Appendix B Shockley–Read Bandgap States

Abstract

No explanation has yet been offered in the literature regarding the reason for this difference in lifetimes, but it raises the important question as to whether the explanation is of a fundamental nature. One obvious difference in the two materials systems concerns the nature of their chemical bonds. The bonding in III-V semiconductors is predominantly of the covalent variety coupled with a degree of iconicity. This was first pointed out by Welker and Weiss as a possible reason for the observation of the larger bandgaps exhibited by III-V materials relative to the corresponding Group IV elemental semiconductors with a similar lattice constant. The formation of the ionic bond draws electrons away from the space between the ions, resulting in a larger periodic potential with a corresponding increase in energy gap, as given by the simple Kronig–Penney model. The ionic bonding in HgCdTe is stronger still, resulting in a significant increase in the energy gap, as registered, for instance, by CdTe, relative to its III-V counterpart, InSb. However, the arguments above would suggest that it is not a distribution in ionic potentials that is required to generate bandgap states, but a distribution in effective lattice constant. Is there any evidence that would suggest that the distribution in the effective equilibrium lattice constant of semiconductors is dependent on the varying degrees of their covalent and ionic bonds? This is an important issue. Room temperature photon detection will only be achieved by utilizing depletion-current-limited photodiodes with long S-R lifetimes. It would be useful to know whether there is a fundamental limitation on this parameter for III-V materials.