Large Inverse Magnetocaloric Effects and Giant Magnetoresistance in Ni-Mn-Cr-Sn Heusler Alloys

Abstract

The magnetostructural transitions, magnetocaloric effects, and magnetoresistance properties of Ni₄₅Mn₄₃CrSn₁₁ Heusler alloys were investigated using X-ray diffraction (XRD), field-dependent magnetization, and electrical resistivity measurements. A large inverse and direct magnetocaloric effect has been observed in Ni₄₅Mn₄₃CrSn₁₁ across the martensitic and Curie transition temperature, respectively. The values of the latent heat (L = 15.5 J/g) and corresponding magnetic (ΔS_M) and total (∆S_T) entropy changes (ΔS_M = 35 J/kg·K for ΔH = 5T and ∆S_T = 39.7 J/kg·K) have been evaluated using magnetic and differential scanning calorimetry (DSC) measurements, respectively. A substantial jump in resistivity was observed across the martensitic transformation. A large negative magnetoresistance (~67%) was obtained at the magnetostructural transition for a field change of 5 T. The roles of the magnetic and structural changes on the transition temperatures and the potential application of Ni₄₅Mn₄₃CrSn₁₁ Heusler alloys for refrigerator technology are discussed.

1. Introduction

Magnetic materials with large magnetic entropy changes (ΔS_M) are potential candidates for use in new refrigerator technology [1]. Large ΔS_M values have been observed for a materials showing first order transitions (FOT) in which sharp changes of magnetization occur across the ordering temperature with simultaneous changes in magnetic and structural phases [2,3]. These materials are characterized by a sharp peak in ΔS_M in a relatively narrow temperature range. In contrast, materials undergoing second order transitions (SOT) exhibit relatively small values of ΔS_M over wide working temperature intervals [4,5]. The advantage of using materials with a large ΔS_M at the SOT is the complete reversibility of the magnetization process. Magnetic refrigerant materials that exhibit both magnetic entropy changes can be used in a refrigeration cycle that exploits both effects [6,7].

Much research has been focused on Ni-Mn-based Heusler alloys in order to find compositions in which the martensitic transformations occur around room temperature [8,9,10,11,12,13,14,15]. It has been shown that the substitution of Co, Cu, and Al in Ni-Mn-Sn Heusler alloys results in large magnetocaloric effects due to the sudden changes of magnetization and resistance near the martensitic transition temperature [16,17,18]. Recently, it was reported that the working temperature of the Ni-Mn-Sn Heusler alloys can be controlled by changing the Sn concentration, making them perfect materials for application in magnetic refrigeration [19]. Sutou et al. observed a large magnetic-field-induced strain in Ni-Mn-based Heusler alloys, and discussed the possible application of these alloys for Ga-free ferromagnetic shape memory alloys [20]. The magnetic order, magnetocaloric effects, and magnetoresistance in Ni-Mn-Sn Heusler alloys were found to depend on the composition and fabrication conditions. Hence, systematic study of these alloys is still needed.

The main objective of this research is to explore magnetocaloric materials based on Mn-based Heusler alloys, and specifically to understand the properties of magnetic and structural transitions in Ni-Mn-Sn. In this paper, we investigate the magnetic, structural, magnetocaloric, thermomagnetic, and magnetoresistance properties of Ni₄₅Mn₄₃CrSn₁₁ Heusler alloys through X-ray diffraction, differential scanning calorimetry, magnetization, and resistivity measurements.

2. Results and Discussion

The XRD pattern for Ni₄₅Mn₄₃CrSn₁₁ at room temperature (Figure 1) indicates that the sample was in mixed martensitic and austenitic phases. The austenitic phase is in a cubic structure and dominates at room temperature, whereas in the martensitic phase, the system is in a tetragonal crystalline state. The XRD peaks from the martensitic phase were characterized by a low intensity, about 30% of the relative intensity of the austenitic phase.

