-H59, incl., molybdenum-base)
High-speed
T
Tungsten-base Molybdenum-base
M Moid
P
Mold steels {P1-P19, incl., lowcarbon; P20-P39, incl., other types)
Special-purpose
L
Low-alloy
F
Carbon-tungsten
IP-
.i i '
; :t
v
.
TABLE 10 1
u
Identification and Type Classification of Tool Steels*
CO
IDENTIFYING ELEMENTS, % -I
3
Type C
Si
Mn
W1
0 60/1.40+
W2
0 60/1.40*
V
Ni
Or
O
W
Mo
Co
Al
C
o
WATER HARDENING TOOL STEELS
H
SYMBOL W
o Z H
.
o
0 25
.
W5
o
o
.
1 10
0 50
.
<
.
SHOCK-RESISTING TOOL STEELS
o
i:
> 1
SYMBOL S
-
K
SI
0 50
0 50
85
0 55
.
.
m
2 50
1 50
.
S2
.
1 00
0 50
2 00
0 40
.
0 80
87
.
>
.
! .
c
.
3
0 50
3 25
.
1 40
.
.
<
COLD-WORK TOOL STEELS
SYMBOL 0, OIL-HARDENING TYPES 01
0 90
1 00
0 90
1 60
.
02
.
.
06?
1 45
07
1 20
0 50
0 50
0 75
1 75
.
.
.
1 00
.
0 25
.
.
.
.
.
SYMBOL A, MEDIUM ALLOY AIR HARDENING TYPES A2
1 00
5 00
A3
1 25
5 00
.
A4
2 00
A6
0 70
2 00
A7
2 25
5 25
0 55
5 00
0 50
5 00
.
.
1 00
I
.
1 00
.
.
4 75
.
1 35
1 80
1 80 .
.
1 25
.
.
1 40
.
.
1 25
.
.
1 00
1 50
.
.
1 25
I
.
i 00
1 00
.
.
.
A10+
.
1 00
.
.
'
1 00
.
1 00
.
.
A9
.
1 00
.
1 00
A8
1 00
.
.
.
1 50
;
.
. V
' --
IDENTIFYING ELEMENTS, %
Type
T
Mn
c
Si
Cr
NI
V
w
Mo
Co
Al
SYMBOL D, HIGH-CARBON HIGH-CHROMIUM TYPES T
1 00
02
1 50
12.00
D3
2 25
12.00
D4
2 25
12.00
1 00
05
1 50
12.00
1 00
D7
2 35
12.00
.
.
.
.
.
.
.
4 00 .
1 00
3 00 .
I
i:
.
.
V
HOT-WORK TOOL STEELS
[
SYMBOL H
H10
0 40 .
H1-H19, INCL., CHROMIUM-BASE TYPES T 0 40 25 3
.
2 50 .
.
0 40 .
1 50 .
r
.
i
-
.:
v v; .
i* -
S7
0 50
1 40
3 25
.
.
.
<
COLD-WORK TOOL STEELS
SYMBOL O, OIL-HARDENING TYPES i
01
0 90
1 00
02
0 90
1 60
06?
1 45
07
1 20
A2
1 00
5 00
A3
1 25
5 00
A4
1 00
2 00
A6
0 70
2 00
A7
2 25
5 25
A8
0 55
5 00
A9
0 50
5 00
.
.
.
0 50
0 50
0 75
1 75
.
.
.
1 00
0 25
.
.
.
.
.
.
SYMBOL A, MEDIUM ALLOY AIR HARDENING TYPES .
.
.
.
.
1 00
.
1 00 .
.
1 00
1 00
.
.
.
4 75
.
.
25
1 25 .
1 25
.
.
.
1 40
1 00
50
.
1 80
1 00
1 00
.
.
.
35
.
1 00
1 00
.
.
A10±
1 00
.
.
.
.
50
80
.
IDENTIFYING ELEMENTS % .
Type
Mn
C
Si
Cr
Mo
V
Mi
Co
Al
SYMBOL D, HIGH-CARBON HIGH-CHROMIUM TYPES
I
1 00
D2
1 50
12.00
D3
2 25
12.00
D4
2 25
12.00
1 00
D5
1 50
12.00
1 00
07
2 35
12.00
.
.
.
.
.
.
.
.
3 00 .
1 00
4 00
.
.
HOT-WORK TOOL STEELS SYMBOL H
H1-H19, INCL CHROMIUM-BASE TYPES ,
H10
0 40
3 25
0 40
2 50
H11
0 35
5 00
0 40
1 50
HI 2
0 35
5 00
0 40
H13
0 35
5 00
1 00
H14
0 40
5 00
H19
0 40
4 25
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1 50 .
1 50 .
1 50' .
.
5 00
.
.
2 00
.
.
4 25
4 25
.
.
H20-H39. INCL. TUNGSTEN BASE TYPES {H27-H39 UNASSIGMED) H21
0 35
3 50
9 00
H22
0 35
2 00 .
11.00
H23
0 30
12.00
12.00
1124
0 45
3 00
15.00
H25
0 25
4 00
H26
0 50
4 00
.
.
.
.
.
.
.
.
.
15.00
.
1 00
.
.
18.00
I
H40-H59, INCL, MOLYBDENUM-BASE TYPES (H40 H44-H59 UNASSIGNED) ,
O O -
r
H41
0 65
4 00
1 00
1 50
8 00
H42
0 60
4 00
2 00
6 00
5 00
m
0 55
4 00
8 00
cn
.
.
.
.
.
05
i
-
H43
.
.
.
.
.
2 00 .
.
.
.
m
CO oa
1
m
,
u CO o
TABLE 10 1
Type
(Continued)
I
C
IDENTIFYING ELEMENTS, % Mn
SI
Qr
Nl
V
H JJ
W
Mo
Co
A!
O O C
o
HIGH-SPEED TOOL STEELS
H
SYMBOL T, TUNGSTEN-BASE TYPES
O z
.
T1
0 70
4 00
1 00
18.00
.
.
.
H
O
T2
0 80
4 00
2 00
18.00
T4
0 75
4 00
1 00
18.00
5 00
T5
0 80
4 00
2 00
18.00
8 00
T6
0 80
4 50
1 50
20.00
12.00
T8
0 75
4 00
2 00
14.00
5 00
T15
1 50
4 00
5 00
12.00
5 00
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
TJ X
<
.
cn
o >
.
r
-
S
.
m
.
>
SYMBOL M, MOLYBDENUM-BASE TYPES Ml M2
0 80 .
4 00
1 00
1 50
8 00
4 00
2 00
6 00
5 00
.
.
0 85/1.00+ .
.
.
.
c jj
a
.
.
<
.
M3
1 05
4 00
2 40
6 00
5 00
M4
1 30
4 00
4 00
5 50
4 50
M6
0 80
4 00
1 50
4 00
5 00
M7
1 00
4 00
2 00
1 75
8 75
M10
0 85
4 00
2 00
M30
0 80
4 00
1 25
2 00
8 00
5 00
M34
0 90
4 00
2 00
2 00
8 00
8 00
M36
0 80
4 00
2 00
6 00
5 00
8 00
M41
1 10
4 25
2 00
6 75
3 75
5 00
M42
1 10
3 75
1 15
1 50
9 50
8 00
M43
1 20
3 75
1 60
2 75
8 00
8 25
M44
1 50
4 25
2 25
5 25
6 25
12.00
M46
1 25
4 00
3 20
2 00
8 25
8 25
M47
1 10
3 75
1 25
1 50
9 50
5 00
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
12.00
.
8 00
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
I
.
:
.
.
.
.
I
IDENTIFYING ELEMENTS, % Type
Mn
C
Si
Cr
Ni
V
W
.
Mo
SPECIAL-PURPOSE TOOL STEELS SYMBOL L. LOW-ALLOY TYPES L2 L3 L6
0 50/1.10t
1 00
0 20
1 50
0 20
.
.
.
1 00
.
.
0 25
.
0 70
0 75
1 5C
.
.
.
.
SYMBOL F, CARBON-TUNGSTEN TYPES 1 25 .
F1
1 00
F?
1 25
.
3 50 .
.
Co
Al
HBO
i uo
1 50
4 00
2 00
B 00
5 00
4 00
2 40
6 00
5 00
4 00
4 00
5 50
4 50
4 00
1 50
4 00
5 00
4 00
2 00
75
8 75
.
M2
0 85/1.OOt .
M3
.
05
M4
.
1 30 .
M6
.
.
.
.
0 80
M7
.
.
.
00
.
.
.
.
.
.
.
.
.
.
.
12.00
.
.
M10
0 85
4 00
2 00
M30
0 80
4 00
1 25
2 00
8 00
5 00
4 00
2 00
2 00
8 00
8 00
4 00
2 00
6 00
5 00
8 00
4 25
2 00
6 75
3 75
5 00
1 15
1 50
9 50
8 00
.
.
.
M34
0 90 .
M36
.
M41
.
.
.
0 80
.
.
10
.
.
M42
1 10
.
3 75
.
8 00
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
M43
1 20
3 75
1 60
2 75
8 00
8 25
M44
1 50
4 25
2 25
5 25
5 25
12.00
1 25
4 00
3 20
2 00
8 25
8 25
10
3 75
1 25
1 50
9 50
5 00
.
.
.
M46
.
M47
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
-
IDENTIFYING ELEMENTS, %
Type
Mn
C
Cr
Si
Ni
V
W
Co
Mo
AI
SPECIAL-PURPOSE TOOL STEELS SYMBOL L. LOW-ALLOY TYPES 1 00
0 20
L3
1 00
1 50
0 20
L6
0 70
0 75
L2
0 50/1.10t
.
.
.
.
.
.
0 25
1 50
.
.
.
.
SYMBOL F, CARBON-TUNGSTEN TYPES F1
1 00
1 25
F2
1 25
3 50
.
.
.
.
MOLD STEELS, SYMBOL P
PI-PI9, INCL, LOW-CARBON TYPES (P7-P19 UNASSIGNED) i
P2
0 07
P3
0 10
.
2 00
0 50
0.60
1 25
.
|
.
P4
0 07
i
5.00
P5
0 10
j
2.25
PS
0 10
.
.
.
1 50
.
0 20 .
.
3 50
.
.
i
_
_
P20-P39, INCL OTHER TYPES (P22-P39 UNASSIGNED) ,
t
P20
0 30
P21
0 20
0 25
1 25
.
.
.
4 00
.
.
.From Steel Products Manual
,
"
Tool Steels," American Iron and Steel Institute
tVarying carbon contents may be available. Contains free graphite in the microstructure to improve machlnability.
,
January, 1970.
1 20 .
O O r
CO H m m
CO w
392
INTRODUCTION TO PHYSICAL METALLURGY
The AISI identification and type classification of tool steels is given in ....
-
From the above discussion
»
Table 10-1. 10-2 Selection of Tool Steels The selection of a proper tool steel for a given | application is a difficult task. The,best approach is to correlate the metal-
I
important selection factors in cations, many other factors rr
lurgical characteristics of tool steels with the requirements of the tool in
the amount of distortion whic
operation.
eration; the amount of surfac
In most cases, the choice of a tool steel is not limited to a single type or even to a particular family for a working solution to an individual tooling problem. Although many tool steels will perform on any given job, they will have to be judged on the basis of expected productivity ease of fabrication and cost. In the final analysis it is the cost per unit part made by the tool
hardenability or depth of hare heat checking; heat-treating mospheres, and equipment; at 10 3 Comparative Properties The c tool steels are given in Table 1
that determines the proper selection. Most tool-steel applications with the exception of those to be made
ance, nondeforming propertie
into machine parts, may be divided into types of operations; cutting shearing, forming, drawing, extrusion, rolling, and battering. A cutting
poor.
tool may have a single cutting edge which is in continuous contact with the
class. The hot-work steels hav
work, such as a lathe or planer tool; or it may have two or more cutting edges which do continuous cutting, such as a drill or a tap; or it may have
ness, but significant differem
a number of cutting edges, with each edge taking short cuts and function-
the high speed steels are ratei this property between the di
,
,
,
resistance to decarburization
,
,
i
hardness, toughness, wear n
and a steel to be selected for h
ing only part of the time, such as a milling cutter or hob. When cutting is the chief function of the tool steel, it should have high hardness as well as
:
good heat and wear resistance.
|
Shearing tools for use in shears, punches, and blanking dies require high wear resistance and fair toughness. These characteristics must
Depth of hardening is r
have been rated relative to eac
taken into account for a partic
10 4 Nondeforming Properties
{
ening temperatures.
he properly balanced, depending on the tool design, thickness of the stock
The
the basis of distortion obtained
Since i
and quenching, the extent to for complex shapes. Intricat their shape after hardening. 7
being sheared, and temperature of the shearing operation. Forming tools are characterized by imparting their form to the part being made. This may be done by forcing the solid metal into the tool impression either hot or cold by using a hot-forging or cold-heading die. This group
ing properties can be machin so that little grinding will be
also includes dies for die casting where the molten or semimolten metal is
\
rather drastic section change;
forced under pressure into the form of the die. Forming tools must have high toughness and high strength, and many require high red-hardness (resistance to heat softening). Drawing and extrusion dies are characterized by substantial slippage between the metal being formed and the tool itself. Deep-drawing dies such as those used for the forming of cartridge cases generally require high strength and high wear resistance. Toughness to withstand outward
s
ject to excessive warpage in
|
the least distortion; those qu
lead to cracking during harde water-hardening steels show associatedx with the tempera
changes during heating and c residual stresses in Chap. 8.
pressures and wear resistance is most important for cold-extrusion dies,
10 5 Depth of Hardening This is r
whereas dies for hot extrusion require, in addition, high red-hardness. Thread-rolling dies must be hard enough to withstand the forces in forming the thread and must have sufficient wear resistance and toughness to adjust to the stresses developed. Battering tools include chisels and all forms of tools involving heavy shock loads. The most important characteristic for these tools is high toughness.
tool steels. The hardenability the recommended quenching as the carbon tool steels (gro and several of the carburizinc
water. The hardenability incr
alloying element which decrc
1
' '
V
V
JRGY
TOOL STEELS
393
.
type classification of tool steels is given lii
of expected productivity, ease of fabricationil it is the cost per unit part made by thetool|
From the above discussion, it is apparent that, for most applications, hardness touqhness. wear resistance, and red-hardness are the most important selection fj|ctorsJjT choosing tool steels. In individual applications, many other factors must be seriously considered. They include the amount of distortion which is permissible in the shape under consideration; the amount of surface decarburization which can be tolerated; hardenability or depth of hardness which can be obtained; resistance to heat checking; heat-treating requirements, including temperatures, atmospheres, and equipment; and finally, the machinability. 1 10 3 Comparative Properties The comparative properties of the most common tool steels are given in Table 10-2. Toughness, red-hardness, wear resist-
with the exception of those to be madd|
ance, nondeforming properties.jnachinability, safety in hardening, and resistance to decarburizatigp have been qualitativeiv rated goQdJFaLil-Or
Sr.
election of a proper tool steel for a givei
| The best approach is to correlate the metail
„
il steels with the requirements of the tool ini |f a tool steel is not limited to a single type off
I
working solution to an individual tooling! jsteels will perform on any given job, they willj I" a
,
J
I
divided into types of operations: cutting,-5 extrusion, rolling, and battering.
A cutting |
I
'
edge which is in continuous contact with thajjs
: ler tool; or it may have two or more cutting
.
'
I
,
iplection.
#%sutting, such as a drill or a tap; or it may havej jth each edge taking short cuts and function-;! as a milling cutter or hob. When cutting is | '
_
P
f
I
steel, it should have high hardness as well as I ji 10-4
'
poor. Depth of hardening is rated shallow, medium, or deep. The steels have been rated relative to each other rather than within anyone particular class. The hot-work steels have been rated good or very good in red-hardness, but significant differences exist in hot strength" of these steels, and a steel to be selected for hot die work requires careful study. Although the high speed steels are rated poor in toughness, there are differences in this property between the different high-speed steels which should be taken into account for a particular application. "
-
shears, punches, and blanking dies require 3 lir toughness.
These characteristics must
iing on the tool design, thickness of the stockS jre of the shearing operation. ;r
.
'
ized by imparting their form to the part being { )rcing the solid metal into the tool impression
f
II
ening temperatures. Since steels expand and contract during heating and quenching, the extent to which dimensions change is most important for complex shapes. Intricately designed tools and dies must maintain their shape after hardening. Those steels rated good or best in nondeforming properties can be machined very close to size before heat treatment,
I
so that little grinding will be required after hardening. Parts that involve "
rather drastic section changes should not be made of steels which are sub-
e form of the die. Forming tools must have.|
;
ject to excessive warpage in heating or quenching, as this will generally lead to cracking during hardening. In general, air-hardening steels exhibit
ngth, and many require high red-hardness J|
s are characterized by substantial slippage | ned and the tool itself.
Deep-drawing dies
orming of cartridge cases generally requirol -
resistance. Toughness to withstand outward 1
re is most important for cold-extrusion dies
'
'
.
.
,
I 105
I
i require, in addition, high red-hardness,
The tool steels in Table 10-2 have been rated on
the basis of distortion obtained in hardening from the recommended hard-
lot-forging or cold-heading die. This group! : ing where the molten or semlmolten metal is I K
:
Nondeforming Properties
hard enough to withstand the forces in form-|j &I sufficient wear resistance and toughness to:
the least distortion; those quenched in oil show moderate distortion; and water-hardening steels show the greatest distortion. The distortion is associated with the temperature gradient and the resulting dimensional changes during heating and cooling, which were discussed in detail under residual stresses in Chap. 8. Depth of Hardening This is related to the hardenability of the individual tool steels. The hardenability ratings in Table 10-2 are based on the use of
-
the recommended quenching medium. The shallow-hardening steels such as the carbon tool steels (group W) the tungsten finishing steels (group F), ,
ped. Battering tools include chisels and all| shock loads. The most important character*! ughness.
;i
and several of the carburizing grades in group P are generally quenched in
water. The hardenability increases with increasing alloy content. The only alloying element which decreases hardenability is cobalt. To develop high
394
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 10-2
Comparative Properties of Some Tool Steels*
I NONfI-: DEFORMING
APPROX STEEL
HARDENING
TYPE
0F
W1
,
1400-1550
QUENCHING
TEMPERING
HARDNESS
MEDIUM
RANGE 0F
ROCKWELLCt
Brine or
,
.
SAFETY IN
TOUGH-
RED-
HARDENING ;: h
PPOPERTIES
HARDENING
NESS
HARD1
Fair
Good
Poor
DEPTH OF
300-650
65-50
Shallow
Poor
300-650
65-50
Shallow
Poor
Fair
Good
Poor Fair
water
W2
1400-1550
Brine or water
S1
1650-1800
Oil
400-1200
58- 40
Medium
Fair
Good
Very good
S5
1600-1700
Oil
350-800
60- 50
Medium
Fair
Good
Best
Fair
01
1450-1500
Oil
300-500
62-57
Medium
Very good
Very good
Fair
Poor
A2
1700-1800
Air
350-1000
62-57
Best
Fair .
Fair
1500-1600
Air
350-800
62- 54
Deep Deep
Best
A4
Best
Best
Fair
Fair
02
1800-1975
Air
400-1000
61- 54
Best
Best
Poor
Good
D3
1700-1800
Oil
400-1000
61-54
Very good
Good
Poor
Good
D4
1775-1850
Air
400-1000
61- 54
H11
1825-1875
Air
1000-1200
54-38
H19
2000-2200
Air or oil
1000-1300
59- 40
H21
2000-2200
Airoroil
1100-1250
54-36
'
H23 H26 H41
2200-2350
2150-2300 2000-2175
T1
2300-2375
T4
2300-2375
T6
2325-2400
M1
2150-2225
M2
2175-2250
M6
2150-2200
M41 L2
Air or oil
Salt, oil, or air Salt, oil, or air
1200-1500
1050-1250 1050-1200
47-30 58-43 60- 50
Deep Deep Deep Deep Deep Deep
B«st
Best
Poor
Good
Very good
Best
Good
Good
Good
Good
Good
Good
Air: good
Good
Good
Good
Deep
Air: good
Good
Fair
Veryg
Good
Fair
Very g
Fair
Poor
Veryg
Good
Good
Poor
Veryg
Good
Fair
Poor
Best
Good
Fair
Poor
Best
Good
Fair
Poor
Good
Fair
Poor
Good
Fair
Poor
Good
Fair
Poor
Veryg Very g Veryg Veryg
Water: poor
Water: poor
Very good
Poor
Oil: fair
Oil: fair
i
Salt, air: good
Deep
1
Salt, air: good
1000-1100
65- 60
1000-1100
66- 62
or salt
1000-1100
65-60
or salt
1000-1100
65-60
Oil, air, or salt Oil, air, or salt
1000-1100
65- 60
1000-1100
66- 61
2175-2220
Oil, air, or salt
1000-1100
70-65
Deep Deep Deep Deep Deep Deep Deep
1450-1550
Water
350-1000
63- 45
Medium
1550-1700
Oil
air, air, air, air,
Oil: fair
Deep
or salt or salt
Oil, Oil, Oil, Oil,
Oil: fair
:
Oil: fair
Oil: fair ;
L6
1475-1550
Oil
350-1000
62- 45
Medium
Good
Good
Very good
Poor
F2
1450-1600
Water or
300-500
66-62
Shallow
Poor
Poor
Poor
Poor
P2
1525-15503
Oil
300-500
64- 585
Shallow
Good
Good
Good
Poor
P20
1500-1600
Oil
900-1100
37-28
Shallow
Geo a
Good
Good
Poor
brine
Adapted from tables tn Sloel Products Manual, "Tool Steels +After tempering tAfter carburizing. "
American Iron and Slee! InsliUite. 1970.
.
.
SCarburized case hardness.
strength throughout a large section, it is important to select a high-alloy steel.
10-6 Toughness The term toughness as applied to tool steels may be thought of as the ability to resist breaking rather than the ability to absorb energy during deformation, as defined in Chap. 1. Most tools must be rigid articles, and usually even slight plastic deformation makes the tool unfit for use, As might be expected, this property is best in the medium- and low-carbon tool steels of groups S and H, which form the basis of the shock-resisting tool steels. Shallow-hardening steels which end up with a relatively soft
m
' .
:-->
tough core are also rated goo which are high in carbon tend 10-7 Wear Resistance
All the tool £ but several are outstanding in fined as the resistance to abra;
tolerances.
Wear resistance it
over the total surface of the pc the hard, undissolved carbide
10 8 Red-hardness
This property,
resistance of the steel to the sc extent in the resistance of the i
r
,
; .
RGY
TOOL STEELS
-
395
-
i
of Some Tool Steels*
RESISTANCE TO
APPROX TEMPERING
HARDNESS
DEPTH OPl
RANGE, 0F
ROCKWELL Ct
HARDENIM8
.FORMING ERTIES
RESISTANCE
MACHINABILITY
DECARBURIZATION
Fair to good
Best
Best
SAFETY IN
TOUGH-
RED-
WEAR
HARDENING
NESS
HARDNESS
Good
Poor
300-650
65-50
Shallow
Fair
300-650
65-50
Shallow
Fair
Good
Poor
Fair to good
Best
Best
400-1200
58- 40
Medium
Good
Very good
Fair
Fair
Fair
Fairto good
350-800
60- 50
Medium
Good
Best
Fair
Fair
Fair
Poor
300-500
62-57
Medium
Very good
Fair
Poor
Good
Good
Good
350-1000
62-57
Deep
350-800
62- 54
400-1000
61- 54
400-1000
61-54
400-1000
61- 54
Deep Deep Deep Deep
1000-1200
54-38
Deep
1000-1300
59- 40
1100-1250
54-36
Desp Deep
47-30
Deep
1200-1500
,
r
irygood
Best
Fair
Fair
Very good
Fair
Fair
Best
Best
Fair
Fair
Good
Fairto poor
Good to fair
B«st
Best
Poor
Good
Best
Poor
Fair
terygood
Good
Poor
Good
Best
Poor
Fair
Best
Best
Poor
Good
Best
Poor
Fair
Vrygooa
Best
Good
Good
Fair
Fair
Fair
Good
Good
Good
Fair
Fair
Fair
lood air
Good
Good
Good
Fair to good
Fair
Fair
!OOd
Good
Fair
Very good
Fair to good
Fair
Fair
J fair JSill, air: good
Good
Fair
Very good
Good
Fair
Fair
Fair
Poor
Very good
Good
Fair
Poor
Very good Very good Very good
Fair
Fair
Poor
Vory good Very good
Fair
Poor
Fair
Fair
Fair
Poor
Fair
Poor Good
'
,
i
1050-1250
58-43
Deep
1050-1200
60- 50
Deep
1000-1100
65- 60
1000-1100
66- 62
Deep Deep
1000-1100
65-60
1000-1100
65-60
1000-1100
65- 60
1000-1100
66- 61
Deep Doop Deep Deep
1000-1100
70-65
Deep
350-1000
63- 45
Medium
350-1000
62- 45
Modiuin
300-500
66-62
Shallow
300-500
R4-58S
900-1100
37-28
|pH:»air JSWt, air: good Mil: fair
Good
Good
PoOr
Very good
d
Fair
Poor
Best
d
Fair
Poor
Best
d
Fair
Poor
d
Fair
Poor
ood Oood
Fair
Poor
Vory good Very good Very good
Fair
Poor
Very good
Very good Very good
IWster: poor
Water: poor
Poor
Good
Good
TOftMli
Very good
Oil: Inii
Ooocl
Voi y cjood
Poor
Oood
Fair
Good
Poor
Poor
Poor
Very good
Fair
Good
Shallow
Good
Good
Poor
Fair
Fair
Good
Shallow
Good
Good
Poor
Fair
Good
Good
Ik 00 u
foor
Fair
:
Fair
Tool Steels," American imn and Steel Institute, 1970.
"
tough core are also rated good in toughness. The cold-work tool steels
which are high in carbon tend toward brittleness and low toughness.
ction, it is important to select a high-alloyIBP0'7 Wear Resistance All the tool steels have relatively good wear resistance, but several are outstanding in this property. Wear resistance may be devj as applied to tool steels may be thou ghtof i '
rather than the ability to absorb energy durChap. 1. Most tools must be rigid articles deformation makes the tool unfit for use,"!
.
"
'
which form the basis of the shock-resisting JI j steels which end up with a relatively softj
'
tolerances. Wear resistance might be required on a single cutting edge or over the total surface of the part. In general, a correlation exists betwaen the hard, undissolved carbide particles and wear resistance.
10-8 Red-hardness This property
jerty is best in the medium- and low-carbon
i:
fined as the resistance to abrasion or resistance to the loss of dimensional
,
also called hot-hardness, is related to the
resistance of the steel to the softening effect of heat. It is reflected to some extent in the resistance of the material to tempering, which is an important
396
INTRODUCTION TO PHYSICAL METALLURGY
selective factor for high-speed and hot-work tools, A tool steel with good , red-hardness is essential when ternperatures at which the tools must opK I
10 10 Resistance to Decarburizatior
of tool steels since it influeni
erate exceed 900oF. Alloying elements which form hard, stable carbides I =
standing in this property are the tool steels that contain relatively large I amounts of tungsten, chromium, and molybdenum.
10-9 Machinability
in a protective atmosphere),
annealed condition, the microstructure of the steel, and the quantity of hardp
are considered fair; and the rr
sistance to decarburization.
When compared with the conventional alloy steels, tool steels are con-
,
10-11 Brand Names
For many year
has a machinability rating of about 30 percent that of B1112 screw stock. It h
particular application would in steel producer and would rece
is therefore usual to compare the machinability of tool steels with W1 at an*'
has led to the use of brand n
arbitrary rating of 100. On this basis, the machinability for each of the different types of tool steel is rated in Table 10-3.
The machinability and general workability of tool steels decrease with in- p creasing carbon and alloy content. Low annealed hardnesses are usually! more difficult to attain as the carbon and alloy content increases. The pres-ft
ence of carbon in combination with strong carbide-forming elements suchp as vanadium, chromium, and molybdenum reduces machinability by the] formation of a large number of hard carbide particles which are out of solution after annealing. TABLE 10-3
Machinability Ratings of Tool Steels
Waler-hardening grades rated al 100
TOOL-STEEL GROUP
.
hardening. Tools that are intr hardening must not show any The straight-carbon tool ste shock-resisting tool steels are
excess carbides.
siderably more difficult to machine. The best machinable tool steel (Wtype!;
v
Decarburization l
1300oF, and unless some metf
carbon. Decarburization will r
The factors that affect machinability of tool steels are the hardness in theft
!
hardening.
"
1 his is the ability of the material to be cut freely and produce
a good finish after being machined. The machinability ratings given in : Table 10-2 merely show the relative difficulty which might be encountered: in machining the steels in question during manufacture of tools and dies.;
1
lected and the amount of me '
generally improve the resistance to softening at elevated temperature. Out-
each of his different types of s to maintain the quality and thi With the rapid growth of inc than one source of supply anc cal composition and some pi time, many tool-steel manufa chemical composition and pt tool steel. Although the AISI h tool steels (Table 10 -1), the us day. Table 10-4 lists approxi various manufacturers.
10 12 Water-hardening Tool Steel; MACHINABILITY RATING
carbon tool steels, although s(
100
small amounts of chromium a
S
85
and wear resistance. The cart
O
90
cent, and the steels may be rc
A
85
D
40-50
H (Or) H (W or Mo)
75
0
50-60
eration, such as hammers, concn
T
40-55
M
45
M
45-60
W
ing to carbon content. 60 to 0.75 percent carbon -for a
.
runs.
75 to 0.95 percent carbon-fot
0
.
i
L
90
F
75
P
75-100
From "Metals Handbook," 8th «d,, American Society lor Metals, Metals Park, Ohio. 1961,
i
\
equally important, such as punche 0
95 to 1.40 percent carbon-for
.
retention of cutting edge are imp
taps, reamers, and turning tools,
In general, the straight-car
LURGY
TOOL STEELS
397
"
-
.
I;
ed and hot-work tools. A tool steel with M lQ
'
0 Resistance to Decarburization This is an important factor in the selection of tool steels since it influences the type of heat-treating equipment se-
en temperatures at which the tools must;
.
elements which form hard, stable carbi
'
lected and the amount of material to be removed from the surface after
nee to softening at elevated temperature:
e the tool steels that contain relatively 1; ium, and molybdenum,
hardening. Decarburization usually occurs when steels are heated above 1300oF and unless some method is used to protect them (such as heating in a protective atmosphere) they are likely to lose some of their surface
Xy of the material to be cut freely and prod
carbon. Decarburization will result in a soft rather than a hard surface after
jachined. The machinability ratings given jeiative difficulty which might be encountei jestion during manufacture of tools and dii .
,
.
,
,
hardening. Tools that are intricately designed and cannot be ground after hardening must not show any decarburization The straight-carbon tool steels are least subject to decarburization
'Ji
.
nability of tool steels are the hardness in tl
sistance to decarburization.
if
Conventional alloy steels, tool steels are goi chine. The best machinable tool steel (W typ ) .
v
ton
about 30 percent that of B1112 screw stock.
,
'
:
! the machinability of tool steels with W1 at 4
asiis basis, the machinability for each of the dig ited in Table 10-3. : ; -
S;ral workability of tool steels decrease with \W; intent. Low annealed hardnesses are usually; carbon and alloy content increases. The pres-
m with strong carbide-forming elements sucW i molybdenum reduces machinability by thej of hard carbide particles which are out of:
r
The
shock-resisting tool steels are poor in this property; the hot-work tool steels are considered fair; and the majority of the other tool steels have good re-
§i?jstructure of the steel, and the quantity of hai
:
.
3i
Brand Names For many years a manufacturer requiring a tool steel for a particular application would indicate the nature of this application to a toolsteel producer and would receive a recommended tool steel. This practice has led to the use of brand names which the tool-steel producer gave to each of his different types of steel. Tool-steel producers made every effort to maintain the quality and thus the reputation of their brand names. With the rapid growth of industry many tool-steel consumers had more than one source of supply and had to issue specifications covering chemical composition and some physical properties of the steel. At the same time, many tool-steel manufacturers published information regarding the chemical composition and physical properties of their various brands of ,
,
tool steel. Although the AISI has standardized the chemical composition of
tool steels (Table 10-1), the use of trade names has persisted to the present day. Table 10-4 lists approximate comparable tool-steel brand names of nability Ratings of Tool Steels i rated at 100
XIP
various manufacturers.
1012 Water-hardening Tool Steels (Group W) MACHINABILITY RATING
These are essentially plain-
100
carbon tool steels, although some of the higher-carbon-content steels have small amounts of chromium and vanadium added to improve hardenability
85
and wear resistance. The carbon content varies between 0.60 and 1.40 per-
i
90
cent, and the steels may be roughly placed into three subdivisions accord-
85 40- 50
ing to carbon content.
-
75
0
50-60
eration, such as hammers, concrete breakers, rivet sets, and heading dies for short
40-55 45
45-60 90 75 75-100 '
k
"
,
.
60 to 0.75 percent carbon -for applications where toughness is the primary consid-
runs.
0
75 to 0.95 percent carbon-for applications where toughness and hardness are equally important, such as punches, chisels dies, and shear blades. .
,
0
95 to 1.40 percent carbon -for applications where increased wear resistance and retention of cutting edge are important. They are used for woodworking tools, drills, taps, reamers, and turning tools. .
8th ed.. American Society lor Metals, Metals
In general, the straight-carbon tool steels are less expensive than the
1 Mi.
my:
u CO 00
H D
O
TABLE 10 4
Comparable Tool-steel Brand Names o
GENERAL
AISI
CLASSIFICATION
NO.
ALLEGHENYLUDLUM
H
FIRTH-
BRAEBURN
BETHLEHEM
CARPENTER
COLUMBIA
CRUCIBLE
o
STERLING
2 H
Water-hardening
o
I
tool steels:
Standard quainy
Pompton
VV1
Extra quality
W1
i Pompton
X
CL
'
XX
Extra
Extra
Extra quality
Shock-resisting
W2
Python
SI
I Seminole
tool steels
Superior 67 Chisel
S5
i AL609
Omega
01
i Saratoga
BTR
Standard
No. 11 Comet
Standard
Labelle Extra
F-S Extra
No. 11 Extra Vanadium
Vanadium Extra
Alva Extra
Extra V
Buster
Atha Pneu
J-S Punch
> c 33
Vibro
No. 481
j Stentor
Airque
j No. 484
Ontario
Lehigh H
Superior 3
No. 610
D3
Huron
Lehigh S
Superior 1
Hampden
H12
Potomac
Cromo-W Cromo-W V
Pressurdie No. 2
No. 345
I Sagamore
air-hardening High-carbon
D2
Deward
.
.
.
Chino
EXL-Die
Ketos
Invaro No 1
H13
Potomac M
Q -
.
Cromo-High \/l Pressurdie
No. 883
Invaro No. 2
Airkool
Airvan
Airdi 150
Chromovan
Smoothcut Atmodie
:
m
Labelle Silicon No. 2
EZ-Die
,
-
i
CEC Smoother
Paragon
high-chromium Hot-work tool steels chromium base
HYCC
Triple Die
Alco Die
Chro-Mow
HWD-1
Vanadium Fire Die
Nudie V
HWD-3
! No. 3
Plastic-mold steels:
Straight iron
Pi
Duramoid C
Mirromold
5% chromium
P4
Duramold A
Super Samson
Crusca cold.
hubbing
air-hardening
Airmold !
TABLE 10 4
(Continued) UNIVERSAL
VANADIUM-
VULCAN
ALLOYS
CRUCIBLE
Red
Fort Pitt
GENERAL
AISI
CLASSIFICATION
NO.
JESSOP
LATROBE
UDDEHOLM \ CYCLOPS
Standard quality
W1
Lion
Standard
UHB
Standard
W1
Lion Extra
UHB Extra
Extra
Extra L
Extra quality
W2
Lion extra Vanadium
UHB-VA
Extra Draco
Elvandi
Extra quality
SI
Top
UHB-711
Alco M, Alco S
Par Exc
'
Water-hardening tool steels:
Shock-resisting tr-inl
ctoolc
Star Tool
Carbon
Notch
Extra Carbon
XL Chisel
>
Kiski S O D
!
A2
<
O
Extra
A-H5
02
Medium-alloy
Sterling
No. 11 Extra
Cold-work tool steels:
Oil-hardening types
Black Diamond
Extra
QA .
.
<
i:
14' rS5
AL609
Omega
No. 481
CEO Smoother
Labelle Silicon No. 2
Chino
EXL-Die
Ketos
Invaro No. 1
Cold-work tool steels:
Oil-hardening types Medium-alloy air-hardening High-carbon high-chromium Hot-work tool steels chromium base
01
Saratoga
02
Deward
A2
Sagamore
A-H5
Airque
No. 484
D2
Ontario
Lehigh H
Superiors
No. 610
D3
Huron
Lehigh S
Superior 1
Cromo-W Cromo-W
Pressurdle
BTR
Kiski
SOD .
.
Stentor
.
Paragon EZ-Die
Potomac
H12 Hi 3
Potomac M
Invaro No. 2
Airkool
Airvan
Airdi 150
Chromovan
HYCC
Triple Die
Smpothcut Atmodie
,
§
:
Hampden I No. 345 .
No. 2
Cromo-High V
Pressurdie
i No. 883
Alco Die
Chro-Mow
HWD-1
vanadium
Nudie V
HWD-3
No. 3
Plastic-mold steels:
Straight iron
PI
Duramold C
5% chromium
P4
Duramold A
;
Fire Die
Mirromold
Crusca cold
Super Samson
air-hardening
hubbing Airmold
' -
'
TABLE 10 4
7
iff-;;?
(Continued)
GENERAL
AISI
UNIVERSAL
VANADIUM-
VULCAN
CLASSIFICATION
NO.
JESSOP
LATROBE
UDDEHOLM
CYCLOPS
ALLOYS
CRUCIBLE
Standard quality
W1
Lion
Standard Carbon
UHB
Standard
Red Star Tool
Fort Pitt
Extra quality
W1
Lion Extra
Extra Carbon
iUHB Extra
Extra
Extra L
Extra
Extra quality
1 W2
iUHB-VA
Extra Draco
Elvandi
|UHB-711
Alco M,
Par Exc
QA
Mosil
487D
Non-shrinkable Colonial No. 6
Oil hardening
Water-hardening tool steels:
Shock-resisting
SI
Lion extra Vanadium
Top Notch
tool steels S5
XL Chisel
No. 259
.
i;
.
Alco S
IUHB
Cyclops
|Resisto
67
;UHB-46
Wando
Cold-work tool steels:
Gil-hardening types
Vedium-alloy air-hardening High-carbon high-chromium
Hoi-work tool steels,
01
Truform
Badger
02
Special oil hardening
Mangano
A2
Windsor
Select B
D2
CNS-1
Olympic
D3
CNS-2
H12
DICA B
Non-shrinkable
Sparta
Air Hard
Vuidie
jTRI-Mo
Ultradie No. 2 and 3
Ohio Die
Alidie
GSN
|TRI-Van
Ultradie No. 1
LPD
iUHB Special
Thermold
UHB-151
B
chromium base
H13
DICA B Vanadium
v D C,
UHBOrvar
Thermold Av
Hi Pro
Hot Form No. 1 and 2
TCM
Hot Form 5
Vulcast O o
Plastic-mold steels:
Straight iron
PI
UHB Forma
5% chromium
P4
UHB Premo
Plastic Die
CO i
air-hardening
m m C/5 W
400
INTRODUCTION TO PHYSICAL METALLURGY
alloy tool steels and with proper heat treatment they yield a hard martens,
m
itic surface with a tough core
These steels must be water-quenched for
.
high hardness and are therefore subject to considerable distortion. They have the best machinability ratings of all the tool steels and are the best in respect to decarburization but their resistance to heat is poor. Because of ,
this low red-hardness
,
carbon steels cannot be used.as cutting tools under
conditions where an appreciable amount of heat is generated at the cutting edge. Their use as cutting tools is limited to conditions involving low speeds and light cuts on relatively soft materials such as wood, brass, alu,
minum, and unhardened low-carbon steels.
Ci*...
. . .
The typical microstructure of a W1 water-quenching tool steel
,
austeni-
tized at 1450oF, brine-quenched and tempered at 3250F, is shown in Fig. 10-1. The low tempering temperature has resulted in a matrix of dark,
etching tempered martensite with some undissolved carbide particles (white dots), and a hardness of Rockwell C 64.
10-13 Shock-resisting l oci Steels (Group S) These steels were developed for those applications where toughness and the ability to withstand repeated shock are paramount. They are generally low in carbon content the carbon varying between 0.45 and 0.65 percent. The principal alloying elements in these steels are silicon, chromium, tungsten, and sometimes molybdenum. Silicon strengthens the ferrite, while chromium increases hardenability and contributes slightly to wear resistance. Molybdenum aids in increasing hardenability, while tungsten imparts some red-hardness to
Fig, 10-2
SI shock-resisting tool steel austeniti;
IZSO F, oil-quenched, and tempered at 800 F. Si lempered martensite with some spheroidal-carbi (Atiiti! dots). Etched in 3 percent nital, 1,000x, Metals Handbook, vol. 7, "Atlas of Microstructur
ican Society lor Metals, 1972 )
,
i
.
t :
these steels.
Most of these st
to be water-quenched to devel The high silicon content tend precautions should be taken ii classed as fair in regard to n
. . .
1
/vai
ability, and hardness is usually group are used in the manul
pneumatic tools, and shear bU '
,
.
V;.
' .
JVC-
Figure 10-2 shows spheroidi matrix typical of an 81 shock oil-quenched, and tempered £ has caused the tempered ms Fig. 10-1. 10 14 Cold-work Tool Steels
This i:
of tool steels, since the majori or more of the steels in this cl
The oil-hardening low-allo smaller amounts of chromium ..
>>
i
Fig. 10-1
Wl water-hardening tool steel austenitized at
1450°F, quenched in brine, and tempered at 3250F. Structure is tempered martensite with some undissolved spheroidal carbide (white dots). Etched in 3 percent nital, 1,000x. (From Metals Handbook, vol. 7, "Atlas of Microstructures," American Society for Metals, 1972.)
"
-
'
V,V
I
" -
forming properties and are le during heat treatment than ar Figure 10-3a shows the nc particles in a tempered martei austenitized at 1500oF oil-qut ,
«S8
0
1GY
.
TOOL STEELS
401
1
?r heat treatment they yield a hard marti
These steels must be water-quonchi Mi$[e subject to considerable distortion.
:
ings of ail the tool steels and are the b
fheir resistance to heat is poor. Becaiis leels cannot be used.as cutting tools ufil
p amount of heat is generated at the cuttli jols is limited to conditions involving f
m
?ly soft materials, such as wood, brass, ali
.
jrbon steels.
a W1 water-quenching tool steel, auste(1t| Zi-M and tempered at 3250F, is shown in Fig berature has resulted in a matrix of dark* ,
10-2
with some undissolved carbide particles| Rockwell C 64. : >
.
SI shock-resisting tool steel austenitized at oil-quenched, and tempered at 800oF. Structure is pored ntiartensite with some spheroidal-carbide particles Stednts). Etched in 3 percent nital 1,000x. (From Betals Handbook, vol. 7 "Atlas of Microstructures," Amer,
roup S) These steels were developed for iness and the ability to withstand repeated
,
i
,
&Wi Society lor Metals
:
\v b generally low in carbon content, the car-
'
,
1972.)
:
.:
65 percent. The principal alloying elements #-|omium tungsten, and sometimes molyb-
;
these steels.
.
Most of these steels are oil-hardening, although some have
ferrite, while chromium increases harden-
to be water-quenched to develop full hardness.
to wear resistance.
The high silicon content tends to accelerate decarburization, and suitable precautions should be taken in heat treatment to minimize this. They are classed as fair in regard to red-hardness, wear resistance, and machinability, and hardness is usually kept below Rockwell C 60. The steels in this group are used In the manufacture of forming tools, punches, chisels,
Molybdenum aids in
tungsten imparts some red-hardness to '
A
t V
I I.
I f
pneumatic tools, and shear blades. Figure 10-2 shows spheroidal carbide particles in a tempered martensite matrix typical of an S1 shock-resisting tool steel austenitized at 1750oF oil-quenched, and tempered at 800"F. The higher tempering temperature ,
.5 ;
has caused the tempered martensite to etch lighter in comparison with
J t
Fi9- 10-1-
;
'
"'
;
1
I
11014 Cold-work Tool Steels of tool steels
,
This is considered to be the most important group since the majority of tool applications can be served by one
or more of the steels in this classification.
The oil-hardening low-alloy type (group O) contains manganese and smaller amounts of chromium and tungsten They have very good nondeforming properties and are less likely to bend sag, twist, distort, or crack .
ized at
J ;
.
,
Struc-
0
.
d spheroi-
' .
during heat treatment than are the water-hardening steels Figure 10-Sa shows the normal microstructure of spheroidal carbide particles in a tempered martensite matrix of an 01 oil-hardening tool steel austenitized at 1500oF oil-quenched, and tempered at 4250F. Raising the
.jl 1,000x. ,
uctures.
"
,
.
402
INTRODUCTION TO PHYSICAL METALLURGY
[ft. I
Fig. 10-3
01 oil-har'tJening tool steel, (a) Austenitized at
1500oF, oil-quenched, and tempered at 425"F. The normal structure consists of spheroidal-carbide particles in a tempered-martensite matrix, (b) Austenitized at ISOCTF, oilquenched, and tempered at 425 F. Structure is coarsemartensite needles (black) in retained-austenite matrix (white), both resulting from overheating. Etched in 3 percent nital, 1,000x. (From Metals Handbook, vol. 7 Atlas of Microstructures, American Society for Metals, 1972.)
-
-
Fiy
10-4
06 oil-hardening tool steel austenitizei
oil-quenched, and tempered at 425°F. The structi consists of graphite (black) and carbide particles a matrix of tempered martensite (gray). Etched in percent nital, 1,000x. (From Metals Handbook, vc Alias of Microstructures," American Society for I 1972.)
"
,
"
I
ible black lines are grain bound; ing properties, good wear resi;
austenitizing temperature to 1800oF results in coarse martensite needles I
and resistance to decarburizatio
(black) in a retained austenite matrix (white), both resulting from over- [ heating (Fig. lO-Sb). These steels are relatively inexpensive and their high I carbon content produces adequate wear resistance for short-run applica-j
steels are used for blanking, for
tions at or near room temperature. The main function of the high silicoim content in 06 steal is to induce graphitization of part of the carbide, there- ! by improving machinability in the annealed condition. Figure 10-4 shows the structure of an 06 oil-hardening tool steel austen-
itized at 1500oF, oil-quenched, and tempered at 4250F. The large black l spots are graphite, white spots are carbide particles, and the matrix is tern- j pered martensite (gray). These steels have good machinability and good re-1 sistance to decarburization; toughness is only fair, and their red-hardness j
is as poor as the straight-carbon tool steels. They are used for taps, solid I treading dies, form tools, and expansion reamers. I The medium-alloy type (group A) with 1 percent carbon, contains up to I 3 percent manganese, up to 5 percent chromium, and 1 percent molybdenum. The increased alloy content, particularly manganese and molybdenum, confers marked air-hardening properties and increased hardenability. Figure 10-5 shows the structure of an A2 tool steel austenitized a!
and tempered at 350oF Rockwell C 63. The stru consists of large alloy-carbide particles (white) a spneroidal-carbide particles (white dots) in a ma
1800oF
tempered martensite. Grain boundaries (black li
,
,
air-cooled, and tempered at 350oF, Rockwell C 63. The large while
spots are alloy carbide particles the small white spots are spheroidal car-* bide particles and the gray matrix is tempered martensite. The barely vij ,
,
..I
\
\
i 5
Fig 10-5
A2 tool steel austenitized at 180&T, ai ,
barely visible. Etched in 3 percent nital, 1,000x. Metais Handbook, vol. 7, "Atlas of Microstruclur
Ameiican Society for Metals, 1972.)
JRGY
TOOL STEELS
lenitized at The normal
'
10-4 06 oil-hardening tool steel austenitized at 1500°F
'
403
,
all-quenched, and tempered at 425°F The structure tonsists of graphite (black) and carbide particles (white) in .
les in a
)t ISOCF, oil.
matrix of tempered martensite (gray)
Etched in 3
.
Iflarcent nital, 1,000x. (From Metals Handbook, vol. 7,
5 coarse-
.e matrix
l
ed in 3 per-
'Atlas of Microstructures 1872.1
,
"
American Society for Metals
,
7 "Atlas ,
als, 1972.)
ible black lines are grain boundaries. This group has excellent nondeformi
ing properties good wear resistance, and fair toughness, red-hardness, ,
1800oF results in coarse martensite neecdies
and resistance to decarburization
but only fair to poor machinability. These steels are used for blanking forming, trimming, and thread-rolling dies.
te matrix (white), both resulting from ove|
,
,
teels are relatively inexpensive and. their hig| squate wear resistance for short-run applicjS -
ature. The main function of the high silicd
Le graphitization of part of the carbide, therl i the annealed condition.
1
iture of an 06 oil-hardening tool steel austell
!id and tempered at 4250F. The large blac ,
:
s are carbide particles, and the matrix isteny|
je steels have good machinability and good nij oughness is only fair, and their red-hardne
bon tool steels. They are used for taps, i expansion reamers.
Dup A), with 1 percent carbon, contains uptS Si 5 percent chromium and 1 percent molybl| content, particularly manganese and mc ,
/
hardening properties and increased hardervll e structure of an A2 tool steel austenitized m
ered at 350UF, Rockwell C 63. The large whltN
les, the small white spots are spheroidal car?!
;
Fig 10'5 A2 tool steel austenitized at 1800°F, air-cooled, i-and
tempered at 350oF Rockwell C 63. The structure < consists of large alloy-carbide particles (white) and small ,
spheroidal-carbide particles (white dots) in a matrix ol tempered martensite Grain boundaries (black lines) are barely visible. Etched in 3 percent nital 1,000x. (From .
,
Metals Handbook vol. 7, "Atlas of Microstructures Amorican Society for Metals 1972.) ,
natrix is tempered martensite. The barely \
.
"
,
!;
m0 404
INTRODUCTION TO PHYSICAL METALLURGY
The high-carbon high-chromium types (group D) contain up to 2.25 percent carbon and 12 percent chromium. They may also contain molybdenum, vanadium, and cobalt. The combination of high carbon and high
chromium gives excellent wear resistance and nondeforming properties. J They have good abrasion resistance, and minimum dimensional change in 1 hardening makes these steels popular for blanking and piercing dies; draw- 'f ing dies for wire, bars, and tubes; thread-rolling dies; and master gauges, ji 10-15 Hot-work Tool Steels (Group H) In many applications the tool is subjected to excessive heat because the material is being worked, as in hot forging and extruding, die casting and plastic molding. Tool steels developed for | these applications are known as hot-work tool steels and have good red,
hardness.
The alloying elements noted for red-hardness are chromium, molyb- Jp denum, and tungsten. However, the sum of these elements must be at least fi 5 percent before the property of red-hardness becomes appreciable.
Rockwell C 46 to 48. Structure consists of a lev
carbide particles in a coarse tempered-martensi Etched in 2 percent nital, 500x. (From Metals H
The hot-work tool steels may be subdivided into three groups: (1) Hot-work chromium-base
vol. 7,
(H11 to H19), containing a minimum of 3.25 percent j
'-
Atlas of Microstructures," American Soc
Metals, 1972.)
chromium and smaller amounts of vanadium, tungsten, and molybdenum. They have .
good red-hardness because of their medium chromium content, supplemented by the addition of the carbide-forming elements of vanadium, tungsten, and molybdenum. The low carbon and relatively low total alloy content promote toughness at
the normal working hardnesses of Rockwell C 40 to 55. Higher tungsten and molyb-
| '
denum contents increase red-hardness but slightly reduce toughness. These steels are extremely deep-hardening and may be air-hardened to full working hardness insections up to 12 in. The air-hardening qualities and balanced alloy content are responsible for low distortion in hardening. These steels are especially adapted to hot die work of all kinds, particularly extrusion dies, die-casting dies, forging dies,
Fig, 10-6 H11 tool steel austenitized at 2050 , quenched, and double-tempered (2 h plus 2 h) £
f '
%
steels have many of the characteri a low-carbon version of T1 high-s for high-temperature applications nickel alloys, and steel. (3) Hot-work molybdenum-base 4 percent chromium, and smaller are similar to the tungsten hot-w(
mandrels, and hot shears.
and uses.
Figure 10-6 shows the structure of an H11 tool steel austenitized at 2050°F, oilquenched, and double-tempered (2 h plus 2 h) at 1100oF Rockwell C 46 to 48. The :s
speed tool steels, although they h Their principal advantage over tt
matrix is coarse tempered martensite with a few spheroidal carbide particles (white
These steels are more resistant to
spots). The high austenitizing temperature has dissolved most of the carbide. An interesting application is the use of the Nil steel for highly stressed structural parts, particularly for supersonic aircraft. The chief advantage of this steel over conventional high-strength steels is its ability to resist softening during continued expo-
but in common with all high-mol treatment, particularly with regarc The hot-work tool steels as a gt
,
content, good to excellent red-f ability. They are only fair to poc hardened they show little or no di
sure to temperatures up to 1000oF and at the same time to provide moderate toughness and ductility at tensile strengths of 250,000 to 300,000 psi.
Another important advantage of H11 for these applications is its exceptional ease of forming while in the austenitic condition by the process called ausforming described m Sec. 9.18. It also has good weldability, a relatively low coefficient of thermal expginsion, and above average resistance to corrosion and oxidation. (2) Hot-work tungsten-base (H21 to H26), containing at least 9 percent tungsten and 2 to l 2 percent chromium. The higher alloy content increases resistance to hightemperature softening compared with the H11 to HI9 steels, but it also makes them more susceptible to brittleness at the normal working hardnesses of Rockwell C 45 to 55. They can be air-hardened for low distortion, or they can be quenched in oil or '
hot salt to minimize scaling.
Although they have much greater toughness, these
S !
In composition, they i
i > '
10 16 High-speed Tool Steels
The:
of the tool steels and usually denum along with chromium, bon content varies between 0.
tain as much as 1.5 percent c;
The major application of hi are also used for making ext
punches and dies.
-
...
.
5Y
TOOL STEEl S
405
i
jm types (group D) contain up to 2.25 per jromium. They may also contain molyb-;
-
The combinalion of high carbon and high-
resistance and nondeforming properties.
i:
nee, and minimum dimensional change iftf
5Jp pularfor blanking and piercing dies; draw-i 5; thread-rolling dies; and master gauges. .
'
In many applications the tool is subjected laterial is being worked, as in hot forging Dlastic molding. Tool steels developed for
I hot-work tool stoels and have good red- ;.i
~
I for red-hardriess are chromium, molyb- |
Fig. 10 6 Hi 1 lool sloel austenitized at 2050"F
ithe sum of these elements must be at least
¥;
,
oll-
quenched, and double-tempered (2 h plus 2 h) at 1100oF
,
Rockwell C 46 to 48. Structure consists of a few spheroidal-
ed-hardness becomes appreciable.
carbide particles in a coarse tempered-martensite matrix % Etched in 2 percent nital, 500x. (From Metals Handbook
.
be subdivided into three groups: .
'
vol. 7
-to H19) containing a minimum ol 3.25 percent
.
5
,
madium, tungsten, and molybdenum. They have / medium chromium content, supplemented by
-
3 l:
,
"
,
Alias ol Microstructures," American Society for
Metals. 197?.)
i eloinonls ol vanadfum, lungslen, and molybfely low total alloy content promote toughness at
steels have many of the characteristics of the high-speed tool steels
pckwell C 40 to 55. Higher tungsten and molyb-
8
less but slightly reduce toughness. These steels inay be air-hardened to full working hardness in Lning qualities and balanced alloy content are
.
,
,
and uses. In composition, they resemble the various types of molybdenum highspeed tool steels, although they have lower carbon content and greater toughness Their principal advantage over the tungsten hot-work steels is lower initial cost These steels are more resistant to heat cracking (checking) than the tungsten grades but in common with all high-molybdenum steels they require greater care in heat treatment, particularly with regard to decarburization.
of an HII tool steel austenitized at 2050" F, oil-
.
h plus 2 h) at 1100"F, Rockwell C 46 to 48. The ;V-:-/je with a few spheroidal carbide particles (white
.
:;:
,
/ /erature has dissolved most of the carbide. '
jse of the H11 steel for highly stressed structural craft. The chief advantage of this steel over con-
The hot-work tool steels as a group have good toughness because of low carbon
ability to resist softening during continued expoind at the same time to provide moderate tough-
content, good to excellent red-hardness and fair wear resistance and machin,
ability. They are only fair to poor in resistance to decarburization
is of 250,000 to 300,000 psi. 111 for these applications is its exceptional ease condition by the process called ausforming de:
hardened they show little or no distortion from heat treating
of the tool steels and usually contain large amounts of tungsten or molyb,
to H26), containing at least 9 percent tungsten higher alloy content increases resistance to high- .
'
ith the H11 to HI 9 steels, but it also makes them
jhe normal working hardnesses of Rockwell C 45 low distortion, or they can be quenched in oil or ough they have much greater toughness, these j -
but when air-
denum along with chromium vanadium, and sometimes cobalt. The car-
resistance to corrosion and oxidation.
.
,
.
10-16 High-speed Tool Steels These steels are among the most highly alloyed
pd weldability, a relatively low coefficient of ther-
.
(3) Hot-work molybdenum-base (H41 to H43), containing 8 percent molybdenum, 4 percenl chromium and smaller amounts of tungsten and vanadium. These steels are similar to the tungsten hot-work steels having almost identical characteristics
ly extrusion dies, die-casting dies, forging dies,
"
In fact, H26 is
.
,
!
[dening. These steels are especially adapted to '
.
a low-carbon version of T1 high-speed steel (see Table 10-1) They have been used for high-iemperaluro applicalions such as mandiols and extrusion dies for brass nickel alloys and steel.
*
bon content varies between 0.70 and 1 percent although some types con:\ tain as much as 1.5 percent carbon The major application of high-speed steels is for cutting tools but they ,
'
.
,
are also used for making extrusion dies
,
burnishing! tools, and blanking
punches and dies.
i :
406
INTRODUCTION TO PHYSICAL METALLURGY
Compositions of the high-speed steels are designed to provide excellent red-hardness and reasonably good shock resistance. They have good nondeforming properties and may be quenched in oil, air, or molten salts. They are rated as deep-hardening, have,good wear resistance, fair machinabllity,
f I
r,7
i
and fair to poor resistance to decarburization.
The high-speed steels are subdivided into two groups; molybdenumbase (group M) and tungsten-base (group T). The most widely used tungsten-base type is known as 18-4-1 (T1), the numerals denoting the content, respectively, of tungsten, chromium, and vanadium in percentages. From the standpoint of fabrication and tool performance, there is little difference between the molybdenum and tungsten grades. The important properties of red-hardness, wear resistance, and toughness are about the same. Since there are adequate domestic supplies of molybdenum and since most of the tungsten must be imported, the molybdenum steels are lower in price, and over 80 percent of all the high-speed steel produced is of the molybdenum type. When better than average red-hardness is required, steels containing co-
5
7
v
i
4
i,
F;g. '0 7 T1 high-speed steel austenitized at 233; quenched to 1 1250F air-cooled, and double-tempt ,
1000"F. Undissolved carbide particles in a matrix
pert'd inartensite. Etched in 4 percent nital, 1,000 % (From Metals Handbook, vol. 7, "Atlas of Microstn Amnrican Society for Metals, 1972.)
bait are recommended. Higher vanadium content is desirable when the material being cut is highly abrasive. In T15 steel, a combination of cobalt
plus high vanadium provides superiority in both red-hardness and abra-sion resistance. The use of high-cobalt steels requires careful protection against clecarburization during heat treatment, and since these steels are more brittle, they must be protected against excessive shock or vibration
wear resistance.
in service.
hardness of Rockwell C 30 to 35
Figure 10 7 shows the structure of a T1 high-speed steel austenitized at 23350F, salt-quenched to 1 1250F, air-cooled, and double-tempered at
large intricate dies and molds. J heat treatment is required, and d
With the exc<
hardness and therefore are use
casting dies and for molds for in Types P20 and P21 are normal!
1000°F. White spots are undissolved carbides in a matrix of tempered mar- i 10-18 Special-purpose Tool Steels tensite.
The presence of many wear-resistant carbides in a hard heat-resistant matrix makes these steels suitable for cutting tool applications, while their
toughness allows them to outperform the sintered carbides in delicate tools and interrupted-cut applications. The tungsten and molybdenum high-
1
more standard steels.
;
speed steels are used in a wide variety of cutting tools, such as tool bits, [; drills, reamers, broaches, taps, milling cutters, hobs, saws, and woodworking tools. 10-17 Mold Steels (Group P) These steels contain chromium and nickel as'the .
V »-
-
"
V.
.
.
principal alloying elements, with molybdenum and aluminum as additives. Most of these steels are alloy carburizing steels produced to tool steel quality. They are generally characterized by very low hardness in the annealed condition and resistance to work hardening; both factors are favorable for hubbing operations. In hubbing, a master hub is forced into a soft
,
::::
:
plex iron-chromium carbides, ar hardenability.
Nickel increases
fine the grain. These steels are dimensional change. Typical u where high wear resistance with rollers, clutch plates, cams, coll are used for arbors, dies, drills,
The carbon-tungsten type (c
water-quenching steels with hig
ft
:
chromium contents promote we
carburized and hardened to a surface hardness of Rockwell C 58 to 64 for
"
m
}
The low-alloy types (group L) element, with additions of vana
blank. After the impression has been formed or cut, the steels are generally
i
i
f
categories and are therefore d They have been developed to he tain application and are more
mmmm
Y
v
TOOL STEELS
407
d steels are designed to provide excellent!
iSMd shock resistance. They have good non-|
i quenched in oil, air, or molten salts. They,| |e good wear resistance, fair machinabiHty,;| arburization.
bdivided into two groups: molybdenum-
se (group T). The most widely used tung4-1
' '
So
,
chromium, and vanadium in percentages; ;.pn and tool performance, there is little dif-: *Slim and tungsten grades. The important
: ;
(T1) the numerals denoting the con- .
r
1
.
5
.
..
|r resistance, nd toughness are about the § te domestic supplies of molybdenum and it be imported, the molybdenum steels are ent of all the high-speed steel produced is
Fig. 107 T1 high-speed steel austenitized at 2335°F saltquenched to 11250F, air-cooled, and double-tempered at ,
|1(KJ0"F. Undissolved carbide particles in a matrix of tem-
;
pered martensite. Etched in 4 percent nital, 1,000x. '
-
r
-
.
JFfem Metals Handbook, vol. 7, "Atlas of Microstructures," hardnoss is required, steels containing co- >Hi| Amftrlcdii Sorloly lor Momls, 197?,)
-
'
r vanadium content is desirable when thd |
;
asive. In T15 steel, a combination of cobalt
juperiority in both red-hardness and abra- 4H|
wear resistance. With the exception of P4 these steels have poor redhardness and therefore are used almost entirely for low-temperature die-
heat treatment, and since these_steels are 'f»J ected against excessive shock or vibration
casting dies and for molds for injection or compression molding of plastics. Types P20 and P21 are normally supplied in a heat-treated condition to a hardness of Rockwell C 30 to 35 so that they can be readily machined into large intricate dies and molds. Since they are prehardened no subsequent heat treatment is required, and distortion and size changes are avoided,
|jh-cobalt steels requires careful protection -mm.
,
,
jre of a T1 high-speed steel austenitized at.
,
I50F, air-cooled, and double-tempered at
olved carbides in a matrix of tempered mar- iB|10-18 Special-purpose Tool Steels Many tool steels do not fall into the usual :
ible for cutting tool applications, while their-
categories and are therefore designated as special-purpose tool steels. They have been developed to handle the peculiar requirements of one certain application and are more expensive for many applications than the
srform the sintered carbides in delicate tools |j
more standard steels.
-
resistant carbides in a hard heat-resistant
5
,
The tungsten and molybdenum high- || a
le variety of cutting tools, such as tool bits, milling cutters, hobs, saws, and woodwork'
vV-v vVv
The low-alloy types (group L) contain chromium as the principal alloying molybdenum, and nickel. The high
element, with additions of vanadium
,
chromium contents promote wear resistance by the formation of hard complex iron-chromium carbides, and together with molybdenum they increase hardenability. Nickel increases toughness while vanadium serves to re-
steels contain chromium and nickel as the
,
'
"
.
5sSth molybdenum and aluminum as additives, 'ji
y carburizing steels produced to tool steel I "
'
'
:
laracterized by very low hardness in the an-
i;:v;be to work hardening; both factors are favor-1|
fine the grain. These steels are oil-hardening and thus only fair in resisting dimensional change. Typical uses are various machine-tool applications where high wear resistance with good toughness is required as in bearings, rollers clutch plates, cams, collets, and wrenches. The high-carbon types ,
,
In hubbing, a master hub is forced into a soft |
are used for arbors
is been formed or cut, the steels are generally j
The carbon-tungsten type (group F) are generally shallow-hardening water-quenching steels with high carbon and tungsten contents to promote
'
surface hardness of Rockwell C 58 to 64 for |
4
,
dies, drills, taps, knurls, and gages. ,
my /r:
'
408
INTRODUCTION TO PHYSICAL METALLURGY
m
high wear resistance. Under some conditions of operation these steels have four to fen times the wear resistance of the plain-carbon group W tool steels, they are relatively brittle so that in general they are used for high-I wear, low-temperature low-shock applications. Typical uses are paper-5
temperatures.
cutting knives
used for hardness and 900 to i;
draws at comparatively low tern Carbon and low-alloy steels ai
,
tween 300 and 500oF while for
,
,
,
wire-drawing dies, plug gages, and forming and finishing;
tools.
*
10-19 Heat Treatment of Tool Steels Proper heat treatment of tool steels is oneof the most important factors in determining how they will perform in ser-
vice. Although the emphasis in heat treatment is usually on the cooling
1
rate, it should be realized that as much damage may be done to the steel on heating as on cooling. % f t
Tool steels should not be heated so rapidly as to introduce large tempera-' | ture gradients in the piece. This may be avoided by slow heating or by pre-m heating the steel at a lower temperature before placing it in the high-heal:
furnace. Some heat treaters prefer placing the tool-steel parts into a co\i
I
furnace and then bringing both the work and the furnace up to temperature :
..
.
C]
33 39
together. It is also important that the tool steel be allowed to remain at the proper temperature for a sufficient time to make certain that the entire section has been heated uniformly. To avoid overheating, tool steels should not be heated to too high a tem-
»4 38
.39 44
S7*
54
8
perature or kept at heat too long. Quenching from excessive temperatijr| may result in cracking.
Overheating causes excessive grain growth and '
: 2
consequent loss in toughness.
It is essential that some means be used to protect the surface of the tool steel from excessive scaling or decarburization during heating. This has been discussed in Sec. 8-18. Any decarburized areas must be removed from tool-steel surfaces to provide satisfactory hardnesses.
i
The manner and media of quenching vary according to the steel being
quenched and the speed required in quenching. The usual quenching; >: : :-:
;
i
.
media are water
brine, oil, and air. Carbon and low-alloy tool steels are; quenched in brine or water; high-alloy tool steels are quenched in oil, air, or molten salts. While still air, fan cooling, and compressed-air blasts are; ,
used for cooling, still air is the preferred method, for It is more likely to prfc
vide uniform cooling. Interrupted quenching is also used for tool steeli | By this method, the steel is quenched in a liquid bath of salt or lead between I'
900 and 1200oF, then cooled in air to about 150oF. * .
.
r
"
f
It is recommended procedure to temper tool steels immediately after f-
'
.
quenching and before they have cooled to room temperature to minimi*; | Fl8 10.8 Dje made 0, carbon l00| stee|. Crackin the danger of cracking due to strains introduced by quenching. The temf pering , or drawing, operation relieves the stresses developed during hard-
ening and provides more toughness. Preferred practice is to utilize long I
1
I I 1
i
in quench because of excessive stresses set up b< thin rim and the body. (At top) Longitudinal secti die. Numbers are Rockwell hardness values. (Be Steel Company.)
/
;:
URGY
:
TOOL STEELS
409
i
some conditions of operation these steels h i;;S;Tesistance of the plain-carbon group W tL
draws at comparatively low temperatures rather than short draws at high temperatures
i gsittle, so that in general they are used for higf
.
Carbon and low-alloy steels are generally tempered at temperatures be
shock applications. Typical uses are pa| dies, plug gages, and forming and finishl
tween 300 and 500oF
while for high-alloy steels the range 300 to 500oF is used for hardness and 900 to 1200"F for toughness The high-speed tool ,
.
|s Proper heat treatment of tool steels is oft \ in determining how they will perform in |s in heat treatment is usually on the cooiii iat as much damage may be done to the stei
"
(!
it
'
:
; jated so rapidly as to introduce large tempei
n
his may be avoided by slow heating or by pre*|| femperature before placing it in the high-hea(|
it*
prefer placing the tool-steel parts into a cold th the work and the furnace up to temperatuNl
ft *4
that the tool steel be allowed to remain attb
f
.
f#jicient time to make certain that the entire SM j
9 44-
nly.
Sv.-'steels should not be heated to too high atem-| ong. Quenching from excessive temperatiij i;
B
\K:
irheating causes excessive grain growth and '
i
s
.
jans be used to protect the surface of the tool or decarburization during heating. This Any decarburized areas must be removed; .
bvide satisfactory hardnesses. quenching vary according to the steel being quired in quenching. The usual quenching,:and
:
air. Carbon and low-alloy tool steels are ;
high-alloy tool steels are quenched in oil, alr.i fan cooling, and compressed-air blasts are-
r,
e preferred method, for it is more likoly to proupted quenching is also used for tool steels. enched in a liquid bath of salt or lead between n air to about 1 SOT.
We to temper tool steels immediately after iave cooled to room temperature to minimize:
Fig. 10'8
D strains introduced by quenching. The tem-i
In quench because of excessive stresses set up between the
,
1 relieves the stresses developed during hard- jl. ,
p,
,
|
...
._
,
Die made of carbon tool steel Cracking occurred .
lhin rim and 1he body (At ,0p) Longitudinal section through -
R die. Numbers are Rockwell hardness values (Bethlehem
ighness. Preferred practice is to utilize long* steel Company) ._
-
' .
I
r
f
t';'
i
410
1
.
INTRODUCTION TO PHYSICAL METALLURGY
Faulty Heal; Treatment
steels are tempered between 950 and 1100 F and the use of a double draw ,
failures.
which repeats the original cycle is common practice. 10-20 Tool Failures The analysis to determine the probable cause for premature
They should be removed from th( immediately to a tempering furm
tool failure is often complex. Thorough investigation, however, will generally reveal good reasons behind every tool failure. It is the purpose ol this section to discuss briefly five fundamental factors that contribute to
section on heat treatment, tool; !
temperature causes grain coarse
Faulty Tool Design This may lead to failure either in heat treatment or during service. When a tool is to be liquid-quenched the use of heavy sections next to light sections should be avoided. During quenching the light sections will cool rapidly and harden before the heavy sections This will set up quenching stresses that often result in cracking. Figure 10-8 shows
face. Evidence of overheating mi
a failed die made of carbon tool steel, while Fig. 10-9 shows a cracked die
this steel was heated to 1800oF im
,
.
made of manganese oil-hardening tool steel.
'
In each case cracking oc,
curred during quenching because of excessive stresses set up by the drastic change in section. This type of failure may usually be avoided by making the tool as a two-piece assembly. The use of square holes is another prime source of tool failure due to faulty design. If it is essential to use adjacent heavy and light sections or sharp corners the use of an air-hardening steel
V ,\
.
mended hardening temperature.
tool failure.
,
'
This fact
Tools should be proper
Figure 10-12 shows a mangani cracked in hardening. The mien
tensite typical of overheated ste of fine tempered martensite and
Improper Grinding
Very high su
tool because of the grinding oper to cause cracks. Light grinding
rection of grinding, while heavy
,
;
is recommended.
1
Faulty Steel Despite the careful control used in the manufacture and inspection of tool steels, occasionally there is some defect in the steel. There may be porous areas resulting from shrinkage during solidification of the ingot which are known as voids or pipe. There may be streaks or Zaps due
to segregation or nonmetallic inclusions, which usually run longitudinally
with respect to the original bar stock. Other defects are tears, which are '
x
transverse surface defects resulting from working the steel under conditions where it does not have sufficient ductility; internal-cooling cracks known as flakes; and surface-coormg cracks as a result of cooling too rapidly after the last forging or rolling operation.
I?
k
Tools made from large bar stock (over 4 in. diameter) of high-chromium
steels generally show a brittle carbide network due to insufficient hot work (Fig. 10-Ha). The use of disks of small bar stock which are upset-forged [ provides additional hot work, which breaks up the brittle carbide network
and ensures a more uniform carbide distribution. Figure 10-10 shows an I
'
it
diameter milling cutter which failed in service, and subsequent micro- [ study showed the presence of a brittle carbide network. This milling cutter 8-in
-
.
was made of bar stock and did not undergo sufficient reduction by hot
working to remove the remnants of the as-cast carbide network. The normal microstructure, with good carbide distribution resulting from a prop-
erly hot-worked high-speed steel after quenching and tempeihng, is illustrated in Fig. 10-1 lb.
i
\
Fig 10-9 Die made of manganese oil-hardening Cracking occurred because of excessive quenchi set up between the heavy body and the small pre section. (Bethlehem Steel Company.)
RGY
TOOL STEELS
50 and 1100"F, and the use of a double draW|
Faulty Heat Treatment This factor is the cause of the large majority of tool failures. Tools should be properly handled during and after the quench They should be removed from the quench while still warm and transferred immediately to a tempering furnace. As was pointed out in the previous section on heat treatment, tools should be quenched from the recommended hardening temperature. The use of excessively high hardening
le is common practice.
.
.f determine the probable cause forprematuri '
Thorough investigation, however, will gen
'
411
lind every tool failure. It is the purpose of| five fundamental factors that contribute to|j
temperature causes grain coarsening which is evident on a fractured sur,
/ lead to failure either in heat treatment or,:?
.
face. Evidence of overheating may usually be found by microexamination Figure 10-12 shows a manganese oil-hardening tool-steel cam which .
to be liquid-quenched, the use of heavy sec* uld be avoided. During quenching, the light harden before the heavy sections. This will often result in cracking. Figure 10-8 shows dI steel, while Fig. 10-9 shows a cracked die
cracked in hardening. The microstructure revealed coarse acicular martensite typical of overheated steel instead of the normal microstructure ,
,
of fine tempered martensite and spheroidal carbides. It is estimated that
this steel was heated to 1800oF instead of the proper temperature of 14750F
.
ling tool steel. In each case, cracking oc-<|
Improper Grinding Very high surface stresses may be set up in a hardened tool because of the grinding operation. These stresses may be high enough
use of excessive stresses set up by the dras-
i of failure may usually be avoided by making g :
i .
'
to cause cracks. Light grinding cracks tend to appear at 90° from the direction of grinding, while heavy grinding cracks present a characteristic
ly. The use of square holes is another prime I
ulty design. If it is essentia! to use adjacent 1 Vp corners, the use of an air-hardening steel |
'
-
;:-ful control used in the manufacture and in-
:
nally there is some defect in the steel. There ij from shrinkage during solidification of the?!
s or pipe. There may be streaks or Zaps due || inclusions, which usually run longitudinally :
'
r stock
: -
.
Other defects are fears, which arejB|?'
ulting from working the steel under condi-|p sufficient ductility; internal-cooling cracksT
ooling cracks as a result of cooling too rap- | '
Mpng operation.
I
:
r;$-.
ock (over 4 in. diameter) of high-chromium|
'
iarbide network due to insufficient hot work;
5 of small bar stock which are upset-forgedl
7
.
.
m
ivhich breaks up the brittle carbide networks
arbide distribution. Figure 1(M0 shows an|
r
ilch failed in service, and subsequent micro-;! .
i brittle carbide network. This milling cutter I id not undergo sufficient reduction by hot4
hi
..
'
;
.
..
J
.
"
-
.
liiiii
ts of the as-cast carbide network. The nor'JjWgy
| carbide distribution resulting from a prop K"?9,10-9 010 mado of manaanese oii-hardoning tool stool, Ki , , is illus*!aBP'Cracklng occurred becauseol excessive quenching stress set up between the heavy body and the small protruding section. (Bethlehem Steel Company.) '
'
.
"
--
_
.
S%ei after quenching and tempering
;
-
: -
-
J
.
,
.
.
.
.
,
......-.
-
412
INTRODUCTION TO PHYSICAL METALLURGY
l?r %& m *':
V
km 1
4« 3
i
1 v
o
7
Fig. 10-10 Failed milling cutters made of M2 high-speed steel. Note chipped teeth. Failure occurred because of poor carbide distribution resulting from insufficient reduction by hot working in the steel plant. (Bethlehem Steel Company.)
Fig, 1011
M2 high-speed steel, (a) Longitudinal
an 8-in. bar showing a carbide network due to i hoi working, (b) Longitudinal section of a 1-in. b; '
0'
!crg>,-<1 showing normal spheroidal-carbide distribi
aiier sufficient hot working. Etched in 4 percent r 'Lairobe Steel Company.)
network pattern (see Fig. 1-33a). The presence of grinding cracks is best revealed by magnetic-particle testing.
SPECIAL CUTTING MATERIALS
Mechanical Overload and Operational Factors Mechanical factors that cause tool failures due to overload may be accidental or are the result of excessive stress concentration or improper clearances and alignment. This type of tool failure is often difficult to determine, since thorough investiga-
10-21 Steflites These are essentially contain from 25 to 35 percent c
tion of the failed tool will not reveal any cause for the short life. Figure 1CM3 shows a failed reamer made of M2 high-speed steel. The bushing on
tent. Microscopically, the alloys
the left was drilled to an undersize hole, and when this reamer attempted
and corrosion, and excellent red
to take an excessively heavy cut, breakage resulted. A common method of failure due to operational factors occurs in tools
makes them very suitable for cu Stelllte metal-cutting tools are
used for hot-work operations. These tools are subjected to repeated thermal stresses because of alternate heating and cooling of the tools. This leads to a network of very fine hairline cracks known as heat checks. Figure
cast steel, stainless steel, brass,
percent carbon, and the remain from Rockwell C 40 to 60, depe
Their outstanding properties ar
be operated at higher speeds th Stellite alloys are usually cast tc fore not so tough as high-speec and more brittle than high-speec machining operations is neces: cutting tools are used as single large inserted tooth cutters, spc
10'14 shows heat checks developed on the surface of a punch tip made of H12 hottwork tool steel. It should be realized that, under these severe op-
erating conditions, eventual failure of the tool is to be expected and thai | there is ino simple solution to the problem of avoiding failure due to heat checking.
A
m .i
fete.
:urgy
TOOL STEELS
413
V
; :
'
J 7
(a) .
(b)
jiigh-speed
.
.
icuuso o( ;
,
i
4 jj Fig. 10-11
Cient reduc-
M2 high-speed steel, (a) Longitudinal section
i | W an 8-in. bar showing a carbide network due to insufficient
m Steel
4 f' hot working, (ti) Longitudinal section of a 1-iir bar upset-
?l |; forged showing normal spheroidal-cnrbido dislribulion
fi I? after sufficient hot working ! |" (Latrobe Steel Company.)
.
Etched in 4 percent nilal, 100x.
*
3a). The presence of grinding cracks is best
SPECIAL CUTTING MATERIALS
testing. Deratlonal Factors ;
10-21 Stellites
Mechanical factors that
rload may be accidental or are the result of |B
i or improper clearances and alignment. This -I .
:
Jicult to determine, since thorough investiga-
lade of M2 high-speed steel. The bushing on jrsize hole, and when this reamer attempted ;ut, breakage resulted. e due to operational factors occurs in tools
.
.
;
1 These tools are subjected to repeated ther- ] |;
r
;
.
This
s:*i:hairline cracks known as heat checks Figure
r
-
.
.
.
:
They
,
t reveal any cause for the short life. Figure
Annate heating and cooling of the tools
These ore ossenlially cobalt-ohiomimn-tungsten alloys.
contain from 25 to 35 percent chromium, 4 to 25 percent tungsten, 1 to 3 percent carbon, and the remainder cobalt. The hardness of stellite varies from Rockwell C 40 to 60 depending upon the tungsten and carbon content. Microscopically, the alloys consist mainly of tungstides and carbides. Their outstanding properties are high hardness, high resistance to wear and corrosion, and excellent red-hardness. This combination of properties makes them very suitable for cutting applications
sloped on the surface of a punch tip made of ould be realized that, under these severe opliure of the tool is to be expected and that 1 I the problem of avoiding failure due to heat .
-
'
.
Stellite metal-cutting tools are widely used for machining steel, cast iron cast steel, stainless steel, brass, and most machinable materials. They may be operated at higher speeds than those used with high-speed-steel tools. Stellite alloys are usually cast to the desired shape and size and are therefore not so tough as high-speed steels. They are also appreciably weaker and more brittle than high-speed steels so that a careful analysis of specific ,
,
machining operations is necessary in selecting the proper tool. Stellite cutting tools are used as single-point lathe tools milling cutter blades for large inserted tooth cutters spot facers, reamers, form tools, and burnish,
,
414
INTRODUCTION TO PHYSICAL METALLURGY
I
a
IJ
f
II
'
.V;J.y5
1
it
v
r
..
f
i /i4
t
-
i fr
r
r
Fig. 10-12 Cam made of manganese oil-hardening tool steel which cracked in hardening because it was quenched
r;
i
from an excessively high temperature (Bethlehem Steel Company.) .
,
ing rollers. Slellite is also used as a hard-facing material for trimming dies on the wear- \
and gauge blocks on plowshares and cultivators for farm use ,
ing parts of crushing and grinding machinery cavating and dredging equipment
,
,
and on bucket teeth for ex-'
?
.
Fig. 10-14
Punch ;ip, 3% in. diameter by 83/4 in.
HI2 hot-work steel. This tool became stuck in he several times and was heated to a much higher t
than it was designed to withstand. Heat checks (Bethlehem Steel Company.)
10-22 Cemented Carbides
i
ft
Fig. 10-13 (Right) Failed reamer made of M2 high-speed I
steel. (Left) Bushing on which failure occurred (Bethlehem Steel Company ) .
.
i :
i
These rru
bide particles of the refractory alloy of the iron group, formir compressive strength. Cemen metallurgy techniques. The pr powder carbides of tungsten, ti these powders with a binder, u powder into compacts of the shapes to achieve consolidatior The blended powders are fo followed by sintering; or by he tering are done at the same tirr tween 5 and 30 tons/sq in., def
1
=10 Y
TOOL STEELS
415
n
8'
l
I I
r
7
"
"
'
ting tool s quenched em Steel
El;!!-
5
d as a hard-facing material for trimming die$ res and cultivators for farm use, on the wear-
ding machinery, and on bucket teeth for ex lent.
10-14 Punch tip, 3% in. diameter by 8% in., made of 112 hot-work steel. This tool became stuck in hot forgings veral times and was heated to a much higher temperature than it wa:; designed to withstand. Hotil checks resulted.
|Bettilehern Steel Company.)
|0-22 Cemented Carbides These materials are made of very finely divided car5
bide particles of the refractory metals, cemented together by a me al or alloy of the iron group, forming a body of very high hardness and high compressive strength. Cemented carbides are manufactured by powdermetallurgy techniques. The process consists essentially in preparing the powder carbides of tungsten, titanium, or tantalum; mixing one or more of these powders with a binder usually cobalt powder; pressing the blended powder into compacts of the desired shape; and sintering the pressed
I
"
,
shapes to achieve consolidation.
i high-speed !d (Bethlehem .
The blended powders are formed into desired shapes by cold pressing, followed by sintering; or by hot pressing, during which pressing and sintering are done at the same time. Pressures used m cold pressing vary between 5 and 30 tons/sq in., depending upon the size and shape of the com-
416
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 10-6
pact. Sintering is carried out at temperatures between 2500 and 2700°Ftor i
Typical Compressive Mechanical F
30 to 60 min. At these elevated temperatures, the cobalt forms a eutectic \ with the carbides
,
MEASURED IN COM MOD
and this eutectic becomes the cementing material. After f
OF
cooling, the sintered compact has its final properties, since it does not re- j spond .to any known heat treatment. The carbides are present as individual J
grains, and also as a finely dispersed network resulting from the precipita- I tion diiring cooling of carbide dissolved in the cobalt during sintering.
'
\
Carbides may be classified into two broad categories: (1) the straighttungsten carbide grades, used primarily for machining cast iron, austenitic r
steel, and nonferrous and nonmetallic materials; and (2) the grades con- | tainingj
major amounts of titanium and tantalum carbides, used primarily
for machining ferritic steel. A more detailed classification basdd on compositioh is shown in Table 10-5. The exceptional tool performance of sintered carbide results from high hardness and high compressive strength combined with unusual red-hardness. The lowest hardness of sintered carbide is approximately the same as the highest hardness available in tool steel, Rockwell A 85 (Rockwell C 67). Typical compressive mechanical properties of sintered carbides are
COMPRESSIVE
ELASTIC
ELAS
STRENGTH PSI
LIMIT PSI
MILL PSI
1 (3% Co)
615,000
500,000
105
1 (6% Co)
614,000
286 000
105
2 (10% Co) 3 (16% Co)
600 000
125,000
87
545,000
95,000
76
5
625,000
230,000
78
6
533,000
97,000
80 82
CARBIDE onoup
,
,
7
635,000
173,000
6
631,000
250 000
81
9
705,000
240,000
86
,
.rrom Metals Handbook 8th ed.. vol. 3, p. 319. American S ,
Values for impact strength are from unnotched specimen: lvalues for fatigue limit are based on 20 million cycles of si
'
-
shown in Table 10-6. TABLE 10 5
Classification of Cemented Carbides*
Microstructure affects hardne
COMPOSITION %
CARBIDE
REMAINDER WC
HARDNESS
TYPICAL
GROUP
Co
R/A
APPLICATIONS
TaC + TiC
particles (grains), their distributr between cobalt and carbide cry tungsten carbide grain size lowt
STRAIGHT TUNGSTEN CARBIDE 1
2 5-6 5 .
.
93-91
0-3
92-85
"
Finishing to medium roughing cuts on cast Iron, nonferrous alloys, and superalloys; low-impact dies
6 5-15
0-2
3
15-30
0-6
ness decreased from Rockwell -
4
3-7
20-42
93 5-92
Light high-speed finishing cuts on steel
5
7-10
10-22
92 5-90
Medium arts and speeds on steel
6
10-12
8-15
92.0-89
Roughing cuts on steel
from fine to intermediate to coe
dies
88-85
Since cemented carbides hav
High-impact dies
usual practice is to braze or mi material (called an insert) to a
ADDED CARBIDE PREDOMINANTLY TiC .
.
ADDED CARBIDE PREDOMINANTLY TaC 7 8
4 5-8 .
8-10
16-25
93-91
Light cuts on steel
12-20
92-90
General purpose and heavy cutting of steel
ADDED CARBIDE EXCLUSIVELY TaC
9
5.5-16
18-30
91.5-84.0
Wear-resistant applications, particularly involving heat
i
From Metals Handbook, 8th ed., vol, 3, p. 316, American Society for Metals, Metals Park, Ohio, 1967.
*
:
m
1
lakes," which are interspersed
lustrated in Fig. 10-15 for a 94 pe
Rough cuts on cast iron: moderate-impact ?
2
.
The coir
stantially more than 500,000 psi, Both properties seem to decrea
under the cutting edge (Fig. 10' tachment is usually preferred. 1
the tool must be removed from t -
r
icon carbide or diamond-imprec tions only one ortwo cutting edc the tip is mechanically secured i to the next cutting edge without this procedure, more cutting ed' have been used, common practi less expensive to replace it thf hazard of damage to the insert
TOOL STEgLS
nov
if temperatures between 2500 and 271 .
Typical Compressive Mechenlcel Properties ot Sintered Carbldeu*
d temperatures, the cobalt forms a e jctic becomes the cementing material.
MEASURED IN COMPRESSION MODULUS OF
has its final properties, since it does nt
ELASTICITY,
POIS-
STRENGTH PSI
LIMIT PSI
MILLION PSI
SON'S RATIO
DUCTILirY,%
6 i 5,000
500,000
105
0 24
0 60
614.000
286.000
105
0 28
0 85
0 73
95
Co)
600.000
125.000
87
0 20
90
1 10
105
Co
545.000
95.000
76
0 22
2 70
1 75
625.000
230.000
78
022
1
00
0 60
90
533.000
97.000
80
0 22
? 00
0 40
90
635,000
173,000
82
0 21
0 90
631.000
250.000
81
1 00
0 60
85
705,000
240,000
86
70
0 60
85
m Co)
nto two broad categories: (1) the stralf primarily for machining cast iron, aus tenil metallic materials; and (2) the grades c ,
used primafil
sgmore detailed classification based on coi 1 1
nance of sintered carbide results from higfl
3 strength combined with unusual red-har
FATIGUE
ELASTIC
jersed network resulting from the prepfj dissolved in the cobalt during sinterlrii
IMPACT
COMPRESSIVE
nent. The carbides are present as indiyi*
lium and tantalum carbides
417
STRENGTH
FT-LB+
,
LIMIT, 1000 PSIt
.
.
,
.
.
,
.
.
.
.
,
.
.
.
.
.
,
.
1
0 22 ,
.
.
.
.ftom MetaK Handbook 8th ed.. vol. 3, p. 319, American Society for Metals, Matals Park, Ohic, 1967. ,
iV«IU(» for impact strength are from unnotchcc) specimens in sections approximalQly 'A in. spuare.
sintered carbide is approximately the samci|
Values fdr fatigue limit are based on 20 million cycles of stress, for specimens, ol the R. R. Moore rotatmg-beam type.
able in tool steel, Rockwell A 85 (Rockwel|||
.
/iChanical properties of sintered carbides arfti shown in Table 10-6. The compressive strength for most grades is substantially more than 500 000 psi, along with vary high modulus of elasticity. ,
les*
3
pSS
TYPICAL
APPLICATIONS
JNGSTEN CARBIDE
Finishing to medium roughing cuts on cast iron, nonferrous alloys, and superalioys; low-impact dies
Rough cuts on casl iron; moderate-impact :M dies if High-impact dies I PREDOMINANTLY TiC
Light high-speed finishing cuts on steel Medium arts and speeds on steel Roughing cuts on steel PREDOMINANTLY TaC
Light cuts on steel
* .
.
General purpose and heavy cutting of steel i
c -
E EXCLUSIVELY TaC
Jl.
"
'
.
0
Wear-resistant applications, particularly involving heat
pociely loi Molals, Mollis Pmk, Ohio, 1967.
Both properties seem to decrease with increasing cobalt content. Microstructure affects hardness and strength. The size of the carbide
particles (grains), their distribution and porosity, and the quality of the bond between cobalt and carbide crystals are important factors. Increasing the tungsten carbide grain size lowers the hardness, because the softer cobalt lakes," which are interspersed between grains, are also larger. This is illustrated in Fig. 10 "t5 for a 94 percent WO, 6 percent cobalt alloy. The hard"
ness decreased from Rockwoll A 93 to 92 lo 91 as the WO grain size went from fine to intermediate to coarse,
Since cemented carbides have low toughness and tensile strength, the usual practice is to braze or mechanically fasten a small piece of carbide material (called an insert) to a steel shank, which provides rigid support under the cutting edge (Fig. 10-16). Where space permits, mechanical attachment is usually preferred. When a tip is brazed to the body or holder, the tool must be removed from the machine for resharpening by using a silicon carbide or diamond-impregnated grinding wheel. Under these conditions only one or two cutting edges of the tip can be used. In contrast, when the tip is mechanically secured in the holder it can be loosened and turned to the next cutting edge without removing the tool from the machine. With this procedure, more cutting edges are available for use. After all the edges ,
have been used
,
common practice is to discard the carbide tip because it is
less expensive to replace it than to recondition it by grinding. Also hazard of damage to the insert by brazing is eliminated.
,
the
418 INTRODUCtjlON TO PHYSICAL METALLURGY
V.
The high hardness and wear
8
well suited for earth drilling an cialized rock bits for drilling in (
1
utilize carbide inserts instead
They are also used for such apf
mills, facings for jaw crushers gage blocks. Cemented carbid sten and molybdenum and for made of steel, copper aluminur
m
,
m
Si
,
I: >
t
Cermets usually contain the nickel or a nickel-base alloy as hard phase is predominantly tit tions and 30 to 70 percent nic high resistance to oxidation r and relatively low density but i used where high-temperature £
J
,
,
When the binder content is le;
;
PIP
JEWS
Fig. 10-15 Cemented carbide, 94 percenl WtJ i
Cam
6 percent Co. Grain size of WC is fine in (a) intermediate in (fe), and coarse in (c). Hard-
ness decreased from Rockwell A 93 in (a) to Ij
Seat
92 in (b) and 91 in (c). Etched in Murakami's reagent, 1,500x. (From Metals Handbook. vol. 7,
|(£l
1b
"
Insert
*
Atlas of Microstructures," American 'in
Society for Metals, 1972.)
More cemented carbides are consumed for metal cutting than for any
(a)
other-type of application. Because of their ability to retain a sharp cut1irjg.|| edge, the straight-tungsten carbide grades are virtually the only tool- material used to cut abrasive materials such as Fiberglas and phenolic resins
Carbides which have the highest hardness are also being used for produc-j
poAl
tion putting of white cast iron at Rockwell C 60. Cemented carbides are
usediifor drills, reamers, boring and facing tools, and saws for the machining of both metals and nonmetals. Cutting speeds and feeds employed vyilli carbide tools are generally higher than those used with high-speed steelorte stellite. '
.
.
i
(b) Fig. 1016
Carbide tools: (a) mechanical holder
hrazed tool. (Courtesy of the DoALL Company.)
IGY
TOOL STEELS
419
The high hardness and wear resistance of cemented carbides make them well suited for earth drilling and mining applications. Various types of specialized rock bits for drilling in extremely hard and abrasive rock formations utilize carbide inserts instead of the conventional hard-faced steel teeth
.
They are also used for such applications as facings for hammers in hammer mills, facings for jaw crushers sandblast nozzles, ring and plug gages and gage blocks. Cemented carbide dies are used for the hot drawing of tungsten and molybdenum and for the cold drawing of wire bar, and tubing ,
1
,
,
made of steel, copper, aluminum, and other materials
.
1
-
L I,
Cermets usually contain the carbides of titanium and chromium with nickel or a nickel-base alloy as the binder. In the most common grade the hard phase is predominantly titanium carbide with chromium carbide addi,
tions and 30 to 70 percent nickel binder. This grade has high hardness
,
Vr
high resistance to oxidation relatively high resistance to thermal shock, ,
and relatively low density but it also has low ductility and toughness. It is used where high-temperature abrasion resistance is the primary objective. When the binder content is less than 20 percent the cermets have been ,
(1
5
,
,
.Insoft
Fig. 10-15 Cemented carbide, 94 percent Wftj s: 0 porconl Co. Grnm slzo of WC Is lino In (a) intermediate in (b), and coarse In (c), Hardness decreased from Rockwell A 93 in (a) to
I
Cam Tool holder Seal
'
92 in (t>) and 91 in (c). Etched in Murakami reagent, 1,500x. (From Metals Handbook
:
,
vol. 7,
Atlas of Microstructures, American
"
"
Society for Metals, 1972.)
3
ire consumed for metal cutting than for annyi jcause of their ability to retain a sharp cutting; carbide grades are virtually the only tool mate-
pSterials such as Fiberglas and phenolic resins. Jj hest hardness are also being used for produc-| '
'
. ..
on at Rockwell C 60. Cemented carbides arejs
OoALl
;':ng and facing tools, and saws for the machin- j
:
etals. Cutting speeds and feeds employed with
igher than those used with high-speed steel orM
Fig 10-16
v:
i '
-
A
-
Carbide tools: (a) mechanical holder and (b)
brazed tool. (Courtesy of the DoALL Company.)
. .
j
.
420
used f6r cutting both steel and cast iron at high speed with medium to light , \ chip loads.
S 50
i".
.
10-23
Ceramic Tools Most ceramic or cemented oxide cutting tools are manu- . factured primarily from aluminum oxide Bauxite (a hydrated form ol aluminum oxide) is chemically processed and converted into a denser
40
crystalline form called alpha alumina. Fine grains are obtained from the {
10
.
10
Cermet v
20
LTJC-Nih
'
T
0 400
precipitation of the alumina or from the precipitation of the decomposed
7 Carbide (C- S)
,
600
alumina compound.
800
1000
1200
14
Cuttinq speed sfm ,
Ceramic tool inserts are produced by either cold or hot pressing. In cold
Fig 10-17
Relation of toollife to cutting speed
pressing, the fine alumina powder is compressed into the required form and I Mrmet'and carbide t00ls when cutting gray iron
then sintered in a furnace at 1600 to 1700°C. Hot pressing combines form- : ZZTZl pL t
ing and sintering, pressure and heat being applied simultaneously. Small amounts of titanium oxide or magnesium oxide may be added for certain types of ceramics to aid in the sintering process and to retard growth. After
the inserts have been formed, they are finished with diamond-impregnated ! grinding wheels. |, Ceramic cutting tools are most commonly made as a disposable insert
whicri is fastened in a mechanical holder. Disposable inserts are available in many styles, such as triangular, square, rectangular, and round. As with when a cutting edge becomes dull, a sharp edge may be obtained by the insert being indexed (turned) in the holder. Ce-
. cemented carbide tools,
'
ramic inserts may also be fastened to a steel shank by epoxy glue. This |
When ceramic cutting tools a used on accurate rigid machinf
1 Machining time is reduced becai ranging from 50 to 200 percent highi (Fig. 10-17). 2 High stock-removal rates and ini of cut can be made at high surface 3 Under proper conditions a ceran carbide tool.
ing in mechanical holders.
5
More accurate size control of tl
wear resistance of ceramic tools.
6
They withstand the abrasion of:
ness, chemical inertness, and resistance to wear. The high hardness and
7
Heat treated materials as hard a
wear resistance of alumina are the main reasons for its use in machining )
8
They produce a better surface fi
The disadvantages of ceramic 1 2 3
.
TABLE 10-7
4
They are brittle and tend to chif Initial cost is approximately twit A more rigid machine is necess Considerably more horsepower
Mechanical Properties of Tool Materials*
PROPERTY
CERAMIC
HIGH-SPEED
C2
STEEL
CARBIOf if QUESTIONS
W
90,000
500,000
230,000
Compressive strength, psi
500,000
600,000
650,000 4
10-1 List the properties most impt cation where each property would
Moduilusof elasticity, psi
60x106
32x106
100x10*
10-2
93
85
92
1780
740
1800
Transverse rupture strength psi ,
Hardness, Rockwell A
MicrOhardness, Knoop lOOg ,
'
,
tt-:-
i
t
8th ed., vol. 3, p. 323, American Society for Metals, Metals Park, Ohio, 1967.
What would be the influence
properties of a tool steel: chromiun ganese, and cobalt? 10-3
.From Metals Handbook :
i
method of holding the insert almost eliminates the strains caused by clamp-
cast iron and hardened steel at high cutting speeds. The inertness of alumina to iron at high temperature prevents welding of the tool to steel or cast iron workplaces and contributes to the production of good surface finish. Table 10-7 lists some mechanical properties of ceramic and other cutting-tool materials.
I"
o 'L'
4 Ceramic tools retain their streng (in excess of 2000oF).
The principal elevated-temperature properties of alumina are high hard-
1
:
INTRODUCTION TO PHYSICAL METALLURGY
Describe the basis for select
on AISI 1020 steel.
am
RGY
TOOL STEELS
cast iron at high speed with medium to ligf
421
50
Grny iron "
<5
~
or cemented oxide cutting tools are mari1
;:
inum oxide.
5 40
Bauxite (a hydrated form
y processed and converted into a densefj
9
\tlumina. Fine grains are obtained from tti
II'
Ceramic -
30
Cermet
10
( ric-Nji .
Carbide (C-8)-
.
10
0 100
r from the precipitation of the decompose
=8 600
800
1000
Cutting speed
,
buced by either cold or hot pressing. In coli per is compressed into the required form and:
?! fig.
1200 sfm
1400
10-17 Relation of tool life to cutting speed for ceramic
,
and carbide tools when cutting gray Iron. (From Metals Handbook 8th ed., vol. 3, p. 323, American Society for Metals Metals Park, Ohio, 1967.) cermet
,
glf&OO to 1700oC. Hot pressing combines form*
,
d heat being applied simultaneously. Smajl
,
I magnesium oxide may be added for certalri| sintering process and to retard growth. Aftef S they are finished with diamond-impregnated :
When ceramic cutting tools are properly mounted in suitable holders and used on accurate rigid machines they offer the following advantages: ,
iI
viost commonly made as a disposable insert nical holder. Disposable inserts are available f '
'
jlar, square, rectangular, and round. As with J n cutting odgo hocomos dull, n sharp edgo lj
'
|
'
\
t being indexed (turned) in the holder. Ce;tened to a steel shank by epoxy glue
2 High stock-removal rates and increased productivity result because heavy depths of cut enn be made at high surfaco speeds. 3 Under proper condilions a ceramic tool lasts liom Ihree to ten times longer than a 4
Imost eliminates the strains caused by clamp-
;
.
More accurate size control of the workpiece is possible because of the greater
wear resistance of ceramic tools.
| oerature properties of alumina are high hard-;
: :
Ceramic tools retain their strength and hardness at high machining temperatures
(in excess of 2000oF).
5
i resistance to wear. The high hardness and t re the main reasons for its use in machining at high cutting speeds. The inertness of aluiture prevents welding of the tool to steel or V .ntributcs to the production of good surface mechanical properties of ceramic and other
Machining time is reduced because of the higher cutting speeds possible. Speeds
ranging from 50 to 200 percent higher than those used for carbides are quite common (Fig, 10-17).
carbide tool.
This;
.
1
6 7
They withstand the abrasion of sand and of inclusions found in castings. Heat treated materials as hard as Rockwell C 66 can be readily machined.
8
They produce a better surface finish than is possible with other cutting tools.
The disadvantages of ceramic tools are as follows: \
1 2 3 4
.
They are brittle and tend to chip easily. Initial cost is approximately twice that of carbide tools. A more rigid machine is necessary than for other cutting tools. Considerably more horsepower is required for ceramic tools to cut efficiently.
of Tool Materials*
CERAMIC
v
;c
.
I
HIGH-SPEED
C-2
STEEL
CARBIDE
'
i \
QUESTIONS
90,000
500,000
230,000
10-1
500,000
600,000
650,000
cation where each property would be required
60x106
32y106
100 X106
10-2 What would be the influence of each of the following alloying elements on the
.
93
85
92
1780
740
1800
List the properties most important for tool steels and gi\>e one industrial appli.
properties of a tool steel: chromium, tungsten, molybdenum, vanadium, silicon
,
10-3 323, American Society for Metals, Metals Park, Ohio, 1967.
man-
ganese, and cobalt?
Describe the basis for selection of a tool steel to be used as athreadrolling die
on AISI 1020 steel.
!
1
422
INTRODUCTION TO PHYSICAL METALLURGY
10-4
m
Describe the heat treatment you would apply to the tool steel selected in Ques-
tion 10-3 and the reasons for this heat treatment.
|.-
10-5 Same as Questions 10-3 and 10-4, for a die to be used to cold-head AIS11020' |h bearing rollers. % % 10-6 Using the equation given below, plot a graph showing the dimensional change,- ( inches per; inches, when spheroidite changes to austenite as the carbon content ; varies from,0 to 1.40 percent. Dimensional change in./in. = 0,1555 + 0.0075 (percent C) ,
10-7
Same as Question 10-6 for the change from austenite to martensite with vary-
ing carbon using the following formula: ,
CAS1
Dimensional change in./in. = 0.0155 - 0.0018 (percent C) ,
10-8 Plot the net dimensional change for the reaction spheroidite -» austenite-* , martensite' as the carbon content varies from 0 to 1.40 percent. (Hint: Use the equa- # tions given in Questions 10-6 and 10-7 to determine an equation for the net dimensional change.) ,
REFERENCES
Cast irons like steels
temperature.
,
most commercially manufactured
,
1970.
carbon.
American Society for Metals: "Metals Handbook," 7th ed., 1948: 8th ed., 1961,1964, [-
and vol 7, 1972, Metals Park, Ohio.
f
American Society of Tool and Manufacturing Engineers: "Tool Engineers Hand-' i book," 2d ed., McGraw-Hill Book Company, New York, 1959.
The ductility of cast iron is ven worked at room temperature. Mc any temperature. However they i cated shapes which are usually m ,
Bethlehem Steel Company: "The Tool Steel Trouble Shooter Handbook 322,". J
Bethlehem, Pa., 1952. |R Palmer, F. R., and G. V. Luerssen: "Tool Steel Simplified," 2d ed., Carpenter Ste E:
Co.. Reading, Pa., 1948.
