The Magnetostriction of a Magnetite Crystal

Abstract

Magnetostriction of an octagonal magnetite crystal along three axes was measured in fields up to 5000 gauss. The curves obtained are all similar in form, first convex towards the $H$ axis, then concave, tending toward a maximum near 5000 gauss. With a longitudinal field of 5000 gauss, the digonal axis expands 30${\mathrm{}\left(10\right)\mathrm{}}^{-4\mathrm{}}$ per cent, the trigonal axis 12${\mathrm{}\left(10\right)\mathrm{}}^{-4\mathrm{}}$ per cent, and the tetragonal axis contracts about 4${\mathrm{}\left(10\right)\mathrm{}}^{-4\mathrm{}}$ per cent. With an equal transverse field the percentage changes are a contraction of 44${\mathrm{}\left(10\right)\mathrm{}}^{-4\mathrm{}}$, a contraction of 28${\mathrm{}\left(10\right)\mathrm{}}^{-4\mathrm{}}$ and an expansion of 4${\mathrm{}\left(10\right)\mathrm{}}^{-4\mathrm{}}$, respectively. Tests indicated that the effect perpendicular to a plane is independent of the direction of the field in that plane. No departure from cubic symmetry was established. To explain these results a model of the magnetic element in magnetite is suggested, similar to Ewing's recent model. The experiments favor the view that the rotating magnetic element is a group of non-parallel electron orbits inside the atom.