Elsevier

Acta Materialia

Volume 106, March 2016, Pages 15-21
Acta Materialia

Full length article
Giant adiabatic temperature change in FeRh alloys evidenced by direct measurements under cyclic conditions

https://doi.org/10.1016/j.actamat.2015.11.054Get rights and content

Abstract

We report on (1) direct measurements of ΔTad for binary Fe49Rh51 during field cycling and (2) maximum possible ΔTad measured under discontinuous protocol. Our results show that the ΔTad is 9.2 K on the first application of magnetic field of Δμ0 H = 1.9 T and it remains as high as 6.2 K during the cycling in alternated field of the same magnitude. In addition, the adiabatic temperature change and magnetic entropy change under the first application of magnetic field and under cyclic conditions were determined indirectly using three different approaches: (1) from magnetic measurements (M(T)H dependences, Maxwell relations), (2) from calorimetry (C(T)p,H, S-T diagram) and (3) from H-T diagram. While the indirectly measured maximum possible ΔTad lies in the range of 10.5–12 K, the reduced value of ΔTad measured directly under cycling (6.2 K) is still extraordinarily high and is 15% higher than in Gd in similar magnetic field. This demonstrates the potential of materials with a first order metamagnetic transition for magnetocaloric applications despite the presence of hysteresis.

Introduction

Due to environmental concerns and the limited efficiency of traditional cooling technology based on the vapour compression cycle, there has been much recent research dedicated to solid state refrigeration based on the magnetocaloric effect (MCE) [1], [2]. This has the potential to increase the efficiency of cooling devices by approximately 20% and to eliminate the need for volatile refrigerants entirely. Magnetic materials with a first-order field-induced phase transition are of particular interest for this emerging technology, since they display both giant magnetic entropy change, ΔSm, and significant adiabatic temperature change, ΔTad. The disadvantage of the first-order type materials is the presence of thermal and magnetic hysteresis which may significantly reduce the effective value of the entropy change over repetitive field applications. For correct evaluation of the MCE in such materials it is crucial to carry out measurements not only under the first but also under the second application of the magnetic field [3]. Unfortunately, such data are rather scant in the literature and this work aims at a comprehensive analysis of the MCE in first-order type materials under cyclic conditions by using three different techniques: (1) direct measurements of ΔTad, (2) calorimetry and (3) measurements of the field or temperature dependences of magnetization with subsequent calculation of ΔSm using the thermodynamic Maxwell relations.

Among the magnetocaloric materials with the first-order magnetic phase transitions, FeRh alloys hold a special place. It has been reported that the ΔTad in FeRh is 12.9 K in Δμ0 H = 1.95 T [4] or 7.9 K in Δμ0 H = 1.9 T [5] obtained by direct measurements, and these are some of the highest values ever recorded for any materials under magnetic field change up to 2 T. This discovery led to the suggestion of using FeRh as a magnetic refrigerant [6]. Although the raw material costs make it unlikely implementing FeRh in bulk form in a real magnetic refrigerator, this alloy is still interesting from the fundamental point of view as a model magnetocaloric material for solid state refrigeration near room temperature. In recent years, the study of FeRh has been intensified due to other potential technological applications which require thin films, such as heat assisted magnetic data storage media [7] and spintronic devices [8], [9]. Due to small amounts of Rh used in thin films and the high price of the final high-tech product, the price of raw materials will be of less significance allowing considering FeRh as a prospective magnetic material for these applications.

The giant MCE observed in FeRh occurs at an antiferromagnetic (AF) to ferromagnetic (FM) phase transition around Tt = 360 K on heating and the reverse FM to AF transition takes place around Tt = 350 K on cooling. With the considerable hysteresis attributable to the AF-FM first-order transition, it is important to know how FeRh performs under repetitive magnetic field application. Indeed, in the literature FeRh is often compared to other prospective magnetocaloric materials, and the MCE value in FeRh considerably exceeds the MCE values in other materials [10], [11], [12]. On the other hand, all the other candidate materials have a narrow temperature hysteresis and a high reversible MCE. So far, direct evidence of ΔTad reversibility in FeRh has not been reported. The giant ΔTad of 6.6 K/T [3] or 4.2 K/T [5] were obtained by using so-called discontinuous protocol [3], [12]: in order to remove the effect of thermal hysteresis, the specimen was brought to the antiferromagnetic state by zero field cooling down to ∼270 K and then the specimen was heated up to a certain target temperature and only then the ΔTad was determined by applying magnetic field. This approach allows determining the maximal possible value ΔTad, however, despite the giant MCE values that were obtained, ΔTad values at repetitive and subsequent field applications (reversible MCE) are unknown.

