Chinese Physics B

TOPICAL REVIEW - MAGNETISM, MAGNETIC MATERIALS, AND You may also like
INTERDISCIPLINARY RESEARCH
- Effect of Magnetic Field and Hydrostatic
Martensitic transformation and related magnetic Pressure on Martensitic Transformation
and Its Kinetics
effects in Ni—Mn-based ferromagnetic shape Tomoyuki Kakeshita, Toshio Saburi, Koichi
Kindo et al.

memory alloys - Martensitic transformation in Cu-doped

NiMnGa magnetic shape memory alloys
Pan-Pan Li, , Jing-Min Wang et al.
To cite this article: Dun-Hui Wang et al 2013 Chinese Phys. B 22 077506
- Evolution of hysteresis characteristic of
shape memory alloys at incomplete phase
transformation cyclic loading
Yajun You and Xin Guo

View the article online for updates and enhancements.

Recent citations
- Magnetic Properties and Magnetocaloric
Effect of Binary Compound NdPd
Y. J. Wang et al

- Martensitic transformation and magnetic

properties in Mn49xCuxNi41Sn10(x = 0–2)
Meng-Zhe Zhao et al

- Structural, magnetic and magnetocaloric

effect studies of Ni42.5(Fe, Co, Ni,
Cu)0.5Mn46Sn11 Heusler alloys
Itegbeyogene P. Ezekiel and Thomas
Moyo

This content was downloaded from IP address 203.110.242.14 on 10/12/2021 at 14:41

 Chin. Phys. B Vol. 22, No. 7 (2013) 077506
TOPICAL REVIEW — Magnetism, magnetic materials, and interdisciplinary research

Martensitic transformation and related magnetic effects in

Ni–Mn-based ferromagnetic shape memory alloys∗
Wang Dun-Hui(王敦辉)† , Han Zhi-Da(韩志达), Xuan Hai-Cheng(轩海成), Ma Sheng-Can(马胜灿),
Chen Shui-Yuan(陈水源), Zhang Cheng-Liang(张成亮), and Du You-Wei(都有为)
National Laboratory of Solid State Microstructures, Department of Physics, Nanjing University, Nanjing 210093, China

(Received 3 May 2013)

Ferromagnetic shape memory alloys, which undergo the martensitic transformation, are famous multifunctional ma-
terials. They exhibit many interesting magnetic properties around the martensitic transformation temperature due to the
strong coupling between magnetism and structure. Tuning magnetic phase transition and optimizing the magnetic effects
in these alloys are of great importance. In this paper, the regulation of martensitic transformation and the investigation of
some related magnetic effects in Ni–Mn-based alloys are reviewed based on our recent research results.

Keywords: magnetocaloric effect, magnetoresistance effect, exchange bias, magnetoelectric effect

PACS: 75.30.Kz, 75.30.Sg, 75.50.Cc, 75.47.De DOI: 10.1088/1674-1056/22/7/077506

1. Introduction alloys have taken on special importance.

FSMAs can be regarded as the combination of ferromag-
Ferromagnetic shape memory alloys (FSMAs), which ex-
netic (FM) martensite and parent phases, and the MT is gener-
hibit ferromagnetism and a shape memory effect simultane-
ally observed between the FM (or paramagnetic (PM)) marten-
ously, have attracted more and more attention for their po-
site phase and the FM (or PM) austenite phase due to the in-
tential applications in many areas. They exhibit many in-
equality of their magnetostructural phase transition tempera-
teresting magnetic properties around the martensitic transfor-
tures. According to the Clausius–Clapeyron relation in the
mation temperature due to the strong coupling between mag-
magnetic phase diagram, making MT occur between the PM
netism and structure. Upon cooling, they undergo a first-order
martensite phase and the FM parent phase in Ni–Mn-based
magnetostructural transition from a parent austenitic phase to
FSMAs helps to decrease the driving magnetic field and gain
a martensitic phase, with abrupt changes of magnetization
large MTREs. [8] On the other hand, FSMAs exhibit large ef-
and lattice constants. The first FSMAs discovered are near-
fects only in the vicinity of the phase transition temperature,
stoichiometric Ni2 MnGa Heusler alloys, in which a giant mag-
which markedly limits their applications. From the application
netostrain effect is observed due to the magnetically induced
point of view, tuning MT and making FSMAs exhibit large
twin boundary motion in the martensitic phase. [1–3] Besides
MTREs at a low magnetic field and desired temperature re-
this property, a large magnetocaloric effect (MCE) was re-
gions is greatly desired. In this paper, we review our recent
ported in Ni51.5 Mn22.7 Ga25.8 alloy by Hu et al. at the In-
progress in the regulation of MT and the investigation of some
stitute of Physics, Chinese Academy of Sciences, which ex-
MTREs in Ni–Mn-based FSMAs.
tends the application field of this FSMA as a potential candi-
date for magnetic refrigeration. [4] In 2004, Sutou et al. found
some new FSMAs, Ni–Mn–Z (Z = In, Sn, Sb) alloys, in which 2. Structure and magnetism in Ni–Mn-based
the martensitic transformation (MT) was observed in their off- FSMAs
stoichiometric systems. [5] Since the magnetostructural transi- The prototype Ni–Mn-based FSMA, an alloy composed
tion in these FSMAs can be induced by an external magnetic of the 2–1–1 Heusler stoichiometry and Ni2 MnGa, is the only
field, some magnetic-transition-related effects (MTREs), such system with stoichiometric composition that undergoes the
as magnetocaloric, magnetoresistance, and magnetostrain ef- MT. In the austenite state, the Ni2 MnGa Heusler alloy has
fects, are obtained in Ni–Mn–Z alloys. [6–8] More and more at- an L21 structure (space group Fm.3m), which consists of four
tention is therefore paid to these multifunctional materials for interpenetrating face-centered cubic (fcc) sublattices. [9] With
promising applications. Both the tuning of the magnetic phase decreasing temperature, it undergoes an MT and acquires a
transition and the optimization of the magnetic effects in these number of structures of martensite phases, such as the 5M and
∗ Projectsupported by the National Basic Research Program of China (Grant No. 2012CB932304) and the National Natural Science Foundation of China (Grant
No. U1232210).
† Corresponding author. E-mail: wangdh@nju.edu.cn

