Materials Characterization 195 (2023) 112486

Contents lists available at ScienceDirect

Materials Characterization
journal homepage: www.elsevier.com/locate/matchar

Effect of low-temperature precipitates on microstructure and

pseudoelasticity of an Fe–Mn–Si-based shape memory alloy
Hesamodin Khodaverdi a, Maryam Mohri b, *, Amir Sabet Ghorabaei a, Elyas Ghafoori b, c,
Mahmoud Nili-Ahmadabadi a, *
a
School of Metallurgy and Materials Engineering, College of Engineering, University of Tehran, Tehran, Iran
b
Empa, Swiss Federal Laboratories for Materials Science and Technology, 8600 Dübendorf, Switzerland
c
Institute for Steel Construction, Faculty of Civil Engineering and Geodetic Science, Leibniz University Hannover, 30167 Hannover, Germany

A R T I C L E I N F O A B S T R A C T

Keywords: Fe–Mn–Si-based shape memory alloys (Fe-SMAs) have attracted much research attention due to their potential
Fe–Mn–Si-based shape memory alloy applications for vibration mitigation, energy dissipation, and re-centering in the construction sector. Because of
Superelasticity the crucial impact of precipitation on the pseudoelasticity (PE) behavior of Fe-SMAs, the equilibrium phase
Precipitation
diagram of an Fe–17Mn–5Si–10Cr–4Ni–1(V-C) (wt%) SMA was used in this study to identify a low-temperature
Equilibrium phase diagram
precipitate and study its effect on the microstructure and PE of the alloy after a low-temperature aging process.
Transmission electron microscopy
Transmission electron microscopy (TEM) studies revealed that aging at 485 ◦ C for 6 h after aging at 750 ◦ C for 6
h led to the precipitation of fresh, parallelogram-shaped, (Cr–V–C)-rich precipitates along with elliptical-shaped,
V-rich precipitates in the austenite grains of the recrystallized samples. Numerous parallel stacking faults (SFs)
were formed due to the presence of the precipitates within the austenite grains. It is postulated that such an
arrangement of SFs can further improve the PE by reducing the activation energy for the nucleation of
ε-martensite laths and inhibiting them from colliding with each other and consequent formation of α'-martensite,
resulting in a residual strain reduction to 2.7% after 4.0% tensile straining.

1. Introduction A combination of the reversible motion of Shockley partial disloca­

tions (SPDs) and back transformation from ε-martensite (with hexagonal
Fe–Mn–Si-based shape memory alloys (SMAs), as one of the most close-packed (HCP) lattice structure) to γ-austenite (with face-centered
practical categories of Fe-based SMAs, have attracted much research cubic (FCC) lattice structure) in unloading is believed to be the micro­
attention [1–4]. Due to their low material cost, high Young's modulus structural reason for the PE effect in Fe–Mn–Si-based SMAs [19]. Several
and strength, exceptional workability, and high stress and strain re­ attempts have been made to enhance the PE of Fe–Mn–Si-based SMAs,
covery [5,6], these alloys have been used for a wide range of applica­ such as controlling the grain size [20,21], texturizing, and precipitation
tions such as coupling devices [7], mechanical tightening, structural [22] in combination with stacking fault energy (SFE) control [23].
elements, active controls, pre-stressing or post-tensioning of structures Among the different approaches, SFE control and precipitation have had
[8], and damping devices [9]. However, low pseudoelasticity (PE) has a greater impact on PE improvement. Precipitates such as Cr23C6 [24],
prevented them from being used in many applications such as dissipa­ TiC [25], VN [26], VC [27], and NbC [28] are used for PE improvement.
tive re-centering systems in civil engineering [10]. It has been revealed that NbC carbides increase the strength of the
A new design for Fe-based SMAs, Fe–17Mn–5Si–10Cr–4Ni–1(V-C) γ-austenite phase and assist in the reversible movement of the γ/ε in­
(wt%), has been developed in Swiss Federal Laboratories for Materials terfaces through a specific crystallographic path. The back-stress on the
Science and Technology (Empa) [11–15]. The alloy can be produced by growing ε-martensite lath due to the stress fields induced by the NbC
a relatively cheap casting process under atmospheric conditions [16] carbides causes this back transformation [28,29]. In addition, VC pre­
and has high potential for large-scale repair and strengthening in the cipitates promote the formation of ε-martensite by reducing the critical
construction section [17] and passive vibration damping [18]. resolved shear stress (CRSS) of austenite [19]. Precipitates also increase

* Corresponding authors.
E-mail addresses: [email protected] (M. Mohri), [email protected] (M. Nili-Ahmadabadi).

https://doi.org/10.1016/j.matchar.2022.112486
Received 26 August 2022; Received in revised form 10 November 2022; Accepted 12 November 2022
Available online 17 November 2022
1044-5803/© 2022 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
H. Khodaverdi et al. Materials Characterization 195 (2023) 112486

