Combining database mining and DFT to create new materials

Abstract

One of the main objectives in modern chemistry is to create new stable materials capable of energy harvesting and energy storage. We can decrease the time it takes to accomplish this by using known materials as the springboard for new materials: combining database mining, first-principles DFT calculations, and concepts borrowed from inorganic chemistry has already proven to be a powerful method of materials discovery and optimization. In this work we use crystallographic databases to obtain families of polar and antipolar materials that can be related to each other by phase transitions, mappable atomic displacements, and crystal chemical symmetry. We choose polar materials because they can boost the charge separation caused by solar illumination, demonstrating some of the highest solar conversion efficiencies in the solid state, and antipolar materials because they have a highly nonlinear change in volume and entropy associated with the phase transition between antipolar and polar states, which means that we can store and release energy by making and breaking chemical bonds without the risk of thermal runaway associated with Li-ion batteries. Our database mining uncovers new examples of polar and antipolar phase transitions and these sets of materials are used as the input for DFT calculations designed to investigate the structure-property relationships of nature’s missing materials. We present results for families of related materials with chemical formula AB₂C₃ that contain previously characterized and potentially synthesizable members with exceptional properties.