© 2013 Chinese Physical Society and IOP Publishing Ltd http://iopscience.iop.org/cpb http://cpb.iphy.ac.cn

077506-1
 Chin. Phys. B Vol. 22, No. 7 (2013) 077506
the 7M modulated structures. But in the case of Ni–Mn–Z FS- Z atom. [10] Therefore, tuning the chemical composition is an
MAs, the MT can be observed only in their off-stoichiometric effective method to adjust the MT of Ni–Mn–Z alloys.
compositions, and mainly in Mn-rich alloys. [5] With the sub-
50
stitution of Mn for Z atoms, the excess Mn atoms occupy the x/
x/
Z sites and the moments are coupled antiferromagnetically to 40 x/
those of the surrounding Mn atoms on the regular Mn sites.

M/emuSg-1
30
In the Ni–Mn-based FSMAs, the magnetic moments are con-
fined to the Mn atoms to a large extent and the magnetic prop- 20
erties can often be described in terms of localized magnetic
moments with indirect exchange interactions, despite the fact 10
that they are metals. [9,10] Figure 1 shows the typical thermo-
0
magnetization curves for Ni–Mn–Cr–Sn FSMAs on cooling
100 150 200 250 300 350
and heating. [11] The characteristic temperatures of FSMSs are
T/K
the martensitic start temperature Ms , martensitic finish temper-
Fig. 2. Temperature dependence of magnetization for Ni50−x Mn39+x Sn11
ature Mf , austenitic start temperature As , austenitic finish tem- alloys (x = 5, 6, 7). [12]
perature Af , and Curie temperatures of the martensitic phase
TCM and austenitic phase TCA . Upon heating, the FM behavior 70
is observed in the martensitic state at first, and then magneti-
60
zation decreases with increasing temperature. Further increas- x/
50 x/
ing the temperature causes a sudden jump of magnetization, x/
M/emuSg-1 x/
corresponding to the reverse MT. Then the austenite state un- 40 x/
dergoes a transition from FM phase to PM phase. An obvious
30
thermal hysteresis between the cooling and heating processes
is observed in Fig. 1, suggesting that the MT is a first-order 20

transition. 10

Ni45Mn42Cr2Sn11 MS 0
6 at 100 Oe 100 150 200 250 300 350 400
T/K
Af Fig. 3. Temperature dependence of magnetization for Ni50−x Cox Mn39 Sb11
M/emuSg-1

4 alloys (x = 0, 7, 8, 9, 11). [13]

Mf

Figure 2 shows the temperature dependence of magne-

tization (M–T ) for Ni50−x Mn39+x Sn11 alloys, in which a re-
2 markable change of MT temperature is observed due to the
As
heating change of Ni/Mn ratio. [12] As for the substitution method, us-
cooling TCA
ing other 3d elements, such as Cr, Co, and Cu, to substitute for
0
150 200 250 300 Ni or Mn has also proven to be a useful way to change the MT
T/K temperature. [11,13,14] Figures 3 and 4 show the M–T curves for
Fig. 1. Typical thermomagnetization curves for Ni–Mn–Cr–Sn FSMAs Ni–Mn–Co–Sn and Ni–Mn–Cu–Sb alloys, respectively. It is
on cooling and heating at 100 Oe. [11]
obvious that the MT temperatures decrease (increase) with in-
creasing Co (Cu) contents, corresponding to the decrease (in-
3. Regulation of MT in Ni–Mn–Z FSMAs
crease) of the e/a values. In order to obtain a large change
3.1. Tuning valence electron concentration of magnetization (∆M) between the martensitic and austenitic
In Ni–Mn-based FSMAs, the change of chemical com- phases, the criterion TCM < Ms < TCA should be satisfied. As
position can alter the electronic structure and the number of shown in Fig. 3, large values of ∆M are observed in the al-
conducting electrons, and then affect the MT through the mod- loys of x = 7, 8, 9. As for Ni39 Co11 Mn39 Sb11 , besides the FM
ification of electron density and exchange interaction. [9,10] Ac- transition at TCA , no obvious MT can be identified in the M–T
cordingly, it is reported that the MT temperatures of these al- curve, since Ms is lower than TCM . Unlike the aforementioned
loys are very sensitive to the valence electron concentration doping methods with 3d-transition metal elements, Han et al.
e/a (electrons per atom). Here the number of valence elec- reported the investigation of MT in Mn50 Ni50−x−y Nbx Sny al-
trons is calculated as the number of 3d and 4s electrons of a loys, in which a 4d-transition metal element is substituted for
3d-transition metal and the number of 5s and 5p electrons of a Ni. [15] Figure 5 shows the characteristic temperatures of phase
077506-2
 Chin. Phys. B Vol. 22, No. 7 (2013) 077506
transitions as a function of Nb content and e/a value for those in a magnetic field of 1 kOe on heating and cooling, in which
FSMAs. Obviously, the values of Ms , Mf , As , and Af decrease an obvious decrease of the MT temperature is observed with
with the increase of Nb content due to the decrease of e/a increasing Ge content. In Ni43 Mn46 Sn11−x Gex alloys, due to
value. For the Curie temperatures, with increasing Nb con- the smaller ionic radius of Ge, the decrease of the cell volume
tent, TCM remains almost a constant, while TCA shows a slight would result in the decrease of the Mn–Mn distance, and as
decrease, suggesting that the influence of Nb on the exchange a consequence, decrease the MT temperature. It is noted that
interaction is small. [15] TCA also changes markedly with the variation of the Ge content,
8 suggesting the effect of the Mn–Mn distance on the exchange
x/ interaction of austenitic phase.
x/
6 x/
50
M/emuSg-1

x/ Ms Af
4 x/
40
x/

M/emuSg-1
2 30

20
0
Mf
150 180 210 240 270 300 330 10
T/K As

Fig. 4. Temperature dependence of magnetization for Ni43 Mn46−x Cux Sn11 0

alloys (x = 1, 2, 3). [14] 100 150 200 250 300 350
T/K
e/a
7.96 7.92 7.88 7.84 7.80 Fig. 6. The M–T curves for Ni43 Mn46 Sn11−x Gex (x = 0, 1, 2) alloys in
320 a magnetic field of 1 kOe on heating and cooling. [17]

