The binding sites for the adsorption of a single Ge atom on the (100) Si surface were identified by using first-principles total-energy calculations. It was found that the theoretical diffusion barriers were in excellent agreement with experimental data. By using a large super-cell, a controversy concerning the binding geometry and migration path of the adatom was resolved, and its effect upon the buckling of Si dimers could be investigated. It was shown that the adatom caused a buckling defect which was frequently detected by using scanning tunnelling microscopy, and that the study of a single adatom was therefore experimentally possible.

V.Milman, D.E.Jesson, S.J.Pennycook, M.C.Payne, M.H.Lee, I.Stich: Physical Review B, 1994, 50[4], 2663-6