CC BY
49
Chelyabinsk Physical and Mathematical Journal. 2020. Vol. 5, iss. 2. P. 186-193.
DOI: 10.24411/2500-0101-2020-15205
MAGNETOCALORIC EFFECT IN Niso.sMnse.sSni^
RIBBON HEUSLER ALLOY IN CYCLIC MAGNETIC FIELDS
A. M. Aliev1", A.B. Batdalov1b, V.V. Koledov2,c, V.G. Shavrov2d,
B. Hernando3,e
1 Amirkhanov Institute of Physics, Dagestan Federal Research Centre,
Russian Academy of Sciences, Makhachkala, Russia 2Kotelnikov Institute of Radio Engineering and Electronics,
Russian Academy of Sciences, Moscow, Russia 3Oviedo University, Oviedo, Spain
"[email protected], [email protected], [email protected], [email protected], e [email protected]
Magnetocaloric properties in ribbon sample of Ni50.3Mn36.5Sn13.2 Heusler alloy are studied by a direct method in cyclic magnetic fields. A strong anisotropy in the magnetocaloric effect is observed in weak magnetic fields, obtaining a large value of MCE when the field is applied parallel to the ribbon plane. The field dependence of the MCE anisotropy is studied and its decrease is clearly revealed in moderate fields. It was found that in cyclic magnetic fields, the MCE value near the magnetostructural phase transition depends on the rate of temperature scanning: a higher rate of the temperature sweep results in greater value of the MCE.
Keywords: magnetocaloric effect, cyclic magnetic field, Heusler alloy, ribbon, anisotropy.
1. Introduction
The interest in the study of materials with the magnetocaloric effect (MCE) is due to the possibility of using them in prospective solid-state magnetic refrigeration technologies. The search for materials with necessary magnetocaloric properties is actually in progress by many research groups in the world [1; 2]. In recent years, much attention is paid to the study of materials with coupled first-order magnetostructural phase transitions (FOMPT). In this case, the total change of entropy is equal to AS = ASs + ASm where ASs is the entropy change due to the structural transition and ASm is the magnetic entropy change. This can give rise to a giant MCE in these materials. For instance, a giant MCE was observed in Ni-Mn-In-(Co) alloy, where the adiabatic temperature change is more than 6 K for p0AH = 1.9 T [2]. But this and similar studies are performed under conditions that significantly differ from those that will be in magnetic refrigerators. In real refrigerators the magnetic material will be subjected to an alternating magnetic field. To improve the efficiency of magnetic cooling, it is needed to use alternating magnetic fields of a high frequency. The frequencies of thermomagnetic cycles, in turn, are limited by the rate of the heat transfer in the material [3]. To increase the rate of the heat transfer, therefore, it is necessary to increase the surface area of the working body, which can be done by making it in the form of a battery of thin plates.
The study was supported by a grant of the Russian Science Foundation, project no. 18-12-00415. Research was carried out using the equipment available at the Analytical Center for collective use of the Dagestan Federal Research Centre, Russian Academy of Sciences.
Magnetocaloric effect in Nis^M^^sSn^^ ribbon Heusler alloy in cyclic magnetic fields 187
However, strong anisotropy effects of MCE can occur in thin samples of alloys produced by different technologies, for example by rapid cooling in the form of melt spun ribbons. The effect of thermomagnetic relaxation processes on the magnetocaloric properties of magnetic materials in alternating magnetic fields can also occur. Particularly, these effects can be strong in the vicinity of the FOMPT. All these peculiarities should be taken into account in designing the prototypes of magnetic refrigerators.
Heusler alloys Ni-Mn-X (X = Ga, In, Sn, Sb) are promising materials for their use in magnetic refrigerators. First, in these alloys, large values of MCE at FOMPT are observed [4-7]. Second, there is an established technology for producing high quality ribbons (thickness is 10-20 microns) of these alloys, with rather high values of MCE [8-11].
