A I D Journal of /All Applied Physics

Impact of structural disorder on the magnetic ordering and magnetocaloric response of amorphous Gd-based microwires

Anis Biswas, Y. Y. Yu, N. S. Bingham, H. Wang, F. X. Qin, J. F. Sun, S. C. Yu, V. Franco, H. Srikanth, and M. H. Phan

Citation: Journal of Applied Physics 115, 17A318 (2014); doi: 10.1063/1.4864143 View online: http://dx.doi.org/10.1063/1.4864143

View Table of Contents: http://scitation.aip.org/content/aip/journal/jap/115/17?ver=pdfcov Published by the AIP Publishing

Articles you may be interested in

Excellent magnetocaloric properties of melt-extracted Gd-based amorphous microwires Appl. Phys. Lett. 101, 102407 (2012); 10.1063/1.4751038

The effect of distributed exchange parameters on magnetocaloric refrigeration capacity in amorphous and nanocomposite materials

J. Appl. Phys. 111, 07A334 (2012); 10.1063/1.3679456

Magnetocaloric effect in Tb 60 Ni 30 Al 10 glass: A material that can either heat or cool upon magnetization J. Appl. Phys. 109, 033914 (2011); 10.1063/1.3544450

Large magnetocaloric effect and enhanced magnetic refrigeration in ternary Gd-based bulk metallic glasses J. Appl. Phys. 103, 023918 (2008); 10.1063/1.2836956

Magnetocaloric effect in Gd-based bulk metallic glasses Appl. Phys. Lett. 89, 081914 (2006); 10.1063/1.2338770

Journal of Applied Physics

Journal of Applied Physics is pleased to announce André Anders as its new Editor-in-Chief

JOURNAL OF APPLIED PHYSICS 115, 17A318 (2014)

CrossMark

Iii «-click for updates

Impact of structural disorder on the magnetic ordering and magnetocaloric response of amorphous Gd-based microwires

Anis Biswas,1a) Y. Y. Yu,1 N. S. Bingham,1 H. Wang,2 F. X. Qin,3 a) J. F. Sun,2 S. C. Yu,4 a) V. Franco,5a) H. Srikanth,1a) and M. H. Phan1a)

1Department of Physics, University of South Florida, Tampa, Florida 33620, USA 2School of Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, China 31D Nanomaterials Group, National Institute for Material Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan

4Department of Physics, Chungbuk National University, Cheongju 361-763, South Korea 5Dpto. Fisica de la Materia Condensada, ICMSE-CSIC, Universidad de Sevilla, P.O. Box 1065, 41080 Sevilla, Spain

(Presented 5 November 2013; received 16 September 2013; accepted 4 November 2013; published online 10 February 2014)

We have studied the impact of structural disorder on the magnetic ordering and magnetocaloric response of amorphous Gd68Ni32 and Gd53Al24Co20Zr3 microwires. We find that the presence of structural disorder significantly broadens the paramagnetic to ferromagnetic (PM-FM) transition and the temperature-dependent magnetic entropy change, while the nature of the second-order magnetic transition and long-range ferromagnetic order are not essentially affected by this effect. The large magnetic moment of Gd and the presence of the long-range ferromagnetic order are believed to result in a large magnetic entropy change, which together with the broadening of the PM-FM transition due to structural disorder contribute to a large refrigerant capacity. The excellent magnetocaloric properties of the amorphous microwires make them very promising candidates for active magnetic refrigeration. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.106371.4864143]

Magnetic refrigeration based on the magnetocaloric effect (MCE) is considered to be a viable alternative to conventional gas compression refrigeration technologies.1 Generally magnetic materials exhibiting large MCE (e.g., the large isothermal magnetic entropy change or the large adiabatic temperature change) over a wide temperature range are promising magnetic refrigerants. In addition, it is desirable for a magnetic refrigerant to have minimal magnetic hysteresis and eddy current losses. In this context, exploring the MCE in soft ferromagnetic amorphous materials is of practical importance,1-4 since these materials exhibit negligible magnetic hysteresis and possess reduced eddy current losses as compared to their crystalline counterparts. In particular, those developed recently in the form of microwires are very interesting, as they show large MCE and large refrigerant capacity (RC).5,6 Relative to their bulk counterparts, these microwires possess enhanced surface areas desirable for heat transfer, and a magnetic bed made of these microwires is highly preferable for engineering actual magnetic regenerators.6-8 Apart from their technological relevance, the magnetic structures in amorphous materials are often complicated, mainly due to the presence of structural disor-der.2 While the MCEs have been reported in various amorphous magnetic systems,1-8 the impact of structural disorder on the magnetocaloric response, such as the broadening of the magnetic ordering transition2 and the enhancement of the RC,6 remains to be investigated. Such knowledge is essential to gain better control over the material performance.

a)Authors to whom correspondence should be addressed. Electronic addresses: biswas.anis@gmail.com, faxiang.qin@gmail.com, scyu@chungbuk.ac.kr, vfranco@us.es, sharihar@usf.edu, and phanm@usf.edu

To shed some light on this important issue, we have performed a systematic study of the impact of structural disorder on the magnetic ordering and magnetocaloric response of amorphous Gd68Ni32 (sample A) and Gd53Al24Co20Zr3 (sample B) microwires.

