Low field induced giant anisotropic magnetocaloric effect in DyFeO3 single crystal*
Ke Ya-Jiao, Zhang Xiang-Qun, Ge Heng, Ma Yue, Cheng Zhao-Hua†
State Key Laboratory of Magnetism and Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China

Corresponding author. E-mail: zhcheng@iphy.ac.cn

*Project supported by the National Basic Research Program of China (Grant Nos. 2010CB934202, 2011CB921801, and 2012CB933102) and the National Natural Science Foundation of China (Grant Nos. 11174351, 11274360, and 11034004).

Abstract

We have investigated the anisotropic magnetocaloric effect and the rotating field magnetic entropy in DyFeO3 single crystal. A giant rotating field entropy change of was achieved from b axis to c axis in bc plane at 5 K for a low field change of 20 kOe. The large anisotropic magnetic entropy change is mainly accounted for the 4f electron of rare-earth Dy3 + ion. The large value of rotating field entropy change, together with large refrigeration capacity and negligible hysteresis, suggests that the multiferroic ferrite DyFeO3 singlecrystal could be a potential material for anisotropic magnetic refrigeration at low field, which can be realized in the practical application around liquid helium temperature region.

Keyword: 75.30.Gw; 75.30.Sg; 75.47.Lx; magnetocaloric effect; rotating field entropy change; magnetocrystalline anisotropy; DyFeO3 single crystal
1. Introduction

Recently, the magnetic refrigeration based on magnetocaloric effect (MCE) that uses an external magnetic field to change the magnetic entropy of magnetic materials, has attracted much attention due to its higher energy efficiency and lower environmental influence than the conventional gas compression refrigeration.[13] Although new materials with giant MCE near room temperature have been studied for applications, materials with giant MCE in the low-temperature region from about 30  K down to sub-Kelvin temperatures are also essential for utilization in certain fields, such as gas liquefaction.[1] Helium plays a significant role in low-temperature scientific research and cryogenic engineering.[4] Generally, helium gas needs to be cooled down to liquid form for its storage and transportation. Therefore, it is significant to develop advanced magnetic refrigerants with large MCE around the liquid helium temperature (4.2  K). In addition, most of the permanent magnets in present market can only provide a maximum field about ∼ 20  kOe. This fact indicates that a large MCE under low magnetic field change is desirable for the fulfillment of a magnetic refrigerator. Low magnetic field induced giant magnetic entropy change has been observed in some rare-earth based intermetallic compounds, such as ErMn2Si2, [5]ErCr2Si2, [6] ErRu2Si2, [7]HoCuSi, [8] ErFeSi, [9] and TmGa, [10] whereas it has not been observed in rare-earth oxides. Alternatively, magnetic refrigeration can also be achieved by a rotating field MCE based on changing the magnetic anisotropy energy in a constant magnetic field, [11] which is attractive due to the simplification and possible miniaturization of the device.[12] Recently, rotating field entropy change due to magnetocrystalline anisotropy has been observed in TbMnO3, TmMnO3, ErFeO3, HoMn2O5, and PrSi at low temperatures.[1317] The exploration of magnetic refrigerant materials with large anisotropic MCE under low magnetic field has practical application value.

In the present work, we report a detailed study on the low-field MCE of DyFeO3 single crystal around liquid helium temperature. The reason for choosing DyFeO3 single crystal is three-fold. Firstly, DyFeO3 shows a complex magnetic transformation. In this system, iron ions order in a weakly canted antiferromagnetic structure below the Né el temperature TN= 645  K, and exhibit characteristic spin-reorientation (Morin) transitions Γ 4(Gx, Ay, Fz) → Γ 1(Ax, Gy, Cz), [1821] at TM∼ 50  K upon cooling, [22] which is only observed in DyFeO3 within RFeO3 (where R is rare-earth elements) series.[21] The Dy3 + sublattice remains paramagnetic until the temperature reaches its ordering temperature . The multiferroic, magnetoelectric properties together with the superfast optomagnetic effect of DyFeO3 single crystal have been extensively investigated.[20, 2325] Unfortunately, the effect of the interaction between the two magnetic sublattices on the MCE has not been understood yet. Secondly, the magnetic moment of Dy3 + ion is large, and consequently a large magnetic entropy change can be expected in DyFeO3 single crystal according to the relation of , where R is the gas constant, and J is the total angular momentum of the magnetic ion. Finally, owing to the strong magnetocrystalline anisotropy, a giant magnetic entropy change along the b axis of DyFeO3 single crystal at 20  kOe around the ordering temperature of Dy3 + is observed. This discovery not only advances our understanding of the MCE anisotropy in the DyFeO3 single crystal, but also opens a new arena for magnetic refrigerator by rotating its magnetization vector at a relatively small magnetic field, instead of moving it in and out of the magnet.

