Abstract
In the first approach considered the vacancy is represented by a spherical hole in an otherwise uniform positive background of charge. The electron density and resulting formation energy, for this jellium model, have been calculated in a non-linear (Kohn-Sham) theory by Nieminen et al. The second approach, originally advocated by Harrison, is based on pseudopotentials and linear screening theory. A third method designated semi-linear theory, which combines features of both the former schemes has recently been proposed by M.W. Finnis. For Na the Harrison model has some relevance and the semi-linear screening approach gives a good estimate of the formation energy. For a high electron density metal like Al, jellium calculations predict unphysical values for the formation energy. None of the existing approaches is adequate for describing the vacancy formation energy of a dense polyvalent metal.