The Effect of Crystal Structure on Magnetostriction

Magnetostriction of steel, soft iron, and magnetite was investigated, using a Weiss electromagnet to produce the field, whose strength was varied up to 10,000 or 20,000 gauss depending on the specimen and an apparatus of the lever type for measuring dimension changes to within about 2${\mathrm{}\left(10\right)\mathrm{}}^{-7\mathrm{}}$cm. Curves are given showing the behavior of a steel sphere, both before and after annealing. A sphere of soft Norway iron gave larger effects than the steel sphere, effects which varied with the orientation of the sphere in the field. A soft iron disk, part of a single grain formed in a plate of iron with 3½ per cent Si, showed larger dimension changes than have yet been found in iron. Along different diameters of the disk quite different results were obtained; in some cases there was a contraction, in others, an expansion. Preliminary work with a magnetite sphere, 1.43 cm in diameter, indicated a lack of cubic symmetry for both the transverse and longitudinal effects. Changing the direction of the field with respect to the crystal axes may change the magnetostriction in a given direction from a contraction to an expansion; also transverse magnetostriction, with a fixed direction of magnetic field, is a contraction for some directions in the crystal, an expansion for others. Theories are briefly discussed. It is suggested that the ordinary magnetostriction curve of iron is the resultant of two or more different types of curves which are characteristic of the small individual crystals in different orientations, the composite curve depending on the heterogeneous arrangement of the crystals. The experiments suggest that the Villari reversal is also probably a consequence of heterogeneous crystal arrangement in the iron. Ewing's latest model of the magnetic atom seems to be capable of explaining the various magnetostrictive effects.