The microwires were fabricated using a home-built-melt-extraction technique.6 The X-ray diffraction patterns confirmed the amorphous nature of the fabricated micro-wires. The average diameter of the wires was determined from scanning electron microscopy to be ^50 im. A quantum design physical property measurement system equipped with a vibrating sample magnetometer probe was used to investigate the magnetic and magnetocaloric properties of the fabricated microwires.

The temperature dependence of magnetic susceptibility [v(T)] reveals a paramagnetic to ferromagnetic (PM-FM) transition at TC ^122 K and ^97 K for sample A and sample B, respectively (Fig. 1(a)). We studied the magnetic field dependence of magnetization [M(H)] at different temperatures (T) in the magnetic field range of 0-30 kOe. As an example, the isothermal M(H) curves of sample A are shown in the inset of Fig. 1(a). From the isothermal M(H) curves, DSM was calculated using the Maxwell's relation1 DSM

= 10 JH"" (fir) dH, where M is the magnetization, H is the

magnetic field, and T is the temperature. Figure 1(b) shows the temperature dependence of DSM for sample A for different magnetic fields, and the inset of Fig. 1(b) shows the DSM(T) curves for both sample A and sample B for 10DH = 30 kOe. Both samples exhibit large DSM around their TC. In fact, for i0DH = 30 kOe, the maximum value of DSM near TC (DSMax) is ^4.5 J/kg K for sample A, while it is

0021-8979/2014/115(17)/17A318/3/$30.00

115, 17A318-1

(©2014 AIP Publishing LLC

master curve, when ASM is normalized to ASM* and the tem-

FIG. 1. (a) Temperature dependence of magnetic susceptibility at a field of 500 Oe for sample A and sample B. Inset shows the isothermal M(H) curves of sample A; (b) temperature dependence of —ASM at different fields for sample A. Inset shows the temperature dependence of —ASM for i0AH = 30 kOe for samples A and B.

perature axis is rescaled as

-(T - 7c)/(7M - 7C) T < TC (T - Tc)/(Tr2 - Tc) T > TC,

where Tr1 and Tr2 are two reference temperatures below and above TC satisfying the relation, ASM(Tr1) = ASM(Tr2) =f x ASM*, with f = 0.5 for this study. It has been pointed out that the existence of a universal behavior of - ASM(T) is a conclusive proof of the SOPT nature.10 For the FOPT, however, such a universal curve cannot be constructed. As shown in Figs. 2(a) and 2(b), a universal behavior does hold for both the microwire samples, confirming the SOPT type of the materials. This result clearly indicates that the presence of structural disorder in the amorphous microwires broadened the PM-FM transition and the -ASM(T) curve, while preserving the nature of the SOPT transition.

It is widely accepted that the magnetism of Gd-based crystalline alloys is dominated by the Ruderman-Kittel-Kasuya-Yosida (RKKY) interactions, which is long-range ferromagnetic in nature and often described by the mean-field theory.2 However, amorphous Gd65Mn35-xGex (x = 0, 5, and 10) alloys were reported to show short-range ferro-magnetism, where the mean-

is therefore essential to understand how the structural disorder impacts the magnetic ordering in our amorphous microwires. In an earlier study, using the mean-field methodology, Oesterreicher and Parker have shown that ASM follows a power law dependence of magnetic field: ASM ~ Hn with n

л cmax ASM

7J/kg K for sample B (inset, Fig. 1(b)). These values of are considerably larger than those reported for other microwires.5,7 For i0AH = 30 kOe, the ASMax for Gd68Ni32 is almost 1.5 times larger than that obtained for a bulk amorphous Gd70Ni30 alloy.2 Besides ASM, RC—an important fig-

ure of calculated as

of a RC =

magnetic refrigerant material—was

JT2ASM(T)dT, where T1 and T2 are the temperatures corresponding the full width at half maximum of a ASM(T) curve. For i0AH = 30 kOe, the RC values of sample A and sample B are determined to be ^322J/kg and ~415 J/kg, respectively. The large values of RC make the present microwires very promising candidates for active magnetic refrigeration.8

Now, we attempt to elucidate the effect of structural disorder on the nature of magnetic phase transition of the microwires. Conventionally, the type of magnetic phase transition is determined from Arrott plots using Banerjee's criterion.9 However, if a first-order phase transition (FOPT) is too weak to have a significant impact on the free energy derivative at the transition, this method can be insufficient to discriminate FOPT from the second-order phase transition (SOPT).10 On the other hand, for ferromagnetic systems undergoing SOPT, a universal curve can be constructed to describe - ASM(T) at different H. All -ASM(T) curves obtained for different H can be collapsed into a universal

FIG. 2. Universal ASM/ASJMax vs. в curve for (a) sample A and (b) sample B. Inset shows the -ASM* vs. curves for (1) sample A and (2) sample B.

