metals

Article
Development of New Cobalt-Free Maraging Steel
with Superior Mechanical Properties via
Electro-Pulsing Technology
Dong Pan, Yuguang Zhao *, Yitong Wang, Xiaofeng Xu and Xueying Chong
Key Laboratory of Automobile Materials, Ministry of Education and School of Materials Science and Engineering,
Jilin University, No. 5988 Renmin Street, Changchun 130022, China; pandong16@mails.jlu.edu.cn (D.P.);
wyj17@mails.jlu.edu.cn (Y.W.); xuxiaofeng@jlu.edu.cn (X.X.); zhongxy17@mails.jlu.edu.cn (X.C.)
* Correspondence: zhaoyg@jlu.edu.cn; Tel.: +86-431-8509-4481




Received: 11 November 2019; Accepted: 26 November 2019; Published: 2 December 2019


Abstract: The ductility of cobalt-free maraging steel is unsatisfactory because of the high content
of Ti. Traditional heat treatment can not effectively improve the ductility of this kind steel. In this
contribution, high-energy electro-pulsing is adopted in a T250 steel to solve this problem efficiently.
It is found that the EPS treatment (electro-pulsing treatment under water-cooling condition) can
accelerate the formation of retained austenite and nano stacking faults. Meanwhile, the microstructure
is also refined by EPS treatment. Then, taking the EPS sample as the initial state, nano-reverted
austenite combined with finer η-Ni3 Ti precipitates are formed during EPA treatment (electro-pulsing
treatment under air-cooling condition), compared with TA (traditional aging) treatment. The results of
mechanical properties indicate that the strength and elongation are both enhanced by electro-pulsing
treatment. Consequently, the electro-pulsing treatment can be a promising technology to devise
cobalt-free maraging steel with better properties.

Keywords: T250 steel; electro-pulsing; retained austenite; stacking faults; mechanical properties

1. Introduction
Since the 1960s, maraging steels have become irreplaceable materials in navigation and aviation
fields by virtue of their excellent strength, toughness, and corrosion resistance [1–3]. Generally,
compared with the quenched carbon steel, the maraging steels usually possess better ductility because
of their high nickel content [4]. The pre-solution treatment will dissolve alloying elements into the
matrix and after the water-cooling procedure, the matrix will be transformed into soft martensitic
laths. Whereas, the soft matrix is only the foundation of the advanced mechanical properties of
maraging steels and appropriate aging regime must be taken into force to guarantee a huge increment
in strength. During the aging process, some inter-metallic compounds, such as Ni3 Ti and Ni3 Mo with
hexagonal crystal structure will be formed, which will lead to the increment in strength [5]. In order to
further enhance the strengthening effect, the element cobalt is added to maraging steel for the dynamic
interaction between cobalt and molybdenum elements, i.e., the prior positions of molybdenum atoms
will be occupied by cobalt atoms so that the precipitation of Ni3 Mo compounds will be promoted [6].
However, accompanied with the gradually increased price and decreased resource reserve, cobalt is
becoming a strategic resource around the global, and under this background, cobalt-free maraging
steels are proposed [7].
Although the production costs of cobalt-free maraging steels decrease significantly, another
problem also come out. To attain the same strength level as the maraging steels containing cobalt, more
titanium is added to cobalt-free maraging steel [8]. In other words, less Ni3 Mo and more Ni3 Ti will be

Metals 2019, 9, 1299; doi:10.3390/met9121299 www.mdpi.com/journal/metals

Metals 2019, 9, 1299 2 of 15

formed after aging regime is applied. Whereas, due to the higher c/a (lattice constant) value of Ni3 Ti,
the lattice coherence between Ni3 Ti and the matrix is also lower than that between Ni3 Mo and the
matrix [9,10]. Thus, compared with maraging steels with cobalt content, the ductility of cobalt-free
maraging steels is unsatisfactory. Recently, plenty of research has concentrated on the enhancement of
ductility of maraging steels. Wang et al. [11] reported that during aging treatment, layered reverted
austenites will be formed. As the reverted austenite can reduce the {111} fiber in martensite [12] and
prevent the crack propagation during tensile deformation, the ductility of maraging steels will be
improved [13,14]. However, the content of reverted austenites cannot be easily controlled so that the
toughening (ductility improvement) effect is unstable. Besides, Pereloma [15] and Li [16] proposed
that by rapid heating treatment, the strength is remarkably enhanced because of the formation of
co-clusters and in contrast to maraging steels under traditional treatment, there is almost no loss in
ductility. However, the R/b (radius/burgers vector) value of the co-cluster (known as the prior series
of GP zone containing Mn, Ni, Ti, and Al) is very sensitive to treating parameters (temperature and
treating procedure). If this value is higher than 15, there will be almost no toughening effect, and
from the viewpoint of developing maraging steels with superior properties, no loss in ductility is
insufficient. Moreover, Lu et al. [17] demonstrated that the ductility of maraging steels can be modified
by the formation of Ni(Al,Fe) precipitations of which the lattice misfit is much lower than Ni3 Ti.
Whereas, the mass percentage of aluminum in this steel even reaches up to 3%. Such a high content of
aluminum will lead to the increment in viscosity of molten steel, and the production cost will also
increase. On the other hand, the grain refinement is considered to be the most effective method to
toughening maraging steels. Unfortunately, there exist some problems for traditional austenitic grain
(γ) refinement technology of maraging steels. Firstly, some severe plastic deformation technology,
such as equal channel angular process (ECAP) [18,19] and high pressure torsion (HPT) can effectively
reduce the grain (γ) size. However, a large amount of crystal defects will also be introduced so
that the toughening effect is weakened. Then, even though the cyclic γ↔α0 transformation can also
refine the grains (γ), there will exist intergranular oxidation because the high content of titanium
enriched at the unstable grain boundaries, and so, the effect ductility improvement will be weakened.
Finally, for recrystallization of maraging steels, not only the processing is time consuming, but also the
parameters (temperature and deformation amount) are rigorous [20]. As a result, a new and more
efficient toughening method is urgently needed for maraging steels.
Currently, as a rapid and effective treating technology, high-energy electro-pulsing processing
has been widely utilized in many fields, such as structural relaxation in amorphous alloys [21],
self-repairing of damage crack [22] and microstructural refinement [23]. For industrial manufacturing,
the electro-pulsing current is also applied on electro-plastic drawing [24] and auxiliary molding of
Ni-based alloys [25]. For steels, many studies have demonstrated the obviously improved strength
and ductility induced by electro-pulsing current [26,27]. Moreover, our previous works also indicated
that some substructures, such as sub-grains, nano-twins, and clusters, which are hard to be formed in
steels by traditional treating strategy and beneficial for the ductility, can be obtained via electro-pulsing
treatment [28,29]. In other words, the electro-pulsing treatment could be a suitable method for the
ductility enhancement of maraging steels.
Thus, this work is aimed at the gradually optimization of solution and aging treatment of
T250 steel by the comparison between traditional heat treatment and electro-pulsing heat treatment.
Besides, the microstructure-property relationship of T250 steels under different treatment will also be
closely investigated. It is expected to develop cobalt-free maraging steels with superior properties
more efficiently.
Metals 2019, 9, 1299 3 of 15

