Computational and data-driven modelling of solid polymer electrolytes

Solid polymer electrolytes (SPEs) have been regarded as a safer alternative for liquid electrolytes in rechargeable batteries, yet they suffer from drawbacks such as low ionic conductivity. Designing SPEs with optimal performance is a challenging task, since the properties of SPEs are influenced by parameters across multiple scales, which leads to a vast design space. The integration of theory-based modeling methods and data-driven approaches can effectively link chemical and structure features of SPEs to macroscopic properties. Machine learning (ML) algorithms are paramount to data-driven modeling. This review aimed to highlight the ML algorithms used for SPE design, and how these algorithms can be employed synergistically with theory-based modelling methods such as density functional theory (DFT), molecular dynamics (MD) and coarse graining (CG). In addition, this work is concluded with our outlook in this young and promising field.