280 ParaA

240 TCA Ni45Mn44Sn11Bx

FerroA
T/K

Af
Mn2B
Ms
200 x/
As
Mf
WMM
Intensity

160 TCM x/

120 FerroM
x/
0 1 2 3
Nb content
x/
Fig. 5. Characteristic temperatures of phase transitions as a function of
Nb content and e/a value for Mn50 Ni50−x−y Nbx Sny alloys. [15]
40 50 60 70 80
2θ/(Ο)
3.2. Modifying Mn–Mn distance
Fig. 7. X-ray diffraction patterns for Ni43 Mn46 Sn11 Bx (x = 0, 1, 3, 5)
Since the magnetic moments are confined to the Mn alloys obtained at room temperature. [19]
atoms and the occupation of Mn atoms affects the exchange
interaction, the interatomic distance between Mn atoms plays
a key role in determining the magnetostructural transition of It is reported that introducing interstitial atoms with a
Ni–Mn-based FSMAs. Actually, the substitution of other el- small atomic radius, such as carbon or hydrogen atoms, is a
ements for Ni, Mn, or Z atoms changes not only the valence useful method to tune the magnetic phase transition temper-
electron concentration but also the cell volume, both of which ature by modifying the distance between magnetic ions. [18]
would affect the MT of Ni–Mn–Z alloys. As for the isoelec- Based on this mechanism, the investigation of MT was per-
tronic substitution, for example in Ni50 Mn34 In16−x Gax [16] and formed in a series of Ni43 Mn46 Sn11 Bx alloys, in which boron
Ni43 Mn46 Sn11−x Gex [17] alloys, Ga or Ge has the same number acts as an interstitial atom. [19] Figure 7 shows the X-ray
of valence electrons as In or Sn, respectively. Thus, in these diffraction patterns of B-doped alloys obtained at room tem-
cases, the effect of e/a on the MT should be neglected and the perature. It is obvious that these alloys retain the L21 -type
variation of the Mn–Mn distance caused by doping elements structure after the introduction of boron atoms. With the in-
with different ionic radii should be taken into account. Fig- crease of the boron content, a foreign phase of Mn2 B ap-
ure 6 shows the M–T curves for Ni43 Mn46 Sn11−x Gex alloys pears. The lattice constants of the alloys are 5.978 Å, 5.981 Å,
077506-3
 Chin. Phys. B Vol. 22, No. 7 (2013) 077506
5.984 Å, and 5.986 Å for x=0, 1,3, and 5, respectively. Fig- icant enhancement of TCA , indicating that the pre-deformation
ure 8 presents the M–T curves for all Ni43 Mn46 Sn11 Bx al- has little effect on the second-order magnetic phase transition
loys on heating in a magnetic field of 100 Oe. Obviously, of the high temperature parent phase. [25]
the transformation temperatures, including the characteristic
temperatures of MT and TCA of Ni43 Mn46 Sn11 Bx alloys, in- meltspun heating
crease with the increasing boron content, which is attributed 30 1.05 GPa
1.95 GPa
to the change of the Mn–Mn distance caused by the interstitial
introduction. [19]

M/emuSg-1
20 A
TC
TM
H=100 Oe
4 x/
x/ 10
x/
3 x/
0
M/emuSg-1

250 300 350 400

2 T/K
Fig. 9. The M–T curves for melt-spun Ni–Mn–Co–Sn ribbons and
pre-deformed Ni–Mn–Co–Sn ribbons compressed under 1.05 GPa and
1 1.95 GPa at 100 Oe on heating. [25]

0 meltspun
30 heating annealed at 1123 K
annealed at 1173 K
150 180 210 240 270 300 330
T/K
Fig. 8. The M–T curves for all Ni43 Mn46 Sn11 Bx alloys on heating in a
M/emuSg-1

20
magnetic field of 100 Oe. [19]