2 Experiment

DyFeO3 ceramic was prepared with the starting material Dy2O3 (> 99.9%)and Fe2O3 (> 99.9%) by using the solid-state reaction method, and was pressed into pellets and sintered in air atmosphere for 48  hours at 1300  ° C. The crystal structure of the sample was examined by powder X-ray diffraction. Seed rods for crystal growth were prepared under the hydrostatic pressure and sintered at 1400  ° C for 48 hours in air atmosphere. DyFeO3 single crystal was grown with four ellipsoidal mirrors (Crystal Systems Inc, FZ-T-10000-H-VI-VP) by the floating zone method. Back-reflection Laue X-ray diffraction experiments were carried out to check the single crystallinity and determine the crystallographic direction. Measurements of magnetization as functions of temperature and magnetic field were performed on commercial superconducting quantum interference device (SQUID) magnetometer (quantum design MPMS-XL).

3 Results and discussion

XRD patterns and back-reflection Laue XRD patterns (not shown) confirmed that DyFeO3 single crystal has an orthorhombically distorted pervoskite structure with Pbnm symmetry, which presented a high single crystallinity of the studied crystals. The Fe magnetic sublattices undergo a paramagnetic-antiferromagnetic transition around .[26] Below the transition temperature, the configuration of Fe  magnetic sublattice is Γ 4(Gx, Ay, Fz), in which the G-type component directly toward the a axis (Gx), but with a canting angle away from the a axis, which causes A-type and weak ferromagnetic (WFM) components along the b and c axis.[23] Figures  1(a)– 1(c) show the tempareture dependence of zero-field cooling (ZFC) and field cooling (FC) magnetizations measured in H= 100  Oe from 2  K to 300  K for the DyFeO3 single crystal along the a, b, and c axes, respectively. We can see an obvious decreasing magnetization along the c axis from 50  K to 45  K, which is followed by a continuous reorientation (Morin) transition from the c to the a axis, corresponding to the spin reorientation of Fe3 + moments from Γ 4(Gx, Ay, Fz) configuration to Γ 1(Ax, Gy, Cz) configuration.[22] The spin reorientation of DyFeO3 is contrary to the case of other rare-earth orthoferrites which show a reorientation transition from Γ 4(Gx, Ay, Fz) to Γ 2(Fx, Cy, Gz).[21] On further cooling, the kink point at 4.5  K indicated by the dotted line of Figs.  1(a) and 1(c) corresponds to the ordering of Dy3 + moments .

Fig.  1. Zero-field-cooled (ZFC) and field-cooled (FC) thermal magnetization curves of DyFeO3 along the (a) a, (b) b, and (c) c axes from 2  K to 300  K under a magnetic field of 100  Oe.

Fig.  2. Isothermal magnetization curves of DyFeO3 single crystal. (a) Magnetization curves along the a axis, (b) magnetization curves along the b axis, (c) magnetization curves along the c axis. The curves at 2  K and 4  K for the b axis are shown in the inset (b) for clarity.

