Establishment of a BaTiO3-based Computational Science Platform to Predict Multi-component Properties

Abstract

Barium titanate (BaTiO3) is considered to be a beneficial ceramic material for multilayer ceramic capacitor (MLCC) applications because of its high dielectric constant and low dielectric loss. Numerous attempts have been made to improve the physical properties of BaTiO3in response to recent market trends by employing multicomponent alloying strategies. However, owing to its significant number of atomic combinations and unpredictable physical properties, finding a traditional experimental approach to develop multicomponent systems is difficult; the development of such systems is also time-consuming. In this study, 168 new structures were fabricated using special quasi-random structures (SQSs) of Ba1-xCaxTi1-yZryO3, and 1680 physical properties were extracted from first-principles calculations. In addition, we built an integrated database to manage the computational results, and will provide big data solutions by performing data analysis combined with AI modeling. We believe that our research will enable the global materials market to realize digital transformation through datalization and intelligence of the material development process.