_)(51 JOURNAL 01'
THE CHEMICAL SOCIETY. ~ o mmittu
H. E.
;
4
ARMSTRONG,
Pu.D., F.R.S.
of
~tthlicntion :
II. F.
M.A., D.Sc. Ph.D., F.R.S.
}!ORLEY,
E ..h' KIN~ON, Ph.D.
w.
A.
.T. :MILLAR TnoMso:o<, l<'.R.S .E. 'L'. E . 'l'ROHPE, PILD., F.R.S.
Bnow!(", J>.Sc., F.R.S. R. Du:-
WYNDIIA.ll
RAMSAY,
,V, P.
WYXNE,
B.Sn.
Qt ~ i1D r ;
C. E. GnovEs, F.R.S .
.~nh-Qt hit or : A. J .
Gm: ENAWAY.
j
!
·~
Vol. LIX.
18tH.
~
-
.
TRANSACTIONS.
- - - - - -·- -- -----···-- - - --
~
LONDON:
G U RNEY & JACKSON, 1, 18()1.
PATEH.NOSTEI~
ROvY.
:1H
THE OSMOTIC PRESSUHES OF SALTS I:-i SOLUTION.
345
I describe the parts in detail, as they are the pt·oduet of considet·able experimental work. X LI.-On the Osmutic Pressures of Salls ·i n Soluti•JII. Hy H. H.
ADJ ll,
i\1.A., Trinity CollP.ge, Cn.mbriugc.
IN a p ftpcr in the Ze·if. phyo"ikttl. Ohern., 1, 481, Vnn't Hoff discusses tue analogy between t.be OSlllOtic pressure of U. S1dt in solution anu the pressure exerted by a gu.s on the walls of it.s containing vesseL The numbers upon whieh he uases the verifieat.ion of his calcuhltions were ohtnineu by Pfeifer, in his Osmot·i:;che r; .. te?·suchw1gen (Leipzig, 1877), aml only ernbrace the figures for· a fe w su.lts, viz., potassium nitrate, poLassinm tartrate, and for the coUlplex organic substances cane sugar, and gum ambic. These were the only independent measurements of osmotic pressures in absolute units, until A. Ladenhurg pnblishl'd (JJ.:r., 22, 1225) a few numbers for tho osmotic pt·essures of solutions of UL·xtrosc, resorcinol, cane sug;11·, and "saccharin." It is nut easy to get n,ctun.l measurements of osmotic pres!' ure~, but n. number of comparisons of tile osmotic pressures of uiffcrent salts in solution were maJc hy J)e V riea (P·rioysluriu~·s Jahr b. , 14 ; Z eit. physikal. Chl'm., 2, 41:1; 3, 1U3), who fouml the conceutration~ of isotouic solutions of these salts, by means of the exosmotie coutmctions whieh they bring about in the protojJ ia.smit: lini.ug membrn.ne of living cell s . lu very dilute solutious, the substances examined appeared to obey the laws of lloyle, Gn.y-Lassac, and Avogadro, but in the case of strung solutiunA, the deviations wcr~ considerable. At the ;mggestion t•f Professor J. J. 'l'h um~on, I took iu hand the · Jircct measurr.ment of osmotic pn:ss ures by Pfeift:r's method, but the length of t.iu1c t·cq uiretl by eac h experiment rcuuers the accumulation of results very slow. '!'he metl1otl consists in Jcpositing a membrane in the subf
FJO. la.
Fw. 1.
:'
;~
41-
4a-
'
.
.:. i
:l {\
I
D
!
c
A
n' £
:i
~
~
~
;
-8
I
~I
~
ll
-!.
t The Pots. The only porous pots found to givo any consistent results are specially ruade of biscuit porcelain, 3 incht's long, l inch external diameter, and -y'6 -inch substonce. In to the open end., a tight ly fitting glass tube E (l!'ig. 1), Jrawn off at one end., is inserted and fixed for a length of about linch~ by Realing wax. No wax is put on the outside, lor experience tihows that, in this cnse, the membrane is fol'med irregularly between the two layers of wax. This is shown by l!'ig. 2, :.:!
c 2
346
ADIE ON THE
OS~!OTIC
PRESSURE!;
347
OF SALTS IN SOLUTION.
' 1·
which is a section of the wall of s uch a pot. As the membrane generally gives way at the top (perhaps the pores of the pot are coarser there, due to the method of manufacture), it is essential to give it a good connection with a solid wall of wax some way down the pot.
~.
.
FIG. 4.
FIG. 3.
I(~ ,~
·-----
Wo.ll of glass
t11l-t~.
Glaze.
d I
j
FIG . 2.
f,
\ ___ .. .. W:u:.
Film.
Wall of glass tube.
!t
Wax.
,. f
Gln1.e not
backed by film. - · .. , _.. _ Film.
Wo.ll of pot. Film.
Some other methoas ·were tried for getting the top Recu re: (i) sopporting the upper end of the membrane by melting paraffin wax into the top of the pot, after a prelimina1·y formation of the membrane and drying tLe pot; this did not give good results; (ii) having the upper ends of the pot glazed with porcelain glaze, but as the glaze could only be applied superficially inside and out, instead of th roughout the mass of the bisc llit, the membrane-forming solutions did not diffuse evenly, and the film was formed irregularly, and even locally, between t.hc two layers of glaze, and when a high interual pl'essure was maintniucd for some time, it forced the solutions through minute c1·acks in the glaze. This is 111iown in section in Fig. 3. Messrs. Morl ent l!'ret·cs, Bn.yeux, kinJ.ly tried to glaze some throughout, but did not succeed. 'l'he 1-i uch long sealing wax joint withstands pressures of to 4~ atmos., depending on the tempemtut·e, and at the same time prevents tlu~ merubraue from approaching tir e top of the pot, since t.he outer solution has Jiff11sed over ! inl'b downwards from the top, before it reaches the inn,.l. solution. This is slwwu iu section in Fig. 4. 'l'he dimensions of thf' pots allow all t.he conuections and fittings to be made of glass. Up to 1Le present., scaling wax is the only cement with which I ho.1·,. been able t.o get any r es ults, with one exception.
3t
For temperat.ure experiments it. is necessary to use another cement. Portland cement iR too porous t.O form a strong film, and the free alkali in it decompose~; the copper sulphate solution used. Th e pressures obtained with thifl cemen t are only. about one-half those got with sPaling wax. The double phosphate of zinc and sodium, and the oxychloride of zinc both make good cements, but only become hard when the zinc oxide used is prepared by heating the nitrate . 'l'his oxide is very hard, and t·equires careful and l)rolonged grinding before it is sufficiently fine to use. When made, the cement sets >cry quickly, and iR difficult to manipulate. However, the only temperature experiment I have been able to complete was made with a pot fixed with the sodiophoRphate cement.
