J. Iron Steel Res. Int.

https://doi.org/10.1007/s42243-024-01226-1 (0123456789().,-volV)(0123456789().
,- volV)

ORIGINAL PAPER

Effect of initial microstructure on microstructure evolution

and mechanical properties of 0.12C martensitic steels
during quenching and tempering
Ya-ru Wang1,2,3 • Zi-yong Hou3 • He Yang3 • Jun Zhao3 • Zhi-yuan Chang1,4 • Fan-mao Meng5 •

Ling Zhang3 • Gui-lin Wu6 • Xiao-xu Huang3

Received: 9 November 2023 / Revised: 9 January 2024 / Accepted: 26 January 2024

 China Iron and Steel Research Institute Group Co., Ltd. 2024

Abstract
The microstructure evolution and mechanical properties of a Fe–0.12C–0.2Si–1.6Mn–0.3Cr–0.0025B (wt.%) steel with
different initial microstructures, i.e., hot rolled (HR) and cold rolled–annealed (CRA), were studied through optical
microscopy, scanning electron microscopy, electron channeling contrast imaging, microhardness and room temperature
uniaxial tensile tests. After water quenching from 930 C to room temperature, a fully martensitic microstructure was
obtained in both as-quenched HR and CRA specimens, which shows a microhardness of 480 ± 5 HV, and no significant
difference in microstructure and microhardness was observed. Tensile test results show that the product of tensile
strength and total elongation (UTS 9 TE) of the as-quenched HR specimen, i.e., 24.1 GPa%, is higher than that of the as-
quenched CRA specimen, i.e., 18.9 GPa%. While, after being tempered at 300 C, the martensitic microstructures and
mechanical properties of the two as-quenched specimens change significantly due to the synergy role of the matrix phase
softening and the precipitation strengthening. Concerning the maximum UTS 9 TE, it is 18.9 GPa% obtained in the as-
quenched CRA one, while that is 24.4 GPa% obtained in the HR specimen after tempered at 300 C for 5 min.

Keywords Initial microstructure  Lath martensite  Quenching  Tempering  Mechanical property

& Zi-yong Hou

[email protected] 1 Introduction
& Zhi-yuan Chang
[email protected] The worldwide movement toward carbon neutrality
1
encourages the automobile industry to develop lightweight
State Key Laboratory of Vanadium and Titanium Resource
and ultra-/high-strength materials [1–4]. Steel is one of the
Comprehensive Utilization, Panzhihua 617000, Sichuan,
China most vital components for automobiles due to its good
2 combination of affordability, exceptional machinability,
State Key Laboratory of Advanced Special Steel, Shanghai
Key Laboratory of Advanced Ferrometallurgy, School of and recyclability. Thus, the application of ultra-/high-
Materials Science and Engineering, Shanghai University, strength steels in car bodies becomes a potential solution
Shanghai 200444, China for lightweight and crashworthiness [4–8].
3
International Joint Laboratory for Light Alloys (Ministry of It is well known that two types of forming methods, i.e.,
Education), College of Materials Science and Engineering, cold stamping and hot stamping, have been widely applied
Chongqing University, Chongqing 400044, China
for car body production. For the cold stamping process,
4
Institute of Steel Technology for Automobile and Household even though it has been widely employed for forming high-
Appliance, Pangang Group Co., Ltd.,
strength steel sheets, there are several drawbacks, such as
Panzhihua 617000, Sichuan, China
5
large springback, high load, low formability, short service
ATB Motor (Shanghai) Co., Ltd., Keen International Center,
life of tools, and severe wear [9]. To overcome these
Shanghai 200041, China
6
drawbacks, the hot stamping process, consisting of heating
Beijing Advanced Innovation Center for Materials Genome
steel blank to a temperature above the finish temperature of
Engineering, University of Science and Technology Beijing,
Beijing 100083, China austenization, holding it for 3–10 min, and then