2. Experimental
Metals 2019, 12, x FOR PEER REVIEW 3 of 15

Metals 2019, 12, x FOR PEER REVIEW 3 of 15

2.1. Materials
2.1. Materials and and Methods
Methods
2.1. Materials
The materials and
The materials Methods
used used in this
in this work work is commercialT250
is commercial T250steel
steeland
andthethe initial
initial state
stateisisannealed
annealed(Figure
(Figure 1a,
1a, containing ferrite, Laves phase (compounds with Ti, Mo, and Al) and γ phase (FeNi 3)), and the
containingTheferrite,
materials Laves
usedphase
in this (compounds
work is commercial withT250
Ti, Mo,steel and
and theAl)initial
and state
γ phase (FeNi
is annealed )), and the
3(Figure
chemical composition (weight percentage: wt.%) of the T250 steel is listed in Table 1. The T250 steel
1a, containing
chemical composition ferrite, Laves phase
(weight (compounds
percentage: wt.%) with
of theTi, T250
Mo, and Al)isand
steel listedγ phase (FeNi
in Table 1.3)),
The and the steel
T250
was machined
chemical into slices
composition (50 × percentage:
(weight 10 × 2.5 mm3wt.%)
3) by CNC wire cutting machine (Jiangsu Dongqing CNC
of the T250 steel is listed in Table 1. The T250 steel
was machined
Machine
intoCo.,
Tool into
slices
LTD, (50
× 10 × 2.5
(50Taizhou, mm )Then,
China).
by CNCall the
wire cuttingwill machine (Jiangsu and Dongqing CNC
was
Machine machined
Tool Co., LTD, slices
Taizhou, ×China).
10 × 2.5 mm 3) by CNC
Then, all the wireSamples
Samples cutting
will machine
be
be traditionally
(Jiangsu Dongqing
traditionally and
electro-
CNC
electro-pulsing
pulsing treated,
Machine Tool Co., respectively,
LTD, Taizhou,and the treating
China). Then,flow
all is
theshown
Samplesin Figure
will be1b. As is seen,and
traditionally the electro-
state of
treated, respectively,
samples for aging and the treating
treatment is flow isaccording
optimized shown intoFigure the 1b. As isproperties
mechanical seen, the of state of samples
samples after for
pulsing treated, respectively, and the treating flow is shown in Figure 1b. As is seen, the state of
agingsolution
treatment
heat istreatment.
optimized according
Traditional to the heat
solution mechanical
treatment properties
(TS) was of samples
carried out at after
820 °Csolution
for 1 h heat
samples for aging treatment is optimized according to the mechanical properties of samples after
under Traditional
treatment. water-cooling condition.
solution heatWhile the corresponding
treatment (TS) electro-pulsing ◦treatment (EPS) was taken
solution heat treatment. Traditional solution heatwas carried
treatment out at
(TS) was820 C forout
carried 1 hat
under
820 °C water-cooling
for 1h
into
condition.force via using a self-made machine containing thyristor circuit (Figure 2b) under water-cooling
under While the corresponding
water-cooling condition. Whileelectro-pulsing
the corresponding treatment (EPS) was
electro-pulsing taken into
treatment (EPS)force
wasvia using a
taken
condition.
self-made
into machine
force containing
via using thyristor
a self-made machine circuit (Figure
containing 2b) under
thyristor water-cooling
circuit (Figure 2b) under condition.
water-cooling
condition.
Table 1. The nominal composition of commercial T250 steel (wt.%).
Table 1. The nominal composition of commercial T250 steel (wt.%).
Element Ni Table 1.CrThe nominal Mocomposition Tiof commercial
Al T250 steelSi(wt.%). Mn Fe
Element
wt.% Ni18.86 Cr
0.37 Mo
3.06 Ti
1.20 Al
0.11 0.01Si 0.01Mn Bal. Fe
Element Ni Cr Mo Ti Al Si Mn Fe
wt.% Note: 18.86
content of0.37
The18.86 C, Co, S, and3.06 1.20
P are all less1.20
than 0.003%,0.11 of0.01
the content0.01 in 0.01
O and N0.01the molten Bal.
wt.% 0.37 3.06 0.11 Bal.
steel
Note: Note: are lessof
The content than 20 ppm.
C, Co,
The content of S,
C,and
Co, PS,are
andallPless
arethan 0.003%,
all less thanthe content
0.003%, theofcontent
O and NofinOthe molten
and N in steel are less than
the molten
20 ppm.
steel are less than 20 ppm.

Figure 1. The microstructure of T250 steel at annealed state (a) and the Schematic diagram showing
the treating flow of this work (b).
Figure
Figure 1. The1. The microstructure
microstructure of T250
of T250 steel
steel at at annealedstate
annealed state(a)
(a)and
andthe
the Schematic
Schematic diagram
diagramshowing
showing the
the treating flow of this work (b).
treating flow of this
2.2. Technological work (b).
Parameters and the Electro-Pulsing Apparatus
2.2. Technological Parameters and the Electro-Pulsing Apparatus

Figure 2. The physical map of the electro-pulsing apparatus (a) and the sketch map of the thyristor
circuit (b).
Metals 2019, 9, 1299 4 of 15

2.2. Technological Parameters and the Electro-Pulsing Apparatus

The electro-pulsing apparatus and thyristor circuit are shown in Figure 2, the samples are fastened
on the Cu–Zr electrodes by the fastening bolts, and the distance between the two electrodes can
be adjusted by the sleeves. The electro-pulsing parameters were set up by an EPT software, and
the detailed parameters (current density and period) are 8.15 × 107 A/m2 and 360 ms, respectively.
For aging process, the parameters of traditional treatment (TA) and electro-pulsing treatment (EPA)
under air-cooling conditions were 480 ◦ C, 2 h and 3.40 × 108 A/m2 , 220 ms and 280 ms, respectively.
The temperature during electro-pulsing treatment is measured by an infrared gun (TBSC).

2.3. Microstructural Analysis and Property Measurement

Samples for optical microscopy (OM) was polished and then etched with ethanol +6.25 wt.% iron
trichloride +20 vol.% hydrochloric acid solution. The Axio Imager M2M optical microscopy (Carl Zeiss
AG, Oberkochen, Germany) was used for grain boundary and microstructure observation. Samples
for transmission electron microscopy (TEM) test were polished to about 30 µm thickness and then the
twin-jet electro-polishing in a 20 vol.% perchloric ethanol electrolyte at −20 ◦ C under a voltage of 20 V
was adopted. The JEM-2100F transmission electron microscopy (JEOL LTD, Tokyo, Japan) is adopted
for substructure and precipitation detection. The X-ray diffraction (XRD) was conducted by using a
D/Max 2500PC XRD machine (Scanning range was 20◦ –110◦ and scanning rate was 2◦ /min) (Rigaku
Coporation, Tokyo, Japan). The tensile test was conducted at room temperature by using a MTS-810
performance testing machine (MTS Industrial System (China) Co., LTD, Shenzhen City, China), and
the tensile speed is 0.003 s−1 .

3. Results and Discussion

3.1. Optimization of Solid-Solution Treatment

Figure 3 shows the microstructure and prior austenitic grain boundary of samples under different
kinds of solution heat treatment. The microstructure of TS sample is composed of lath-like martensite.
As for the EPS sample, the corresponding microstructure is also mainly comprised of lath-like martensite
(the temperature is measured to be 807 ◦ C, which is above Ac3 temperature of T250 steel [30]). Whereas,
the width of martensitic laths in the EPS sample is obviously smaller than that of the TS sample. Besides,
the prior austenitic grains of EPS sample are also finer than the TS sample (Figure 3c,d). The above
mentioned results indicate that the martensitic laths and prior austenitic grains of T250 steel are refined
by electro-pulsing treatment, and the refined martensite is a consequence of the refined prior austenitic
grains [31]. This can be attributed to the enhanced nucleation rate of austenitic transformation induced
by electro-pulsing current [32]. Furthermore, the growth of austenitic grains are also inhibited due
to the extremely short treating procedure of EPS treatment. In this way, finer microstructure can be
inevitably obtained by electro-pulsing treatment.
Metals 2019, 9, 1299 5 of 15
Metals 2019, 12, x FOR PEER REVIEW 5 of 15

Figure
Figure 3.
3. Optical
Optical microscopy
microscopy (OM)
(OM) images
images showing
showing the
the microstructure
microstructure of
of the
the traditional
traditional solution
solution heat
heat
treatment
treatment (TS)
(TS) sample
sample (a),
(a), the electro-pulsing treatment
the electro-pulsing treatment (EPS)
(EPS) sample
sample (b),
(b), the prior austenitic
the prior austenitic grain
grain
boundary
boundary ofof the
the TS
TS sample (c), and
sample (c), and the
the EPS
EPS sample
sample (d).
(d).