"
§. American Society||
Seabright, L. H.: "The Selection and Hardening of Tool Steels," McGraw-Hill Book Company, New York, 1950.
Therefore cast iro
carbon. Since high carbon contei
American limn and Steel Institute: Steel Products Manual, "Tool Steels," New Yorli
Roberts, G. A., J. C. Hamaker, and A. R. Johnson: "Tool Steels. for Metals, Metals Park, Ohio. 1962.
,
In relation to the iron-iron carbk amount of carbon than that necej
:
Allegheny Ludlum Steel Corporation: "Tool Steel Handbook," Pittsburgh, 1951.
i. /
r
1
11-1 Introduction
ing is the only suitable process a| casf irons.
Although the common cast ire
properties than most steels, they steel, and have other useful pro[ good foundry control, and approp type of cast iron may be varied ova in foundry control have led to the f whose properties are generally cc 112 Types of Cast Iron The best met! to metallographic structure. Th( which lead to the different types the alloy and impurity content th ,
and the heat treatment after castir
of the carbon and also its physics
!
-
,
iron carbide in cementite, or it m
shape and distribution of the free
i ;
:
.
: ; r
-
-
j
.
I
i
3Y
i
i '
u would apply to the tool steel selected in Que*
'
at treatment.
r-ifyO-4
,
for a die to be used to cold-head AISI 1020
plot a graph showing the dimensional change,,.
.
s changes to austenite as the carbon content
"
/in. = 0.1555 + 0.0075 (percent C)
change from austenite to martensite with
vary-
ula:
CAST IRON
l/in. =0.0155 -0.0018 (percent C)
ige for the reaction spheroidite austenite ! -ies from 0 to 1.40 percent. (Hint: Use the equa i-7 to determine
an equation for the net dimen
1M
Tool Steel Handbook, Pittsburgh, 1951.
i:
"
"
yiv eel Products Manual, s Handbook,
"
Tool Steels," New York, |
"
Cast irons like steels, are basically alloys of iron and carbon. In relation to the iron-iron carbide diagram, cast irons contain a greater amount of carbon than that necessary to saturate austenite at the eutectic temperature. Therefore, cast irons contain between 2 and 6.67 percent carbon. Since high carbon content tends to make the cast iron very brittle, most commercially manufactured types are in the range of 2.5 to 4 percent ,
carbon.
7th ed., 1948; 8th ed., 1961,1964;
lufacturing Engineers:
Introduction
The ductility of cast iron is very low, and it cannot be rolled, drawn, or
Tool Engineers Hand-|
"
Company, New York, 1959.
Tool Steel Trouble Shooter Handbook 322,
Tool Steel Simplified, 2d ed.. Carpenter Steel
worked at room temperature. Most of the cast irons are not malleable at any temperature. However, they melt readily and can be cast into complicated shapes which are usually machined to final dimensions. Since casting is the only suitable process applied to these alloys, they are known as
"
"
A
.
'
12.
R. Johnson: "Tool Steels,
casf irons.
American Society
i Hardening of Tool Steels, McGraw-Hill Book| "
Although the common cast irons are brittle and have lower strength properties than most steels, they are cheap, can be cast more readily than steel, and have other useful properties. In addition, by proper alloying, good foundry control, and appropriate heat treatment, the properties of any type of cast iron may be varied over a wide range. Significant developments in foundry control have led to the production of large tonnages of cast irons whose properties are generally consistent.
|vi1-2 Types of Cast Iron The best method of classifying cast iron is according to metallographic structure. There are four variables to be considered which lead to the different types of cast iron namely the carbon contnnt, the alloy and impurity content the cooling rate during and after freezing. ,
,
*
and the heat treatment after casting. These variables control the condition of the carbon and also its physical form. The carbon may be combined as iron carbide in cementite, or it may exist as free carbon in graphite. The shape and distribution of the free carbon particles will greatly influence the
i
4;!4
INTRODUCTION TO PHYSICAL METALLURGY
physical properties of the cast iron. The types of cast iron are as follows. White ca,st irons, in which all the carbon is in the combined form as cementite. ft
Malleable cast irons, in which mosj or all of the carbon is uncombined in the form of irregular round particles known as temper carbon. This is obtained by heat treatment of white cast iron. Gray cast irons, in which most or all of the carbon is uncombined in the form of graphite flakes. Chilled cast irons, in which a white cast-iron layer at the surface is com-
..
oustemre
4ustenite
bined with a gray-iron interior. Nodular cast irons, in which by special alloy additions, the carbon is
S 1566
,
largely uncombined in the form of compact spheroids. This structure differs from malleable iron in that it is obtained directly from solidification and the round carbon particles are more regular in shape. Alloy cast irons, in which the properties or the structure of any of the above types are modified by the addition of alloying elements.
-
7; Z 7'
Fernte
+
+ Dustemte
eutectic
133;
to
Fernte
11-3 White Cast Iron The changes that take place in white cast iron during
"
iquid
Fernte
solidification and subsequent cooling are determined by the iron-iron carbide diagram discussed in Chap. 7. All white cast irons are hypoeutectic alloys and the cooling of a 2.50 percent carbon alloy will now be described. The alloy, at x, in Fig. 11-1, exists as a uniform liquid solution of carbon
-Peorlite +
F
-
peorlite
,
> >»; -
- -j
-
,
1
dissolved in liquid iron. It remains in this condition as cooling takes place
until the liquidus line is crossed at Xa. Solidification now begins by the formation of austenite crystals containing about 1 percent carbon. As the
temperature falls, primary austenite continues to solidify, its composition moving down and to the right along the solidus line toward point C. The liquid in the meantime is becoming richer in carbon, its composition also
r
0
I
0.8 1
U-
-
2
2.5
3
Steels
Weight pe
Fig. 1M
The metastable iron-iron carbide phase d
moving down and to the right along the liquidus line toward point E. At the Jj eutectic temperature, 2065oF
,
the alloy consists of austenite dendrites con-1
taining 2 percent carbon and a liquid solution, containing 4.3 percent car-
bon. The liquid accounts for (2.5-2.0)/(4.3-2.0), or 22 percent of the alloy i
by weight. This liquid now undergoes the eutectic reaction isothermally tcv|
form the eutectic mixture of austenite and cementite known as ledeburite.
*
Since the reaction takes place at a relatively high temperature, ledeburite |
"
.
.
v
ing behind layers of massive, free cementite.
-]
As the temperature falls, between x3 and x,, the solubility of carbon in
Since white cast ire
remaining .austenite containing 0.8 percent carbon and constituting
cause of this brittleness and lack
!
T
'
compound.
cementite as a continuous interd
i
>0: V:v|
(Fig, 11 -26) reveals that the dark ; It was pointed out in Chap. 7 tl
austenite decreases, as indicated by the Acm line CJ. This causes precipitation of proeutectoid cementite, most of which is deposited upon the cementite already present. At the eutectoid temperature, 13330F, thel
-
'
reaction isothermally to form pear
temperature, the structure remain The typical microstructure of w transformed austenite (pearlite) in tite, is illustrated in Fig. 11-29. h
tends to appear as a coarse mixture rather than the fine mixture typical ol many eutectics. It is not unusual for ledeburite to be separated completely, with the eutectic austenite added to the primary austenite dendrites, leav,
(6.67-2.5)/(6.67-0.8), or 70 perci
hard and wear-resistant but exti "
Completely white" cast irons an
SSrgy
'
v
CAST IRON
425
iron. The types of cast iron are as fol|0
all the carbon is in the combined forrM
.
?800
Liquid
h most or all of the carbon is uncomblnrol jticles known as temper carbon. This isi
Liquid
Liquid
I\ aurJclnlfi
ilte cast iron.
[st or all of the carbon is uncombinedih l
comcnlilp
2065
D
F
r AiiMcnilf
k white cast-iron layer at the surface Is CQm
1
Eutectic
Austenite
WtW' special alloy additions, the carboa:j| jn of compact spheroids. This structured!
1656
Fernte
C
(it is obtained directly from solidificatioh a|
+
+
oulectic
cementitn
/Acm
\0u5len1le /
'"
'
/
+
J
111
.
> more regular in shape
,
ne properties or the structure of any ot lh
-
ertite
ie addition of alloying elements.
that take place in white cast iron durinj|
'
.
f;fntf:
vM cooling are determined by the iron-ih
I
-
Chap. 7. All white cast irons are hypoeutectl
..
V;
c 11 li t'
'
I
r.nmen ilif
-
pearlite
percent carbon alloy will now be describll|
exists as a uniform liquid solution of carbonj lains in this condition as cooling takes placftf d at Xj. Solidification now begins by thefor*! lontaining about 1 percent carbon. As thfl tenite continues to solidify, its compositlpr3 :
along the solidus line toward point C.
T
5
„
U
0.8 1
'l
'I..'
FejC Steels
Cost irons
JVeiqM percent carbon
is
)ming richer in carbon, its composition also|
i
Fig. 11 1
The metastable iron-iron carbide phase diagram.
along the liquidus line toward point E. Atthefj
jhe alloy consists of austenite dendrites coi,:e v« liquid solution containing 4.3 percent ca$ .
....
: ;:
-
,
2 5-2 .
.
0)/(4.3-2.0). or 22 percent of the al|
dergoes the eutectic reaction isothermally t<| ustenite and cementite known as ledeburit
(6.67-2.5)/(6.67-0.8), or 70 percent of the alloy, undergoes the eutectoid reaction isothermally to form pearlite. During subsequent cooling to room temperature the structure remains essentially unchanged. ,
3 at a relatively high temperature, ledeburitel
The typical microstructure of white cast iron, consisting of dendrites of
lixture rather than the fine mixture typical ofi
transformed austenite (pearlite) in a white interdendritic network of cemen-
jal for ledeburite to be separated completely,!
tite, is illustrated in Fig. 11-23. Higher magnification of the same sample (Fig. 11-26) reveals that the dark areas are pearlite. It was pointed out in Chap. 7 that cementite is a hard brittle interstitial compound. Since white cast iron contains a relatively large amount of
ded to the primary austenite dendrites, leavp
free cementite.
,
.
v::>tween x3 and X4, the solubility of carbon irtll
,
ited by the Acm line CJ. This causes precipif ptite, most of which is deposited upon the|
.
cementite as a continuous interdendritic network it makes the cast iron ,
'
iAt the eutectoid temperature 13330F, ,
th
ing 0.8 percent carbon and constituti
!
hard and wear-resistant but extremely brittle and difficult to machine Completely white" cast irons are limited in engineering applications because of this brittleness and lack of machinability. They are used where .
"
r
.
426
INTRODUCTION TO PHYSICAL METALLURGY
The purpose of malleabiliza
4«
Li*
mi it
mm
in white iron into irregular nod Commercially, this process is
Pearlite
and second stages of the annt
Cementite
White irons suitable for com
A
range of composition: Percent
J
5«J N
Carbon
2 00-2.65
Silicon
0 90-1.40
Manganese Phosphorus
0 25-0.55
Sulfur
0 05
.
.
.
.eir#-
.
Less than 0. .
i2
m
i
In the first-stage annealing
,
t
temperature between 1650 and verted to austenite at the lower
* .y
.
(b) Fig. 11-2 T e microstructure of white cast iron, (a) Dark areas are primary dendrites of transformed austenite (pearlite) in a white interdendritic network of cementite, 20x. (M Same sample at 250x, showing pearlite (dark) and cementite (white). Etched in 2 percent nital.
2800
Liquid T
austenite 1 98 ,
resistance to wear is most important and the service does not require ductility, such as liners for cement mixers, ball mills, certain types of drawing dies, and extrusion nozzles. A large tonnage of white cast iron is used as a starting material for the manufacture of malleable cast iron. The range mechanical properties for unalloyed white irons is as follows: hardness Brinell 375 to 600, tensile strength 20,000 to 70,000 psi, compressive strength 200,000 lo 250,000 psi. and modulus of elasticity 24 to 28 million psi. 11-4 Malleable Cast Iron It was pointed out in Sec. 7-4 that cementite (iron carbide) is actually a metastable phase. There is a tendency for cementite
IrorvgraDr Austenite
1666 3
'
0 69 ,
/
i in
Iernle
to decompose into iron and carbon, but under normal conditions it tends
|
to persist indefinitely in its original form. Up to this point, cementite has
i
been treated as a stable phase; however, this tendency to form free carbon
s
is the basis for the manufacture of malleable cast iron.
i
The reaction FejC
iron-ce
rernle
3Fe + C is favored by elevated temperatures, the
existenqe of solid nonmetallic impurities, higher carbon contents, and the presence of elements that aid the decomposition of FeiC. On the iron-iron carbide equilibrium diagram for the metastable system,
Wei(
shown in Fig. 11-3, are superimposed the phase boundaries of the stable iron-carbon (graphile) system as dotted lines.
Fig. 11 3 The stable iron-graphite system (dott superimposed on tho molaslnble iron-iron carl
i
I
I
-
-
-
1
3Y
CAST IRON
,
427
5
The purpose of malleabilization is to convert all the combined carbon
i
Peorllt«i*
in white iron into irregular nodules of temper carbon (graphite) and ferrite
.
Commercially this process is carried out in two steps known as the first and second stages of the anneal ,
5?5
Cementite
.
m
White irons suitable for conversion to malleable iron are of the following range of composition:
i
Percent
1
Carbon
2 00-2.65
Silicon
0 90-1.40
.
.
pJ t Manganese It:/
1
<3h
0 25-0.55 .
Phosphorus
Less than 0.18
Sulfur
0 05 .
In the first-stage annealing the white-iron casting is slowly reheated to a ,
temperature between 1650 and 1750QF During heating, the pearlite is con.
verted to austenite at the lower critical line The austenite thus formed dis.
mm w
.
'
(a) Dark
mte (pearl5
,
;
1
2800
20X. (b)
_
iquid
rl cementite
Liquid
Liquid
+
austenite
cementite
1 98
iportant and the service does not requl ent mixers, ball mills, certain types of dravi A large tonnage of white cast iron is us(|
,
Iron-grnghite (2075°) Austenite 20
Iron-cemenfile 2065°)
V
13
.
.
knufacture of malleable cast iron. The ranj
i
,
+ cementite
0 69
d
%ength 20,000
Austenite
1666
illoyed white irons is as follows: hardrie|L
,
/**
and modulus of elasticity 24 to 28 millio i
Iron-graphite
1360°
Iron-cementite (1333°)
0 80 .
)inted out in Sec. 7-4 that cementite (iro(| le phase. There is a tendency for cementit irbon, but under normal conditions it tendsy
Ferrite
Ferrite
+
cementite
ginal form. Up to this point, cementite haj| however, this tendency to form free carbortl
e of malleable cast iron. V:;
'
-
.
:
-M
] C is favored by elevated temperatures, tKe|
1 impurities, higher carbon contents, andth|| :
"
7
-
;
]the decomposition of Fe,C.
L
_
0
uilibrium diagram for the metastable systeii
mposed the phase boundaries of the stabj|i as dotted lines.
M
I
3
1
Weight percent carbon Fig. 11-3
The stable iron-graphite system (dotted lines)
superimposed on the metastable iron-iron carbide system
,
5
6
6 67 ,
428
INTRODUCTION TO PHYSICAL METALLURGY
solves some additional cementite as it is heated to the annealing temperature.
Figure 11-3 shows that the austenite of the metastable system can :
dissolve more carbon than can austenite of the stable system. Therefore, a driving force exists for the carbon to precipitate out of the austenite as
t
;
free graphite. This graphitization starts at the malleableizing temperature. The initial precipitation of a graphite nucleus depletes the austenite of carbon, and so more is dissolved from the adjacent cementite, leading to further carbon deposition on the original graphite nucleus. The graphite nuclei grow at approximately equal rates in all directions and ultimately appear as irregular nodules or spheroids usually called temper carbon (Fig. 11-49). Temper carbon graphite is formed at the interface between primary, carbide and saturated austenite at the first-stage annealing temperature, with growth around the nuclei by a reaction involving diffusion and carbide decomposition. Nucleation and graphitization are accelerated by the presence of submicroscopic particles that can be introduced into the iron by the proper melting practice. High silicon and carbon contents promote nucleation and graphitization, but these elements must be restricted to certain maximum levels since the iron must solidify as white iron. Therefore, graphitizing nuclei are best provided by proper a nnealing prac-
i
tice. The rate of annealing d tendency, and temperature < annealing exerts considerab particles produced. Increasii
decomposition of primary a per unit area.
However, higl
excessive distortion of castin
ening operations after heat ti ed to provide maximum prac and are therefore controllec
casting is held at the first-si carbides have been decompc process, the casting must ht
large loads may require as n
f
6
first-stage graphitization coi throughout the matrix of saU After first-stage annealing, to about 1400 in preparatic ment. The fast cooling cycl( equipment used. In the second-stage anneal
5 to IftT/h through the critic r*r
'
fi,
ft
take place. During the slow is converted to graphite on 1
remaining austenite transfer plete, no further structural ch perature, and the structure c matrix (Fig. 11 b). This type The changes in microstructL schematically in Fig. 11-5. In the form of compact noo
m
continuity of the tough ferritin ductility than exhibited by gr to lubricate cutting tools, wjof malleable iron. Ferritic ma
Hi \
i?:vi
.
(a) Fig. 11-4 (a) Malleable iron, unetched. Irregular nodules of graphite called temper carbon. lOOx. {b) Ferritic malleable iron, temper carbon (black) in a ferrite matrix. Etched in 5 percent nital, 100x,
i
if
five, agricultural, and railro casting on bridges; chain-ho and many applications in gei Alloyed malleable irons arc tion of alloying elements noferritic malleable iron.
Since
malleableized, their influenc
principal kinds are copper-all alloyed malleable iron. Thee
'
rnr r
;
nViinMiiini
-.-
'
5
rgy
CAST IRON
ite as it is heated to the annealing temps
429
tice. The rate of annealing depends on chemical composition nucleation tendency, and temperature of annealing. The temperature of first-stage annealing exerts considerable influence on the number of temper-carbon particles produced, increasing annealing temperature accelerates the rate ,
austenite of the metastable systGm|B
i austenite of the stable system. Therefoj arbon to precipitate out of the austeniter
on starts at the malleableizing temperatui jraphite nucleus depletes the austenitil /ed from the adjacent cementite leadinggml ,
he original graphite nucleus. The grapjil||» equal rates in all directions and ulllmqjijS .V r spheroids usually called temper cart)(M '
:
"
>
'
jraphite is formed at the interface betweej austenite at the first-stage annealing WMm
the nuclei by a reaction involving diffusi6l|| jcleation and graphitization are acceleratwfl
copic particles that can be introduced IrtW
V practice. High silicon and carbon conterrtlg .:3
.
:
v izatlon, but these elements must be restrict ;since
the iron must solidify as white irorBl
%Bire best provided by proper annealing prsffiS
decomposition of primary carbide and produces more graphite particles per unit area. However, high first-stage annealing temperatures result in excessive distortion of castings during annealing and the need for straightening operations after heat treatment. Annealing temperatures are adjusted to provide maximum practical annealing rates and minimum distortion and aro thoreforo controlled botwoon 1650 and 1750oF,
The white-iron
easting is held at the first-stage annealing temperature until all massive carbides have been decomposed. Since graphitization is a relatively slow process, the casting must be soaked at temperature for at least 20 h, and
large loads may require as much as 72 h. The structure at completion of
firsl-stacjo graphltizntion consists of tompor-carbon nodules distributed throughout the matrix of saturated austenite. After first-stage annealing the castings are cooled as rapidly as practical to about 1400oF in preparation for the second stage of the annealing treatment. The last cooling cycle usually requires 2 to 6 h, depending on the equipment used. ,
In the second-stage annealing, the castings are cooled slowly at a rate of 5 to 1 ST/h through the critical range at which the eutectoid reaction v/ould take place. During the slow cooling, the carbon dissolved in the austenite is converted to graphite on the existing temper-carbon particles, and the remaining austenite transforms into ferrito.
Once graphitization is com-
plete, no further structural changes take place during cooling to room tem-
perature, and the structure consists of temper-carbon nodules in a ferrite matrix (Fig. 11 . Ab). This type is known as standard or lerritic malleable iron. The changes in microstructure during the malleableizing cycle are shown
m iFerriter F ritenr MkjC OJMS
schematically in Fig. 11-5. In the form of compact nodules the temper carbon does not break up the ,
continuity of the tough ferritic matrix. This results in a higher strength and ductility than exhibited by gray cast iron. The graphite nodules also serve to lubricate cutting tools which accounts for the very high machinabiiity of malleable iron. Ferritic malleable iron has been widely used for automotive, agricultural and railroad equipment; expansion joints and railing casting on bridges; chain-hoist assemblies industrial casters; pipe fittings; ,
si Wmmm
5
,
,
nodules Tr nodule;
and many applications in general hardware. Alloyed malleable irons are those whose properties result from the addition of alloying elements not normally present in significant quantities in ferritic malleable iron. Since those alloyed malleable irons a/ e completely
fitic mal-
malleableized
jtrix.
Etched
,
their influence is largely on the ferritic matrix.
The two
principal kinds are copper-alloyed malleable iron and copper-molybdenumalloyed malleable iron. The effect of copper is to increase corrosion resist-
.
1
430
INTRODUCTION TO PHYSICAL METALLURGY
3000 110 -o
25C0
r
135~ ;
2 130
r
Total corbon Silicon
2 40%
Manganese
0 31% 0 075% 0 144%
.
1 01 % .
,
Sulfur
1
m 125
.
Phosphorus
.
IQ00 i
lO
0
30
40
.
.
Percent copper
v
500
2,0
.
Fig. 11-6 0
Relationship between copper content anc
ness for a malleable kron of the composition shown. (Malleable Founders Society.)
.6
-
Fig. 11-5 The changesiin microstructure as a function of the malleableizing cycle-resulting in temper carbon in a
ferrite matrix'. (From "Malleable Iron Castings
"
,
Malleable
Founders Society, Cleveland Ohio, 1960.)
\
,
ance, tensile strength, and yield point at very slight reduction in ductility
of pearlite formed depends upon tl and the rate of cooling. High que (air blast) result in greater amount quench produces a fast enough c the matrix will be completely peai
Hardness is1 also increased, as shown in Fig. 11-6. The addition of copper and molybdenum in combination produces a malleable iron of superior corrosion resistance and mechanical properties. The mechanical properties of a copper-molybdenum-alloyed malleable iron are as follows: Tensile strength, psi Yield point, psi Elongation, % in 2 in. BHN
hi
m
58,000-65,000 40,000-45,000
15-20
\
wm
135-155
Compare these properties with those given for terrific malleable iron in Table 11-1.
0
ft
I
.
11-5 Pearlitic Malleable Iron
\\\
\
v
If a controlled quantity of carbon, in the order of >
3 to 0.9 percent, is retained as finely distributed iron carbide, an entirely
.
different set of mechanical properties results. The strength and hardness j of the castings will be increased over those of terrific malleable iron by an amount which is roughly proportional to the quantity of combined carbonremaining in the finished product. First-stage graphitization is a necessary prerequisite for all methods of
|
mm V
A
s1
i
Tempe corbon ,
manufacturing malleable-iron castings. If manganese is added, the regular \
cycle can bie maintained to retain combined cgrbon throughout the matrix, or the second-stage annealing of the normal process may be replaced by, a quench, usually air, which cools the casting3 through the eutectoid range;
fast enough to retain combined carbon throughout the matrix. The amount
;
1?;
Fig. 117 Pearlitic malleable iron. Nital etch, 500x (Malleable Founders Society.)
-m
v
-JY
CAST IRON
431
M0 X3
3
135. . .
H
r
Total carbon Silicon
130-
°
irrilc
Manganese Sulfur 125-
2 40% 1 01% 0 31% .
.
.
0 075% 0 144% .
Phosphorus
.
!2()
0 .
:v
t
5
-
60
.
1.0
2.0
3.0
4.0
fer'' Percent copper %0li-1"!'6 Relationship between copper content and hardffliftss for a malleable iron of the composition shown. (Malleable Founders Society.)
ction of n in a
ir
v
ilalleable
r
and the rate of cooling. High quench temperatures and fast cooling rates (air blast) result in greater amounts of retained carbon or pearlite. If the air quench produces a fast enough cooling rate through the eutectoid range,
r?-} point at very slight reduction in ductilityl
the matrix will be completely pearlitic (Fig. 11 -7).
hown in Fig. 11-6. The addition of coppeifl on produces a malleable iron of superior; anical properties. n
65,000
-
45.000
-
I*
The mechanical pn
-alloyed malleable iron are as followlf
-
of pearlite formed depends upon the temperature at which the quench starts
ii
V
m
mm
m
20 155
-
th those given for ferritic malleable iron I
3.
" .
|
jitrolled
4
quantity of carbon, in the order < s finely distributed iron carbide, an entire
1
[Derties results. The strength and hardnei a over those of ferritic malleable iron by fi rtional to the quantity of combined carbon?
5
i necessary prerequisite for all methods Of; astings If manganese is added, the regula
V i
V
i 1
is
1
Temperj
carbon 3
i m
4
.
.
.
V
jjin combined carbon throughout the matrikj j of the normal process may be replaced b
i
bis the castings through the eutectoid rang|| j carbon throughout the matrix. The amoulil
TIB-IW :
Pearlitic malleable iron. Nltal etch, 500x.
(Malleable Founders Society.)
; ;
432
INTRODUCTION TO PHYSICAL METALLURGY
If the cooling rate through the critical range is not quite fast enough to retain all the combined carbon
,
1
'
.
,
the more carbon will be dissolved'I |
'
cooling.
The higher the temperature
,
from the graphite nodules. Subsequent cooling will retain the combined
I
t j;
carbon and develop the desired properties. It is common practice to temper most pearlitic malleable irons after air ; t cooling. Those having coarse pearlitic structures are tempered at relatively ;? f '
high temperatures (between 1200 and 1300oF) to spheroidize the pearlite [ ,
(Fig, 11-9), improve machinability and toughness, and lower the hardness. £| If it is desired to increase the mechanical properties of the matrix, it is
necessary to reheat for 15 to 30 min at 1550 to 1600oF to re-austenitize and j
homogenize the matrix material. The castings are then quenched in heated
I
and agitated oil, which develops a matrix of martensite and bainite with a 1 r hardness of Rockwell C 55 to 60.
tmMmmi
The amount of martensite formed will
i liRg. 119 Microstructure of a pearlitic malleablf :;
f!
'
;: tempered to obtain a spheroidite matrix. Nital e
5 I (Malleable Founders Society.) r
I
depend upon the quenching si
ture from which the work is q thickness of the casting, and malleable iron is then temper
h4 W
desired properties. The matri>
tempered martensite, dependii Welding of pearlitic malleab
the formation of a brittle, lov A',
t5*k
i
f
Fe rr il
\
bead, caused by the melting a
\
pearlitic structure adjacent to altered through the redissolvir The tensile properties of fen Table. 11-1.
Temper carbon
PI
m
f '-
the areas surrounding the temper-carbon \ nodules will be completely graphitized while those at greater distance' I from the nodules will be pearlitic (Fig. 11-8). Because of its general ap-i pearance, this is referred to as a £itv// s-eye structure | h A fully ferritic malleable iron may be converted into pearlitic malleable\ I iron by reheating above the lower critical temperature followed by rapid! f ,
;
TABLE 11-1
Tensile Properties of Mi TENSILE STRENGTH,
Fig. 11-8 Typical appearance of a "bull's-eye" structure. Temper carbon nodules surrounded by ferritic areas (white),iwith lamellar pearlite (dark) located between the bull's eyes. Nital etch, 100X. (Malleable Founders Society.)
1 \ l
I ;
TYPE
1 000 PSI
Ferritic
50-60
Pearlitic
65-120
,
S3Y
CAST IRON
433
i critical range is not quite fast enough
jhe areas surrounding the temper-carb6|| '
S jhitized while those at greater distan i ,
c (Fig. 11-8). Because of its general a] '
bull s-eye structure
.
nay be converted into pearlitic malleabf Jer critical temperature, followed by rapl; .
ature, the more carbon will be dissolvoi
sequent cooling will retain the combined properties.
V vijer most pearlitic malleable irons after alr| ..
iarlitic structures are tempered at relatively
.
|o and 1300oF>. to spheroidize the pearlitl iy and toughness, and lower the hardness mechanical properties of the matrix, it Is ;
Xnin at 1550 to 1600oF to re-austenitize and
The castings are then quenched in heated
I immmmm
a matrix of martensite and bainite with a The amount of martensite formed will
0
.
11-9 Microslruclure ol a pearlitic malleable iron tempered to obtain a spheroldite matrix. Nital etch, 500x
-
(Malleable Founders Society.)
depend upon the quenching speed of the particular oil used the temperature from which the work is quenched, the time at that temperature, the thickness of the casting, and the chemistry of the iron. The martensitic ,
malleable iron is then tempered between 'ISO and 1320"F to develop the
desired properties. The matrix microstrucluie consists of various types of tempered martensite, depending on the final hardness of the castings. Welding of pearlitic malleable iron is seldom recommended because of the formation of a brittle low-strength white-iron layer under the weld bead, caused by the melting and rapid freezing of the malleable iron. The ,
pearlitic structure ndjacont to the white iron in the welding zone also is
altered through the redissolving of some temper carbon
.
The tensile properties of ferritic and pearlitic malleable iron are given in i
Table 11-1.
.
TABLE 111
Tensile Properties of Malleable Cast Iron TENSILE STRENGTH
YIELD STRENGTH
1 000 PSI
1 000 PSI
% IN 2 IN.
BHN
Ferritic
50-60
32-39
20-10
110-145
Pearlitic
65-120
45-100
16-2
163-269
,
TYPE
"
.s-eye structure. erritic nroas
ied between the | Founders Society.)
,
,
.
ELONGATION
,
I..
.I
-.
1
434
INTRODUQ ION TO PHYSICAL METALLURGY
Alloyed pearlitic malleable castings are made from white irons that con- i tain one or more alloying elements so that the regular malleableizing anneal
will not result in a ferritic matrix. The alloy additions usually do not affect
i
first-stage graphitization but serve as carbide stabilizers during the eutectoid range or subeutectoid tempering treatments. Many of the alloying elements also increase hardenability and strengthen the matrix. Manganese and sulfur may be added in quantities not normally found in standard malleable iron. Copper may be added to improve strength, corrosion resistance, and graphite distribution. Suitably alloyed pearlitic malleable
fig. 11-11
Space models of flake graphite. (Afl
Mackenzie.)
iron may be fully martensitic in sections as heavy as 2 in. after air quenching from 1600'F, Some of the industrial applications of pearlitic malleable iron are for
These alloys solidify by first pearance of combined carbon
*
axle and differential housings, camshafts and crankshafts in automobiles:
tic reaction at 2065oF. The gra content, high temperature and notably silicon.
for gears, chain links, sprockets, and elevator brackets in conveyor equipment; for rolls, pumps, nozzles, cams, and rocker arms as machine parts, for gun mounts, tank parts, and pistol parts in ordinance; and finally for a
,
position and cooling rates. Most gray cast irons are hypoeutectic alloys §
There is experimental evidei factors, the alloy will follow th (Fig. 11 -3), forming austenite i 2075oF. At any rate any cemer The graphite appears as man\ plates which give gray cast iro
containing between 2.5 and 4 percent carbon.
ture (Fig. 11-10).
variety of small tools such as wrenches, hammers, clamps, and shears.
11-6 Gray Cast Iron
This group is one of the most widely used alloys of iron. In
the manufacture of gray cast irons, the tendency of cementite to separate into graphite and austenite or ferrite is favored by controlling alloy com- i
,
It should be
represents their appearance dimensional particles. They a nected, and may be represente
During continued cooling, th cause of the decrease in solut
precipitated as graphite or a
m
graphitizes. 5:
:
-
.
V
The strength of gray cast ire which the graphite is embedde
.
-
.
it ii
the matrix may be varied fron ferrite in different proportions,
r
ite-ferrite mixture is the softe;
l '
1
hardness increase with the inc1 mum with the pearlitic gray iro gray cast iron with the matrix ;
i
| I t17 Silicon in Cast Iron i Fig. 11-10 Graphite flakes in gray cast iron. Unetched, IGOx. .1
1
:
m .
..
.
are such that the eutectoid cet
be entirely ferritic. On the othe mentite is prevented, the matrix
1
'
condition of the eutectoid cerr
of gray iron.
r
m
Silicon is
It increases flui
molten alloy. The eutectic cor
IRGY
:
:
435
astings are made from white irons that nts so that the regular malleableizing am
-j
.
:
CAST IRON
vix. The alloy additions usually do not al erve as carbide stabilizers during the eutt impering treatments. Many of the alio;
liability and strengthen the matrix. Marl! iin quantities not normally found in stand; e added to Improve strength, corrosion
ution.
.
11-11 Space models of flake graphite. (After ienzie.)
Suitably alloyed pearlitic malleab)
} sections as heavy as 2 in. after air quencl v
.
.
.
j
d pistol parts in ordinance; and finally fori
These alloys solidify by first forming primary austenite. The initial appearance of combined carbon is in the cementite resulting from the eutectic reaction at 2065oF<. The graphitization process is aided by high carbon content, high temperature, and the proper amount of graphitizing elements, notably silicon. There is experimental evidence that with proper c&ntrol of the above
vrenches, hammers, clamps, and shears.
factors, the alloy will follow the stable iron-graphite equilibrium diagram
Judications of pearlitic malleable iron
are f
camshafts and crankshafts in automoblli
ts, and elevator brackets in conveyor equlj 3,
.
cams, and rocker arms as machine pai
% vine of the most widely used alloys of iron. | rons, the tendency of cementite to se ferrite is favored by controlling alloy cOl
lost gray cast irons are hypoeutectic all< percent carbon.
,
(Fig. 11-3), forming austenite and graphite at the eutectic temperature of 2075oF. At any rate, any cementite which is formed will graphitize rapidly. The graphite appears as many irregular generally elongated and curved ,
plates which give gray cast iron its characteristic grayish or blackish fracture (Fig. 11-10). It should be emphasized that while the microstructure represents their appearance on a plane surface, the flakes are threedimensional particles. They are, in effect, curved plates sometimes con-
nected, and may be represented by the space models shown in Fig. 11-11. During continued cooling, there is additional precipitation of carbon because of the decrease in solubility of carbon in austenite. This carbon is
precipitated as graphite or as proeutectoid cementite which promptly
1r.
graphitizes.
The strength of gray cast iron depends almost entirely on the matrix in which the graphite is embedded. This matrix is largely determined by the condition of the eutectoid cementite. If the composition and cooling rate are such that the eutectoid cementite also graphitizes then the matrix vili be entirely ferritic. On the other hand if graphitization of the eutectoid cementite is prevented, the matrix will be entirely pearlitic. The constitution of the matrix may be varied from pearlite through mixtures of pearlite and ,
/
,
,
ferrite in different proportions down to practically pure ferrite. The graphite-ferrite mixture is the softest and weakest gray iron; the strength and hardness increase with the increase in combined carbon reaching a maxi,
i w
,
mum with the pearlitic gray iron. Figure 11-12 shows the microstructure of gray cast iron with the matrix almost entirely pearlitic.
ill-7 Silicon in Cast Iron Silicon is a very important element in the metallurgy etched, 100x.
of gray iron. It increases fluidity and influences the solidification of the molten alloy. The eutectic composition is shifted to the left approximately
;7:V
4
i
p
.
436
INTRODUCTION TO PHYSICAL METALLURGY
1>
-
.
.
'
. .
>
1
.
.
Fernte
0
10
r
.
-
1
-
1
I [\
1
II
White \- PearliMc
L3
f
-
cast iron\ casl ifon
4}'
l C .
ZO
0 -
V
1.0
5.0
Sihcon percMT ,
flakes
I Fig. 11-13 Relation of structure to carbon anc content of cast iron. (After Maurer.)
!
.
JO
.
ophite
DearlitG
-
0* -
Silicon is a graphitizer
:
,
and
elements, it favors solidificatk tem.
it
f
.
.
.
,
,
>5 .
Therefore
during solidi precipitated as primary graphi Ite has formed its shape cai
4
weak graphite flakes that bre notch effect at the end of the
I
.
V
and low ductility of gray iron. The relation of the carbon ar
ill
tions of cast iron is shown in F
i!
the structure will be white cas A
J
'
cause graphitization of all the
..
.
The microstructure will consis
pearlite, as in Fig. 11-12. In r complete dissociation of the result in a ferritic gray cast ire
C
.
9i r
i 0
-
in
Fig. 11-12
'
,
5
u
r ft,
4
Microstructure of gray cast iron. Graphite
flakes in a pearlitic matrix with a small amount of ferriie
(white areas). JEtched in 2 percent nital. (a) 100x; (h) 500> 1
.
30,:percent carbon for each 1 percent silicon, which effectively depresses the temperature at which the alloy begins to solidify. As the silicon content
0
.
is inicreased, the austenite field decreases in area, the eutectoid carbon-
\
2
S
.
i.cc"
3 .
,
pe"~e"'
content is lowered, and the eutectoid transformation occurs over a broad-
Fig. 11 14 Relation of tensile strength to carbi content of oast iron. (Coyle, Trans. ASM, vol. 1
ening range.
1927.)
,
1
t 1
1 .
is
5 is
[ :. JRGY
CAST IRON
4
.
Ei.O i
437
r
4 3 40 .
.
2 r
30 .
-
.
g n 20 .
White
Pearlitic
cast
cast iron
iro n
Ferntic ast iron
10 .
0
1.0
2.0
Graphite! flakes
3.0 1.0 5.0 Silicon, iierceii
60 .
70 .
01'13 Relation of structure to carbon and silicon sent of cast iron. (After Maurer.)
Silicon is a graphitizer and if not counterbalanced by carbide-promoting elements, it favors solidification according to the stable iron-graphite sys-
r -
,
tem. V »
Therefore
during solidification in the presence of silicon, carbon is
,
precipitated as primary graphite in the form of flakes. Once primary graphite has formed its shape cannot be altered by any method. It is these weak graphite flakes that break up the continuity of the matrix and the notch effect at the end of these flakes that accounts for the low strength
.
,
V
and low ductility of gray iron. The relation of the carbon and silicon content to the structure of thin sec4
tions of cast iron is shown in Fig. 11-13. In region I, cementite is stable, so the structure will be white cast iron. In region II there is enough silicon to cause graphitization of all the iron carbide except the eutectoid cementite. ,
ML ass if
The microstructure will consist of graphite flakes in a matrix that is largely pearlite, as in Fig. 11-12. In region III, the large amount of silicon causes complete dissociation of the cementite to graphite and ferrite. This will result in a ferritic gray cast iron of very low strength.
r
'
V ..
ft
& "
V i
1
i
Mr 3
00 35
fs
.
3raphite ir--.r. i[ of ferrite -
.
.
?o
1
o
Pao
a
e 30 .
So
)0x: (6) 500x 9>
1 percent silicon, which effectively depresses alloy begins to solidify. As the silicon conteflfi
eld decreases in area, the eutectoid carbon utectoid transformation occurs over a broad-J
2
3
Silicon, percent
11-14
Relation of tensile strength to carbon and silicon
content of cast iron. (Coyle, Trans. ASM vol. 12, p. 446, 1927.) ,
V
438
INTRODUCTION TO PHYSICAL METALLURGY
i
1
The influence of carbon and silicon content on the tensile strength is
,
shown in Fig. 11-14. The highest tensile strength is obtained with a carbon
sj
content of 2.75 percent and a silicon content of 1.5 percent. These peK; C centages in Fig. 11-13 will result in a pearlltic gray cast iron. If the percent- - | ages of carbon and silicon are such as to yield eitheir a white cast iron era | ,
ferritic gray cast iron the tensile strength will be low. < | Careful control of the silicon content and cooling rate is required to graphitize the eutectic and proeutectoid cementite, but not the eutectoid J: ,
cementite, in order to end up with a pearlitic gray iron of high strength. 11-8 Sulfur in Cast Iron Most commercial gray irons contain between 0.06 and.
'
'
J?
;
0
0
12 percent sulfur. The effect of sulfur on the form of carbon is the reverse of silicon. The higher the sulfur content, the greater will be the amount of combined carbon, thus tending to produce a hard, brittle white iron. Aside from producing combined carbon, sulfur tends to react with iron to form iron sulfide (FeS). This low-melting compound, present as thin inter- Bdendritic layers, increases the possibility of cracking at elevated tempera-.*
Si
.
v
.
'
tures (red-short). High sulfur tends to reduce fluidity and often is responsible for the presence of blowholes (trapped air) in castings. f1 .
'
Fortunately, manganese has a greater affinity for sulfur than iron, forming \manganese sulfide (MnS).
IV .i
The manganese sulfide particles appear as: tv Fig. 11-15 Gray iron showing steadite areas (arro\
small, widely dispersed inclusions which do not impair the properties of the - f JJ J f percent nital' 500x casting. It is common commercial practice to use a manganese content of
c
two to three times the sulfur content.
p
(The lntemational
11-9 Manganese in Cast Iron Manganese is a carbide stabilizer, tending to: ? increase the amount of combined carbon, but it is much less potent than sulfur. If manganese is present in the correct amount to form manganese sulfide, its effect is to reduce the proportion of combined carbon by re-
moving the effect of sulfur. Excess manganese has little effect on solidili;]l|
cation and only weakly retards primary graphitization. On eutectoid graphi-|
tization,:however, manganese is strongly carbide-stabilizing. 11-10 Phosphorus in Cast Iron Most gray irons contain between 0.10 and 0.9D; percent phosphorus originating from the iron ore. Most of the phosphorus.; combines with the iron to form iron phosphide (Fe3P). This iron phosphide '
forms a ternary eutectic with cementite and austenite (pearlite at room
and phosphorus content is low. where a less fluid iron may not t If the silicon, sulfur mangan trolled at proper levels, the only of a pearlitic gray iron is the gn soft and weak, its size, shape, chanical properties of the cast i graphite flakes and the increase for the improvement in the quali ,
temperature). The ternary eutectic is known as steadite and is a normal |: iin Heat Treatment of Gray Iron i V
:
.
;:::
' .
:
-
i
.
feature in the microstructure of cast irons (Fig. 11T5). Steadite is relatively-; f brittle, and with high phosphorus content, the steadite areas tend to form, a continuous network outlining the primary austenite dendrites. The con-. dition reduces toughness and makes the cast iron brittle, so that the phosr, phorus content must be carefully controlled to obtain optimum mechanical properties.
Phosphorus increases fluidity and extends the range of eutectic freezing,;; thus increasing primary graphitization when the silicon content is high:
quently applied heat treatment fo
dition usually contains residual: ferent rates throughout various S( stresses may reduce strength, ca even result in cracking. The terr below the transformation range i lief of stress with minimum decoi
of 1000 to 1050oF is desirable.
I
m
\
-
1
I
i
'
A
JRGY
CAST IRON
439
d silicon content on the tensile streng
Spjast tensile strength is obtained with a cai
,
"
silicon content of 1.5 percent. These
'
.
jit in a pearlitic gray cast iron. If the perce
i such as to yield either a white cast irpttr
iile strength will be low.
- M
jon content and cooling rate is required
,
joeutectoid cementite, but not the eutectoiS
b with a pearlitic gray iron of high streng||
...
inercial gray irons contain between 0.06 a
..
iM'jt of sulfur on the form of carbon is the revlL fur content, the greater will be the amount? '
ig to produce s hard, brittle white iron, lined carbon, sulfur tends to react with iroml iow-melting compound, present as thin inter
4 possibility of cracking at elevated tempert"
'
"
;
tends to reduce fluidity and often is resp Hvhoies (trapped air) in castings, iji s a greater affinity for sulfur than iron, forrtit
.
.
4
.
iV-SdThe manganese sulfide particles appeai j lions which do not impair the properties of th ercial practice to use a manganese content*
i 11-15 Gray iron showing steadite areas (arrow)
.
Shed in 2 percent nital 500x. (the International Nickel ,
Jmpany.)
content.
anganese is a carbide stabilizer, tending
'
bined carbon, but it is much less potent t
.
snt in the correct amount to form mangan1
and phosphorus content is low. It is therefore useful in very thin castings
ie the proportion of combined carbon by f
where a less fluid iron may not take a perfect impression of the mold If the silicon sulfur, manganese, and phosphorus contents are controlled at proper levels the only remaining variable affecting the strength .
Excess manganese has little effect on solid! > s primary graphitization. On eutectoid grapf
,
.
.
e is strongly carbide-stabilizing.
,
M
of a pearlitic gray iron is the graphite flakes. Since graphite is extremely
lost gray irons contain between 0.10 and Q| ng from the iron ore. Most of the phosphor
soft and weak, its size shape, and distribution will determine the me,
chanical properties of the cast iron It is the reduction of the size of the graphite flakes and the increase in their distribution that have accounted .
m iron phosphide (FeaP). This iron phosphil
:
h cementite and austenite (pearlite at rool of cast irons (Fig. 11 -15). Steadite is relativ
norus content, the steadite areas tend to for
ing the primary austenite dendrites. The cilr id makes the cast iron brittle, so that the phfi
for the improvement in the quality of gray cast iron Heat Treatment of Gray Iron Stress relieving is probably the most frequently applied heat treatment for gray irons. Gray iron in the as-cast condition usually contains residual stresses because cooling proceeds at different rates throughout various sections of a casting The resultant residual stresses may reduce strength cause distortion, and in some extreme cases .
sutectic is known as steadite and is a norrri|
1-11
I
'
,
.
,
jfully controlled to obtain optimum mechanic
dity and extends the range of eutectic freeziri| aphitization when the silicon content is hi(j 4
mi
even result in cracking. The temperature of stress relieving Is usually well below the transformation range of pearlite to austenite For maximum relief of stress with minimum decomposition of carbide a temperature range of 1000 to 1050oF is desirable. Figure 11-16 indicates that from 75 to 85 .
,
I
440
INTRODUCTION TO PHYSICAL METALLURGY
with flame or induction hardenir
As-quenched gray iron is britt strength and toughness but de
TOOT is required before the im|:
1
-
-
After tempering at 700oF for r 1 '
-
matrix is still about Rockwell C
tempering temperature of 300 t is equivalent to Rockwell C 55 1 commercially to increase the s strength of the as-cast metal ca
Fig. 11-16 Effect of stress-relieving temperature and time on residual stress in gray iron. (After G. N. J. Gilbert, from Metals Handbook," vol. 2, American Society for Metals. Metals Park, Ohio: 1964.)
silicon and total carbon content
"
is usually quenched and tempe abrasion by increasing the har embedded in a hard martensitic
percent of the residual stress can be removed on holding for 1 h in this range. When almost complete stress relief (over 85 percent) is required, a minimum
process can replace the white-ir
can be applied where chilling is
temperature of 1 lOOT can be employed. Annealing of gray iron consists of heating it to a temperature that is high enough;to soften it and thus improve the machlnability. For most gray Irons
large castings. The combinatior lubricant results in a surface wi
tions such as farm implement g
an annealing temperature between 1300 and 1400 F is recommended. Up | °
automotive camshafts. Thus he ,
to approximately 1100oF the effect of temperature on the structure of gray iron is insignificant. As the temperature increases above 1100oF, the rate at which iron carbide decomposes to ferrite plus graphite increases markedly, reaching a maximum at about 1400oF for unalloyed or low-alloy iron. ,
i !
The casting must be held at temperature long enough to allow the graphi- \ tizing process to go to completion. At temperatures below 1300UF, an excessively long holding time is usually required. Gray iron is normalized by being treated to a temperature above the trans-
:
are therefore generally preferrei Graphite-flake sizes are usual sizes prepared jointly by the AFJ
ASTM (American Society for Tes
formation range, held at this temperature for a period of about 1 h/in. of maximum section thickness, and cooled In still air to room temperature.
of gray iron as an engineering r 1112 Size and Distribution of Graphi interrupt the continuity of the pe and ductility of the gray iron. S
tion and measurement of flake ; 1
1971 Book of ASTM Standards
,
The temperature range for normalizing gray Iron is approximately 1625 to 1700°F. Normalizing may be used to enhance mechanical properties, such as hardness and tensile strength, or to restore as-cast properties that have been modified by another heat treating process, such as graphitizing orthe preheating and postheating associated with repair welding. Gray iron, like steel, can be hardened when cooled rapidly or quen'cheij from a. suitable elevated temperature. The quenched iron may be tempered by reheating in the range from 300 to 1200oF to increase toughness and
' -
.-vj
relieve stresses. Ordinarily gray iron is furnace-hardened from a tempera-} ture of
,
;
1575 to 1600oF. The quenching medium may be water, oil, hot salt,
or air, depending on the composition and section size.
Oil is the usual 1
J
,
lengths of the largest graphite iron at 100x. Numbers are assic
; TABLE 11 2
Graphite Flake Sizi
AFS-ASTM
L
FLAKE SIZE NUMBER 1
2 3 4
5
quenching medium for through-hardening. Quenching in water maybe
6
too drastic and may cause cracking and distortion unless the castings arei
7
massive and uniform in cross section. Water is often used for quenchin|
8
i
.
li .
I
ii
i
J1' IRGY
m
J iron '
-;
.
CAST IRON
.
.
.
|; strength and toughness but decreases hardness. A temperature of about "
700oF is required before the impact strength is restored to the as-cast level After tempering at 700oF for maximum toughness the hardness of the matrix is still about Rockwell C 50. Where toughness is not required and a .
78 095 -
,
,
.
.
flame or induction hardening where only the outer surface is hardened
pAs-quenched gray iron is brittle. Tempering after quenching improves
I
37% 1,56 .
.
'
S With
441
98
*
tempering temperature of 300 to 500oF is acceptable the matrix hardness IS equivalent to Rockwell C 55 to 60 Heat treatment is, not ordinarily uaed cdfhmercially to increase the strength of gray-iron castings because the strength of the as-cast metal can be increased at less cost by reducing the silicon and total carbon contents or by adding alloying elements. Gray iroh is usually quenched and tempered to increase the resistance to wear and ,
r
' .
:; i
.
r and time
,
;
bert, from Metals,
n
t
I n be removed on holding for 1 h in this ranga ilief (over 85 percent) is required, a minimum
. § WMts of heating it to a temperature that is hlgli|
,
large castings. The combination of high matrix hardness and graphite as a lubricant results in a surface with good wear resistance for some applica-
"
'
tions such as farm implement gears sprockets, diesel cylinder liners, and ,
V/een 1300 and MOOT is recommended. U| "
automotive camshafts. Thus
feet of temperature on the structure of gr8l| mperature increases above 1100oF
,
the rati
,
mperature long enough to allow the grapM ,
.
an
ing treated to a temperature above the tranjj!
temperature for a period of about 1 h/in. mM
r nd cooled in still air to room temperatur
.
,
1971 Book of ASTM Standards Part 31. The measurement is made of the ,
'"
. Vernalizing gray iron is approximately 1625! .
lengths of the largest graphite flakes in a unetched section of the gray iron at 100X. Numbers are assigned as indicated in Table 11-2.
f sed to enhance mechanical properties, sue!
th, or to restore as-cast properties that hav(| treating process, such as graphitlzing orth7.,
'
?
TABLE 112
isociated with repair welding. J hardened when cooled rapidly or quenchetf| '
: ira
"
:
-
4n
v-ay
FLAKE SIZE NUMBER
ture. The quenched iron may be temperedl
300 to 1200oF to increase toughness andj iron is furnace-hardened from a temper
"'
iposition and section size. Oil is the us gh-hardening. Quenching in water may W
acking and distortion unless the castings ar||
! section. Water is often used for quenchin
Graphite Flake Sizes
AFS-ASTM
iv-iienching medium may be water, oil, hotsalti
m
Large graphite flakes seriously
interrupt the continuity of the pearlitic matrix thereby reducing the strength and ductility of the gray iron. Small graphite flakes are less damaging and are therefore generally preferred. Graphite-flake sizes are usually determined by comparison with standard sizes prepared jointly by the AFS (American Foundrymen's Society) and the ASTM (American Society for Testing Materials). The procedure for preparation and measurement of flake size is given in ASTM Designation A247-67
bout 1400oF for unalloyed or low-alloy irom
tion. At temperatures below 1300oF jsually required.
,
heat treatment extends the field of application
of gray iron as an engineering material. 12 Size and Distribution of Graphite Flakes
)ses to ferrite plus graphite increases matl| .
.
process can replace the white-iron surface usually produced by chilling. It can be applied where chilling is not feasible as with complicated shapes or
Employed.
jjrove the ma chinability. For most gray iro
abrasion by increasing the hardness. A structure consisting of graphite embedded in a hard martensitic matrix is produced by heat treatment This
fi-
IN,
MM
1
4 or more
2
2-4
64
3
1-2
32
4
Va-I
16
5
8
7
V4-V2 ' /a-V-i V16- Va
8
Vii or less
1
6
ll
LENGTH OF LONGEST FLAKES AT lOOx
128
4
2
442
INTRODUCTION TO PHYSICAL METALLURGY
Size 1 i Longest flakes 4 in. or more in length.
Size 2
Longest flakes 2 to 4 in. in length.
Size 5
Longest flokes ViteVi in. in A
.
V "
f( i'x' -r
'1 y ~?
i
i
t
sr. -~J
/
/v
l- i
Y
-
L Siie'3 v
-
.
Longest flakes 1 to 2 in. in length.
Size 4
Longest flakes Yi to 1 in. in length
Size 7 Longest flakes Vm to Vt in. in
,
J
.-. i
i
'
-
'
.
.
i /
-v .-:
i
Fig. 11-17 Graphite-fiake size chart illustrated by typical fields showing as nearly as possible the sizes represented. (Prepared jointly by ASTM and AFS.)
thus reducing flake size.
The flake lengths are illustrated in Fig. 11 -17 by typical fields showing as nearMas possible the various sizes.
Slow cooling of hypoeutectic irons to favor graphitization also produces 1
i
Hov strong graphitizing influence, in a weak casting. The best method of reducir
the graphite flakes seems to be
large crystals of primary austenite. This restricts the eutectic mixture or
known as an/nocu/anf. Inocul
graphite to the grain boundaries and results in graphite flakes that are relatively few in number and coarse.
are metallic calcium aluminu
Increasing the carbon content to increase the amount of eutectic also increases the amount of graphite formed. This may weaken the cast iron more than a smaller flake size can strengthen it. Increasing the silicon content increases the amount of eutectic formed
they operate is not clearly unc tion of primary austenite rest and improves the distribution The way in which the graph
,
,
cium silicide, or combination
,
wis
i
v
I
m
mm®. CAST IRON
443
v
th. Size 2 Longest flakes 2 to 4 in. in length.
Size 5
Longest flakes
to K in. in length.
Size 6
Longest flakes % to
in. in length.
\
7 1
1
n Size 4 Longest flakes Vt to 1 In. in length.
Size 7
Longest flakes
fo
Size 8 Longest flakes K« in. or less in length.
in. in length.
wmmmifm
wmmm
mmmmmM
IS
mi
wmmm
IIIHI
V
/
by typical jpresented.
thus reducing flake size. However since high silicon content has such a ,
6;
-
: :>
.
.
ated in Fig. 11-17 by typical .3 SizeS.
fields showing ai|H| p-
WM irons to favor graphitization also produce«| enite. This restricts the
eutectic mixture..«||
4ies and results in graphite flakes that are rej p' .
arst!. arse.
«
ent to increase the amount of eutectic also in* !e formed. This may weaken the cast H-on morel
||
strengthen it. ent increases the amount of eutectic formed
f -
strong graphitizing influence, the matrix will probably be terrific, resulting in a weak casting. The best method of reducing the size and improving the distribution of the graphite flakes seems to be by the addition of a small amount of material known as an inoculant. Inoculating agents that have been used successfully are metallic calcium aluminum, titanium, zirconium, silicon carbide, cal,
cium silicide
or combinations of these. The exact mechaniS|Ti by which they operate is not clearly understood. They probably promote the nuclea,
tion of primary austenite resulting in small grains, which reduces the size and improves the distribution of the graphite flakes. The way in which the graphite flakes are arranged in the microstructure ,
r
'
i
444
INTRODUCTION TO PHYSICAL METALLURGY
11-20). The cooling rate in thi graphitization. The few large, straight graph that the iron is hypereutectic ir
of gray cast iron is usually indicated as one or more types that have been jointly prepared by the AFS and the ASTM. The five flake types are shown in
Fig. 11-18. Type D and type E flake patternsjjsually result from the graphitization of
?
a normal ,eutectic structure. These types appear in irons of very high purity
f
i
or in commercial irons that have been cooled rather rapidly during solidification. Although the graphitic flake size is small, the interdendritic pattern and high graphite content weaken the material. Therefore, types D and E _ i flake patterns are undesirable in gray irons. When the cooling rate is slower \ most commercial gray irons show complete divorcement of the eutectic so that types D and E flake patterns do not occur. The most desirable flake pattern in gray iron is represented by the uniform distribution and random orientation of type A. This results from a completely divorced eutectic structure. As was mentioned earlier, the size of the individual graphite flakes is determined by the size of the austenite
alloying elements reduce the c< present in sufficient amounts t below 3.5 percent carbon.
11-13 Mechanical Properties and Af
,
portant classification of gray it employed in the ASTM Spec classed in seven classes (Nos.: mum tensile strength of test bi For example, class 20 gray in of 20,000 psi; class 30, 30,000 chanical properties of standan
,
crystals around which they form.
Tensile strength is importar
The rosette pattern of type B graphite flakes is common only in the intermedipte region of a chilled cast iron. This region is known as the mottled region and consists of a mixture of gray and white cast iron (Fig.
subjected to static loads in ind pressure vessels, housings, va
psi in tensile strength are usu
i
somewhat more expensive to f irons do not exhibit a well-defi stress-strain curve does not s
modulus of elasticity cannot b
5 fe vj -
Type A
"
Type B
,
/
mine the
y
or the
"
"
relative
modulus at
"
tangent" modulus by d The percent elongation is sme
percent, and the reduction of ai Compressive strength is im
l Type C
chinery foundations or suppori
strength of gray iron is much g a function of the shearing strei
i i ,
J V-t TypeD '
Fig. 11-18
Graphite-flake types.
Type E
Type A-uniform distribu-
tion, random orientation; type B-rosette groupings, random orientation; type C-superimposed flake sizes, random orientation; type D-interdendritic segregation, random orientation; type E-interdendritic segregation, preferred orientation. (Prepared jointly by ASTM and AFS.)
"-i
'
along an oblique plane unless by buckling. Many grades of gray iron hav grades of steel. This character
gray iron a suitable material fc The hardness of gray iron is iron and the metallic matrix. will cause wide variations in
The Brinell tester, covering a hardness value than the Rock
Figure 11-19 shows the ge
i
'
i
V
GY
;
CAST IRON
445
)ated as one or more types that have bi
11-20). The cooling rate in this region is the maximum that would permit
the ASTM. The five flake types are shown!
graphitization. The few large straight graphite flakes present in type C always indicate that the iron is hypereutectic in carbon content. Silicon and several other alloying elements reduce the carbon content of the eutectic and if they are ,
ns usually result from the graphitizationlt jse types appear in irons of very high purii s been cooled
,
rather rapidly during soiidifl*
present in sufficient amounts the eutectic composition may be reduced to below 3.5 percent carbon.
iake size is small, the interdendritic pattern) en the material. Therefore, types D and
;
ft 13 Mechanical Properties and Applications of Gray Cast Iron The most im-
gray irons. When the cooling rateisslowefi '
portant classification of gray irons, from an engineering standpoint, is that
;ow complete divorcement of the eutectic;
.
JSaerns do not occur.
employed in the ASTM Specification A48. The gray-iron castings are classed in seven Classes (Nos. 20, 25, 30,35,40,50,60) which give the minimum tensile strength of test bars in thousands of pounds per square inch. For example, class 20 gray iron would have a minimum tensile strength of 20,000 psi; class 30, 30,000 psi; and so on. Table 11 -3 gives typical mechanical properties of standard gray-iron test bars, as cast. Tensile strength is important in selecting a gray iron for parts that are subjected to static loads in indirect tension or bending. Such parts include
-j
rn in gray iron is represented by the unifofi ation of type A. This results from a coi ure.
As was mentioned earlier, the size
is determined by the size of the austenite* | .
M
B graphite flakes is common only in thO id cast iron. This region is known as the
::,
r;;i a mixture of gray and white cast iron (Fig)
iff.
pressure vessels, housings, valves, fittings, and levers, irons above 40,000 psi in tensile strength are usually considered high-strength irons and are somewhat more expensive to produce and more difficult to machine. Gray irons do not exhibit a well-defined yield point as do most mild steels. The stress-strain curve does not show a straight-line portion; thus a definite modulus of elasticity cannot be determined. Usual methods are to determine the relative modulus at 25 percent of the expected tensile strength, "
or the
"
"
tangent" modulus by drawing a tanqent at some given stress value.
The percent elongation is small for all cast irons, rarely exceeding 3 to 4
Type E
,
percent, and the reduction of area is too slight to be appreciable. Compressive strength is important when the gray iron is used forjna-
Type C
v
Type E ' '
.:::\
'
:
orm distribu-
: / :\Dings, .
random
flake sizes, random
segregation, random Segregation, preferred >TM and AFS.)
chinery foundations or supports. Like all brittle materials, the compressive strength of gray iron is much greater than its tensile strength and is largely a function of the shearing strength. Failure in compression usually occurs along an oblique plane unless the specimen is long enough to allow failure by buckling. Many grades of gray iron have higher torsional shear strength than some grades of steel. This characteristic, along with low notch sensitivity, makes
gray iron a suitable material for various types of shafting. The hardness of gray iron is an average result of the soft graphite in the iron and the metallic matrix
.
Variation in graphite size and distribution
will cause wide variations in hardness (particularly RockweH hardness)
.
The Brinell tester covering a larger area, tends to give a more accurate hardness value than the Rockwell tester ,
.
Figure 11-19 shows the general correlation between Brinell hardness
446
INTRODUCTION TO PHYSICAL METALLURGY
i CD 10
X m
m
-
t-
t
t- C\J o t-
IT) C\J co co
C\J o
OJ
CM
CO
C\J
C\J
5
-
2
-
v
-
X
h
g- 12G
-
o
x
H CD i!
111 < -I
w 5 z
x 9 5
z
T CO
Z
m
"T
£o2
i !
.
z
o en in o in m oo m r--
o o o o
co
co
in cq t-
160
Fig 11-19 I
I'
£ Q O |-' X
o T-
O
CVj
z o
co o cd O) co o
in
CO
in
CD
CD
Is--
CO
X
I I I O) CD CVJ
I CO
I
CD
co
<
o >-'
if
_
z
3 o z
(13
o
3 P Q
w
C/D
o
0)
o 2 UJ
O
_
T-
in in
co
CO
I I CVJ CO
)
o co -
o co
sf cd
r-
r-
I
I
t
I
t-
C\J
CVJ
CVJ
I
I
I
I
cd in
o
in o co
UJ
CT)
CO
CD
r-
in
o oO co
-
CO
-
y x z
TO
C/D X O I-
3?
< UJ X CO
UJ X I- CO CO X
O O O O O O O o o o o o o o
o o p in p o in '
'
cd
cm' o cd Is--" co co"
cm co
- t in
co
(A OJ
> CO
X
uj X 5 "
o
a
m
o
s cc
m
O CO X
-
O h- CO
o o
o o o o
o
o o o "
co
en
o o
o o o o
o o
CO CD O CM
o -t i CD CO
0)
X
LU O
O
d Z CO UJ
o o
Z X LU h- CO
o p o in in in in CM CO t-" CD" CM CM CM
.
-
j-
CO
X
o o o o o o o o o o o o
"
CM
CM
CO
co
in
CD
CO UJ
5 co
CO
I- < CO -I
<
: v;;
4 . .
\
mm
f
tents will result in decreasing he
!
I
< o
o m o m o o o CM CM CO CO CO CO
bers obtainable with various iron s
marked on hardness as it is on ter
Because gray iron is the least ex be considered first when a cast r
should be chosen only when the
gray iron are inadequate. Examph TABLE 111
t i
in _
o
j
I
!
o
5
a considerable effect on the hardn
!
X
General correlation between Brinell hard
anri compressive strength. (After Donoho.)
I
,
°
D
36
s
i
Z (
<
5
1-
CO
CO h- X 3
m
.
1-
cd co" tC\J
o
I
_
'
'
CD LL
320
and compressive strength. This ( tween tensile strength and hardn steadily increases with increasing by microstructural variations as is The microstructure is the prima: gray iron. Table 11-4 indicates fi-
o o o o o o o o o o o o o o o m o
111 OJ <
280
_ "
cm' cm' cm co co' co'
'
-
t
1
> Z p 5
200
riflrd-w-;, i"inf:| "jfrce'
The Brinell Hardi
by General Microstructure TYPE
Ferritic (annealed) gray iron
1
Austenitic irons
i
Pearlitic alloy iron of low allc Tempered martensitic irons
I
White iron
1
Soft gray iron Pearlitic irons
Martensitic irons
unalloyed (accor carbon content) Alloy white iron ,
Martensitic white iron
Nitrided iron (surface only)
CAST IRON
RGY
.
447
I
i
.A
.
,
t-
i- CVJ CM CM
.
r
160 K
l
CM CO t j
a
,
I'
¥ "
o in oo
80-;--
60 0
lo m o in o o m r** o o t- in co co Is-
160
200
240
280
320
360
riaraness brmel! rjrpbe' ,
"
oJ cvj cvi co co co
j, 11-19 General correlation between Brinell hardness compressive strength. (After Donoho.)
o o o o o o o o o o o o o o o in o o in *
"
o
and compressive strength. This correlation is much better than that be-
in in
(D co i-
t-
-
t-
t
CM -
cq p co
oq p in
cd co cd r'- oo go
i i i i o> co cvj oo
tween tensile strength and hardness because the compressive strength steadily increases with increasing hardness and is not greatly influenced by microstructural variations as is the tensile strength. The microstructure is the primary factor in determining the hardness of gray iron. Table 11-4 indicates the wide range of Brinell hardness numbers obtainable with various iron structures. The composition also exerts a considerable effect on the hardness. Increasing carbon and silicon contents will result in decreasing hardness although the effect is not as marked on hardness as it is on tensile strength. ,
CVJ
I i i J; c\j oo
co
in in cd r- Is-
o oo
cm O oq in (0 N O CM CO
T-
T-
T-
t-
C\l
00
CM
I
I
I
I
I
I
1
5 a o
O n V
co in o in p co
TO
Ol t- CO
0)
CD CO o
Because gray iron is the least expensive type of casting, it should always be considered first when a cast metal is being selected. Another metal should be chosen only when the mechanical and physical properties of gray iron are inadequate. Examples of applications requiring a bare minl-
S o
o o o o o o o o o o o o o o
,
f
o o o in p o_ in '
,
"
co cm o' 00 N co oo cm co .
.
m h-
CO
oo
TABLE 11-4 The Brinell Hardness of Iron Castings Classed by General Microstructure
bd
sV .
u
4
-
5
< c
TYPE
BHN
0)
o o O O O O O
S
Ferrltic (annealed) gray iron
110-140
jxf
Austenitlc irons
140-160
S3
Soft gray iron
140-180
o o o o o o o
o o o o p o_ in "
co
h-' a)"
OOO
o
OtM
_ "
Is-" g
CDOO
x tn
03
o o o o o o o o o o o o o o o o in in in in
E
S
_ "
CM CD t- CD" CM' CM CM* CM CM CO CO
J" m CD
o a to
r
I cMCMcoco- -inco
Pearliticlrons
160-220
Pearlitic alloy iron of low alloy content
200-250
Tempered martensitic irons
260-350
Martensitlc irons
350-450
White iron, unalloyed (according to carbon content) Alloy white iron
280-500 450-550
Martensitic white iron
550-700
Nitrided iron (surface only)
900-1 000 ,
1
Mi 448
INTRODUCTION TO PHYSICAL METALLURGY
mum of casting properties and lowest possible cost are counterweights for elevators and industrial furnace doors. Gray iron is widely used also for
guards and frames around hazardous machinery.
Many types of gear
housings, enclosures for electrical equipment, pump housings, and steam b turbine housings are cast in gray iron because of its low cost. Other similar I
gray-iron castings are used for motor frames, fire hydrants, and sewer I covers.
\
11-14 Chilled Cast Iron Chilled-Iron castings are made by casting the molten metal against a metal chiller, resulting in a surface of white cast iron. This
WHITE
hard, abrasion-resistant white-iron surface or case is backed up by a softer
:
gray-iron core. This case-core structure is obtained by careful control of
;, *
the overall alloy composition and adjustment of the cooling rate. Freezing starts first, and the cooling rate is most rapid where the molten J-
metal is in contact with the mold walls. The cooling rate decreases as the ij center of the casting is approached. A chilled-iron casting may be produced 1
by adjusting the composition of the iron so that the normal cooling rate 1 at the surface is just fast enough to produce white iron while the slower | cooling rate below the surface will produce mottled or gray iron (Fig. 11 -20), If only selected surfaces are to be white iron, it is common practice to i use a composition which would normally solidify as gray iron and employ metal liners (chills) to accelerate the cooling rate of the selected areas, The depth of the white-Iron layer is controlled by using thin metal plates, t
whenever a thin white-iron layer is desired and heavier metal plates where ft
a deeper-chill is necessary.
t\
'
it.
9
; 15
,
MOTTLED
jm
The depth of chill decreases and the hardness of the chilled zone ifr.'BE
creases with increasing carbon content. Since silicon is a graphitizer, theJIt depth of chill is decreased with-increasing silicon content. The addition of manganese decreases the depth of chill until the sulfur
'
y
has been neutralized by formation of manganese sulfide. Above thislj amount, manganese increases chill depth and hardness.
Phosphorus decreases the depth of chill. With carbon and silicon conv |»
stant, an increase of 0.1 percent phosphorus will decrease the depth of | chill about 0.1 in.
Nickel reduces the chill depth, and its influence is about one-fourth that;
of silicon. The reduction In chill depth is accompanied by a gradual in crease in hardness until the nickel content reaches about 5 percent. Nickel
GRAY
also refines the carbide structure of the chill and the gray-iron structur8| below the chill.
Chromium is used in small amounts to control chill depth: 0.01 percent-
r
chromium will neutralize about 0.015 percent silicon. Because of the forr: mation of chromium carbides
,
chromium is used in amounts of 1 to 4 perl Fig 11-20 Fracture of a chilled-iron casting showing 1 improve abrasion resistance! J white, mottled, and gray portions, 3x. I
cent in chilled irons to increase hardness and
.
Si
URGY
P
.
CAST IRON
ind lowest possible cost are counte
a
mace doors. Gray jron is widely u$ed
vlhazardous machinery.
-
:
it
Many typ#
trical equipment, pump housings,jnj||
jray iron because of its low cost. Othe|w
mm
for motor frames, fire hydrants,jjha; 32
on castings are made by casting the
WHITE
resulting in a surface of white cast irortH
f3%3-iron surface or case is backed up byaloj
:
5 V"Ve structure is obtained by careful conti and adjustment of the cooling rate. ie cooling rate is most rapid where the
noid walls. The cooling rate decreasesai ached. Achilled-iron casting may be prodtt MPn :
of the iron so that the normal cooling
?;{ ough to produce white iron while the sloi b will produce mottled or gray iron (Fig. 11
M;:-re to be white iron
,
it is common praetii
_
uld normally solidify as gray iron and efh] erate the cooling rate of the selected ai
0
layer is controlled by using thin metal p||l
.
syer is desired and heavier metal plates
MOTTLED
A
ses and the hardness of the chilled zone
ion content. Since silicon is a graphitizer,! nth-increasing silicon content. ;,
.
.
,
:;:
.
,
ie decreases the depth of chill until the su
|;ormation of manganese sulfide. os chill depth and hardness.
Above fi„ ;|
3 depth of chill. With carbon and silicon d rcent phosphorus will decrease the depth t
3pth, and its influence is about one-fourth tl ri chill depth is accompanied by a gradual
i
|f|nickel content reaches about 5 percent. Nici| jcture of the chill and the gray-iron struct!
GRAY
.
V .11 amounts to control chill depth; 0.01 percd'
m
out 0.015 percent silicon. Because of the fi
'
.
ps, chromium is used in amounts of 1 to 4 bse hardness and improve abrasion resistar *
1
.
11'20 Frarluro ol a ohilled-iron raRlinq showing Iho
Ha, molllod, and gray porlions
,
3>.
449
450
INTRODUCTION TO PHYSICAL METALLURGY
It also stabilizes carbide and suppresses the formation of graphite in heavy
.
sections. When added in amounts of 12 to 35 percent chromium will impart resistance to corrosion and oxidation at elevated temperatures. Copper, in additions of less than 4 percent decreases the depth of chill ,
,
,
but in excess of this amount the chill depth and hardness increase. Copper also reduces the ratio of the mottled portion to the white-iron portion. Molybdenum is only about one-third as effective as chromium in increasing the.chill depth; however it improves the resistance of the chilled face to spalltng pitting, chipping, and heat checking.
41
'
,
,
m
A constant chill depth may be obtained by using a combination of alloying elements that have opposite effects. Since nickel reduces chill depth ,
it is common practice to add chromium which increases chill depth to ,
,
neutralize the nickel and result in a constant chill depth. The normal ratio
employed iot this purpose is 3 parts of nickel to 1 of chromium. Chilled-iron casting is used for railway-car wheels crushing rolls, stamp shoes and dies, sprockets, plowshares and many other heavy-duty ma,
,
chinery partis.
#
Table 11-5 gives the composition and hardness of typical 1
chilled-iron castings. 11 15 Nodular Cast Iron
Nodular cast iron, also known as ductile iron, spheroidal, graphite iron and spherulitic iron, is cast iron in which the graphite is present as tiny balls or spheroids. The compact spheroids interrupt the continuity of the matrix much less than graphite flakes; this results in highet strength and toughness compared with a similar structure of gray iron. Nodular cast iron differs from malleable iron in that it is usually obtained
#
,
1
fig. 11 Nodular iron, unelched, showing graphite sfiieioids, 125x. (The International Nickel Company.)
as a result of solidification and does not require heat treatment. The sphe roids are mbre rounded than the irregular aggregates of temper carbon found in malleable iron (Fig. 11 -21). The total carbon content of nodular iron is the same as in gray cast iron.
Spheroidal graphite particles form during solidification because of the presence of a small amount of certain alloying elements. The noduleforming addition, usually magnesium or cerium, is made to the ladle just
before casting. Since these elements have a strong affinity for sulfur, the ,| base iron-alloy sulfur content must be below 0.015 percent for the treat-
ment to be effective, and the alloys ar The amount of ferrite in the as-cast
rate of cooling. Nodular irons with a cent pearlite are known as ferritic iro maximum ductility, toughness, and m A matrix structure which is largely by normalizing. Normalizing is carr
perature of 1600 to 1650C,F. Pearliticc but less ductile than ferrite irons.
TABLE 11-5
i .
Si
Mn
3 35
0 50
0 55
3 40
1 35
0 60
Mold boards
3 50
1 00
Sprockets
3 30
1 80
APPLICATION
C
Car wheels Plowshares
By
.
.
.
.
permission from
Ohio, 1961.
t
t
?!
'
m
Composition and Hardness ot Typical Cnilled-iron Castings*
} .
!
.
.
.
.
.
HARDNESS
62 scleroscope
levels.
Austenitic ductile irons are highly s
514Brinell
tenitic structure down to at least -750F
0 60
534Brinell
of their relatively high corrosion resis
0 65
477 Brinell
elevated temperatures..
.
.
.
Metals Handbook," 8th ed., American Soc'ety for Metals. Metals Park,
"
A
by quenching in oil or water from 160C are usually tempered, after hardening,
The tensile mechanical properties of in Table 11-6.
AS
URGY
1
'
i
oppresses the formation of graphite in hi -
.
aunts of 12 to 35 percent, chromium will li
v
.>ft4id
?
oxidation at elevated temperatures. 0 '
#
than 4 percent, decreases the depth of chi|
* f
le chill depth and hardness increase. Cop| nottled portion to the white-iron portion,
I!
one-third as effective as chromium- in in*!
jver, it improves the resistance of the chilled ing, and heat checking.
M t
pe obtained by using a combination of alloy*.;
i|te effects. Since nickel reduces chill depthf
chromium, which increases chill depth, l| t in a constant chill depth. The normal r8tf0|||
;
4
«
3 parts of nickel to 1 of chromium.
for railway-car wheels, crushing rolls, stam|
:
y
and many other heavy-duty mai
lowshares
,
#
ues the composition and hardness of typic8f|
:
r r
past iron, also known as ductile iron, sp/lAsi
#
jrulitic iron, is cast iron in which the graphitef '
eroids. The compact spheroids interrupt the
ess than graphite flakes; this results in highfi
pared with a similar structure of gray iroi| i malleable iron in that it is usually obtali d does not require heat treatment. The spLT
i
,
i the irregular aggregates of temper carbol| x
5 .,
form during splidificay(3yn ,
-
.
H'
bnesium or cerium, is made to the ladlejuf Elements have a strong affinity for sulfur, tl
and the alloys are described as "desulfurized." The amount of ferrite in the as-cast matrix depends on composition and rate of cooling. Nodular irons with a matrix having a rpaximum of 10 percent pearlite are known as ferritic irons (Fig. 11 -22). This structure gives maximum ductility, toughness, and machinability. A matrix structure which is largely pearlite can be produced as cast or by normalizing. Normalizing is carried out by air cooling from a tem,
perature of 1600 to 1650oF. Pearlitic ductile irons (Fig. 11-23) are stronger but less ductile than ferrite irons. A martensitic matrix may be obtained by quenching in oil or water from 1600 to 1700oF. The quenched structures are usually tempered after hardening, to the desired strength and hardness
t must be below 0.015 percent for the trei ardness of Typical Cnilled-iron Castings*
,
ft ment to be effective
J}ecause. of i\
j of.certain aiioying..ei.ements,..„]7ie_nodui
Nodular Iron unetched, showing graphite
folds, 125x. (Tho International Nickel Company.)
nodular iron is the same as in gray cast irorig
' .
'
11-21
I
,
Mn
HARDNESS
0 50
0 55
1 35
0 60
514Brinell
1 00
0 60
534 Brlnell
1 80
0 65
477 Brinell
.
.
.
.
.
.
.
.
62 scleroscope
i
levels.
Austenitic ductile irons are highly alloyed types which retain their austenitic structure down to at least -750F. These irons arebf interest because
of their relatively high corrosion resistance and good fereep properties at f elevated temperatures. The tensile mechanical properties of basic types of nodular iron are given .
"
8th ed., American Society for Metals, Metals Park,
in Table 11-6.
i
i
452
INTRODUCTION TO PHYSICAL METALLURGY
breaker parts; mining-hoist drums buckets; steel mill-work rolls furr
i
,
tool and die-wrenches levers, han ,
miscellaneous dies for shaping steel A summary of the cast-iron micrc ,
I
5
.
-
r
m
j 11-16 Alloy Cast Irons An alloy cast ire '
4
at various temperatures is shown in
-
)
.!
added element or elements in suffic
.
modification in the physical or mech obtained from
raw
materials
,
such
V
Si
phosphorus, are not considered alio
. .
Alloying elements are added to ce resistance to corrosion, heat, or we
erties. Most alloying elements in ca: tization, and this is one of the impc common alloying elements are chn and vanadium. Chromium increases combined c;
'
I
V
-
1
mium carbides that are more stable
chromium increase strength, hardn' Fig. 11-22
Ferritic nodular iron showing graphite spheroids
in a ferrite matrix. Etched in 2 percent nital, 125x. (The International Nickel Company.)
kllii lilii
Some" typical applications of nodular iron are agricultural-tractor andi
implement parts; automotive and diesel-crankshafts, pistons, and cyl-j inder heads; electrical fittings, switch boxes, motor frames, and circuit-K TABLE 11 6
Mechanical properties of Basic Types of Nodular Iron* -
TYPE Ferritic
Pearlitic
ALLOY CONTENT
i ~
TENSILE STRENGTH,
YIELD STRENGTH.
ELONGATION,
PSI
PSI
% IN 2 IN.
BUS 4
130
210
Low
55,000
35,000
25
High
90,000
70,000
12
Low
80,000
60,000
10
20D
Low'
130,000
90,000
7
?75
High
130,000
110,000
2
27
100,000
80,000
10
213
150,000
130,000
2
320
60,000
30,000
40
I3C-
60,000
40,000
10
Quenched
Austenitic 5
"
-
:
-
By
permission from Metals Handbook.' 1954 Supplement, American Society (or Metals, Metals Park, Ohio. "
i-Normalized.
t3.00 percent C. 2.50 percent Si, ZO O percent Ni, 2.0 percent Mn. S3.00 percent C, 2.0 percent Si, 20.0 percent Ni. 1 percent Mn, 1.5 percent Or. i
i
i t
s
i
i
I fig 1123 Pearlitic nodular iron showing graphite spneroids in a pearlitic matrix. Etched in 2 percent nit 500v (The International Nickel Company.)
.
~
:
,
r LURGY /
>
'
CAST IRON
453
'
v
-
.
.
"
r
-
breaker parts; mining-hoist drums drive pulleys, flywheels, and elevator buckets; steel mill-work rolls, furnace doors, table rolls, and bearings; ,
tool and die-wrenches, levers handles', clamp frames, chuck bodies, and miscellaneous dies for shaping steel aluminum, brass, bronze, and titanium. ,
,
A summary of the cast-iron microstructures and the phases coexisting at various temperatures is shown in Fig. 11-24. 6 Alloy Cast irons Ah alloy cast iron is one which contains a specially
.
*
1
'
added element or elements in sufficient amount to produce a measurable modification in the physical or mechanical properties. Elements normally obtained from raw materials, such as silicon, manganese, sulfur, and phosphorus, are not considered alloy additions. Alloying elements are added to cast iron for special properties such as
resistance to corrosion, heat, or wear, and to improve mechanical properties. Most alloying elements in cast iron will accelerate or retard graphitization, and this is one of the important reasons tor alloying. The most common alloying elements are chromium, copper, molybdenum, nickel, and vanadium.
Chromium increases combined carbon by forming complex iron-chro5
t
-
,
mium carbides that are more stable than iron carbide. Small amounts of
.
i
chromium increase strength, hardness, depth of chill, and resistance to
jphite spheroids 125x. (The
of nodular Iron are agricultural-tractor e and diesel-crankshafts, pistons, and gs, switch boxes, motor frames, and circ
m 5
.
m
Wei.
'
.
:
clypcsct ini:N(iiii.
Nodular Iron* yii i.nnim-.Ncviii.
I I ONfiAriON,
PSI
% IN 2 IN.
5i,
-
7
35,000
25
70,000
12
60,000
10
90,000
7
110,000
2
RO oon
10
130,000
2
30,000
40
40,000
10
,
mmmmsm
am
21!
m
V"
mmmmi&Sii7a
ment, American Society (or Metals, Metals Park Ohio. ,
11-23
Pearlltlc nodular Iron showing graphite
llherolds In a pearlltlc matrix. Etched in 2 percent nital,
percent Mn. rcent Mn, 1.5 percent Or.
(The International Nickel Company.) <:
w
454
INTRODUCTION TO PHYSICAL METALLURGY
wear and heat but decrease macl-
i
tendency of chromium is illustratec structure of a soft gray iron as the
.
a
1
T
Comtnerciai ;as' io1 -jiui
Ft;
Fas I coo! t
r
Moaerale
y I- /
/
n !U
11
.
+ 6,
a
I c,
ui
f 0 C 3
"0
a
ft
White C,l
FfnrliliC d' ). C.I
Reheat; hold in zone
II 30 1 hou's
Fast caol
Ferniic
Pearlil i.
aray C i
cluciileC.I
J Slow cool
ff/
floke qrophile
'
qraphne-temper carDon
Ji
n
'
f; -
P\ G,
-
'
\ 3*
q'ophite sphcoids
-
.
peariile
n - fernte
4 1
y - auslenite :
/WBHiillE
'
Pearl itic malleable
Ferritic malleable
Fig. 11-24 Summary;;Of cast-iron microstructures and the phases coexisting at various temperatures. (From Moffatt, F'earsall, and Wulff, "The Structure and Properties of
Materials," John Wiley & Sons, Inc., New York, 1964.)
fig. 11-25
Low-nickel cast iron. As-cast, graphite, p
and a small amount of ferrite (white areas). Etched i cent nital. Top 100X, bottom 500x (The Internation
i
Nickel Company.)
-J5»-»-
-
iv '
1 LURGY
-
CAST IRON
wear and heat but decrease machinability.
455
The strong carbide-forming
tendency of chromium is illustrated by the following effects on the microI
structure of a soft gray iron as the chromium content is increased:
v:
i /,-
I
1
1 h-
FejC
Oesulfurize()
low cool
Moderate
S
slow cool
V
+ /
n -I fv
0.
I*
\
1 . Ferntic
Pearhhc
gray C.I
duclileC.I
'
6/
s
i Ferrilir ductileC.I.
sv*"
-
3
,
r
flake graphite '|i(ipliil(; lempdr cnihon
I'l
i
Gs - graphite spheroids P ; poorlite a 1 terrije -
y
auslenile
Ml.
'
Grophlte Ferrlfe Mtures and the
S From Moffatt '
,
' oriios of M964.,
;
|Flg. 11-25 Low-nickel cast iron. As-cast, graphite, pearlite,
r *
a small amount of ferrlte (white areas). Etched in 2 per-
i
leent nital. Top lOOx, bottom 500x. (The International Nickel Company.) "
5
-
v
456
INTRODUCTION TO PHYSICAL METALLURGY
Percent 0 03 .
06 .
Structure
1
Ferrite and coarse graphite Less ferrite, some pearlite, and finer graphite Fine graphite and pearlite
3
0 0
Fine graphite, pearlite, an( Disappearance of graphite
.
.
50
Much massive carbide
.
10-30
Fine carbide
Chromium additions of less than
I
I
1
-
mechanical properties. For resistai peratures, as much as 35 percent i other alloying elements. Copper is a graphitizer but is onl this respect. For general engineer between 0.25 and 2.5 percent. Copf and strengthen the matrix. Molybdenum improves mechanic of carbides. Molybdenum is added
'1
r
i
IS*
and its effect is similar to that in st
transverse strength, heat resistance improved. Molybdenum also retard increasing hardenability and freedoi denum is always used in combinatii
t .A
.
.
.
.
Alloy
m %
..VP.
v
*4
7 i
..
Austenitef.
3
matrix
AT
Martensite
4
needle
}
Fig. 11-26 Nickel-molybdenum cast iron (1.50 percent Ni, percent Mo). As-cast, graphite flakes in a bainitic matrix. Etched in 2 percent nital. Top 100X, bottom 500x (The International Nickel Company.) 1 62 .
if [
Fig. 11 27
Ni-Hard, 3.95 percent Ni, 1.57 percent Cr.
maitensite needles in austenite plus alloy carbidi Etched in 2 percent nital, 250x. (The International Nic Company.) casi
,
CAST IRON
:
X:/::itpMe v
-jriiirlite
,
and finer graphite
1
.
3
.
0 0
Fine graphite, pearlite, and small carbide Disappearance of graphite
50
Much massive carbide
,
10-30
irlite
457
Fine carbide
Chromium additions of less than 1 percent give a general improvement In mechanical properties. For resistance to corrosion or for use at high temperatures, as much as 35 percent chromium is used in combination with other alloying elements.
Copper is a graphitizer but Is only about one-fifth as potent as silicon In this respect. For general engineering applications the copper content Is between 0.25 and 2.5 percent. Copper tends to break up massive cementite and strengthen the matrix.
Molybdenum Improves mechanical properties and Is a mild stabilizer of carbides. Molybdenum is added in quantities from 0.25 to 1.25 percent,
and its effect is similar to that in steel. Fatigue strength, tensile strength, transverse strength heal resistance, and hardness o{ the cast iron are all improved. Molybdenum also retards the transformation of austenite, thus increasing hardenability and freedom from cracking and distortion. Molybdenum is always used in combination with other alloying elements. ,
«>.,> , IMS
J
Si
r .
IP
Austenite
matrix
"
5}
2* J
,
4"
' .
Martensite
needle i
percent Ni, bainltic
ottom 500x.
jg-ifl. 11-27 IMI-Hard, 3.95 percent Ni, 1.57 percent Cr. Ast tast, martensite needles in austenite plus alloy carbides Etched in 2 percent nital 250x. (The International Nickel .
,
fcCompany.)
1
1
458
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 11-7
Typical Low-alloy Cast Irons*
Silicon, %
2 25
2 09
2 23
2 12
2 1 1
2 57
Total carbon, %
3 53
3 33
3 38
3 44
3 41
2 81
Combined carbon, %
0 57
0 89
0 64
0 74
0 72
0 73
Sulfur, %
0 08
0 06
0 06
0 06
0 06
0 06
0 14
0 16
0 14
0 16
0 16
0 16
0 59
0 62
0 60
0 60
0 60
0 60
.
.
.
.
Phosphorus % Manganese, %
.
,
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
,
.
,
.
,
,
,
,
,
.
,
Nickel, %
0 71
1 77
46
1 00
2 00
2 00
Chromium, %
0 25
0 74
0 15
0 25
0 20
0 20
0 60
0 90
0 90
2 400
2 625
26,5
3 000
3 300
3 700
28.000
36,000
30,000
39,000
44,000
66,000
180
214
187
212
250
300
.
.
Molybdenum, % Transverse strength, lb Tensile strength, psi BHN *
.
.
,
.
.
.
,
,
,
,
1
,
.
.
,
,
,
v
,
By permission from A, W, Grosvenor, "Basic Metaliurgy," vol. 1 American Society (or Metals, Metals Park, Ohio, 1954, ,
Vanadium is a very powerful carbide former stabilizes cementlte and reduces graphitization. Vanadium additions between 0.10 and 0.25 percent, increase tensile strength transverse strength, and hardness. Nickel is a graphitizer but only about one-half as effective as silicon in ,
.
,
.
,
9
,
this respect. The purpose of nickel (0.5 to 6.0 percent) in the engineering gray irons is to control the structure by retarding austenite transformation, stabilizing pearlite, and maintaining combined carbon at the eutectoid quantity. Thus the microstructure of a low-nickel cast iron shows graphite, pearlite, and very little ferrite (Fig. 11-25). In combination with about 1
I
I
percent molybdenum, the matrix tends to be bainitic (Fig. 11-26). This structure has a hardness of about 385 BHN.
For excellent abrasion resistance, about 4 percent nickel in combination with about 1.50 percent chromium is added to white cast iron. The structure is shown in Fig. 11-27. The primary dendrites, originally austenite have been partially transformed to martensite. The combination of iron carbides in a martensitic matrix results in high hardness (600 to 750 BHN) along with good strength and toughness. The addition of 14 to 38 percent nickel to gray irons results in high heat resistance, high corrosion resistance, and low expansivity. Because of the large amount of nickel, the matrix will be austenitic (Fig. 11-28). The composition and mechanical properties of some low-alloy cast irons are given in Table 11-7.
<4&
,
(
Fig. 11-28
Ni-Resist, 21.06 percent Ni, 2.20 pe
06 percent Mg. As-cast, nodular graphite ani in an austenitic matrix. Etched in 2 percent ni 100X, bottom 500x. (The International Nickel C
0
.
11-3
QUESTIONS .
'
,
Discuss the effect of the ar
iron.
;
11-1
Explain the difference in microstructure and properties of white and gray
11-4
How may the properties of c
cast Irort
11-5
Differentiate, in microstruc
11-2
iron.
Differentiate between free and combined carbon.
i
;
.
"
'
Sr:
1
JRGY
CAST IRON
2 23
2 12
2 11
2 57
3 38
3 44
3 41
2 81
0 64
0 74
0 72
0 73
.
.
.
189
.
.
.
.
) 06
0 06
0 06
0 06
0 06
0 14
0 16
0 16
0 16
162
0 60
0 60
0 60
0 60
77
1 46
1 00
2 00
2 00
174
0 15
0 25
0 20
0 20*
0 60
0 90
0 90
.
,
.
.
.
.
.
.
.
.
.
.
,
.
.
.
3 300
3 700
000
30,000
39,000
44,000
66,000
214
187
212
250
300
i
4>
.
3 000
-
i
.
.
.
26.5
. .
.
.
.
,
f
.
.
;625
,
*
.
,
116
.
.
.
.
.
4
,
-
vol. I.American Society for Metals, Metals Park, Ohio, 1954.
'
s
f
jl carbide former, stabilizes cementite, and dium additions
,
;
m
between 0.10 and 0.25 per-:
Mh transverse strength '
4:
,
and hardness.
V;;;) nIy about one-half as. effecitve as siI icon in; .
.
4
nickel (0.5 to 6.0 percent) in the engineering
vSSjicture by retarding austenite transformation, iitaining combined carbon at the eutectold
Graphite
jture of a low-nickel cast iron shows graphite,J|
s (Fig. 11'25). In combination with about if ktrix tends to be bainitic (Fig. 11-26). This
Lout 385
\
BHN.
stance, about 4 percent nickel in combing Ichromium is added to white cast iron. The;
Alloy
r
carbide Austenite
The primary dendrites, originally aus Vl isformed to martensite. The combination Of .27
matrix
.
-
matrix results in high hardness (600 to 75Q;
-
th and toughness.
d
cent nickel to gray irons results in high heat
sistance, and low expansivity.
Because o£
e matrix will be austenitic (Fig. 11-28).
anical properties of some low-alloy castlroifl !
Flfl. 11 i?8
Ni-Rosisl 21.06 porcont Ni, ?,?0 poroonl Cr percent Mg. As-cast nodular graphite and carbides In an austenitic matrix Etched in 2 percent nital. Top ,
,
0 06 .
,
.
100X
,
bottom 500x. (The International Nickel Company ) .
4
11-3 iron.
microstructure and properties of white and gray? md combined carbon.
Discuss the effect of the amount of free carbon on the properties of gray cas
11-4
How may the properties of gray cast iron be varied?
11-5
Differentiate
iron.
,
in microstructure, gray cast iron
,
malleable iron and nodulai ,
460
INTRODUCTION TO PHYSICAL METALLURGY
11-6 Why are graphite flakes in gray iron very often surrounded by ferrite areas? 11-7 Why should the iron-iron carbide diagram not be used to determine the structures in gray iron? 11-8 Why is malleable iron made only from hypoeutectic white iron? 11-9 Is it possible to make nodular irorvby heat treatment? Explain. 10- 10 Why should the sulfur content be low in the manufacture of nodular iron? 11- 11 In the manufacture of nodular iron, why are inoculants added only just before casting? 11-12 Why is welding of chilled cast irons not recommended? 11 -13 Why is welding of pearlitic malleable iron not recommended? 11 -14 What is the disadvantage of too high a first-stage annealing temperature for malleable cast iron? Too low a temperature? Explain. 11 -15 Assume that a C clamp is to made of cast iron. Select a suitable type of cast
A
iron and explain the reasons for the selection.
REFERENCES
12-1 "
American Foundrymen's Association. "Cast Metals Handbook Chicago, 1944. American Society for Metals: "Metals Handbook," 7th ed., 1948; 8th ed. vol. 1, 1961,
Introduction
Metallic materials
,
be divided into two large grot
,
materials are iron-based and th
,
,
vol. 2, 1964, vol. 7, 1972, Metals Park, Ohio.
American Society for Testing Materials: "Gray, Ductile and Malleable Iron Castings," Special Technical Publication no. 455, Philadelphia, 1969.
other than iron as the principa materials is made up of the alloy
Boyles A/ "The Structure of Cast Iron," American Society for Metals, Metals Park
tin, lead, and zinc.
,
,
Ohio, 1947.
Climax Molybdenum Company: "The Uses of Molybdenum in Nodular Irons," New York 1964.
I
Other nonf(
lesser extent include cadmium
,
titanium, tantalum, and the pre
,
Gray Iron Founders' Society: "Gray Iron Castings Handbook," Cleveland, 1958. Malleable Founders Society: "Malleable Iron Castings," Cleveland, 1960. Merchant H. D. (ed.): "Recent Research on Cast Iron," Gordon and Breach Science ,
group.
This chapter will beconcernec and alloys.
Publishers New York, 1968. ,
Walton C. F, (ed.): "Gray and Ductile Iron Castings Handbook," Gray and Ductile Iron Founders' Society, Cleveland, 1971.
COPPER AND COPPER ALLOYS
12-2 Copper The properties of copp trical and thermal conductivity strength and ease of fabricatiot a pleasing color, can be welded, by jDlating or lacquering. Certain by suitable alloying. Most of th( tors contains over 99.9 percen tough-pitch copper (ETP) oroxy Electrolytic tough-pitch copper spouts, automobile radiators ar and distillery and other process per contains from 0.02 to 0.05 copper as the compound cuproi copper form an interdendritic e ,
ri
:
.
i
core.'
-
.
.V
„!>
.
1GY
iy iron very often surrounded by fe rrlte areas? bide diagram not be used to determine the *1 '
I
r
only from hypoeutectic white iron?
r iron by heat treatment? Explain. \ it be low in the manufacture of nodular iron? j
NONFERROUS METALS AND ALLOYS
ar iron, why are inoculants added only just bef<
t irons not recommended?
iff
alleable iron not recommended? '
too high a first-stage annealing temperature iperature? Explain. vil
.
made of cast iron. Select a suitable type of ci
i selection.
:
:M
12-1 Introduction v
" .
Cast Metals Handbook," Chicago 1944. Handbook, 7th ed., 1948; 8th ed., vol. 1.196l j;il ,
'
s
,
materials are iron-based and the nonferrous materials have some elemen ,
ark Ohio. ,
jron, American Society for Metals, Metals Parftifff "
-
,
be divided into two large groups ferrous and nohferrous. The ferrous
"
als: "Gray, Ductile and Malleable Iron Castings, 455, Philadelphia, 1969. .
Metallic materials when considered in a broad sense, ma}
lie Uses of Molybdenum in Nodular Irons," New "
ron Castings Handbook, Cleveland, 1958. able Iron Castings, Cleveland, 1960.
.
"
arch on Cast Iron,
"
Gordon and Breach Science
other than iron as the principal constituent. The bulk of the nonferrous materials is made up of the alloys of copper, aluminum, magnesium, nickel tin, lead, and zinc. Other nonferrous metals and alloys that are used to e lesser extent include cadmium, molybdenum, cobalt, zirconium, beryllium titanium, tantalum, and the precious metals gold, silver, and the platinun group.
This chapter will be concerned with the more important nonferrous metalf and alloys.
ile Iron Castings Handbook," Gray and Ductile Si .
1971.
COPPER AND COPPER ALLOYS
12-2 Copper The properties of copper that are most important are high elec Irical and thermal conductivity good coriosion resistance, machinability strength, and ease of fabrication. In addition, copper is nonmagnetic, has a pleasing color, can be welded brazed, and soldered, and is easily finishec by plating or lacquering. Certain of these basic properties may be improvec by suitable alloying. Most of the copper that is used for electrical conductors contains over 99.9 percent copper and is identified as electrolytic tough-pitch copper (ETP) or oxygen-free high-conductivity copper (OFHC) Electrolytic tough-pitch copper is also used for roofing gutters, down ,
,
,
spouts automobile radiators and gaskets, kettles, vats, pressure vessels ,
and distillery and other process equipment. Electrolytic tough-pitch cop per contains from 0.02 to 0.05 percent oxygen, which is combined with copper as the compound cuprous oxide (Cu20). As east copper oxide anc copper form an interdendritic eutectic mixture (Fig. 12-1). After working ,
462
INTRODUCTION TO PHYSICAL METALLURGY
i r 9
.J
-
..
'
'
.
: .
. r
ft /
j
/
si -
m
Fig. 12-1 Copper-copper oxide eutectic in cast toughpitch copper. Lightly etched with sodium dichromate, 500x.
IS.
fig. 12-2 Particles of copper oxide (black spots) in 1 toufih-pitch copper. Lightly etched in ammonium hy and hydrogen peroxide 250x. (Revere Copper and E
(Revere Copper and Brass Company.)
Company.)
and annealing, the interdendritic network is destroyed (Fig. 12-2) and the strength is improved. Oxygen-free copper is used in electronic tubes or similar applications because it makes a perfect seal to glass. Arsenical copper containing about 0.3 percent arsenic has improved resistance to special corrosive conditions and is used for certain condenser and heat-exchanger applications. Free-cutting copper with about 0.6 percent tellurium has excellent machining properties and is used for bolts studs, welding tips, and electrical parts such as contact pins, switch gears, relays, and precision electrical equipment. ,
reduction for strip is based on thi
i
I r
,
difference.
Annealed tempers are used for described by range of grain size 01 age grain diameter in millimeters. ized by ASTM E79 Estimating the ,
,
TABLE 12 1
Cold-worked T
Silver-bearing copper has a silver content of 7 to 30 oz/ton. Silver raises
the recrystallization temperature of copper thus preventing softening during soldering of commutators. It is preferred in the manufacture of
DESCRIPTION
electric motors for railroad and aircraft use.
Quarter hard
,
.y .
.
.
12 3 Temper" Designation of Copper and Copper Alloys Since copper and most copper alloys are homogenous single phases, they are not susceptible to heat treatment and their strength may be altered only by cold working. ,
There are two general classes of temper for non-heaMreatable wroughtcopper alloys: cold-worked and soft or annealed.
The different cold-worked tempers shown in Table 12-1 are obtained by cold-working the annealed material a definite amount. The percentage
-T-
Half hard
Three-quarter hard Hard i;
Extra hard
Spring Extra spring *
From "Metals Handbook
"
,
Metals Park, Ohio.
1961
«.
i.ijii
'
mii mi
RGV
NONFERROUS METALS AND ALLOYS
46J
r.
'
ii .
:
JSC
.
r
V
».
v
,
.
s.,
i
i
-
55; pugh-
-
«f % t2-2 Particles of copper oxide (black spots) in wrought
;
late, 500x.
.
'
Bfc' tough-pitch copper. Lightly etched in ammonium.hydroxide and hydrogen peroxide 250x. (Revere Copper and Brass $ t Company.) '
.
I
network is destroyed (Fig. 12-2), and the ree copper is used in electronic tubes or lakes a perfect seal to glass. oout 0.3 percent arsenic has improved re;
reduction for strip is based on thickness difference and for wire on area difference.
Annealed tempers are used for forming at room temperature and are described by range of grain size or nominal grain size expressed as average grain diameter in millimeters. Measurement of grain size is standard-
iditions and is used for certain condenser
,
jt 0.6 percent tellurium has excellent mabr bolts, studs, welding tips, and electrical ?::;':tch gears relays, and precision electrical
ized by ASTM E79 Estimating the Average Grain Size of Wrought Copper ,
,
TABLE 12-1
Cold-worked Temper Designations*
iver content of 7 to 30 oz/ton. Silver raises
:
;:;
: : ;
e of copper, thus preventing softening rs. It is preferred in the manufacture of
aircraft use.
Quarter hard
10.9
20.7
Half hard
20.7
37.
r
STRIP
and Copper Alloys Since copper and % ous single phases, they are not susceptible * ngth may be altered only by cold working.
'
: ::
:,
...
'
APPROXIMATE % REDUCTION BY COLD WORKING
DESCRIPTION
-
WIRE
Three-quarter hard
29.4
50.0
Hard
37.
60.5
)f temper for non-heat-ireatable wrought- «
Extra hard
50.0
75.0
Spring
60.5
84.4
soft or annealed.
Extra spring
68.7
90.2
|
\
ipers shown in Table 12-1 are obtained by is "
terial a definite amount.
The percentagej
.From "Metals Handbook Metals Park, Ohio.
"
,
1961 ed.. p 1006. American Society for Metals .
I
,
1
464
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 12 2
Commonly Used Grain Sizes
GRAIN SIZE, MM
RECOMMENDED FOR
0 015
Slight forming operations;
.
100
1000
best-polishing
Easy drawing, good polishing such
0 025 .
L
a
900
as hubcaps o
0 035
Good drawing and polishing as for headlight reflectors
Sf 800
0 050
Heavy drawing and spinning; more difficult to polish
I 700
0 100
Severe draws on heavy material
.
.
a .
600
and Copper-base Alloys. The grain size best suited for a particular application depends upon the thickness of the metal, the depth of draw, and the type of surface required after the draw. Commonly used grain sizes are
500
400 Cu
shown in Table 12-2.
12-4 Copper Alloys 1
sified as follows:
Fig. 12-3
g
Cu-rich portion of the Cu-Zn phase diagi
iTiorn "Metals Handbook
I
20
Weight perce
"
:
10
The most important commercial copper alloys may be clas-
Brasses-alloys ot copper and zinc
" ,
1948 ed. p. 1206, Amen ,
Society for Metals, Metals Park Ohio.) ,
A
Alpha brasses-alloys containing up to 36 percent zinc
1 Yellow alpha brasses-20 to 36 percent zinc 2 Red brasses S to 20 percent zinc B Alpha plus beta brasses-54 to 62 percent copper Bronzes-up to 12 percent of alloying element
II
A
Tin:bronzes
B
Silicon bronzes
C
unit cubes. The ordering reactio prevented by quenching.
The effect of zinc on the tensili shown in Table 12-3.