2. Results and Discussion
The aim of this paper is to study the anisotropy and time dependence of MCE in ribbon samples of a Heusler alloy in alternative magnetic fields of low frequencies. The use of a direct method is necessary for the correct estimation of the MCE value at MSPT, since the use of indirect methods around first-order transitions can give untrue, often overestimated values. For example, the giant MCE reported in [12-14] was not further confirmed by direct measurements [15]. In addition, at present, except for the direct method proposed by authors, there are no other methods to measure the MCE in alternating magnetic fields [16]. The frequency of the cycles of the magnetic field used in our experiments is less than 1 Hz, which is, of course, much lower than the operating frequency of the prototypes of magnetic refrigerators, but such studies can indicate trends in the behavior of the magnetocaloric effect at higher frequencies.
In the present work, the MCE is studied in melt-spun ribbon of Ni-Mn-Sn Heusler alloys by direct technique, in a low (up to 4 kOe) and a moderate (18 kOe) magnetic field by the modulation method [16]. The frequency of the alternating magnetic field in the experiment was 0.3 Hz. The alternating magnetic field was applied to the sample during the experiment. This method has been already tested on many material samples, including Gd, manganites and Heusler alloys [16-19]. The specific heat was measured by an ac-calorimetery technique. Samples with 1 x 3 x 0.015 mm3 dimensions 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 [9].
According to the magnetization data (Fig. 1a) Curie point occurs at 311 K. FOMPT begins at MS = 256 K, and ends at Mf = 252 K. Corresponding temperatures in the austenite phase are AS = 275 K, and Af = 278 K, respectively. With further cooling, the Ni50.3Mn36 5Sn132 undergoes a phase transition from low to high magnetization state in the martensite phase. The application of a 20 kOe magnetic field shifts the temperature of FOMPT to lower temperatures, and the temperature ferromagnetic-paramagnetic transition to a higher one.
Fig. 1b shows the temperature dependence of the specific heat for Ni50.3Mn36 5Sn13 2 in a zero magnetic field and in the magnetic field of 18 kOe. High-temperature specific heat anomaly displays a maximum at T = 312.5 K, corresponding to the ferromagnetic-paramagnetic phase transition, which takes place in the austenite phase. At temperatures of 250 K in cooling run and 273.5 K in heating one, a second anomaly in the specific heat corresponding to the FOMPT is observed. The width of the hysteresis is about 16 K.
188
A.M. Aliev, A.B. Batdalov, V.V. Koledov, V.G. Shavrov, B. Hernando
0.016
0.008
0.004
o.ooo
- —•— H=1 kOe —O— H=20 kOe ; 0.9 - # * •• /•
- 3 0.8
- 1\ d
1 . 1 1 1 . 1 1 1 . 1 1
100 150
300 350 40
220
240
260
280 T (K)
300
320
340
200 250
H (Oe)
Fig. 1. a) Magnetization M(T) of Ni50.3Mn36.5Sn13.2 in heating and cooling runs in magnetic fields of 1 and 20 kOe; b) Specific heat of Ni50.3Mn36.5Sn13.2 at heating and cooling (zero-field) and in a magnetic field (18 kOe) in heating run (red line)
To study the anisotropy of the MCE, measurements were performed in two geometries: 1 — the ribbon plane being perpendicular to the magnetic field, and 2 — the ribbon plane being parallel to the magnetic field. All other conditions were identical and the same differential thermocouple was used in both experiments. Figure 2 shows the temperature dependence of the adiabatic temperature change ATad in Ni50.3Mn36.5Sni3.2 when H is applied parallel to the ribbon plane. There are a direct MCE due to the PM-FM phase transition, and inverse MCE at FOMPT. Thermal hysteresis is observed in the magnetostructural transition region. The maximum value of the direct effect is equal to ATad = 0.6 K. An abrupt change in sign of the MCE is observed at magnetostructural phase transition, being the value of the inverse effect much smaller than the corresponding to the direct one, either in heating and cooling runs. At a further cooling of the martensite phase, again a direct magnetocaloric effect is observed, with a maximum at T = 240 K. This effect is due to the transition to the ferromagnetic state in the martensite phase.
It is known that usually thin magnetic films exhibit anisotropy associated with the shape of the sample: the magnetization along the plane of the sample is much higher than in the perpendicular direction of magnetization due to the shape anisotropy [20], and with a field increasing the anisotropy decreases.