In order to deeply
deeply investigate
investigate the substructure
substructure of martensite in the EPS EPS sample,
sample, TEMTEM observation
observation
was conducted. As depicted in Figure 4, the width of martensitic laths in the TS sample is larger than
the EPS sample (Figure 4a,e), which is consistent with the results of Figure 3. In addition, compared
theTS
with the TSsample,
sample,thethe
EPSEPS presents
presents severer
severer dislocation
dislocation pile-ups
pile-ups (Figure
(Figure 4b,f). Furthermore,
4b,f). Furthermore, the
the results
results
of Figure of 4d
Figure 4d demonstrate
demonstrate thatalso
that there there also
exist existstacking
some some stacking faults
faults in in the
the EPS EPS sample.
sample. It is
It is worth
worth
noting noting
that somethatlayered
some layered
retainedretained
austeniteaustenite is also in
is also detected detected in EPS
EPS sample sample
(Figure (Figure
4a,c), 4a,c),
and the and
crystal
orientation between retained austenite and martensite is the K–S relationship [33]: (011)α //(111)γ ,
the crystal orientation between retained austenite and martensite is the K–S relationship [33]:
(011)
[–100] α //[1–10]
α//(111) γ, γ
[–100] α//[1–10]
. whereas γ. whereas
in the TS sample, in almost
the TSnosample,
retainedalmost no can
austenite retained austenite can
be distinguished. be
To get
distinguished. To get ridofof
rid of the scale limitation TEMtheobservation
scale limitation
in theof TEM observation
detection in the detection
of retained austenite of retained
in different samples,
austenite
XRD detectionin different samples,
is necessary XRD
to be detection is necessary to be conducted.
conducted.
Metals 2019, 9, 1299 6 of 15
Metals 2019, 12, x FOR PEER REVIEW 6 of 15

Figure 4.4.Transmission
Figure Transmission electron
electron microscopy
microscopy (TEM)(TEM)
imagesimages
showingshowing the martensitic
the martensitic laths (a),
laths (a), dislocations
dislocations in martensitic laths (b), retained austenite (c), nano stacking faults (d)
in martensitic laths (b), retained austenite (c), nano stacking faults (d) in the EPS sample and in the EPS sample
the
and the martensitic
martensitic laths (e),laths (e), dislocations
dislocations in martensitic
in martensitic laths
laths (f) in the (f)
TS in the TSthe
sample; sample;
insert the
maps insert
showmaps
the
show theorpatterns
patterns HRTEMorimageHRTEM image
of the of thearea
selected selected
in thearea in the corresponding
corresponding figures. figures.

The results of the XRD measurement

measurement are are shown
shown in Figure
Figure 5. Characteristic peaks of austenite are
found in the XRD patterns of the EPS sample. However, no characteristic peaks of retained austenite
can be detected in the TS sample. The XRD results imply that the electro-pulsing treatment promotes
the formation
formationof ofretained
retainedaustenite.
austenite.ForForthetheformation
formation mechanisms,
mechanisms, thethehomogeneously
homogeneously distributed
distributed Ni
content in samples
Ni content at initial
in samples state,
at initial the crystal
state, defects
the crystal in austenitic
defects grains
in austenitic and the
grains andgrain refinement
the grain effect
refinement
after
effectEPS treatment
after should should
EPS treatment be takenbeinto consideration.
taken Firstly, because
into consideration. Firstly,the treating
because theperiod is extremely
treating period is
short, there’s
extremely no sufficient
short, there’s notime for Ni time
sufficient contentfor to
Nihomogeneously dissolve intodissolve
content to homogeneously the matrix.
intoBecause
the matrix.the
Ni is an austenitic
Because the Ni is an stabilizing
austeniticelement, the element,
stabilizing retained the
austenite
retainedis easier to beisformed
austenite easier toinbe
theformed
Ni enriched
in the
area in the EPS
Ni enriched areasample [34].sample
in the EPS Secondly,
[34].under the intensely
Secondly, under theimpact of the
intensely electron
impact of thewind, there
electron will
wind,
exist
theremore crystal
will exist moredefects in austenitic
crystal defects ingrains during
austenitic EPSduring
grains treatment,
EPS and as a result,
treatment, and the
as amartensitic
result, the
transformation will be inhibited
martensitic transformation will [35]. Finally, because
be inhibited the average
[35]. Finally, because grain
the size of thegrain
average EPS sample (9.7 EPS
size of the µm,
Figure
sample3d,(9.7calculated
μm, Figureby3d, Image Pro Plus
calculated 6.0 software)
by Image Pro Plus (6.0
6.0version,
software) Media Cybernetics,
(6.0 version, MediaMD, USA) is
Cybernetics,
remarkably
MD, USA) issmaller than that
remarkably of the
smaller thanTSthat
sample
of the(38.2 µm, Figure
TS sample (38.23c,
μm,calculated
Figure 3c, bycalculated
Image ProbyPlus Image6.0
Pro Plus 6.0 software), the interface energy of EPS sample between austenite and martensite is higher
Metals 2019, 9, 1299 7 of 15

Metals 2019, 12, x FOR PEER REVIEW 7 of 15

software), the interface energy of EPS sample between austenite and martensite is higher than the TS
sample, thus, the growth of martensitic in the EPS sample will stop earlier than the TS sample when
than the TS sample, thus, the growth of martensitic in the EPS sample will stop earlier than the TS
the growth is about to attach the prior austenitic grain boundary. As a result, more retained austenite
sample when the growth is about to attach the prior austenitic grain boundary. As a result, more
is formed in the EPS sample [36].
retained austenite is formed in the EPS sample [36].

Figure
Figure5.5.X-ray
X-raydiffraction
diffraction(XRD)
(XRD)patterns
patternsof
ofsamples
samplesunder
underdifferent
different kinds
kinds of
of treatment.
treatment.

Generallyspeaking,
Generally speaking,thethestacking
stackingfaults
faultsare
arehard
hardtoto form
form in in ferrous
ferrous materials
materials with
with bcc
bcc structure
structure
due to the relatively higher stacking fault energy. As for the formation mechanisms
due to the relatively higher stacking fault energy. As for the formation mechanisms of nano stacking of nano stacking
faults in
faults inthe
theEPS
EPSsample
sample(Figure
(Figure4d),
4d),the
theauthors
authorsbelieve
believethat
thatthethethermal
thermalcompressive
compressivestress
stressinduced
induced
by electro-pulsing
by electro-pulsing isis necessary
necessary toto be
be investigated,
investigated, based
based on on the
the previous
previous research
research [29]. During
During EPS
EPS
treatment,
treatment, due to the extremely rapid heating speed, the thermal expansion will be delayed
to the extremely rapid heating speed, the thermal expansion will be delayed [37], [37], which
will result
which will in the generation
result of thermal
in the generation compressive
of thermal stress (F
compressive T ) in (F
stress the
T) EPS sample.
in the The Fvalue
The value
EPS sample. T of can
FT
be estimated by [38]:
of can be estimated by [38]:
FT = Eλ∆T (1)
FT  EλΔT (1)
where λ is the thermal expansion coefficient, ∆T refers to the temperature rise, and E refers to Young’s
where λ is the thermal expansion coefficient, ΔT refers to the temperature rise, and E refers to Young’s
modulus. Whereas, in fact, there exist a critical force (FC ) for the formation of stacking faults, and its
modulus. Whereas, in fact, there exist a critical force (FC) for the formation of stacking faults, and its
value is governed by [39]:
value is governed by [39]: 2µαbp µa2
FC = + (2)
2μα
D bp μa 2 p
24πωb
FC   (2)
where D means the grain size of EPS sample,Dbp is the 24πω bp
partial dislocation vector, µ refers to shear
modulus, a denotes the lattice constant, and ω is the width of nano stacking faults in Figure 4d. If the
where D means the grain size of EPS sample, bp is the partial dislocation vector, μ refers to shear
actually driven force is smaller than this critical force, no stacking faults will be formed. During EPS
modulus, a denotes the lattice constant, and ω is the width of nano stacking faults in Figure 4d. If the
treatment, the thermal compressive stress can be regarded as the driven force for the formation of
actually driven force is smaller than this critical force, no stacking faults will be formed. During EPS
stacking faults. According to Equations (1) and (2), the calculated thermal compressive stress is about
treatment, the thermal compressive stress can be regarded as the driven force for the formation of
1.79 GPa, which is obviously higher than the calculated critical force (0.54 GPa). Therefore, it can be
stacking faults. According to Equations (1) and (2), the calculated thermal compressive stress is about
concluded that the formation of nano-stacking faults is attributed to the thermal compressive stress
1.79 GPa, which is obviously higher than the calculated critical force (0.54 GPa). Therefore, it can be
during EPS treatment.
concluded that the formation of nano-stacking faults is attributed to the thermal compressive stress
during EPS treatment.
The tensile properties of solution heat treated samples are shown in Figure 6. The yield strength
and elongation of TS sample are 794 MPa and 15.8%. For the EPS sample, the corresponding values
are 1045 MPa and 19.8%, which are apparently higher than that of the TS sample. The higher tensile
Metals 2019, 9, 1299 8 of 15