the number of dislocations and stacking faults (SFs) in the γ-austenite was considered to analyze Kα radiation of the alloying elements. The
matrix, which act as embryos for ε-martensite nucleation [30–34]. It has EDS spectra were quantified by the TIA software based on the Cliff-
recently been found that having a fine-grained austenite matrix along Lorimer method [37] with k-factors calculated with respect to silicon.
with uniform distribution of precipitates facilitates the γ ⇌ ε trans­ A maximum relative uncertainty value of ~5% was detected for the
formation and enhances the PE behavior of the alloy [35]. reported chemical compositions. Samples for the OM and SEM obser­
In a recent study on the newly designed Fe–Mn–Si-based SMA [36], it vations were mechanically polished and etched with a solution of H2O2
has been shown that (Cr–V–C)-rich precipitates uniformly form inside (35%), HNO3 (65%), and HCl (32%) (7:30:9). The HR-TEM specimens
the austenite grains by high-temperature aging at 750 ◦ C, resulting in a were prepared by standard mechanical thinning followed by electro­
significant improvement of PE. To recognize any new low-temperature polishing in a solution of HClO4 and C2H5OH (1:9) at − 20 ◦ C and 22 V.
precipitates, it is essential to investigate the equilibrium phase dia­ To evaluate the mechanical properties of the heat-treated samples,
gram of the alloy. In the present work, the effect of low-temperature monotonic loading–unloading experiments were performed with a
aging at 485 ◦ C on the evolution of a new type of precipitate is stud­ Zwick/Roell Z020 tensile test machine. During the loading–unloading
ied, and its combination with the primary high-temperature precipitates tests, the strain evolution was measured with a calibrated sensor-arm
is employed to further enhance the PE of the alloy. Low-temperature extensometer type BTC-EXMULTI.011 made by Zwick/Roell Company.
precipitates, which may improve shape memory and superelasticity Longitudinal tensile specimens were machined from the samples and
properties, are highly desirable because lowering the aging temperature prepared in a dog-bone shape with a reduced section length of 32.0 mm
decreases manufacturing costs and associated problems like oxidation and a cross-section of 1.0 × 0.8 mm2 (Fig. 1). Furthermore, to study the
and grain growth. It is shown that after 4.0% loading in tension, a re­ microstructural evolution of the samples after loading–unloading tests,
sidual strain of 2.7% can be achieved. The underlying mechanisms of the sub-size tensile specimens with a gauge length of 25.00 mm and a width
observed improvement in PE after low-temperature aging at 485 ◦ C are of 6.25 mm were prepared according to the ASTM E8M-04 standard.
studied by detailed microstructural characterizations, which provide Tensile loading to a maximum strain of 4% was carried out at room
new insights into the processing of this alloy for vibration mitigation and temperature with a crosshead speed of 0.5 mm/min followed by
seismic damping applications. unloading with the same crosshead speed until a force of 10 N was
attained.
2. Materials and methods
3. Results
Rebars with a diameter of 18 mm and a nominal composition of
Fe–17Mn–5Si–10Cr–4Ni–1(V-C) (wt%), provided by re-fer AG, 3.1. Equilibrium phase diagram of Fe–17Mn–5Si–10Cr–4Ni–1(V-C) (wt
Switzerland, were used in this study. The material was produced by hot %) alloy
rolling at 1000 ◦ C after casting. The microstructure of the as-received
rebars consisted of equiaxed austenite grains and fine precipitates at Fig. 2 shows the equilibrium phase diagram of the studied alloy
the grain boundaries [35]. The as-received alloy was cold–caliber-rolled calculated by using Thermo-Calc software with TCFE7 database [38].
to an octagonal-shaped speciemen with equal sides of 14.2 mm. The cold The liquid phase is thermodynamically stable at high temperatures
rolling was performed in 16 consecutive steps to reach an equivalent down to approximately 1320 ◦ C, where it starts to solidify to form the
strain of 0.25 at ambient temperature, using a caliber rolling machine γ-austenite phase in a temperature range of about 95 ◦ C until 1225 ◦ C.
with a roller diameter of 110 mm and a rolling speed of 800 mm/min. Further cooling to 1000 ◦ C promotes the formation of M7C3 carbide via a
The cold-rolled specimens were recrystallized in a vacuum furnace at precipitation reaction of γ → γ + M7C3. Subsequently, the Sigma1 phase
925 ◦ C for 50 min followed by air cooling to room temperature [35]. is formed by a precipitation reaction of γ → γ + Sigma1 at 665 ◦ C, and its
High-temperature aging was performed on the recrystallized specimens molar fraction reaches a maximum of ~0.15 at 485 ◦ C (inset of Fig. 2).
at 750 ◦ C for 6 h in a muffle furnace followed by air cooling to ambient Additionally, the thermodynamic calculations predict that the Sigma1
tempearture. Subsequently, a set of the aged specimens was double aged phase transforms into the Sigma2 and A2 phases below 485 ◦ C. In the
at a relatively lower temperature of 485 ◦ C for 6 h and then cooled down present study, the possible precipitation of the Sigma1 phase and its
to the room temperature under atmospheric conditions. The recrystal­ relation to PE evolution were investigated through double aging of the
lized, single aged, and double aged specimens are hereafter named as Rex-Aged samples at 485 ◦ C.
Rex, Rex-Aged, and Rex-Double aged, respectively.
Dilatometry analysis with a cooling/heating rate of 1 ◦ C/s was 3.2. Microstructure development after high- and low-temperature aging
conducted on cylindrical samples with a length of 10 mm and a diameter
of up to 2 mm in an Adamel DT1000 dilatometer. A schematic view of Figs. 3a–c show the FE-SEM micrographs of the Rex (a), Rex-Aged
the heat treatment cycles and the optical microstructures of the as- (b), and Rex-Double aged (c) samples. The microstructure of the Rex
received material before and after the cold rolling are provided in the sample mainly consisted of recrystallized austenite grains with an
Supplementary Material File (see Fig. S1). average grain size of 5 μm and reversed austenite regions with a diam­
The volume fractions of phase constituents were measured using X- eter of approximately 1 μm. Since grain boundary junctions act as
ray diffraction (XRD) analysis with a Cu Kα radiation source (λ = 0.154 preferential nucleation sites for the martensitic transformation, most
nm) at an acceleration voltage of 45 kV and a tube current of 200 mA. strain-induced martensite forms in these locations [39]. Therefore, the
The microstructures of the specimens were studied using optical mi­ reversed austenite grains usually appeared in the triple junctions of
croscopy (OM; ZEISS Axioskop 2 MAT), secondary electron analysis by recrystallized austenite grains (Fig. 3a). Such formation of reversed
field-emission scanning electron microscopy (FE-SEM; FEI Nova Nano­
SEM 450), and high-resolution transmission electron microscopy (HR-
TEM; FEI Tecnai F20 series). The chemical compositions of precipitates
were analyzed using energy-dispersive X-ray spectroscopy (EDS; Oxford
X-MaxN 80 T silicon drift detector with a take-off angle of 14.6◦ and a
resolution of 134 eV FWHM at reference energy of 5.9 keV) by the TEM.
Standard samples of pure elements were used for calibrating the TEM
imaging and analysis (TIA) interface of the microscope software.
Moreover, the multi-polynomial background correction for the X-ray Fig. 1. Geometry of the dog-bone specimens used for the tensile tests (di­
signals along with thin foil thickness correction for the TEM specimens mensions in mm).

2
H. Khodaverdi et al. Materials Characterization 195 (2023) 112486

(Fig. 3c). In addition, the number of (Cr–V–C)-rich precipitates (poly­

hedral-shaped) was reduced, but some new needle-shaped precipitates
formed inside the austenite grains. Similar to the Rex and Rex-Aged
samples, the grain size of recrystallized austenite for the Rex-Double
aged sample did not change, but the grain size of reversed austenite
slightly decreased (Fig. 3c).
Fig. 3d illustrates the XRD diffractograms of the specimens from
which the volume fractions of α'-martensite and ε-martensite were
calculated according to the method provided in Ref. [41]. The Rex
sample contained 14 vol% ε-martensite and 7.9 vol% α'-martensite,
while the Rex-Aged sample possessed a much higher volume fraction of
ε-martensite of about 43% and, interestingly, had almost zero volume
fraction of α'-martensite. The disappearance of α'-martensite and the
increase in the volume fraction of ε-martensite in the Rex-Aged spec­
imen could be related to the formation of precipitates, which can change
the chemical composition of the austenite matrix and inhibit the for­
mation of randomly oriented ε-martensite laths inside the austenite
grains [42]. Furthermore, the XRD diffractogram of the Rex-Aged sam­
ple showed peak splitting (indicated by the small arrow in Fig. 3d),
Fig. 2. Calculated equilibrium phase diagram of the studied which can be related to the presence of precipitates, SFs, and
Fe–17Mn–5Si–10Cr–4Ni–1(V-C) (wt%) alloy at 1 atm pressure by the Thermo- ε-martensite that reduce the symmetry of the microstructure [43]. In
Calc software. comparison to the Rex-Aged sample, the volume fraction of ε-martensite
reduced to nearly 16% and the degree of peak splitting decreased for the
austenite at the triple junctions of recrystallized austenite grains during Rex-Double aged sample.
the annealing of cold-rolled austenite has also been reported elsewhere
[39,40]. Moreover, there were some constituents with sizes of less than
3.3. Loading–unloading experiments
800 nm located inside the austenite grains and also at the grain
boundaries of austenite, which were (Cr–V–C)-rich precipitates ac­
The engineering stress-strain curves after loading–unloading tests for
cording to our previous findings [35]. In comparison to the Rex sample,
the As-Received aged [35] and the heat-treated samples are shown in
the Rex-Aged sample showed a sharp increase in the number of pre­
Fig. 4. The values of PE and other tensile properties of the samples
cipitates with a higher volume fraction inside the austenite grains than
inferred from Fig. 4 are summarized in Table 1. For this particular alloy,
at the grain boundaries. The average austenite grain size of the Rex
the 0.1% offset yield stress (σY0.1%) was considered for estimating the
sample remained close to 5 μm after aging (Fig. 3b). After low-
onset stress of the martensitic transformation [44]. The elastic modulus
temperature aging of the Rex-Aged sample at 485 ◦ C for 6 h, the vol­
of this alloy can vary between 40 and 170 GPa under different micro­
ume fraction and continuity of reversed austenite regions decreased
structural conditions [45]. Since the elastic moduli of the samples were

Fig. 3. SEM micrograph of the (a) Rex, (b) Rex-Aged, and (c) Rex-Double aged samples. (d) XRD diffractograms of the specimens.

