See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/309545582

Strengthening effect of nanoscale precipitation and transformation induced

plasticity in a hot rolled copper-containing ferrite-based lightweight steel

Article  in  Scripta Materialia · March 2017

DOI: 10.1016/j.scriptamat.2016.10.025

CITATIONS READS

12 41

6 authors, including:

Xiaodong Wang Xj Jin

Shanghai Jiao Tong University Shanghai Jiao Tong University
72 PUBLICATIONS   1,176 CITATIONS    167 PUBLICATIONS   2,414 CITATIONS   

SEE PROFILE SEE PROFILE

Li Wang
University of Westminster
41 PUBLICATIONS   842 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

High-entropy Alloys Strengthened by Coherent Precipitates View project

Toughening Mechanism and Microstructure Design of Ultrahigh Stength Press Hardening Steels for Automobiles View project

All content following this page was uploaded by Xiaodong Wang on 28 March 2021.

The user has requested enhancement of the downloaded file.

 Scripta Materialia 129 (2017) 25–29

Contents lists available at ScienceDirect

Scripta Materialia

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

Regular article

Strengthening effect of nanoscale precipitation and transformation

induced plasticity in a hot rolled copper-containing ferrite-based
lightweight steel
Junfeng Wang a, Zhaoguang Wang a, Xiaodong Wang a,⁎, Qi Yang b,c,⁎⁎, Xuejun Jin a, Li Wang b,c
a
State Key Laboratory of Metal Matrix Composites, School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
b
Baoshan Iron & Steel Co., Ltd., Shanghai 201900, China
c
State Key Lab of Development and Application Technology of Automotive Steels (Baosteel), Shanghai 201900, China

a r t i c l e i n f o a b s t r a c t

Article history: Effects of copper addition on microstructure and tensile property of a hot rolled ferrite-based lightweight steel
Received 29 July 2016 are investigated. Copper addition delays bainitic transformation, resulting in higher volume fraction of retained
Received in revised form 20 October 2016 austenite with lower stability in the steel. Addition of 1.0 wt% copper increases the yield strength and tensile
Accepted 20 October 2016
strength of the steel by 17% and 19%, respectively, and maintains high ductility. The yield strength enhancement
Available online xxxx
is predominantly attributed to nanoscale copper precipitation. The higher work hardening rate due to precipita-
Keywords:
tion hardening and the transformation-induced-plasticity mechanism leads to higher tensile strength (805 MPa)
Ferrite-based lightweight steel and remarkable elongation-to-failure (~30%) for the copper-containing steel.
Bainitic transformation © 2016 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
Copper precipitation strengthening
Mechanical property