The isothermal magnetization curves as functions of field along different crystal directions were measured to investigate the anisotropic magnetic entropy change in DyFeO3 single crystal. Figures  2(a)– 2(c) illustrate the isothermal magnetization curves along the a, b, and c axes of the DyFeO3 single crystal in the temperature range of 2– 40  K with an interval of 2  K, respectively. The isothermal magnetization curves along the a and c axes of the DyFeO3 single crystal in the Fe3 + moments spin-reorientation temperature range of 40– 70  K are also measured (not shown). There is no obvious increase or peak in the curves. This is because the magnetization change in the Morin transition region is quite small since the magnitude of the WFM component is about 0.01 percent of the G-type component in an Fe sublattice. The magnetization curves for these three directions are different either in shape or in magnetization values. A spin-flip phenomenon can be observed along the a, b, and c axes of the DyFeO3 single crystal at 2  K due to the antiferromagnetic interaction of RR being similar to TbFeO3.[2729] A complete reversibility of MCE requires that no hysteresis of either the temperature or the magnetic field appears in the vicinity of a magnetic ordering temperature. As shown in the inset of Fig.  2(b), a negligible hysteresis loop was observed only at 2  K and 4  K for the b axis. When the temperature is above 4  K, no hysteresis loops can be observed from the isothermal magnetization curves. From the data we can see a significant difference in the isothermal magnetization curves along the a, b, and c axes of the DyFeO3 single crystal, especially between the b axis and the c axis at low field, which implies that a giant anisotropy of MCE can be expected in the bc plane. Magnetic field-induced entropy change -Δ SM was calculated from the isothermal curves with the following equation based on a Maxwell relation:

where the slope of two adjacent data points is approximatively used for the numerical calculation of the gradient of (∂ M/∂ T)H. In this case, Δ T = 1  K and Δ H = 2  kOe.

The calculated -Δ SM as a function of temperature is shown in Figs.  3(a)– 3(c) for a field along the a, b, and c axes, respectively. Peaks appear at 9  K for the a axis with a value of 21.10  J/kg· K and at 11  K for the b axis with a value of 21.48  J/kg· K in a field of 70  kOe, whereas the values of -Δ SM for the c axis are smaller than those of the other two axes in the temperature range above the ordering of Dy3 + moments . We can clearly see a giant entropy change with a value of 16.51  J/kg· K along the b axis in a low magnetic field of 20  kOe near the ordering temperature of Dy3 + moments. Furthermore, a large anisotropy of the MCE is observed in the DyFeO3 single crystal along the b axis and c axis driven by a low magnetic field. The temperature of the corresponding peaks along the b axis of DyFeO3 single crystal is close to the magnetic transition of a Dy sublattice, which implies that Dy– Dy interaction plays an important role in giant MCE in a low magnetic field.

Fig.  3. Magnetic entropy changes of DyFeO3 single crystal. (a) Magnetic entropy changes along the a axis, (b) magnetic entropy changes along the b axis, and (c) magnetic entropy changes along the c axis.

Fig.  4. Calculated -Δ SM and refrigeration capacity (RC). Field dependence of (a) magnetic entropy changes of DyFeO3 and (b) refrigeration capacity along the a, b, and c axes at 5  K.

The field-dependent -Δ SM of different axes of DyFeO3 single crystal at 5 K is shown in Fig.  4(a). A large anisotropy of MCE is observed in DyFeO3 single crystal among the ab plane and c axis. It is noteworthy that there is a giant change below 20  kOe along the b axis at 5  K. For magnetic refrigeration application, not only a large entropy change value, but also a large refrigeration capacity is needed. The characteristic parameter that determines the magnetic cooling efficiency of a magnetocaloric material is the refrigeration capacity (RC) and is defined as

where T1 and T2 are the temperatures corresponding to both sides of the half-maximum value of -Δ SM peak. The RC is the measurement of the amount of heat transfer between the cold and hot reservoirs in an ideal refrigerator as a function of field. The field-dependent refrigeration capacity of different crystallines of the DyFeO3 single crystal is shown in Fig.  4(b). The three directions of DyFeO3 single crystal manifest obvious anisotropy with RC value, which is equal to 260.48  J/kg, 586.08  J/kg, and 10.1  J/kg in a field of 70  kOe for a, b, and c axes, respectively. It is interesting that DyFeO3 single crystal exhibits a strong magnetocrystalline anisotropy between b and c axes. Single crystal with giant anisotropic MCE at a relatively small magnetic field is a promising choice for the new type of magnetic refrigeration by simply rotating the magnetic field or refrigerants.