3
H. Khodaverdi et al. Materials Characterization 195 (2023) 112486

3.4. TEM study on the microstructures after aging at high and low
temperatures

Figs. 5a–b show TEM images from the substructure of the Rex sam­
ple. Particles with sizes between 15 and 20 nm were uniformly distrib­
uted in the austenite matrix (white arrows in Fig. 5a). It has recently
been indicated that these particles are rich in vanadium and can be
considered as embryos for the V-rich precipitates [35]. Additionally, a
number of parallel SFs (blue arrows in Figs. 5a–b) could be observed in
the microstructure of the Rex sample. The TEM images of the Rex-Aged
sample are shown in Figs. 5c–d, indicating the formation of a higher
number of SFs and ε-martensite laths after aging in comparison to the
Rex sample. The TEM-EDS analysis showed that the polyhedral-shaped
precipitates were rich in Cr, V and C elements (Fig. 5c). These parti­
cles with approximately similar chemical compositions but with spher­
ical morphologies and relatively smaller sizes were also present in the
substructure of the Rex sample (white arrows in Fig. 5b). The details of
TEM-EDS analyses are provided in the Supplementary Material File
(Figs. S2–S9). A dark-field TEM image obtained from the [0110]
Fig. 4. Loading–unloading curves of the As-Received aged [35], Rex, Rex- reflection of ε-martensite is shown in the inset of Fig. 5d. It was observed
Aged, and Rex-Double aged samples strained to 4% in tension. that ε-martensite laths generally grew parallel to each other in separate
groups in the samples.
calculated by using Hooke's law, the calculated values corresponded to Figs. 6a–d illustrate TEM images from the substructure of the Rex-
mixed microstructures of austenite, ε-martensite, and α’-martensite with Double aged sample. Fig. 6a shows an austenite grain with a large
different volume fractions of the constituents. Furthermore, it has been number of SFs that are all located in the same orientation inside the
shown that mainly the stress-induced martensitic transformation occurs grain. Such a parallel orientation of SFs, but with a lower degree of
until the stress level reaches the yielding point of this alloy [44]. orientation, was also detected in the TEM micrographs of the Rex and
Therefore, the mixture of phases is changed by increasing (decreasing) Rex-Aged samples (Fig. 5). In contrast to the other samples,
the strain in the elastic part, which justifies the nonlinear deformation parallelogram-shaped precipitates were observed in addition to
behavior during loading (unloading). The PE strain was calculated ac­ polyhedral-shaped precipitates in the microstructure of the Rex-Double
cording to Eq. (1): aged sample (Figs. 6b–c). As shown in the inset of Fig. 6c, the
parallelogram-shaped precipitates indicated a range of 170–330 nm in
σ4% length. It seemed that these precipitates were the same as the needle-
εtotal = εres + εpse + εE = εres + εpse + (1)
Eunloading shaped precipitates observed in the FE-SEM image (Fig. 3c), which a
portion of their cross-section could be seen in the TEM images
where εres, εpse, and εE are the residual strain, PE strain, and elastic strain,
(Figs. 6b–c). Furthermore, ε-martensite laths with a limited number of
respectively. σ 4% denotes the strength at the peak tensile strain, and
variants were observed in the Rex-Double aged sample (Fig. 6b). An HR-
Eunloading is the elastic modulus in unloading.
TEM image from a section of one of the ε-martensite laths with a width
The Rex sample exhibited an elastic modulus of 160 GPa in loading
of nearly 2.2 nm is shown in Fig. 6d. By applying the fast Fourier
which decreased to 105 and 80 GPa for the Rex-Aged and Rex-Double
transform (FFT) and then the inverse FFT (IFFT) to a portion of the
aged samples, respectively. The value of σY0.1% was 440 MPa for the
image (dashed white square), the displacement of atomic planes for the
Rex sample, while it slightly increased and reached 475 MPa for the Rex-
formation of ε-martensite from γ-austenite was revealed as shown with
Aged sample. After performing the aging heat treatment at 485 ◦ C for 6
red lines in the inset of Fig. 6d.
h, the σY0.1% increased to 525 MPa for the Rex-Double aged sample.
Figs. 7a–b show TEM micrographs from the substructure of the Rex-
Similarly, the σ4% increased from 670 MPa for the Rex sample to 737 and
Double aged sample after performing a loading–unloading test to a peak
750 MPa for the Rex-Aged and Rex-Double aged samples, respectively.
strain of 2% in tension. Unlike the substructures of the non-deformed
Furthermore, in comparison to the Rex sample with a residual strain of
samples (Figs. 5 and 6), parallel SFs crossed each other after applying
3.09%, the Rex-Aged and Rex-Double aged samples showed relatively
the deformation (blue arrows in Fig. 7a). Moreover, the crossing SFs
lower residual strains of 2.87% and 2.70%, respectively. Such behavior
were trapped by the polyhedral-shaped, (Cr–V–C)-rich precipitates,
could be mainly related to the higher σ4% values, lower Young's moduli
resulting in the formation of highly distorted areas around the pre­
and higher PE strains of the aged samples compared with those of the
cipitates in the austenite matrix (yellow arrows in Fig. 7a). Fig. 7b in­
Rex sample (Table 1 and Eq. (1)). It is worth mentioning that the ob­
dicates sparse V-rich precipitates, ranging from 70 to115 nm in size,
tained residual strain of 2.70% for the Rex-Double aged samples after
inside a variant of annealing twins in the austenite phase. No SFs were
4% loading in tension is the lowest residual strain value reported so far
observed in the twinned crystals that could tangle around the elliptical-
for this particular alloy.
shaped, V-rich precipitates, compared with the polyhedral-shaped

Table 1
Mechanical characteristics of the As-Received aged [35], Rex, Rex-Aged, and Rex-Double aged samples after loading–unloading experiments with a peak strain of 4%
in tension. εres: residual strain, εpse: PE strain, εE: elastic strain, σY0.1%: 0.1% offset yield stress, σ4%: strength at peak tensile strain, Eloading: elastic modulus in loading,
Eunloading: elastic modulus in unloading.
Sample name E loading (GPa) σY0.1% (Mpa) σ4% (Mpa) Eunloading (GPa) εE (%) εpse (%) εres (%)
As-Received aged [35] 160 275 600 130 0.46 0.54 3
Rex 160 440 670 135 0.50 0.41 3.09
Rex-Aged 105 475 737 115 0.64 0.49 2.87
Rex-Double aged 80 525 750 140 0.53 0.77 2.70

4
H. Khodaverdi et al. Materials Characterization 195 (2023) 112486

Fig. 5. (a–b) Bright-field TEM images of the Rex sample illustrating SFs and precipitates in the austenite matrix. (c) Bright-field TEM image of SFs and (Cr–V–C)-rich
precipitates in the Rex-Aged sample. (d) Bright-field and dark-field TEM images from ε-martensite laths in the Rex-Aged sample; the [0110] reflection of ε-martensite
was used for dark-field imaging as shown in the corresponding selected area electron diffraction pattern.