Aluminum-enriched transformation-induced-plasticity (TRIP)
steels, also called δ-TRIP steels or ferrite-based lightweight steels,
could possess lower densities, excellent combinations of strength and
ductility, and appreciable producibility [1–10], and therefore have
great potential for automotive applications. As a characteristic constitu-
ent phase, δ-ferrite forms during solidification and persists during sub-
sequent thermo-mechanical processing and heat treatment due to the
addition of a large amount of aluminum [2,4,6]. Moreover, it is hardly
refined by the austenite-to-ferrite transformation and recrystallization.
Correspondingly, the δ-ferrite is oftentimes pretty coarse, and softer
than α-ferrite and fine austenite [11]. Strengths of ferrite-based light-
weight steels are therefore relatively low owing to the presence of a
large fraction of the δ-ferrite phase in the microstructure.
It is well known that precipitation strengthening is a useful ap-
proach to improving the strengths of steels. Microalloying with strong
carbide-forming elements such as Ti, Nb, Mo and V could remarkably in-
crease yield strengths and/or tensile strengths of ferritic steels [12–15]
and TRIP-assisted multiphase steels [16,17] through precipitation
strengthening as well as other microstructural modifications (such as
grain refinement, bainitic transformation, etc.), however the ductility
of the steels may be deteriorated. It has been another important strate-
gy to strengthen the above types of steels by utilizing copper (Cu)-rich
precipitates [18–21]. Because of limited solubility in Fe, Cu solutes could
⁎ Corresponding author.
⁎⁎ Correspondence to: Qi Yang, Baoshan Iron & Steel Co., Ltd., Shanghai 201900, China.
E-mail addresses: xdwang77@sjtu.edu.cn (X. Wang), yangqi@baosteel.com (Q. Yang).
undergo continuous nanoscale precipitation during a proper isothermal
hold or aging process, with the crystal structure of precipitates
transitioned from coherent body-centered cubic (BCC) through
twinned 9R to face-centered cubic (FCC) [22]. The effects of Cu addition
on microstructure as well as strength enhancement without noticeable
ductility deterioration in TRIP-assisted steels had been studied [23],
however the microstructural features including the Cu precipitation,
bainitic transformation, and retained austenite (RA) stability need to
be further elaborately clarified to interpret the correlation between mi-
crostructure and mechanical behavior. This is one of the objectives of
the present investigation which is to be performed on hot rolled
ferrite-based lightweight steels containing δ-ferrite. The other objective
is to confirm the feasibility of utilizing Cu precipitates to strengthen
ferrite-based lightweight steels and to attain an excellent combination
of strength and ductility in the meantime.
Two ferrite-based lightweight steels were fabricated for the investi-
gation. The chemical compositions of the steels are Fe-0.35C-1.1Mn-
4.1Al-0.4Si-(0\1.0)Cu (in wt%) with traces of nitrogen, sulfur and phos-
phor. The two steels were referred to as “Cu-free” and “1.0Cu”, respec-
tively hereafter for brevity. Steel ingots were homogenized at 1200 °C
for 2.5 h, and rough rolled from 180 mm to 40 mm, followed by air
cooling. The slabs were then reheated to 1200 °C for 60 min and hot-
rolled into plates of 3.6 mm with a finishing temperature of approxi-
mately 880 °C. The rolled plates were immediately water quenched to
450 °C, and then held at this temperature in a furnace for 60 min, and
finally furnace cooled to room temperature in order to simulate a coiling
procedure. After cooling, the hot-rolled plates were pickled in a dilute

http://dx.doi.org/10.1016/j.scriptamat.2016.10.025
1359-6462/© 2016 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
26 J. Wang et al. / Scripta Materialia 129 (2017) 25–29
HCl solution to remove the surface oxide film for microstructural and
mechanical analyses.
Tensile test samples were cut along the rolling direction (RD) with a
gauge length and width of 50 mm and 12.5 mm, respectively (according
to the ASTM E8-04 standard). Tests were performed on a computer-
controlled 8801 Instron machine with a constant crosshead speed of
3.0 mm/min (corresponding to an initial strain rate of 8.8 × 10−4/s) at
room temperature. At least three samples were tested to obtain average
tensile property for each steel.
Microstructures on the plane of the plate RD and normal direction
(ND) were observed using optical microscope (OM) and field-
emission scanning electron microscope (SEM; FEI Versa 3D). Metallo-
graphic specimens were mechanically polished and etched in a 4%
nital solution. Electron back-scattered diffraction (EBSD) analysis was
carried out on the RD-ND plane of the hot-rolled steel plates as well,
using Hitachi Su-70 Schottky field emission SEM at an acceleration volt-
age of 20 kV. EBSD data were post-processed by the HKL CHANNEL 5 fla-
menco software. Fine microstructures were further characterized by
transmission electron microscope (TEM; JEOL ARM-200F) equipped
with a Cs corrector, operating at an acceleration voltage of 200 kV.
TEM foils were prepared by twin-jet electropolishing punched disk
specimens (3 mm in diameter and 80 μm in thickness) in a 4 vol%
perchloric acid ethanol solution at −30 °C and with a voltage of about
50 V. Phase identification was made by an X-ray diffractometer (XRD,
Bruker-AXS D8 Discover) with Cu Kα radiation operated at 40 kV and
40 mA at room temperature. Volume fractions of retained austenite
(RA) were measured from the Rietveld refinement of XRD profiles to
take account of the texture effect. The carbon concentration of the RA
was determined using the following equation [24], assuming that the
partitioning effect of substitutional elements can be neglected,
aγ ¼ 3:578 þ 0:033X C þ 0:00095X Mn þ 0:0056X Al þ 0:00157X Si
þ 0:0015X Cu
where aγ is the austenite lattice parameter in Å, and XC, XMn, XAl, XSi and
XCu are the concentrations of C, Mn, Al, Si and Cu in steels, respectively,
in wt%. The austenite lattice parameter (aγ) was determined from the d-
spacing of (220)γ positions. To evaluate the effect of tensile strain on the
Fig. 1. OM and SEM micrographs and EBSD phase maps of the hot rolled lightweight steels: (a–c) Cu-free steel; (d–e) 1.0Cu steel. SEM micrographs and EBSD maps show transformation
products of prior austenite in darker bands in more detail. The step size for EBSD mapping is 70 nm. The area in red color depicts RA. (For interpretation of the references to color in this
figure legend, the reader is referred to the web version of this article.)
variation of the volume fraction of RA, XRD measurements were made
in the uniformly-deformed regions of fractured specimens and the
necked regions which are 3 mm away from the fracture surfaces. Tensile
true strains, ε, were calculated as follows:
 
