A Simulation Study of Oxygen-Vacancy Behavior in Strontium Titanate: Beyond Nearest-Neighbor Interactions

We investigate the effect of acceptor-dopant cations on oxygen-vacancy migration in the perovskite oxide SrTiO3 by static lattice simulation techniques. We focus on two themes: dopant cations modifying the activation energies for vacancy migration, and dopant cations binding oxygen vacancies in binary associates. In both cases a variety of defect configurations exceeding the scope of first nearest neighbor (1NN) interaction is examined, i.e. INN, 2NN, 3NN, 4NN, and 5NN. The behavior of four divalent dopants (Ni2+, Fe2+, Co2+, and Mn2+) and one trivalent dopant (Al3+) is compared. The simulations predict that the binding energy of an oxygen vacancy to any of these dopants is negative at INN sites, but only converges to zero for 4NN sites. In addition the simulations show that the migration energy of a vacancy is affected by an acceptor-dopant cation over a length scale of several unit cells. A simple analytical model is used, together with the calculated site and activation energies, to predict how in general the ionic conductivity is affected by the interactions between oxygen vacancies and acceptor-dopant cations. The model predicts that the magnitude of these effects depends on the specific dopant, its concentration, and the temperature.