~~
.' i ·1
l .~
i
t
·~ 'I
J
1 I
The llfem brane.
Pfeffer in his researches used a membrane of copper ferrocyanide formed by precipitation from 3 per cent. solutions of copper sulphate, and potassium ferrocyauide, when these were allowed to diffuse from each side of a porous pot which ·had been immersed in the copper sulphate solution, and partially dried. He also experimented with membranes of calcium phosphate, ferric hydroxide, f'mToas carbonate, and ferric ferrocyanide, a11d found that th ey did not prevent the passage of salts so completely as the copper compound. I also tried the same membrane and, in addit.ion, aluminium hydroxide, and silicic, and tungstic acids, but could not get satisfactory results with the commercial pots, e:~~cept with ferric ferrocyanide. The specially manufactured pots gave excellent results with the copper salt, and so SA.'I'ed further experiment.
} ~
''
1
t f
I·
348
ADIE
0~
OF SALTS IX SOLUTIOX.
THE OS:IIOTIC PRESSU RES
'l'be cl1i e£ ohjPction to th e f errocyanide film is that the solut.ions which surround it must be neutral, or at any rate not dissociated* in solution. On the othe1· hand, the d estruction of the film is an excellent, but trying proof that a salt is dissociated in solution.
il-l~
k<-cp tl1 eir maximum prcsRures constant for a longer time than t be nitrate. H a stror.gcr Rolntion is used :tt first, th e film very often bursts. Oniy about 25-30 per cent. of the pots se t n p g i vc sati;;factory r~sult.s.
Pot s wlrich gave r eR olts n early a mnximnm on tl1eir trials were
Th e Gauge.
u~cd for testing Hoyle's law in so lutions . lwt only those which gave
A simple form of manometer was found to give g-ood results, since it could be aHached to the pot b y the help of a blowpipe, n.fter the film was made (Fig. 1, p. :)45). Cis a capillary tube of uniform bore about 4.') em. long. The rest of th e t.ubing is about 131i' inch bore. The h eights were read sufficiently accurately by the millimcti·e scale D.
th r ma"imom were used for oLscr vati ons which had to be co mpared with th ose fr om other pot:,;. vVhcn not in nse, the p ots were fill erl with. anrl stood in, distilled water. '!'bat this docs n ot affect the film was shOI\'11 by the fuct that a p ot (<' ltn!'l ~") which was k ept stand in g in wo.ter for about. a .r en t· gave pr:u.:tic:~lly the same results aftt·r· th is iuten•al as it did
Mode of Preparing the Membranes and Setting up the Apparatus. At first I tried Pfeffer's method of depositing the membranes, but could not get a stl'ong film by it. The following, based on a suggeHtion of Professor J . J . Thomson's assistant, Mr . K Everett, was, however, found to giYe excellent r esults:The clean nod dry pot with its glass to be attached is gradually lowered int o the copper sulph at e solution, while the ferrocyanide solution is b eing poured into the in side. The pots at·e theu placed under the air pump and k ept in a vacuum for some drtys till all the air i;; drawn out of their pores. Th ey are then allowed to stand about three weeks in more d ilute solutions, to thicken up any weak places in the film . 'l'oo long standing in a nwuum is apt to cause a precipita.tion of copper oxide on the outside of the p ot, and thi s seems to interfere with the free p assage of water. The film thus formed at the first contact of the two liquid s is rigllt in the middle of the pot wall,;, and standR pressures up to atmos. Wh en ready for use, the pots are attached to their g auges by the blowpipe. It is easy to empty the p ot.s through the dra wn-out tube A, by means of a piece of tube drawn out into a long ca.pillary (Fig. la) , and a s uc tion pump. They are filled by a wash-bottl e, or pipette with a capillary deliv ery tube. In order to select tire pots which can be used, they are filled with a salt solution which will get up a pressure of about 1 atmo. A solution of potassium nitrate 2 \, gram mol. per litre does for thi .-; purpose, but solutions of -41ii gram mol. of potassi um sulphate per· litre or of •'•J" gram mol. of potash alum per litre are better, as they
4t
• Throughout this paper th e term " di s~oc iation" is used in its ordinnr.v of dissociation into constituents. Wh en dissociation into ions id meant., the words " into ions" ha> c been inse rt ed in the text.
n~ceptation
hdo re.
The
O~sen·alions.
Owing to the thickness of t,he pot n.nd th e di sb~n ce of tile membrane from thfl inner surfa.cc, washing, &c., whid1 requires the help of diffusion, tak es a long- t.irne. A~ t.be solutions USL'd a rt! vel'.)' rlilnt.e (about o·.j-8 per cent.), it i;.; necessary to remove any excess of salt left in the pot 11·:t ll s by the lust ~olntiou. 'l'be pot is fir~t thoroughly was hed witlr distillt:d wn.tet·, iu order t.o r emove any salt solutirm ad lw riug- to the siues nf the pot ami tubeR. but to extract tbe sa.lt contained in t.lte p ores it is necess<~ry to ti ll the pot with distilled watc1· and
s mall percentage of salt in tl1 e pores causes a current of water throngh the tilm from tl1e outside, w}u ch asHists the w11 ~ hing as well as the diffusion. This treatment is g enci·ully 1wtlicieut to removt' traces of the la.Rt ot:cnpnnt, though it i~ ;ttl visn hlt• to wash fut· :l- :J, dH_\'~, ·wh en tirue a llow s, and the couseeuti1'e solutious !1 l'C not i~ o to uie . (~·or e:l'am pl .;R Rec I.) · 'l'h e pot. is now filled with the n ew solution, allowed t o stand for 24 hours, and t hen r efilled. \.Yh en the tempcratul'e is t:ons t.ant., read ings are takeu of t!te top of the air co lumn ofg
·l
:-350
I
I.
i
I [
II
l .
!)