Aluminum bronzes
In most cases the addition of s
D Beryllium bronzes Cupronickels-alloys of copper and nickel Nickel silvers-alloys of copper, nickel, and zinc
III IV
12-5 Brasses-General Some tin
with the copper atoms at the con
,
,
ductility as strength increases. 12-3, increases ductility along w strength and ductility is obtained
Brasses are essentially alloys of copper and zinc. 1 ;
of these alloys have small amounts of other elements such as lead ;; ,
or aluminum. Variations in composition will result in desired color,;! | J ble 12 3 Effect of zinc on Properties of copp<
strength, ductility, machinability, corrosion resistance, or a combination -; [
I
f
of such properties.
if
The portion of the binary copper-zinc phase diagram which is applicable f to commercial alloys is shown in Fig. 12-3. The solubility of zinc in the J r .
'
alpha (a) solid solution increases from 32.5 percent at 1657
to about 39
;
k
percent at 850CF. Since copper is f.c.c. (face-centered cubic), the a solidj | solution is f.c.c. The beta (f3) phase is a b.c.c. (body-centered cubic) elec-fp tron compound and undergoes ordering indicated by a dot-dash line, in! the region of 850 to 875CF. On cooling in this temperature range the b.c.c. ,
j3 phase, with copper and zinc atoms randomly dispersed at lattice points.i changes continuously to the ordered structure fi1, which is s'!ill b.c.c. but
f
,
f
i
TENSILE STRENGTH
,
Zn,%
PSI
0
32,000
5
36,000
10
41 000
15
42 000
2J
43,000
25
45,000
,
,
30
46,000
35
46,000
iO +6')
54,000
*0ata !rom Chase Brass & Copper Co. for commercial alloys
t
*
"
RGY
NONFERROUS METALS AND ALLOYS
465
1 Used Grain Sizes
I100
2000
RECOMMENDED FOR v
" v
-
Slight forming operations;
1000
1800
best polishing
0 + a
Easy drawing, good polishing such
+
900
1600
as hubcaps
Good drawing and polishing as (or headlight reflectors
BOO
1400 3
/3 a
Heavy drawing and spinning; more difficult to polish
e 700 1200 i-
3
Severe draws on heavy material
600
ain size best syited for a particular applica|
-
1000
500
3S of the metal, the depth of draw, and thl
a + /3
|he draw. Commonly used grain sizes af$
0
-
800
400
20
10
Cu
ant commercial copper alloys may be clas-Sl
30
40-
50
Weight percent zinc
Pi
.
12-3
Cu-rich portion of the Cu-Zn phase diagram.
om "Metals Handbook," 1948 bd,, p. 1206, Americnn :inc
So.Clely
tor Meltils, Metals Park, Ohio.)
ining up to 36 percent zinc ) to 36 percent zinc
with the copper atoms at the corners and zinc atoms.at the centers of the unit cubes. The ordering reaction is so rapid that it cannot be retarded or prevented by quenching.
cent zinc
to 62 percent copper
loying element
The effect of zinC on the tensile properties of annealed copper alloys is shown in Table 12-3.
In most cases, the addition of solid-solution elements tends to decrease
ductility as strength increases. The addition of zinc, as shown in Table 12-3 increases ductility along with strength. The best combination of
and nickel
i nickel, and zinc :
,
essentially alloys of copper and
zinc.
strength and ductility is obtained in 70Cu-30Zn brass.
"
il amounts of other elements such as lead,
i composition will result in desired color,-t
TABLE 12 3
y corrosion resistance, or a combination
TENSILE STRENGTH,
ELONGATION,
BHN,
PSI
%1N?IN.
10 MM, 500 KG
0
32,000
46
38
5
36,000
49
49
10
41,000
52
54
15
42,000
56
58
20
43,000
59
56
,
Zn,%
)er-zinc phase diagram which is applicable in Fig. 12-3. The solubility of zinc in the i
es from 32.5 percent at IGST 'F to about 39 i '
is f.c.c. (face-centered cubic) the « solid
Effect of Zinc on Properties of Copper Alloys*
,
hase is a b.c.c. (body-centered cubic) elec-
| ordering, indicated by a dot-dash line, In cooling in this temperature range the b.c.c. atoms randomly dispersed at lattice points
,:
[dered structure 0', which is still b.c.c. butj
*
; 1
|
25
45 000
62
54
30
46,000
65
55.
35
46 000
60
55
40 (-I fi)
54,000
45
75
,
,
'
Data from Chase Brass & Copper Co for commercial alloys of moderate grain size. .
466
INTRODUCTION TO PHYSICAL METALLURGY
i
The commercial brasses may be divided into two groups, brasses for cold ; ,
working (a brasses) and brasses for hot working (a plus p brasses). 12'6 Alpha Brasses
Alpha brasses containing up to 36 percent zinc possess
relatively good corrosion resistance nd good working properties. The color J i
1
of a brasses varies according to copper content from red for high-copper 1 L
alloys to yellow at about 62 percent copper. The a brasses may be divided
i
into two groups yellow a brasses and red brasses. Yellow a Brasses These contain 20 to 36 percent zinc, combine good strength with high ductility and are therefore suited fordrasticcold-working
IP
,
operations. It is common practice to stress-relief anneal these brasses after severe cold working to prevent season cracking. Season cracking or stress-corrosion cracking is due to the high residual stresses left in the brass as a result of cold working. These stresses make the brass more susceptible to intergranular corrosion, particularly in ammonia atmospheres
(Fig. 12-4). Stress relieving in the recovery range (up to about 500°F) orthe
i
9
.
J -
| F19 l?-5 Plug-type dezincification in brass, unetche
3
>)
(Movere Copper and Brass Company.)
r
i
2
0
.
s
3
substitution of a less susceptible c stress-corrosion cracking. Yellow « brasses are also subject tion. This type of corrosion usual sea water or with fresh waters that
-
bon dioxide.
A
.
i
4 5
1
V
.
r
/
aV
Dezincification invo
sequent deposition of porous nor unless stopped will eventually pen lead to leakage through the porou: area it is known as plug-type dezir tin or antimony minimize dezincifk The most widely used yellow a t and yellow brass (65Cu-35Zn). Ty radiator cores tanks, headlight lamp fixtures socket shells, sere' grommets, rivets, springs; plumbi ,
,
,
'
ponents.
V."" vi
The addition of 0.5 to 3 percer leaded brass is used for screw-ma Fig. 12-4 Stress-corrosion cracking in « brass, 150x, (Revere Copper and Brass Company.)
i
1
parts, tumblers, gears, and watch f Two variations of yellow « brass
1
'
.
URGY
NONFERROUS METALS AND ALLOYS
467
y be divided into two groups, brasses forc ll
:
..
. . . . .
ses for hot working (a plus /? brasses).
Mjs containing up to 36 percent zinc possesi tance and good working properties. Thecotpr to copper content from red for high-copi
srcent copper. The a brasses may be divided]
.
ises and red brasses.
htain 20 to 36 percent zinc, combine g6b|| I are therefore suited for drastic cold-workirt|| actice to stress-relief anneal these brasstt.
AySjrevent season cracking. Season cracking© ;
'
1 R 5?>
'
iue to the high residual stresses left in thl idng. These stresses make the brass mortflf|| "
IP "
Tosion, particularly in ammonia atmosphef%|r
:
.
the recovery range (up to about 500CF) ortheib
1]
12-5 Plug-type dezincification in brass unetched, (Revere Copper and Brass Company.) ,
substitution of a less susceptible copper alloy will minimize the danger of stress-corrosion cracking. Yellow a brasses are also subject to a pitting corrosion called dezincification. This type of corrosion usually occurs when brass is in contact with sea water or with fresh waters that have a high content of oxygen and carbon dioxide. Dezincification involves dissolution of the alloy and a subsequent deposition of porous nonadherent copper. Action of this kind,
4
.
unless stopped, will eventually penetrate the cross section of the metal and lead to leakage through the porous copper layer. If it occurs in a localized area it is known as plug-type dezincification (Fig. 12-5). Small amounts of
r
1?
tin or antimony minimize dezincification in yellow brasses. The most widely used yellow a brasses are cartridge brass (70Cu-30Zn) and ye//ow brass (65Cu-35Zn). Typical applications include automotiveradiator cores tanks, headlight reflectors; electrical-flashlight shells, lamp fixtures socket shells, screw shells; hardware-eyelets, fasteners, ,
V
,
grommets,_uv€ts, springs; plumbing accessories; and ammunition com-
g;. P:
ponents.
The addition of 0.5 to 3 percent lead improves machinabilitysso that leaded brass is used for screw-machine parts engraving plates, keys, lock parts tumblers, gears, and watch parts. ,
hsox.
& '
I?
,
Two variations of yellow a brasses have been developed for special ap-
1 -
468
INTRODUCTION TO PHYSICAL METALLURGY
plications. Admiralty metal (71Cu-28Zn-1Sn). with the addition of 1 per cent tin for improved strength and corrosion resistance, is used for condenser and heat-exchanger tubes in steam power plant equipment. Better
!
corrosion resistance is obtained in aluminum brass (76Cu-22Zn-2AI). This
alloy forms a tenacious, self-healing film which protects the tube against
high cooling-water velocities in marine and land power stations. Red Brasses
i
These contain between 5 and 20 percent zinc. They gener-
ally have better corrosion resistance than yellow brasses and are noil
susceptible to season cracking or dezincification. The most common low-1 4?
zinc brasses are gilding metal (GSCu-SZn), commercial bronze (90Cu-10Zn|, red brass (85Cu-15Zn), and low brass (80Cu-20Zn). Gilding metal (95Cu-5Zn) has higher strength than copper and is used for coins, medals, tokens, fuse caps, primers, emblems, plaques, and as a base for articles to be gold-plated or highly polished. 12
Commercial bronze (90Cu-10Zn) has excellent cold-working and hot-li f'9v " w ' '
-
working properties and is used for COStume jewelry, compacts, lipstick I
-
; / Muntz metal water-quenched from isist /, ls 9 ' " Phase 13 dark- Etched in ammonium hydroxide and hydrogen peroxide 50x. (By permission .
,
cases, marine hardware, forgings,, » rivets, and screws. Leaded commercial |1 to ™ R M- Brick. R- B Gordon, and a. Phillips, strucu .. , , , and Properties of Alloys, 3d ed., McGraw-Hill Book -
"
-.r-
.
r-,i
>
.
bronze (1.75 percent Pb) is used for screws and other parts for automatic .
New York 1965)
screw-machine work.
fled brass (85Cu-15Zn) is used for ets, hardware condenser and heat ,
stick cases compacts, nameplates, t ,
Low brass (80Cu-20Zn) is used fc thermostat bellows musical instrun ,
drawn articles.
12-7 Alpha Plus Beta Brasses These co Consideration of Fig. 12-3 shows that a and jS The (3' phase is harder £ than a; therefore these alloys are n brasses. At elevated temperatures tl since most of these alloys may be h i they nave excellent hot-working prof ft, The most widely used a + fi' bras has high strength and excellent hot '
.
,
i
Si-
structure of annealed muntz metal i
from the p region may suppress the -
v :
-
.-
Figure 12-7 shows the microstructure
i
-
from 15150F. The quench preserved i particularly at the grain boundaries. perature will allow more of the a to coi
Fig. 12 6 Two-phase structure of annealed muntz metal.
W:
Fin ammonium appears lighthydroxide surrounding the darker«grains. Etched and hydrogen peroxide, 150x.
» Mb-
(Revere Copper and Brass Company.)
.1
'
thus it is possible to heat-treat this
form for ship-Sheathing tural work-
i
I
1;
i
r
' .
v
i
I
15
,
condenser t"
0 used for valve
-
-
:
-
'
"
-
m
i
NONFERROUS METALS AND ALLOYS
469
m j
-28Zn-1Sn) with the addition of ,
d corrosion resistance, is used f6r;
g in steam power plant equipment..:,B|
.
1
i aluminum brass (76Cu-22Zn-2A|)f|i
-
ng film which protects the tube agf arlne and land power stations, rmt :
/een 5 and 20 percent zinc. They flt
l
»
ance than yellow brasses and are:
i
Idezincification. The most common lol ju-SZn), commercial bronze (90Cu-10ZA , Jass (80Cu-20Zn).
S aps, jgherprimers, strengthemblems, than copper and is plaques,
»
'
id or highly polished.
i) has excellent cold-working and
or costume jewelry, compacts,
V
.
iM-l Muntz metal water-quenched from 1515°F. /3'
hotv
tee Is light, « phase is dark. Etched in ammonium
lipstii
Ikoxide and hydrogen peroxide, 50x. (By permission
S3 rivets, and screws. Leaded commerdi
%R. M. Brick, R. B. Gordon and A. Phillips, "Structure I Properties of Alloys 3d ed., McGraw-Hill Book ,
iSyor screws and other parts for automal
"
.
,
'
ipany, New York 1965.) ,
Red brass (85Cu-15Zn) is used for electrical conduit screw shells sockcondenser and heat exchanger tubes plumbing pipe, lip,
ets, hardware
,
,
,
stick cases compacts, nameplates, tags, and radiator cores. ,
Low brass (80Cu-20Zn) is used for ornamental metalwork medallions, thermostat bellows, musical instruments flexible hose, and other deep,
,
,
p
Si
.
drawn articles.
|7 Alpha Plus Beta Brasses These contain from 54 to 62 percent copper.
l i
1
\ Consideration of Fig. 12-3 shows that these alloys will consist of two phases,
|, a and /3 F *
'
The p' phase is harder and more brittle at room temperature
.
than a; therefore these alloys are more difficult to cold-work than the a ,
f brasses. At elevated temperatures the p phase becomes very plastic, and
I since most of these alloys may be heated into the single-phase p region
>
,
they have excellent hot-working properties
5
.
The most widely used a + /3' brass is muntz metal {60CU"40Zn)
,
which
II has high strength and excellent hot-working properties. The two-phase I- structtiro
>
of annonlod munlz molnl is shown in I hi 12 (3. Rapid cooling
from the p region may suppress the precipitation of most of the
a phase.
Figure 127 shows the microstructure of muntz metal after water quenching
from 15150F. The quench preserved most of the p
,
particularly at the grain boundaries
.
netal
ichec bx.
H p
I
bul some
,
Subsequent reheating to a low tem-
perature will allow more of the n to come out of the supersaturated solution; thus it is possible to heat-treat this alloy. Muntz metal is used in sheet
form for ship-sheathing, condenser heads, perforated metal, and architectural work.
It is also used for valve stems brazing rods, and condenser ,
i
470
INTRODUCTION TO PHYSICAL METALLURGY
tubes. Leaded muntz metal containing 0.40 to 0.80 percent lead has improved machinability.
Free-cutting brass (61.5Cu-35.5Zn-3Pb) has the best machinability ol any brass combined with good mechanical and corrosion-resistant prop-
"
A
l
erties. It is used for hardware, gears and automatic high-speed screw machine parts.
Forging brass (60Cu-38Zn-2Pb) has the best hot-working properties of
any brass knd is used for hot forgings, hardware, and plumbing parts.
-
7
.
Architectural bronze (57Cu-40Zn-3Pb) has excellent forging and free
matching [broperties. Typical applications are handrails, decorative moldings grilles, storefronts, hinges, lock bodies, and industrial forgings. ,
Naval brass (60Cu-39.25Zn-0,75Sn), also known as tobin bronze, has increased resistance to salt-water corrosion and is used for condenser plates,
F!<3. 12 9
welding ro;d, propeller shafts, piston rods, and valve stems. Leaded naval
(light) in a /3 matrix (dark). Etched in NH.OH + H20.,, I
.
,
'
Fed , lOOx.
(From Metals Handbook vol. 7 "Atlas ol Microstruclures." American Sociely lor Melals 197?,) .
brass with
Manganese bronze as sand cast, a needle
.
the addition of 1.75 Pb for improved machinability is used for
,
,
,
.
marine hardware. Figure 12-8 shows the two-phase structure of hot-extruded naval brass. The /3 phase is darkened while the alpha phase is light. Manganese bronze (58.5Cu-39Zn-1.4Fe-1 Sn-0.1 Mn), really a high-zinc brass hasjhigh strength combined with excellent wear resistance and is used for clutch disks, extruded shapes, forgings, pump rods, shafting rod,
valve stems and welding rod. ,
ganese bronze as sand cast.
Fic It cc
solution in a matrix of the /3 phase
,
.
12-8 Cast Brasses
The previous disci wrought brasses which are mainly ,
cast brasses are similar in name to 1
appreciable amounts of other alloyi
i
1 to 6 percent and lead from 1 to 10 *5
1
3 .
7*\
manganese nickel, and aluminum.
1
,
An example of a casting brass is p.
mm
mm i
m
1
.
6
V '
1
mmmm
5*
i
7m.
4*
[M 6
11111111 y f.
-
mm
r
9r
4
Fig. 12 8 Hot-extruded naval brass, fi phase is dark, n phase is light. Etched in FeCl,, 75v. (Revere Copper and Brass Company.)
t
i
Fiy. 12'10 Leaded red brass as continuous cast. Con dendrites with uniformly distributed lead particles (bla( <30ts). Etched in NH,OH + H20„ 200\'. (From Metals H !)00K, vol, 7, ' Atlas of Microstructures," American Soci
!c( Metals. 1972.)
.
...
:
NONFERROUS METALS AND ALLOYS
471
nmg 0.40 to 0.80 percent lead hasini g '
y
: .
I
: -
C!n-3Pb) has the best machinability chanical and corrosion-resistant proi
:
ars and automatic high-speed scre#; .
las the best hot-working properties igs, hardware, and plumbing parts i-3Pb) has excellent forging and fre$L» nations are handrails, decorative moW mm. v;uipk bodies, and industrial forgings. c T
.
m&n), also known as tobln bronze, hasW|!
tr
OSion and is used for condenser plateSr ypfl. 12-9 Manganese bronze, as sand cast, a needles W in * P matri* (dark) Etched in NH.OH + h2o2, then n rods and valve stems. Leaded navat '
'
'
.
.
r.apfeCI,, lOOx. (From Metals Handbook
,
for improved machinability IS used TOj:
.
.
darkened while the alpha phase is light?
.Sj3i-1 4Fe-1Sn-0.1Mn), really a higlwinf .
/r 'i with excellent wear resistance and i uV-Vlpes forgings, pump rods, shafting rO ,
"
Atlas ot
,
,
)ws the two-phase structure of hot-exl .
vol. 7
icros,ructures American Society for Metals
1972 )
: a!
-
valve stems, and welding rod.
Figure 12-9 shows the structure of man-
ganese bronze as sand cast.
It consists of white needles of the a solid
solution in a matrix of the p phase.
§12-8 Cast Brasses The previous discussion was concerned primarily with
4 .
;>
.
wrought brasses which are mainly binary alloys of copper and zinc. The ,
cast brasses are similar in name to the wrought brasses but usually contain appreciable amounts of other alloying elements. Tin may be present from 1 to 6 percent and lead from 1 to 10 percent; some alloys may contain iron
1
ii
,
manganese, nickel and aluminum. ,
An example of a casting brass is leaded red brass (85Cu-5Sn-5Pb-5Zn)
,
A
.
OF
m
>t6
mm
v
y
M
0
s
3
mm 5
m m ml v
f
.
V
i
'-
s
;
1
1
R
i
f
-
'
0
2
\1 I; '
a
'
! and
112-10 Leaded red brass as continuous cast Cored idrltes with uniformly distributed lead particles (black .
«
J»). Etchort in NII.OII 1 H,0,i 200X. (From Molfils llnndibk, vol. 7. "Alias of Microstructures," American Society or Metals, 1972.)
1 4
1 \
I I m
I
472
INTRODUCTION TO'PHYSICAL METALLURGY
which is used for general castings requiring fair strength, soundness, and
i
of phosphorus content is between 0
'
good machinj ng properties, such as low-pressure valves, pipe fittings, small
.
1 and 11 percent The copper-rich portion of the co 12-11. The /3 phase forms as the re .
gears, and small pump castings. The structure of leaded red brass, as continuous cast, is shown in Fig. 12-10. Ij consists of cored or segregated a dendrites, due to the relatively rapid cooling rate, and small, uniformly distributed lead particles (black dots). 12-9 Bronzes-General The term bronze was originally applied to the coppertin alloys; however the term is now used for any copper alloy, with the exception of copper-zinc alloys, that contains up to approximately 1.2 percent of the principal alloying element. Bronze, as a name, conveys the idea of a higher-class alloy than brass, and as indicated in the proceeding sections, it has been incorrectly applied to some alloys that are really special brasses. Commercial bronzes are primarily alloys of copper and tin, aluminum, silicon, or beryllium. In addition, they may contain phosphorus,
r f
,
. ;
f
.
''
4,j,
lead, zinc, or nickel.
12-10 Tin Bronzes These are generally referred to as phosphor bronzes since phosphorus is always present as a deoxidizer in casting. The usual range
5S
m
noo
2000
1000
Mm
Si
m
i
I800
I (a) .iir
a + L
900
1500
800
1400 O)
I
O)
3
3
700
1200 E
a
I1J
v
600
% 1 +
r
1000
a + 8
800
a
500
400
a
:.: .
+
e
500
500 5
Cu
10
15
Weight percent tin
Fig. 12-11
Copper-rich portion of the copper tin phase
diagram. (From "Metals Handbook," 1948 ed., p. 1204,
American Society for Metals, Metals Park, Ohio.)
i
20
25
(b)
30
fig. 1212 Structure of a cast 10 percent phosphor-bron
|i»i' toy. (a) Rapidly cooled, particles of in a dendritic a
juatrlx, 75x, (b) detail of the 5 phase, 1,000x. (American Srass Company.)
NONFERROUS METALS AND ALLOYS
473
.I i5
gs requiring fair strength, soundness}
'
of phosphorus content is between 0 01 and 0.5 percent and of tin between 1 and 11 percent .
as low-pressure valves, pipe fittings-,$
,
.
: £jThe structure of leaded red brass, as
The copper-rich portion of the copper-tin alloy system is shown in Fig
.
M 12-11. The /3 phase forms as the result of a peritectic reaction at 14680F
MO. It consists of cored or segregate
.
rapid cooling rate, and small, unlfofi bots). \nze was originally applied to the copj now used for any copper alloy, with; that contains up to approximately 1.2
nent. Bronze, as a name, conveys the 1 -
5, and as indicated in the proceeding
~
.
-
i llied to some alloys that are really spscl) |re primarily aljoys of copper and tin; ill i addition, they may contain phosphoi
mmm
mm
Hi
m
m
.
a deoxidizer in casting. The usual ran<
m
K
m
ly referred to as phosphor bronzes sift .-
i
'
si
r
1
«
2000
m 1800
(a)
5;
m
A
'
1600
i'loo
i r
.
1
i
Y
1000
+
20
O
800
i + 8
a
1
t
500 25
(b)
30
It tin
pg. 12-12 Structure of a cast 10 percent phosphor-bronze foy. (a) Rapidly cooled, particles of 5 in a dendritic a
t
phase
inalrlx
1204,
,
75x; (b) detail of the 6 phase
grass Company.) i -
: v v -
.
,-.
-
.
1,000x. (American
474
INTRODUCTION TO PHYSICAL METALLURGY
At 1087JF, the ft phase undergoes a eutectoid reaction to form the eulectoid mixture (tv -I- y). At 9680F, gamma (y) also undergoes a eutecloid transformation to (a + 6). The diagram also indicates the decomposition of the delta (S) phase. This takes place by a eutectoid reaction at 662°F form-
provement in the casting properti wear resistance
.
ing {a + e). This reaction is so sluggish that in commercial alloys the epsilon (e) phase is nonexistent. The slope of the solvus line below 968f
Lead is often added to tin bron resistance. High-lead tin bronze m The leaded alloys are used for bu;
,
' '
light loads.
shows a considerable decrease in the solubility of tin in the n phase. The precipitation of the
12 H Silicon Bronzes The copper-rich system is shown in Fig 12-13. Th. 3 percent at 15650F and decrease .
slow that, for practical purposes, the solvus line is as indicated by the ver- }
5
.
tical dotted line below 968°F. For this reason slow-cooled cast tin bronzes
action at 1030oF is very sluggish
,
il
containing below 7 percent tin generally show only a single phase the i a solid solution. There is some of the 6 phase in most castings contain- i ing over 7 percent tin. The structure of a rapidly cooled cast 10 percent
,
so
,
1100
phosphor bronze (Fiq. 12-12a) shows small particles of the S phase in a J. fine dendritic a matrix. Detail of the 5 phase is shown in Fig. 12-12b The phosphor bronzes are characterized by high strength toughness,
.
,
i
I
high corrosion resistance, low coefficient of friction, and freedom from I coacnn aro ncoH nr HI a nh ranmc hollnvA/c season nranU\nn cracking. Thnw They are used ovtonciwolw extensively ffor diaphragms, bellows, Irtrt lock washers, cotter pins, bushings, clutch disks, and springs. Z\np is sornetimes used to replace part of the tin. The result is an irn-
1000
i
000
noo
=12000
1000
1800
L )
3'Xj a a1
a
/3
+ L
900
1600 3
ID
800 a
1400
-
+ K
700
K
600
1200
600
1000
-
a + Y 500
Cu
fl
2
3
4
5
sac Cl
6
Weight percent silicon
Fig. 12-13 Copper-rich portion of the copper-silicon alloy system. (From Metals Handbook," 1948 ed, p. 1203, Amorican Society for Metals, Molals Park, Ohio.) .
"
"
i
\
"
I
i
b 7.S 10 Weighl percenl olummuir
,
'
>
:
.
fe! Fig. 12-14 Copper-rich portion of the copper-aluminum <||Moy system. (From "Metals Handbook 1948 ed., p. 11 l tmicsn Society for Metals, Metals Park, Ohio.) "
2
.
2 1:
-
1
NONFERROUS METALS AND ALLOYS
475
m d
.
a eutectoid reaction to form the euti
provement in the casting properties and toughness with ilittle effect on
na (y) also undergoes a eutectoid trt
wear resistance.
/iilso indicates the decomposition of
Lead is often added to tin bronze to improve machinability and wear resistance. High-lead tin bronze may contain as much as 25 percent lead The leaded alloys are used for bushing and bearings under moderate or
by a eutectoid reaction at 6620F foi
.
uggish that in commercial alloysygtl lie slope of the solvus line below 9w. ihe solubility of tin in the n phase."
11
flue to this change in solubility is,SO;
ie solvus line is as indicated by theve?|
light loads.
i
Silicon Bronzes The copper-rich portion of the copper-silicon alloy system is shown in Fig. 12-13. The solubility of silicon in the a phase is
11
5
3 percent at 15650F and decreases with temperature. The eutectoid reaction at 1030oF is very sluggish so that commercial silicon bronzes which .
lis reason, slow-cooled cast tin bron2
,
.
.nera
lly show only a single phasetj f
,
:
X the 5 phase in most castings contaff
.
re of a rapidly cooled cast 10 pW)
1100
vvs small particles of the 8 phase tit
it
2000
4
r
the 8 phase is shown in Fig. 12-126,; cterized by high strength, toughness
4
.
ifficient of friction, and freedom frpL tensively for diaphragms, bellows, lo
1000
"
.
I ttOO
ch disks, and springs. :
.
i
i
g;ie part of the tin. The result is an i
:
-
i
900
1600 Ll.
2000
t
a>
-
1
L a
BOO
l
.
1800
+
a
(3
a + Z
-
1400
i
i
5
1600 o 1
700
i
(3 + y, 1400 E
1200
+
K
1200
5
i
son
-
-
a + r 4
I
5
1000
i + r2
1000 500 Cu
6
0"
25 .
5
7.5
10
12.5
15
Weight percenl aluminum
Rg. 12-14 Copper-rich portion of the copper-aluminum i»oy system. (From "Metals Handbook 1948 ed, p. 1160,
illoy
"
'
,
9
American Society for Metals, Metals Park Ohio.) ,
1
1
476
INTRODUCTION TO PHYSICAL METALLURGY
generally contain less than 5 percent silicon, are single-phase alloys.
Silicon bronzes are the strongest of the work-hardenable copper alloys. , They have mtechanical properties comparable to those of mild steel and corrosion resistance comparable to that of copper. They are used for tanks pressure vessels, marine construction, and hydraulic pressure lines.
'
,
S: '
12-12 Aluminum Bronzes The copper-rich portion of the copper-aluminum
%
alloy system is shown in Fig. 12-14. The maximum solubility of aluminum in the a. solid solution is approximately 9.5 percent at 1050t'F. The $ phase undergoes a eutectoid reaction at 1050oF to form the (a-t- ) mixture.
Most commercial aluminum bronzes contain between 4 and 11 percent aluminum. Those alloys containing up to 7,5 percent aluminum are generally single-phase alloys while those containing between 7.5 and 11 per,
cent aluminum are two-phase alloys. Other elements such as iron nickel, ,
c£ ***
and silicon are frequently added to aluminum bronzes. Iron (0.5 to 5.0 percent) increases strength and hardness and refines the grain; manganese
,
|
Wot+y2) gi
nickel (up to 5 percent) has the same effect as iron but is not so effective:
silicon (up to 2 percent) improves machinability; manganese promotes soundness in castings by combining with gases and also improves strength, The single-phase aluminum bronzes show good cold-working properties and good strength combined with corrosion resistance to atmospheric and water attack.: They are used for condenser tubes cold-work forms, corrosion-resistant vessels nuts and bolts, and protective sheathing in marine applications. The a + /3 aluminum bronzes are interesting because they can be heattreated to obtain structures similar to those In steel. Figure 12-15a shows the structure of primary a and granular eutectoid (a + 72). representative of an as-cast 10 percent aluminum bronze. On furnace cooling from above the eutectoid temperatures, a lamellar structure resembling pearlite is formed (Fig. .12-15b), If the two-phase alloy is quenched from 1500 to 1600°F, a needlelike structure resembling martensite is formed (Fig. 12-15c). The quenched alloys are tempered between 700 and 1100aF to increase strength and hardness. Heat-treated aluminum bronzes are used for gears
m
m
11
,
,
'
,
1
1
3:
i .1
1
is
Hi 557/
i
is 1
m
i iiiiiiK I
\ 5
2 percent beryllium, A typical heatsolution-anneal at 1450oF water-qi ,
%
hardening possibilities.
The theory of age hardening is discussed in Sec. 6-16. The optimum J
m
1
0:
and drawing.and forming dies.
mechanical properties are obtained in an alloy containing approximately
m
m
'
propeller hubs, blades, pump parts, bearings, bushings, nonsparking tools, 12-13 Beryllium Bronzes The copper-rich portion of the copper-beryllium alloy system is shown in Fig, 12-16. The solubility of beryllium in the a solid solution decreases from 2.1 percent at 1590-T to less than 0.25 percental room temperature. This change in solubility is always indicative of age-
f
ft
600oF.
Figure 12-17 shows the microstru ing a nominal 1 92 percent beryllium 12-17a shows the structure after a £ .
the dark-etching y phase primarily at ing from the proper annealing tern
NONFERROUS METALS AND ALLOYS
1 E
jrcent silicon, are single-phase alloys i gest of the work-hardenable copper alldp. '
m
i
::
§A jes comparable to those of mild steel-ijf e to that of copper. They are used forta|] iuction
,
and hydraulic pressure lines;!
3
mmft
m
*
.
i .
.
ier-rich portion of the copper-aluminr ,.
477
"
Primary
112-14. The maximum solubility of alui
i
°
approximately 9.5 percent at 1050
F
7
.
i
1 m
ning up to 7.5 percent aluminum are those containing between 7.5 and 11
.
Primary «
m
M
d reaction at 1050CF to form the (« + .
mm ii Eutectoid
Hoys. Other elements such as iron, nid
quently added to aluminum bronzes. If.
mm
(a)
'
rength and hardness and refines the grai same effect as iron but is not so effectivii y&j&fates machinability; manganese promotr ning with gases and also improves strength & ;s:-J;jronzes show good cold-working properti "
;
Eutectoid
(b)
j;
"
;
ith corrosion resistance to atmospheric a'hj 5r condenser tubes, cold-work forms,
id bolts, and protective sheathing in mari 1
s are interesting because they can be heafe lilar to those in steel. Figure 12-15asho\
granular eutectoid {a + yz), representative urn bronze. On furnace cooling from above
lamellar structure resembling pearlite m :
V;:|wo-phase alloy is quenched from 1500 t( embling martensite is formed (Fig. 12-15c)i )ered between 700 and IIOO
'
Fig. 12 15 Structures of aluminum bronze All samples etched with ferric nitrate, (a) Ascast 10 percent aluminum bronze showing primary a and granijlar eutectoid, 750x; (b) furnace-cooled aluminum bronze showing lamellar eutectoid 500X; (c) quenched 10.7 percent aluminum bronze showing a marlensitic /> structure lOOx. (Ampco Metal, Inc.) .
*
F to increase!
5
13
!?5 .
4
i(c)|
IS
.
IK
1its
,
eated aluminum bronzes are used for gear8i| arts, bearings, bushings, nonsparking tools,| 2 percent beryllium. A typical heat-treating cycle for this alloy would be:
-rich portion of the copper-beryllium alloy !i
r
The solubility of beryllium in the « solidl
rcent at 1590oF to less than 0.25 percent at!
Jge in solubility is always indicative of age-|| '
1 -
!
'
solution-anneal at 1450"F
,
water-quench, cold-work, and finally age at ,
600oF.
Figure 12-17 shows the microstructures of a beryllium bronze containing a nominal 1.92 percent beryllium and 0 20 to 0.30 percent cobalt. Figure .
St
12-17a shows the structure after a slow cool has allowed precipitation of
3 is discussed in Sec. 6-16. The optimutfl;;
the dark-etching y phase primarily at the grain boundaries Water quench-
ained in an alloy containing approximatelyl
ing from the proper annealing temperature will result in a single-phase
.
'
8
INTRODUCTION TO PHYSICAL METALLURGY
The cupronickel alloys have high re high resistance to the corrosive and
water. They are widely used for cond
1000
T
exchanger tubes for naval vessels ant
/
A copper-25 percent nickel alloy i; electrolytic tough-pitch copper in th dime as shown in Fig. 12-18
SO: :
1600
,
,
1100
n ¥ (i B + 7 SCO
v . KM
ft
s5
JO
-
i
3 + X IOC
i d 600
3< '0 3
4
5
6
'
2;
10
e
Weight percent beryllium
Copper-rich portion of the copper-beryllium illoy system. (Frorji Metals Handbook," 1948 ed., p. 1176, American Society for Metala, Metals Park, Ohio,) ig
.
12-16
"
1
(a) 1
3.-
structure, and subsequent aging will allow precipitation of the y phase as very fine particles throughout the a matrix (Fig. 12-17Jb). The use of too high an aging temperature causes grain-boundary coarsening typical of
r
1 3; .
:
1
,
5*
overaging (Fig. 12'17c). Under certain conditions, the (3 phase comes out
of solution during solidification of the ingot. The primary p usually per- «|
:
sists during subsequent processing and appears as densely populated Ja| bands known as /3 stringers (Fig. 12-17 d). The mechanical properties Oi -W
i
.
two beryllium-bronze alloys are shown in Table 6-3.
iBl
Beryllium bronzes are used for parts requiring a combination of excellent IbI formability in the soft condition with high yield strength, light fatigueoMp strength, and creep resistance in the hardened condition (many springs); JB
V
S
t
US
"
ft.
parts requiring corrosion resistance, high strength, and relatively high
Microstructures of beryliium bronze
electrical conductivity (diaphragms, contact bridges, surgical instruments, »yhaqueous SO|Ulion of ammonium
.
Etched
persuifate and ammc
bolts, and screws); hard parts that will wear well against hardened steel i-apawm hydroxide, aoox. (a) siow-cooied and aged for 3 h
(firing pins, dies,-ru nonsparking tools).iih 1 . .
*u *
.
.
or>
/
12-14 Cupromckels These are copper-nickel alloys that contain up to 30 percent ,
»
IIll1600 7 phase (dark) mainly at * 9rain boundaries, (b) MfiSolution-treafed and aged for 3 h at 600oF, Fine particles yin
an a „ (c) solution-treated and aged for 2 h
nickel. The copper-nickel binary phase diagram (Fig. 6-11) Shows complete lypiTOO F. Grain-boundary coarsening typical of an overag solubility, so that all cupronickels are single-phase alloys. They are not:IBl',ructljre- (O P stringers" shown as densely populated "
susceptible to heat treatment and may have their properties altered only by;;!
cold working.
If
IWidS. The phase, originating in the cooling of the ingc S»not attacked by the etchant and appears white against ; iWtened a matrix. (The Beryllium Corporation.)
i
mmm
.
m 1
-
:
; t a.
.
-
'
URGY
.
NONFERROUS METALS AND ALLOYS
479
The cupronickel alloys have high resistance to corrosion fatigue and also
= 2000 1
|g H'
L
1800
(3 + i
high resistance to the corrosive and erosive action of rapidly moving sea water They are widely used for condenser, distiller, evaporator, and heat.
exchanger tubes for naval vessels and coastal power plants.
A copper-25 percent nickel alloy is used for cladding on both sides of electrolytic tough-pitch copper in the manufacture of the United States
1600
dime, as shown in Fig. 12-18. 1400
ft
?00 E
r
'
/
1000
V
4r
I
V
.
r-7 7
Vi
,
p
...
1
3
+ y -
800
i
v
.
5
9
d 600 10
v
>1
7: '
if;
aercenl beryllium
1/
v
Ijoiylllum ed p. 1176.
i
t
>: .
1
.
»
.
'
1 -i
:»
.
sty,
ing will allow precipitation of the y phase a»
V
t the a matrix (Fig. 12-17b). The use of toS| auses grain-boundary coarsening typical
sr certain conditions, the (3 phase comes Oji
on of the ingot. The primary f3 usually pef?
f
messing and appears as densely populatedif eft Flg. 12-17 d). The mechanical properties p|
re shown in Table 6-3.
for parts requiring a combination of excelled
tion with high yield strength, light fatigujf (d)
e in the hardened condition (many spring;
iistance, high strength, and relatively hig| agms, contact bridges, surgical instrument
|:-
S that will wear well against hardened stew
6m hydroxide, 300x. (a) Slow-cooled and aged for 3 h
:
oer-nickel alloys that contain up to 30 percenl
.
,
ry phase diagram (Fig. 6-11) shows completl
'
-
,
5
ckels are single-phase alloys. They are no]
and may have their properties altered only
I
If
p ibn-treated and aged for 3 h at 600oF. Fine particles j ifin an a matrix, (c) Solution-treated and aged for 2 h
Hd0°F. Grain-boundary coarsening typical of an overaged Jieture. (d) /3 "stringers" shown as densely populated
Ms. The /3 phase, originating in the cooling of the ingot
,
I Mi
|ji:12-17 Microstructures of beryllium bronze. Etched Pi aqueous solution of ammonium persulfate and ammo-
S?00'F. y phase (dark) mainly at a grain boundaries. (6)
tools).
f
||t0t attacked by the etchant and appears white against a kened a matrix. (The Beryllium Corporation ) .
5
IS
ii
480
INTRODUCTION TO PHYSICAL METALLURGY
ALUMINUM AND ALUMINUM ALLOYS ;
3*i
12'16 Aluminum
The best-known characti
the density being about one-third tf
aluminum alloys have a better strem strength steels. Aluminum has good rosion resistance and high electrica
mmmmm&i
,
pure form of aluminum is used for phc of its high light reflectivity and nonta i
I*
Aluminum is nontoxic nonmagnetic ,
characteristic makes aluminum usel
such as bus-bar housings or enclos Although the electrical conductivi
aluminum is about 62 percent that of i suitable as an electrical conductor fo Fig. 12-18
Pure aluminum has a tensile strengl stantial increases in strength are ob Some alloys properly heat-treated, a
Copper-25 percent nickel cladding on both
sides of electrolytic tough-pitch copper in the United States
dime. Etched in K,Cr207;+ H2S0, + HCI, 50x. (From Metals
,
Handbook, vol. 7, "Atlas of Microstructures," American
psi.
Society for Metals, 1972.)
One of the most important charac ability and workability. It can be cast 12-15 Nickel Silvers These are essentially ternary alloys of copper, nickel, and zinc. Commercial alloys are produced with the following range of composition: copper 50 to 70 percent, nickel 5 to 30 percent, zinc 5 to 40 percent. The nickel silvers containing over 60 percent copper are single-phase alloys that show only fair hot-working properties but are ductile and easily worked at room temperature. The addition of nickel to the copper-zinc
alloy gives it a pleasing silver-blue white color and good corrosion resistance to food chemicals, water, and atmosphere.
These alloys make ex-
cellent base metals for plating with chromium, nickel, or silver. They are used for rivets, screws, table flatware, zippers, costume jewelry, nameplates, and radio dials. The nickel silvers containing between 50 and 60 percent copper are twoi
phase a + j8 alloys. They have a relatively high modulus of elasticity and, like the a + /3 brasses, are readily hot-worked. Nickel silvers are less susceptible to stress corrosion than binary copper-zinc alloys of the same zinc content.
Typical applications of the a + 0 nickel silvers include springs and con- f? tacts in telephone equipment, resistance wire, hardware, and surgical and dental equipment.
The composition and typical mechanical properties of copper and some
copper alloys are summarized in Table 12-4 on pages 482 and 483. i '
i
L
desired thickness, stamped, drawn sp to almost any conceivable shape. Commercially pure aluminum 110C for applications where good formabil sion (or both) are required and where been used extensively for cooking ut nents, food and chemical handling i ,
,
assemblies.
12 17 Alloy Designation System The des wrought aluminum alloys was standai in 1954. It follows a four-digit numbe
the alloy group (Table 12-5). The se the original alloy or impurity limits; ze integers 1 through 9 indicate alloy rr minimum aluminum purities of 99.00 p are the same as the two digits to the r mum aluminum percentage when it i: cent.
Thus 1060 indicates a material
purity and no special control on indiv In the 2xxx through 8xxx alloy groi identify the different aluminum alloys
ILURGY
NONFERROUS METALS AND ALLOYS
481
INUM AND ALUMINUM ALLOYS ;
>
Aluminum The best-known characteristic of aluminum is its light weight
,
the density being about one-third that of steel or copper alloys. Certain
aluminum alloys have a better strength-to-weight ratio than that of high?.
strength steels Aluminum has good malleability and formability,high corIkosion resistance, and high electrical and thermal conductivity. An ultraptlreform of aluminum is used for photographic reflectors to take advantage of its high light reflectivity and nontarnishing characteristics. ft Aluminum is nontoxic nonmagnetic, and nonsparking. The nonmagnetic
i5
:
.
,
characteristic makes aluminum useful for electrical shielding purposes
such as bus-bar housings or enclosures for other electrical equipment.
Mthough the electrical conductivity of electric-conductor (EC) grade aluminum Is about 62 pei cent that of copper, its light weight makes it more
i
V!
suitable as an electrical conductor for many industrial applications.
i
'
Pure aluminum has a tensile strength of about 13,000 psi. However, sub-
; ng on both « United Slalos
stantial increases in strength are obtained by cold working or alloying.
M (From Metals
Some alloys, properly heat-treated, approach tensile strengths of 100,000
American '
;
,
'
.
'
: :; r -
psi.
One of the most important characteristics of aluminum is its machinT ability and workability. It can be cast by any known method, rolled to any desired thickness, stamped, drawn, spun, hammered, forged, and extruded
d
ssentially ternary alloys of copper, nickel produced with the following range of com it, nickel 5 to 30 percent, zinc 5 to 40 pen ng over 60 percent copper are single-phasd
to almost any conceivable shape.
working properties but are ductile and ei e The addition of nickel to the copperjr-blue white color and good corrosion n iter, and atmosphere. These alloys makes
sion (or both) are required and where high strength is not necessary. It has been used extensively for cooking utensils, various architectural components, food and chemical handling and storage equipment, and welded
Commercially pure aluminum, 1100 alloy (99.0+ percent AI), is suitable for applications whefe good formability or very good resistance to corro-
*
.
-
.
imping with chromium nickel, or silver. They; ,
ile flatware, zippers, costume jewelry, na
assemblies.
17
Alloy Designation System The designation of wrought aluminum and wrought aluminum alloys was standardized by The Aluminum Association in 1954. It follows a four-digit numbering system. The first digit indicates
ng between 50 and 60 percent copper are
ave a relatively high modulus of elasticity ai eadily hot-worked. Nickel silvers are less Si
than binary copper-zinc alloys of the same ill v;::-8 a + nickel silvers include springs and nt, resistance wire, hardware, and surgical
.
..
cal mechanical properties of copper and Si ed in Table 12-4 on pages 482 and 483 .
,
.
:
i
1
the alloy group (Table 12-5). The second digit indicates modification of the original alloy or impurity limits; zero is used for the original alloy, and integers 1 through 9 indicate alloy modifications. In the 1xxx group for minimum aluminum purities of 99.00 percent and greater, the last two digits are the same as the two digits to the right of the decimal point in the minimum aluminum percentage when it is expressed to the nearest 0.01 percent. Thus 1060 indicates a material of 99.60 minimum percent aluminum purity and no special control on individual impurities.
In the 2xxx through Sxxx alloy groups, the last two digits serve only to identify the different aluminum alloys in the group.
IB
;
00
-I 33
o O c
TABLE 12 4 Chemical Composition and Typical Mechanical Properties of Copper and Some Copper Alloys*
o H
WROUGHT ALLOYS
o z
MATERIAL
YIELD STRENGTH
TENSILE STRENGTH
COMPOSITION, %
0 5% OFFSET
HARDNESS
% IN 2 IN.
ROCKWELL B
,
o
,
.
,
FORM
-i
ELONGATION
-
,
1 000 PSI
1 000 PSI
"
D
,
-
Cu
OTHERS
Sn
Zn
HARD
SOFT
HARD
SOFT
<
CO
HARD
SOFT
HARD
4
50
58
SOFT
O >
Copper Gilding metal
Sheet
99.9+
Sheet
95.0
Commercial bronze
Sheet
90.0
Red brass
Sheet
85.0
Low brass
Sheet
80.0
55
32
48
50
55
35
45
11
5
38
68
7
10.0
67
37
53
11
3
40
75
10
15.0
80
45
55
15
4
43
85
10
20.0
85
43
65
15
4
50
86
11
15
.
Rod
80.0
20.0
80
45
60
15
5
50
Spring brass
Sheet
75.0
25.0
80
47
60
15
5
45
87
Brass
Sheet
70.0
30.0
86
45
65
15
4
50
87
Sheet
89.0
31.0
85
46
65
15
4
58
87
22
Sheet
65.0
35.0
90
45
70
15
5
60
85
30
Muntz metal
Sheet
60.0
40.0
80
57
60
15
9
48
87
42
Phosphor bronze
Sheet
96.0
40
0 25 P
90
45
75
18
4
50
90
30
Sheet
92.0
80
+P
110
60
85
25
3
55
99
45
Tube
76.0
2 Al
83
62
75
16
17
52
86
33
8 Al
134
76
100
30
13
55
99
69
85
60
50
25
20
45
90
25
100
53
98
18
3
60
95
13
62
54
39
15
30
40
55
110
55
83
12
4
60
Aluminum brass
"
.
22.0
92.0
Aluminum bronze
1 Al 1 Mn, 1 Fe
Manganese bronze Admiralty metal
Rod
68.0
29.0
Tube
71.0
21.0
10
Naval brass
Rod
60.0
39.0
0 75
Silicon brass
78.0
20.0
Tin brass
88.0
10.0
,
.
.
0 25 Pb .
2 0 Si .
20
85
.
"
r
33
<
-
Cartridge brass
.
>
Q
Yellow brass
.
m
3
86
CASTING ALLOYS COMPOSITION %
MATERIAL
BHN
,
TENSILE STRENGTH
,
,
YIELD STRENGTH
.
Cu
Zn
Sn
OTHERS
1 000 PSI
ELONGATION
,
500 kg,
1 000 PSI
%
30
17
45
28
16
22
65
30
18
18
75
,
,
10 mm
Cond. copper
99.85
Brass
70.0
30.0
Tin brass
63.0
36.0
Silicon brass
81.0
15.0
4 Si
90
45
Aluminum brass
16
63.0
32.5
2 5 Al
62
35
Bronze
88.0
18
12.0
+P
40
22
11
81.0
19.0
+P
70
35
25
12
135
Gear bronze Leaded red brass
10 .
.
40
120
88.0
40
55
2 5 Ni
42
17
32
85.0
50
50
5 Pb
75
34
17
25
60
.
.
.
.
.
"
"
rreu
r
ur ci&t>
'
i-a
W.KJ
Sheet
80.0
20.0
85
43
65
15
4
50
Rod
80.0
20.0
80
45
60
15
5
50
Spring brass
Sheet
75.0
25.0
80
47
60
15
5
Brass
Sheet
70.0
30.0
86
45
65
15
4
Cartridge brass
Sheet
89.0
31.0.
85
46
65
15
Yellow brass
Sheet
65.0
35.0
90
45
70
15
Muntz metal
Sheet
60.0
40.0
80
57
30
Phosphor bronze
Sheet
96.0
90
45
Sheet
92.0
Tube
76.0 .
Low brass
86
11
45
87
15
50
87
4
58
87
22
5
60
85
30
15
9
48
87
42
75
18
4
50
90
30
"
Aluminum brass
40
0 25 P
80
+P
.
.
.
22.0
92.0
Aluminum bronze Rod
88.0
29.0
Tube
71.0
21.0
10
Naval brass
Rod
0 75
80.0
39.0
78.0
20.0
Tin brass
88.0
10.0
60
B5
25
3
55
99
45
83
62
75
16
17
52
86
33
8 Ai
134
76
100
30
13
55
99
69
85
60
50
25
20
45
90
25
100
53
38
18
3
60
95
13
62
54
39
15
30
40
55
110
55
83
12
4
60
1 Al, 1 Mn, 1 Fe
Manganese bronze Admiralty metal Silicon brass
110
2 Al
.
.
0 25 Pb .
2 0 Si .
20
3
85
.
i
86
m
& g
CASTING ALLOYS
BHN,
COMPOSITION, % MATERIAL Cu
Zn
Cond. copper
99.85
Brass
70.0
30.0
Tin brass
63.0
36.0
ELONGATION
1 000 PSI
1 000 PSI
0/
30
17
45
28
16
22
30
18
18 16
,
10 .
,
,
500 kg
,
10 mm
/o
40 65
75
.
,
.
Silicon brass
81.0
15.0
4 Si
90
45
63.0
32.5
2 5 Al .
62
35
18
Bronze
88.0
12.0
+P
40
22
11
70
81.0
19.0
+P
35
25
12
135
120
Gear bronze
88.0
40
55
2 5 Ni
42
17
32
75
Leaded red brass
85.0
50
50
5 Pb
34
17
25
60
10 Pb
35
17
20
65 90
Silicon bronze Aluminum bronze
Manganese bronze .
OTHERS
YIELD STRENGTH,
Aluminum brass
.
.
Nickel silver
.
Cupronlckel
.
.
10.0
80.0
.
Sn
TENSILE STRENGTH,
.
95.0
10
4 Si
55
22
35
93.0
40
2 5 Si, 0.5 Fe
50
18
20
.
.
.
89.0
10 Al, 1 Fe
67
32
88.0
9 Al, 3 Fe
80
35
-
68.0
20.0
4 Mn 5 Al, 2.5 Fe
110
70
15
24.0
4 Mn, 5 Al 3 Fe
115
70
15
60.0
20.0
20 Ni
45
20
35
70.0
,
,
64
33
35
40
140
m
3)
o C CO
210 .
30 Ni
.
z
25
64.0
z
o
55 120
S m
> 05 >
.From
S. L Hoyt
,
"
"Metal Data.
Van Nostrand Reinhold Company, New York, 1952.
Z
o > r;
O <
-
O)
09 U
. . .
I
484
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 12 5
Designation for Alloy Groups* -
ALUMINUM
Si Aiuminum, 99.00% and greater, major alloying element Copper Manganese
H3: Strain-hardened and Then £
training magnesium which are given their properties. The degree of str?
ASSOCIATION NO.
2xxx
bilizing treatment is indicated in the W: Solution Heat-treated An ur
3xxx
which spontaneously age at room
Silicon
4xxx
Magnesium
5xxx
ment. This designation because o the period of aging is indicated: for
r
Ixxx
-
,
Magnesium and silicon
6xxx
Zinc
7xxx
Other element
8xxx
Unused series
9xxx
T: Thermally Treated Applies t without supplementary strain hardei T is followed by the numerals 2 thn cific combination of basic operation tions resulting in significantly differf indicated by adding one or more dig T2: Annealed (cast products only) -
-
'
The Aluminum Association. <
1;M8 Temper Designation The temper designation follows the alloy designation and is separated from it by a dash. The Aluminum Association Temper Designation System, adopted in 1948, is used for wrought and cast alu
-
T3: Solution heat-treated and then
-
minum and aluminum alloys. It is based on the sequences of basic treat- }
ments used to produce the various tempers.
,
|
T4: Solution heat-treated and na
-
condition
T5: Artificially aged only. Applies 1 after an elevated-temperature rapid-c -
The standard temper designation system consists of a letter indicating j the basic temper. Except for the annealed and as-fabricated tempers, it is more specifically defined by the addition of one or more digits. There are four basic tempers: F as fabricated, 0 annealed, H strain-hardened, andT
ing or extrusion T6; Solution heat-treated and then
-
T7: Solution heat-treated and then
-
heat-treated. -
F: As Fabricated
the temperature and time conditions
Applied to products which acquire some temperas
the result of normal manufacturing operations. There is no guarantee o!
5
mechanical properties.
is carried beyond the point of max! growth and/or residual stress T8: Solution heat-treated cold-wo
-
-
0: Annealed, Recrystallized
This is thesoftesttemperof wroughtalloy
,
T9: Solution heat-treated artificial
-
,
products. ;. '
H: Strain-hardened This applies to products which have their mechani-
-
cal properties increased by cold working only. The -H is always followed by
T10: Artificially aged and then cc lowed by cold working to improve stre -
two or more digits. The first digit indicates the specific combination oi |
1219 Aluminum-Copper Alloys (2xxx Ser aluminum copper equilibrium diagrar mum solubility of copper in aluminu cold work performed, with the numeral 8 representing the full-hard condi-: | solubility decreases to 0.45 percent e tion. Therefore, half hard is -H14, quarter hard is -H12, etc. Extra hard I between 2.5 and 5 percent copper w tempers are designated by numeral 9. A third digit is often used to indicate | hardening. The theta (fl) phase is an position corresponds closely to the ( properties. is carried out by heating the alloy in H2: Strain-hardened Then Partially Annealed Applied to products thai/j| followed by rapid cooling. Subsequ are cold-worked to a harder temper and then have their strength reduce will allow precipitation of the 0 phasi to the desired level by partial annealing. The residual amount of cold wofo alloy. These alloys may contain small is designated by the same method as the -HI series.
basic operations as follows: HI: Strain-hardened Only -
;-: The second digit designates the amounto! ;,
the degree of control of temper or to identify a special set of mechanic | -
um, manganese, chromium and zinc. ,
i all*-
li
-
1
JRGY
NONFCRROUS METALS AND ALLOYS
for Alloy Groups*
485
2xxx
H3: Strain-hardened and Then Stabilized Applied only to alloys contraining magnesium which are given a low-temperature heating to stabilize their properties. The degree of strain hardening remaining after the stabilizing treatment is indicated in the usual way by one or more digits. W: Solution Heat-treated An unstable temper applicable only to alloys
3xxx
which spontaneously age at room temperature after solution heat treat-
-
ALUMINUM
I!; -
y*
ASSOCIATION NO.
' -
i it
greater,
St
Ixxx
-
itient. This designatlOh because of natural aging, is specific only when the period of aging is indicated: for example, 2024-W (V2 h).
4xxx
,
5xxx S
6xxx
.
t: Thermally Treated Applies to products thermally treated, with or
-
7xxx
without supplementary strain hardening, to produce stable tempers. The
8xxx
r -
9xxx
tions, resulting in significantly different characteristics for the product, are indicated by adding ohe or more digits to the basic designation.
smper designation follows the alloy de?! by a dash. The Aluminum Association Te
1
'
T is followed by the numerals 2 through 10, inclusive, designating a spe-
cific combination of basic operations. Deliberate variations of the condi-
r
-
-
Mxd in 1948, is used for wrought and cast l It is based on the sequences of basic ti
T2: Annealed (cast products only) t3: Solution heat-treated and then cold-worked
-
"
v;
X;
various tempers.
.
Ignation system consists of a letter indlcafi
-
,
T4: Solution heat-treated and naturally aged to a substantially stable
condition
t5; Artificially aged only. Applies to products which are artificially aged
after an elevated-tempei-ature rapid-cool fabrication process, such as casting or extrusion
r the annealed and as-fabricated tempers, It the addition of one or more digits. There ricated, 0 annealed, H strain-hardened, andl
-
T6: Solution heat-treated and then artificially aged T7: Solution heat-treated and then stabilized: applies to products where
-
ed to products which acquire some temper
icturing operations. There is no guarantee
5.
the temperature and time conditions for stabilizing are such that the alloy is carried beyond the point of maximum hardness providing control of growth and/or residual stress T8: Solution heat-treated cold-worked, and then artificially aged T9: Solution heat-treated artificially aged, and then cold-worked ,
-
ized
This is the softest temper 01 wrought al
,
-
,
T10: Artificially aged and then cold-worked lowed by cold working to improve strength -
is applies to products which have their mech cold working only. The -H is always followed irst digit indicates.the specific combinatiort
ily The second digit designates the amouiiti the numeral 8 roprosenting the full-hard corii is -HI4, quarter hard is -HI2, etc. Extra hi numeral 9 A third digit is often used to indlCi
'
119 Aluminum-Copper Alloys (2xxx Series)
,
the same as -T5 but fol-
The aluminum-rich end of the
aluminum copper equilibrium diagram is shown in Figure 12-19
.
The maxi-
mum solubility of copper in aluminum is 5 65 percent at 1018oF and the solubility decreases to 0.45 percent at 572' F Therefore alloys containing .
,
.
,
between 2.5 and 5 percent copper will respond to heat treatment by age
.
v Imper or to identify a special set of mechani ;;
{ien Partially Annealed Applied to products t!
;
ier temper and then have their strength redu ;ial annealing. The residual amount of cold i method as the -H1 series.
hardening. The theta (()) phase is an intermediate alloy phase whose composition corresponds closely to the compound CuA .; Solution treatment
is carried out by heating the alloy into the kappa (k) single-phase region followed by rapid cooling Subsequent aging, either natural or artificial, will allow precipitation of the 0 phase thus increasing the strength of the alloy. These alloys may contain smaller amounts of silicon iron, magnesi.
,
,
um, manganese chromium, and zinc. ,
486
INTRODUCTION TO PHYSICAL METALLURGY
Figure 12-20 shows the structu
800 1400
after solution heat treatment and
1200
and Al in a matrix of aluminum-rit
1000 o
tion as a result of the heat treatment
mainly of black insoluble particles
700
particles of CuAI2 (white, outlined)
k + L
500
5 65 .
» 500
The three most widely used wro
3
2017
Z3
800 g E 400
.
,
and 2024.
The oldest of all
duralumin (2017) containing 4 perc(
a:
rivets in aircraft construction Sine* .
+ 0
600
tion treatment it is refrigerated to p
300
solution-treated condition it has gi ,
I
400
be easily formed. Subsequent retui causes precipitation of the 0 phase creasing the nardness and strength Alloy 2014 has higher copper anc susceptible to artificial aging. In tt higher tensile strength much highc
200
00
.
2
3
4
5
s
6
3
10
Weight percent copper
Fig. 12-19 Aluminum-rich portion of the copper-aluminum alloy system. (From Metals Handbook, 1948 ed., p. 1160,
,
American Society for Metals, Metals Park, Ohio.)
than 2017. This alloy is used for he truck frames. Figure 12-21 shows thi and thermal transformation typical containing 4.5 percent copper and I
highest strengths of any naturally
Anneoled - 0 50
T S
35,000 psi slonqI 18% conductivity 44 - 4£ .
.J
t
"
I
46
1
44
O
48
.
. ft
-
i
\
42 0.-0
£| 40
Note; Pure alun
38
elonq 35
T S I6,0( .
-
A
S 31 34
65 % IAC _
-
32 -
Fig. 12-20 Alloy 2014-T4 closed-die forging, solutiontreated at 935°F for 2 h and quenched in water at 150
°
F
.
Lor.gitudinal section. Structure consists mainly of dark, insoluble particles {A) of. a complex compound of Fe, Mn, Si, and Al, and a few particles (B) of CuAl., (white, outlined)
I
30 -
Solution heat f reotment - W T S .
.
45,000 psi
elong 24 % conductivity 33% 1ACS
28 1
_
35
40
45
Atlas of
"
Microstructures," American Society for Metals, 1972.)
_
50
Tensile strength
'
in a matrix of a solid solution (C). Etched in Keller s
reagent, 100X. (From Metals Handbook, vol. 7,
.
conducts
Fig. 12-21 Relationship of conductivity strength, and thermal transformation typical of a 2014 aluminum alloy ,
sSIlurgy
NONI ERROUS METALS AND ALLOYS
.
487
Figure 12-20 shows the structure of alloy 2014-74 closed-die forging 1-100
after solution heat treatment and quenchincj in warm water. It consists mainly of black insoluble particles of a complex compound of Fe Mn, Si,
1200
and Al in a matrix of aluminum-rich solid solution. There are just a few particles of CuAI2 (white, outlined) since most of this compound is in solu-
,
tion as a result of the heat treatment.
1000 ii-
5 65 .
The three most widely used wrought aluminum-copper alloys are 2014 and 2024. The oldest of all the heat-treatable aluminum alloys is
s
5
,
2017
,
800 S
K + B
duralumin (2017) containing 4 percent copper. This alloy is widely used for rivets in aircraft construction. Since this is a natural-aging alloy after solution treatment it is refrigerated to prevent aging. As a single phase in the solution-treated condition it has good ductility so that the rivethead may ,
600
,
01
,
ft?
'
400
be easily formed. Subsequent return of the material to room temperature causes precipitation of the 0 phase as small submicroscopic particles increasing the hardness and strength. Alloy 2014 has higher copper and manganese content than 2017 and is ,
4
5
6
7
9
10
ght percent copper
susceptible to artificial aging. In the artificially aged temper, 2014 has a higher tensile strength, much higher yield strength and lower elongation
>per-aiuminum
S;yied., p. 1160
,
no.)
than 2017. This alloy is used for heavy-duty forgings, aircraft fittings, and
truck frames. Figure 12-21 shows the relationship of conductivity, strength, and thermal transformation typical of a 2014 aluminum alloy. Alloy 2024, containing 4.5 percent copper and 1.5 percent magnesium, develops the highest strengths of any naturally aged aluminum-copper type of alloy.
B
1
Anneoled - 0 T S 35,000 psi
50
.
S
a
.
18% conduclmty 44 - 48.5 % IACS
48
Gueroqed
46
<
flT50,000 psi elong 5 % conductivity 42.5-45% IACS
44
42 <1>
v
'
* '
c:
-
T6- artif iciol oqe
a
TS
Notcj Pure nluminum -0
o
.
i s
S
.
38
:
.
Te.odo psi
CO
36.5-40%
II
elonq 35%
a>
58,000 psi
elong 13 % conductivity;
_
IACS
nduchvity
65% ACS
A
34 "
I
-
32
solutlon-
;
30
er at 150T. '
'
nly of dark, :: .
Si;
'\6 of Fo. Mn, ;
-
2
Solution heat treatment-W T S .
.
45,000 psi
TS
33% IACS
.
L
-
_
40
45
1972.)
6
_
_
12-21 Relationship of conductivity strength, and mal transformation typical of a 2014 aluminum alloy. ,
J
1
50
Tensile strength 103 psi
I Keller's s,
65,000 psi elong 14%
28 -
7, "Atlas of
.
concuctivity 28-32% IACS 35
hite, outlined)
14 - natural ac|e
elonq 24 % conductivity
55
60
65
488
INTRODUCTION TO PHYSICAL METALLURGY
The higher magnesium content, compared with 2017, makes it more diffi-
percent nickel (2218) has been developed for applications involving elevated temperatures such as forged cylinder heads and pistons. The only binary aluminum-copper casting alloy is 195, containing 4 per-
" "
-
.
.
.
.
-
.
"
:
cult to fabricate. A combination of strain hardening and aging will develop the maximum yield strength attainable in high-strength alloy sheet. Typical uses of 2024 alloy are aircraft structures, rivets, hardware, truck wheels, and screw-machine products. An aluminum-copper alloy containing 2
i
*
%
.
m
m
cent copper.;: When properly heat-treated, this alloy has an excellent combination of strength and ductility. Alloy 195, sand-cast, is used for flywheel and rear-axle housings, bus wheels, aircraft wheels, and crankcases. Several casting alloys are produced that contain approximately 8 percenj copper. These alloys, 112, 113, and 212, may contain substantial controlled additions of silicon, as well as iron and zinc. The presence of silicon increases fluidity, so that alloys 113 and 212 are preferred forthin-sectioned castings such as housings, cover plates, and hydraulic brake pistons. A series of casting alloys such as 85, 108, 319, and 380, classed as alumi-
.
80
6G0 a
+
I
il 650
I
V ;
ii
630
1
620
Al
!
2 Weight percent m<
fig. 12.23 Aluminum-rich portion of the aluminummanganese alloy system. (From "Metals Handbook IE ed, p 1163. American Society for Melals Metals Park, "
,
1
,
num-copper-silicon alloys, have been developed containing less than 5 | Ohio) percent copper and from 3 to 8 percent silicon. Figure 12-22 shows the 12 20 typical structure of a 380 alloy, die-cast, that has desirable properties. The copper provides higher strength and better machining properties than the straight aluminum-silicon alloys, while the silicon provides better casting and pressure-tightness properties than the aluminum-copper alloys. Typical applications include brackets, typewriter frames, manifolds, valve bodies, oil pans, and gasoline and oil tanks.
Aluminum-iVlanganese Alloy 5 (Sxxx of the aluminum-manganese alloy s; maximum solubility of manganese in tectic temperature of 12160F Althoi creasing temperature alloys in this gr .
,
Because of the limited solubility mat ing element in any casting alloys anc One of the alloys in this group is th( ,
formability, very good resistance to c cal applications are uxensils food equipment gasoline and oil tanks, pi 12 21 Aluminum-Silicon Alloys (4xxx Serie ,
,
,
. . ,
vwi
s**-.
,
»< /
fC .
m "
.
vX".S
aluminum-silicon alloy system is sh solubility of silicon in the a solid sc temperature of 1071 . Although the lower temperatures these alloys are c ,
alloy 4032, containing 12.5 percent j coefficient of thermal expansion It i Aluminum-silicon casting alloys ha to corrosion. Alloy 13 (12 percent si .
»
Fig. 12-22 Alloy 380 die casting. Area near a machined surface {A) shows structure typical of a casting that has desirable properties; interdendritic particles of eutectic
silicon (8) and CuAI2 (C) in a matrix of aluminum solid solution (O). Etched in 0.5 HF, 260x. (From Metals Handbook, vol. 7, "Atlas of Microstructures," American Society
are used for intricate castings fittings.
,
foe
12 22 Aluminum-Magnesium Alloys (Sxxx of the aluminum-magnesium system the solvus line shows a considerable
for Metals, 1972.)
i' ;
mm
i .
-
URGY
NONFERROUS METALS AND ALLOYS
489
f
v
9nt, compared with 2017, makes it more, ion of strain hardening and aging willdeye -x-
attain able in high-strength alloy sheet. Tj
68C 1250
4'
570
ift structures, rivets, hardware, truck wh( „ s An aluminum-copper alloy containirtflp; len developed for applications involving w
P + i 66C
.
3
jrged cylinder heads and pistons. ' ;||| Icopper casting alloy is 195, containing 4 pS, ility. Alloy 195, sand-cast, is used for flywhi '
a
'
/
13
.
1190 S10
a + B
,,
V.-lheels, aircraft wheels, and crankcases.
:" Produced that contain approximately 8 perr
+
Ll_
S50
[heat-treated., this alloy has an excellent cbf .
1220
530 '
-
'
M60
: '
113, and 212; may contain substantial c6fl||
520
well as iron and zinc. The presence of slliwHT
'
Al
$
2
A
S
Weight percent manganese
s 113 and 212 are preferred forthin-sectioi
12'23 Aluminum-rich portion of the aluminum-
ver plates, and hydraulic brake pistons, :
1
gshono mloy syalom (Fiom "Moiohs iinnclbook," 1940
,ch as 85, 108, 319, and 380, classed as alU
,
(). 1163, American Society for Melals, Metals Park,
ave been developed containing less thartf
p 8 percent silicon. Figure 12-22 shows flj Pf-ty, die-cast, that has desirable properties. THn gth and better machining properties than tfl :
26 Alurhinum-Manganese Alloy 5 (3xxx Series)
The aluminum-rich portion
of the aluminum-manganese alloy system is shown in Figure 12-23. The maximum solubility of manganese in the a solid solution is 1.82 at the eu-
jys, while the silicon provides better casting,
tectic temperature of 12160F. Although the solubility decreases with decreasing temperature, alloys in this group are generally notage-hardenable. Because of the limited solubility, manganese is not used as a major alloying element in any casting alloys and is used in only a few wrought alloys. Ohe of the alloys in this group is the popular 3003 alloy, which has good formability very good resistance to corrosion, and good weldability. Typi-
rties than the aluminum-copper alloys. Typj|| '
1
f
,
cal applications are utensils, food and chemical handling and storage oquipmonl qnsolino ;in(l oil Innks, profismo vrssols, and piping. ,
1J-21 Aluminum-Silicon Alloys (4xxx Series) The aluminum-rich portion of the aluminum-silicon alloy system is shown in Figure 12-24. The maximum ,
solubility of silicon in the a solid solution is 1.65 percent at the eutectic temperature of 1071oF. Although the solvus line shows lower solubility at lower temperatures these alloys are generally not heaWreatable. Wrought alloy 4032 containing 12.5 percent silicon, has good forgeability and low ,
3
,
coefficient of thermal expansion. It is used for forged automotive pistons Aluminum-silicon casting alloys have excellent castability and resistance to corrosion. Alloy 13 (12 percent silicon) and alloy 43 (5 percent silicon) are used for intricate castings food-handling equipment, and marine .
machined
I that has
,
fittings.
eutectic im solid letals Hand-
can Society
1
2-22
Aluminum-Magnesium Alloys (5xxx Series) The aluminum-rich portion of the aluminum-magnesium system is shown in Figure 12-25. Although Ifu; solvus lino shows n consklomblo diop in iho soluhilily ol magnoslum
*. .
:
490 INTRODUCTION TO PHYSICAL METALLURGY
700
700i 1200
;
.
i 11,6
1 65
/
10O0
.
a
L
i
L
600
I
a
« 500
+
600
3 ,
a
e
8 a)
a)
I
400
r
500
300
"
1200
liOOk1
85
oT =3
Hiooo ?
Q
.
a + MgoSi
500
900
d 400
'
> > -
.
200
2
Al
6
4
8
12
10
16
14
Weight percent silicon 800
Fig. 12-24 Aluminum-rich portion of the aluminum-silicon
400 Al
Metals Handbook, 1948 ed, p. 1166, "
alloy system. (From
"
American Society for Met&ls, Metals Park, Ohio.)
in aluminum with decreasing temperature, most commercial wrought alloys in this group contain less than 5 percent magnesium, and, with low J '
heat-treatable.
silicon content, they are not
4
8
12
16
Percent magnesium siliade
The wrought alloys are characterized by good weldability,
Fig. 12-26 Aluminum-rich portion of the aluminummagnesium silicide system. (From "Metals Handbool< 1948 ed., p. 1246 American Society for Metals Metals ,
,
Park, Ohio.)
good cor-
rosion resistance, and moderate strength. Alloy 5005 (0.8 percent mag nesium) is used for architectural extrusions; alloy 5050 (1.2 percent mag nesium) for tubing and automotive gas and oil lines; alloy 5052 (2.5 percent -
magnesium) for aircraft fuel and nesium) for marine and welded s (5.2 percent magnesium) for insect
-
use with magnesium alloys. BOOr
The aluminum-magnesium castir magnesium) alloy 218 (8 percent r magnesium). The first two are used fittings for chemical and sewage ut brake shoes. Alloy 220 is the only c able, resulting in the highest mecha casting alloys. The casting properti they require careful foundry practice 12 23 Aluminum-Silicon-Magnesium Alloy con combine to form a compound turn forms a simple eutectic syster ,
l
_
700
u-
1300
L -
=moo
500
m4
:
+
i
L
i 900
a 500 3 -
a
1:
TRi
i
e 400
300
5«;
+ 13
portion of the AI-Mg2Si system is shi
200 -
100 Al
4
2
6
8 10 12 14 Weight percent magnesium
Fig. 12-25 Aluminum-rich portion o! the aluminum-
"
"Metals Handbook. 1948 for Metals. Metals Park, ed., p. 1163, American Society Ohio.)
magnesium alloy system. (From
i
r
i:
16
18
300
20
of the Mg2Si after artificial aging (te reach their full strength. The wroug Magnesium and silicon are usually f
silicide. The structure of alloy 606 FeaSiAl,-, (gray, scriptlike) and IV ; solution matrix, Figure 12-27. These corrosion resistance and are more i
-7-
URGY
NONFERROUS METALS AND ALLOYS
.
49
700 L
= 1200
L
fl + z -
--
r
11.6
Mq2Si l /
1000 LL.
+ a)
+ /3
7
1200
.
L
500
= 1100
i /i.es
800 2
a
QJ
1000 s
600
a + Mg? Si
500 -
m e
,io
'
-lOO
JOO
16
12
I percen* silicon 800
inum-sihcon
ed., p. 1166, 5 )
400 AI
4 ti | Percent magnesium silicide
.
ig temperature, > most commercial wrought M%s than 5 percent magnesium, and, with tog ;
bat-treat able.!
IG
jjfrg. 12-26 Aluminum-rich portion ol (he aluminummagnesium silicide system. (From "Metals Handbook American Society for Metals, Metals
;
"
,
i|l!M8 ed., p. 1246 jiirk, Ohio.)
,
s 'riaracterized by good weldability, good ::
'v tate strength. Alloy 5005 (0.8 percent mall
magnesium) for aircraft fuel and oil lines; alloy 5083 (4.5 percent magnesium) for marine and welded structural applications; and alloy 5056 (5.2 percent magnesium) for insect screens, cable sheathing, and rivets for use with magnesium alloys. The aluminum-magnesium casting alloys include alloy 214 (3.8 percent magnesium), alloy 218 (8 percent magnesium), and alloy 220 (10 percent magnesium). The first two are used for dairy and food handling equipment,
ural extrusions; alloy 5050 (1.2 percent mag
*
otive gas and oil lines; alloy 5052 (2.5 percent
1300
L
fittings for chemical and sewage use, fittings for marine use, and aircraft brake shoes. Alloy 220 is the only one in this group which is age-hardenable, resulting in the highest mechanical properties of any of the aluminum casting alloys. The casting properties of alloys in this group are poor, and
1100 a + z a>
900 -t
they require careful foundry practice.
14.9
2-23
700
500
a + /3
Aluminum-Silicon-Magnesium Alloys (6xxx Series) Magnesium and silicon combine to form a compound magnesium silicide (Mg2Si), which in turn forms a simple eutectic system with aluminum. The aluminum-rich portion of the AI-Mg2Si system is shown in Figure 12 26. It is precipitation of the Mg2Si after artificial aging (temper T6) which allows these alloys to reach their full strength. The wrought alloys include 6053 6061, and 6063. Magnesium and silicon are usually present in the ratio to form magnesium silicide. The structure of alloy 6061 plate, hot-rolled shows particles of Fe3SiAI12 (gray scriptlike) and Mg2Si (black) in an aluminum-rich solidsolution matrix. Figure 12-27. These alloys are characterized by excellent *
-
300
,
10
12
4
16
18
20
percent magnesium '
,
i v linum'
'
-
,
Jbook," 1948
pis Park,
4
,
corrosion resistance and are more workable than other heat-treatable al-
. .
..
492 INTRODUCTION TO PHYSICAu METALLURGY
)
A
.
i
m
B
J
mmmmm
Fig. 12-27 Alloy 6061 plate, as hot-rolled, longitudinal section. Particles {A) of Fe.,SiAI,, (gray, scriptlike) and
Mg,Si (black) (S) in a matrix of aluminum-rich solid solut
Etched in 0.5 percent HF, 250x. (From Metals Handbook
ion.
Fig. 12-29
vol. 7.
Alloy 7075-0 sheet annealed. Structure ,
of fine and coarse particles (A) of MgZn, (black) anc insoluble particles (B) of FeAI., (light gray outlined)
.
Atlas of Microstructures, Ameiican Society for "
"
,
Metals, 1972.)
matrix of aluminum-rich solid solution. Etched In 2
cent nitric acid, 500x. (From Metals Handbook vol. Atlas of Microstructures American Society for Me 1972.) ,
V
furniture,
i
vacuum-cleanertubing, bridge railings, and architectural applications. The aluminum-silicon-magnesium casting alloys 355, 356, and 360 pro-
!
loys. Typical applications include aircraft landing mats, canoes,
vide a desirable combination of castability, pressure-tightness, strength,
and corrosion resistance. In the heat-treated condition their mechanical
properties approach those of the aluminum-copper alloys. They are widely used in aircraft applications, machine-tool parts, and general-purpose castings.
12-24 Aluminum-Zinc Alloys (7xxx Series) The aluminum-rich portion of the aluminum-zinc alloy system is shown in Figure 12-28. The solubility of zinc i 700 L
-
U200
t
500
H1000
500 800 p
a
400 Q
500
.
i
E300
i
31,6
200
400
/
a+B
100 10
20
30
40
DO
"
"
,
in aluminum is 31.6 percent at 52 Commercial wrought alloys cont smaller additions of manganese i zinc, 2.5 percent magnesium, 1.5 zinc, 3.3 percent magnesium, 0.6 | cent zinc, 2.7 percent magnesium, tensile strengths obtainable in alu structure of alloy 7075-0 sheet, a particles of MgZn2 (black) and afe1 outlined) in a matrix of a(aluminurr The susceptibility of these alloys by the addition of chromium and t in applications requiring high stren as aircraft structural parts. The aluminum-zinc casting alloy zinc, 0.6 percent magnesium, 0.5 tanium, provides high mechanical This alloy also has fair casting cha and very good machinability. It is u and radio equipment. 12 25 Corrosion Resistance of Aluminum
Weight percent zinc
Fig. 12-28 Aluminum-rich portion of the aluminum zinc
sion resistance of aluminum is du
-
alloy system. (From Metals Handbook," 1948 ed., p. 1167, American Society for Metals, Metals Park, Ohio.) "
ii
oxide film that forms immediately o
j
NONFERROUS METALS AND ALLOYS
4i
GY
*
ML:
m
m udinal
3) and )lid solution. handbook
Fig. 12-29
Alloy 7075-0 sheet
,
annealed.
Structure consists
|;0f fine and coarse particles [A) of Mg2n2 (black) and a few Insoluble particles (S) of FeAI (light gray, outlined) in a
,
iety for
.
,
'
' .
matrix of aluminum-rich solid solution. Etched in 25 perMhl nitric acid, 500x. (From Metals Handbook, vol. 7, "
-
:
Allns of Microslrucluros
ide aircraft landing mats, canoes, furniture, railings, and architectural applications.
" .
Amoncan Socioly lor Motals.
1972.)
-
;
sium casting alloys 355, 356, and 360 prog
in aluminum is 31.6 percent at 527CF, decreasing to 5.6 percent at 257CF. Commercial wrought alloys contain zinc, magnesium, and copper with smaller additions of manganese and chromium. Alloy 7075 (5.5 percent zinc, 2.5 percent magnesium, 1.5 percent copper), alloy 7079 (4.3 percent zinc, 3.3 percent magnesium, 0.6 percent copper), and alloy 7178 (6.8 percent zinc, 2.7 percent magnesium, 2.0 percent copper) develop the highest
Df castability, pressure-tightness, strength;' he heat-treated condition their mechanical
ie aluminum-copper alloys. They are widely
V
machine-tool parts, and general-purpose
ieries) The aluminum-rich portion of
the
tensile strengths obtainable in aluminum alloys. Figure 12-29 shows the structure of alloy 7075-0 sheet, annealed. It consists of fine and coarse particles of MgZn2 (black) and a few insoluble particles of FeAla (light gray,
ihown in Figure 12-28. The solubility of zinc
outlined) in a matrix of a (aluminum-rich) solid solution.
The susceptibility of these alloys to stress corrosion has been minimized by the addition of chromium and by proper heat treatment. They are used in applications requiring high strength and good corrosion resistance, such as aircraft structural parts.
of
The aluminum-zinc casting alloy known as 40E containing 5.5 percent zinc, 0.6 percent magnesium 0.5 percent chromium, and 0.2 percent titanium, provides high mechanical properties without solution treatment. This alloy also has fair casting characteristics good, corrosion resistance, and very good machinability. It is used for aircraft fittings turret housings, and radio equipment.
O)
,
,
mo
,
,
1.
12-25 Corrosion Resistance of Aluminum and Aluminum Alloys The high corrosion resistance of aluminum is due to the self-protecting, thin, invisible oxide film that forms immediately on exposing surfaces to the atmosphere.
ium-zinc
!d., p. 1167, )
L
'
494
INTRODUCTION TO PHYSICAL METALLURGY
This film protects the metal from further corrosion. If the oxide film is re-
TABLE 126
moved, in many environments, a new film will form immediately and the metal remains fully protected. In certain strongly acid or alkaline solutions, or in contact with moist corrosive materials that prevent access of oxygen to the aluminum surface, the protective film does not form readily. Therefore, the aluminum should be adequately protected or not used at all. A relatively thick oxide coating on aluminum and aluminum alloys may be produced by placing the metal into an aqueous solution containing 15 to 25 percent sulfuric acid. This process, known as anodizing, produces a clear, transparent coating containing submicroscopic pores that are usually sealed before use to prevent absorption and staining. Sealing may be accomplished by suitable heating in hot water. The corrosion resistance of aluminum-copper alloys and aluminum-zinc alloys is satisfactory for most applications but is generally lower than that of the other aluminum alloys. Under certain corrosive conditions they are subject to intergranular corrosion. Therefore, these alloys in the form of
Nominal Composition and Typical Me(
Alloys*
NOMINAL COMPOSITION % ,
ALLOY AND TEMPER
T( Si
Cu
Mn
Mg
Or
Ni
SI 1 1 ,
EC
(99.45 ( % aluminum) -
0
12
H14
16
1100
(99.0 i % aluminum) 0
-
13
H14
-
-
18
HI 8
24
2014
3 8
44
,
.
08 .
0.4 27
-
14
62
76
70
2017
08
4 0
.
.
05 .
0.5
0.1
0
-
26
74 2024
62 0 5 ,
4 5 ,
06 .
1.5
01 .
O
-
27
14
-
221 8
68 0 2 ,
4 0
.
1,5
2 0
161
3U03
5,
59 0 6 .
-
.
1.2
O
16 22
1032
7i'.
12.5 -
mm
7
>
09 ,
.
1.0
09 .
7c
55
,
5005
3 4
08
,
.
18 H34
23
-
5050
3 4
12
,
.
0
2'
-
H34 -
28
303"
-
25 ,
0.25
0
28
H34
38
-
0 1 .
9
. -
sr.
-
5.2
0 1 ,
0
42
H18
63
5083
0 7 .
4.5
0 3313
42 3 3
0 25
1 0
.
.
0.25
0
IS
16
45 04 ,
0 1 ,
0 1 .
07 .
0.1
0.1
0
13
T6
35
-
-
?075
Fig. 12-30 Full cross section of Alclad 2024-74 sheet showing white cladding layer of commercially pure aluminum on top and bottom of sheet, 125x. (Research Laboratory, Aluminum Company of America.)
33-c3"-7 3
-
-;
,
-
,
...
2.5
0.3
5.5 33 83
05 ,
20 .
.
2.7
0.3
6.8
33
33
76
88
-
-
1 5
76
7178
'Compiled
from information in Metals Handbook;' 1961 edition ivUStuelspecimens Vuin. thick, ,
i
3 5
O
Ik.
Ar
I
NONFFnnOUR ME7ALS AND ALLOYS
i further corrosion. . fea
If the oxide film is re-
new film will form immediately and th#
18LE 12-6 Nominal Composition and Typical Mechanical NOMINAL COMPOSITION % AILOY AND
jss of oxygen to the aluminum surface, the
TEMPER
at all.
g on aluminum and aluminum alloys may;| al into an aqueous solution containing 15 '
process, known as anodizing, produces
.
EC
1100
BHN
23
24
22
15
44
44 .
0.8
04 .
0
27
14
18
45
T4
62
42
20
105
70
60
13
135
40 .
0.5
0.1
0
26
10
22
45
62
40
22
105
3 5
45
,
.
0.6
1.5
0.1
C
27
11
22
47
T4
68
47
19
120
59
44
13
02
4.0
...
06
...
1.2
.
1.5
..
...
0
16
6
40
28
H14
22
21
16
40
55
46
9
120
12.5
0.9
...
1.0
09 .
T6
8005
04
0.8
.
O
18
6
H34
23
20
30t 8t
-
-
5050
04 ,
..
21
8
24t
36
H34
28
24
8t
53
5052
2.5
0.25
0
28
13
30
47
H34
38
31
14
68
-
5056
0 1 .
5.2
0.1
O
42
22
35
65
H18
63
59
10
105
42
21
22
-
-
B083
0.7
4.5
0
6061
06
0.25
.
...
10 .
0.25,
0
18
8
30
30
T6
45
40
17
95
-
6063
04
0 1
.
.
0.1
0.7
01 .
0.1
0
13
7
-re
35
31
-
;7075
05
15
.
.
...
2.5
03 .
73
5.5 33
15
ie
60
T6
83
73
11
150
7178
0 5
20
,
-
25
1 ai-
o
-
bli Lnborn-
41
O
-
-
28
1.2
-
-
2.0
T61 .
-
0.5
T4
4032
iheet
14
105
0 0 -
s
4
16
20
T6
m
12
45
-
-
ire alumi-
/HN.-DIAM
5
»17
-
I
SPECIMEN
,
,
17
3003
m
'
1 000 PSI
13
.
-
p
STRENGTH
18
2218
m
,
H14
2024
m
STRENGTH
0
0 8
on. Therefore, these alloys in the form of
m
Ni
H18
plications but is generally lower than thai .
Zn
% IN 2 IN.,
(99.0 + % aluminum)
jminum-copper alloys and aluminum-zinc nder certain corrosive conditions they are
Cr
.H14
-
:
Mg
YIELD
(99.45+% aluminum)
2014
n hot water.
Mn
0
bsorption and staining. Sealing may be ac-:;|
J
Cu
,
Moling submicroscopic pores that are usual- ij .
,
1 000 PSI
-
'
ELONGATION TENSILE Si
jidily. Therefore, the aluminum should beJ
Properties of Some Wrought-aiuminum
Typical mechanical properties
,
ine solutions, or in contact with moist cor-
495
.
...
2.7
0.3
6.8
0
;)3
Ifi
16
T6
88
70
11
tmpiled from information in "Metals Handbook 1961 edition, American Society for Metals "
,
specimens '/u in thick. .
,
Metals Park, Ohio
496 INTRODUCTION TO PHYSICAL METALLURGY
sheet are usually clad with a highnum (1100) or a magnesium-silicoi slabs are mechanically attached to accomplished by hot rolling. The usually IV2 or 21/2 percent of the t
] 5 5
Ul O O O ifi ominOLOLoOLnoo in
CQ
t-
in o o o o o o r- r- co co o o)
I
IO -r-
ft
1 EC HI
c Q
z
I
5?
I
z
c
-
:
.
.
12-30 shows the full cross section
cladding layer is visible at the edge sively used for aircraft applications t high strength and high resistance to The nominal chemical compositi of some wrought- and cast-alumim
SpSSSSSSSSSSSSSSSSS
z
12-7. Table 12-8 shows the relations X
X
ness for some aluminum alloys.
5m
o
5
d tr o
Q
M t-
CM
TABLE 12 8 Relationship of Electrical Con Some Aluminum Alloys*
I
>
LU I
CONDUCTIV
CD CO Q
z
_
ALLOY
M LU o Z X Q
HI H 0.
J
I- W T-
2 o
;S2
z
TEMPER
% lACSt
EC
0
62
1100
0
59
H18
57
0
50
T4
30
T6
40
0
50
T3,T4
30
2014
1 5 a
:
: .
2
Q
a
t
.
2024
N
o
2
:
5 O
3003
co oo o o co co oo o
5
C
9
O O LO p p p -
-
1
2
<
-
a
40
35
H38
35
0
45
T4
40
T6
40
7075
T6
30
B195
T4
33
6061
O z
55
o
S :3
o p LO iri co tri
SS2
q
:
CO
co q
CNJ .
CM
p en oi
in
42
H18 CO
CO
O
50
H12
0
5052
CO
O
.
a
N
3
LO
0
1
11 I'.
T6
33
T51
43
+1-
CD
C
355
O 5
x LU
<
o 0 D Q
oo5ooo5ooo522oo
2 o c
-
i
x il J
4
>
O
Q z
< <
5 LU
_
_
.H-
2 2
LO
00
00
o
CO UD CD CD CO
"T "T 'T
"
CO
.
c
LO
>7 4-
LU
-
LO
-
(NCJCMCOCMCO
CO
CO
i Is
356
J 1
T6
36
T7
42
T51
43
T6
39
T7
40
ooll m
* .
Compiled Irom "Metals Handbook," vol. 1, American Soc tlnternational Annealed Copper Standard .
i if
NONFEPSO'JS METALS AND ALLOYS
'
RGY
1
o o
mi
sheet are usually clad with a high-purity alloy such as commercial aluminum (1100) or a magnesium-silicon alloy of the 6000 series. The coating slabs are mechanically attached to the alloy core ingot, and the bonding is
2
accomplished by hot rolling. The nominal cladding thickness per side is usually Vk or 21/2 percent of the thickness of the base material. Figure
o o o _ o
oo co o> i -
i
12-30 shows the full cross section of Alclad 2024 sheet.
lOooooooooooqqioqoCJ.; : CD uicri cvioih~oo-
I I
iq co c\j d
TABLE 12-8
co co q o
CO
co co oo d
(JO
'O
IO
lO
.
.
d co
in
co
d
: d
CONDUCTIVITY,
BHN
ALLOY
TEMPER
% lACSt
500 KG, 10MM
EC
0
62
19
1100
0
59
23
H18
57
44
0
50
45
T4
30
105
T6
40
135
0
50
47
T3,T4
30
120
0
50
20
H12
42
35
HIS
40
55
0
; u>
47
H38
35
77
0
45
30
T4
40
65
T6
40
95
T6
30
150
3003
o
IO
CO
in
cb
co
i-
(0
!)() >;'
6061 q
cn
q
q
o o
CD
C\J
.
0) 01
7075
B195 355
520052002000 0020200 D
Relationship of Electrical Conductivity and Hardness for
Some Aluminum Alloys*
2014
r d
CL(/)(/)Q-CLCO(/)CLQ(/)(/)i/)(/)a.(/)a.QQ
.
T4
33
75
T6
33
90
T51
43
65
T6
36
80
T7
42
85
T51
43
60
T6
39
70
T7
40
75
.
356 '
;:: ; >
'
-
tJ-
p ? <£ ? p
CD
cj j- j-ooocn i
< 5
I
-
t-
t-
C\l
t-
t--i-CMC\JCMC\JC\JCO
CD
CD
CD
CD
lo
co
o O
CO
CO
COW
LO
LO
COOT
The clear white
cladding layer is visible at the edges of the sheet. Alclad alloys are extensively used for aircraft applications because of the excellent combination of high strength and high resistance to corrosion. The nominal chemical composition and typical mechanical properties of some wrought- and cast-aluminum alloys are given in Tables 12-6 and 12-7. Table 12-8shoWs the relationship of electrical conductivity and hardness for some aluminum alloys.
2024
Cvl
497
'
Compiled from "Metals Handbook vol. 1, American Society lor Mclals, 1961, tlnlemational Annealed Copper Standard. "
,
'
J
498
INTRODUCTION Tb PHYSICAL METALLURGY
MAGNESIUM AND MAGNESIUM ALLOYS
12 26
-
.
and 3). The full name of the base meta
Table 12-9.
alloys.
c
On the basis of equal volumes, aluminum weighs 1 V2 times more, iron and steel weigh 4 times more, and copper and nickel alloys weigh 5 times more than magnesium. Magnesium has a c.p.h. (close-packed hexagonal) crystal structure and plastic deformation takes place at room temperature by slip along the basal planes. The ductility of magnesium is lower than that of f.c.c. metals since there are fewer slip systems available for plastic deformation. Above 400oF, however, additional planes become active, and the plasticity of magnesium and many of its alloys is improved. Commercially pure magnesium or primary magnesium has a minimum purity of 99.8 percent and usually contains small amounts of aluminum, iron, manganese, silicon, and copper. Approximately half the magnesium produced is used in alloy form for structural purposes, primarily in the aircraft and missile industries. Magnesium is used as an alloying element in aluminum, zinc, lead, and other nonferrous alloys. It has found increasing use in photo-engraving because of its light weight and rapid but controlled etching characteristics. Magnesium has a great affinity for oxygen and other chemical oxidizing agents. It is used as a deoxidizer and desulfurizer in the manufacture of nickel and copper alloys, also as a getter in the manufacture of vacuum tubes. Because of its high chemical activity, it finds use in the production of uranium and zirconium by thermal reduction with magnesium. Magnesium anodes provide effective corrosion protection for water heaters, underground pipelines, ship hulls, and ballast tanks.
I
for brevity when the base metal being r6
Magnesium The chief advantages of magnesium are its light weight, ease of machinability, and the high strength-to-weight ratio obtainable with its
b
"
"
12-27 Alloy Designation and Temper The American Society for Testing Materials (ASTM) has published a system of alloy nomenclature (Specification B275-61) and temper designation (Specification B296-61) for light metals and alloys. This system has been officially adopted by The Magnesium Association for all magnesium alloys. The temper designation is the same as that adopted by The Aluminum Association for aluminum alloys and is covered in Sec. 12-18. The designation for alloys and unalloyed metals is based on their chemical-composition limits as follows (from ASTM Designation B275-61 by permission of the American Society for Testing Materials): ALLOYS
a Designations for alloys consist of not more than two letters representing the alloying elements (Note 1) specified in the greatest amount, arranged in order of decreasing percentages, or in alphabetical order if of equal percentages, followed by the respective percentages rounded off to whole numbers and a serial letter (Not8s2
T
' "
' "
'
rV j
T
s
i
!
<
_
1
In rounding-off percentages
,
the near
mal is followed by a 5 the nearest even ,
d When a range is specified for the allc be used in the designation .
e When only a minimum percentage rounded-off minimum percentage shoul<
,
.1
The letters used to represent alio
t
NOTE 1
For codification
an alloying el
,
the base metal) having a minimum contei or computed in accordance with the pen amount present is the mean of the range
specified) before rounding off NOTE 2 The serial letter is arbitrarily a with A (omitting I and O) and serves to dif A serial letter is necessary to complete e; NOTE 3 The designation of a casting a .
position specified for the corresponding a ingot designation may consist of an all letters
,
one for each product compositic
designations
.
UNALLOYED METALS
Designations for unalloyed metals consis'
retained but dropping the decimal point full name of the base metal precedes thi ,
when the base metal being referred to is As an example for magnesium alloy AZi .
,
element specified in the greatest amount; specified in the second greatest amount; aluminum percentage lies between 8 6 a .
mean zinc percentage lies between 1 5 .
ar
that this is the first alloy whose composi tion AZ92.
TABLE 12-9
Letters Repress
A
Aluminum
M
l\
B
Bismuth
N
h
C
Copper
P
L
D
Cadmium
Q
£
E
Rare earths
R
C
F
Iron
S
£
G
Magnesium
T
T
H
Thorium
Y
A
K
Zirconium
Z
Z
L
Beryllium
1
4
,
-
-
LURGY
NONFERROUS METALS AND ALLOYS
LLOYS
and 3). The full name of the base metal precedes the designation
,
.
Table 12-9.
c
les, aluminum weighs 1V2 times more, ironantt|
.
d When a range is specified for the alloying element
,
be used in the designation. ,
.
e at room temperature by slip along the baa
,
ignesium is lower than that of f.c.c. metafcl
or computed in accordance with the percentages specified for other elements
ggytems available for plastic deformation. Abov
.
NOTE 2 The serial letter is arbitrarily assigned in alphabetical sequence starting with A (omitting I and 0) and serves to differentiate otherwise identical designations A serial letter is necessary to complete each designation NOTE 3 The designation of a casting alloy in ingot form is derived from the composition specified for the corresponding alloy in the form of castings. Thus, a casting ingot designation may consist of an alloy designation having one or more serial
slum or primary magnesium has a minimutt)|
.
sually contains small amounts of aluminum,
.
;| |
v:;-ygnesium produded is used in alloy form
ontrolled etching characteristics.
;
nity for oxygen and other chemical oxidizing; idizer and desulfurizer in the manufacture 5 as a
"
getter in the manufacture of vacuiii "
lemical activity, it finds-use in the producti' hermal reduction with magnesium.
letters, one for each product composition or it may consist of one or more alloy designations. ,
3
UNALLOYED METALS
Designations for unalloyed metals consist of the specified minimum purity, all digits retained but dropping the decimal point followed by a serial number (Note 2). The full name of the base metal precedes the designation but it is omitted for brevity when the base metal being referred to is obvious. As an example, for magnesium alloy AZ92A, "A" represents aluminum, the alloying element specified in the greatest amount; "Z" represents zinc, the alloying element specified in the second greatest amount; "9" indicates that the rounded-off mean aluminum percentage lies between 8.6 and 9.4; 2" signifies that the rounded-off mean zinc percentage lies between 1.5 and 2.5; and "A" as the final letter indicates that this is the first alloy whose composition qualified assignment of the designa,
,
"
1 effective corrosion protection for water heat;
hip hulls, and ballasttanks. .3 per The American Society for Testing Ml a system of alloy nomenclature (SpecificatiCj! ation (Specification B296-61) for light meta jeen officially adopted by The Magnesium alloys. The temper designation is the same lum Association for aluminum alloys and;!
designation for alloys and unalloyed metals ii iposition limits as follows (from ASTM Desifl; i:j;Jf \ of the American Society forTesting Materials)!
tion AZ92.
TABLE 12-9
Letters Representing Alloy Elements
A
Aluminum
M
Manganese
B
Bismuth
N
Nickel
C
Copper
P
Lead
D
Cadmium
Q
Silver Chromium
E
Rare earths
R
F
Iron
S
Silicon
G
Magnesium
T
Tin
.;{$;?ist of not more than two letters representing the Sr£pd in the greatest amount, arranged in order of
H
Thorium
Y
Antimony
K
Zirconium
Z
Zinc
tabetical order if of equal percentages, followed
L
Beryllium
;
;
jJed off to whole numbers and a serial letter (NoteSj
The
amount present is the mean of the range (or the minimum percentage if only that is Specified) before rounding off.
olanes become active, and the plasticity of iloys is improved.
increasing use in photo-engraving because
the rounded-off mean should
e When only a niinimum percentage is specified for the alloying element the rounded-off minimum percentage should be used in the designation NOTE 1 For codification an alloying element is defined as an element (other than the base metal) having a minimum content greater than zero either directly specified
ose-packed hexagonal) crystal structure, arid
.
In rounding-off percentages, the nearest whole number shall be used. If the deci-
mal is followed by a 5, the nearest even whole number shall be used
i copper and nickel alloys weigh 5 times mow
in the aircraft and missile industries. MagnisS ement in aluminum, zinc, lead, and othernof|i
but it is omitted
for brevity when the base metal being referred to is obvious ; b The letters used to represent alloying elements should be those listed in
Mintages of magnesium are its light weight, easft S ?lh strength-to-weight ratio obtainable withH$
copper.
499
S
500
INTRODUCTION TO PHYSICAL METALLURGY
12-28 Magnesium Alloys
Although most magnesium alloys are ternary alloys,
The solubility of zinc in solid m
they may be considered as based upon four binary-alloy systems. These are magnesium-aluminum, magnesium-zinc, magnesium-rare earths, and
6440F to 1.7 percent at 300oF. Alio: percent zinc show the most potent of the magnesium-based binary syst Magnesium-Aluminum-based Alio: sium-aluminum-manganese (AM) an
magnesium-thorium. In each case the solvus line on the magnesium-rich side shows a decrease in solubility of the alloying element in solid magne- I slum as the temperature decreases. This indicates that certain alloy com- i
positions may be strengthened by age hardening. |v For example, in Fig. 12'31 showing the magnesium-rich portion of the ]
casting alloys.
aluminum-magnesium alloy system, the maximum solubility of aluminum
castings with a good combination o elongation.
in magnesium is 12.7 percent at 8180F, decreasing to 3.2 percent at 400°F. * Alloys containing over 6 percent aluminum, which include all the Mg-AI *
7Vj
casting alloys, are therefore heat-treatable.
The AMI 00A alloy is popular for pre
The sand-casting alloys AZ63A temperature applications If the ope .
may give satisfactory service at temf used where maximum toughness or yield strength are required. AZ92A is plus good pressure-tightness are req alloy are shown in Fig. 12-32 The
Aiomic percent aluminum 40
30
20
10
.
1200
I V
most of the eutectic has "separated" grains, leaving behind the compound may be retained in solution by the r treatment (Fig. 12-32b). Subsequen
.
.
Liq -
+
8 +Liq.
|
864°
of the compound as fine particles wit of the more rapid cooling in permar
818° 40.2 %
12.7%
_
(Fig. 12-32a) shows a network of m? originally composed of 5 and e solid s
p
sand casting, the structure (Fig. 12-3 pound and no lamellar constituent.
placing AZ63A for applications requir 8
strength.
S +
AZ91A and AZ91B alloys are especi ings are pressure-tight with good yiel
The wrought alloy M1A (containing relatively low-strength magnesium al rosion resistance
200 0
?0
Weiqhl pcrccnl aluminum
Fig. 12.31 Magnesium-rich portion of the aluminummagnesium alloy system. The 8 phase is also known as the compound Mg,7AI,;, a designation based on crystallographic evidence rather than composition, since Mg AL would be simpler and within the fi homogeneity field. (From "
Metals Handbook," 1948 ed., p. 1163, American Society
for Metals, Metals Park, Ohio.)
0
and hot formability The AZ31 alloys are widely used having good strength and formability aluminum, they are not heat-treate strain hardening. AZ31 with low cal widely used for magnesium photot ,
dustry.
AZ61A has excellent strength and c trusion and forging alloy.
AZ80A alloy, containing 8.5 percen
URGY
NONFERROUS METALS AND ALLOYS
.
h most magnesium alloys are ternary all leased upon four binary-alloy system8.: ' lagnesium-zinc, '
.
magnesium-rare earthsl
i case the solvus line on the magnesiu
iubility of the alloying element in solid mB
5
eases. This indicates that certain alloy d by age hardening.
showing the magnesium-rich portion ofl
castings with a good combination of tensile strongth,,yield strength, and
cent aluminum, which include all the M heat-treatable. 9 s;
elongation. The sand-casting alloys AZ63A and AZ92A are used for normaltemperature applications. If the operating stresses are not too high, they may give satisfactory service at temperatures as high as 350°F. AZ63A is used where maximum toughness or ductility along with moderately high
yield strength are required. AZ92A is used where maximum yield strength plus good pressure-tightness are required. The micrdstructures of AZ92A alloy are shown in Fig. 12-32. The structure in the sand-cast condition (Fig. 12-32a) shows a network of massive 8 (Mg17AI,2): The eutectic was originally composed of fi and e solid solution rich in magnesium. However, most of the eutectic has separated ; the e phase has joined the primary e
1200
"
Liq
This group includes the magne-
sium-aluminum-manganese (AM) and the magnesium-aluminum-zinc (AZ) casting alloys. The AM100A alloy is popular for pressure-tight sand and permanent-mold
system, the maximum solubility of alumlff t at 8180F, decreasing to 3.2 percent ai 4l
iminum
.
The solubility of zinc in solid magnesium varies fnom 8.4 percent at 644°F to 1.7 percent at 300oF. Alloys in the composition range of 4 to 8 percent zinc show the most potent precipitation-hardening effects of any of the magnesium-based binary systems.
Magnesium-Aluminum-based Alloys
,
501
+
81R0 \2.rVc
1
"
grains, leaving behind the compound as a white network. The compound may be retained in solution by the relatively rapid cooling after solution treatment (Fig. 12-325). Subsequent artificial aging causes precipitation of the compound as fine particles within the grains (Fig. 12-32c). Because of the more rapid cooling in permanent-mold casting as compared with sand casting, the structure (Fig. 12-32d) shows finer particles of the compound and no lamellar constituent. AZ91C and AZ81A are gradually replacing AZ63A for applications requiring good ductility and moderate yield strength.
