Comment on “Use of a variable-charge interatomic potential for atomistic simulations of bulk, oxygen vacancies, and surfaces of rutileTiO2”

Abstract

In a recent article [Phys. Rev. B 73, 165406 (2006)], Hallil et al. presented two empirical variable-charge models for $\mathrm{Ti}{\mathrm{O}}_{2}$. Both models are based on the QEq charge equilibration scheme, which we have recently shown [B. S. Thomas et al., Phys. Rev. B 74, 214109 (2004)] is inappropriate for this system due to charge shielding and the subsequent suppression of electrostatic forces. It is therefore questionable whether the extensions proposed by Hallil et al. provide transferability beyond that available with conventional partial charge models. We recommend that Hallil et al. test their method on other ${\mathrm{Ti}}_{x}{\mathrm{O}}_{y}$ systems to evaluate its utility.