Materials Science and Engineering A 367 (2004) 317–321
Effect of carbonitride particles formed in austenite on the strength of microalloyed steels M.D.C. Sobral a , P.R. Mei b , H.-J. Kestenbach c,∗ a Centro Federal de Educação Tecnológica da Bahia, Salvador, BA, Brazil Department of Materials Engineering, Universidade Estadual de Campinas, SP, Brazil Department of Materials Engineering, Universidade Federal de São Carlos, Rod. Washington Luis, km 235, 13565-905 São Carlos, SP, Brazil b
c
Received 17 July 2003; received in revised form 15 October 2003
Abstract Precipitation strengthening from two different carbonitride populations was investigated in a niobium and vanadium microalloyed hot-rolled steel. Interphase precipitation as well as carbonitride particles formed in austenite were found to contribute to precipitation strengthening. After normalizing, austenite precipitation strengthening remained effective while the strengthening contribution of interphase precipitation was strongly reduced. © 2003 Elsevier B.V. All rights reserved. Keywords: Microalloyed steels; Carbonitride precipitation; Precipitation strengthening; Thermornechanical processing; Normalizing; Electron microscopy
1. Introduction Precipitation hardening by fine carbonitride particles has been used for many years in order to increase the strength of hot-rolled microalloyed steels. During thermomechanical processing, carbonitrides may nucleate in austenite during rolling, at the ␥/␣ interface during transformation (interphase precipitation), or in supersaturated ferrite during final cooling [1]. It is frequently believed, however, that a direct strengthening effect can only be obtained from particles which have nucleated in ferrite during or after the phase transformation, and that prior precipitation in austenite, although important for grain refinement, would adversely affect the yield strength by reducing the amount of microalloy elements in solution which would otherwise be available for precipitation in ferrite [2–4]. A long-standing argument in support of this view has been the idea that particles nucleated in ferrite would be partially coherent with the matrix, whereas particles nucleated in austenite would loose coherency during transformation to ferrite [5]. It has ∗ Corresponding author. Tel.: +55-16-260-8501; fax: +55-16-261-5404. E-mail addresses:
[email protected] (M.D.C. Sobral),
[email protected] (P.R. Mei),
[email protected] (H.-J. Kestenbach).
0921-5093/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.msea.2003.10.280
been argued theoretically, however, that coherency should not enhance the strengthening effect of carbonitride particles in microalloyed steels [6], due to the very low volume fractions involved (typically between 10−4 and 10−3 ). In addition, it has now been shown that all carbonitride particles encountered in commercial microalloyed steels are incoherent [7]. Finally, extensive transmission electron microscopy (TEM) observations of commercial hot strip steels conducted in the laboratory of one of the present authors suggested that fine carbonitride particles which nucleate during or immediately after rolling in the unrecrystallized austenite may also cause precipitation strengthening [8–10]. A very interesting situation with respect to precipitation hardening occurs in hot strip steels during coiling. In general, prior to coiling, carbonitrides will have formed either in austenite during finish rolling or in the form of interphase precipitation during transformation. Due to accelerated cooling on the run-out table, a relatively large amount of microalloy elements may still be in solution at the beginning of coiling, with both particle distributions competing to use this portion for particle growth. As shown by detailed investigations in our transmission electron microscopy laboratory, such competition may result in similar contributions of the two types of carbonitride particles to strengthening in commercial hot strip steels [10].
318
M.D.C. Sobral et al. / Materials Science and Engineering A 367 (2004) 317–321
In order to evaluate more precisely the role of deformationinduced carbonitride particles on strengthening, an austenite holding period just below the finish rolling temperature was now included during an experimental rolling regime which was designed to approximate thermomechanical processing on a hot strip mill. The extra time before transformation was expected to favour the volume fraction of austenite-nucleated particles and reduce the extent of interphase precipitation. Microstructure and mechanical properties were compared for samples with and without this holding period. The results are shown in the present paper.
