Materials Science & Engineering A 648 (2015) 51–56

Contents lists available at ScienceDirect

Materials Science & Engineering A

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

Combination of ductility and toughness by the design of fine

ferrite/tempered martensite–austenite microstructure in a low carbon
medium manganese alloyed steel plate
Jun Chen a,n, Meng-yang Lv b, Zhen-yu Liu a, Guo-dong Wang a
a
State Key Laboratory of Rolling and Automation, Northeastern University, Shenyang 110819, People's Republic of China
b
School of Materials and Metallurgy, Northeastern University, Shenyang 110819, People's Republic of China

art ic l e i nf o a b s t r a c t

Article history: Based on the mechanisms of plasticity and toughness, a low carbon medium manganese alloyed steel
Received 13 June 2015 plate with the outstanding combination of ductility and toughness was designed. The microstructure
Received in revised form characteristics, strain hardening behavior and toughening mechanisms were investigated in detail. The
6 September 2015
microstructure is composed of ferrite/tempered martensite and austenite after heat treating at 650 °C for
Accepted 8 September 2015
2 h. The retained austenite shows multi-type morphology, multi-scale size and relatively poor me-
Available online 11 September 2015
chanical stability, resulting in multi-peak strain hardening behavior and sufficient TRIP effect. The ex-
Keywords: cellent strain hardening capacity and ductility are hence achieved. The excellent low temperature
Medium manganese steel toughness is also achieved due to the relatively lower concentrations of Mn and C, the formation of fine
Retained austenite
ferrite grains and a small volume fraction of fine retained austenite. The steel hence possesses the ex-
Strain hardening behavior
cellent ductility (the total elongation is
37.3%) and low temperature toughness (the Charpy V-notch
Mechanical properties
impact energy is
158 J at  80 °C).
& 2015 Elsevier B.V. All rights reserved.

1. Introduction austenite increase with increasing the concentrations of Mn and C

Recently, medium Mn transformation induced plasticity (TRIP)
steels containing 5–8 mass% Mn have been attracting more at-
tentions due to the outstanding combination of strength and
ductility [1–8]. Moreover, Since Miller proposed 0.11C–5.7Mn steel
[9], the reverse transformation behavior, stabilization mechanisms
of retained austenite, strain hardening behavior and tensile
properties had been further understood [1–4,9–14]. It has been
recognized that only sufficiently stable retained austenite in gen-
eral leads to strong toughening [15,16]. Whereas the high stable
retained austenite has few contributions to strain hardening ca-
pacity, resulting in the dramatic reduction in ductility. The role of
retained austenite in improving ductility and toughness is incon-
sistent. However, it has been demonstrated that the high strain
hardening capacity and ductility can be effectively enhanced by
the TRIP effect. The sufficient TRIP effect of metastable austenite
indicates that its contribution to low temperature toughness is
very poor. Hence, some other toughening mechanisms should be
introduced. It is well known that the Mn and C are strong auste-
nite stabilizers and the volume fraction and stability of retained
n
Corresponding author. Fax: þ 86 24 2390 6472.
E-mail address: cjer19841011@163.com (J. Chen).
[4,10,14], resulting in the attractive strain hardening capacity and
ductility. However, the high Mn and C steels with a body-centered
cubic crystal structure always have relatively high ductile brittle
transition temperatures [17]. Therefore, the Mn and C concentra-
tions should be reduced. On the other hand, we also add some Ni
and refine grains to enhance toughness.
The objective of the present work was therefore to investigate
whether the excellent combination of ductility and low tempera-
ture toughness can be achieved by the design of fine ferrite/tem-
pered martensite–austenite microstructure in a low carbon med-
ium manganese alloyed steel plate. Particular attentions were to
elucidate strain hardening behavior and toughening mechanisms.
This study was done by a quantitative analysis of the morphology
and size of retained austenite.
2. Experimental procedure
The chemical composition of the steel is Fe–0.05C–0.23Si–
3.3Mn–0.003P–0.001S–1.4Ni (in mass%), which has relatively
lower Mn and C concentrations by comparison with other medium
manganese steels. The steel was melted by means of a high-fre-
quency vacuum induction furnace in an Ar protective atmosphere
and cast into an iron mold. The ingot was homogenized at 1200 °C