Initially, it was stated that the MCE in FeRh is totally irreversible and shows only on the first application of the field [1]. This idea was later questioned in Ref. [13] and investigated with magnetic measurements indirectly; it was shown that the giant ΔSm in FeRh does not vanish after the first field cycle. These considerations were confirmed directly by calorimetry measurements [14] where it has been shown how to determine the region of reversibility of the ΔSm. The high value of ΔSm is indispensable but not sufficient for a magnetic material to be a good magnetic refrigerant. Therefore, the adiabatic temperature change ΔTad should be taken into account. Also it is necessary to note that the timescale of calorimetric and magnetic measurements used for ΔSm evaluation cannot compare with real performance of a magnetic cooling device (∼10 T/s). Direct measurements of the adiabatic temperature change (ΔTad) are, however, in the right range, and essentially the only method to reveal true values of the MCE as in a working magnetic refrigerator.

In this work, we demonstrate with a quasi-dynamic method of direct measurements of ΔTad and conventional determination of ΔS from quasi-static magnetic measurements that even under cyclic conditions, although ΔTad in FeRh is reduced compared to the initial value, it is still higher than in Gd, the benchmark magnetocaloric material commonly used in prototypes of magnetic refrigerators.

Section snippets

Experimental details

A bulk sample of Fe49Rh51 composition was prepared by arc melting of pure elements Fe (99.98%), Rh (99.8%) in helium atmosphere (10−4 mbar) in a water-cooled Cu-crucible. The sample was subjected to homogenizing annealing in an evacuated quartz ampoule for a week at 1000 °C followed by air quenching. For further experiments a plate of approximate size 5*3*1 mm3 was cut from the ingot.

Bruker D8 Advance diffractometer was used for the X-ray diffraction (XRD) analysis at room temperature. Scanning

Magnetization measurements

According to X-ray diffraction, scanning electron microscopy and energy dispersive x-ray spectroscopy the sample contains 97 vol% of CsCl-type phase of the composition Fe49.3Rh50.7 with the lattice parameter 2.99 Å. A small amount of an fcc phase of the composition Fe37.5Rh62.5 with the lattice parameter 3.76 Å is also present. This second phase often appears to have been ignored in other publications on FeRh.

The temperature dependences of magnetization, M(T)H, in various magnetic fields up to

Conclusion

Correct estimation of the large magnetocaloric effect (MCE) in materials experiencing a first order magnetic transition is important for development of magnetocaloric refrigerants to be used in solid state cooling. Experiments under cyclic conditions close to those in a potential magnetic refrigerator are necessary in order to avoid the overestimation that happens when only the first application of the magnetic field is considered. In this work, using indirect methods, the maximal possible

Acknowledgements

This work was partly supported by the Program of the Ural Branch of RAS (Project No 15-17-2-22). K.S. gratefully acknowledges the financial support of the Ministry of Education and Science of the Russian Federation in the framework of Increase Competitiveness Program of NUST ‘MISiS’ (K3-2015-029). O.G. thanks the DFG (SPP1599).

References (27)

  • I.A. Radulov et al.

    On the preparation of La(Fe,Mn,Si)13Hx polymer-composites with optimized magnetocaloric properties

    J. Magn. Magn. Mater.

    (2015)
  • O. Gutfleisch et al.

    Magnetic materials and devices for the 21st century: stronger, lighter, and more energy efficient

    Adv. Mater

    (2011)
  • K.P. Skokov et al.

    Magnetocaloric materials with first-order phase transition: thermal and magnetic hysteresis in {∖mathrm{LaFe}_{11.8}∖mathrm{Si}_{1.2}} and {∖mathrm{Ni}_{2.21}∖mathrm{Mn}_{0.77}∖mathrm{Ga}_{1.02}} (invited)

    J. Appl. Phys.

    (2012)
  • Cited by (161)

    View all citing articles on Scopus
    View full text