Materials Science & Engineering A 731 (2018) 149–155

Contents lists available at ScienceDirect

Materials Science & Engineering A

journal homepage: www.elsevier.com/locate/msea

Effect of hot rolling temperature on the microstructure and mechanical T

properties of ultra-low carbon medium manganese steel
⁎
S. Yua, L.X. Dua, , J. Hua, R.D.K. Misrab
a
The State Key Laboratory of Rolling and Automation Northeastern University, Shenyang 110819, China
b
Laboratory for Excellence in Advanced Steel Research, Department of Metallurgical, Materials and Biomedical Engineering, University of Texas at El Paso, TX 79968-
0521, USA

A R T I C LE I N FO A B S T R A C T

Keywords: The effect of hot rolling temperature on microstructural evolution and mechanical properties of 0.05C-5Mn (wt
Medium manganese steel %) steel was investigated. With the decrease of hot rolling temperature from 1150 °C to 900 °C, the average size
Reversed austenite of prior austenite grain was decreased from ~ 40 µm to ~20 µm, and average width of martensite lath was
Hot rolling temperature refined from ~ 400 nm to ~ 250 nm. The morphology of reversed austenite was changed from film-like shape to
Mechanical properties
block-like shape, and the volume fraction of reversed austenite was increased from 14.75% to 16.28% because of
TRIP effect
the stimulated nucleation in refined prior austenite grains by rolling at lower temperature. The yield strength,
tensile strength and the total elongation were simultaneously increased from 673 MPa, 805 MPa, and 27.5% to
704 MPa, 846 MPa, and 31%, respectively. The low temperature impact toughness was significantly enhanced
from 133 J to 222 J at − 40 °C. The significant improvement in mechanical properties by lower hot rolling
temperature were attributed to both the refined tempered martensite lath with higher density of high mis-
orientation grain boundaries and the active TRIP effect of reversed austenite with suitable mechanical stability.

1. Introduction
Light weight and safety is an important requirement of the industry
[1–3]. Enhanced strength and toughness is usually obtained by addition
of expensive alloying elements and adjustment of thermomechanical
controlled processing, which unfortunately increases the cost and limits
the development [4–7]. In comparison with conventional high strength
steels, the medium manganese (5–8 wt%) steels have attracted sig-
nificant attention because of excellent combination of high strength,
excellent ductility and impact toughness, especially at low temperature.
It has potential applications in automobiles, offshore platform, bridges
and other structural component.
The superior mechanical properties of medium manganese steel are
mainly attributed to the TRIP effect and the cooperative deformation of
ferrite. For example, the Fe-0.2C-11Mn as-cold-rolled steel with a high
volume fraction of austenite after intercritical annealing exhibited an
excellent combination of TE of 43–70% and UTS of 900–1087 MPa [8].
It is recognized that stable retained austenite is beneficial to toughness
by restricting brittle fracture without undergoing transformation
[9–11]. For example, an excellent combination of ductility (total
elongation of ~37.3%) and toughness (Charpy v-notch impact energy
of ~ 158 J at −80 °C) was obtained by the design of fine ferrite-
austenite microstructure [12]. Zou et al. [13] suggested that the me-
tastable reversed austenite significantly improved the low-temperature
toughness by relieving the local stress concentration and consuming the
tip energy of the propagating crack. However, the mechanism of re-
tained austenite in improving toughness and ductility is still unclear.
Although there have been extensive studies on medium manganese
steel which focused on chemical composition and heat treatments, there
are limited studies on hot rolling temperature. Therefore, the objective
of this paper is to study the effect of hot rolling temperature on the
microstructure and mechanical properties of ultra-low carbon medium
manganese steel. Excellent properties and desired microstructure were
obtained via low temperature rolling and annealing treatment.
2. Experimental procedure
The chemical composition of the steel was Fe-0.05C-0.2Si-5Mn-
0.38Cr 0.28Cu-0.28Ni-0.22Mo (wt%). The experimental steel was
melted in a high-frequency vacuum induction furnace and cast into
ingot of thickness ~ 0 mm. The ingot was homogenized at 1200 °C for
2 h, and hot rolled to 12 mm thickness at 1150 °C and 900 °C (desig-
nated as HR1150 and LR900) via seven passes and cooled to room
temperature in water. The two plates were intercritically annealed at