Figure 2a shows the zero-field-cooled (ZFC) and field-cooled (FC) M(T) curves for Ni₄₅Mn₄₃CrSn₁₁ in a magnetic field of 100 Oe. In the ZFC case, the sample was initially cooled to 5 K in zero magnetic field, and then data were taken in 100 Oe magnetic field as the temperature increased from 5 K. In FC mode, data were collected during cooling in a field of 100 Oe. The M(T) data show that the sample undergoes three transitions (i) at the Curie temperature of the martensitic phase (T_CM); (ii) at the martensitic transition temperature (T_M) from a low magnetic state to the FM austenitic phase; and (iii) at the Curie temperature (T_C) from the FM austenitic to the PM austenitic phase. The sample shows a thermal hysteresis in magnetization about T_M. The presence of hysteresis and the jump-like change in magnetization at T_M are typical for first-order structural (martensitic) transitions observed in such systems [21]. There is splitting in the ZFC and FC magnetization curves below the blocking temperature (T_B). Similar ZFC–FC splitting has been observed in References [9,22], which is attributed to exchange bias phenomena. The high field (H = 5 T) ZFC M(T) curves are shown in Figure 2b. A larger difference in the magnetization of about 30 emu/g was observed during the cooling at T = 235 K for H = 5 T (Figure 2b). Additionally, the shift in temperature (ΔT) was around 20 K at H = 5 T due to the observed field-induced transition [21].

The thermal properties of Ni₄₅Mn₄₃CrSn₁₁ were investigate by DSC measurements and are shown in Figure 3. The first-order nature of the phase transitions in Ni₄₅Mn₄₃CrSn₁₁ was confirmed by the temperature hysteresis of the heat flow transition peaks (see Figure 3). The large endothermic/exothermic peaks observed during heating/cooling cycles are related to the latent heat of the first-order magnetostructural transition (MST) from the low magnetization martensitic state to the ferromagnetic state. The temperature hysteresis of the heat flow of about 18 K between heating and cooling cycles detected from DSC measurements is consistent with magnetization results. The latent heat (L) was calculated from the endothermic peak using the relation:

$$L={{\displaystyle \int }}_{T_s}^{T_f}\frac{\mathrm{d}Q}{\mathrm{d}T}\mathrm{d}T$$

(1)

where dQ/dT is the rate of change of heat flow with temperature, and T_s and T_f are the initial and final temperatures of the magnetostructural phase transitions on heating, respectively. The latent heat of the magnetostructural phase transition and corresponding total entropy changes (∆S_T) estimated from DSC curves were 15.5 J/g and 39.7 J/kg·K, respectively.

Isothermal magnetization M(H) curves for the Ni₄₅Mn₄₃CrSn₁₁ at T_M and T_C are shown in Figure 4. The metamagnetic-like behavior was observed in the magnetization curves at T_M due to a field-induced transition. The magnetic susceptibility of the sample calculated from the results shown in Figure 2 demonstrates a Curie-Weiss behavior that is characteristic of a parametric phase for T > 318 K. In the low magnetic field region, the initial susceptibility increases drastically with temperature starting at 305 K.

The magnetic entropy changes (ΔS_M) were estimated from the magnetization isotherms measured at different temperatures calculated using the Maxwell relation from [23]. It has been calculated that the entropy changes estimated by the Clausis-Clapeyron equation and Maxwell’s relation are in good agreement in the case of Mn-based Heusler alloys [6]. Figure 5 shows the ΔS_M as a function of temperature for various magnetic fields (ΔH) for Ni₄₅Mn₄₃CrSn₁₁. The sample exhibits an inverse magnetocaloric effect with ΔS_M = +34 J/kg·K at the MST (T_M) for a magnetic field change of 5 T. These values of ΔS_M are much larger than those reported in Ni-Mn-Sn based Heusler alloys (17 J/kg·K) [17,18]. This large value of ∆S_M was due to a large ∆M (as seen in Figure 2b). Additionally, the entropy change associated with the magnetic transition is over a relatively wider range of temperature, even though the values are comparatively smaller (~6.5 J/kg·K). From the inset of Figure 5, one can see that the ∆S_M increases almost linearly with ∆H at T_M and T_C.

