Canadian Metallurgical Quarterly

The Canadian Journal of Metallurgy and Materials Science

ISSN: 0008-4433 (Print) 1879-1395 (Online) Journal homepage: http://www.tandfonline.com/loi/ycmq20

Microstructural evolution of a hot-rolled

microalloyed complex phase steel

H. Scott, G. Sidhu, F. Fazeli, A. K. Pilkey & J. D. Boyd

To cite this article: H. Scott, G. Sidhu, F. Fazeli, A. K. Pilkey & J. D. Boyd (2016): Microstructural
evolution of a hot-rolled microalloyed complex phase steel, Canadian Metallurgical Quarterly,
DOI: 10.1080/00084433.2016.1252555

To link to this article: http://dx.doi.org/10.1080/00084433.2016.1252555

Published online: 07 Nov 2016.

Submit your article to this journal

Article views: 5

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=ycmq20

Download by: [Laurentian University] Date: 28 November 2016, At: 22:30

Microstructural evolution of a hot-rolled
microalloyed complex phase steel
H. Scott1, G. Sidhu1, F. Fazeli2, A. K. Pilkey*1 and J. D. Boyd1
The current study examines a grade of hot-rolled and continuously cooled complex phase sheet
steel comprised of polygonal ferrite (PF), granular bainite (GB) and lath bainite (B). The quantity
of each constituent phase depends on the thermomechanical processing conditions, which vary
between commercially produced sheets. In this study, the effects of cooling rate and austenite
grain morphology on microstructure are determined through a series of dilatometry experiments.
The resulting CCT diagrams show a progression in the order PF → GB → B with increasing
cooling rate, and that a Pancaked (unrecrystallised) austenite condition promotes the formation
of PF to higher cooling rates and the formation of GB to higher temperatures. Application of the
CCT results to industrially produced sheet provides a useful approach for interpreting the
evolution of microstructure during controlled-cooling and coiling. However, direct comparison is
limited by the moderate level of austenite pancaking that can be achieved through laboratory
dilatometry experiments in comparison to an industrial hot mill. Notable differences in
microstructure are observed between the leading and trailing edges of industrially produced
sheets due to relatively small variations in cooling schedules.
L’étude en cours examine une nuance de tôle d’acier laminée à chaud et refroidie en continu à phase
complexe comprenant de la ferrite polygonale (PF), de la bainite granulaire (GB) et de la bainite
massive (B). La quantité de chaque phase constituante dépend des conditions de traitement
thermomécanique (TMP), qui varient parmi les tôles produites commercialement. Dans cette étude,
on détermine les effets de la vitesse de refroidissement et de la morphologie du grain d’austénite
sur la microstructure par une série d’expériences de dilatométrie. Les diagrammes de CCT qui en
résultent montrent une progression de l’ordre de PF → GB → B avec l’augmentation de la vitesse
de refroidissement, et qu’une condition d’austénite en galette (non recristallisée) favorise la
formation de PF à des vitesses de refroidissement plus élevées et la formation de GB à des
températures plus élevées. L’application des résultats de CCT aux tôles produites industriellement
fournit une approche utile à l’interprétation de l’évolution de la microstructure lors du
refroidissement contrôlé et de l’enroulement. Cependant, la comparaison directe est limitée par le
niveau modéré de formation en galette de l’austénite obtenu lors des expériences de dilatométrie
en laboratoire par rapport à un laminoir industriel à chaud. On observe des différences
remarquables de la microstructure entre les bordures de tête et de queue des tôles produites
industriellement en raison des variations relativement faibles des programmes de refroidissement.
Keywords: High strength steel, Microstructural characterisation, CCT behaviour

1
Mechanical and Materials Engineering, Queen’s University, Kingston
K7L3N6, ON, Canada
2
CanmetMATERIALS, Natural Resources Canada, Hamilton L8P0A5, ON,
Canada
*Corresponding author, email keith.pilkey@queensu.ca

© 2016 Canadian Institute of Mining, Metallurgy and Petroleum

Published by Taylor & Francis on behalf of the Institute
Received 21 May 2016; accepted 18 October 2016
DOI 10.1080/00084433.2016.1252555 Canadian Metallurgical Quarterly 2016 1
Scott et al. Microstructural evolution of a hot-rolled microalloyed complex phase steel

Table 1 Chemical composition of steel under study (wt-%)