I: :1
.1
1. j ~
~I
ADJE 0:-.1 THE OS:\IOTIC PRF.SSURES
When C'Onst~-tnt. , the din1inishPd volume of the air, v, in the ga.n ge is read, a11d t.he pr:essure calcul~tted hy Boyle's law. p., is ob tai11ed by COJTecting the atmospht>ric pt·essurfl for the differcnee of level of the mercury in the two limbs of the U -tube, aud also for the column of solution in A. p 1 is conect.ed for th e difference of the mercnry levels, fot• the atmospheric pre~Rure on the water suJ·I'Onnding the pot, nud for t.h£: iliffercnce of level between this water and the mercury column in commullication wtth the pot. Tbe hiRt correction may be obviated by making the wate1·-level of the veRse! in which the pot is immersciOciated; socii urn J,itrate gradually attacks the film. Potassium, calcium, ammo11ium, and aluminium sulpb11.tes are not dif:>sociated; sodium, magnesium, and copper snlphates are dissociated. Potassium ferricyanide is dissoc:iated, whilst the cob11.lticyanide is not. Potassium hydrogen carbonate gradually we~tkens the film, and trisodinm pho~phate, as might be expected, dest-roys it at once. In fact, it appears that the potassium salts of the stt·ong acids are the least ·dissociated, whilst the sodium salts of these acids are much more generally dissociHted. '!'his might. almo~t be expected from tl1e position of :;odium in the periodic :tl'rnngement of tile t-lemeuts, in the futme suction of the first g1·oup as l1ydroge n, and fl'om tho behaviour of aeids towards the film, but at the same time it is possible that the action is not due to simple dissociat.ion, but to a specitic action of the sodium salts, C11used by their being relatively less completely saturated than the correspondmg pota~sium salts. l£ tLis be tl1e c:Lsc, there ought to be some difterence between the two in their action as soh·e11t~ of more complex compounds. The only caHe which suggests it.self is the incomplete precipitation of tLe paraglobulins from solution by means of sodium :;alts as compared with other neutral salts, though I am not aware of any compa1·ative trials of the pot~ssium salt:;,
OF' 8.<\LTS IN
;)5l
SOLUTIO~.
The dest.r uction anrl hre~king down of tlte film may he eitht•t• sudden, or gr~tdual. The gradual breaking down is shown in t,he lower curves of K ., SO, in p 2 , in Fig. 7 (p. 354), and of potnsh alnm in -1, in Fig. 15 (p. 061), where the curve takes the ~Rme fonn as in the case of known diQsociation of chrome alum in ~,ig. ] 6 (p. :·362). The phenomena of dissociat.ion of simple, and of double ~nits in solution has been investigated by F. Riidorff (1Je1·., 21, 4, :3044). H<: :;h0ws that many of these are dissociated, and may be sepnmJed i11to their constit.uPnts by diffusion through a porous cell into water. In the following tables, the temperature at which the o~motic pressures were taken varies from 15° to 19°: this does 11ot i11troduce any appreciable error in tLe pressures est.imated in atmospheres, :t1ul in the Boyle's law observations t.he variation of temperature of tlw eonsecuti ve ohserva tions is less than ~ • The quotient. P/71', wLieh is the same numericn.lly ns Arrheniu,.,' coefficient of
_l_ g-ram-mol. per litt·f! = 0·14 atmo. ltiO ~ ~tml
for
1
n
X
lli~
"
"
=
n
X
0·14 atmo.
This should be conected for temperature, n.nd at 15° becomes X (\·14-75. 1t may also be co11venient t,o t·emember that nt 15° a pressure of 1 atmo. would be produced by 0·424 gram-moleeule pet· litre.
J/.
I. Boyle's Law applied to Salts in Solu11:on. In the following tables column I is the strength of the ~olution in formula weights per lib·e; colnron II thtl osmotic pressul'e in >ltmospheres; column III the quotient P j ;r, or ·i, the cli:;;socin.tion coefficient. The results will be found plotted out in Figs. 5 and 10. gt~am
..... .,
ADIE 0:\ THE OS:\IOTIC PRESSUilES
;);)_
OF SALTS
I~ S OLUTIO~.
b Solution of NaNO, in different pots.
Sl•lut.ion of KKOa in
Fi~ .
•'J.
~
1
r--+ L
:::
·
! 3
u. ''·
2 ·87
0·(16
40
1 10
2·:w
1·07
I
I"!"> C.~
J·:u
1
1\n~O .,.
] ·;)9
0 ·89'
i
.s 2 0 :::
;;.
0
"/
v/
/
vb.;
.l~~v Vi /~ I~
- ~~
,,
i
.
1
I I. I v
v
i
v
I
.
li
1
I' I
lI
r, G 4 ll 2 C'onrr ntrnt ions in 4'~ grnm form . wt.. per lit•·r .
i
H
• 'J'h ~~e fi g nre~ were obtoinrd for tho nitrate wl•en used ns th (' firs t so lution, fvr ryinr: the pots. llence thr,r differ morr than the others.
~~
I
''i1
I I
:;.j .. I
iJ
§
).i'
:;::
1\
.,
·\
_Lj
JJITl i i I I FI I I I 1-tl
""VV l1 vv ./'/
1
1
1
1
I
'·11
l
I
~'V.fll I /V
·~j
:::
\f/ V ~"~ ~/ ~
~
t
· ~/ · ~'"> ...
i
~
r-1 hf I I .~
1-- 1
· '
5.
~'vt-~;·
1'
FIG. G.
l ·GG
0·4G(i
'
~i\
I ·Gi
1:1
u
}' ro.
0·60 • l\fea.n of 1·79, 1·59.
r
~;j
I
1
Mean of 2·39, 2 ·41 , 2·3';'. 1fpa n of 1·7o, 1·48. 1 :i8, l ·G1, 1·M), 1·G4."" Mean of O·!Jti, 0·82.
,
}·.jl
1 7·5
i
! ~
1"()94
20
r--: : 11-r--r--r--r-:---,----,~~~y ~i I ! 1 V / ~ a'
1·:39
1•04
~--
1
:n1
4•!0
h\"0,..
t
1 II)
1
su ~
III.
-· 5 ( n<'O.rl .r)
1 40 1
2
II.
IT I.
2 tJ
1
I.
IT.
I.
Fig. (;. ~··
1. 8imple Salts. Q.
:l .i :)
2
3 4 5 -.!6 grant form . wt. per litr!'.
fi
7
!-1
'.
fl. '.
I •. ,.!'
.t ~ ·.
Snlur.ion of K ,SO,. Ji' ig. 7 ( p. 354). f'<·1·i ps (i ) in one pot p, . ln ord er starting with strongest.
c.
I.
IL
,'i flU
JIT.
1·895
1·:~.-,
4
1·68
tiO
U>O
1,; ·:. .,
'• •!i i'
~
80
] ·38
} ·(I;";
1·02'
1·82
0·505
1·8 1
,r ~
I ·'"
2
su ( 1 80 1
Mean of 1·05, 0·99_
I ·~,
· ~
'1
ADIE
ox
TRE 0Sllt0TIC PRESSURES
1\/W.