⁎
Corresponding author.
E-mail address: dulx@ral.neu.edu.cn (L.X. Du).

https://doi.org/10.1016/j.msea.2018.06.020
Received 6 March 2018; Received in revised form 4 June 2018; Accepted 5 June 2018
Available online 06 June 2018
0921-5093/ © 2018 Elsevier B.V. All rights reserved.
S. Yu et al.
Fig. 1. Optical micrographs of prior austenite: (a) HR1150; (b) LR900.
Fig. 2. Martensite lath microstructure: (a) HR1150; (b) LR900.
650 °C for 50 min, and air cooled to room temperature. The specimens
were mechanically polished and etched by a mixed solution composed
of 0.5% xylene, 3% dodecyl-benzene sulfonic acid, and 96.5% saturated
picric acid at 65 °C to reveal the boundaries of prior austenite grains.
The mean grain sizes of the specimens were determined by a linear
intercept method. The specimens with dimensions of 6 mm diameter
and 30 mm length were cut from the annealed plate along the rolling
direction. The tensile tests were conducted on a Shimadzu AG-X uni-
versal testing machine at a speed of 3 mm/min. The impact test were
performed in the temperature range of 20 to − 80 °C with standard
Charpy impact specimens (dimensions: 10 × 10 × 55 mm3) machined
along the rolling direction using Instron Dynatup 9200 series in-
strumented drop weight impact tester, consistent with ASTME23 spe-
cification.
The specimens used for EBSD were electrolytically polished with a
solution composed of 8% perchloric acid and 92% alcohol at room
temperature, and EBSD analysis was carried out using a Zesis Ultra55
scanning electron microscope (SEM). For substructure observation, the
3 mm diameter disks were ground to thickness of 40 µm and then twin-
jet electropolished using a solution consisting of 8% perchloric acid and
92% ethanol at − 30 °C, and their microstructures were examined by
FEI Tecnai G2 F20 transition electron microscope (TEM). The volume
fraction of reversed austenite was determined by a D/max 2400 XRD
using a Cu-Kα at room temperature and the integrated intensities of
150
Materials Science & Engineering A 731 (2018) 149–155
(200)γ, (220)γ, (311)γ, (200)α, and (211)α peaks were used to quantify
the amount of reversed austenite using Eq. (1) [14]:
Vγ = 1.4Iγ /(Iα + 1.4 Iγ) (1)
where Vγ is the volume fraction of reversed austenite, Iγ is the in-
tegrated intensity of the austenite peaks, and Iα is the integrated in-
tensity of the ferrite peaks.
3. Results and discussion
3.1. Microstructure
Fig. 1 provides prior austenite microstructure for the HR1150 and
LR900 steels, showing the equiaxed prior austenite grains. Hence, it can
be deduced that the recrystallization has occurred at both 1150 °C and
900 °C. The average prior austenite grain size is refined from ~ 40 µm
to ~ 20 µm as the hot rolling temperature decreases from 1150 °C to
900 °C.
Fig. 2 exhibits the morphology of as-hot-rolled steel plate subjected
to different rolling temperature. Both steels show fully martensite
lathes, and the average values of lath width are ~ 400 nm and ~
250 nm for the HR1150 and LR900 steels, respectively. Moreover, the
dislocation density in the LR900 steel is pronouncedly higher than that
in the HR1150 steel.
S. Yu et al. Materials Science & Engineering A 731 (2018) 149–155

Fig. 3. TEM microstructure of experimental steels: (a) bright-field image of specimen from annealed HR1150; (b) dark-field image of specimen from annealed
HR1150; (c) bright-field image of specimen from annealed LR900; (d) dark-field image of austenite in specimen from annealed LR900.

