Prediction of high-strain polar phases in antiferroelectric PbZrO₃ from a multiscale approach

PbZrO₃ (PZO) is the first known antiferroelectric perovskite whose properties have been investigated since last century. It has been well established both experimentally and computationally that such polar phase of PZO is R3c, quite independent of the direction of the applied electric field. But one may wonder if this is the only polar metastable phase of
PbZrO₃? The computational and experimental evidence this far suggest that if metastable polar phases other than R3c for PZO do exist they are likely to be much higher in energy and, therefore, require high stabilizing electric field. While large electric fields are challenging experimentally, computational investigation offers an excellent tool to overcome this challenge and provide a valuable insight. Here we designed a multiscale first-principles computational approach for predictions of polar phases and applied this approach to predict previously overlooked polar phases of PZO. We also validated some of our computational predictions by comparison with experimental data from the literature and proposed an experimental route to induce such polar phases through an application of electric fields.