123
 Y.R. Wang et al.

transferring the steel sheet from the furnace to the hot Table 1 Chemical composition of studied steel (wt.%)
stamping mould for simultaneous forming and quenching
C Si Mn Cr B Fe
[10], has been developed. For example, the tensile strength
of 22MnB5 steel, after subjected to hot-rolling, cold-roll- 0.12 0.2 1.6 0.3 0.0025 Balance
ing, and annealing, showing a ferritic–pearlitic
microstructure, is approximately 600 MPa, while the ten-
sile strength can reach 1600 MPa after hot stamping [11]. martensitic dual-phase microstructures with different sizes,
The tensile strength together with the absorbed energy of morphology, and distribution of martensite by altering the
the hot stamping component was significantly enhanced. initial microstructures using heat treatment and thermo-
The hot stamping process was first applied to the impact- mechanical processing routes. Yan et al. [27] found that
resistant side members of the Swedish SAAB 9000 model initial microstructure has an impact on both microstructural
in the 1980s [12]. However, till now, the produced com- evolution and mechanical properties. Detailed information
ponents by the hot stamping process take only 15%–38% concerning the effect of initial microstructure on the as-
of the total amount of components in body-in-white [13] quenched and tempered microstructures and mechanical
due to the strength–ductility trade-off in steels. properties of low-carbon lath martensitic steels still needs
It is a historical issue to achieve a good combination of to be included.
ultra-high strength and toughness in steel. Lath martensitic Therefore, the purpose of present study is to investigate
steel containing less than 0.35 wt.% carbon, in general, was the effect of initial microstructure on the microstructure
widely used. In the past decades, considerable progress has evolution and the mechanical properties of low-carbon
been made in developing ultra-high strength steels in such steel. The as-quenched and tempered microstructure and
a low-carbon martensite [7, 14–18]. A significant body of mechanical properties of 0.12C (wt.%) steel with different
research on low-alloy high-strength steels has demon- initial microstructures were studied by using a combination
strated that the final microstructure of steel plays a crucial of optical microscopy, scanning electron microscopy
role in determining its mechanical behavior. Meanwhile, (SEM), electron channeling contrast imaging (ECCI),
the strength and toughness of steels can be tuned by adding microhardness and room temperature uniaxial tensile tests.
trace elements, such as Nb, V, Ti, Mo, and W, to form The role of initial microstructure on the microstructure
dispersion of nano precipitates and refine prior austenite evolution and mechanical properties of lath martensitic
grains by pinning effect [15, 16, 19–22]. By adding Nb steel was also further discussed.
element, a 2000 MPa grade ultra-high strength steel
with the product of tensile strength and total elongation
(UTS 9 TE) * 17.7 GPa% has been developed by Liang 2 Materials and methods
et al. [15]. Furthermore, the optimal processing parameters
are also equally important. An Fe–Mn–Si–Nb alloyed steel The chemical compositions of the studied steel are listed in
with a UTS 9 TE, i.e., * 31.4 GPa%, has been produced Table 1. The alloy was made by vacuum induction melting
through a novel quenching–partitioning–tempering process and cast into an ingot. The forged ingot was cut into a
[17]. 100 mm 9 100 mm 9 L mm slab (L is the length of the
The microstructure of martensitic steel, such as mor- ingot). Then, the slab was homogenized at 1250 C for
phology, size, and type, is related to the processing 2.5 h and hot-rolled to 3.5 mm in thickness (hereafter
parameters and chemical composition, which determine the referred to as HR). The hot-rolled plate was then pickled
mechanical properties. The role of the initial microstruc- and cold-rolled to 1.5 mm with a thickness reduction of
ture of low-carbon lath martensitic steel on the subsequent 57.1%, after which it was annealed for 5 min at 750 C
as-quenched and tempered microstructure and mechanical (hereafter referred to as CRA).
properties has yet to receive attention. Luo et al. [23] The equilibrium phase fractions of the studied steel were
suggested that the substructure of martensite, including calculated using the Thermo-Calc software with the
packet, block, and lath, was refined when the size of prior TCFE12 database (see in Fig. 1) for a temperature range
austenite grains was decreased. Zhang et al. [24] found that from 0 to 1400 C, which was relevant for the solid phase
adding Nb into 42CrNiMoV bolt steel can significantly observation. The start temperature of austenization (Ac1)
refine the size of prior austenite grain and the corre- and finish temperature of austenization (Ac3) are 684 and
sponding as-quenched martensitic microstructure. Järvinen 823 C, respectively, as seen in Fig. 1a. From the detailed
et al. [25] suggested that the austenitization parameters and magnification of Fig. 1a, austenite can completely trans-
the initial microstructure of 22MnB5 steel influence the form into ferrite as the temperature drops below 684 C.
grain size and morphology of austenite and its mechanical When the temperature continues to decrease, various types
properties. Mirzadeh et al. [26] produced ferritic–

123
Effect of initial microstructure on microstructure evolution and mechanical properties…

Fig. 1 Equilibrium phase diagram (a) and detailed magnification in a (b) of studied steel