http://dx.doi.org/10.1016/j.msea.2015.09.032
0921-5093/& 2015 Elsevier B.V. All rights reserved.
52 J. Chen et al. / Materials Science & Engineering A 648 (2015) 51–56

for 2 h, then hot rolled in six passes to a 12 mm thickness plate, (220)γ, (113)γ, (200)α, (112)α and (220)α diffraction peaks [18,19].
and finally cooled to room temperature in water. The plate was Transmission electron microscopy (TEM) specimens were me-
further isothermally treated at 650 °C for 2 h to obtain a con- chanically thinned to
50 μm thickness from both sides using
siderable volume fraction of retained austenite, and then water silica papers, followed by punching to prepare round disks with
quenched to room temperature. 3 mm diameter. Subsequently, the disks were further thinned
Standard round tensile samples with the gauge length of using a twin-jet electropolisher (Struers TenuPol-5) under a vol-
40 mm and diameter of 8 mm were sampled from above men- tage of
32 V at a temperature of
 10 °C. The electrolyte con-
tioned tempered steel plate along the rolling direction. Interrupted sisted of
9% perchloric acid and
91% absolute ethyl alcohol.
tensile tests were conducted on a CMT-5105 tensile tester at room The thin foils were examined on a field-emission transmission
temperature at a cross beam speed of 3 mm/min. The standard electron microscopy (FEI Tecnai G2 F20) operated at 200 kV and
Charpy V-notch (CVN) impact samples with the size of well investigated using selected area electron diffraction (SAED).
10  10  55 mm3 were also prepared along the rolling direction,
and the impact tests were conducted on a JBW-500 impact tester
at the temperatures of 15,  40 and 80 °C. Moreover, the average 3. Results and discussion
yield strength, ultimate tensile strength, total elongation and CVN
impact energy were determined using three tensile or CVN impact 3.1. Mechanical properties and multi-peak strain hardening behavior
test data.
Specimens were cut from the tempered steel plate and their The true stress–strain curve and the corresponding strain
surfaces along thickness direction and rolling direction were me- hardening rate of the steel are shown in Fig. 1. The steel shows an
chanically polished and then etched in 4% nital solution for optical ultimate tensile strength (UTS) of
763 MPa, lower yield strength
microscopy (OM, LEICA DMIRM) and scanning electron microscopy of
530 MPa and total elongation (TE) of
37.3%, exhibiting a low
(SEM, Zeiss Ultra 55). The quantitative X-ray diffraction (XRD) yield ratio of
0.69 and superior combination of UTS and TE. It is
analysis was used to determine the volume fractions of retained interesting to note that the clear steps (indicated as A to G in the
austenite. Meanwhile, for accurate measurement, the specimens inset in Fig. 1a) in true stress–strain curve are visible below the
were cut from the gauge section of different tensile samples with true strain of 0.08. Beyond this point, the true stress continuously
different strains, mechanically polished, and then electropolished increases as the true strain increases. On the other hand, the strain
in a mixture consisting of
12.5% perchloric acid and
87.5% hardening rate was derived from the true stress–strain curve, as
absolute ethyl alcohol at room temperature to remove surface shown in Fig. 1b, exhibiting that many peaks in strain hardening
strain layer. Furthermore, the volume fractions of retained auste- rate curve are observed and the peak value is far higher than that
nite were measured using the integrated intensities of (200)γ, reported in previous studies [20,21]. A steep increase in true stress

