Computational search for novel solid state electrolytes

The increasing global energy usage in portable electronic devices and automobiles make a demand for high-power energy storage systems. All solid state batteries (ASSB), with solid state electrolytes (SSE), open a floodgate of prospective characteristic properties for future energy storage systems. Although, known solid state electrolytes (sulfide, oxides, borates, and so on) do not satisfy all performance requirements, such as high ion conductivity or compatibility with anode materials, for wide implementation in various applications. Now, the major effort is to find novel solid state electrolytes. In contrast to experimental methods, computational methods can drastically accelerate novel material search. The goal of the proposed project is to develop machine learning methods for accelerating a property prediction of solids and perform computational search for novel solid state electrolytes using machine learning and evolutionary structure prediction algorithms. We aim to develop computational methods that will have high impact on material science, especially for energy storage systems.