Role of tight-binding parameters and scaling laws on effective charges in semiconductors

Abstract

We compute the transverse effective charges for several compound semiconductors by using the empirical tight-binding model and the Berry-phase approach. We compare different parametrizations showing that a suitable tuning of the scaling laws for the radial part of the hopping parameters provides a fairly good prediction of the effective charges even with the minimal ${\mathrm{sp}}^{3}$ basis set. In contrast new and refined parametrizations that reproduce very well the dispersion of the conduction band at equilibrium by including d and/or ${s}^{*}$ polarization orbitals may underestimate the effective charges. We suggest that the root of such a discrepancy may rest with the difficulty of determining the scaling laws for polarization orbital.