Fig.  5. Isothermal magnetization curves within the bc plane of DyFeO3 single crystal at (a) 4  K and (b) 6  K. 0° and 90° correspond to the direction of b and c axes, respectively.

The rotating magnetic entropy change can be obtained by rotating the crystal from the b axis to the c axis and measuring the corresponding isothermal magnetization curves. The representative isothermal magnetization curves at different angles for temperatures of 4  K and 6  K for DyFeO3 single crystal in the magnetic field range of 0– 20  kOe with an interval of 1  kOe are shown in Figs.  5(a) and 5(b), respectively. Taking the b axis as the initial angle, we can obtain the rotating magnetic entropy change as a function of angle by using Eq.  (1). As shown in Fig.  6, the largest values of under 20  kOe can be achieved at temperature of 5  K for DyFeO3. The large reversible anisotropic magnetic entropy change at a relatively small magnetic field suggests that a promising candidate for new type magnetic refrigeration can be achieved at low field. Low-field-induced giant anisotropic magnetocaloric effect can be realized in the practical application by simply rotating the DyFeO3 single crystal or magnet, rather than moving the magnetic refrigerant in and out of the magnet.

Fig.  6. Rotating field entropy changes -Δ SR(α ) from the b axis to the c axis versus magnetic field DyFeO3 single crystal at 5  K.

4 Conclusion

In conclusion, a large anisotropic entropy change and RC were observed in DyFeO3 single crystal. Owing to the strong magnetocrystalline anisotropy between b and c axes, there was a giant rotating magnetic entropy change under 20  kOe at 5  K. This discovery not only gives us a deeper insight into the understanding of the MCE anisotropy in spin canting anti-ferromagnetic single crystal, but also moves forward the practical application for rotary magnetic refrigerator by rotating its magnetization vector at a relatively small magnetic field, instead of moving it in and out of the magnet.

