Magnetocaloric effect and refrigeration cooling power in amorphous Gd7Ru3 alloys

In this paper, we report the magnetic, heat capacity and magneto-caloric effect (MCE) of amorphous Gd7Ru3 compound. Both, temperature dependent magnetization and heat capacity data reveals that two transitions at 58 K and 34 K. MCE has been calculated in terms of isothermal entropy change (ΔSM) and adiabatic temperature change (ΔTad) using the heat capacity data in different fields. The maximum values of ΔSM and ΔTad are 21 Jmol−1K−1 and 5 K respectively, for field change of 50 kOe whereas relative cooling power (RCP) is ∼735 J/kg for the same field change.


INTRODUCTION
Magnetocaloric refrigeration and power generation are amongst the latest solid-based refrigeration technologies used for cryogenic applications and an ideal substitute for the existing gas-based refrigeration in terms of environmentally benign, adaptability and efficiency.Also, the technology is solid state based which makes it attractive for applications such as space missions, food storage, air conditioning, gas liquefaction etc. [1][2][3][4] Magnetocaloric refrigeration is manifests as isothermal entropy change ∆S M or adiabatic temperature change ∆T ad (heating or cooling) of magnetic solids due to a varying magnetic field.][3][4] Large value of MCE spreads over a wide temperature range is considered as one of the most important requirements of a practical magnetic refrigerant system.][7][8][9][10] FOT and strong magnetocrystalline coupling present several disadvantages, i.e., high hysteresis loss and hard magnetic behavior reduce the efficiency of refrigerating cooling power (RCP) [= ∆S M (T) * δT FWHM ] and structural changes promotes cracks nucleation which may cause several damages in the refrigeration materials during cycles. 1 First order transitions are sharp reducing the operating temperature range and refrigerating cooling power.
On the other hand, metallic glasses/ amorphous alloys display very unique properties, such as, (i) soft magnet with second order magnetic transition, reduced coercivity, and high permeability; (ii) tunable transition temperature by composition or annealing which is useful in application of magnetic refrigeration [11][12][13][14][15] taking these parameters into consideration, we study the magnetic, heat capacity and magnetocaloric effect in amorphous Gd 7 Ru 3 alloys.We found that large ∆S M and ∆T ad spreads over 2 to above 120 K temperature range in 50 kOe field required for practical magnetic refrigerant system.

EXPERIMENTAL DETAILS
The 20-30µm thick ribbons were prepared by melt spinning technique under argon atmosphere on rotating copper wheel with 42 ms −1 speed.Structural information was received by X-ray diffraction using Cu K α radiation.Magnetization was measured using Quantum design SQUID whereas, heat capacity was measured in PPMS in the temperature ranges of 2-200 K. MCE has been calculated using M-H isotherms near transition temperature and heat capacity data in different field.Before measurement at each temperature the specimen was zero-field cooled from 60 K.