OF SALTS JX SOLUTIO:\',
Fr a. 7.
f.
Solution of KI iu one pot I.
8
80 7
.,;
§
80 6 80 5 80
""1
,,
• 1
3
a
,.'0
I;
4 5 gram form. wt. per Jit.J•e.
II.
81)
2·82
5
.. ,;
I
4 HO
:~
'/
HI.
~!
219
2
so
1·01 (J·(i6
80
d. Solution of (NH,),SO. in one pot
t
It
:::o
2·G4
1
81J 1 80
•
40 Solution of CaSO, in one
.!.J_~ 80
2
pot"~·
n. .f.ti7
()· •
()•574
~0
~N
~~~
2
J;l ~~·
1·5:2
~ ;~
2·21
1·57
1·77 5~
1·58
.
:[
1·64
0·4~(j
1 •77
.~ ~
If, = 3·31, ( 1 = 2·.:~ . 3 c = 1'<31.
2 '01
KI.
2 ·0:3
Fw . 8.
·U ff
4·
I
:'1 ~
:2 ·;3!:)
-<
r +t;: :hI
2
2·::13
ll
2
This is a saturated solution.
I
;!.,iJ: i•.
;. :!'' .
ITT.
j '' .• 1. f ·
1·.)9 1
0·245
t
~.~
)r
ITt
1·6i
2 1
00
3 4 5 gram form . wt. per litre.
G
I
~r
I ·G l
092
t
j·~ '::,:I~
7,
1·30
r.
0
2·36
1
1:'.
~i
f:
] ·81
1
H
I 5G
, ,
] '9ti
1·705
80
.f
3·0;)
. ~j.
2
:~
'f
l'.'iO
2
8 1
T.
nr.
a·371
1 ·3[>
8u
7
Series (ii) in one pot [ 1, in following order: 2, 4, 1, :3, 5 x "
·' t:rr .{'
i:'
6
If.
Jl
4
.ll
o,. Fig. 8.
2·5()
80 1
X,l),')
7
.I!'' ;
8
I
. ·i,.: ,~..,,
'' li'
. ,:3 ·f
:t
''. 1~:
();"J( i
ADIE OX THE OS::\IOTIC PRESSURfi:S
OF SALTS IN x,~~HPO, .
:i:l 7
SOLUTIO~.
F ro. 10.
2. Hydrngen Salts. n. Solution of KHCO, in one pot pz.
0 t)Q
4
Fig. 9.
JI.
III.
2·0:{
0·01
I.
l ·O P
ov
0·90
·id
!
·~,n
:.1
I"
7: 2
j'l
~
);I
~;
~
2
0·59
8V
2
1·05
~!
•b•I
.., >!
1
Mea.u of 0 ·95, 1·07. Duplit at-c from o = 1·:12. ~ Duplicate from o = Cr65ii.
::'~I
1 2 a 7I'u gmm forn . wt. per litre.
lfiO. !l.
c. :::lolutiun of Na"HCi in une pt t
1\TT Gn.,.
,.
·,'~'i
:.1
,oi'/
._!/ ~-,v 'It'
.,
::
~
.;; ~
1
~
(//
~
.~
t~ ~
l .t·~
l)J ~
~v
v
·1
i. l
·I,,
rr". .Fig. ll.
I.
If.
I H.
l 20 l .J.V l
-±·:)2
:3·86
:!•1:-:l
:ng
1·~:3
4 •40
~0
X u..21ICi.
' .!
·'
::j
-~ I
.. ,I
'I
}IJG. 11.
vY 2
·1
G
8
,!,
, •., gram form. w t. per litre.
[,. ~olution of Na"HPO, in ore pot v. I.
4
BO
-•>
8v 1 t!V
Fig. 11) (p. 3iJI).
II.
IIL
1·82
1·62
1·50
2·GS
/i.
~ 3
.§
.
~
:
2 :,
l
1
O·.'J1
I
1·40 2 3 0'0 gru.m form . wt. pLr litr~.
4
:~o8
OS~IOTIC
AlliE VN THE
PRESSURES
OF SALTS IN SOLUTION.
3. St.tlts of Cumponnd Aci..Z;.
a. Solut.iuus uf K~FeC~N6 • Fig. 12.
f, wot·kiog uowu from strJng to weak.
b. Solutions of K.Co,C, 2N,,. strong to weak. Fig. 13. II.
H.
. r.
I.
8 1tiU {i
1ti6 4
~
'
4a
0;: .
Me om.
lli. For weun.
:N9
-
3'38
3'44
3•07
4 160 3
2·-±9
-
2•65
2•57
3·30
lliO
2·05
!{ ;{
H>O ~·(; = ~ ltiO
tiu
1·4!)
2·;, 1tiV
2
1 160
o·;,u
1·78
l ·Gt:
1'\)2
3·42
-
1 ·GS
3·64.·
-
-
1·4:)
3 ·!)~)
1·27
-
]·27
;~·G4
16v- oO 2 Hi5 1 160 0•5 160
0·9.5
:1·40
KsCo 2 C 1 ~N 1 ~.
0·50
:l
-
0·0H
0 4ti
O·ii;)
0·!10
}l;(i
-
l
71'
working down from
2
HI.
r.--~.
I.
2·6
1tiU
In two pots,
3MJ
Mean.
,-.A.-, 0. "~·
3·46
3·28
3·37
6·1!)
5•85
--
2·!9
2·49
-
5·95
l ~!
-
2'17
5·82
1•77
1•70
1·74
6·35
C·OS
0·91
0•90
0·91
6•50
(J-43
.;~
0·515
0·514
/·:31
7·36
j
2·17
0·512
·, ,I :) <
;,;
~
1 :1
FIG. 13.
~:
:,~
4
:K, FeC,,:\,;.
F1n. 1:1.
--r
j;fY
4
3
-~·r
E
~ ~
~
~
ttY!I II II I J~ ..~ _Jt·~
j .. ';./]
u:·
3
"'
,,«1
-""~~~
1
~dz1soc 2
1
l~l~~
.~~~ ~ .J?/1 I I I I I I I .-l--t-x=PTI I ~ 1
2
'4 3 TJ v gr>IUt form. wt.
5 J:ll'L'
G
7
1
2
r; 3 4 Th gram form. wt. per litre.
c. Solutions of K(SbO)C~H~0 6 • dilute to strong. Fig. 14.