3.3. Altering internal or external stress

10
It is well known that the MT of Ni–Mn-based FSMAs
can be driven not only by temperature and magnetic field, but
also by stress, which is a distinctive property of these multi- 0
functional materials. Applying stress on Ni–Mn–Z alloys can 200 220 240 260 280 300 320
modify the Mn–Mn distance and affect the relative stability be- T/K
tween the high temperature cubic phase and the low tempera- Fig. 10. The M–T curves under a magnetic field of 100 Oe for the
ture martensitic phase, thereby affecting the magnetostructural melt-spun and annealed Ni–Mn–Sn ribbons on heating. [28]
transition. It has been reported that a hydrostatic pressure has
a remarkable effect on the MT in many Ni–Mn-based FSMAs, Rapid quenching by melt spinning is reported as an ef-
such as Ni–Mn–In, Ni–Mn–Co–Sb, and nickel-rich Ni–Mn– fective method to achieve homogeneous Ni–Mn–Z FSMAs,
Ga alloys. [10,20,21] We all know that pre-deformation can in- avoiding or at least reducing the time of annealing, and making
crease the MT temperature in some shape memory alloys (such it easy to obtain highly textured polycrystalline ribbons. [26,27]
as TiNi, TiNiNb), as has been verified by thermal and electri- In this way, melt-spun ribbons have been obtained by rapid
cal measurements. [22–24] Accordingly, Ma et al. reported the cooling from a liquid alloy. They possess a metastable struc-
effect of pre-pressure on the MT in Ni–Mn–Co–Sn ribbons. [25] ture and a higher energy, because the atoms, during the quick
Figure 9 shows the M–T curves for melt-spun ribbons and pre- transition from the liquid to the solid state, do not take up
deformed ribbons, compressed under 1.05 GPa and 1.95 GPa, the equilibrium positions with lower energy. Annealing the
upon heating. It is observed that the MT temperature increases as-quenched ribbons would modify the short-range order of
with increasing pressure, suggesting that the relative stability atoms, thereby affecting the exchange interaction. Based on
between the parent and the martensitic phases can be affected this mechanism, Xuan et al. reported the effect of heat treat-
by pre-deformation. On the other hand, due to the lower vol- ment on the MT in Ni–Mn–Sn melt-spun ribbons. [28] The ex-
ume of the martensitic phase in contrast to the cubic phase, periment results show that, after the annealing process, the MT
the pre-pressure would favor the martensitic phase from the temperature increases obviously in the annealed ribbons, as
thermodynamic point of view and support the martensite to a shown in Fig. 10. As mentioned above, there is a residual in-
higher temperature, which is also responsible for the increase ternal stress in melt-spun Ni–Mn–Sn ribbons due to the rapid
of the MT temperature. As shown in Fig. 9, there is no signif- cooling. After annealing, the as-quenched ribbons undergo
077506-4
 Chin. Phys. B Vol. 22, No. 7 (2013) 077506
structural and stress relaxations, which may change the Mn– martensitic phase transition should be a two-step process
Mn distance, resulting in the increased MT temperature in Ni– and an intermediate phase may exist in HPA Ni–Mn–Co–Sn
Mn–Sn ribbons. Therefore, altering the internal stress by heat alloy. [32] This phase transition can be further verified by the
treatment is also an effective way to tune the MT temperature electrical resistance and calorimetric measurements, which are
in Ni–Mn-based FSMAs. shown in Fig. 12. The application of external stress may in-
duce an enhanced magnetoelastic coupling between the mag-
3.4. Intermediate phase transition in Ni–Mn–Sn alloys netic and the structural degrees of freedom in HPA Ni–Mn–
Ni–Mn–Ga Heusler alloys have been extensively inves- Co–Sn alloy, which would be responsible for the appearance
tigated as prototypical FSMAs. More interestingly, some of the intermediate phase. [32]
of these alloys with composition close to the stoichiomet-
ric Ni2 MnGa exhibit a weak first-order phase transition from (a)
the austenitic phase to the intermediate phase prior to the heating
MT. [29–31] This intermediate phase transition is often charac-
terized by anomalies in the elastic, thermal, resistivity, and
As
magnetic properties. It is widely believed that the intermedi- Af

ρ
ate phase transition originates mainly from the magnetoelastic
TI
coupling between the magnetic and the structural degrees of HPA
freedom in the Ni–Mn–Ga alloy system. [29–31] However, there
are very few reports about the intermediate phase in Ni–Mn–Z
alloys that were discovered very recently. Considering that 200 240 280 320 360
T/K
the high-pressure synthesis is a useful method to enhance
the magetoelastic coupling in magnetic materials, we stud- heating
ied the magnetic phase transitions in an Ni–Mn–Co–Sn alloy cooling TCA
prepared by high-pressure annealing (HPA) at 4 GPa. [32] Fig-
ure 11 shows the M–T curves for Ni–Mn–Co–Sn melt-spun Af

ribbon and HPA alloy in a field of 100 Oe on heating. It is

As
Endothermic

observed that the MT temperature increases with HPA, in-

TI
dicating the effect of external stress on the phase transition TI

temperature. It is worth noting that an anomaly of the ther- TCA

momagnetic behavior is observed in the HPA alloy, occurring Mf (b)
Ms
between MT and the FM transition of austenite. As shown 200 250 300 350 400 450
in the inset of Fig. 11, magnetization above MT displays a T/K
bump at a temperature around 334 K. The rise in magnetiza- Fig. 12. (a) Variation in the resistivity of HPA alloy versus temperature
and (b) DSC curves upon cooling and heating at a rate of 10 K/min for
tion between Af and 334 K is somewhat similar to the jump HPA alloy. [32]
of magnetization observed around the MT, suggesting that the
4. MT related effects in Ni–Mn–Z alloys
100 Oe
4.1. Magnetocaloric effects
12 heating
30 Magnetic materials with a large MCE have attracted in-
M/emuSg-1

TI
Af
8 creasing attention in recent years for their potential applica-
As
100 Oe tions as magnetic refrigerants. Compared with the conven-
M/emuSg-1

4
20 tional refrigerators based on the compression–expansion of
200 280 360
T/K gas, the magnetic refrigerators have many advantages, such
meltspun
as high efficiency, small volume, and freedom from harmful
10 HPA
gas leakage. Therefore, magnetic refrigeration is considered
heating to be an environmentally friendly technology.
0 The Ni–Mn-based FSMAs are reported to exhibit a large
150 200 250 300 350 400
T/K MCE due to the large change of magnetization around the
Fig. 11. Temperature dependence of the magnetization for Ni–Mn–Co– magnetic phase transition, which makes them potential candi-
Sn melt-spun ribbon and HPA alloy in the field of 100 Oe on heating. dates for magnetic refrigeration. For a special review about
The inset presents a partly enlarged view of the M–T curve for the HPA
alloy. [32] the MCE of FSMAs, readers are referred elsewhere, [9] and
077506-5
 Chin. Phys. B Vol. 22, No. 7 (2013) 077506
hereafter we just introduce some MCE results from our recent TCA , which also gives rise to the change of magnetic entropy.
work. Therefore, a coexistence of negative and positive MCEs can
From the application point of view, magnetic refrigera- be observed in these alloys. Figure 14 shows the temper-
tion at room temperature is obviously highly desirable. By ature dependence of MCE for Ni43 Mn46−x Cux Sn11 alloys at
tuning the composition, especially substituting Co for Ni or 10 kOe. [14] The values of ∆SM around the MT temperature are
Mn, the MT temperature can be effectively adjusted to room very large and the MT can be tuned by adjusting the concen-
temperature. [13] Figure 13 plots the temperature dependence tration of Cu, which is important for the application of mag-
of magnetic entropy changes (∆SM ) for Ni50−x Cox Mn39 Sb11 netic refrigeration. In the case of MCE around TCA , the magni-
(x = 7, 8, 9) alloys at 10 kOe. These FSMAs all show a large tude of ∆SM is relatively small, but the refrigeration capacity
MCE at low magnetic fields, and the working temperature of is large. [14]
the Ni42 Co8 Mn39 Sb11 alloy is just at the room temperature. [13] For magnetic refrigeration, a wide operating temperature
range is of great importance. The HPA Ni–Mn–Co–Sn al-
10 loy shows a two-step phase transition process, and as a con-
x/
x/
x/
sequence, two successive positive MCE peaks are observed
8
around room temperature, i.e., 284 K and 320 K, correspond-
DS/JSkg-1SK-1