Fig. 2 Schematic of heat treatments used for each condition

of precipitates gradually form. When the temperature Fig. 3 Schematic illustration of tensile test specimens and metallo-
decreases to 447, 401 and 334 C, the M2B, M7C3, and graphic specimens. ND—Normal direction
M23C6 carbides start to form, respectively. At 300 C, the
predominant precipitates are M7C3 (volume fraction of
1.06%), M23C6 (volume fraction of 0.48%), and M2B performed using SHIMADZU AG–X Plus (10 kN) with a
(volume fraction of 0.04%). specific crosshead speed of 0.5 mm/min, and three speci-
Based on the calculated phase diagram, specimens with mens were used for each condition.
dimensions of 40 mm 9 25 mm (rolling direction The metallographic and microhardness test specimens
(RD) 9 transverse direction (TD)) were first cut from the were ground by SiC paper, mechanically polished with
HR and CRA plates by wire electrical discharge machining 0.5 lm diamond spray, and then etched in 4 vol.% Nital.
(Wire-EDM). The schematic of heat treatments is shown in The average grain size of ferrite and pearlite is calculated
Fig. 2. The specimens were austenitized at 930 C for using an intercept method, and the nano measurer software
5 min, followed by water quenching (hereafter referred to is used. For each condition, * 500 grains were taken into
as HR-Q and CRA-Q, respectively). Then, the as-quenched account for an accurate statistical effect. The microstruc-
specimens were tempered at 300 C for 5, 10, 30, 120, and ture observations were carried out by MODEL DM2000X
600 min, respectively, which were hereafter referred to as optical microscope and JEOL JSM-7800F field emission
HR/CRA-T 5, HR/CRA-T 10, HR/CRA-T 30, HR/CRA-T SEM, operating at 15 kV. The microhardness measure-
120, and HR/CRA-T 600, respectively, for easy discussion. ments were conducted under a constant loading condition
All the heat treatments were conducted using a GSL- using a load of 1.961 N for a dwell time of 15 s. An
1400X tube furnace. average value of 10 points for each specimen was taken,
The dog-bone tensile specimens were cut along with RD and the distance between neighboring indents was set
by Wire-EDM, as seen in Fig. 3. The tensile tests were longer than 150 lm to avoid interference.

123
 Y.R. Wang et al.

3 Results 3.2 Microstructure evolution during quenching

and tempering
3.1 Thermodynamic calculations
The initial microstructure is composed of polycrystal fer-
Within the temperature range of 0–800 C, the chemical rite and pearlite in both HR and CRA specimens, as seen in
composition in the matrix BCC, FCC, M2B, M7C3, and Figs. 5a, b and 6a, b. The different features in
M23C6 is shown in Fig. 4a, b, c, d, respectively, indicating microstructure between the HR and CRA specimens are the
that the chemical composition in the matrix and precipi- size, morphology, and distribution of ferrite and pearlite. It
tates changes to some extent with temperature. Notably, is clear that, in the HR specimen, the elongated pearlite and
with the temperature decreasing, the Mn content of BCC ferrite, along the hot rolling direction, are observed. A
and three types of precipitates, i.e., M2B, M7C3, and M23C6 typical full recrystallization microstructure, instead of a
increases first and then decreases, while the Mn content of deformed microstructure, of the CRA specimen is
FCC as a whole is on a downward trend. The content of Cr observed. The cold rolling and annealing are the main
increases slowly in M2B and increases significantly in reason for such microstructure in the CRA specimen. In
M23C6. However, the change of Cr content in BCC and addition, the average grain size of ferrite in the CRA
FCC is minor due to the low adding amount in total. specimen, i.e., 7.5 ± 2.1 lm, is much smaller than that of
Overall, the calculated results shown in Figs. 1 and 4 can the HR specimen, i.e., 10.5 ± 4.2 lm.
guide heat treatments and alloy design and optimization, to Figures 5c, d and 6c, d depict the as-quenched
minimize the addition of alloying elements and reach the microstructure of HR and CRA specimens. Both specimens
expected phase and microstructure at the end. show a full lath martensitic microstructure. While, the fine
and complicated martensitic microstructure, including prior
austenite grain (PAG), packet, block, and lath, is somehow

Fig. 4 Variation of chemical composition in M2B (a), M7C3 (b), M23C6 carbides (c) and BCC and FCC with temperature (d)

123
Effect of initial microstructure on microstructure evolution and mechanical properties…

448 and 288 C, respectively. Thus, during quenching to

room temperature, carbon could diffuse to a distance from
a few nanometers to tens of nanometers, depending on the
quenching rate above Ms [30]. This phenomenon is called
auto-tempering. For the present steel, the Ms and Mf are
423 and 314 C, respectively, and the auto-tempering will
also occur. A similar phenomenon has been reported by
Morsdorf et al. [31] and Cai et al. [32] in other types of
steels.
The microstructures of the tempered CRA and HR
specimens are shown in Figs. 7 and 8, respectively. No
significant change occurs after tempering for 5 min in both
steels. When the tempering time is prolonged to 10 min,
the precipitates can be found inside the thin laths, other
Fig. 5 OM microstructures of specimens. a CRA; b HR; c CRA-Q;
d HR-Q than in the coarse laths only. After tempering for 120 and
600 min, the volume fraction and the size of the precipi-
tates are increased. Meanwhile, the coarsening of thin lath
also occurs due to the migration and elimination of low-
angle lath boundaries, as seen in Figs. 7 and 8. Further-
more, with increasing tempering time [33] or temperature
[34], the sizes of inter-lath carbides increase noticeably due

