Proposed explanation of the<i>p</i>-type doping proclivity of ZnTe

An explanation is proposed for the fact that ZnTe is unique among the II-VI compound semiconductors in that it can be doped p type rather easily: a p-like deep-level resonance lies within the valence band of ZnTe and emerges into the fundamental band gap with increasing Se content x in ${\mathrm{ZnTe}}_{1\mathrm{\ensuremath{-}}\mathit{x}}$${\mathrm{Se}}_{\mathit{x}}$ random alloys. This level generates free holes when it lies below the Fermi energy in the valence band, making its parent defect a shallow acceptor. When the level moves into the gap, the impurity becomes a deep hole trap. The native and foreign antisite defects ${\mathrm{Zn}}_{\mathrm{Te}}$ and ${\mathrm{Li}}_{\mathrm{Te}}$ are suggested as possible parent defects of the relevant deep level; they are predicted to be shallow acceptors in ZnTe, while the corresponding defects are deep traps in other II-VI compound semiconductors. Tests of this proposal are suggested and the substitutional s- and p-bonded deep levels of ZnSe and ZnTe are predicted, extending the theory of Hjalmarson et al. [Phys. Rev. Lett. 44, 810 (1980)]. The possibility of doping ZnSe p type with (antisite) Be is also proposed and discussed.