6 ing to the MT and the intermediate phase transition, respec-

tively, which are shown in Fig. 15. [32] As we know, two suc-
4 cessive MCE peaks with the same sign have been discov-
ered in some second-order phase transition materials, such as
2
PrNiCo5 and Ho2 In, [33,34] which enlarge the working temper-
0 ature range effectively. In the case of HPA Ni–Mn–Co–Sn al-
270 280 290 300 310 320 330 loy, two ∆SM peaks with the same sign are partly overlapping
T/K as well, giving rise to a wide refrigerant temperature interval
Fig. 13. Temperature dependence of ∆SM in a magnetic field of 10 kOe of about 67 K with appreciable MCE, which is indicated by
for Ni50−x Cox Mn39 Sb11 alloys. [13] the shaded area in Fig. 15. [32]

20
(a) 1.2
x/ 10 kOe HPA
x/
DS/JSkg-1SK-1

15 x/
0.9
DS/JSkg-1SK-1

10
0.6

5
0.3

0
210 225 240 255 270 0
220 240 260 280 300 320
T/K T/K
Fig. 15. Temperature dependence of ∆SM for HPA alloy in a magnetic
(b) x/
x/ field of 10 kOe. The shadow area indicates the refrigerant temperature
-0.3 x/ interval. [32]
DS/JSkg-1SK-1

4.2. Magnetoresistance effect

-0.6
The Ni–Mn–Z alloys undergo a magnetostructural tran-
sition from the high-temperature austenitic phase to the low-
-0.9
temperature martensitic phase with decreasing temperature.
Since the crystallographic symmetry changes through the MT,
-1.2
240 260 280 300 320 the band structure varies correspondingly, leading to different
T/K densities of state near the Fermi level and a sharp change in
Fig. 14. Temperature dependence of ∆SM in a magnetic field of 10 kOe resistivity near the MT temperatures.
for Ni43 Mn46−x Cux Sn11 (x = 1, 2, 3) around (a) MT temperature and
(b) TCA . [14] Figure 16 shows the temperature dependence of the elec-
trical resistivity for Ni45 Mn42 Cr2 Sn11 alloy at H = 0 and
Besides the magnetostructural transition in Ni–Mn-based 50 kOe on heating, in which a typical resistivity behavior
FSMAs, there is another second-order FM transition around of Ni–Mn-based FSMA is observed. [11] In the martensite
077506-6
 Chin. Phys. B Vol. 22, No. 7 (2013) 077506
state, the resistivity decreases with increasing temperature, netoresistance (MR) effect is expected. Figure 17 plots the
and around the MT temperature of 230 K, a sudden drop of the temperature dependence of MR for the high-Mn content al-
resistivity is observed. Further warming leads to the increase loys, Mn47+x Ni43−x Sn10 , at a magnetic field of 50 kOe. [36]
of resistivity in the austenite state. It seems that the resistiv- The maximum values of MR for x = 1, 2, and 3 alloys under
ity exhibits a semimetal-like behavior in the martensitic state, the magnetic field of 50 kOe are 23%, 25%, and 21%, respec-
while it is metallic for the austenitic state. As we know, the tively. The inset of Fig. 17 shows the magnetic field depen-
first-order MT in Ni–Mn-based FSMAs often develops with dence of MR for the Mn49 Ni41 Sn10 alloy at 235 K. The value
a region of metastability with austenite and martensite co- of MR decreases gradually with increasing magnetic field and
existing. With increasing temperature, especially around the becomes almost saturated under a field of 35 kOe. It is noted
MT temperature, more and more martensitic fractions in the that the MR remains almost unchanged as the magnetic field
sample are transformed into the austenitic phase. Since the decreases from 50 kOe to 0 Oe, implying that the alloy remains
martensite phase has higher resistivity as compared with the in a kinetically arrested austenite phase and the return transfor-
austenite phase, the overall resistivity decreases with increas- mation from austenite to martensite cannot be realized. [36]
ing temperature, showing the semimetal-like behavior around
the MT temperature despite the metallic nature of martensite 4.3. Exchange bias effect
and austenite states. [35] In addition, a downshift of 15 K in The exchange bias was discovered at the interface be-
the MT temperature is observed in Ni45 Mn42 Cr2 Sn11 alloy tween the FM Co core and the antiferromagnetic (AFM) CoO
at 50 kOe, suggesting that the transition from weak-magnetic shell by Meiklejohn and Bean in 1956. From then on, ex-
martensite to FM austenite can be realized in advance with a tensive investigations have been carried out on this subject in
sufficiently strong applied magnetic field. many systems for its applications in ultrahigh-density mag-
netic recording, giant magnetoresistance, sensors, and spin-
550
tronic devices. The exchange bias is usually described as the
0 Oe
50 kOe shift hysteresis loops along the field axis for coupled AFM–
FM systems cooled though the Néel temperature in the pres-
450
ence of a magnetic field, which is due to a unidirectional mag-
ρ/mWScm

netic anisotropy caused by the coupling between the AFM and

heating
the FM at the interface. To date, the exchange bias has been
350
investigated in many systems, such as small oxide particles,
multilayer films, and inhomogeneous materials.
For Mn rich Ni50 Mn25+x Sn25−x Heusler alloys, the ex-
250
140 180 220 260 cess Mn atoms occupy the Sn sites. In such spatial configu-
T/K rations, the Mn–Mn neighbors have a smaller separation than
Fig. 16. Temperature dependence of electrical resistivity in the mag- those in the stoichiometric compound, which would introduce
netic fields of 0 Oe and 50 kOe for Ni45 Mn42 Cr2 Sn11 alloy. [11]
an AF exchange in these off-stoichiometric alloys. [9] It is well
known that the coexistence of FM and AFM phases in the
0 martensitic phase and the coupling at the AFM/FM interface
result in the exchange bias behavior, which has been observed
x/
x/ in some Ni–Mn–Z alloys, such as Ni50 Mn25+x Sb25−x [37] and
x/ Ni50 Mn36 Sn14 . [38] However, the exchange bias properties in
-10 @ 50 kOe
MR/%

the alloys with high contents of Mn, especially when the con-
0 x/
@ 235 K tent of Mn is higher than that of Ni, have rarely been inves-
MR/%

