Electronic Structure and the Properties of Metals. II. Application to Zinc

The method developed in the preceding paper for computing properties of metals from first principles is applied to zinc. The OPW form factors which determine many electronic properties and the characteristic function of wave number which determines many atomic properties are computed and applied to sample properties. Those properties treated are the Fermi surface, electronic specific heat and cyclotron resonance, the resistivity due to vacancies, the resistivity of the liquid, the electron-phonon interaction, the crystal structure and ca ratio, the energy change on melting, the structure and energy of formation of vacancies, the elastic constants, the "stabilization" of ordered structures in alloys, phonon structure and dispersion, and the Kohn effect. Where comparison of calculated electronic properties with experiment was possible the agreement was good. Agreement with experiment was more limited for the atomic properties, though the discrepancies appeared to be consistent with the uncertainty in the interpolations used in obtaining the energy-wave-number characteristic. Such a discrepancy was finding the fcc structure lower in energy than the hcp structure; this also gave rise to instability against the formation of certain phonons. Otherwise the agreement for the atomic properties was semiquantitative. It is suggested that irregularities found by Brockhouse, Rao, and Woods in the phonon spectrum of lead are not images of the Fermi surface, but images of the energy-wave-number characteristic. Such fluctuations depend upon the detailed structure of the atom and are found to be much larger than those associated with the Kohn effect; in zinc they occur at wave numbers near, but not exactly at, 2k₅. The appearance of these irregularities suggests the possibility of computing the energy-wave-number characteristic, and therefore a wide range of properties, from such measurements of the phonon spectrum along symmetry directions. The possibility of obtaining the essential results of the theory with simpler approximations is considered, as well as the possibility of improving on the method.