2. Materials and methods A commercial niobium and vanadium microalloyed steel was remelted and cast in the form of a 42 mm thick slab in a thin-slab casting simulator, cooled to room temperature, and cut into two identical samples. These were reheated to 1250 ◦ C and thermomechanically processed on a laboratory-scale rolling mill. The chemical composition of the thin slab was 0.07% C, l.32% Mn, 0.08% Si, 0.010% P, 0.009% S, 0.03% Al, 0.02% Nb, 0.05% V and 0.010% N. Hot rolling was conducted in the form of three roughing passes between 1150 and 1050 ◦ C to 25 mm thickness, followed by five finishing passes between 900 and 840 ◦ C to a final thickness of 10 mm. After rolling, one sample was rapidly cooled to 600 ◦ C before the industrial coiling operation was simulated by holding for 2 h in a tubular furnace and final air cooling to room temperature. The other sample, after rolling, was subjected to an additional austenite holding period of 15 min at 820 ◦ C before starting the same simulated coiling treatment. In addition, part of the “coiled” strip was submitted to a normalizing heat treatment at 900 ◦ C for 30 min. Tensile tests and microstructural observations by optical microscopy were conducted on all the samples. Round tensile specimens were machined from transverse sections (tensile stress normal to the rolling direction), selecting a 4 mm gauge diameter and 20 mm gauge length in accordance with ASTM E8M standard subsize samples. The state of carbonitride precipitation was determined for the “as-coiled” samples by transmission electron microscopy, with the particle origin determined by noting the presence or absence of a Baker–Nutting orientation relationship on the electron diffraction pattern [11]. When investigating the strengthening mechanisms in low-carbon steels, it is common practice to use the structure–property relationship developed many years ago by Pickering and Gladman [12], and which only takes into account the effects of solid solution and grain size hardening: σPG = 15.4(3.5 + 2.1%Mn + 5.4%Si + 23%Nf + 1.13d −1/2 )
(1)
where σ PG is the predicted yield strength in MPa, %Mn, %Si and %Nf are the contents of manganese, silicon and free
nitrogen in weight percent, and d is the ferrite grain size in millimeters. Due to the presence of the microalloy elements, %Nf was assumed to be zero. The difference between the predicted values, σ PG , and the experimental results derived from tensile testing, σ exp , can then be related quantitatively to the presence of additional strengthening mechanisms.
3. Microstructure and mechanical properties Very similar ferrite + pearlite microstructures with frequent banding were observed for all processing conditions under the optical microscope, with some examples being shown in Fig. 1. Average yield strength values as determined from three different tensile tests and the mean ferrite grain size after (simulated) coiling are shown in Table 1, together with the yield strength predictions σ PG and an additional strengthening value, σ ppt , taken simply as the difference between σ exp and σ PG . As shown in a previous paper [9], additional strengthening in hot strip steels can be expected to come from either carbonitride precipitation or dislocation hardening. In the present case, very low dislocation densities of only about 108 cm−2 were observed in the transmission electron microscope, suggesting that the difference between σ exp and σ PG should have been caused entirely by precipitation strengthening. As can be seen from Table 1, a drop in yield strength of about 40 MPa occurred due to the holding period in austenite. Since part of that drop was caused by an increase in ferrite grain size (from 6.3 to 8.1 m), a loss of only 18 MPa (from 112 to 94 MPa) should correspond to the effect of holding in austenite on reducing the amount of precipitation strengthening. Examples of the state of carbonitride precipitation after the simulated coiling operation are shown in Figs. 2 and 3. Since microalloy elements are known to segregate during solidification, different sample locations may show different precipitate distributions and particle sizes. All TEM samples were therefore taken from the same central mid-thickness position within the original hot-rolled strip. Due to the limited number of TEM observations (each condition was observed in a minimum of three ferrite grains only), Figs. 2 and 3 may not give representative particle distributions, but should give realistic particle sizes. As would be expected, therefore, holding in austenite increased the average size of the carbonitride particles which had been nucleated during rolling in austenite (see Fig. 2), and decreased the average size of the carbonitride population Table 1 Mechanical properties and individual strengthening contributions after “coiling” Processing conditions
σ exp (MPa)
Grain size (m)
σ PG (MPa)
σ ppt (MPa)
Direct coiling
434 ± 10
6.3 ± 0.4
322
112
Holding before coiling
391 ± 11
8.1 ± 0.5
297
94
M.D.C. Sobral et al. / Materials Science and Engineering A 367 (2004) 317–321
319
Fig. 1. Optical microstructures of samples processed with holding period before coiling. As-coiled condition in (a), after normalizing in (b). Magnification 300×.
Fig. 2. Carbonitride precipitation in austenite. Without austenite holding time in (a), with austenite holding time in (b), magnification 70,000×. Diffraction pattern in (a) shows absence of a Baker–Nutting orientation relationship between carbonitrides and ferrite matrix.
320
M.D.C. Sobral et al. / Materials Science and Engineering A 367 (2004) 317–321
Fig. 3. Carbonitride particles formed by interphase precipitation. Without austenite holding time in (a), with austenite holding time in (b), magnification 70,000×. Diffraction pattern in (a) shows presence of a Baker–Nutting orientation relationship between carbonitrides and ferrite matrix.
which had been formed afterwards during interphase precipitation (see Fig. 3). In order to understand the correlation between these carbonitride populations and the precipitation strengthening in Table 1, it is important to recognize that interphase precipitation would normally be associated with a higher strengthening potential, due to its finer particle distribution when compared to the carbonitrides formed in austenite [10]. Thus, the loss of only 18 MPa of precipitation strengthening due to holding can readily be explained by a partial substitution of the volume fraction of finer interphase precipitation by that of coarser austenite precipitation.