Fig. 4. XRD spectra of annealed HR1150 and annealed LR900.
The HR1150 and LR900 steels were further annealed at 650 °C for
50 min, and their annealed microstructures are shown in Fig. 3. The
annealed HR1150 and annealed LR900 steels consist of sub-micrometer
reversed austenite and tempered martensite. However, the morphology
of reversed austenite in the annealed HR1150 and annealed LR900
steels is different. The film-like reversed austenite with average width
of ~ 100 nm and small amount of block-like reversed austenite can be
observed in the annealed HR1150 steel. Whereas the irregular block-
like reversed austenite with average width of ~ 160 nm and small
amount of film-like reversed austenite can be observed in the annealed
LR900 steel. When the hot-rolled steel was intercritically annealed, the
reverse transformation from tempered martensite to austenite was
mainly along the tempered martensite boundaries, resulting in lath-
shaped grain morphology [15,16]. However, the recrystallization of
martensite matrix and reverse transformation from martensite to aus-
tenite may occur simultaneously due to the high dislocation density
introduced by low temperature rolling, resulting in bock-like reversed
austenite [16,17]. During intercritical annealing process, the disloca-
tion density significantly decreased due to climb and annihilation.
The XRD spectra of annealed HR1150 and annealed LR900 steels
are presented in Fig. 4. The volume fraction of reversed austenite was