RESULTS AND DISCUSSION
Figure 1 shows the XRD patterns of Gd 7 Ru 3 alloy.Obtained result shows that Gd 7 Ru 3 is a fully amorphous alloy.Two very broad peaks have been observed with a maximum of 2θ values between 30 • and 50 • , for the first coordination shell.Another peak in the range of 55 • to 75 • was observed.M-T data of Gd 7 Ru 3 compound collected in various applied magnetic fields both under zero field cooled (ZFC) and field cooled conditions (FCC) are shown in figure 2, whereas, inset of figure 2(a) shows the Curie-Weiss fit of the inverse (d.c.) magnetic susceptibility and inset of figure 2(b) shows the thermal hysteresis below ordering temperature.It can be seen from the figure 2(a) that M-T data shows two magnetic transitions, the first one at high temperature and the second one at roughly half of the high temperature transition.The high temperature transition seen in this compound corresponds to the onset of long range magnetic order and is denoted by T ord .Low temperature transition is denoted by T 1 and has ferromagnetic nature.The M-T data collected in H= 100 Oe reveals that T ord and T 1 of this compound are 58 K and 34 K respectively.Also, M-T data collected in higher fields, i.e., H = 1 kOe and 3 kOe, and 5 kOe reveals that the T ord increases with increase in magnetic field.Therefore, field dependence of T ord indicates that transition at T ord is ferromagnetic in character.Apart from the multiple magnetic transitions, another interesting feature seen in this compound is the existence of thermomagnetic irreversibility between the ZFC and FCC magnetization data.Generally, such thermomagnetic irreversibility between the ZFC and FCC magnetization data is seen in the compounds with narrow domain wall systems. 16t can be seen from the figure 2(a) that at high temperatures, the inverse of susceptibility of this compound follows the Curie-Weiss law.The effective moment (µ eff ) and paramagnetic Curie temperature (θ p ) thus obtained from the fit are ∼8 µ B and 159 K respectively.Although effective moments in excess of 7.94µ B /Gd are currently interpreted in crystalline materials as being the result of conduction electron polarization, as explained by Buschow et.al., in the present material an interpretation in terms of small ferromagnetic clusters seems more appropriate. 15The distribution of nearest neighbor distances mentioned above in other extreme can lead to small regions in which the ferromagnetic coupling is stronger than the average.In such regions ferromagnetically ordered clusters may present at temperatures higher than ordering temperature where they behave as super-paramagnetic particles.
Figure 3 shows the field dependent magnetization isotherms, obtained at 5 K and up to a maximum field of 120 kOe.The M-H isotherms at different temperature in step of 5 K is shown in inset.It can be seen from the figure that saturation of magnetization is obtained above 40 kOe.At 5 K saturation magnetization is found to be 7.24 µ B in an applied field of 120 kOe.However, the gJ values corresponding to Gd 3+ ion is 7 µ B .The difference between the gJ values and the experimentally observed values, may be due to coupling between 4f electron spins proceeding to a large extent indirectly via 5d electron polarization.The 5d electronic states will split by the local FIG.3. M -H isotherm at 5K. Inset shows the M-H isotherms of Gd 7 Ru 3 alloys near ordering temperature.Before measurements at each temperature the specimen was zero-field cooled from 100 K. crystal field and only low lying crystal field states will be occupied by 5d electrons.It is reasonable, therefore to assume that 4f-5d coupling involves only some particular 5d orbital's and in this way leads to an anisotropic exchange interaction.
In order to further understand the nature of the magnetic state of this compound, heat capacity measurements, both under zero-field as well as in various applied fields, have been performed.The representative C vs. T plot for Gd 7 Ru 3 is given in figure 4(a).It can be seen from this figure that zero-field C-T data shows two anomalies at 58 K and 34 K, which are close to the magnetic transitions seen at T ord and T 1 in the M-T data of this compound (see Figure 2(a)).It may also be seen from the figure that with increase in field, the anomaly at T ord gets rounded off and shift towards higher temperatures.Therefore, the C-T data also indicates that T ord anomaly is ferromagnetic in character.
In order to analyze the magnetic behavior of this compound, C mag (magnetic part of heat capacity) has been resolved from the zero-field heat capacity.The C mag was resolved from temperature variation heat capacity data (C-T) by deduction of nonmagnetic contribution from it.The C lattice and C ele contribution to the heat capacity was determined using equation (1).In this sample, Debye model is not a good approximation for the calculation of lattice contribution to the heat capacity.Therefore, a modified expression while taking into account the Debye and the Einstein models, as represented by the second and third terms respectively of equation ( 1) was used to analyze the C-T data. 17 Here, γ is the coefficient of electronic specific heat, R is the universal gas constant, α E 's and α D 's are the anharmonicity coefficients for the optical branches and the acoustic branches, respectively; T where θ E 's and θ D are the Einstein and Debye temperatures, respectively.In equation (1), the first term corresponds to electronic contribution to heat capacity whereas, the second and third terms are due to the phonon contribution corresponding to the Einstein and Debye models, respectively.4][25] It may further be noticed from equation ( 1) that in second term corresponding to the Einstein model, the summation extends from i=1 to 27, which is due to 27 different optic branches expected in this compound. 17   It may be mentioned here that C mag at low temperatures, neither shows T 3/2 nor T 3 dependence, and hence indicates that low temperature magnetic state in this compound is neither purely TABLE I. Calculated values of γ (coefficient of electronic specific heat), θ E 's (Einstein temperatures), θ D (Debye temperature), α E 's (anharmonicity coefficients for the optical branches) and α D 's (anharmonicity coefficients for the acoustic branches) from zero field heat capacity data.ferromagnetic nor antiferromagnetic.It can be seen from the inset of the figure 4(b) that magnetic entropy shows saturation tendency towards the theoretical value Rln(2J+1) at temperatures well above T ord , which implies that only few levels of the ground state multiplet are involved in the ordering process. 26he MCE of this compound have been determined in terms of −∆S M and ∆T ad using the heat capacity data in 0, 20 and 50 kOe fields using the equations. 1

