Predictive estimation of electro-optic properties of crystals: A step toward high-throughput screening

Effects involving interactions between electromagnetic waves and applied electric field in materials, commonly known as electro-optic effects, have remarkably diverse applications ranging from telecommunications to next-generation photonic devices. Identification of materials with large electro-optic response is highly desirable for these applications; however, associated experiments remain challenging. Ab initio simulations provide a convenient avenue for evaluating the strength of electro- and elasto-optic effects in crystals, as well as other related properties. Here, we introduce an efficient workflow for predicting tensorial electro-optic coefficients (Pockels and Kerr) in non-magnetic crystals of most symmetries, including all the necessary ingredients, such as dielectric, elasto-optic, and electrostrictive tensors. This approach can also be utilized for evaluating descriptors in high-throughput evaluations of optical properties. We benchmark our simulations on tetragonal BaTiO3 and wurtzite AlN crystals, obtaining, where available, reasonably good agreement with the results of prior experimental and theoretical reports.