The refrigeration capacity (RC) is another useful parameter used to quantify MCE properties. It has been calculated by integrating the ΔS_M (T, H) curve over the full width at half maximum using the relation [23]:

$$\mathrm{RC}={{\displaystyle \int }}_{T_1}^{T_2}\mathrm{\Delta }{S}_M\left(T\right)\mathrm{d}T$$

(2)

The maximum RC values for Ni₄₅Mn₄₃CrSn₁₁ was 380 J/kg and 220 J/kg at the SOT and FOT, respectively, for a magnetic field change of 5 T. These RC values are comparable to those observed in Ni-Mn-Sn systems near room temperature [16]. Since magnetic and thermal hysteresis have been observed at the FOT—which are disadvantageous for magnetic refrigeration—the magnetocaloric properties at SOTs are more favorable for refrigeration.

The ZFC resistivity ρ(T) curves of Ni₄₅Mn₄₃CrSn₁₁ at magnetic fields of 0 and 5 T are shown in Figure 6a. At low temperature, the resistivity curves remain relatively constant until the step-like drops in the resistivity were observed in the vicinity of T_M (see Figure 6a). The large jump near T_M is due to a metamagnetic transition from a low magnetization state with strong antiferromagnetic correlations to a ferromagnetic austenitic phase. The application of a magnetic field shifts the jump in resistivity toward lower temperatures, resulting in the giant MR. The maximum MR was found to be ~ −67% at T ≈ 251 K when the system passes from the low magnetization state of the martensitic phase to the ferromagnetic austenitic phase (see Figure 6b). The observed MR value for Ni₄₅Mn₄₃CrSn₁₁ is 20% larger than those reported for Sn-based Heusler alloys [16], and similar to those observed in Ni₅₀Mn₃₄In₁₆ Heusler alloys [12].

3. Materials and Methods

A stoichiometric polycrystalline ingot (3 g) of Ni₄₅Mn₄₃CrSn₁₁ was fabricated by conventional arc-melting in an ultra-high purity argon atmosphere using high-purity (Ni: 99.9%; Mn, Cr: 99.99%; and Sn: 99.9999%) elements. For homogenization, the sample was wrapped in tantalum foil and annealed at 1123 K for 24 h in high vacuum (≈10⁻⁴ torr) and slowly cooled to room temperature. X-ray diffraction (XRD) measurements were done with an X-ray diffractometer (GMC-Mini Materials Analyzer, Hampshire, IL, USA) using Cu-Kα radiation to determine the phase purity and crystal structure. The magnetic properties were measured in a temperature interval of 5–400 K, and in magnetic fields up to 5 T, using a superconducting quantum interference device (SQUID) magnetometer by Quantum Design, Inc., San Diego, CA, USA. The resistance and magnetoresistance (MR) of the sample was studied using the four probe method within the temperature interval of 5–400 K and in magnetic fields up to 5 T. All magnetotransport measurements were carried out in zero-field-cooled (ZFC) conditions. The MR(H,T) values were calculated using MR = [{R(H,T) − R(0,T)}/R(0,T)] × 100%. The differential scanning calorimetry (DSC) measurements were carried out employing a Perkin-Elmer DSC 8000 instrument (with the ramp rate of 20 K/min during heating and cooling) (PerkinElmer, Santa Clara, CA, USA) in the temperature range 123–473 K.

4. Conclusions

We have investigated the magnetic, structural, thermal, transport, and magnetocaloric properties of Ni₄₅Mn₄₃CrSn₁₁ alloys. This compound exhibits large inverse and direct magnetocaloric effects of 35 J/kg·K and 6.5 J/kg·K at T_M and T_C, respectively. The application of a magnetic field shifts the jump in resistivity towards lower temperatures, resulting in a giant MR (~−67%). The maximum RC values for a magnetic field change of 5 T were found to be 380 J/kg and 220 J/kg at the SOT and FOT, respectively. The large magnetocaloric effect, magnetoresistance, and relatively low cost of the constituents of the Ni-Mn-Cr-Sn alloys make these compounds potential working materials for magnetic refrigeration.