SURFACE-SCIENCE. MAR 19 1999; 424 (1) : 109-116

Monte Carlo simulations of the Si(111) surface at equilibrium have been performed. No evidence is found for the existence of a disordered flat phase, as previously reported in literature. Examination of the specific heat as a function of temperature, the height-height correlation function and an order parameter for the Si(111) surface reveals only one phase transition, the familiar Kosterlitz-Thouless roughening transition. It is shown that the argument for the existence of a disordered flat phase is based on a spurious effect on the specific heat that can be attributed to the presence of isolated adatoms. The presence of such adatoms, with no bonds to the bulk of the crystal at all, is excluded in our Monte Carlo simulations.