An atomistic modelling investigation of the defect chemistry of SrTiO3 and its Ruddlesden-Popper phases, Srn+1TinO3n+1 (n = 1–3)

Abstract Atomistic modelling is routinely used to simulate the structure and properties of materials. As classical techniques rely on the accuracy of potential models (or force fields), it is important to demonstrate the reliability and transferability of potential models in describing the defect chemistry of materials. We present a comprehensive study on the defect chemistry of SrTiO3 and for the first time of its Ruddlesden-Popper phases, Srn+1TinO3n+1 (n = 1–3). We have used atomistic simulations based on a partial charge rigid ion potential model. As the usage of partial charge rigid ion potential models presents challenges when dealing with doping schemes that require electronic compensation, we present an approximation that does not rely on experimental ionization energies and electron affinities. We compare defect solution energies with previous computational literature and discuss our data within experimental evidence of doping SrTiO3, demonstrating that we can represent the defect chemistry of the material as well as potential models based on the shell models, which inherently include the effect of polarizability. Finally, we present the defect solution energies of Ruddlesden-Popper phases and compared them with the sparse computational literature and discuss our data within experimental evidence of doping Srn+1TinO3n+1 (n = 1–3).