Fig. 6 SEM microstructures of specimens. a CRA; b HR; c, e CRA-

Q; d, f HR-Q

clearly presented as shown in Fig. 6c, d, e.g., packet con-

sists of several well-developed parallel blocks [28], even
though the quantification for each type of substructure is a
challenge so far. Compared with the as-quenched
microstructure, the width of the lath in the HR specimen is
slightly coarser than that in the CRA one. This could be
because the growth of martensite is hindered in the CRA
specimen by relatively dense boundaries, which are
inherited from the initial condition. Interestingly, some fine
carbides are distributed inside the coarse laths, while car-
bides are rarely observed inside the thin laths, as seen in
Fig. 6e, f. Hutchinson et al. [29] pointed out that the
martensite starting temperature (Ms) and the martensite
Fig. 7 SEM microstructure of CRA specimen tempered at 300 C. a,
finishing temperature (Mf) of 0.12C (wt.%) mild steels are b CRA-T 5; c, d CRA-T 10; e, f CRA-T 120; g, h CRA-T 600

123
 Y.R. Wang et al.

Fig. 9 ECCI microstructure of CRA specimens. a CRA-Q; b CRA-T

5; c CRA-T 600

martensite is in the range of 1015 to 1016 m-2, which

mainly depends on the carbon content [37] as well as
weakly related with other factors, e.g., alloying elements. A
large number of studies have been conducted on the
changes in dislocation density during tempering at different
temperatures and time [35–41]. A similar trend in dislo-
cation evolution, i.e., decreases sharply in the first few
Fig. 8 SEM microstructure of HR specimen tempered at 300 C. a,
seconds and goes down slightly afterward, and then keeps
b HR-T 5; c, d HR-T 10; e, f HR-T 120; g, h HR-T 600 almost constant at the end, is always observed, no matter
how the chemical composition varies [35–39]. Therefore,
to the faster grain boundary diffusion. In contrast, the sizes here we did not present the conventional X-ray diffraction
of intra-lath carbides remain nearly constant. (XRD) measurements to estimate the dislocation density,
To further understand the as-quenched and tempered even though it could provide an approximate value and
microstructure, typical ECCIs are also given in Fig. 9. The strongly depends on the data process and equipment set-
lath boundaries are straight in the as-quenched specimen. tings. Meanwhile, the supersaturated carbon atoms escape
Two types of laths, i.e., thin (* 80 nm) and coarse from the matrix, and various defects form carbides, leading
(* 640 nm), are found, which is similar to the earlier to the relaxation of lattice distortion and reduction of solid
reports in Fe–C–Cr alloys by Hou et al. [35, 36]. After solution strengthening [37, 38]. These microstructural
tempering for 5 min, the lath boundaries become curved changes would contribute to the variation of mechanical
instead of straight all the way, as seen in Fig. 9b. After properties, including strength and ductility, referred to as
tempering for 600 min, a large number of circular precip- microhardness, as seen in Fig. 10. Notably, there is a small
itates are coarsened to a size of 34 ± 6 nm in diameter, increment in the HR-Q specimen after tempering for
and a few elongated precipitates are also observed (seen in time longer than 10 min, which could be related to the
Fig. 9c). The width of laths also coarsens from 209 ± 43 to precipitation strengthening caused by a considerable
788 ± 103 nm. number of fine precipitates formed during tempering
[21, 40, 41]. With the prolonged tempering time, the
3.3 Mechanical properties microhardness for both types of specimens, i.e., CRA-T
600 and HR-T 600, are almost constant. The lath marten-
After quenching, the microhardness of the two specimens site is stabilized during the tempering due to the strong
is 480 ± 5 HV, which mainly depends on the carbon pinning effect of fine precipitates [33, 35, 42, 43] and the
content in a specific range, e.g., \ 0.65 wt.% [36]. Nor- low driving force of boundary migration at such low
mally, the dislocation density of the as-quenched temperatures, i.e., 300 C [37].

123
Effect of initial microstructure on microstructure evolution and mechanical properties…

Table 2 UTS 9 TE of CRA and HR specimens under different

conditions
Specimen UTS 9 TE/GPa%

CRA-Q 18.9 ± 0.85

CRA-T 5 15.3 ± 1.13
CRA-T 120 16.3 ± 2.03
CRA-T 600 15.7 ± 0.90
HR-Q 24.1 ± 0.59
HR-T 5 24.4 ± 0.98
HR-T 120 17.7 ± 0.90
HR-T 600 23.2 ± 1.19

Fig. 10 Microhardness of as-quenched and tempered HR and CRA

specimens (1158 ± 13 MPa) 9 TE (21.08% ± 0.6%), i.e., 24.4
GPa%, is obtained after 5 min of tempering. After tem-
pering for 120 min, the UTS 9 TE of the CRA specimen
are 853 ± 15 MPa and 20.77% ± 2.0%, respectively,
while the tensile strength of the HR specimen reached
799 ± 20 MPa at 120 min of tempering, and the elonga-
tion of the HR specimen reached a maximum value of
28.45% ± 0.6% at tempering for 600 min.