1000 50k
80.0k B

800 40k 60.0k

Strain hardening rate, MPa
True stress, MPa

800
D
40.0k
600 750 30k C
G A
E
700 F
20.0k
400 20k F
650 E G
D 0.0
600 0.00 0.02 0.04 0.06 0.08 0.10

200 550
C 10k
B
A
500
0.00 0.02 0.04 0.06 0.08 0.10
0 0
0.00 0.04 0.08 0.12 0.16 0.20 0.24 0.00 0.05 0.10 0.15 0.20 0.25
True strain True strain

660
Strain hardening rate, MPa True stress, MPa

650 D1 D2 D3 D4
640
630
620
610

40.0k

20.0k

0.0
0.036 0.038 0.040 0.042 0.044 0.046
True strain
Fig. 1. Tensile stress–strain curve and corresponding strain hardening rate curve of the steel. (a) True stress–strain curve; (b) multi-peak strain hardening behavior and
(c) local magnification map taken from representative region D.
 J. Chen et al. / Materials Science & Engineering A 648 (2015) 51–56
and corresponding high instantaneous strain hardening rate are
mainly attributed to relatively rapid deformation induced mar-
tensite transformation of a considerable volume fraction of re-
tained austenite at each step. After that, the true stress gradually
increases and the strain hardening rate nearly remains unchanged
due to the pause of TRIP effect or the deformation induced mar-
tensite transformation of a little retained austenite. However,
when the true stress reaches another critical stress to activate the
transformation of a considerable volume fraction of retained aus-
tenite, a new round of the TRIP effect vigorously occurs, resulting
in the formation of a new step in true stress–strain curve and a
new peak in strain hardening rate curve.
In addition, the representative region D was magnified and
shown in Fig. 1c, showing that the step D can be further divided
into four stages. In the first stage (D1), both true stress and strain
hardening rate gradually increase with increasing the true strain,
followed by a dramatic decrease, which is attributed to local stress
relaxation caused by a considerable volume fraction of retained
austenite transformed into martensite. After that, the strain
hardening rate steeply increases up to a peak value (
40 GPa) and
then dramatically decreases as the true strain increases at the
second stage (D2). The third stage (D3) is similar to the second
stage (D2), only the peak value decreases to
20 GPa. In the fourth
stage (D4), the true stress gradually increases in a low strain
hardening rate until it reaches another critical stress to activate
the transformation of retained austenite with relatively high sta-
bility as the true strain increases.
The CVN impact energies of 247 J (at 15 °C), 243 J (at  40 °C)
and 158 J (at 80 °C) were measured, exhibiting superior low
temperature impact toughness. It is commonly thought that the
impact toughness can be improved by enhancing the stability of
retained austenite, but its contribution to tensile properties are
usually weakened [22,23]. Therefore, the combination of excellent
ductility and low temperature toughness can be hardly achieved.
However, in the present work, the steel possesses not only su-
perior combination of UTS and TE, strain hardening capacity and
lower yield ratio of 0.69, but also excellent low temperature im-
pact toughness.
3.2. Ferrite/tempered martensite–austenite microstructure
Fig. 2 provides the morphology of the steel after heat treating
at 650 °C for 2 h and Fig. 3 presents X-ray diffraction in as-hot
rolled steel and the heat treated steel. Fig. 3 shows that the mi-
crostructure is full martensite before heat treating. As we know
Fine ferrite grains
Tempered martensite
5 μm
Fig. 2. Morphology of the steel after heat treating at 650 °C for 2 h. (a) Optical microstructure and (b) SEM image.
53
(011) α
(112) α
(002) α
(022) α
(111) γ
(002) γ
(022) γ
(113) γ
Intensity, A.U.
After heat treating
As-hot rolled
40 50 60 70 80 90 100 110
2θ, degree
Fig. 3. X-ray diffraction in as-hot rolled and heat treated steels.
that the dislocation density in martensite is very high. Hence, the
dislocation density decreases due to climb and annihilation of
dislocation and some fine ferrite grains form due to coalescence of
sub-grains and migration of sub-grain boundaries at a relatively
high heat treating temperature. In addition, the full-width at half-
maximum values of (200)α, (112)α and (220)α diffraction peaks of
as-hot rolled steel are
0.74,
0.74 and
0.98 degree, respec-
tively. But these values of above mentioned three diffraction peaks
of heat treated steel are
0.34,
0.39 and
0.51 degree, re-
spectively. These results imply that the dislocation density sig-
nificantly decreases after heat treating and the strong recovery has
occurred. On the other hand, the steel was isothermally treated at
650 °C for 2 h, where the Thermo-Calc calculation results show
that the two phases (ferrite and austenite) coexist. The austenite
reversion transformation occurred during heat treating and some
austenite phase can be remained at room temperature owing to
the partition of Mn, Ni and C between ferrite and austenite and
size effect. So the austenite diffraction peaks of (200)γ, (220)γ and
(113)γ can be observed in Fig. 3. Therefore, the ferrite/tempered
martensite–austenite microstructure was attained after heat
treating.
The TEM images of ferrite/tempered martensite–austenite mi-
crostructure are presented in Fig. 4. Note that both block-form and
lath-form retained austenite can be observed. Moreover, it is
worth noting that the block-form retained austenite is always
Austenite
1 μm
54 J. Chen et al. / Materials Science & Engineering A 648 (2015) 51–56

