Polar-on-nonpolar epitaxy: Sublattice ordering in the nucleation and growth of GaP on Si(211) surfaces

Abstract

When a binary compound with two different atoms per primitive cell is grown on a lattice- matched elementary substrate in which the two atoms are identical, there exists an inherent ambiguity in the nucleation of the compound, with two different possible atomic arrangements, distinguished by an interchange of the two sublattices of the compound. For defect-free growth, one of the two possible nucleation modes must be suppressed. The problems involved in doing so depend very strongly on the crystallographic orientation of the substrate; for most orientations they are greatly complicated by the additional problem of electrical neutrality at the resulting polar/nonpolar interface. For zincblende-on-diamond growth, the (211) orientation is potentially one of the most promising orientations to overcome both classes of problems. The simplest possible atomic configuration for this orientation exhibits two different classes of bonding sites, some with one dangling bond, some with two, in the correct arrangement to create a strong preference for one of the two atomic configurations. We have grown GaP on Si (211) by MBE. Etching studies have indicated that (under the proper nucleation conditions) the growth is largely free of antiphase disorder, and that the sublattice ordering corresponds to P-atoms bonded to those interface sites that have two back bonds to Si. The likely nucleation sequence is discussed.