precipitates in the non-twinned regions. It was realized that these V-rich accomplished by forming SFs on (111) planes of the FCC crystal in every
precipitates evolved from the fine particles observed in the Rex sample two layers at pre-existing SFs or as newly developing SFs [47,48].
(Fig. 5a) such that their size increased at the expense of the reduction in Applying external stress will generate a dislocation loop two layers away
their volume fraction by double aging the sample (Figs. 6a and 7b). from the existing SFs to produce a four-layer ε-martensite plate. When
Table 2 lists the reported chemical compositions of observed precipitates the thickness of the ε-martensite plate reaches a threshold value, a new
in several FeMnSi-SMAs subjected to different heat treatment conditions martensite plate will then be formed at a certain distance to the first
and the average chemical compositions of observed precipitates in this martensite plate to relax the lattice strain of the initially formed
study. martensite plate. Therefore, the SFs are considered as embryos for the
nucleation of ε-martensite [49,50]. It has also been reported that the
4. Discussion essential factor for obtaining good PE behavior in Fe-based SMAs is the
propensity to produce very thin ε-martensite plates by applying external
Several studies have investigated the mechanism of PE in Fe–Mn–Si- stress [51]. Having a high density of SFs in the austenite grains is
based SMAs. In the first attempts, Sawaguchi et al. [46] justified the PE necessary to form extremely thin and single-variant ε-martensite plates
behavior by the reverse transformation from ε-martensite to γ-austenite with a uniform distribution within each grain of the austenite phase.
in unloading caused by the back-stress experienced by SPDs residing at Given the fact that α'-martensite, which weakens the PE behavior of Fe-
the top of the ε-martensite plates. Leinenbach et al. [19] later showed based SMAs, can nucleate at the intersection of two ε-martensite laths
that the microstructural reason for PE is a combination of the back [52], the formation of SFs in random orientations should be minimized
transformation from ε-martensite to γ-austenite and reversible motions in the microstructure.
of SPDs. In low SFE materials, a dislocation typically can dissociate into Although the microstructure of the Rex specimen contains relatively
two extended dislocations, creating an SF bounded by SPDs. Therefore, equiaxed austenite grains with a small grain size, the tensile test does not
SFs play an important role in the PE response of the studied Fe–Mn–Si- show a significant PE response for this specimen (Figs. 3 and 4). The
based SMA with an SFE value of 5.75 mJ/m2 [35]. presence of 7.9 vol% α'-martensite in the microstructure is one of the
Generally, the nucleation of ε-martensite in γ-austenite can be reasons for this inferior PE behavior. Furthermore, the TEM images

5
H. Khodaverdi et al. Materials Characterization 195 (2023) 112486

Fig. 6. Bright-field TEM images of the Rex-Double aged sample illustrating (a) a high number of parallel SFs in an austenite grain, (b) polyhedral- and parallelogram-
shaped precipitates along with parallel SFs and ε-martensite laths in the austenite matrix, and (c) the distribution of parallelogram-shaped precipitates in the
austenite matrix. (d) HR-TEM image and corresponding FFT and IFFT images of an ε-martensite lath formed parallel to the {111} planes of austenite. Selected lattice
spots for producing the IFFT image are shown by the red circles in the FFT pattern. (For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)

show elliptical-shaped particles with a size of 15–20 nm in the micro­ M7C3 carbides in the microstructure. By comparing these data with the
structure of the Rex sample (Fig. 5a). Stanford et al. [53] have shown precipitation behavior of the studied alloy at three different tempera­
that fine particles with a size of less than 30 nm can be counterpro­ tures of 925, 750, and 485 ◦ C (Figs. 5 and 6), the polyhedral-shaped and
ductive to achieving superior PE behavior by pinning stress-induced parallelogram-shaped precipitates in the Rex-Double aged sample could
ε-martensite and inhibiting its reversion to γ-austenite. be related to M7C3-type and sigma-type precipitates in the phase dia­
The εres value of the Rex-Aged sample (2.87%) shows about 7.1% gram, respectively. Although there are some studies that support these
improvement over that of the Rex sample (3.09%). This enhancement is arguments [55,56], the crystallography of the parallelogram-shaped
mainly related to the more formation of (Cr–V–C)-rich precipitates in­ precipitates with the average chemical composition of
side the austenite grains with the same austenite grain size (5 μm) Fe–14Mn–5Si–26Cr–3Ni–6V–12C (wt%) needs to be investigated in
compared with the Rex sample. These polyhedral-shaped precipitates, detail in the future.
which create lattice strains (Fig. 5c), are uniformly distributed in the The observed improvement of εpse in the Rex-Double aged sample is
austenite matrix (Fig. 3b), leading to the formation of a significant thought to be partly caused by the formation of the new precipitates,
number of SFs accompanied by a high fraction (43 vol%) of ε-martensite which can induce the development of a higher number of SFs grouped in
(Figs. 5c–d). The high number of SFs not only facilitates the γ → ε the same orientation inside the austenite grains compared with the other
martensitic transformation but also results in the significant reduction of samples (Figs. 6a–b). The generation of a high number of precipitate/
Young's modulus compared with the Rex sample (Table 1). The reduc­ matrix interfaces in the double-aged sample can play a role in the for­
tion in Young's modulus is related to the change in the interatomic mation of new SFs in the microstructure. It has been shown that dislo­
bonding configuration at the SFs [54]. cations form at the precipitate/matrix interface to accommodate the
In comparison to the εpse value of the Rex-Aged sample, an increase elastic strain fields (represented in Figs. 5c and 6b) accompanied by the
of 57% in εpse is observed for the Rex-Double aged sample, resulting in a formation of the precipitate [35,57]. Consequently, the SFs can be
higher PE strain of 0.77%. According to the equilibrium thermodynamic formed by the dissociation of the dislocations associated with the pre­
data at 485 ◦ C (Fig. 2), sigma-type precipitates can form along with cipitate/matrix interface [58]. In addition, the formation of SFs after

6
H. Khodaverdi et al. Materials Characterization 195 (2023) 112486

Fig. 7. Bright-field TEM images of the Rex-Double aged sample after a loading–unloading experiment to a peak strain of 2% in tension. (a) Groups of parallel SFs
crossing each other (blue arrows) around a (Cr–V–C)-rich precipitate in non-twinned regions of the austenite matrix, creating distorted zones around the precipitate
(yellow arrows). (b) Three elliptical-shaped, V-rich precipitates (inset) in a variant of annealing twins inside the austenite matrix with no observation of SFs around
them. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 2
Chemical compositions of various kinds of precipitates formed under different heat treatment conditions in several FeMnSi-SMAs and average chemical compositions
of precipitates observed in this study (*).
Alloy composition (wt%) Heat treatment Comments Precipitate composition (wt%) Ref.

Fe–30Mn–6Si–5Cr Hot-rolled GB-phase Fe–29.5Mn–5.9Si–6.3Cr [70]

