Transition-metal impurities in semiconductors and heterojunction band lineups

The validity of a recent proposal that transition-metal impurity levels in semiconductors may serve as a reference in band alignment in semiconductor heterojunctions is positively verified by using the most recent data on band offsets in the following lattice-matched heterojunctions: ${\mathrm{Ga}}_{1\mathrm{-}\mathrm{x}}$${\mathrm{Al}}_{\mathrm{x}}$As/GaAs, ${\mathrm{In}}_{1\mathrm{-}\mathrm{x}}$${\mathrm{Ga}}_{\mathrm{x}}$${\mathrm{As}}_{\mathrm{y}}$${\mathrm{P}}_{1\mathrm{-}\mathrm{y}}$/InP, ${\mathrm{In}}_{1\mathrm{-}\mathrm{x}}$${\mathrm{Ga}}_{\mathrm{x}}$P/GaAs, and ${\mathrm{Cd}}_{1\mathrm{-}\mathrm{x}}$${\mathrm{Hg}}_{\mathrm{x}}$Te/CdTe. The alignment procedure is justified theoretically by showing that transition-metal energy levels are effectively pinned to the average dangling-bond energy level, which serves as the reference level for the heterojunction band alignment. Experimental and theoretical arguments showing that an increasingly popular notion on transition-metal energy-level pinning to the vacuum level is unjustified and must be abandoned in favor of the internal-reference rule proposed recently [J. M. Langer and H. Heinrich, Phys. Rev. Lett. 55, 1414 (1985)] are presented.