AZ91A and AZ91B alloys are especially suited for die casting. The castings are pressure-tight with good yield strength and ductility.
The wrought alloy M1A (containing 1.2 percent manganese) is a low-cost, relatively low-strength magnesium alloy. It has excellent weldability, corrosion resistance and hot formability. The AZ31 alloys are widely used as general-purpose extrusion alloys having good strength and formability. Since they contain only 3 percent aluminum, they are not heat-treatable and attain their properties by strain hardening. AZ31 with low calcium content known as PE alloy, is widely used for magnesium photoengraving sheet in the printing industry. AZ61A has excellent strength and ductility and is used mainly as an ex,
10
jiluminum-
,
so known as '
"
'
.
;
d on crystalsince Mg )AI2 leity field. (From rican Society
1,
0
Mg
oluminum '
-J POO
.
.
trusion and forging alloy. AZ80A alloy containing 8.5 percent aluminum, is a heat-treatable alloy ,
i
1
502
v:-
I
INTRODUCTION TO PHYSICAL METALLURGY
used for extruded products and
combination of high strength and r Magnesium-Zinc-based Alloys T zirconium (ZK) and magnesium-zi effect of zirconium additions up to is one of grain refinement. Zircon grained, columnar cast structure, tht The casting alloys ZK51A and Zk ,
I
P
tensile strength and ductility of any to the strong age-hardening effect and the fine-grain effect of zirconiur
requires very careful foundry contr microporosity and hot cracks. Thee duce these problems with little effet addition of a rare-earth metal {ZE4
i
p
with some reduction in mechanica
(b)
a
characteristic lamellar form of eute
Mi
;;
'
boundaries of magnesium-rich solid The wrought alloy ZK60A is the hii and hollow shapes. It has good tc improve its properties further. The earth metal (ZE10A) results in a che Magnesium-Rare Earth-based Al nesium-rare earth-zirconium (EK)
:v j
. .
0
<3 *
p
5)
f
J?
,
I 4*
A-92A
Fig. 12-32 The microstructure of magnesium alloy 250X. (a) Sand-cast, etched in phosphopicral; (b) sand
S
/
9
,
-
cast, solution-treated, etched in glycol; (c) sand-cast, solution-treated, artificially aged, etched in acetic glycol;
(d) permanent-mold cast, etched in phosphopicral (The Dow Metal Products Company.)
*
.
Fig. 12-33 Alloy ZH62A-T5 sand casting. Lamellar eutf Mg-Th-Zn compound at th s grain boundaries of magnes nch solid solution. Etched in 2 percent nital, 250x. (Fr Melals Handbook, vol. 7, "Atlas of Miorostructures." Am
ican Society for Metals, 1972.)
mi
m
.
LURGY
NONFERROUS METALS AND ALLOYS
Bs.
W
used for extruded products and press forgings.
503
II offers an excellent
combination of high strength and moderate elongation. Magnesium-Zinc-based Alloys These are essentially magnesium-zinc-
T
zirconium (ZK) and magnesium-zinc-thorium (ZH) casting alloys.
The
effect of zirconium additions, up to about 0.7 percent, to magnesium alloys is one of grain refinement. Zirconium completely eliminates the coarse-
grained, columnarcast structure, thus increasing the mechanical properties. The casting alloys ZK51A and ZK61A attain the highest combination of '
tensile strength and ductility of any magnesium casting alloys. This is due to the strong age-hardening effect of the magnesium-zinc binary system and the fine-grain effect of zirconium. Unfortunately the high zinc content
SI
,
requires very careful foundry control to produce sound castings free of microporosity and hot cracks. The addition of thorium (ZH62A) helps to reduce these problems with little effect on mechanical properties, while the addition of a rare-earth metal (ZE41A) also reduces the above problems with some reduction in mechanical properties. Figure 12-33 showd the
"
(b)
characteristic lamellar form of eutectic Mg-Th-Zn compound at the grain boundaries of magnesium-rich solid solution in a ZH62A-T5 alloy, sand cast.
The wrought alloy ZK60A is the highest-strength extrusion alloy for solid
1
and hollow shapes. It has good toughness and may be heat-treated to improve its properties further. The replacement of zirconium by a rareearth metal (ZE10A) results in a cheaper and tougher alloy. Magnesium-Rare Earth-based Alloys This group includes the magnesium-rare earth-zirconium (EK) and magnesium-rare earth-zinc (EZ)
4f
:
4
V
(d)
L alloy A-92A
,
1; (b) sandand-cast, : cetic glycol; picral. .
!lg 12-33 Alloy ZH62A-T5 sand casting. Lamellar eutectic Ktg-Tn-Z.- compound at the grain boundaries of magnesium.
fich solid solution. Etched in 2 percent nital, 250x. (From Metals Handbook vol. 7, "Atlas of Microstructures," Amer,
icam
5
: V;. .
\
Society for Metals, 1972.)
4
504
INTRODUCTION TO PHYSICAL METALLURGY
casting alloys. The rare-earth elements are atomic numbers 58 71 (see -
temperature uses where high stre preferred where long-time lower-sti
Table 2.3).
Improvement in elevated-temperature properties is obtained by a high recrystallizatfon temperature and by precipitates at the grain boundaries that are stable at high temperature to minimize creep. The addition of rareearth elements to magnesium satisfies both requirements, so that these
HK31A is also used as a wrougf plications. Other wrought alloys HM31XA and the sheet alloy HM21A ,
manganese alloys These have the I wrought-magnesium alloy .
alloys are suitable for use up to 500oF
.
.
Casting alfbys EK30A, EK41A, and EZ33A all have similar mechanical
The nominal composition and t] cast- and wrought-magnesium allo 12 29 Corrosion Resistance of Magnesii
properties, but EZ33A shows poorer corrosion resistance because of the presence of iinc. The structure of EZ33A-T5 alloy, sand cast, is shown in
Fig. 12-34. jit consists of a network of massive magnesium-rare earth
slum alloys to atmospheric corrosic the amount present indoor or out
compound (dark) in a magnesium-rich solid solution (light). When the fare-earth addition contains 50 percent cerium, it is known as
,
'
mischmetal and is used as a low-cost commercial alloy. The rare-e arth elements are not used as the principal addition in any
4
values of humidity (below 10 perce rosion resistance but the resistanc ,
During indoor exposure the magne
,
,
wrought-magnesium alloys.
sistance then those containing alum
Magnesium-Thorium-based Alloys The group includes the magnesiumthorium-zirconium (HK) and magnesium-thorium-zinc (HZ) casting alloys. /
door exposure conditions
With al
.
zinc content above 5 percent
Thorium, like the rare-earth elements, greatly improves the elevated- |
,
corrc
mental results indicate that the cor
temperature properties of magnesium.
pares favorably with that of alumir low-carbon steel in industrial atmo!
Casting alloys HK31A and HZ32A are used for applications in the rangeof 350 to 7000F where properties better than those of the rare earth-contain-
purity of the alloy seems to be a cc
ing alloys are needed. HK31A is particularly good for short-time elevated-
sion resistance. Controlled-purity ce resistance under marine conditions
The corrosion resistance of magni is dependent upon the presence of i and cobalt.
The effect of heat treatment on th
alloys in salt solutions varies with 1 cooling alloying elements, and irr manganese alloy Ml heat treatmen,
,
increases the corrosion rate and the
sr
i1
ment at about 1050oF
followed by pitting and increases corrosion resis
0
1 *
A
,
12 30 Joining Magnesium Alloys
I
1
Magne
joined by most of the common fusio V
-
These include shielded-metal arc we
seam and spot welding riveting, be hesive bonding. Arc welding spot wi ,
Fig. 12-34 Alloy EZ33A-T5 sand casling. Structure consists of a network of massive.magnesium- rare earth compound
(dark) in a magnesium-rich solid solution (white). Etched
in glycol, lOOx. (From Metals Handbook, vol. 7,
"
Atlas of
Mictostructures," American Society for Metals, 1972.)
i
T t
J
i i
,
monly used methods
.
In all magnesium alloys the solidifi( point and shrinkage decrease with ar ,
.-
3Y
NONFERROUS METALS AND ALLOYS
slements are atomic numbers 58-71 (i :
temperature uses where high stresses are encountered, while HZ32A is
preferred where long-time lower-stress properties are important. HK31A is also used as a wrought sheet alloy for high-temperature applications. Other wrought alloys in this group are the extrusion alloy
< oerature properties is obtained by a hi
d by precipitates at the grain boundalii re to minimize creep. The addition of rai atisfies both requirements, so that th
HM31XA and the sheet alloy HM21 A, both of which are magnesium-thorium-
manganese alloys. These have the best high-temperature properties of any
500°F. ,
505
wrought-magnesium alloy.
and EZ33A all have similar mechani
The nominal composition and typical mechanical propSrties of some
iorer corrosion resistance because of tKI
cast- and wrought-rtiaghesium alloys are given in Table 12-10.
of EZ33A-T5 alloy, sand cast is showrt l|| twork of massive magnesium-tare eai m-rich solid solution (light)
li'SS Corrosion Resistance of Magnesium Alloys The resistance of magne-
,
sium alloys to atmospheric corrosion depends upon the alloying element, the amount present ihdoor or outdoor exposure, and humidity. At low values of humidity (below 10 percent) magnesium alloys show good corrosion resistance, but the resistance decreases with increasing humidity. During indoor exposure the magnesium alloys show better corrosion resistance then those containing aluminum, but the reverse is true under outdoor exposure conditions. With aluminum content above 9 percent and zinc content above 5 percent, corrosion resistance is decreased. Experimental results indicate that the corrosion resistance of magnesium compares favorably with that of aluminum alloys and is superior to that of low-carbon steel in industrial atmospheres. Under marine exposure, the
.
,
pontains 50 percent cerium, it is known
,
-cost commercial alloy not used as the principal addition in artl
/
.
,
pys
y
The group includes the magnesium
gnesium-thorium-zinc (HZ) casting alloy$| lements, greatly improves the elevatei
.
2A are used for applications in the range Of
ptter than those of the rare earth-contain , particularly good for short-time elevated !
purity of the alloy seems to be a controlling factor in determining corrosion resistance. Controlled-purity cast alloys show a far superior corrosion resistance under marine conditions than uncontrolled-purity alloys.
The corrosion resistance of magnesium alloys in aqueous salt solutions is dependent upon the presence of impurities such as iron, copper, nickel, and cobalt.
The effect of heat treatment on the corrosion resistance of magnesium alloys in salt solutions varies with the heat-treating temperature, rate of cooling alloying elements, and impurity content. In the magnesiummanganese alloy Ml heat treatment in the range of 500 to 900"F greatly increases the corrosion rate and the tendency to pit. However heat treat-
'
f
.
,
,
m
,
ment at about 1050 F
,
(ollowed by rapid cooling, completely eliminates
pitting and increases corrosion resistance.
12'30 Joining Magnesium Alloys
Magnesium and magnesium alloys may be
,
joined by most of the common fusion and mechanical fastening methods. These include shielded-metal arc welding gas welding, electric resistance seam and spot welding riveting, bolting, self-fastening devices, and adhesive bonding. Arc welding spot welding, and riveting are the most com,
ire consists '
ompound : rinluid Atlas ol
J72.)
m
,
,
monly used methods.
In all magnesium alloys the solidification range increases and the melting point and shrinkage decrease with an increase in alloy content. Alloys con,
506
INTRODUCTION TO PHYSICAL METALLURGY
taining up to 10 percent aluminu structure. Alloys containing more
crack when hot {hot-short) and ma; CO g
have no tendency toward hot-shor
oil
Failure of welded joints usually c
cm
O >- T"
1
t-t
z
t-
.
CJ) Tj-
O CD
CO
CM
ID
to
(£) cm
CT) oj
CO co
CD
CO
CO
CJ
CM
O CO
to
the weld rather than the weld itsell
OJ
growth of the base metal. Welds in some magnesium alloy susceptible to stress-corrosion era stresses set up in the welding pro relieved by a suitable stress-relief t Adhesive bonding is a comparati drilling is required there is less s strength in adhesive-bonded joints hesive fills the spaces between cont between any dissimilar metals in tl
CO
CM CO
ID
lO
ID
CO
CO
CO
CO
CO
CO
h-COCpO
CO
(D
ID
CO
ID
CO
O
CO
,
>
o _
I
!
I- CI i-
CO LU
z
use of thinner materials and takes CMCnCMOCM
CO LU
CO CM
CMi-i-T-C\JCMCNJ
CO
CO
<
>
-
_
<
l
_
CO OJ
OJ CO
UJ H O
CO
CO
I- C/3
C\J
CVJ
55 w o Z QC O
0)
I
O z
z
co
ui
<
z
u
CO CO
cr <
-
<
z
l
J co co
co
lt) - f
lo lo
co lO
a
.
t
z
NICKEL AND NICKEL ALLOYS
UJ LU
X CO
12-31 Nickel Nickel is characterized by gc tion. It is white in color and has gi properties. It forms tough, ductile : common metals. Approximately 60 in stainless and nickel-alloy steels. nickel alloys and for electroplating.
LU co
0
co CD
r
<
r
h
r
iDcocMiDr
r--
o
O
oooooooooo
CO
5
m
oj q
O O .
CO
CO
CO
CM
l
1
Z
CO
5 O
Or r-OLnqor S COOOCM tCDCDLn- J-
N
o .
c\i
LO
oi oi
in
-
T-
O
ID
CM
O
tance and hardness, nickel makes a
OJ
tO
o
6
O
p ID CD
o
ai
I CO ID CO
CO
p o lo p q
-
ld
CM
v
oj
o
d d
CM
O
-
o 7
CM
o (p
<
d
o o o o o
cxj
2
'
-
corrosion and wear.
due to the heavy nickel undercoat. Cast nickel is sometimes used for
ui
UJ
a
.
5 UJ
a z
<
< ui
UJ
z < < X
>
o i
m
lu
lu
UJ
_
<
I I CO CO
cppcocojOcDgpcnggp p gi<6<<<<<<<<<<<
gSSSSSSSSjcogSg <9<<
CO
<
i
n
N
<
i
ID
LD
H
H
i
i
m < < < i
OJ
larly where contamination with cop amounts of silicon and manganese of sound ductile castings. Wrought nickel Is not adversely a heating. Its mechanical properties a tains its strength at elevated temper; at low temperatures. The electrical ,
6
a
z cr
Although nic
chromium to increase wear resistanc
a: (E
Riveting is the most common m gives joints of good strength and ef does not require highly skilled laboi
Cj
><
z
N
.
DC <
cm q p p
Z
Q
o Q
H
QC
ID
Q
M
UI
<
Q
Q
.
5
z
1X1
a
UJ
-
<
possible with magnesium. This me joining stiffeners to sheet and is us'
Q
UJ
cc
V
-
J
CD
LU
o
o
CO
5
<
<
< X
Q
j
Q
UJ C3
CL
CO
O
CO
O
o
z
g
J
o
UJ H P.
1
r-j
-
Z (T O
<
-
Tt
co
<
_
'
CO CD
in CO
o
o
UJ C3 W)
I
v Vv;
T-
CO
I "
CM
T-
Q
i
.
T-
<
I
'
cm
CO
H
UJ
5
'
-OJC\JOJOJ
I/)
CD
>f
I
t
CD
O
ill s?
I
-
CO
>-
x-
O
O
< CO CO CO CD
5 N N N
5 < < < N
5 5 g2 gB N
,
< i ii5IP
LURGY
.
NONFERROUS METALS AND ALLOYS
o|
507
,
taining up to 10 percent aluminum aid weldability by refining the grain structure. Alloys containing more than 1 percent zinc have a tendency to crack when hot (hot-short) and may result in weld cracking. Thorium alloys have no tendency toward hot-shortness.
i !> (O 00 (O
]J
-
LO to
i lo o o cd
>
cj
- T- T-
o> 10
in
CO
in to to to
CO
CO
"J
"
CTJ
O
O
C\J
TC
TC
to
CO
CO
"
ID
'-
S J312
g
1
li
Failure of welded joints usually occurs in the heat-affected zone next to the weld rather than the weld itself. This is most likely due to some grain growth of the base metal. Welds in some magnesium alloys, particularly the Mg-AI-Zn series are susceptible to stress-corrosion cracking. This is due to the high residual stresses set up in the welding process. These residgal stresses may be ,
relieved by a suitable stress-relief treatment. | CO CO CO 00 Tj-
CO
csj
in
co
Adhesive bonding is a comparatively new method of joining. Since no drilling is required, there is less stress concentration and better fatigue strength in adhesive-bonded joints as compared with other types. The ad-
LO CO O
hesive fills the spaces between contacting surfaces and acts as an insulator between any dissimilar metals in the joint. Adhesive bonding allows the use of thinner materials and takes better advantage of the weight saving possible with magnesium. This method of joining is particularly suited for
co LU
t
C\J
CD OJ CO O tJCU CM CO Tl- Tl-
a
.
< x
>
h-
r;
<
a
CM OJ OJ CO OJ
-
CO
o
CO LO CO C\J
CO
o
z
<
<
CD
CO
z
CO CO
H
LU
H
<
ir
"
<
co co
o
CO lO lO CO in co
m
CO
a
CM h- 5 <0
.
CO CO ;W
a UJ
<
LU
Q
H
z
LU
NICKEL AND NICKEL ALLOYS
Ul
X co
3 q p
.
co co
;
-
si
4
|q oj in n. k
.
p o o o o
.
in
o
.
o -
m .
CO
c\i
p p in in o
cvj
does not require highly skilled labor.
a
< O Q
*
joining stiffeners to sheet and is used in aircraft and radar applications. Riveting is the most common method of joining magnesium, since it gives joints of good strength and efficiency, is a fairly simple method, and
t
CM
-
»-
CM
CM
o
m
o .
.
o « «
9
-
CO CM
.
:
;
"
:
in.
o
. o
N
d d
o
9
Cvl
:
tion. It is white in color and has good workability and good mechanical properties. It forms tough, ductile solid-solution alloys with many of the common metals. Approximately 60 percent of the nickel produced is used in stainless and nickel-alloy steels. Most of the remainder is used in highnickel alloys and for electroplating. Because of its high corrosion resistance and hardness, nickel makes an ideal coating for parts subjected to corrosion and wear. Although nickel is often given a flash coating of chromium to increase wear resistance, most of the corrosion protection is
o
co m oo
|2-31 Nickel Nickel is characterized by good resistance to corrosion and oxida-
co
1
due to the heavy nickel undercoat. Cast nickel is sometimes used for corrosion-resistant castings, particu-
larly where contamination with copper or iron must be avoided. Small
~
:
;
.
o
r-roy l
amounts of silicon and manganese are added to facilitate the production of sound, ductile castings.
CD
-
.
a u
a
t < < < <
<
.
u
.
u. [2 P
m < < < t- o
m ai
if !
o
CO CO
N
<
5 < <
N < tM
S S jo m <
< 11 pi 4
1-F
Wrought nickel is not adversely affected by cold working, welding, or heating. Its mechanical properties are similar to those of mild steel. It retains its strength at elevated temperatures and its ductility and toughness at low temperatures. The electrical conductivity of nickel, while not so
508
f
INTRODUCTION TO PHYSICAL METALLURGY
2
o o o o
I
O
o o o
o
o
o
CD r-
CO CO
o
O
u)
ID
-
CO
i
high as that of copper or alumir leads and terminals in many electt The most important commercial E nickel, permanickel, and duranic A nickel is the basic material
,
2
Including cobalt. Cast commercial
5s
silicon to improve fluidity and cast
C J
z
1
o
O
O
O
UJ
CO
-si-
-sf
CM
o
in
in
m
CM
O
u)
CO
CM
Cvl
<*
CM
combined with resistance to corro
in
si-
CM
nickel is used by the chemical an of evaporators, jacketed kettles equipment. ,
X
D nickel and E nickel conform gi the important difference being the respectively of manganese replacir
CD Z
m m X
i
55 in o
00
s s ss§
in
in
in
m
CM
CM
*
CM
,
of manganese improves the resist;
in
temperatures. The mechanical stre
6 T-
elevated temperatures, is somewh; has better resistance to attack by ; spark-plug electrodes, ignition tub
z
LU CD W Z
A Z
_
boiler refractory bolts. Since E nick
L
o o
X
LU I- °
in CD
n
CD
.
in
o
o
en
o
O
m
m
m
i -
o
-
CM
h-
r
in o
m r -
D nickel, its mechanical properties a Typical uses are for spark-plug wi
CM
o
furnaces.
5
Si
a;
CO
z T3 "O
O a
1 SS z
2
c
V go?
COOO
<
a>
(0
Duranickel is a wrought age-hu aluminum alloy. It offers a combin that of heat-treated steels) and the e Duranickel springs are used as lai frames. This alloy is also used for bellows, snap-switch blades, and i parts of fishing tackle. Permanickel is an age-hardenabl( properties and corrosion resistance tion, good electrical and thermal co softening at elevated temperatures nickel and it should be used in pi; where higher electrical conductivit'
v
,
a; C5) 03
c
iimi miii
.
o
c
E
CO
c «!. 3? COO
in en
< <
(f) w
2 i
o
0)
Z
to
a i
d
C
H Z
o
D
CD
d 5
te
n
U.
c
2 c
LL
5
C
<
(J
_.
f - CO r-
C
+°i
5 C
s
±d
+ o o
2
,
Z
5 M
o « E
.
-
1 m
.
O
6
c
c c
.
O
(L)
CD
< Q
2
m CM
-
Z 4i.3S
ex
LO
C
O
C
S CD
m
OT CO
(-)
z
CD
O
T-
en
in o m m cm o o "*
o o o c
.
CD o o cm m
o o m S o o o o
essential.
c
z
m
a3
0)
en
si
T-
<
Q
CO
AC
c LU
m
<
<
2
5
a:
CD
o CO
E
n
CL
E 1 z
<
per, iron, chromium, silicon, molybd
c
.
"
UJ
o
The chemical composition and typ mercial grades of nickel are given in 12-32 Nickel Alloys The most common a
Nickel-Copper-based Alloys Copp is added to increase formability dec
2
CD
c
5
,
!
-
4
-
aim
-..
NONFERROUS METALS AND ALLOYS
509
high as that of copper or aluminum, is satisfactory for current-carrying --
-I
-
i- CO
CO CO
leads and terminals in many electronic applications.
i- CI
The most important commercial grades of nickel are A nickel, D nickel, E nickel, permanickel, and duranickel. A nickel is the basic material, containing 99 percent minimum nickel including cobalt. Cast coftimercial nickel contains approximately 2 percent
silicon to improve fluidity 6nd castability. A nickel is used where strength O
ID
lO
LO
O
in
LO
combined with resistance to corrosion and oxidation is required.
o
Rolled
CNJ
nickel is used by the chemical and soap industries for the construction of evaporators, jacketed kettles, heating coils, and other processing equipment. D nickel and E nickel conform generally to the composition of A nickel,
t
in
CNJ
m en
in
1-
CNJ
oo or
uom r o
mm or
oo oo f-
-
i-CNJ
i
-
t-CNJ
-
i
i
-
i
the important difference being the inclusion of about 4.5 and 2.0 percent, respectively, of manganese replacing a like amount of nickel. The addition of manganese improves the resistance to atmospheric attack at elevated temperatures. The mechanical strength of D nickel, both at normal and at elevated temperatures, is somewhat greater than that of A nickel, and it has better resistance to attack by sulfur. D nickel is used extensively for spark-plug electrodes, ignition tubes, radio-tube grid wires, and marineboiler refractory bolts. Since E nickel has a lower manganese content than D nickel, its mechanical properties are intermediate between A and D nickel.
Typical uses are for spark-plug wires and as electrical lead-in wires for furnaces.
8.
g -
"
D
"D
O
O
c
(0
0)
C
- 0)
conj ocncn-o '
O "O O O en rara
"
0)
C
O)
0)
c 0)
TO .
C 0)
Duranickel is a wrought, age-hardenable, corrosion-resisting, nickelaluminum alloy. It offers a combination of high strength (comparable to that of heat-treated steels) and the excellent corrosion resistance of nickel.
'
g cn cn "O Q Q. C
s
.
< <
CO
CO
< <
CO CO
1
6
6 m
O
OJ
O
.
CO < - to
3ro
-
-
-; - cn io
m
z CO 3 « S m S m m ._
1
i
,
°°
n l9
00
J>Ot-
0)0000
o
i
to
nickel, and it should be used in place of duranickel only in applications where higher electrical conductivity and better magnetic properties are
I
essential.
o
o
cnj
m
9 P "
oo CD
O
O
O
O
0) c
ra
E
o
0)
Q
Permanickel is an age-hardenable, high-nickel alloy having mechanical properties and corrosion resistance similar to those of duranickel. In addition, good electrical and thermal conductivuy is present. Its resistance to softening at elevated temperatures is somewhat inferior to that of dura-
i
0)
laj
parts of fishing tackle.
Z|C0 =|> m m
i .
Duranickel springs are used as laundry clips, jewelry parts, and optical frames. This alloy is also used for instrument parts such as diaphragms, bellows, snap-switch blades, and in the sports field for fish-hooks and
Q
.
i
The chemical composition and typical mechanical properties of the commercial grades of nickel are given in Table 12-11. 12-32 Nickel Alloys The most common alloying elements with nickel are copper, iron, chromium, silicon, molybdenum, manganese, and aluminum.
Nickel-Copper-based Alloys
Copper is completely soluble in nickel and
is added to increase lormability decrease price, and still retain the corro,
:
V:
510
i
INTRODUCTION TQ PHYSICAL METALLURGY
ing, and antiseizing characteristics tack. Both alloys have similar mec taining less silicon has better mach
sion resistance of nickel. Monel is the most important of the nickel-copper
alloys, containing approximately two-thirds nickel and one-third copper.
Monel has high corrosion resistance to acids, alkalies, brines, waters, food products, and the atmosphere. It has mechanical properties higher than
,
valve seats pump liners, and impell ,
those of the brasses and bronzes, but lower than those of alloy steels. It
Constantan (45 percent nickel
,
also has good toughness and fatigue strength and finds considerable use in elevated-temperature applications. It does not oxidize at a destructive
.
rate below approximately 1000oF in sulfur-free atmospheres and for some
applications may be used at temperatures up to 1500 F Monel has widespread use in the chemical, pharmaceutical, marine, power, electrical, laundry, textile, and paper-equipment fields. The microstructure of Monel "
.
i
.
\
'
alloy age-hardenable. Figure 12-35 shows the fine precipitate of NiaCAl.Ti) black spots in Monel K-500 alloy after aging for 4 h at 1300oF. Since the
as
last property is desirable for thermc
iron-constantan thermocouples wen Nickel-Silicon-Copper- based Alloy in this group is Hastelloy D It cont
annealed and cold-drawn was shown in Fig. 3-16.
. R Monel is a nickel-copper alloy which contains high sulfur to improve mach mability. It is produced primarily for automatic screw-machine work. K Monel contains approximately 3 percent aluminum, which makes the
55
trical resistivity, the lowest tempera highest thermal emf against platim The first two properties are import
copper. It is a casting alloy which It can be machined only with difficult Its most important characteristic is concentrated sulfuric acid at elevati orators
'
,
reaction vessels pipelines, ,
precipitate is resolvable with the optical microscope, it indicates that the J
Nickel-Chromium-Iron-based Alloys
alloy is overaged. Thus it is possible to obtain a nonmagnetic corrosion-
and ternary nickel-chromium-iron al
resistant material with extra strength and hardness. Some typical applications of K Monel are marine pump shafts, springs, aircraft instruments, ball bearings, and safety tools. H Monel and S Monel, containing 3 and 4 percent silicon, respectively,
alloys. Some nominal compositions
are casting alloys that combine high strength, pressure-tightness, nongall-
heater pads hair driers and hot-w
V
,
and others) used as electric heatir
and industrial furnaces; 60Ni-16Cr-24
used as electrical heating elements 1 ,
,
rheostats for electronic equipment an and 35Ni-20Cr-45Fe used for heavyalloys show good resistance to oxk
gases.
They are widely used in cas
equipment furnace parts, carburizini ,
pots, and other equipment that must 1800oF. ..
.
.
.
<
.
.
/k
1
Inconel
,
with a nominal compositi
inherent corrosion resistance strong ,
extra resistance to high-temperaturt applications for Inconel were in fooders, coolers
regenerators, pasteurize ing milk. Inconel is outstanding in its and cooling in the range of zero to 16 i
Fig. 12-35
Monel K-500, field 1 h at 2200LT. transferred to
a furnace at 1300oF
and aged 4 h, water-quenched. Black
spots are Ni (AI,Ti) precipitate. Etched in NaCN, (NH,)..S,0Si .
1
,
000x. (From Metals Handbook, vol. 7, "Atlas of Micro-
,
"
structures,
American Society for Metals, 1972.)
I
,
used for exhaust manifolds and heate
tensively in the furnace and heat-trt carburizing boxes retorts, muffles, ar ,
Inconel X is an age-hardenable Inc tions of titanium (2.25 to 2 75 percent .
:
.
iY
NONFERROUS METALS AND ALLOYS
is the most important of the nickel-copper .:| two-thirds nickel and one-third copper. S;-'::fice to acids, alkalies, brines waters, food It has mechanical properties higher than s but lower than those of alloy steels. II
r
ons.
It does not oxidize at a destructive
in sulfur-free atmospheres and for some oeratures up to 1500 F. Monel has widearmaceutical, marine power, electrical.
valve seats
if
-
W
'
I v
. .
The microstructure of Monei
concentrated sulfuric acid at elevated temperatures.
.
springs, aircraft instruments, ball
ling 3 and 4 percent silicon respectively. igh strength, pressure-tightness nongall,
,
Nickei-Chromlum-lron-based Alloys
A variety of binary nickel-chromium
and ternary nickel-chromium-iron alloys are used as electrical-resistance alloys. Some nominal compositions are 80Ni-20Cr (Chromel A, Nichrome V and others) used as electric heating elements for,household appliances and industrial furnaces; 60Ni-16Cr-24Fe {Chromel C, Nichrome, and others) used as electrical heating elements for toasters, percolators, waffle irons, heater pads hair driers, and hot-water heaters, also in high-resistance rheostats for electronic equipment and as dipping baskets for acid pickling; and 35Ni-20Cr-45Fe used for heavy-duty rheostats. Many of the above alloys show good resistance to oxidation, heat fatigue,' and carburizing gases. They are widely used in cast and wrought form for heat-treating equipment, furnace parts, carburizing and nitriding containers, cyaniding pots, and other equipment that must withstand temperatures up to about ,
I .. >
vA '
.
It is used for evap-
orators, reaction vessels, pipelines, and fittings in the chemical industry.
Sc-"/; optical microscope, it indicates that the .< i sible to obtain a nonmagnetic corrosionJ. igth and hardness Some typical appiica- j ,
,
Nickel-Silicon-Copper-based Alloys The best-known commercial alloy in this group is Hastelloy D. It contains 10 percent silicon and 3 percent copper. It is a casting alloy which is strong, tough, and extremely hard. It can be machined only with difficulty and is generally finished by grinding. Its most important characteristic is its excellent corrosion resistance to
own in Fig. 3-16 y which contains high sulfur to improve fnarlly for automatic screw-machine work. ily 3 percent aluminum which makes the 35 shows the fine precipitate of NijfAl.Ti) oy after aging for 4 h at 1300t'F. Since the
> shafts
,
,
,
:
pump liners, and impellers. Constantan (45 percent nickel 55 percent copper) has the highest electrical resistivity, the lowest temperature coefficient of resistance and the highest thermal emf against platinum of any of the copper-nickel alloys. The first two properties are important for electrical "resistors while the last property is desirable for thermocouples. The copper-constantan and iron-constantan thermocouples were discussed in Chap. 1. ,
,
v Tient fields.
,
taining less silicon, has better machinabifity. Typical applications include
,
igue strength and finds considerable use
ing and antiseizing characteristics along with resistance to corrosive attack. Both alloys have similar mechanical properties but H Monel; con,
;
,
511
V
.
,
1800'F.
Inconel, with a nominal composition of 76Ni-16Cr-8Fe, combines the strength, and toughness of nickel with the extra resistance to high-temperature oxidation of chromium. The first inherent corrosion resistance
,
applications for Inconel were in food-processing equipment such as heaters, coolers, regenerators, pasteurizers and holding tanks for pasteurizing milk. Inconel is outstanding in its ability to withstand repeated heating and cooling in the range of zero to 1600"F without becoming brittle and is used for exhaust manifolds and heaters of airplane engines. It is used extensively in the furnace and heat-treating field for nitriding containers carburizing boxes, retorts muffles, and thermocouple-protection tubes. Inconel X is an age-hardenable Inconel. Hardening is secured by additions of titanium (2.25 to 2.75 percent) and aluminum (0.4 to 1 percent). A ,
: .
¥;¥;Wred to
;
Black
,
,
>
. . .
y>.
A'Cro-
m 512
INTRODUCTION TO PHYSICAL METALLURGY
X (47Ni-9Mo-22Cr-18Fe) has outstan. up to 2200L F
It is used for many ir
'
0
.
m
aircraft parts such as jet-engine tai and vanes.
Nickel-Chromium-Molybdenum-Cop group were originally developed as r nitric acids over a wide range of co
0
0
Two casting alloys are Illium B (50Ni5Cu). They provide
22.5Cr-6.5Mo-6 c
.
chinable high-strength casting alloy; rotary bearings and pump and valve
Jj
-
in corrosive environments Illium R
o
.
wrought alloy that provides heat anc pump and valve shafting, hardware h
f
Fig. 12-36 Hastelloy B, solution-annealed at 2150'F and water-quenched. Globular constituent is carbide: matrix is fee. y solid solution. Etched in chrome regia, 500\. (From Metals Handbook, vol. 7, 'Atlas of Microstructures," Amer-
ican Society for Metals, 1972.)
1233
The nominal chemical compositio of some nickel alloys are given in Tai Nickel-Iron Alloys The nickel-iron Accurate phase boundaries below at
lished because of the sluggishness c tures. Nickel and iron are completely as solid solutions. Nickel lowers proc
t
considerable portion of its high room-temperature strength is retained ai temperatures up to 1500- F. Typical applications include parts that require high strength and low plastic-flow rate at temperatures up to 1500 F such ,
as gas turbihe supercharger, and jet-propulsion parts, and springs for temperatures up to 10000F
1600
Nickel-Molybdenum-lron-based Alloys
1300
1
.
-
.
Hasfe//ojM(57Ni-20Mo-20Fe)and
r+L -
Hastelloy B (62Ni-28Mo-5Fe) are the two best-known alloys in this group.
Figure 12-36 shows the structure of Hastelloy B after solution annealing anc! water quenching. It consists of globular carbides in a matrix of f.c.c.y
1400
solid solution. These alloys are austenitic and therefore do not respond to age hardening. By cold working, it is possible to obtain strength and ductility comparable to those of alloy steel. These alloys are noted for their high resistance to corrosion by hydrochloric, phosphoric, and other nonoxiaiz-
1300 =
;
V
'
-
80
ing acids. They are used in the chemical industry for equipment to handle,
|
transport, and store acids and other corrosive materials.
Nickel-Chromium-Molybdenum-lron-based Alloys
1000
The remainder of the
Hastelloy alloys fall into this group, the best-known one being Hastelloy 0 (54Ni-17Mo-15Cr-5Fe-4W). These alloys are characterized by their high
600 \
3 +
\
400 Q
corrosion resistance to oxidizing acids such as nitric, chromic, and sulfuric acids. They generally have good high-temperature properties and are
200 -
e
0
20
resistant to oxidizing and reducing atmospheres up to 2000oF. They are J
30
used in the chemical industry, when dealing with strong oxidizing acids | ,
50
fig 12-37 The iron-nickel alloy system (From Metals
.
for pump and valve parts spray nozzles, and similar applications. Hastelloy;
40
Weight percenloc
-
,
;
Handbook, 1948 ed., p. 1211 American Society for Metals ,
. ..
J -
NONFERROUS METALS AND ALLOYS
513
X (47Ni-9Mo-22Cr-18Fe) has outstanding strength and oxidation resistance up to 2200,T. It is used for many industrial-furnace applications and for aircraft parts such as jet-engine tail pipes, afterburners, turbine blades, and vanes.
Nickel-Chromium-Molybdenum-Copper-based Alloys The alloys in this group were originally developed as materials resistant to both sulfuric and nitric acids over a wide range of concentration and exposure conditions. Two casting alloys are HWum B (50Ni-28Cr-8.5Mo-5.5Cu) and Illium G (56Ni22.5Cr-6.5Mo-6.5Cu). They provide superior corrosion resistance in machinable high-strength casting alloys. Typical applications are thrust and
(
rotary bearings and pump and valve parts where high hardness is required in corrosive environments. Illium R (68Ni-21Cr-5Mo-3Cu) is a machinable wrought alloy that provides heat and corrosion resistance. It is used for pump and valve shafting, hardware items, tubing, sheet, and wire. The nominal chemical composition and typical mechanical properties of some nickel alloys arc given in Table 12-1;.'
o 30 F and
X Eiiiilrix is .
()()
'
as,
"
(hoin
H12-33 Nickel-Iron Alloys The nickel-iron alloy system is shown in Fig. 12-37.
Amer-
.
Accurate phase boundaries below about IIOO' F have not yet been estab-
lished because of the sluggishness of structural changes at low temperatures. Nickel and iron are completely soluble in the liquid state and solidify as solid solutions. Nickel lowers progressively (he yto tv transformation in
h room-temperalure strength is retained at
[pica! applications include parts that require j
I
"
-
ow rate at temperatures up to 1500' F, such d jet-propulsion parts, and springs for lem-
1600 2800
d Alloys
l?00
Hastelloy A (57Ni-20Mo-20Fe) and «
e the two best-known alloys in this group, f
"
e of Hastelloy B after solution annealing and f iof globular carbides in a matrix of f.c.c.y v austenitic and therefore do not respond to it is possible to obtain strength and ductilsteei. These alloys are noted for their high ochloric phosphoric, and other nonoxidiz-
r
-
r + l
y+ l
L
2600
i
-
; : ;
,
5 1000
1600 1 .
BOO
'
chemical industry for equipment to handle.
2400
Dther corrosive materials.
200
Iron-based Alloys The remainder of the ' ;)up. the best-known one being Hastelloy C -
'
. .
"
A:
,
ng acids such as nitric chromic, and sul-
good high-temperature properties and are i cing atmospheres up to 2000' F. They are When dealing with Strong oxidizing acids X Fig ,
-
,
.
400
;:00
te
'
,
800
a
,
,
nozzles
a + y
\
ese alloys are characterized by their high
IC
20
30
40
50
Weight percentage nickel 12 37
The iron-nickel alloy system. (From Metals
and Similar applications. Hastelloy 1 Handbook. 1948 ed., p. 1211, American Socisty for Metals.)
1
60
80
90
N
Ill
'
:j
.
' .
.
V-J-
' '
.
.
'
' .
.
2
H
TABLE 12 12
X
Nominal Composition and Typical Mechanical Properties of Some Nickel Alloys'1
o o
MATERIAL
(ESSENTIAL ELEMENTS),
YIELD STRENGTH 0 2% OFFSET
TENSILE
NOMINAL COMPOSITION
STRENGTH,
CONDITION
.
ELONGATION, %
,
IN 2 IN.
1 000 PSI
1 000 PSI
%
c ,
o
BHN
z
,
,
H
o H
.
Monel
(wrought)
66.15 Ni(+Co), 31.30 Cu,
Annealed
75
35
40
125
1 35 Fe, 0.90 Mn
Hot-rolled
90
50
35
150
Cold-drawn
100
80
25
190
Cold-rolled
110
100
5
240
75
35
35
140
o -
.
o X c
-
CO
5 I
Monel
64.0 Ni(-rCo), 31.5 Cu:
(cast)
1 5 Si
K Monel
65.25 Ni(+Co), 29.60 Cu,
As cast
2 75 Al, 0.45 Ti .
63.0 NI(+Co), 30.5 Cu,
H Monel
3 2 Si
Annealed
100
45
40
155
Annealed"
155
100
25
270
i
150
140
5
300
o
185
160
10
335
As cast
115
70"
10
265
Inconel
77.0 Ni( Co), 15.0 Cr,
Annealed
85
35
45
150
(wrought)
7 0 Fe
Hot-rolled
100
60
35
180
Cold-rolled1
135
110
5
260
Inconel X
72.85 Ni(+Co), 15.15 Cr,
Annealed
115
50
50
150
6 80 Fe, 2.50 Ti,
Annealed"
175
115
25
300
.
.
0 75 Al .
"
00
Properties of Some Metals and Alloys.
The international Nickel Co.;
Metals Handbook, 1961 ed., American Society for Metals, Metals Park. Ohio. "
5 percent extension.
.
Hard temper
.
"
Age-hardened.
«»:
iijMiiijniii iiiiip im u,iiiiiiffa
.
,
TABLE 12-12
.
.
s,,
ii
ii .iiiniiiiiDMiMiu ai .
(Continued)
NOMINAL COMPOSITION MATERIAL
(ESSENTIAL ELEMENTS),
TENSILE
CONDITION
%
STRENGTH 1
,
000 PSI
YIELD STRENGTH ,
0 2% OFFSET
.
ELONGATION % ,
.
1 000 PSI
,
IN 2 !N.
BHN
,
Hastelloy alloy A
Bal. Ni 22 Mo.
As cast
73
44
22 Fe, Mn Si
10
Rolled6
180
115
50
44
210
Hastelloy alloy B
Bal. Ni 28 Mo, 5 Fe, Mn Si
Sand-caste
Hastelloy alloy C
Bal. Ni 16 Mo.
Sand-caste
16 Cr 5 Fe,
Rolled6
130
71
,
,
,
,
,
,
Rolled6
80
50
9
199
120
56
50
215
78
50
4 W, Mn Si
5
199
45
204
0-2'
321
,
Hastelloy alloy D
Bal. Ni 10 Si, 3 Cu, Mn
Sand-cast6
118
118
Hastelloy
47 Ni, 9 Mo
Wrought sheet
114
52
alloy X
43
22 Cr, 8 Fe C. W
90s
Sand-cast
65
42
11'
89s'
,
,
,
t
Spring temper Spring temper0 .
.
'
S m
.
.<
mi,
fjif: r
H Monel
63.0 NI(+Co), 30.5 Cu,
.
-
vi ii iccfixro
Spring temper Spring temper"
150
140
5
300
185
160
10
335
As cast
115
70 b
10
265
150
>w;w. ... . ?
r
3 2 Si .
Inconel
77.0 Ni(+Co), 15.0 Cr.
Annealed
85
35
45
(wrought)
7 0 Fe
Hot-rolled
100
60
35
180
Cold-rolledc
135
110
5
260
.
Inconel X
72.85 NI(+Co), 15.15 Cr,
Annealed
115
50
50
150
6 80 Fe, 2.50 Ti.
Annealed"
175
115
25
300
.
0 75 Al .
a"'Properties ''
of Some Metals and Alloys
" .
O
S percent extension. cHard temper
The International Nickel Co.: -Metals Handbook," 1961 ed. American Society for Metals Metals Park Ohio .
.
,
.
.
dA9e-hardaned.
TABLE 12-12
(Continued)
NOMINAL COMPOSITION MATERIAL
(ESSENTIAL ELEMENTS),
CONDITION
Hastelloy alloy B
Hastelloy alloy C
000 PSI
,
.
1 000 PSI
,
ELONGATION % ,
IN 2 IN.
BHN
,
As cast
73
44
10
180
22 Fe, Mn, Si
Rolled15
115
50
44
210
Bal. Ni, 28 Mo, 5 Fe, Mn, Si
Sand-caste
Bal. Ni, 16 Mo.
Sand-caste
16 Cr, 5 Fe, 4 W. Mn, Si
Rolled'
Bal. Ni, 10 Si.
Sand-caste
118
118
114
65
Rolled8
3 Cu, Mn
Hastelloy
47 Ni, 9 Mo,
Wrought sheet
22 Cr, 8 Fe, C: W
Sand-cast
50 Ni, 28 Cr,
IlIiumB
0 2% OFFSET.
Bal. Ni, 22 Mo.
Hastelloy alloy D alloy X
YIELD STRENGTH
STRENGTH, 1
%
Hastelloy alloy A
TENSILE
8 5 Mo, 5,5 Cu .
Grade B1
80
50
9
199
120
56
50
215
78
50
5
199
130
71
45
204
0-2 f
321
52
43
90s
42
11'
89"
1 0-4.5
200-240
61-67
50-62'
.
(2.5-4.5% Si)" Grade B4
z
45-51
45-51'
0 5 max .
325-360
56 Ni, 22.5 Cr.
Cast
m
33 33
(6.1-6.3% Si)" Illium G
z
O
O
68
50
32
200
6 5 Mo, 6.5 Cu
C cn
S
.
m
Illium R
68 Ni, 21 Cr, 5 Mo, 3 Cu
Solution-treated
112.8
50.2
45.7
162
20% cold-worked
142.3
128.1
11. 5
238
>
> z a
eAnnealed ' In 1 in.
.
> -
Rockwell B. hGrade defined ' Elastic limit.
i
'
by silicon content.
«j
"
i
1
4f> 516
INTRODUCTION TO PHYSICAL METALLURGY
iron. Alloys that contain up to 6 percent nickel are ferritic. As the nickel content Increases, the alloys have an increasing tendency to air-harden on slow cooling. Alloys with 6 to about 30 percent nickel are martensitic after fast cooling. After slow cooling or reheating, they decompose into « plus) phases. The amount of each phase present is dependent upon the nickel
percent nickel alloy covered with ar
Dumet wire and is used to replace pi tubes.
An alloy containing 36 percent nil as Elinvar has a zero thermoelastic elasticity is almost invariable over a
content, heat treatment, and amount of cold working. Alloys containing more than 30 percent nickel are predominantly austenitic and nonmagnetic.
is used for hair springs and balance \
Alloys of iron and nickel, containing 20 to 90 percent nickel, have wide
in precision instruments Permalloys include several nickel.
t
application because of their useful thermal expansion and magnetic and thermoelastic properties. As the nickel content is increased above 25 percent, thermal expansion falls off sharply, becoming almost invariable, for ordinary ranges in temperature, at 36 percent nickel. Further additions of nickel result in an increase in thermal expansion. Figure 12-38 shows the effect of nickel on the coefficient of linear thermal expansion of iron-nickel alloys at room temperature. The 35 percent nickel alloy is known as Invar, meaning invariable, and is used where very little change in size with change in temperature is desirable. Typical applications include length standards, measuring tapes, instrument, parts, variable condensers, tuning forks, and special springs.
nickel that have high magnetic pen weak magnetizing forces They als( electrical resistivity Permalloy parts .
.
communication circuits
taining 8 to 12 percent aluminum
permanent magnets in motors, gene ceivers
microphones, and galvanomi By variation in the percentage of r chromium copper, and molybdenum ,
LEAD AND LEAD ALLOYS
12-34 Lead The major properties of lead softness, malleability low melting poir
static bimetals, thermoswitches, and other temperature-regulating devices.
,
Alloys containing approximately 28 percent nickel, 18 percent cobalt, and 54 percent iron have coefficients of expansion closely matching those of standard types of glass. They are used for matched glass-to-metal seals
properties, low electrical conductivity high corrosion resistance.
By far the largest tonnage of lead is followed by the use of tetrae in high-test gasoline. Lead compou
under the trade names of Kovar and Fernico. A 46 percent nickel alloy,
batteries
,
called Platinite, has the same coefficient of expansion as platinum. A421 ! i
£2 4
J
,
3 20
,
30
10
50
50
70
B0
100
Percem nickel
Fig. 12-38
Effect of nickel on the coefficient of linear
thermal expansion of iron-nickel alloys at room temperature.
(After Guillaume from "Metals Handbook," vol. 1, American Society for Metals, Metals Park, Ohio, 1961.)
m
m
its high density for shieldin ness for gaskets and for calked joint; for cable sheathing. As a coating on w Advantage is taken of the high corro; equipment in the chemical industry as ing industry as pipe for transporting v, improve the machinability of bronzes : t2-35 Lead Alloys Antimony and tin are the lead. The antimony-lead phase diagr simple eutectic system with the eutect f,
Si
-
many high-grade paints. The high weight of lead makes it sui balances
1 12
. .
,
properties can be secured.
'
6
14
,
cobalt, have outstanding magnetic p
In, the range of 30 to 60 percent nickel, it is possible to select alloys ol appropriate expansion characteristics to fit particular applications. Alloys containing 68 percent iron, 27 percent nickel, and 5 percent molybdenum. or 53 percent iron, 42 percent nickel, and 5 percent molybdenum have high coefficients of thermal expansion. They are used in combination with a low-expansion alloy to produce movement. Applications include thermo-
o
.
The aluminum-nickel-cobalt-iron e
,
Hi
k -f
NONFERROUS METALS AND ALLOYS
517
|
percent nickel are ferritic. As the nlc an increasing tendency to air-harden ( )ut 30 percent nickel are martensitici reheating, they decompose into aplus se present Is dependent upon the nic
percent nickel alloy covered with an oxidized copper plating is known as
Dumet wire and is used to replace platinum as the "seal-in" wire in vacuum
.
i
tubes.
'
An alloy containing 36 percent nickel and 12 percent chromium known as Elinvar has a zero thermoelastic coefficient; that is the modulus of ,
junt of cold working. Alloys containif»9| edominantly austenitic and nonmagnetic.
elasticity is almost invariable over a considerable range in temperature It is used for hair springs and balance wheels in watches and for similar parts .
ling 20 to 90 percent nickel, have wid*||
in precision instruments.
ul thermal expansion and magnetic andj
«
Permalloys include several nickel-iron alloys in the range of 78 percent nickel that have high magnetic permeability under the influence of very weak magnetizing forces. They also have low hysteresis losses and low electrical resistivity. Permalloy parts are used as loading coils in electrical
ed above 25 percent, thermal expansion invariable, for ordinary ranges in tempersr additions of nickel result in an increase
communication circuits.
.
38 shows the effect of nickel on the co
isers, tuning forks, and special springs.? | nickel, il is pos iblo to solocl alloysotj
The aluminum-nickel-cobalt-iron alloys commonly called Ainico, containing 8 to 12 percent aluminum 14 to 28 percent nickel, 5 to 35 percent cobalt, have outstanding magnetic properties. These are widely used as permanent magnets in motors, generators, radio speakers, telephone receivers, microphones, and galvanometers. By variation in the percentage of nickel and proper additions of cobalt, chromium, copper, and molybdenum, different combinalions of magnetic
ptics to fit particular applications. Alloy* i
properties can be secured,
,
on of iron-nickel alloys at room tempera-.|
,
/ is known as Invar, meaning invariabffi.J -Jge in size with change in temperatureis| '
.
.
;; plude length standards, measuring tapes,| |: '
;
rcent nickel, and 5 percent molybdenum,
eel, and 5 percent molybdenum have high
:
They are used in combination with 3
i
.
AND LEAD ALLOYS
novement. Applications include thermo- J;12.34 Lead The major properties of lead include heavy weight high density, ,
nd other temperature-regulating devices
y 28 percent nickel, 18 percent cobalt, and of expansion closely matching those of
je used for matched glass-to-metal seais i and Fernico. A 46 percent nickel alloy, \ efficient of expansion as platinum.
softness, malleability, low melting point, and low strength. It has lubricating properties, low electrical conductivity, high coefficient of expansion, and high corrosion resistance.
By far the largest tonnage of lead is used in the manufacture of storage batteries, followed by the use of tetraethyl lead as Iho antiknock ingredient in high-test gasoline. Lead compounds are used in the manufacture of many high-grade paints.
The high weight of lead makes it suitable for use as weights and counterbalances, its high density for shielding against /j rays and y rays, its softness for gaskets and for calked joints in cast-iron pipe, and its flexibility for cable sheathing. As a coating on wire, lead acts as a drawing lubricant. Advantage is taken of the high corrosion resistance of lead by its use for equipment in the chemical industry as a roofing material, and in the plumbing industry as pipe for transporting water and chemicals. Lead is used to improve the machinability of bronzes brasses, and free-machining steels. 12-35 Lead Alloys Antimony and tin are the most common alloying elements of ,
100 "
4
,
ar
.
iperature HUM ic.'lll
v
.
lend. Tho ;inlimony-lend phnsn dinqram is shown in Fiq fi ?9. This is a simple euteclic system with the eulectic composition at 1 1.2 percent anli-
.v
.
4:
518
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 12-13
Properties of Cast Lead-Antimony
Alloys*
i ;
TENSILE BHN
STRENGTH,
ANTIMONY, %
,
PSI '
!
used solders are those containing a
o
2 500
40
or 50 percent each with or without
i
3 400
70
2
4 200
80
3
4 700
9 1
metal, a lead-tin alloy containing fn steel sheets for roofing and automc
4
5 660
10.0
5
6 360
11.0
6
6 840
11.8
7
7 180
12.5
8
7 420
13.3
9
7 580
14.0
10
7 670
14.6
11
7 620
14.8
12
7 480
15.0
13
7 380
15.2
14
7 000
15.3
,
.
,
.
,
.
,
.
,
,
Lead alloys containing bismuth point eutectic. These alloys are use and boiler plugs. Lead-tin-antimony alloys are wide metals. The lead base provides lov casting; additions of antimony prov also lower the casting temperature duce brittleness, and impart afinersl only as a backing material for the e quired to resist wear, contains the Ic Foundry type metal on the other ha hand composition. Since the cast t ,
,
,
,
,
i ,
I
,
,
,
,
,
Lead in Modern Induslry. Lead Induslries Associa"
From
,
tion, New York, 1952.
'
r-V-1
diagram, it is also a simple eutectic at 61.9 percent tin and 3610F. Althc used for their melting characteristii ness and stre ngth as shown by the
quires the hardest, most wear-resista llization
mony. Antimony is generally added to lead to raise the recrysta temperature and to increase hardness and strength, as shown by the values
7>
mIS
m
in Table 12-13. Lead-antimony alloys contain from 1 to 12 percent antimony and are used for storage-battery plates, cable sheathing, collapsible tubes,
MS
as
and for building construction.
The lead-tin alloy system is shown in Fig. 6-30. Like the lead-antimony
E
i
s TABLE
12 14
Properties of Lead-Tin
t
!
Alloys*
to
TENSILE TIN. %
STRENGTH,
3
BHN
V,
2£
PSI
-
5
3 200
80
10
4 100
11.5
15
4 900
12.0
20
5 400
11.7
30
6 200
12.4
40
6 600
13.0
50
7 000
14.3
50
7 200
10.7
.
ml
,
,
,
,
,
<
m
m m
i
m
i mm m:2
m
,
,
m
,
fig 12 39 Linotype metal: 12 percent antimony 4 perc tin, 84 percent lead. Almost entirely a ternary eutectic Sliucture. (American Smelting and Refining Company.) ,
'
From -'Lead in Modern Industry." Lead Indus-
tries Association. New York, 1952.
i
m
i
SB
NONFERROUS METALS AND ALLOYS
519
diagram, it is also a simple euteclic system with tho ouloctic point located
s ot Cast Lead-Antimony
at 61.9 percent tin and 361'F. Although lead-tin alloys are most commonly used for their melting characteristics as in solder, tin also increases hard,
TRENGTH,
BHN
ness and strength, as shown by the values in Table 12 -14. The most widely
40
or 50 percent each, with or without small percentages of antimony. Terne
si
used solders are those containing about 40 percenl tin and 60 percent lead
,
2 500
.
,
b 400
70
200
80 .
,
[1 700
.
10.0
360
11.0
,
,
6,840
11.8
SW,180
12.5
M20
13.3
7 580
14.0
7 670
14.6
7 620
14.8
,
,
,
/ iBO
lb 0
7 380
15.2
000
15.3
.
.
.
,
.
.
,
i
.
steel sheets for roofing and automotive fuel-tank applications. Lead alloys containing bismuth tin, and cadmium form a low-meltingpoint eutectic. These alloys are useful in electric fuses, sprinkler systems,
9 1
,
B 660
p
metal, a lead-tin alloy containing from 10 to 25 percent tin, is used to coat
.
,
,
and boiler plugs.
!
Lead-tin-antimony alloys are widely used in the printing industry as type metals, The lead base provides low cost low melting point and ease in casting; additions of antimony provide hardness and wear resistance and also lower the casting temperature; additions of tin increase fluidity, re,
duce brillleness, and impart alinerstructure. Electrotype metal, being used only as a backing material for the electroformed copper shell and not required to resist wear, contains the lowest percentages of tin and antimony.
i
Foundry type metal, on the other hand, is used exclusively to cast type for
::
liy," Load liuiusliica Ashocih-
hand composition. Since the cast type is used over and over again, it re-
::
Jed to lead to raise the recrystallization ;
quires the hardest, most wear-resistant alloy that is practical to use. Found-
i
less and strength, as shown by the values .\ U
lwn in Fig
:
.
,
collapsible tubes, Jt;
6-30. Like the lead-antimony |
.
-
>
. .
:
ferties ot
:
S5 4
.
;:
.
m
i
Lead-Tin
m "
is
mi
ys contain from 1 to 12 percent antimonyl }
jlates, cable sheathing
vlLE NGTH,
Si
BHN
7i
1
ft
80 .
11.5
i
12.0
i
11.7 ;':
12.4
" :-.:
13.0
mm
10.7
fig. 1£-39 3rk, 1952.
.
m
4 5
si
1
mm
14.3
Industry," Lead Indus-
,
Si
as*
m
(if
Linotype metal: 12 percent antimony 4 percent tin, 84 percent lead. Almost entirely a ternary eutectic slructure. (American Smelting and Refining Company.) ,
"
.
V:
520
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 12.15
Nominal Composition and Typical Mechanical Properties of Some Lead Alloys* NOMINAL COMPOSITION, %
ALLOYS Pb
Chemical lead
Corroding lead
I
Sb
Sn
Table 1 .15 (Continued) TENSILE
YIELD
STRENGTH
STRENGTH
PSI
PSI
%
2385
1 180
29
Sand-cast
1 800
800
30
s;
Chill-cast
2 000
47
4
4
CONDITION
OTHERS
.
Rolled sheet
99.9-1-
99.73 +
ELONGATION
,
,
,
,
0 10
Arsenical lead
Bal.
Calcium lead
Bal.
Soft solders
97.5
1
95
5
.
0 15 As. 0.10 Bi
Extruded sheath
2 500
40
0 028 Ca
Extruded and aged
4 500
25
.
,
.
1
.
,
5 Ag
20
50
50
13 1 500
50
8
5 800
3 650
16
11.
6 100
4 800
60
14.
,
.
t
Antimonial lead
Hard lead
Type metal
.
,
1
Extruded and aged
3 000
50
7
91
9
Chill-cast
7 500
17
15.
,
,
96
4
Cold-rolled 95%
4 020
48.3
94
6
Cold-rolled 95%
4 100
47
95
2 5 .
,
,
25 6
86
11
3
78
15
7
61
25
2 Cu
SAE 13
85
10
5
flAE 14
75
15
10 .
SAE 15
83
15
1
G
83.5
12.75
0 75
12.
t:
.
14
:
i
,
99
80 A-;
23
Cast
19 24
,
I
Lead-base babbitt;
"
Chill-cast ;
1 As .
,
3 As
19
4
22
10,350
2
erally includes up to 2 percent copper as an additional hardener. Linotype g and Intertype casting machines die-cast an entire line of type characters at each casting. It is important that the alloy used should have a low melting M
.
:-
22
The nominal composition and typi alloys are given in Table 12-15. TIN AND TIN ALLOYS
point and a short temperature range during solidification. Therefore, the Ir 12-36 Tin ternary eutectic alloy or compositions near this are preferred. Figure 12-39 shows the microstructure of linotype metal containing 12 percent antimony,; 4 percent tin, and 84 percent lead. This alloy has a liquidus temperature oU 4630F and a solidus temperature of 4620F, and the structure is almost en-
tirely a ternary eutectic mixture.
;|
Lead-base bearing alloys are known commercially as babbitts or white metal alloys. One group includes the alloys of lead-tin-antimony and usiH
ally arsenic, while the other group includes alloys of lead and tin with smatt,
percentages of calcium, barium, magnesium, and sodium. Figure 12-40 shows the microstructure of a lead-base bearing alloy. It consists of cubes;
of primary antimony-tin compound in a binary eutectic mixture of lead amj:
tin solid solutions. These alloys are used for automotive connecting rod l main and camshaft bearings, diesel-engine bearings, railroad-car joun bearings, and many electric-motor bearings.
m
20
15
Metals Handbook," 1961 ed., American Society for Metals, Metals Park, Ohio.
'
.
5
10,500
,
.
.
10.000
9 800
ry type metal contains the largest amounts of tin and antimony, and gen- M
-
.
3 400 ,
80
,
I
Tin is a white, soft metal that hi
lubricating properties. It undergoes normal tetragonal structure (white tii ature of 55.80F. This transformation from 7.30 to 5.75
and the resulting metal to coa'se powder known as ti very sluggish and considerable ur Common impurities in tin tend to de ordinary conditions the transformal ,
,
,
Over half the primary tin used in other metals primarily steel in the ,
copper tubing is useful for handlin centages of carbon dioxide and o; element in copper aluminum, and ,
sections.
WW
"
NONFERROUS METALS AND ALLOYS
Mechanical Properties of Some Lead Alloys*
521
e 12.15 (Continued) TENSiLE
YIELD
ELONGA-
STRENGTH
STRENGTH,
TION,
PSI
%
1 180
29
COMPOSITION, % CONDITION
'
' .
|psi
OTHERS Rolled sheet
2 3B5 .
,
TYPICAL USES
BHN
Material of construction in the
chemical industry Sand-cast
1,800
Chill-cast
J 000
47
42
Storage batteries, cable sheathing, paint, calking, antiknock fluid, liquid metal for heat treating
2 500
40
49
Cable sheathing
4 500
25
800
.
0 10 .
0 15 As 0.10 Bi .
Extruded sheath
,
.
0 028 Ca
Extruded and aged
.
1
1
.
.
Cable sheath and creep-resistant pipe
3 400
1 500
50
8
20
5 800
3 650
16
11.3
6 100
4 800
60
14.5
.
,
50
Chill-cast
7 500
17
,
,
'
7
15.4
4
5 r 4 020 i f 4,100 .
-
.
12.4
n::
::: Cast
;
r
23
19
.
24
2 Cu
5
Chill-cast
19
4
22
Moderate loads: blowers, pumps
10 350
2
20
High loads; diesel-engine bearings
22
Elevated-temperature bearing; trucks
.
.
3 As
15
9 800
.
.
3
,
Metals Park, Ohio.
The nominal composition and typical mechanical properties of some lead
3t amounts of tin and antimony and genopper as an additional hardener. Linotype ,
SS£ilie-cast an entire line of type characters a( t the alloy used should have a low melting '
ange during solidification. Therefore the ,
tions near this are preferred Figure 12-39 .
ype metal containing 12 percent antimony.
alloys are given in Table 1215, TIN AND TIN ALLOYS
12-36 Tin Tin is a white, soft metal that has good corrosion resistance and good lubricating properties. It undergoes a polymorphic transformation from the normal tetragonal structure (white tin) to a cubic form (gray tin) at a temper-
1 This alloy has a liquidus temperature of ;
ature of 55.8"F. This transformation is accompanied by a change in density from 7.30 to 5.75, and the resulting expansion causes disintegration of the
of 462"F, and the structure is almost en-
metal to coarse powder known as tin pest. However, the transformation is very sluggish, and considerable undercooling is necessary to initiate it. Common impurities in tin tend to delay or inhibit the change so that, under
m known commercially as babbitts or white
is the alloys of lead-tin-antimony and usup includes alloys of lead and tin with small '
'v
,
magnesium, and sodium. Figure 12-40
jad-base bearing alloy
.
It consists of cubes
ind in a binary eutectic mixture of lead and
are used for automotive connecting rods 1 iesel-engine bearings railroad-car journal ; tor bearings. .
,
....
i
Light loads, car journal bearings
5
10 500 ,
1 As
0 75
Electrotype Stereotype Linolype Monotype Foundry type
10 000 ,
1
Cable sheathing Storage-battery grids
47
.
25
10
Coating and joining metals, body solder
Rolled sheet and extruded pipe
48,3
,
Cold-rolled 95%
mi?
,
3 000
Cold-rollpd 95%
6
,
Extruded and ag«S s -
v-v
.
5 50
-
.
13 ,
-
.
.
5 Ag
,
i
3 2-4 5
30
_
J L
;
ordinary conditions, the transformation is of no practical importance. Over half the primary tin used in this country goes into the coating of other metals, primarily steel in the manufacture of tin cans. Tin-coated copper tubing is useful for handling fresh waters that contain large percentages of carbon dioxide and oxygen. The use df tin as an alloying element in copper, aluminum, and lead has been discussed in preceding sections.
522
INTRODUCTION TO PHYSICAL METALLURGY
The most common alloying elei produce pewter and the tin-base
bearing applications. Figure 12' base babbitt.
There are CuSn roc
large cubes of SnSb compound
,
The SnSb cubes are extremely hu resistance of babbitt.
I
The nominal composition and t;
>
alloys are given in Table 12-16.
Fig. 12-40 Lead-base bearing alloy: 15 percent antimony 5 percent tin, 80 percent lead. Cubes of primary antimonytin compound (white) in a binary euteclic mixture of lead and tin solid solutions, 75x. (American Smelting and Refining Company.) ,
O
Fig. 12-41 Tin-base hard babbitt of 84 percent tin, 1 cent copper, and 9 percent antimony, 50x. Star-shaf CuSn compound and rectangular crystals of SnSb C( pound in a ductile ternary euteclic matrix. (By perm
12-37 Tin Alloys Lead is alloyed with tin to produce several soft solders that have higher strength than the lead-base solders. Tin solders containing 5 percent antimony or 5 percent silver are preferred for electrical equipment because these solders have higher electrical conductivity than the high-
Irom R. M. Brick, R. B. Gordon, and A. Phillips, "Stn and Properties of Alloys, 3d ed., McGraw-Hill Book "
lead alloys. TABLE 12-16
Company, New York, 1965.)
Nominal Composition and Typical Mechanical Properties of Some Lead Alloys3
Table 12.16 (Continued i
51; c-:-?":
NOMINAL COMPOSITION, %
ALLOY Sn
Tin (pure)
99.8 min
Hard tin
99.6
Cu
Sb
Pb
CONDITION
Ag Casl
TENSILE
YIELD
STRENGTH,
STRENGTH,
PSI
PSI
3 100
.
/c IN 2 IN
.
551
,
80% reduction
04
ELONGATION °
4 000 ,
Antimonial tin solder
95
Tin-silver solder
95
Cast
5
1
5 900
38 i
,
Sheet
4 600 .
30
70
Soft solder
Cast
3 600 ,
49
5 800 .
37
63 91
Tin babbitt
45 .
Cast Chill-cast
4 5 .
,
9 300
4 400:;:
11.200
6 100 +
,
89
75 .
Chill-cast
35 .
84
8
8
65
15
2
92
8
Die-cast
321
7 500 ,
2
,
10 000 ,
18
Die-cast
7 800 ,
White metal
Chill-cast
15 .
7 200 ,
91
Pewter
7
2
Annealed sheet
3 600 ,
."Metals Handbook, tin 4 in.
"
1961 edj. American Society for Metals, Metals Park, Ohio.
tCompressive yield (0.125 perifent offset).
!
I
f
i
I
40
'
I
-
_
SY
.
NONFERROUS METALS AND ALLOYS
523
The most common alloying elements for tin are antimony and copper to produce pewter and the tin-base babbitts that are used for high-grade bearing applications. Figure 12-41 shows a typical microstructure of tinbase babbitt. There are CuSn rods arranged in a star-shaped pattern and large cubes of SnSb compound, all in a ductile tin-rich ternary eutectic. The SnSb cubes are extremely hard and contribute to the excellent wear resistance of babbitt.
The nominal composition and typical mechanical properties of some tin alloys are given in Table 12-16.
f
t
;
v;v/i)timony Intimony/ ,
'
'
'
-
'
'
-
of lead ind I
-
Fig 12-41
Tin-base hard babbitt of 84 percent tin, 7 perStar-shaped CuSn compound and rectangular crystals of SnSb compound in a ductile ternary eutectic matrix. (By permission from R. M. Brick R. B. Gordon, and A. Phillips, "Structure and Properties ol Alloys, 3d ed., McGraw-Hill Book Company. New York 1965.)
n to produce several soft solders that have
.
nc-nt copper and 9 percent antimony, 50x. ,
se solders. Tin solders containing 5 persr are preferred for electrical equipment ner electrical conductivity than the high-
,
''
,
.-
f
Mechanical Properties ol Some Lead Alloys'
COMPOSITION % ,
Cu
Pb
Table 12.16 (Continued) TENSILE
CONDITION
STRENGTH
Ag
u
_
04
37
Chill-cast Chill-cast
.
.
8 2
,
3 600 ,
53 .
49
6 800
12
,
7 500
321
,
9 300
4 400t
11,200
6 100*
,
!0,000
Die-cast
7 800 ,
Chill-cast 2
BHN
381
,
4 600
Die-cast 18
551
,
5 900
Cast
35
t IN ? IN.
,
Cast
45
ELONGATION, 0
4 000
Sheet 30
,
3 100
80% reduction Cast
.
STRENGTH PSI
If ~
Cast
YIELD ,
,
2
1
15 .
,
8 600 ,
Automotive applications, better
24
corrosion and wear resistance than
30
lead-base bearing alloys
23
Castings for costume jewelry
20
7 200
Annealed sheet
40
Electrotinning, alloying Collapsible tubes and foil Solder for electrical equipment Solder for electrical equipment For joining and coating of metals
14 17
,
TYPICAL USES
95 .
Vases, candlesticks, book ends
b Melals Park, Ohio. .
i
4
a
.
v
1 524
INTRODUCTION TO PHYSICAL METALLURGY
Commercially pure titanium is lo\ and less expensive than titanium a high ductility for fabrication but li piping, valves and tanks, aircraft fi
TITANIUM AND TITANIUM ALLOYS
12-38 Titanium
The process of producing titanium sponge by the magnesium
reduction of titanium tetrachloride was discovered by W. J. Kroll in 1938. Shortly afterward the United States ar/ned services became interested in
12-39 Titanium Alloys
the metal primarily because of its high melting point (3035oF). There was the possibility of developing titanium alloys with strength at elevated
The addition of
ence the alpha to beta transformal
to refer to alloying elements as alp
temperatures which might substitute in military equipment for nickel-base and cobalt-base alloys. Titanium has a density of about 0.16 Ib/cu in. com-
pared with steel at 0.28. Therefore, titanium alloy structures have a high strength-weight ratio and are particularly useful for aircraft parts. Titanium has excellent corrosion resistance up to approximately 1000CF. While titanium is the fourth most abundant element in the earth's crust, it is relaWeight percent
tively expensive to obtain from its ores. Titanium has a strong affinity for the gases hydrogen, nitrogen, and oxygen, all of which form interstitial solid solutions with titanium. They all
2 4 6 8 10 mo
i
_-
-
i-j
i
i
20
30
1
_
1720° 1700
have a marked strengthening effect, as illustrated for nitrogen in Fig. 12-42. When the amount of absorbed oxygen, nitrogen, or hydrogen exceed speci-
1600
fied limits, they embrittle titanium, reducing impact strength, and cause
V
brittle failure under sustained loads at low stresses. 1500
Titanium metal has a close-packed hexagonal crystal structure, called
48.5
alpha, at room temperature. This structure transforms to body-centered cubic beta at 16250F(8820C).
1400
1300
427
I
(29)/l2400
40 :
at
1200
100
o
noo
Tensile strength
80
a)
1000
400
60
a
2
300
40 -
900 882°
(13
20 -
200 ?
-
a
5
Hardness
56,5 300
100
40 30
4j 35
124.5)
'
'
00
I
-
Elongation
20
600
10
0
Fig. 12-42
\ 01 .
0 2 i0.3 0 4 0 5 0 6 Weight-percent nitrogen ,
.
,
,
07
0 Ti
,
20
30
40
50
Atomic percent
Increase in strength resulting from the presence
of nitrogen in otherwise pure titanium. (From Brick, Gordon, and Phillips, "Structure and Properties of Alloys," 3d ed., McGraw-Hill Book Company, New York, 1965.)
10
Fiy. 12-43 "
The Ti-AI phase diagram. (From Max Han
Constitution of Binary Alloys," 2d ed., McGraw-Hill E Company, New York 1958.)
1 a
NONFERROUS METALS AND ALLOYS
I
Commercially pure titanium is lower in strength, more corrosion-resistant, and less expensive than titanium alloys. It is used for applications requiring high ductility for fabrication but little strength, such as chemical process
vS?Sg titanium sponge by the magnesium was discovered by W. J. Kroll in 1938.
piping, valves and tanks, aircraft firewalls, tailpipes, and compressor cases.
5 armed services became interested in
iigh melting point (3035oF). There was )ium alloys with strength at elevated te in military equipment for nickel-base as a density of about 0.16 Ib/cu in. comtitanium alloy structures have a high ularly useful for aircraft parts. Titanium
2,
s#|up to approximately 1000oF
.
While ti-
i
12-39 Titanium Alloys The addition of alloying elements to titanium will influence the alpha to beta transformation temperature. It |s common practice to refer to alloying elements as alpha or beta stabilizers. An alpha stabilizer
'
J
l i
jt element in the earth's crust, it is relabres.
::
Weight percent aluminum 2 4 6 8 10
the gases hydrogen, nitrogen, and oxysolid solutions with titanium. They all as illustrated for nitrogen in Fig. 12-42.
1800
-
i
i
-
j
j
15
i
-
20
30
40
50
60
/0
80
90
1
_
,,
M-vjen nitrogen, or hydrogen exceed speci,
,
reducing impact strength, and cause
V s at low stresses.
.
.
.
M60g 53
.
500
ked hexagonal crystal structure, called structure transforms to body-centered
48,5
0 42 48 5
i 100 I1
4M
V
(V).51
II
r
-
li
65°
660°
* ;
:: -
,
;J
-
-
-
.
1
1
li
50 presence
Fig. I2 '13 Tho Ii-AI phase diagram (( rem Max Hanson, Constitution of Binary Alloys 2d ed.. McGraw-Hill Book Company, New York 1958.) ,
1
fU
Aiomic percent aluminum
"
-
"
Mloys."
"
,
4
i
525
30
!00 Al
526
INTRODUCTION TO PHYSICAL METALLURGY
I
means that as solute is added, the alpha to beta transformation temperature
cause the beta phase to persist dc stronger than alpha alloys. The alloying additions in solution is ;
is raised; similarly, a beta stabilizer lowers the transformation temperature. Aluminum is an alpha stabilizer, as may be seen from the Ti-AI phase diagram (Fig. 12-43). Important beta stabilizers are chromium, molybdenum,
,
alpha phase in alpha-beta alloys is beta alloy is stronger yet especially
vanadium, manganese, and iron. Ti-Mo and Ti-V alloy systems (see Fig. 12-44) show complete solid solubility, forming the beta solid solution over
,
shows the alpha-beta microstruct
the entire range. The alpha-phase field is severely restricted, the maximum extent being 1.8 percent Mo and 3.5 percent V. The Ti-Mn phase diagram, shown in Fig> 12-45, illustrates a beta stabilizer by means of a eutectoid reaction. In eutectoid systems intermetallic compounds always occur. Tin is substantially neutral in the amount present in commercial alloys. The relative amounts of alpha and beta stabilizers in an alloy, and the heat treatment, determine whether its microstructure is predominantly one-phase alpha a mixture of alpha and beta, or the single-phase beta over its useful
alpha-beta alloys can be further si
Weighl peicerr V
10
50
,
temperature range.
Properties are directly related to microstructure. Single-phase alloys are
weldable with good ductility; some two-phase alloys are also weldable, but their welds are less ductile. Two-phase alpha-beta alloys are stronger than the one-phase alpha alloys, primarily because b.cc. beta is stronger than . alpha. \ Most important, two-phase alloys can be strengthened by heat treatment because the microstructure can be manipulated by controllc.p.h
ing heating, quenching, and aging cycles. Alpha Alloys Most of the alpha alloys contain some beta-stabilizing alloying elements/ The compositions of these alloys are balanced by high aluminum content so that the alloys are essentially one-phase alpha. Figure 12-46 shows coarse, platelike alpha in aTi-5AI-2.5Sn alloy after hot working and annealing. The alpha alloys have two main attributes: weldability and retension of strength at high temperatures. The first results from the onephase microstructure, the second from the presence of aluminum. Alloying elements in solution stiengthen the alpha-phase alloys, and aluminum is the most effective strengthener of alpha alloys. Especially important, its effect persists to high temperatures. Hot working of alpha alloys contain'
1
ing more than about 6 percent aluminum is difficult. Hot workability of high-aluminum alpha is improved by additions of beta-stabilizing alloying elements in amounts small enough so that the beta phase is present in small quantities in the annealed microstructure. Some applications of Ti-5AI-2.5Sn alloy include aircraft tailpipe assemblies and other formed sheet components operating up to QOCTF, and missile fuel tanks and structural parts operating for short times up to 1100oF. Alpha-Beta Alloys These contain enough beta-stabilizing elements to
[ In-Ill
;
600
0
?
10
u
r
-
iQ
40
50
Atomic percent \
Fig. 12-44 The Ti-V phase diagram. (From Max Hans Constitution of Binary Alloys," 2d ed,. McGraw-Hill B Company, New York, 1958.)
m
?0
NONrEHROl IS METALS AND ALLOYS
;
iha to beta transformation temperature )wers the transformation temperature.
cause the beta phase to persist down to room tomperatute, and they are
stronger than alpha alloys.
gpv£fiay be scon (rem \hp Ti-AI phase dia-
"
.
forming the beta |olid solution over
Id is severely restriisted, the maximum . I percent V. The Ti-Mn phase diagram, la stabilizer by means of a eutectoid i letallic compounds always occur. Tin it present in commercial alloys. .Sg
If the
alpha phase in alpha-beta alloys is strengthened by aluminum, the alphabeta alloy is stronger yet, especially at elevated temperatures. Figure 12-47 shows the alpha-beta microslructure in an annealed Ti-8Mn alloy. The alpha-beta alloys can be further strengthened by heat treatment. Essen-
Mo and Ti-V allo systems (see Fig
-
,
The beta phase, as strengthened by beta-
alloying additions in solution, is stronger than the alpha phase.
abilizcrs are chromium molybdenum, >
527
Wi-niM pri i en! v.lll-li!ii;iii
The
II/
,M
',(1
'U
/I
Sibilizers in an alloy, and the heat treattructure is predominantly one-phase r the single-phase beta over its useful icrostructure. Single-phase alloys are yo-phasc alloys are also weldable, but
...
/iSe alpha-beta alloys are stronger than y because b.c.c. beta is stronger than vV:;Ahase alloys can be strengthened by '
1620"
'
I Oil!)
"
,
.
.
28.niO
v jicture can be manipulated by controll/cles.
S contain some beta-stabilizing alloyf these alloys are balanced by high re essentially one-phase alpha. Figure i a Ti-5AI-2 5Sn alloy after hot working e two main attributes: weldabiiity and '
(
" .
1'
.
atures.
I ,'00
The first results from the one-
'
I rlO
rn the presence of aluminum. Alloying
.gsjSalpha-phase alloys, and aluminum is ' "
.
.
.
Vjlpha alloys. Especially important
,
1000
its l
Hot working of alpha alloys containninum is difficult. Hot workability of additions of beta-stabilizing alloying so that the beta phase is present in icrostructure. Some applications of
a-ti
bilpipe assemblies and other formed V-30"F, and missile fuel tanks and struc-
;v
- <
31 3 25 .
:
.
600 0
10
?0
M
ll
up to 1100"F
.
enough beta-stabilizing elements to
10
so
U\
Alomic jjCiCfiil vunodiuin (From Max Hansen,
Fig. 12-44 The Ti-V phase diagram Constitution of Binary Alloys," 2d ed,. McGraw-Hill Book .
"
Company, New York, 1958.)
I
/()
«n
oo
loo i/
528
f
INTRODUCTION TO PHYSICAL METALLURGY
Weight perceni mongonese 20
10
30
10
50
1700
60
ll
70
,
1800
1300
1200
1660°
1600
- 1100 \
1500
I 1400
I
1330°
\
1300
=
r
\
1200
f
1175°
i
30
/3
39.2
v/7.
:33) (42.5)
1100
Fig. 12-46
TI-5AI-2.5Sn hot-worked below the a transl
(nation temperature, annealed 30 rnin at 2150°F (1177-'( which is above the /3 transformation temperature and
1000
9501
,
lumaoe-oooled. 900
7
.
:
t
382°
Structure is coarse platelike a. ,
Kroll's reagent 10Cx. (From Metals Handbook ,
,
Atlas of Microstructures
"
r,
300
_
J
i
-
"
,
Etchi vol. 7
,
American Society for Metal:
1972.)
v - ;j
700
some cases
,
quenched titanium alio;
of alpha designated alpha prime and
600 -
0 44
-
.
550°
nation originally was borrowed fron
i8
500 (~0 5r(-20) .
Or,
400
o Ti
10
20
30
40
50
60
70
mm
Atomic percent manganese
Fig. 12-45 The Ti-Mn phase diagram. (From Max Hansen, Constitution of Binary Alloys," 2d ed., McGraw-Hill Book Company, New York, 1958.)
n
'
A
"
ft
1
i
SI
5
-
tially, this is accomplished by quenching from a temperature in the alpha-
0
beta field followed by aging at moderately elevated temperature. In contrast , \ to the usual age-hardening procedure, a homogeneous beta solid solution \ \ is not formed in the first step. If an all-beta structure were formed, the beta
a
m
grain size would be excessively large, and the subsequent formation of !
v. .
-l
'
J'
P
i
.
.J)
m 5)
alpha would be mainly at the beta grain boundaries. These two factors re- 3 \ duce the ductility of the aged alloy. Quenching suppresses the transforma- ! tion of the elevated-temperature beta phase that would occur on slow cooling. Aging at elevated temperature causes precipitation of fine particles of alpha in the volumes that were beta grains prior to quenching. This fine structure is stronger than the coarse, annealed alpha-beta structure, In
w
f i
Fig, 12-47 Ti-8Mn alloy annealed 2 h at 1300-'F (704oC) furnace-cooled to 1100°F (5930C) and held 1 h. « grain ,
,
(gray) in fi matrix (light); also a at the prior beta grain boundaries. Etched in Kroll's reagent 500'-:. (From Me ,
Handbook, vol. 7, "Atlas of Microstructures
Society for Metals, 1972.)
.
"
,
American
NONFERROUS METALS AND ALLOYS
70
529
7i
j
1300
1200 .
V
1100 i
j
/ri
1330°
I
' ,
I
Iff.
/
fit
l*-
I
Fig. 12-46 Ti-5AI-2.5Sn hot-worked below the a transformation temperature, annealed 30 min at 2150oF (1 1770C) ! vvhich is above the fi transformation temperature, and ' lurnace-cooled. Structure is coarse, platelike a. Etched in Kroll's reagent, lOOx. (From Metals Handbook, vol. 7, ,
Atlascof Microstructures," American Society for Metals, 1972.)
"
some cases, quenched titanium alloy structures may be of an unstable form of alpha designated alpha prime and called titanium martensite. This desig nation originally was borrowed from steel metallurgy, where martensite is
-
i.O
70
Innson
V
VP
.
Book
.
m
r
'
inching Irom a temperature m the alphaierately elevated temperature In contrast dure a homogeneous beta solid solution .
0
,
n all-beta structure were formed the beta ,
large, and the subsequent formation o' grain boundaries. These two factors rey. Quenching suppresses the transforma'
"
.
iy beta phase that would occur on slow viature causes precipitation of fine particln;;
:':
jbeta grains prior to quenching. This fine arse, annealed alpha-beta structure
.
In
Fig. 12 '!/ li fJMn alloy, annealed ? h al ISOO'F (70'1 C). iurnace-cooled to I I00 F (SGa'C). and held I h. n grains (grny) in /) ninlnx (liqhl); also ir al the pnoi beta qinin boundanob, Llchud in Kiull's ruagcnl, ')00 Society (oi Molals 197P ) .
m
{I loin Mobile
Handbook, vol. 7 "Atlas of Microstructures," American
530
INTRODUCTION TO PHYSICAL METALLURGY
a metastable structure formed by a diffusionless phase transformation when steel is quenched from a high temperature. In modern terminology, however, martensite is a word for any needle-like metallic structure formed by diffusionless shear, usually upon rapid cooling. Figure 12-48 shows the
/
>
titanium martensite structure in aTi-6AI-4V alloy which was water-quenched from 1950°F. Strength of alpha-beta alloys can be increased about 35 percent by heat treating, compared with the properties of annealed material. Typical applications of the Ti-6AI-4V alley include aircraft gas turbine compressor blades and disks; forged airframe fittings; and sheet metal airframe parts. The Ti-8Mn alloy has been used for aircraft skins and primary structural parts subject to temperatures in the range of 200 to 600 F
)
1
\
0
1
''
.
Beta Alloys
'
Unlike the alpha alloys, beta alloys can be strengthened by
[
j
I
heat treatment. Figure 12-49a shows the all-beta microstructure in a Ti\ (a) i . 3AI-13V-11Cr alloy after solution treatment for 10 min at 1450oF. This alloy hg. 12-49 T;-13V-11Cr-3AI sheet (a) Solution-treated is weldable in both the annealed and heat-treated conditions. Aging at ;or 10 min al 1450oF (7880C) and air-cooled Structure elevated temperature arter solution treatment results in the precipitation of consists of equiaxed grains of metastable beta {b) Same as (a), except aged for 48 h at 9003F (482 0) Structure fine particles of alpha and TiCr2 compound. The microstructure after aging \ consists of dark particles of precipitated n in /j grains for 48 h at 900oF shows dark particles of precipitated alpha in beta grains Etched in 2HF, 10HNO 88H,0, 250x. (From Metals Hand ,
.
,
.
.
:
.
.
(see Fig. 12-49b). Ultimate strengths up to 215,000 psi with 5 percent elongation,are possible after heat treatment. This is an increase over the annealed strength of at least 50 percent. Beta alloys have been used for highstrength fasteners and for aerospace components requiring high strength
oook, vol. 7,
,,
"
Atlas of Microslructures
"
.
American Society
'
or Metals, 1972.)
Typical mechanical properties of sot
at moderate temperatures.
12-17.
ZINC AND ZINC ALLOYS .v
12-40 am
The principal use of zinc is as a coati is more highly anodic than steel and coating acts as the sacrificial anode ,
.
i
.
protecting the steel from any attack. IV by various methods such as hot-dip gs ing metallizing or spraying of molten r tation. Steel products that are galvan hardware pipe and tubing, screws, sh,
m
vr
mm
,
in sherardizing parts to be coated ar airtight container which is then revc ,
,
Fig. 12-48 Ti-6AI-4V alloy, held for 1 h at 1950oF (1066°C) which is above the j3 transformation temperature, and waterquenched. Structure is a' (titanium martensite); prior j3 grain boundaries are also visible. Etched in 10HF, 5HNO:,, 250x. (From Metals Handbook, vol. 7, "Atlas of Microstructures," American Society for Metais, 1972.) ,
i:
slightly below the melting point of zin
i
impregnates the surface and diffuses in coating. Zinc oxide is used in the manufactu
tilting glass, glazes, matches, paint, po Zinc can be easily worked into variou ,
NONFERROUS METALS AND ALLOYS
531
)y a diffusionless phase transformation
gh temperature, in modern terminology, W-Sjany needle-like metallic structure formed I l-
.
.
:
.
.;
)n rapid cooling. Figure 12-48 showsthe i-6AI-4V alloy which was water-quenched '
|
eta alloys can be increased about 35 perwith the properties of annealed material
.
kl-4V alloy include aircraft gas turbine jed airframe fittings; and sheet metal air- f '
\
*
| been used for aircraft skins and primary j
Pi
-1
v . tures in the range of 200 to 600CF.
t
apvys, beta alloys can be strengthened by
j>
'
iows the all-beta microstructure in a Tr-
[a)
eatment for 10 min at 1450oF This alloy and heat-treated conditions Aging at
-
.
'
Soliilion-tronlotl I fig, IZ IO Ti-IW I ICr-nAI shoot. I W 10 min al 1450 K {788' C) and air-cooled. Slmcturo
.
n treatment results in the precipitation of impound. The microstructure after aging
p iiicles of precipitated alpha in beta grains
f;Jonsists of equiaxed grains of metastable beta (b) Same 3S(a), except aged for 48 h at 900'? (482'C). Structure
-
[Bched in 2HF, 10HNO:„ SSH.O, 250x. (From Melals Hand-
ths up to 215 000 psi with 5 percent elon,
Catment.
'
This is an, increase over the an-
5<'0k, vol. 7, "Atlas of Microstructures," American Society br Metals 1972.)
.
ient. Beta alloys have been used for high-
,
;
ce components requiring high strength
r
-
Typical mechanical properties of some titanium alloys are given in Table 12-17,
ZINC AND ZiNC ALLOYS
\
12-40 The principal use of zinc is as a coating for steel to prevent corrosion. It is more highly anodic than steel, and in a corrosive atmosphere the zinc coating acts as the sacrificial anode. Thus the zinc is consumed while
prolocling the steel from nny allack. Metallic zinc coalings may be applied by various methods such as hot-dip galvanizing; electrogalvanizing, painting metallizing or spraying of molten metal, and by sherardizing or cementation. Steel products that are galvanized include bolts,-chains, fencing, ,
:
pipe and tubing, screws, sheets, tanks, wire, and wire cloth. In sherardizing, parts to be coated are tightly packed with zinc dust in an airtight container, which is then revolved and heated to a temperature slightly below the melting point of zinc. In the presence of heat, the zinc impregnates the surface and diffuses into the steel, providing a thin uniform hardware
,
V
a 066-C).
.
.
.
KSsfld water'
or p 5HNO:„ ro-
i
m
coating. Zinc oxide is used in the manufacture of dental cement, enamels, floor
tilting, glass, glazes, matches, paint, pottery, rubber goods, tires, and tubes. Zinc can be easily worked into various shapes and forms by common fab-
532
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 12-17
Typical Mechanical Properties of Some Titanium Alloys*
die castings. The zinc die-casting al
ROOM TEMPERATURE CONDI-
ALLOY
TION
TENSILE
YIELD
ELON-
SJRENGTH,
STRENGTH,
GATION, |-
«
PSI
PSI
%
m
and have greater strength than all d alloys. They can be cast to close din minimum cost; their resistance to sur range of applications They are usl .
COMMERCIALLY PURE TITANIUM Annealed
Comm. purity (99.0%)
79,000
below 200°F
63,000
since above this tempen 30 percent and their hardness 40 perc
27
The Al-Zn phase diagram is showr
ALPHA TITANIUM ALLOYS Ti-5AI-2.5Sn
Annealed
125,000
120,000
18
¥
Ti-6AI-4Zr-1V Ti-8AI-1Mo-1V
Annealed
143,000
138,000
17
:
HTt
147,000
135,000
16
-
Annealed
138,000
125,000
15
Ti-4AI-4Mn
Annealed
148,000
133,000
16
HTt
162,000
140,000
9
TI-6AI-4V
Annealed
135,000
120,000
11
HTt
170,000
150,000
7
Ti-7AI-4Mo
Annealed
160,000
150,000
15
HTt
190,000
175,000
12
.
,
I
.
:
temperature a precipitation reaction (
BETA TITANIUM ALLOY 180,000
above 5270F (2750C) At that tempe reaction into a and phases Comrr fast enough to prevent the eutectoid tr mixture of a' and Figure 12-51a AG40A (Zamak-3) die-cast. The struc mary p (zinc-rich solid solution) surro Slower cooling in a permanent mold v much coarser (Fig. 12-51 fc>) When die castings are aged at roc .
Ti-8Mn
HTt
forms at 7200F (3820C) and 5 percent solutions. The a' constituent of the ei
ALPHA-BETA TITANIUM ALLOYS
Ti-3AI-13V-11Cr
,
,
170,000
tion.
6
The /3 may contain about 0 35 .
freshly made die casting. During a 5-wi this will decrease to about 0 05 perce,
Compiled from data in Metais Handbook, voi. 1, American Society for Metals, 1961.
'
.
tHT is usually after age-hardening.
particles of a within the /3 structure (Fi The two die-casting alloys in genera AG40A, SAE 903) and Zamak-5 (ASTM ,
ricating methods. Pure zinc has a recrystallization temperature below room temperature, so that it "self-anneals and cannot be work-hardened at "
room temperature. The presence of natural impurities or added elements , raises the recrystallization temperature. Therefore, the less pure grades of wrought zinc will show an increase in hardness and strength with working. For deep-drawing purposes a relatively pure zinc should be used. Typical ?
applications include drawn and extruded battery cans, eyelets, grommets, laundry tags, and address plates. The addition of lead and cadmium re-: suits in higher hardness, stiffness, and uniform etching quality. It is used; for weatherstrips, soldered battery cans, and photoengraver's plate. Fo[! added stiffness, good creep resistance, and easy work hardening, alio]
containing from 0.85 to 1.25 percent copper are recommended. A wrought
Atomic percentage zinc V
10
700
20
500
3 + L
j
300r
275°
copper alloy has been used in the form of heavy rolled plate in theaircra industry and for dies in the blanking of aluminum-alloy sheet and thin st
The major use of zinc as a structural material is in the form of alloys
fa1
3
JI.6
200
04-/3
gated roofing, leaders, and gutters.
The 4 percent aluminum, 0.04 percent magnesium, and up to 3.5 percen|
40
600-
zinc alloy containing from 0.50 to 1.50 percent copper and 0.12 to 1.i
percent titanium has outstanding creep resistance and is used for corru-J
30
560°
100
Al
10
20
30
40
50
60
Weight percentage zinc
12.50 The Al-Zn phase diagram (From Metals HandoK, 1948 ed., p. 1167, American Society for Metals ) .
.
m
:
i
.
..
,v
-
-
Y
NONFERROUS METALS AND ALLOYS
irties of Some Titanium Alloys*
% r
ROOM TEMPERATURE
11
YIELD
STRENGTH,
STRENGTH,
GATIOfi $
PSI
PSI
%
I I
\LLY PURE TITANIUM 79,000
3 I-
63,000
'
120 000 ,
18
143,000
138 000 ,
17
147,000
135 000
16
,
11 F
I I
TITANIUM ALLOYS 138 000
125 000
15
148,000
133,000
16
162 000 ,
140 000
9
135 000
120,000
11
,
,
,
,
170 000 ,
150,000
7
160 000
150,000
15
190 000
175 000
12
170 000
6
,
,
.
ft
.
-
125,000
,
I
,
,
i
.
,
The Al-Zn phase diagram is shown in Fig. 12-50. A lamellar eutectic forms at 720"F (382<'C) and 5 percent aluminum, containing a' and p solid solutions.
'
The a
constituent of the eutectic is stable only at temperatures
above 5270F (2750C).
and cannot be work-hardened at S
|e of natural Impurities or added elements g jrature. Therefore
,
Atomic pctcenloqe line
the less pure grades of
se in hardness and strength with working. J
700
ielatively pure zinc should be used. Typical sxtruded battery cans eyelets, grommels.
600
"
/
.
At that temperature, it transforms by a eutectoid
reaction into a and p phases. Commercial die-casting alloys are cooled fast enough to prevent the eutectoid transformation and retain the eutectic mixture of a' and /3. Figure 12-51a shows the microstructure of alloy AG40A (Zamak-3), die-cast. The structure consists of white grains of primary j8 (zinc-rich solid solution) surrounded by the dark eutectic mixture. Slower cooling in a permanent mold will cause the lamellar eutectic to be
this will decrease to about 0.05 percent, the excess appearing as minute particles of « within the (3 structure (Fig. 12-51 c). The two die-casting alloys in general use are known as Zamak-3 (ASTM AG40A, SAE 903) and Zamak-5 (ASTM AC41 A, SAE 925). They both contain
i recrystallization temperature below room * "
range of applications. They are usually limited to service temperatures since above this temperature their tensile strength is reduced
below 200oF
When die castings are aged at room temperature or slightly elevated temperature, a precipitation reaction occurs in the zinc-rich 3 solid solution. The (3 may contain about 0.35 percent aluminum in solution in a freshly made die casting. During a 5-week room-temperature aging period,
j
,
American Sociel/loi Moials 1961.
neals
minimum cost; their resistance to surface corrosion is adequate for a wide
much coarser (Fig. 12-51 fo).
ITANIUM ALLOY 180 000
die castings. The zinc die-casting alloys are low in cost and easy to cast and have greater strength than all die-casting metals except the copper alloys. They can be cast to close dimensional limits and are machined at
30 percent and their hardness 40 percent.
27
riTANIUM ALLOYS
:A
1.
*
ELON- ; I
TENSILE
20
10
30
°
F
30
40
-
.
80 90
z
11200
,
s
ilOOO
The addition of lead and cadmium re-
.
;s and uniform etching quality. It is used t 1 sry cans and photoengraver's plate. For
o00
J
t-
V
,
,
istance
and easy work hardening alloys ent copper are recommended. A wrought,
;
-
.
. .
>
-
,
.
9b
- o
27
300'
a
to 1.50
percent magnesium, and up to 3.5 perceo! |
he form of heavy rolled plate in the aircraft | ing of aluminum-alloy sheet and thin steel
_
100. 4
10
20
30
40
50
60
70
Weight percentoqe jinc
.
fig. 12-50 The Al-2n phase diagram. (From Metals Hand-
i
600
99.2
2001
|i|!rs.
uctural material is In the form of alloys for
,
78
31.6
g creep resistance and is used for corru-
0
1
1
i
..
;
800
382
400
,
|«|; to 1.50 percent copper and 0 12
533
book, 1948 ed. p. 1167, American Society for Metals.) ,
j
i
30
90
400
Zn
534
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 12 18 Typical Mechanical Properties of Some jr.*
MATERIAL COMPOSITIONS
fie
,
APPROX. %
TE
TREATMENT
ST PS
' .
s
ku if -
-
WROUGHT A
.
1
Commercial rolled zinc
Hot-rolledt
19
(deep drawing),
Hot-rolledj Cold-rolledt Cold-rolledt
23,
[ 0.08 max Pb, -
7
bal. Zn
Commercial rolled zinc
<7)
,
;
l 0.05-0.10 Pb, I 0.05-O.O8 Cd, i bal. Zn .
I Commercial rolled zinc,
ft
0.25-0.50 Pb,
.
0 25-0.45 Cd, .
'
bal. Zn
Cold-rolledt Cold-rolledt
29
Hot-rolledt
23,1
Hot-rolledt Cold-rolledt
25,(
22, i ,
29,1
31, ( 24,(
j [tel.Zn
Hot-rolledt Cold-rolledt
32,(
Cold-rolledt
40,C
.
Hot-rolledt
28 C
Hot-rolledt
36, C
Cold-rolledt Cold-rolledt
48 0
Copper-hardened rolled i zinc alloy,
i 0.85-1.25 Gu, Rolled zinc alloy
,
> [ 0.85-1.25 Cu, y f0.006-O.O16 Mg % *bal Zn
,
if:
32, (
,
37, C ,
CASTING ALL
Zamak-3, SAE 903,
Die-cast
41 0
Die-cast
47,6
,
A8TM AG40A
cooling; 250x. (c) Zamak-3 alloy, die-cast then aged for 10 days at 203oF (95°C). Aging has caused additional pre-
xni),
|3.5-4.3 Al, | 03-O.O8 Mg,
cipitate of a (black dots) In the light grains of p (zincrich) solid solution; 1,000x. All samples etched in 50g CrO:„ 4g NajSO,, 1 liter H20. (From Metals Handbook, vol. 7, Atlas of Microstructures," American Society for Metals,
21,
25,
Cold-rolledt
:
(a), but the lamellar eutectic is coarser due to the slower
Hot-rolledt
Hot-rolledt
'
*
'
surrounded by the dark eutectic mixture; I.OOOx, (b) Zamak-5 alloy, permanent-mold-cast. Same structure as
21, 27,
Hot-rolledt
"
Fig. 12-51 (a) Zamak-3 alloy, as die-cast. Structure consists of white grains of primary /3 (zinc-rich solid solution)
,
.
1:0.15 max Cu,
"
1
1972.)
about 4 percent aluminum and 0.04 percent magnesium. Zamak-3 h.
Zn (99.99%)
.
iiniak-5,SAE 925, ITM AC41A (XXV), 5-4 3 Al, 1 03-0.08 Mg, 75-1.25 Cu, .
,
slightly higher ductility and retains its impact strength better at sligh
fir
elevated temperature. Zamak-5, containing about 1 percent copper, somewhat harder and stronger and has slightly better castability. They
i Zn (99.99%)
used for automotive parts, household utensNs, building hardware, padlocks, toys, and novelties. The maximum composition limit of certain
iTM (XX1),
purities such as lead (0.007 percent), cadmium (0.005 percent), and tin (Gj percent) must be strictly observed to minimize intergranular corrosion.? The composition and typical mechanical properties of some zinc alio; are given in Table 1218.
I I :::::
i
} £
i
I
I
ak-2, SAE 921,
Sand-cast
20,0( 30,0(
4 5 Al,
-
.
?-0 10 Mg, .
3 5 Cu,
-
.
Zn (99.99%) ak-5
Sand-cast
as above)
20,0C 30,0(
ican Zinc Institute, New York.
ijtudinal direction. rerse direction
.
NONFERROUS METALS AND ALLOYS
RGY
A
:
,,
WBLE 12 18
"
V
i
Typical Mechanical Properties of Some Zinc Alloys'
||aterial compositions, .APPROX. %
TENSILE TREATMENT
% r*"
.
P: .
«5
.
ELONGATION,
BHN,
% IN 2 IN..
10-MM BALL
WROUGHT ALLOYS
i
A Si
STRENGTH, PSI
.
t5
I Commercial rolled zinc
Hot-rolledt
19,500
65
38
% |{deep drawing),
Hot-rolled i
23,000
50
38
Cold-rolledi
21,000
40
f '/JO.OS max Pb, tel. Zn
Cold-rolledl.
27,000
40
Commercial rolled zinc,
Hot-rolledi
21,000
5?
43
a05-0.1O Pb,
Hot-rolled:!:
25,000
30
43
0 05-0.08 Cd
Cold-rolledl
22,000
40
bal. Zn
Cold-rolled:]
29,000
30
Commercial rolled zinc,
I lol-rolled !
23 000
50
47
0 ?6 t).!>0 I'!.,
I lol mllod I
;'<) ()()()
.
1:'
47
Cold-roiled I
25,000
4b
bal. Zn
Cold-rolled I"
31,000
28
Copper-hardened rolled Jinc alloy, 0 85-1.25 Cu,
Hot-rolledi
24,000
20
1)2
l-lol rolledi:
32,000
15
60
Cold-rolled I'
32,000
5
Cold-rolledi
40,000
3
Hot-rolled .!.
28,000
20
61 80
!
.
,
i
1
,
J;«.25-0.45 Cd,
,
.
M Zn .
-
,
.
Rolled zinc alloy 085-1.25 Cu, 0 006-0.016 Mg
Hot-rolledi
36,000
10
Cold-rolledl
37.000
20
bal. Zn
Cold-rolled j
48,000
2
,
"
.
,
dure con-
i
.
J solution) < (6)
8
-
:
ucture as
'
Die-cast
41,000
10
82
Die-cast
47,600
7
91
ASTM AG40A
aged for itional prei (zincin 50g CiO,
IXX111), 3 5-4 3 Al, .
.
0 03-
0.08 Mg, 0-0 15 max Cu, bal. Zn (99 99%) .
,
vol. 7
CASTING ALLOYS
2amak-3, SAE 903,
ne slower
,
.
it Metals,
Zamak-5,SAE 925, ASTM AC41A (XXV),
nd 0.04 percent magnesium.
Zamak-3 has I =0.03j 0 08fj;Mg, ,
.
etams its impact strength better at slightly \ 25 cu k-5 containing about 1 percent copper, is ;;bai. Zn (99.99%) lillfr and has slightly better castability. They ares j2amak-2 SAE 921, -
,
,
'
household utensils :::
,
Sand-cast
building hardware, pad- [ASTM(XX1),
ie maximum composition limit of certain im-
-;
20,000-
70-100
30,000
3-5-4.5 Al,
|;;fi cent), cadmium (0.005 percent), and tin (0.005 jl gf 1 9' rved to minimize intergranular corrosion, bal Zn (99 99%) '
535
al mechanical properties of some zinc a"oys 2amak 5
(same as above)
American Zinc Institute, New York. jjlongiludirial direction.
Transverse direction.
Sand-cast
20,000-
30,000
70-100
536
INTRODUCTION TO PHYSICAL METALLURGY
I
some use as a brazing solder. Coin s
PRECIOUS METALS
coins and for electrical contacts
The precious-metals group includes silver, gold, andthesix platinum metals platinum, palladium, iridium, rhodium, ruthenium, and osmium. This group is characterized by softness, good electrical conductivity, and very high
contain cadmium and tin. In brazing is similar to that of soft soldering ex temperature. There is no melting of the is achieved by interfacial penetration property of these alloys is the tempe freely into a joint. By suitable variati obtain brazing alloys that melt anywhe ing alloys are used for many applicatio
corrosion resistance to common acids and chemicals.