151
S. Yu et al.
Fig. 5. Characteristics of annealed experimental steels analyzed by EBSD: (a) reversed austenite distribution of annealed HR1150; (b) grain boundary misorientation
distribution of annealed HR1150; (c) reversed austenite distribution of annealed LR900 (d) grain boundary misorientation distribution of annealed LR900.
Fig. 6. Engineering stress-strain curves of annealed experimental steels.
estimated to be 14.75% and 16.28% for annealed HR1150 and an-
nealed LR900 steels, respectively. The reversed austenite volume frac-
tion of annealed LR900 is higher than annealed HR1150. The reversed
austenite preferentially nucleated at prior austenite boundaries and
then at martensite lath boundaries [18,19]. On the one hand, the nu-
cleation sites for reverse transformation can be strongly increased by
lowering rolling temperature because of the refinement of prior auste-
nite grains and martensite laths. On the other hand, relatively high
dislocation density can promote the diffusion of C and Mn, which en-
hances the kinetics of reversed austenite transformation and increases
the austenite volume fraction.
EBSD micrographs of annealed experimental steels are presented in
Fig. 5. The blue lines represent high angle grain boundaries with the
152
Materials Science & Engineering A 731 (2018) 149–155
misorientation larger than 15°. The annealed steels consist of small
packets with fine lath and reversed austenite. The reversed austenite is
mainly distributed at the packets boundaries and prior austenite
boundaries because of their high interfacial energy. The packet
boundaries and prior austenite boundaries are primarily characterized
by high misorientation grain boundaries (Fig. 5a and c). On the other
hand, the deformation band or substructure is partly remained at re-
latively lower finishing rolling temperature, and these defects can dis-
appear by recrystallization during annealing, resulting in the increase
of high misorientation grain boundaries (Fig. 5b and d). The high
misorientation grain/packet boundaries are effective barrier which ar-
rest the propagation of cleavage microcracks [20,21], whereas the low
misorientation grain boundaries have less ability to deviate the clea-
vage crack [22].
3.2. Tensile properties
Fig. 6 shows the engineering stress-strain curves of annealed ex-
perimental steels. The curve of annealed HR1150 presents continuous
yielding. The yield strength (YS), ultimate tensile strength (UTS), and
total elongation (EL) is 673 MPa, 805 MPa, and 27%, respectively.
However, the curve of annealed LR900 exhibits a yield platform. The YS
and UTS increase from 673 MPa and 805–704 MPa and 847 MPa, re-
spectively. The total elongation increases from 27.5% to 31%.
As seen in the Fig. 7, the volume fraction of reversed austenite be-
fore and after tensile deformation is 14.75%, 10.08% and 16.28%,
10.70% for annealed HR1150 and annealed LR900, respectively. The
reversed austenite transformation ratio of annealed LR900 is higher
than that of annealed HR1150 during tensile deformation. This is be-
cause that the block-like reversed austenite is easily experienced TRIP
effect at identical strain due to its poor stability. However, the relatively
S. Yu et al.
Fig. 7. XRD spectra of (a) annealed HR1150 and (b) annealed LR900 before and after tensile deformation.
Fig. 8. Schematic diagram of typical deflection-load curve.
Fig. 9. The relationship between total energy, crack initiation energy and test
temperature during Charpy impact test.
higher reversed austenite transformation ratio is not the only reason for
the enhanced strength, the refinement of tempered martensite matrix is
also a crucial factor. Hence, the outstanding tensile property is mainly
achieved by the combination of finer tempered martensite matrix and
TRIP effect of reversed austenite which can further relax stress con-
centration and hence delay the onset of necking.
153
Materials Science & Engineering A 731 (2018) 149–155
3.3. Impact property
The typical curve of load versus deflection consists of five stages:
elastic deformation stage, plastic deformation stage, ductile deforma-
tion stage, brittle propagation stage and ductile fracture stage, as shown
in Fig. 8. The fracture process consists of crack initiation and crack
propagation. The crack initiation occurred when the load reached the
Pm (as shown in Fig. 8). Thus, the crack initiation stage consists of
elastic deformation stage and plastic deformation stage. The crack in-
itiation point Pm represents the difficulty of crack formation and the
sum of corresponding absorbed energy is regarded as the crack initia-
tion energy. Similarly, the sum of relevant absorbed energy in sub-
sequent three deformation stages is considered as the crack propagation
energy.
The total impact energy and crack initiation energy of annealed
steels as a function of test temperature is presented in Fig. 9. The total
energy of annealed LR900 and annealed HR1150 is 259 J and 233 J at
room temperature, respectively. The impact absorbed energy of an-
nealed LR900 decreases gradually as the testing temperature decreases
from 20 °C to − 40 °C, and it reaches 222 J even at − 40 °C. However,
there is steep decrease in impact absorbed energy for the annealed
HR1150, and it is only 133 J at − 40 °C. With further lowing testing
temperature, the impact absorbed energy decreases for the both steels,
but the impact absorbed energy of annealed LR900 is greater than that
of annealed HR1150.
Fig. 10 shows the load-deflection and absorbed energy-deflection
plots for annealed HR1150 and annealed LR900 at − 40 °C. The peak
load and crack initiation energy is 22.6 kN, 51 J and 22.5 kN, 52 J,
respectively. It indicates that the variation in the elastic deformation
and plastic deformation is unchanged. However, there is obvious dif-
ference in the subsequent crack propagation process. The crack pro-
pagation energy of annealed HR1150 and annealed LR900 is 81 J and
181 J, respectively. It indicates that the crack propagation is sig-
nificantly inhibited during the ductile propagation stage for annealed
LR900.
The curve of ductile propagation sharply decreases in the range
2.5–6 mm for annealed HR1150 (Fig. 10a). The ductile propagation
stage is reduced, and the brittle propagation is extended, resulting in
the deterioration of toughness. Compared with the curve of annealed
HR1150, the ductile propagation stage is extended in the curve of an-
nealed LR900 (Fig. 10b). The deflection of ductile propagation stage is
mainly in the range of 6–12.5 mm and 12.6–23 mm. When the deflec-
tion reaches ~ 5 mm, the curve of ductile deformation drastically de-
creases to ~ 12.5 kN. In the range of 6–12.5 mm, the decreasing rate of
the curve slows down until the deflection reaches ~12.5 mm, where the
curve drastically drops again, and after that, in the range of
12.6–23 mm, the decreasing rate of the curve further slows down, until
S. Yu et al. Materials Science & Engineering A 731 (2018) 149–155

Fig. 10. The relationship of between load, impact energy and deflection at − 40 °C: (a) annealed HR1150; (b) annealed LR900.

Fig. 11. XRD spectra of (a) annealed HR1150 and (b) annealed LR900 before and after impact deformation.