In one pot
G
71'2 ,
8
7
worl;ing up from
!)
litl'l'.
VOL. LIX.
ll
360
ADIE ON THE OSMOTIC PRESSURES
II.
1·35
1·20
1·03
1·2:3
0·747
1·3!1
I.
8' 160 6 160 4 160
,---A---~
I.
160
1•84
;~
3
4
5
G
7
8
l
'·
~~
2 160
-
1·08
-
3·8G
0·61
-
3·63
-
.·,
-
-
-
-
:'l ~j
0·37
-
4·39
-
·~0.')
-
~·7:2
3'.50
-
~ '
..'
(~
I)
·'..
~:' ,i
.:!-)
,,
.I I
'l'his is a nearly saturated solution.
l\2AI 2 (SO, ) 4 .2-1-Aq.
2
~
-
3·G4
!l :
1
1
-
-
2
1
t
:3·6.'>
ltio
lUO-
$::
.;;
--
1·18
1 166 0•602
3
2·04
1•56
Too-
~
:3'44
-
1 4
-
'I
160 2'41
1\:(SbO)c.u.oo.
-<
-
ltiO -
Fro. 14.
III. r--"---,
2•32
160 0•616
This solution is nearly saturated.
;a·
I'·
4'82 T6o
1'14
0·32
3t>l
II.
4
~
1
0!" SALTS IN SOLUTIO:-i.
III.
..1
..
Fro. 15.
.I '
4
Tlru gram form. wt. per litre.
4. D ouUle Salts.
a. Solutions of K 2 Al,(S0,),,24 H20. In two pots, ~, and fa, wot·king from strong to dilute in f3· Fig. 15. II. ,--~
I.
6·023 1
l606
160 ;,
1UO
I'·
lia·
:1
3
1/
1I;
,,'t
-< 2
.(
i· ..J:;.
III.
,----"----, 1
3·20
2·7o'
~
'it~ I
3•86
4 ·00 1
2·39.)
3·85
~~
: !f'
2
3 4 5 1 ;};; gram form. wt. per litre.
6
: .~~,,~
7
~ ~
;
,:: 2
D
2
:; ,
,.
31i2
ADIE:
~.
0~
THE
OS~10TIC
Solutiom; of K 1 Cr,(S0,) 4 ,24Aq in two pots. ,-----"--,
5 160 4
2·10
-
2··46
--
Green'
:3
2·08"
-
1·62
16iJ
,
-
2
,-----,
-
Green
4·40
-
t!
1·3:~
-
-
3·96
I
0·47
-
3·3G
FIG .
'I
~j
3·17
~·t
'I i,
: ·~
StLtUrt~.tivll
16.
3
9
2
1
--riro
4 3 5 gram form. wt. per litre.
6
~l
JJ
4
::
:~Ij
8·72
'l'hc green solution is prepared by boiling the ord inary bluegreen solution, and is in a condition of dissociation, since it attacks th e tilm. 2 'fhis observation may be too low, since it followed (3), which w£>akened the film. K 2Cr2 (S0.).,24Aq .
d
'n
::3-86
I·lO
1
solution at D, dilution dirninishes the osmotic pressure until the point Cis reached. Frorn C toll, the curve produced passes through 0, and may rept·esent the dur.Ltion of existence of some complex
...,
1
;.:"
31)3
8·85
2 160 160
,;
OF SALTS IN SOLUTIO~.
FIG. 17.
lou
.,
Fig. 16. liT.
II. J.
'l
PRESSURES
7
8
The examination of these curves leads to the observation that they are n early continuous, and take the form OABCD, Fig. 17. 'rhis ma.y be divided int.o four parts; starting with a satura•ed
1
point.
molecnle of the salt, or of a definite hytlr-ate. The part CB may be long or short, and above or below the line of gaseous p1·essure of an undi ssociated molecule. At B, the specific action of the sol vent may be supposed to begin dissociating this complex molecule of the salt. This goes on increasing through A towards 0, but does not puss beyond the line of complete dissociation into ions. In mrtny cases, dilution does uot appear to make this part of the prcssut·e curve approa.ch the line of complete dissociation into ions. The portion OABC may be well seen in the KN0.1 curve, but in the K2SO, and the KI cutves, only the portion OA, and a little beyond A towards B. The port.ion CD is well seen at the more concentrated end of tue alum, and better in the tartar emetic curve, both of whieh are nea1· ly saturated. OA.BC is nearly straight for KI, for the double salts, and for the salts of the compound acids. In the case of alum, t.he change of direction at C takes place at the concentration T~u gram for·m. wt. per litt·e. Wh ere the part OA is "traight, it cuts the complete dissociation 1ine wheu produced backwards, at a point between 'g'~o anti · h coucent.ratious. 'l'he presence of hydrogen in t.he acid snits does not appear to have any specific e:ffed on the form of the curves. As the c ur·ves are so nearly continuous, we cannot infer the existence of definite hydrates in t he soluti ons of the salts examined. Fo1· t.his purpose, one must get a larger mnge of observations, which can on ly be procured by some less lengthy method, such as the lowering of the freezing point. The part DC shows a loss of kinet.ic energy ou tlilution; this may be due to the 1·etat·dation of tho ruolL·<.:Ltles iu the state in which they
:.1' .,
:.,_
j'
j i .,
/:; :,
;) j) ' J ·,
.~
3!i4
ADIE ON THE OSMOTIC PRESSURES
exist in !>olution, being greater thnn that of those which are ju::;t entet•iJtg the solid state.
OF SALTS IN SOLUTIO~.
a . Sa.lts which cowbine to form well-marked double salts :In pot
Salt solution.
£1m .
1 _ (NH,)~so~ . • .......•
Osmotic Sum of pres$ure. compo11ents.
.).j,O
~~~ K~FeC 6 X6 gave "
.,
a pressure of 0·77 atmo.
"
l"l3
"
Assuming ·Cilarles' law 11pplies to salts in solution, and that K,Fl'GsNGbehaves as a single u.nit, the pressure should ha.ve been 0·77 x 327 = 0·87 atmo. <:!90
Tlutt i8, the calcu,lated rise on this assumption
= 0·87- 0·77 = 0·10 atmo. Rut the obserred rise
= 1·1:3- 0·77 = 0·36 atmo. Now, on reference to the preceding results, I, 3, a., we fintl tltat the ratio of observed osmotic pressure to t.hat calculated on the unclissociated molecule hypothesis is about ::H:i5* for the above solution. 'l'herefore, the calcnla.ted rise, instead of being 0·10 atmo., should be 0·10 - ::Hi5 = o·:16;i atmo., which is the value fouud . That is, so fu.r as a single experiment C!l.n uphold the point, Charles' law does appear to apply to salts in solution. It must be nott·d that the factor 1~, w batever be its interpret~ttion, appeat·s here as well as in the application of Boyle's law.