C Mn Si Nb Ti Cr Al N

Transfer bar 0.049 1.51 0.56 0.03 0.16 0.61 0.05 0.007
Sheet 11 0.043 1.44 0.57 0.03 0.16 0.63 0.05 0.007
Sheet 18 0.041 1.43 0.57 0.03 0.16 0.63 0.05 0.008

Introduction forms: (1) 30-mm thick transfer bar material for dilat-
ometer specimens and (2) hot-rolled, controlled-cooled,
Microalloyed complex phase (CP) steels are a class of coiled sheet. Samples representing a high (sheet 18) vs.
advanced high strength steels which have been developed low (sheet 11) finish rolling temperature were provided.
for applications where stretch flangeability is a critical For each finish rolling temperature, samples from the
property. Several microstructural features have been leading edge (L) and trailing edge (T) of the as-processed
shown to influence the stretch flangeability of CP steels sheet were provided. The processing details of the four
as evaluated by the hole expansion test. First, coarse industrially produced sheets are given in Table 2.
bands of ferrite grains, often attributed to Mn segre-
gation, are detrimental to hole expansion;1–5 therefore,
a uniform microstructure is preferable. The volume frac- Dilatometry
tion6 and grain size7,8 of the various microstructural con-
stituents have also been shown to influence hole Solid 5 mm diameter × 10 mm long cylindrical specimens
expansion, along with the morphology and distribution were machined from the 30-mm thick transfer bar with
of (Ti,Nb)(C,N) precipitates on the scale of ∼2– the cylinder axis (compression direction) oriented in the
100 nm.9–11 Despite this collective knowledge, the opti- through-thickness direction. Some hollow cylindrical
mal microstructure for highest strength and hole expan- samples (5 mm OD × 10 mm long with 1.5-mm wall
sion properties has not yet been determined. Moreover, thickness) were also prepared to achieve higher cooling
in order to control the effects of these microstructural fea- rates in the dilatometer.
tures, a thorough understanding of the evolution of CCT characteristics were determined by means of a
microstructure during the processing of CP steel sheet is Bähr DIL805A/D quenching and deformation dilat-
first required. ometer. Two different thermomechanical schedules were
Previous studies have identified various features and employed to provide two different starting austenite con-
relationships among the microstructures and processing ditions. All test specimens were held for 15 min at 1200°C,
conditions for microalloyed CP steels. The final micro- cooled to 1050°C, deformed in uniaxial compression to
structure generally comprises polygonal ferrite (PF), gran- 0.3 strain at a strain rate of 1 s−1 and cooled to 850°C.
ular bainite (GB), conventional lath bainite (B) and This was the starting point for the CCT experiments for
martensite/retained austenite (M/A), with size, distribution the Recrystallised austenite condition (Fig. 1). A second
and morphology dependent upon the composition and set of samples was deformed at 850°C to an additional
processing conditions.8,12,13 Large, elongated bands of fer- 0.3 strain at a strain rate of 1 s−1 to produce the Pancaked
rite tend to develop in steels cooled at slow rates and/or austenite condition (Fig. 1).
those with high Mn content.1 Dilatometry experiments For each of the two austenite conditions, individual
carried out on CP steels indicate a tendency of deformed samples were cooled at constant rates of 0.5–80°C s−1
austenite to displace continuous-cooling-transformation and sample dilation was recorded as a function of temp-
(CCT) curves to higher temperatures and shorter times erature. A hollow sample was gas quenched from 1200°C
when compared with recrystallised austenite.13–15 An at a rate of approximately 200°C s−1 to capture the as-
analysis of a Nb–Ti microalloyed CP steel by Hu et al.16 reheated austenite grain structure.
showed that an increase in cooling rate and a decrease in
coiling temperature caused M/A constituents to become
small, dispersed and granular, the grain size to be refined Microstructural characterisation
and a decrease in the amount of PF. The solid and hollow dilatometry specimens were sec-
In the present work, CCT diagrams for recrystallised tioned along the longitudinal axis by means of a Struers
and pancaked austenite conditions of a Nb–Ti microal- Accutom-5 automatic precision cut-off machine. The
loyed CP sheet steel were determined using laboratory
dilatometry experiments. The CCT diagrams were then
Table 2 Processing details of industrially produced sheet
used to interpret the evolution of microstructure in four samples
industrially produced sheets.
Finish rolling Coiling
Sample Location temperature/°C temperature/°C
Experimental
11L LEAD 869 506
Materials 11T TRAIL 890 482
The chemical composition of the steels under study is 18L LEAD 931 533
18T TRAIL 929 495
given in Table 1. Sample material was provided in two