-10
-20 tigated. In these alloys, the excess Mn atoms would occupy
-20 not only the Sn sites but also the Ni sites, and the moments of
0 20 40 Mn at the tetrahedral (Ni) sites are antiparallel to those of Mn
H/kOe
-30 atoms at the octahedral (Mn) sites. Through the MT, the Mn–
180 220 260 300
T/K Mn distance decreases due to the twinning of the martensitic
Fig. 17. Temperature dependence of MR for Mn47+x Ni43−x Sn10 (x = 1, phase, leading to the enhanced strength of the AF exchange
2, 3) alloys at a magnetic field of 50 kOe. Inset: field dependence of coupling in Ni–Mn–Sn alloys with high Mn content. [39]
MR for Mn49 Ni41 Sn10 alloy at 235 K. [36]
Figure 18 shows the temperature dependence of zero
Since the magnetic field can induce the magnetostruc- field cooled (ZFC) and field cooled (FC) magnetization curves
tural phase transition in Ni–Mn-based FSMAs, a large mag- for Mn50 Ni40 Sn10 alloy. [39] With decreasing temperature, the
077506-7
 Chin. Phys. B Vol. 22, No. 7 (2013) 077506
PM–FM transition occurs at a TCA of about 280 K. Then a sud- Mn50 Ni40 Sn10 alloy occupy both the Ni and Sn sites and their
den decrease in magnetization is observed, corresponding to moments are coupled antiferromagnetically to those of Mn
the transition from austenite to martensite. Further decreasing atoms on the regular Mn sites. After the MT, the AFM interac-
the temperature leads to an FM transition of the martensitic tion in the martensitic phase is enhanced due to the change in
phase. Below the Curie temperature of the martensite TCM , a the Mn–Mn distance, which is strong enough to pin the FM do-
splitting is observed between the ZFC and FC curves and be- mains. It is worth noting that, unlike the Ni–Mn based FSMAs
comes much more pronounced with decreasing temperature, with high Ni content, another AFM interaction between Mn
showing the coexistence of AFM and FM exchange interac- atoms at the Ni sites and the ones at the regular sites is intro-
tions in the martensite state. duced in the present alloy. Consequently, the AFM regions are
increased while the FM regions are reduced in Mn50 Ni40 Sn10
20 alloy, which give rise to the considerably large exchange bias
Mn50Ni40Sn10
effect. [39]
ZFC
15
FC
4.4. Magnetoelectric effect
M/emuSg-1

10 Magnetoelectric (ME) multiferroics, which combine FM

and ferroelectric properties in the same system, are particu-
5 M
TC TCA larly appealing for their abundant physical properties and po-
tential applications in devices, such as ME memories, trans-
0
ducers, actuators, and sensors. The ME effects can be classi-
0 100 200 300 fied as direct ME (DME) and converse ME (CME) effects,
T/K which are characterized as magnetic-field-induced polariza-
Fig. 18. ZFC and FC magnetization curves as a function of temperature tion and electric-field-induced magnetization, respectively. In
for Mn50 Ni40 Sn10 alloy under a magnetic field of 100 Oe. [39]
recent years, laminated ME composites have been extensively
investigated owing to their large DME or CME effect. [40]
2K The ME effect in these laminates is the result of the elas-
20 35 K
75 K
tic coupling between piezoelectric and magnetostrictive lay-
10 125 K ers, so most studies of the ME composites are focused on
M/emuSg-1

the Terfenol-D based laminates for their outstanding magne-

0 tostrictive property. [40] However, this rare-earth iron alloy is
quite expensive and a fairly large magnetic field is required
-10
to obtain a large ME in this composite. Therefore, a ME com-
posite made of a low-cost magnetic alloy and exhibiting a large
-20
ME effect is still greatly desired.
-4000 -2000 0 2000 4000
Ni–Mn-based FSMAs exhibit a large magnetostrain due
H/Oe
to the magnetic field-induced twin boundary motion or mag-
Fig. 19. Magnetic hysteresis loops for Mn50 Ni40 Sn10 alloy measured at
different temperatures. [39] netic field driven MT. [1–3,8] Based on this effect, these alloys
can be utilized as the FM layers of laminated composites to
Figure 19 shows the magnetic hysteresis loops of obtain a large DME effect. When an external magnetic field is
Mn50 Ni40 Sn10 alloy at different temperatures measured after applied on Ni–Mn-based FSMAs, a large strain is induced and
the FC process from 320 K. It is obvious that the magnetiza- then transferred as a stress in the piezoelectric layer. Finally,
tion curve at 2 K shifts completely to the negative field axis a voltage is produced in the ceramic due to the piezoelectric
with a considerably large exchange bias field (HE ) of 910 Oe. effect. This mechanism has produced large DME effects in
Here the value of HE is calculated using HE = −(H1 + H2 )/2, Ni2 MnGa/Pb(Zr0.52 Ti0.48 )O3 (PZT), and Ni–Mn–Co–Sn/PZT
where H1 and H2 are the left and right coercive fields, respec- laminates. [41,42]
tively. With increasing temperature, the value of HE decreases As for the CME effect in the laminates composed of mag-
and the shift in the hysteresis loop almost disappears at 75 K. netostriction alloy and piezoelectric material, the mechanical
Therefore, the temperature of 75 K can be referred as the ex- force (stress) generated from the ceramic is transferred to
change blocking temperature, TB , where the AFM cannot pin the magnetic alloy to induce a variation of magnetic flux by
the FM regions and the FM domains begin to rotate. Above the piezomagnetic effect. [43,44] It is known that, in Ni–Mn–
TB , an asymmetrical hysteresis loop is observed at 125 K. As Z alloys, the stress can be used as an external force to drive
mentioned above, the excess Mn atoms in high-Mn-content MT, leading to a sharp change of magnetization. Thus, a gi-
077506-8
 Chin. Phys. B Vol. 22, No. 7 (2013) 077506
ant CME effect can be expected in the FSMAs/piezoelectric 2.0
laminate. [45] In order to investigate the room-temperature

