Direct measurements of the magnetocaloric effect in
of Heusler alloys Ni - Mn - M (M = In, Sn)
A. M. Alieva,*, A. B. Batdalova,
V. V. Koledovb, V. G. Shavrovb,
V. D. Buchelnikovc,
J. Garcíad, V. M. Pridad, B. Hernandod
(a) Amirkhanov Institute of Physycs of Daghestan Scientific Center, RAS,
Makhachkala 367003, Russia
(b) Kotelnikov Institute of Radio Engineering and Electronics, RAS, Moscow
(c) Chelyabinsk State University, 454001 Chelyabinsk, Russia
(d) Depto. de Física, Facultad de Ciencias, Universidad de Oviedo, Calvo Sotelo
s/n, 33007 Oviedo, Spain
Direct measurements of the magnetocaloric effect in samples of rapidly quenched
ribbons of Mn50Ni40In10 and Ni50Mn37Sn13 Heusler alloys, with potential applications in
magnetic refrigeration technology, are carried out. The measurements were made by a precise
method based on the measurement of the oscillation amplitude of the temperature in the
sample while is subjected to a modulated magnetic field. In the studied compositions both
direct and inverse magnetocaloric effects associated with magnetic (paramagnet - ferromagnet
- antiferromagnet) and structural (austenite - martensite) phase transitions are found.
Additional inverse magnetocaloric effects of small value are observed around the
In recent years the magnetic compounds with significant magnetocaloric effect (MCE)
near room temperature are intensively studied. On the basis of these studies the possibility of
ecological, economically viable solid-state refrigerators is proposed . Recently, among the
promising materials for magnetic refrigeration, a considerable interest is attracted to the
Heusler alloys family Ni-Mn-X (X = Ga, Sn, Sb, In), in which a giant MCE is found [2, 3].
These alloys are characterized by the fact that near room temperature they exhibit a sequence
of magnetic (paramagnetic - ferromagnetic - antiferromagnetic) and structural (austenite-
martensite) phase transitions. In some cases, the magnetic and structural transitions merge
into magnetostructural transitions. In the region of second order PM-FM phase transition, near
TC, the direct MCE, while at the martensitic transition of first order and near the
metamagnetic transition of the FM-AFM, as well as merged metamagnetostructural transition,
the inverse MCE is observed. The inverse MCE in bulk Heusler alloys samples is studied in
In the most of MCE studies, indirect methods based on the measurement of isothermal
magnetization curves are used. However, in many cases the use of indirect methods for
obtaining data on magnetocaloric properties of the materials is insufficiently substantiated.
This particularly concerns to the cases of first order magnetic phase transitions and
magnetostructural transitions [7-9]. Therefore, it needs to develop direct methods for studying
magnetocaloric properties, with high sensitivity, simplicity and adaptability.
In the present work, the MCE is studied in melt spun ribbons Ni-Mn-M (M = Sn, In)
Heusler alloys by the direct technique, in modulating magnetic fields of small-amplitude, as it
has been proposed in . The use of the magnetocaloric materials in the form of thin films
or ribbons in the manufacture of refrigerators can optimize the heat transfer between the
working body and heat-exchange fluid, and thus improving the technical characteristics of
refrigeration unit. However, it is difficult to investigate MCE in small samples by indirect
methods, and the classical direct measurements are impossible to be conducted. The proposed
method allows a high precision investigation of the magnetocaloric properties of samples with
small size and mass performed in weak magnetic fields. The measurements were performed at
frequencies ranging 0.3-0.5 Hz. The magnetic field during the experiment was always
directed along the plane of the sample. The heat capacity was measured by an AC-
calorimeter. Gd was used as the test sample for the calibration of the set-up.
With the help of the technique, we have investigated the MCE in Ni50Mn37Sn13 and
Mn50Ni40In10 Heusler alloys. Samples with typical dimensions 1x3x0.015 mm3 were cut from
ribbons obtained by rapid quenching from the melt. Ribbons have a textured microcrystalline
structure, with elongated column grains perpendicularly oriented to the ribbon plane [11, 12].
According to , the Mn50Ni40In10 sample transforms at 311 K into the ferromagnetic
state. The martensitic phase transition starting and finishing temperatures are MS = 213 K, Mf
= 173 K, respectively, whereas the corresponding ones to the austenite transition are AS = 222
K, and Af = 243 K, respectively. The Ni50Mn37Sn13 sample transforms into the ferromagnetic
state at TC = 313 K, and the martensitic transformation and austenite transition temperatures
are MS = 218 K, Mf = 207 K and AS = 224 K, Af = 232 K, respectively .
It is worthwhile mentioning that characteristic features of magnetostructural
transitions in Heusler alloys quite clearly reveal themselves in the magnetization and
susceptibility behavior under weak magnetic fields [11 - 13]. At the same time, the MCE
study is usually carried out in strong magnetic fields, when the mentioned features are
suppressed. Concerning with this, the MCE investigation in weak magnetic fields is of
particular interest for studying the nature of phase transitions in Heusler alloys.