α
α 5 1/nm

α
γB
γB
α
200nm 200nm

αL
γ α
5 1/nm 111γ
αL γL 101α

αL γL 111γ
011α
γL
200nm 200nm Zone axis :[111]α//[011]γ

Fig. 4. TEM images of ferrite/tempered martensite–austenite microstructure. (a) and (c) Bright-field TEM images; (b) and (d) the corresponding dark-field TEM images;
(e) schematic illustration of selected area diffraction patterns: the solid and blank circles indicate the diffraction spots for austenite (γ) and ferrite (α), respectively; γB is the
block-form retained austenite, γL the lath-form retained austenite, α the ferrite and αL is the tempered martensite lath.

surrounded by fine ferrite grains, while the lath-form retained 16

Volume fraction of retained austenite, %

austenite is always displayed between tempered martensite laths. Measured

The fine ferrite grains are formed by climb and annihilation of Calculated
12
dislocations, coalescence of sub-grains and migration of sub-grain f γ = 15.6 exp(−15.65ε )
boundaries, resulting in a greatly decrease in yield strength of
ferrite matrix, which cannot sufficiently suppress the transfor-
8
mation from retained austenite to martensite, indicating that the
block-form retained austenite has relatively poor stability. How-
ever, the lath-form retained austenite possesses relatively better 4
stability [24]. In addition, the selected area electron diffraction
patterns indicate that both block-form and lath-form retained
austenite have an identical orientation relationship with α matrix, 0
0.00 0.04 0.08 0.12 0.16 0.20
namely the Kurdjumov–Sachs (K–S) orientation relationship of
True strain
(  11 1)γ//(0  11)α and [011]γ//[111] α[25–27].
Fig. 5. Volume fraction of retained austenite as a function of true strain during
tensile deformation at room temperature.
3.3. Stability of austenite phase

mechanical stability of retained austenite and the rapider de-

Fig. 5 presents the volume fraction of austenite phase in tensile
sample with different true strains, showing that the austenite
fraction decreases as the true strain increases and this relation can
be well fitted by the exponent decay equation reported in the
previous work [28–30].
fγ = f γ0 exp( − kε)
where, fγ0 is the initial volume fraction of austenite phase, fγ the
volume fraction of austenite phase at a true strain ε, and k is the
constant. Moreover, k is defined to be the stability coefficient of
retained austenite, and a higher k value indicates the lower
formation induced transformation from retained austenite to
martensite. The k value is about 15.65 in the present work, in-
dicating that the mechanical stability of retained austenite is poor
due to relatively lower carbon and manganese concentrations in
retained austenite. The vigorous TRIP phenomenon hence takes
(1) place at early stage during tensile deformation and the volume
fraction of austenite phase sharply decreases from
15.6% to
4%
by increasing the true strain from 0.0 to 0.08. The actual volume
fraction of about 12% in the sample has transformed into mar-
tensite, and this may be a reason why high instantaneous strain
hardening rate can be observed. However, beyond the true strain