It would be reasonable to expect that the same effect is inherent to MCE.
Indeed, the anisotropy of the MCE is observed in the studied sample, but in a field of 18 kOe, this effect is very weak (see inset in Fig. 2, shown for the heating run only). Stronger anisotropy of MCE appears in weak fields (Fig. 3), and the anisotropy is observed in the whole temperature range in which the MCE occurs. The MCE value is larger when the magnetic field is applied along the plane of the sample. In addition, in weak applied fields a more significant difference between the values of MCE at FOMPT in heating and cooling runs is observed.
cooling runs at magnetic field change of 18 kOe. Inset — anisotropy of MCE
Magnetocaloric effect in Ni50.3Mn36.5Sni3.2 ribbon Heusler alloy in cyclic magnetic fields 189
In view of the possible use of Heusler alloys in magnetic refrigerators the study of the temporal dependence of the magnetocaloric effect is a very interest issue. In Fig. 4 the temperature dependences of the MCE in the heating run at three different heating rates are shown.
In the FM-PM magnetic phase transition region, it is seen that the MCE value the same at all modes. Nevertheless, there is a strong dependence of the MCE on the heating rate around the FOMPT. At a sample heating rate of 3.6 K/min, the inverse MCE value is equal to ATad = "0.17 K, and the inverse and direct effects are nearly equals in modulus. At 2.2 K/min rate of temperature scan, the maximum value of the effect is equal to A Tad = "0.07 K, and at 0.6 K/min the effect is only A Tad = "0.027 K. The longer time the sample is in the transition area, the MCE is smaller. In addition, the temperature of the maximum of MCE is shifted to higher temperatures with the rate of temperature scan. This may be partly due to the fact that the lock-in amplifier that is used to measure the differential thermocouple signal has no time to instantly adjust to the maximum useful signal, as a consequence that in this region there is a strong phase change between the variation of the sample temperature and the applied magnetic field. But the increase of the MCE with rate of temperature scan is not explained by the instrumental effect. In this case, the measured signal should be less because the lock-in amplifier does not have time to tune the true value of MCE. Further confirmation of the truth of these results is the measurement of the MCE at the same conditions around the FM-PM transition, where a time dependence of MCE is not observed.
The observed behavior significantly differs from published reports of time dependence of MCE [21-24]. In general, these results are related to the so-called magnetic field first application effect, when a large MCE near the MSPT originates. The first application of a magnetic field results in an irreversible phase transition. With further magnetic field on/off cycling the MCE decreases and becomes similar to the normal time-independent MCE. The irreversible character of the magnetic field-induced martensitic transformation makes a sensitive contribution of the structural subsystem to ATad to be only appreciated during the first application of the magnetic field. Thereafter, the behavior of A Tad is essentially determined by the magnetic subsystem only. Combination of these factors results in the observed behavior of the adiabatic temperature change during subsequent actions of the magnetic field, i. e., becomes reversible and typical of conventional MCE. In contrast to these findings [21-24], in our experiment we did not study the effect of the first run, as the alternating magnetic field is applied to the sample during all the experiment (as it would be in real magnetic refrigerators), and during the experiment there are thousands of field on/off cycles. Since the contribution of the structural subsystems only appears at the first field application, but the MCE value continues to decrease, we are actually observing some relaxation processes in the subsystem. It is appropriate to compare these times (tens of seconds) with relaxation times in FeRh [25], where the magnetization changes slightly over time more than 103 s.
at magnetic field changes of 3 kOe applied perpendicular and parallel to the ribbon plane
190
A.M. Aliev, A.B. Batdalov, V.V. Koledov, V.G. Shavrov, B. Hernando
The reason for the unusual behavior of the MCE, in the studied Heusler alloy, can be its independence of either the external magnetic field long-time relaxation processes, and/or relaxation processes that take place under the influence of an alternating magnetic field. Effect of alternating magnetic fields on the magnetostructural phase transitions is little studied, and further researches are needed.