Fe–15Mn–7Si–9Cr–5Ni Aged at 900 ◦ C for 1 h Intermetallic Fe–17.5Mn–13.5Si–11.2Cr–7.6Ni [71]
Fe–15Mn–6Si–9Cr–5Ni–1.5Ti–0.16C Aged at 850 ◦ C for 0.5 h Rod-like Fe–16Mn–8.5Si–10Cr–4Ni [72]
Fe–(13–27)Mn–5.5Si–8.5Cr–5Ni As-cast Sigma-phase Fe–17.7Mn–8.1Si–15.8Cr–3Ni [73]
Chi-phase Fe–31.5Mn–15.8Si–7.8Cr–10.6Ni
Lathy-phase Fe–12.5Mn–7Si–12.1Cr–3Ni
Fe–17Mn–5Si–10Cr–4Ni–1(V-C) Aged at 750 ◦ C for 6 h Polyhedral-shaped Fe–14Mn–6Si–30Cr–2Ni–6V–8C *
Fe–17Mn–5Si–10Cr–4Ni–1(V-C) Aged at 485 ◦ C for 6 h Parallelogram-shaped Fe–14Mn–5Si–26Cr–3Ni–6V–12C *
Fe–17Mn–5Si–10Cr–4Ni–1(V-C) Aged at 485 ◦ C for 6 h Elliptical-shaped Fe–7Mn–10Cr–2Ni–50V–0.2Si − 0.2C *

aging can also stem from the segregation of solute atoms at the newly In the Rex-Aged sample with fewer oriented SFs, in comparison with
generated dislocations during aging, resulting in the local lowering of the Rex-Double aged sample, multiple martensite variants form in
the SFE at the dislocations and causing the dislocations to dissociate to different orientations within each austenite grain (Fig. 5d). However,
nucleate new SFs [30]. The formation of SFs in one orientation inhibits due to the presence of enough large precipitates (70–330 nm) along with
the development of new SFs in random orientations and reduces the the higher number of oriented SFs in the Rex-Double aged sample,
chance of the intersection of SFs with each other. Furthermore, the in­ ε-martensite laths almost form in one orientation within each austenite
crease in the size of the pre-existed V-rich particles from less than 30 nm grain (Fig. 6b). As a result, a relatively lower volume fraction of ther­
(15–20 nm) to about 70–115 nm can have a synergistic effect on the PE mally formed ε-martensite develops in the Rex-Double aged sample
improvement of the Rex-Double aged sample compared with the Rex (16%) compared with that of the Rex-Aged sample (43%), which may be
and Rex-Aged samples. beneficial for PE improvement because the chance of intersection be­
The shape and morphology of the precipitates are determined by the tween thermally formed and stress-induced formed ε-martensite is
interplay between the elastic strains and interfacial energy minimization reduced.
during precipitation. The elastic energy is dependent on the elastic To investigate the role of martensite-start (MS) and austenite-start
stiffness, misfit strain, precipitate size, and applied stress, and the (AS) temperatures of the γ ⇌ ε phase transformation in the evolution
interfacial energy is dependent on imbalanced atomic forces in the of thermally formed ε-martensite, dilatometry tests were performed on
precipitate/matrix interface [57,59]. As shown in Fig. 6b, the SFs form the heat-treated samples. As shown in Fig. 8, the measured MS tem­
around the polyhedral-shaped, M7C3-type precipitate, and after peratures for the Rex, Rex-Aged, and Rex-Double aged samples are 240,
applying tensile strain they are multiplied in different orientations and 255, and 260 ◦ C, respectively, which are well above the room temper­
trapped by the precipitate (Fig. 7a). However, the stress fields around ature. The increases in the MS values after aging can be related to the
the parallelogram-shaped, sigma-type precipitate can to some extent formation of the precipitates and the resultant changes in the chemical
prevent the SFs from colliding with the precipitate (Fig. 6b), resulting in composition of the austenite matrix near the precipitates and also to the
a more oriented distribution of the SFs in the austenite matrix that may lattice strains generated during the evolution of the precipitates
reduce the probability of their trapping by the precipitate during the [60–62]. Moreover, the measured AS temperatures for the Rex, Rex-
deformation. It should be noted that possible local chemical fluctuations Aged, and Rex-Double aged samples are 155, 140, and 115 ◦ C, respec­
in the matrix near the precipitates and their effects on local SFE varia­ tively, which are lower than the corresponding MS temperatures. These
tions may also influence the formation of SFs, which require further results can indicate the facilitation of martensite-to-austenite trans­
investigations in the future. formation by precipitation; obstacles such as precipitates can increase

7
H. Khodaverdi et al. Materials Characterization 195 (2023) 112486

can be drawn:

1. The Rex specimen with a γ-austenite grain size of 5 μm has the lowest
measured PE strain of 0.41% among the other specimens. Although
the microstructure of the Rex specimen contains fine, equiaxed
austenite grains, the presence of 7.9 vol% α'-martensite along with
small V-rich particles with a size of 15–20 nm reduces the PE effect.
2. A significant improvement of approximately 20% in the PE strain is
achieved by aging the Rex specimen at 750 ◦ C for 6 h. This
enhancement of PE for the Rex-Aged specimen is mainly due to the
precipitation of polyhedral-shaped, (Cr–V–C)-rich precipitates
(M7C3-type) with a uniform distribution inside the austenite grains,
which causes the formation of a large number of stacking faults (SFs)
and ε-martensite laths in the microstructure.
3. The highest measured PE strain of 0.77% for this alloy is attained by
performing double aging heat treatment on the Rex-Aged specimen
at 485 ◦ C for 6 h. The formation of fresh, parallelogram-shaped,
(Cr–V–C)-rich precipitates (sigma-type) with the average chemical
composition of Fe–14Mn–5Si–26Cr–3Ni–6V–12C (wt%) and an in­
Fig. 8. Dilatation curves versus temperature at the cooling/heating rate of
crease in the size of pre-existed V-rich particles to 70–115 nm with
1 ◦ C/s for the Rex (C1 and H1), Rex-Aged (C2 and H2), and Rex-Double aged
the chemical composition of Fe–7Mn–10Cr–2Ni–50V–0.2Si–0.2C (wt
(C3 and H3) samples. The samples were first cooled from the heat treatment
temperature and then heated to measure the γ ⇌ ε phase transformation
%) are believed to be responsible for the PE improvement in the Rex-
temperatures. Double aged specimen.
4. The formation of sigma-type precipitates in the microstructure of the
Rex-Double aged specimen causes a large number of SFs to be ori­
the elastic energy storage during the γ → ε phase transformation,
ented in the same direction inside the austenite grains. It is postu­
assisting the reverse transformation [63–66]. These observations sug­
lated that the parallel alignment of these SFs may further improve
gest that the type of precipitates and the orientation of SFs are the
the γ ⇌ ε martensitic phase transformation by reducing the proba­
decisive factors in determining the volume fraction of thermally formed
bility of the collision of ε-martensite laths with each other and
ε-martensite in the microstructures.
consequent formation of α’-martensite during tensile loa­
Jian et al. [67] have shown that during applying strain, more SFs
ding–unloading experiments. This results in a residual strain of 2.7%
would form on a specific set of {111} γ planes and combine with the pre-
after 4% loading in tension for the Rex-Double aged specimen, which
existing SFs on these planes. The accumulation and extension of SFs on
shows about 13% improvement compared with that of the Rex
the same plane lead to the coalescence of individual SFs, generating
sample.
longer SFs. The long SFs tend to be spaced regularly with the increase in
the applied strain (Fig. 7a). When the spacing between adjacent long SFs
reaches twice the spacing between adjacent {111} γ planes, the local Declaration of Competing Interest
regions will transform into ε-martensite. Similar to the role of pre­
cipitates, having more oriented groups of SFs may also assist in the The authors declare that they have no known competing financial
formation of ε-martensite laths just on a specific set of {111} γ planes, interests or personal relationships that could have appeared to influence
thereby reducing the possibility of the intersection of ε-martensite var­ the work reported in this paper.
iants with each other and consequent formation of α'-martensite, which
is detrimental to PE. Additionally, the activation energy required for the Data availability
γ → ε martensitic transformation is reduced by forming a high number of
SFs in the matrix, resulting in better PE behavior [19]. These conditions The raw/processed data required to reproduce the findings of this
can be met in the Rex-Double aged sample due to the formation of the study are available from the corresponding authors upon reasonable
sigma-type precipitates alongside the M7C3-type precipitates in the request. The acquired EDS spectra used for calculating the chemical
austenite matrix compared with the Rex-Aged sample (Figs. 6 and 7). compositions of the precipitates are provided in the Supplementary
Another factor that improves the PE of the aged samples is the Material File.
absence of coarse annealing twins in the microstructures. To obtain a
good PE response, the formation of annealing twins should be sup­ Acknowledgment
pressed in Fe–Mn–Si-based SMAs [21,68,69]. The interactions between
annealing twins and stress-induced ε-martensite not only distort the The authors acknowledge the support from re-fer AG, Switzerland,
twin boundaries but also significantly inhibit the γ → ε martensitic for providing the material for this research study.
transformation. The annealing twins may also prevent the formation of
thermally induced ε-martensite because a rise in the fraction of the twins Appendix A. Supplementary data
lowers the MS temperature [69].
Supplementary data to this article can be found online at https://doi.
5. Conclusions org/10.1016/j.matchar.2022.112486.