4 Discussion

4.1 Effect of initial microstructure on quenching

and tempering microstructure

No matter with the CRA and HR specimens, polycrys-

talline ferrite and blocky pearlite are the initial
microstructure. The only main difference between the CRA
and HR specimens is the morphology and size of the ferrite
grain, as seen in Fig. 5a, b and Fig. 6a, b. This is mainly
because a significant recovery and even recrystallization
occur during the hot rolling at elevated temperatures
[40, 44]. The CRA specimen shows a typical full recrys-
tallization microstructure, instead of a deformed feature,
which is observed due to cold rolling and annealing, while
Fig. 11 Tensile properties of CRA and HR steels under different in the HR specimen, the elongated pearlite and ferrite along
conditions. a, b Stress–strain curve; c, d ultimate tensile strength, the hot rolling direction are obtained.
yield strength, and elongation; e, f UTS 9 TE
After water quenching, the width of the lath in the HR-Q
specimen is coarser than that in the CRA-Q, as seen in
Fig. 6c, d. The finer the initial microstructure in the steel,
The mechanical properties of the HR and CRA speci- the finer the austenite grains and thus the as-quenched
mens after tempering for different time are shown in martensitic microstructure [23, 24, 45–47]. Thus, it is
Fig. 11. The tensile properties with the tempering time concluded that the initial austenite grains can refine the
shown in Fig. 11a, d are consistent with the microhardness substructure of martensite, such as block and lath [48, 49].
with the tempering time shown in Fig. 10. According to The Ms temperature in the steels, i.e., below 0.1–0.2 wt.%
Fig. 11e, f and Table 2 (listing the UTS 9 TE), the max- C, is always higher than the room temperature, and the
imum UTS (1248 ± 1 MPa) 9 TE (15.1% ± 0.7%), i.e., carbon can partition carbon from the super-saturated fresh
18.9 GPa% is obtained in the as-quenched state for CRA martensite to various defects, such as dislocation, to form
specimen, while in the HR specimen, the maximum UTS carbides after quenched below the Ms temperature [50, 51].

123
 Y.R. Wang et al.

Fig. 12 Schematic of microstructure evolution in studied steels microstructure (fresh martensite, thicker black lines represent prior
during quenching and tempering of CRA (a–c) and HR (d–f). a, austenite grains); c, f tempered martensite obtained by long-term
d Initial microstructure (ferrite and pearlite); b, e as-quenched tempering

During the auto-tempering, the width of primarily formed obtained laths are coarser. Quantity studies have shown that
laths becomes coarse due to boundary migration and merg- the finer the initial grain size, the more potential nucleation
ing. The diffusion distance of carbon is also longer than the sites of the precipitate phase, and therefore the higher the
lath formed at the late stage, which is always regarded as the volume fraction of the precipitates forms in a short time [53].
more prone auto-tempering area, as shown in Fig. 6e, f. From SEM taken in randomly selected areas (Figs. 7 and 8
After being tempered for a short time, the as-quenched represent that of the CRA and the HR specimens, respec-
martensite has very high density dislocations, which pro- tively), it can be seen that the precipitates in the CRA
vides plenty of potential nucleation sites for the precipitate specimens occupy a larger volume fraction.
[49]. Then, a large number of dispersed precipitates are
rapidly formed at intra/inter laths. With the extension of the 4.2 Effect of initial microstructure on mechanical
tempering time, the size of the precipitates gradually grows, properties
which leads to the weakening of the pinning effect of the
precipitates, and the laths are also coarse [52]. The difference After quenching, full lath martensite is observed in both
in the initial microstructure of CRA and HR specimens CRA and HR specimens, showing a microhardness of
would inhabit the as-quenched microstructure, leading to 480 ± 5 HV. This is consistent with the earlier report on
microstructure differences, e.g., the density of dislocation steels with similar carbon content [36] and mainly depends
and solid solution of elements in the matrix phase. Then, on the carbon content rather than other alloying elements.
during the tempering, the microstructure evolution of After tempering at 300 C for various time, the micro-
martensite, including migration of various grain boundaries, hardness changing trend in these two specimens is roughly
nucleation, and growth of precipitates, are strongly related to the same. The microhardness decreases abruptly in the first
the as-quenched one, which is the main reason for the 5 min tempering and then keeps constant within a specific
microstructure difference in the CRA and HR specimens. range, i.e., from 480 ± 5 HV to 410 ± 5 HV (CRA) and
The schematic of microstructure evolution during the heat- 401 ± 8 HV (HR) as tempering time changes from 10 to
ing treatment is summarized in Fig. 12. In the final tempered 600 min. During the first 5 min tempering, the lattice dis-
microstructure, the volume fraction of the precipitates in the tortion is released, and the dislocation density is signifi-
CRA specimen is larger, and precipitates are evenly dis- cantly decreased due to the recovery of as-quenched
persed across the matrix, while the precipitates in the HR martensite [37, 53]. Thus, the microhardness and tensile
specimen appear to be primarily at lath boundaries, and the strength of the two specimens rapidly reduced for both the

