Dielectric materials are insulators that can screen an external electric field. These materials are used in capacitors, high power applications, computer chips, and sensors, making them an essential part of today's electronic technology. The screening in these materials can originate from dipoles of various length scales, but the electronic and ionic contributions have attracted special attention as they are intrinsic to a material.
Surprisingly, a microscopic description of polarisation was only formulated two decades ago, and its calculation became computationally feasible only recently. In light of this progress, it is timely to use these to develop new dielectric materials. In contrast to the conventional characterisation approach, where theoretical works follow experiments, this ``materials design'' approach aims to suggest new dielectric materials or new usage of conventional dielectric materials a priori to experimental findings. In this thesis, I show that a combination of conventional electronic structure methods and data science tools can be used synergistically to design dielectric materials.
This thesis consists of three main results. Firstly, new properties of conventional materials are probed using electronic structure theory, in particular the dipole induced by F-doping in non-polar Ruddlesden-Popper phase Sr3Ti2O7 is discussed. Next, electronic structure calculation are combined with data science methods, to analyse the permittivity of 1364 materials. Finally, a fully data-oriented method for analysing distortions in dielectric materials is shown. Together these studies show the strength of combining conventional electronic structure and data science techniques and that their balance can be adjusted according to the problem of interest. These studies not only open a new path towards next-generation dielectric materials, but the methodologies are also significant as many of them are either newly developed or newly imported from different scientific fields. Therefore, works in this thesis are likely to promote further material discoveries.