Studies of the Propagation Velocity of a Ferromagnetic Domain Boundary

Abstract

Experimental results are given on the velocity of propagation of a single domain boundary in a crystal of silicon iron with a simple domain structure. In weak applied magnetic fields (\ensuremath{\sim}0.003 oersted) the velocity is given by a relation of the form $v=G(H\ensuremath{-}{H}_{0})$, where $G$ is a constant \ensuremath{\sim}4 cm/sec./oersted in this crystal, and ${H}_{0}\ensuremath{\cong}0.003$ oersted is the starting field. Calculation of the eddy current losses accompanying the motion of a plane boundary gives a theoretical expression for $G$ in good agreement with experimental values; the predicted linear dependence on the resistivity was approximately verified by measurements at 78\ifmmode^\circ\else\textdegree\fi{}, 194\ifmmode^\circ\else\textdegree\fi{}, and 293\ifmmode^\circ\else\textdegree\fi{}K. In stronger fields (g5 oersteds) there is evidence that the wall closes on itself, and the experimental velocity of collapse of the wall as deduced from flux changes agrees with the theoretical result based on a model of eddy current losses accompanying a collapsing cylindrical boundary. The results have a bearing on the well-known eddy current anomaly, namely, the fact that the total loss in a ferromagnetic material undergoing a.c. magnetization is often two or three times larger than the eddy-current and hysteresis losses calculated in the usual way assuming a spatially uniform and isotropic classical permeability.