A new mechanism was proposed to account for the observed 2 x 1 reconstruction of the perfect Si<111> surface. It did not refute other explanations but showed that a simple unrestricted Hartree-Fock treatment could lead to two types of instabilities. The first one was of a magnetic nature as encountered in some transition metals; the other one was a charge instability likely to be coupled with atomic displacements. It was shown that the surface parameters were such that both types of instabilities could occur simultaneously. Furthermore direct calculations of the energy strongly suggest that the (2 x 1) arrangement should be the most stable at least in the magnetic case.

A New Mechanism for the 2 x 1 Reconstruction of the Silicon <111> Surface. G.Allan, M.Lannoo: Surface Science, 1977, 63, 11-20