II I. Dalton's Law of Partial Pteswres applied to Salts in Solution.
1 have arranged the values obtained under three heads:-
0 Htlloti!J prc•surc.
. A. .-, Sum of
,)
compon~nt• .
!!
1 ·264
-
1·295
;:! ...
1·~6;)
-
H!2
.~l
I K~sol· ............. 40
1·2!)
-
1·40
__!_ (NHt) 2 Al 2 (S0,) 1 ,::!4Aq
2·37
::! · 5;~
1·98
2·39
2·1)6
40 1
40
Al2(S0,)3 .. . ........
40
At 17',
In pot r .
IC.
, - -..A. .-----...
If. Charles' Law appl/etl to Salts in S olution. In t.his case the difficulty is to procllre a cement which will allow the formation of ~L film, but which will neither ct'ack, not· melt. I found that scaling-wax joints will not allow the use of a temperature above :)0°. Using the sodiophosphate of zinc, I have got one observation . Unfortnnately, th e solution used, potassium fcrrocyanide, attacked the film at a tempt:rature of 70°, and I have not been able to get another
3G5
1
- K2Al 2 (S0,),,24Aq ....
40
~r
~ ·~
2·515
J
·; 1·9G
2·32
b. Salts of the same acids, which do not form double salts:i . Iodides in pot o,:-
· l·
. .' j
Osmotic pressure.
Salt solut.ion.
_!_ KI
0·9::!
-1 Cait. .... . . .... .. ..
0 .65
80
1
KI
+
:o
Cal, . . . . . . .
u. Sulphates in pot
Ul:J
'j I
1·57
+ .·:1
r
" . I!
o~ : -
Salt solution.
Osmotic l'ressut·c.
40 CaSOt .. · • · · · · · ·
0·4G7
0'
cowpone11ts.
·. f·
40
40
Sum of
574
Sum of component.s.
-
r:. '~'
_!_ K 2 SO, .•.. · · · · · · · · · · 40
1·4:1
+ 40 _!_ K SO,
1·88
0·.574 CaSO - .':t\) -~
2
.;·· l·SU7
c. Salts of diffe1·ent acids which do not form double salts: -
. ,,I
i .l ; :j : ~I t '•!
. j· ·'
. ·.i
• The factor, 3·65 is tlmt found for the above solution at 17°. De Vries ha~ shown (foe. cit.) thnt the rotios of ·i arc nearly mdependent of tcmp e t~•ture.
"t
.
'
3()()
.ADIE ON THE OSMOTIC PRESSURES
In pot r.
In pot
1 40
K2so~
................
410 KN03....•..•........ 1
160
K~Fe(CN)a
..........
1 1 - K 2 SO~ + - KN0 3 • • • • 40 41J 1 1 - K2SO, + - OK,Fe(CN) 6 • 40 16 1 1 - KN0 3 +-K,Fe(CN) 6 40 160
pres~ure.
0·96
0·50
-
0·48
2·05
2·21
2·37
2·3\)
1•80
1·90
1·9J5
1·91
1·23
1·31
1·495
1•44
1
1
20
40
GO
o. p.
K(SbO)C-~.H~06
K 4 Fe(CN)6 ... KsCoz(CN),2·.
i.
-
-
2·30 2·50 1·35 1·21 3·4!:) :3'12
·-
-
GO
~--"----,
o. P·
i.
0.
1·59 2·21 1·82
1·42 1·9.5 1·6:3
1·50
-
-
-
,..- ...... - .,
i.
p.
1·:~ o
o. p.
2·:13 2·68
0•51 0·97
-
P~·
1
1•64 1•47 Hll 1·6:d 1·n 1•55 1·26 1·U 2·]6 1·9:-3 2·26 2·02 1·32 1·18
40
,..--'--,
c. Solutions of potassium salts of org:tnic acids. pots P2 and 7 for 2-'0 form. wt. per litre : -
~
o. P·
i.
0·96
1·72 1·ti4
0 ·9 ~
1•25 2·24, 1•81 2·34 0·66 1·17 0·79 1•4l 1•17 2·10 0·75 1·34 2·05 :~·67 :)'46 t3'18
,-..A-, i. o. P· - - --
80
,---"-, o. P·
i.
0·47 1•6() 0·4()6 1·7 7
KFo . . . . . . . . . . . KAc . . . . . . . . . • KPr ........... Kl3z .. . .. .. . .. K20x..........
4~) KzOx. • "
'l.
1 · 4~
2·68
Two series m
r.
,..---"----, p. i.
,--...A..--, o. p.
i.
1·09 1·42 1·46 Hl5 2·26 (;i)
1·4:3 1•66
1·~8
0.
IV. Influence of the Acid. a. Solutions of potassium salts of different strengths, mostly in one pot o:-
KN0 3 • • • • • • • KI .......... KCl0 3....... KC,H30 z ... .• K2SO, ....... KzC20,....... KHC0 3 •••••• KHS03 ...... K zC.H.Os ....
26
NaNOs .... NAzS~Oa ... · NB.e!HPO, .. Na3Ci ......
As a whole, these results show that when the two components can combine to form a doable salt, Dalton's law does not hold accurately, and that the observed pressut·e for the compound is less than the sum of the pressut·es of the constituents. In t.be enses when the components do not combine to form a double salt whi ch may be crystallised out, the observed pressure of the mixture of the two salts is about equal to the sum of the pressureH of the constituents. In the case of the iodides, IJ, i, tho observed pressut·e is eYen collsiderably greater than tho sum. In the results of pot.,. inc, as in t.hose of pot o., in the sn.me class, the deviations, wheu in the same direction, are smaller than those iu class u.
,_...A-,
1
components.
1·48
-
0·814
3G7
b. Solutions of sodium salts of different strength~, most.ly in il : -
Osmotic
-
1·40
o~.
,--...A..--, Sum of
,- ~
Osmotic Sum of pressure. components.
Salt solution.
Ql.' l:iALTS 1~ SOLUTIO:\'.
0·975 1·2 7 1·33 1·74 2·0::l
• • • • • • " • • · • • • .. " .. · · · · {
1 7tl
1•48 1·59
;:~;
}2·52
These results are discussed after the next scl'ies.