Fig. 12. SEM fractures of experimental steel impact tested at − 40 °C: (a) annealed HR1150; (b) annealed LR900.

the deflection reaches ~ 23 mm. The brittle stage decreases due to the
ductility enhancement occurs in the ductility stage. It indicates that the
crack propagation is effectively inhibited by high misorientation grain
boundaries in annealed LR900. The metastable reversed austenite also
plays a crucial role in enhancing the toughness. Comparing the volume
fraction of reversed austenite before and after impact test in Fig. 11, the
metastable block-like reversed austenite in annealed LR900 remarkably
induced TRIP effect. TRIP effect can relax the residual stress (RS) within
marteniste transformation, which can moderate RS distribution [23].
Moreover, the TRIP effect can also consume energy during crack

154
S. Yu et al.
propagation. Thus, the enhanced toughness of annealed LR900 is at-
tributed to the low temperature deformation which is beneficial to re-
fine prior austenite grain and increase the density of high misorienta-
tion grain boundaries, and suitable stability of reversed austenite can
relax RS and consume the energy during crack propagation.
The SEM fractographs of annealed steels impact tested at − 40 °C
are showed in Fig. 12. The fracture of specimen annealed HR1150
mainly comprises of small and flat uniform dimples. The fracture of
specimen annealed LR900 reveals large and deep dimples, indicating
better low-temperature toughness.
4. Conclusions
In this study, the microstructure characteristics and mechanical test
were performed to reveal the effect of hot rolling temperature on the
mechanical properties of ultra-low carbon medium steel, and the fol-
lowing conclusions are drawn.
(1) The prior austenite grains were refined from ~ 40 µm to ~ 20 µm as
the hot rolling temperature was reduced from 1150 °C to 900 °C.
Consequently, there was also a decrease in martensite lath width by
lowering hot rolling temperature. After annealing at 650 °C for
50 min, the morphology of reversed austenite was mainly film-like
shape with volume fraction of 14.75% in annealed HR1150 steel,
and was mainly block-like shape with volume fraction of 16.28% in
annealed LR900 steel. Compared to annealed HR1150 steel, an-
nealed LR900 steel possessed higher volume fraction of reversed
austenite due to refinement of quenched martensite lath which of-
fered more nucleation sites for the transformation of reversed
austenite.
(2) There was a slight increase in strength and ductility by lowering hot
rolling temperature. However, the impact absorbed energy can be
strongly enhanced. Especially for the testing temperature of
− 40 °C, it increased from 133 J to 222 J by lowering hot rolling
temperature from 1150 °C to 900 °C.
(3) The high density of high misorientation grain boundaries and re-
versed austenite with suitable stability significantly improved the
toughness of the experimental steel. The enhanced toughness was
attributed to the increased crack propagation energy from 81 J to
181 J, and the impact fractures was changed from small and flat
dimples to large and deep dimples.
Acknowledgments
The authors appreciate the financial support by National High-tech
R&D Program (863 Program) [Grant No. 2015AA03A051], the National
Natural Science Foundation of China [Grant No. 51604072]. R.D.K.
Misra also acknowledges continued collaboration with Northeastern
University as Honorary Professor in providing guidance in research.
References
[1] Z.B. Jiao, J.H. Luan, M.K. Miller, Precipitation mechanism and mechanical
155
Materials Science & Engineering A 731 (2018) 149–155
properties of an ultra-high strength steel hardened by nanoscale NiAl and Cu par-
ticles, Acta Mater. 97 (2015) 58–67.
[2] J.G. Speer, E.D. Moor, K.O. Findley, Analysis of microstructure evolution in
quenching and partitioning automotive sheet steel, Metall. Mater. Trans. A 42
(2011) 3591–3601.
