Search

The first project was an attempt to test experimentally the conditions under which the impulse approximation (IA) is valid for the dynamic structure factor of a real crystal, where anharmonic effects are significant, and determine the high energy motions in solids. Neutron scattering measurements for a polycrystalline lithium sample were made for wavevector transfers in the range 14 A^0-1 to 49 A^0-1 at the Rutherford Appleton Laboratory. The wavevector scaling properties of the position of the maximum and width of the scattered neutron distribution suggest that the IA is a good approximation for Q > 25 A^0-1 at a sample temperature of 100K. However the width was found to be 14% larger than that calculated using the harmonic model density of phonon states determined by experiment. Numerical calculations for the incoherent dynamic structure factor predicted widths that were in good agreement with those calculated assuming the IA to be valid. It is concluded that either the model density of phonon states is incorrect or a systematic experimental error is present due to the approximation made for the instrumental resolution function. Several theoretical predictions for the power law divergence of the wavevector, magnetic field and energy dependence of the longitudinal magnetic susceptibility for an isotropic ferromagnet below the Curie temperature remain untested. Inelastic neutron scattering measurements were performed at the Institute Laue Langevin for a single crystal of EuO, which is an almost ideal isotropic Heisenberg ferromagnet, in order to test the theoretical predictions. The wavevector dependence of the longitudinal susceptibility was found to be well described by the mean field result and the divergent term concluded to be absent or negligibly small. The energy dependence was found to be quasielastic and was represented empirically by a Lorentzian. The statistical accuracy of the data precluded a detailed test of the longitudinal susceptibility field dependence.