Reference
1 Gschneidner Jr K A, Pecharsky V K and Tsokol A O 2005 Rep. Prog. Phys. 68 1479 DOI:10.1088/0034-4885/68/6/R04 [Cited within:2] [JCR: 13.232]
2 Gutfleisch O, Willard M A, Bruck E, Chen C H, Sankar S G and Liu J P 2011 Adv. Mater. 23 821 DOI:10.1002/adma.v23.7 [Cited within:1] [JCR: 14.829]
3 Shen B G, Sun J R, Hu F X, Zhang H W and Cheng Z H 2009 Adv. Mater. 21 4545 DOI:10.1002/adma.v21:45 [Cited within:1] [JCR: 14.829]
4 Kapitza P 1934 Proc. R. Soc. Lond. A 147 189 DOI:10.1098/rspa.1934.0214 [Cited within:1]
5 Li L, Nishimura K, Hutchison W D, Qian Z, Huo D and NamiKi T 2012 Appl. Phys. Lett. 100 152403 DOI:10.1063/1.4704155 [Cited within:1] [JCR: 3.794]
6 Li L, Hutchison W D, Huo D, Namiki T, Qian Z and Nishimura K 2012 Scripta Mater. 67 237 DOI:10.1016/j.scriptamat.2012.04.028 [Cited within:1] [JCR: 2.821]
7 Samanta T, Das I and Banerjee S 2007 Appl. Phys. Lett. 91 152506 DOI:10.1063/1.2798594 [Cited within:1] [JCR: 3.794]
8 Chen J, Shen B G, Dong Q Y, Hu F X and Sun J R 2010 Appl. Phys. Lett. 96 152501 DOI:10.1063/1.3386536 [Cited within:1] [JCR: 3.794]
9 Zhang H, Shen B G, Xu Z Y, Shen J, Hu F X, Sun J R and Long Y 2013 Appl. Phys. Lett. 102 092401 DOI:10.1063/1.4794415 [Cited within:1] [JCR: 3.794]
10 Mo Z J, Shen J, Yan L Q, Tang C C, Lin J, Wu J F, Sun J R, Wang L C, Zheng X Q and Shen B G 2013 Appl. Phys. Lett. 103 052409 DOI:10.1063/1.4816729 [Cited within:1] [JCR: 3.794]
11 Nikitin S A, Skokov K P, Koshkid'ko Y S, Pastushenkov Y G and Ivanova T I 2010 Phys. Rev. Lett. 105 137205 DOI:10.1103/PhysRevLett.105.137205 [Cited within:1] [JCR: 7.943]
12 Kuz'min M D and Tishin A M 1991 J. Phys. D: Appl. Phys. 24 2039 DOI:10.1088/0022-3727/24/11/020 [Cited within:1] [JCR: 2.528]
13 Das P K, Bhattacharyya A, Kulkarni R, Dhar S K and Thamizhavel A 2014 Phys. Rev. B 89 134418 DOI:10.1103/PhysRevB.89.134418 [Cited within:1]
14 Balli M, Jand l S, Fournier P and Gospodinov M M 2014 Appl. Phys. Lett. 104 232402 DOI:10.1063/1.4880818 [Cited within:1] [JCR: 3.794]
15 Jin J L, Zhang X Q, Li G K, Cheng Z H, Zheng L and Lu Y 2011 Phys. Rev. B 83 184431 DOI:10.1103/PhysRevB.83.184431 [Cited within:1]
16 Jin J L, Zhang X Q, Ge H and Cheng Z H 2012 Phys. Rev. B 85 214426 DOI:10.1103/PhysRevB.85.214426 [Cited within:1]
17 Huang R, Cao S, Ren W, Zhan S, Kang B and Zhang J 2013 Appl. Phys. Lett. 103 162412 DOI:10.1063/1.4825274 [Cited within:1] [JCR: 3.794]
18 Berton A 1968 J. Appl. Phys. 39 1367 DOI:10.1063/1.1656307 [Cited within:1] [JCR: 0.71]
19 Prelorendjo L A, Johnson C E, Thomas M F and Wanklyn B M 1980 J. Phys. C: Solid State Phys. 13 2567 DOI:10.1088/0022-3719/13/13/012 [Cited within:1]
20 Rajeswaran B, Sanyal D, Chakrabarti M, Sundarayya Y, Sundaresan A and Rao C N R 2013 Europhys. Lett. 101 17001 DOI:10.1209/0295-5075/101/17001 [Cited within:1] [JCR: 2.26]
21 White R L 1969 J. Appl. Phys. 40 1061 DOI:10.1063/1.1657530 [Cited within:3] [JCR: 0.71]
22 Zhao Z Y, Zhao X, Zhou H D, Zhang F B, Li Q J, Fan C, Sun X F and Li X G 2014 Phys. Rev. B 89 224405 DOI:10.1103/PhysRevB.89.224405 [Cited within:2]
23 Tokunaga Y, Iguchi S, Arima T and Tokura Y 2008 Phys. Rev. Lett. 101 097205 DOI:10.1103/PhysRevLett.101.097205 [Cited within:2] [JCR: 7.943]
24 Kimel A V, Kirilyuk A, Usachev P A, Pisarev R V, Balbashov A M and Rasing T 2005 Nature 435 655 DOI:10.1038/nature03564 [Cited within:1] [JCR: 38.597]
25 Tokunaga Y, Taguchi Y, Arima T H and Tokura Y 2012 Nat. Phys. 8 838 DOI:10.1038/nphys2405 [Cited within:1] [JCR: 19.352]
26 Gorodetsky G, Sharon B and Shtrikman S 1968 J. Appl. Phys. 39 1371 DOI:10.1063/1.1656309 [Cited within:1] [JCR: 0.71]
27 Gordon J D, Gorodetsky G and Hornreich R M 1976 J. Magn. Magn. Mater. 7 280 DOI:10.1016/0304-8853(76)90094-9 [Cited within:1] [JCR: 1.826]
28 Bidaux R, Bouree J E and Hammann J 1975 J. Phys. Chem. Solids 36 655 DOI:10.1016/0022-3697(75)90083-9 [Cited within:1] [JCR: 1.527]
29 Bouree J E and Hammann J 1975 J. Phys. 36 391 DOI:10.1051/jphys:01975003605039100 [Cited within:1] [JCR: 0.71]