Title:	Pair distribution functions in amorphous Fe-Ni-Al-Zr alloys
DNr:	SNIC 2020/5-599
Project Type:	SNIC Medium Compute
Principal Investigator:	Gabriella Andersson <gabriella.andersson@physics.uu.se>
Affiliation:	Uppsala universitet
Duration:	2020-12-01 – 2021-12-01
Classification:	10304
Keywords:

Abstract

Amorphous Fe-Zr-based alloys are suitable candidates for magnetic refrigeration due to the relationship between their magnetic and microstructural properties. Previous studies include FeZr with additions of Al, Si, Ga, Ge and Sn [1,2]. These systems have the advantage of low cost, being free of rare-earth elements. They can exhibit large changes in magnetic entropy close to room temperature at relatively low applied fields (< 1.5 T), while being free from magneto-structural phase transitions. Furthermore, they exhibit higher resistivity, larger values of refrigerant capacity, and more tunable Curie temperature, compared to rare-earth based alloys. In our recent experimental studies, Fe-Ni-Al-Zr (FNAZ) amorphous thin films with quaternary composition gradients were characterized by macroscopic magnetization measurements. For a specific FNAZ amorphous film, very promising values of order-disorder transition temperature and magnetic entropy change were observed. Thus, the material is a candidate for future magnetic refrigeration applications. In order to understand the role of atomic arrangements for the magneto-caloric behavior we need to link the short-range order and medium-range order of the constituents to the entropy change. The aim of the project is to, in an ab initio approach, investigate the structural properties of amorphous Fe-Ni-Al-Zr alloys of different compositions to examine how the distribution of atomic pairs influences the magnetic qualities of the material. To do this, we will use the stochastic quenching method to computationally produce atomistic models of the amorphous structures. This is accomplished by creating dozens of randomly generated starting structures which are then fed to the electronic structure code (VASP in this case) to determine from first principles the interatomic forces. Those are then utilized to alter the structure (i.e., atomic positions) to create a new structure with lower potential energy and overall reduced forces. The process is repeated until the maximal force on any atom is below a predefined threshold. The ensemble average of the thus prepared structures (based on different random starting conditions) is then taken to analyse and predict structural properties, such as the total and pair decomposed pair distribution function (PDF) and the atoms’ relative bond angle distribution. The gathered information will not only be vital in understanding the formation and structure of the amorphous Fe-Ni-Al-Zr alloy, but also to interpret the existence of long-range magnetism in a material with short-range order. With the distribution of interatomic distances of e.g. Fe-Fe, Fe-Ni and Ni-Ni at hand, the magnetic properties can be modelled to compute the composition-dependent magneto-caloric properties. Theoretical determination of the transition temperature and its closely related magnetic entropy change enables us to directly compare and assess our experimental findings with theory. [1] Y. K. Fang, et al., J. Appl. Phys. 105 (2009) 07A910 [2] Hiroshi Maeda et al., J. Japan Inst. Metals 47 (1983) 688–691