123
Effect of initial microstructure on microstructure evolution and mechanical properties…

CRA and HR specimens. After tempering for 30 min, the Table 3 Precipitation contributions as a function of tempering time
microhardness of the CRA specimen is almost unchanged Tempering Diameter/ Shear Burgers Precipitation
due to the synergy precipitation strengthening and matrix time/min nm modulus/ vector/ strengthening/
softening. At the same time, the microhardness of the HR MPa nm MPa
specimen is slightly increased. With the prolonged tem-
0 1 8.3 9 104 0.248 173
pering time, the matrix softening and precipitation
600 34 8.3 9 104 0.248 690
strengthening reach an equilibrium condition, leading to a
stable microhardness for both specimens. It should be
noticed that the tensile strength of both types of specimens to 1158 ± 13 MPa (HR). When the tempering time is
decreases with the tempering time prolonging. prolonged to 120 min or even 600 min, the tensile strength
Unlike the microhardness, the tensile strength and total of CRA specimen decreases by 368–395 MPa, and the
elongation of CRA-Q specimens are slightly higher and elongation increases by 1.7%–5.7%. Meanwhile, the tensile
lower than those of the HR-Q specimen, respectively. This strength of HR specimen decreased by 349–370 MPa, and
may be caused by the combined effect of grain boundary the elongation increased by 2.3%–8.4%. The findings pave
strengthening [40] due to the fine substructure in the CRA the way for high-performance steel design by optimizing
specimen. When tempering at 300 C for various time, the the initial microstructure before quenching and tempering.
softening of the matrix leads to a decrease in the micro-
hardness, especially at the very early stage, owing to the
rapid release of internal stress and annihilation of high 5 Conclusions
mobility of dislocation [37]. This leads to a decrease in the
tensile strength and an increase in the total elongation. 1. The microstructure of as-quenched CRA and HR
Meanwhile, the supersaturated alloying elements will dif- specimens is fully martensite, showing a microhard-
fuse to the place with carbon segregation or defects to form ness of 480 ± 5 HV for both specimens, no matter the
carbides. The Ashby-Orowan model is adapted for cubic morphology and grain size of the initial microstructure.
polycrystalline materials with spherical precipitates and is The different features in microstructure between the
used to evaluate the precipitation strengthening, HR and CRA specimens are the size, morphology, and
1
!   distribution of ferrite and pearlite, leading to a differ-
0:538Gbf 2 X
Drppt ¼ ln ð1Þ ence in the as-quenched microstructure.
X 2b
2. After tempering at 300 C for different time, a compet-
where Drppt is the amount of precipitation strengthening; itive process of precipitation strengthening and matrix
G is the shear modulus, with a value of 8.3 9 104 MPa; phase softening occurs. At 5 min tempering, the micro-
b is the magnitude of the Burgers vector taken by hardness of the HR and CRA specimens decreased
0.248 nm; f is the volume fraction of the precipitates; and rapidly due to the matrix softening during the very early
X is the diameter of the precipitates expressed in nm [54]. stage of the fast recovery, and at 10 min tempering, a
According to the equilibrium phase diagram as shown in large number of precipitates were nucleated at inter/intra
Fig. 1, the volume fractions of M2B, M7C3 and M23C6 laths. The microhardness of the CRA specimen
precipitates at 300 C are 0.04%, 1.06% and 0.48%, remained stable, while the microhardness of the HR
respectively. Thus, we assume that the volume fractions are specimen increased slightly. At 120 min tempering, the
0.79% (half of the equilibrium value) at the initial stage microhardness of the two specimens remained stable.
and 1.58% after tempering at 300 C for 600 min, 3. The mechanical properties of the studied steel can be
respectively. We assume that the initial size of precipitates tuned in an extensive range via initial microstructure
is 1 nm (at the very early stage of nucleation and growth, control together with tempering parameters. The
i.e., as-quenched condition), while that is about 34 nm after maximum UTS 9 TE, i.e., 18.9 GPa%, of the CRA
tempered at 300 C for 600 min according to the specimen, is obtained in the as-quenched specimen. In
microstructure observation. Thus, the strengthening con- contrast, for the HR specimen, that is 24.4 GPa% after
tribution by the precipitates varies in the range of being tempered at 300 C for 5 min.
173–690 MPa, as shown in Table 3. Therefore, with tem-
pering time prolonging, the size of precipitates gets coar- Acknowledgements This research was funded by the Chongqing
ser, as seen in Fig. 9, and the precipitation-strengthening Natural Science Foundation (No. CSTB2022NSCQ-MSX1394),
Graduate Research and Innovation Foundation of Chongqing, China
effect weakens [37]. When tempering for 5 min, the tensile (Grant No. CYS22008), Open Project of State Key Laboratory of
strength of the two specimens decreases, i.e., from Advanced Special Steel, Shanghai Key Laboratory of Advanced
1248 ± 1 to 1159 ± 28 MPa (CRA) and from 1176 ± 21 Ferrometallurgy, Shanghai University (SKLASS 2023-10) and the