2 Canadian Metallurgical Quarterly 2016

 Scott et al. Microstructural evolution of a hot-rolled microalloyed complex phase steel

1 Thermomechanical schedules for dilatometer experiments illustrating procedures to produce Recrystallised and Pancaked
starting austenite conditions

industrial sheet samples were sectioned to observe the The sectioned samples were mounted and polished up
longitudinal – through-thickness (RD – TT) plane. to a 0.06 μm colloidal silica solution on a Struers DAP-

2 Optical micrographs of industrially produced sheet microstructures

Canadian Metallurgical Quarterly 2016 3

Scott et al. Microstructural evolution of a hot-rolled microalloyed complex phase steel

V Pedemin 2 automatic polisher. Ammonium hydroxide Hardness tests

was added to the colloidal silica at a concentration of Vickers hardness tests were conducted at the centre region
5% to reduce pitting in the final microstructure. All of the solid dilatometer specimens using a 10 kg load. The
samples were etched in 2% Nital. Samples examined by average values of five measurements are reported.
electron backscattered diffraction (EBSD) mapping
were polished with alumina and ion milled for 10–
20 min before scanning.
Microstructures were observed by optical microscopy Results
(OM) using a Zeiss Axioskop 2 MAT and by scanning Microstructures of industrially produced
electron microscopy (SEM-SE) on a JEOL 840. The sheets
vol.-% of various microstructural constituents was deter-
The microstructures of the four industrially produced
mined by a point-counting procedure as per ASTM E562.
sheets are shown in Figs. 2 (OM) and 3 (SEM). They
A grid consisting of 475 points was overlaid on SEM
were characterised as mixtures of PF, GB and B. Follow-
images, and the number of points for a particular con-
ing the classification of Krauss and Thompson,12 these
stituent was recorded manually for five fields. In consider-
three microstructural constituents are defined as follows:
ation of the non-uniform strain distribution, all
microstructural observations of the deformed dilatometer PF – equiaxed and featureless grains
samples were limited to an area of 2 mm × 2 mm at the GB – generally equiaxed grains containing dispersed M/
centre of the specimen. The hollow sample was examined A islands
halfway along its length at the midpoint between the B – packets of ferrite laths with interlath M/A.
outer edge and inner edge of the specimen.
Selected dilatometer specimens were characterised in Examples of each type of microstructural constituent
further detail through EBSD imaging on an FEI FEG- are labelled in Fig. 3. Measurements of vol.% of the
Nova NanoSEM. The magnification and step size were microstructural constituents present in the industrially
varied depending on the resolution required for each produced sheets and Vickers hardness values are reported
microstructure, resulting in a total scanned area between in Table 3. The microstructures of all four sheets comprise
200 μm × 200 μm and 500 μm × 500 μm. predominantly PF and GB, although their relative

3 SEM micrographs of industrially produced sheet microstructures, indicating examples of PF, GB, bainite (B) and M/A
particles

4 Canadian Metallurgical Quarterly 2016

 Scott et al. Microstructural evolution of a hot-rolled microalloyed complex phase steel

Table 3 Measurements of vol.-% of microstructural constituents and hardness for industrially produced sheets

Vol.-% PF Vol.-% GB Vol.-% B Hardness/HV

Material Average +/− Average +/− Average +/− Average +/−

11L 35 3.0 63 3.9 2 1.1 281 13

11T 22 2.0 69 2.8 9 1.5 268 13
18L 45 1.8 54 2.0 1 0.3 273 9
18T 80 7.1 18 6.6 2 0.7 258 6

4 EBSD maps showing prior austenite grains for solid dilatometer samples cooled at 80°C s−1: a Recrystallised and b Pan-
caked (compression axis is vertical)

amounts vary considerably. It can be seen that sheet 18L significant given the relatively large variability of these
has roughly equal amounts of PF and GB, while the measures.
microstructure of 18T is mostly PF. By contrast, GB is
the major constituent in sheets 11L and 11T. One
additional feature evident in Fig. 3 is the prevalence of Microstructures of dilatometer samples
M/A particles in sheet 18L. EBSD maps were produced using Bruker Espirit 2.0 soft-
The hardness values given in Table 3 indicate that the ware for solid dilatometer samples cooled to room temp-
leading edge sheets (11L, 18L) generally exhibit higher erature at 80°C s−1. Grains are defined as regions with a
levels of hardness than the trailing edge sheets (11T, maximum misorientation of 5° and are illustrated in the
18T); however, the differences are not statistically maps in Fig. 4 for a Recrystallised and b Pancaked

