Novel Solid-State Electrolytes for Li Ion Batteries by Computational Design and High-throughput Ab Initio Calculation

Solid-state electrolytes (SSEs) can alleviate many of the issues of Li-ion batteries arising from the utilization of the organic liquid electrolytes. Up to date, several SSEs have been proposed from previous experiments or computational screening from publicly available materials databases, however, none of those SSEs are fully satifactory. In this work, we take one step further; instead of simply exploring a pre-existing materials database, we try to design new materials with aliovalent substitution of cations. This aliovalent substituion is known to facilitate kinetics of Li diffusion. We first screen potential host materials for generating new SSEs by aliovalent substituion, considering fundamental properties such as the presence of transition metals, thermodynamic stability, and band gap. Afterward, we crudely examine the potential energy surface (PES) around the Li ion at interstitial sites. The materials with the most smooth PES are then chosen for the possible candidate for the host material. Finally, we choose some materials from the candidate list and demonstrate that indeed they become a Li ion conductor after aliovalent doping. Starting from 42,337 structures, we find a number of SSEs with predicted Li ionic conductivity comparable to the state-of-the-art Li₁₀GeP₂S₁₂.