Abstract

Metamagnetic NiMn-based Heusler alloys are one of the promising rareearth-free magnetocaloric materials due to their attractive large magnetocaloric effect (MCE) arising from their first order phase transition (FOPT). Both austenitic and martensitic states of Ni 0.50 Mn 0.5-y In y are ferromagnetic for composition of $0.15 \le \mathrm {y}\le 0.16$ [1,2]. For Ni 0.50 Mn 0.34 In 0.16 , it exhibits an inverse MCE of ~ 12 J kg-1 K -1 at 190 K, which is far away from room temperature [2]. On the other hand, for y = 0.15 in Ni 0.50-z Mn 0.5+z In 0.15 , its field-induced martensitic transformation shifts to 285 K[3]. Most efforts in optimizing the performance of this alloy series have been dedicated to compositional studies and mainly reported only the temperature range of its inverse MCE due to the interest in giant MCE for practical applications. However, these ferromagnetic alloys exhibit at least three temperature-induced phase transitions: a low temperature ferromagnetic-paramagnetic (FM-PM) transition followed by a first order martensitic structural transition at higher temperatures and the FM-PM transition of high temperature austenitic phase. In this work, we aim to study, in a broad temperature range, the influence of these three overlapping transitions on the field dependence of MCE in Ni 0.49-x Mn 0.36+x In 0.15 . Fig 1(a) shows the magnetization vs. temperature (M(T)) data of Ni 48.1 Mn 36.5 In 15 which reveals that it exhibits three phase transitions occurring at different temperatures as observed from the $\partial M/ \partial T$ slopes. The first one corresponds to the Curie transition of the martensitic phase, producing a small gradual decrease of M in the range of Ts C of the martensite phase. At higher temperatures, a gradual decrease of $\partial M/ \partial T$ is observed. These are in agreement with the 3D MCE (magnetic entropy change, ΔS M ) plot shown in Fig 1(b) whereby two direct MCE and an inverse MCE are observed. A small direct MCE is observed at low temperatures up to ~ 220 K, followed by a large inverse MCE at higher temperatures up to ~ 268 K due to the subsequent competing phase transitions followed by a direct MCE peak. While the martensitic-austenitic transition does not significantly overlap with the Curie transition of the austenite, the direct MCE of that FM-PM transition exhibits a magnitude of the same order of that of the inverse MCE (7.13 J kg-1 K-1 versus -2.83 J kg-1 K-1 respectively). However, the low temperature direct MCE is rather diminished due to the overlapping of its Curie transition and the martensitic transition, whereby the martensite phase disappears before its T C . Fig 2 shows that the variant electron/atom (e/a) ratio (due to alteration of Ni/Mn ratio) shifts the different magnetic phase transitions in the Ni 49+x Mn 36-x In 15 alloys. With increasing e/a ratio, the T pk of 1st direct MCE (1st transition), inverse MCE (2nd transition) and direct MCE at higher temperatures (3rd transition) regimes exhibit an increase, increase and decrease respectively. The T pk of the inverse MCE agree with the temperature at which the austenite phase begins to form, as observed from DSC data. For both 2nd and 3rd transitions, it can be observed that both exhibit similar trends of increased T pk for e/a ratio ≥ 7.86, indicating their phase competition, which also affected the visibility of their peaks as shown in the inset of Fig 2 (their peak Δ S M remain rather small for e/a ratio ≥ 7.86). Hence, the peak Δ S M values of the 2nd and 3rd transitions are comparable (i.e. same order of magnitude) when their temperatures are well separated, as observed for e/a ratio ≤ 7.84. Although it is previously found that the entropy change of the phase transition from DSC data correlates well with the e/a ratio[4], there is no direct correlation between the ΔS M of the martensitic transition and the e/a ratio partly due to this overlap. The previously described behavior for Ni 48.1 Mn 36.5 In 15 agrees well with the observed trend of exponent n(as in ΔS M α Hn [5]) controlling the field dependence of MCE. For its first transition, n gradually decreases from n~1 and abruptly increases as the sign of ΔS M switches at the FOPT (2nd transition). For the 3rd transition, the typical n behavior for a second order phase transition is observed and the main features of the field dependence of each transition are essentially maintained.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call