5 SEM micrographs showing PF, GB and B in a Recrystallised sample cooled at 1°C s−1 and b Pancaked sample cooled at 3°
C s−1

Canadian Metallurgical Quarterly 2016 5

Scott et al. Microstructural evolution of a hot-rolled microalloyed complex phase steel

starting austenite conditions. For the Recrystallised start- boundaries, often surrounding regions of GB and
ing austenite condition (Fig. 4a), there is evidence of B. The GB constituent contains equiaxed M/A particles
equiaxed austenite grains approximately 75 μm in diam- (∼1 μm) distributed throughout the ferritic matrix. In
eter. For the Pancaked starting condition (Fig. 4b), the B constituent, elongated M/A particles lie on bainitic
elongated austenite grains approximately 70 μm × 30 μm ferrite lath boundaries.
can be seen, which are generally oriented perpendicular The variation in final microstructure with cooling rate
to the compression axis. The microstructures of the con- for the two starting austenite conditions is shown in
tinuously cooled dilatometer samples consisted of PF, Fig. 6. For the Recrystallised austenite condition, there
GB and B in varying proportions. Examples are shown is a mixture of PF and GB at the slowest cooling rate
in Fig. 5. PF is found along prior austenite grain (Fig. 6a). At higher cooling rates, the microstructure is

6 SEM micrographs showing variation in final microstructure with cooling rate for a, c, e Recrystallised and b, d, f Pancaked
starting austenite conditions cooled at a, b 0.5°C s−1; c, d 3°C s−1; e, f 30°C s−1

6 Canadian Metallurgical Quarterly 2016

 Scott et al. Microstructural evolution of a hot-rolled microalloyed complex phase steel

a mixture of GB and B (Fig. 6c and e). For the Pancaked

austenite condition at the slowest cooling rate (0.5°C s−1),
the microstructure is predominantly PF (Fig. 6b). With
increasing cooling rate, there is an increasing amount of
GB (Fig. 6d and f ). A small amount of B is observed at
the highest cooling rate (Fig. 6f ).

CCT diagrams
The dilatometer data were analysed by plotting sample
length change (ΔL) vs. temperature. Deviation from the
linear thermal contraction lines for austenite and ferrite
indicates the start and finish, respectively, of an auste-
nite decomposition reaction. The data points in Figs.
8a and 9a represent all of the start-and-finish tempera-
tures measured at each (constant) cooling rate (includ-
ing repeat runs) for the two starting austenite
conditions. Guided by the final microstructures for
each starting austenite condition and cooling rate
(Fig. 6), full CCT diagrams were generated (Figs. 7a 8 a CCT diagram and b vol.-% of PF, GB and B vs. cooling
and 8a). rate for Pancaked starting austenite condition
The CCT diagram obtained for the Recrystallised
starting austenite condition is shown in Fig. 7a.
The martensite (M) region on the CCT diagram vs. 600°C) when compared with the Recrystallised auste-
indicates possible transformation of residual nite condition. Also, the amount of GB formed at the 30°
austenite to martensite, resulting in the M/A constituent C s−1 cooling rate is much higher for Pancaked austenite
in GB and B. The measurements of final vol.-% (∼80%) in comparison to Recrystallised austenite
PF, GB and B are plotted for each cooling rate in (∼30%).
Fig. 7b.
The CCT diagram obtained for the Pancaked starting
Hardness measurements
austenite condition is shown in Fig. 8a, along with the
measurements of final vol.-% PF, GB and B for each cool- Hardness increases with increasing cooling rate from 0.1
ing rate in Fig. 8b. Comparing Figs. 7 and 8, there are to 5°C s−1, then remains constant at approximately
some important differences in the CCT behaviour for 250 VHN up to the highest cooling rate (70–80°C s−1).
the Recrystallised vs. Pancaked prior austenite conditions. There is no effect of starting austenite condition (Fig. 9).
The Pancaked austenite condition promotes the for-
mation of PF to higher cooling rates (10 vs. 1°C s−1)
and the formation of GB to higher temperatures (650
Discussion
The results of the dilatometer experiments provide a use-
ful starting point for understanding the microstructural
evolution that occurs during industrial hot rolling, con-
trolled-cooling and coiling of the microalloyed CP steel
investigated. With Recrystallised austentite as the starting
condition (Fig. 4a), the final microstructure is predomi-
nantly PF at low cooling rates, GB at intermediate rates
and B at high rates (Fig. 7). When the initial austenite