,
12-41 Silver and Silver Alloys
The photosensitivity of silver and certain silver salts, coupled with their ease of reduction, forms the basis for photography.
,
Silver-clad copper, brass, nickel, and iron are used for electrical con- I ductors, contacts, and chemical equipment. A recent development is the
use of silver coatings on glass, ceramics, and mica to provide a conducting ;::: vi _
.
.
Silver-Copper-Zinc Silver alloys in th or silver brazing alloys. In addition tc
base for subsequent electroplating of electronic devices. The high reflectivity and ease of electroplating make silver useful in reflectors silverware, and jewelry. ,
Silver-Copper Alloys
ferrous materials.
12-42 Gold and Gold Alloys Aside from the u and dental products they have many in corrosion resistance nontarnishing c ductivity, and ease of electroplating rr trical applications. Electroplated gold i on contacts on vibrating components
The silver-copper alloy system is shown in Fig
12-52. It is a simple eutectic-type system, with the eutectic point located at 28.1 percent copper and 14350F. The maximum solubility of copper in silver is 8.8 percent, and the slope of the solvus line indicates the possibility of age-hardening certain alloy compositions. Sterling silver (7.5 per cent copper) and coin silver (10 percent copper) are age-hardenable alloys,
,
,
,
but little commercial use is made of this heat treatment because of the close
thermal-limit fuses to protect electrice
temperature control required. The 28 percent copper eutectic alloy finds
1100
paratus, as a freezing-point standard, f< and as a high-melting solder for vacuu A 70 percent gold-30 percent platini
32000
1000
1800
900
1600 i
-
a
0
800 "
/8.8i
92
28.1
1400
,
tive light filters. Other industrial appli
of 22420F, is used as a high-melting-pc Gold-palladium-iron alloys develop v
treatment and are used primarily for pc the highest resistivity contains 49.5 pe and 10 percent iron. k12 43 Platinum and Platinum Alloys Platini abundant metal in the platinum group. num are high corrosion resistance hi ductility. It forms extensive ductile so the unalloyed form platinum is used ,
700 1200
,
600 1000
I;
jr
500
.
800 40
Ag
Fig. 12-52
10;
[
20
30
40
50
60
70
80
Weight percent copper
The silver-copper alloy system. (From "Metals
Handbook," 1948 ed., p.Tl148, American Society for Metals, Metals Park, Ohio.)
i E: '
mam
-
"'«
.
t
i
1
'
;
/
90
Cu
thermometer elements electrical conte ,
dental foil, electrodes, heat- and corn
jewelry. It is also used as a catalyst in t mins, and high-octane gasolines. Most of the important binary platinurr solubility, so that the increase in hardn ing is due to solid-solution hardening additions upon platinum is shown in
hi
S:V :::;
3Y
NONFERROUS METALS AND ALLOYS
537
some use as a brazing solder. Coin silver is used for United States silver coins and for electrical contacts
v v es silver, gold and the six platinum me
.
,
' " '
Silver-Copper-Zinc
Silver alloys in this group are known as silver solders or s/7ver brazing alloys. In addition to silver copper, and zinc, they often contain cadmium and tin. In brazing the physical mechanism of bonding is similar to that of soft soldering except that it takes place at a higher temperature. There is no melting of the material being joined and the bond is achieved by interfacial penetration of the brazing alloy. The important property of these alloys is the temperature at which they melt and flow freely into a joint. By suitable variation of composition it is possible to obtain brazing alloys that melt anywhere from 1100 to ISSO'F. Silver brazing alloys are used for many applications in the joining of ferrous and non-
plium, ruthenium, and osmium, Thisgraii pd electrical conductivity, and very hlgS
,
jacids and chemicals. :i iotosensitivity of silver and certain silver; j "
,
,
jduction, forms the basis for photography.
,
and iron are used for electrical coth equipment. A recent development is the ,
\
'
and mica to provide a conducting ;iV g of electronic devices The high reflex :
ramiCS,
,
.
|
ake silver useful in reflectors silverware; ,
i
(
h-copper F
.
2-42
alloy system is shown in Rg.
e system :,
ferrous materials.
'
,
Gold aiid Gold Alloys
Aside from the use of gold alloys forcoinage, jewelry
,
and dental products they have many industrial applications. The very high
with the eutectic point located
,
corrosion resistance
The maximum solubility of copper ift
,
nonlarnishing characteristics, good electrical con-
.3
i;
ductivity and ease of electroplating make gold coating suitable for electrical applications. Electroplated gold is used in wave guides on grid wires, on contacts on vibrating components, and as a thin film on glass for selec-
of the solvus line indicates the possi-
,
by compositions. Sterling silver {7 5 petv;y-::kcent copper) are age-hardenable alloys
,
.
,
,
|f this heat treatment because of the close
tive light filters. Other industrial applications of high-purity gold include
\e 28 percent copper eutectic alloy finds
thermal-limit fuses to protect electrical furnaces, as a target in x-ray apparatus, as a freezing-point standard, for the lining of chemical equipment, and as a high-melting solder for vacuum-tight pressure weids.
A 70 percent gold-30 percent platinum alloy, with a solidus temperature of 22420F, is used as a high-melting-point platinum solder.
32000
Gold-pallndium-iron alloys develop very high resistivity after proper heat treatment and are used primarily lor potentiomolei wire. The alloy having the highest resistivity contains 49.5 percent gold, 40.5 percent palladium, and 10 percent iron.
moo
L
600 +
-
92
12-43
03
j81400
Platinum and Platinum Alloys Platinum is the most important and most abundant metal in the platinum group. The important properties of platinum are high corrosion resistance high melting point. White color, and ,
(incliiily
?(J0
II lorms oxlon.'iivc dnr.lilo solid fiolulionr, with olhor motals.
In the unalloyed form platinum is used (or thermocouple and resistancethoimomolor olomonts oloch ical conlacls crucihlos and laboratory ware, ,
i-
,
10()0
,
dental loil, electrodes, heal and coirosion-icsisianl oqutpmenl and lot jewelry. It is also used as a catalyst in the production of sulfuric acid, vitamins, and high-octane gasolines. Most of the important binary platinum alloy systems show complete solid solubility so that the increase in hardness and strength obtained by alloying is due to solid-solution hardening The hardening effect of alloying additions upon platinum is shown in Fig 12-53 Of the metals shown ,
800
70
80
90
Cu
'
UMii copper
: ' '
Metals
,
Metals
.
.
.
1
'
' .
.
V:
,
538
INTRODUCTION TQ PHYSICAL METALLURGY
260
I
240
i 220
Ay 200
7
i
catalyst for the oxidation of ammoni alloy shows excellent resistance to glassworking equipment. The 10 alloy with its composition carefully of the widely used rhodium-platinu percent rhodium alloy is used for c ,
num. Platinum-rhodium alloys cc rhodium are used as windings in 3275°F.
Platinum-lridium Alloys
180 0s :
O)
S 160
Platinum
is employed for crucibles and othei mechanical properties and excelle ,
the 5 to 15 percent iridium alloys ma Electrical contacts for dependable s
Cu
5 140 a
stats generally contain between 10 percent iridium alloys are used for for aircraft spark plugs.
Au
I 120 CD
Platinum-Ruthenium Alloys
ir
O
The al
applications similar to those of the harder to work than iridium the prac percent ruthenium. The 5 percent in medium-duty electrical contacts. tacts in aircraft magnetos and the 1 Platinum-ruthenium alloys are also trades, hypodermic needles, and pe Platinum-Nickel Alloys This group nickel, has good strength at elevat
100
,
>
80 Rh
3d
60
I
40
alloy is used for long-life oxide-co Platinum-Tungsten Alloys The mo 4 and 8 percent tungsten. Typical 8
20
0 0
21
4
6
8
10
14
12
16
18
2C
Alloy addlllon, weight percent
Fig. 12.53 Hardening effect of alloying additions on platinum. (The International Nickel Company.)
nickel produces the greatest hardening effect and palladium the least, jflj general, the effect of alloying additions on the tensile strength parallelstj effect on hardness.
Platinum-Rhodium Alloys These contain between 3.5 and 40 percej rhodium. Rhodium is the preferred alloying element to platinum for applicationssat high temperatures under oxidizing conditions. The 10 cent rhodium alloy is the most popular one in this group. It is the stand
:
I
Ik
tp2 '
.
-
-
fy
:
-
electrodes, electrical contacts, grid: eter wire, strain gauges, and hard cc
A platinum 23 percent cobalt alio; is used as a permanent magnet in magnet is essential. 12.44 Palladium and Palladium Alloys P respects and is second to it in impor ladium as compared with platinum i of palladium is in telephone relay c( remove oxygen from heat-treating a tion of hydrogen gas. Palladium lea binding, glass signs, and trim. Like platinum, palladium forms ci
NONFERRCUS METALS AND ALLOYS
catalyst for the oxidation of ammonia in the manufacture of nitric acid
.
539
This
alloy shows excellent resistance to molten glass and is used for nozzles in glassworking equipment. The 10 percent rhodium-90 percent platinum alloy, with its composition carefully controlled serves as the positive side of the widely used rhodium-platinum vs. platinum thermocouple The 3.5 percent rhodium alloy is used for crucibles as an alternative to pure platinum. Platinum-rhodium alloys containing bolween 10 and 40 percent ,
.
rhodium are used as windings in furnaces operating between 2800 and 3275 F.
Platinum-lridium Alloys Platinum alloyed with 0 4 to 0.6 percent iridium is employed for crucibles and other laboratory ware. The rich color high .
,
mechanical properties and excellent corrosion and tarnish resistance of ,
4
the 5 to 15 percent iridium alloys make them the preferred metal for jewelry. Electrical contacts lor dependable service in magnetos relays, and thermostats generally contain between 10 and 25 percent iridium. The 25 or 30 percent iridium alloys are used for hypodermic needles and as electrodes for aircraft spark plugs. ,
an
Platinum-Ruthenium Alloys The alloys in this group have properties and applications similar to those of the platinum-iridium group. Ruthenium is harder to work than iridium the practical limit of workability being about 15 percent ruthenium. The 5 percent ruthenium alloy is used in jewelry and
i
,
in medium-duty electrical contacts. The 10 percent alloy is used (or contacts in aircraft magnetos and the 14 percent alloy in heavy-duty contacts. Platinum-ruthenium alloys are also employed for aircraft spark-plug electrodes hypodermic needles, and pen nibs.
Pd
,
Platinum-Nickel Alloys This group of alloys, containing up to 20 percent nickel, has good strength at elevated temperature. The 5 percent nickel alloy is used for long-life oxide-coated cathode wires in electron tubes.
10
12
14
Platinum-Tungsten Alloys The most popular alloys in this group contain 4 and 8 percent tungsten. Typical applications include aircraft spark-plug electrodes electrical contacts, grids in power tubes for radar, potentiometer wire, strain gauges, and hard corrosion-resisting instrument bearings. ,
16
18
20
on, weignl p(;rcf;ni
ening effect and palladium the least
.
'
In
tions on the tensile strength parallels the
Air.-
contain between 3 5 and 40 percent .
'
)6 alloying element to platinum for most I
under oxidizing conditions The 10 per- I ular one in this group It is the standard-W m .
.
m
A platinum 23 percent cobalt alloy has unusual magnetic properties and is used as a permanent magnet in small instruments where a very short magnel is essential. 12-44 Palladium and Palladium Alloys Palladium resembles platinum in many respects and is second to it in importance. The principal advantage of palladium as compared with platinum is its lower cost. The major application of palladium is in telephone relay contacts. It is also used as a catalyst to
remove oxygen from heat-treating atmospheres and as a filter for purification of hydrogen gas. Palladium leaf is used for decorative effects in bookbinding, glass signs, and trim. Like platinum, palladium forms complete solid solutions with almost all
:
540
INTRODUCTION TO PHYSICAL METALLURGY
12-45 Iridium
Iridium is the
most corro
iridium has been used for crucibles
m
180
temperatures, and as extrusion dies
160
amounts of iridium, up to about 0.1 |
The main use of iridium is as a h
I
I
Ru
140
r
120
size and Improving the mechanica casting alloys. 12 46 Osmium Osmium has a high meltin very high temperatures. Osmium ai wear resistance
and good corrosioi elude fountain-pen nibs phonograp strument pivots. 12'47 Rhodium Rhodium is similar to pic higher reflectivity. It has exception: equal to that of iridium. Rhodium with high reflectivity. It is used as a for reflectors for motion-picture pr thin rhodium plate is sometimes use ,
Cu
T3
,
100
5
Ag
80 Au
>
60 PI
:
contacts.
40
The main use of rhodium
and palladium. 12-48 Ruthenium
20
3
2
4
6
8
10
12
14
16
8
Alloy oddiiion iveighi percent ,
Fig. 12-54 Hardening effect of alloying additions on palladium. (The International Nickel Company.)
in Table 12-19.
alloying elements. The hardening effect of alloying additions on palladium is shown In Fig. 12-54. Of the metals shown, ruthenium and nickel are very effective hardeners, while platinum Is least effective. Palladium-Silver Alloys Alloys containing 1,3, 10, 40, 50, and 60 percent palladium are widely used for electrical contacts. The 60 percent palladium alloy Is employed tor electrical contacts operating at reasonably high currents and for precision resistance wires. The lower-palladium-content 4
alloys are used for contacts in low-voltage relays and regulators. Palladiumsilver alloys are used for brazing stainless steel and other heat-resistant J alloys.
Palladium-silver alloys that have additions of copper gold, zinc, aind platinum are age-hardenable and yield high mechanical properties after . ,
heat treatment.
'
This element cannot bi
temperatures above 2800CF. The gen approaches that of iridium. The met£ as a catalyst for the synthesis of certa as a hardener for platinum and palla Typical mechanical properties of s 12-49 Electrical Contacts Since many of are widely used for electrical conta discussed in greater detail (The H. A
The properties of an ideal contact High electrical conductivity for maximum
Low temperature coefficient of resistanci form as possible High thermal conductivity to decrease te dency toward oxidation Low surface contact resistance to utilize
High melting point to prevent the formati surface roughening High boiling point to prevent local vapot High corrosion resistance to prevent an i
High nonwelding and nonsticking charac
i
:
v
I
jGY
NONFERROUS METALS AND ALLOYS
12-45 Iridium :
Iridium is the most corrosion-resistant element known
.
541
Pure
iridium has been used for crucibles in studying slag reactions at very high temperatures and as extrusion dies for very high-melting glasses. ,
The main use of iridium is as a hardening addition to platinum Small amounts of iridium up to about 0.1 percent, are used for refining the grain size and improving the mechanical properties of gold- and silver-base
I
.
,
Mi
casting alloys.
G 12-46 Osmium
Osmium has a high melting point and cannot be worked even at
very high temperatures. Osmium and its alloys have high hardness, high
Os
r
,
i
'
wear resistance
,
ij /
and good corrosion resistance. Typical applications in-
elude fountain-pen nibs, phonograph needles
,
electrical contacts and in,
strument pivots.
Ag
12-47 Rhodium Rhodium is similar to platinum in color and has considerably higher reflectivity. It has exceptionally high corrosion resistance almost equal to that of iridium. Rhodium provides a nontarnishing electroplate with high reflectivity. It is used as a finishing plate in the jewelry field and
Rh
,
Au
for reflectors for motion-picture projectors and aircraft searchlights. A thin rhodium plate is sometimes used on the surfaces of sliding electrical contacts. The main use of rhodium is as an alloying addition to platinum and palladium. 12-48 Ruthenium This element cannot be cold-worked but may be forged at temperatures above 2800"F. The general corrosion resistance of ruthenium approaches that of iridium. The metal is rarely used in the pure form except 10 "
as a catalyst for the synthesis of certain hydrocarbons. It is employed mainly
!;"'
as a hardener for platinum and palladium.
Jrlilion, weiyhl pcrcem
'
.
.
Typical mechanical properties of some metals and their alloys are given
on
in Table 12-19.
12-49 Electrical Contacts Since many of the precious metals and their alloys are widely used for electrical contacts, this particular application will be
jig effect o( alloying additions on pallaciium
discussed in greater detail (The H. A. Wilson Company, Union, N.J.).
ietals shown ruthenium and nickel are very
The properties of an ideal contact material are:
,
um is least effective.
containing 1
,
3
High eleclrical conductivity lor maximum current-carrying capacity
10, 40, 50, and 60 percent a
,
Low temperature coefficient of resistance to keep contact resistance as nearly uni-
electrical contacts. The 60 percent palla- |
form as possible
rical contacts operating at reasonably high J i
Stance wires. The lower-palladium-content !l f jw-voltage relays and regulators Palladium- | | jig stainless steel and othef heat-resistant - -
:
-
y "
.
J
gold, zinc, and mL
High boiling point to prevent local vaporization and loss of material during arcing
.
.
"
copper
,
Low surface contact resistance to utilize minimum contact pressure High melting point to prevent the formation of molten bridges, loss of material, and surface roughening
'
Shave additions o l
High thermal conductivity to decrease temperature rise ol contact and reduce tendency toward oxidation
ind yield high mechanical properties after
High corrosion resistance to prevent an increase in contact resistance High nonwelding and nonsticking characteristics
m mm
. V:
.
542
INTRODUCTION TO.PHYSICAL METALLURGY
High hardness and toughness to preven in parts that operate at high frequencies o H
It is apparent, from the properties I a universal contact material. The pr is based upon combining two or mo ing the less advantageous properties The contact materials may be clas; in the following groups;
CO
<
m
tu
O to
Q 0-
i°
IT)
in
d
d
CO CO
O
CM
i-
CM
c\j
CM
in
CO
CM
CO
CO i-
oominooajco -incn rcnoocD'-
X CO
r
CM CM
CDh
LO h-
O m
tCO
O CD
CD
r-~
co
cd
t-
CM
lO
t-
t-
moooooinco
.oh-
COOlCMCDT-HO TCMOCDCDOCDh-i-inOl l'h-.COh-T-OOlinOlinCM T" TT- TT1-T-TCMl11- T- 3 CO CM T-
.ON-
"
.
High conductivity; silver and silver alloys
CM
m
Corrosion- and oxidation-resistant: platin
.
Refractory and arc-resistant: tungsten am High conductivity plus arc-resistant: powt
o I
-
< . o z
Silver and Silver Alloys Silver has tl ductivity of all the contact materials relatively low temperatures because
z CO
O _
l
HI -
. .
lACO
COOO.OSOJOCMOOCMCOCOCOCMCO
Tl-
CMCO-
CM
CM
CO
lO
CVJ-CO.-fl
CO
o ir) r - co d cm t-
lO
co
"
-fl-
CO
CO
J
thus maintaining low contact resista
H C3
tacts under light and intermediate p if current and voltage do not becom tages of silver are its low hardness, sulfide films, and tendency to build
z DC
t _
m1
to cm
co
o
o
m m
-
tj
m
cm
f--
t
-
S
3ai
:5
in t-
in
- f co
o
in
in
o
o
o
f
-
in
j
o
to t-
current conditions. These disadvanti C3 z
a
LU
CC
(0
I
_
UJ
J
d D-
and increase corrosion resistance.
t/3 o >>
z g LU 0.
co
S
SSw
ScMSinSS
oin
o
ooooooto
Soo
I
OT r
Platinum, Palladium, and Gold
l£)»-Oa>CO
-
i-
CD
< .
o
n
s o
S o
a
to maintain low contact resistance o Z
sistance to oxidation at the high lo Since these metals are soft, they are r
o H
to y
Q Z
10
V
a
_
o o
alloyed to obtain higher hardness wit
to
5
sistance or surface-contact resistan(
o o
<
i 1
i
t
E
o
E
E
a
5
.
o>
T-
UJ CD
<
The
this group is their corrosion resistar the sureness of making contact unde
4
D
£
alloying elements, principally copper platinum, palladium, and iron. The < ness, raise the melting point, reduc resistance to welding or sticking, inc
< o
0
!= CO
i
CO
=2I
3
11.
£
0 ? e 3
liisl
o
o
E E
e e
E
o
Q
.
E E
E
=> - £ E =
I + to .5 c s a.
o a
-
§ a. S
fl
ments are iridium, ruthenium, osmiurr
and iron are also used as alloying ele Tungsten and Molybdenum Tungsti property of high resistance to arc e points, high boiling points, and hig They are used in applications requirii the contacts have to operate freque
5';.,.V T'....J"',H1«»
i
I
.VY
NONFERROUS METALS AND ALLOYS
543
v
High hardness and toughness to prevent mechanical wear and failure particularly ,
in parts that operate at high frequencies or under high contact pressures
It is apparent, from the properties listed, that no one metal or alloy can be
1 in
co
.T
CD
CO
-t
in
CvilOOi-OCD
i-
CM
LO CO CD - t
CO CD
CM
i-
LDi-i
I-
-
l"
a universal contact material. The practical selRCtion of a contact material is based upon combining two or more desirable properties while minimizing the less advantageous properties for a particular application.
The contact materials may be classified according to contact properties in the following groups: High conductivity: silver and silver alloys
CO
CM
t-
.
Corrosion- and oxidation-resistant: platinum and related alloys Refractory and arc-resistant: tungsten and molybdenum
t
High conductivity plus arc-resistant: powder-metallurgy compacts
O
lO
h-
CO
TJ-
CO
CO
CO
CO
O
CM
Silver and Silver Alloys
Silver has the highest electrical and thermal con-
ductivity of all the contact materials. The oxides of silver decompose at relatively low temperatures because of arcing reverting to metallic silver, thus maintaining low contact resistance. Silver is used for sensitive con-
CO
,
"
"
tacts under light and intermediate pressure. They operate satisfactorily if current and voltage do not become excessive. The principal disadvantages of silver are its low hardness low melting point, tendency to form sulfide films, and tendency to build up on one electrode under excessive
CD
,
lO
O
lO
m
O
O
CD
o Tt- r--
t-
-
t
O m
current conditions. These disadvantages are minimized by the addition of alloying elements, principally copper, cadmium, zinc, nickel, manganese, platinum, palladium, and iron. The effect of alloying is to increase hardness, raise the melting point, reduce material loss or transfer, increase resistance to welding or sticking increase resistance to erosion by arcing,
5
,
13
o> m
o
§ §
!
o?§?o§o t-
i-
i-
CD
and increase corrosion resistance.
9 13
Platinum, Palladium, ahd Gold
O
53
5
o -o -o B
I
,
S -
ra c 5 n ip
3 r
ctocraccciiocooiccracc
5
I
,
a
ST + £
-
Ei|| £EoQ11 If 1 ,
p-5C
.
3
.
o) =
'r' £ £ -
E
S E
E
.
o
II E o
it? f
to maintain low contact resistance over long periods because of their resistance to oxidation at the high local temperatures reached in arcing. Since these metals are soft, they are rarely used in the pure form. They are alloyed to obtain higher hardness without undue sacrifice of corrosion resistance or surface-contact resistance. The most common alloying elements are iridium ruthenium, osmium, rhodium, and silver. Copper, nickel,
3
The outstanding property of the alloys in
this group is their corrosion resistance to surrounding atmospheres and the sureness of making contact under light pressures. The metals are able
+>--
and iron are also used as alloying elements, Tungsten and Molybdenum Tungsten and molybdenum have the unique property of high resistance to arc erosion along with very high melting points, high boiling points, and high resistance to welding and pitting. They are used in applications requiring high contact pressures and where the contacts have to operate frequently or continuously. The chief dis-
i
ft
;
544
INTRODUCTION TO PHYSICAL METALLURGY
advantage of tungsten and molybdenum is their tendency to form oxides particularly where severe arcing takes place. In some applications these disadvantages may be overcome by using very high contact pressures, by : f incorporating a wiping action in the gontacts, or by use of protective cir- f cuits to suppress excessive arcing.
Powder-metallurgy Compacts
Silver and tungsten do not alloy with each
other by the ordinary process of melting and casting.
These metals, how-
12-14
Differentiate between the terms b
12-15 12-16
Why are tin bronzes suitable for i Why is beryllium suitable for tool;
12-17 What are the outstanding propert 12-18 What are the outstanding propert 12-19 Why do long-range electrical tre aluminum shell?
r
12-20
Explain the meaning of the digits
2107-T4, 5056-H16 7075-T6, 6061-0.
ever, may be combined by using the techniques of powder metallurgy (dis- I cussed in Chap. 16) to produce alloys which are homogeneous and have high physical properties. It is thus possible to take advantage of the high electrical conductivity of silver or copper and the high arc resistance of tungsten or molybdenum. These materials are designed to operate under ; conditions of heavy current and voltage and are also often used in low- * current highrfrequency applications. Typical combinations in this group are silver-tungsten, silver-molybdenum, silver-tungsten carbide, silver-molybdenum carbide, coppertungsten carbide, and copper-tungsten. Other compositions are manufactured by powder metallurgy to take advantage of the high electrical conductivity of silver and copper and the semirefractory properties of cadmium oxide, iron, graphite, and nickel. Many of these compositions can be produced in strip and wire form and i may be drilled, rolled, swaged, drawn, formed, bent, and extruded without difficulty. They may be used as replacements for fine silver or silver alloys in some applications, since they give greater resistance to sticking and welding, have improved mechanical properties, and possess greater re-
Hastelloy C, and lllium G 12-33 Give the composition special prop( iron-nickel alloys: Invar Kovar, Platinite, a
sistance to electrical erosion.
12-34 What are the outstanding propertie
1
,
12-21
Why do many aluminum alloys re
amples.
12-22 Why do aluminum alloy 2017 rivet 12-23 What outstanding properties are ( nickel?
12-24
What outstanding properties do a
typical applications
.
12-25
Which aluminum casting alloy dev
Why?
12-26 What is meant by anodizing alumi 12-27 What are the outstanding property 12-28 12-29
What is the effect of zirconium ad' What is the effect of a rare-earth r
addition?
12-30 12-31
Compare aluminum and magnesiu Discuss the methods used to join
12-32
Give one application and the reas
alloys: duranickel permanickel, Monel, K ,
.
,
,
property.
QUESTIONS
12-35 What property is important for fu suitable alloy for this application 12-36 What properties are important in al .
12-1 12-2 12-3 12-4
What is.the most important property of copper? Explain why copper is a suitable material for automobile radiators. Explain the reasons for the difference in microstructure of Figs. 12-1 and 12-2. What would be the temper of 0.25-in.-diameter copper wire cold-drawn from
0 50-in.-diameter soft wire? .
12-5 12-6 12-7 12-8 Why?
What is season cracking'? How may it be minimized? What \s\dezincificationl How may it be minimized? How does the addition of lead to brass improve its machinability? Which copper alloy would be best for the tubes in a marine heat exchanger?*
12-9
Why are most copper-zinc alloys not age-hardenable?
12-10 Discuss the effect on corrosion resistance of copper by increasing addition?;
'
jc-V
steel?
12-42 12-43
What are the outstanding propertie Which of the platinum metals has t
has the highest modulus of elasticity? 12-44 Name two important properties of t 12-45 What is one disadvantage of titaniu 12-46 What are the two crystal structures tion temperature?
12-47
Explain what is meant by an alpha
12-13
12-48
Name one alpha stabilizer and thre
Why is "manganese bronze" a misnomer?
:
.
Which type of soft solder is preferr
12-39 Compare the lead-base and tin-ba applications. 12-40 Why is it difficult to work-harden le 12-41 Why is tin-coated steel used for tir
12-12 What properties would be important in the choice of a copper-ahoy spring!
>
v
Why is "white metal" suitable for b
of (a) zinc, (£>)'tin, (c) nickel. 12-11 Why is muntz metal heat-treatable? Describe atypical heat treatment and tl resulting microstructure.
f
.
12-37
12-38
r
NONFERROUS METALS AND ALLOYS
iY
bdenum is their tendency to form oxides|
takes place. In some applications thesi
8
by using very high contact pressures bj the contacts, or by use of protective cli ,
g
-
,
he techniques of powder metallurgy (di!
.
-jr
V
12-22 12-23
copper and the high arc resistance
12-24 typical 12-25 Why? 12-26 12-27 12-28 12-29
voltage and are also often used in |Q'
>ns.
vgroup are silver-tungsten, silver-mo! .
,
:
;:
] silver-molybdenum carbide copped j-tungsten. Other compositions ,
,
gy to take advantage of the high electrical.
and the semiref ractory properties of cadtt
What is meant by anodizing aluminum?
What are the outstanding properties of magnesium? Of nickel? What is the effect 6f iirconium additions to magnesium alloys? What is the effect of a rare-earth metal addition to magnesium? Of thorium
12-30 Compare aluminum and magnesium with regard to corrosion resistance. 12-31 Discuss the methods used to join magnesium alloys. 12-32 Give one application and the reasons for selection of the following nickel
n be produced in strip and wire form ai ,
What outstanding properties do aluminum-silicon alloys have? Give some applications. Which aluminum casting alloy develops the highest mechanical properties?
addition?
lickel.
awn, formed
Why do aluminum alloy 2017 rivets have to be refrigerated until used? What outstanding properties are generally given an alloy by the addition of
nickel?
! materials are designed to operate untf '
Why do many aluminum alloys respond to age hardening? Give some ex-
amples.
js possible to take advantage of the higl -
Why are tin bronzes suitable for use as bearings? Why is beryllium suitable for tools in the petroleum industry? What are the outstanding properties of cupronickel alloys? What are the outstanding properties of aluminum? Why do long-range electrical transmission lines use a steel core and an
12-21
alloys which are homogeneous and ha'
.
Differentiate between the terms brass and bronze.
12-15 12-16 12-17 12-18 12-1!)
12-20 Explain the meaning of the digits in the following aluminum specifications: 2107-T4, 5056-H16, 7075-T6, 6061-0.
melting and casting. These metals ho'
-
12-14
aluminum shell?
Silver and tungsten do not alloy with eac: :
545
bent, and extruded withoijl
alloys: duranlckel, permanickel, Monel, K Monel, constantan, Inconel, Inconel X,
eplacements for fine silver or silver all6y|
Hastelloy C, and lllium G. 12-33 Give the composition, special properties, and one application of the following
a
give greater resistance to sticking and ical properties, and possess greater
Iron-nickel alloys: Invar, Kovar, Platinite, and Elinvar.
12-34 What are the outstanding properties of lead? Give one application for each property.
12-35 What property is important for fusible plugs? Give the composition of a suitable alloy for this application. '
iperty of copper?
?rence in microstructure of Figs. 12-1 and 12-2. 1 25-in.-diameter copper wire cold-drawn front:
12-36 What properties are important in alloys for type metals? 12-37 Why is "white metal" suitable for bearing applications? 12-38 Which type of soft solder is preferred for electrical equipment and why? 12-39 Compare the lead-base and tin-base babbitts with regard to properties and
/ may it be minimized? nay it be minimized?
applications. 12-40 Why is it difficult to work-harden lead, tin, or zinc at room temperature? 12-41 Why is tin-coated steel used for tin cans to hold food and not zinc-coated
to brass improve its machinability?
steel?
est for the tubes in a marine heat cxchangerf!
12-42
What are the outstanding properties of (a) silver, (b) gold, and (c) platinum?
12-43
Which of the platinum metals has the highest corrosion resistance? Which
le material for automobile radiators.
.
ys not age-hardenable?
has the highest modulus of elasticity? 12-44 Name two important properties of titanium. 12-45 What is one disadvantage of titanium?
m resistance of copper by increasing addilion$ ?| .
;
able? Describe a typical heat treatment andth*
12-46 What are the two crystal structures of titanium, and what is the transforma-
iortant in the choice of a copper-alloy spring?
tion temperature? 12-47 Explain what is meant by an alpha or beta stabilizer. 12-48 Name one alpha stabilizer and three beta stabilizers.
-
a misnomer?
!
546
INTRODUCTION TO PHYSICAL METALLURGY
i
12-49
m
Explain why the two-phase titanium alloys are stronger than the s;lngle-phas«
alpha alloys.
J
12-50 What are the two main attributes of alpha alloys?
/
12-51
-j
How may alpha-beta alloys be strengthened?
12-52 How may beta alloys be strengthened? 12-53 Give at least two applications for alpha alloys, alpha-beta alloys, and beta!
I
alloys.
ME" HIG
S
REFERENCES
Alice, J.: "Introduction to Magnesium and Its Alloys," Ziff-Davis Publishing, Chicago,! "
1945.
V
'
Aluminum Company of America: "Casting Alcoa Alloys," Pittsburgh, Pa., 1951. American Society for Metals: "Magnesium," Metals Park, Ohio, 1946. Metals Handbook," Metals Park, Ohio, 1948,1961,1972.
:
"
:
"
Physical Metallurgy of Aluminum Alloys," Metals Park, Ohio, 1949.
Bunn, E.S., and R. A. Wilkins: "Copper and Copper-base Alloys," McGraw-Hill Book? |13-1 Introduction E
Company, New York, 1943.
1955.: I
Dow Chemical Company: "Magnesium Alloys and Products," Midland, Mich Ellis, O. W.: "Copper and Copper Alloys," American Society for Metals, Metals Park,'
Ohio, 1948. 1 Gibbs, L E.: "Cold Working of Brass," American Society for Metals, Metals PariCl '
k
will exhibit characteristics at low tern
I
higher temperatures for other metals
Internationar Nickel Co.: "Nickel and Its Alloys," "Nickel and Nickel-base Alloys."' Age-hardening Inco Nickel Alloys," "Engineering Properties of Duranickel," "En- f "
"
"
gineering Properties of Inconel, "Engineering Properties of Monel, "Engineering Properties of Nickel," New York. Lead Industries Association; "Lead In Modern Industry," New York, 1952.
.
1000oF may be necessary to recrysta and lead will recrystallize at or near rc metals are usually determined at room
!
metals is based on their behavior at n(
Liddell, D. M.: "Handbook of Nonferrous Metallurgy," McGraw-Hill Book Company.
p
New York, 1945.
perature is changed, becoming either havior of metals often occur which ma
Mathewson, C. H. (ed ): "Modern Uses of Nonferrous Metals," 2d ed., American In-. f, stltute of Mining and Metallurgical
melting metals such as tin and lead m
for a high-melting metal such as tungsl
"
Ohio, 1946-
The terms high and lowtt
own natural environment. What is co
'
Engineers, New York, 1958.
| |
a particular application
.
Mondolfo, L. P.: "Metallography of Aluminum Alloys," John Wiley & Sons, Inc., New.J York, 1943
National Bureau of Standards (U.S.): Zinc and Its Alloys, Natl. Bur. Std. (U.S.) Cirtj 395, 1931.
;
|
Raudebaugh, R. J.: "Nonferrous Physical Metallurgy," Pitman Publishing Corpora-/
tion. New York, 1952.
|
1959.
I
Raynor, G. V.: "Physical Metallurgy of Mg and Its Alloys," Pergamon Press, New Yoriu
METALS AT HIGH TEMPERATURES 3-2 Elevated-temperature
r
Tests The beha at elevated temperatures depends upor life expectancy of machine parts is usl test for many years to determine what
Reynolds Metals Company: "Heat Treating Aluminum Alloys," 1948; "AluminuiB;
necessary to extrapolate from shorter-1
Forming," 1952; "Finishes for Aluminum," 1951, Louisville, Ky. -i Roberts, C. S:: "Magnesium and Its Alloys," John Wiley & Sons, Inc., New York, I960, i
ever, must be done with great care fror
Samans, C. H;: "Engineering Metals and Their Alloys," The Macmillan Company, Nswl York, 1964. i
This is especially difficult for high-temp in behavior will occur with time at tern|
Van Horn, Kent R.: "Aluminum," 3 vols., American Society for Metals, Metals Part, Ohio, 1967;' Vines, R. F.: "The Platinum Metals and Their Alloys," The International Nickel Cft,
ultimate strength (rupture strength) anc
New York, 1941.
In high-temperature tests it is necessi the time application of the stress In ii metals at elevated temperatures it is c( .
,
u
f i
the test specimen. While this procedure
J LURQY
j
se titanium alloys are stronger than the single-ph;
Attributes of alpha alloys? '
'
:
ys be strengthened?
:
strengthened?
ations for alpha alloys, alpha-beta alloys, and -
s\um and Its Alloys, Ziff-Davis Publishing, Chi "
it
|
i
"Casting Alcoa Alloys," Pittsburgh, Pa., 1951.
METALS AT HIGH AND LOW TEMPERATURE c
jagnesium," Metals Park, Ohio, 1946. lals Park, Ohio, 1948, 1961, 1972.
Mumlnum Alloys," Metals Park, Ohio, 1949.
-l Introduction The terms high and low temperature are entirely relative to our own natural environment. What is considered a high temperature for low-
kipper and Copper-base Alloys," McGraw-Hill 8
l
lesium Alloys and Products, Midland, Mich,, 1 r Alloys, American Society for Metals, Metals Pi "
melting metals such as tin and lead may be considered a low temperature for a high-melting metal such as tungsten. Therefore, lower-melting metals will exhibit characteristics at low temperatures that will require relatively higher temperatures for other metals. For example a temperature of about 1000°F may be necessary to recrystallize iron after cold working but tin and lead will recrystallize at or near room temperature. The properties of metals are usually determined at room temperature, and our thinking about metals is based on their behavior at normal temperatures. When the temperature is changed, becoming either higher or lower, changes in the be-
"
Brass," American Society for Metals, Metals Part
,
"
and Its Alloys,
"Nickel and Nickel-base
Alloy*
,
loys," "Engineering Properties of Duranickel," "Engineering Properties of Monel, "Engineff "
"
,
k
.
ad in Modem Industry, New York, 1952. nferrous Metallurgy, McGraw-Hill Book Comp "
"
"
n Uses ol Nonlerrous Metals, ;
havior of metals often occur which may seriously affecl their usefulness in a particular application.
2d ed., American
rgical Engineers, New York, 1958. of Aluminum Alloys," John Wiley & Sons, Inc.
ALS AT HIGH TEMPERATURES
ji.S.): Zinc and Its Alloys, Natl. Bur. Std. (U.S.) d
.2 Elevated-temperature Tests The behavior of metals observed by stressing at elevated temperatures depends upon the length of the test period. Since life expectancy of machine parts is usually high it is not possible to run a test tor many years to determine what to use in current construction. It is necessary to extrapolate from shorter-time tests This extrapolation, however, must be done with great care from tests that will provide useful data. This is especially difficult for high-temperature applicatiohs since changes in behavior will occur with time at temperature
Physical Metallurgy," Pitman Publishing Cori gy of Mg and Its Alloys, Pergamon Press, New V "
,
sat Treating Aluminum Alloys, 1948; "Alumlfl Aluminum," 1951, Louisville, Ky. "
.
Its Alloys," John Wiley & Sons, Inc., New York, 1
,
"
als and Their Alloys, The Macmillan Company,
:
i
.
In high-temperature tests it is necessary to determine the dependence of ultimate strength (rupture strength) and yield strength (Creep strength) on the time application of the stress. In investigating the plastic behavior of
3 vols., American Society for Metals, Metals Pk 1
als and Their Alloys, The International Nickel "
metals at elevated temperatures it is convenient to apply a tensile load on ,
the test specimen. While this procedure in many cases, does not duplicate
5
,
i
543
INTRODUCTION TO PHYSICAL METALLURGY
META
service conditions, it is possible, by careful interpretation, for the data
to provide useful iriformation which can be applied to combined stres$ [ >
tinues at an approximately constant n between the rate of work hardening
.
:
conditions:
:-
.
I
Many tests have been developed jor high-temperature studies, but th* f three most widely used ones are; 1
I
Creep tests at small deformations: low stresses and low strain rates for long timi ;
periods 2 Stress-rupture (creep-rupture) tests at larger deformations: higher stresses and; larger strain rates for shorter periods of time i 3 Short-time tensile tests at large deformations: high stresses and high strain rates; r | available with the usual tension-testing equipment '
i '
13-3 Creep Tests Creep is a property of great importance in materials used for; ?; high-temperature applications. It may be defined as a continuing slow plas- : ,
tic flow under constant conditions of load or stress. Creep is generally associated with a time rate of deformation continuing under stresses well
below the yield strength for the particular temperature to which the metal Isubjected. It occurs at any temperature, though its importance depend* |v ,
upon the material and the degree to which freedom from continuing de formation is desired. IBv
temperature. There is a means of measuring the elongation of the specif
men very accurately and a means of heating the specimen under closelyf
controlled conditions.
The total creep or percent elongation is plotfe(J|
.
stage may continue for a very long ti the stress is sufficiently high there h ,
rate accelerates until fracture occurs There is little or no correlation betv ical properties of a material and its c greatly affected by small variations in .
grain size of the metal is an important acteristics. Whereas at room tempe higher yield and ultimate strengths th verse is true at elevated temperatures
is
A creep test is simply a tension test run at constant load and constanil
recovery or recrystallization In some creep rate may continue to decrease i
.
tures the grain boundaries may act as tions which cause creep .
The presence of solute atoms ever creep by interfering with the motion ,
A more potent factor in retarding cree
finely dispersed second phase 13-4 Stress-rupture Tests These tests are c .
a material to resist fracture at elevated \
against time for the entire duration of the test. Two typical creep curve
the loads are high enough to cause cc involved is usually between 10 and 40(
are shown in Fig. 13-1.
long as 1,000 h,
The various stages of creep are illustrated by curve A. When the loadi
first applied, there is an instantaneous elastic elongation, then a primar|| stage of a transient nature during which slip and work hardening take plac in the mosHavorably oriented grains. The creep rate (tangent to the cur is initially high and gradually slows to a minimum. This is followed by. secondary stage of steady-state creep, during which the deformation cqij
A series of specimens are broken at constant load, the stresses being sele few minutes to several hundred hours
log-log coordinates and if no struct: ,
period, the relationship of rupture st Typical stress-rupture data for S-590 (i are shown in Fig. 13-2. Discontinuitie:
with changes in the alloy and indica ure from transcrystalline low-temper;
To fracture
Primory i Secondory stoge i Tertiory
t
creep |steady state creep: creep
temperature type. The principal differences between tl the testing time, the stress or strain rat
I I
g o
I
and measurement of temperature Ioe fracture to determine the elongation a dinary creep test. From these data the at very high stresses may also be deter
B
,
a>
,
}Elos1ic strain Time-
Fig. 13-1
minimum creep rate for S-590 alloy is For some applications such as super
Typical creep curves illustrating the stages of
,
creep.
;
i
i
LLURGY
METALS AT HIGH AND LOW TEMPERATURES
549
tinues at an approximately constant rate. During this stage a balance exists
)ssible, by careful interpretation, for the on which can be applied to combined sti
between the rate of work hardening and the rate of softening because of recovery or recrystaliization. In some cases, under moderate stresses the
:
li
,
creep rate may continue to decrease at a very slow rate, and the secondary stage may continue for a very long time (curve S, Fig. 13-1). However if the stress is sufficiently high, there is a tertiary stage in which the creep
iveloped for high-temperature studies, but s are:
,
jtions: low stresses and low strain rates for long
rate accelerates until fracture occurs.
There is little or no correlation between the room-temperature mechanical properties of a material and its creep properties. Creep seems to be greatly affected by small variations in microstructure and prior history. The
e) tests at larger deformations: higher stresses sriods of time
rge deformations: high stresses and high strain
|-testing equipment
"
p perty of great importance in materials used
|ns.
It may be defined as a continuing slowpti
fiditions of load or stress. Creep is genei iof deformation continuing under stresses
grain size of the metal is an important factor in determining its creep characteristics. Whereas at room temperature fine-grained materials show higher yield and ultimate strengths than coarse-grained materials the reverse is true at elevated temperatures. It is believed that at high temperatures the grain boundaries may act as centers for the generation of disloca| tions which cause creep ,
.
the particular temperature to which the
.
jy temperature, though its importance depti
;:
.
.
-
v;V;;|jegree to which freedom from continuing
A more potent factor in retarding creep is the presence of a strong, stable, finely dispersed second phase.
1
jnsion test run at constant load and coffi
.
The presence of solute atoms, even in minor amounts, tends to retard creep by interfering with the motion of dislocations through the crystal.
v;.;; ans of measuring the elongation of the 9]
Stress-rupture Tests these tests are conducted to determine the ability of
means of heating the specimen under cfi
a material to resist fracture at elevated temperatures. In stress-rupture tests
total creep or percent elongation is p
| the loads are high enough to cause comparatively rapid rupture. The time
iuration of the test. Two typical creep
involved is usually between 10 and 400 h, although some tests may run as I;
p are illustrated by curve A. When the loi tantaneous elastic elongation, then a pri ring which slip and work hardening take id grains. The creep rate (tangent to the cil S;;J:ay slows to a minimum. This is followed tate creep, during which the deformation ;
'
:
I
long as 1,000 h.
A series of specimens are broken at each temperature of interest, under constant load, the stresses being selected so fractures will occur from a few minutes to several hundred hours. The results are usually plotted in log-log coordinates, and if no structural changes occur during the test period, the relationship of rupture stress and time to rupture is linear. Typical stress-rupture data for S-590 (cobalt-chromium-nickel-base) alloy are shown in Fig. 13-2. Discontinuities in the straight lines are associated
with changes in the alloy and indicate a change in the method of failure from transcrystalline low-temperature type to intercrystalline high-
To racture
temperature type.
The principal differences between the stress-rupture and creep test are the testing time, the stress or strain rate level, and the sensitivity of control and measurement of temperature load, and strain. It is possible before
8
,
fracture to determine the elongation as a function of the time as in an ordinary creep test. From these data the steady-state or minimum creep rate at very high stresses may also be determined. The log-log plot of stress vs ,
,
:
.
ie stages of
minimum creep rate for S-590 alloy is shown in Fig 13'3. For some applications such as superheater tubes, still tubes, piping pipe .
,
V
.J
,
I! *
?
: .
W
'
550
if
i
INTRODUCTION TO PHYSICAL METALLURGY
MET
at lower loads. As a method of ratii IOC
two different lots of the same alloy
,
t
correlation with creep tests at usabl
1200
In some applications the design signed for a life of 1 h, and turboje 1 000-h life. In these cases the test p
1350 °F .1500 "F o
,
1600 0F
2
and stress-rupture data can sometii
IC
1700 °F
For applications of longer design lifi with a design life of 100 000 h (13 ye£ ,
obtained during shorter time period;
1900 "F
.
0 00 .
0 1
00
1
.
.
10C
IC
Materials for high-temperature u; minimum creep after a stated peric quantity used in design, is the stress
10,000
I00C
deformation in a stated time. Some
Rupture time, hours
Fig. 13-2
dard of 0.1 percent creep in 10 000
Log-log plot of stress vs. rupture time for S-590
,
High Temperature Properties of Metals," American Society for Metals, Metals Park, Ohio, 1951.) alloy. {From
"
in 1,000 h.
fittings, sheet-metal parts, nozzle guide vanes, and boilers, only ruptura* dataware important. For other applications, such as bolts, steam valves,!
13-5 Short-time Tension Tests
steam-turbine blading, turbine rotors, turbine casings, and valve sterns, creep data are considered most important.
Considering the fact that the load in the stress-rupture test is much higherj than design values and that the test continues to fracture, there is sore f
doubt as to the usefulness of the data obtained. Metal behavior at high
*
loads, high rates of deformation, and short life may not be indicative of thah
'
These test
ing a sample and testing under strain tensile-testing machine. Elastic pre not real, since their values depend tions, and their accuracy depends o The duration of testing is usually oi effects of time at temperature are nc test fails to predict what will happen and therefore has very little applica' rapid estimation of materials which short-time tensile strength is frequet
100-
ture curve.
The variation of the she
Inconel X with temperature is shown 13-6 Creep Properties of Various Alloys widely used for moderate-temperat
in
O
900oF. An increase in carbon contenl in
10
vfoo
temperatures where the carbides are may be true at higher temperatures 1 The recommended structure of plalr
5
'
o9 L
2
0 001 .
0.01
01
.
.
-
1
10
100
Minimum creep rote, percent/hour * :
:
n;
:
.
Fig. 13-3 Log-log plot Of stress vs. minimum creep rate for S-590 alloy. (From "High Temperature Properties of Me-als," American Society for Metals, Metals Park, Ohio, 1951.)
1000
10,000
service is the normalized one.
The a
stable and tends to spheroidize mo The use of aluminum as a deoxidizer
produce fine grain, which lowers c should be kept low, and their effect is ( of manganese and molybdenum.
,
i v . -
i
i
I
r
r
URGY
METALS AT HIGH AND LOW TEMPERATURES
551
at lower loads. As a method of rating different alloys or in comparison of two different lots of the same alloy the stress-rupture test seems to show a ,
correlation with creep tests at usable loads.
1200 0F
In some applications the design life is short.
1350 "F
Guided missiles are de-
signed for a life of 1 h and turbojet engines are frequently designed for ,
1500 °F
1
000-h life, in these cases the test period can be as long as the design life and stress-rupture data can sometimes be used directly in such designs
1600 "F
,
,
.
1700 F
For applications of longer design life-for example steam or gas turbines ,
with a design life of 100 000 h (13 years)-it is necessary to extrapolate data ,
obtained during shorter time periods.
.
1900 °F '
1
Iture time
10 ,
100
1000
Materials for high-temperature use are usually designed for a certain
1
minimum creep after a Stated period. The creep strength, which is the quantity used in design, is the stress required to produce a definite percent
10,000
deformation in a stated time. Some gas-turbine designers have ef a Stan-
hours
ne for S-590
dard of 0.1 percent creep in 10 000 h. This is equivalent to 0.01 percent
Melals,"
in 1 000 h.
,
,
1951..) :
v ozzle guide vanes, and boilers, only rupttJi V wsr applications, such as bolts, steam val\ -
.
..
13-5 ShOrt-time Tension Tests
These tests are used to study the effect of heat-
ing a sample and testing under strain rates that are available in the ordinary tensile-testing machine. Elastic properties at elevated temperatures are not real, since their values depend upon the time between load applica-
Vie rotors, turbine casings, and valve sfe Dst important.
3 load in the stress-rupture test is much higl
and their accuracy depends on the sensitivity of the extensometer. The duration of testing is usually only a few minutes, and the important tions
,
the test continues to fracture, there is 3
if the data obtained. Metal behavior at hlj
effects of time at temperature are not measured. The short-time tension test fails to predict what will happen in a shorter or longer period of time
on, and short life may not be indicative of thi
and therefore has very little application. The test is sometimes used for rapid estimation of materials which may warrant further study, and the short-time tensile strength is frequently used as the 0.1-h point on a rupture curve. The variation of the short-time yield and tensile strength of Inconel X with temperature is shown in Fig. 13-4.
13-6 Creep Properties of Various Alloys Plain-carbon and low-alloy steels are widely used for modemle-temperalure applications particularly below
f
,
900oF. An increase in carbon content improves the creep strength at lower temperatures where the carbides are present in lamellar form. The reverse may be true at higher temperatures where the carbides are spheroidized. The recommended structure of plain-carbon steels for high-temperature .
1
.\
5
10
100
urn creep rote, percent/hour
1000
10,000
service is the normalized one. The annealed structure appears to be less stable and tends to spheroidize more rapidly reducing creep strength. ,
.
creep rate
jperties of Park, Ohio,
The use of aluminum as a deoxidizer in the manufacture of steel tends to
produce fine grain, which lowers creep strength. Aluminum additions should be kept low, and their effect is considerably reduced by the presence of manganese and molybdenum.
552
MET
INTRODUCTION TO PHYSICAL METALLURGY
parts up to 1000oF. Type 422 is use( Both the above types are hardenable
180
pered about 100° above the operatii
150
structure. Tensile strengfh
Type 430 (16 percent Cr) and 446 (
140
ferritic. These grades generally usee creep strength than the hardenable sistance. Type 430 is used for heat-e
-
,
120
ing, and furnace parts operating at used for similar applications up to 2i The austenitic stainless steels (3x)
o
g 100 rield strength
in in
than the 4xx series and better corrosi
80
(25 percent Cr, 20 percent Ni) is us thermocouple wells aircraft-cabin h Type 347 (18 percent Cr 11 percent N superheater tubes and exhaust systc turbines operating up to 1600oF. Another group of alloys that have c ,
60
,
,
40-
ture range of 1200 to 1400°F are esser iron alloys. Many contain small amot
20-
0
o
200
400
500
800
1000
1200
1400
1600
1800
2000
m
Temperature, 0F Fig. 13-4 Short-time yield strength and tensile strength of Inconel X at elevated temperatures. (From "Metals Handbook," 1954 supplement, p. 42, American Society for Metals, Metals Park, Ohio.)
m -
4 m
In low-alloy steels, containing less than 10 percent alloy, molybdenum ||p
and vanadium are most effective in raising the creep resistance. The car- lii-
'
bon content is usually kept below 0.15 percent. The 0.5 percent molyb- |B|
denum steel is used for piping and superheater tubes up to 850oF. Above 'Jr
this temperature, spheroidization and graphitization tend to take place, |H| with a reduction in creep strength. The addition of 1 percent chromium JMK to this steel increases the resistance to graphitization, and the steel is now I I used for piping and boiler tubes at temperatures to 1000 :'| ;
oF
.
The chromium-molybdenum-yanadium steels containing up to 0.50 per- IHI
cent carbon are used in the normalized and tempered,-or quenched and :M|
1
i
m
trade names for alloys in this groi 16-25-6, and D-979. These alloys are various other components of gas turl and exhaust equipment. The nickel-base alloys such as IV R-235
,
Inconel 700, Udimet 500, am
aircraft applications. They are intern of 1400 to 1800oF. These alloys cont percent chromium, up to 10 percent percent cobalt, and titanium and alur collector rings, and exhaust valves o form for combustion liners, tail pipe jet engines. The cobalt-chromium-nickel-base
and N-155 are suitable for applicatic base alloys but have lower rupture str buckets of gas turbines.
tempered condition. .They have relatively high yield and creep strengths .'iBjand are suitable for bolts, steam-turbine rotors, and other parts operating [ M-
All the commercial alloys mentiom rapidly when heated above about 1700
at temperatures up to 1000oF. The straight-chromium (4xx series) stainless steels are used for elevatedtemperature applications that require increased corrosion and oxidation resistance. Type 410 is used for bolts, steam valves, pump shafts, and other
these metals will raise the allowable o| present limit. This limit is related to a the base metals. Most promising bas alloys are molybdenum (melting point
i
J?.
. -
.
„_-.
URGY
METALS AT HIGH AND LOW TEMPERATURES
553
parts up to 1000' F. Type 422 is used for similar applications up to 1200oF. '*
Both the above types are hardenable to a martensitic structure and are tem-
pered about 100° above the operating temperature to promote stability of structure.
rensile strength
Type 430 (16 percent Cr) and 446 (25 percent Cr) are nonhardenable and ferritic. These grades, generally used in the annealed condition have lower creep strength than the hardenable types but show greater oxidation resistance. Type 430 is used for heat-exchange equipment, condensers, pip,
ing, and furnace parts operating at temperatures to 1550°F. Type 446 is used for similar applications up to 2000oF. The austenitic stainless steels (3xx series) show better creep properties than the 4xx series and better corrosion and oxidation resistance. Type 310 (25 percent Cr, 20 percent Ni) is used for furnace linings, boiler baffles, thermocouple wells, aircraft-cabin heaters, and jet-engine burner liners. Type 347 (18 percent Cr, 11 percent Ni, + Cb andTa) is used for steam lines, superheater tubes, and exhaust systems in reciprocating engines and gas turbines operating up to 1600oF. Another group of alloys that have good creep properties in the tempera% 1000
Imperature
1200 ,
MOO
16C0
1800
2000
°F
Irength of lis Hand-
ture range of 1200 to 1400oF are essentially chromlum-nickel-molybdenumiron alloys. Many contain small amounts of titanium and aluminum. Some trade names for alloys in this group are A-286, Discaloy, Incoloy 901, 16-25-6, and D-979. These alloys are used as forgings lor turbine wheels, various other components of gas turbines, sheet-metal casings, housings, and exhaust equipment.
for Metals
,
the nickel-base alloys such as M-252, Waspaloy, Rene 41, Hastelloy Inconel 700, Udimet 500, and Unitemp 1753 are widely used for aircraft applications. They are intended for use in the temperature range R-235
jj less than 10 percent alloy
'ow 0.15 percent. The 0 5 percent molyb*| ind superheater tubes up to 850°F Above|
of 1400 to 1800oF. These alloys contain 50 to 70 percent nickel, about 20 percent chromium, up to 10 percent molybdenum or tungsten, up to 20 percent cobalt, and titanium and aluminum. They are used for manifolds, collector rings, and exhaust valves of reciprocating engines, and in sheet
m and graphitization tend to take
form lor combustion linors, tail pipos, nnd cnslngs of gas turbines and
molybdenum 3 in raising the creep resistance The carik ,
.
i
.
.
place
th. The addition of 1 percent chromlum|
mce to graphitization and the steel is now at temperatures to 1000' F ,
.
anadium steels containing up to 0.50 per/malized and tempered or quenched and| -
.......
,
; .,
,
jet engines. The cobalt-chromium-nickel-base alloys such as S-816, S-590, L-605, and N-155 are suitable for applications in the same range as the nickelbase alloys but have lower rupture strength. They are used for wheels and buckets of gas turbines.
relatively high yield and creep strengths ;|-turbine rotors, and other parts operating
All the commercial alloys mentioned above tend to lose their strength rapidly when heated above about 1700rjF. It seems unlikely that alloys using these metals will raise the allowable operating temperature much above the
ries) stainless steels are used for elevated-J; iquire increased corrosion and oxidation j)olts steam valves pump shafts and othef
present limit. This limit is related to a great extent by the melting points of the base metals. Most promising base metals for future high-temperature
,
,
,
alloys are molybdenum (melting point 4730oF) and tungsten (melting point
'
iA.'
.
554
ME
INTRODUCTION TO PHYSICAL METALLURGY
f
6170oF). These metals are relativ( purity form. A recently developed n titanium has higher rupture strenc any other commercial alloy. Two high density and their great susc advantage may be overcome by th resistant coating.
> m
>
x i -
CO
o
d
§ t-
i
m o
: d d
:
5 -G S -Q S -Q S
in
nj
N
2
CD CD' CD 5
m
jz iz i: i= S
o
-CD
.
. CD
:
: d
.
:'
o
o
i-
cm
ca
in
-
d d
<
'
.
E E
G
N
CD
8
: cn
I t-
05
-
CO
15
CO
05
CO
a
J3
3-
O O
ID
The nominal composition of sonr given in Table 13-1, while some ty for several of these alloys are giver
00
! d t- oJ oi f-- co :
o
.
o
in
o
cn
cn
3
METALS AT LOW TEMPERATURES
S : CN CO CO CO CO Cvi
13-7 Effect of Low Temperature on Propc below normal room temperature 1
m > o -
_
J3
O
,
j
>
]
o
_
<
] 1
CN
.
f 1-
j-
cn
in
co
TABLE 13-2
_
High-temperature Strength Values for I
-
< => z
]
;
o o
LU
! oo
! co CD
O
CD
>
I
:
Q -
-
l
O
13,8 8 8
I
CO
2
a>
O
.
CD
2
g-
CO
.
o
?
cc
LU X
X o
o
CO
E
z
.
o) i
t- in
CN
a
800
55,000 36,500
1200
20,000
Carbon-molybdenum
800
60,000
steel, 0.15 0, 0.55 Mo
1000
45,000
1200
27,000
Type 410 stainless
1000
48,000
steel, 13 Or
1200
25,000
1300
17,000
O
cc
I
CO
X o
03
03
03
n
n
n
o CN
o CN
ra
I
LU
in <
o 2 3
O
2
Z
CD
CD
ui
CN
CN
C\J
consnJcocOconJninJnJnJ
in
n
,
PSI
1000
annealed
9
o
.
o
o
Carbon steel, 0.150, 5 (7)
Jj
n '
STRENGTH
o
in in
TENSILE
!
<
CD
TEMP, F
< CD
*
ra
ALLOY
J
LU
_
O
SHORT-TIME
111
in
CO
J3
J3
CD
O O
o
o
o
o
o
CM
CM
i-
00
CN
o cc a)
E o CO
in
ID
6
CD
IT)
'
St
in
ocDiricNinocncDO)-*
o
in
o
o
i-
CNi-->-CJi-->-i-'--r--
C\J
CN
CN
CM
CN
Type 304 stainless
1000
60,000
steel, 19 Or, 9 Ni
1200
46,000
1300
37,000
1200
83,000
o c
p; N-155 (low 0),
o '
tn
o
£ o
o
o
X
X
CO
TO
EE
*
CO C
in i-
in 1-
m t-
o
Si
§88 8
1
d d d d
d d d d d d d d d o o
-
1-
o
in in o
m
CM
T
t-
1
o 00
Ul I
o > )
_
j
<
:' :
>, >, >,x
< Q 5= Q
m O
/
1350
99 000
20 Ni, 4 Mo,
1500
78 000
bal. Fe
SiliM W, 4 Ob, 4 Fe, bal. Oo
r-
in
:
V
S-816 0.4 0, 20 Or,
3 Mo, 3 W, 1 Ob,
O N
llilli|lfl|
-
40,000
.
a
I CD
O _
.
-
o O CC X
1500
CO
CN
e*5
60,000
20 Ni, 20 Oo,
,
in co
o
z
1350
-
d d d d d
15
1
0.15 0, 21 Or,
\
cd
in o m
- cd c? O) in co co S in ti
i
i
i
i
W > _l CO 2
E
Inconel X, 0.04 0, 15 Or, 73 Ni, 1 Ob, 2.5 Ti, 0.9 Al, 7 Fe
,
G ,
r
,
,
1600
60,000
1200
120 000
1350
93,000
1500
60 000
,
,
9 GY
i
METALS AT HIGH AND LOW TEMPERATURES
6170oF). These metals are relatively abundant and are available in higt purity form. A recently developed molybdenum alloy containing 0.5 percei titanium has higher rupture strength in the range of 1600 to 2000oF the any other commercial alloy. Two disadvantages of these metals are the high density and their great susceptibility to oxidation. This latter di; advantage may be overcome by the development of a suitable oxidatioi
>
i co
- - s .
2
. m m m m 5 '
.
i: i; i:
d
.
co
.
o
ra
ra
J3
J3
v
resistant coating.
.
The nominal composition of some wrought heat-resistant superalloys given in Table 13-1, while some typical high-temperature strength value for several of these alloys are given in Table 13-2.
k = V
; O W o> O w oq . O t- C\J C\J CO
o lO
o M
METALS AT LOW TEWlPERATURES
o CO
CO
CO
00
13-7 Effect of Low Temperature on Properties co
As the temperature is decreas*
below normal room temperature the hardness, yield strength, and, wi ,
o
cm
.
*
TABLE 13 2 High-temperature Strength Values for Several Alloys
<
SHORT-TIME
LU
oo
; co
cm
<
in
co
TEMP
ALLOY
I
F
"
TENSILE ,
STRENGTH, PSI
CREEP STRENGT STRESS-RUPTURE
y-
.
O
CD
1 000 H ,
1 000 H
;
Carbon steel, 0.15C,
i co p co to ni cu oi 1
N. J3 XI
3 J3 J3
DC
o
c
o
nJ
nj
ro
n
£i
a
a! s
o o o o o
CM C\] t-
C\J CM
o
01
55,000
26,800
18,50
36,500
12 000
5 750
2 7C
1200
20,000
2 200
620
2e
Carbon-molybdenum
800
60,000
30,000
21,DC
steel, 0.15 C, 0.55 Mo
1000
45,000
25,000
10,800
6 7C
1200
27,000
4 400
2 000
7C
O lO O O tCM CM
CM CM CV
01
o o d d d d
.
48,000
19,000
12,000
1200
25,000
7 000
2 200
1300
17,000
Type 304 stainless
1000
60,000
35,000
17,000
12,0(
steel, 19 Cr, 9 Ni
1200
46,000
14,000
7 000
4 0(
1300
37,000
9 000
3 900
(low C)
,
,
20 NI, 20 Co,
o ip io o u: cm t-; - to d d d d
,
,
.
1000
.
| -
,
Steel, 13 Cr
N-155
O
,
.
Type 410 stainless
0 15 C, 21 Cr IT
1 00C
800
s s
! CM i2 05 05
.
PER
1000
annealed
2
O
< m
0 01°/
to
5
o
.
,
-
.
2
- ro S. t- ir
0 1% PER ,
z
5
:
PSI
STRENGTH, PSI,
O TO
,
,
,
,
,
,
1200
83,000
37,000
19,000
16,01
1350
60,000
22,000
14,500
10,51
1500
40,000
13,000
8 000
5 0i
,
,
3 Mo, 3 W, 1 Cb, bal. Fe
I I1.:
in
O I
X
S-816 0.4 C, 20 Cr, ,
1350
99,000
30,000
18,000
1? 0
20 Ni, 4 Mo,
1500
78,000
17,000
11,500
81
4 W, 4 Cb 4 Fe, bal. Co
1600
60,000
9 500
6 500
50
Inconel X, 0.04 C,
1200
120,000
69,000
60,000
48,0
1350
93,000
42 000
37,500
30,0
1500
60,000
18,000
18,000
15,0
,
t-
ra
E
i"
5 =5 |£ |8||. X £ § D 3
£
5
,
co
,
,
'
CD c
,
,
in CD
oo
]
o
5
o
CO
1i
2
15 Cr, 73 Ni, 1 Cb
o Li.
2 5 Ti, 0.9 Al, 7 Fe .
,
,
556
INTRODUCTION TO PHYSICAL METALLURGY
few exceptions, the ultimate strength and modulus of elasticity of all metals and alloys increase. The variation of yield and tensile strengths of iron, nickel, and copper with temperature is shown in Fig. 13-5. In regard to the effect of temperature on ductility, metals fall into two distinct groups, those which remain ductile at low temperatures and those which become brittle. An indication of the amount of ductility, or plastic deformation, before fracture may be obtained from a study of the fracture surface. A cup-cone type of fracture is typical of a ductile material which has failed in shear after plastic deformation when tested in tension. A brittle material fails by cleavage with no evidence of plastic deformation. As the temperature is decreased, face-centered-cubic metals fracture only by shear and show a gradual and continuous decrease in ductility. Metals with other crystal structures may fail by shear at room temperature, but with decreasing temperature the mode of fracture changes from shear (ductile) to cleavage (brittle). The change in fracture often appears as a sharp drop in ductility. The effect of temperature on the ductility of iron, cop per, and nickel is shown in Fig. 13-6. The tensile properties of some steels and nonferrous materials at low temperatures are given in Table 13-3. Cleavage fractures of structural members are often sudden and unexpected and usually result in catastrophic brittle failure of the part. Great interest
ME
Temperature, 328
148
-
32
-
212
100
Ni
30 Cu 9
60 a>
/ CL
/
40
-
/ 20
/
-
"
0 200
-
100
0
-
IOC
Temperature °C ,
Fig. 13-6 Variation of ductility of iron copper, and ni with temperature. (From Behavior of Metals at Low 1 peratures," American Society for Metals, Metals Park, Ohio, 1953.) ,
"
Temperature, °F 328
-
-
100
148
32
212
r s
\
.
.
-
in this problem developed during V ships failed in a brittle manner wi:
\ \
S 80
\ \\
cases, the ship was split in two. The normal temperatures, yet the failure
-
\\\
2 CT1
deformation.
g 60
\
The tendency of steel to fail in a bi centration, increased speed of load
5
ture. 40
20
are held constant.
Mi
Cu ;
V:a
9
These three factors are inter
temperature is the easiest one to me to study the change from ductile t perature, provided that stress cona
00
-
100
0
100
Temperature, °C
Fig. 13-5 Variation of yield and tensile strengths of iron, copper, and nickel with temperature. (From Behavior of "
Metals at Low Temperatures," American Society for Metals, Metals Park, Ohio, 1953.)
:;
i
i r
These conditior
or Izod notched-bar impact test (see degree of correlation with room-ter ness behavior is shown by the note come so common that most of thi
obtained by it. 13-8 Effect of Temperature on Notched-t many temperatures, a plot of energ;
METALS AT HIGH AND LOW TEMPERATURES
h and modulus of elasticity of all metals of yield and tensile strengths of iron, 3 is shown in Fig 13-5. ure on ductility metals fall into two dis-v
m
557
lempefOlure, 'f 328
-
-
I48
32
2I2
.
lOOh
,
(uctile at low temperatures and thosej
n of the amount of ductility
,
i obtained from a study
i
Mi
or plastic
of the fracture I
e is typical of a ductile material which formation when tested in tension. A
Cu o
y 60\-
h no evidence of plastic deformation: :>
S
;e-centered
cubic metals fracture only mtinuous decrease in ductility Metal8= til by shear- at room temperature, but
/ /
-
.
I
/
.
lode of fracture changes from
4C
/
shear
/
/
hange in fracture often appears as a,.
;
20
)f temperature on the ductility of iron,; -
13-6. The tensile properties
::
/ /
of soma
\/temperatures are given in Table 13-3. bers are often sudden and unexpected ,5
.
-
I
0 200
100
0
-
100
Temperature, "C
ittle failure of the part Great interest .
Fig 13 6 Variation of ductility of iron copper, and nickel .with temperature (From "Behavior of Metals at Low Tem,
.
peratures," American Society for Metals, Metals Park, Ohio, 1953.)
in this problem developed during World War II when a number of welded ships failed in a brittle manner with almost explosive rapidity. In some cases, the ship was split in two. The steel used for ship plate was ductile at normal temperatures, yet the failure was of a brittle nature with little plastic deformation.
The tendency of steel to fail in a brittle manner is increased by stress concentration, increased speed of load application, and decrease of temperature. These three factors are interrelated, and the effect of lowering the temperature is the easiest one to measure quantitatively. It is often possible to study the change from ductile to brittle fracture with decreasing temperature, provided that stress concentration and speed of load applicatior are held constant. These conditions are satisfied in the ordinary Charp or Izod notched-bar impact test (see Sec. 1 -33). A rough approach to sorm
\ :
i
degree of correlation with room-temperature and low-temperature tough ness behavior is shown by the notched-bar impact test. This test has be come so common that most of the information on toughness has beer obtained by it.
f
Is
,
|;
13-8 Effect of Temperature on Notched-bar Test If tests on steel are made a many temperatures, a plot of energy absorbed vs. temperature will usually
11
Ill
m
....
TABLE 13-3
31
Tensile Properties of Some Steels and Nonferrous Materials at Low Temperature*
O a
MATERIAL AND
COMPOSITION Low-carbon steel, 0
13-0.1* C
YIELD
CONDITION
As-rolled As-rolled
.
TEMP, 0F
As-rolled
-
Annealed Annealed
NI steel, 0.13 C, 5 13 Ni, 0.41 Mn,
Oil-quenched
0 19 Cr, 0.15 SI
tempered
.
.
and
PSI
PSI
54,700
85
67,700
292
Annealed
70 -
-
-
-
42 700 ,
296
423
c
STRENGTH
70 -
TENSILE
STRENGTH,
155 000 ,
,
ELONGATION,
REDUCTION
% IN 2 IN.
IN AREA, %
o H
o 2
66,300
29.7
71.8
H O T3
80,700
33.6
70.3
121,300
26.5
55
co
45,700
27.5
77.5
o
137,000
75 25
m
155,000
X < -
>
.
03 .
.
68
103.000
25
74
242
153,000
25
57
319
175,000
21
50
>
3)
CD
1200oF
Nl-Cr-Mo steel,
Oil-quenched
0 33 C, 0.67 Cr,
and
2 45 Ni, 0.64 Mo
tempered
.
.
11850F
V
Fe-Ni alloy,
Water-
0 16 C, 35.8 Ni,
quenched
.
.
<
-
70
137,700
152,000
14
6
141,000
154,500
15.6
64 ,
90
145,000
163,000
15.6
62
292
183,500
201,500
17
63
70
52,400
81,100
32
58
423
127,000
144,000
20
60
78
-
-
65
0 86 Mn .
Commercially pure NI
24,600
65,500
42
27,600
76,400
43
73
292
27,600
97,000
53
74
-
-
Commercially
70 112
Annealed
75
19,700
23,500
16
112
21,350
24.700
18
Hard-rolled
pure Al
-
-.„
TABLE 13 3
COMPOSITION
i
YIELD CONDITION
TEMP, °F
STRENGTH, PSI
Al alloy 2017,
Solution-
4 Cu, 0.5 Mn,
treated
0
,,
-
iitii
(continued)
MATERIAL AND
I
..
.. .
1
-
TENSILE
STRENGTH, PSI
ELONGATION,
REDUCTION
% IN 2 IN.
IN AREA %
75
45,500
68,000
15
112
46,500
70 000
16
14
,
,
5 Mg
.
Al alloy 2052, 2 5 Mg, 0.25 Cr Pure copper
Hard-rolled Annealed -
38,600
43,500
39,200
45,600
18.5
75
8 600
31,400
48
76
112
10,100
38,500
47
74
58
77
15,000
50,800
75
125,000
187,000
2 6
5
112
147,000
202,000
4
5
OQO
1
.51 a nnn
-
Water-
quenched and anprl
-
-
,
292
-
Cu-Be, 2.56 Be
75
112
-
.
nnn
.
3
s m
'
' -
-
w .
,
~
iV-,-
-
,-,1
1
-.
1200oF
Ni-Cr-Mo steel
Oil-quenched
0 33 C, 0.67 Cr 2 45 Ni 0.64 Mo
and
,
,
.
70
152 000
4
141,000
154,500
15.8
145,000
54
163 000
15.6
183 500
62
201.500
17
53
52,400
81,100
32
127,000
144.000
20
,
tempered
,
6
90
-
,
1185=F
292
,
Fe-Ni alloy
Water-
,
0 16 C 35.8 Ni 0 86 Mn
quenched
Commercially
Annealed
.
,
,
-
22
<
-
137 700 ,
.
65
58 60
,
pure Ni
-
70
24,600
65.500
:i2
42
27,600
78
76 400
43
73 74
,
292
27,600
97,000
53
75
19 700
23.50C
16
21,350
24,700
18
-
Commercially
Hard-rolled
,
pure Al
112
-
mMM hMl!ilMI>ll|lli(lr
TABLE 13 3
ri:iMi
mum
mm.
(continued)
MATERIAL AND
CONDITION
TEMP, =F
COMPOSITION
AI alloy 2017,
Solution-
4 Cu, 0.5 Mn,
treated
0
-:t|i|iit(ii
-
YIELD
TENSILE
STRENGTH,
STRENGTH,
PSI
PSI
ELONGATION,
REDUCTION
% IN 2 IN.
IN AREA. %
75
45,500
68,000
15
112
46,500
70,000
16
14
5 Mg
.
Al alloy 2052, 2 5 Mg, 0.25 Cr
Hard-rolled
Pure copper
Annealed
-
75
38,600
43,500
112
39,200
45,600
18.5
75
8 600
31,400
48
76
112
10,100
38,500
47
74
292
15 000
50,800
58
77
.
-
quenched and aged
75
125,000
187,000
2 6
5
147,000
202,000
4
5
292
155,000
214,000
3
6
75
28,200
51,100
49
77
112
27,300
57,100
60
79
292
29,600
73,500
75
73
77
28 000
39,000
35
110
30,900
52,000
3
-
-
Annealed
Cu-Zn, 30.5 Zn, 0 10 Fe
,
112
Water-
Cu-Be, 2.56 Be
,
-
.
2 >
.
Extruded
Magnesium alloy Ml, 1 Mn
-
Magnesium alloy
Cast, heat-
AZ63, 6 Al, 3 Zn
treated, and
Ni alloy Monel,
Annealed
-
,
> z
77
21,500
38,900
45
58
110
24,400
35,400
25
3 4
75
20,900
70,800
41
75
112
27,100
85,300
40
74
292
29,600
113,000
51
72
.
.
.
.
aged
-
r
o
S tn
28.86 Cu,
-
0 28 Mn
-
.
-
o
m 3)
> c 33
Metals Handbook, 1948 ed., Amer:can Society for Metals. Metals Park. Ohio "
Compiled
5
.
from data in
-
m
en
i
.
560
INTRODUCTION TO PHYSICAL METALLURGY
MET,
show a temperature range in which the impact values drop sharply as the temperature is lowered. At the same time, the mode of fracture changes from a predominantly fibrous-shear type to a crystajline-cleavage type. This is shown graphically in Fig. 13-7. Values in the transition range are often erratic since slight changes in conditions will affect the values. The temperature at which some specified level of energy absorption or fracture appears is defined as the transition temperature. In ASTM specifications it is defined as the temperature at which specimens show a fracture of 50 percent shear and 50 percent cleavage. The lower the transition j temperature the better is the steel able to resist the embrittling effect of stress concentration, high loading rate, or low temperature. A study of available data for iron and steel indicates that their low-temperature behavior is affected by two classes of variables, namely, metallurgical factors and mechanical factors.
250
200
I
Percent C 0 01 .
CD
150
CD
o ; .
T 100
50
0
300
-
-200
-100
&
13-9 Metallurgical Factors The important interrelated metallurgical factors affecting the low-temperature behavior of iron and steel are composition, debxidation, heat treatment and microstructure, surface condition, and
oji
31
0
100
2
Testing temperature
Fig. 13-8
Effect of carbon on the shape of the transitioi 1961 ed., p. 227
curve. (From "Metals Handbook
"
,
,
I American Society for Metals, Metals Park, Ohio.)
grain size. Increasing carbon content decreases the notched impact strength at room temperature and raises the transition temperature (Fig. 13-8). The physical form of the carbon is also important. When the cementite is spheroidized, it seems to be less harmful to low-temperature properties. A manganese content of up to 1.5 percent lowers the transition temperature, as shown in Fig. 13-9, but does not change the shape of the transition
The use of aluminum in addition t
seems to have a beneficial effect c
carbon steels. The room-temperatur the transition temperature is lowere creased up to about 0.10 percent
.
.
usually found in aluminum-treated <
largely to the improved toughness
.
curve.
Silicon, in amounts up to 0.30 percent used to deoxidize steels, lowers transition temperature and improves notch toughness because a cleaner
120
steel and a more uniform ferritic grain are produced. Larger amounts have
100
the reverse effect, and the presence of 4 percent silicon results in a brittle structure even at room temperature.
1=
r
80
5
<
Percent Mn 1 55 .
60 1 01 .
Tough
0 39 .
I
0 30
40
.
Transition zone
20 O1
Brittle
ft
0 -
Temperature
150
-100
-50
0 50 100 Testing temperature
c
,
Fig. 13-7 Typical curve of impact strength vs. testing temperature for a ferritic steel, showing transition temperature zone in which erratic values may be expected. (The International Nickel Company.)
i
13-9 Effect of manganese on Charpy V-notch value; |flf a 0.30 percent carbon steel. (From "Metals Handbook .
r''
ed., p. 227, American Society for Metals, Metals Par Ohio.)
V
1 METALS AT HIGH AND LOVV TEMPERATURES
/hich the impact values drop sharply as thl| -
561
2'jO
3 same time, the mode of fracture change* i
-
_
_
vgZhear type to a crystalline-cleavage type. Thisf, .7
.
s
Values in the transition range are often "
conditions will affect the values.
Percent C
200
0 11
0 01
,
t:
' |
ime specified level of energy absorption orl
0 22 ,
the transition temperature. In ASTM sped- j hperature at which specimens show afrac-.
100
0 31 .
'
| percent cleavage. The lower the transition iJR
0 43 ,
[teel able to resist the embrittling effect of ft' Jing rate or low temperature. ,
0 53 .
50
3 63 .
3 67
A study of
,
«]el indicates that their low-temperature be-
0
s of variables,-namely, metallurgical factors ,ff
-
300
-200
0
-100
100
200
300
400
500
,
Testing temperature, 0F
m
13-8 Effect of carbon on the shape of the tnansmon
jrtant interrelated metallurgical factors af-
savior of iron and steel are composition. ir.'i'hd microstructure, surface condition and Mi
'
-
,
The use of aluminum in addition to silicon for the deoxidation of steel seems to have a beneficial effect on the notch toughness of mediumcarbon steels. The room-temperature impact resistance is improved and the transition temperature is lowered as the amount of aluminum is in-
lecreases thfe notched impact strength at '
he transition temperature (Fig. 13-8). The ')so important. When the cementite is sphe- il
) 1.5 percent lowers the transition temperaifl
creased up to about 0.10 percent The relatively fine ferritic grain size usually found in aluminum treated cast and wrought steels contributes
does not change the shape of the transition
largely to the improved toughness.
rmful to low-temperature properties.
i$KSt
0 percent used to deoxidize steels, lowers
.
-
r I20i-
roves notch toughness because a cleaner ; grain are produced. Larger amounts have
bnce of 4 percent silicon results in a brittle : ture.
V
.
80H
5
Percent 1 55
i
,
60H
101
O
0 39 ,
I
0 30 ,
40 5
20
i
0 150 '
-
sting
) temper-
ed. (The
.
a 0,30 percent carbon steel. (From Metals Handbook," "
,
i
0 50 100 Testing temperature, 0F
11961 ed, p, 227, American Society for Metals, Metals Park, Ohio)
J
- 50
Fig, 13-9 Effect of manganese on Charpy V-notch values
'
1
100
-
150
200
250
"
V 1 ?? -
562
INTRODUCTION TO PHYSICAL METALLURGY
MET/
For a particular type of steel and strength level, fine-grained steels have higher notch toughness than coarse-grained steels. The transition temperature is lowered as the grain size decreases. This is illustrated by Fig. j 13-10. The fine-grained condition is usually due to a deoxidation practice
Temperoiure
'
,
200
150
-
-
-
100
that uses silicon, aluminum, or vanadium.
Nickel is the most effective alloying element for increasing resistance to :
low-temperature embrittlement in steel and is one of the few alloying ele- . ments which improve the low-temperature ductility of iron. Nickel additions \ to steel increase room-temperature toughness, lower the transition tem- J
50
40
perature, and widen the transition-temperature range. The effects of \. variation in nickel content and temperature on the Charpy keyhole impact values of low-carbon steels are shown in Fig. 13-11. The curves indicate
5
13% Ni -
/ /
that the transition temperature for 1020 killed steel is slightly below 0oF. \
Bt% Ni
/
i
2 % Ni
0 15 %C
The 2 percent nickel steel retains considerable toughness down to -100°F.
0
.
The 31 and 5 percent nickel steels seem to be best between -100 and |;
/
5% Ni
2pooF. The SVa percent nickel steel shows only a gradual decrease in ,:
/3 %Ni /
-
!0
toughness with decrease in temperature. The 13 percent nickel steel shows j [.
7 /
no transition temperature, and its room-temperature toughness remains at L 0
almost the same value over the entire test range.
-300
-200
100
-
Temperature °l Molybdenum, vanadium, and titanium have a similar effect on notch; | toughness. Small amounts tend to raise the transition temperature, but as-i jfo-13-11 Ef'ect of nickel content on the resistance to ,
the amount is increased, transition temperature is lowered.
| t
Z
The best microstructure for low-temperature toughness is that of tem p;-, .
pered martensite. This structure gives est transition temperature compared \ cific steel (Fig 13-12). The notch touc
IOC
.
increasing amounts of bainite Retain on transition temperature Carburized or nitrided surfaces ten
Grain size 5-o
.
i
t
75
.
carbon and alloy steels This is due to that resist plastic bending under shock tic bending and may slightly increase r duce fatigue strength Notched-bar impact values of some low temperatures are given in Table 1c .
OJ
o
50
.
5
25
3
10
1 -
50
. 0
+50
100
150
Testing temperature, °F
Fig. 13-10 Effect of grain size on notch toughness of 1030 steel. (From Metals Handbook," 1961 ed., p. 236, American Society for Metals, Metal? Park, Ohio.) "
;:.
-
200
i
Mechanical Factors The mechanical pact results are the stress concentrati
concentration is determined by the sha radius of the notch increases the stress
duce brittle behavior at higher tempera by the striking velocity of the pendulu sensitive to the striking velocity when
1
i
LURQY
METALS AT HIGH AND LOW TEMPERATURES
563
1 h
te\ and strength level fine-grained steels hev6s
Temperature, "C
,
.
.
I
n coarse-grained steels. The transition te
: ::;; .
.
-
"
150
200
-
-
100
-50
0
70
grain size decreases. This is illustrated by p|| dition is usually due to a deoxidation practk*
6C
or vanadium.
) alloying element for increasing resistancetb|
'
.
v
int in steel and is one of the few alloying ele-| temperature ductility of iron. Nickel additional
50
'
rature toughness
,
5
lower the transition tern-? The effects of
10
'
msition-temperature range
.
/
'
gjgd temperature on the Charpy keyhole impacl l
13% Ni
re shown in Fig. 13-11. The curves indicate! | re for 1020 killed steel is slightly below 0oF | |
3C
3 2 7= Ni
.
ains considerable toughness down to
1 steels seem to be best between - 100 and| | skel steel shows only a gradual decrease in I mperature. The 13 percent nickel steel showi| | ;
0
.
0 207oC
I5%C
.
/
J%N/ /3i%Ni 2
:
i
0%Ni
2 7oNi
s
IGO'F.i I 10
7
;
'
iMjd its room-temperature toughness remains at| ie entire test range.
'
;
| I
.
id titanium have .a similar effect on notchj f ind tO raise the transition temperature but as
sition temperature is lowered. r
j
,
0
-300
-200
-100
Temperature,
+ 100
0f
h 13-11 Effect of nickel content on the resistance to
1'3
TTT em i"l!™'lt 0, "ormaMzed low-carbon
Seels (keyhole notch). (The International Nickel Company.) ,
r low-temperature toughness is that of tern- 3 ;
I )
pered martensite. This structure gives the highest toughness and the lowest transition temperature compared with other microstructures of a specific steel (Fig. 13-12). The notch toughness of martensite decreases with increasing amounts of bainite. Retained austenlte has only a slight effect on transition temperature.
Carburized or nitrided surfaces tend to lower the notch toughness of
carbon and alloy steels. This is due to the hard, less ductile surface layers that resist plastic bending under shock loads. Decarburization favors plastic bending and may slightly increase notch toughness; however, it will reduce fatigue strength. Notched-bar impact values of some steels and nonferrous materials at low temperatures are given in Table 13-4. 1310 Mechanical Factors
The mechanical factors that affect notched-bar im-
pact results are the stress concentration and the strain rate. : :J
100
150
.
:: : .
-
.
temperature, 0F v .
less of 1030 .
p6, American
200
The stress
concentration is determined by the sharpness of the notch. Decreasing the radius of the notch increases the stress concentration, which tends to produce brittle behavior at higher temperatures. The strain rate is determined by the striking velocity of the pendulum, and Ihe energy absorbed is very sensitive to the striking velocity when the steel is near the transition tern-
-
L
_..-.
564
METAI
INTRODUCTION TO PHYSICAL METALLURGY
120 "
a
o
.
5
o
o
1-
1-
1-
CM CD CM
I
I
I
I
I
O)
o
o
o
o
o
o
o
O O)
I
100 -
i CO
*
CM
i
Manensue
i
80C\J
o
o
o
I
I
I
o
CO
in
"f
I
o i-
o i-
CO
CO
I
I
CO
CO
CM
O
00
GO i-
Ol OJ
I
I
I
I
CM Ol CM
;
<
60-
> i
a
H £0
o
OJ OD
W ai t-
O
CM CM CM g g CM CM
CM O O
CM
TTTTTTT
TTT
T
CD i-
-
o <
-
CO
CO
Pe(
i
CO
.
S
5 40-
tr
COO
<
O
O
m
o
20-
UJ
i
o H
i
o
CD
CM
CO
CM
CM
CO
O
CO
h- O CM CM
CO CO t- UO
CO tJ
CD
lO
CM
-
CD
-
z -
1
0 250
200
-
-
150
-100
-50
Testing temperature
5
,
°F
.
0)
o
2
[ Fig. 13-12 ;
ra .
I
t 3
0)
i Ol CO
CO CO CM Ot-CO
CD
»0.17 percent carbon, Cr-Mo steel. (From "Metals Hand1961 ed., p. 235, American Society for Metals
Q) CD
CO CJ) LO CO CO CM CM CM i- i- LO tJ -
CD CO i-
ibook
o
CD
O)
'
3
(
-
0)
«>
UJ CD r
I
d Z
D-
CO W o 5 z tr o UJ UJ 1a: I-
I- CO Q
_
O
o o
o
o o
cm O)
i-
O
i-
CM
o CO
o o o o
o o
o o
cm in in o co ai t- i- r-" "
in in
t
-
in co
co
o o
o
O)
CO
7-
O)" co
o
in
co
o o
o
o
co
8 i
3
O
«
„
o
ra
o
ra a)
o
z
ra
D
o
Z
n
o o
i I 01
a;
u o
Z
(A
g
CD
o
d o
o
00 CM
.= "
o
ai
a
Q)
c
5
cn
O o
-
Ol
CD
d o
-
a z <
m mm
h-
|
o CO o
<
OJ
E »
Ol
at
aj
CD
«
«
<
(D
5
88
CO
CE
UJ
'
5
lar kind of service.
3
_
0) CO
3
m CO
sign and serve only to compare differe Any attempt to use these values for stri
d
cy
d Q
.
I
It must be emphasized that notchedconditions under which the test was rui
cant for design only when correlated wil
T
o
o
,
strain rates must be done with extremi
U)
E
i
perature. High striking velocity has tl temperature and tends to exaggerate t To summarize, the best notch-tough lecting a fine-grained low-carbon, fi has been quenched to a fully martens desired hardness level.
ra
a> o
Z
:
_
*
_
co
o o
E
a
,
»
2
„
-
"
,
Metals Park, Ohio.)
5
£ o
to
Effect of microstructure on notch toughness c
S 6
"
co
S N
5 CO "
.
z
" 1 §" & O
-
[5
>i
>i
o
o
ra
ra
>1
a
1 Ol Ol
SPUESTIONS 13-1
o
o
o d
1 r»
E
CN
" - CD
CO
6
.
c <
d 5 CD CD
.
CO z
o
z Z „ £o5
ra
z
t
What factors should be considered »
as compared with testing at room temperatu 13-2 Define creep. Why is this properly impc 13-3 Give a specific application where cr design.
I URGY
METALS AT HIGH AND LOW TEMPERATURES
.
&6S
120 CM
OJ
I
100
fllOl loiistlc
Murlensile
? ? CO CO
5 i-
t-
III CO
S
CM
II
I
I
? "
s
s
80 Bainile
CM
50
CO
Pearlite
S
r S S
O
C\JCJCSJggc\JC\J
CMOO
TTT7TTT
7 1
05
d 40
20 cd
in o>
-
0 250
-
200
-
150
5
2 CO .
ooooo
ooo
oj un in
t- o o
o
CO Oi t- t- h>»
*- O) CO
t
tn
.
-uocococo
o
-
co
o
co
8 73
"
D
CD
100
i t .i
i
perature.
High striking velocity has the same effect as lowering the test
temperature and tends to exaggerate brittle behavior. To summarize, the best notch-toughness properties are obtained by selecting a fine-grained, loW-carbon fully-killed nickel-alloy steel which has been quenched to a fully martensitic structure and tempered to the ,
desired hardness level.
il iiiliilpil:ro
ta
It must be emphasized that notched-bar impact values apply only to the
_
.
50
j ilOk," 1961 ed., p. 235, American Society for Metals, jetals Park, Ohio.) .
o
0 0F
|g, 13-12 Effect of miorostruoture on notch toughness of i| (0.17 percent carbon, Cr-Mo steel. (From "Metals Hand-
.
3
-50 ,
o
o>
-100
Testing temperature
9
.
o
0
conditions under which the test was run. The results cannot be used in de1
ra
sign and serve only to compare different factors under those conditions. Any attempt to use these values for structures of different size or different
<
2
o
d d
2
strain rates must be done with extreme care. The values become significant for design only when correlated with a particular structure in a particu-
(0
'
coo 5
lar kind of service.
C\J
d
CD o
d
0)
N
5
3
CD
ESTIONS
CO
CO CM
E
a
E IJ-20<
0
.
OSS 2
2
13-1
What factors should be considered when testing at elevated temperatures
fis cornpnrod with testing at room tomporaluro?
o
5
3
13-2 Define creep. Why is this properly important for high-temperature application? 13-3 Give a specific application where creep properties would be important in design.
5
561)
INTRODUCTION TO PHYSICAL METALLURGY
13-4 13-5
Draw a typical creep curve and explain the stages of creep. What metallurgical factors affect the creep characteristics of metals?
13-6 What are the limitations on the use of stress-rupture data? 13-7 Which alloying elements are most effective in raising the creep strength oli I steels? 1 I ,
13-8 What is the effect of low temperatures on the mechanical properties of metalsf f 13-9 What is meant by transition temperature How is it measured? < il
i
13-10
What factors influence brittle failure?
13-11
What are the limitations on the use of notched-bar data for the performancsi %:
.
i
WE ME
of actual parts? 13-12 Differentiate between a shear fracture and a cleavage fracture. 13-13 What alloying element is most effective in improving the low-temperatursM toughness of steel?
;
t
REFERENCES
American Society for Metals: "Behavior of Metals at Low Temperatures,'
Metals
14-1
Park, Ohio, 1953.
m
:
"
:
"
:
"
Creep and Recovery," Metals Park, Ohio, 1957. High Temperature Properties of Metals," Metals Park, Ohio, 1961. Metals Handbook," 1948 and 1961 editions. Metals Park, Ohio.
American Society for Testing Materials: "Evaluation of Metallic Materials in Desii
be defined as unintentional deterio ment. It may be considered essenti one of the most destructive influent the importance of wear resistance m
for Low-temperature Service," Special Technical Publication no. 302, Philadelphia 1962. 4 Clark, F. H.: "Metals at High Temperatures," Van Nostrand Reinhold Company, Neft
York, 1950.
|
The displacement and detachmen surface may be caused by contact
Clauss, F. J.: "Engineer's Guide to High-temperature Materials," Addison-Weslejf: Publishing Company, Inc., Reading, Mass., 1969.
j
Conway, S. B.: "Numerical Methods for Creep and Rupture," Gordon & Breach Sci;
ence Publishers, New York, 1967.
metallic wear) (2) a metallic or a n ,
||
Dorn, J. E. (ed.): "Mechanical Behavior of Materials at Elevated Temperatures,'
McGraw-Hill Book Company, New York, 19i31.
moving liquids or gases (erosion)
1
.
|
New York, 1959.
.1
Wear involving a single type is ra
Parker, E. R.: "Brittle Behavior of Engineering Structures," John Wiley & Sons, lr
and adhesive wear occur
New York, 1957.
.
Each for
conditions, including environment mating parts lubricant, temperature of foreign particles and compositi
Savitsky, E. M.: "The Influence of Temperature on the Mechanical Properties of Me
,
als and Alloys, Stanford University Press, Stanford, Calif., 1961. "
,
Seigle, L, and R. M. Brick: "Mechanical Properties of Metals at Low Temperatun
,
A Survey, Trans. ASM., vol. 40, p. 813, 1948.
1
parts involved. Since in most machi
Sully, A. H.: "Metallic Creep," Interscience Publishers, Inc., New York, 1949. Wigley, D. A.: "Mechanical Properties of Materials at Low Temperatures," Pleiv
avoided completely even with the be to use a hard metal and a relatively s is used (as in a bearing) for the part i
Press, New York, 1971.
m
Er
form of corrosion. The above three wear under rolling friction or slidin whether lubrication can or cannot be
Heheman and Ault (eds.): "High Temperature Materials," John Wiley & Sons,'
International Nickel Company: "Nickel Alloy Steels," 2d ed.. New York, 1949.
Introduction The quality of most me of their surfaces and on surface deti oration is also important in engineer! limiting the life and the performanct
m-2 Mechanism of Wear
1
In adhesive wea
and scuffing tiny projections produc ,
with the relative motion of contactii
further movement. If the driving fore
i
>
4 1
'
r
11 i j
f
the interlocked particles are deformec
;: .
LURGY
'
..
1 !
/e and explain the stages of creep is affect the creep characteristics of metals?
i
.
.
iV Apn the use of stress-rupture data?
are most effective in raising the creep strength-
jmperatureson the mechanical properties of metal ?n temperature? How is It measured?
Srittle failure?
Ion the use of notched-bar data for the perfbrmafV
WEAR OF METALS
? shear fracture and a cleavage fracture s most effective in improving the low-temperatur | .
"
II Behavior of Metals at Low Temperatures
"
,
/letals Park
Metal!
I
Ohio, 1957.
,
dirties of Metals
,
"
J i
Metals Park, Ohio, 1961.
|| |
J0B and 1961 editions. Metals Park, Ohio.
i |
:
iterials: "Evaluation of Metallic Materials in De8igJ| | .
.
Special Technical Publication no. 302, Philadelphttji "
iperatures, Van Nostrand Reinhold Company, Ni
1 |.
) to High-temperature Materials," Addison-Westej ding, Mass., 1969.
)ds for Creep and Rupture,
Gordon & Breach
1
Introduction
The quality of most metal products depends on the condit'on
of their surfaces and on surface deterioration due to use. Surface deteri-
oration is also important in engineering practice; it is often the major factor limiting the life and the performance of machine components. Wear may be defined as unintentional deterioration resulting from use or environment. It may be considered essentially a surface phenomenon. Wear is one of the most destructive influences to which metals are exposed, and the importance of wear resistance needs no amplification. The displacement and detachment of metallic particies from...a.metalljc surface may be caused by contact with .(l) another metal (adhesive or
metallic wear), (2) a metallic or a nonmetallic abrasive (abrasion), or (3) moving liquids or gases (erosion). Erosion is usually accompanied by some
17.
behavior of Materials at Elevated Temperatures*!
lew York, 1961.
"
form of corrosion. The above three types of wear may be subdivided into wear under rolling friction or sliding friction and, further, according to
;3|
Temperature Materials," John Wiley & Sons, Inc '5
whether lubrication can or cannot be used. :
jlickel Alloy Steels," 2d ed., New York, 1949 f Engineering Structures
"
Wear involving a single type is rare, and in most cases both abrasive
John Wiley & Sons, lnc.,1
i
and adhesive wear occur. Each form of wear is affected by a variety of conditions, including environment, type of loading, relative speeds of
|
maling parts, lubricant, temperature, hardness, surface finish, presence
lanical Properties of Metals at Low Temperatures;";
of foreign particles, and composition and compatibility of the mating parts involved. Since in most machinery applications wear can raroly bo avoided completely even with the best lubrication, it is common practice to use a hard metal and a relatively soft one together. The softer material
,
Temperature on the Mechanical Properties of Melt; rsity Press p
.
,
Stanford, Calif., 1961.
813, 1948.
orscionco Publishors
.
Inc., Now York, 1949.
srties of Materials at Low Temperatures
,
|
| Plenum
s
is used (as in a bearing) for the part which is most economical to replace. In adhesive wear also called scoring, galling, seizing
14-2 Mechanism of Wear
,
and scuffing, tiny projections produce friction by mechanical interference with the relative motion of contacting surfaces increasing resistance to further movement. If the driving force is sufficient to maintain movement ,
,
the interlocked particles are deformed. If they are of a brittle material, they
_.-
" ,
568
.
.-
i-l--*
.
.V
INTRODUCTION TO PHYSICAL METALLURGY
r
i .
adhesive wear on the surface of an in(
IP
under heavy load. Lubrication with resulting in metal-to-metal wear. Abrasive wear occurs when hard
across a surface, or when a hard s
r
The abrading particles from the har
the softer material. These hard pa metal and cause the tearing off of n the main journal bearing of a cranks Hard dirt particles with sharp project cutting and scratching the journal I obvious solution is to Improve the ef
1
The ease with which the deformed
the toughness. Therefore hardness ,
that influence adhesive wear also de
Fig. 14-1 Surface of a truck drive unit showing galling caused by inadequate lubrication. (From R. D. Barer and B F. Peters. 'Why Met&ls Fail," Gordon and Breach Science
,
.
14-3
Publishers, New York, 1970.)
factors, hardness is probably the moi Factors Influencing Wear Lubricatio to wear resistance particularly in ad ,
tion, a sufficiently thick lubricating
may be torn off. This leads to the conclusion that wear resistance will be improved by preventing metal-to-metal contact and by increasing the hard-
contact, and metallic wear is reduced
ness to resist initial indentation, increasing the toughness to-resist tearing
out of metallic particles, and increasing the surface smoothness to eliminate the projections. Figure 14'1 shows galling or gouging of the surface of a truck drive unit. Investigation indicated that the galling was due to in-
r
adequate lubrication because of leaky oil seals. Figure 14-2 shows severe
iiniiuiiiiiiiiiiiiiuiiiiiiiiiii .
it \
i
m r
wmm
i
fig. 14.3 The main journal bearing of a crankshaft was tljadly scratched by dirt particles which contaminated the
Fig. 14-2 Adhesive wear on the pinion gear of a tractor due to inadequate lubrication. (Courtesy of D. J. Wulpi,
Wiricant
.
mpany.)
International Harvester Company.)
U
..
I
I .
1
.
:
1 m /
(Courtesy of D. J. Wulpi, International Harvest(
UPIOY
WEAR OF MEIALS
569
f
adhesive wear on the surface of an induction-hardened gear tooth operated under heavy load. Lubrication with a light engine oil was inadequate ,
resulting in metal-to-metal wear.
Abrasive wear occurs when hard particles slide or roll under pressure \
i
across a surface, or when a hard surface rubs across another surface.
The abrading particles from the harder object tend to scratch or gouge the softer material. These hard particles may also penetrate the softer metal and cause the tearing off of metallic particles.
Figure 14-3 shows
the main journal bearing of a crankshaft which was damaged by dirty oil. Hard dirt particles with sharp projections broke through the lubricant film ,
'
cutting and scratching the journal bearing s relatively soft surface. obvious solution is to improve the efficiency of the oil filter.
An
The ease with which the deformed metal may be torn;off depends upon the toughness. Therefore, hardness and toughness, the same properties 3 galling
that influence adhesive wear, also determine abrasive wear.
Barer and
,
each Science
:{.;) the conclusion that wear resistance will b6 '
'
y
to-metal contact dnd by increasing the hard-*
n, increasing the toughness to-resist tearinffi increasing the surface smoothness to elimi*
[HHlii
AA shows galling or gouging of the surfa itlon indicated that the galling was due to ifL of leaky oil seals. Figure 14-2 shows severii
i
UK
1
>
1
1
14-3 The main Journal bearing of a crankshaft was
.
a tractor J
.
Wulpi,
Of these two
hardness is probably the more important one. 3 Factors Influencing Wear Lubrication is an important contributing factor to wear resistance, particularly in adhesive wear. In "thick-film" lubrication, a sufficiently thick lubricating film completely eliminates metallic contact, and metallic wear is reduced to a negligible amount. This is, howfactors
idly scratched by dirt particles which contaminated the cant. (Courtesy of D. J. Wulpi International Harvester ,
flipany.)
;
570 INTRODUCriON TO PHYSICAL METALLURGY
ever, the ideal condition, and more frequently
"
boundary lubrication"
14-4 Methods of Testing for Wear Resista nomenon wear resistance is represe other engineering properties It is (
occurs. This is the condition of intermittent metallic contact that occurs
when the oil film cannot be continuously maintained.
,
Under boundary
.
conditions, the amount of wear depends upon speed, pressure, nature of
wear test is not feasible
the mating surfaces, and efficiency of'the residual oil film. In many cases,' however, lubrication is impractical or is not wanted, as in braking. Although actual melting of the metal occurs only in rare instances, the
designed to simulate actual service proved reproducibility, should be abli
sideration, and most important
effect of heat produced by dry wear can reduce wear resistance in several
service data.
ways. It may temper hardened structures, cause phase changes that in I
14-5 Protection against Wear
.
Therefore
,
,
she
Many matt
crease hardness and brittleness, and decrease mechanical properties, and
protection against wear. The selectio
it accelerates corrosion reactions.
requires a thorough analysis of the edge of applicability and limitations c and data concerning the cost involve
The dominant frictional factor for metallic materials is believed to be
welding. Atoms of the same or crystallographically similar metals have very strong forces of cohesion. When two clean surfaces of the same metal actually touch each other, they will weld together because of atomic at-; traction. If, by friction, sufficient pressure is applied to break through;' anyl residual separating material such as oil, dirt, or adsorbed moisture,;
able for comparisons imposes a neei or technician who selects materials ti
Various techniques for providing follows:
and the surfaces are in sufficient contact to have elastic or plasitic defoK V
.
Electroplating
mation occur, then seizing or welding takes place. The softening of metals; by high temperatures increases the ease of plastic deformation and facili-' tates welding. Seizing may cause complete stoppage, or if relative motion: is not prevented, pieces of the opposite face may be pulled out. The result ant projection then may cause scoring, galling, and excessive local wear.
Anodizing Diffusion
Metal spraying Hard facing
Many methods may be used to minimize the danger of seizing. One is to,
use thin layers of hard surfacing material. The use of at least one metaf that forms some sort of lubricating film or thin, tightly adherent oxide,?
