Ferri- to ferro-magnetic transition in the martensitic phase of a Heusler alloy

https://doi.org/10.1016/j.jallcom.2012.02.062Get rights and content

Abstract

During the past decade the magnetic properties of Heusler alloys have been extensively studied, motivated in part by the observation of large magnetocaloric effects (MCEs) displayed by these alloys near room temperature. We present new data and develop a consistent mechanism to explain the complex hysteretic behavior of a Ni50Mn35In15 Heusler alloy. The magnetization of this alloy is characterized by two critical temperatures. Below the lower critical temperature, the alloy is a ferrimagnetic martensite. Between the two critical temperatures, the alloy is a ferromagnetic martensite. Above the higher critical temperature, it is a paramagnetic austenite. The transitions at both critical temperatures are first order. The ferri-to-ferromagnetic transition and the crystallographic martensite-to-austenite transition explain the various facets observed in the MZFC and MFC vs. T plots and their variations with increasing magnetic field. The model successfully explains the isothermal M vs. H loops near room temperature, whose behavior is strongly dependent on the initial magnetic state.

Highlights

► Ni50Mn35In15 Heusler alloy exhibits large MCE peaks near room temperature. ► The alloy's magnetic properties have been studied for refrigeration applications. ► New magnetic data leads to a new model to explain the alloy's complex behavior. ► This paper is controversial because our model differs from the previous model. ► We provide new perspectives on the Heusler alloys complex magnetic behavior.

Introduction

The magnetic properties of Heusler shape memory alloys have been the subject of many studies. These studies have been motivated in part by the observation of large magnetocaloric effect (MCE) peaks displayed by these [1], [2], [3], [4], [5], [6] and related [7], [8], [9] alloys near-room temperature. The structure and magnetic properties of the stoichiometric Ni50Mn25Ga25 alloy have been of particular interest to a number of research groups. They concluded that this alloy, on heating, undergoes a first-order magnetocrystalline transition from tetragonal martensite to a cubic austenite structure at a transformation temperature, TM, ranging from 175 K to 220 K; followed by a second-order ferromagnetic–paramagnetic transition identified with a Curie temperature of the austenite phase, TC, between 375 K and 380 K; the transitions are reversible with temperature [10], [11]. Later it was found [12], [13], [14], [15], [16] that the supposed transition temperatures, TM and TC, could made nearly coincident either by doping the alloy with Co or Cu or by slightly varying the alloy composition in the off-stoichiometric form of Ni50+XMn25−XYGa25+Y, with X  5 and Y  l–2. The isothermal magnetization versus field loops displayed large hysteresis losses and the magnetization characteristics indicative of a field-induced magneto-structural phase transformation [17], [18], [19], [20], [21], [22].

More recently, the structure and the magnetic properties of the off-stoichiometric Heusler alloy and related alloys have generated much interest because their unusual and complex magnetic properties [23], [24], [25], [26], [27], [28], including the presence of a larger inverse magnetocaloric effect (MCE) peak [29] and a smaller conventional peak both occurring near room temperature as well as the display of large magnetoresistance at moderate field values [30]. Similar to what had been previously proposed for the Co or Cu-doped Ni50Mn25Ga25 and the off-stoichiometric Ni50Mn25Ga25 alloys, the presence of the inverse and conventional MCE peaks observed in the Ni50Mn35In15 alloy is believed to be the result of the near-coincidence of two transformation temperatures both close to room temperature. That is, on heating, the Ni50Mn35In15 alloy has been assumed to undergo a first-order magneto-structural martensitic transformation from tetragonal martensite to ferromagnetic cubic austenite at TM, followed by the ferromagnetic-to-paramagnetic second-order magnetic transition of the austenite phase at its Curie temperature, TC [29], [30], [31], [32], [33].

