History dependence of directly observed magnetocaloric effects in (Mn, Fe)As

Abstract

We use a calorimetric technique operating in sweeping magnetic field to study the thermomagnetic historydependence of the magnetocaloric effect (MCE) in Mn 0.985 Fe 0.015 As. We study the magnetization history for which a "colossal" MCE has been reported when inferred indirectly via a Maxwell relation. We observe no colossal effect in the direct calorimetric measurement. We further examine the impact of mixed-phase state on the MCE and show that the first order contribution scales linearly with the phase fraction. This validates various phase-fraction based methods developed to remove the colossal peak anomaly from Maxwell-based estimates.The first order nature of the Curie transition in MnAs has been of interest for many years and was the motivation behind Bean and Rodbell's well-known model of first order magnetic phase transitions 1 . Recently MnAs has been in the spotlight again after the report of a "colossal" magnetocaloric effect (MCE)-an order of magnitude larger than usual 2 . It, along with anomalous peaks in the MCE of other systems, was inferred indirectly from magnetization measurements via a Maxwell relation and disputed because of a comparison with calorimetric measurements showing no colossal effect 3,4 . The calorimetric measurements, however, follow a significantly different thermomagnetic history. Here we use a calorimetric technique operating in a field-sweeping mode to reproduce several thermomagnetic histories and thus investigate the possibility of observing an MCE that depends on the specific magnetization history.The MCE can be quantified as an adiabatic temperature change, ∆T ad , or an isothermal entropy change, ∆S, for a certain magnetic field variation. The latter can be estimated via a Maxwell relation:where M is the magnetization and B is the magnetic flux density, which we later assume to be equal to µ 0 H. The use of this relation requires a knowledge of magnetization as a function of B and T and it is usually obtained from isothermal magnetization measurements taken in discrete temperature steps. It is important to note that the Maxwell relation is based on equilibrium thermodynamics, i.e. assuming that magnetization and entropy are single-valued functions of state. This is true in materials with a continuous phase transition, however, first order phase transitions usually display finite hysteresis due to the presence of multiple local minima in the free a) m.bratko09@imperial.ac.uk energy separated by an energy barrier. This enables the existance of metastable, non-equilibrium states. Nevertheless, the Maxwell relation is often used, assuming that the effect of a small and finite region of irreversibility is negligible.A "colossal" MCE was first reported in MnAs under hydrostatic pressure 2 followed by its observation in Mn 1−x Fe x As 5 and Mn 1−x Cu x As 6 at ambient pressure. The effect was later reported even in MnAs at ambient pressure 7 , in disagreement with previous reports 2,8 . The authors in Ref. 7 claimed the capture of a colossal effect was possible due to smaller t...

Determining the first-order character ofLa(Fe,Mn,Si)13

“…27 In order to estimate the latent heat contribution to the total entropy change from the magnetization data we have used the Clausius-Clapeyron equation:…”

Effect of grain constraint on the field requirements for magnetocaloric effect in Ni45Co5Mn40Sn10 melt-spun ribbons

“…This procedure is known as discontinuous heating and prevented erroneously large entropy changes from the post-processing in the magnetization data as described elsewhere. 22,34–36 …”

Strong magnetocaloric effect induced by spin reorientation transitions in epitaxial Ho thin films

“…The value of the MCE strongly depends on the measurement procedure of the isothermal M(H) curves [23]. Driving the sample through the phase transition in a mixed-phase state or stable state leads to a different value of MCE [23].…”

“…The value of the MCE strongly depends on the measurement procedure of the isothermal M(H) curves [23]. Driving the sample through the phase transition in a mixed-phase state or stable state leads to a different value of MCE [23]. For instance, in this case, when the sample is abruptly cooled down to a lower temperature by 5 K or when heated to room temperature before next measurement the values of M(H), and MCE differ by about 8 times [23].…”