The microstructure and pseudoelasticity (PE) of an References

Fe–17Mn–5Si–10Cr–4Ni–1(V-C) (wt%), Fe-based shape memory alloy
have been studied under recrystallized (Rex), recrystallized and aged [1] H. Koohdar, M. Nili-Ahmadabadi, F. Javadzadeh Kalahroudi, H.R. Jafarian, T.
G. Langdon, The effect of high-pressure torsion on the microstructure and
(Rex-Aged), and recrystallized and double aged (Rex-Double aged) outstanding pseudoelasticity of a ternary Fe–Ni–Mn shape memory alloy, Mater.
conditions. According to experimental investigations and detailed Sci. Eng. A 802 (2021), https://doi.org/10.1016/j.msea.2020.140647.
microstructural and mechanical assessments, the following conclusions [2] M. Vollmer, C. Segel, P. Krooß, J. Günther, L. Tseng, On the effect of gamma phase
formation on the pseudoelastic performance of polycrystalline Fe–Mn–Al–Ni shape

8
H. Khodaverdi et al. Materials Characterization 195 (2023) 112486

memory alloys, Scr. Mater. 108 (2015) 23–26, https://doi.org/10.1016/j. [27] L. Chengxin, W. Guixin, W. Yandong, L. Qingsuo, Z. Jianjun, Effect of addition of V
scriptamat.2015.06.013. and C on strain recovery characteristics in Fe–Mn–Si alloy, Mater. Sci. Eng. A
[3] A. Bauer, M. Vollmer, T. Niendorf, Effect of crystallographic orientation and grain 438–440 (2006) 808–811, https://doi.org/10.1016/J.MSEA.2006.01.098.
boundaries on martensitic transformation and Superelastic response of [28] S. Kajiwara, D. Liu, T. Kikuchi, N. Shinya, Remarkable improvement of shape
Oligocrystalline Fe–Mn–Al–Ni shape memory alloys, Shape Mem. Superelast. 7 memory effect in Fe-Mn-Si based shape memory alloys by producing NbC
(2021) 373–382, https://doi.org/10.1007/S40830-021-00340-3. precipitates, Scr. Mater. 44 (2001) 2809–2814, https://doi.org/10.1016/S1359-
[4] Z. Zhang, J. Zhang, H. Wu, Y. Ji, D. Kumar, Iron-based shape memory alloys in 6462(01)00978-2.
construction: research, applications and opportunities, Materials. 1723 (2022), [29] A. Baruj, T. Kikuchi, S. Kajiwara, N. Shinya, Improvement of shape memory
https://doi.org/10.3390/ma15051723. properties of NbC containing Fe–Mn–Si based shape memory alloys by simple
[5] X.L. Gu, Z.Y. Chen, Q.Q. Yu, E. Ghafoori, Stress recovery behavior of an Fe-Mn-Si thermomechanical treatments, Mater. Sci. Eng. A 378 (2004) 333–336, https://doi.
shape memory alloy, Eng. Struct. 243 (2021), 112710, https://doi.org/10.1016/J. org/10.1016/j.msea.2003.10.357.
ENGSTRUCT.2021.112710. [30] M.J. Lai, Y.J. Li, L. Lillpopp, D. Ponge, S. Will, D. Raabe, On the origin of the
[6] H. Peng, G. Wang, S. Wang, J. Chen, I. MacLaren, Y. Wen, Key criterion for improvement of shape memory effect by precipitating VC in Fe–Mn–Si-based shape
achieving giant recovery strains in polycrystalline Fe-Mn-Si based shape memory memory alloys, Acta Mater. 155 (2018) 222–235, https://doi.org/10.1016/j.
alloys, Mater. Sci. Eng. A 712 (2018) 37–49, https://doi.org/10.1016/j. actamat.2018.06.008.
msea.2017.11.071. [31] J. van Aswegen, R.D.W. Honeycombe, Precipitation on stacking faults in Cr-Ni
[7] A.V. Druker, A. Perotti, I. Esquivel, J. Malarría, A manufacturing process for shaft austenitic steels, Acta Metall. 12 (1965) 1–13, https://doi.org/10.1016/0001-6160
and pipe couplings of Fe–Mn–Si–Ni–Cr shape memory alloys, Mater. Des. 56 (2014) (64)90048-3.
878–888, https://doi.org/10.1016/j.matdes.2013.11.032. [32] F.H. Froes, B.W.K. Honeycombe, D.H. Warrington, Conditions controlling matrix
[8] W.J. Lee, B. Weber, C. Leinenbach, Recovery stress formation in a restrained and stacking fault precipitation, Acta Metall. 15 (1967) 157–159, https://doi.org/
Fe–Mn–Si-based shape memory alloy used for prestressing or mechanical joining, 10.1016/0001-6160(67)90170-8.
Constr. Build. Mater. 95 (2015) 600–610, https://doi.org/10.1016/J. [33] K. Kamei, Y. Maehara, Y. Ohmori, Effect of stacking fault precipitation on hot
CONBUILDMAT.2015.07.098. deformation of austenitic stainless steel, Trans. Iron Steel Inst. Jpn. 26 (1986)
[9] L. Janke, C. Czaderski, M. Motavalli, J. Ruth, Applications of shape memory alloys 159–166, https://doi.org/10.2355/ISIJINTERNATIONAL1966.26.159.
in civil engineering structures—overview, limits and new ideas, Mater. Struct. 38 [34] J.J. Irani, R.T. Weiner, Precipitation of vanadium carbide on stacking faults,
(5) (2005) 578–592, https://doi.org/10.1007/BF02479550. Nature 205 (4973) (1965) 795, https://doi.org/10.1038/205795a0.
[10] H. Otsuka, K. Nakajima, T. Maruyama, Superelastic behavior of Fe–Mn–Si–Cr [35] H. Khodaverdi, M. Mohri, E. Ghafoori, A. Sabet Ghorabaei, M. Nili-Ahmadabadi,
shape memory alloy coil, Mater. Trans. 41 (2000) 547–549, https://doi.org/ Enhanced pseudoelasticity of an Fe–Mn–Si-based shape memory alloy by applying
10.2320/MATERTRANS1989.41.547. microstructural engineering through recrystallization and precipitation, J. Mater.
[11] Z. Dong, U.E. Klotz, C. Leinenbach, A. Bergamini, C. Czaderski, M. Motavalli, Res. Technol. 21C (2022) 2999–3013, https://doi.org/10.1016/j.
A novel Fe-Mn-Si shape memory alloy with improved shape recovery properties by jmrt.2022.10.092.
VC precipitation, Adv. Eng. Mater. 11 (2009) 40–44, https://doi.org/10.1002/ [36] M. Mohri, I. Ferretto, C. Leinenbach, D. Kim, D.G. Lignos, E. Ghafoori, Effect of
adem.200800312. thermomechanical treatment and microstructure on pseudo-elastic behavior of
[12] W. Wang, A. Hosseini, E. Ghafoori, Experimental study on Fe-SMA-to-steel Fe–Mn–Si–Cr–Ni-(V, C) shape memory alloy, Mater. Sci. Eng. A 855 (2022),
adhesively bonded interfaces using DIC, Eng. Fract. Mech. 244 (2021), 107553, 143917, https://doi.org/10.1016/J.MSEA.2022.143917.
https://doi.org/10.1016/J.ENGFRACMECH.2021.107553. [37] G. Cliff, G.W. Lorimer, The quantitative analysis of thin specimens, J. Microsc. 103
[13] E. Ghafoori, M. Neuenschwander, M. Shahverdi, C. Czaderski, M. Fontana, (1975) 203–207, https://doi.org/10.1111/j.1365-2818.1975.tb03895.x.
Elevated temperature behavior of an iron-based shape memory alloy used for [38] Thermo-Calc Documentation Set, Version, 2022.
prestressed strengthening of civil structures, Constr. Build. Mater. 211 (2019) [39] T. Song, B.C. de Cooman, Martensite nucleation at grain boundaries containing
