GROWTH-KINETICS SIMULATION OF THE AL-GA SELF-ORGANIZATION ON GAAS(100) STEPPED SURFACES

We present results of a stochastic kinetic simulation of Al and Ga segregation during the mobility enhanced epitaxial deposition of an Al & Ga monolayer on the As face of a stepped GaAs(100) surface. The simulation deposits the atoms on the surface and moves them from site to site so that the frequency of each event is proportional to its rate. Since the rate constants needed in the simulation have not been measured we use guesses based on the available information and analogies with similar systems. The simulations generate exactly the growth pattern corresponding to a given set of rates. We have found a set of rate constants which generates Al & Ga growth patterns having all the properties observed experimentally. The predicted aggregation patterns are stable with respect to reasonable changes in the numerical values of the kinetic parameters. By varying the corresponding rates we find how the elementary kinetic steps affect segregation. One of the more interesting results is that to obtain good segregation with smooth Al-Ga borders it is essential that the Al-Al interactions are anisotropic and are stronger than the Ga-Ga interactions.