Block
500
Equivalent diameter, nm
400
300
200
100
1 2 3 4
Fig. 6. Equivalent diameter and length width ratio of retained austenite.
of
0.08, the volume fraction of austenite phase gradually de-
creases to
0.8% with further increasing the true strain from

0.08 to
0.24. At this stage, the actual volume fraction of about
3% in the sample has transformed into martensite, indicating that
the stability of austenite phase is relatively better. The relatively
stable retained austenite has few contributions to strain hardening
rate, but it can further relax stress concentration and hence delay
the onset of necking, resulting in a higher total elongation.
3.4. Effects of austenite characteristics on deformation behavior
It is well known that the strain hardening behavior is related to
TRIP effect, while the TRIP effect depends on the stability of
retained austenite, which strongly depends on the chemical
composition, morphology and size of retained austenite
[10,13,21,31–33]. In order to reveal multi-peak strain hardening
behavior, about thirty TEM pictures obtained from different re-
gions of the thin foil were chosen to measure the equivalent
diameter and the length width ratio of retained austenite, and the
measured results are shown in Fig. 6. It can be seen that the length
width ratio is from
1.0 to
12.0 and the size is from
100 nm to

500 nm, indicating that the retained austenite shows multi-type
morphology and multi-scale size. For the same heat treating
temperature, the C, Mn and Ni concentrations in different retained
austenite grains can be thought to be similar, implying that the
chemical driving forces of the transformation for different retained
austenite grains are similar. Hence, the stability of retained aus-
tenite strongly depends on the morphology and size.
In the case of transformation from austenite to martensite by a
single variant mode, the increase in elastic strain energy can be
estimated using the follow equation [33,34].
UTS×TE, GPa%
Fig. 7. Volume fraction of retained austenite and UTS  TE in various steels reported in published papers [35-49].
J. Chen et al. / Materials Science & Engineering A 648 (2015) 51–56 55
Lath 1
ΔE v = {E1ε12(x/d)2 + (E2ε22 + E3ε32)(x/d)}
Coarse
2 (2)
where, E is the Young’s modulus, ε the elastic strain in each lattice
plane, x the thickness of thin martensite plate and d is the auste-
nite grain size. If each Young’s modulus and elastic strain are
substituted by 132.1 GPa (E1), 220.8 GPa (E2, E3), 0.139 (ε1), -0.07
(ε2) and -0.014 (ε3), the following equation can be obtained.
ΔE v = 1276.1(x/d)2 + 562.6(x/d) (3)
Therefore, the ΔEv decreases from
1838 MJ/m3 to 163 MJ/m3
Fine
as the retained austenite grain size increases from
100 nm to
5 6 7 8 9 10 11 12