V. lnfinen.:e of Base. a.
Solutiom of nitrates of different bases in t~·o pots, IJ anll :'3 : -
-~l)1
,-...A..-, KN0 3•••••• Na~0 3 •• ••• NH 4 NU~ ... 1\fg(NOah · • Ca(N0:1)z . .• Sr(NOaJ~ .. · Ba(N0:1 )~ •• Co(N0~)2 · · Cu(NO:c)2 ·.
'I·
{j.
1 ·u1
1 '76 1'79
2·12 1'67 1•58
i. 1·50 (Mean) 1"60
1'48 1'32 1 ·so - - 1'40
2•14
2 '04 -
2 ' 14
--
1 40
,..---'-- - ,
~.
0 '!:!9 (Mean)
0 '60
o·6o u·us
-
1 ·07
1•41
1•82 1 '91 1·93
a.
-
1 '35 0'93
--
~
1·59 1•07 1 ·1s
1 '72 1 ·os 1 ·IJ2 (Mean) 1'38 2•47 1'33 2·38 - 2 ·~2 1 ·GG
--
b. Solutions of sulphn~es of different bases in two pot s, "and .,., with O.P. for '.l'o iu different pots :-
-
-
1·49
4·11
2•17
5'98
(1·:12
1'14
6·9() 1•77
:~ · H
G·:).j
368
ADIE
0~
THE OSMOTIC PRESSURES 1
1
26 ,--"- ·--,
40
o.p.
K.so, .... . .. . Na2SO,. (NH,)2S O, ... . MgSO, ... . .. . CaSO, . • .. ... . AI2 (S 0,) 3 . ...
2•19 3·13 2•64 O·i8 1
OF SALTS IN
,-----·-"-------.., "'·
i.
i for mean.
r.
1·97 2·79
1"139
2·38 0·70
1·2G
2•40
1·40 1·29:>
2·20
-
-
-
1·41 2·:23
0·79 2 b!65 1·22
Probably due to destru ction of film. by the salt. Extrapolation vA.lue beyond saturated solution, and has no interpt·etation. Iuserted to show different order of O.P. for an nndissociat.ed solution. 1
2
c. Solutions of alums of different composition of per lit.re; Kat 16°, Tat 14u :0. P· _J__,
K,Al 2(S 0,),,24H20 . . . . K,Cr,(S0,),,24Hf) ... : . (NH,) 2 Al,(S0,),,24H,O .
(IC,
T.
2·:39
1·96
i.
r---"----, K. r.
4·27 4·40 2·25
~:l-46
2·:~7
·io gram form. wt.
1·98
;)·50
3·54
d. Solutions of normal and hydrogen salts:-
I_
40
Na 3 Ci .......... .•
_1 Na 2 H('; I' 40
.•• • • •••• •
[ 1·4(3 calcd.J
3G9
firmed by the fact that the values of i are more nearly constant for than for -io solutions in the case of the bases, and that this point is in BC . It must be noted that sodium salts generally m·e more clisRociatcd than potassium salts, particularlr wheu the acids luwe high affinity. This is also borne out by the fact that sodium salts gonemlly de:'ltroy the fi lm, sooner or later, whi lst potassium salts do not. The su lpbates nlso are much mor·e dissociated in watet·y solution tbnn the nitrates, except when they form double salts, such as the alums. The base values do not appear to be periodic, but rather to be connected with thtl acidity of the base. No general principle governs the action of hydrogen in acid salts; in some cases it lowers the 0. P., in others it raises it. 'J'aking the salts of the organic acids first, we see that an increase in the molecular weight of the acids is accompanied by an increase of the O.P., and also that the O.P. seems.to depend on the basicity of the acid. This may be interpreted that the carboxyl group grauually I0ses its unsaturated condition with increal!e ' of molecular weight of the attached carbon group, so that t.he retardation whic:h occul's from the action of the water on the carboxyl group gmdnally loses ittl prominence. In ordet· to compare my valnes of osmotic pressures with those of the electric conductivities, I append the following values of KohlJ·ausch's (B.A. Reports) for tlt e molecular conductivities of solutions 1 af 1 ll' gmm form. wt.. per litre ( 108 k/m, whet·e k the condnctivity and m = the molecule in grams). The osmotic pr·essures are those deter·mined for ... ~ gram form. wt. per litre. -;/0
=
=
2·12 Kohlrausch. KNO:~
1 - K 2 0x . . . . . . . . . . . .
1·41
1 - KHOx .. . .. . . . . . . 40
0·77
4V
1 = 0·97. 60
SOLUTIO~.
Na.JPO, KHSO,, and K~S 0 3 are decomposed by water, and dissoh-e the tilm. Magnesium, copper, ammonium and sodium nitrates, and ammonium sulphate Jm,·e smaller values of i for dilut.e solutionR than for concentJ"Uted. This may be explained by the existence of A. hydrate, but it is not practicable to define them, by this method. In examining these results, it mnst be remembm·ed that they aro taken at any points in the curve OABCD, §I. In order to be strictly comparable they should be taken from OA, or from BC; this is con-
...... . .... .. KI .. . ..... . . ... .. KCIO, . . . .. ...... . KAc .... .... ......
983 1069 927 784
~K2SO, . . ... . . .... .
897
KNOa ......... . . .. NaNOa . .......... .
98::1 817
~ Ba(NO")., .. .. ... . .
755
iK2S01 . . ..... . ... .
897
iNa~so, .. . ....... .
7at
o. p .
1·(3± 1·81 1·7~
1·2(3
1
40
1·25 1·71) 1·79
I
41) 1
20 1 21j
1·3:3
'
;
.
:I
2·19 3 ·13
370
ADIE ON THE OSlllOTIC PRESSURES
OF SALTS IN SOLUTIOX.
The values in the two cases appear to take a simila.r direction, though in the case of sodium, my values are always higher than the conesponding values for potassium. Kohlrausch ulso gives the relative velocities of the 10ns : -
per litre the limits are 1·13 to 2·02 for simple salts, and for ·lo form. wt. 1·17 to 2·34. In the table in V, the values of i do not differ so much: 1·33-1·93 for 210 form. wt. per litre, and 1·66-2·47 for ·i'-o· This confirms the fact that the osmotic pressures do not furnish any guide to the determination of the molecular weights of inorganic salts in solution in water, at least for ~omparison with ga11eous molecular weights. It is possible tl1at tbe results may be mot·e consistent with organic than with metallic baseR. In the case of the acids, the osmotic pressures of the solutions which ~·ontain 0 \- equivalent of potassium, or sodium per lit,·e agree much better than the corresponding proportion of the fo1·mula weight. The sulphates are more consistent than the nitr·ates, though they dissociate more easily; the latter m~ty be connected with the formation of water compounds of the acid formed by dissociation, giving af! result a basic salt in solution, mixed with dilute snlphuric acid. The limiting value of i, as found by An·henius' hypothe."is into ions is as follows-
K.
