Accelerated Materials Design of Lithium Superionic Conductors Based on First‐Principles Calculations and Machine Learning Algorithms

Abstract

A method for efficiently screening a wide compositional and structural phase space of LISICON-type superionic conductors is presented that utilizes a machine-learning technique for combining theoretical and experimental datasets. By iteratively performing systematic sets of first-principles calculations and focused experiments, it is shown how the materials design process can be greatly accelerated, suggesting potentially superior candidate lithium superionic conductors.