w0 t 0
ε ¼ ln ð2Þ
wt
where w0 and t0 are the width and thickness of tensile specimens before
tensile test, respectively; w and t are the width and thickness of tensile
specimens in the uniformly-deformed regions or necked regions after
tensile test, respectively. It should be noted that the necked region is
in the three-dimensional stress state (rather than the uniaxial tensile
stress state as in the uniformly-deformed region), and the strain value
calculated by Eq. (2) roughly represents tensile limit strain.
Fig. 1 shows typical OM and SEM micrographs, and EBSD phase maps
of the Cu-free and 1.0Cu steels, respectively. The hot-rolled microstruc-
tures are basically banded structures in which lighter bands are aggre-
gates of recovered/recrystallized δ-ferrite grains, and darker bands are
transformation products of prior austenite during the isothermal hold
and furnace cooling [25]. Darker bands show different variants and mor-
phologies, depending on whether Cu is added. In the Cu-free steel, the
darker bands mainly include carbide-free bainite which is composed of
lath-shaped bainitic ferrite and austenite, in addition to polygonal α-
ferrite grains and some blocky austenite particles (Fig. 1(a, b)). Martensite
is hardly observed in the microstructure, as indicated by the EBSD phase
map (Fig. 1(c)). It is thus implied that the RA after bainitic transformation
reaction is quite thermally stable. In the 1.0Cu steel, the darker bands are
mainly composed of blocky austenite/martensite packets (Fig. 1(d, e)).
Some of prior austenite transform into feathery bainitic structure during
the simulated coiling process, which is illustrated in black regions in
Fig. 1(d) and by the inset magnified view in Fig. 1(e), and will be further
ð1Þ confirmed by TEM analysis. The formation of martensite (M/A islands) as
verified by the EBSD phase map (Fig. 1(f)) is ascribed to the transforma-
tion of large and relatively thermally-unstable austenite when the tem-
perature drops below Ms. (martensite start temperature) during furnace
cooling. According to EBSD statistical measurements from at least 100
grains for each phase, the grain sizes for δ-ferrite, α-ferrite and RA are
 J. Wang et al. / Scripta Materialia 129 (2017) 25–29
11.5 ± 5.8 μm, 2.8 ± 0.8 μm and 2.2 ± 1.1 μm, respectively in the Cu-free
steel, and 11.1 ± 5.3 μm, 2.4 ± 0.9 μm and 2.5 ± 1.0 μm, respectively in
the 1.0Cu steel. The presence of RA is attributed to ‘incomplete reaction
phenomenon’ of bainitic transformation when the austenite carbon con-
tent approaches the T0 line [16,26]. XRD measurements show that the vol-
ume fraction of RA is 17.1% in the 1.0Cu steel, and 9.6% in the Cu-free steel.
From Eq. (1), the average carbon concentration of RA is 2.08 wt% in the
Cu-free steel, and 1.85 wt% in the 1.0Cu steel.
Cu is an austenite stabilizer. A simple thermodynamic calculation in-
dicates that the addition of 1.0 wt% Cu does not cause an appreciable
change in the volume fraction of austenite at high temperatures. Thus,
Cu addition increases thermal stability of high-temperature austenite.
The prior austenite with increased thermal stability could be more read-
ily retained in a blocky shape by delaying bainitic reaction or without
experiencing substantial bainitic reaction during an isothermal hold
(subsequent to finish rolling and water quenching), although some of
the austenite with larger grain sizes would eventually transform into
martensite as the temperature falls below Ms. (Fig. 1(f)). This could be
the reason why a larger volume fraction of blocky RA is present in the
1.0Cu steel. In comparison, the Cu-free steel could undergo adequate
bainitic transformation reaction during an isothermal hold because of
the lower thermal stability of high-temperature austenite. Thus, the
RA is more carbon enriched due to the carbon partitioning from the
neighboring bainitic ferrite, and its volume fraction is lower in the Cu-
free steel. Compared to the Cu-free steel, the 1.0Cu steel possesses
higher volume fraction of RA with lower carbon content after the simu-
