CPM Seminar

Magnetic systems such as spin glasses, permanent magnets, and the trapped flux in superconductors exhibit a nonexponential response to a change in an applied magnetic field. This time dependence is usually characterized as quasilogarithmic or stretched exponential. Historically, this type of relaxation has been attributed to disorder in the system and experimental results have been phenomenologically modeled with a distribution of energy barriers. We have recently developed a magnetic model which uses interactions and not disorder to generate a nonexponential relaxation. The model consists of a planar array of magnetic dipoles or spins. The dipoles are constrained by an anisotropy energy to orientate perpendicular to the plane. With this constraint, the magnetic dipolar interactions between thespins tend to demagnetize the system. In the model, the dipolar interactions are treated with a mean field approximation, the demagnetization field. This approximation assures that interactions, and not disorder, dominate the system dynamics. The remarkable result is that, even when treated in the mean field limit, the system responds quasilogarithmically in time and can mimic a stretched exponential over six decades in time. This nonexponential relaxation is due to the dipolar or demagnetizing field which decreases as the system approaches equilibrium, i.e., as the system relaxes the driving force diminishes. It also predicts the quasilogarithmic decay and the nonmonotonic temperature dependence of the decay slope observed in the decay of the remanent magnetization observed in high temperature superconductors. Our recent experiments of the remanent magnetization in the archtypal spin glasses CuMn and AgMn are consistent with the presence of macroscopic interactions [2]. The observed interactions are of the correct sign to be responsible for the time variation of the magnetization in spin glasses.

[1] "A Model System for the Study of Slow Dynamics," D.K. Lottis, R.M. White, and E. Dan Dahlberg, Phys. Rev. Lett. 67, 362 (1991) and E. Dan Dahlberg, D.K. Lottis, R.M. White, M. Mattson, and "Slow Relaxation: An Effect of Interactions? " E. Engle, INVITED J. Appl. Phys. 76, 6396 (1994).

[2] "Macroscopic Interactions in Spin Glasses," Emily Engle, Thesis, University of Minnesota, 1996, unpublished.

Thursday, April 9th, 13:15
Ernest Rutherford Physics Building, Boardroom (room 105)