4. Effect of normalizing As has been known for a long time [13], the yield strength of a microalloyed hot-rolled steel is usually reduced by normalizing. This effect was originally attributed to the loss of coherency which those very fine carbonitride particles supposedly formed in ferrite after rolling would suffer during the phase transformation [5]. As suggested recently [9], however, the effect of normalizing should only be related to a precipitate coarsening effect, and should thus depend upon the type of carbonitride population present after rolling. In fact, recent results have shown that normalizing leads to very intensive coarsening of the finer interphase particles, while coarser carbonitride distributions which have formed on the deformation substructure in austenite may
Table 2 Mechanical properties and individual strengthening contributions after normalizing Processing conditions Direct coiling
σ exp (MPa) 367 ± 4
Grain size (m) 7.2 ± 0.7
σ PG (MPa) 309
σ ppt (MPa) 58
Holding before coiling
364 ± 1
8.9 ±1.1
288
76
survive the normalizing treatment without major changes [14]. Yield strength values after normalizing are shown in Table 2, together with an estimate of the remaining amount of precipitation strengthening after particle coarsening. When comparing equivalent data in Tables 1 and 2, it can be seen that normalizing has led to a major loss of precipitation strengthening (from 112 to 58 MPa) after direct coiling, whereas a much smaller loss (from 94 to 76 MPa) occurred when austenite holding was included before coiling. Such results are consistent with the two types of carbonitride distributions shown in Figs. 2 and 3, both contributing to precipitation strengthening of the hot-rolled samples but subject to different degrees of particle coarsening during normalizing: • Austenite precipitation whose strengthening contribution was favoured by the austenite holding period and remained effective after normalizing. • Interphase precipitation whose strengthening contribution was favoured by direct coiling but strongly reduced after normalizing.
M.D.C. Sobral et al. / Materials Science and Engineering A 367 (2004) 317–321
In agreement with the arguments presented above, two previous investigations of one of the present authors have already shown that the amount of precipitation hardening found in as-rolled samples could be maintained after normalizing [8,9]. Significantly, in those two cases, the only important precipitation strengthening contributions had come from carbonitride particles formed in austenite, due to a rather high transformation temperature which prevented the occurrence of interphase precipitation on a finer scale.
5. Conclusion It has been shown that carbonitride particles formed in austenite during finish rolling may give an important contribution to precipitation strengthening in hot strip steels. Their strengthening contribution was essentially maintained after normalizing, contrary to the strengthening effect of interphase precipitation which was strongly reduced by normalizing.
Acknowledgements This work received financial support from the São Paulo State Agency FAPESP (contract 2000/11506-6) and the Federal Government Agency FINEP (contract 82/97). H.-J.K.
321
would like to thank the Brazilian National Research Council CNPq for a research fellowship.
References [1] R.W.K. Honeycombe, in: J.M. Gray, et al. (Eds.), HSLA Steels: Metallurgy and Applications, Metals Park, Ohio, ASM, 1986, pp. 243–250. [2] L. Meyer, F. Heisterkamp, K. Hulka, W. Müschenborn, in: T. Chandra, T. Sakai (Eds.), THERMEC 97, TMS, Warrendale, PA, vol. 1, 1997, pp. 87–97. [3] T. Gladman, The Physical Metallurgy of Microalloyed Steels, The Institute of Materials, London, 1997, pp. 298–303. [4] A.J. DeArdo, Mater. Sci. Forum 284–286 (1998) 15–26. [5] W.B. Morrison, J. Iron Steel Inst. 201 (1963) 317–325. [6] M.F. Ashby, in: A.S. Argon (Ed.), Physics of Strength and Plasticity, MIT Press, Cambridge, MA, 1969, pp. 113–131. [7] E.V. Morales, J. Gallego, H.-J. Kestenbach, Phil. Mag. Lett. 83 (2003) 79–87. [8] A. Itman, K.R. Cardoso, H.-J. Kestenbach, Mater. Sci. Technol. 13 (1997) 49–55. [9] S.S. Campos, E.V. Morales, H.-J. Kestenbach, Metall. Mater. Trans. A 32A (2001) 1245–1248. [10] H.-J. Kestenbach, J. Gallego, Scripta Mater. 44 (2001) 791–796. [11] A.T. Davenport, L.C. Brossard, R.E. Miner, J. Met. 27 (1975) 21–27. [12] F.B. Pickering, Physical Metallurgy and the Design of Steels, Applied Science Publishers, London, 1978, pp. 50. [13] W.B. Morrison, J. Woodhead, J. Iron Steel Inst. 201 (1963) 43–46. [14] S.S. Campos, E.V. Morales, H.-J. Kestenbach, Mater. Sci. Forum 426–432 (2003) 1517–1522.