123
 Y.R. Wang et al.

Science and Technology Commission of Shanghai Municipality (No. [25] H. Järvinen, M. Isakov, T. Nyyssönen, M. Järvenpää, P. Peura,
19DZ2270200), and the Open Project of the Large Casting and Mater. Sci. Eng. A 676 (2016) 109–120.
Forging Manufacturing Technology Engineering Center of Shanghai [26] H. Mirzadeh, M. Alibeyki, M. Najafi, Metall. Mater. Trans. A 48
Institute of Mechanical and Electrical Engineering, State Key Labo- (2017) 4565–4573.
ratory of Vanadium and Titanium Resources Open Fund [27] S. Yan, X. Liu, T. Liang, Y. Zhao, Mater. Sci. Eng. A 712
(No. 2022P4FZG04A). (2018) 332–340.
[28] S. Morito, H. Tanaka, R. Konishi, T. Furuhara, T. Maki, Acta
Mater. 51 (2003) 1789–1799.
Declarations [29] B. Hutchinson, J. Hagström, O. Karlsson, D. Lindell, M. Torn-
berg, F. Lindberg, M. Thuvander, Acta Mater. 59 (2011)
Conflicts of interest Zi-yong Hou is a youth editorial board mem- 5845–5858.
ber for Journal of Iron and Steel Research International and was not [30] L. Morsdorf, E. Emelina, B. Gault, M. Herbig, C.C. Tasan, Acta
involved in the editorial review or the decision to publish this arti- Mater. 205 (2021) 116521.
cle. The authors declare no conflict of interest. [31] L. Morsdorf, C.C. Tasan, D. Ponge, D. Raabe, Acta Mater. 95
(2015) 366–377.
[32] H.L. Cai, P. Chen, J.K. Oh, Y.R. Cho, D. Wu, H.L. Yi, Scripta
References Mater. 178 (2020) 77–81.
[33] Z. Hou, R.P. Babu, P. Hedström, J. Odqvist, Mater. Sci. Tech-
[1] E. Scharifi, T. Schade, A. Ademaj, S.V. Sajadifar, U. Weidig, T. nol. 36 (2020) 887–893.
Niendorf, K. Steinhoff, Steel Res. Int. 92 (2021) 2000633. [34] Y.M. Zhu, Q. Zhang, Q. Yuan, J.S. Guan, Y.B. Yang, G. Xu,
[2] H.X. Yin, Z.Z. Zhao, A.M. Zhao, X. Li, H.J. Hu, J.T. Liang, J. Arab. J. Sci. Eng. 47 (2022) 9061–9073.
Iron Steel Res. Int. 22 (2015) 622–629. [35] Z.Y. Hou, R.P. Babu, P. Hedström, J. Odqvist, J. Mater. Sci. 53
[3] W. He, B.L. Wang, Y. Yang, Y.H. Zhang, L. Duan, Z.P. Luo, (2018) 6939–6950.
C.J. Song, Q.J. Zhai, J. Iron Steel Res. Int. 25 (2018) 830–838. [36] Z. Hou, P. Hedström, Y. Xu, W. Di, J. Odqvist, ISIJ Int. 54
[4] M. Tisza, I. Czinege, Int. J. Light. Mater. Manuf. 1 (2018) (2014) 2649–2656.
229–238. [37] G. Krauss, Steel Res. Int. 88 (2017) 1700038.
[5] H. Karbasian, A.E. Tekkaya, J. Mater. Process. Technol. 210 [38] F. Liu, K. Chen, C. Kang, Z. Jiang, S. Ding, J. Mater. Res.
(2010) 2103–2118. Technol. 19 (2022) 779–793.
[6] C.P. Couto, R.I. Revilla, R. Politano, I. Costa, Z. Panossian, I. De [39] W. Ding, T. Wu, L. Sun, C. Yan, X. Lu, X. Lei, Steel Res. Int.
Graeve, J.L. Rossi, H. Terryn, Corros. Sci. 190 (2021) 109673. 93 (2022) 2100627.
[7] W. Huang, H. Gu, Q. Liu, T. Si, Mater. Chem. Phys. 256 (2020) [40] A. Dutta, D. Ponge, S. Sandlöbes, D. Raabe, Mater. Sci. Forum
123729. 941 (2018) 198–205.
[8] Y.X. Qiao, Z.B. Zheng, H.K. Yang, J. Long, P.X. Han, J. Iron [41] B. Dou, H. Zhang, Y. Tao, Q.S. Ma, S. Guo, Tungsten 5 (2023)
Steel Res. Int. 30 (2023) 1463–1476. 189–197.