S(T, H)
Similarly, based on the M-H isotherms near ordering temperatures, the change in −∆S M of sample was calculated using the integrated Maxwell relation 1 Where T av = (T i+1 +T i ) /2 means average temperature and ∆T = T i+1 − T i means temperature difference between two magnetization isotherms measured at T i+1 and T i with the magnetic field H i to H f .The MCE behavior of amorphous Gd 7 Ru 3 has been determined in terms of ∆S M as well as ∆T ad using heat capacity data in field change of 20 and 50 kOe as shown in figure 5(a) & 5(b). 1 It can be seen from figure 5(a) that ∆S M vs. T plot shows a maximum near T ord , and ∆S M do not die out even at temperatures well above 120 K.This may be due to the presence of short range magnetic correlations or spin fluctuations in the paramagnetic state.
The maximum isothermal entropy change (∆S max M ), for ∆H = 50 kOe and 20 kOe, of this compound is ∼21 J mol −1 K −1 and 10 J mol −1 K −1 respectively.It is of interest to note that the ∆S max M values of (Er/Dy)Al 2 compounds, which are promising magnetic refrigerants in the temperature range of 13 to 60 K, varies in the range of 4.5 to 8.0 J mol −1 K −1 , for ∆H=50 kOe. 1 The ∆S max M of RNi 2 compounds, which have ordering temperature less than 80 K, varies from 3-8 J mol −1 K −1 , for the same field change. 27 ∆H = 40 kOe). 28Therefore, the comparison of maximum ∆S M of present compound with that of the potential magnetic refrigerant materials indicates that, this material is suitable for the refrigeration application below 60 K.In figure 4(b), ∆T ad vs. T plot of this compound is shown at 20 and 50 kOe.The maximum value of ∆T ad ∆T max ad at 20 and 50 kOe are ∼2 and ∼5 K respectively.The temperature variation of ∆T ad is similar to that of temperature dependence of ∆S M .Furthermore, the ∆T max ad values HoNiAl are 4 and 8.7 K, for ∆H = 20 and 50 kOe, respectively, whereas for DyNiAl, for the same field changes, these values are found to be 3.5 and 6.8 K, respectively.It may be mentioned here that ∆T max ad value, for ∆H = 50 kOe, of RNi 2 compounds varies in the range of 3.5-9 K 27 whereas for (Er,Dy)Al 2 compounds, for the same field change, this value varies between 7 and 11 K. 29 In the material Gd 7 Ru 3 , large ∆S M and ∆T ad values persists in a wide temperature range around ordering temperature range.This feature is important to obtain relatively high cooling capacity (RC parameter) and is characteristic of amorphous alloys.Relatively high values of RC parameter, i.e., 180 and 735 J/mol respectively, were obtained for ∆H=20 and 50 kOe.This RC of investigated alloys is comparable to the values determined for Gd 55 Fe 30 Al 30 and Gd 55 Fe 20 Al 25 . 30[32][33][34]

CONCLUSIONS
The magneto-thermal properties of amorphous Gd 7 Ru 3 have been studied.The magnetocaloric property of this compound is found to be comparable to that of many potential refrigerant materials like (Er/Dy)Al 2 , RNi 2 , Gd 2 PdSi 3 etc.Large RCP, large ∆T ad change, soft magnetic behavior and wide operating temperature range (>120 K) make it an attractive candidate as magnetic refrigerant in low temperature region.The magnetic state of Gd 7 Ru 3 has been studied by heat capacity measurements, both, under zero-field as well as in various applied fields.

ACKNOWLEDGMENT
One of the authors (Pramod Kumar) thanks DST, Govt. of India for proving financial support for this work.

- 22
However, in the calculation of C lattice , only three different θ E 's, each one corresponding to a group of nine optic branches, are taken into consideration.All article content, except where otherwise noted, is licensed under a Creative Commons Attribution 3.0 Unported license.See: http://creativecommons.org/licenses/by/3.0/Downloaded to IP: 14.139.60.97On: Tue, 13 Oct 2015 05:17:23

FIG. 4 .
FIG. 4. (a) Temperature dependent heat capacity in applied fields of 0, 20, 50 kOe.The arrows in the figure indicate the ordering temperature (T ord ) and the low temperature transition seen at T 1 .(b)Temperature variation of zero field heat capacity.Open circles represent experimental data and solid red line is the calculated nonmagnetic contribution.Filled spheres represent magnetic contribution.The insets show magnetic entropy vs. temperature in zero fields.

Figure 4 (
Figure 4(b)shows temperature variation of total heat capacity, nonmagnetic and magnetic contribution of the Gd 7 Ru 3 compound whereas; inset shows the temperature variation of magnetic entropy of these compounds.The parameters used for calculating the nonmagnetic contribution to the heat capacity are given in table I.It may be mentioned here that C mag at low temperatures, neither shows T 3/2 nor T 3 dependence, and hence indicates that low temperature magnetic state in this compound is neither purely

7 P
FIG. 5. Temperature dependence of calculated (a) ∆S M and (b) ∆T ad in applied fields of 20 and 50 kOe.