Abstract
A theoretical analysis of the helium diffraction experiments performed on the GaAs(110) surface by Cardillo, Becker, Sibener, and Miller is presented. Dynamic scattering calculations for several model potentials approximating that of the surface are presented. These suggest that the surface is a smoothly corrugated soft wall with a peak-to-trough distance along the [001] direction of ∼ 1 Å and an attractive well depth of ∼ 7 meV. Existing methods for generating van der Waals potentials are reviewed and an explanation is suggested for their poor performance in the case of helium. New semiempirical rules fit to the known helium rare-gas potentials and consistent with the recent results of Esjberg and Nørskov are proposed. These are used to construct an approximate potential for GaAs(110), which we show to agree qualitatively with the empirically fit potential. The classical turning-point surface of this potential lies 3.5 Å above the As nucleus. Scattering calculations performed on this potential are shown to agree poorly with experiment. We conclude from this that the scattering is very sensitive to nuances in the potential, and that very accurate surface charge densities will be required before a potential capable of scattering in agreement with experiment can be constructed.