lated coiling process.
Fine microstructure formed during the bainitic hold in the 1.0Cu
steel is further characterized. Fig. 2(a) illustrates a TEM bright-field
image of a bainitic sheaf. As shown in the corresponding center dark-
field images (Fig. 2(b, c)) and by the selected area electron (SAED) pat-
tern (Fig. 2(d)), RA films exist in between bainitic ferrite plates, and
there are no carbides precipitating within or between bainitic ferrite
plates. This implies that the supersaturated carbon in the bainitic ferrite
diffuses into the surrounding austenite during the transformation reac-
tion and then stabilizes the remaining austenite. The crystallographic
orientation relationship between bainitic ferrite and RA film follows
the N-W relationship: ð111Þγ ==ð110ÞB F and [011]γ//[001]BF. Bainitic fer-
rite plates are less than 200 nm in thickness, and RA films are about a
few tens of nanometer in thickness. In morphology, this bainitic struc-
ture is basically carbide-free bainite, except that laths of bainitic ferrite
Fig. 2. TEM micrographs of the 1.0Cu steel: (a) bright-field image of bainite; (b) g200Bf dark-field image of bainitic ferrite; (c) g200γ dark-field image of film-like retained austenite; (d) SAED
pattern of bainite.
27
and RA in the 1.0Cu steel are finer than those in the Cu-free steel.
From TEM analysis, the above fine structure is unlikely to be originated
from martensitic transformation due to a thermal effect, considering
that no twinned substructure is observed within prior austenite having
such high carbon content. It is unclear that the bainitic structure in the
1.0Cu steel is finer than that in the Cu-free steel. Imaginably, the coordi-
nation of bainitic transformation and interphase Cu precipitation might
slow down the growth of bainitic ferrite.
Cu clustering/precipitation inside ferrite grains in the 1.0Cu steel is
investigated.
Fig. 3(a) shows the high-angle annular dark-field scanning transmis-
sion electron microscopy (HAADF STEM) image of δ-ferrite grains. High-
density ultra-fine particles and sparse dislocation lines are observed in-
side δ-ferrite. The sparse distribution of dislocations inside the δ-ferrite
grain interior indicates that δ-ferrite undergoes quite adequate recov-
ery/recrystallization at high rolling temperature. As seen in Fig.
3(b) which depicts a magnified view of the boxed region in Fig. 3(a), a
large number of precipitating particles (shown by white dots) at a nano-
meter scale gather along the dislocation line. Cu precipitates have a
closely-packed face-centered cubic (FCC) crystal structure, as demon-
strated by the atomic-resolution HAADF STEM image and the inset cor-
responding fast Fourier transform (FFT) pattern in Fig. 3(c). There exists
a N-W orientation relationship between the precipitates and matrix:
ð111ÞFCC−Cu ==ð110Þα and [011]FCC − Cu//[001]α. The inverse FFT (IFFT)
image by masking {111} FCC reflection is shown in Fig. 3(d), and statis-
tical measurements indicate that the sizes of Cu precipitates are mainly
smaller than 3–5 nm. The densely distributed Cu precipitates inside the
ferritic matrix would undoubtedly affect mechanical property of the
steel.
Room temperature tensile stress-strain curves and strain hardening
rate curves (as an inset) of the two steels are shown in Fig. 4(a). The
1.0Cu steel possesses the yield strength of 563 MPa and tensile strength
of 805 MPa, higher than those of the Cu-free specimen by 83 MPa
(i.e., 17% increment) and 130 MPa (i.e., 19% increment), respectively,
while the elongation-to-failure of these two steels are comparable,
about 30%. In addition, the 1.0Cu steel maintains higher strain hardening
rate than the Cu-free steel at true strains ranging from 0.03 to 0.24. Me-
chanical properties of the two steels are listed in Table 1 in detail.
The strain hardening rate is closely related to the transformation be-
havior of RA during tensile deformation in TRIP steels. Fig. 4(b) shows
the variation of the volume fraction of RA as a function of true strain
28 J. Wang et al. / Scripta Materialia 129 (2017) 25–29