The tensile properties of solution heat treated samples are shown in Figure 6. The yield strength
and elongation of TS sample are 794 MPa and 15.8%. For the EPS sample, the corresponding values
are 1045 MPa and 19.8%, which are apparently higher than that of the TS sample. The higher
tensile properties of the EPS sample are attributed to its refined prior austenitic grains. Based on the
Metals 2019, 12, x FOR PEER REVIEW 8 of 15
Hall–Petch effect, with the decrease of prior austenitic grain size, the yield strength is bound to increase.
Furthermore,
propertiesthe of thedeformation
EPS sampleprocess will also
are attributed to itsbecome moreaustenitic
refined prior homogeneous grains. because
Based on of thethe
Hall–refined
priorPetch
austenitic
effect, grains.
with theMoreover,
decrease ofexcept for the refined
prior austenitic grain
grain size, the size,
yield the stacking
strength faults
is bound and retained
to increase.
Furthermore,
austenite the deformation
are also favorable for theprocess
better will alsoproperties.
tensile become more homogeneous
Under the pinningbecause
effectofofthe refinedfaults
stacking
prior austenitic grains. Moreover, except for the refined grain size, the stacking
on dislocation movement, the strength will be enhanced. As for the effect of retained austenite on faults and retained
austenite
mechanical are also favorable
properties of the EPSfor the better tensile
sample, it canproperties.
be reflectedUnder
by the
thepinning effect of stacking
work-hardening faults (Θ,
rate curves
on dislocation movement, the strength will be enhanced. As for the effect of retained austenite on
(dσtrue /dεtrue )/σtrue , σtrue is true stress and εtrue is true strain). As is depicted in Figure 6b, because of
mechanical properties of the EPS sample, it can be reflected by the work-hardening rate curves (Θ,
the finer grain size and martensitic laths, the dislocation are easier to be entangled with each other
(dσtrue/dεtrue)/σtrue, σtrue is true stress and εtrue is true strain). As is depicted in Figure 6b, because of the
in thefiner
EPSgrain
sample.
size andTherefore,
martensitic thelaths,
initial
thework-hardening
dislocation are easier rate
toofbe the EPS sample
entangled with eachis higher
other inthan
the the
TS sample. Moreover,
EPS sample. Therefore, in contrast
the initial to the disappeared
work-hardening work-hardening
rate of the EPS sample isability in the
higher than theTS
TS sample
sample. at a
true strain of 0.038,
Moreover, the EPS
in contrast sample
to the possesses
disappeared more significant
work-hardening continuous
ability work-hardening
in the TS sample capability
at a true strain of
(Figure 6b).theThis
0.038, EPScan be attributed
sample to the
possesses more prevention
significant of crackwork-hardening
continuous propagation incapability
the EPS(Figure
sample6b). during
This
tensile can be attributed
deformation, whichtois the prevention
better of crack propagation
for the improvement in the[40].
of ductility EPSDuring
sample this
during tensilewhen
process,
deformation,
the crack extends which is better retained
to the layered for the improvement
austenite, the of martensitic
ductility [40]. During this process,
transformation will bewhen the so
promoted
crack extends to the layered retained austenite, the martensitic transformation
that the further plastic deformation will be guaranteed and the work-hardening can be continued will be promoted so [40].
that the further plastic deformation will be guaranteed and the work-hardening can be continued
The above-mentioned effect of retained austenite on mechanical properties of the EPS sample is known
[40]. The above-mentioned effect of retained austenite on mechanical properties of the EPS sample is
as theknown
transformation induced plasticity (TRIP) effect.
as the transformation induced plasticity (TRIP) effect.

Figure
Figure 6. Tensile
6. Tensile properties
properties of samples
of samples under different
under different kinds ofkinds of solid-solution
solid-solution treatment:
treatment: (a)
(a) engineering
engineering tensile curves, (b) normalized work-hardening
tensile curves, (b) normalized work-hardening rate curves. rate curves.

3.2. Optimization
3.2. Optimization of Aging
of Aging Treatment
Treatment
In order
In order to develop
to develop T250
T250 steelwith
steel withhigher
higher strength
strength andandductility, thethe
ductility, EPSEPS sample withwith
sample betterbetter
properties is taken as the initial state of the aging treatment, according to the results
properties is taken as the initial state of the aging treatment, according to the results of Section of Section 3.1. 3.1.
The microstructure of the EPS + EPA samples with different treating period is given in Figure 7. After
The microstructure of the EPS + EPA samples with different treating period is given in Figure 7.
220 ms of the EPA treatment, a large number of rod-like η-Ni3Ti precipitates with a length of 32 nm
After 220 ms of the EPA treatment, a large number of rod-like η-Ni3 Ti precipitates with a length of
were found. Based on the SAED (Selected Area Electron Diffraction) results (the left insert in Figure
32 nm were
7a), found.
retained Basedison
austenite the
also SAED
found (Selected
in the EPS + EPAArea Electron
(220 Diffraction)
ms) sample. results
In contrast to the(the
EPSleft insert in
sample
Figure 7a), retained
(Figure 4c), the austenite
width of is also found
retained in theinEPS
austenite the +EPS
EPA + (220
EPA ms)
(220sample.
ms) sampleIn contrast to the EPS
was smaller.
sample (Figure 4c),
Furthermore, theare
there width
almost ofnoretained austenite
characteristic peaksin of the + EPA (220
EPS austenite
retained ms)
in the sample wasXRD
corresponding smaller.
Furthermore, there indicates
patterns, which are almosttheno characteristic
decomposition peaks ofaustenite
of retained retainedduring
austenite in of
220 ms thethecorresponding
EPA treatment XRD
werewhich
patterns, found.indicates
For the EPS
the+decomposition
EPA (280 ms) sample, the η-Ni
of retained 3Ti precipitates
austenite during 220are also
ms of detected
the EPA and the
treatment
were found. For the EPS + EPA (280 ms) sample, the η-Ni3 Ti precipitates are also detected the
average length was about 47 nm. This indicates that with the increase in the EPA treating period, and the
η-Nilength
average 3Ti precipitates present significant growth. Whereas, it is worth noting that some nano-globular
was about 47 nm. This indicates that with the increase in the EPA treating period, the
phase with a size less than 10 nm are observed in the EPS + EPA (280 ms) sample (Figures 7b–f). The
η-Ni3 Ti precipitates present significant growth. Whereas, it is worth noting that some nano-globular
results of SAED patterns presented in Figure 7c,d, and f demonstrate that the globular phase is
austenite. Also, it is seen from the results of XRD patterns (Figure 5), characteristic peaks of austenite
Metals 2019, 9, 1299 9 of 15

phase with a size less than 10 nm are observed in the EPS + EPA (280 ms) sample (Figure 7b–f).
The results of SAED patterns presented in Figure 7c,d, and f demonstrate that the globular phase is
Metals 2019,
austenite. Also,12,it
x FOR PEERfrom
is seen REVIEW
the results of XRD patterns (Figure 5), characteristic peaks of9austenite
of 15

appear again when the EPA treating time arrives 280 ms. Consequently, the nano austenite in the
appear again when the EPA treating time arrives 280 ms. Consequently, the nano austenite in the EPS
EPS + EPA (280 ms) sample can be considered as nano-reverted austenite.
+ EPA (280 ms) sample can be considered as nano-reverted austenite.