CME coefficient/GSV-1
293 K
CME effect in Ni–Mn–Z based composite, we first prepared a 1.0 288 K

Ni43 Mn41 Co5 Sn11 (Ni–Mn–Co–Sn) alloy, whose MT temper-

ature is just around 293 K, as shown in Fig. 20. The CME mea- 0
surement of the Ni–Mn–Co–Sn/PZT laminate was performed
by using an induction method. [44] During the measurement, -1.0
the DC magnetic bias field is applied in the plane of the lam-
inate, and two pickup coils are wrapped around the sample -2.0
to measure the time dependence of the magnetization in that -1000 -500 0 500 1000
direction. Figure 21 shows the frequency dependences of the Hbias/Oe

CME coefficients for the composite in different temperatures Fig. 22. CME hysteresis loops of the laminated composite at different
temperatures. [45]
under an optimum bias magnetic field (Hbias ) of 660 Oe and an
AC electric field of 80 V/mm. At 293 K, giant peak values of in Fig. 22. For T = 293 K, the CME coefficient increases with
CME coefficients are 1.26 G/V, 2.02 G/V, and 2.91 G/V at the the increasing Hbias and reaches up to the maximum value of
frequencies of 8.5 kHz, 52.0 kHz, and 91.5 kHz, respectively, 2.0 G/V at 655 Oe. As mentioned above, the MT can be in-
corresponding to the electromechanical resonance frequen-
duced by a magnetic field, i.e., Hbias . So the behavior of the
cies. It is observed that the temperature has an important in-
CME coefficient is determined by the cooperation of Hbias and
fluence on the CME effect, i.e., the values of CME coefficients
stress, and the peak at 293 K means that, with the assistance of
at 293 K are much larger than those at 288 K, which can be
Hbias of 655 Oe, the stress-induced change of magnetization is
attributed to the fact that 293 K is very close to the MT temper-
the largest. Further increasing Hbias leads to a decrease of the
ature and the stress induced magnetic changes are large there.
magnetic change. In the case of 288 K, the value of Hbias cor-
The Hbias dependence of CME coefficients at different temper-
responding to the maximum CME coefficient is 682 Oe, which
ature in an AC electric field of 80 V/mm and 54 kHz is shown
is somewhat higher than that in 293 K. [45]

40
Ni43Mn41Co5Sn11 5. Conclusion
30 Ni–Mn–Z (Z = In, Sn, Sb) Heusler alloys are newly dis-
covered FSMAs and have been extensively investigated owing
M/emuSg-1

20 heating 293 K to their abundant physical properties and promising applica-

tions in magnetic refrigeration, actuating devices, and sensors.
In this review, we have mainly summarized our recent research
10
288 K
results about the regulation of MT and the investigation of
some related magnetic effects in Ni–Mn–Z FSMAs.
0
200 240 280 320 The number of conducting electrons, the distance be-
T/K tween Mn atoms, and the relative stability between the parent
Fig. 20. Thermomagnetization curve for the Ni43 Mn41 Co5 Sn11 alloy and the martensitic phases are the main factors for determin-
under a magnetic field of 1 kOe. [45] ing the MT in Ni–Mn–Z alloys. Accordingly, some effective
methods, such as (i) tuning the composition, (ii) modifying
3.0 the Mn–Mn distance, and (iii) altering external and internal
293 K
CME coefficient/GSV-1