Fig. 3. (a) STEM image of δ-ferrite grain; (b) magnified view of the boxed region in (a); (c) atomic resolution HAADF STEM image (FFT pattern inset); (d) IFFT image acquired by masking
{111} FCC reflections.

for the present two steels. As deformation proceeds, in particular
throughout uniform strain, more RA particles are transformed at a faster
rate in the 1.0Cu steel. This means that RA is mechanically less stable in
the 1.0Cu steel, although the transformed volume fraction of RA is
higher than that in the Cu-free steel. The mechanical stability of RA in
the 1.0Cu steel lower than that of RA in the Cu-free steel could be as-
cribed to the joint effect of primarily blocky shape, slightly larger grain
size and lower carbon concentration of RA in the 1.0Cu steel. However,
the higher transformed volume fraction together with lower mechani-
cal stability of RA, probably assisted by Cu precipitation, causes the
1.0Cu steel to maintain a higher strain hardening rate up to high strains
(Fig. 4(a)) and thus achieve a remarkably high elongation-to-failure.
Furthermore, as generally accepted, tailored mechanical stability of RA
ensures continuous martensitic transformation at varied stress levels,
thus providing continuous strain hardening of TRIP steels [27,28]. That
is to say, transformation of RA with larger grain sizes and/or a blocky
shape (i.e., with lower mechanical stability) primarily provides strain
hardening at a lower stress level, while transformation of RA with
finer grain size and/or a lath shape (i.e., with higher mechanical stabili-
ty) provides strain hardening at a higher stress level. This is believed to
be true for the present two steels.
As a TRIP steel is loaded, yielding starts from the softer phase, ferrite
when RA has fine grain size and good mechanical stability [3,11,29]. Fur-
thermore, δ-ferrite is believed to be the softer ferrite since it has a larger
grain size (compared to α-ferrite) and undergoes quite adequate recov-
ery/recrystallization (Fig. 1 and Fig. 3). The contributions to the

Fig. 4. (a) Room-temperature engineering stress-strain relationship of the hot-rolled steels (strain hardening rate curves inset); (b) variation of RA volume fraction as a function of true
strain.
 J. Wang et al. / Scripta Materialia 129 (2017) 25–29
Table 1
Mechanical properties of the both steels.
Yield Tensile Uniform
strength/MPa strength/MPa elongation/%
Cu-free 480 675 24
1.0Cu 563 805 24.6
difference in the yield strengths of the present two steels would only
arise from the Cu solid-solution strengthening and Cu precipitation
strengthening of δ-ferrite, since the two steels undergoing adequate re-
covery/recrystallization have comparable δ-ferrite grain sizes (the ef-
fects of grain boundary strengthening and dislocated substructure
strengthening can thus be ignored). The Cu solubility limit in iron in
the range of 500–700 °C is given in the following equation [30]:
 
