Mastering a 1.2 K hysteresis for martensitic para-ferromagnetic partial transformation in Ni-Mn(Cu)-Ga magnetocaloric material via binder jet 3D printing

Abstract

Magnetocaloric (MC) materials have gained traction in the research and industry communities for their prospects in solid state magnetic refrigeration. Important to the commercialization of MC materials are: (1) establishment of a fabrication method that can combine high surface area for heat transfer and geometric freedom for designing an efficient heat exchanger which has low pressure drop for the coolant and (2) advancement of low cost alloys with appropriate MC properties. In this regard, additive manufacturing may provide the geometric freedom necessary for adapting designs to solid state cooling, and the Ni-Mn(Cu)-Ga Heusler ferromagnetic shape memory alloys (FSMAs), exhibiting a martensitic para-ferromagnetic transformation at T_ms=304 K, can provide a low-cost MC material, very promising for magnetic cooling. In this study, a Ni_49.5Mn_19.1Cu_6.6Ga_24.8 (at.%) alloy is additively manufactured using powder bed binder jet 3D printing with subsequent sintering. This printed and sintered material enabled a large change of magnetization during partial transformation cycles with the smallest temperature hysteresis recorded for FSMAs, equal to about 1.2 K, regardless of the value of magnetic field applied. Under 2 T and at 304 K it exhibits an adiabatic temperature change (ΔT_ad) of 2 K and a stable cycling behavior of ΔT_ad = │1.7│ K for 100 cycles. The maximum of magnetic field-induced entropy change |ΔS_{m, 2T}| ≈ 12.0 J/kg·K was estimated at 304 K. These results demonstrate the viability of powder bed binder jet 3D printing as an effective fabrication method for functional magnetocalorics, as well as the outstanding MC characteristics of a low-cost Ni-Mn(Cu)-Ga Heusler-type FSMA.

Introduction

Worldwide energy consumption is increasing, and a large portion of this growth is generated by cooling technologies such as refrigerators and air conditioners [1]. Currently, refrigeration is produced through a vapor-compression cycle technology that has a maximum Carnot efficiency of 30–40 %, can be loud, and uses chlorofluorocarbons (CFC’s) and hydrochlorofluorocarbons (HCFC’s), hazardous gases which contribute to ozone depletion [2,3]. Recently, the possibility of using solid state magnetic refrigeration, based on the magnetocaloric effect (MCE), has been proposed as an alternative technology. Magnetic refrigeration presents several advantages compared to the vapor-compression cycle: it can be 20–30 % more efficient than the current technology, it is nearly silent because of the limited need for moving parts like the compressor in a typical system, and it is eco-friendly [2]. Nevertheless, many of the established MCE materials contain rare earth elements, such as Gd and La, which are economically strategic materials, and also present potential health risks to populations near to their mining sites and other occupational exposures [4,5].

Recently, two categories of the rare-earth-free Ni-Mn-based (In, Sn, Sb, Ga) Heusler-type magnetic shape memory alloys (SMAs), exhibiting a giant MCE due to the first order martensitic transformation (MT) near room temperature, have been researched: Mn-rich Ni-Mn-(In, Sn, Sb) metamagnetic SMAs (MetaMSMA) [[6], [7], [8], [9]] and Ni-Mn-Ga ferromagnetic SMAs (FSMAs) [10]. MetaMSMAs display a large drop of magnetisation (ΔM) in the temperature range of MT from the high-temperature ferromagnetic austenite to the low-temperature antiferromagnetic martensite, whereas in FSMAs the large drop of ΔM is obtained as a result of MT from the ferromagnetic martensite to the paramagnetic austenite. The adiabatically applied magnetic field induces these magnetostructural transformations producing cooling of MetaMSMAs, as a result of inverse MCE, and heating in the case of FSMAs resulting from a conventional MCE. The large values of the field-induced isothermal entropy change (ΔS_m) and corresponding adiabatic temperature change (ΔT_ad) are the main parameters characterizing the MCE.

On the other hand, the typically large thermal hysteresis of MT (around 10 K) is a serious disadvantage of these materials since a very high magnetic field is required to achieve reversibility of a magnetostructural transformation [[11], [12], [13]]. Therefore, an important research goal to make these materials competitive is a narrowing of the thermal hysteresis of MT, which can be achieved primarily by tuning compositions. The other possible method, still not explored extensively in literature, is the exploration of partial transformations (minor loops) that can increase reversibility and decrease hysteresis [14].

Numerous sources compiling the current state of magnetocaloric cooling assert the need for a thermally efficient heat exchanger that has a high surface-to-volume ratio, in order to maximize heat transfer capabilities [13,[15], [16], [17]]. Such a heat exchanger is a challenge when fabrication or assembly techniques affect materials’ functionality, leaving some alloys that show promise in laboratory testing to exhibit only a modest MCE after fabrication [18,19]. Currently, some of the attempted fabrication routes for magnetocaloric heat exchangers are: selective laser melting [18], powder packed bed [20], composite compaction [21], or polymer bonding [19,22]; each method has some limited success but none of them has emerged as the sole solution to the fabrication question. Typically, Ni-Mn-based Heusler SMAs are manufactured as single crystals, thin films, foams, and polycrystals. Additive manufacturing (AM) can add the great possibility of fabricating complex shapes with controlled, designed porosity as explored through binder jet printing and laser-based manufacturing. Research studies in the area of additively manufactured MCE materials are sparse. Moore et al. studied the selective laser melting of magnetocaloric alloy La(Fe, Co, Si)₁₃ through comparison of two different bulk geometries [18]. Laser powder bed fusion of Ni-Mn-Ga is described in the recent work by Laitinen et al. [23,24] and Nilsen et al. [25]. Mostafaei et al. and Caputo et al. studied binder jet 3D printing of Ni-Mn-Ga with compositions more suitable for purposes of the magnetic shape memory actuators than for potential magnetocaloric applications [[26], [27], [28], [29]]. Taylor et al. reported on the sintering of particle-based ink 3D printed elemental Ni, Mn, and Ga powder particle-containing inks [30]. Stevens et al. have introduced both direct laser deposition of Ni-Mn-Co-Sn MetaMSMAs and preliminary binder jet printing results for Ni-Mn(Cu)-Ga FSMA [31,32].

The present study explores a magnetocaloric material fabricated via binder jet 3D printing (BJ3DP), focusing on the Ni-Mn(Cu)-Ga FSMA, transforming martensitically near room temperature from the paramagnetic austenite into the ferromagnetic martensite. We found that the MT in this material can be realized with an extremely narrow thermal hysteresis of 1.2 K through a partial transformation still involving large values of ΔM, ΔS_m and ΔT_ad. The relatively small values of both hysteresis and transformation interval gave rise to the stable cycling amplitude of ΔT_ad equal to about 2 K under 2 T at 304 K.