500 nm, showing a greater transformation resistance needed to
Length width ratio
be overcome for fine retained austenite grains. On the other hand,
for a given retained austenite grain size, the lath-form retained
austenite commonly possesses relative high stability [24]. These
results indicate that the stability of retained austenite in the steel
greatly varies with the size and morphology. When the stress in-
creases to a critical stress, only some retained austenite with si-
milar stability can transform into martensite. With further in-
creasing stress to another critical stress, another retained austenite
with similar stability can also transform into martensite, and a
new round of TRIP can occur. This phenomenon was repeated
many times during tensile deformation, resulting in multi-peak
strain hardening behavior.
3.5. Toughening mechanisms
It has been recognized that only stable retained austenite can
greatly enhance toughness [15,16,22], but the more stable retained
austenite in general leads to poor strain hardening capacity, re-
sulting in higher yield ratio and lower uniform ductility. So in the
present work, the heat treating temperature was chosen at 650 °C
to produce relatively unstable retained austenite to obtain
expected strain hardening capacity and excellent ductility.
Meanwhile, other toughening mechanisms were used to obtain
excepted toughness. The probable reasons why the steel possesses
excellent low temperature toughness are as follows.
(1) The concentrations of interstitial carbon atom and sub-
stitutional Mn atom which deteriorate low temperature impact
toughness are relatively low.
(2) Figs. 5 and 6 show that a small volume fraction of fine
retained austenite possesses relatively high stability, which has
some contributions to enhance toughness.
(3) Fig. 2a shows that fine ferrite grains have formed during
heat treating due to relative high tempering temperature and
lower Mn concentration which lead to strong recovery. It is well
known that the fine grains can greatly arrest crack propagation.
50 Medium Manganese steels
Fe-0.05C-9Mn [35]
Fe-0.2~0.4C-5~7Mn [21]
40 Fe-0.26C-10.1Mn-6.3Al [36]
Fe-0.12~0.22C-5~10Mn-3Si-3Al [37]
Fe-0.2C-5Mn [38]
Fe-0.092C-4.6Mn [39]
30 Modified TRIP steels
Fe-0.4C-2.0Mn-1.8Si-0.5Cr [40]
Fe-0.21C-1.5Mn-1.5Si-0/0.04Nb [41]
20 Fe-0.17C-1.65Mn-0.38Si-1.11Al-0.08P [42]
Fe-0.1C-1.5Si-1.5Mn-0.5Cu [43]
Fe-0.4C-2.0Mn-1.7Si-0.4Cr [44]
Fe-0.18C-1.48Si-1.44Mn-0.15Al [45]
10 Fe-0.14C-1.9Si-1.7Mn [46]
This work Conventional high strength steels
Fe-C-Si-Mn-Ni-Cr-Mo bainite steels[47]
0 Cu-bearing HSLA steels[48]
0 10 20 30 40 50 60 C-Mn steels[49]
Volume fraction of retained austenite, %
56 J. Chen et al. / Materials Science & Engineering A 648 (2015) 51–56
This may be the main reason why the steel possesses outstanding
low temperature toughness.
3.6. Comparison with other steels
In addition, the products of UTS and TE vs. volume fractions of
retained austenite in some typical high strength steels reported in
published papers have been summarized, as shown in Fig. 7. The