Fig. 1 shows the temperature dependence of the specific heat for Ni50Mn37Sn13 in a
zero magnetic field and in the field of 11 kOe. High-temperature specific heat anomaly
displays a maximum at T = 309.5 K, corresponding to the FM-PM phase transition, which
takes place in the austenite phase. At temperatures of 217 K (cooling mode) and 233 K
(heating mode) a second anomaly in the heat capacity corresponding to a first order
metamagnetostructural phase transition is observed. The heat jump ΔC is different for both
regimes and correlates with the magnetization behavior . The presence of strongly
pronounced hysteresis (ΔT = 16K) indicates the phase transition as a first order one.
Fig. 2 shows the temperature dependence of MCE in Ni50Mn37Sn13 alloy measured
with an amplitude of ΔH = 500 Oe for the modulated magnetic field. MCE shows a maximum
at T = 312.5 K, near TC, while around the martensitic transition temperature – the inverse
MCE exhibits two maxima: around T = 224 K (cooling mode) and at T = 236 K (heating
mode). In addition, in the temperature range 284-302 K, below the ferromagnetic transition
there is another peak of inverse MCE with a maximum at T = 297 K. All these features
indicate the existence of a rather complicated picture of the MCE. As it is shown in  the
temperature dependence of the low-field magnetic susceptibility of melt spun ribbons has two
peaks. Such magnetization behavior can be explained as follows. As is known, the low-field
susceptibility is caused by process of domain walls displacements . Susceptibility is
directly proportional to the saturation magnetization and inversely proportional to the
anisotropy of the magnetic substance [14, 15]. When approaching the Curie point the degree
of reduction of the anisotropy intensity is higher than the corresponding to the saturation
magnetization. In this regard, the susceptibility near Curie point shows a peak (Hopkinson
effect). In the low-field range, magnetization displays a second peak that can be explained by
the model proposed in  for Heusler alloys, where there is a possibility of the sign
inversion of the exchange interaction. According to , in Heusler alloys, the phase
transition from ferromagnetic phase to the antiferromagnetic (or paramagnetic) phase takes
place accompanied by structural changes. In this case, a decrease in magnetization at the
transition from cubic austenite phase to the martensitic phase is due to the simultaneous FM-
AFM transition. It is also possible, that such a decrease in magnetization may be due to the
fact that in Ni-Mn-X Heusler alloys the structural phase transition that is observed at low
temperatures is accompanied by the origin of AFM ordering of manganese atoms in the
ferromagnetic matrix (i.e. the mixed ferro-antiferromagnetic phase) in the martensitic phase
Clearer picture of direct and inverse MCE is observed in Mn50Ni40In10 (Fig. 3). Near
TC, direct MCE peak at T = 314 K, and around the martensitic phase transition temperatures
two inverse MCE peaks are observed. As can be seen, similarly to the case for Ni50Mn37Sn13,
when the temperature goes down, in the austenite phase near the PM-FM transition there is a
direct MCE. At further cooling the first order structural austenite - martensite phase transition
is observed with the second negative peak (TM = 201K), which is a consequence of the
appearance of antiferromagnetic interactions in the martensite phase. In heating mode, the
maximum of inverse MCE is shifted toward higher temperatures (TM = 237K) in accordance
with the structural transition temperature shift. Inverse MCE is manifested at the structural
phase transition austenite - martensite, due to the coexistence in the martensitic phase of FM
and AFM interactions between manganese atoms [2, 15, 17]. The peak value of ΔT at the
direct martensitic transformation temperature, ΔTcooling = 0.00166 K (for magnetic field
change ΔH=370 Oe), is less than the value at the reverse transformation, ΔTheating=0.0042 K.
This difference was to be expected because the magnetization change rate is larger upon the
reverse martensitic transformation (insert in Fig. 3), though the magnetization change ΔM is
larger upon the direct transformation .
Fig. 3 shows that near the magnetic phase transition in the austenite phase there is a
peak of inverse MCE, despite the absence of any features on the low-field magnetization in
Mn50Ni40In10 ribbons at this temperature range . The observed inverse MCE may be
probably due to an additional phase shift, as a consequence of the presence in the austenitic
phase, of both ferromagnetic and antiferromagnetic exchange, simultaneously. Such
antiferromagnetic exchange is due to the additional number of Mn atoms in the crystal
structure in comparison with the stoichiometric composition Ni2MnX [2, 17]. To determine
the nature of inverse MCE above the Curie temperature of austenitic phase further careful
measurements of magnetic properties in Mn50In40In10 melt spun ribbons are required.