The data presented by [34] is more appropriately interpreted (on heating) as a first order ferri-to-ferro-magnetic phase transition within the martensitic phase, followed by a magneto-structural transition from a martensitic to an austenitic phase at which the material becomes paramagnetic. The more recent studies by Bhobe et al. [35], [36] on the phase structure and magnetic properties of the Ni50Mn35In15 alloy by X-ray absorption fine structure (EXAFS) and SQUID magnetometry measurements seems to confirm that this alloy displays a cubic austenite B2 structure above 305 K and a tetragonal L21 martensite structure below 302 K. The Ni50Mn35In15 alloy undergoes a crystallographic phase transition, the details of which are affected by minute differences in the alloy composition, sample preparation, and heat treatment; however, the assumed near coincidence of TM and TC temperatures does not adequately account for the details of the complex magnetic behavior observed in this alloy as a function of both temperature and field. In this paper we present new magnetic data and provide a new a mechanism that accounts for the various features of this complex behavior.

Section snippets

Experimental methods

The Ni50Mn35In15 alloy samples used for this study were prepared by arc melting appropriate amounts of the component elements using a water-cooled copper hearth in an argon atmosphere under ambient pressure. The sample was then homogenized for 2 h at 800 °C in an evacuated quartz tube and then quenched in ice water.

The microstructure, chemical composition, and phase structure of the alloy were then examined by energy dispersive spectroscopy (EDS), and X-ray diffraction, respectively. The EDS

Results and discussion

Fig. 1 is a summary of the X-diffraction data obtained on the Ni50Mn35In15 alloy between 273 K and 388 K. The room temperature (298 K) data reveal a primary austenite phase with some amount of the martensite phase (Fig. 1a and b). The martensite phase persists in the sample on heating at least up to 363 K (Fig. 1b), whereas cooling the sample below room temperature results in the near disappearance of the austenite phase around 273 K (Fig. 1a). Therefore, the X-ray data presented in Fig. 1 clearly

Summary and conclusions

Based on our experimental data, a new interpretation for the magnetic behavior of the well-studied Ni50Mn35In15 Heusler alloy is presented. A magnetization peak previously identified as a transformation to a ferromagnetic austenite phase is reinterpreted as a spin flipping peak occurring within the martensite phase and representing a Kittel transformation [38]. This Kittel transformation, resulting in a rapid increase in the magnetization value, is the same phenomenon that gives rise to the

References (43)

  • R. Tickle et al.

    J. Magn. Magn. Mater.

    (1999)
  • E. Brück et al.

    Physica B

    (2003)
  • I. Babita et al.

    J. Alloys Compd.

    (2007)
  • O. Tegus et al.

    Physica B

    (2002)
  • A.O. Pecharsky et al.

    Alloys Compd. J.

    (2002)
  • V.K. Pecharsky et al.

    J. Magn. Magn. Mater.

    (1999)
  • K. Prokeš et al.

    J. Alloys Compd.

    (2002)
  • T.T.M. Palstra et al.

    J. Magn. Magn. Mater.

    (1983)
  • V. Basso et al.

    Physica B: Condens. Matter

    (2008)
  • F-X. Hu et al.

    Appl. Phys. Lett.

    (2000)
  • P.J. Webster et al.

    Philos. Mag. B

    (1984)
  • A. Sozinov et al.

    Appl. Phys. Lett.

    (2002)
  • X. Zhou

    J. Phys. Condens. Matter

    (2004)
  • T. Krenke et al.

    Phys. Rev. B

    (2006)
  • Y. Tokura et al.

    J. Appl. Phys.

    (1996)
  • W.H. Wang et al.

    IEEE Trans. Magn.

    (2001)
  • M. Khan et al.

    J. Appl. Phys.

    (2003)
  • S. Stadler et al.

    Appl. Phys. Lett.

    (2006)
  • F.-X. Hu et al.

    Phys. Rev. B

    (2001)
  • L. Pareti et al.

    Eur. Phys. J. B

    (2003)
  • F. Albertini et al.

    Appl. Phys. Lett.

    (2002)
  • Cited by (0)

    View full text