Predicted growth mode for metal homoepitaxy on the fcc(111) surface

In an attempt to understand homoepitaxial growth on fee (111) surfaces, the potential energy barriers for both inter- and intralayer diffusion are calculated for Ag, Au, Cu, Ni, Pd and Pt using a corrected effective medium theory in its simplest form, namely MD/MC-CEM theory. These metals exhibit two different characteristic behaviors, classified by the relative magnitude of the barriers between the two kinds of diffusion: (i) for Ag, Au, Cu and Ni, the interlayer diffusion barrier is always larger than the intralayer one, even for small islands or near kink sites at step edges; and (ii) for Pd and Pt, the reverse situation is true, with the interlayer diffusion barrier decreasing further for small islands or near a kink site. We therefore predict that under ''natural'' conditions, the first group of metals will grow three-dimensionally over the whole temperature range until the step-flow regime is reached. In particular, re-entrant growth is not expected to occur at low temperatures. By contrast, for the second group of metals we predict that a two-dimensional growth regime exists below step-flow, and re-entrant layer-by-layer growth is possible at low temperature due to the extremely low barrier for small islands or near kink sites. These predictions are consistent with all available experiments, which have investigated the AE, Cu and Pt systems. We suggest that further experiments are desirable, in particular for Pd growth on Pd(111). We also point out that the reduced barriers for interlayer diffusion, due to the small size of the island and to the presence of kinks, have a common physical origin.