:XH,.
No..
H.
tD~.
~Mg.
6~
50
:32
274
:30
26
I.
NO:..
C!Oa.
A c.
55
48
44
26
These velocities take the same direction as the osmotic pressure valueR, except in the C
For
M'X'
•
•••••••
0
=2
•••••
M"X' 2 or JI.1' 2 X" . .... l'vl"'X'3 or M' 3 { or M" 2 X' 2
=3
X"'}
· where
::n
= 4
M = any metal, X = any acid radicle.
These valueM are not exceeded in my observations, ann it is only the stronger acids and bases which approach this limit. On the other hand, for the organic acids, the stronger are less di~sociated. If dissocintion into ions takeR place, why do tbey not get through the film rather thnn tbe water molecules As a matter of fact, the dissociations which can be obsened arc those of the salt. into an acid and a basic portion, of which the acid part gets through, and destroys the film. This passage of the ions through the film should bP. mot·e distinctly noticeable in the case of the acid salts, where, according to Kohlrau11ch 's numbers, the velocit.y of H is 274; while that of K is ouly 62. However· we do not find the O.P. of these salts less than it ought to be, from this cause, V, d. On the other hand, the va1ue of i may be conditioned by the amount of dissociation of tho molecules of the salt, p1·oduced by the action of the solvent, as described above. On thiR l1ypothcsis, the vstloe of i at its limit represP.nts the maximum amount of dissociation of each molecule of the solid in the state in which it exists in solution, produced by the action of the solvent. There is no doubt that the va!ucs of 1·, obtained by any of the
r
;
~
il •:'!
'~i ;.I!
:,I
'•I . ··:~
j:
·~.f
!.rI'
·ji;
r!l
l.f
~l,·!,l I
.d
l;; :1
.,' ''
I;i~i! :·;,~
.\r! :'I ~ ~r
~~
.;~ ~.
"
?.JELDOL.A AXD HUGHES : NOTES ON THE
AZO-DERIVATIVES OF f:I-~APHTHYLAMINE.
various methods in use, agree with one another, are fairly definite, and have a meani1;1g, but I incline to the explanation that the action of the solvent is twofold, and may be expressed asi. Combination of the molecules of the dissolved substance with the solvent to form complex molecules . ii. Separation of t.he complex molecules of the dissolved substance, under the action of the soh·ent, into simpler molecules, not necessarily ns small a~ the limiting gaseous molecu le. In conclusion, I have great pleasure in thanking Professor J. J. ThoHlson for the use of the CaYcndish Laboratory, Cambridge, where this 1·esearch wa.s conducted, and for his kind assistance, and advice during the progress of the wm-k.
pounds derived from ~-naphthyla.mine, viz., that t.ho latter contain the greup
372
XLIL-Notes on the Azo-de?·it·atives of {3-Naphthylamine.
By
RAPHAEL MELDOLA, F.R.S.,
and
FltAKK BuGHF: S.
AT the brginning of the prrsent session we commencecl n.11 investigation hn.ving for its object ihe extension of the r esnlts which had previously been arrived at respecting the eonstitution of t.he azoderivatives of ,8-napht.hol by one of the anthot·s in conjunction with J!'. J. East ('J'rans., 53, 1888, 4GO) and Gilbert. '1'. Morgan ('l'ra.nR., 55, 1889, 114 and G03). 'l'he work was carried on till one of ns (F. H.) recei 1·ed n.n appointmeut necessitating removal from London, R.nd R.lthongh we cannot regard ihc pt'esent eom.munication as throw ing any new light on the qul"stiuu of the constitution of the azo-de1·ivatives of [:l-n;
:N /'-..
N-O·H . - ,
and analo3y would thercfot·e lead to a similar formula for the com·
373
N /'-.
~-~Hz.
:Much evidence has been adduced in f,wour or the view that Ll•ese compounds contain the NH. group, and is so well known tll'Lt it need not be reca.p itulated here; but that this evidence still nppenrs insufficient to some chemists is shown by t.he fact that in a paper published last year by H. Goldschmidt and G. Rosell (Be·r., 23, 1890, 487) the quinonoirl formula is adopted by these autbol's u"th for the ,8-naphthol and /3-naphthylamine a:zo-derivatives. It is fo1· the purpose of obtaining further evidence as to whether these last compounds contain two ltydrogen atoms attached to the same nitmgc n atum that the present experiments have been commenced. As in t.he pi'evious part of the work on the azo-{J-naphtho l derivatives the most satisfactory results were obtainPd when the aromatic nucl eus linked by the a.zo-gt·oup to the naphthalene r e:>idue contained a stro ngly acid radicle (NOz), it was thought desirable to begin with the analogous nitroazo-componnds of {:1-naphthylamine, the more especi:dly as the simpler benzeneazo-f:l-naphtbylamine has already been made the sabject of numerous experimental investigatio ns by Zincke and Lawson, Goldschmidt, &c. O?·thnn·it-roben.teneazo-{3-naphthylamine.
Para- ancl. meta-nitrobenzeneazo-f3-naphtbylamine wm·e uescJ·ibeJ
by one of us many years ago (Trans., 1883, 43, 430; and 1884, 45 , 116). In order to complete the series of isomerides, we l1a\e now prepared the isomeric compound f1·om rliR.zotised orthonitmniline and The orthonitmniline was diazot.ised in tho usual way with the calculated quantity of sodium nitrite in hydrochlOJ·ic acid solution, and the diazo-chloride poured into the cold, dilute aqueous solut.ion of ,8-uaphthylamine hydrochlor·ide. The aw-compound n.t once separates as a dull, b1·onzy powder, and uJte1· being collected and washed well with wat.er, it was cJ·ystallisvtl hum alcohol aud finally from glacial acetic acid nutil the analysis ohowed that the substa11cc was pure. The following results were o biained : ~-naphtbylamine.
I. 0·1233 gram gave 0·2987 gram C0 2 and 0·05G8 gram H 2 0. II. O·l Gl9 , , 0·3904 , , 0·072G , , III. 0·0991 , , 16 c.c. moist N fi.t, 9° and 757·8 mm. hal'. IV. 0·1574 , , 25·8 c.c. moist N at 11·5° and 75:2·6 mm bar.
