Thomas and Marks had criticized the present authors’ use of empirical models to describe oxides. Using arguments that they had developed previously, they concluded that charge equilibration-based models were inappropriate for oxides because they were not transferable to different stoichiometries (TixOy, for example) and had recommended the use of pure ionic models with partial charges. However, contrary to ionic models, charge equilibration-based models could capture the ionocovalent character of oxides. The idea was defended here that these models could become transferable at the cost of improving the short-range potentials involved in order to describe the modifications of the chemical bonds with stoichiometry.

Reply to “Comment on `Use of a Variable-Charge Interatomic Potential for Atomistic Simulations of Bulk, Oxygen Vacancies, and Surfaces of Rutile TiO2. Hallil, A., Tétot, R., Berthier, F., Braems, I., Creuze, J.: Physical Review B, 2007, 76[16], 167402