1
Selective heat treatment
i14-6 Electroplating The wear resistance c electroplating a harder metal on its si
sulfide, or phosphide coating is frequently helpful. Aluminum oxide is| i
very effective in preventing welding. For parts that operate under such|
on base materials are Chromium nicl1
high pressures that elastic deformation permits intimate contact, the best|
been used to reduce the wear of lead
preventive method is a lubricant that combines with the metal surface ti):|
Two types of chromium plating u
form a "corrosion" product of sufficient strength to keep the surfaces;
chromium and porous chromium. TI
separated. The use of materials of high elastic limit will minimize seizure
as that used for decorative purposes t to 0.010 in. Porous chromium plate h
due to intimate contact produced by plastic deformation.
Impact is a factor in wear, since the suddenly applied load may causel plastic flow Eind a change in shape. Proper design should provide a surfai compressive yield strength above the compressive stress produced b
impact and Sufficient support so that subsurface flow does not occur. Fatigue failure is included in a discussion of wear since it is a gradual! J
,
pits or channels to hold lubricants.
the specially prepaTed surface if is i ,
chromium plate. The hardness of ch 1050 Vickers. Another factorLCQnirib low coefficient of friction of chromii
deterioration due to use. Proper design to eliminate stress concentrations
in the cyTin
at notches aind sharp angles will increase fatigue strength. Since fatigue
galling is another useful property of
failures are always due to tensile stress, residual compressive stress al
mium-plated steel parts may be asser without seizing or galling. The higl is helpful in reducing wear under con
the surface will provide additional protection. This may be accomplish! by case hardening, such as carburizing, and by shot peening.
l
3
r :
J
i: i
.
'
'
er ar
ton rmgsjjL
1
WEAR OF METALS
571
i
md more frequently "boundary lubricatioif
.4 Methods of Testing for Wear Resistance
i of intermittent metallic contact that occur continuously maintained Under bound sar depends upon speed pressure, nature c ciency of the residual oil film In many cai
I
,
.
the metal occurs only in rare instances
:;|| ,
tr
y wear can reduce wear resistance in several ed structures
,
;
.
,
4-5
?|
requires a thorough analysis of the actual service conditions, a knowledge of applicability and limitations of the particular material and process, and data concerning the cost involved. The lack of engineering data avail
-
When two clean surfaces of the same meti
able for comparisons imposes a need for good judgment on the engineer
ley will weld together because of atomic
or technician who selects materials to withstand wear.
jcient pressure is applied to break through! V-:\:}rial such as oil dirt, or adsorbed moisturaji
Various techniques for providing surface protection to wear are as
;
,
,
0i$ienX contact to have elastic or plastic defor|K I welding takes place. The softening of metals||| es the ease of plastic deformation and facifM ause complete stoppage or if relative motiort| opposite face may be pulled out The result
;
,
One is tdft
ping material. The use of at least one metallic ,
I ( '
Metal spraying
[e scoring, galling and excessive local Weariflp leafing film or thin
Electroplating
Diffusion
.
.
follows;
Anodizing
,
a to minimize the danger of seizing
Many materials and methods are available for
protection against wear. The selection of a particular material and process
3 or crystallographically similar metals ha .
Protection against Wear
ancf
v-ctor for metallic materials is believed to bif i
"
service data.
cause phase changes that in|
jss, and decrease mechanical properties ions.
universal
designed to simulate actual service conditions. These tests should have proved reproducibility, should be able to rank various materials under consideration, and most important, should be validated by correlation with
.
as In braking.
"
wear test is not feasible. Therefore, equipment for wear testing must be
,
,
is not a s|mple phe-
nomenon, wear resistance is represented by fewer stanqiardizecrtests than other engineering properties. It is generally accepted that a
.
ctical or is not wanted
Since wear
0
Iff
t
1
"
I
Hard facing Selective heat treatment
tightly adherent oxide mM-e Electroplating The wear resistance of a metal part can be improved by
g is frequently helpful.
electroplating a harder metal on its surface. The metais most often
Aluminum oxide $
plated
welding. For parts that operate under such p
on base materials are Chromium, nickel, and rhodium. Indium plating has
iformation permits intimate contact
been used to reduce the wear of lead bearings.
the best® cpSant that combines with the metal surface tofp ,
"
of sufficient strength to keep the surfaces;*
als of high elastic limit will minimize seizure m iced by plastic deformation .
since the suddenly applied load may causefc
hape. Proper design should provide n surfacefp
above the compressive stress produced by B .
so that subsurface flow does not occur
.
T
Two types of chromium plating used industrially are known as hard chromium and porous chromium. The hard-chromium plate is the same as that used for decorative purposes but much thicker, usually from 0 0001 to 0.010 in. Porous chromium plate has on its surface carefully controlled pits or channels to hold lubricants. The term is misleading, since below .
thfi specially prepared surface, it is no more porous than ordinary hardchromium plate. The hardness of chromium plate is equivalent to 950 to 1050 Vickers. Another factor contributing to the reduction of vwajijsjhe low coefficient of friction of chromium plate. Chromium plating is used ,
in a discussion of wear since it is a gradual * per design to eliminate stress concentrations m
will increase fatigue strength Since fatigue J" jnsile stress residual compressive stress aim .
,
:ional
This may be accomplished!
protection arburizing and by shot peening .
,
i
m
.
in the cylinders and piston rings of internal-combustion engines. Nongalling is another useful property of chromium plate. Force-fitted chro-
mium-plated steel parts may be assembled and disassembled many times without seizing or galling. The high corrosion resistance of chromium is helpful in reducing wear under corrosive conditions.
572 INTRODUCTION TO f'HYSICAL METALLURGY
The first four of these processes we only the last two will be discussed he Chromizing consists of the introdL
The hardness of nickel plate is from 140 to 425 Vickers depending upon the nickel-plating solution used. Nickel plate is a good deal softer than chromium plaje, but in many cases it is hard enough for the purpose and
more economical. A nickel, plate may be finished by machining, while a
layers of the base metal. The process
chromium plate must be ground.
and may be applied to nickel cobalt, rr
The better throwing power of nickel- i
plating solutions as compared with chromium-plating solutions is an
,
\-
When it is applied to iron or steel
advantage in plating parts that have recesses. The hardness of rhodium plate is from 540 to 640 Vickers, and its wear '
,
stainless-steel case.
plate has high reflectivity, high heat resistance, and nontarnishing prop- i erties along with good hardness and wear resistance. The use of rhodium :
plate for refleptors of high-intensity light sources, for electrical contacts, ; [ m
'
and for slip riijigs and commutators has been mentioned in Chap. 12.
! :
.
14'7 Anodizing Th formation of an oxide coating by anodizing may be used to >. improve the wear resistance of certairr'meTalsriTie ahbdizi process is usually applied to aluminum, magnesium zinc, and their alloys. In anod-
the newest oxide layer always forms next to' the base metal \
Since
,
.
peratures may produce some distortio
Chromized high-carbon steels have a a low coefficient of friction.
Chrom
tools, hydraulic rams, pistons and pu Siliconizing or Ihrigizing, consists ( terial with silicon. The process is can ,
,
izing, the work is the anode, and oxide layers are built up on the base 1 metair
If the steel cont
(above 0.60 percent), chromium carbi resistance. The chromizing process ciple of transfer of chromium through tures." The temperatures used range fr
resistance is between those of nickel plate and chromium plate. Rhodium j
;
corrosion resistance and heat resistat
,
in order for the process to continue, the previously formed oxide layers
1700 to 1850oF. The work is heated ii
must be porous enough to allow the oxygen ions to pass through them. Anodizing aluminum is simply a method of building up a much thicker
terial such as silicon carbide and chl
oxide coating than mav be obtainediw-exposure to air. The Alumilite pro-
content of the base material.
cdss developed by the Aluminum Company of America uses sulfuric acid
14 percent silicon and is essentially
as an electrolyte for anodizing. The films produced !: eJxanspare.nt,..thicker, and more porous than those produced by other electrolytes. Continue
conized cases are difficult to mach
,
case depth ranges from 0.005 to 0.1 i
The ca;
.
Rockwell B 80 to 85.
due to a low coefficient of friction am
development resulted in Alumilite hard coatings, which are thicker and harder than prdinary anodic coatings. Aircraft parts such as hydraulic
pistons, guide tracks, gears, cams, screws, swivel joints! and friction locks
are xn de of hard-coated aluminum ajjiiys.
Bl4-9
"
The productiorTof a hard wear-resistant surface by anodizing has greatly extended the uses of magnesium and its alloys. Flash anodic coatings are
The increase i
cases have been used on pump shafts, valve guides, valves, and fittings for th Metal Spraying Metal spraying or flan years in production salvagejo build u to repair worn surfaces.
It has four
"
often used as-a base for paint adherence
appli cations.
Anodizing zinc produces a coating which has greater resistance to weaf
than chromate films. Anodic zinc coatings are used for cartridge cases,! airplane propeller blades, wire-screen cloth, and refrigeratorstrelves.
14-8 Diffusion
Sevpral processes improve wear resistanee-by diffusion of some|
element into the surface layers. These are:
Sprayed coatings can be applied by plating, which is used to deposit tun plasma arc spraying, which can depos
Metallizing fs usually done by auto controlled rate j3f sj3ged through thej: _
Air, oxygen, and a combustible gas ai
Carburizing
hoses and form a high-temperature J;
Cyaniding
tip. The wire tip is continuously melte a/e directed at the work by the high-\ surface, these particles flatten out to f same time they are forced into surfaa some mechanical interlock with previc
Carbonitriding Nitriding
Chromizing
f
U i
.
Siliconizing rl *
'
T ; \ -
.
"
*
i
' .
TOTfn
irr'irjiijTPiririiririirfTi i
j :
! i
i
WEAR OF METALS
te is from 140
to 425 Vickers depending upcm sed. Nickel plate is a good deal softer than
.
The first four of these processes were discussed in detail in Chap. 8, and only the last two will be discussed here. i Chromizing consists of the introduction of chromium into the surface
.
i'
jV cases it is hard enough for the purpose Plate may be finished by machining white'l
layers of the base metal. The process is not restricted to ferrous materials and may be applied to nickel, cobalt, molybdenum, and tungsten to improve
,
5und. The better throwing power fnicket Ired .with chromium-plating solutions is at have recesses
corrosion resistance and heat resistance.
When it is applied to iron or steel, it converts the surface layer into a stainless-steel case. If the steel contains appreciable amounts of carbon (above 0.60 percent), chromium carbides will precipitate, increasing wear resistance. The chromizing process most widely used employs the prin-
.
3late is from 540 to 640 Vickers and its wi ,
)f nickel plate and chromium plate Rhodiali| .
gh heat resistance, and nontarnishing pro j sss and wear resistance
tensity light sources
,
The use of rhodiurf
.
ciple of transfer of chromium through the gas phase al elevated temperatures. The temperatures used range from 1650 to 2000oF. These high temperatures may produce some distortion and grain growth during treatment. Chromized high-carbon steels have a hardness of 800 to 1000 Vickers and a low coefficient of friction. Chromizing is used on drop-forging dies, tools, hydraulic rams, pistons, and pump shafts. Siliconizing, or Ihrigizing, consists of impregnation of an iron-base ma-
for electrical contact
tators has be n mentioned in Chap 12 n oxide coating by anodizing may be usecftO .
_
bf certailTTnetals. The anodizing process f| magnesium zinc, and their alloys. ln afiod-( ,
and oxide layers are built up on the base ,
]e layer always forms next to the base mBta|
.
iontinue
.
;
terial with silicon. The process is carried out in the temperature range of
the previously formed oxide layers How the oxygen ions to pass through theffu
1700 to 1850oF. The work is heated in contact with a silicon-bearing ma-
,
terial such as silicon carbide, and chlorine gas is used as a catalyst. The case depth ranges from 0.005 to 0.1 in., depending mainly on the carbon content of the base material. The case produced contains approximately 14 percent silicon and is essentially an iron-silicon solid solution. Siliconized cases are difficult to machine, although the hardness is only Rockwell B 80 to 85. The increase in wear resistance by siliconizing is due to a low coefficient of friction and nongalling properties. Siliconized cases have been used on pump shafts, conveyor chain links, cylinder liners, valve guides, valves, and fittings for the chemical and oil industries. Metal Spraying Metal spraying or flame spraying has been used for many
.
.
Ply a method of building up a much thicket;
:
bmesdJay-expoam Jo air The Alumilite prcf .
lum Corhpany of America uses sulfuric ad
The films produced are transparent Jhlfiki produced b other electrolytes. Continui ,
,
nilite h&rd coatings which are thicker a ,
.
coatings. Aircraft parts such as hydraull ams, screws
,
swivel joints; and friction lock
ji num alloys. .
;
y anodizing has greatiyl jum and its alloys. Flash anodic coatings am
;;-;!ar-resistant surface b '
adherence.
;
oating which has greater resistance to weali
-screen cloth and refrigeratdrshelves.
!
,
.
:;l
iprove wear resistanee by diffusion of sonrii These are:
years in production gaiuarp tn h iif to repaix w orn surfaces. applications. .
i
zinc coatings are used for cartridge easel
573
[T
mmTRmns that are unclPrsi7ft anrl
It has found increased use for wear-resistant
_
"
'
Sprayed coatings can be applied by several methods; metallizing; flameplating, which is used to deposit tungsten carbidej d_aluminum oxide; plasma arc spraying, which can deposit almost all inorganic materials. Metallizing Fs usually done by automatically feeding a metal wire at a
cpntrolled rate of speed thrpugh the mfitallizing tool or
"
gun" (Fig. 14-4).
Air, oxygen, and a combustible gas are supplied to the gun by means of hoses and form a high-temperature high-velocity flame around the wire tip. The wire tip is continuously melted off, and the liquid-metal particles are directed at the work by the high-velocity flame. When they strike the surface, these particles flatten out to form irregularly shaped disks. At the
same time they are forced into surface pores and irregularities
provide
some mechanical interlock with previously deposited material. Cooling is 1
-
-
1 -
574 .V
INTRODUCTION TO PHYSICAL METALLURGY
'
:
.
-
-
r
foi spraying hard
4 min
Wire ond qas nozzle
10 max
Flame cone
corrosion-resista
silicon. Aluminum,) tinr and zinc p powder gun. j
i
Burning gases
Wire
,
nickel-base or cobalt-base material
Plasma is a luminous stream of io
I
Oxy- fueigas
Melting envelope '
.Air cap
I
wire
Atomized spray-
|
Sprayed coating
Compressed air
I
throu£h
f
genjiad-combuotion gas, as well asji
I
from the gas stream. Detailsof a plas ly in Fig. 14-6. The disadvantage of ful safety measures that must be enhigh temperatures high noise level, to
JI
"
spraying., (From "Metals Handbook, vol. 2, 1964, American Society for Metals, Metals Park, Ohio.)
\ "
deposited particles. The nature "orthe oxides formed
,
Various-methods have been devel(
h
the base material.
uniaerfnelanizinff
'
,
and alloys but will not bond to copp Other bonding methods prepare the rough threading. The Fusebond me on the surface to provide a good stn
properties, s|uch as chromium steel, aluminum bronze, or silicon-aluminumMr '
alloys, shovy relatively high strength in the sprayed form. Metals whicHBp
but it cannot be used on brass bronz ,
form loose, friable oxides, such as the brasses and copper, produce coat-sp;
In general, a sprayed metal coating than equivalent cast or wrought me of hardness by wrought wires and sp metallizing, while Table 14-2 gives the arc sprayed coatings. High-chromium (13.5 percent) stait plications where a hard coating is re ance is not necessary. The material h shrinkage, and little tendency to cracl armature shafts, cylinder liners, pisto
ingsjof low strength. Another method of metallizing employs an oxyacetylene powder gun, any.
contoured surface. It is necessary to postheat the coated surface to fusi
the sprayed deposit. The powder gun can apply thin coatings and is uselul
Spray powaer (rem comsien
Aspirating gos
:.;
.
! min 10 max
p . .
No:zle
|
iff
i
I
3l
,
Elurning goses
'
:
.,
/
Oxy-fuel gas-
n Nylon casingElectrode adiustment
Ji
1
n
-
/
d
'
/
Spray si ream-
One of the mosl
bonding coat of molybdenum 0.001 i alloy layer. Molybdenum adheres to n
deposit. Metals that form dense, tenacious oxides having good physicaiSK
Fig. 14-5. The gun can spray metal powder in place or over almost
_
melting metals. An additional advant.
Substrate-
Tig. 14-4 Cross-sectiort diagram of a wire gun for metal
electricjarc. Temperatu
an
obtainable; thus the use of plasma fla
-
vCooling water jacket
Plosr
top
MIS
r3
Electrode
.
Sprayed coating-''jj
Powder spray gun-
Substrate 1
Fig. 14-5
i
spraying, (From
.
-
Cross-section diagram of a powder gun for metal "
Metals Handbook," vol. 2, 1964, American
Society for Metals, Metals Park, Ohio.)
"
1
r
I '
" -
-
-i
Electrical connection
'
I
and cooling water inlet
fig 14-6
-Plasmo
gas inlet
Elecli and cool
Cross-section diagram of a plasma gun for mel aying. (From Metals Handbook," vol. 2, 1964, Americ ciety for Metals, Metals Park, Ohio.) "
IURGY
WEAR OF METALS
4 mm
!0 max
mi
-
S iVrjK . .
for spraying hard, corrosion-resistant alloys. Mostj)Mhese alloys are nickel-base or cobalt-base materials containing chromium, boron, and silicon. Aluminum, trnrand zinc powders can also be ptayed-by the
i
flome cone
'
-Binning
goses
_ -B .
Melting wire
pov/der gun.
- h.
.
Plasma is a luminous stream of ionized gas produced by passing a gas through an electric arc. Temperatures up to 30,000oF are economically obtainable; thus the use of plasma flame permits deposition of the highest-
"
melting metals. An additional advantage of the plasma process is that oxy-
V
i Atomized spray-
gen and combustkjn-gas, as well as their combustion products, are absent
Sprayed coating-
from the gas stream. Details'of a plasma spray gun are srhown schematical-
mi substrat
in Fig. 14-6. The disadvantage of plasma-arc spraying is the very careful safety measures that must be employed due to the added hazards of
for metal '
64, American
_
high temperatures, high noise level, toxic waste products, and radiation. Various-methods have beerTdeveloped for bonding .sprayed metals to the base material.
alloy layer. Molybdenum adheres to nearly all steels and-many other metals and alloys but will not bond to copper, brass, bronze, or nitrided steels. Other bonding methods prepare the surfaces by abrasive blasting or by rough threading. The Fusebond method deposits rough, porous nickel on Ihe surface lo provide a good strong anchorage for sprayed coatings,
_
ure Of {he oxides rmed uTTaer me allizin! arge degree the physical properties of the| '
nse, tenacious oxides having good physicsll
:
steel, aluminum bronze, or silicon-aluminum' Irength in Ihe sprayed form
Metals which|
.
h as the brasses and copper
;
,
:
but it cannot be used on brass, bronze, or copper.
produce coat-;!
ing employs an oxyacetylene powder
One of the most widely used methods is to spray a
bonding coal of molybdonurn O.OOt to 0.003 in. thick which forms a thin
film (orms on l|ie exposed surfaces of thtf "
S75
In general, a sprayed metal coating is harder and mord brittle and porous than equivalent cast or wrought metal. Table 14-1 shows a comparison of hardness by wrought wires and sprayed particles of various metals by
gun.;
/ metal powder in place or over almost mfi
molalli/ing, while Table 14 -2 gives the Vickors hardness ol typical plasma-
ssary lo posthoal the coated surface to fus&
arc sprayed coatings. Hiqh-chromium (13.5 pnrcnnt) stainless stonl can be used in many ap-
.
der gun can apply thin coatings and isuselullf
pliciiioiu; whoio (i hntcl cojiIuhj is roqtmo
ance is not necessary. The material has high strength and elongation, low
mm
I
r
I
armature shafts, cylinder liners, pistons, valve stems, and hydraulic rans.
lil tiiih
/Burning gases
luink,'K)o and litllo tondoncy lo crack on r.haflr.. Typical npplicnlions nro
.
"
,
1 Nylnl! l.UMIu)
;
.
X;J
-
.
.
-
V
<
Fleclrode .
-
.
.
.
iC.oolini-i wnliii iiilKo!
I'll!'.,inn
(tii iniii'i .ri
Powder "ii'.-i
BIT
Copper olloy no.-vlo
iid|ir.l inHil
.
Spray si ream-''
Electrode
Spmycri CiKilmq I Id.11lull rnnncclion
Sub'ili n lo
un for metal i'l Amnricnn ,
uiid i.nolnui wnlei inlel
Fig, 14-6
'
ir.id')
m:. mlel
if! 1' 1 I II.ill i.onii'-' i Hill
mil toulirn] wiitiii iMtlrt
Cross-section diagram of n plasma gun for metal
IJprnyimj (l iom "Moifilii iltmdbook," vol. ;'. lOG'l, Amorii.an .
vSociety lor Metals, Metals Park, Ohio.)
/
576
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 14-1
The development of hlgher-tempe
Comparison of Hardness of Wrought Wires '
and Sprayed Particles of Various Metals'1
opened the door to many new coatin standing resistance to wear and abrs
KNOOP HARDNESS
!
METAL
VALUES (50-g LOAD)
positions, applied by flame-plating,
WROUGHT
SPRAYED
WIRE
PARTICLES
either wear resistance alone or wear sion resistance is needed. The basil
i
,
taining 6 to 8 percent cobalt is recomi valve plates, bearings cobalt (13 to 17 percent) wear abou
AlLtmlnum bronze Molybdenum
229
408
404
1535
Monel
338
326
1010
307
445
1025
370
504
nickel, and mixed tungsten-chromiu
1 080
398
664
wear resistance and resistance to heal
Type 304
360
381
shown in Table 14-3 a chromium c;
type 420
372
757
can also be used for the same kind c
such as seals
,
Steels:
,
better resistance to mechanical shod
.
,
.From "Metals Handbook
' .
may be deposited by flame-plating
vol. 1. 1964, American Society for Metals
Metals Park, Ohio
Tl
.
under corrosive or high-temperature about one-half the wear resistance to
Molybdenum coatings combine a hard wearing surface with good adhesion. They have good wear and abrasion resistance and have been used.i
successfully to build up the worn surfaces of aluminum pulleys and alu| "
1
I
provide an excellent bearing surface and are particularly good when run-;
'
-
I
sion or where highly localized loads a
for applications involving light abrasiv
ning against bronze bearings under severe abrasion conditions. Copper
ary lubrication exists and to provide a surface to inexpensive base materials
parts as pump impellers, bronze castings, split motor bearings, and
to apply,
brake valves. Monel and nickel coatings are used where wear resistanp| combined with corrosion resistance is needed, as in pump plungers, shaft||
and
and hydraulic pumps.
wear-resistantcflmpounds are avai laB
"
,
| alloys can be used for general-purpose wear applications. Aluminum (4.ip Hard Facing The production of a h bronze is very wear-resistant and machines easily. It performs well on sucl»|||A .7 metals ing isJ
ci-.1
-
minum and iron brake drums used on elevators and presses. They also-
The wear resistance of sprayed met lizing is not suitable for service invol
' * ' . .
TABLE 14-2
Molybdenum wire As sprayed Molybdenum powder As sprayed ]
!_-J
168-183 321-368
Tantalum powder 443
Hafnium carbide
ma rials for ixj abilit
.
1000 1000
168-205
fer. (5xyacetylene-gas welding produ
Chromic oxide
1000
Chromium-aluminum oxide
287-455
Hafnium oxide
251-313
positioned mpje pxedse Electric-arc welding is less expensive i to automatic equipment Arc deposits a to be porous; they tend to develop crac ,
Zirconium oxide 246-263
638-805
.
gradients due to rapid heating ancf cc 330-390 .
i
i
;
xidatroTrpr
Osmium oxide
Tantalum carbide
3
.:
flux-coated for electric gnr
of impurities, thermal and electricaiTin
r
i.
_
and jire
From Materials Engineering Manual no. 201, Van Nostrand Reinhold Company, New York, 1962.
f
quiring qnl Jhe_bard-far;ir
CARBIDES AND OXIDES
.
As sprayed Tungsten wire As sprayed Tungsten powder As sprayed
.
expensive alloys and protection in def The hard-facing material is provided ing rod. These rods are generally use
Vickers Hardness of Typical Plasma-sprayed Coatings*
METALS
re
._
an oxvacetylene flame or electric are that (1) it may be applied to locali;
melted and spread over the base met surface ranging from 1/16 to 1/4 in tl .
1
LLURGY
WEAR OF fytETALS /577 4 <
Ison of Hardness of Wrought Wires
The development of higher-temperature flame-spraying processes has opened the door to many new coating materials, many of which have out-
i of Various Metals*
KNOOP HARDNESS
standing resistance to wear and abrasion. Several tungsten carbide compositions, applied by flame-plating, can be used for applications where either wear resistance alone or wear resistance plus shock heat, or corrosion resistance is needed. The basic tungsten carbide composition containing 6 to 8 percent cobalt is recommended for general-wear applications such as seals, valve plates, bearings, and shafts. Alloys containing higher cobalt (13 to 17 percent) wear about 30 to 40 percent faster but provide better resistance to mechanical shock. Compositions of tungsten carbide nickel and mixed tungsten-chromium carbides are also available where wear resistance and resistance to heat or corrosion (or both) are needed. As shown in Table 14-3, a chromium carbide nickel-chromium composition can also be used for the same kind of service. Aluminum oxide coatings
VALUES (50-g LOAD) I
WROUGHT
SPRAYED
WIRE
PARTICLES
229
408
404
1535
338
326
307
445
370
504
398
664
360
381
372
757
vol. I, 190-1
..
\
-
.
.
Aiiiciicnii Socioly loi Mol.ils
,
,
,
?!
may be deposited by flame-plating. They are superior to tungsten carbides under corrosive or high-temperature oxidizing conditions but have only
.
about one-half the wear resistance to wet abrasion.
(
k
ine a hard wearing surface with good adhe*!>
The wear resistance of sprayed metals is generally very good, but metallizing is not suitable for service involving heavy impact or extreme abra-
and abrasion resistance and have been used '
v
]
worn surfaces of aluminum puileyr. nnd alu-
sion or where highly localized loads are applied. Metallizing is best suited for applications involving light abrasive wear, for conditions where boundary lubrication exists, and to provide a wear-resistant or corrosion-resistant
' .
.
i
is used on elevators and presses
.
They also
surface and are particularly good when run-:
Coppefl surface to inexpensive base materials. Aluminumj )4.-|p Hard Facing The production of a hard wear-resistant surface layer on and machines easily. It performs well on sucti| |> J metals by welding is known as hardjaping. This method is relatively easy )nze castings, split motor bearings, and air-l to apply, requiring onlyjheJiard lacma_aJJoys in the fdrm of v eMjig rods
s under severe abrasion conditions
.
eral-purpose wear applications.
-
_
kel coatings are used where wear resistancaf
stance is needed
,
as in pump plungers shafts.
are that (1) it may be applied to localized areas subjected to wear, (2) hard
,
.
_
and an oxvacetylene flame or elexMcarc. The advantages of hard facing
a f
wear-resistant compounds are availaBTe, and (3) it provides effective use of expensive alloys and protection in depth.
CARBIDES AND OXIDES
The hard-facing material is provided in the form of an electrode or welding rod. These rods are generally used bare for oxyacetylene-gas welding and are fIux-coatedjor ejectric airtrweIclingr The fluiTcoating centalns
Hafnium carbide
1000
materials for arc stability, oxidation protection of the molten weld, fluxing
Tantalum carbide
1000
Plasma-sprayed Coatings*
.
.
Osmium oxide
168-205
Chromic oxide
1000
Chromium-aluminum oxide
287-455
Hafnium oxide
251-313
Zirconium oxide
638-805
1 :
of impurities, thermal and electrical insulation, and control of metal transfer. Oxyacetylene-gas welding produces smoother deposits that can be positioned mgr precisely, whHle the heating and cooling rates are slower, Electric-arc welding is less expensive may be faster, and lends itself better \ to automatic equipment. Arc deposits are generally rougher and more likely to be porous; they tend to develop cracks because of the sharp temperature __
_
,
gradients due to rapid heating ancf cooling. The hard-facing material is 5
Jostrand Remhold Company New York, 1962. ,
i
melted and spread over the base metal and bonds with it to form a new surace ranging from 1/16 to 1/4 in. thick, depending on the application.
578 INTRODUCTION T0! PHYSICAL METALLURGY
)
o
0)
oj
to 3
C
ni
Only the surface of the base metal is prevents mixing of the alloy with tc changing the properties of both the < Hard facings can be applied to mos
ia)
V
-
.
-
V
"'"
.
5 S
J
« 5 o
| "S f |
< s
i§ o
eg
cd
1-
CNJ
T-
T-
X
LU
c
5
tions it is not advisable to face nor
X
nJ
below 2,000oF.
O
Carbon steels are re
below 0.35 percent carbon.
8 S 2 i TO 3 S c
Weldi
"
creasing carbon content and the hig
w 2 E o.
i 3
"
,
heated before and postheated after h
e s g
oj
ductile iron, and high-speed steels ( appropriate welding practices are o faced but copper, brass and bronzt of their low melting points and high Hard facing is most extensively usee abrasion is impossible as on oil-well
S
moving equipment mining tools, eni
o CM
I 5 S 5 7
il „ o i-
m r-
. o
ID
-
n
E 0
I Io> i5 T
in -
Tf
<-
O
2? S
*-
2 £ BI
o t
'
si
-
-
O
i
CO
o
UJ
corrosion-resistant than case-harden
CD O)
critical local areas of large componei be impractical or impossible to hard' Since hard facing is a welding proce location without dismantling heavy ( crease operating efficiency by extend ing replacement cost and loss of prod of a low-cost base metal for parts th<
O
" 2 £
| 8|5
O
processing equipment. Hard facinc parts such as metalworking dies anc high wear rate. Hard-faced surface: CM
?
£III
O
in
,
w B 5>
to
TO
W
TO
>.
Q
.
O
CO UJ
U
a
S; a, o
o
<
co c E c a; " o E
!S8£
m m
O)
n o
O
CO M
CM 9
2 3
ills
o
o
5
"
CO
CO
O
I a S ? £ 1 i? S
There are "
O
o
Z
Br-
or
2?
.
o
a>
o
re u
CO to
.
I O
M O
5
5
11 i So I-
CO
o
c
LLI
(0
,
of inserts only where needed, as in t mium irons containing 17 to 32 perc different alloy compositions. The <
c o
a i
c .
a
c
LU
a
|§-o-:
<0
5
'
\
UJ
2
m
<
tungsten carbide. A simplified classi alloys is given ip Table 14-4.
factor because the material has a loi
I8 x
150 differei
Tungsten carbide hard facings hav resistance. Although it is relatively
o
CO
E
more than
terials commercially available, rangii 2 percent total alloy content to nicl
CO
>
a
,
,
=
.
!§£§ '
!
a)
Sis
CM
1
,
k s "
i
i
£
A 6
I 8
*-
S S
t
5I
W)
8
O 13 o K
111 .s .s I
0)
2
o
i
E o
CD
O
J;..
and are best for metal-to-metal wear
such as farm equipment used in sar tougher than the austenitic grades Additions of tungsten, molybdenum, help increase hot-hardness and add
URGY
:
-
WEAR OF METALS
o
0)
Only the surface of the base metal is brought to melting temperature; this
S" s~ siS II t n
prevents mixing of the alloy with too much of the base metal and thus changing the properties of both the coating and the base metal. Hard facings can be applied to most ferrous metals but with a few excep-
a)
a; e a
S o
CM o
§ o O 3
OJ OJ
-
'
T-
,0
,
fill!
CD 1-
579
tions it is not advisable to face nonferrous alloys having melting points below 2,000oF. Carbon steels are relatively easy to hard-face particularly beiow 0.35 percent carbon. Welding becomes more difficult with incrcjasing carbon content, and the high-carbon and alloy steels must be preheated before and postheated after hard facing. Stainless steels castTron, ductile iron, and high-speed steels can also be hard-faced provided that appropriate welding practices are observed. Monel can easily be hardfaced, but copper, brass, and bronze are relatively difficult to do because ' of their low melting points and high conductivity. ,
S S .2 i
3
s
m
2
c
c
t; a
.
0
,
o
5 =
S in eg »
-
cj
5 « o£ i
o
J
S 2 S S a. o
! s
Hard facing is most extensively used where systematic lubrication against abrasion is impossible as on oil-well drilling tools agricultural and earthmoving equipment, mining tools engine valves, and refinery or chemical processing equipment. Hard facing also extends the life of lubricated parts such as metalworking dies and areas of machine parts that have a high wear rate. Hard-faced surfaces are usually more wear-, heat-, and ,
,
,
J; . : V g
is.- Jill i- 21
i
11 J(0
o
0)
o o
OJ o 2 0)
o
S lis v
C
-C W
II 2 g
o O r}O N
O)
1_
5 01 5 o m
CO
O
LLl
,
CD
tfl
CO
till . . . . .
ill
J«?
"
.
R
g S S
11
n
m S
C
,
U
o
corrosion-resistant than case-hardened or flame-hardened surfaces. Also, 1
|
i
i
'
4
3
crease operating efficiency by extending the life of equipment, by decreasing replacement cost and loss of production time, and by permitting the use
i I
of a low-cost base metal for parts that wear or corrode. There are more than 150 different compositions of hard-facing maInrials commercially available ranging from steels containing only about ,
2 percent total alloy content to nickel-base and cobalt-base alloys and tungsten carbide. A simplified classification of wear-resistant hard-facing alloys is given ip Table I4-4.
>
i
Tungsten carbide hard facings have the highest hardness and best wear
o
CM P-* '0
Q -t
O O
c
=
.
o
g 2 -
» «
.
is
-
-
w CD
CD
5 o o a
«|| I | > 5 5 o.
resistance.
I
lactor
Although it is relatively costly this may not be an important ,
because the material has a long lite and can be applied in the form of inserts only where needed as in the case of rock drill bits. High-chro,
,
Js?
*
critical local areas of large components can be hard-faced where it would be impractical or impossible to harden the component by heat treatment. Since hard facing Is a welding process, it may be used to repair parts on location without dismantling heavy equipment. Hard facing serves to in-
;
5
I I
J
mium irons containing 17 to 32 percent chiomium are available in many different alloy compositions. The austenitic types are relatively cheap and are best for metal-to-metal wear or low-stress abrasion applications ,
such as farm equipment used in sandy soil The hardenable grades are tougher than the austenitic grades and have excellent wear resistance Additions of tungsten molybdenum, and vanadium ar sometimes used to .
I
.
,
1
help increase hot-hardness and add to abrasion resistance
.
v. .
4
Martensitic
::
::y
:
.:
; <:
580
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 14 4
for hard surfacing. The austenitic st grade, have been used for moderate
A graded series of wear-reslstant alloys'"
A 1
.
Tunpsten carbide
Maximum abrasion resistance; worn
ing and grinding of coal limestone, a steels are very tough and work-harde This alloy is used as a base for hard f£ as for overlays. ,
surfaces become rough 2
.
High-chromium irons
ExceHent erosion resistance; oxidation resistance
3 Martensitic irons .
o
Excellent abrasion resistance;
; 14-11 Selective Heat Treatment
high compressive strength 4
Cobalt-base alloys
Oxidation resistance, corrosion resist-
5
Nickel-base alloys
Corrosion resistance; may have oxidation and creep resistance
.
'
m
ance, hot strength, and creep resistance
CD
CO CO
c
CD c
o
.
'
C
m
-
ca
O) D
o
n ca
O)
O)
'
ui ca
'
6 Martensitic steels .
ui ca cu
7 Pearlitic steels .
o
o
c
c
core. These methods were discusse QUESTIONS
Good combinations of abrasion and
14-1
impact resistance; good compressive strength
14-2
Inexpensive; fair abrasion and impact resistance
8 Austenitic steels .
Work hardening
Stainless steels
Corrosion resistance
Manganese steel
Maximum toughness with fair abrasion resistance; good metal-to-metal wear
143
Explain methods that may be used
14-4
What factors should be considered t
n
14-5
i
Surface Protection against Wear," American Society for Metals, Metals Park, Ohio 1954. ,
Differentiate between metallic wear Differentiate between "thick-film" li
n
resistance under impact '"
The metl
are induction hardening and flame h low-hardening methods to produce
ra
1
i
What would be a good method of rei how may wi 14-7 What are the advantages and disa plating for wear resistance? 14-8 List the various diffusion processes a practical application of each. 14-6
Aside from lubrication
14-9
What is the principle of metallizing
,
applications. 14-10 Give some applications of hard faci
irons are mainly chromium-nickel, chromium-molybdenum, or chromium- f The combination of rrartensite and a carbide matrix REFERENCES '
tungsten alloys.
1 %'
provides a hard composite structure with good abrasion resistance. The cobalt-base alloys contain from 45 to 63 percent cobalt 24 to 29 ,
American Society for Metals: "Metals Ham vol. 2 1964, Metals Park, Ohio. ,
percent chromium, 5.50 to 13.50 percent tungsten, and 1.10to3.20 percent!
:
"
"
carbon. These alloys are used where wear and abrasion resistance com-. bined with resistance to heat and oxidation is required. They have been used as hard-facing materials for ladle linings and pouring spouts to re- > sist hot gases and liquids. The nickel-base alloys contain 70 to 80 percent nickel, 11 to 17 percent chromium, 2.5 to 3.7 percent boron, and smaller amounts of cobalt and silicon. They have slightly better wear and oxidation resistance than the cobalt-base alloys and have been used for applications up to 1800oF. Some typical applications are for hard facing of hot-heading dies, piercing mandrels, and shear blades exposed to hot solid metals. Martensitic steels have the advantage of low initial cost good hardness,
:
,
strength, abrasion resistance, and toughness. They have been used mainly as buildup layers under other alloy compositions that have better abrasion
Surface Protection against Weai
Surface Treatment of Metals
,
York, 1947. -: Hard
Facing for Impact Welding ,
Burns and Bradley: "Protective Coatings I pany, New York, 1967.
Burwell, J. T. (ed.): "Mechanical Wear
" ,
f
Ohio, 1950.
Gabe, D. R.: "Principles of Metal Surfac Press New York, 1972. ,
Ingham and Shepard: "Metallizing Handt Riddihough M.: "Head Facing by Welding Stoody Company: "Stoody Hard Facing ( ,
.
f
.
IV
Avery H. S.: "Hard Surfacing by Fusion V
resistance or higher toughness. Pearlitic steels have relatively low hard-;; ness and w6ar resistance and are therefore infrequently used as facings. They are used mainly as build-up metal for welding or as the bast]
V
"
,
WEAR OF METALS
LURGY
for hard surfacing. The austenitic steels, particularly the high-manganese grade, have been used for moderate-service conditions such as the crushing and grinding of coal, limestone, and aggregates. Austenitic manganese steels are very tough and work-harden rapidly under impact (see Sec. 9-10). This alloy is used as a base for hard facing because of its tQughness, as well
ies o( wear-resistart alloys* Maximum abrasion resistance; worn
surfaces become rough Excellent erosion resistance; oxidation resistance
as for overlays.
.
Excellent abrasion resistance:
11
high compressive strength
.
and creep resistance
ESTIONS
Good combinations of abrasion and
.
a
impact resistance; good compressive strength
4
Inexpensive; fair abrasion and impact
I
.
-
v
.
'
14 1
Differentiate between metallic wear, abrasive wear, and erosion.
14 2
Differentiate between "thick-film" lubrication and boundary lubrication. Explain methods that may be used to minimize seizing, What factors should be considered to set up a good wear test?
14-3 14-4
rosislanco
14 5
Work hardening
14 6 14-7
Corrosion resistance
What would bo a ()Ood molhod ol rortucinri woai in a diavj/ing din? Why? Aside (rom lubrication, how may wear be reduced in an automotive cylinder? What are the advantages and disadvantages of nickel plating vs. chromium
plaling for wear resistance?
Maximum toughness with fair abrasion resistance; good metal-to-metal wear resistance under impact
!
The methods used for selective heat treatment
core. These methods were discussed in Chap. 8.
Corrosion resistance: may have oxidation and creep resistance
'
Selective Heat Treatment
are induction hardening and flame hardening. These are essentially shallow-hardening methods to produce a hardened case and relatively tough
Oxidation resistance, corrosion resistance. hot strength,
581
14'8
List the various dillusion processes (or increasing wear resistance, and give
a practical application of each.
14-9 What is tho princlplo of mnialli/inn lo incroosn wonr msislnnco? Givo somo
American Society lor Moluls, Motuls Paik, Ohio \ iA.
applications.
14-10
Give some applications of hard facing to improve wear resistance.
ckel, chromium-molybdenum or chromium-1 f talion ol martensite and a carbide matrix J REFERENCES ucture with good abrasion resistance ,
.
tain from 45 to 63 percent cobalt 24 to 29 | % ,
50 percent tungsten
,
and 1.10 to 3.20 percent'| \
d where wear and abrasion resistance com-
4 for ladle linings and pouring spouts to re- J
"
:
"
Surface Treatment of Metals," Metals Park, Ohio, 1941. ;
Hard Facing for Impact, Welding J. fW.YJ, vol. 31. no. 2, pp. 116-143, 1952. Rurns and Bradley; "Proloclivo Coalings lor Moials," Von Noslrond Roinhold Com:
-
ie nickel-base alloys contain 70 to 80 percent ,
Surface Protection against Wear and Corrosion," Metals Park, Ohio, 1954.
:
York, 1947.
.
nium, 2.5 to 3.7 percent boron
vol. 2, 1964, Metals Park, Ohio.
Avory, H. S.: "Hard Surfacing by Fusion Welding." American- Brake Shoe Co., New
and oxidation is required. They have been
;
American Society for Metals "Metals Handbook," Vlh ed., 1948; 8th ed., vol. 1, 1961,
and smallerlj
pany, New York, 1967,
They have slightly better wear and oxidatiorij
Burwell, J. T. (ed.): "Mechanical Wear," American Society for Metals, Metals Park,
e alloys and have been used for applications i pplications are for hard facing of hot-head-.j|
Gabe, D. Ft.: "Principles of Metal Surface Treatment and Protection," Pergamon
./nd
Ohio, 1950. Press, New York, 1972.
shear blades exposed to hot solid metals J advantage of low initial cost good hardness, :W. .
.
,
and toughness. They have been used mainly
Ingham and Shepard; "Metallizing Handbook," Metco, Inc., Westbury, N.Y., 1965. Riddihough, M.: "Head Facing by Welding," lliffo & Sons, Ltd., London, 1948, Stoody Company: "Stoody Hard Facing Guidebook," 2d ed., Whitter, Calif., 1966.
alloy compositions that have better abrasion .
is. Pearlitic steels have relatively low hard-1
id are therefore infrequently used as hardf as build-up metal for welding or as the baseT
i <-x: ; ; -
-
I
i r \ :
\ |
i
cc OF 15-1
introduction
In the broad sense, corr
tion of a material by chemical, electn between the environment and the rr --
sistent in character.
i
In some instan
thin adherent film which merelyjstaii as a retardant to further corrosive a ;
corrosion are bulky and porous in cl One of the most serious problems in the billions of dollars each year. a great deal is known; yet despite exl there is still a lot to learn.
In some i
corrosion is highly obvious, but in o rosion, it is less obvious but just as i
The basjc cause of corrosion is i\ J
forms.
The metals tend to revert to
esses of corrosion. ;
15-2 Electrochemical Principles Corrosic process resulting in part or all of tl metallic to the ionic state. Corrosion certain areas of a metal surface thrc
any solution that contains ions, lo groups of atoms. Pure water, for ( hydrogen ions {H+) and negatively (
amounts. The electrolyte, therefore, or alkaline solutions of any concen cuit, there must be two electrodes, ai
be connected. The electrodes may bi may be different areas on the same tween the anode and the cathode mi
\
'35
1
1
CORROSION OF METALS
5
5-1
.
ji
Introduction In the broad sense, corrosion may be defined as "the destruction of a material by chemical, electrochemical, or metallurgical interacton between the environment and the material." Generally it is slow but persistent in character.
In some instances the corrosion products exist as a
thin adherent film which merelyjstains or tarnishes the metal and may act as a retardant to further corrosive action. In other cases, the products of 1
corrosion are bulky and porous in character, offering no protection. One of the most serious problems of industry, corrosion causes damage in the billions of dollars each year. It is a complex problem about which a great deal is known; yet despite extensive research and experimentation, there is still a lot to learn. In some cases, such as direct chemical attack, corrosion is highly obvious, but in other cases, such as intergranular corrosion, it is less obvious but just as damaging. The basic cause of corrosion is the instability of metals in their refined
forms. The metals tend to Revert to their natural states through the processes of corrosion.
15-2 Electrochemical Principles
Corrosion is essentially an electrochemical
process resulting in part or all of the metal being transformed from the
metallic to the ionic state. Corrosion requires a flow of electricity between certain areas of a metal surface through an electrolytp. An electrolyte is
any solution that contains ions. Ions aro olectrically charged atoms or groups of atoms. Pure water, for example, contains positively charged hydrogen ions (H+) and negatively charged hydroxyl ions (OH-) in equal amounts. The electrolyte, therefore may be plain water, salt water, or acid ,
or alkaline solutions of any concentration To complete the electric circuit, there must be two electrodes an anode and a cathode and they rnust A .
,
,
be connected. The electrodes may be two different kinds of metals or they H may be different areas on the same piece of metal :The connection be- ! tween the anode and the cathode may be by a metallic bridge but in cor,
.
,
1
.
i
584
INTRODUCTION TO PHYSICAL METALLURGY
H+ +
H
+
Fe + +
Anode
Anoae
Fe+ + Fe e
H
+
H
6 +
H+
e
"2
e
n2
ciCathode
Cathode
Hp
H
'
'
i*r
"
-
r
|
" -
H2
+
| Fig. 15-2 Polarization of the local cathode by a film of |; hydrogen (The International Nickel Company.)
Fig. ' 5-1 Illustration of the formation of ions at the anode and hydrogen at the cathode in local cell action. (The International Nickel Company.)
.
Urnrosion it is usually achieved simply by contact.
In
The products of anode and cath
order for electricityap
enter into further reactions that yie rosion products. For example with the cathodic reaction in their migrati
,
to flow tnere musfBe a potential difference between thfe electrodes. 11 3 piece of ordinary iron is placed in a solution of hydrochloric acidJK
,
,
vigorous bubbling of hydrogen gas is observed. On the surface of theJBI
anode encounter ferrous ions movi
ij sions metal there are numerous tiny anode and cathode areas caused by inclu»;jB| in the metal, surface imperfections, localized stresses, orientation 11
ions combine to formjerrpus hydro; oxidized by oxygen in solution to fori
I of the grains, or perhaps variations in the environment. This condition i$|B||
Vshown schematically in Fig. 15-1. At the anode, positive-charged iron atomsJHI
'
detach themselves from the solid surface and enter the solution as positive!
,
.
in the metal. At the cathode the electrons meet and neutralize some posi-S (§15-3 Factors Influencing Corrosion
One
tively charged hydrogen ions which have arrived at the surface through the v
fluencing corrosion is the differeno
/ electrolyte. In losing their charge, the positive ions become neutral atomsif agaTrTirKJ combine to form hydrogen gas. So, as this process continues,! oxidation and orrosion of iron occurs at the anodes, and plating out ofi
tial is due to the chemical natures of th
;
metals when coupled together ancf inr indication of which metals may be a given by the standard electromotive
hydrogen occurs at the cathodes. The amount of metal which dissolves
r
agitatioiTof the solution this rust ma face or right next to it where it ca' further progress of corrosion ,
f .
ions, while the negative charges, in the form of electrons, are left behind ,
'
r as a form of iron rust. Depending o
is proportional to the number of electrons flowing, which in turn is depen- j dent upon the potential and the resistance of fhe metal.
In order for corrosion to continue it is necessary to remove the corrosion.J products from the anode and the cathode. In soifie cases, the evoiutfon .
of the hydrogen ga% at the cathode is very slow, and the accumulation of a|
layer of hydrogen on the metal slows down the reaction. This is known as'|
OH '
catholic polarization (Fig. 15-2). However, oxygen dissolved in the elecyMI trolyte can react with accumulated hydrogen to form water, thus allowing all,
1
OH
Fe++
'
FelOHK
Fe"|-+ OH
ism
OH
.
corrosion to proceed. For iron and water, the rate of film removal d»JWjj|;
-
I
/ pends on the effective concentration of dissolved oxygen in water adjacent Mt to the cathode. This effective concentration, in turn, depends upon tbtfF™55
'
degree of aeration, amount of motion, temperature, presence of dissolved; salts, and other factors.
pron. (The International Nickel Company.)
s
AW
m
{fig. 15-3 Formation of ferrous hydroxide in the rusting
™
.
'
:
'
CORROSION OF METALS
\\v.\k
is
t the anode
Fig. 15-2
585
/....; i-
Polarizalion of the local cdlhode by a film of
hydrogen. (The International Nickel Company.)
mm (The t
simply by contact. :
The products of anode and cathode processes frequently meet and
In order for electricity
ial difference between the electrodes
enter into further reactions that yield many of our common visible cor-
rosion products. For example, with iron in water the hydroxyl ions from
.
s placed in a solution of hydrochloric acid
the cathodic reaction in their migration through the electrolyte toward the anode encounter ferrous ions moving in the opposite direction. These
,
. v n gas is observed. On the surface of the anode and cathode areas caused by inclu- "J '
ions combine to form ferrous hydroxide (Fig. 15-3). This soon becomes oxidized by oxygen in solution to form ferric hydroxide, which precipitates as a form of iron rust. Depending on the alkalinity, oxygen content, and agitation of the solution, this rust may form either awa from the iron surface or right next to it, where it can exert more of an influence on the .
-
jiperfections, localized stresses, orientation | ?tions in the environment. This condition is .1
.
At the anode, positive-charged iron atoms j
lid surface and enter the solution as positive * es, in the form of electrons are left behind
further progress of corrosion.
,
ie electrons meet and neutralize some posk hich have arrived at the surface through the
15-3 Factors Influencing CbrCbsion One of the most important factors in inf fluencing corrosion is the difference in electrical potential of dissimilar metals when coupled together and immersed in an electrolyte. This poten-
'
rge, the positive ions become neutral atoms drogen gas. So as this process continues,
tial is due to the chemical natures of the anodic and cathodic regions. Some
,
n occurs at the anodes, and plating out of J
indication of which metals may be anodic as compared with hydrogen is
ides. The amount of metal which dissolves J of electrons flowing which in turn is depen- f ;
given by the standard electromotive-force series (Table 15-1). The stan-
"
v
-
-
,
i
e resistance of the metal
.
inue it is necessary to remove the corrosion | Vne cathode.
In sorrie cases the evolution i lode is very slow and the accumulation of a ,
,
This is known as | 2). However, oxygen dissolved in the elec- | ated hydrogen to form water, thus allowing ' %.
I slows down the reaction
.
.
Ull
H OH
pn and water, the rate of film removal der
/ration of dissolved oxygen in water adjacent 3 concentration motion
,
in turn, depends upon the temperature, presence of dissolved ,
tFig. 15-3
Formation of ferrous hydroxide in the rusting of
bron. (The International Nickel Company.) i
1
M
m
586
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 15-1
Electromotive-force Series*
ELECTRODE REACTION
STANDARD
STANDARD
ELECTRODE
ELECTRODE
POTENTIAL £",
ELECTRODE REACTION
VOLTS, aS'C
K = K+ + e-
Co = Co++ + 2eNi =Ni++ +2e~
2 712
Sn =Sn++ +2e-
2 34
Pb = Pb++ + 2e-
1 70
V2H2 = H' + e
1 67
Cu = Cu++ + 2e-
0 345
.
Na = Na+ + e"
-
.
Mg = Mg++ + 2e-
-
.
Be = Be++ + 2e-
-
.
Al = Al3+ + 3e-
-
.
Mn == Mn++ +2e-
Cr = Cr3+ + 3e-
-
-
Fe = Fe++ + 2e-
-
.
-
0 126 .
0 000
0 800
0 71
Pd = Pd++ + 2e-
0 83
0 52
Hg = Hg++ + 2e-
0 854
0 440
Pt = Pt++ + 2e-
12
0 402
Au = Au34 + 3e
1 42
0 340
Au = Au+ + e
1 68
.
the formation of oxides and as a ca tion removes metallic ions from tf The effect of oxide film on the mi oxygen acts to remove hydrogen fr
.
0 522
,
The effect of dissolved oxygen or
.
Cu =Cu+ +e-
.
Ga = Ga3+ + Se"
0 136
-
,
to the corroding solution complete
counts for their superior corrosion coating tends to increase corrosioi to alternating periods of immersion
0 250 .
-
If the metallic ion is removed by tl whichjs precipitated on the anode i Oxide films of this type are formed i
0 277
Ag = Ag+ + e"
.
I
.
-
1 05 .
-
-
0 762
-
Zn=Zn+++2e-
POTENTIAL £0,
2 922 .
-
.
VOLTS, 250C
2 87
-
Ca = Ca++ + 2e-
tion is increased relative to the rec a reduction in potential
.
.
.
.
"
.
Cd = Cd++ + 2e-
-
.
In = in34- +3e-
-
.
Tl =TI+ +e-
-
be increased The effectiveness o fluenced by the amount of cathode ;
.
.
.
0 336
TABLE 15-2
.
Courtesy of The International Nickel Company
Galvanic Series of M(
Anodic (Corroded) End
.
Magnesium Zinc
dard hydrogen cell is assigned a value of zero, and the potential developed by a half cell of the metal in question coupled to a standard half cell is compared with that of the hydrogen cell. This listing in Table 15-1 is in decreasing order of activity. The more active metals at the.top of the list ex- * hibit a stronger tendency to dissolve than those at the bottom. A metal i
Aluminum
Cadmium
Aluminum alloys Low steel
Alloy steel Cast iron
higher in the series will displace a metal lower in the series from solution, j The electromotive series holds only for metals under conditions for which
Stainless steel (active)
the series was determined. The electrolytes contained particular concen-
Yellow brass
'
trations of salts of the same metal that was heing studied. Under actual conditions, in other electrolytes, their behavior may be different. Instead of the electromotive series, a somewhat similar galvanic series is used
which is based on experience with combinations of metals in a great variety of environments. Table 15-2 gives such a series for a number of metals and alloys in sea.water moving at high velocity. In any couple, the metal near the top of this series will be anodic and sgffer corrosion, while the one nearer the bottom will be cathodic and receive some galvanic protection,; The difference in electrical potential between twp metals is related to distance between them in the galvanic series. A metal coupled with another close to it on this list will usually corrode mors slowly than when coupled
Muntz metal
Aluminum breiss Red brass
Copper Aluminum bro'pze
Copper-nickel alloys Monel
Nickel (passive) Inconel
Silver
Stainless steel (passive) Titanium
Gold Platinum
.
with one further below it.
Catholic (Protected) End
The relative concentration of both ions involved in the reaction has
.
definite influence on the electrical potential. If the metaliic-ion concentri '
/
j
.
J
Courtesy
of The International Nickel Comp
1
: IRGY
CORROSION OF METALS
587 c
tion is increased relative to the reducible ion concentration, there will be
a reduction in potential.
STANDARD
ELECTRODE REACTION
If the metallic ion is removed by the formation of an insoluble compound i
ELECTRODE
which is precipitated on the anode, and this filrnis adherent and impervious |
POTENTIAL E"
,
to the orroding solution, complete insulition results and orrosLoa stops. \
VOLTS 250C ,
I
Co =Co++ + 2eNi = Nl++ + 2e-
Pd = Pd" -I 2fe
0 83 0 854
oxygen acts to remove hydrogen from around the cathode, corrosion will
Pt = Pt++ + 2e-
12
Au = Au3+ + 3e-
1 42
Au = Au+ + e-
1 68
VzHs = H' + e"
v
.
.
;
.
Hg = Hg'1 + 2e
Pb = Pb++ + 2e-
.
Oxide films of this type are formed on aluminum ancfcfirorniunri, which ac- 1
0 277
counts for their superior corrosion resistance. A porous oxide or metallic coating tends to increase corrosion, especially when- the part is exposed to alternating periods of itW feron ancf drying. The effect of dissolved oxygen on the corrosion rate is twofold: it acts in the formation of oxides and as a cathodic depolarizer. If the oxide fprmation removes metallic ions from the metal, corrosion will bejnqreased. The effect of oxide film on the metal was mentioned previously, if the
Sn = Sn++ + 2e-
.
-
-
0 250 .
0 136
-
.
0 126
'
-
.
-
v
0 000 .
Cu = Cu++ + 2e-
0 345
Cu = Cu+ + e-
0 522
Ag = Ag+ + e-
0 800
.
.
.
.
.
be increased. The effectiveness of oxygen in fertioving hydrogen is in-
.
'
.
fluenced by the amount of cathode area. With a large cathode the hydro-
.
y "j
.
TABLE 15-2
Galvanic Series of Metals and Alloys In Sea Water*
Anodic (Corroded) End
Magnesium !a value of zero
Zinc ,
and the potential developed
pstion coupled to a standard half cell is com-
Cadmium
Jfi cell. This listing in Table 15-1 is in de-:|R
Aluminum alloys
more active metals at the -
Aluminum
.
top of the list flx- ;.
dissolve than those at the bottom
.
Low steel
A metafMp
ie a metal lower in the series from solution il
s only for metals under conditions for which5 L, e electrolytes contained particular concen- jwp etal that was'being. studied. Under actual 1m :
Muntz metal Yellow brass
'
s,
Aluminum brsiss Red brass
their behavior may be different. Instead
Goppor
somowhul slmllnt galvanic series Is used
Aluminum bronze
i\th combinations of metals in a great variety
Copper-nickel alloys
ues such a series for a number of metala and|k iigh veloelly. In any couple
:
Ihe metal nMr||B;
,
inodic and suffer corrosion while the one mm,
f ..o6\c and receive some galvanic protection. ential between two metals is related to dis- m
i/anic series. A metal coupled with another W
.
0:
1 iy corrode more slowly than when coupled
Nickel (passive) Inconel Silver
Stainless steel (passive) ,
Titanium
Gold
Catholic (Protected) End .
,
cal potential. If the metallic-ion concentra- m
f
i
Monol
Platinum
f both ions involved in the reaction has a i
m
Cast iron
Stainless steel (active)
.
:
Alloy steel
'
Courtesy of The International Nickel Company.
i
588
INTRODUCTION TO PHYSICAL METALLURGY
gen that reaches it will spread out
by reaction with oxygen. This is \ "
cathode with a small anode. TRis
ample. If steel plates are joined by for several months the copper m ,
there will be no significant accele the rivets. If however, copper pi ,
mersed under the same condition
severely (Fig. 15-4). Agitation acts to increase the co
1 i
x
i
m
solution into contact with the meta
m
cause corrosion known as local a
BTfferences in potential from pc
5%
the surface or differences in surfact in environment such as the differ ,
the corroding solution at one poi with another point on the metal action. V.'
.
.
,
,
{
point than at another.
: ::\:'\'-
:
'
-
:
-
This difference in metal-io
metal Is in contact with a solutioi
5
>
This situate
.; .
.
m l7
Si 3
'
3
i '
;
'
v.,
s
.
V
ft 1
fl
'
:
>-
.
vsi-
. .,
:
Fig. 15-4 Influence of area relationship between cathode and anode illustrated by copper-steel couples after immer
-
. .
. -
" .
sion in sea water, (a) Copper rivets with small area in steel plates of large area have caused only slight increase in corrosion of steel, (b) Steel rivets with small area in
copper plates of large area have caused severe corrosion of steel rivets. (The International Nickel Company.)
.
Hi
-
jTMW>H l -
~
-y.J-l"- '
"
"
ir-'
15
l
v
Fig. 15-5 Severe corrosion in region of high velocity oi admiralty-brass disk after rotation in sea water. (The flnternational Nickel Company.)
IHGY
CORROSION OF METALS
589
gen that reaches it will spread out and will be more accessible for removal by reaction with oxygen. This is why it is poor practice to couple a large "
cathode with a small anode. Tfiis may be illustrated by the following example. If steel plates are joined by copper rivets and immersed in sea water for several months, the copper rivets will remain in good condition and there will be no significant acceleration of the corrosion of the steel near
A
m lis
the rivets. If, however, copper plates are joined by steel rivets and immersed under the same Conditions, the steel rivets wIN be attacked very severely (Fig. 15-4). Agitation acts to increase the corrosion rate by bringing fresh corroding ; solution Into contact with the metal. "
Slfferences in potential from point to point on a single metal surface
cause corrosion known as local action and may be due to impurities on the surface or differences in surface structure or environment. A difference 5;
In environment, such as the difference in concentration of metal ions in
the corroding solution at one point on the metal surface as compared with another point on the metal surface, will cause' corrosion by local action.
This difference In metal-ion concentration can be set up when a metal is in contact with a solution where the velocity is greater at one
point than at another. This situation can be created by spinning a metal
m 1!
i I.
i '
v
!
m L
m
cathode jr immer;a
in steel
isc in in
irrosion of
\: Fig, 15-5 Sovere corrosion in region of high velocity on an admiralty-brass disk after rotation In sea water. (The
£lnternationai Nickel Company.)
I INTRODUCTION TO'. PHYSICAL METALLURGY
590
i disk througH| salt water. Since the metal nearer the center of the disk
moves more Slowly than at the edge, this allows metal ions to accumulate near the center and be swept away near the edge. At the edge, the region i if of highest velocity, metal-ion concentration will be the least, and severe i if A x., :_ . r-> .u. \ i corrosion wiJJ take place in this region (Fig. 15-5). For this application, a metal must bl,e chosen that will be able to hold its protective film right up | to the outer edge under these conditions. i Other factors-such as the presence of other ions in solution, the tem'
..
'
-
f; ii'.' .
! .''V
'
'
SI
'
[ perature of the solution, and the existence of stray electric currents-may Wnater'a''y afffect the corros'on rate.
:
15-4 Spec,f,c Corrdjs,on Types Spec,f,c descr,pt,ons are genera,,y used for
F,g. 15-7
certain types bf industrially important corrosion. When the entire surface of
P,tt,ng corros,on due to mar,ne organ,sms
metal was immersed in sea wafer. (The International Nickel Company.)
the metal is attacked to the same degree, it is known as uniform corrosion. 3 '
This type is qnusual in metals, since they are rarely so homogeneous that the surface will be evenly corroded.
Pitting corrosion is an example of nonuniform corrosion resulting from ; ] inhomogeneities in metal due to inclusions, coring, and distorted zones. I i
surface resulting from localized c(
These inhomogeneities set up differences of potential at localized spots to j I-
base metal occurs when there is a b
cause deep isolated holes. Figure 15-6 shows an electrolytically formed 1
example, when the chromium plate of the exposed steel takes place.
pit. The Iarg6 pit formed when the surface was penetrated in a small area
became attached to the surface whi
1
then grew rapidly into a large cavity under the surface. Progressive growth : of the cavity caused further penetration of the surface from below. Part of ;
Cavitation corrosion illustrated s ,
the collapse of bubbles and cavitiei tween a surface and a liquid is sucf surface, causing very high stresses regularly. These collapses produc remove particles of the surface, eve and pockmarks. Figure 15-9 show:
the roof of the cavity then collapsed. Figure 15-7 shows pitting on a metal , t
mam A i
tation corrosion on the surface of a
£.1
may be minimized or eliminated by or by using a protective coating. I
mm. m
Si
some stainless steels, and certain tc s
1
Liquid mi
m
» '
Stru clure : i ' -
iff
v
-
..
Ms %
m
'
mm Fig. 15-6
m
a
|*Fig 15-8 Cavitation corrosion caused by collapsing ac |of bubbles which form at points where the local pressu
.
-
>
!'
m
ma
tis equal to or below the vapor pressure of the liquid. (I jMaterials Engineering Special Report No. 202, January ,
Electrolitically formed pit, 350x. (M. A. Streicher,
1963.)
Du Pont Research Laborajtory.) .
i
I
A
J
i
\ l:
/
™.
J
1
CORROSION OF METALS
JRGY
591
ce the metal nearer the center of the di8R|
edge, this allows metal ions to accumulate|
Iplaway near the edge. At the edge, the region concentration will be the least
,
and several
is region (Fig. 15-5). For this application II be able to hold its protective film right
,
"
conditions.
m
the tem|| ie existence of stray electric currents-mayll rate' If Dresence of other ions in solution
.
if
,
?j:;;:5cific descriptions are generally used ftjg : : '
-
ig. 15-7 Pitting corrosion due to marine organisms while netal was immersed in sea water. (The International
ortant corrosion. When the entire surface ofl
*;;
pe degree, it is known as uniform corrosion.J§ since they ar$ 3ded.
"
llckel Company.)
rarely so homogeneous thdj
pie of nonuniform corrosion resulting frortilj
surface resulting from localized corrosion under marine organisms that
$:-}fato inclusions, coring, and distorted zones.%
became attached to the surface while immersed in sea water. Pitting of the
differences of potential at localized spots t<3 gure 15'6 shows an electrolytically formed
example, when the chromium plate in a steel auto bumper is broken, pitting
s v) the surface was penetrated in a small arftraBp Mty under the surface. Progressive growth
of the exposed steel takes place. v Cavitation corrosion, illustrated schematically in Figi 15-8, is caused by
Part of..-'«!
the collapse of bubbles and cavities within a liquid. Vibrating motion between a surface and a liquid is such that repeated loads are applied to the
letration of the surface from below
.
ipsed
.
|
Figure 15-7 shows pitting on a metal
i I
-
m
base metal occurs when there is a break in the protecting layer orfilm. For
surface, causing very high stresses when these bubbles form and collapse regularly. These collapses produce high stress impacts which gradually remove particles of the surface, eventually forming deep pits, depressions, and pockmarks. Figure 15-9 shows numerous small pits formed by cavi-
I
m
i
I
tation corrosion on the surface of a cast-iron sleeve. This type of corrosion
may be minimized or eliminated by switching to a more resistant material or by using a protective coating. In general, aluminum bronzes, Stellite,
I
some stainless steels, and certain tool steels have good resistance to cavi-
f Liquid
-
-
9 '
Slcu clure
-
.
Fig. 15-8 Cavitation corrosion caused by collapsing action ol bubbles which form at points where the local pressure |;ls equal to or below the vapor pressure of the liquid. (From Streicher,
|Materials Engineering, Special Report No. 202, January 1963 ) .
| i
11
592 INTRODUCTION TO) PHYSICAL METALLURGY
. p
m1 :; -
iii
"
Liquid
Relatively higher oxygen
\
concentration
,
,
.
,..
-
-
. .
fry!1
Low oxygen concentration area
'
lo) Oxygen concentration type r
Fig. 15-10 Crevice corrosion caused by: (a) differenc oxygen concentration and (£>) difference in metal-ion
centration. Notice that in each case corrosion takes [ where there is a deficiency of either oxygen or metal (From Materials Engineering, Special Report No. 202
5
,.
:x
v
I
,
January 1963.)
Fretting corrosion is a common I Fig. 15-9
1 In
vibration which results in striking fitting highly loaded surfaces. Sui clamped or press fits splines, keyv ject to minute relative movement destroys dimensions and reduces f ,
Cavitation corrosion in the form of numerous
small pits in local areas cjii the surface of a cylinder sleeve
,
of gray cast iron. (Courtesy of D. J. Wulpi, International
Harvester Company.)
1}
,
is a mechanical-chemical phenome tation damage. Materials such as cast iron, bronze and steel castings, and
gether, adhesive forces cause small
steel plate hpve relatively poor resistance to cavitation damage, but they
continued slight motion the weldec
'
,
can be protefted by welded layers, sprayed-metal coatings, and some nonmetallic Goatlings.
Crevice corrosion is a general term including accelerated attack at the I junction of two metals exposed to a corrosive environment. We know from :s
experience that corrosion is more likely to occur in crevices which retain »
surfaces and react chemically with t der in the joint. Figure 15-11 show oil pump gear during fatigue testi fretting corrosion may be overcome the source of vibration by tighter cl;
solutions and take longer to dry out. It is also possible for corrosion to j occur at crevices even when they are completely immersed. Accelerated attack can opcur because of a differential in oxygen concentration (Fig.
15-10a). Oxypen has relatively easy access to the outside of the joint, which j f
is cathodic. JThe metal in the joint is relatively anodic. Thejdeposit of in- j j?
soluble corrosion product around the anodic center tends to more com- | ;
pletely exclude oxygen, resulting in a low oxygen concentration area and j \ increased electrical potential. If the action continues, a pit forms in the \ i
ft
center. Corrosion always occurs in the region of oxygen deficiency. Crev- j \
.
;
r '
ices can also lead to differences in metal-ion concentrations at different ]
3
locations.
For example, it is possible for an area in a crevice to have a '
1
'
higher metaljic-ion concentration than the area outside. Thus, corrosion
can take place in the region of metal-ion deficiency at the edge of a mechanical joint, as illustrated in Fig. 15-1 Ob. The best way to avoid this type of corrosion is to eliminate crevices entirely by changing the design or filling in joints that are liable to cause trouble.
.1
r
IS /
Figure. 15.11
Fretting corrosion of the shaft of an oil-p
drive gear during fatigue testing (Courtesy of D. J. Wul International Harvester Company.) .
3
; .
URGY
CORRpSION OF METALS
1 iquid
Heldlivcly higher oxygen concentrotion
593
Liquid'
Areo ol rPkilivcly lower" melol ion concentratioji
"
-:-
-
5
-
z
Areaiof higher metal --S.-_
Low oxygen "'
v
-
"
5 .
' -
<
'
.» ».
- ion concentration
concentration oreo
(n) Oxygen concentrotion type
*
[b] Metal ion concentrotion cell type
Fig. 15 10 Crevice corrosion caused by: (a) diflcrence in oxygen concentration and (b) difference in metal-ion concentration. Notice that in each case corrosion takes place
!.«.. « '
where there is a deficiency of either oxygen or metal ions. (From Materials Engineering, Special Report No. 202, January 1963.)
Fretting corrosion is a common type of surface damage produced by vibration which results in striking or rubbing at the interface of close-
fitting, highly loaded surfaces. Such corrosion is common at surfaces of clamped or press fits, splines, keyways, and other close-fitting parts sub-
jmerous
ider sleeve ' .
.
.
I
.
.
.
innlionnl
jocl lo minute rolnlivo movomonl. Frolting corrosiop ruins bearings, destroys dimensions, and reduces fatigue strength. This type of corrosion is a mechanical-chemical phenomenon.
as cast iron, bronze and steel castings and 1 r resistance to cavitation damage but they :. ers, sprayed-metal coatings and some non- ' ]
I to a corrosive environment. We know from
tore likely to occur in crevices which retain l
the source of vibration by tighter clamping or more rigid mounting. Other
,
,
I ral term including accelerated attack .
-
at the
iry out. It is also possible for corrosion to hey are completely immersed. Accelerated
:
differential in oxygen concentration (Fig.
basy access fo the outside of the joint '
When two components rub to-
gether, adhesive forces cause small particles of the surface to weld. With continued slight motion, the welded particles tear away from the opposing surfaces and react chemically with the atmosphere, forming debris or powder in the joint. Figure 15 11 shows fretting corrosion on the shaft of an oil pump gear during fatigue testing. There are several ways in which fretting corrosion may bo overcome. The most obvious way is to remove
,
;
which
;
oint is relatively anodic. The deposit of inund the anodic center tends to more com-
i
,
.
ng in a low oxygen concentration area and I A t-
If the action continues
,
a pit forms in the :.
rs in the region of oxygen deficiency Crev- j
;vi
.
;es in metal-ion concentrations at different
y-v.
.possible for an area in ..
;
,
a crevice to have a j (s
jon than the area outside. Thus, corrosion
f
mm
£4$ metal-ion deficiency at the edge of a me- V
'
wiFig. 15-105. The best way to avoid this type % vices entirely by changing the design or by.m ure- 15 Frf ;ng c°rrfon » a a ' r;Jp- drive gear during fatigue testing. '
'
.
,
) cause trouble.
H
'
International Harvester Company.)
shaft of an oil-pump
(Courtesy of D. J. Wulpi,
594
INTRODUCTION TO PHYSICAL METALLURGY
i
methods include raising hardnesses of mating surfaces, inserting rubber gaskets in joipts (to absorb motion), lubricating with a dry medium (molybdenum disulfide held in a binder), and sealing the entire area with a ma-
?<31
terial such as rubber cement to exclude atmospheres. Strengthening treat- j 1
ments such ab nitrlding, shot peening, surface rolling, chromium plating, or
flame and ineluctlon hardening also lessen fracture resulting from fretting in shafts.
/ '
' ' .
'
.
.
li
-
r \
Intergranutpr corrosion is another example of nonuniform corrosion when a potential difference exists between the grain boundaries and the rest of the alloy. This type of corrosion usually takes place when precipitation of a phase from a solid solution occurs. Since precipitation usually takes place faster at the grain boundaries, the material in the vicinity of the grain boundary becomes depleted of the dissolved element, creating a difference of ;potential and the grain boundary will dissolve preferentially (Fig. 15-12). Often a visual examination of the part will not reveal the extent of the damag 0 and in most cases there is an appreciable loss in mechanical properties. I Stress corkosion is acceleration of corrosion in certain environments Fig. 15-13 Stress-corrosion cracking in type 304 aust when metalsj are externally stressed or contain internal tensile stresses stainless steel. A bend was exposed to chloride-cont due to cold working (see Sec. 12-6). The cracks may be transgranular or 2 ; water, 250x. (M. A. Streicher Du Pont Research Laboratory.) intergranular or a combination of both. The magnitude of stress necessary ,
'
,
c
,
to cause faildre depends on the corrosive medium and. on the structure of i
Stress corrosion
.
corrosion because it can occur in ! be attacked in certain environments in one metal will not cause crackin diet where attack will occur The
9 v
the base metal
,
.
3
: .
.
Steels containin
.
stress corrosion because the alum
s
S v
tends to make them more prone nitrate solutions
.
.
aluminum nitride. Some stainless st
-
cracking in the presence of chloride
__
V
4
calcium chloride
and several oth
,
been recognized as the major cau in processes involving a chloride-c shows a stress-corrosion crack in t
;
m
i
m
Fig. 15-12 Intergranular corrosion in type 316 stainless steel after a 27-h exposure to boiling sulfate-sulfuric acid solution, 500x. (M. A. Streicher, Du Pont Research :i Laboratory.) '
H
11
f -i-
1
m
\ i
_
though they will pit badly in the pr less steels are less likely to fail froi martensitic grades
.
Stress-corrosio
copper alloys when they are expose pecially in the presence of oxygen cracking can be minimized by avoid five coatings and, in the case of br ,
rgy
CORROSION OF METALS
IP
595
p
.
.
,,
jesses of mating surfaces, inserting rubber f Jion), lubricating with a dry medium (molyb-:i
V
,
t jler), and sealing the entire area with a ma<| i exclude atmospheres
.
s
Strengthening treat-'I
r yi 1
9ening, surface rolling chromium plating, or ,
i?
i also lessen fracture resulting from fretting |
!I
n
7.
mother example of nonuniform corrosion %
U ists between the grain boundaries and the|
rrosion usually takes place when precipita*|i lution occurs. Since precipitation usuallj f boundaries, the material in the vicinity o j spleted of the .dissolved element, creating tjkm \ grain boundary will dissolve preferentially Sp )
i
vo
0
,
L
Tiination of the part will not reveal the exten|||B| ;
s there is an appreciable loss in mechanlcaf K1
'
;
'
v
k
>
Stress-corrosion cracking in type 304 austenitic
|water, 250x. (M. A. Streicher, Du Pdnt Research
bf both. The magnitude of stress necessaiya laboratory.)
"
"
V
|stainless steel. A bend was exposed to chloride-containing
f
e jfl2. 6). The cracks, may be transgranular of'
r '
I-
essed or contain internal tensile Stresses W9: 15 13
.
,
IS
,
i Mon of corrosion in certain environments iP -
'
'
\ corrosive medium a'nd, on the structure of > '
the base metal. Stress corrosion is one of the most important types of corrosion because it can occur in so many metals. Almost any metal can be attacked in certain environments, yet the conditions that cause cracking in one metal will not cause cracking in another. Thus, it is difficult to pre-
9
i
dict where attack will occur. The presence of nitrogen in iron and steel
s
tends to make them more prone to stress-corrosion cracking in some nitrate solutions. Steels containing aluminum have better resistance to stress corrosion because the aluminum combines with nitrogen to form
aluminum nitride. Some stainless steels are susceptible to stress-corrosion
cracking in the presence of chlorides, e.g., in solutions of sodium chlorde, calcium chloride, and several others. Stress-corrosion cracking has been recognized as the major cause of austenitic stainless steel failure
am
in processes involving a chloride-containing environment. Figure 15-13 shows a stress-corrosion crack in type 304 austenitic stainless steel. Although they will pit badly in the presence of chlorides, the ferritic stainless steels are less likely to fail from stress corrosion than the austenitic
».
0
ainless
|uric acid
f
martensitic grades. Stress-corrosion cracking can also occur in stressed copper alloys when they are exposed to ammonia and its compounds especially in the presence of oxygen and carbon dioxide. The danger of cracking can be minimized by avoiding residual stresses, by using protec,
tive coatings, and, in the case of brasses, by keeping zinc content below
>v
- '-.
.
!
i!
" .
' -
'
<
' ' * .
1
"
1*
596
INTRODUCTION TO-;PHYSICAL METALLURGY
{
TABLE 15-3
A type of corrosion that has be
Effect of C6rro3\on Type on Properties*
Pitting Intergranular
07
5
7
5
15
02
15
25
20
80
metal corrosion. In certain types o' atomic power, liquid metals such a heat-transfer medium. The path of one leg at high temperature in the i temperature in a heat exchanger usually increases with temperature solid to dissolve up to its solubility li
Strers
0 1
100
100
100
100
deposited because of the lower sol
LOSS OF PROPERTIES, % DEPTH OF
WEIGHT loss;
TYPE
%
PENETRATION, .
i
%
YIELD
TENSILE STRENGTH
STRENGTH
ELONGATION
.
Uniform
1 .
.
.
1
1
1
.
1
.
,
leg is continually corroded
,
By permission from L. F. Mondolfo and O. Zmeskal, "Engineering Metallurgy, McGraw-Hill Book Company, New York, "
*
and th
deposited cfi bsion products
1965.
Th
.
mass transfer which leads to gradu zone.
15 percent. Brasses containing 20 to 40 percent zinc are highly suscep-
by use of inhibitors in the liquid me fective inhibitor in liquid bismuth t
tible to attack.
Intergranular corrosion and stress corrosion have a very serious effect on the mechanical properties of the metal. The reduction in strength is
of iron by mass transfer. 15
not due to the amount of metal removed but rather to the stress concentra-
tion produced by the fine cracks. Table 15-3 gives an idea of the effect of .
*
'
* .
'
.|
: : :V
.
-
.
"
v
-
.
.
the type of corrosion on properties.
5 Methods for Combating Corrosion to prevent corrosion by selection o surface protection of a given materi
Preferential corrosion of one of the components may even occur in
1
Use of high-purity metals
single-phase solid-solution alloys. Dezincification in brass (see Sec. 12-6)
2
Use of alloy additions Use of special heat treatments Proper design Cathodic protection
'
]
,
The most effective method (
.
.
3
'is an exampl| of this kind of corrosion (Fig. 12-5).
.
4
.
Galvanic corrosion occurs at the interface where two metals are in con-
5
.
tact in a corroding medium. This type of corrosion, illustrated in Fig. 1514,
6
Use of inhibitors
7
Surface coatings
.
was discusseid in Sec. 15-2.
.
] si
In most cases the use of high-| corrosion by minimizing inhomoge ,
resistance.
1
Alloy additions may reduce corros
'
V
V;.
austenitic stainless steels when coi
.
,
about 900 to 1400oF
1 .
"
*
1
;
-
-
precipitate chroi
This precipitation depletes the bbun more susceptible to intergranular ci
be avoided either by reducing the c cent) or by converting the carbide to is more widely used and involves th
..1 -
-
.
,
These elements have a great affinity
\
bides that are not soluble in austenit(
very little carbon available for comb
1
what is known as a stabilized staihles Fig. 15-14
Galvanic corrdslon of magnesium where it is in
close contact with a steel icore around which the mag-
nesium was cast. (The International Nickel Company.)
i
r
/ !
I
i
corrosion resistance by forming
,
or
nonporous surface oxide films This .
-"
URGY
CORROSION OF METALS
m
jpertles* LOSS OF PROPERTIES % ,
)N,
,
TENSILE
YIELD
STRENGTH
STRENGTH
1
1
1
7
5
15
25
20
80
100
100
100
Engineering Metallurgy
.
:
ELONGATION
'
McGraw-Hill Book
597
A type of corrosion that has become increasingly important \s liquidmetal corrosion. In certain types of nuclear reactors for the production of atomic power, liquid metals such as bismuth and sodium are used as the heat-transfer medium. The path of the liquid metal is a closed loop with
one leg at high temperature in the reactor core and the other leg at loweH
temperature in a heat exchanger. The solubility of a solid in the iftjlfflp usually increases with temperature. Therefore, there is a tendency for fffisp
5
solid to dissolve up to its solubility limit in the high-temperature leg and be deposited, because of the lower solubility limit, in the cooler leg. The hot
Company
.
leg is continually corroded, and the cold leg becomeb plugged with the
New York.
g 20 to 40 percent zinc are highly suscep*
deposited c4(m)siOn products. This phenomenon is essentially one of mass transfer which leads to gradual deterioration of the metal in the hot '
zone. The most effective method of controlling this type of corrosion is by use of inhibitors in the liquid melal. Zirconium has been used as an effective inhibitor in liquid bismuth to decrease the liquid-metal corrosion
i
stress corrosion have a very serious effect j
of the metal. The reduction in strength is | 155
romovod Ixil rnlhor lo Iho slross conoontra-
;! i
V0e of the components may even occur in -j jys. Dezincification in brass (see Sec. 12-6) S prrosion (Fig. 12-5). J the interfaco whero two metals are in con-
'
1
.
2
.
3
.
'
I
5
.
is lype ol corrosion illustrated in Fig. 15-14, ,
Use of high-purity metals Use of alloy additions Use of special heat treatments Proper dnsifjn Oalhodic piolccllon
6
Use of inhibitors
7
Surface coatings
.
i
Mirny m. lluuls an') ufuxl iiKlusliially
to prevent corrosion by selection of the proper alloy and structure, or by surface protection of a given material. The most important are:
s Sks. Table 15'3 gives an idea of the effectt)f j Vties.
of iron by mass transfer. Methods for Combnling Corroftion
.
\
In most cases, the use of high-purity metals tends to reduce pitting 1 corrosion by minimizing inhomogeneities, thereby improving corrosion
j
resistance.
Alloy additions may reduce corrosion by several methods. For example, austonilic stainless steels, when cooled through a temperature range of
about 900 to 1400 F, precipitate chromium carbides at the grain boundaries. This precipitation depletes the boundaries of chromium and makes them
more susceptible to inlergranular corrosion. This lype of corrosion rray / be avoided either by reducing the carbon to a low value (below 0.03 per- / cent) or by converting the carbide to a more stable form. The latter method ' is more widely used and involves the addition of titanium or columbium. These elements have a great affinity for carbon, producing
very stable car- i [
.
bides that are not soluble in austenite at elevated temperature. This leaves | j very little carbon available for combination with chromium and results in
1
sre il is In
what is known as a stabilized stainless steel. Some alloy additions improve j
jany.)
nonporous surface oxide films. This is particularly true of manganese and
nag-
corrosion resistance by forming, or helping the formation of, adherent, | i
i
1
"
'
598
INTRODUCTION TO PHYSICAL METALLURGY
aluminum additions to copper alloys, molybdenum additions to stainless
Anodic oret crevices
steels, and magnesium additions to aluminum.
|
.
Heat treatment which leads to homogenization of solid solutions, es- :
il
J pecially in cast alloys that are subject to coring, tends to improve corro- & sion resistance. Stress-relief treatments following cold working are widely i f.
m
used to improve the resistance of alloys susceptible to stress corrosion.
Proper design should keep contact with the corrpding agent to a mini- ; mum. Joinfe should be properly designed to reduce the tendency for ]
liquids to erfter and be retained. Contact between materials far apart in"! ;
.
j the electromfotive series should be avoided. If this cannot be done, they -
ia) Bad
I should be separated by rubber or plastic to reduce the possibility of gal-j I
H
i vanic corrosion. For example, Fig. 15-15a shows two cases of galvanic j
[ qorrosion wiien using dissimilar metals. Since aluminum is the anode with | t respect to steel, corrosion of an aluminum rivet can be expected when it isM j used to fasten steel sheets together. Similarly, if a steel rivet is used to 1' '
/ fasten alumihum sheets, then undercutting galvanic corrosion of the alu-1
[6] Good
ilminum sheel will result in loose rivets, slipping, and possible structural! -
Idamage. This type of corrosion can be prevented by applying a nonhard-1 l
.
ZZ
TZT
ning insulating joint compound in the area vyhere the sheet and the rivet -j | to fasten the plates by welding. (From Materials Engin ! or bolt are in contact, or by applying a zinc chromate primer to all con- j j in9' sPecial Report No. 202, January 1963.) '
tacting surfaces and then coating the primed area with an aluminum paint.
; <
Where the fasteners are not subject to high stresses, the contact points | can be insuldted with plastics or other nonmetallic sleeves, shims, washers, ? and similar piarts, as shown in Fig. 15-15/?.
It was pointed out earlier that cre> They are fre as shown in Fig 15-16a, where two p
tration-cell corrosion
.
.
Aluminum
Aluminum rivet (corrodes)
Sfee! rivet
1
solution. No matter how much torqu impossible to eliminate crevices int Cre' instead of mechanical fasteners as sulating gaskets between surfaces tf
MTlp
trates and becomes stagnant. .
,
Undercutting
[a) Bod
Cathodic protection is obtained t Insuloling washer
Stee1-
Copper
Insulating gosket
-
mally corrode in electrical contact w series The more active metal thus b a galvanic battery in which the corrx! .
.
Aluminum
1
Insulating washer
Insuloting bushing [b) Good
Fig. 15-15 Design suggestion to minimize corrosion when fastening two dissimilar metals, (a) Corrosion of aluminum
.
rivets in steel plates or aluminum plates when fastened with a steel rivet. (t>) Recommended practice of using insulating materials between the steel bolt and copper and aluminum plates. (From Materials Engineering, Special
jtPeport No. 202, January 1963.) i
;.:
.
f i
! 1
mi
I
,
cathode. The metals generally used zinc and magnesium In some case: tained by an external voltage source .
,.
.
sists of a relativelyjnert material su The structures most frequently protei pipelines, hulls of ships, and boijers, pipe, anodes arCTauried some 8 to 1 hole should be sufficient to locate t
jURQY
CORROSION OF METALS
er alloys, molybdenum additions to stainlessl
599
Anodic tireos in
jns to aluminum.
crevices
IvSjs to homogenization of solid solutions, es-f
i
e subject to coring tends to improve corrO'? ,
itreatments following cold working are widely ,
p of alloys susceptible to stress corrosion. contact with the corroding agent to a mini-|
.
perly designed to reduce the tendency forf led. Contact between materials far apart iff| ild be avoided. .
:;
If this cannot be done
,
(ff) Bad
thej|
3r or plastic to reduce the possibility of gall
ggge, Fig. 15-15a shows two cases of galvanl f ar metals. Since aluminum is the anode witf|j
Weld
an aluminum rivet can be expected when it 19;
.
sgether. Similarly, if a steel rivet is used W; undercutting galvanic corrosion of the alui| Jose rivets, slipping, and possible structurl
[ti\ Good ii*.-
flg. 15-16 (a) Crevice corrosion between plates when
on can be prevented by applying a northaf|| nd In the area where the sheet and the rive*
.
ilted together, (b) Recommended (Jractice if possible, is ,
'
to fasten the plates by welding. (Frotn Materials Engineer-
applying a zinc chromate primer to all c6$$ ing, Special Report No. 202, January 1963.) ing the primed area with an aluminum painty
subject to high stresses, the contact pointsf ar other nonmetallic sleeves, shims
,
washers,!
It was pointed out earlier that crevices are a potential source of concentration-cell comosion. They are frequently encountered in sections such as shown in Fig. IS-16a, where two plates are bolted together in a corrosive solution. No matter how much torque is applied to the bolt, it is practically impossible to eliminate crevices into which the solution gradually pene-
Fig. 15-15i>.
"
Stec! tivel
trates and becomes stagnant. Crevices can be avoided by using wf lds 5>
Undercullmy
ling washer -
Insulating gasket
S3
instead of mechahldal fasteners, as shown in Fig. 15-166, orjjy using insulating gaskets between surfaces that are machined parallel. Cathodic protection is obtained by placing the metal that would normally corrode in electrical contact with one that is above it in the galvanic series
. ,
um
Insulalinq washer
-
The more active metal thus becomes the anode. This is essentially/..
a galvanic batteryTrV which the corrodjng metal is made to function as the/?11 cathode. The metals generally used to provide this type of protection are' In some cases, the protoctive direc.t.currentjs obtained by ari external voltage source. The anode in this case usually con- h zinc and magnesium.
psion when :
-
( aluminum astened
:: v
;
.
using copper and Special
.
sists of a relativelyjnert material such as carbon, graphite, or platinum. If
The structures most frequently protected by this meth6d are underground'
'
,
pipelines, hulls of ships, and boilers. For the protection of underground j! pipe, anodes artTBuried some 8 to 10 ft from the pipe. The depth of the
hole should be sufficient to locate the anode in permanently moist soilajiia
i
600
INTRODUCTION Tff) PHYSICAL METALLURGY
Individual a'nodes are connected to a collector wire which in turn is brazed
diffusion layer is properly controlle the operations of pickling with inh
'
to the pipelipel The current discharges from the anode to the soil, collects ; on the pipeline, and returns to the anode through the connecting wire, i
'
r-i- r v -
-
-
.
.
"
y
fluxing to facilitate wetting; dippii
For the catfiodic protection of ship hulls, zinc or magnesium anodes are i fastened to -the rudder and to the hull itself in the region around the pro- \
trolled composition; and wiping
,
protection fh domestic and industrial water heaters and elevated water- ; storage tanks.
of zinc coating on steel products
peller. Magpesium anodes have become widely used to provide cathodic , "
V
si
to regulate the thickness and unite Zinc coating or galvanizing is af ucts than any other method of mi
'
Inhibitors
,
are chemicals which, when added to the corrosive solution, ji
,
and wire products for all outdoor e
reduce or eliminate its corrosive effect. In most cases, the inhibitor will form a protective layer on the metal surface. Inhibitors are added to the
steel such as boilers pails, cans, ,
pipe and conduit; and exposed sti
antifreeze mixtures used in automobile radiators. Oxidizing agents when : added to the corrosive solution will produce oxide films on aluminum, chromium, and manganese. j Surface coatings include paints, salt and oxide films, and metallic coatings. Paints and other organic coatings are primarily used to improve the appearance of the surfaces and structures. The use of paint for corrosion , protection only is secondary and of little economic importance to the paint
in the familiar tin can.
Terne metJ
which tin has been added to ensu
zinc, but much cheaper than tin roofing material. Aluminizing hot used for applications that require a and heat such as mufflers and tail| Electroplated coatings are used ( for industrial applications the mos .
,
,
industry. Paint provides a protective film to the metal and is effective only as long as the film is unbroken. Salt and oxide films are obtained by reacting the metal with a solution =
rosion protection.
In addition to
electroplated coatings are applied 1 as wear resistance high electrical high or low light-reflecting ability wear resistance (see Chap. 14); sil ,
which produces the desired film. Some examples are: A chromate pickle [ ]
.
protects magnesium by forming a film of magnesium chromate, Parkerizing
or Bondeniing for ferrous alloys protects by forming a.phosphate film,
if t
'
'"
:
work oilTrbifi and steel because steel is cathodic to them and hence is protected electrochemically in spite of any porosity or minor voids in the
coating. Grit or sand blasting is almost universally used for cleaning and :S
preparing tine surface prior to spraying. Most coating systems employ-
jl supplementary organic sealers pr top coats. The sealers are usually vinyl ;:i 5U ppiei liei lie
if: chlorinatedjrubber applied over inhibiting primers.
[j/
Hot dippiijig is used mainly to apply a coating; of zinc, tin, cadmium,"|jLs aluminum, £r lead to steel. The hot-dip process has a wide range of ap-
plicability, lut the coating applied must contend with a brittle diffusion I
'
layer of int rmetallic compounds at the interface. This may result in pod
adhesion aijid a tendency to flake on bending, unless the thickness of thdl s
,
copper, tin,
Qhromium and rhodium have high r Zinc coatings are used on iron am corrosion protection. These includ and electrical conduit.
Metallizing was discussed in Chap. 14 in regard to surface coating for | wear resistance. Practically all metallized coatings for corrosion use are either zinc tor aluminum. These metals are used primarily for corrosion
H
conductivity; silver
anodizing for aluminum and magnesium forms a thick oxide film, and pas- li sivating for stainless steels also forms an oxide film. Metallic coatings may be obtained by a variety of methods such as ' metallizing,:hot dipping, electroplating, diffusion, and cladding.
Cadmium ci
l coatings but are not quite so good f ;
is also used for radio chassis and i
s lderable. Chromium coatings are '
tures, hardware, and appliances. 7 porous and normally offer little pre they are usually applied over thicke coatings are used principally as an and rhodium plating. Nickel has go(
by the atmosphere. Copper coatings chromium coatings particularly'On use of electroplated tin coatings is fc be made thinner than those made bj ,
The use of chromium and silicon a
has been discussed in Chap
.
11 '
-
I!-!'.
j
/
.
14.
/:
m
IGY
CORROSION OF METALS
d to a collector wire which in turn is br
charges from the anode to the soli, collet the anode through the connecting wmu ship hulls, zinc or magnesium anodes the hull itself in the region around thepr
e become widely used to provide cathodfc| justrial water heaters and elevated water*!
sh, when added to the corrosive solution i/e effect.
;
,
.
.
In will produce oxide films on aluminum
aints, salt and oxide films
,
and metallic
.
,
agnesium forms a thick oxide film and pas,
) forms an oxide film.
btained by a variety of methods such as opiating, diffusion and cladding. i Chap. 14 in regard to surface coating for * ,
I metallized coatings for corrosion use are '
se metals are used primarily for corrosion J ; steel is cathodic to them and hence is pro- J 5ite of any porosity or minor voids in the is almost universally used for cleaning and /
:
Most coating systems employ
yj or top coats. The sealers are usually vinyl
s4r inhibiting primers
*
.
to apply a coating of zinc tin, cadmium, ,
£gne hot-dip process has a wide range of ap-
;
jplied must contend with a brittle diffusion ids at the interface
.
This may result in poor|L
iake on bending, unless the thickness of the W-
>
.
Aluminizing, hot-dipped aluminum poating on steel, is ;
used for applications that require a combination of resistance to corrosion ; and heat, such as mufflers and tailpipes of automotive:engines.
Electroplated coatings are used extensively for decorative purposes, but for industrial applications the most important single function is for corrosion protection. In addition to corrosion protection and appearance,
Zinc coatings are used on iron and steel products which require primarily corrosion protection. These include nuts, bolts, screws, nails, hardware,
and electrical conduit. Cadmium coatings are used as substitutes for zinc coatings but are not quite so good for outdoor exposure. Cadmium plating is also used for radio chassis and electronic equipment since it is readily
solderable. Chromium coatings are used on automobile trim, plumbing fixtures, hardware, and appliances. These coatings are very thin and rather porous and normally offer little protection against corrosion. Therefore, they are usually applied over thicker coatings of nickel or copper. Nickel coatings are used principally as an underplate for chromium, silver, gold, and rhodium plating. Nickel has good corrosion resistance but is tarnished by the atmosphere. Copper coatings are used as an undercoat for nickel and chromium coatings, particularly dTl zinc-base die castings. The greatest '
'
: -
pipe and conduit; and exposed structural steel. Tin plate is widely used in (he familiar tin can. Terne metal is sheet steel plated In a lead bath to
woar resistance (see Chap. 14); silver and copper are best for electrical conductivity; silver, copper, tin, and cadmium improve solderabiiity; chromium and rhodium have high resistance to tarnish.
,
sys protects by forming a phosphate film,
;
steel such as boilers, pails, cans, and tanks; hardware for outdoor use;
electroplated coatings are applied to obtain other surface properties such as wear resistance, high electrical conductivity, good solderabiiity, and high or low light-reflecting ability. Nickel and chromium plates provide
lined by reacting the metal with a solution m Some examples are: A chromate pickle a film of magnesium chromate Parkerizing
) spraying.
ucts than any other method of metallic coating. The major applications
roofing material.
HSijl structures. The use of paint for corrosion ;| d of little economic importance to the paint | | #ctive film to the metal and is effective only i
.
to regulate the thickness and uniformity of the coating.!; Zinc coating, or galvanizing, is applied to a greater tonnage of steel prod-
which tin has been added to ensure bonding. It is more expensive than zinc, but much cheaper than tin. It is used for gasoline tanks and as a i
,
jatings are primarily used to improve the
'
fluxing to facilitate wetting; dipping the article in a molten bath of controlled composition: and wiping, shaking, or oentrifuging the dipped piece
and wire products for all outdoor exposure; articles fabricated from sheet .
In most cases the inhibitor will)
fnetal surface. Inhibitors are added to the'I omobile radiators Oxidizing agents when fl
.y-
diffusion layer is properly controlled. The hot-dip process usually includes the operations of pickling with inhibited acid to produce a clean surface;
of zinc coating oh Steel products include roofing and siding sheets; wire
'
,
601
use of electroplated tin coatings is for food containers. These coatings may be made thinner than those made by the conventional hot-dipping method. The use of chromium and silicon as diffusion coatings for wear resistance has been discussed in Chap. 14. In addition to the above two metals,
: ;:p :
m i
602
t
INTRODUCTION TO PHYSICAL METALLURGY
aluminum and zinc diffusion coatings are also used to provide corrosion ,
15-5
What determines whether a given
protection. All diffusion-coating processes follow essentially the same ; procedure and are based on the same principles. The part to be coated is placed inicontact with a powder containing the metal to provide the f
15-6
What Is the limitation on the use
the corrosion of metals?
15-7
coating. At elevated temperature there is a transfer of the metal to the base i \ material, through the vapor phase, usually by means of a suitable catalyst. : v
i Holding at temperature or reheating without the powder after the initial
!i| penetration will allow further diffusion to the desired depth
.
The process \ 1
| of alloying st el with aluminum by diffusion is called calorizing. Calorized J ', steel is higWyj'esistant to ox[dgtlon and corrosion by hot gases
,
. -
-
-
,
particu-
15-8
Why do aluminum and chromium
15-9
What is meant by anodizing? Ho
15-10
What three factors are necessan
15-11
Explain and describe the mecha
15-12
Discuss the differences and si
and stress corrosion.
;
15-13
larly sulfurous gases. Calorized parts are used in furnaces that employ ;
Describe each of the following:
tion corrosion, fretting corrosion. 15-14 Explain cathodic protection.
fuels high in sulfur; bolts for use up to 1400°F; salt cyanide, and lead pots; ,
and oil refineries. Zinc impregnation is obtained by a process known as f sherardizing.
How does the distance between
affect corrosion rate?
.
15-15
List seven methods of corroslor
is effective.
The principal application of sherardizing is for small steel
parts, such s bolts, nuts, and washers, or for castings exposed to the ;
atmosphere. >
.;
REFERENCES
Cladding isl a method by which the coating becomes an integral part of
Ailor, William H. (ed): "Handbook on Co
the material. This may be accomplished by casting or hot working. Casting is best sujtedfwhen there is a considerable difference in the melting points > of the cjaddi;ng material and the base material. Hot rolling is the more ; widely used method for cladding. Slabs or sheets of the cladding material are strapped fto an ingot of the base material. After heating to the rolling temperature, the straps are removed and the entire assembly is rolled. The
& Sons, Inc., New York, 1971.
American Society for Metals: "Corrosior :
"
:
"
Metals Handbook," 1948, 1961,
,
heat and prepsure during rolling weld the two materials together.
Noble, New York, 1970.
R. M., and W. W. Bradley: "Prote ical Society Monograph 129, Van Nos
Burns,
The
cladding may! be of the same base material as the core. Alclad is the name ? applied to aluminum alloys which are clad with pure aluminum to improve corrosion resistance (see Fig. 12-30). Steel may be clad with nickel, nickelchromium, or nickel-copper alloys. Aside from corrosion resistance, cladding is sometimes done to obtain a combination of properties that are not available by any other method. Copper-clad steel wire is a good example. The copper exterior provides high electrical conductivity and good cor- ; rosion resistance, while the steel core provides high tensile strength. This j! wire is produced by pouring molten copper into a mold containing a round, heated steel billet. A weld that is able fo withstand cold working is pro- i duced between the two metals.
Surface Protection against We; Bosich, Joseph F.: "Corrosion Preventi .
Evans, U. R.: "An Introduction to Meta New York, 1963.
.
:
"
Metallic Corrosion Passivity ar
Ltd., London, 1946. Fontana and Greene: "Corrosion Engin York, 1967, «
International Nickel Company: "Corrosic LaQue, F. L, and H. R. Copson: "Com ed., Van Nostrand Reinhold Company
Scully, J. C: "Fundamentals of Corrosio Speller, F. N.: "Corrosion," 3d ed., McG Uhlig, H. H.: "Corrosion and Corrosion New York, 1971. : The Corrosion Handbook," Joh
After solidification, the composite billet
"
is hot-rolled to a rod and then drawn to the desired size.
"
* .
-
...
-
..
-
;
.
QUESTIONS 15-1
Define corrosion.
15-2
How does corrosion differ from erosion?
15-3 15-4
Explain the mechanism of electrochemical corrosion. Why doep corrosion generally occur at the anode? i
-
iiiiiiiUIJBMI
\ I
i mm
I /
i
URGY
CORROSION OF METALS '
603
:
V
.
coatings are also used to provide corrosi
15-5 15'6
jiting processes follow essentially the
. .
?:
the corrosion of metals?
the same principles. The part to be CO; )owder containing the metal to provide
;
ure there is a transfer of the nietal to the b
15-7 .
15-8 Why do aluminum and chromium show superior corrosion resistance? 15-9 What is meant by anodizing? How does it affect corrosion resistance? 15-10 What three factors are necessary to form a galvanic cell? 15-11 Explain and describe the mechanism of dezincification. 15 12 Discuss the differences and similarities between infergranular corrosion
Heating without the powder after the -inrtL
. ,
diffusion to the desired depth. The process'
,
h by diffusion is called calorizing. Calorized|
and stress corrosion.
bation and corrosion by hot gases particu
15-13
,
j ed parts are used in furnaces that emplof ie up t6 1400oF; salt cyanide, and lead poS|
-
How does the distance between metals in the galvanic series (Table 15-2)
affect corrosion rate?
Iiase, usually by means of a suitable catal;
j
What determines whether a given area is anodic or cathodic? What is the limitation on the use of the eleclromolive series {Table 15-1) in
Describe each of the following: pitting corrosion, crevice corrosion, cavita-
tion corrosion, fretting corrosion. 15-14 Explain cathodic protection.
'
,
15-15
pnation is obtained by a process known as
List seven methods of corrosion protection and explain why each method
is effective.
pplication of sherardizing is for small stee( d washers or for castings exposed to the ,
IKftEFERENCES ch the
coating becomes an integral part df
"
Allor, William H. (ed): "Handbook on Corrosion Testing and Evaluation, John Wiley
yTiplished by casting or hot working.. Castln$|
;
& Sons, Inc., New York, 1971.
onsiderable difference in the melting points':
American Society for Metals: "Corrosion of Metals, Metals Park, Ohio, 1946. : Metals Handbook,'1 1948, 1961, and 1964 ed., Metals Park, Ohio. : Surface Protection against Wear and Corrosion," Metals Park, Ohio, 1954. "
SSgpe base material. . Hot rolling is the mor&i '.
:
"
"
"
:
|g. Slabs or sheets of the cladding material : base material. After heating to the rolling
Bos-.ich, Joseph P.: "Corrosion Prevention for Practicing Engineers,
,
pved and the entire assembly is rolled ng weld the two materials together
.
.
Noble, New York, 1970.
The
The
"
Barnes and
i, "
Burns, Ft. M., and W. W. Bradley: "Protective Coatings for Metals, American Chem-
: .
se material as the core Alclad Is the nartHP
o-
r
.
ical Society Monograph 129, Van Nostrand Reinhold Company, New York, 1955. Evans, U. R.: "An Introduction to Metallic Corrosion, 2d ed., St. Martin's Press, "
,
:
h are clad with pure aluminum to ImprovK' L-
!-30)
.
Steel may be clad with nickel
,
nickel-
)ys. Aside from corrosion resistance
,
clad- -'ymk
in a combination of properties that are not jl | Copper-clad steel wire is a good example
. ,
;;
t:
igh electrical conductivity and good cor-M | l core provides high tensile strength This Iten copper into a mold containing a round is able to withstand cold working is proAfter solidification the composite billet ;
.