[3] A.G. Kostryzhev, O.O. Marenych, C.R. Killmore, Strengthening mechanisms in
thermomechanically processed NbTi-Microallyed steel, Metall. Mater. Trans. A 46
(2015) 3470–3480.
[4] P. Cizek, B.P. Wynne, C.H.J. Davies, Effect of composition and austenite deforma-
tion on the transformation characteristics of low-carbon and ultralow-carbon mi-
croalloyed steels, Mater. Trans. A 33 (2002) 1331–1349.
[5] S. Lee, B.C. De Cooman, On the selection of the optimal intercritical annealing
temperature for medium Mn TRIP steel, Metall. Mater. Trans. A 44 (2013)
5018–5024.
[6] C.L. Zhang, D.Y. Cai, Y.H. Wang, Effects of deformation and Mo,Nb,V,Ti continuous
cooling, transformation in Cu-O-Cr-Ni-Mo weathering steels, Mater. Charact. 59
(2008) 1638–1642.
[7] N.Y. Zolotorevsky, S.N. Panpurin, A.A.Z. isman, Effect of ausforming and cooling on
the orientation relationship in martensite and bainite of low carbon steels, Mater.
Charact. 107 (2015) 278–282.
[8] Z.H. Cai, H. Ding, R.D.K. Misra, Austenite stability and deformation behavior in a
cold-rolled transformation-induced plasticity steel with medium manganese con-
tent, Acta Mater. 84 (2015) 229–236.
[9] J. Chen, M.Y. Lv, S. Tang, Correlation between mechanical properties and retained
austenite characteristics in a low-carbon medium manganese alloyed steel plae,
Mater. Charact. 106 (2015) 108–111.
[10] G. Lacroix, T. Pardoen, P.J. Jacques, The fracture of TRIP-assisted multiphase steels,
Acta Mater. 56 (2008) 3900–3913.
[11] J. Hu, L.X. Du, G.S. S, The determining role of reversed austenite in enhancing
toughness of a novel ultra-low carbon medium manganese high strength steel, Scr.
Mater. 104 (2015) 87–90.
[12] J. Chen, M.Y. Lv, Z.Y. Liu, Combination of ductility and toughness by the design of
fine ferrite/tempered martensite-austenite microstructure in a low carbon medium
manganese alloyed steel plate, Mater. Sci. Eng. A 648 (2015) 51–56.
[13] Y. Zou, Y.B. Xu, Z.P. Hu, Austenite stability and its effect on the toughness of a high
strength ultra-low carbon medium manganese steel plate, Mater. Sci. Eng. A 675
(2016) 153–163.
[14] Z. Li, D. Wu, Effects of deformation and subsequent austempering on the me-
chanical properties of Si-Mn TRIP steels, ISIJ Int. 46 (2006) 121–128.
[15] M. Calcagnotto, D. Ponge, E. Demir, Orientation gradients and geometrically ne-
cessary dislocations in ultrafine grained dual-phase steels studied by 2D and 3D
EBSD, Mater. Sci. Eng. A 527 (2010) 2738–2746.
[16] J. Han, A.K. Silva, D. Ponge, The effects of prior austenite grain boundaries and
microstructural morphology on the impact toughness of intercritically annealed
medium Mn steel, Acta Mater. 65 (2014) 215–228.
[17] J. Han, S.J. Lee, J. Gil, The effects of the initial martensite microstructure on the
microstrcture and tensile properties of intercritically annealed Fe-9Mn-0.05C steel,
Acta Mater. 78 (2014) 369–377.
[18] C. Wang, J. Shi, C.Y. Wang, Development of ultrafine lamellar ferrite and austenite
duplex structure in 0.2C5Mn steel during annealing, ISIJ Int. 51 (2011) 651–656.
[19] J. Han, Y.K. Lee, The effects of the heating rate on the reversed transfomation
mechanism and the phase stability of reversed austenite in medium Mn steels, Acta
Mater. 67 (2014) 354–361.
[20] A. Lambert, X. Garat, T. Sturel, Application of acoustic emission to the study of
cleavage fracture mechanism in a HSLA steel, Scr. Mater. 43 (2000) 161–166.
[21] L.Y. Lan, C.L. Qiu, D.W. Zhao, Microstructural characteristics and toughness of the
simulated coarse grained heat affected zone of high strength low carbon bainitic
steel, Mater. Sci. Eng. A 529 (2011) 192–200.
[22] A.F. Gourgues, Electron backscatter diffraction and cracking, Mater. Sci. Technol.
18 (2002) 119–133.
[23] W.C. Jiang, W. Chen, W. Woo, S.T. Tu, X.C. Zhang, V. Em, Effects of low-tem-
perature transformation and transformation-induced plasticity on weld residual
stresses: numerical study and neutron diffraction measurement, Mater. Des. 147
(2018) 65–79.