288 K stresses, are used to adjust the MT in Ni–Mn-based FSMAs.

Due to the special magnetic and crystallographic struc-
2.0
tures and the strong coupling between magnetism and struc-
ture, Ni–Mn–Z FSMAs exhibit a series of interesting mag-
1.0
netic properties, such as magnetocaloric, magnetoresistance,
and exchange bias effects. It is known that the MT of these
alloys can be driven by an external stress, leading to a large
0 change of magnetization. Based on this distinctive property,
0 20 40 60 80 100
giant CME effects are observed in Ni–Mn–Z alloys/PZT lam-
Frequency/kHz
inates, which extends the application area of these magnetic
Fig. 21. Frequency responses of CME coefficient at different tempera-
tures with Hbias of 660 Oe. [45] multifunctional materials.
077506-9
 Chin. Phys. B Vol. 22, No. 7 (2013) 077506
References [24] Liu Y and Favier D 2000 Acta Mater. 48 3489
[1] Murray S J, Moarioni M, Allen S M, O’Handley R C and Lograsso T [25] Ma S C, Xuan H C, Zhang C L, Wang L Y, Cao Q Q, Wang D H and
A 2000 Appl. Phys. Lett. 77 886 Du Y W 2010 Chin. Phys. B 19 117503
[2] Wang W H, Wu G H, Chen J L, Yu C H, Gao S X, Zhan W S, Wang Z, [26] Llamazares J L S, Sánchez T, Santos J D, Pérez M J, Sanchez M L,
Gao Z Y, Zheng Y F and Zhao L C 2000 Appl. Phys. Lett. 77 3245 Hernando B, Escoda L, Suñol J J and Varga R 2008 Appl. Phys. Lett.
[3] Sozinov A, Likhachev A A, Lanska N and Ullakko K 2002 Appl. Phys. 92 012513
Lett. 80 1746 [27] Hernando B, Llamazares J L S, Santos J D, Escoda L, Suñol J J, Varga
[4] Hu F X, Shen B G and Sun J R 2000 Appl. Phys. Lett. 76 3460 R, Baldomir D and Serantes D 2008 Appl. Phys. Lett. 92 042504
[5] Sutou Y, Imano Y, Koeda N, Omori T, Kainuma R, Ishida K and [28] Xuan H C, Xie K X, Wang D H, Han Z D, Zhang C L, Gu B X and Du
Oikawa K 2004 Appl. Phys. Lett. 85 4358 Y W 2008 Appl. Phys. Lett. 92 102503
[6] Krenke T, Duman E, Acet M, Wassermann E F, Moya X, Mañosa L and [29] Planes A, ObradóE, Gonzàlez-Comas A and Mañosa L 1997 Phys. Rev.
Planes A 2005 Nat. Mater. 4 450 Lett. 79 3926
[7] Yu S Y, Liu Z H, Liu G D, Chen J L, Cao Z X, Wu G H, Zhang B and [30] Gonzàlez-Comas A, ObradóE, Mañosa L, Planes A, Chernenko V A,
Zhang X X 2006 Appl. Phys. Lett. 89 162503 Hattink B J and Labarta A 1999 Phys. Rev. B 60 7085
[8] Kainuma R, Imano Y, Ito W, Sutou Y, Morito H, Okamoto S, Kitakami [31] Cui Y T, Chen J L, Liu G D, Wu G H and Wang W L 2004 J. Phys.:
O, Oikawa K, Fujita A, Kanomota T and Ishida K 2006 Nature 439 957 Condens. Matter 16 3061
[9] Planes A, Mañosa L and Acet M 2009 J. Phys.: Condens. Matter 21
[32] Ma S C, Xuan H C, Zhang C L, Wang L Y, Cao Q Q, Wang D H and
233201
Du Y W 2010 Appl. Phys. Lett. 97 052506
[10] Mañosa L, Moya X, Planes A, Gutfleisch O, Lyubina J, Barrio M,
Tamarit J L, Aksoy S, Krenke T and Acet M 2008 Appl. Phys. Lett. [33] Zhang Q, Cho J H, Li B, Hu W J and Zhang Z D 2009 Appl. Phys. Lett.
92 012515 94 182501
[11] Wang D H, Zhang C L, Han Z D, Xuan H C, Gu B X and Du Y W 2008 [34] Rocco D L, Amaral J S, Leitão J V, Amaral V S, Reis M S, Fernandes
J. Appl. Phys. 103 033901 R P, Pereira A M, Araújo J P, Martins N V, Tavares P B and Coelho A
[12] Han Z D,Wang D H, Zhang C L, Xuan H C, Gu B X and Du Y W 2007 A 2009 Phys. Rev. B 79 014428
Appl. Phys. Lett. 90 042507 [35] Xuan H C, Deng Y, Wang D H, Zhang C L, Han Z D and Du Y W 2008
[13] Han Z D, Wang D H, Zhang C L, Xuan H C, Gu B X and Du Y W 2008 J. Phys. D: Appl. Phys. 41 215002
J. Appl. Phys. 104 053906 [36] Xuan H C, Zheng Y X, Ma S C, Cao Q Q, Wang D H and Du Y W
[14] Wang D H, Zhang C L, Xuan H C, Han Z D, Zhang J R, Tang S L, Gu 2010 J. Appl. Phys. 108 103920
B X and Du Y W 2007 J. Appl. Phys. 102 013909 [37] Khan M, Dubenko I, Stadler S and AliN 2007 Appl. Phys. Lett. 91
[15] Han Z D, Chen X, Zhang Y, Chen J, Qian B, Jiang X F, Wang D H and 072510
Du Y W 2012 J. Alloy. Compd. 515 114 [38] Li Z, Jing C, Chen J P, Yuan S J, Cao S X and Zhang J C 2007 Appl.
[16] Aksoy S, Krenke T, Acet M, Wassermann E F, Moya X, Mañosa L and Phys. Lett. 91 112505
Planes A 2007 Appl. Phys. Lett. 91 241916 [39] Xuan H C, Cao Q Q, Zhang C L, Ma S C, Chen S Y, Wang D H and Du
[17] Han Z D, Wang D H, Zhang C L, Xuan H C, Zhang J R, Gu B X and Y W 2010 Appl. Phys. Lett. 96 202502
Du Y W 2009 Mater. Sci. Eng. B 157 40
[40] Nan C W, Bichurin M I, Dong S X, Viehland D and Srinivasan G 2008
[18] Shen J, Gao B, Zhang H W, Hu F X, Li Y X, Sun J R and Shen B G J. Appl. Phys. 103 031101
2007 Appl. Phys. Lett. 91 142503
[41] Zhao K, Chen K, Dai Y R, Wan J G and Zhu J S 2005 Appl. Phys. Lett.
[19] Xuan H C, Wang D H, Zhang C L, Han Z D, Gu B X and Du Y W 2008
87 162901
Appl. Phys. Lett. 92 242506
[20] Nayak A K, Suresh K G, Nigam A K, Coelho A A and Gama S 2009 [42] Chen S Y, Ye Q Y, Miao W, Wang D H, Wan J G, Liu J M, Du Y W,
J. Appl. Phys. 106 053901 Huang Z G and Zhou S Q 2010 J. Appl. Phys. 107 09D901
[21] Mandal K, Pal D, Scheerbaum N, Lyubina J and Gutfleisch O 2009 J. [43] Jia Y M, Or S W, Chan H L W, Zhao X Y and Luo H S 2006 Appl.
Appl. Phys. 105 073509 Phys. Lett. 88 242902
[22] Lin H C, Wu S K, Chou T S and Kao H P 1991 Acta Metall. Mater. 39 [44] Zhou J P, Guo Y Y, Xi Z Z, Liu P, Lin S Y, Liu G and Zhang H W 2008
2069 Appl. Phys. Lett. 93 152501
[23] Piao M, Otsuka K, Miyazaki S and Horikawa H 1993 Mater. Trans. [45] Chen S Y, Wang D H, Han Z D, Zhang C L, Du Y W and Huang Z G
JIM 34 919 2009 Appl. Phys. Lett. 95 022501

077506-10