log10 ½Cuwt% ¼ 5771323 T2
−15763:84 T
þ 9:944961
where T is the absolute temperature. By extrapolation, the Cu solubility
at 450 °C is about 0.15 wt%, and becomes even less during furnace
cooling. Therefore, an increase in the yield strength due to the Cu
solid-solution strengthening (equal to 38 MPa/wt% × Cu solubility in
wt% [31]) is less than 5.7 MPa. From the above analysis, it is clearly dem-
onstrated that the higher yield strength of the 1.0Cu steel predominant-
ly arises from the Cu precipitation. It should be pointed out that the yield
strength enhancement due to precipitation hardening is theoretically
calculated to be more than 138 MPa according to the dislocation-
precipitate cutting model (shown by eq. (3) in reference [20]) and to
be more than 317 MPa according to the dislocation-precipitate bowing
model (shown by Eq. (4) in reference [20]), both much higher than the
measured value of about 80 MPa. The fact that the strengthening of Cu
precipitates can not be predicted by the dislocation-precipitate interac-
tion models might be attributed to the modulus difference between Cu
precipitate and ferrite or an interaction between screw dislocation cores
and precipitates with the sizes of a few nanometers [20,32]. The higher
tensile strength of the 1.0Cu steel is due to the composite effect of pre-
cipitation hardening and higher strain hardening rate (Fig. 4).
In summary, the effects of Cu addition on microstructure evolution
and mechanical behavior in a hot rolled ferrite-based lightweight steel
are investigated. Cu addition improves thermal stability of high-
temperature austenite and subsequently delays bainitic transformation
at 450 °C. As a result, a higher volume fraction of RA primarily in a
blocky shape but with relatively lower thermal/mechanical stability re-
mains in the ferrite matrix of the Cu-containing steel during furnace
cooling to room temperature. Addition of 1.0 wt% Cu increases the
yield strength and tensile strength of the steel by 83 MPa (i.e., 17% incre-
ment) and 130 MPa (i.e., 19% increment), respectively, but maintains
high ductility. The increment of the yield strength is predominantly at-
tributed to nanoscale Cu precipitation. The cooperation of precipitation
hardening and the TRIP mechanism results in the higher strain harden-
ing rate, and therefore higher tensile strength (805 MPa) and remark-
able elongation (~30%) for the 1.0Cu steel.
View publication stats
29
Acknowledgment
Total The authors are grateful for financial support from the National Nat-
elongation/% ural Science Foundation of China (grant no. 51271112) and Research
30.5 Fund from Baoshan Iron & Steel Co., Ltd.
29.8
Appendix A. Supplementary data
Supplementary data to this article can be found online at doi:10.
1016/j.scriptamat.2016.10.025.
References
[1] H.L. Yi, K.Y. Lee, H.K.D.H. Bhadeshia, Phys. Eng. Sci. 467 (2010) 234–243.