Simple linear extrapolation of our low-fields data yields 0.24 K and 0.41 K change in
temperature ΔT with 1 T magnetic field variation for Mn50Ni40In10 and Ni50Mn37Sn13
respectively. More realistic value of the MCE can be extract from a field dependence of MCE
(Fig. 4). The experimental data can be fitted by expression ΔT=aHn , where n=0.88. From
this follow ΔT=0.16 K at field change of 1 T. As we can see, the magnitude of MCE in the
melt spun ribbons is small. Nevertheless, melt spun ribbons can be effectively employed in
magnetic cooling devices through more efficient heat transfer. One can expect that in
magnetic refrigerators the thermodynamic efficiency will be greatly enhanced by increasing
the frequency of cycles. It is expected that films and melt-spun ribbons having faster heat
transfer as compared with bulk samples, would be preferable. Thus, Heusler alloys in ribbon-
shape samples, having the optimal combination of thermodynamical and magnetic
characteristics, can compete with other magnetic coolant materials.
In conclusion, this paper reports on the application of a MCE measurement method
using modulated magnetic field, for the direct MCE investigation in samples of melt-spun
Heusler alloys ribbons, with small mass and thickness. The findings point out the complex
nature of structural and magnetic phase transition in these alloys, which needs further
This work was supported by RFBR (09-08-96533, 09-08-01177, 10-02-92662), FANI
(02.513.12.3097), Research Program of DPS RAS, MAT2009-13108-C02-01 and FC09-
1. A. M. Tishin and Y. I. Spichkin. 2003, The Magnetocaloric Effect and its
Applications, Institute of Physics, New York.
2. A. Planes, L. Manosa and M. Acet. J. Phys.: Condens. Matter 21, 233201 (2009).
3. A.K. Pathak, I. Dubenko, J.C. Mabon Sh. Stadler and N. Ali. J. Phys. D: Appl. Phys.
42, 045004 (2009).
4. T. Krenke, E. Duman, M. Acet et al. Nat. Mater. 4, 450 (2005).
5. V.V. Khovaylo, K.P. Skokov, O. Gutfleisch, H. Miki, T. Takagi, T. Kanomata, V.V.
Koledov, V.G. Shavrov, G. Wang, E. Palacios, J. Bartolomé, and R. Burriel. Phys.
Rev. B 81, 215506 (2010).
6. V.K. Sharma, M.K. Chattopadhyay, R. Kumar, T. Ganguli, P. Tiwari and S.B. Roy. J.
Phys.: Cond. Matter. 19, 496207 (2007).
7. J. S. Amaral and V. S. Amaral. J. Appl. Phys. Lett. 94, 042506 (2009).
8. M. Balli, D. Fruchart, D. Gignoux and R. Zach. Appl. Phys. Lett. 95, 072509 (2009).
9. D. Bourgault, J. Tillier, P. Courtois, D. Maillard, and X. Chaud. Appl. Phys. Lett. 96,
10. A.M. Aliev, A.B. Batdalov, V.S. Kalitka. JETP Letters 90, 663–666 (2009).
11. J. D. Santos, T. Sanchez, P. Alvarez, M.L. Sanchez, J.L. Sánchez Llamazares, B.
Hernando, Ll. Escoda, J.J. Suñol, and R. Varga. J. Appl. Phys. 103, 07B326 (2008).
12. J.L. Sánchez Llamazares, T. Sanchez, J.D. Santos, M.J. Pérez, M.L. Sanchez, B.
Hernando, Ll. Escoda, J. J. Suñol, and R. Varga. Appl. Phys. Lett. 92, 012513 (2008).
13. I. Babita, S.I. Patil and S. Ram. J. Phys. D: Appl. Phys. 43, 205002 (2010).
14. S.V. Vonsovskiy. Magnetism. Moscow. Nauka. 1971. 1032 p. (in Russian).
15. Kaiyuan He, Hui Xu, Zhi Wang, Lizhi Cheng. J. Mater. Sci. Technol. 16, 145 (2000).
16. V.D. Buchelnikov, S.V. Taskaev, M.A. Zagrebin, V.V. Khovailo, P. Entel. JMMM
320, e175-e178 (2008).
17. P.J. Brown, A.P. Gandy, K. Ishida, R. Kainuma, T. Kanomata, K-U. Neumann, K.
Oikawa, B. Ouladdiaf and K.R. A Ziebeck. J. Phys.: Cond. Matter 18, 2249 (2006).
Figure 1. Specific heat of Ni50Mn37Sn13 at heating and cooling and in magnetic field of 11 kOe.
Figure 2. MCE in Ni50Mn37Sn13 at heating and cooling at magnetic field change of 500 Oe.
Figure 3. MCE in Mn50Ni40In10 at heating and cooling at magnetic field change of 370 Oe. Inset –
dM/dT at heating and cooling run.
Figure 4. Field dependence of MCE in Mn50Ni40In10.