Perhaps the long-time relaxation processes are characteristic to all materials, although in bulk materials the characteristic relaxation times can be of tens of hours, and in thin films and ribbons, as we have seen, these times are tens of seconds. It was recently found, that continuously application of cyclic magnetic filed within the temperature hysteresis loop in Ni47Mn4oSni2.5Cuo.5 Heusler alloy results in «inverse MCE — direct MCE» crossover due to irreversible martencite-austenite transition [26]. The observed in weak fields in the Ni5o.3Mn36.5Sni3.2 Heusler alloy cannot be explained in a similar way, since the applied weak magnetic fields cannot cause magnetostructural phase transitions. Therefore, the observed phenomenon of the dependence of the MCE on the temperature sweeping rate can be due to relaxation processes.
3. Conclusion
In conclusion, the study of relaxation processes may provide the key to understanding the difference of MCE in the heating and cooling runs observed in Heusler alloys [24]. This difference may be due to various relaxation processes in these protocols, even if the same rate of the temperature is scanned. Thus, research of magnetocaloric properties of materials in the real conditions, in which they will be used in actual magnetic refrigerators, need to discuss about the prospects of the practical use of these materials in magnetic refrigeration technology. The results of this work show that the relaxation processes and anisotropy effects of MCE need to be taken into account at the design of magnetic refrigerators to get the highest cooling power efficiency of the machine.
References
1. PecharskyV.K., Jr GschneidnerK.A., MudrykY., PaudyalD. Making the most of the magnetic and lattice entropy changes. Journal of Magnetism and Magnetic Materials, 2009, vol. 321, pp. 3541-3547.
2. LiuJ., GottschallT., SkokovK.P., MooreJ.D., GutfleischO. Giant magnetocaloric effect driven by structural transitions. Nature Materials, 2012, vol. 11,
pp. 620-626.
3. SilvaD.J., BordaloB.D., PereiraA.M., VenturaJ., AraujoJ.P. Solid state magnetic refrigerator. Applied Energy, 2012, vol. 93, pp. 570-574.
4. AlievA., BatdalovA., BoskoS., BuchelnikovV., Dikshteinl., KhovailoV., Koledov V., Levitin R., Shavrov V., Takagi T. Magnetocaloric effect and magnetization in a Ni-Mn-Ga Heusler alloy in the vicinity of magnetostructural
at 272 K
Magnetocaloric effect in Ni50.3Mn36.5Sn13.2 ribbon Heusler alloy in cyclic magnetic fields 191
transition. Journal of Magnetism and Magnetic Materials, 2004, vol. 272-276, pp. 20402042.
5. SharmaV.K., Chattopadhyay M.K., Kumar R., GanguliT., TiwariP.,
RoyS.B. Magnetocaloric effect in Heusler alloys Ni50Mn34In16 and Ni50Mn34Sn16. Journal of Physics: Condensed Matter, 2007, vol. 19, p. 496207.
6. NayakA.K., Sureshand K.G. NigamA.K. Giant inverse magnetocaloric effect near room temperature in Co substituted NiMnSb Heusler alloys. Journal of Physics D: Applied Physics, 2009, vol. 42, p. 035009.
7. Basso V., SassoC.P., SkokovK.P., GutfleischO., KhovayloV.V. Hysteresis and magnetocaloric effect at the magnetostructural phase transition of Ni-Mn-Ga and Ni-Mn-Co-Sn Heusler alloys. Physical Review B, 2012, vol. 85, p. 014430.
8. Hernando B., Sanchez-Llamazares J.L., Santos J.D., EscodaLl., SunolJ.J., VargaR., BaldomirD., SerantesD. Thermal and magnetic field-induced martensite-austenite transition in Ni50.3Mn35.3Sn14.4 ribbons. Applied Physics Letters, 2008, vol. 92, p. 042504.
9. Santos J.D., Sanchez T., Alvarez P., Sanchez M.L., Sanchez-Llamazares J.L., HernandoB., EscodaLl., SunolJ.J., VargaR. Microstructure and magnetic properties of Ni50Mn37Sn13 Heusler alloy ribbons. Journal of Applied Physics, 2009, vol. 103, p. 07B326.
10. GuanW., LiuQ.R., GaoB., YangS., WangY. Large magnetocaloric effect at low magnetic field in Ni50-xCoxMn35In15 ribbons. Journal of Applied Physics, 2011, vol. 109, p. 07A903.