[2] S. Chatterjee, M. Murugananth, H.K.D.H. Bhadeshia, Mater. Sci. Technol. 23 (2007)
819–827.
ð3Þ [3] H.L. Yi, K.Y. Lee, H.K.D.H. Bhadeshia, Mater. Sci. Eng. A 528 (2011) 5900–5903.
[4] H.L. Yi, K.Y. Lee, H.K.D.H. Bhadeshia, Mater. Sci. Technol. 27 (2011) 525–529.
[5] H.L. Yi, J.H. Ryu, H.K.D.H. Bhadeshia, H.W. Yen, J.R. Yang, Scr. Mater. 65 (2011)
604–607.
[6] Y.J. Choi, D.W. Suh, H.K.D.H. Bhadeshia, Phys. Eng. Sci. 468 (2012) 2904–2914.
[7] H.L. Yi, P. Chen, Z.Y. Hou, N. Hong, H.L. Cai, Y.B. Xu, D. Wu, G.D. Wang, Scr. Mater. 68
(2013) 370–374.
[8] H.L. Yi, P. Chen, H.K.D.H. Bhadeshia, Metall. Mater. Trans. A 45 (2014) 3512–3518.
[9] S.S. Sohn, B.J. Lee, S. Lee, N.J. Kim, J.H. Kwak, Acta Mater. 61 (2013) 5050–5066.
[10] S.J. Park, B. Hwang, K.H. Lee, T.H. Lee, D.W. Suh, H.N. Han, Scr. Mater. 68 (2013)
365–369.
[11] K. Lee, S.-J. Park, Y.S. Choi, S.-J. Kim, T.-H. Lee, K.H. Oh, H.N. Han, Scr. Mater. 69
(2013) 618–621.
[12] U. Brüx, G. Frommer, Steel Res. Int. 73 (2002) 534–548.
[13] A. Zargaran, H.S. Kim, J.H. Kwak, N.J. Kim, Scr. Mater. 89 (2014) 37–40.
[14] S.G. Hong, H.J. Jun, K.B. Kang, C.G. Park, Scr. Mater. 48 (2003) 1201–1206.
[15] H.-W. Yen, C.-Y. Huang, J.-R. Yang, Scr. Mater. 61 (2009) 616–619.
[16] X.D. Wang, B.X. Huang, L. Wang, Y.H. Rong, Metall. Mater. Trans. A 39 (2008) 1–7.
[17] G.K. Tirumalasetty, M.A. van Huis, C.M. Fang, Q. Xu, F.D. Tichelaar, D.N. Hanlon, J.
Sietsma, H.W. Zandbergen, Acta Mater. 59 (2011) 7406–7415.
[18] B. Mishra, K. Kumbhar, K.S. Kumar, K.S. Prasad, M. Srinivas, Mater. Sci. Eng. A 651
(2016) 177–183.
[19] J. Takahashi, K. Kawakami, Y. Kobayashi, Mater. Sci. Eng. A 535 (2012) 144–152.
[20] Y.R. Wen, Y.P. Li, A. Hirata, Y. Zhang, T. Fujita, T. Furuhara, C.T. Liu, A. Chiba, M.W.
Chen, Acta Mater. 61 (2013) 7726–7740.
[21] M. Kapoor, D. Isheim, G. Ghosh, S. Vaynman, M.E. Fine, Y.-W. Chung, Acta Mater. 73
(2014) 56–74.
[22] Y.R. Wen, A. Hirata, Z.W. Zhang, T. Fujita, C.T. Liu, J.H. Jiang, M.W. Chen, Acta Mater.
61 (2013) 2133–2147.
[23] S.-J. Kim, C.G. Lee, T.-H. Lee, C.-S. Oh, ISIJ Int. 42 (2002) 1452–1456.
[24] D. Dyson, B. Holmes, J. Iron Steel Inst. 208 (1970) 469–474.
[25] J. Wang, Q. Yang, X. Wang, L. Wang, 2016. Metall. Mater. Trans. A. , http://dx.doi.org/
10.1007/s11661-016-3752-8.
[26] H.K.D.H. Bhadeshia, J.W. Christian, Metall. Trans. A 21 (1990) 767–797.
[27] Z.H. Cai, H. Ding, X. Xue, J. Jiang, Q.B. Xin, R.D.K. Misra, Scr. Mater. 68 (2013)
865–868.
[28] J.H. Ryu, J.I. Kim, H.S. Kim, C.-S. Oh, H.K.D.H. Bhadeshia, D.-W. Suh, Scr. Mater. 68
(2013) 933–936.
[29] J.H. Ryu, D.-I. Kim, H.S. Kim, H.K.D.H. Bhadeshia, D.-W. Suh, Scr. Mater. 63 (2010)
297–299.
[30] M. Perez, F. Perrard, V. Massardier, X. Kleber, A. Deschamps, H. de Monestrol, P.
Pareige, G. Covarel, Philos. Mag. 85 (2005) 2197–2210.
[31] Q.L. Yong, Secondary Phases in Steels, Metallurgy Industry Press, Beijing, 2006.
[32] M.E. Fine, D. Isheim, Scr. Mater. 53 (2005) 115–118.