« 2/3 at a transition temperature. A more general framework has been formulated later by Franco et al. to find out the local exponent n for any ferromagnetic materials obeying the mean-field theory.11 According to this approach, n can be

associated with d and b as n = 1 + — jj . According to

the mean-field theory, the values of b, C, and d should be 0.5, 1, and 3, respectively, yielding n = 2/3.13 We have examined the magnetic field dependence of ASM near TC for both samples and obtained a linear relationship between —ASJ

H2/3 with an intercept at the —ASMax axis (see inset of Fig. 2(b)). This result suggests that our present samples follow the mean-field theory, with long-range ferromagnetic interactions.

Using the mean-field approach, the magnetic entropy S(r) of a ferromagnetic system can also be expressed as14

-AS(r)

1 NKB(a2

where N is a number of spins, J is a spin value, kB is the Boltzmann constant, r is the reduced magnetization, rspont is the reduced spontaneous magnetization. From Eq. (2), it is obvious that AS vs. r2 plots below TC must have a horizontal

FIG. 3. Comparison between the spontaneous magnetization (Msp) at different temperatures calculated from the magnetocaloric data and Arrott plots for (a) sample A and (b) sample B. Insets show the —ASM vs. M2 plots for selected temperatures.

drift from the origin, which corresponds to aspnnt. Similarly, if a —ASM vs. M2 plot for a ferromagnetic system shows a linear dependence with a constant slope and a horizontal drift from the origin below TC, it can be assumed that Eq. (2) and so the mean-field theory is valid for that system. Furthermore, it is possible to determine its spontaneous magnetization (Msp) from the horizontal drift of — ASM vs. M2 curves from the origin.

For samples A and B, the — ASM vs. M2 curves at different temperatures are linear with nearly constant slope in the entire temperature range below TC (the non-linearity of the curve is only observed at very low fields when magnetic domains start to be formed), as shown in the insets of Figs. 3(a) and 3(b) for selected temperatures. We calculated Msp at different temperatures from the horizontal shifts of the curves, the results of which are in excellent agreement with those obtained from the conventional method using Arrott plots (see Figs. 3(a) and 3(b)). It appears that in the paramagnetic region (T > TC), the — ASM vs.

curves pass through the origin as Msp do not exist. This further proves that the present microwires obey the mean-field theory and that the long-range ferromagnetic interactions occur in these systems.

In summary, we have demonstrated that the presence of structural disorder significantly broadens the magnetic transition and the temperature-dependent magnetic entropy change in amorphous Gds8Ni32 and Gd53Al24Co20Zr3 microwires. The large magnetic moment of Gd and presence of the long-range ferromagnetism are believed to retain the large magnetic entropy change, which, together its large temperature distribution caused by the structural disorder, contributes to the large refrigerant capacity in the amorphous microwires.

Research at USF was supported by the U.S. Department of Energy under Award No. DE-FG02-07ER46438 (magnetic studies). Research at CBNU was supported by the Converging Research Center Program funded by the Ministry of Education, Science and Technology under Award No. 2013K000405. V.F. acknowledges the support of the Spanish Ministry of Science and Innovation and EU FEDER (Project No. MAT2010-20537), the PAI of the Regional Government of Andalucía (Project No. P10-FQM-6462), and the United States Office of Naval Research (Project No. N00014-11-1-0311). F.X.Q. was supported under JSPS Fellowship Program.

'V. Franco et at, Annu. Rev. Mater. Res. 42, 305 (2012). 2X. Y. Liu et at., J. Appl. Phys. 79, 1630 (1996). 3Q. Luo et at, Appl. Phys. Lett. 89, 081914 (2006). 4R. Caballero-Flores et at., Appl. Phys. Lett. 96, 182506 (2010). 5R. Varga et at, Scr. Mater. 65, 703 (2011). 6F. X. Qin et at., Acta Mater. 61, 1284 (2013). 7C. R. H. Bahl et at., Appl. Phys. Lett. 100, 121905 (2012). 8M. D. Kuz'min, Appl. Phys. Lett. 90, 251916 (2007). 9S. K. Banerjee, Phys. Lett. 12, 16 (1964). 10C. M. Bonilla et at, Phys. Rev. B 81, 224424 (2010). 11V. Franco et at., Appl. Phys. Lett. 89, 222512 (2006). 12X. C. Zhong et at, J. Appl. Phys. 112, 033903 (2012). 13H. Oesterreicher and F. T. Parker, J. Appl. Phys. 55,4334 (1984). 14J. S. Amaral et at., J. Magn. Magn. Mater. 322, 1569 (2010).