Figure
Figure 7. TEM
7. TEM images
images of of
EPSEPS
+ +EPA
EPA(220
(220ms)
ms) sample
sample (a)
(a)and
andEPS
EPS+ EPA (280
+ EPA ms)ms)
(280 (b–f); The The
(b–f); insertinsert
maps show the SAED patterns of the selected area in the corresponding figures; The (f) is the dark
maps show the SAED patterns of the selected area in the corresponding figures; The (f) is the dark field
field image of (e); REA—retained austenite, RTA—reverted austenite.
image of (e); REA—retained austenite, RTA—reverted austenite.
The results of Figure 5 have shown that the characteristic peaks of austenite disappeared in the
The results of Figure 5 have shown that the characteristic peaks of austenite disappeared in the
EPS + EPA (220 ms) sample. Whereas, after the EPA treatment for 280 ms, the characteristic peaks of
EPS + EPA (220 ms) sample. Whereas, after the EPA treatment for 280 ms, the characteristic peaks
austenite appeared again. This indicates that during EPA treatment for 220 ms, there exists the
of austenite appeared
decomposition again.
of the layeredThis indicates
retained thatinduring
austenite EPA As
EPS sample. treatment for treatment
for the EPA 220 ms, there
for 280exists
ms, the
decomposition of thethe
there also exists layered retained
formation austenite in EPS
of nano-reverted sample.
austenite. EvenAsthough
for thetheEPA treatment
crystal for 280
structure of ms,
thereretained
also exists the formation of nano-reverted austenite. Even though the crystal structure
austenite and reverted austenite are crystallographically identical, it is still necessary to of retained
discuss
austenite andthe formation
reverted mechanism
austenite of nano-reverted austenite
are crystallographically identical,because of necessary
it is still its totally to
different
discuss the
morphology
formation and size
mechanism offrom the retainedaustenite
nano-reverted austenite.because
At first, the
of local enrichment
its totally of Ni
different should be firstly
morphology and size
from discussed. Compared
the retained austenite.with
Atthe EPSthe
first, + EPA (220
local ms) sample,
enrichment ofthe
Ni η-Ni 3Ti precipitates in the EPS + EPA
should be firstly discussed. Compared
with (280 ms) sample show more significant growth. This needs higher local enrichment of Ni in the
the EPS + EPA (220 ms) sample, the η-Ni3 Ti precipitates in the EPS + EPA (280 ms) sample show
matrix. Then, because the Ni is austenitic stabilization element, there will be more reverted austenite
more significant growth. This needs higher local enrichment of Ni in the matrix. Then, because the Ni
in the EPS + EPA (280 ms) sample. In other words, the chemical conditions of α’→γ reverse
is austenitic stabilization element, there will be more reverted austenite in the EPS + EPA (280 ms)
Metals 2019, 9, 1299 10 of 15

sample. In other words, the chemical conditions of α’→γ reverse transformation in the EPS + EPA
(280 ms)
Metalssample
2019, 12, xare
FORmore sufficient than the EPS + EPA (220 ms) sample.
PEER REVIEW 10 of 15
The dislocation morphology, retained austenite and nano stacking faults in the EPS sample are
transformation in the EPS + EPA (280 ms) sample are more sufficient than the EPS + EPA (220 ms)
also responsible for the formation of nano-reverted austenite, and the detailed mechanisms are given
sample.
in Figure 8. As shown in Figure 8a,b, the dimension characteristic of a single dislocation line is
The dislocation morphology, retained austenite and nano stacking faults in the EPS sample are
anisotropic, and affected by this, the second phase will grow into a rod-like shape [41]. Whereas,
also responsible for the formation of nano-reverted austenite, and the detailed mechanisms are given
the dislocation
in Figure 8. pile-ups
As shownare more isotropic
in Figure than
8a and b, the the dislocation
dimension line, of
characteristic thea morphology
single dislocation of the
linesecond
is
phaseanisotropic,
will be globular underby
and affected this condition
this, the second [41]. Based
phase will on
growtheinto
results of Figure
a rod-like shape4b,f,
[41].the nano-reverted
Whereas, the
austenite will finally
dislocation pile-upsgrow
are into
moreaisotropic
globularthanshape. Moreover,line,
the dislocation considering the reduction
the morphology in thephase
of the second energy of
will be
austenitic globular
reverse under this condition
transformation, [41]. Based
the existence on the austenite
of retained results of Figure 4b,f, the
and stacking nano-reverted
faults can accelerate
austenite will
the formation finally grow into
of nano-reverted a globular
austenite. shape.toMoreover,
Owing the same considering the reduction
crystal structure of retainedin the energy and
austenite
of austenitic reverse transformation, the existence of retained austenite and
reverted austenite, the reverted austenite tends to nucleate on the retained austenite. In this regard, stacking faults can the
accelerate the formation of nano-reverted austenite. Owing to the same crystal structure of retained
nucleation energy is relatively lower than nucleate in other regions so that the nucleation on retained
austenite and reverted austenite, the reverted austenite tends to nucleate on the retained austenite.
austenite is easier to occur. This can also be verified by the results of Figure 7c, the distribution of
In this regard, the nucleation energy is relatively lower than nucleate in other regions so that the
nano-reverted
nucleation austenite
on retainedataustenite
layered isretained
easier toaustenite
occur. Thisis can
clearly
also detected.
be verified Moreby theimportant, the α’→γ
results of Figure
reverse transformation
7c, the distribution ofwill generallyaustenite
nano-reverted undergoattwice-lattice
layered retained shear [42]. isAs
austenite depicted
clearly in Figure
detected. More 8d,
because the nano
important, thestacking faultstransformation
α’→γ reverse have already possessed
will generally theundergo
a/6<111> plane, which
twice-lattice shearis [42].
the same
As as
the hexagonal
depicted inclosely
Figure arranged
8d, because surface
the nanoafter the first
stacking lattice
faults haveshear,
already thepossessed
reverse transformation
the a/6<111> plane, energy
which
can also be is the same
reduced as the
[42]. hexagonal the
Thereupon, closely arranged
existence of surface
stacking after the is
faults first lattice
also shear, thefactor
a favorable reversefor the
transformation
formation energy can
of nano-reverted also be reduced
austenite. The above[42]. Thereupon,
mentionedthe existenceis
discussion ofmainly
stackingrelated
faults isthe
also a of
effect
favorable factor for the formation of nano-reverted austenite. The above
the substructure on the distribution and morphology of nano-reverted austenite. As for the nano size mentioned discussion is
mainly related the effect of the substructure on the distribution and morphology of nano-reverted
and intensive distribution of reverted austenite, it is related with enhanced nucleation rate [32] and
austenite. As for the nano size and intensive distribution of reverted austenite, it is related with
extreme-treating time for growth induced by the EPA treatment.
enhanced nucleation rate [32] and extreme-treating time for growth induced by the EPA treatment.

Figure
Figure 8. Sketch
8. Sketch mapsmaps showing
showing theeffects
the effectsof
ofdislocation
dislocation morphology
morphology (a,b), retained
(a,b), austenite
retained (c) and
austenite (c) and
nano stacking faults (d) on the formation of nano-reverted austenite.
nano stacking faults (d) on the formation of nano-reverted austenite.
Metals 2019, 9, 1299 11 of 15

Metals 2019, 12, x FOR PEER REVIEW 11 of 15

For the EPS + TA sample, the existence of η-Ni3 Ti precipitates with a length of 54 nm is also
detected For(Figure
the EPS9a,b).
+ TA Because
sample, the
the existence of η-Ni3Ti precipitates
η-Ni3 Ti precipitates have enoughwithtime
a length of 54 nm
for growth is alsoTA
during
treatment, their size is also larger than that in the EPS + EPA sample. And also, compared with
detected (Figure 9a,b). Because the η-Ni 3Ti precipitates have enough time for growth during TAthe
EPS + EPA sample,
treatment, their size
theisovergrowth
also larger of
than 3 Ti in the EPS + TA sample will lead to more Ni consumption
Nithat in the EPS + EPA sample. And also, compared with the
so EPS
that+theEPA sample,
local the overgrowth
enrichment of Ni
of Ni does 3Ti in the EPS + TA sample will lead to more Ni consumption
not exist in the EPS + TA sample. Consequently, the reverted
so that the
austenite local be
cannot enrichment
detectedofinNi does
EPS + not
TA exist in the
sample EPS + TA
(Figures sample.
5 and 9a). Consequently,
What’s more,the reverted to
according
austenite cannot be detected in EPS + TA sample (Figures 5 and 9a). What’s
Figure 9c,d, some globular Fe2 Mo precipitates were also observed. The Fe2 Mo precipitatemore, according to Figure
is an
9c,d, some globular Fe2Mo precipitates were also observed. The Fe2Mo precipitate is an overaging
overaging product. During the early precipitation stage, the structure of η precipitate is a core-shell
product. During the early precipitation stage, the structure of η precipitate is a core-shell structure
structure [43], i.e., the growth of Ni3 Ti core of can be restricted by the Ni3 Mo shell. Whereas, with the
[43], i.e., the growth of Ni3Ti core of can be restricted by the Ni3Mo shell. Whereas, with the extension
extension of aging time, the shell (Ni3 Mo) will be gradually dissolved into the matrix so that the size of
of aging time, the shell (Ni3Mo) will be gradually dissolved into the matrix so that the size of the core
the core (Ni3 Ti) will become larger (the so called overaging state). Meanwhile, the dissolved Ni3 Mo
(Ni3Ti) will become larger (the so called overaging state). Meanwhile, the dissolved Ni3Mo shell will
shell will also be transformed into Fe2 Mo precipitates [43].
also be transformed into Fe2Mo precipitates [43].