,
,
Irawn to the desired size
.
'
'
'
'
M
/
New York, 1963. "Metallic
:
Corrosion Passivity and Protection, Edward Arnold (Publishers) "
Ltd., London, 1946. Fontnna and Greene:
"
Corrosion Pnqinoering," McGraw-Hill Book Company, New
York, 1967.
International Nickel Company: "Corrosion in Action, New York, 1955. "
LaQue, F. L, and H. R. Copson: "Corrosion Resistance of Metals and Alloys,
2d
ed.. Van Nostrand Reinhold Company, New York, 1963.
Scully, J. C: "Fundamentals of Corrosion," Pergamon Press, New York, 1966. Speller, F. N.: "Corrosion," 3d ed., McGraw-Hill Book Company, New York, 1951. Uhlig, H. H.: "Corrosion and Corrosion Control," 2d ed., John Wiley & Sons, Inc., New York, 1971. :
"
The Corrosion Handbook," John Wiley & Sons, Inc., New York, 1948.
m erosion?
.
ictrochemical corrosion. ! occur at the anode? '
m
"
>
-
. .
r
-
-
.
i
f
i
1
PC ME :
16-1 Introduction Powder metallurgy ma metal powders and using them to metallurgy principles were used as to make iron implements. The use powders for ornamental purposes \ ages.
In 1829, Woolaston published a | producing compact platinum from p the first scientific work in the field of
dations for modern techniques. It is interesting to note that in the ments were produced in powder fori The invention of the incandescent
development of a suitable filament lum, and tungsten were used, but t long been evident that tungsten wou trie lamp, but to work tungsten intc conventional metallurgy at the begii mained for Coolidge, in 1909, to ma
sten can be worked in a certain tem|
tility at room temperature. Finely divii into small ingots which were sintere point of tungsten. These sintered inbut could be worked at elevated tern
ture. Subsequent working at the elev; until a stage was reached where the r and could be drawn into wire with ten
The Coolidge method led to a nev metals such as molybdenum tantalu
I
,
1 n
\
'
v
-
V
POWDER METALLURGY ' ,
16-1 Introduction
Powder metallurgy may be defined as the art of producing
metal powders and using them to make serviceable objects. Powder metallurgy principles were used as far back as 3000 B.C. by the Egyptians
A '
.
to make iron implements. The use of gold silver, cdpper, brass, and tin ,
!
powders for ornamental purposes was commonplace during the middle
;:
ages.
j ;i
In 1829, Woolaston published a paper which described a process for producing compact platinum from platinum sponge powder. Considered the first scientific work in the field of powder metallurgy, this laid the foundations for modern techniques. It is interesting to note that in the nineteenth century more metallic ele-
j
ments were produced in powder form than in any other form.
-
The invonlion ol the inr.nndosoenl oloclric light by Edison required the development ol a suitable filament maturml. Powders of osmium, tantalum, and tungsten were used but the filaments were very brittle. It had ,
long been evident that tungsten would make an ideal filament for the electric lamp, but to work tungsten into the necessary fine wire was beyond conventional metallurgy at the beginning of the twentieth century. It remained for Coolidge in 1909, to make the important discovery that tungsten can be worked in a certain temperature range and will retain its duc,
' .
-
.I
1
tility at room temperature Finely divided tungsten powder was compressed into small ingots which were sintered at temperatures below the melting point of tungsten. These sintered ingots were brittle at room temperature but could be worked at elevated temperatures near the sintering temperature. Subsequent working at the elevated temperature improved its ductility until a stage was reached where the metal was ductilerat room temperature and could be drawn into wire with tensile strength approaching 600 000 psi. The Coolidge method led to a new method of fabrication for refractory metals such as molybdenum tantalum, and columbiu'm. It also led to the ,
,
i
i v
.
:
I
606
INTRODUCTION TO PHYSICAL METALLURGY
development of cemented carbides and composite metals. At about the
same time, porous-metal bearings were manufactured using the technique i
Row moteriois
Elemento
metal p.
of powder metallurgy. These and other applications will be discussed later | , in this chapter.
r
; i
>
Mix
Initially all P/M (symbol used to represent powder metallurgy) parts i were small, and the mechanical properties were more or less comparable ]
with conventional materials.
Si
Today, however, parts greater than 1 ft in
J
Pressnreless molding
Farm
Slip costing
Solid
green condition it was ISVz in. in diameter by 44 in. long. An important ;
store
Hot
fus,on
press,ng
development has been the increasing use of large P/M parts by the auto-
I
L
diameter and weighing more than 50 lb are being produced in large quanti- i ties. Figure 16-1 shows a picture of a molybdenum-powder billet which is ) believed to be the heaviest made to date. It weighed 3,000 lb, and in the ;
Gas pressure bonding
or
nhotive industry. Materials with mechanical properties far exceeding those 1
]
banding
Spark
sintering
(sintering)
of conventional materials have been developed by improving heat treat-
-
ments, powder compositions, and processing methods to achieve higher densities. High strength, ductility, and toughness may be obtained in P/M parts, so that the old idea of brittle fragile parts is no longer valid. 16-2 Powder Metallurgy Processes The two main operations of the powder metal process are compacting and sintering. Compacting, or pressing, consists in subjecting the suitably prepared
Extrusion
.
Optiono! or
1
r
Optional finishing
llneat hent, plote tumble it [ Ireot, oil or plastic im(
secondary
.
operations
I
Fimsned product
powder mixtures, at normal or elevated temperature, to considerable pres-
Fig. 16-2 Powder metallurgy processes. (Courtesy of t Metal Powder Industries Federation ) .
sure. The resulting powder compac in this form
to be "green." It can Sintering is an operation in which heat, usually in an inert atmosphere point of the solid metal. Sintering will as well as other desired properties In addition to compacting and sin tion, other accessory operations ma elude presintering sizing, machininc the powder metallurgy processes is si showing steps in the manufacture c ,
.
,
I
i
4
Fig. 16-3. 16-3 Preparation of Metal Powders
There
ular method of powder production metallurgy products. Many mechanic produce powders for specific applici methods are atomization reduction c
Fig. 16-1 Molybdenum-powder forging billet, shown in Jhe green condition, measuring ISVe in. in diameter by 44 in.
long. {Courtesy of GTE ylvania.)
,
,
MM
HUliliiH...J
IIU4 I L
,
Atomization is the method most fr
melting points such as tin, lead zin ,
UIIUB
i /
,
9ft'
!
RGY
POWDER METALLURGY
bides and composite metals. At about ih$i
jigs were manufactured using the technltjujiy
Raw
Elemenlol or olloy
materials
metal powders
ijnd other applications will be discussed lato
1
led to represent powder metallurgy) parts||
r
,
JM. in diameter by 44 in. long. An important1
easing use of large P/M parts by the aulomechanical properties far exceeding those
Form
.
'
h\ The two main operations of the powder
Roll compacting High energy forming
Presintering
for rnKtalL
Solid Mol
> 10 I 0
pressing
fusion or
Spaik sinlcrmg
hoiulini]
Isinlermg)
1
Sinlrating
Sinl crmi)
_
-
J
fxlrusion
Optional finishing steps (heat treat, plate,tumble.machme,steam treat, oil or plastic impregnate)
Optional or secondary operations
Optional .manufacturing steps (metal infiltrotion,size,coin,forge or reform, repress resinter) '
1
F imshed
nsists in subjecting the suitably prepared
product
,
-
Gas pressure bonding
and sintering.
ilevated temperature
3
Isostatic compacting
Die compacting (mechanical or hydraulic)
I
been developed by improving heat treat-J and processing methods to achieve higher ity, and toughness may be obtained in P/M ittle fragile parts is no longer valid.
'
-
Pressui nlcss molding Slip casting
jre of a molybdenum-powder billet which is de to date. It weighed 3 000 lb, and in the
Other additions
[ (die lubricarjts, graphite) T Mixing
Mix
al properties were more or less comparabtelT Today, however, parts greater than 1- ft-inaB: an 50 lb are being produced in large quantk||
|
607
Finished P/M products
to considerable preSr|Jiig. 16-2 Pov;der metallurgy processes. (Courtesy of the Wetal Powder Industries Federation.)
sure. The resulting powder compact is known as a briquette and is said, in this form, to be "green." It can be handled, but it is relatively brittle. Sintering is an operation in which the green briquettes are subjected to heat, usually in an inert atmosphere, at a temperature below the melting point of the solid metal. Sintering will give the required mechanical strength as well as other desired properties.
In addition to compacting and sintering, and depending upon application, other accessory operations may be added to the process. These include presintering. sizing, machining, and impregnation. A summary of the powder metallurgy processes is shown in Fig. 16-2. Typical flow sheets showing steps in the manufacture of powder metal parts are shown in Fig. 16-3. 16-3 Preparation of Metal Powders There is a definite relation between a partic ular method of powder production and desired properties of powder
-
.
metallurgy products. Many mechanical and chemical methods are used to
pfi jwn in the
jy 44 in.
produce powders for specific applications, but the three most important methods are atomization
,
reduction of oxides, and electrolytic deposition.
Atomization is the method most frequently used for metals having low melting points such as tin, lead, zinc, cadmium, and aluminum. As the ,
1
,
608
-
' -
.
-
.
v
.
INTRODUCTION TO PHYSICAL METALLURGY
liquid metal is forced through a snr causes the metal to disintegrate ar Atomized products are generally ir (Fig. 16'4a). A wide range of parti by varying the temperature of the n atomizing gas, rate of flow of met? of the orifice and nozzle. The pi process is its flexibility. It will produi ness, and in the production of a giv distribution can be closely maintainThe reduction of compounds of t
.
2
in
S
s
Li
'
ill :
i5
i
a convenient, economical and fie: ,
T S
The largest volume of metallurgica
c7>
oxide reduction.
gH OO e s
"
ct> q
,
,
-
CO
terial and the conditions of reduct a
3
LJI
,
with carbon monoxide or hydroger quently ground. The nature, particle
tvicvj
So
Mill scale or che
9
LU
UJ
CO
9
a
CD
.
9
1
S
,
a
Of
s
5
(?)
m
<
deposited particles. If the oxide pow degree of size uniformity can be oi particles produced by oxide reductic ideal for molding. The shape is gene and the particles are porous. This is producing powders of the refractory
LU
00
cn
.
LxJ
DO II
CD LU
LU
o
a
,
UJ
UJ
num. Oxide reduction is also an ecoi o =
2
o
5
ms it
U
in
3
o
9
E
"
CO
5 7,
,
LJI
e2l 5 (1)
52 s 1 .
5
of iron, nickel, cobalt, and copper. The method of electrolytic depositi of extremely pure powders of princ is essentially an adaptation of elec
density, temperature, circulation of '
'
'
*
'
1 3
--
,
'
'
/: :
10
;
ft 5 SE 5 E i
S
E
-
fi
o
3:
111
ill
o
II -
o
1?
CO
q3
If
(2
5
>> i r
a
o
sat
g-1 (2 g
si
o
o
3
W
K
C
electrolyte, the powder may be dired deposit may be a soft spongy substj powder, or the deposit may be a h from hard, brittle electrodeposits a purposes. Most of the powder pre commercial applications is of the sf powder is generally dendritic (Fig. 1 has low apparent density, the dendri ing properties because of interlockir 6-4 Characteristics of Metal Powders
o
f
111
11
8
material during processing as well aJ depends to a large extent upon the material. Aside from the chemical ci
I
7* I
.IPtGY
POWDER METALLURGY
609
liquid metal is forced through a small orifice, a stream of compressed air
causes the metal to disintegrate and solidify into finely divided particles.
. .
.
Atomized products are generally in the form of sphere-shaped particles (Fig. 16-4a). A wide range of particle-size distributions may be obtained by varying the temperature of the metal, pressure and temperature of the atomizing gas, rate of flow of metal through the orifice, and the design
of the orifice and nozzlfe.
The principal advantage of the atomization
process is its flexibility. It will produce powders of different degrees of fineness and in the production of a given fineness, uniformity of particle-size distribution can be closely maintained. The reduction of compounds of the metals (usually an oxide) provides a convenient, economical, and flexible method of producing powders. ,
m
The largest volume of metallurgical powder is made by the process of oxide reduction. Mill scale or chemically produced oxides are reduced with carbon monoxide or hydrogen, and the reduced powder is subsequently ground. The nature, particle size, and distribution of the raw material and the conditions of reduction greatly influence the form of the
ft
11
deposiled particles. If the oxide powder is graded before reduction, a high degree of size uniformity can be obtained in the reduced powder. The particles produced by oxide reduction are spongelike rn structure and are
null
CO
ideal for molding. The shape is generally jagged and irregular (Fig. 16 -4fa), and the particles are porous. This is the only practical method available for producing powders of the refractory metals such as tungsten and molybdemim Oxldo induction if; also an economical method of producinr) powdors ol iron, nickel, cobalt, and copper. The method of electrolytic deposition is most suitable for the production
or
r «
a,
1
M «
2! 2 *
i
of extremely pure powders of principally copper and. iron. This process is essentially an adaptation of electroplating. By regulation of current
S 2-
a el
density, temperature, circulation of the electrolyte, and proper choice of
2 u. "
11
fg
a
electrolyte, the powder may be directly deposited from the electrolyte. The deposit may be a soft spongy substance which is subsequently ground to powder, or the deposit may be a hard, brittle metal. Powders obtained from hard brittle electrodeposits are generally not suitable for molding purposes. Most of the powder produced by electrolytic deposition for commercial applications is of the spongy type. The shape of electrolytic powder is generally dendritic (Fig. 1.6-4c). Although the resulting powder has low apparent density, the dendritic structure tends to give good molding properties because of interlocking of the particles during compacting
111 o
<
,
Ms E
d
I a
II
s
i
III r
in
III
.
CO CD
16-4
Characteristics of Metal Powders In all cases, the performance of the material during processing as well as the properties of the finished product depends to a large extent upon the basic characteristics of the powder
material. Aside from the chemical composition and purity
,
m
the basic char-
I
610
INTRODUCTION TO PHYSICAL METALLURGY
The size of these powders may be s sions as determined by microscopi Particle-size distribution is impoi will influence its behavior during purposes, the selection of a desire
#>
put 15
plication is usually based upon exf preferred over a coarser powder, £
j
26 A" *
size and larger contact areas which erties after sintering. Particle-size sieve analysis that is, the amount i ,
ljllll|l ll{lll l |lll falll|l lpl|llll|l$|llll|lW
,
etc. mesh sieves. It should be appa icant results regarding particle siz tides are spherical in shape. Inac the particles are irregular or flaky
i
.
.
iu*
'
St
The nature of the surface of inc
powder characteristic. Powders pro (Fig. 16-46) usually have a highly t served, whereas atomized particles of surface roughness. The characte tional forces between particles whic ,
or settling or during compaction. or between the powder and its envi amount of surface area per unit of p area is very high for powders made the data in Table 16-1.
Particle shape is important in infl teristics of powders. Spherical-sha qualities and result in uniform physi however irregular-shaped particles
i
i
,
'
Fig. 16-4 Shape of powders produced by different methods, (a) Atomizing; (b) oxide reduction; (c) electrodeposition. (Courtesy of the Metal Powder Industries Federation.)
(el
; ;; '
r
:
I .
acteristics of a metal powder are particle size and size distribution, particle | shape, apparent density, and particle microstructure. '
:V \v
-
.
,
Metal powders may be divided into sieve and subsieve size ranges. Those d |
Typical Specific Sut
POWDER Reduced Fe Fine, 79%-325 mesh Normal blend
Coarse, 1%-325 mesh
Sponge Fe-normal blend Atomized Fe -normal blend
in the sieve-size class are usually designated according to the finest mesh
Electrolytic Fe- normal blend
through which all the powder will pass. If all the powder passes througMl
Reduced Tungsten 0.6 micror
a 200-mesh screen, it is designated as a minus 200-mesh powder, etc. The||
.From J
subsieve-size powders all pass through a 325-mesh sieve used in practice.
Powder Metallurgy Institute New York,
:
I
I
/
I
TABLE 16 1
.
S. Hirschhorn, "An Introdu ,
?
-
'
:;:
URGY
POWDER METALLURGY
J
v
.
€
.
:i .
*.
The size of these powders may be specified by averaging the actual dimen-
'
sions as determined by microscopic examination. Particle-size distribution is important in the packing of the powder and will influence its behavior during molding and sintering. For practical purposes, the selection of a desirable size distributiofi for a specific application is usually based upon experience. In general, a finer powder is preferred over a coarser powder, since finer powders have smaller pore
i 4a
size and larger contact areas, which usually results in better physical prop-
'
i
llll Hill
erties after sintering. Particle-size distribution is specified in terms of a sieve analysis, that is, the amount of powder passing through 100-, 200-, etc. mesh sieves. It should be apparent that sieve analysis will yield significant results regarding particle size and distribution only when the par-
r .
:f
.
611
9
I
ticles are spherical In shape. Inaccurate information will be obtained if the particles are irregular or flaky. The nature of the surface of individual particles is also an important lt>),'
.
powder characteristic. Powders produced by chemical reduction of oxides (Fig. 16'4;b) usually have a highly roughened surface which is easily observed, whereas atofnized particles (Fig. 16-4a) have a much finer degree of surface roughness. The character of the surface will influence the fri c tional forces between particles, which is important when powder is flowing or settling or during compaction. Since any reaction between particles or between the powder and its environment is initiated at the surface, the amount of surface area per unit of powder can be significant. The surface are:a is very high for powders made by reduction techniques, as shown by -
.
5
-
the data in Table 16-1.
3
3
Particle shape is important in influencing the packing and flow characteristics of powders. Spherical-shaped particles have excellent sintering qualities and result in uniform physical characteristics of the end product; however, irregular-shaped particles have been found superior for practical
TABLE 16-1 Typical Specific Surface Areas of Commercial Powders*
Fig. 16-4 Shape o( powders produced by different methods
,
SPECIFIC SURFACE
(a) Atomizing; (b) oxide
reduction; (c) electrodeposition (Courtesy of the Metal Powder Industries Federation ) .
.
v
1
v::;;e particle size and size distribution particle ;
.
,
particle microstructure. /
.
jd into sieve and subsieve size ranges. Those
Stilly designated according to the finest mesh "4 |will pass. If ail the powder passes through
POWDER
AREA (cnrVgm)
Reduced Fe
Fine, 79%-325 mesh
5160
Normal blend
1500
Coarse, 1%-325 mesh
516
Sponge Fe-normal blend
800
Atomized Fe-normal blend
525
Electrolytic Fe- normal blend Reduced Tungsten 0 6 micron .
400 5000
ated as a minus 200-mesh powder etc. The a ,
through a 325-mesh sieve used in practice
.From J :
-
.
W
iw
S. Hirschhorn, "An Introduction to Powder Metallurgy Powder Metallurgy Institute New York, 1969.
"
.
,
,
American
-.
:
-::
.
.-
612
INTRODUCTION TO PHYSICAL METALLURGY
TABLE 16 2 MATERIAL
Apparent Densities of Commercial Metal Powders* SPECIFIC
APPARENT
GRAVITY
DENSITY
Aluminum
2 70
Antimony
6 68
Cadmium
8 65
Chromium
7 1
2 5-3 5
Silver
Cobalt
89
1 5-3
Tin
8 93
0 7-4
Tungsten
.
0 7-1
MATERIAL
2-2 5 .
.
.
.
.
4-6
11.3
Lead
.
.
.
Nickel
Silicon
3
.
Copper
Molybdenum
.
.
SPECIFIC
APPARENT
GRAVITY
DENSITY
10.2
3-6 5
89
2 5-3 5
2 42 .
0 5-0 8
10.50
1 2-1 7
.
.
.
.
.
.
.
5 75
.
1-3
.
19.3
5-10
Zinc
7 14
2 5-3
Iron and steel
7 85
1-4
.
i
Fig. 16-6 Model illustration of the "bridging
.
"
Magnesium
m
*
1 74
0 3-0 7
.
.
.
.
effect cai
by small particles.
By permission from C. G. Goetzel, "Treatise on Powder Metallurgy," vol. 1, Interscience Publishers, Inc.. New York, 1949.
completely fill the pores It is even i tides to decrease the apparent dens illustrated in the model of Fig 16-6. .
molding. The mechanism of packing involves three processes: the filling of gaps between larger particles by smaller ones, breakdown of bridges or arches, and rtiutual sliding and rotation of particles. These processes are
.
the powder influenced the effect on <
important in loading die cavities with metal powders. v
der is very effective in increasing the 1 reduce the apparent density sharply
Apparent density may be defined as the weight of a loosely heaped quan-
.
.
tity of powder necessary to fill a given die cavity completely. A list of apparent-density ranges of a number of metal powders is given in Table 16-2. Naturally, increasing the specific gravity or density of the solid ma- ; terial increases the apparent density of the powder. As was just pointed J
a property of great importance for b(
Powders with low apparent density and deeper cavities to produce a brii tendency of the compact to shrink du increasing apparent density
out, the packing of powder particles is greatly influenced by particle sizej and shape. For example, a given space can be completely filled only by the 4 |.l6-5 Mixing Proper blending and mixing .
same size of jsubes exactly aligned. Any particle shapes that are curved or 5
i
formity of the finished product tained by blending in advance the dil Dei
.
irregular canhot completely fill a space, and this leads to the presence of » porosity. The importance of packing of spheres was discussed in Chap. 2 under crystal structures, where it was shown that face-centered-cubic and close-packed-hexagonal structures had high packing factors. An effective way to increase the apparent density is to fill the spaces among particles
ing powders lubricants, and volatiliz ,
of porosity are added to the blended
40
with smaller ones. This leads to a filling arrangement known as interstitial
Spherical
35
packing (Illustrated in Fig. 16-5); however, even the smaller particles cannot
rreqular
3C
25 20 a
v.,
.
.
15
Flake
05 0
I
Fig. 16-7
10
20 30 40 50 60 70' 80 90 I00 Pereent-325 mesh powder
The effect of -325 mesh additions to a +325
sh distribution of 316 stainless-steel powder on appan |ensity for three different shapes of the addition (From 4s. Hirschhorn Introduction to Powder Metallurgy .
"
Fig. 16-5 An example of interstitial packing, that is, voids among large particles being filled by smaller ones.
#3
) mi
"
,
American Powder Metallurgy Institute
,
;
,
New York 1969.) ,
;JRGY
POWDER METALLURGY
613
lol Motel Powdors* A
MATERIAL
Molybdenum Nickel
SPECIFIC
APPARENT
GRAVITY
DENSITY
10.2
3-6 5
89
2 5-3 5
2 42
0 5-0 8
10.50
1 2-1 7
.
.
Silicon
.
.
Silver Tin
.
.
.
. .
.
5 75 19.3
5-10
Zinc
7 14
2 5-3
Iron and steel
7 85
1-4
Metalluigy
,
i
1-3
.
Tungsten
M
.
Fig. 16-6
.
.
Model illustration of the "bridging" effect caused
by small particles.
vol. I. Inlorscience Publishers Inc.. New York, 1949. ,
completely fill the poros. It is ovon possible lor tho addition of small paricking involves three processes: the filling of by smaller ones
,
ticles to decrease the apparent density by an effect known as "bridging," illustrated in the model of Fig. 16-6. As shown in Fig; 16-7, the shape of the powder influenced the effect on apparent density. Fine spherical powder is very effective in increasing the apparent density, while flake additions reduce the apparent density sharply. The apparent density of a powder is
breakdown of bridges or j
I rotation of particles. These processes are | '
S with metal powders. | ed as the weight of a loosely heaped quan- j .
II a given die cavity completely. A list of %
a property of great importance for both molding and sintering operations. Powders with low apparent density require a longer compression stroke and deeper cavities to produce a briquette of given size and density. The tendency of the compact to shrink during sintering seems to decrease with
lumber of metal ppwders is given in Tabls \
\ specific gravity or density of the solid ma- | lensity of the powder. As was just pointed j
"
rticles is greatly influenced by particle size
increasing apparent density.
n space can be completely filled only by the §16-5 Mixing Proper blending and mixing of the powders are essential for uniformity of the finished product. Desired particle-size distribution is obtained by blending in advance the different types of powders used. Alloying powders, lubricants, and volatilizing agents to give a desired amount of porosity are added to the blended powders during mixing. The time for
ned. Any particle shapes that are curved or a space, and this leads to the presence of 5
icking of spheres was discussed in Chap. 2 \ it was shown that face-centered-cubic and
Jures had high packing factors. An effective density is to fill the spaces among particles |o a filling arrangement known as interstitial
40 -
however, even the smaller particles cannot j
;
3
Sphericnl Irregular
'
30
I
at
1
£ 20 lb
i
o;i. . i ...i. ...i -i-ii. 0 10 20 30 40 50 60
;
1 t is, voids
Floke
10 i
.
70 80 90 100
Percent-325 mesh powder
.
, '
Fig. 167
The effect of -325 mesh additions to a +325
j mesh distribution of 316 stainless-steel powder on apparent
J density for three different shapes of the addition. (From S. Hirschhorn, "Introduction to Powder Metallurgy," American Powder Metallurgy Institute, New York 1969.)
"
s
.
,
1 614
v
.
INTRODUCTION TO PHYSICAL METALLURGY
.
\
|
:
mixing may vary from a few minutes to severai days, depending upon ex-
perience and'the resuits desired. Overmixing should be avoided in many cases, since it may decrease particle size and work-harden the particles.
'
16-6 Compacting the most important operation in powder metallurgy is com- l pacting or pressing. The ability to obtain a satisfactory pressed density ; often determifies the feasibility of manufacture by powder metallurgy. Most ;
compacting ii done cold, although there are some applications for which compacts areihot-pressed.
The purpos of compacting is to consolidate the powder into the desired :
shape and as: closely as possible to final dimensions, taking into account : c-v.v:.-
,
;
i
i
any dimensiohal changes that result from sintering; compacting is also designed to impart the desired level and type of porosity and to provide adequate strength for handling. Compacting techniques may be classified into two types: (1) pressure : techniques, such as die, isostatic, high-energy-rate forming, forging, ex-
7]
mm
trusion, vibratory, and continuous, and (2) pressureless techniques such .
!
as slip casting, gravity, and continuous.
,
m
Die compahtion is the most widely used method. The usual sequence of operations in|die compacting consists in filling the die cavity with a definite
volume of pdjwder; application of this required pressure by movement of ; the upper antji lower punches toward each other; and finally, ejection of the ; green compact by the lower punch. These operations are shown schematically in Fig. 16-8. The pressures commonly employed range from 19 to 50 tons/sq in. The
Fig, 16-9 Multiple-action 20-ton powder metal mechanl press. (F. J. Stokes Corporation.)
pressure may be obtained by either r chanical presses are available with p
Upper punch
speeds of 6 to 150 strokes/min. Compact' Die
m 3
r
-
m
Tl
presses are high-speed production and economy in operation, and relati costs. A mechanical press is shown i motion press combining one motion 1 chanlcally linked motions from belc motion allows the production of multi and can also be used as a movable c
thin-walled pieces.
Lower
Hydraulic presses have higher pre
punch
slower stroke speeds generally less 1 for higher-pressure more complicate Dies are usually made of hardened i the powder to be compacted consists generally constructed of two parts. 1 ,
a),
a)
,
Fig. 16-8 Schematic operations for the die compacting to form powdered parts. (By permission from L. F. Mondolfo
and 0. Zmeskal,
"
Engineering Metallurgy," McGraw-Hill
Book Company, New Yor , 1955.) I
?
1
/
i
,
JRGY
POWDER METALLURGY
v y
615
-
.
Wx
.
ninutes to several days depending upon ex-v ed. Overmixing should be avoided in many*
f
,
|
.
|;5?:iparticle size and work-harden the particles.1 .
ant operation in powder metallurgy is com»| , lity to obtain a satisfactory pressed densityi
of manufacture by powder metallurgy, Mottl ough there are some applications for which-!
1
,
\
I
%
is to consolidate the powder into the deslred|
jble to final dimensions, taking into account| % j& t result from sintering; compacting is alsQ| %
1
$W-di level and type of porosity and to provide| i
i r
3
.
-
y be classified into two types: (1) pressure j |
'
r
atic, high-energy-rate forming, forging, ex-j |
s
.
r
;
jous, and (2) pressureless techniques sucfl| I
ntinuous.
,
i
1
'| f
1
i
5
-
gwidely used method. The usual sequence bt| t bonsists in filling the die cavity with a definite
r
of the required pressure by movement of J :
? |oward each other; and finally ejection of the J ,
..
r
ounch. These operations are shown schawl <£
I f iployed range from 19 to 50 tons/sq in. The ' j Fig
16-9 Mjltiple-action 20-ton powder metal mechanical press. (F. J. Stokes Corporation.) : .
1
pressure may be obtained by either mechanical or hydraulic presses. Me-
un ch
chanical presses are available with pressure ratings of 10 to 150 tons and Compnct _
1
m
1
.
v
ipacting to
'
.
motion press combining one motion from above with two independent mechanically linked motions from below (Fig. 16-10). The secondary lower motion allows the production of multilevel compacts with simplified tooling and can also be used as a movable core rod to assist in the production of
thin-walled pieces. Hydraulic presses have higher pressure ratings, up to 5,000 tons, but slower stroke speeds, generally less than 20/min. These presses are used for higher-pressure, more complicated powder metal parts. Dies are usually made of hardened ground, and lapped tool steels. When the powder to be compacted consists of hard abrasive: particles, the die is generally constructed of two parts. The tough outer section supports the
[c] '
.
speeds of 6 to 150 strokes/min. The important features of mechanical presses are high-spfeed production rates, flexibility in design, simplicity and economy in operation, and relatively low investment and maintenance costs. A mechanical press is shown in Fig. 16-9. This is a 20-ton multiple-
Mondolfo
,
!raw-Hill
,
1
j
i16
INTRODUCTION TO PHYSICAL METALLURGY
Forging and extrusion techniques tent. In either case the powder is "c< container. The sealed container is h-
extruded.
:
Figure 16-11 shows the i
powdered metal. After forging or e moved either mechanically or chemii of extremely high density and usuall In vibratory compaction pressure ously to a mass of powder in a rigid (
1
3
,
paction, this method allows the use
given level of densification. A major vibration to practical tooling and pre Continuous compaction is appliec rod, sheet, tube and plate. Most c
M
,
flowing loose powder between a se gap is adjusted to give a compact ( speed of powder rolling is much U
i
operations.
S//p casting is widely used for cer metals. The process consists of fi powder suspended in a liquid vehi settling. The slip is then placed in i
Fig. 16-10 The multiple actions of the press shown in Fig. 16-9. (F. J. Stokes Corporation.)
terial (such as plaster of paris) to for the mold, the slip casting is dried anc for materials that are relatively incomj tion, but the process does not lend i of the long time required for the liqu of the mold. Figure 16-12 shows a casting molybdenum powder. In gravity compaction, the die is f
i
hardened, polished, wear-resistant insert which is the working surface of the die. These replaceable liners are discarded when worn and reduce the
cost of die upkeep. The punches are made of die steel heat-treated to be slightly softer than the die, since they are usually easier to replace than the die. They must be
perfectly aligned and very closely fitted.
I
In isostatic compacting, pressure is applied simultaneously and equally J
sintered in the die.
in all directions. The powder is placed in a rubber mold which is immersed |
The die is usua
graphite. Since pressure is not us Commercially this method is used fo
in a fluid bath within a pressure vessel, so that the fluid may be placed .1 under high pressures. Since pressure is applied uniformly, it is possible to i
obtain a very uniform green density and a high degree of uniformity in 1 properties. This method has been used extensively for ceramic materials .;
Extrusion container
rather than metals.
.
.
'i .
i
High-energy-rate techniques may be either mechanical, pneumatic, or 1
Mild srei
explosive- or spark-discharge methods applied in a closed die. The ad-
vantage of these methods is the short time and high pressures that can be attained. It is also possible to use low-grade and very cheap powders, and M.
some parts, due to increased strength of the green compacts, may be used|B
POWO!
without subsequent sintering. Disadvantages include high punch and dieM Fig wear, limited-tolerances, and high cost.
i
I /
J
Method of extruding wire and rods from p0
dered metal.
m
i
Ham
.
IGY
POWDER METALLURGY
617
>v '
Forging and extrusion techniques have been used only to a limited extent. In either case the powder is "canned" or placed in some kind of metal container. The sealed container is heated or evacuated and then forged or extruded. Figure 16-11 shows the method of extruding wire or rod from powdered metal. After forging or extruding, the container material is re-
moved either mechanically or chemically. Both techniques yield compacts of extremely high density and usually do not require sintering In vibratory compaction, pressure and vibration are applied simultaneously to a mass of powder in a rigid die. Compared with ordinary die com.
paction. this method allows the use of much lower pressures to achieve a
given level of densification. A major problem is equipment design to apply vibration to practical tooling and presses.
Continuous compaction is applied primarily for simple shapes such as rod, sheet, tube, and plate.
Most of the commercial techniques involve
flowing loose powder between a set of vertically oriented rolls. The roll
gap is adjusted to give a compact of desired properties. In general, the
speed of powder rolling is much less than that of conventional rolling operations.
S//p casting is widely used for ceramics but only to a limited extent for metals.
The process consists of first preparing a
"
slip
"
containing the
powder suspended in a liquid vehicle and additives to prevent particle settling. The slip is then placed in a mold made of a fluid-absorbing material (such as plaster of paris) to form the slip casting. After removal from the mold, the slip casting is dried and sintered. This technique is attractive for materials that are relatively incompressible by conventional die compac-
ivn in
ant insert which is the working surface of s are discarded when worn and reduce the
tion, bul Ihe process does not lend itself to high production rates because
steel heat-treated to be slightly softer than
of the long time required for the liquid to be removed through the porosity of the mold. Figure 16-12 shows a molybdenum crucible made by slip-
asier to replace than the die. They must be
casting molybdenum powder.
v sly fitted.
In gravity compaction, the die is filled with loose powder which is then
sure is applied simultaneously and equally placed in a rubber mold which is immersed
i
re vessel, so that the fluid may be placed j essure is applied uniformly it is possible to ;
sintered in the die. The die is usually made of an inert material such as graphite. Since pressure is not used, parts are generally more porous. Commercially this method is used for the production of P/M filters.
,
ensity and a high degree of uniformity in een used extensively for ceramic materials
\
Die
Extrusion container
may be either mechanical pneumatic, or
Mild steel
,
riethods applied in a closed die The ad.{ g3 short time and high pressures that can be -j '
v
can
.
ise low-grade and very cheap powders and '!
r
,
rength of the green compacts may be used ,
Disadvantages include high punch and die gh cost.
ZjZZZZZZZ 2ZZZZZZ P\ Powder
Ram
Fig, 16-11
Method of extruding wire and rods from pow-
dered metal.
I
518
1
INTRODUCTION TO PIHYSICAL METALLURGY
90 -
i
i
80 -
70 -
20
30
40
Apparent density (percent theoretical)
Fig. 16-13 Dependence of green density (at 30 tons/s on apparent density for different iron powders. (From Hirschhorn, "Introduction to Powder Metallurgy," Am; Powder Metallurgy Institute, New York, 1969.)
16-7 Sintering The sintering process is below the highest melting constitu high enough to form a liquid cons
Fig. 16-12 . Molybdenum crucible made by slip-casting molybdenum powder of 2 to 3 particle size, followed by air drying and sintering 73 h_at:3270oF in hydrogen; unetched, 3x. (Courtesy of T. I. Jones, Los Alamos Scientific Laboratory.)
cemented carbides, where sinterir
the binder metal. In other cases, nc
Continuous pressureless compaction is used to produce porous sheet for electrodes in nickel-cadmium rechargeable batteries. The powder may be applied in the form of a slurry (similar to the slip in slip casting) to r be coated on a metal screen or solid metal sheet to produce unusual composites. The green compact density is a most useful property, since it is an indication of the effectiveness of compaction and also determines the be- A ,
10 000 ,
Q
8 000 ,
.
Tl
tin
6 000 ,
havior of the material during subsequent sintering. Green density seems to | i
4 000 ,
increase with i'ncreasing compaction pressure, increasing particle size or apparent density, decreasing hardness and strength of particles, and de-
2,000 -
creasing compaction speed. The dependence pf green density at 30 tons/
0
sq in. on apparent density is shown in Fig. 16-13. The strength gf green
compacts resujits chiefly from mechanical interlocking of the irregularities
0
40
80
120
160
Compaction pressure (tsi)
green densityjncreases green strength because more particle movement
Fig. 16-14 Dependence of green strength on compact pressure for electrolytic iron powder. (From H. H. Hau Powder Metallurgy," Chemical Publishing Co. New Y
and deformation will promote greater interlocking.
1974.)
on the particle surfaces. Increasing compaction pressure (Fig. 16-14)
,
I
f
t
i
or
"
,
v GY
POWDER METALLURGY
619
1 90
I 1
I
,
1
I ::;
70 :
20
i
i
30
10
Apporent density (percent theoretical)
Fig. 16-13 Dependence of green density (at 30 tons/sq in.) on apparent density for different iron powders. (From J. S. l
;
: Hirschhorn, "Introduction to Powder Metallurgy," American
\ ' Powder Metallurgy Institute, New York, 1969.) .
.J '
4
.
16-7 Sin1:ering The sintering process is usually carried out at a temperature below the highest melting constituent. In some cases the temperature is high enough to form a liquid constituent, such as in?the manufactura of
iting «ed by air
cemented carbides, vvhere sintering is done above the melting point of the binder metal. Irt other cases, no melting of any constituent takes place.
An
letched
,
taction is used to produce porous sheet
Ssj-y rechargeable batteries. The powder may rry (similar to the slip in slip casting) to
1
10 000 ,
r solid metal sheet to produce unusual
s 0,000
t
1
a most useful property since it is an in,
compaction and also determines the be-
6 000 ,
p
equent sintering. Green density seems to Jtion pressure increasing particle size or ;
,
' -
dness and strength of particles
,
2 000
and de-
,
5 dependence of green density at 30 tons/
ol 0
own in Fig. 16-13. The strength of green jchanical interlocking of the irregularities sing compaction pressure (Fig 16-14), or .
"
trength because more particle movement
bater interlocking.
40
80
120
1G0
Compoction pressure Itsil
Fig. 16-14 Dependence of green strength on compaction pressure for electrolytic iron powder. (From H. H. Hausner, "
Powder Metallurgy," Chemical Publishing Co., New York,
1974.)
i- -v;-;1 .
.
-. ..
.
.
I
1 i
....
620
I
INTRODUCTION TO PHYSICAL METALLURGY
0
Sintering furnaces may be either the electric-resistance type or gas- or
0 -
oil-fired type. |Close control of temperature is necessary to minimize varia5?
tions in final dimensions. The very uniform and accurate temperature of the electric furnace makes it most suitable for this type of work. Since bonding between particles is greatly affected by surface films the
CI
,
1 XT
formation of undesirable surface films, such as oxides, must be avoided.
-
1
This may be accomplished by the use of a controlled protective atmosphere. Another function of the atmosphere is to reduce such films if they are present on the powders before mixing and briquetting. The protective atmosphere should not contain any free oxygen and should be neutral or reducing to the metal being sintered. A dry hydrogen atmosphere is used in the sintering of refractory carbides and electrical contacts but most ,
7
commercial siiptering atmospheres are produced by the partial combustion of various hydrocarbons.
purpose.
Natural gas or propane is often used for this
1
Sintering is- essentially a process of bonding solid bodies by atomic
forcfes.
5
-
Sintefing forces tend to decrease with increasing temperature,
but obstructions to sintering-such as incomplete surface contact, presence of surface films, and lack of plasticity-all decrease more rapidly with Fig. 16-16 (a) Tungsten powder of 10 particle size plasma-arc-sprayed on a copper mandrel. Note distor structure of molten tungsten drops as splashed on e other, with appreciable voids (black) between each. E 10 g CuSO, - 5H20, 20 ml NH,OH, 40 ml H.,0, 500x. (fa same material after sintering 24 h at 3460 F in hydroge Note growth of equiaxed tungsten grains and substanl reduction in porosity. Same etch; 500x. (Courtesy of
increasing terriperature. Thus elevated temperatures tend to favor the sin-
tering process The longer the time of heating or the higher the tempera-
"
"
ture, the greater will be the bonding between particles and the resulting tensile strength.
°
Despite a great deal of experimental and theoretical work on the funda-
mental aspect1 of sintering, there is still much of the process that is not
Jones, Los Alamos Scientific Laboratory.)
understood. Ishe sintering process starts with bonding among particles as
the material Kpats up. Bonding involves diffusion of atoms where there is intimate contact between adjacent particles leading to the development
of grain boun|Jaries. This stage results in a relatively large increase in ]
strength and Hardness, even after short exposures to an elevated tempera-
ture. During tfie next stage, the newly formed bond areas, called
a
O
I [b) i<7i j Fig. 16-15 Schematic illusj ation of a three-sphere sintering model: (a) original poin| contacts, (b) neck growyi, (c) "
and (d) pore rounding. (Fr<5m J, S. Hirschhom, Introduction to Powder Metallurgy," American Powder Metallurgy Institute, New York, 1S69.)
5
i
/
"
;
.
"
necks,
I
grow in size, followed by pore roui terms of a three-sphere model in Fi age and eventual elimination. Th temperatures and times necessary a the distorted as-sprayed structure voids, whereas Fig. 16-165 shows th °
sintering for 24 h at 3460
F
.
16-8 Hot Pressing This method consists simultaneously. Molding and sintei results in higher densities and greal pressing as compared with cold co in gas content and shrinkage effects elongation, and density. Hot pressi marily for the production of very har shows the equiaxed single-phase st titanium carbide powder made in a
POWDER METALLURGY
621
.a
ier the electric-resistance type or
gas- or
mperature is necessary to minimize varia-
.
:
t
MM®
3
'
ery uniform and accurate temperature of |
1A
5t suitable for this type of work |L es is greatly affected by surface films the 9 .
3
,
films such as oxides must be avoided. ,
,
use of a controlled protective atmosphere lere is to reduce such films if they are .
mixing and briquetting
The protective my free oxygen and should be neutral or red. A dry hydrogen atmosphere is used bides and electrical contacts but most s are produced by the partial combustion i\ gas or propane is often used for this .
;
-;
v
' .
L i
IT
f
s
v
.
v
,
,
ess of bonding solid bodies by atomic decrease with increasing temperature
"
1
.
jch as incomplete surface
?
if-**
,
contact, pres-
slasticity-all decrease more rapidly with r ;: .
::
valed lomperalures tend lo favor the sin-
: -
M Fig. Hi H, (.i) Iiii)i|Mi>ii powdoi nl ID /< |>(lilii:lf m/
[ne of heating or the higher the tempera-
m plasma-arc-sprayed on a copper mandiot. Nole dislorled
iing between particles and the resulting
m other, wilh approcinblo voids (black) between each. Etch
m structure of molten
"
tungsten drops as "splashed
9 tO g CuSO, . 5HzO, 20 ml NH.OH, 40 ml H,0, 500x. (b) The in hydrogen.
'
antal and theoretical work on the funda-
is still much of the process that is not s starts with bonding among particles as
'
involves diffusion of atoms where there
jnt particles leading to the development results in a relatively large increase in
jfjgshort exposures to an elevated temperat
'
ewly formed bond areas called "necks ,
on each
same material after sintering 24 h at 3460
Note growth of equiaxed tungsten grains and substantial % leriuclinn in poiositv Snnm elnh. fiOfl- (C.ouil(!SV of I I
' Jones,
1
Lo;. AUnni1. .Smeiililic l.nlioinloiy )
grow in size, followed by pore rounding. This is shown schematically in terms of a three-sphere model in Fig. 16-15. The last stage is pore shrink age and eventual elimination. This stage is rarely complete, since the -
temperatures and times necessary are too impractical. Figure 16-16a shows the distorted as-sprayed structure of tungsten powder with considerable voids, whereas Fig. 16-16£) shows the substantial reduction in porosity after
)
sintering for 24 h at 3460
F
"
o
Id) Sinler-
i 16-8 Hot Pressing This method consists in applying pressure and temperature simultaneously. Molding and sintering take place at the same time, which results in higher densities and greater productions. The advantages of hot pressing as compared with cold compacting and sintering are a reduction in gas content and shrinkage effects, along with higher strength, hardness, elongation, and density. Hot pressing is used only to a limited extent, pri marily for the production of very hard cemented-carbide parts. Figure 16 17 shows the equiaxed single-phase structure and black pores of hot-pressed titanium carbide powder made in a graphite die at 4l60oF and a pressure .
-
fh
j
.
(c)
o-
etal-
-
.
I
i
1
-
!
622
INTRODUCTION TO PHYSICAL METALLURGY ;
)
'
/
r
This may require another sizing op
i
mm
the part.
>
Depending upon the particular ap heat-treated to obtain certain desi
r
may be a stress-relief or annealinc
r
compositions may be age-hardenei case-hardened by carburizing cyar
V 7
,
Various finishing operations may facture of powder metal parts Thes .
ing, burnishing straightening, debt ,
:T::Vr;
Protective surface coatings ma
spraying, and many of the other me
Of the various joining methods for powder metallurgy products.
c
,
;
i
f /1 i .'
J
» '
-
:
Fig. 16-17
Hot-pressed titanium carbide made from powder
in a graphite die cit 4150°F:knd at a pressure of 2,000 psi, showing equiaxed single-phase structure with black voids.
,
Etch one part HP arid fourjparts HNO:,; lOOx. (Courtesy of T
.
I. Jones, Los Alamos Scjentiflc Laboratory.) I
,
I
of 2,000 psi, jThe principal disadvantage of this method is dies to stand up under pressure at elevated temperatures. .
the high cost of
16-9 Supplemental Operations For applications that require higher density or ]
sections that are difficult to fill bef(
operation known as coining or repressing. Coining serves the purpose of condensing the sintered compact. It is possible to obtain considerable J
density after pressing. Also, since |: zontal, grooves and cutouts perpenc not be formed conventionally and r are also limited by press capacity, I Tolerances of P/M parts are influe sintering temperature, sintering time
may be made directly from powder. The restricted plastic deformation J within the die also allows, in many cases, close dimensional tolerances to I be held without the necessity of costly machining. j Heating forsintering may be interrupted at some intermediate tempera- | ture. This is known as presintering. At this point, the compact may have |
good machinability or be sufficiently soft to allow the use of operations that .| are not feasible after sintering.
I I
In some cases, resintering after repressing will increase the mechanical j >. properties considerably. As compared with straight-sintered metal, resin--j p tering may increase the tensile strength of copper by about 60 percent and *
of iron by about 30 percent. Despite the increase in strength, resintering| may result in? large grain size and loss of dimensions due to shrinkage.-
.
j -
! - -
-
I
,
close dimensional tolerances, sintering is followed by a cold-working
plastic deformation within the die, resulting in more complex shapes than |
1
Impregnation is the means used compact. This is carried out prim; as in the self-lubricating bearings. > by dipping the parts in a container i of the pores by vacuum and then fo sure. Waxes and greases may also low-melting metal impregnant sue spongy matrix of nonferrous alloys of the metal. Impregnation with liq specific gravity of iron-base parts. 16-10 Design of P/M Parts The configur largely on the compacting operatic freely, sharp corners long, thin si
i
'
have marked effects on tolerances
tolerances cost considerably less 1 tolerances. Precision-tolerance pi
curacy, maintenance, and replaceme not exceed what is absolutely requii
Figure 16-18 shows several desigr is most ideally suited for the pro shapes, that do not have large vari Surface indentations or projections
:
POWDER METALLURGY
623
f
This may require another sizing operation
,
which will increase the cost of
the part. Depending upon the particular application the sintered compact may be heat-treated to obtain certain desirable properties. The heat treatment ,
r
may be a stress-relief or annealing treatment. Suitable nonferrous alloy compositions may be age-hardened. Steels may be quench-hardened or case-hardened by carburizing, cyaniding, or nitriding. Various finishing operations may be carried out to complete the manufacture of powder metal parts. These include machining shearing, broaching, burnishing, straightening, deburring, grinding, and sandblasting. Protective surface coatings may be applied by electroplating, metal spraying, and many of the other methods described in Chap. 14.
»
,
Of the various joining methods, only brazing has been used.extensively for powder metallurgy products.
'
.
Impregnation is the means used to fill the internal pores in the sintered compact. This is carried out primarily to improve antifriction properties, as in the self-lubricating bearings. Oil impregnation may be accomplished by dipping the parts in a container of hot oil, or by first drawing the air out of the pores by vacuum and then forcing the oil into the pores under pressure. Waxes and greases may also be used as impregnants. The use of a low-melting metal impregnant, such as tin and lead babbitt alloys in a spongy matrix of nonferrous alloys, tends to improve the bearing properties of the metal. Impregnation with liquid lead has been used to increase the specific gravity of iron-base parts.
1
om powder
'
V 000 psi, ck voids.
urtesy of
.
i L
dvantage of this method is the high costof at elevated temperatures
.
.
,
ie, resulting in more complex shapes than -1 wder. The restricted plastic deformation J my cases close dimensional tolerances to I ,
:
costly machining.
j
interrupted at some intermediate tempera- J
ring. At this point the compact may have } ,
££3ntly soft to allow the use of operations that 1 er repressing will increase the mechanical
g;:|:npared with straight-sintered metal resin- | ,
strength of copper by about 60 percent and I espite the increase in strength resintering ;; > ,
md loss of dimensions due to shrinkage
ie io Design of P/M Parts The configurations possible with P/M parts depend -
mg: jpplications that require higher density or w sintering is followed by a cold-working 1 repressing. Coining serves the purpose of ' :iact. It is possible to obtain considerable I .
largely on the compacting operation.
Since metal powders do not flow
freely, sharp corners, long, thin sections, and large variations in cross sections that are difficult to fill before pressure is applied can have lower
density after pressing. Also, since press action is vertical rather than horizontal, grooves and cutouts perpendicular to the direction of pressing cannot be formed conventionally and must be made by machining. Designs are also limited by press capacity, length of stroke, and platen work area.
Tolerances of P/M parts are influenced by many variables. Alloy, density, sintering temperature, sintering time, and coining or resizing operations all have marked effects on tolerances. P/M parts manufactured to regular tolerances cost considerably less than those manufactured to precision
tolerances. Precision-tolerance parts are more demanding in tool accuracy, maintenance, and replacement so that specified tolerances should not exceed what is absolutely required. Figure 16-18 shows several design tips for P/M parts Powder metallurgy is most ideally suited for the production of cylindrical or rectangular shapes, that do not have large variations in cross-sefctional dimensions ,
.
.
.
Surface indentations or projections can easily be formed on the tops or
{
.
624 INTRODUCTION TO WsiCAL METALLURGY
Avoid
Preferred
Preferred
Avoid
to make
,
'
YZZZZZZZZZZZA
I
Les s
Greater
mm
mm
than
,
- 030 in. 0 .
VrnwA
T
mm
than
IP m
0.030 in.
16-11 Applications of Powder Metallurc used for the production of refractor
Should be machined
T
als
Sidewalls should be thicker than 0.030 in.
,
Part has lower strength if all steps are pressed
.
and metal-nonmetal combinati
method for certain parts Table 16-; tion, properties and applications c The high melting points of the ref
j h f
but have greater tender
tenance costs.
.
,
JL
o
o
4°
A tapered counterbofe strengthens the tool
the conventional melting and casti
II
and aids inltool removal
method for the manufacture of tu
beginning of this chapter This tec of producing molybdenum tantalu One of the outstanding uses of p
Avoid deep, narrow splines
.
,
hard materials in a metallic matrix
'
,
carbide products.
Eliminate separate key and keywoy, a keyed bushing can be pressed Rounded corners permit uniior'm powder flow in the die 0 005 10 0 008 in .
1i/
In the producti
a suitable mixture of the carbides c
cobalt as a binder is compacted a materials can be cut machined, an ,
,
Undercut must be machined
(
1 1
Undercut can
.
1
Li.
V ! .
j
.
be pressed
pact is then subjected to a high-terr ation during which the liquid coba a solid piece. Cemented-carbide strength, red-hardness and wear ,
End a taper with a small land area
Flange relief caiji be pressed to save machining
brittle, they are usually employed as
Fig. 16-18 Design tips foi P/M parts. (From Machine
also used as liners for wear-resistai
Design, Metals Referen£e;lssue, The Penton Publishing
Other examples in this classificat and dressing
Co., Cleveland, 1970.)
wheels, drill-core bits
i,
,
bedded in cemented carbides or mi Metal-nonmetal combinations ha\
1
bottoms of parts, and flanges can be formed at either end. Splines, gear
friction materials such as clutch fac
teeth, axial holes, counterbores, straight knurls, slots, and keyways are
contain a metallic matrix of copper graphite to form a smoothly engag or emery for frictional purposes. friction and prevent seizing. Copp current-collector brushes and in po
easily formed. However, undercuts, holes at right angles to the direction of
pressing, reverse tapers, reentrant angles, and threads cannot be pressed. \ In many case parts requiring features that cannot be pressed directly can still be produced economically by pressing to semifinished shapes and machining the desired details. Thin walls, narrow splines, and sharp
/ V/
-
"
:
Composite metals are metal comt
corners shoujd be avoided. Side walls of varying thickness should not be
of each metal for particular applicati
less than 0.030 in,, and uniform walls should not be less than about 0.060
useful for alloys of metals that are
in. for parts o\ any appreciable length. Also, abrupt changes in wall thickness and feather edges should be avoided, and corners should be rounded.
Small, long lloles should not be used, and a minimum flange overhang of about 1/16 in. at heads or shoulders is preferred to no overhang. These
precautions will facilitate production as well as minimize tool costs. Sharp edges and Idhg, thin sections on punches or core pins not only are costly k
\ i
/ \
monotectics.
Casting tends to pro
techniques are used whereas home from powders. The electrical Indus ,
the production of heavy-duty contac to abrasion and arcing of a refractor conductivity of silver or copper. Sir
l!
I
iGY
POWDER METALLURGY
Avoid -
Preferred
cnlor
to make, but have greater tendency to break and require higher main-
13
tonance costs.
-OiOin,
-
i
r
1
Should be machined
16-11 Applications of Powder Metallurgy
H
Powder metallurgy techniques are
used for the production of refractory metals composite metals, porous metals and metal-nonmetal combinations, and as a more efficient production method for certain parts. Table 16-3 on pages 626-629 shows the composition, properties, and applications of some powdered metals. ,
Part has lower strcnqth if all sfeps are pressed
C Uu ]
625
3
,
The high melting points of the refractory metals make-it impossible to use the conventional melting and casting techniques. The use of the powder
O
method for the manufacture of tungsten filaments was described at the
beginning of this chapter. This technique offers the only practical method of producing molybdenum tantalum, and other metals of the same group. One of the outstanding uses of powder metallurgy is the combination of
Avoid deep narrow splines ,
m
,
hard materials in a metallic matrix which serves as the basis for cefnented,
carbide products. In the production of cemented-carbide cutting tools, a suitable mixture of the carbides of tungsten, tantalum and titanium with cobalt as a binder is compacted and presintered. In this condition, the ,
Rounded cornns porrni! unifoi'm powder flow inth'Mlip 0 005 to 0 008 in.
materials can be cut, machined and ground to the final shape. The com-
.
,
.
pact is then subjected to a high-temperature (about 2750oF) sintering oper-
m\m
ation during which the liquid cobalt binds the hard carbide particles into
j
h i
a solid piece. Cemented-carbide tools are noted for high compressive strength, red-hardness, and wear resistance. Since they are relatively brittle, they are usually employed as brazed-on tips to a steel tool. They are also used as liners for wear-resistant applications. ,
[ nil n luper with 0 snuill land ineo ;hine
S h 1R g
Other examples in this classification are diamond-impregnated grinding wheels
,
drill-core bits, and dressing tools. These consist of diamonds em-
bedded in cemented carbides or more plastic metals and alloys. Metal-nonmetal combinations have found wide use in the manufacture of ,
ian be formed at either end. Splines, gear
friction materials such as clutch facings and brake linings
.
s,
straight knurls, slots, and keyways are \
'
uts, holes at right angles to the direction of ant angles, and threads cannot be pressed. I ;atures that cannot be pressed directly can
These materials
contain a metallic matrix of copper or bronze for heat conductivity, lead or graphite to form a smoothly engaging lining during operation, and. silica or emery for frictional purposes. Iron is sometimes added to increase friction and prevent seizing Copper-graphite combinations are used as current-collector brushes and in porous bronze and iron bearings .
by pressing to semifinished shapes and Thin walls, narrow splines and sharp } ,
\e walls of varying thickness should not be ; v
'i walls should not be less than about 0.060
;
length. Also, abrupt changes in wall thickDe avoided, and corners should be rounded.
HS's used, and a minimum flange overhang of jlders is preferred to no overhang. These *
ction as well as minimize tool costs. Sharp if n punches or core pins not only are costly ;«
.
Composite metals are metal combinations that retain the characteristics
of each metal for particular applications. Powder metallurgy is particularly useful for alloys of metals that are not soluble in the liquid state or form monotectics. Casting tends to produce a two-layer alloy unless special techniques are used, whereas homogeneous mixtures are easily produced from powders. The electrical industry makes use of composite metals in the production of heavy-duty contacts which combine the high resistance to abrasion and arcing of a refractory metal such as tungsten with the high conductivity of silver or copper. Similarly the lubricating qualities of lead ,
m
:
;
-
'
i1' I
s TABLE 16 3
Properties and Applications of Some Powder Metals*
MATERIAL AND CONDITION
SPECIFICATION NUMBERS
H
TYPICAL
COMPRES-
DENSITY
TENSILE
SIVE YIELD
G PER
PCT,
STRENGTH,
STRENGTH,
CuCm
THEO.
PSI
PSI
HARDNESS'
SURFACE
CORE
ELASTIC
JJ
-
o
MODULUS
ELONGA-
(TENSION),
TION,
PSI
%
o c
o H
APPLICATIONS
o z
6
.
.
0
. .
76
2C.0P0
15,000
83
30,000
25,000
.
R1.38 Rf 55
,
£138.
12.70.0,000
Rf55
15,500,000
3 0,
Structural (lightly loaded gears), magnetic (motor pole pieces).
o
projectile driving bands, selfAs sintered
Unalloyed iron, 99.9% Fe
65 .
10
MPIF 35: F-0000 N through T SAE: 850, 851.852. 853,
7 5 .
96
7 5
96
40,000
35,000
Rf 95
Rf95
25,500,000
35,000
Rc65
Rt9b
25,500,000
lubricating bearings-may be
<
steam treated for wear resistance
w
o >
20
855;J471C Carbonitrided
ASTM: B310, B439
.
100,000
(core)
0 040 in. case
to
0 80 C
140,000
.
.
0 5 case .
to 60
20 core
Structural, wear-rebisting
(small levers and cams)
m
> r;
c
As sintered
62
79
31,000
24,000
68
87
44 000
30,000
68
87
64,000
.
Rb 15
Rbl5
13 300,000
.
Structural (moderately loaded gears, levers and cams)
.
Iron-carbon sinteredsteel
MPIF 35: F-0001-Pthrough
As sintered
.
,
3)
05
,
08 C
Rb65
Rb65
18,300,000
1.0
Rc3S
Rb65
18 300,000
05
Structural (moderately loaded gears, levers, and cams requiring wear resistance)
Rh80
11,600,000
05
Structural (pump housings.
1.0
support brackets, tightly loaded gears), self-lubricating bearings -may be steam-treated for
0 8C
F-0008-P
.
SAE: 852 Heat-ireated
ASTM:B310
As sintered
Low-carbon iron-copper alloy,
,
,
to 45
(farge parts)
58 .
74
29,500
20,000
Rh80
.
10.5 Cu
0 25 C max, 7.0 to 25.0 Cu, .
Otol.O Mn.Otol.OZn.
As sintered
balFe
10.5 Cu
62 .
80
34,000
21,000
Rh90
Rh90
13,800.000
wear resistance
MPLF 35: FC0300 N through T. FX2000-T (infiltrated)
Copper-
SAE: 870, 872
infi It rated,
ASTM: B303A (infiltrated),
25 Cu up
High-impact structural parts, socket adaptors-may be
to4 Ni
carbonitrided for wear resistance
B222
75 .
96
60,000
55,000
Rb60
Rb60
25,500,000
10
iirn Timiliii i nii '
TABLE 16-3
-
'
nfinil
(Continued) DENSITY
MATERIAL AND
SPECIFICATION
CONDITION
NUMBERS
COMPRES-
TYPICAL
TENSILE
SIVE YIELD
HARDNESS
G PER
PCT
STRENGTH
STRENGTH.
Cu Cm
THEO.
PSI
PSI
63
81
T
ELAST1C
MODULUS
ELONGA-
(TENSION)
TION %
SURFACE
CORE
PSI
Rb50
Rb50
14 400 000
15
Rb70
13.800 000
08
.
APPLICATIONS
As sintered 3 0 Cu, .
.
45 000 ,
,
,
.
08C .
Iron-copper-carbon alloy
,
3.0
to 25.0 Cu 0.3 to 1.0 C.
As sintered
62 .
80
60,000
54,000
.
.
5 0 Cu 0.8 C
.
.
,
bal Fe plus up to 0 40 Mn .
or 1.5 Zn. or 1.5 Co.
heat-treated
FX2010-T (infiltrated)
5 0Cu 0 8C
.
SAE:864-B
866-A, 867-B
,
872
,
.
.
62 .
80
65,000
80,000
Rc30
Rb 100
13 800,000 ,
05 .
brackets, levers
and ratchets)can be heat-treated to a high ,
.
,
,
Infiltrated
ASTM; B303 B. C (infiltrated) B426
Structural (medium loads including gears cams, support
Sintered and
MPIF35:FC0308 FC0508,
74 ,
95
80,000
degree of wear resistance 70 000 ,
Rb90
20.0 Cu 0.8C .
,
Infiltrated heat treated ,
74 .
95
100.00(1
flo nnn
Or- AO
Rb90
24.600 000 .
07 .
o <
i
Vl'
'
"
Mbsimvjreu-
m
'
*
'
a
Structural (moderately loaded
0 8C iron-carbon-sintered steel
<
gears, levers and cams) As sintered
MPIF 35: F-0001 -P-nrough
68
87
44,000
68
87
64.000
,
icaoo
flb65
Rb65
18 300.000
10
Rc35
Rb65
18.300.000
05
,
,
.
1 1
"
1
,
0 8C
F-0008-P
.
SAE: 852 Heat-treated
ASTM: B310
.
.
to 45
(large parts)
Structural (moderately loaded gears, levers, and cams
requiring wear resistance) As sintered
58 ,
7*
29.500
20 000
Rh 80
Rh 80
11 600.000
05
,
,
Structural (pump housings support brackets lightly loaded gears), self-lubricating bearings .
iO,5Cu
Low-carbon iron-copper altoy
,
0 25 C max. 7,0 to 25,0 Cu, .
0to 1.0 Mn, Otol.OZn.
As sintered
balFe
10,5 Cu
6 2 ,
8C
34,000
2
'
-
000
Rh 90
Rh 90
13 800,000
10
,
,
-ma
y be steam-treated for
wear resistance
MPIF 35: FC030O N ihrough T. FX2000'T (infiltrated)
Copper-
SAE: 870, 872
infiltrated,
ASTM: B303A (infiltrated),
25Cuup
socket adaptors-may be
lo4Ni
carbonitrided for wear resistance
7 5 ,
96
60,000
55.000
Rb 60
Rb60
25,500.000
10
High-impact structural parts
.
B222
TABLE 16-3
(Continued) COMPRES-
TYPICAL
TENSILE
SIVE YIELD
HARDNESS
STRENGTH.
STRENGTH
PSI
PC-
DENSITY ' ' .
ATERIAL AND
CONDITION
S==CIFICAT10N JMBERS
' .
G PER
POT
Cu Cm
THEC
6 3
61
45,000
62
80
60,000
,
ELASTIC MODULUS
ELONGA-
(TENSION)
TION, %
SURFACE
CORE
PSI
Rb 50
Rb50
14,400.000
15
Rb 70
13,800,000
08
APPLICATIONS -
As sintered 3 0 Cu, .
.
.
08C ,
As sintered
iron-copper-carbon alloy 3,0 ,
.
54.000
.
5 0Cu,0.8C
to 25.0 Cu, 0.3 to 1,0 C.
.
Structural (medium loads including gears, cams, support
bal Fe plus up to 0.40 Mn Sintered and
or I .SZn. or 1.5 Co.
62 .
60
65.000
sc :oo
Rc30
Rb 100
13,800 000
05
74
93
80,000
7C,CC0
Rb90
Rb 90
24,600,000
07
74
95
100 000
50 000
Rc 42
Rb 100
24 600,000
05
72
92
110,000
Rc45
Rc20
22.500.000
07
Structural, wear-resisting oil pump gears)
As sintered
72
92
60,000
R0 68
30
Structural (couplings)
Heai-lreated
7 2
92'
145,000
RC44
1 1
Structural, wear-resisting (oil pump gears and heavily loaded support brackets)
MPIF 35: FC0308. FC0508.
neat-treated
?X2010-T (infiitrated)
5 0Cu,0.8C
,
.
-
SAE: 864-B. 866-A. 5-57-6. Infiltrated
372
.
rackets, levers and ratchets)can be heat-treated to a high degree of wear resistance ,
.
20.0 Cu, 0.8 C
ASTM: B303 S, C (infil.
trated), 8426 Infiltrated, heat treated
.
.
,
.
20.0 Cu, 0.8 C I
Low-alloy steel, cherrucally AISU630,0.30 C com-
Dined).0.25 Mc
"
55 Ni
Coined,
resintered, '
ieat-treated
.
iprealloy type) "
ow-ailoy steel, cnemicaliy
-
.
.
.
22
500,000
,
.
AISI4650,0.45 C (com-
bined), 0.50 Mo. 2.0Ni (diffusional alloy type) Coined
ASTM: 8484
resintered,
76 .
97
200,000
heat-treated
M
m
f m
MPIF 35: FM-0205-S and T ,
0
O
etal Progress Data Sheet, American Society for Metals, Metals Park, Ohio. April 1971.
Rc56
15 ,
Structural, wear- and impactresisting (oil pump gears to 3 000 psi and heavily loaded transmission gears) ,
> r c 3J
O < O) ro -
4
:
,
'f,:
'
.
'
<;
.
;: v >. -
; ;.
.
:
Mm.
lf;
>;;'':'
i:
w
:
to GO
(Continued)
TABLE 16 3 DENSITY TENSILE
MATERIAL AND CONDITION
SPECIFICATION
G PER Cu Cm
NUMBERS
PCT..
THEO.
STRENGTH,
COMPRES-
TYPICAL
SIVE YIELD
HARDNESS
STRENGTH
ELASTIC
JO
MODULUS
ELONGA-
(TENSION),
TION,
PSI
%
SURFACE
PSI
PSI
CORE
o c
o
APPLICATIONS
H
o
68
87
100 000
65.000
Rc35
Rc20
18,300.000
05
Structural, wear-resisting, highstress (planetary differential and transmissiorrgears up to 6 hp)
74
95
175.000
125.000
RC45
Rc30
24,600,000
2 5
Structural, high-stress, impact resisting (shifter lugs and clutches)
Heat-treated '
4
.
0 Ni, 1.0 Cu
.
.
.
"
Nickel ailoy steel, 4.0 to
0 70 C
'
7
o
.
0 Ni,a0to2.0Cu.
.
Heat-treated
0 0 to 0.80 C .
Nearest equivalent: AISI 2517
7 0Ni,2.0Cu,
.
.
.
o 5:c
(diffusional alloy type)
z -
i
o
< cn
o > i
.
MPIF 36: FN-0408-S and T, Carbonitrided
FN-0705-S and T
0 040 in case
ASTM: B484
.
4
89
69 .
Rc45
90.000
Rb 65
19,600,000
35 .
(core)
0 Ni, 1.0 Cu,
.
00 .
C (core)
Structural, wear-resisting, highstress, and requiring welded assembly (welded assembly of pinion and sprocket)
rn
> c
JO
o <
-
Prealioyed, as sintered
62
80
50,000
20,000
Rb42
66
86
60.000
45,000
Rb60
.
Rb42
40
Rb60
10
.
(dis. NHj) Sintered austenitic stainless
steei, type 316L
Prealioyed,
MPIF35:SS-316L, P
as sintered
.
Structural, corrosion-resisting. nonmagnetic (small gears, levers, cams and other parts for
(dis. NH3)
through R
,
exposure to salt water and
ASTM: B525
specific industrial acids)
Prealioyed, as sintered
12.5
66
86
45,000
62
80
65.000
35,000
Rb90
.
(dry hydrogen) Prealioyed, as sintered
.
35
Rb90
.
Sintered martensitic stainless
(dis. NH,)
steel, type 410L MPIF 35: SS-41 ON through P
Heat-treated
62
80
70.000
45 000
Rc30
08
Heat-treated
68
88
100 000
80.000
Rc30
08
TABLE 16 3
.
.
,
.
.
Structural, corrosion-resisting. nonmagnetic (small gears, levers, cams, and other parts for exposure to salt water and specific industrial acids where applications require heat treating for wear resistance)
,
(Continued) DENSITY
MATERIAL AND SPECIFICATION
CONDITION
NUMBERS
COMPRES-
TYPICAL
TENSILE
SIVE YIELD
HARDNESS
G PER
PCT
STRENGTH,
STRENGTH,
CuCm
THEO
PSI
PSI
73
82
28,000
79
89
35,000
ELASTIC
MODULUS
ELONGA-
(TENSION),
TION,
PSI
%
SURFACE
CORE
20,000
Rh 74
Rh 74
40 000
Rh84
Rh 84
15
Rh 61
Rh 61
10
APPLICATIONS
Copper-nickel-zinc alloy (nickel silver), 62.0 Cu,
.
18.0 Zn 18.0 Ni, 2.0 Sn
MPIF35: BZN-1818U
W
as sintered
,
Sintered brass. 77 to 80 Cu
Prealioyed
,
0 to 2.0 Pb, bal Zn
,
MPIF35:BZ0218-T and U SAE: 890, 891 ASTM:B282
Sintered bronze 86 to90 Cu 9 5 to 10.5 Sn 1.0 max ,
.
,
Mechanical components
76
87.5
resintered
80
92
31,000
Re 49
Blended
68
76
16 000
Rh50
.
29,600
,
copper-tin
.
,
,
atmos-
pheric-corrosion-resisting (builders hardware, mechanism housings lock parts, pump housings)
Coined and .
resisting (gears levers, chuck jaws, parts for marine exposure) ,
,
as sintered
.
Structural nonacid corrosion,
.
ASTM: B458
1
7
Single-pressed
,
,
24
2 5 .
11
"
a <
Prealloyed, as sintered
62
80
50,000
20,000
Rb42
Rb42
4 0
66
86
60,000
45,000
Rb60
Rb60
10
.
.
(Cis. NH3) Sintered austenitic stainless
steel, type 316L
P-ealloyed,
MPIF35:SS-316L, P
as sintered
Structural, corrosion-resisting
.
.
nonmagnetic (small gears levers, cams and other parts for ,
through R
idis. NH3)
,
ASTM: B525
exposure to salt water and
Prealloyed
specific industrial acids)
,
as sintered
66
86
45,000
62
80
65,000
.
12.5
(dry hydrogen) Prealloyed, as sintered
Structural .
35,000
Rb90
Rb90
35
,
corrosion-resisting,
nonmagnetic (small gears levers cams, and other parts for
.
,
Sintered martensitic stainless
(dis. NH3)
steel, type 410L MPIF35: SS-410Nthrough P
Heat-treated
62
80
70,000
45 000
Rc30
08
exposure to salt water and specific industrial acids where
Heat-treated
68
88
100.000
30,000
Rc30
08
applications require heat treating for wear resistance)
,
.
.
,
.
.
i
mm
mm
TABLE 16-3 (Continued) COMPRES-
TYPICAL
TENSILE
SIVE YIELD
HARDNESS
DENSITY MATERIAL AND SPECIFICATION
CONDITION
NUMBERS
Copper-nickel-zinc alloy (nickel silver), 62.0 Cu, 18.0 Zn, 18.0 Ni, 2.0 Sn MPIF35: BZN-1818U, W
G PER
PCT.
STRENGTH.
STRENGTH.
Cu Cm
THEO.
PSI
PSI
73 .
82
28,000
79
89
35,000
76
87.5
29,600
ELASTIC
MODULUS
ELONGA-
(TENSION).
TION
PSI
%
SURFACE
CORE
20,000
Rh74
Rh 74
7
40 000
Rh84
Rh84
15
Rh61
Rn 61
10
24
APPLICATIONS
Structural, nonacid corrosion-
resisting (gears, levers, chuck jaws, parts for marine exposure)
Single-pressed as sintered
.
,
,
ASTM: B458
Sintered brass, 77 to 80 Cu. 1 0 to 2.0 Pb, balZn .
Mechanical components, atmospheric-corrosion-resisting (builders hardware, mechanism housings, lock parts, pump housings)
Prealloyed, as sintered
.
MPIF35:BZ0218-T and U SAE: 890, 891
Coined and
ASTM: B282
resintered
80
92
31,000
Re 49
Blended
68
76
16,000
Rh50
25
72
81
30,000
Rh65
30
Rr* 57
25
Sintered bronze, 86 to 90 Cu, 9 5to10.5Sn, 1.0 max
copper-tin
Fe, 1.75 max C
powders,
.
MPIF35: BT0010N through R
single-pressed
ASTM: B255
as sintered
As sintered
.
.
.
.
(8,000 graphited)
2 59 .
28,300
Rh57
I
Structural, atmospheric-corrosionresisting, bearings (journal bearings, thrust bearings, load-carrying bearing plates)
.
and coined
Lightly loaded gears and ratchets, camera parts, circuit board, heat sink.
O
Sintered,
cabinet hardware
D
Sintered
Alcoa Type 201 AB4 4 Cu, 0.8 Si, 0.4 Mg,
.
2 63 .
30
32,200
.
i
0
.
bal Al
coined,
heat-treated
2 63 .
49,400
Re 87-
Re-87-
20 .
2 m
>
JO
O -
<
;
630
INTRODUCTION TO PHYSICAL METALLURGY
are combined with the load-carrying ability of copper In the copper-lead bearings. Controlled porosity of powder metal parts has led to the production of porous bearings, gears, and filters. S,elf-lubricating bearings are made of bronze powder with controlled porosity after sintering. The pores are subsequently filled with oil. In operation, the load on the bearing and the increased heat set up by the moving part within the bearing force the oil out ]
of the pores to provide automatic and uniform lubrication. Self-lubricating bearings are used extensively in the automotive industry and in washing machines, refrigerators, electric clocks and many other types of equipment. Porous-metal gears are used in oil pumps for their lubricating prop-
I
3
,
0
erties. Metal filters, used in the chemical industry, are similar to the ceramic type but have higher mechanical strength and resistance to both
mechanical |and thermal shock. Finally, in many applications the use of powder metallurgy techniques results in more economical manufacture of the part. Where load conditions are not severe, small gears, cams, levers, sprockets, and other parts
%
of iron, steel, brass, or bronze may be molded from powders to reduce ;
greatly or completely eliminate expensive and time-consuming machining I and other forming operations.
For example, the gears of a gear-type oil , \
Fig. from i iy. 16-19 kj-ic/ Typical i y iudi parts pans produced p
pump must, have accurately formed involute teeth or the pump will be in- ] (p. j. stokes Corporation.)
powder metals
efficient. The machined gear is cut from a cast blank by a skilled machinist i
with about 64 percent of the metal lost in chips. On the other hand, any semiskilled man can fill a hopper and operate a press which can turn out hundreds of these gears with dimensional accuracy and with less than 1
16-4 Contrast mechanical and hydraul
vantages disadvantages, and application ,
percent of the metal as waste. | Small Alniico permanent magnets containing aluminum, nickel, cobalt, | and iron may be made from powders or by casting. The cast alloy is dif- J
16-5
Why is sintering carried out in a ci
16-6 Why do elevated temperatures tei sintering forces tend to decrease with im
grinding. These magnets may be molded of powders directly to size and
16-7 What are the advantages and disac cold compacting and sintering? 16-8 Give three specific applications o
shape and their dimensions held to acceptable tolerances during sintering. ;|
these parts may be manufactured by othe
ficult to ma'chine, and finishing to dimensions must be done by tedious In addition, a finer grain size and greater mechanical strength are obtained..| in the sintered magnets.
s
Some typical parts produced by powder metallurgy techniques are shown » in Fig. 16-19.
t
powder metallurgy method.
16-9
Why is pore size important in the
How may pore size be controlled?
16-10
Why are canned powders often e
techniques?
REFERENCES QUESTIONS
16-1
Why isiparticle-size distribution important in the packing of powders?
"
16-2 Discuss the importance of particle shape on the properties of sintered cpm-,J pacts? ?! 16-3 List the three common methods of powder production and discuss their in-j fluences on the properties of the final product.
J
/
American Society for Metals: "Metals Han : Powder Metallurgy In Nuclear En
i
American Society for Testing Materials: " Products," Philadelphia 1953. Clark, F. H.: "Advanced Techniques in Pc ,
New York 1963. ,
1 URGY
1
POWDER METALLURGY
631
>;
carrying ability of copper in the copper-lead | i
. -
; der metal parts has led to the production of ;|
-
e
filters. Self-lubricating bearings are made of |
d porosity after sintering. The pores are sub-J
peration, the load on the bearing and the in*| . '
oving part within the bearing force the oil out.
0
'
atic and uniform lubrication. Self-lubricatingJ ?
o
&
o
m
y in the automotive industry and in washing ; v v/ ;
v
:
trie clocks, and many other types of equip3 used in oil pumps for their lubricating prop- <
-
-the chemical industry are similar to
"
"*
0
,
9
*
y mechanical strength and resistance to botll l |
*0
ok.
ns the use of powder metallurgy techniques manufacture of the part. Where load condiiars, cams, leverk, sprockets, and other parte:
I
ze may be molded from powders to reduce,| te expensive and time-consuming machinin f s For example, the gears of a gear-typfi m '
'
.
formed involute teeth or the pump will be lrf|
o
+m O O 16-19 Typical parts produced from powder metals. J. Stokes Corporation.) .
: .
is cut from a cast blank by a skilled machinist metal lost in chips. On the other hand, ari$
jpper and operate a press which can turn ou p dimensional accuracy and with less than!
16-4
Contrast mechanical .and hydraulic compacting presses with regard to ad-
vantages, disadvantages, and applications.
i
16-5
Why is sintering carried out in a controlled-atmosphere furnace?
16-6 Why do elevated temperatures tend to favor the sintering process although
magnets containing aluminum, nickel, cobali| powders or by casting. The cast alloy Is dtH> Jng to dimensions must be done by tedioui
sintering forces tend to decrease with increasing temperature? 16' 7 What are the advantages and disadvantages of hot pressing as compared with cold compacting and sintering?
ay be molded of powders directly to size ah]
16-8 Give three specific applications of powder metallurgy parts
leld to acceptable tolerances during sintering
these parts may be manufactured by other methods and give the advantages of the
and greater mechanical strength are obtained! id by powder metallurgy techniques are shown*
.
Describe how
,
powder metallurgy method.
16-9 Why is pore size important in the manufacture of self-lubricating bearings? How may pore size be controlled?
16-10 Why are canned powders often evacuated in powder extrusion and forging techniques?
S
REFERENCES
American Society for Metals: "Metals Handbook
" ,
1948 ed., Metals Park Ohio. ,
oution important in the packing of powders? If particle shape on the properties of sintered com-
American Society for Testing Materials: "Testing Metal Powders and Metal Powder
ethods of powder production and discuss their in-
Clark, F. H.: "Advanced Techniques in Powder Metallurgy
:
"
Powder Metallurgy in Nuclear Engineering
" ,
Metals Park, Ohio, 1958.
Products," Philadelphia 1953. ,
e final product.
New York, 1963
.
" ,
Rowman and Littlefield,
632 INTRODUCTION TO PHYSICAL METALLURGY
Goetzel, C. G.: "Treatise on Powder Metallurgy, vols. 1 to 3, Interscience Publishers, „, v , Inc., New York, 1949-1952. Hausner, H. H.: "Powder Metallurgy, Chemical Publishing Company. Inc., New York, "
m
j
"
1947. "
(ed.): "Modern Developments In Pewder Metallurgy, York, 1966. K
.
,
H. Roll, and P. K. Johnson (eds.):
Plenum Press, New
Iron Powder Metallurgy,
"
Plenum
Press, New York, 1968.
'
Hirschhorn, J. S.: "Introduction to Powder Metallurgy, American Powder Metal- i lurgy Institute, New York, 1969. The Iron and Steel Institute: Symposium on Powder Metallurgy, .Special Report38,
FA AN
"
London, 1947.
Jones, W. D.: 'I.Fundamental Principles of Powder Metallurgy," Edward Arnold Publishers Ltd., London, 1960.
Leszynski, W. J. (ed.): Powder Metallurgy, Interscience Publishers, Inc., New York, "
"
1961.
"
Metal Powder Industries Federation; Powder Metallurgy Equipment Manual, New "
17-1
York, 1968.
Poster, A. R. (ed.): "Handbook of Metal Powders, Van Nostrand Reinhold Com-
Introduction
When one considers th
"
are fabricated and placed In servia prematurely. Simply from a statistic present engineering practice, to expt the number of failures of a particuli important because they may affect liability. In some cases particularly injury or death, it will lead to expensi motive manufacturers under proddini
pany, New York, 1966. Schwartzkopf, P.: "Powder Metallurgy, The Macmillan Company, New York, 1947. "
i
,
I
;
dogs to recall millions of cars to coi even though the actual number of fa The purpose of this chapter is to bri failure and to illustrate some of the f.
illustrations in this chapter were take failure-"How Components Fail" by for Metals 1966) and "Why Metals F (Gordon and Breach Science Publishe 17'2 Procedure In any failure analysis it is as possible from the failed part itself £ ,
1 t
ditions at the time of failure. Some o
1 1 1
1
.
2
.
3
.
4
.
5
.
6 7
.
.
How long was the part in service? What was the nature of the stresses at
Was the part subjected to an overload Was the part properly installed? Was it subjected to service abuse?
Were there any changes in the environ Was the part properly maintained?
A study of the fractured surface she t
1 «
I t
1
.
2
.
Was the fracture ductile, brittle, or a cc Did failure start at or below the surface
r
-
Mallurgy," vols. 1 to 3, Interscience Publishers,
hemical Publishing Company, Inc., New York, | in Powder Metallurgy," Plenum Press, New -l r
)n (eds.);
Iron Powder Metallurgy," Plenum \
"
owder Metallurgy, American Powder Metal- f "
'
\ ium on Powder Metallurgy," Special Report 38, | s of Powder Metallurgy
" ,
FAILURE ANALYSIS
Edward Arnold Pub-
iurgy," Interscience Publishers, Inc., New York, ;
\ Powder Metallurgy Equipment Manual, New 1 "
"
.
|
.
17-1 Introduction When one considers the many millions of metallic parts that are fabricated and placed in service, it is not unusual that some will fail prematurely. Simply from a statistical viewpoint it is not reasonable, with The Macmillan Company, New York, 1947. \ present engineering practice, to expect no failures. However, even though "
r letal Powders, gy,
"
Van Nostrand Relnhold Com-
the number of failures of a particular component may be small, they are important because they may affect the manulacturer's reputation for re-
liability. In some cases, particularly when the failure results in personal injury or death, it will lead to expensive lawsuits. It is nol unusual for automotive manufacturers under prodding and publicity from consumer watchdogs to recall millions of cars to correct a design or heat-treating defect even though the actual number of failures was very small,
The purpose of this chapter is to briefly explain the basic causes for metal failure and to illustrate some of the failures by case histories. Most of the illustrations in this chapter were taken from two excellent books on metal failure-"How Components Fail" by Donald J. Wulpi (American Society
1966) and "Why Metals Fail" by R. D. Banjr and B. F. Peters (Gordon and Breach Science Publishers, 1970). ' ?!
for Metals
,
17'2 Procedure
In any failure analysis it is important to get as much information as possible from the failed part itself along with an investigation of the conditions at the time of failure. Some of the questions to-be asked are: How long was the part in service?
1
2
.
3
.
I
4
.
5
.
6
.
7
.
What was the nature of the stresses at the time of failure?' Was the part subjected to an overload? Was the part properly installed? Was it subjected to service abuse? Were there any changes in the environment? Was the part properly maintained?
A study of the fractured surface should answer the following questions: 1
Was the fracture ductile, brittle, or a combination of the tWo?
2
Did failure start at or below the surface?
.
.
634
1
INTRODUCTION TO PHYSICAL METALLURGY
3
.
4
.
failures by steel companies auto rr manufacturers nearly 50 percent c design the rest being distributed be
Did the failure start at one point, or did it originate at several points? Did the crack start recently or had it been growing for a long time?
,
,
i
It should b6 apparent that no suitable solution may be prescribed unless information regarding how the part performed and failed is available. Laboratory and field testing permit the evaluation of the effects of material, design, and fabrication variables on performance of the part under controlled conditions. Failure analysis, on the other hand, is concerned with parts returned from service and thus gives results of actual operating conditions. By combining the information from tests with the results from analysis, a clear picture of the causes of failure can be obtained. Rarely are failures assigned to a single cause. Usually they result from the com-
,
17 3
\
Modes of Fracture As was pointed ture often yields much information
identify the type of failure Ductile Sec. 3-7 but it will be useful to rev .
,
Ductile fractures are the result o
formation (slip or twinning) along c brittle fractures are due to tensile fractures
,
both types are present ir
basic mechanism often determines By the same token a knowledge of h whether a particular failure was c
bined effects of two or more factors that are deterimental to the life of the
part or structure.
,
When studying a failure, care must be used to avoid destroying important evidence. Detailed studies usually require documentation of the service history (time/temperature, loading, environment, etc.) along with chemical analysis, photomicrographs, and the like. Further study of the sequence of events leading up to the failure, plus knowledge of the location, markings, and condition of all adjacent parts at the time of failure, is necessary to confirm analysis. There always exists the possibility of unforeseen
nature.
Figure 17-1 shows two bolts pu ductile and brittle behavior The on .
it failed in a ductile manner by she mation. The bolt on the left was re in a brittle fashion with no apparer ,
by a single load are dull gray and deformed plastically. Small cayitie;
loading, unreported collision, or unanticipated vibration that may have contributed to premature failure. The procedure for investigating a failure covers four areas as follows:
together and eventually grow to torn
i
crack spreads with the aid of stres;
generally moving perpendicular to t
1 Initial observations. A detailed visual study of the actual component that failed" should be made as soon after the failure as possible. Record all details by many
a
"
shear lip" at the surface (see Fig
photographs for later review. Interpretation must be made of deformation markings, fracture appearance, deterioration, contaminants, and other factors. 2 Background data. Collect all available data concerned with specifications and drawings, component design, fabrication, repairs, maintenance, and service use.L!
3 Laboratory studies. Verify that the chemical composition of the material is within | t
i
specified limits. Check dimensions and properties of the component. Supplementary! |
tests may be made as needed-for example, hardness and determination of micro-;| f structure to check heat treatment, nondestructive tests to check for processing de- . f fects or existing cracks, composition of corrosion products, a free-bend test to check J /
ductility, etc. Very often, examination of a fracture surface with a low-power binoc- j \
ular microscope can reveal the type and cause of failure. J 4 Synthesis qf failure. Study all the facts and evidence, both positive and negative, ] and answers to the typical questions given earlier. This, combined with theoretical 1
'
analysis, should indicate a solution to the problem of failure.
:
i
Extensive sjudies of carburized and hardened gears for heavy-duty trucks, s machine tool s mining machines, diesel engines, etc. showed that 38 per- „ '
,
,,
, j.
,
.,
. ....
,..
Fi9-17,1
.- M
Two bolts intentionally pulled to failure in
lenslon to demonstrate brittle and ductile behavior. Th
cent of the failures resulted from surface problems (pitting, spallmg, crush- ,| brittle bolt left| was hard Rockwell c 57. the ductile b( ing, and scoring), 24 percent from bending fatigue, 15 percent from impact, J ( was soft, Rockwell c 15. (Courtesy of d. j wuipi, interi
i
and 23 percent from miscellaneous causes. From a detailed analysis of I | national Harvester Company.) m
i !
.
4
FAILURE ANALYSIS
Y
635
I
-
i
failures by steel companies auto manufacturers, and eiectrical equipment ' manufacturers, nearly 50 percent of all failures can be attributed to faulty
prmit the evaluation of the effects of ma-* || jriables on performance of the part under ;
ture often yields much information on the contributing flactors and helps to identify the type of failure. Ductile and brittle fractures were discussed in
inalysis, on the other hand is concerned .J and thus gives results of actual operating :
Sec. 3-7, but it will be useful to review the fracture modes.
i0ji, or did it originate at several points? yad it been growing for a long time?
-
,
design, the rest being distributed between production and service problems. {uitable solution may be prescribed unless J jart performed and failed is available. . % 17'3 Modes of Fracture As was pointed out earlier, proper analysis of the frac'
'
-
,
;
,
.
;.
iformation from tests with the results froni | I
Ductile fractures are the result of shear forces that produce plastic deformation (slip or twinning) along certain crystallograpfiic planes whereas ,
Clauses of failure can be obtained. Rarely | cause. Usually they result from the com-
brittle fractures are due to tensile forces that producejcleavage. In most fractures, both typfeS are present in varying degrees, identification of the
tors that are deterimental to the life of the :
basic mechanism often determines the type of load that initiated fracture.
nust be used to avoid destroying important i
By the same token, a knowledge of load application can help in determining whether a particular failure was ductile (shear) or brittle (cleavage) in nature.
ally require documentation of the service ng, environment, etc.) along with chemical ;
-
.
Figure 171 shows two bolts pulled to fracture in tension to illustrate ductile and brittle behavior. The one on the right was soft (Rockwell C15); it failed in a ductile manner by shear, resulting in extensive plastic deformation. The bolt on the left was relatively hard (Rockwell C 57) and failed
j the like. Further study of the sequence plus knowledge of the location, markings,
1
parts at the time of failure, is necessary A
in a brittle fashion, with no apparent plastic flow. Shear fractures caused by a single load are dull gray and fibrous, with edges which are usually deformed plastically. Small cayities are initially formed by slip. They join together and eventually grow to form a crack under continued loading. The crack spreads with the aid of stress concentration at the tip of the crack, generally moving perpendicular to the tensile force and eventually forming a shear lip" at the surface (see Fig. 3-14).
ivays exists the possibility of unforeseen
or unanticipated vibration that may have j ig a failure covers four areas as follows:
:
visual study of the actual component that failed" failure as possible. Record all details by many 1 iretation must be made of deformation markings,
"
contaminants, and other factors.
V
.
ivailable data concerned with specifications and ication, repairs, maintenance, and service use. '
he chemical composition of the material is within J and properties of the component. Supplementary j '
example, hardness and determination of micro- |
nondestructive tests to check for processing de-1
)n of corrosion products, a free-bend test to check j
tion of a fracture surface with a low-power binoc- I ,
~
1
)e and cause of failure. he
facts and evidence, both positive and negative ] ,
ns given earlier. This combined with theoretical n to the problem of failure. ,
:
v ;:
I
ed and hardened gears for heavy-duty trucks ies diesel engines etc. showed that 38 per|m surface problems (pitting spalling, crush-
1
,
,
Fi9-171 Two bolts intentionally puiied to failure in
,
r
,
rom bending fatigue 15 percentfrom impact, neous causes. From a detailed analysis of
was soft
,
«V1
,
T
Z<
Rockwell C 15. (Courtesy of D. J. Wulpi, Inter-
national Harvester company.) ,
4
:
1
i
1 .
i
1
636
INTRODUCTION TO PHYSICAL METALLURGY 1
Brittle (cleavage) fractures generally appear bright and crystalline. Each crystal tends to fracture on a single cleavage plane, and this plane varies only slightly from one crystal to the next in the aggregate. For this reason
it follows that" a cleavage fracture in ajDolycrystalline specimen will generally sparklejin the light when rotated in the hand. Surfaces of brittle
r
-
.
-
r : .
.
fractures sometimes have distinctive appearances. From the origin of fracture, a characteristic "chevron" or "herringbone" pattern is formed which points to the fracture origin (Fig. 17-2). Since (as pointed out in Chap. 3) slip and cleavage occur on a different set of crystallographic planes, the nature of individual fractures can often be determined by metallographic exahnination in the laboratory. Fractures are rarely either cleavage or shear. The variable stresses that usually exist in a structure, the changing of stress patterns during the progress of fracture, or the microscopic differences in orientation of grains produce fractures composed of both shear and cleavage areas. Considera-
]
tion of combinations of fracture modes can often give information regard-
ing the naturfe of the fracture.
Figure 17-3 shows three samples of the
same materialjas they reacted to notched-impact tests at different tempera-
tures. On theijeft, the fracture surface is mainly dull gray and fibrous; the edges are curved, indicating plastic deformation, so that the fracture mode is mostly shear. In the center the mode was mixed shear and cleavage, since the surface is both shiny and dull with some evidence of plastic deformation at the edges.
The fracture on the right is by cleavage alone.
.
mm
V.
.
Fig. 17-3
Combinations of fracture modes are shown I
fracture surfaces of three impact test specimens which L broken at different temperatures. On the left fracture I f mostly shear; in the center, combined shear and cleava and on the right, cleavage. (Courtesy of D J. Wulpi 1 International Harvester Company.) ,
.
,
The entire surface is bright and the dence of plastic deformation
.
17-4 Stress and Strength The solution tc stressing of parts depends on the di on the part and the strength requirec
the type of load and the geometry i stress or a complex system of multij elude internal residual stresses fron as stresses from external loads.
The basic stresses in a part undet 3-2
.
The most important are the nc
the plane of the cross section) and the cross section). Normal stresse shear stresses tend to produce plaf .
'
'
.
7
maximum shear stress occurs at a 4:
'
' .
.
' ,
1
.
When a part is under load yielding greater than the shear yield strengt when the shear strength is overcomf tures occur when the tensile (cohesi\ ,
6. 1
X:
-
r'
<
.
m i
m
iKM mm
i
Fig. 17-2
J
is also apparent in the upper right-hand corner. (Courtesy of D, J. Wulpi, Inlernalionai Horvofiior f ompnny.) '
.
;
r,
i ' .
i
.
. K
h
stress.
Consideration should be given to gating a particular mode of failure. Fc fracture at a gear tooth root the sig bending stress at that location. Cc ,
would not be significant in this case
"Chevron" pattern points to the origin of the
brittle fracture (arrow) in tijiis specimen. A fatigue fracture
1
>
h4x
.
\
"
.)
/
the gear tooth, the reverse would be 17-5 Types of Loading In many cases the ,
failure. There are essentially five typ
m FAILURE ANALYSIS
GY
herally appear bright and crystalline
637
i i
Each | f vingle cleavage plane, and this plane variesl | .
'
the next in the aggregate
.
For this reason
/re in a polycrystalline specimen will gen- ; ; rotated in the hand. Surfaces of brittle ; nctive appearances. From the origin of |
or "herringbone" pattern Is formed 1 gin (Fig. 17-2). Since (as pointed out in 1
ron
"
.
1
ST
.V
.
ur on a different set of crystallographic M / . ,. .', a Fig. 17-3 Combinations of fracture modes are shown by racuires can often be determined by metal- . fracture surfaces of three Iriipact test SpeCimens which were ;
,
;
v
,
S&S|oratOry.
;? broken at different temperatures. On the left, fracture is
avage or shear. The variable stresses that 1 ,TI0Stly shear; in ,he cen,er' combinecl sheiir and cleavage; and on the right, cleavage. (Courtesy of D. J. Wulpi, hanging of Stress patterns during the proginternational Harvester Company ) popic differences in orientation of grains '
...
,
.
.
ooth shear and cleavage areas. Considerai modes can often give information regard- i
WS* Figure 17-3 shows three samples of the i
notched-impact tests at different tempera- j
urface is mainly dull gray and fibrous; the
i
..
t
so that the fracture mode i he mode was mixed shear and cleavage | ind dull with some evidence of plastic de- |
?w - stic deformation
,
,
acture on the right is by cleavage alone.
.
|
The entire surface is bright and the edges are straight showing no evidence of plastic deformation. Stress and Stren9th The solution to failure problems resulting from overstressing of parts depends on the determination of two factors: the stress on the part and the strength required to support that stress. Depending on ,
"
*
the type of load and the geometry of the part, there may be simple axial stress or a complex system of multlaxial stresses. The total stress can include internal residual stresses from fabrication or heal: treatment as well as stresses from external loads.
The basic stresses in a part under external load were discussed in Sec. 3 2. The most important are the normal stresses (those perpendicular to the plane of the cross section) and shear stresses (those in the plane of
/
the cross section). Normal stresses tend to produce, separation, while shear stresses tend to produce plastic flow. It was pointed out that the maximum shear stress occurs at a 45
°
m 1
V
f :
f
.
.
..3
angle to the initiating tensile stress.
When a part is under load, yielding will occur when the shear stress is greater than the shear yield strength; ductile or shear fractures develop when the shear strength is overcome by the shear stress; and brittle fractures occur when the tensile (cohesive) strength is exceeded by the tensile stress.
mm
fti
i
Consideration should be given to the significant stresses when investigating a particular mode of failure. For example, if failure is due to afatigue
AM.
111 7>
5
the
fracture
courtesy
fracture at a gear tooth root, the significant stress would be the repeated bending stress at that location. Contact stress acting on the gear face would not be significant in this case. For a pitting or wear-type tailure of the gear tooth, the reverse would be true. 17-5 Types of Loading In many cases, the type of Load Is a contributing factor to failure. There are essentially five types of loads illustrated in Table 17-1 -
i
638
INTRODUCTION TO PHYSICAL METALLURGY
axial, bending, torsional, direct sh load is applied coincident with the uniform across the cross section
,
cables. Bending loads are produce the center line.
Across the cross si
at the outermost fibers to zero at tfv
of gear teeth. Torsional loading in\ in a plane normal to the center line (0
(D (D
to
i
XI
LU
V:::.:;:.:J
_
1
a
.
O)
0) 0) CO
<
E
o
LU
I£
ed to torsion are shafts and coil spi
si
to the rest
I S,
stress distribution is uniform acros
o
r
CO
o
CQ
c
mum at the surface to zero at the n
<0 (0
ra
.
(0
-
c
o
CO
o
>
o
spaced parallel planes and tend to similar to a cutting ad
CD
,
1
compressive loads perpendicular ti forces between the surfaces. The s 8?
z
g
I
i
-
CQ
11.2
"
force direction. Examples of conta
24
o
Sl.ii!
teeth.
oil x
o
o r a (D
Q W W
K5 Ira
All these types of loads induce no balanced by the material's cohesive
5 c
o
Si
overload fractures to occur when
$ £ o
5
UJ
; !
-5
EC I w
values. The failed main bearing of £ overhaul. The cage was broken, o balls were heavily scored, and both
-
i
fi
*5
c
o
f
c
0) c
5
5 3
0
f jl t f %
.
CD
!
D <
O
'
(0 c
o
ra
C
jn
c
0)
O
5
CQ
2
2»
ft!
o
o
? Fig 17-4 Both halves of inner ball race showing dame i ! on one side only due to misalignment. (From R. D. Bai .
,
.
.| | and B
F. Peters, "Why Metals Fail," Gordon and Breac I ; Science Publishers, New York, 1970.) .
-
i vie . . *
,
FAILURE ANALYSIS
639
axial, bending, torsional, direct shear, and contact. In axial loading the load is applied coincident with the center line of the part and the stress is uniform across the cross section, as in tensile test bars and supporting cables. Bending loads are produced by couples of forces coincident with -
,
I
the center line. Across the cross section, the stress varies from maximum at the outermost fibers to zero at the neutral axis, as in beams and the root
of gear teeth. Torsional loading involves the application of a force couple in a plane normal to the center line. The shear stress varies from a maxi-
mum at the surface to zero at the neutral axis. Examples of parts subjected to torsion are shafts and coil springs. Direct shear loads act on closely jQ
spaeed parallel planes and tend to move part of the material with respect
a)
to the rest, similar to a cutting action, as in rivets and bolts. The shear
<0
JZ
o
>
o
stress distribution is Uniform across the cross section. Contact loads are
compressive loads perpendicular to two surfaces, combined with sliding
i
forces between the surfaces. The stress distribution varies with depth and
force direction. Examples of contact loading are roller bearings and gear teeth.
u
IJOR
2
o
s
c
S § §o (0
SI
overload fractures to occur when the applied load reaches excessive values. The failed main bearing of an air compressor was found during an
51
overhaul. The cage weis broken, one ball was split in two, several other balls were heavily scored, and both inner and outer racesi showed signs of
£f o
.
All these types of loads induce normal and shear stresses which must be balanced by the material's cohesive and shear strengths. It is possible for
a>
11 ?5
(0
5 C .
.
(0
o
- .
.-
o
-
13 « c
o
1
-
O
Fig. 17-4 Both halves of inner ball race showing damage on one side only due to misalignment. (From R. D. Barer and B. F. Peters, "Why Metals Fail," Gordon and Breach Science Publishers, New York, 1970.) }
i 1 m
~
[
v
" .
r
640
INTRODUCTION TO PHYSICAL METALLURGY
being badly overheated. Figure 17-4 shows both halves of the inner race. The failed surface is to one side of the central track and extends about half
way around the race. Subsequent examination indicated that the bearing was subjected to misalignment in its casting. This misalignment caused overloading and overheating of the bearing along with an end thrust which
:
resulted in the failure on one side of the central track. Figure 17-5 shows
a failed phosphor-bronze spring with numerous cracks on the inside surface outlined by fluorescent penetrant. The maximum tensile stress for phosphor bronze is about 50,000 psi. Calculations indicated that the spring was subjected to a stress of 76,900 psi in service; therefore, the failure was the result of overload. Figure 17-6 shows a gear with two broken teeth. :
.
.
Tooth A failed in service, while tooth B was fractured in the laboratory in an overload test. The similarity in the appearance of both fracture surfaces indicates that a high overload probably caused tooth A to break.
.J y i .
.
Fig. 17 6
17-6 Fatigue Fractiires Fatigue failures are the most common types of fracture in machines and probably constitute about 90 percent of all fractures. Such fractures develop after a large number of load applications, generally at a stress level below the yield strength of the material. Fatigue testing was discussed briefly in Sec. 1 -34. Fatigue stresses develop in three principal ways, as shown in Fig. 17-7. The upper diagram Illustrates the stress pattern under reversed loading, typical of a rotating shaft under a bending load, where tension, compres-
In this gear tooth A broke in service ,
,
and tc
B was fractured by a single blow in the laboratory
.
Th
similarity in fracture surfaces indicates that tooth A w£ also broken by a sharply applied load (Courtesy of D. .
Wulpi International Harvester Company.) ,
typical of a punch or gear teeth Th dition under unidirectional loading .
varies from minimum to maximum
head bolts and connecting-rod bolt Since a fatigue fracture is progn of time, the fracture surface usually i shell markings. Figure 17-8 show
sion, or shear stresses of the same magnitude alternate. The middle dia-
gram shows the stress variation under unidrectional loading where the load varies from zero to a maximum either in tension, compression, or shear
i
"
threaded axle; an arrow indicates
originates at the surface of the pari the shear strength. In this case tht continuity on the surface and as si ,
nearly across the section before fin? Fatigue fractures initiate in shea *
i
work hardening, eventually formin develop into cracks (see Fig. 3-15).
i
growth depends upon the stress rr limit of the material notch sensitiv ,
V
structural flaws and inclusions. Fig has formed the stress is reduced to I ,
Fig. 17-5 Failed phosphor-bronze haich spring showing '
'
'
'
: : :] .
numerous cracks on the inside surface outlined by dye penetrant. (From R. D. Barer and B. F. Peters, Why Metals Fail," Gordon and Breach Science Publishers, New York, "
1970.)
;
m !
I
! 4
i t
the crack may not propagate any fu increase in strength due to strain hai load is large enough the crack will ai tensile stress. Variation in the cycli marks to develop on the fracture si ,
n .
K
1 f
7-4 shows both halves of the inner race
FAILURE ANALYSIS
64':
.
the central track and extends about half«
I
t examination indicated that the bearing ii its casting. This misalignment caused J e bearing along with an end thrust whicliil t
} of the central track. Figure 17-5 showsS i with numerous cracks on the inside siirstrant.
The maximum tensile stress for 4
1
isi. Calculations indicatedthat the spring v
O psi in service; therefore the failure was
-
.
.
,
.6 shows a gear with two broken teeth.
-
)oth B was fractured in the laboratory in | in the appearance of both fracture sur- -W .
jad probably caused tooth A to break. s are the most common types of fracinstitute about 90 percent of all fractures.
