Based upon purely geometrical considerations, it was pointed out that a polar semiconductor surface with reconstruction did not satisfy the necessary condition for the existence of an Ehrlich-Schwoebel step-edge barrier. It was further shown, by explicit Monte Carlo calculations, that the kinetic surface roughening observed on a GaAs(001) surface could be accounted for by the stability and complexity of the (2 x 4) reconstruction, and hence an Ehrlich-Schwoebel barrier was unnecessary in order to account for it. It was pointed out that it was incorrect to use the solid-on-solid model with an Ehrlich-Schwoebel barrier for surface growth studies or for atomistic growth simulations.

Absence of a Step-Edge Barrier on a Polar Semiconductor Surface with Reconstruction. M.Itoh, T.Ohno: Physical Review B, 2000, 62, 1889-96