First-Principles Modeling in the Context of Dielectric Materials Science and Design

Abstract

First-principles calculation and data-centric approach have become powerful tools for materials design. In this study, we introduce our latest research outcomes that are related to computational dielectric materials science. The main contents are as follows: (1) first-principles-based multiscale modeling of electronic charge transfer and molecular dynamics simulation of ionic carrier transfer in polymer dielectrics, (2) first-principles modeling of the inorganic filler/polymer interface and charge injection from the (metal) electrode to polymer dielectrics, and (3) a combined first-principles and machine learning approach for predicting dielectric properties of various materials. The recent advances in the atomistic understanding of the electrical properties of dielectric materials is highlighted. In addition, we show that, with this knowledge of the underlying physics, one can develop a machine learning model that can accurately predict the physical properties, albeit with only a small dataset.