7 a CCT diagram and b vol.-% of PF, GB and B vs. cooling 9 Variation in Vickers hardness with cooling rate for Recrys-
rate for Recrystallised starting austenite condition tallised and Pancaked starting austenite conditions

Canadian Metallurgical Quarterly 2016 7

Scott et al. Microstructural evolution of a hot-rolled microalloyed complex phase steel

is deformed (Pancaked) (Fig. 4b), the formation of PF is more prevalent in the 11L microstructure, such as (Nb,
extended to higher cooling rates (10 vs. 1°C s−1) and the Ti)(C,N) precipitates.
formation of GB to higher temperatures (650 vs. 600°C)
as seen in Fig. 8. In addition, the PF grain size is smaller
at the 0.5°C s−1 cooling rate (Fig. 6a vs. b), and the Conclusions
amount of GB is higher at the 30°C s−1 cooling rate
(i) CCT diagrams have been produced for a microalloyed
(Fig. 7b vs. 8b). These observations generally agree with
CP steel with Recrystallised and Pancaked prior austenite
recently published work on microstructure evolution
grain structure. Deformation of the austenite grain struc-
and CCT diagrams for similar steel compositions.8,10,12,13
ture promotes the formation of PF to higher cooling rates
The CCT diagram for the Pancaked starting austenite
(10 vs. 1°C s−1) while the formation of GB is extended to
condition (Fig. 8a and b) can be used to interpret the final
higher temperatures (650 vs. 600°C). More GB is present
microstructures observed in the industrially produced
in the pancaked austenite structure at the highest cooling
sheets. However, it must be recognised that the exper-
rate examined (30°C s−1).
imental CCT diagram only approximates the transform-
(ii) An increase in cooling rate results in an increase in
ation conditions for the industrial sheet because it is not
hardness; however, the prior austenite grain condition
possible to reproduce the highly pancaked state of the
has no discernible influence on hardness despite signifi-
austenite in a laboratory dilatometer. This difference is
cant differences in the final microstructure.
apparent from the ‘aligned microstructure’ observed for
(iii) The CCT diagram for the Pancaked starting austenite
the industrially produced sheet (Figs. 2 and 3), which is
condition provides a useful tool for interpreting the
absent in the microstructures of the dilatometer samples
microstructures observed in industrially produced sheet
(Fig. 6). Despite this notable difference, the two-step
subject to different cooling paths. However, direct com-
cooling curves for the industrially produced sheets have
parison is not possible given the significantly higher
been superimposed on the Pancaked CCT diagram in
level of austenite pancaking obtained in an industrial
Fig. 10 for interpretive purposes. Data above 850°C are
hot mill.
not included in order to compare the industrial process
to the dilatometry results; therefore, differences in finish
rolling temperature have not been accounted for. It can Acknowledgements
be seen in Fig. 10 that there are small differences between
The financial support of AUTO21 and ArcelorMittal
the leading (L) and trailing (T) edges for each of the two
Dofasco (AMD) is appreciated, as well as materials pro-
industrial cooling schedules; however, the most signifi-
vided by AMD.
cant difference is that the 10–20s ‘hold’ occurs at a
much higher temperature for sheet 18 (700–750°C) than
for sheet 11 (∼600°C). Recognising that the CCT dia-
gram for the austenite condition exiting the hot mill
References
would be displaced to the left (shorter times), Fig. 10 1. R. D. K. Misra, S. W. Thompson, T. A. Hylton and A. J. Boucek:
‘Microstructures of hot-rolled high-strength steels with significant
can be used to explain the higher levels of PF in sheet
differences in edge formability’, Metall. Mater. Trans. A, 2001, 32,
18 and GB in sheet 11. However, the much lower quan- (13), 745–760.
tity of PF in 18L relative to 18T is an unexpected result. 2. M. Calcagnotto, D. Ponge and D. Raabe: ‘On the effect of manga-
One possible explanation is that the ∼750°C ‘hold’ temp- nese on grain size stability and hardenability in ultrafine-grained fer-
erature for the 18L sheet may have been too high, effec- rite/martensite dual-phase steels’, Metall. Mater. Trans. A, 2012, 43,
(1), 37–46.
tively remaining above the transformation start 3. M. Takahashi: ‘Development of high strength steels for auto-
temperature for PF. An additional discrepancy in the mobiles’, Shinnittetsu Giho, 2003, 2–6.
results is the higher hardness value for 11L vs. 11T in 4. J. Drillet, N. Valle and T. Iung: ‘Nanometric scale investigation of
spite of the higher PF content in the 11L sheet. This find- phase transformations in advanced steels for automotive appli-
cation’, Metall. Mater. Trans. A, 2012, 43, (13), 4947–4956.
ing suggests that additional hardening features may be
5. J. Dykeman: ‘Advanced high strength steel – recent progress,
ongoing challenges, and future opportunities 71–83’, Int. Symp.
New Dev. Adv. High-Strength Sheet Steels. AIST, 2013.
6. K. Hasegawa, K. Kawamura, T. Urabe and Y. Hosoya: ‘Effects of
microstructure on stretch-flange-formability of 980 MPa grade
cold-rolled ultra high strength steel sheets’, ISIJ Int., 2004, 44, (3),
603–609.
7. E. Yasuhara, A. Tosaka, O. Furukimi, T. Uchiyama and N.
Yamada: ‘High-strength hot-rolled steel sheet having excellent
stretch flangeability, and method of producing the same’,
US6364968 B1, 2002.
8. R. D. K. Misra, H. Nathani, J. E. Hartmann and F. Siciliano:
‘Microstructural evolution in a new 770 MPa hot rolled Nb–Ti
microalloyed steel’, Mater. Sci. Eng. A, 2005, 394, (1–2), 339–352.
9. K. Kamibayashi, Y. Tanabe, Y. Takemoto, I. Shimizu and T.
Senuma: ‘Influence of Ti and Nb on the strength-ductility-hole
expansion ratio balance of hot-rolled low-carbon high-strength
steel sheets’, ISIJ Int., 2012, 52, (1), 151–157.
10 Industrial cooling schedules superimposed on CCT dia- 10. E. Joon Chun, H. Do, S. Kim, D.-G. Nam, Y.-H. Park and N. Kang:
gram for Pancaked starting austenite conditions ‘Effect of nanocarbides and interphase hardness deviation on