combination of UTS and TE in our work is also plotted in Fig. 7.
Note that the combination of UTS and TE in our work is superior to
conventional high strength steels, most modified TRIP steels and
some 0.2–0.4% C  5 to 10% Mn steels. Besides its superior com-
bination of UTS and TE, the steel also possesses excellent low
temperature impact toughness.
4. Conclusions
Based on the mechanisms of plasticity and toughness, a low
carbon medium manganese steel plate was designed. The micro-
structure characteristics, deformation behavior and toughening
mechanisms of the steel after heat treating at 650 °C for 2 h were
investigated in details, and the following conclusions were drawn.
1. The steel treated by a simple tempering process possesses
outstanding combination of UTS and TE (
28.5 GPa%) and low
temperature impact toughness (
158 J at  80 °C).
2. The microstructure is composed of ferrite/tempered marten-
site and austenite. The fine ferrite grains have formed due to strong
recovery. The retained austenite shows multi-type morphology
(the length width ratio is from
1.0 to
12.0), multi-scale size
(the equivalent diameter is from
100 nm to
500 nm) and rela-
tively poor mechanical stability (the k value is about 15.65).
3. The excellent ductility is obtained due to the sufficient TRIP
effect, and the multi-peak strain hardening behavior is mainly
related to retained austenite characteristics. The outstanding low
temperature toughness is also achieved due to a small volume
fraction of fine stable retained austenite and the formation of fine
ferrite grains.
4. The combination of UTS and TE in our work is superior to
conventional high strength steels, most modified TRIP steels and
some 0.2–0.4% C  5 to 10% Mn steels.
Acknowledgments
This work is supported by Project funded by China Postdoctoral
Science Foundation (2014M560217, 2015T80260) and Funda-
mental Research Funds for Central Universities (N120807001).
References
[1] N. Nakada, K. Mizutani, T. Tsuchiyama, S. Takaki, Acta Mater. 65 (2014)
251–258.
[2] C. Wang, J. Shi, C.Y. Wang, W.J. Hui, M.Q. Wang, H. Dong, W.Q. Cao, ISIJ Int. 51
(2011) 651–656.
[3] B.C. De Cooman, P. Gibbs, S. Lee, D.K. Matlock, Metall. Mater. Trans. A 44 (2013)
2563–2572.
[4] S.J. Lee, S. Lee, B.C. De Cooman, Scr. Mater. 64 (2011) 649–652.
[5] P.J. Gibbs, B.C. De Cooman, D.W. Brown, B. Clausen, J.,G. Schroth, M.J. Merwin,
D.K. Matlok, Mater. Sci. Eng. A 609 (2014) 323–333.
[6] E.J. Seo, L. Cho, B.C. De Cooman, Metall. Mater. Trans. A 46 (2015) 27–31.
[7] R. Zhang, W.Q. Cao, Z.J. Peng, J. Shi, H. Dong, C.X. Huang, Mater. Sci. Eng. A 583
(2013) 84–88.
[8] D.W. Suh, J.H. Ryu, M.S. Joo, H.S. Yang, K. Lee, H.K.D.H. Bhadeshia, Metall.
Mater. Trans. A 44 (2013) 286–293.
[9] R.L. Miller, Metall. Trans. 3 (1972) 905–912.
[10] J.J. Wang, S. van Der Zwag, Metall. Mater. Trans. A 32 (2001) 1527–1539.
[11] T.F. Jing, F.C. Zhang, Mater. Lett. 31 (1997) 275–279.
[12] J. Han, Y.K. Lee, Acta Mater. 67 (2014) 354–361.
[13] J.H. Ryu, D.I. Kim, H.S. Kim, H.K.D.H. Bhadeshia, D.W. Suh, Scr. Mater. 63 (2010)
297–299.