Figure
9. 9.
TEMTEMimages
imagesofofprecipitates
precipitates(a),
(a), dark
dark field
field image
image of
of Ni
Ni33TiTi(b),
(b),dark
darkfield
fieldimage
imageofof
FeFe
2Mo (c)
Figure 2 Mo (c)
and HRTEM image of precipitates (d) in EPS + TA sample. The insert maps
and HRTEM image of precipitates (d) in EPS + TA sample. The insert maps show the SAED or show the SAED or FFT
FFT
patterns of the selected area in the corresponding
patterns of the selected area in the corresponding figures.figures.

TheThe traditional
traditional peak
peak aging
aging regime
regime forfor maraging
maraging steel
steel atatTSTS state
state is is holding
holding atat
480480
◦ C°C
forfor
2–32–3 h
h [28].
[28]. Whereas, compared with TS sample, the EPS sample possesses
Whereas, compared with TS sample, the EPS sample possesses higher energy (higher entanglement higher energy (higher
entanglement
degree degree and
of dislocation of dislocation and the
the existence of existence of nanofaults)
nano stacking stackingforfaults) for the following
the following aging aging
process.
process. Therefore, in contrast to traditional peak aging regime, the peak aging time should be less
Therefore, in contrast to traditional peak aging regime, the peak aging time should be less than 2–3
than 2–3 h. Thus, the EPS + TA sample is in the overaging state. Besides, the measured temperature
h. Thus, the EPS + TA sample is in the overaging state. Besides, the measured temperature of the
of the EPS + EPA (220 and 280 ms) sample are 295 °C and 393 °C, which are lower than the
EPS + EPA (220 and 280 ms) sample are 295 ◦ C and 393 ◦ C, which are lower than the temperature of the
temperature of the traditional peak aging regime. Based on classical nucleation theory, it seems that
traditional peak aging regime. Based on classical nucleation theory, it seems that the thermal conditions
the thermal conditions for the formation of η-Ni3Ti precipitates in the EPS + EPA sample are
forinadequate.
the formation of η-Niin3 Ti
Whereas, precipitates
fact, the nucleation EPSof+η-Ni
in thesites EPA3Tisample are inadequate.
precipitates must be some Whereas,
crystalindefects
fact, the
nucleation sites of η-Ni
(such as vacancy 3 Ti precipitates
or interstitial mustdue
atom). Thus, be some
to thecrystal
higher defects (such
resistivity as vacancy
of these or interstitial
crystal defects, the
atom). Thus, due to the higher resistivity of these crystal defects, the temperature
temperature around the nucleation sites will be higher than the measured values (295 and 393 around the nucleation
°C) in
sites
thewill
EPSbe higher
+ EPA than[44].
sample the In
measured values
a word, even (295 and
though 393 ◦ C) in the
the temperature EPS
of the + EPA
whole sample
samples are[44].
lowerIn a
word,
thaneven though the temperature
the temperature of traditionalof the aging
peak wholeregime,
samplesthe aretemperature
lower thanof thenucleation
temperatureareaof
oftraditional
η-Ni3Ti
Metals 2019, 9, 1299 12 of 15
Metals 2019, 12, x FOR PEER REVIEW 12 of 15

peak
will be aging
evenregime, the temperature
more higher of nucleation
than the samples area
so that the of η-Ni3 Ti will
precipitation be even
of η-Ni more higher than the
3Ti still possesses sufficient
samples so that the precipitation
thermal conditions. Moreover, due to the of η-Ni Ti still possesses sufficient thermal conditions.
3 extremely short treating period of the EPA treatment, Moreover,there
due
to
is the extremely time
no sufficient shortfortreating periodof
the growth of η-Ni
the EPA treatment, there is no sufficient time for the growth of
3Ti. Consequently, compared with the EPS + TA sample,
η-Ni
the size
3 Ti. Consequently, compared with the EPS
of η-Ni3Ti precipitate is smaller in the EPS + TA+ sample, the size of η-Ni3 Ti precipitate is smaller
EPA sample.
in theThe + EPA properties
EPStensile sample. of aging treated samples are displayed in Figure 10. The yield strength
and The tensile of
elongation properties
the EPS of aging
+ TA treated
sample are samples
1870 MPa areand
displayed
16.1%. Asin Figure
for the 10. The yield strength
electro-pulsing treated
and elongation of the EPS + TA sample are 1870 MPa and 16.1%. As for
sample, the corresponding values are 1987 MPa, 17.5%, and 2123 MPa, 20.1% for the EPS + EPA the electro-pulsing treated
(220
sample, the corresponding values are 1987 MPa, 17.5%, and 2123 MPa, 20.1%
ms) and EPS + EPA (280 ms) samples, respectively. These results indicate that under the effect offor the EPS + EPA (220
ms) and EPS + EPA
electro-pulsing, the (280 ms) and
strength samples, respectively.
ductility of aging These
treatedresults
sample indicate that under
are improved the effect of
simultaneously.
electro-pulsing, the strength and ductility of aging treated sample are improved
Take the precipitation strengthening mechanism into consideration, the contribution of precipitatessimultaneously. Take
the precipitation
to the strengthening
yield strength mechanism
(Δσp) is governed into consideration, the contribution of precipitates to the
by [45]:
yield strength (∆σp ) is governed by [45]:
0.4 MGb ln(2r / b)
Δσ p  0.4MGb ln(2r/b) , (3)
∆σp = π√ 1  υ • λP , (3)
π 1−υ λP
where, M is the Taylor factor, G is the shear modulus, b means the Burgers vector, ν is Poisson’s ratio,
and λpMrefers
where, is the Taylor
to the factor,
average G isspacing
the shear of modulus,
the rod-likeb means the Burgers
precipitates. vector,the
Because ν is size
Poisson’s ratio,
of η-Ni 3Ti
and λ p refers to the average spacing of the rod-like precipitates. Because the size
precipitates (r, considered as a sphere, increase with the increment in the length of rod-like precipitate)of η-Ni 3 Ti precipitates
(r,
in considered
the EPS + EPA as asample
sphere,isincrease
smaller with
than thethatincrement
in the EPSin the sample,
+ TA length oftherefore,
rod-like the precipitate) in the
contribution of
EPS + EPA sample is smaller than that in the EPS + TA sample, therefore,
precipitates to yield strength in the EPS + EPA sample is higher than the EPS + TA sample. the contribution of precipitates
to yield strength
Consequently, inEPS
the the +EPS
EPA + sample
EPA sample is higher
possesses than
higher yield EPS + TA
thestrength. Onsample.
the other Consequently,
hand, due tothe the
EPS + EPA sample possesses higher yield strength. On the other hand,
bigger size of η-Ni3Ti, the lattice mismatch between precipitates and matrix in the EPS due to the bigger size of η-Ni
+ TA 3 Ti,
sample
the lattice mismatch
is higher than that between
in the EPS precipitates
+ EPA sample,and matrix
which inwill lead+toTA
the EPS sample
crack is higher
initiation at than that in the
the interface of
EPS + EPA sample, which will lead to crack initiation at the interface of precipitate
precipitate and matrix [46]. Consequently, the ductility of the EPS + EPA sample is relatively higher. and matrix [46].
Consequently,
It is worth noting the ductility
that, the of the EPS
tensile + EPA sample
properties of the EPSis relatively
+ EPA (280 higher. It is worth
ms) sample are noting that,than
also better the
tensile
that ofproperties
the EPS +ofEPA (220+ ms)
the EPS EPA sample.
(280 ms)Thissample canare
bealso better than
attributed thatexistence
to the of the EPS of+nano-reverted
EPA (220 ms)
sample.
austenite in the EPS+EPA (280 ms) sample. Compared with the EPS + EPA (220 ms) sample,(280
This can be attributed to the existence of nano-reverted austenite in the EPS+EPA ms)
despite
sample. Compared with the EPS + EPA (220 ms) sample, despite the relatively
the relatively bigger size of η-Ni3Ti in the EPS + EPA (280 ms) sample, the nano-reverted austenite bigger size of η-Ni 3 in
Ti is
the
alsoEPSan +effective
EPA (280strengthening
ms) sample, the nano-reverted
phase. Hence, theaustenite
EPS + EPA is also an effective
(280 ms) sample strengthening
possesses phase.
higher
Hence,
strength. EPS + EPAthe
theMoreover, (280TRIP
ms) sample possesses
effect induced byhigher strength. Moreover,
nano-reverted austenite are the also
TRIPhelpful
effect induced
for the
by
modification of ductility. Therein, compared with the EPS + EPA (220 ms), the EPS + EPA (280with
nano-reverted austenite are also helpful for the modification of ductility. Therein, compared ms)
the EPS + EPA
possesses (220
better ms), the EPS + EPA (280 ms) possesses better ductility.
ductility.