8 Canadian Metallurgical Quarterly 2016

 Scott et al. Microstructural evolution of a hot-rolled microalloyed complex phase steel

stretch-flangeability in 998 MPa hot-rolled steels’, Mater. Chem. transformation in Nb-Mo microalloyed steels’, Metall. Mater.
Phys., 2013, 140, (1), 307–315. Trans. A, 2011, 42, (12), 3729–3742.
11. H. Scott, L. Siqueira, G. Sidhu, F. Fazeli, X. Wang, A. K. Pilkey and 14. R. Y. Zhang and J. D. Boyd: ‘Bainite transformation in deformed
J. D. Boyd: ‘Microstructure evolution during thermomechanical austenite’, Metall. Mater. Trans. A, 2010, 41, (6), 1448–1459.
processing of a high strength sheet steel 1171–1178’, Proc. Int. 15. N. Isasti, D. Jorge-Badiola, M. L. Taheri and P. Uranga: ‘Phase
Conf. Solid-Solid Phase Transform. Inorg. Mater. 2015, 2015. transformation study in Nb-Mo microalloyed steels using dilatome-
12. G. Krauss and S. W. Thompson: ‘Ferritic microstructures in continu- try and EBSD quantification’, Metall. Mater. Trans. A, 2013, 44, (8),
ously cooled low- and ultralow-carbon steels’, ISIJ Int., 1995, 35, 3552–3563.
(8), 937–945. 16. J. Hu, L.-X. Du and J.-J. Wang: ‘Effect of cooling procedure on
13. N. Isasti, D. Jorge-Badiola, M. L. Taheri, B. López and P. Uranga: microstructures and mechanical properties of hot rolled Nb–Ti bai-
‘Effect of composition and deformation on coarse-grained austenite nitic high strength steel’, Mater. Sci. Eng. A, 2012, 554, 79–85.

Canadian Metallurgical Quarterly 2016 9