437–452, https://doi.org/10.1016/J.CONBUILDMAT.2019.03.098. intrinsic grain boundary dislocations, ISIJ Int. 54 (2014) 2394–2403, https://doi.
[14] D.I.H. Rosa, A. Hartloper, A. de Castro e Sousa, D.G. Lignos, M. Motavalli, org/10.2355/ISIJINTERNATIONAL.54.2394.
E. Ghafoori, Experimental behavior of iron-based shape memory alloys under [40] M.J. Sohrabi, M. Naghizadeh, H. Mirzadeh, Deformation-induced martensite in
cyclic loading histories, Constr. Build. Mater. 272 (2021), 121712, https://doi.org/ austenitic stainless steels: a review, Arch. Civil Mech. Eng. 20 (2020) 1–24, https://
10.1016/J.CONBUILDMAT.2020.121712. doi.org/10.1007/s43452-020-00130-1.
[15] E. Hosseini, E. Ghafoori, C. Leinenbach, M. Motavalli, S.R. Holdsworth, Stress [41] J.H. Robertson, IUCr, Elements of X-ray Diffraction by B. D. Cullity, Urn:Issn:0567-
recovery and cyclic behaviour of an Fe–Mn–Si shape memory alloy after multiple 7394 35, 1979, p. 350, https://doi.org/10.1107/S0567739479000917, 1-509.
thermal activation, Smart Mater. Struct. 27 (2018), 025009, https://doi.org/ [42] M.M. Pan, X.M. Zhang, D. Zhou, R.D.K. Misra, P. Chen, Fe–Mn–Si–Cr–Ni based
10.1088/1361-665X/AAA2C9. shape memory alloy: thermal and stress-induced martensite, Mater. Sci. Eng. A 797
[16] C. Leinenbach, H. Kramer, C. Bernhard, D. Eifler, Thermo-mechanical properties of (2020), https://doi.org/10.1016/j.msea.2020.140107.
an Fe–Mn–Si–Cr–Ni–VC shape memory alloy with low transformation temperature, [43] G. Nolze, PowderCell: A Mixture between Crystal Structure Visualizer, Simulation
Adv. Eng. Mater. 14 (2012) 62–67, https://doi.org/10.1002/adem.201100129. and Refinement Tool, International School on Powder Diffraction, 2002.
[17] J. Vůjtěch, P. Ryjáček, J.C. Matos, E. Ghafoori, Iron-based shape memory alloy for [44] W.J. Lee, B. Weber, G. Feltrin, C. Czaderski, M. Motavalli, C. Leinenbach, Phase
strengthening of 113-year bridge, Eng. Struct. 248 (2021), 113231, https://doi. transformation behavior under uniaxial deformation of an Fe–Mn–Si–Cr–Ni–VC
org/10.1016/j.engstruct.2021.113231. shape memory alloy, Mater. Sci. Eng. A 581 (2013) 1–7, https://doi.org/10.1016/
[18] R. Fosdick, Y. Ketema, Shape memory alloys for passive vibration damping, j.msea.2013.06.002.
J. Intell. Mater. Syst. Struct. 9 (1998) 854–870, https://doi.org/10.1177/ [45] M. Beßling, C. Czaderski, J. Orlowsky, Prestressing effect of shape memory alloy
1045389X9800901009. reinforcements under serviceability tensile loads, Buildings 11 (2021) (2021) 101,
[19] C. Leinenbach, A. Arabi-Hashemi, W.J. Lee, A. Lis, M. Sadegh-Ahmadi, S. van https://doi.org/10.3390/buildings11030101.
Petegem, T. Panzner, H. van Swygenhoven, Characterization of the deformation [46] T. Sawaguchi, T. Kikuchi, S. Kajiwara, The pseudoelastic behavior of Fe–Mn–Si-
and phase transformation behavior of VC-free and VC-containing FeMnSi-based based shape memory alloys containing Nb and C, Smart Mater. Struct. 14 (2005)
shape memory alloys by in situ neutron diffraction, Mater. Sci. Eng. A 703 (2017) S317, https://doi.org/10.1088/0964-1726/14/5/022.
314–323, https://doi.org/10.1016/j.msea.2017.07.077. [47] N. Bergeon, G. Guenin, C. Esnouf, Microstructural analysis of the stress-induced ε
[20] T. Omori, M. Okano, R. Kainuma, Effect of grain size on superelasticity in Fe-Mn- martensite in a Fe–Mn–Si–Cr–Ni shape memory alloy: part I—calculated
Al-Ni shape memory alloy wire, APL Mater. 1 (2013), https://doi.org/10.1063/ description of the microstructure, Mater. Sci. Eng. A 242 (1998) 77–86, https://
1.4820429. doi.org/10.1016/S0921-5093(97)00511-X.
[21] G. Wang, H. Peng, C. Zhang, S. Wang, Y. Wen, Relationship among grain size, [48] N. Bergeon, G. Guenin, C. Esnouf, Microstructural analysis of the stress-induced ε
annealing twins and shape memory effect in Fe–Mn–Si based shape memory alloys, martensite in a Fe–Mn–Si–Cr–Ni shape memory alloy: part II: transformation
Smart Mater. Struct. 25 (2016), 075013, https://doi.org/10.1088/0964-1726/25/ reversibility, Mater. Sci. Eng. A 242 (1998) 87–95, https://doi.org/10.1016/
7/075013. S0921-5093(97)00512-1.
[22] A. Arabi-Hashemi, W.J. Lee, C. Leinenbach, Recovery stress formation in FeMnSi [49] E. Gartstein, A. Rabinkin, On the f.c.c. → h.c.p. phase transformation in high
based shape memory alloys: impact of precipitates, texture and grain size, Mater. manganese-iron alloys, Acta Metall. 27 (1979) 1053–1064, https://doi.org/
Des. 139 (2018) 258–268, https://doi.org/10.1016/j.matdes.2017.11.006. 10.1016/0001-6160(79)90193-7.
[23] O. Matsumura, T. Sumi, N. Tamura, K. Sakao, T. Furukawa, H. Otsuka, [50] H. Li, D. Dunne, N. Kennon, Factors influencing shape memory effect and phase
Pseudoelasticity in an Fe–28Mn–6Si–5Cr shape memory alloy, Mater. Sci. Eng. A transformation behaviour of Fe–Mn–Si based shape memory alloys, Mater. Sci.
279 (2000) 201–206, https://doi.org/10.1016/S0921-5093(99)00644-9. Eng. A 273–275 (1999) 517–523, https://doi.org/10.1016/S0921-5093(99)00391-
[24] Y.H. Wen, L.R. Xiong, N. Li, W. Zhang, Remarkable improvement of shape memory 3.
effect in an Fe–Mn–Si–Cr–Ni–C alloy through controlling precipitation direction of [51] S. Kajiwara, Characteristic features of shape memory effect and related
Cr23C6, Mater. Sci. Eng. A 474 (2008) 60–63, https://doi.org/10.1016/J. transformation behavior in Fe-based alloys, Mater. Sci. Eng. A 273 (1999) 67–88,
MSEA.2007.05.043. https://doi.org/10.1016/S0921-5093(99)00290-7.
[25] N. Stanford, D.P. Dunne, Effect of NbC and TiC precipitation on shape memory in [52] L. Bracke, L. Kestens, J. Penning, Transformation mechanism of α’-martensite in an
an iron-based alloy, J. Mater. Sci. 41 (2006) 4883–4891, https://doi.org/10.1007/ austenitic Fe–Mn–C–N alloy, Scr. Mater. 57 (2007) 385–388, https://doi.org/
S10853-006-0050-7. 10.1016/j.scriptamat.2007.05.003.
[26] K. Li, Z. Dong, Y. Liu, L. Zhang, A newly developed Fe-based shape memory alloy [53] N. Stanford, D.P. Dunne, Effect of second-phase particles on shape memory in
suitable for smart civil engineering, Smart Mater. Struct. 22 (2013), 045002, Fe–Mn–Si-based alloys, Mater. Sci. Eng. A 454 (2007) 407–415, https://doi.org/
https://doi.org/10.1088/0964-1726/22/4/045002. 10.1016/j.msea.2006.11.084.