Figure10.
Figure 10. Tensile
Tensile properties
properties of
ofsamples
samplesunder
underdifferent
differentkinds
kindsofofaging
agingtreatment:
treatment:(a)
(a)Engineering
Engineering
tensilecurves,
tensile curves,(b)
(b)Normalized
Normalizedwork-hardening
work-hardeningrate
ratecurves.
curves.

Figure 9b shows the work-hardening rate curves of aging treated samples. As is seen, due to the
relatively bigger size of austenitic grains, the work-hardening rate of EPS + TA sample is the lowest.
Besides, the EPS + EPA (280 ms) sample possesses the highest initial work-hardening rate, which is
Metals 2019, 9, 1299 13 of 15

Figure 9b shows the work-hardening rate curves of aging treated samples. As is seen, due to
the relatively bigger size of austenitic grains, the work-hardening rate of EPS + TA sample is the
lowest. Besides, the EPS + EPA (280 ms) sample possesses the highest initial work-hardening rate,
which is induced by the co-strengthening effect of η-Ni3 Ti and nano-reverted austenite. Furthermore,
unlike the continuously decreased work-hardening rate of EPS + TA and EPS + EPA (220 ms) samples,
and peak is observed in work-hardening rate curve of EPS + EPA (280 ms) sample. This interesting
phenomenon is related with the TRIP effect induced by nano-reverted austenite in EPS + EPA (280 ms)
sample [36]. During this process, the gradually concentrated local stress around the nano-reverted
austenite will lead to the increase in work-hardening rate [40]. Whereas, after the γ→α’ transformation,
the concentrated local stress will be released so that the work-hardening rate will decrease again [40].
Based on the above mentioned results, the optimized electro-pulsing treatment can simultaneously
enhance the strength and ductility of T250 steel very efficiently. However, there still exist some
deficiencies of this technology. For example, the formation mechanism of retained austenite in EPS is
related with the heterogeneously distributed Ni element. Whereas, if this technology is applied in the
microstructure refinement of materials at semi-solid state or as cast state, there may exist elemental
segregation, which is harmful to the mechanical properties. Besides, the grain refinement effect is
sensitive to the electro-pulsing parameters. If the parameters are not appropriate, there will be no
refinement effect or the grains will be coarser. So, the authors will continue to study the detailed
effect of electro-pulsing current on the structure-property relationship of metallic materials so that the
application of electro-pulsing technology will be more universal and more advanced.

4. Conclusions
In summary, high-energy electro-pulsing current is applied in cobalt-free maraging steel to obtain
more advanced mechanical properties. Compared with TS treatment, after EPS treatment within only
360 ms, substantial retained austenite and nano-stacking faults are formed. Moreover, finer prior
austenitic grains and martensitic lath are also acquired. Based on this, the following EPA treatment
can promote the formation of finer η-Ni3 Ti precipitates and a large amount of nano-reverted austenite
within only 280 ms. Under the same initial state, the tensile strength and elongation are both modified
via electro-pulsing treatment. Eventually, the best tensile properties are optimized to be 2123 MPa
(Yield strength) and 20.1% (Total elongation, TE%) by EPS + EPA (280 ms) treatment. To summarize, the
electro-pulsing treatment can be a novel and effective method to developing new cobalt-free maraging
steel with superior properties.

Author Contributions: Conceptualization, D.P.; methodology, Y.Z. and X.X.; formal analysis, Y.W., and Y.Z.;
investigation, D.P. and Y.W.; data curation, D.P. and X.C.; writing—original draft preparation, D.P.; writing—review
and editing, Y.W. and X.C.; visualization, Y.W. and X.X.; supervision, Y.Z.; project administration, X.X. and Y.Z.
Funding: This research was supported by the National Natural Science Foundation of China (Grant No. 51701080)
and Talent Development Program-Outstanding Youth Fund of Jilin Province (Grant No. 20190103053JH).
Acknowledgments: TEM observation by Yitong Wang is sincerely acknowledged.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Guo, Z.; Sha, W.; Vaumousse, D. Microstructural evolution in a PH 13–8 stainless steel after aging. Acta Mater.
2003, 51, 101–116. [CrossRef]
2. Luo, H.; Shi, J.; Wang, C.; Gao, W.; Sun, X.; Dong, H. Experimental and numerical analysis on formation of
stable austenite during the intercritical annealing of 5Mn steel. Acta Mater. 2011, 59, 4002–4014. [CrossRef]
3. Liu, H.W.; Liu, J.W.; Luo, C.P.; Liu, Z.J. Microstructure, crystallography of phase transformation and multiple
precipitations in PH 15-7 Mo stainless steel. J. Alloys Compd. 2016, 672, 386–392. [CrossRef]
4. Xu, W.; Rivera-Díaz-del-Castillo, P.E.J.; Yan, W.; Yang, K.; San Martín, D.; Kestens, L.A.I.; van der Zwaag, S.
A new ultra-high strength stainless steel strengthened by various coexisting nanoprecipitates. Acta Mater.
2010, 58, 4067–4075. [CrossRef]
Metals 2019, 9, 1299 14 of 15