[14] C. Zhao, C. Zhang, W.Q. Cao, Z.G. Yang, H. Dong, Y.Q. Weng, ISIJ Int. 54 (2014)
2875–2880.
[15] M. Lacroix, T. Pardoen, P.J. Jacques, Acta Mater. 56 (2008) 3900–3913.
[16] J. Chen, M.Y. Lv, S. Tang, Z.Y. Liu, G.D. Wang, Mater. Charact. 106 (2015)
108–111.
[17] M. Niikura, J.W. Morris Jr, Metall. Trans. A 11 (1980) 1531–1540.
[18] N.H. van Dijk, A.M. Butt, L. Zhao, J. Sietsma, S.E. Offerman, J.P. Wright, S. van
der Zwaag, Acta Mater. 53 (2005) 5439–5447.
[19] A.K. De, D.C. Murdock, M.C. Mataya, J.G. Speer, D.K. Matlock, Scr. Mater. 50
(2004) 1445–1449.
[20] J.B. Seol, J.E. Jung, Y.W. Jang, C.G. Park, Acta Mater. 61 (2013) 558–578.
[21] J. Shi, X.J. Sun, M.Q. Wang, W.J. Hui, H. Dong, W.Q. Cao, Scr. Mater. 63 (2010)
815–818.
[22] Q. Zhou, L.H. Qian, J. Tan, J.Y. Meng, F.C. Zhang, Mater. Sci. Eng. A 578 (2013)
370–376.
[23] J.B. Seol, D. Raabe, P.P. Choi, Y.R. Im, C.G. Park, Acta Mater. 60 (2012)
6183–6199.
[24] X.C. Xiong, B. Chen, M.X. Huang, J.F. Wang, L. Wang, Scr. Mater. 68 (2013)
321–324.
[25] G. Kurdjumov, G. Sachs, Phy. Z (in German) 64 (1930) 325–332.
[26] N. Nakada, T. Tsuchiyama, S. Takaki, S. Hashizume, ISIJ Int. 47 (2007)
1527–1532.
[27] X.D. Li, Z.D. Yin, Mater. Lett. 24 (1995) 239–242.
[28] K.I. Sugimoto, Mater. Sci. Technol. 25 (2009) 1108–1117.
[29] K.I. Sugimoto, M. Kobayashi, S.I. Hashimoto, Metall. Trans. A 23 (1992)
3085–3091.
[30] K.I. Sugimoto, N. Usui, M. Kobayashi, S.I. Hashimmoto, ISIJ Int. 32 (1992)
1311–1318.
[31] S. Lee, S.J. Lee, B.C. De Cooman, Scr. Mater. 65 (2011) 225–228.
[32] R. Ding, D. Tang, A.M. Zhao, Scr. Mater. 88 (2014) 21–24.
[33] S. Takaki, K. Fukunaga, J. Syarif, T. Tsuchiyama, Mater. Trans. 45 (2004)
2245–2251.
[34] S. Yamamoto, K. Asabe, M. Nishiguchi, Y. Maebara, Tetsu-to-Hagané 82 (1996)
771–776.
[35] J. Han., S.J. Lee, J.G. Jung, Y.K. Lee, Acta Mater. 78 (2014) 369–377.
[36] C.Y. Lee, J. Jeong, J. Han, S.J. Lee, S. Lee, Y.K. Lee, Acta Mater. 84 (2015) 1–8.
[37] H. Aydin, E. Essadiqi, I.H. Jung, S. Yue, Mater. Sci. Eng. A 564 (2013) 501–508.
[38] C. Wang, W.Q. Cao, J. Shi, C.X. Huang, H. Dong, Mater. Sci. Eng. A 562 (2013)
89–95.
[39] A. Arlazarov, M. Gouné, O. Bouaziz, A. Hazotte, G. Petitgand, P. Barges, Mater.
Sci. Eng. A 542 (2012) 31–39.
[40] G.H. Gao, H. Zhang, Z.L. Tan, W.B. Liu, B.Z. Bai, Mater. Sci. Eng. A 559 (2013)
165–169.
[41] I.B. Timokhina, P.D. Hodgson, E.V. Pereloma, Metall. Mater. Trans. A 35 (2004)
2331–2341.
[42] E. De Moor, S. Lacroix, A.J. Clarke, J. Penning, J.G. Speer, Metall. Mater. Trans. A
39 (2008) 2586–2595.
[43] C.G. Lee, S.J. Kim, T.H. Lee, S. Lee, Mater. Sci. Eng. A 371 (2004) 16–23.
[44] G.H. Gao, H. Zhang, X.L. Gui, P. Luo, Z.L. Tan, B.Z. Bai, Acta Mater. 76 (2014)
425–433.
[45] J. Zhang, H. Ding, R.D.K. Misra, C. Wang, Mater. Sci. Eng. A 611 (2014) 252–256.
[46] A. Itami, M. Takahashi, K. Ushioda, ISIJ Int. 35 (1995) 1121–1127.
[47] J. Chen, S. Tang, Z.Y. Liu, G.D. Wang, Mater. Sci. Eng. A 559 (2013) 241–249.
[48] S.K. Dhua, A. Ray, D.S. Sarma, Mater. Sci. Eng. A 318 (2001) 197–210.
[49] X.G. Zhou, Z.Y. Liu, S.Y. Song, D. Wu, G.D. Wang, J. Iron Steel Res. Int. 21 (2014)
86–90.