1
:
;
:
.
i
f
'j...ie number of load applications, generally
:
Fig. 17-6 In this gear, tooth A broke in service, and tooth B was fractured by a'slngle blow In the laboratory. The similarity in fracture surfaces indicates that tooth A was also broken by a sharply applied load. (Courtesy of D, J. Wulpi, International Harvester Company.)
ength of the material. Fatigue testing was
mpe
principal ways, as shown in Fig. 17-7. b stress pattern under reversed loading i bending load, where tension compres,
,
]
typical of a punch or gear teeth. The diagram on page 642 shows the condition under unidirectional loading with a preload. In this case the stress varies from minimum to maximum without reaching zero, as in cylinder-
rie magnitude alternate. The middle dia-
head bolts and connecting-rod bolts.
ider unidrectional loading where the load either in tension, compression or shear
Since a fatigue fracture is progressive, developing over a long period of time, the fracture surface usually shows characteristic beach" or "clamshell markings. Figure 17-8 shows a bending fatigue fracture of a large threaded axle; an arrow indicates the origin of fracture. Generally, failure originates at the surface of the part where the shear stresses first exceed the shear strength. In this case the fracture started at the indicated discontinuity on the surface and, as shown by the beach marks, proceeded nearly across the section before final separation. Fatigue fractures initiate in shear by a mechanism involving slip and work hardening, eventually forming microscopic discontinuities which develop into cracks (see Fig. 3-15). Once a crack is formed, its rate of growth depends upon the stress magnitude, stress gradient, endurance limit of the material, notch sensitivity and the presence or absence of structural flaws and inclusions. Figure 17-9 shows that if, after the crack has formed, the stress is reduced to below the value necessary to initiate it, the crack may not propagate any further. This is probably caused by an increase in strength due to strain hardening at the crack tip. If the applied load is large enough, the crack will advance perpendicular to the maximum tensile stress. Variation in the cyclic load causes small ridges or beach marks to develop on the fracture surface They indicate the position of
,
"
"
,
;
.
.
.owing
.
.
dye hy Metals
.
l York,
.
i
4
.
I
-
642 v.;
1
INTRODUCTION TO PHYSICAL METALLURGY
-
08 .
24,000 psi +
-
0.5
S
+
0 Time
o
01 .
17
,
500 psi
0 Time
a)
0 2 14 000 ,
psi
la] 0
12
3
1
Life, cycles x lO6 Fig. 17-9
Relationship of part life to crack length (F Machine Design, The Penton Publishing Co Clevelar .
.
-
-
\
.
-
v
.
-
,
.
,
November 13, 1969.) +
When fatigue originates in seven bending, the progressing cracks n ratchet marks" as shown in Fig 1
0
Time
10
Fig. 17-7 Tvhe basic fatigue stress conditions, (a) Reversed
"
1
.
17-7 Effect of Stress Raisers
stress, (b) unidirectional stress, (c) unidirectional stress with a preload.
stresses occur most often at fillets
\
root of the'advancing crack at a given time. As the section gradually weakens, the crack grows faster, and the clam-shell markings get further
In machir
larities that concentrate and incre stress raisers.
Is
apart, larger, and more distinct. Therefore, when these markings are pres- j . ent, they provide a means of locating the origin of fracture accurately.
:
i : V
.
v
8
1
1
..5 4
j
V,!
"
Fig. 17-8 The presence of "beach marks usually indicates that failure was caused by fatigue. Here fracture began at
a discontinuity (arrow). (Courtesy of D. J. Wulpi, International Harvester Company.) t
'
;
I :
\ ;
.
' . .
i
i
.
i
Fig. 17-10 "Ratchet marks" around edges of fatigue failures indicate that fracture began at several points
.
(Courtesy of D. J. Wulpi, International Harvester Comp;
f
5Y
-
FAILURE ANALYSIS
643
08 .
24,000 psi -
0,6
I 04 .
1
17
I
Time
500 psi
,
-
0 2 .
14,000 psi
(6)
005
J;
o
i
jm
2
"
3
4
Life, cycles x I06
:
M':ig.17.9 Relationship of part life td Crack length. (From n Machine Design, The Penton Publishing Co., Cleveland,
If November 13, 1969.)
;
-
,
.
,
.
When fatigue originates in several locations of a filleted shaft in rotating bending, the progressing cracks run into each other, usually resulting In ratchet marks as Shown in Fig. 17-10.
)
: :
"
"
Reversed
-?; stress with -
5
1
i
17-7 Effect of Stress Raisers In machine and structural members, the highest stresses occur most Often at fillets, holes, and similar geometrical irregularities that concentrate and increase surface stress. stress raisers.
a given time. As the section gradually| jr, and the clam-shell markings get further | Therefore, when these markings are presating the origin of fracture accurately
;
.
:1
A ft
4
i
1
-
1 v
ly indicates began at Inter-
Fig. 17-10 "Ratchet marks" around edges of fatigue failures indicate that fracture began at several points (Courtesy of D. J. Wulpi, International Harvester Company.) .
'
These are called
644
INTRODUCTION TO
'
PHYSICAL METALLURGY
The majority of stress raisers fall into one of the following broad groups: 4
Si V.' - V.
.
1
.
Those caused by changes in the geometry of a part, such as holes, keyways,
Or,g,n
threads, steps, or changes in diameter in shafts and bolt heads, etc. 2 Surface discontinuities, such as nicks, notches, machining marks, pitting, cor-
f
.
T
.
rosion, etc. 3
.
Defects Inherent in the material, such as nonmetallic inclusions, minute cracks,
0
voids, etc.
&
Primary stress raisers are usually of the first group, although those of
the second ahd third groups may play secondary, related roles. Even ordim4
i
nary tool marks act as notches which tend to concentrate stresses, particularly at the root of the notch. They are especially damaging when they '
occur at section discontinuities such as fillets.
Under a static load, the highly stressed metal yields plastically at a notch root or hole edge, thereby passing the high stresses to other sections until fracture occurs. However, under fatigue, or repeated loads, where the stress is below the elastic limit, yielding is more localized, and a crack may start before the stress pattern changes to relieve the stress concentration.
7
Figure 17-1T shows the effect of severe notches in steel specimens. The importance of stress raisers can be shown in a single example. If a small hole is drilled in a wide strip of elastic material and the strip is subjected to axial tension, the stress at the edge of the hole reaches a maximum of 1
-.
Direction of Rotatk
:
3 times the normal stress.
Fig. 1712 The offsetting effect of rotation on the zom final fracture reveals the direction that the shaft rotatec during operation. (Courtesy of D J. Wulpi Internation. Harvester Company.) .
In rotating machine parts, the final rupture area is not directly opposite the start of fracture but is slightly offset by the effect of rotation. This is
,
:
illustrated by Fig. 17-12 which show; Fracture originated due to stress co
i
,
120
#1
and the beach marks swing aroun rupture because of the clockwise ro
100
Sharp corners are always stress possible. The valve spindle on the
§ 80
sharp corner, whereas the one on tl e
corner and no sign of failure. Figure
60
20
at the sharp corner of a valve bonne viding a generous radius at the inne stud which failed by fatigue at the f stud, Fig. 17-15£), shows the charact«
0
indicates the spot where fatigue sta Fig. 17-15c indicates very rough ma
0
40
m//m/////////mfn Severely notched specimens
\
l
,
0
50 ;
100
150
200
250
on the thread face, and the radius at
Tensile strength, 1000 psi
Fig. 17-11
In general, hard materials show a
In specimens with severe notches, fatigue limits
level out at 25,000 to 45,000 psi and drop off slightly as tensile strengths rise above 180,000 psi.
I
1 i '
v../
i
t \
terials, and this property will affect cyclic loading. In a high notch-sens
in
-
.
'
FAILURE ANALYSIS
.
ly
645
1 into one of the following broad groups: I
geometry of a part, such as holes
,
keyways, ;;
Ongi n
I in shafts and bolt heads, etc.
[icks, notches, machining marks
,
s
pitting, cor-
ch as nonmetallic inclusions, minute cracks,
5 i
s
i
illy of the first group, although those of |
play secondary, related roles. Even ordi- | lich tend to concentrate stresses partlc-J ,
s
hey are especially damaging when they f
'
-
jch as fillets.
1
tressed metal yields plastically at a notch la
3
g the high stresses to other sections until .| r fatigue, or repeated loads, where the .
"
siding is more localized, and a crack may |
1
Final
vjnges to relieve the stress concentration. |
;:
Rupture
severe notches in steel specimens. The ; ,
.
c;:v3e shown in a single example. If a small .;
"
Direction of Rotation
lastic material and the strip is subjected i i edge of the hole reaches a maximum i | Jig. 7M
a
i \ -
i
17-12 The offsetting effect of rotation on the zone of final fracture reveals the direction that the shaft rotated during operation. (Courtesy of D. J. Wulpi, International
final rupture area is not directly opposite \ Harvester company ) y offset by the effect of rotation. This is
illustrated by Fig. 17-12, which shows the fracture surface of a broken shaft. Fracture originated due to stress concentration at a corner of the keyway, and the beach marks swing around counterclockwise toward the final .
rupture because of the clockwise rotation.
.
.
Sharp corners are always stress raisers and should be avoided when possible. The valve spindle on the left in Fig. 17-13 shows a crack at the sharp corner, whereas the one on the right has a generous radius in the corner and no sign of failure. Figure 17-14 shows a crack which originated at the sharp corner of a valve bonnet. Cracking could be avoided by providing a generous radius at the inner corner. Figure 17-15a shows a steel stud which failed by fatigue at the first thread. The fractured end of the .
[
J
I
stud. Fig. 17-15b, shows the characteristic fatigue markings, and the arrow indicates the spot where fatigue started. Examination of the thread root, Fig. 17-15c indicates very rough machining by the presence of torn metal on the thread face, and the radius at the thread root is very small. In general hard materials show a higher notch sensitivity than soft materials, and this property will affect the appearance of a fracture under cyclic loading. In a high notch-sensitive material, the crack tends to grow ,
ue limits
ly as
,
i
i
646
INTRODUCTION TO PHYSICAL METALLURGY
i
<7)
I
i ] '
Fig. 17-13 Valve spindle at left is cracked at the corner (arrow). The other spindle has a generous radius at this
location. (From R. D. Bafer and B. F. Peters, "Why Metals Fail," Gordon and Breach Science Publishers, New York, 1970 ) .
v
i
more rapidly; along the highly stressed surface than toward the center.
Therefore, the beach marks curve away from the origin of fracture, as il- i
lustrated in ''Fig. 17-16a. In a less notch-sensitive material, such as an- \
nealed steel, the crack moves more rapidly toward the center than along the | surface and will produce concave beach marks around the origin of frac- ] '
ture, as illustrated in Fig. 17-16fc).
i
Fig. 1745 (a) Steel stud which failed as a result of fat at the first thread. (6) Fractured end of the stud, show fatigue markings. Fatigue initiated at arrow, (c) Root i
5
.
thread of stud. Note the torn metal, from machining, a
t
1
arrows; 200x, (From R. D. Barer and B. F. Peters,
"
Wh
Metals Fail," Gordon and Breach Science Publishers, \
York, 1970.)
5
-
;
r
f
Fig. 17-14 Sectional valve bonnet showing crack which originated at the sharply machined corner. (From R. D. Barer and B. F. Peters, "Why Metals Fail," Gordon and Breach Science Publishers, New York, 1970.)
some cases, particularly when the k( is transmitted through the key. Fra keyway and progresses in shear p peeling fracture. Sometimes th( the shaft, forming a separated shell "
mm
I
Internal corners in longitudinal g act as stress raisers. Fatigue cracks mum stress. In spline shafts, multip together, producing a "starry" tract crack path from the inner corner o
"
St
i
FAILURE ANALYSIS
647
I -
4
a]
.
5
4 -
r
r
corner
> at this
m
m
hy Metals iw York,
1 stressed surface than toward the center
.
iyXirve away from the origin of fracture as il- if. less notch-sensitive material such as an- 1 ,
m
,
bre rapidly toward the center than along the i ave beach marks around the origin of frac- i \ ;
Si
\i>)
(c)
:
| : Fig. 17-15 (a) Steel stud which failed as a result of fatigue i
at the first thread, {b) Fractured end of the stud, showing
1; fatigue markings. Fatigue initiated at arrow, (c) Root of
M thread of stud. Note the torn metal, from machining, at "
arrows; 20Cx. (From R. D. Barer and B. F. Peters, Why Metals Fail,
"
;
Gordon and Breach Science Publishers, New
York, 1970.1
n
I
I
v-
Internal corners in longitudinal grooves such as splines and keyways, act as stress raisers Fatigue cracks that develop follow the paths of maxi,
<
.
mum stress. In spline shafts, multiple cracks form (Fig. 17-17a) and grow
together, producing a "starry" fracture appearance (l|ig. 17-18). Atypical crack path from the inner corner of a keyway is shown in Fig. 17-17b. In
some cases, particularly when the key is loosely fittedl nearly all the torque k which
.
'
n R. D.
i
is transmitted through the key. Fracture starts at thefbottom corner of the
keyway and progresses in shear parallel to the SLjrface
,
>n and
"
resulting in a
peeling" fracture. Sometimes the peeling action goes entirely around the shaft, forming a separated shell (Fig. 17-19). J I.
548
INTRODUCTION TO PHYSICAL METALLURGY
[Ruplure areo
m
i
(
Fatigue area
Origin a]
Fig. 17-16
The degree of notch sensitivity affects the
manner in which beach marks develop. In notch-sensitive
:
alloys, such as high-strength steel, these marks curve away from the source of failure (left). The reverse is true in
notch-insensitive material (right). (From D. J. Wulpi,
"
How
Components Fail," American Society for Metals, Metals Park, Ohio, 1966.)
Fig. 17-18 When spline shafts fail in fatigue (from rev
i
torsional loading) they generally develop "starry" frac surfaces. (Courtesy of D. J. Wulpi International Harve ,
,
17 8 ;: ;:.>
.
;
Effect of Strength Reducers In addition to stress raisers, certain metallurgical condifions may act to lower the strength of the metal and lead to fracture. Such conditions include overheating, grinding burns, poor heat
Company.)
J Microscopic examination of the rr
treating, and jipoor casting practice. c
showed the normal microstructure
I
steel (Fig. 17-20£)). A sample take
Figure 17-2ba shows a ruptured tube from a marine boiler. Visual ex-
martensite (Fig. 17-20c).
This stru
amination indicated that:
steel to 1,700 to 1 800oF and then w
1
a temporary cause of poor circulatic
,
The split wefe at the center line of the tube-facing the fire, and therefore at a
fer through the wall. This caused tl
high heat input,zone.
2
The thin edges of the split and the
"
"
stretcher
marks on the Inner surface at the
break are indicative of plastic flow of the metal.
3 Only a small amount of internal deposit was noted in the immediate area of the split.
i.
to 1,800oF in a few minutes.
At th
ductile, so that a gentle bulging of thin for the internal pressure. At n effectively quenched the overheated Figure 17-21a shows the inner £
cooling pipe with pits at a and a cra( structure some distance from the
Crack
5 (j
(»)
Crack
(*i
Fig. 17-17 Fatigue cracks;tend to follow paths of maximum stress concentration. Circular lines indicate stresses. In
splines and keyways, the presses concentrate at inner
Fig. 17-19
comers, (a) Spline, (b) ke|way. (From D. J. Wulpi,
members may peel around the shaft under the surface. (From D. J. Wulpi, "How Components Fail," American Society for Metals, Metals Park Ohio, 1966.)
"
How
Components Fail," Americjan Society for Metals, Metals Park, Ohio, 1966,)
t
Fatigue cracks in keyways of loosely fitting
,
E
1111
i
/
i
L
i
FAILURE ANALYSIS
649
i
m
i
J
I
he
' A
insitiv€ ve away in i "How ,
;tals
;
1
addition to stress raisers certain metal-
:
,
wer the strength of the metal and lead to le overheating grinding burns, poor heat
Fig. 17-18 When.spline shafts fail in fatigue (from reversed torsional loading), they generally develop "starry" fracture surfaces. (Courtesy of D. J. Wulpi, international Harvester Company,)
,
be
Microscopic examination of the material some distan;ce from the rupture
.
I
showed the normal microstructure of ferrite and pear|,te typical of a 1025
id tube from a marine boiler.
steel (Fig. 17-206). A sample taken from the lip of trie fracture showed martensite (Fig. 17-20c). This structure could only gfise by heating the
Visual ex-
,
steel to 1,700 to 1,800 F and then water quenching. Aibparently there was '
a temporary cause of poor circulation in the tube, thus feducing heat transfer through the'wall. This caused the metal of the tub6 wall to reach 1,700
f the tube-facing the fire, and therefore at a
.
"
i
to 1,800oF in a few minutes. At this temperature th metal is weak but
stretcher" marks on the inner surface at the
I the metal.
ductile, so that a gentle bulging of the wall took placej until it became too
i
leposlt was noted In the immediate area of the
thin for the internal pressure. At rupture, the flow of jwater from the tube
.
-
,.
..
>
.
effectively quenched the overheated steel to give the martensitic structure.
Figure 17-21a shows the inner surface of a failed copper-nickel-iron cooling pipe with pits at a and a crack at b. The norma| fine-grained micro-
structure some distance from the failure is shown ijn Fig. 17-21 b. The
Crock
Crock
maximum es.
In
iner "
Fig. 17-19
How
eta Is
i
Fatigue cracks in keyways of loosely fitting
members may peel around the shaft under the surface. (From D. J. Wulpi, "How Components Fail," American
Society for Metals, Metals Park, Ohio, 1966.)
i
m i:
10
INTRODUCTION TO PHYSICAL METALLURGY
extended the hardened zone into
shown as the white area through tf
&
the crack which started near the
area.
4
Figure 17-22c shows a pn
m r
3
1
m
I t'L-
!
i
5
i
.
9
Fig. 17-20 (a) A burst boilej- tube. Note the thin edges and
{a)
the "stretcher" marks on th inside surface, (b) Normal microstructure of a 0.25 percent carbon steel tube showing
a mixture of pearlite (dark) and ferrite (white), 100x. (c) Microstructure at the lip of the burst specimens. This shows
only martensite and would only result from a drastic quench of the tube; 200x. (From RljD. Barer and B. F. Peters, Why Metals Fail," Gordon and Breach Science Publishers, '
-i
New York, 1970.) '
X
microstructure at a section near the pitted area is shown in Fig. 17-21c.
Notice the relatively large grain size and the presence of cracks. The large
7P-
grain size wasj due to overheating in that area, probably when the pipe was bent to shape. The pits were confined to this zone of overheated and enlarged grains, since the corrosion resistance of this alloy is considerably reduced if it is heated to temperatures much above 1000oF Hardened steel can be severely damaged by improper grinding. Exces-
r
' ,
5V-
.
sive heat causes damaging residual stresses, but more significantly surface areas may transform to hard, brittle martensite which may result in a pattern of fine hairlike cracks easily revealed by magnetic particle inspection (see Fig. 1 -SSa). Faulty heat/treatment may often be the cause of failure. The possibility
i
of cracking during heat treatment of shallow-hardened and throughhardened steels was discussed in Sec. 8-31. Figure 17-22a shows a bandsaw blade which cracked after being in use 30 min. Microscopic examina-
tion showed fiat in hardening the teeth of the saw the manufacturer also I f ,
1
i
3
?
1
v
Si
Fig. 17-21 (a) Inner surface of a copper-nickel-iron C( pipe showing pits at a and a crack at b. (b) Normal fi grained microstructure at some distance from the heal
area, 150x. (c) Section cut through the pipe near the
area. The large grain size confirms exposure to high t perature. Black lines are cracks; 150x. (From R. D. B '
and B. F. Peters "Why Metals Fail Gordon and Breai Science Publishers New York, 1970.) "
,
,
,
FAILURE ANALYSIS
iY
651
I
extended the hardened zone into the blade.
The extent of hardening is
shown as the white area through the tooth and below in Fig. 17-22£). Notice the crack which started near the root of a tooth in the brittle, hardened area. Figure 17-22c shows a properly heat-treated, blade from another
mm.
1
1
m
sr.,,:
.
m
LA
v
v
w
i
I I idges and
((7)
-
prmal showing (c)
X
3
.
This shows
tic quench fers,
ublishers,
>
.X:£ir the pitted area is shown in Fig. 17-21c. size and the presence of cracks. The large g in that area, probably when the pipe was
5 v,
jnfined to this zone of overheated and en-
on resistance of this alloy Is considerably natures much above 1000oF.
ly damaged by improper grinding. Excesidual stresses, but more significantly sur:;; rd, brittle martensite which may result in a asily revealed by magnetic particle inspec-
V:
.
.
,
ten be the cause of failure. The possibility ment of shallow-hardened and throughin Sec. 8-31. Figure 17-22a shows a bandying in use 30 min. Microscopic examina-
:
he teeth of the saw, the manufacturer also
mm
Fig. 17-21 (a) Inner surface of a copper-nlckel-iron cooling pipe showing pits at a and a crack at b. (b) Normal finegrained microstructure at some distance from the heated area, 150x. (c) Section cut through the pipe near the pitted area. The large grain size confirms exposure to high temperature. Black lines are cracks; 150x. (From R. D. Barer and B. F. Peters, Why "
'
Metals Fail," Gordon and Breach
Science Publishers, New York, 1970.)
=1= f
1 i
i
I*
1
! INTRODUCTION TO PHYSICAL METALLURGY
.
Faulty foundry practice may oft «-.
m
cracks at the bolthole of a cast U croscopic examination Fig. 17-2,
4
,
coarse shrinkage (black areas)
,
whi
A double-ended piston from a cc
1
a crack in the wall of one of th
3
showed the presence of a very th in
confirmed by sectioning (Fig
] i
I i
.
17
which accounted for the unusually 17-9
Effect of Residual Stresses Resic part independent of any external fo tion will result in residual stresses i table presents the tendencies to b(
la)
I
Mm;% :'' =ig
.
la)
:T:;
17-22 (a) Band-saw blide cracked after 30 min of
jse. (b) CracKed blade shoeing that teeth were hardened Afell past the teeth roots. Notice the vertical crack in the
Hardened zone; 12x. (c) Pr|perly hardened blade-only :
he teeth have been hardened; 12x. (From R. D. Barer and
3
.
F. Peters, "Why Metals Fail," Gordon and Breach Science
Publishers
,
New York, 197o| f
supplier in whfch the root of the thread is not hardened, thereby giving the blade greater toughness and flexibility.
Figure 17-2 a shows a fractured carburized steel shell support part. It
was made of a low-alloy steel, carburized to a depth of 0.030 in. and hard-
ened to Rockyfell C 52. Microscopic examination of the case. Fig. 17-23/3,
shows a whiti carbide network around tempered martensite. The hard, brittle carbide? network is a frequent cause of failure in carburized parts. It could have peen avoided by use of a proper diffusion cycle during car-
400x. (From R. D. Barer and B. F
burizing to reduce the surface carbon content, or by proper heat treatment after carburizi ng to break up the carbide network (see Sec. 8-26).
Fail," Gordon and Breach Science Publishers New Yo 1970.)
'
-
1
.
Fig. 17-23 (a) Fractured carburized steei shell support 2x, (b) Microstructure of the carburized area showing white, brittle carbide network around tempered marten
i
m i
.
Peters "Why ,
,
Metal;
i;
f FAILURE ANALYSIS
653
Faulty foundry practice may often cause failure. Figure 17-24a shows cracks at the bolthole of a cast, leaded gun- metal waveguide flange. Microscopic examination, Fig. 17-246, revealed the presence of excessive coarse shrinkage (black areas), which is due to inadequate foundry practice. A double-ended piston from a condensate pump was leaking water from a crack in the wall of one of the pistons. Radiographic investigation
i
-
showed the presence of a very thin wall at the crack location. This was confirmed by sectioning (Fig. 17-25). During casting the core shifted, which accounted for the unusually thin wall.
2?« St
17 9
EJfect of Residual Stresses
Residual slressos are sirossos thai exist in a
part independent of any external force. Neatly every manufacturing operation will result in residual stresses in varying degrees (see Table 17-2). This table presents the tendencies to be expected for surface residual stresses
a
»
;)in of
;
l¥S'ardened i s In the ' -
.
-only 3arer and
ch Science
thread is not hardened, thereby giving the Jxibility. ired carburized steel shell support part. carburized to a depth of 0.030 in. and hard;
< around tempered martensite.
The hard,
quent cause of failure in carburized parts. use of a proper diffusion cycle during car;arbon content, or by proper heat treatment le carbide network (see Sec. 8-26).
m
mx
copic examination of the case. Fig. 17-236,
i
Fig. 17-23 (a) Fractured carburized steel shell support part, 2x. (b) Microstructure of the carburized area showing white, brittle carbide network around tempered martensite;
400x. (From R. D. Barer and B. F. Peters, "Why Metals Fail," Gordon and Breach Science Publishers, New York,
1970.)
354 ,NTRODUCT,ON TO PHYS,CAL METALLURG
j
i
i
i
i
n
si
Wm Ik
Fig, 17-25 Sectioned piston rod showing off-center
and crack (arrow) outlined by magnetic
m
mm
(From R. D. Barer and B. F. Peters
,
particle inspe Metals Fail
"Why
Gordon and Breach Science Publishers
i
,
>
[a]
Fig. 17-24 (a) Cracks at bplthole of a waveguide flange. (t)) Mlcrostructure near cracks showing gross shrinkage
,
New York
,
li
failures. The behavior of metals ; cussed in Chapter 13 Heating a i .
lower its yield strength tensile stre ing increase in ductility Failure re be related to excessive creep stre tunately components serving at t
:
,
voids (black areas), 23x. (from R, D. Barer and B. F.
.
Peters, "Why Metals Fail,"j-Gordon and Breach Science Publishers, New York, 1976.)
,
,
TABLE 17 2
only; The origjn of residual stresses due to heat treatment was discussed in Sec. 8-31. In general, residual stresses are beneficial when they are oppo-
site to the applied load.
Since cracks are propagated only by tensile
stresses, surface residual compressive stress would be most desirable.
Heat-treating processes that usually produce compressive residual stress are the shallow-hardening ones such as nitriding, flame hardening, induction hardening, and usually carburizing. Welding usually produces residual tensile stresses due to the contraction of the weld metal during cooling
A
Residual Stresses Cau
TENSILE
COMPRESSI\
STRESSES
STRESSES
Welding Grinding Straightening
Nitriding Shot peening Flame and int
hardening Heat and que:
i
Single-phase
from the weld jtemperature. The effect o| residual stresses varies with material hardness and with the presence of stress raisers. In general, soft materials with no stress raisers experience almost complete fading of residual stress while operating under
i
almost all of their residual stress (Fig. 17-26).
J
.
reversing loadlp, while notched parts made from very hard materials retain
17-10 Other Variables
Aside from stress raisers strength reducers, and residual ,
stresses, other variables may have to be considered when investigating
»'
!
i i
.From
"Machine Design," The Penton Publisi
FAILURE ANALYSIS
655
I
I
I
li
1 3
1
la
m i
Is
it
f
Si 5
4
s
Sit
mm
s>'''KJ- '*&
.
.
I 5
t;.
1
.
,1
3 T*
V
'
.
«
»yj;! ! M Ffg. 17-26 Sectioned piston rod showing off-fcenter core 1 '..*'f$ and crack (arrow) outlined by magnetic particle inspection. jS
. .
i
mk (From R. D. Barer and B. F. Peters, "Why Metals Fail," 3 Gordon and Breach Science Publishers, New York, 1970.) failures.
L
le
.
i
!
i
s
5 due to heat treatment was discussed in
.
.
sses are beneficial when they are oppo- -i cracks are propagated only by tensile ssive stress would be most desirable.
ly produce compressive residual stress ch as nitriding, flame hardening, induczing. Welding usually produces residual iction of the weld metal during cooling . .
<
The behavior of metals at low and high temperatures was discussed in Chapter 13. Heating a metal above room temperature tends to lower its yield strength, tensile strength, and hardness with a corresponding increase in ductility. Failure resulting from elevated temperature may be related to excessive creep, stress rupture, or thermal fatigue. Unfortunately, components serving at high temperatures often .deteriorate by TABLE 17-2
Residual Stresses Caused by Manufacturing Operations*
TENSILE
COMPRESSIVE
STRESSES
STRESSES
Welding Grinding Straightening
Nitriding Shot peening
EITHER
Flame and induction
hardening Heat and quenching Single-phase materials
Carburizing Rolling Casting Abrasive metal cutting (tensile stresses most common) Nonabrasive metal
i
cutting Heat and quenching
aries with material hardness and with the
r jeral, soft materials with no stress raisers g of residual stress while operating under Irts made from very hard materials retain
materials that
undergo phase transformation
'
ig. 17-26). ss raisers, strength reducers and residual -
(tensile stresses most common)
,
ive to be considered when investigating
'
From "Machine Design," The Penton Publishing Co. Cleveland, Oct. 16, 1969. ,
v
PHi;
5 INTRODUCTION TO PHYSICAL METALLURGY
Soil materials I Unnotched
Imporlonce of residual s|fesi
SS:
Hard materials
n Notched
Unimportunl
Stress condition
I Notched
Unnotched
I Wodera.e
Cose
g
.
too %
Low
High
overstress
overstress
Very
One-way bending lood
I;
Percent fading of residull stress:
No stress concentration
50%
0%
17-26 Relationship of material hardness and residual
ess fading. (From MachlniDesign, The Penton Pubhing Co., Cleveland, Oct. 1 , 1969.)
Two - way bending load
m
|
some form of hqt corrosion or instability. Performance depends more upon Reversed bending
resistance to this type of attack than upon the material's basic properties.
and rotation lood
Generally, decreasing the temperature of a metal raises the yield strength tensile strength!| and hardness at the same time reducing ductility. While at ,
,
ordinary temperatures a material may show a ductile
,
Fig. 17-27 Fracture appearances of bending-fatigue f ures. Final fracture zones are shown as crosshatched
shear fracture, below
the "transition |emperature" the basic mode of fracture changes to brittle,
(From Machine Design The Penton Publishing Co CI ,
low-energy cleavage.
land Nov. 27 ,
,
1969.)
In some cases, the rate of loading may determine whether a part will fail and the type of failure. Under extremely low rates of loading, ductile metals
show a large drjbp in strength, but stronger steels show little change. Under
stress which occurs on one side ol
rapid loading rates, the apparent strength is somewhat higher. When the
rate of loading'approaches what may be considered impact loads, brittle fracture may be induced in a normally ductile material because of the lack
where the maximum applied i
i
.
fillets and tool marks tend to
of time for flow to occur.
The composjtion and microstructure of a material will also influence the m
.
Reversed bending fatigue without
Tensile surface
/
i
ductile shear, whereas below it, brittle cleavage fractures predominate. Transition temperatures appear to be lowered by the addition of nickel and molybdenum and raised by carbon, manganese, and chromium.
17-11 Bending Fractures Bending is one of the common causes of fracture in machine and structural parts. Failure may be from a single application of a load greater than the overall strength of a part or can be due to a reversing load that results in a bending fatigue fracture.
failures are sfpwn in Fig. 17-27, with final fracture zones as cross-hatched areas. Ordinaiily, bending fatigue cracks are perpendicular to the tensile
Tensile surface Crack*-
0
I
Compressive surface
Fig. 17-28
Bending fractures usually develop on surfar and normal to the stress direction Sharp fillets concer .
bending stresses
,
causing cracks to develop more rapic
Arrows indicate bending direction (From D J. Wulpi How Components Fail American Society for Metals E Metals Park, Ohio, 1966.) .
.
"
,
"
,
m
I
i
Compressive surface
I
In many cases| the pattern on the fracture surface indicates the forces that
caused bendi|g fracture. Typical fracture appearances of bending-fatigue
Crack
V
The surfacelappearance of fatigue fractures wgs described in Sec. 17-6. V:V.-:i
initia
est there.
type of fracture. Generally, in heat-treated steel, the best combination of mechanical properties is obtained with a tempered-martensite structure (see Fig. 8-20). As with the rate of loading, the effect of composition and microstructure is determined primarily by their influence on the transition
temperature. Above the transition temperature, the fractures are usually
stress
cracks are oriented in cylindrical a one-way bending load As discuss(
,
-
FAILURE ANALYSIS
Hard materials
Stress
>
No stress concentration
condition
I ;
Unnotched
hed
I
Case
Notched
Low
High
overstress
overstress
Mild stress concentration Low overstress
High overstress
657
High -stress concentration Low overstress
High overstress
1 Moderate
Very
I
I
50%
t
One-way bendinq load
0%
Two-way bending load
I
ty. Performance depends more upon
ft
Reversed bending
|jpon the material's basic properties.
and rotation load
ire of a metal raises the yield strength, same time reducing ductility. While at Fig. 17-27
v v ;show a ductile, shear fracture, below
ures.
V c mode of fracture changes to brittle,
Fracture appearances of bending-fatigue fail-
Final fracture zones are shown as crosshatched areas
.
(From Machine Design, The Penton Publishing Co., Cleveland, Nov. 27 1969.) ,
S ?may .
determine whether a part will fail
kly low rates of loading, ductile metals
onger steels show little change. Under
stress which occurs on one side of the bend and originate at the surface
[y be considered impact loads, brittle
where the maximum applied stress is located Figure 17-28 shows how cracks are oriented in cylindrical and filleted shafts overloaded due to a one-way bending load. As discussed earlier stress raisers such as sharp
kngth is somewhat higher. When the
.
'
ly ductile material because of the lack
J 3
,
fillets and tool marks tend to initiate cracks because the stress is highest there.
ire of a material will also influence the
Reversed bending fatigue without rotation will usually cause cracks on
treated steel, the best combination of
jwith a tempered-martensite structure loading, the effect of composition and
,
jrily by their influence on the transition
Tensile surface
I temperature, the fractures are usually
Crack
i
rittle cleavage fractures predominate
ie lowered by the addition of nickel and manganese, and chromium.
Compressive surface
e of the common causes of fracture in -
Crock*-
iviire may be from a single application of a '
.
th of a part or can be due to a reversing
.
:
ue fracture.
3) Compressive surfoce
plie fractures was described in Sec. 17-6.
Fig. 17-28 Bending fractures usually develop on surfaces and normal to the stress direction. Sharp fillets concentrate bending stresses causing cracks to develop more rapidly. Arrows indicate bending direction. (From D J. Wulpl
racture surface indicates the forces that
fracture appearances of bending-fatigue
,
iith final fracture zones as cross-hatched
.
"
,
How Components Fail American Society for Metals Metals Park Ohio, 1966.) "
,
i cracks are perpendicular to the tensile
,
i
I
i
Tensile surface
J
-
,
1
658
INTRODUCTION TO PHYSICAL METALLURGY
B
A
1
,
i
I 1:
i 1
n
1
1
i
i if-,
4
A
i
\ i
j
1 i
i
Fig- 17-30 Some rotating bending-fatigue failures be beneath surfaces In this induction-hardened axle sh; fracture started at A and moved into the cross sectior
Fig. 17-29 This 1050 shaft! 1.94 in. in diameter, broke in reversed bending fatigue. A sharp fillet concentrated the bending stresses, causing a crack to develop on opposite
.
meeting another subsurface crack that started at B n
sides with final fracture in Jhe middle. (Courtesy of D. J. Wulpi, International Harvester Company.)
ing in final fracture (Courtesy of D. J Wulpi Internal .
.
,
Harvester Company ) .
tion before meeting another small £ point to both origins indicating bri 17-12 Torsional Failures Torsional failui
opposite sicle| of the shaft since each side undergoes alternate tensile and
,
compressive ptresses. Figure 17-29 shows a 1050 shaft which broke in
ing crankshafts torsion bars and
reversed bending fatigue. A sharp fillet concentrated the bending stresses, causing failure to start at opposite sides of the shaft with final fracture
,
,
fatigue fracture is quite different
in the middle;
Crack an longitudinal, [ \ Cra shear plane
Under reversed bending with a rotational load, the final fracture area tends to be offset from the initial crack due to the effect of rotation (see
Fig. 17-12). j While most bending fatigue cracks originate at the surface, it is possible under certain;conditions for the crack to originate below the surface. This condition may arise due to the presence of a microcrack or other metallurgical discontinuity, usually arising from fabrication processes that cause the strength iin its immediate vicinity to be considerably lower than the surface strength. Figure 17-30 shows an induction-hardened axle shaft of
1
i
I?
a
T
s
i Fig. 17-31
shear plane
Crack perpendicular to principal
j
/ /
45°
tensile stress
Torsional fatigue can develop parallel to th
principal shear stresses (top) or perpendicular to the ,
principal tensile stresses (bottom) (From D. J. Wulpi
1041 steel which failed in rotating bending fatigue. Fracture A started
below the hardened zone and moved nearly half-way across the cross sec-
an
Crack an /erse transverse
y
.
"
How Components Fail
"
i Metals Park Ohio, 1966.) i :
,
,
'
American Society for Metals
iY
FAILURE ANALYSIS
j
659
n
A
i
mm
4
tm
*»5
:
m
\
.
w
:
i
r
s
4 '
4
A
§
' :
.
m
Fig. 17-30 broke in '
ated the
i
opposite of D. J.
e each side undergoes alternate tensile and 1:17-29 :
shows a 1050 shaft which broke in
meeting another subsurface cfack that started at B, resulting in final fracture. (Courtesy of D. J. Wulpi, International Harvester Company.)
tion before meeting another small subsurface fracture (6). Chevron marks point to both origins, indicating brittle, sudden final failure. 17-12 Torsional Failures
Torsional failures are most common in shafts, includ-
ing crankshafts, torsion bars, and axles. The appearance of a torsionlatigue fracture is quite different from that caused by bending fatigue.
arp fillet concentrated the bending stresses, ijosite sides of the shaft with final fracture ;h
Some rotating bending-fatigue failures begin
beneath surfaces, in this induction-hardened axle shaft, fracture started at A and moved into the cross section,
Crack on longitudinal hnal shear plane
a rotational load, the final fracture area
tial crack due to the effect of rotation (see
-
Crock on i ransverse
shear plone
cracks originate at the surface, it is possible
( 11 Cifick pcrpendiculnrT1
Se crack to originate below the surface. This .
y I
presence of a microcrack or other metallur-
££sing from fabrication processes that cause w vicinity to be considerably lower than the
lotensile principal stress
S
1
A
'
P shows an induction-hardened axle shaft of
itating bending fatigue. Fracture A started moved nearly half-way across the cross sec-
Fig. 17-31 Torsional fatigue can develop parallel to the principal shear stresses (top), or perpendicular to the
principal tensile stresses (bottom). (From D. J. Wulpi, "
How Components Fail," American Society tor Metals
,
Metals Park, Ohio, 1966.)
: ;V -
-
-
7
1 i
60
INTRODUCTION TO PHYSICAL METALLURGY
Torsional-fatigue failures occur along the planes of maximum shear or
along the plane of maximum tension. Maximum shear stress occurs along the axis of the shaft and at right angles to it, as in Fig. 17-31 (top), while the maximum tensile stress acts at an angle of 45 to the two shear stresses, as °
in Fig. 17-31 (bottom). Figure 17-32 shows schematically the basic types of torsional fractures. Torsional cracks may follow the transverse or longitudinal shear plaies, the diagonal planes of maximum tensile stress, or a
[a)
combination of these.
4
In a shaft subjected to torsion, the maximum shear stress is equal to the maximum tensile stress.
4
In a part without stress raisers, which fracture
n
will occur will bepend upon the relative values of the shear strength and
tensile strengtt|. The strength values are a function of the material and its condition. [In steel, the shear strength is approximately one-half the
tensile strengtfi. Therefore, the shear stress will reach the shear strength of the steel lohg before the tensile stress will reach the tensile strength, and a shear-type failure will result. Transverse cracks are more prevalent
Fig. 17-33
(a) Transverse shear failure in a 1045 steel b
tested in torsion
,
(b) Tensile-type failure along a spiral
angle in a gray cast-iron bar tested in torsion.
than longitudinal cracks, because grinding or machining marks are oriented in a transverse direction. Figure 17 33a shows a transverse shear failure in a 1045 steel ba tested in torsion. In brittle materials such as gray cast iron, however, the tensile strength is less than the shear strength. -
Therefore, the tensile stress will reach the tensile strength of the cast iron before the shear stress will reach the shear strength, and a tensile-type
failure will result. Figure M-33b sh by a combination of tension and lone
i
In torsional fatigue stress raisers bending fatigue. Oil holes fillets, ar trate the stresses and produce a tensi ,
i
,
Type of foilure
Voriotionsol bosic potterns .iBosic pattern
ib) Saw tooth due to stress
Stor pattern tensile
I
concentration at fillet
.
.
\\\/
45°
,
times the shear stress. Therefore th strength before the shear stress will n
5°
,
Small step shear 2
,
raise the tensile stress to more than 3
stress remains essentially the same twice the shear strength the applie
J
I
Ironsverse
verse hole the stress concentration f.
Large step
ran O CD » CD
type fracture along a 45° spiral angle
I 17-13 Summary Other types of failure in i
covered in Chap. 14 and surface d Chap. 15. ,
To determine the true cause of a f
D
Lanqitudinol shear 3
m
consideration to the interplay of des .
environment, and service loads The .
Fig. 17-32 Basic torsional fractures. (From Machine Design, The Penton Publishing Co., Cleveland, Dec.
of the categories outlined in the folio involve redesign change of material trol, protection against environment c
11,1969.)
restrictions on service loads or servic
i
I
,
,
I FAILURE ANALYSIS
661
along the planes of maximum shear or :
; ision. Maximum shear stress occurs along '
angles to it as in Fig. 17-31 (top), while the
-
- -
,
in angle of 45° to the two shear stresses, as .32
'
shows schematically the basic types of j
racks may follow the transverse or longitual planes of maximum tensile stress, or a
[o)
h the maximum shear stress is equal to the
'
i
,
.
s
part without stress raisers, which fracture 3 relative values of the shear strength and
. ,
.
.
values are a function of the material and
'
'
sar strength Is approximately one-half the %
3 shear stress will reach the shear strength
}
nsile stress will reach the tensile strength, suit. Transverse cracks are more prevalent
i Fig. 17-33 (a) Transverse shear failure in a 1045 steel bar tested in torsion. (b) Tensile-type failure along a spiral angle in a gray cast-iron bar tested in torsion
a/Viuse grinding or machining marks are oriFigure 17-333 shows a transverse shear Jj in torsion. In brittle materials such as gray .
.
i
strength is less than the shear strength.
II reach the tensile strength of the cast iron
ach the shear strength, and a tensile type
failure will result. Figure 17-336 shows the failure in a gray cast-iron bar by a combination of tension and longitudinal shear when tested in torsion. In torsional fatigue stress raisers are nearly as serious as they are in bending fatigue. Oil holes fillets, and grooves in a shaft tend to concentrate the stresses and produce a tensile-type fracture. In the case of a transverse hole, the stress concentration factor may be above 3. The effect is to
\
-
,
,
Vorialions ol basic patterns U) Saw tooth due to stress concentration at fillet Star pattern
raise the tensile stress to more than 3 times its normal value while the shear ,
1
! 1:1°
Smoll slep
stress remains essentially the same Although the tensile strength is still twice the shear strength the applied tensile stress is now more than 3 .
,
times the shear stress.
Therefore the tensile stress will reach the tensile ,
strength before the shear stress will reach the shear strength and a tensiletype fracture along a 45" spiral angle will result (see Fig 17-34)-
Larye slep
,
.
CD
17-13 Summary Other types of failure include surface damage dup to wear ,
To determine the true cause of a failure the investigator must give full consideration to the interplay of design fabrication, material properties, environment, and service loads. The cause will usually be classified in one of the categories outlined in the following list Appropriate solutions may involve redesign change of material or processing (or both), quality control, protection against environment changes in maintenance schedules, or ,
D 45-
vj
,
90'
.
.
,
(lachine ,
,
covered in Chap. 14 and surface damage due to corrosion, covered in Chap. 15.
Dec.
,
restrictions on service loads or service life
i
'
m m I
,
.
662
INTRODUCTION TO PHYSICAL METALLURGY
5
4
Defects due to welding (porosity
,
ur
tration, undeibead cracking heat affec ,
1
.
i
6 Abnormalities due to heat treating ( growth, excessive retained austenlte, c
7
Flaws due to case hardening (int
cycles).
a
-
51
8 Defects due to surface treatments (( hydrogen embrittlement). 9 Careless assembly (mismatch of ma stress, gouges or Injury to parts, and th 10 Parting line failures in forging due
II. Failures due to Faulty Design Consic 1 Ductile failure (excess deformation 2 Brittle fracture (from flaw or stress r 3 Fatigue failure (load cycling, strain rolling contact fatigue fretting fatigue). 4 High-temperature failure (creep, oxi( 5 Static delayed fractures (hydrogen er mentally stimulated slow growth of flaw 6 Excessively severe stress raisers inhi 7 Inadequate stress analysis, or impo: complex part. 8 Mistake in designing on basis of sta cant material properties that measure th ,
,
illP I 1
i
failure mode.
III. Failure Due to Deterioration During .
Fig. 17-34 Holes in shafts sometimes
4
concentrate stresses, resulting in torsional-
fatigue failures. The spllned shaft (top) is
carburized alloy steel at Rockwell C 60, and
the crankshaft (bottom) is of 1045 steel, induction-hardened and tempered to Rockwell C 55. (Courtesy of D. J. Wulpi, International Harvester Company.)
1
1 Overload or unforeseen loading con< 2 Wear (erosion, galling, seizing, goug 3 Corrosion (including chemical attack, ification, graphitization of cast iron, con
4
Inadequate or misdirected maintena
punching holes, cold straightening, and
5
Disintegration due to chemical attac
elevated temperatures. 6 Radiation damage (sometimes must
I
CLASSIFICATION OF FAILURE CAUSES I
.
1
and dosage. 7 Accidental conditions (abnormal opei
Failures Due to Faulty Processing
Flaws due to faulty composition (inclusions, embrittling
terial).
vibrations, impact or unforeseen collisio
impurities, wrong ma-
|
\
2 Defects originating in ingot making and casting (segregation, unsoundness, po rosity, pipes, rionmetallic inclusions). 33 Defects due to working (laps, seams, shatter crocks, hot-short splits, delamina-
tion, and exceps local deformation).
44 Irregularitlefs and mistakes due to machining, grinding, or stamping (gouges,
4
'
'
.
"
November, 1972.
11
i
1!
Metals Engineering Ouarter/y.Americarl Society for Metals.
It is obvious from this chapter that ment of primary and secondary caus problem. Knowledge of each type o mize future problems. Every service obtain the maximum amount of infoi
burns, tearing,-fins, cracks, embrittlement). .From T J. Dolan, "Analyzing . Failures of Metal Components.
destroy vital evidence of cause of failure),
;
1 .
A
lurgical and visual examination stres will also add a great deal of knowlec ,
I
FAILURE ANALYSIS
663
5 Defects due to welding (porosity, undercuts, cracks, residual stress, lack of penetration, underbead cracking heat affected zone). 6 Abnormalities due to heat treating (overheating, burning, quench cracking grain growth, excessive retained austenite, decarburization, precipitationk ,
,
7
Flaws due to case hardening (intergranular carbides, soft coreT~"WFong heat
cycles). 8 Defects due to surface treatments (cleaning, plating, coating, chemical diffusion,
hydrogen embrittlement). 9 Careless assembly (mismatch of mating parts, entrained dirt or abrasive, residual stress, gouges or injury to parts, and the like). 10 Parting line failures in forging due to poor transverse properties.
II. Failures due to Faulty Design Considerations or Misapplication of Material 1 Ductile failure (excess deformation, elastic or plastic;-tearing or shear fracture). 2 Brittle fracture (from flaw or stress raiser of critical size). 3 Fatigue failure (load cycling, strain cycling, thermal cycling, corrosion fatigue, rolling contact fatigue, freitting fatigue). 4 High-temperature failure (creep, oxidation, local melting, warping).
5
Static delayed fractures (hydrogen embrittlement, caustic embrittlement, environ-
mentally stimulated slow growth of flaws). 6 Excessively severe stress raisers inherent in the design.
7
Inadequate stress analysis, or impossibility of a rational stress calculation in a
cdmplex part.
8 Mistake in designing on basis of static tensile properties, instead of the significant material properties that measure the resistance of the material to each possible
I
failure mode.
III. Failure Due to Deterioration During Service Conditions
Fig. 17-34 Holes in shafts sometimes
1 Overload or unforeseen loading conditions. 2 Wear (erosion, galling, seizing, gougirig, cavitation). 3 Corrosion (including chemical attack, stress corrosion, corrosion fatigue), dezincification, graphitization of cast iron, contamination by atmosphere. 4 Inadequate or misdirected maintenance or improper repair (welding, grinding,
concentrate stresses, resulting in torsional
-
fatigue failures. The splined shaft (top) is carburized alloy steel at Rockwell C 60, and the crankshaft (bottom) is of 1045 steel, Induction-hardened and tempered to .1
Rockwell C 55. (Courtesy of D. J. Wulpi,
punching holes, cold straightening, and so forth).
International Harvester Company.)
5
Disintegration due to chemical attack or attack by liquid metals or platings at
elevated temperatures.
6 Radiation damage (sometimes must decontaminate for examination which may destroy vital evidence of cause of failure), varies with time, temperature, environment,
£S*
and dosage.
(inclusions, embrittling impurities, wrong ma-
S ng and casting (segregation, unsoundness, po)
.
.
iams, ' ,
shatter cracks, hot-short splits, delamina-
to machining, grinding, or stamping (gouges,
'
ement).
7 Accidental conditions (abnormal operating temperatures, severe vibration, sonic vibrations, impact or unforeseen collisions, ablation, thermal shock, and so forth).
It is obvious from this chapter that the analysis of failure and proper assignment of primary and secondary causes of failure are often a very complex problem. Knowledge of each type of failure is important to avoid or minimize future problems. Every service failure should be carefully studied to obtain the maximum amount of information concerning its failure Metal.
ils." Metats Engineering Quarter/y American Society for Metals .
,
lurgical and visual examination stress analysis and intelligent questioning ,
,
will also add a great deal of knowledge regarding the failure Application .
i
'
*|
664
INTRODUCTION TO PHYSICAL METALLURGY
of this accurjiulated knowledge to the prevention of future failures is the
Grover
,
H. J., S. A. Gordon and L. F ,
.
-
;
;
1
Bureau of Aeronautics Navy Dept
goal of failure analysis.
,
Heywood
,
1962.
Larson
QUEiSTIONS
,
R. B.: "Designing Against'
F. R., and F. L. Garr: How Fa
for Metals Metals Park Ohio, Man ,
Lipson
17-2 What w(j!uld you look for in a visual examination of the above gear?
Parker
17-3
New York 1957. Polushkin 1956.1
,
,
,
Wulpi
,
.
17-13 Once a fatigue crack forms, what does its rate of growth depend upon? 17-14 Explain the relation between the origin of fracture and the final fracture zone of a part subjected to a one-way bending load and one subjected to a rotating bending load. ;: 17-15 Give three examples of stress raisers. 17-16 How will the beach marks curve with respect to the origin of failure in a hardened steel and annealed steel under cyclic loading? Explain. 17-17 Describe the kind of fracture which may occur as a result of a loose-fitting
r
I;
key on a shaft.
17-18 Explain the effect of three strength reducers. Explain why residual stresses are important in failure analysis. Under which conditions are residual stresses most significant with regard to
the material?
17-21 What will be the difference in fracture surface appearance between a low |
overstress and a high overstress on a material?
j
17-22 What is the difference in appearance between bending fatigue and torsion jj
fatigue?
|
17-23 What will be the difference in fracture between steel and cast iron subjected | '
'
to torsion?
;
i
;
REFERENCES
Alban, Lester .£.: Why Gears Fail, Metal Progress, Aroerican Society for Metals, Metals $ .
Park, Ohio, November 1970. | ; R. D., |nd B. F. Peters: "Why lyletals Fail," Gordon and Breach Science Pub- f } lishers, New York, 1970. i|| '
"
"
.
.
-. .
.
.
-
-
I-j
» '
. .
m
Barer
,
Bennett, J. A.|, and G. W. Quick: Mechanical Failures of Metals in Service, Nat. Bur. Std. Circ. 550, Washington, D.C., 1954. Forrest
,
Peter-R.: "Fatigue of Metals," Addison-Wesley Publishing Co., Inc., Reading, .
Mass., 1964
U
;
mmmm
I i
0
!
N.J., 1967
Ohio 1966.
How may the origin of a fatigue fracture be determined?
17-20
A. M.: "Materials Considerati
ic
.
D. J.: "How Components Fail
.
Describe the five types of loading and the type of stresses produced. What is the most common type of fracture in machine parts?
17-19
E. P.: "Defects and Failure,
Republic Steel Corporation: "Analysi
Ruskin Cliffs
"
r..
,
E. R.: "Brittle Behavior of Em ,
17-10 Describe the three principal ways in which fatigue stresses develop. .Nl Describe the development of beach marks" on the surface of a fatigue failure. ,
,
,
why ductile fractures occur when the shear strength is exceeded.
17-12
Charles: Basic Course in Faili .
What tests would you perform on the above gear? Explain the reasons for
17-8 17-9
,
lishing Co Cleveland Ohio Oct 1
selecting the tests. 17-4 Explain the goal of failure analysis. 17'5 Explain the difference in appearance between ductile and brittle fractures. 17 6 Describe the most important basic stresses in a part under external load. 17'7 Explain .why brittle fractures occur when the tensile strength is exceeded, and
'
,
17-1 Suppose you were given a gear with several broken teeth and asked to investigate the reason for failure. Make a list of the questions you would ask.
FAILURE ANALYSIS
665
i
i M e prevention of future failures is the ,|
Grover, H. J., S. A. Gordon, and L. R. Jackson: "Fatigue of Metals and Structures," Bureau of Aeronautics, Navy Dept., Washington, D.C., 1954. Heywood R. B.: "Designing Against Fatigue of Metals," Barnes & Noble, New York, ,
1962.
Larson, F. R., and F. L. Carr: How Failures Occur
,
Metal Progress, American Society
for Metals Metals Park, Ohio, March 1964.
, Lipson, Charles: Basic Course in Failure Analysis Machine DeAion The Penton Publishing Co. Cleveland, Ohio, Oct. 16, 1969. \ Parker E. R.: "Brittle Behavior of Engineering Structures," John; Wiley & Sons Inc., ,
jth several broken teeth and asked to inves;
,
!of the questions you would ask.
1 examination of the above gear?
,
,
Polushkin, E. P.: "Defects and Failures of Metals 1956.1
mt '
fracture in machine parts?
Cliffs, N.J., 1967.
Wulpi, D. J.: "How Components Fail," American Society for Metals, Metals Park, Ohio, 1966.
i
:{:S in which fatigue stresses develop.
,
i'ach marks" on the surface of a fatigue f ail ure
v/
.
fracture be determined?
does its rate of growth depend upon?
. V-? ;j origin -
\
i
of fracture and the final fracture zone
jg load and one subjected to a rotating bend-
aisers.
\ie with respect to the origin of failure in a sr cyclic loading? Explain. hich may occur as a result of a loose-fitting
.
igth reducers. re important in failure analysis. iidual stresses most significant with regard to Vi -
fracture surface appearance between a low material?
larance between bending fatigue and torsion
racture between steel and cast iron subjected
/ Progress, American Society for Metals
,
.
Metals
I
Vletals Fail," Gordon and Breach Science Pub-
jianical Failures of Metals in Service Nat. Bur. ,
Elsevier Publishing Co., New York,
,
.
;f
"
,
Republic Steel Corporation: "Analysis o f Service Failures," Cleveland, Ohio, 1961. Ruskin, A. M.: "Materials Considerations in Design," Prentice-Hall, Inc., Englewood
inco bolwoen ductilo and brittle fractures,
i; stresses in a part under external load when the tensile strength is exceeded, and ear strength is exceeded. and the type of stresses produced.
..
,
New York 1957.
the above gear? Explain the reasons for
;
,
,
|
954.
i Addison-Wesley Publishing Co. Inc., Reading, ,
m
'
f
INDEX n .I
Il Abrajsion of metals, 567
APPENDIX
nil
.tfiml : .
1:
tual cooling rate 233-287 determination of 233 ,
;
,
,
effect on microstruoture 290-292 effect of size and mass of piece 289-296
A
,
quenching medium 284-287
A
,
A
,
surface area to mass ratio 289-290 surface condition of piece 288-289
Temperature-conversion Table
,
°
.F
C
0C
"
F
"
F
,
i 273
-
250
-
200
-
459 418
-
-
-
-
150
.
-
"
-
-
238 -148 - 58
100 50 40
'.
-
3°
-
20 10
-
+
400
752
1000
770
1016
1832 1850
1600
410
1610
2912 2930
420
788
1020
1868
1620
2948
! 430
806
1886
1630
2966
440 450 460
824 842
1030 1040 1050
1904
1640 1650
470 480
878
490
914
;
-
-
-
328
510
950
1
520
968
1060 1070 1080 1090 1100 1110 1120
!)
530
986
1130
(
540
1004
1140
2084
1740
[j 550
1022
1150 1160 1170
2102 2120 2138
1750
1180 1190
2156
1780
2174
1790
1 I 12
1200
2192
1800
3272
1130
1210
2210
1810
620 630 640 650 660 670 680 690 700 710 720 730 740 750
1148
1220
2228
1166
1230
2246
1820 1830
3290 3308 3326
1184
1240 1250 1260
2264 2282 2300
1840 1850
1270 1280 1290 1300 1310 1320 1330
1364
1340
2444
1930 1940
1382
1350
2462
1950
760
1400
1360
2480
770 780
1418 1436
1370
2498
1960 1970
1380
2516
1980
790 800
1454
1390 1400
2534
810
1490
1410 1420 1430
2570
2588 2606
1990 2000 2050 2100 2150
.
22
1
1
;
4
.
14
860 896
0
32
i
500
932
5
41
f1
10 15
50 59
20s
68
25
77
30 35
86
560 570
1040
95
40
104
580
1076
45
113
590
1094
50
122
600
55
131
61.0
60
140
65
149 158 167 176 185
70 75. 80
85 90 95 100. 110
' ,
; "
194
.
'
203' 212
;
230 248 266
120 130 140 150 160
302 320
170
338
284
180
356
190
374
200 210 220 230 240 250 260 270 280 290 300 310 320 330
392 410
1058
1202 1220
1238 1256 1274 1292
1310 1328 1346
870 880 :
554 572
.
590
:
608 il 626
V
340
644
.-
350
662 f
1680 1690
2012
1700
2030
1710
2048
1720
2066
1730
.
-
212
450
184 157
-
-
129
101
-
-
-
-
-
-
-
-
73
46 40 34 29 23
350
-
150
-
100
-
50
-
40
-
-
18
15
5
12
10 15
1
30
2
35
4
40
7
45
10
50
3344
13
55
16 18
60
1860
3362 3380
2318
1870
3398
21
70
2336
1880 1890
3416 3434
24 27
75 80
3452
29
85
3470
32
90
2408 2426
1920
3488 3506
35
95
38
,
silicon 361 stainless 361-376
re
tungsten 361 vanadium 361
01
,
,
,
Alloy systems:
aluminum-copper 485-486 ,
aluminum-lead 206
1470 1480
2678
1634
1490
2714
2696
2732 2750
1520
2768
1706
940 950
1724
1530 1540
2786 2804
100
aluminum-manganese 489
110
aluminum-silicon 179, 490
54
130
aluminum-zinc
60 65
140
71
160
93 121 149
4082
,
492, 533 cobalt-tungsten 219 ,
,
.
'
76
170
83
180
88
190 200
copper-antimony, 222 copper-beryllium 478 ,
;
copper-lead 205 copper-nickel 168
\
copper-palladium 211 copper-silicon 474
177
me
,
,
copper-tin 472 copper-zinc 465
350
204
400
4352
232
450
2450 2500
4442 4532
260 288
500 550
iron-nickel 209
2550 2600
4622
316
4712
343
600 650
iron-silicon 222
2650
4802
371
700
2700
399 427
750 800
1742 1760
1550 1560
2822
2750
4892 4982
2840
2800
5072
454
850
1778
1570
2858
2850
5162
90 0
980
1796
1580
2876
2900
5252
482 510
. 990
1814
1590
2894
3000
5432
538
1000
CO'
ion
,
4172 4262
2350 2400
Arch Arsei Atom
,
,
2300
Appj
,
,
150
250 300
Anm Anot
-
120
2250
1598 1616
aluminum-magnesium, 490 aluminum-magnesium silicide 491
49
var
Atom tab
Atom 1 Atomi
,
gold-copper, 210
iron-chromium-carbon 362-364 ,
elei neu
,
iron-nickel-chromium
-
carbon 366
,
] : iron-tin, 222
lead-antimony, 187
-
, "
3
lead-tin 188 magnesium-aluminum 500 nickel-iron 513 ,
,
.
950
,
I . Silver-copper, 536 U silver-platinum 197 ,
titanium-aluminum 526 ,
-
I
r(
,
,
f
9
355-358 nickel-chromium 358-359 ,
43
I
I
e
9
nickel
3524
'
.
,
,
3542 3560
2200
1510
734;
Alu Alu Anr
,
,
1500
390
t
,
,
1910
1688
716 I-
All
,
,
65
1900
1670
960 970
,
,
2390
1652
698 !
I
I
moiybdenum 457 nickel 458 vanadium 458 Alloy steels 349-383 chromium 358 definition of 349 manganese 359-360 molybdenum 360-361
0
25
920 930
680 il
,
copper 457
10
20
2552
244-245
,
30 20
9
910
370 380.
'
Allotropy 85, 208-209 Alloy cast iron 454-458 chromium 457 ,
-
4
2354 2372
AlcladS 60 SyStem f0r S,eelS
200
-
-
1
AISI (American Iron and Steel Institute)
250
-
-
-
,
300
-
7
890 900
360
,
400
-
-
-
2624
1580
268 240
,
aging process 191-194 property changes due to 193-194 solution treatment 190-191
-
-
1760 1770
2642 2660
'
,
1994
1440
1526
518 536
1976
1450 1460
1508
830 840 850 860
;
1670
1544
820
446 464
500
1660
1958
1562
428
482
1922 1940
3578 3596 3614 3632 3722 3812 3902 3992
1472
-
A
Age hardening 190-194 ,
2984 3002 3020 3038 3056 3074 3092 3110 3128 3146 3164 3182 3200 3218 3236 3254
40
-
,
Admiralty metal 468
°
C
prol qua
Atomi: Ausfor Auster Auster defii horn micr
Austen on
c
INDEX
7
Abrasion of metals 567
Alloy systems: titanium-manganese 528
,
itual cooling rate
283-287 determination of 283 ,
effect on microstructure, 290-292
i
effect of size and mass of piece 289-296 ,
quenching medium, 284-287
1
titanium-vanadium Alnico, 517 630 Aluminizing 601 Aluminum 481-485 ,
anodizing, 494
,
corrosion resistance 493-497
,
1
°
°
C
F
°
F
C
,
,
Aluminum alloys 485-497 ,
aluminum-copper 485=488 i aluminum-magnesium 489-491 aluminum-manganese 489 '
,
1832
1850 1868 1886 1904 1922 1940 1958 1976 1994 2012 2030 2048 2066
2084 2102 2120 2138 2156 2174 2192 2210 2228 2246 2264 2282
i rig
4 s
S L
2300 2318 2336 2354 2372 2390 2408 2426 2444 2462 2480 2498 2516 2534 2552 2570 2588 2606 2624 2642 2660 2678 2696 2714 2732 2750 2768 2786 2804 2822 2840 2858 2876 2894
1600 1610 1620
2912 2930 2948
1630 1640 1650 1660 1670 1680 1690 1700 1710 1720
2966
2400 2450 2500 2550 2600 2650 2700 2750 2800 2850 2900 3000
-
-
-
184
-
73
-
48
-
3092
,
,
-
-
-
-
-
-
-
40 34 29 23 18 15 12 9'
classification system for steels
alloy designation, 481, 484 Aluminum brass 468 Aluminum bronze 476 ,
,
1
manganese, 359-360 molybdenum, 360-361 nickel, 355-356 nickel-chromium 358-359
10 13 16
60 ?>
18
85
2J
70 i
-
24 27
75
,
. ,
*
343
4802
371 399 427 454 482 510 538
650 700 750 800 850 900 950 1000
88 93 121
149 177
'
129-130 ,
temperature 132-135 of steel: full, 249-252 process, 254
stainless, 361-376
4712
83
,
,
600 550 600
35 38 43 49 54 60 65 71 76
recovery
recrystalllzation 131-135
,
silicon, 361
260 288 316
32
grain size, 136-137
-
4442 4532 4622
29
Annealing, 129-144 effect on properties, 137 grain growth, 135-136
definition of 349
.
tungsten, 361 vanadium, 361
J
properties, table, 252
Alloy systems:
'
spheroidizing 252-254 ,
aluminum-copper, 485-486
stress-relief
aluminum-lead, 206
254
,
Annealing twins 118 Anodizing 572 ,
aluminum-magnesium 490 aluminum-magnesium-sillclde. 491 alumirium-manganese, 489 aluminum-silicon, 179, 490 aluminum-zinc 492, 533 cobalt-tungsten 219 copper-antimony, 222 ,
,
,
copper-beryllium, 478
"
.
,
Apparent density of metal powders
,
table, 612
Architectural bronze 470 ,
Arsenical copper 462 Atom binding 76-78
M
,
,
covalent, 77 ionic 76-77 ,
| copper-lead, 205 copper-nickel, 168
metallic, 77-78 van der Waals 78 Atomic diameter 78-79
]
,
icopper-palladium 211 ,
,
copper-silicon, 474 copper-tin, 472 icopper-zinc 465
table, 79
'
Atomic numbers and weights 'table, 68 ,
Atomic structure 65-70
,
,
»gold-copper 210 ,
plron-chromlum-carbon. 362-364 Iron-nickel, 209 Iron-nlckel-chromium-carbon 368 ,
Iron-silicon 222 ,
Iran-tin, 222 lead-antimony 187 ,
lead-tin 188 ,
!,magnesium-aluminum, 500
| nickel-iron, 513
electrons, 65 neutrons, 65
protons, 65 quantum numbers, 69 Atomization, 607-609 Ausforming 383 Austempering, 313-315 Austenlte: definition of, 234 ,
homogeneity, effect of 282-283 ,
|:8llver-copper 536 ,
iSllver-platinum 197 ,
illltanium-aluminum 626 ,
4
temper designation, 484-485
,
204 232
5072 5162 5252 5432
Aluminum Association:
nickel, 458 vanadium, 458 Alloy steels 349-383
80 85 90 95 100 110 120130 140 150 160 170": 180 . 190 200 250 300 350 400,, 450
4892 4982
,
30 35 40 45 50 55
7
,
temper designation 484-485
25
2 4
,
designation system 481, 484
copper, 457 molybdenum, 457
5
aluminum-zinc 492-493 ,
chromium, 457
10
,
i
,
\ Alloy cast ifon, 454-458
0
aluminum-silicon-magnesium 491-492 corrosion resistance 493-497
,
il
244-245
Compdsition and properties of table, 496
,
r Allotropy, 85 208-209
4
+
,
Alclad, 497 602
7
-
,
,
chromium, 358
-
aluminum-silicon 489
,
IS 20
-
.
solution treatment 190-191
3
101
-
,
,
AISI (American Iron and Steel Institute)
300 250
-
-
3074 3110 3128 3146 3164 3182 3200 3218 3236 3254 3272 3290 3308 3326 3344 3362 3380 3398 34163434 345? 347(1 3488 350B 3524 3542 3560 3578 3596 3614 3632 3722 3812 3902 3992 4082 4172 4262 4352
-
157 129
. '
1940 1950 1960 1970 1980 1990 2000 2050 2100 2150 2200 2250 2300 2350
240 212
,
,
5
268
-
2984 3002 3020 3038 3056
1730 1740 1750 1760 1770 1780 1790 1800 1810 1820 1830 1840 1850 1860 1870 1880 1890 1900 1910 1920 1930
-
tf
,
surface condition of piece 288-289 Admiralty metal 468 Age hardening 190-194 aging process 191-194 property changes due to, 193-194
.
527
,
,
surface area to mass ratio 289-290 i
,
,
,
microstructure of 235 Austenlte transformation: ,
on continuous cooling 273r276 f ,
n 690
INDEX
Austenite transformation: isothermal, 261-273 :
'
r
\
,
,
pearlitic, 430-434
Austenitic stainless steels, 371-374 Axial ratio, 84
nodular, 450-453 microstructure, 450-454
Babbitt:
properties of, 452 types of, 423-424
Austenitizing temperature, 281-282
copper-beryllium 478 copper-nickel 478-479 copper-silicon 474 copper-tin 472
manufacture of, 426-429
slow cooling, 237t-241 '
Copper alloys:
Cast iron, malleable: ferritic, 429-430
,
,
mechanical properties of table, 482-4 ,
lead-base, 520-521 tin-base, 521 , Bain, E. C, 260 Bainite:
austempering, 313-315 microstructure, 268-269
,
temper designation 463 Copper-constantan 7 ,
white, 424-426
,
Copper oxide in copper
Castings, defects in, 98-100
Bending fractures, 656-659 Beryllium bronze, 476-479
age hardening, effect of, 478-479 microstructure, 479
phase diagram, 478 Bonderizing, 600
,
pipe, 98 porosity, 98-99 shrinkage, 98
,
definition of 583 ,
dezincification 467 ,
effect on properties table, 596 ,
electrochemical principles 583-585
Cavitation corrosion, 591-592
,
electromotive-force series 586
Cemented carbide tools, 415-419, 625
,
factors in 585-590
Cementite, definition of, 234
,
galvanic series, 587 inhibitors 600
Ceramic tools, 420-421 Cermets, 419-420
,
methods of combating 597-602
Charpy impact specimen, 43
,
season cracking 466 specific types 590-597 Creep properties of, table 555 Creep tests 45, 548-549 Crevice corrosion 592
Chemical elements:
Bragg equation, 90,
,
atomic number, atomic weight, and symbols, table, 68
Brasses, 464-472
alpha, 466-469
,
,
definition of, 65
.
yellow, 466-468 alpha plus beta, 469-471 cast, 471-472
Breaking strength, 40-41 Bhnell hardness test, 26-27 Briquette, 607 Bronzes, 472-478 ,; aluminum, 476 >
,
,
periodic table, 71
,
Critical cooling rate definition of, 260 Critical resolved shear stress, table 115
Chilled cast iron, 448-450
,
Chromel A, 511 Chromel-alumel, 7 Chromol C, 511
,
,
interstltials 96-97
Chromium in cast iron, 453-457
,
vacancies
Chromium steels, 358 Chromizing, 573
96
,
Crystal structure 80-85 body-centered cubic 80-81 ,
Classification of failure causes, 662-663 Coherent lattice theory, 192
tin, 472-475
Coin silver, 536 Coining, 622
1 i
space lattice 80
Commercial bronze, 468
i
unit cell 80
Bulls-eye structure, 432 Calorizing, 602
Carbon-concentration gradient, 318-319 Carbon solubility in iron, 234-236 Carbonitrlding, 326-328 Carburizing, 317-326
case depth by, table, 321 equation, 319 gas, 322-323 heat treatment after, 325-326
liquid, 323-325 |
pack, 319-321
1
Constantan, 511 Coolincj rate:
types of, 444 malleable, 426->434
4
m
1
461
I
3
Diffusion 163-165 Dislocation 97-98 in slip 113-115
614-616
,
Ductile iron (see Nodular cast Iron) Ductility elongation 41 ,
reduction in area 41 Dumet wire 517 ,
,
Eddy current Inspection 57-60
effect of zinc on, table, 465
Elastic hardness 24-25 Elastic limit 38-39
bronzes, 472-480
,
,
aluminum, 476
,
Electrolytic deposition of
beryllium, 476-479
.
166
,
,
alpha plus beta, 469-471
1 r'.
,
Dezincification 467 ,
alpha, 466-469
"
,
,
Die compaction of powders,
.5 482
brasses, 464-472
copper-aluminum, 476
in tool steels 397 Dendrite 93
Dendritic segregation 1
free-cutting, 462
r
,
,
arsenical, 462 In cast iron, 457
silicon, 475-476 tin, 472-475
,
,
,
mechanical properties of, table,
chilled, 448-45(5 gray, 434-448 graphite flake, 441-445 size chart, ,i»42
Cyaniding 326-328
,
Davenport E. S., 260 Decarburization 319
Copper, 461-464
Case hardening. 315-336 carbonitrlding, 326-328
alloy, 453-458 :!
1
I
actual (see Actual cooling rate)
temper designation, 462-463 Copper alloys, 464-480
,
,
1
critical, 260
oxygen-free high-conductivity,
,
,
-
Congruent-melting alloy, 170, 195
silver-bearing, 462
face-centered cubic 81-83
Crystal systems table, 81 Crystallization 91-94 Crystal log raphic planes 87-89 Cupronickels 478-479
1
Comparison method to measure grain size, 101-102 Composite metals, 625
electrolytic tough-pitch, 461
Cast Iron, 423-4E&
,
close-packed hexagonal 84-85 ,
Cartridge brass, 467 carburizing, 316-326 cyaniding, 326-328 flame hardening, 332-333 gas cyaniding, 327 Induction hardening, 333-336 nitrlding, 328-332
,
Crystal imperfections 94-98 dislocations 97-98
beryllium, 476-473 silicon, 475-476 i
|
461-462
,
Coring 165-166 Corrosion of metals 583-602
hot tears, 99-100
Cathodic polarization, 584
Bakelite, 15
red, 468 *69
nickel silvers 480
J
*
Electrolytic tough-pitch
1
powders, 609
copper 461 ,
Electron configuration table, 75 Electroplating 571-572 ,
,
'
pp.
*4
INDEX
Copper alloys:
Cast iron, malleable; territic, 429-430
,
nickel silvers 480
,
,
Copper oxide in copper
,
461-462
Coring 165-166 Corrosion of metals 583-60 definition of 583 dezinclfication 467 effect on properties table, 596 electrochemical principles 583-585
hot tears, 99-100 pipe, 98
,
,
porosity, 98-99
,
i
shrinkage, 98
Cathodic polarization, 584 Cavitation corrosion, 591-592
,
,
4
electromotive-force series
Cemented carbide tools, 415-419, 625 Cementite, definition of, 234
factors in 585-590 galvanic series, 587 inhibitors 600
Ceramic tools, 420-421 Cermets, 419-420
,
586
,
reagents, table 22 ,
Eutedlic reaction 175 equation 178, 183, .
218
,
Eut ctlc systems
171-179 mlcrostructures: AI-SI 180 Pb-Sb 187 ,
,
,
Pb-Sn, 188 properties, 189-190
Eutectoid reaction 212-214 ,
equation 212
,
i 1
,
Equilibrium diagrams (see Phase diagrams) ,
,
Castings, defects in, 98-100
Endurance limit 44 Erosion of metals 567 Etching 18-19
,
temper designation 463 Copper-constantan 7
white, 424-426
,
Extractive metallurgy vili ,
,
methods of combating 597-602
Charpy impact specimen, 43
,
Chemical elements:
atomic number, atomic weight, and symbols, table, 68
season cracking 466 specific types 590-597 Creep properties of, table 555 Creep tests 45, 548-549 Crevice corrosion 592 ,
1
,
,
effect of residual stresses 653-654
,
,
effect of strength reducers
,
periodic table, 71
,
Critical cooling rate definitidfl of, 260
Chilled cast iron, 448-450
,
Critical resolved shear stress table, 115
Chromel A, 511
Crystal imperfections 94-98 dislocations 97-98 interstitials 96-97
Chromel-alumel; 7
,
Chromium steels, 358 Chromizing, 573
,
96
,
,
,
663
-
i
body-centered cubic 80-81 close-packed hexagonal 84-85 face-centered cubic 81-83
,
Ferrlte:
,
Coherent lattice theory, 192
definition of 234 microstructure of 235 Ferritic irons 451
,
1
Coin silver, 536 Coining, 622
1
Commercial bronze, 468 Comparison method to measure
grain size, 101-102 Composite metals, 625
.
,
,
space lattice 80 unit cell 80
,
,
,
Ferritic stainless steels 369-370 , Flame hardening 332-333 Flame-plated coatings properties of,
,
Crystal systems table, 81 Crystallization 91-94 Crystallographic planes, 87-89 Cupronickels 478-479
,
,
,
,
Congruent-melting alloy, 170, 195 Constantan, 511
Cooling rate:
actual (see Actual cooling rate)
critical, 260
i
1 M
1
,
table
,
Floe process 330
Davenport E. S. 260 Decarburlzation 319 in too/ steels 397
Forging brass 470 Foundry type metal
,
,
,
Fluorescent-penetrant inspection, ,
Fracture modes 635-636 ,
,
in cast Iron, 457
temper designation, 462-463 Copper alloys, 464-480 brasses, 464-472
alpha, 466-469
alpha plus beta, 469-471 effect of zinc on, table, 465
,
Free-cutting brass 470
166
,
Dezinclfication 467 Die compaction of powders, 614-616 ,
Diffusion 163-165 Dislocation 97-98 ,
,
reduction in area
,
41
Fusebond method of spraying
575
Eddy current inspection 57-60
Elastic hardness 24-25
Galvanizing 601 Gas carburizing 322-323 ,
,
Gilding metal
,
468
Gold, 543 properties of, 537 Gold alloys 537
,
,
Grain size in castings
,
,
100-101
Grain size measurement 101-104
,
Elastic limit 38-39
,
,
Electrolytic deposition of powders
Electrolytic tough-pitch copper,
Electron configuration table, 75 Electroplating 571-572 ,
,
462
,
,
,
Ductile iron (see Nodular cast iron) Ductility elongation 41 Dumet wire 517
,
Fretting corrosion 593-594 Galvanic corrosion 596
in slip 113-115
i
Free-cutting copper
,
,
,
oxygen-free high-conductivity, 461 silver-bearing, 462
519
,
Fracture 118-121
,
Dendritic segregation
electrolytic tough-pitch, 461 free-cutting, 462 mechanical properties of table, 482
51-54
,
,
Dendrite 93
arsenical, 462
578
,
Cyaniding 326-328
,
Copper, 461-464
,
Fatigue tests 44-45 Fernico 516
Crystal structure 80-85
Classification of failure causes, 662
,
,
,
vacancies
648
effect of stress raisers 643-647 fatigue fractures 640-643 loading types of, 637-640 modes of fracture 635-636 torsional fractures 659-661 Fatigue fractures 640-643 ,
,
Chromium in cast iron, 453-457
,
,
,
Chromel C, 511
Failure analysis, 633-664 bending fractures 656-659 classification of causes 662-663 ,
,
49
definition of, 65
copper-aluminum, 476
,
,
properties of, 452 types of, 423-424
silicon, 475-476 tin, 472-475
End-quench hardenability test 297-302 correlation with C-T diagram 300
472
,
mechanical properties of table, 482-483
microstructure, 450-454
beryllium, 476-479
,
,
copper-tin
nodular, 450-453
bronzes, 472-480 aluminum, 476
,
Elinvar 517
,
,
pea-litic, 430-434
m
Electrotype metal 519
copper-beryllium 478 copper-nickel 478-479 copper-silicon 474
manufacture of, 426-429
891
,
461
609
Gravity compaction of powders, Gray cast iron 434-448 ,
graphite flake, 441-445 size chart 442 types of 444 ,
,
617
ii;: - . 692
.'..:/v-
INDEX
Gray cast Iron:
.
. 1
M,
Inconel X 511-512
Induction hardening 333-336
manganese In, 438
Ingot Iron 225 Intergranular corrosion 594 Intermediate alfoy phases 148-150, 195-198
,
\
microstructure, 436
Magnaflus test, 49-51
,
Magnaglo test, 49-51 Magnesium, 498-499 Magnesium alloys, 500-507 composition and properties of, table, 506
,
phosphorus In, 438-439 properties of, 446 silicon In, 435-438 sulfur in, 438
,
electron compound
150
,
intermetalllc compound, 149 Interstitial compound 149-150 Intermetalllc compound 149 Interstitial compound 149-150
corrosion resistance, 505
,
designation and temper, 498-499
,
H Monel, 510
j
/y Hadfleld manganese steel, 360
\/ Vl Hard Hadfleldfacing, silicon577-;581 sieeL 361 _
Hardenability, 296-302 definition of, 293
specification for, 301-302 test for, 297-289
Hardenability bands, 301 Hardenable carbon steels, 343-345 Hardness, 24-37
Interstitial solid solution 153-154 ,
Interstitials 96-97 Invar 516 Iridium 541 Iron: ,
,
,
alpha, 225
Malleable cast iron, 426-434
beta, 225
copper-alloyed, 434
cooling curve, 226
ferritic, 429-430 manufacture of, 426-429
dslta, 225
conversion, table, 28-29 cylindrical correction chart, 36
Iron-iron carbide phase diagram
pearl itic, 430-434 ,
delta region 231-232
tests for: applications, table, 34 Brinell, 26-27
effect of alloying elements on
Knoop, 32-733 Vickers, 32
Rockwell: scales, table, 30
superficial tester, 30-31 scleroscope, .24-25 scratch, 25-26 Vickers, 31-32
Hardness-penetration diagram, 293
Hastelloy A, 512!j Hastelloy B, 512i:'
Hastelloy C, 512
Hastelloy D, 511. Hastelloy X, 512-513
230-234
,
352-353
,
eutectic reaction, 232 eutectoid reaction, 233 peritectic reaction, 231 structures: austenite 234 cementite 234 ferrite, 234 ledeburlte, 234 ,
,
pearlite, 234
Isostatic compaction of powders Isothermal annealing 270
,
,
(See also Metal spraying)
determination, 260-265
etching 18-19 reagents, table, 22 fine polishing, 16-17 Intermediate polishing, 16 mounting, 15-16 ,
Isotopes 74-76 ,
izod impact specimen 43 ,
Heat of fusion, 92
Heyn method to rneasure grain size, 102-103 Holography, 60-62
Jeffries method to measure grain
Homogenization, 166-168
Jominy hardenability test, 297-302
size, 103-104
Hot pressing of powders, 621-622 /Hot working, 137-140 Hume-Rothery, William, 150, 152 Hypereutectic alloys, 174 Hypereutectic cast irons, 234 Hypereutectoid alloys, 213 Hypereutectoid steels, 234 Hypoeutectlc alloys, 174 Hypoeutectic cast irons, 234 Hypoeutectoid alloys, 213 Hypoeutectold steels, 234
,
Knoop hardness test
,
32-33
Kovar, 516
Metalloids, 76
Metallurgical microscopes, 19-24
Lead, 517
table, 620-521
lead-antimony, 517-518 lead-tin, 51 8-520 terne metal, 519 .
,
,
Liquidus line, 158
Impact test, 42-44
specimen: Charpy, 42-43
Low brass 469 Lucite, 15-16 ,
Inconel, 511
M, temperature
,
i
.
KM'
i
t
at high temperatures, 547-555 elevated-temperature test, 547-548 properties of, table, 555 at low temperatures 555-565 properties of, table, 558-559 ,
Metastable phase, 230 Miller indices, 87-89
Modules of elasticity, 41 Molybdenum: ,
257
effect of carbon, 259
I
Metals:
in cast iron, 457 electrical contacts 543-544
Izod, 43-44
Impregnation, 623
nil nil ni(
nil nli Nick Nick Nick Nitri ca
Nodi mi pr
Non( ec
eh
mi
rai su
ull Norn
Optii
optical, 19-20 characteristics of, 76
,
lllium B, 513 lllium G, 513 lllium R, 513
pe
Nick
electron, 19-24
Lead alloys, 517-521 composition and properties of,
Ledebufite, definition of 234 Linotype metal 520 Liquid cgrbgrizing, 323-325 Liquid-metal corrosion 597
Ihrigizing, 573
I
efl
he
rough grinding 15 sampling, 15 Metallography, definition of, 14 ,
K Monel, 510
E efl
flu
in Bakelite 15 in Lucite, 15-16
Hot hardness, 395
du
Fh
Metallographic sample preparation, 15-19
,
CO
D
nil
Metallograph, 19
,
Nav Nich Nich Nick A in
nil
Metallizing, 600
260-266
alloy element effect of, 277-279 austenitic grain size effect of, 280-281
Navi
definition of, 257 microstructure of, 258-260 Martensite transformation, 257-260 Martensitic stainless steels, 366-368 Metal "whiskers," 95
,
Isothermal-transformation diagram
Mon Mon Mun
Martensite:
Metal spraying, 573-577
616
Molv
co
Manganese bronze, 470 Manganese steel, 359-360 Maraging steels, 377-383 composition of, table, 378 properties of, table, 380 Martens, A., vii
,
definition of, 2f
microhardness, 32-33
joining, 505, 507 magnesium-aluminum-based, 501-503 magnesium-rare earth-based, 503-504 magnesium-thorium-based, 504-505 magnesium-zinc-based, 503 Magnetic-particle inspection, 49-51
,
gamma, 225 Iron-constantan 7
file, 26
temperature, 257-259
effect of carbon, 259 formula for calculation, 259
,
heat treatment, 439-441
properties of, 542
Orde Osm
Oxyc Pack Palle pr Palls pr Park Paul: Pear de mi
-
-
- - .
INDEX
Ma
Inconel X 511-512 ,
induction hardening 333-336 ,
Ingot iron 225 Intergranular corrosion 594 Intermediate alloy phases 148-150, 195-181 electron compound 150 ,
,
,
,
intermetalllc compound 149 interstitial compound 149-150 Intermetalllc compound 149 Interstitial compound 149-150 ,
corrosion resistance, 505
,
designation and temper, 498-499 joining, 505, 507
,
magnesium-aiuminum-based, 501-503 magnesium-rare earth-based, 503-504 magnesium-thorium-based, 504-505
,
Interstitials 96-97 ,
Invar, 516 Iridium 541
magnesium-zinc-based, 503 Magnetic-particle Inspection, 49-51
,
Iron:
alpha, 225
Malleable cast Iron, 426-434
beta, 225
copper-alloyed, 434
cooling curve, 226 ,
;
-.
345
delta
pearlitic, 430-434
Iron-constantan 7 ,
6
Icon-Iron carbide phase diagram 230-234 ,
delta region 231-232 ,
34
effect of alloying elements on
352-353
,
eutectic reaction, 232 eutectoid reaction, 233
peritectlc reaction, 231 :V:k':.
,
composition of, table, 378 properties of, table, 380 Martens A., vil
,
pearlite, 234
Isostatic compaction of powders Isothermal annealing 270
,
616
,
Isothermal-transformation diagram 260-:266 ,
alloy element, effect of, 277-279 austenitic grain size, effect of 280-281 ,
determination, 260-265
Isotopes 74-76 Izod Impact specimen 43
Jeffries method to measure grain size, 103-104
2
Metallographlc sample pre aratioh, 15-19 etching, 18-19 fine polishing 16-17 intermediate polishing, 16 mounting, 15-16 in Bakelite 15 In Lucite, 15-16
Metallurgical microscopes 19-24 ,
composition and properties of table, 520-521 lead-antimony, 517-518 lead-tin, 518-520 terne metal, 519
Ledeburite, definition of, 234
Linotype metal 520 Liquid carburlzing 323-325 Liquid-metal corrosion 597 Liquidus line, 158 ,
,
,
Low brass, 469 Lucite, 15-16
A
-
Mj temperature, 257 effect of carbon, 259
m
,
Floe process
,
330
Nodular cast iron, 450-453 microstructure, 450-452
properties of, 452' Nondestructive testing, 45-62 eddy current, 57-60 elements of, 46
fluorescent penetrant, 51-54 holography, 60-62 magnetic particle, 49-50 radiography, 46-49 summary, table, 58-59 Normalizing, steel, 254-256
electron, 19-24
Lead alloys, 517-521
: ;..:
case depth by 328
ultrasonic, 54-57
Metalloids, 76
Lead, 517
nickel-sillcon-copper-based, 511
Nitrlding, 328-332
(See a/so Metal spraying) Metallograph, 19
,
Kovar, 516
nickel-molybdenum-iron-based, 512
Metallizing, 600
sampling, 15 Metallography, definition of 14
Knoop hardness test, 32-33
nickel-chromium-rtiolybdenumIron-based, 5 2-513 nickel-copper-based, 509-511
Nickel silvers, 480
rough grinding, 15
K Monel. 510
nickel-chromlum-molybdenumcopper-based, 513
Nickel steels, 355-358
,
Jominy hardenabiilty test, 297-302
effect of low temperature, 562 effect on transition temperature, 562
Metal spraying, 573-577 Mefdl "whiskers," 95
,
ze, 102-103
D nickel, 509 duranickel, 509 E nickel, 509
Nickel-chromium steels, 358-359
,
,
in cast Iron, 458 L composition and properties, table, 508
Martensite transformation, 257-260 Martensitic stainless steels, 366-366
reagents table, 22
,
Navy steel, 361 Nichrome, 511 Nichrome V, 511 Nickel, 507-509 A nickel, 509
nickel-Iron, 513, 516-517
microstructure of, 256-260
,
93
Manganese steel, 359-360 Maraging steels 377-383
Martensite: definition of, 257.
,
Naval brass, 470
514-515 nickel-chromlum-lron-based, 511-512
Manganese bronze, 470
,
structures: austenlte 234 cementlte 234 ferrite 234 iedeburite, 234
'
Muntz metal, 469
Nickel alloys, 509-517 composition and properties, table,
manufacture of, 426-429
gamma, 225
Molybdenum steel,. 360-361 Monel, 170, 510 Monotectic reaction, equation, 204
permanlckel, 509
ferritlc, 429-430
225
,
Magnaflus test, 49-51 Magnaglo test, 49-51 Magnesium, 498-499 Magnesium alloys, 500-507 composition and properties of, table, 506
,
Interstitial solid solution 153-154
i
temperature, 257-259 effect of carbon, 259 formula for calculation, 259
693
optical, 19-20 ,
.
i
i
Optical pyrometer, 11-14
Metals:
characteristics of 76 ,
at high temperatures 547-555 elevated-temperature test; 547-548 ,
properties of, table, 555 at low temperatures 555-565 properties of, table, 558-559 ,
Metastable phase 230 ,
Miller Indices, 87-89
Modules of elasticity
,
41
Molybdenum: In cast Iron, 457 electrical contacts 543-544 properties of, 542 ,
Order-disorder transformation, 209-212 Osmium, 541
Oxygen-free high-conductivity copper, 461 Pack carburlzing, 319-321 Palladium, 543
properties of, 539 Palladium alloys, 539-540 properties of, table, 542 Parkerizing, 600 Pauli exclusion principle, 67 Pearlite: definition of, 234 microstructure, 235
4
694
ij
INDEX
Peltier effect, 5
Periodic table of'elements, 70-74 Perltectic reaction, 196-202
sintering, 61,9-621
Preferential corrosion, 596 Preferred orientation, 122
definition of, 147
types, 148-154 Intermediate, alloy, 148 150 electron gompound, 150
intermetajilc compound, 149
interstitial;! compound, 149-150
pure metal, r148 solid solution, 150-154
Proportional limit, 38 Pseudoeutectic alloy, 171 Pyrometers, 5-14 ji optical. 11-14
substitutional, 152-153
Phase diagrams/ 155-220 coordinates of, 156
determination of, 156-157 reactions on: eutectic, 171-178 eutectoid, 212-214
\ monotecticj 202-205 '
perltectic, 196-202 peritectoid, ! 214-216
Phpsphor bronze (see Tin bronze) Physical metallurgy, viii Piezoelectric effect, 54
Pitting corrosion, 591 effect on properties, 124-126
polycrystalllng material, 121-124 by slip, 107-116 in different lattice structures, 115-116 mechanism, 110 115
resolved shear stress, 110
by twinnlng,.116-118 -
platlnum-lrldium, 539 platinum-nickel, 539
platlnum-rtioidium, 538-539 platinum-ruthenium, 539 platinum-tungsten, 539
r properties of, table, 542
? ; Polymorphism, 85
.
-
622
"
n
623
JS ix & & 610-611
:
,
®
m>
1
I j
5
n
s
factors in 152-153 interstitial 153-154 mlcrostructures Cu-NI, 169 order-disorder 209-212 substitutional 152-153 terminal 182 types of 152 Solidus line 158 Solvus line 182 Sorbite 309
in i Isot
,
Isot
,
,
Quantum numbers, 69
,
Quenching, 283-287 actual cooling rate, 284 mechanism, 283-284
mar
,
mar
,
mar
mo I nicl.
,
medium, 284-286
,
circulation, 287
,
temperature, 286-287 type of, 284-285. stages in, 283-284
Sorby Henry C, vii ,
Spheroidal graphite Iron (see Nodular cast iron) Spheroidite 254
Radiography of metaii,
'
£-49
Spherulltlc iron ''-?e Nodular cast iron) Stainless stee'
Rockwell hardness test:
cyllndrjcal correction chart, 36 scales, table, 30
Rockwell superficial hardness test, 30-31
pi silic stair
,
composition and properties of
,
table, 370-371 ferritic, 369-370 .
temf tool
manganese in 374 martensitic 366-368
tung
,
vane
.
from heat treatment, 336-343
precipitation-hardening, 374-376
Stelllte
Steadite In cast Iron 438-439
Steriini
,
Steels:
Stiffnes
alloy composition of, 247 alloying elements in effects of, table, 356-357 . annealing: full 249-252 ,
Strain, Stress,
,
true, Stress StressStress-i
-
,
.
Ruthenium, 541
process, 254
S cgrves (see Isothermal-transformation
spheroidizlng 252-254
properties of. table 252
diagram)
.
'
,
stress-relief, 254
Substiti
S Monel, 610
austempering 313-315
Sauveur, Albert, vll
austenite,
§ciJl*s on sjeel, 288 protection against, 288-289 Scleroscope hardness test, 24-25
ausfenitizing temperature 281-282
Season cracking In brass, 466
,
homogeneity of, effect of
,
282-283 ,
balnite, 268-269 Phromlum 358 ,
critical cooling rate 260 decarburizatlon 319 effect: of manganese 248
electrical contacts, 543
har
,
Silipon bronze; 475 476
Sliver alloys, 536-537
relati relati
Surface
classification of 244-245
Sherardidnfl, 531, 602 Short-time tension tests, 551
Siiippntjlng. 573 , .it Silver, properties of*;536
chenr crysti
Superla
jpicrostructure, 268-269 eoldrshorl 248 ;
Sillpon steel, 361
quei resli seal
,
temperature. 132-135 Red brass, 469 Red hardness, 395-396
Resilience. 42 Rhodium. 541
2r376
austenitic 37 i-j/4 classification of 362
Recrystallizatlon, 131-135
norr
plaii
252-254
,
Radiation pyrometerjv.l0-11
nicl. nitn Pi
Spheroidize annealing
R Monel, 510
properties of. table, 542
mm
c
,
SelMi rlcellno bearings, 930
Y % . vi;5acUteal contacts, 544
S"
8
,
Pyrometry, 1
c
hyp
,
Seebaek pffect. 5
Powder, matalluray
y.t
,
.
platinum alloys, 537-539
.
Slip casting 617 Slip system 115 alloys, properties of 168-170
radiation, 10-11 thermoelectric, 5-10
hyp
r
,
'
Platinum. 537,; 543
r
619-621
,
Slip, 107-116
Residual stresses, 653-654
Plastic deformation:
.
,
Sintering in powder metallurgy
Red-short, 438
Plasma spraying, 575
.V
Sintered carbide tools 415-419
Raoult's law, 172
Platlnlte, 516
462
,
,
,
159 161
I;
Silver-bearing copper Silver solders 537
Solders, 519 522 Solid solutions ;S0-154
table, 218 ;
rules for: chemical composition of phases, 159 relative amounts of each phase,
-
hai hai he: hoi
,
,
interstitial! 153-154
4
t
silver-copper 536-537 silver-copper-zinc 537
374-376
Pewter, 523 Phase:
ii
,
Precipitation hardening (see Age hardening) Precipitation-hardening stainless steeJe,
equation, 214 Permalloys, 517
-
Steel
silver brazing 537
properties and applications of,
,
Peritectoid reaction, 214-216
:
sir Silver alloys:
table, 626-629
equation, 196
m
V
Powder metallurgy: processes, 606-607
,
Pooling-transformation diagram ,
,
,
of phosphorus 248 ,
Of silicon 248 ,
of sujfur, 248
-
hardenablllty 296-302 .
,
273-276
Temper
Tempen tab)
Tempen by co
by py opti radi
recc
1 '
4
Powder metallurgy:
processes, 606-607
properties and applications of, table, 626-629
sintering, 619-621
Precipitation hardening (see Age hardenln£ Precipitation-hardening stainless steels, '
* 374-376 Preferential corrosion, 596 Preferred orientation, 122 Proportional limit, 38 Pseudoeutectic alloy, 171 Pyrometers, 5-14 5i '
Quenching, 283-287
actual cooling rate, 284 mechanism, 283-284 medium, 284-286 circulation, 287
temperature, 286-287 type of, 284-285, stages In, 283-284
iase,
ie)
-
Radiography of metSiU,.
"
; r
v-}.
.">
hypereutectold, 234 ..
- .
Sintering in powder metallurgy. 619-621 Slip, 107-116
critical temperatures, 240-241 A;
Slip casting, 617 Slip system, 115 Solders, 519, 522
slow cooling, 240-241 , ; :
microstructures of, 242 .
'
,.
.
'
hypoeutectoid, 234
.,y
critical temperature /237.;,:- . '
microstructures of, :;239
; 50-154
-
slow cooling, 237-240 .
,
microstmctures, Cu-Ni, 169
in iron-iron carbide diagram, isothermal annealing„270 isothermal-transformation diagram,
order-disorder, 209-212 substitutional, 152-153 terminal, 182
260-266 manganese, 359-360 maraging, 377-383
types of, 152
martensits, 257
Red hardness, 395-396
molybdenum, 360-361 nickoly SSS dSS
Spheroidal graphite iron (see Nodular
normalizing, 254-256 properties of, table, 252 -
nickel-chromium, 358 359 nitralloys, 330
Resilience, 42
|s, 115-116
Rockwell hardness test;
.2t376
Rhodium, 541
cylindrical correction chart,
36
scales, table, 30
protection against, 288-289
coftiposition and properties of,
stainless (see Stainless steels)
table, 370-371 fefritic, 369-370
tempering, 305-313
manganese in, 374
tungsten, 361
tool (see Tool steels) vanadium, 361
Rockwell superficial hardness test, Ruthenium, 541
S curves (see Isothermal-transformation diagram) S Monel, 510
Stellite, 413-414
SteadIte in cast iron, 438-43S
Sterling silver, 536
Steels:
Stiffness, 41
alloy, composition of, 247
Strain, 37
alloying elements in, effects of,
Stress, 37 true, 41-42
table, 356-357
30-31
annealing: full,'249-252 process, 254 properties of, table, 252 spheroidizing, 252-254
Stress corrosion, 594-596
Stress-relief annealing, 254 Stress-rupture tests, 54i?-552 Substitutional solid solution 152-153 chemical-affinity factor 152
stress-relief, 254
,
austempering, 313-315
,
austenite, homogeneity 282-283
protection against,
288-289
Scleroscope hardness test,
24-25
Season cracking in brass, 466 Seeback effect, 5
Self-lubricating bearings, 630 Sherardizlng, 531, 602
Short-time tension tests, 551 Silicon bronze, 475-476 . .'!.
properties of, table,
of, effect of, ,
bainite, 268-269 microstructure, 268-269 chromium, 358 classification of, 244-245 cold-short, 248
cooling-transformation diagram critical cooling rate 260 ,
decarburlzation 319 ,
542
,
of silicon, 248 of sulfur, 248
hardenability 296-302 ,
,
,
austenitizing temperature 281-282
effect: of manganese of phosphorus 248
crystal-structure factor 152 relative-size factor 152
relative-valence factor 152-153 ,
Superlattice 209 ,
Surface heat treatment (see Case hardening) Temper brittleness 306 Temperature versus electromotive force ,
,
273-276
table, 6
Temperature measurement 1-14
,
Silver, properties ofj:536 Silver alloys, 536-537 electrical contacts, 543
residual stresses, 3oti-343 scale on, 288 silicon, 361
pfecipitation-hardening, 374-376
from heat treatment, 336-343
jl24
plain carbon, composition of, 246 quenching, 283-287 . '
martensitic, 366-368
Residual stresses, 653-654
'
Solidus line, 158 Solvus line, 182 Sorbite, 309 Sorby, Henry C, vli
austenitic, 37 i-J74 classification of, 362
Recrystallization, 131-135 temperature, 132-135 Red-short, 438
9
hot shortj 248
Silver Solders, 537 Sintered carbide tools, 415-419
Stainless stee'
6-49
Raoult's law, 172
Silicon steel, 361 Siliconizing, 573
V*-,.
hardening temperature. 281-282 heat, treatment, 249-345 y ..;
Sphetulitic iron '-9e Nodular Cast Iron)
Radiation pyrometer ,,)0-11
Red brass, 469
60&-613
test for, 297-299' 5 ,'ix
cast iron) Spheroidite, 254 Spheroldize annealing, 252-254
R Monel, 510
Sauveur, Albert, vli Scale on steel, 288
s, hardenability: 8peciflcatloniy301-302
'
Quantum numbers, 69
of
,
hardenable carbon, 043-345
factors in, 152-153 interstitial, 153-154
radiation, 10-11
'
;
Steel
Silver alloys: silver brazing, 537 silver-copper, 536-537 silver-copper-zlnc, 537 Silver-bearing copper, 462
alloys, properties of, 168-170
thermoelectric, 5-10 Pyrometry, 1
|n
I
Solid solutions,
optical, 11-14
49 150
i
,
248
by color 1-2 by pyrometers 5-14, ,
,
optical, 11-14
;
radiation, lO- -ll
recording and controlling 10 ,
,
r
.
Temperature measurement, by pyrometers:
Tool steels:
thermoelectric, 5-10
distortion 393 ,
by thermometers, 2-4 gas or vapor pressure, 3-4
failures, 410-412 hardenabllity 393-394 heat-treatment 408-410 high-speed, 405-406 ,
liquid-expansion, 3 metal exparjsion, 2
,
(resistance, 4 Temperature scales, 1 Tempered martenslte 311 Tempering:
.
,
alloying elements, effect of
,
shock-resisting, 400-401 special-purpose, 407-408
353-354
secondary hardness, 354 steel, 305-313
toughness, 394-395 water-hardening, 397-400
effects: on microstructure, 308-311
on properties, 305-307 of time and temperature, 311 Tensile properties, 38-42 breaking strength, 40-41 ductility: elongation, 41
Torsional fractures, 659-661
Toughness, 42 at low temperatures table, 564 ,
of tool steels, 394-395 Transducer, 66 Transition carbide 308
reduction in area, 41 elastic limit, 38-39
,
Transition temperature 560 ,
modulus of elasticity, 41
Troostite, 309
proportional limit, 38
Tungsten:
resilience, 42
'
electrical contacts, 543-544
toughness, 42
properties of, 542
true stress-strain, 41-42
steel, 361
ultimate.strength, 40
Twinning, 116-118
yield point, 39
Twins:
yield strength) 39-40
annealing, 118 deformation, 118
Tensile test, 37}-42 Terne metal, 51(9
Testing:
Ultimate strength, 40 Ultrasonic inspection 54-57
|!
creep, 45 ; fatigue, 44-45
,
Uniform corrosion 590 ,
hardness, 24-r37
nondestructive, 46-62 impact, 42-44 Thermocouple, ;5-8
/
construction,;;7-8
A
,
,
Vibratory compaction of powders Vickers hardness test 31-32
,
617
,
:
protecting tubes, 8
Thermometers, | 1-4 f
Thermometry, "I?
,
Thermopile, 11i;
J
Thomson effeci:
Tin alloys, oza-Jia1 composition and properties of, table, 522-523
Tin bronze, 472-475
Tin pest, 521 Titanium, 524-525
Titanium alloys,: 525-531
alpha alloys,lj526 i
Vacancies, 96 Vanadium in cast iron 458 Vanadium steels 361
''
materials, 7
i
hot-work, 404-405
machinability, table 396 mold, 406-407 selection of, 392-393
,
i
i
i
'
«96 INDEX
alpha-beta alloys, 527-530
beta alloys, 530-531
composition and properties of, table, 532
Wear of metals, 567-581
.
V
factors, 569-570
\ i\
mechanism, 567-569 protection against, 571-581
\vhite cast iron, 424-426
microstructure, 426 White metal, 520-521 -Work hardening 123 ,
Wrought iron, 226-230 composition of 228 ,
manufacture of, 226-228
microstructure, 229 nickel, 22»-?30
properties of, table, 229 X-ray
I
89-90
diffraction for radiography 46-47 ,
X-rays
,
Titanium martepsite, 529
Yellow brass, 467
Tobin bronze, 470 Tool failures, 4t0-412
Yield point, 39 Yield strength, 39-40
Tool steels, 38''-421
A1SI classification, 387 brand names, table, 398-399
cold-work,
1-404
comparative|properties of, table, 394-395 decarburizat|on, 397
Zamak-3, 533 Zarnak-5 533 ,
Zinc alloys, 531-535
composition and properties of, table, Zyglo, 51-54
i
I; '
i
A
%
:
'
>
.