11. Sanchez-Llamazares J.L., Flores-ZunigaH., Sanchez-Valdes C., RossC.A., Garcia C. Refrigerant capacity of austenite in as-quenched and annealed Ni51.1 Mn31.2In17.7 melt spun ribbons. Journal of Applied Physics, 2012, vol. 111, p. 07A932.
12. de Campos A., RoccoD.L., Carvalho A.M.G., Caron L., CoelhoA.A., GamaS., da Silva L.M., Gandra F.C.G., Santos A.O., Cardoso L.P., Ranke P.J., Oliveira N.A. Ambient pressure colossal magnetocaloric effect tuned by composition in Mn1-xFexAs. Nature Materials, 2006, vol. 5, pp. 802-804.
13. von Ranke P.J., GamaS., CoelhoA.A., Campos A., Carvalho A.M.G., Gandra F.C.G., OliveiraN.A. Theoretical description of the colossal entropic magnetocaloric effect: Application to MnAs. Physical Review B, 2006, vol. 73, p. 014415.
14. RoccoD.L., Campos A., Carvalho A.M.G., Caron L., CoelhoA.A., GamaS., Gandra F.C.G., Santos A.O., Cardoso L.P., Ranke P.J., Oliveira N.A. Ambient pressure colossal magnetocaloric effect in Mn1-xCuxAs compounds. Applied Physics Letters, 2007, vol. 90, p. 242507.
15. BratkoM., MorrisonK., Campos A., GamaS., CohenL.F., SandemanK.G.
History dependence of directly observed magnetocaloric effects in (Mn, Fe)As. Applied Physics Letters, 2012, vol. 100, p. 252409.
16. Aliev A.M., Batdalov A.B., Kamilovl.K., KoledovV.V., ShavrovV.G., BuchelnikovV.D., GarciaJ., PridaV.M., HernandoB. Magnetocaloric effect in ribbon samples of Heusler alloys Ni-Mn-M (M=In, Sn). Applied Physics Letters, 2010, vol. 97, p. 212505.
17. Aliev A.M., Gamzatov A.G., Batdalov A.B., KalitkaV.S., Kaul A.R. Direct and inverse magnetocaloric effects in A-site ordered PrBaMn2O6 manganite. Journal of Alloys and Compounds, 2011, vol. 509, pp. L1-L3.
18. Aliev A.M., Gamzatov A.G., KalitkaV.S., Kaul A.R. Low field magnetocaloric effect and heat capacity of A-site ordered NdBaMn2O6 manganite. Solid State Communications., 2011, vol. 151, pp. 1820-1823.
192
A.M. Aliev, A.B. Batdalov, V.V. Koledov, V.G. Shavrov, B. Hernando
19. Aliev A.M., Gamzatov A.G., KamilovK.I., KaulA.R., Babushkina N.A.
Magnetocaloric properties of Ьао.7Сао.зМп16Оз and Lao.7Cao.3Mn18O3 manganites and their “sandwich”. Applied Physics Letters, 2012, vol. 101, p. 172401.
20. Caballero-Flores R., Franco V., Conde A., Kiss L.F. Influence of the demagnetizing field on the determination of the magnetocaloric effect from magnetization curves. Journal of Applied Physics, 2009, vol. 105, p. 07A919.
21. Khovaylo V.V., SkokovK.P., GutfleischO., MikiH., KainumaR., KanomataT.
Reversibility and irreversibility of magnetocaloric effect in a metamagnetic shape memory alloy under cyclic action of a magnetic field. Applied Physics Letters, 2010, vol. 97, p. 052503.
22. Booth R.A., MajetichS.A. The magnetocaloric effect in thermally cycled polycrystalline Ni-Mn-Ga. Journal of Applied Physics, 2012, vol. 111, p. 07A933.
23. Cakir O., Acet M. Neutron diffraction study of the magnetic-field-induced transition in Mn3GaC. Applied Physics Letters, 2012, vol. 100, p. 202404.
