Prediction models of both the electronic and ionic contributions to the static dielectric constants have been constructed using data from density functional perturbation theory calculations of approximately 1200 metal oxides via supervised machine learning. We developed two types of random forest regression models for oxides with the ground-state crystal structures: one model requires only compositional information and the other model also uses structural information. Although the training data included various atomic frameworks, the prediction models performed well even when only compositional information was used as feature descriptors. In prediction of the electronic contributions to the dielectric constants, the accuracies of the regression models with and without structural information were comparable, while the structural descriptors more clearly improved the prediction accuracy for the ionic contributions. We also analyzed the feature importance for prediction of the dielectric constants. The mean atomic mass and mass density were determined to be significant features in prediction of the electronic contributions without and with structural information, respectively. The standard deviation of the principal quantum number and mean neighbor distance variation were found to be important for the respective prediction models of the ionic contributions. The correlations between the dielectric constants and these features are discussed, along with the underlying physical mechanisms.

No abstract

No abstract

No abstract

Pseudo Ruddlesden–Popper-type Li2SrNb2O7 undergoes a phase transition between paraelectric and weak ferroelectric phases at T c (=217 K), whereas Ta-counterpart Li2SrTa2O7 stays paraelectric though its room temperature structure is identical to that of Li2SrNb2O7. In the present study, the ferroelectric phase transition of Li2Sr­(Nb1–x Ta x )2O7 is investigated as a function of Ta-concentration x. As the Ta-concentration x increases, T c is found to decrease monotonously, leading to the disappearance of the ferroelectricity at x = 0.4. This phase-transition suppression with increasing x in Li2Sr­(Nb1–x Ta x )2O7 is supported by first-principles calculations, showing that the soft mode in Li2SrNb2O7 or the increase of x in Li2Sr­(Nb1–x Ta x )2O7 enhances their bonding states. In the composition of x = 0.4, dielectric permittivity gradually increases on cooling to become independent of temperature around 0 K, indicating an incipient ferroelectricity of the system. Empirical analyses suggest existence of a quantum paraelectric state in a composition range of 0.3 < x < 0.4. The present study provides a novel playground to investigate a quantum para/ferroelectricity in layered-perovskite-type compounds.