[9] K. Mori, P.F. Bariani, B.A. Behrens, A. Brosius, S. Bruschi, T. [42] A. Kostka, K. Tak, R. Hellmig, Y. Estrin, G. Eggeler, Acta
Maeno, M. Merklein, J. Yanagimoto, CIRP Ann. 66 (2017) Mater. 55 (2007) 539–550.
755–777. [43] Y.H. Yu, Z.D. Liu, Z. Liu, Z.Z. Chen, H.S. Bao, C. Zhang, Z.G.
[10] S.S. Li, H.W. Luo, Int. J. Miner. Metall. Mater. 28 (2021) Yang, J. Iron Steel Res. Int. 30 (2023) 1279–1290.
741–753. [44] B. Hu, X. Tu, H. Luo, X. Mao, J. Mater. Sci. Technol. 47 (2020)
[11] H. So, D. Faßmann, H. Hoffmann, R. Golle, M. Schaper, J. 131–141.
Mater. Process. Technol. 212 (2012) 437–449. [45] X. Liu, K.Y. Zhao, Y.H. Zhou, D. Ye, W. Jiang, Q.L. Yong, J.
[12] T. Taylor, A. Clough, Mater. Sci. Technol. 34 (2018) 809–861. Su, Adv. Mater. Res. 393–395 (2011) 440–443.
[13] I. Wróbel, A. Skowronek, A. Grajcar, Symmetry 14 (2022) 969. [46] M.A. Shtremel’, V.I. Lizunov, V.V. Shkatov, Met. Sci. Heat
[14] B. Hu, H. Luo, F. Yang, H. Dong, J. Mater. Sci. Technol. 33 Treat. 21 (1979) 747–750.
(2017) 1457–1464. [47] N.T. Egorov, M.S. Podgaiskii, L.I. Razumova, Met. Sci. Heat
[15] J. Liang, H. Lu, L. Zhang, F. Li, R. Cao, K. Liu, H. Pan, H. Treat. 21 (1979) 861–865.
Teng, X. Li, A. Guo, Z. Zhao, Mater. Sci. Eng. A 801 (2021) [48] D. Ning, C.R. Dai, J.L. Wu, Y.D. Wang, Y.Q. Wang, Y. Jing, J.
140419. Sun, Mater. Charact. 176 (2021) 111111.
[16] Y.T. Lin, H.L. Yi, Z.Y. Chang, H.C. Lin, H.W. Yen, Front. [49] M.S. Mohsenzadeh, J. Mater. Eng. Perform. 32 (2023) 164–169.
Mater. 7 (2020) 611390. [50] G. Badinier, C.W. Sinclair, X. Sauvage, X. Wang, V. Bylik, M.
[17] K. Zhang, P. Liu, W. Li, Z. Guo, Y. Rong, Mater. Sci. Eng. A Gouné, F. Danoix, Mater. Sci. Eng. A 638 (2015) 329–339.
619 (2014) 205–211. [51] T. Kohne, J. Fahlkrans, A. Stormvinter, E. Maawad, A.
[18] Y.J. Li, G. Yuan, L.L. Li, J. Kang, F.K. Yan, P.J. Du, D. Raabe, Winkelmann, P. Hedström, A. Borgenstam, Metall. Mater.
G.D. Wang, Science 379 (2023) 168–173. Trans. A 54 (2023) 1083–1100.
[19] J.B. Dong, K. Li, L.X. Gao, C.S. Li, Tungsten 5 (2023) 512–521. [52] Z. Liu, Z. Liu, X. Wang, C. Dong, Z. Chen, H. Bao, Mater. Res.
[20] Y. Zhang, C. Zhao, M. Sato, G. Miyamoto, T. Furuhara, ISIJ Int. Express 6 (2019) 116527.
61 (2021) 1641–1649. [53] C.L. Miao, C.J. Shang, H.S. Zurob, G.D. Zhang, S.V. Subra-
[21] Y. Tian, H. Yu, T. Zhou, K. Wang, Z. Zhu, Mater. Charact. 188 manian, Metall. Mater. Trans. A 43 (2012) 665–676.
(2022) 111919. [54] T. Gladman, Mater. Sci. Technol. 15 (1999) 30–36.
[22] Z.B. Liu, Z. Yang, J.X. Liang, Z.Y. Yang, G.M. Sheng, J. Iron
Steel Res. Int. 29 (2022) 207–214. Springer Nature or its licensor (e.g. a society or other partner) holds
[23] H. Luo, X. Wang, Z. Liu, Z. Yang, J. Mater. Sci. Technol. 51 exclusive rights to this article under a publishing agreement with the
(2020) 130–136. author(s) or other rightsholder(s); author self-archiving of the
[24] C. Zhang, W. Hui, X. Zhao, Y. Zhang, X. Zhao, Eng. Fail. Anal. accepted manuscript version of this article is solely governed by the
135 (2022) 106144. terms of such publishing agreement and applicable law.

123