Recent tight-binding calculations of bulk electronic total energies by Koiller, Davidovich, and Falicov (KDF) [Phys. Rev. B 41, 3670 (1990)] indicated the tendency for ${\mathrm{Al}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Ga}}_{\mathit{x}}$As alloys to form ordered structures. The stablest structure they predicted was the monolayer (AlAs${)}_1$(GaAs${)}_1$[001] superlattice, which was recently observed in homogeneous vapor-phase growth. In light of these results we have examined the possibility that bulk energetics can explain this ordering. We have subjected KDF’s tight-binding total-energy calculations and, separately, our own first-principles pseudopotential total-energy calculations to a statistical-mechanics analysis of order-disorder transitions. We find that bulk thermodynamics is inconsistent with the observed ordering; hence, explanations must be sought elsewhere (e.g., surface thermodynamics or kinetic effects).

• Received 4 June 1990

DOI:https://doi.org/10.1103/PhysRevB.43.1584