24. SkokovK.P., KhovayloV.V., MullerK.-H., MooreJ.D., LiuJ., GutfleischO. Magnetocaloric materials with first-order phase transition: thermal and magnetic hysteresis in LaFe11.8Si1.2 and Ni2.21Mn0.77Ga1.02. Journal of Applied Physics, 2012, vol. 111, p. 07A910.
25. FengY., LeeY.-H., FukudaT., KakeshitaT. Time dependent nature of first order magnetostructural transition in FeRh. Journal of Alloys and Compounds, 2012, vol. 538, pp. 5-7.
26. Gamzatov A.G., Aliev A.M., Ghotbi Varzaneh A., KameliP, Sarsaril.A.,
YuFengS.C. Inverse-direct magnetocaloric effect crossover in Ni47Mn40Sn12.5Cu0.5 Heusler alloy in cyclic magnetic fields. Applied Physics Letters, 2018, vol. 113, p. 172406.
Accepted article received 10.05.2020 Corrections received 25.05.2020
Magnetocaloric effect in Ni50.3Mn36.5Sn13.2 ribbon Heusler alloy in cyclic magnetic fields 193
Челябинский физико-математический журнал. 2020. Т. 5, вып. 2. С. 186-193.
УДК 537.622 DOI: 10.24411/2500-0101-2020-15205
МАГНИТОКАЛОРИЧЕСКИЙ ЭФФЕКТ В ЛЕНТОЧНОМ ОБРАЗЦЕ СПЛАВА ГЕЙСЛЕРА Ni5o.3Mn36.5Sni3.2 В ЦИКЛИЧЕСКИХ МАГНИТНЫХ ПОЛЯХ
А. М. Алиев1", A. Б. Батдалов1,6, В. В. Коледов2,с, В. Г. Шавров2^,
Б. Эрнандо3,е
1 Институт физики им. Х. И. Амирханова,
Дагестанский федеральный исследовательский центр РАН, Махачкала, Россия 2Институт радиотехники и электроники им. В. А. Котельникова РАН, Москва, Россия 3 Университет Овьедо, Овьедо, Испания
"[email protected], [email protected], [email protected], [email protected], e [email protected]
Исследованы магнитокалорические свойства ленточного образца сплава Гейслера Ni5o.3Mn36.5Sni3.2 прямым методом в циклических магнитных полях. В слабых магнитных полях наблюдается сильная анизотропия магнитокалорического эффекта: магнитокалорический эффект в образце больше в случае приложения магнитного поля параллельно плоскости ленты. С ростом магнитного поля наблюдается уменьшение анизотропии эффекта. Обнаружено, что в циклических магнитных полях значение магнитокалорического эффекта вблизи магнитоструктурного фазового перехода зависит от скорости изменения температуры образца: чем выше скорость развёртки температуры, тем выше значение магнитокалорического эффекта.
Ключевые слова: магнитокалорический эффект, циклическое магнитное поле, сплав Гейслера, лента, анизотропия.
Поступила в редакцию 10.05.2020 После переработки 25.05.2020
Сведения об авторах
Алиев Ахмед Магомедович, кандидат физико-математических наук, заведующий лабораторией, Институт физики им. Х. И. Амирханова Дагестанского федерального исследовательского центра, Махачкала, Россия; e-mail: [email protected].
Батдалов Ахмед Батдалович, доктор физико-математических наук, главный научный сотрудник, Институт физики им. Х. И. Амирханова Дагестанского федерального исследовательского центра, Махачкала, Россия; e-mail: [email protected].
Коледов Виктор Викторович, доктор физико-математических наук, ведущий научный сотрудник лаборатории магнитных явлений, Институт радиотехники и электроники им. В. А. Котельникова РАН, Москва, Россия; e-mail: [email protected].
Ш@вров Владимир Григорьевич, доктор физико-математических наук, профессор, заведующий лабораторией, Институт радиотехники и электроники им. В. А. Котельникова РАН, Москва, Россия; e-mail: [email protected].
Эрнандо Бланка, профессор, отделение физики, научный факультет, Университет Овьедо, Овьедо, Испания; e-mail: [email protected].
Работа поддержана грантом РНФ, проект № 18-12-00415. Исследование проведено с использованием оборудования Аналитического центра коллективного пользования Дагестанского федерального исследовательского центра РАН.
