Bulk combinatorial analysis for searching new rare-earth free permanent magnets: Reactive crucible melting applied to the Fe-Sn binary system

Abstract The reactive crucible melting method is known to be an efficient and low-cost bulk combinatorial synthesis technique to search for new phases. In this work, we tested this technique by investigating the Fe-Sn binary system. For high-throughput characterization, this synthesis technique was combined with energy dispersive x-ray spectroscopy as well as magneto-optical Kerr microscopy. The latter was used for identification of desired phases with uniaxial magnetic anisotropy. Reliability of the reactive crucible method was evaluated by comparison of the phase composition forming in the reactive crucible with phases appearing in conventionally melted samples and with the phases represented in the reported Fe-Sn phase diagram. It was found that the Fe5Sn3 phase, existing in the equilibrium phase diagram at 800 °C and forming in conventionally melted alloys, does not exist in the diffusion zone of the reactive crucible. The problem of ‘missing phases’ is discussed. In addition, we have shown that the anisotropy energy obtained by quantitative analysis of the uniaxial domain structure of Fe3Sn2 phase, gives the value of 1 MJ/m3, whereas K1 evaluated by conventional magnetometry is ∼ 0.07 MJ/m3 only, by this demonstrating how erroneous the assessment of K1 from domain structure can be. Finally, a new unit cell is proposed for Fe5Sn3 phase, which is commensurately modulated by the orthorhombic unit cell with lattice parameters of a = 4.221 A, b = 7.322 A, c = 5.252 A and the space group Pbcm(α00)0s0 with a modulation vector of q = (½,0,0).