9
H. Khodaverdi et al. Materials Characterization 195 (2023) 112486

[54] Y. Chen, T. Burgess, X. An, Y.-W. Mai, H.H. Tan, J. Zou, S.P. Ringer, C. Jagadish, [64] D.P. Dunne, C.M. Wayman, The effect of austenite ordering on the martensite
X. Liao, Effect of a high density of stacking faults on the Young’s modulus of GaAs transformation in Fe-Pt alloys near the composition Fe3Pt: II. Crystallography and
nanowires, Nano Lett. 16 (2016) 1911–1916, https://doi.org/10.1021/acs. general features, Metall. Trans. A. 4 (1) (1973) 147–152, https://doi.org/10.1007/
nanolett.5b05095. BF02649613.
[55] A.I. Gorunov, Investigation of M7C3, M23C6 and M3C carbides synthesized on [65] E. Acar, G.P. Toker, H. Kurkcu, H.E. Karaca, High temperature shape memory
austenitic stainless steel and carbon fibers using laser metal deposition, Surf. Coat. behavior of Ni47.3Ti29.7Hf20Pd3 alloys, Intermetallics (Barking) 111 (2019),
Technol. 401 (2020), 126294, https://doi.org/10.1016/J. 106518, https://doi.org/10.1016/J.INTERMET.2019.106518.
SURFCOAT.2020.126294. [66] E. Acar, M. Kok, I.N. Qader, Exploring surface oxidation behavior of NiTi–V alloys,
[56] F. Ernst, D. Li, H. Kahn, G.M. Michal, A.H. Heuer, The carbide M7C3 in low- Eur. Phys. J. Plus 135 (1) (2020) 1–9, https://doi.org/10.1140/EPJP/S13360-019-
temperature-carburized austenitic stainless steel, Acta Mater. 59 (2011) 00087-Y.
2268–2276, https://doi.org/10.1016/J.ACTAMAT.2010.11.058. [67] L. Jian, C.M. Wayman, On the mechanism of the shape memory effect associated
[57] R. Shi, N. Ma, Y. Wang, Predicting equilibrium shape of precipitates as function of with γ(fcc) to ε(hcp) martensitic transformations in Fe-Mn-Si based alloys, Scr.
coherency state, Acta Mater. 60 (2012) 4172–4184, https://doi.org/10.1016/J. Metall. Mater. 27 (1992) 279–284, https://doi.org/10.1016/0956-716X(92)
ACTAMAT.2012.04.019. 90512-D.
[58] J.R. Patel, K.A. Jackson, H. Reiss, Oxygen precipitation and stacking-fault [68] H. Peng, G. Wang, Y. Du, S. Wang, J. Chen, Y. Wen, A novel training-free processed
formation in dislocation-free silicon, J. Appl. Phys. 48 (2008) 5279, https://doi. Fe-Mn-Si-Cr-Ni shape memory alloy undergoing δ → γ phase transformation,
org/10.1063/1.323558. Metall. Mater. Trans. A Phys. Metall. Mater. Sci. 47 (2016) 3277–3283, https://doi.
[59] A.M. Jokisaari, S.S. Naghavi, C. Wolverton, P.W. Voorhees, O.G. Heinonen, org/10.1007/S11661-016-3521-8.
Predicting the morphologies of γ’ precipitates in cobalt-based superalloys, Acta [69] Y.H. Wen, H.B. Peng, D. Raabe, I. Gutierrez-Urrutia, J. Chen, Y.Y. Du, Large
Mater. 141 (2017) 273–284, https://doi.org/10.1016/J.ACTAMAT.2017.09.003. recovery strain in Fe-Mn-Si-based shape memory steels obtained by engineering
[60] D.L. Rittermann, A. Kyrolainen, P.J. Ferreira, Influence of annealing treatment on annealing twin boundaries, Nat. Commun. 5 (1) (2014) 1–9, https://doi.org/
the formation of nano/submicron grain size AISI 301 Austenitic stainless steels, 10.1038/ncomms5964.
Metall. Mater. Trans. A 37 (8) (2006) 2325–2338, https://doi.org/10.1007/ [70] H. Lin, C. Lin, K. Lin, Y. Chuang, An investigation of grain-boundary phase in
BF02586207. Fe–30Mn–6Si–5Cr shape memory alloy, J. Alloys Compd. 27 (2001) 279–284.
[61] A. Sato, K. Soma, T. Mori, Hardening due to pre-existing ϵ-Martensite in an Fe- [71] B. Maji, M. Krishnan, The effect of microstructure on the shape recovery of a
30Mn-1Si alloy single crystal, Acta Metall. 30 (1982) 1901–1907, https://doi.org/ Fe–Mn–Si–Cr–Ni stainless steel shape memory alloy, Scr. Mater. 48 (2003) 71–77,
10.1016/0001-6160(82)90030-X. https://doi.org/10.1016/S1359-6462(02)00348-2.
[62] V. Kokorin, L. Kozlova, A. Titenko, Temperature hysteresis of martensite [72] D. Bu, H. Peng, Y. Wen, N. Li, Influence of ageing on wear resistance of an
transformation in aging Cu–Mn–Al alloy, Scr. Mater. 47 (2002) 499–502, https:// Fe–Mn–Si–Cr–Ni–Ti–C shape memory alloy, Mater. Des. 32 (2011) 2969–2973,
doi.org/10.1016/S1359-6462(02)00136-7. https://doi.org/10.1016/j.msea.2017.11.071.
[63] D.P. Dunne, C.M. Wayman, The effect of austenite ordering on the martensite [73] H. Peng, Y. Wen, Y. Du, Q. Yu, Q. Yang, Effect of manganese on microstructures
transformation in Fe-Pt alloys near the composition Fe3Pt: I. Morphology and and solidification modes of cast Fe-Mn-Si-Cr-Ni shape memory alloys, Metall.
transformation characteristics, Metall. Trans. A. 4 (1) (1973) 137–145, https://doi. Mater. Trans. B Process Metall. Mater. Process. Sci. 44 (2013) 1137–1143, https://
org/10.1007/BF02649612. doi.org/10.1007/S11663-013-9880-2.

10