5. Rajkumar, K.V.; Rao, B.P.C.; Sasi, B.; Kumar, A.; Jayakumar, T.; Raj, B.; Ray, K.K. Characterization of aging
behaviour in M250 grade maraging steel using eddy current non-destructive methodology. Mater. Sci. Eng. A
2007, 464, 233–240. [CrossRef]
6. Tharian, K.T.; Sivakumar, D.; Ganesa, R.; Balakrishnan, P.; Sinha, P.P. Development of new low nickel, cobalt
free maraging steel. Mater. Sci. Technol. 1991, 7, 1082–1088. [CrossRef]
7. Li, K.; Wei, L.; An, B.; Yu, B.; Misra, R.D.K. Aging phenomenon in low lattice-misfit cobalt-free maraging steel:
Microstructural evolution and strengthening behavior. Mater. Sci. Eng. A 2019, 739, 445–454. [CrossRef]
8. Squire, R.; Wilson, E.A. Effect of cobalt on impact toughness of steels. Mater. Sci. Technol. 1994, 10, 52–55.
[CrossRef]
9. Li, Y.C.; Yan, W.; Cotton, J.D.; Ryan, G.J.; Shen, Y.F.; Wang, W.; Shan, Y.Y.; Yang, K. A new 1.9 GPa maraging
stainless steel strengthened by multiple precipitating species. Mater. Des. 2015, 82, 56–63. [CrossRef]
10. Shekhter, A.; Aaronson, H.; Miller, M.; Ringer, S.; Pereloma, E. Effect of aging and deformation on the
microstructure and properties of Fe-Ni-Ti maraging steel. Metall. Mater. Trans. A 2004, 35, 973–983.
11. Wang, C.C.; Zhang, C.; Yang, Z.G. Austenite layer and precipitation in high Co-Ni maraging steel. Micron
2014, 67, 112–116. [CrossRef] [PubMed]
12. Masoumi, M.; de Barrous, I.F.; Herculano, L.F.G.; Coelho, H.L.F.; de Abreu, H.F.G. Effect of microstructure
and crystallographic texture on charpy impact test for maraging 300 steel. Mater. Charact. 2016, 120, 203–209.
[CrossRef]
13. Du, P.J.; Yang, D.P.; Bai, M.K.; Xiong, X.C.; Wu, D.; Wang, G.D.; Yi, H.L. Austenite stabilisation by two step
partitioning of manganese and carbon in a Mn-TRIP steel. Mater. Sci. Technol. 2019, 17, 2084–2091. [CrossRef]
14. Harding, I.; Mouton, I.; Gault, B.; Raabe, D.; Kumar, K.S. Carbon partitioning and microstructure evolotion
during tempering of an Fe-Ni-C steel. Scripta Mater. 2019, 172, 38–42. [CrossRef]
15. Pereloma, E.; Shekhter, A.; Miller, M.; Ringer, S. Aging behaviour of an Fe-20Ni-1.8Mn-1.6Ti-0.59Al (wt.%)
maraging alloy: Clustering, precipitation and hardening. Acta Mater. 2004, 52, 5589–5602. [CrossRef]
16. Li, K.; Yu, B.; Misra, R.D.K.; Han, G.; Liu, S.; Shang, C.J. Strengthening of cobalt-free 19Ni3Mo1.5Ti maraging
steel through high-density and low lattice-misfit nanoscale precipitates. Mater. Sci. Eng. A 2018, 715, 174–185.
[CrossRef]
17. Jiang, S.H.; Wang, H.; Wu, Y.; Liu, X.J.; Chen, H.H.; Yao, M.J.; Gault, B.; Ponge, D.; Raabe, D.; Hirata, A.;
et al. Ultrastrong steel via minimal misfit and high-density nanoprecipitation. Nature 2017, 544, 460–465.
[CrossRef]
18. Yang, M.X.; Yang, G.; Liu, Z.D.; Wang, C.; Huang, C.X. Significant enhancement of strength in a lamellar-type
nanostructured maraging steel subjected to equal-channel angular pressing for 12 passes. Mater. Sci. Eng. A
2012, 550, 429–433. [CrossRef]
19. Yang, M.X.; Yang, G.; Liu, Z.D.; Wang, C.; Hu, C.; Huang, C.X. Effect of equal-channel angular pressing on the
microstructures and properties of 18Ni(C250) maraging steel. Acta Metall. Sin. 2012, 48, 164–169. [CrossRef]
20. Zhang, W.; Zou, D.N.; Qiao, G.J. Hot deformation behavior and microstructural evolution of as-cast co
containing maraging stainless steel. Mater. Res. Express 2019, 6, 086586. [CrossRef]
21. Mizubayashi, H.; Okuda, S. Structural relaxation induced by passing electric current in amorphous Cu50 Ti50
at low temperature. Phys. Rev. B 1989, 40, 8057–8060. [CrossRef] [PubMed]
22. Song, H.; Wang, Z.J.; He, X.D.; Duan, J. Self-healing of damage inside metal triggered by electropulsing
stimuli. Sci. Rep. 2017, 7, 7097. [CrossRef] [PubMed]
23. Zhao, Y.G.; Zhang, J.T.; Tan, J.; Ma, B.D. Microstructure refinement and property improvement of metastable
austenitic manganese steel induced by electro-pulsing. J. Iron Steel Res. Int. 2014, 21, 685–689. [CrossRef]
24. Zhou, Y.; Chen, G.Q.; Fu, X.S.; Zhou, W.L. Effect of electropulsing on deformation behavior of Ti-6Al-4V
alloy during cold drawing. Trans. Nonferrous Met. Soc. China 2014, 24, 1012–1021. [CrossRef]
25. Zhang, X.; Li, H.W.; Zhan, M. Mechanism for macro and micro behaviors of the Ni-based superalloy during
electrically-assisted tension: Local Joule of heating effect. J. Alloys Compd. 2018, 742, 480–489. [CrossRef]
26. Zhao, Y.G.; Ma, B.D.; Guo, H.C.; Ma, J.; Yang, Q.; Song, J.S. Electropulsing strengthened 2GPa boron steel
with good ductility. Mater. Des. 2013, 43, 195–199. [CrossRef]
27. Ma, B.D.; Zhao, Y.G.; Bai, H.; Ma, J.; Zhang, J.T.; Xu, X.F. Gradient distribution of mechanical properties in the
high carbon steel induced by the detour effect of the pulse current. Mater. Des. 2013, 49, 168–172. [CrossRef]
Metals 2019, 9, 1299 15 of 15

28. Pan, D.; Zhao, Y.G.; Xu, X.F.; Wang, Y.T.; Jiang, W.Q.; Chong, X.Y. A novel strengthening and toughening
strategy for T250 maraging steel: Cluster-orientation governed higher strength-ductility combination induced
by electropulsing. Mater. Des. 2019, 169, 107686. [CrossRef]
29. Pan, D.; Zhao, Y.G.; Wang, Y.T.; Xu, X.F.; Yin, P.L.; Chong, X.Y. Formation of ultrafine pearlites in tempered
42CrMo steel induced by electro-pulsing. Mater. Lett. 2019, 251, 73–77. [CrossRef]
30. Gui, B.C.; Yang, Z.Y.; Jiang, Y.H.; Ding, Y.L. Effect of solution temperature on mechanical properties of
corrosion-resistant maraging steel. J. Iron Steel Res. 2012, 24, 45–48.
31. Loewy, S.; Rheingans, B.; Meka, S.R.; Mittemeijer, E.J. Modulated martensite formation behavior in
Fe-Ni-based alloys; athermal and thermally activated mechanisms. J. Mater. Res. 2015, 30, 2101–2107.
[CrossRef]
32. Pan, D.; Wang, Y.T.; Zhao, Y.G.; Xu, X.F.; Chong, X.Y.; Yin, P.L. Completely wetted α/α grain boundaries
in electro-pulsing treated DP600 steel under different cooling conditions. Mater. Lett. 2019, 252, 264–267.
[CrossRef]
33. Zhang, M.X.; Kelly, P.M. Accurate orientation relationship between ferrite and austenite in low carbon
martensite and granular bainite. Scripta Mater. 2002, 47, 749–755. [CrossRef]
34. Wang, Y.T.; Pan, D.; Zhao, Y.G.; Sun, L.Y.; Xu, X.F.; Yin, P.L.; Chong, X.Y.; Jiang, W.Q.; Wang, Z. Enhanced
stability of retained austenites in quenched 25SiMn2MoV steel by electro-pulsing current. Mater. Sci. Eng. A
2019, in press. [CrossRef]
35. Strife, J.R.; Carr, M.J.; Ansell, G.S. The effect of austenite prestrain above the Md temperature on the
martensitic transformation in Fe-Ni-Cr-C alloys. Metall. Trans. A 1977, 8, 1471–1484. [CrossRef]
36. Yang, H.S.; Bhadeshia, H.K.D.H. Austenite grain size and the martensite-start temperature. Scr. Mater. 2009,
60, 493–495. [CrossRef]
37. Zhang, W.; Sui, M.L.; Zhou, Y.Z.; He, G.H.; Guo, J.D.; Li, D.X. Electro-pulsing induced evolution of
microstructures in materials. Acta Metall. Sin. 2003, 39, 1009–1018.
38. Tang, D.W.; Zhou, B.L.; Gao, H.; He, G.H. Thermal stress relaxation behavior in thin films under transient
laser-pulse heating. J. Appl. Phys. 1993, 73, 3749–3752. [CrossRef]
39. Huang, C.X.; Wang, K.; Wu, S.D.; Zhang, Z.F.; Li, G.Y.; Li, S.X. Deformation twinning in polycrystalline
copper at room temperature and low strain rate. Acta Mater. 2006, 54, 655–665. [CrossRef]
40. Pan, D.; Zhao, Y.G.; Xu, X.F.; Wang, Y.T.; Jiang, W.Q.; Ju, H. Effect of high-energy and instantaneous
electropulsing treatment on microstructure and properties of 42CrMo steel. Acta Metall. Sin. 2018,
54, 1245–1252.
41. Wang, Y.T.; Zhao, Y.G.; Xu, X.F.; Pan, D.; Jiang, W.Q.; Yang, X.H.; Wang, Z. Superior mechanical properties
induced by the interaction between dislocations and precipitates in the electro-pulsing treated Al-Mg-Si
alloys. Mater. Sci. Eng. A 2018, 735, 154–161. [CrossRef]
42. Zhu, J.; Zhao, Y.W.; Pan, T.X.; Cai, Q.G. Investigation on reverse austenite in 18Ni(250 grade) maraging steel.
Iron Steel 1981, 16, 41–45.
43. Tian, J.L.; Li, Y.C.; Wang, W.; Yan, W.; Shan, Y.T.; Jiang, Z.H.; Yang, K. Alloying element segregation effect in a
multi-phase strengthened maraging stainless steel. Acta Metall. Sin. 2016, 52, 1517–1526.
44. Zhang, X.; Li, H.W.; Zhan, M.; Shao, G.D.; Ma, P.Y. Extraordinary effect of the δ phase on the electrically-assisted
deformation responses of a Ni-based superalloy. Mater. Charact. 2018, 144, 597–604. [CrossRef]
45. Srivatsan, T.S. Dispersion strengthened aluminum alloys. Mater. Manuf. Process. 1991, 6, 565–568. [CrossRef]
46. Ashby, M.F. Work hardening of dispersion-hardened crystals. Philos. Mag. 1966, 14, 1157–1178. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).