Abstract
The phonon dispersion relations of magnesium are calculated from a local form of the Heine-Abarenkov model potential by means of second-order perturbation theory. It is found that the calculated phonon frequencies do not agree well with those observed experimentally. The possible reasons for the discrepancies are discussed. It is shown that the inclusion of the non-pair interionic forces, implied by higher orders of perturbation theory, is not necessary to explain the phonon spectrum, but that satisfactory agreement could be obtained with a different form of model potential. The role of the Born-Huang conditions in hexagonal close-packed simple metals is considered, and it is shown that the satisfaction of these conditions is an insensitive test of the suitability of a model potential.