Accelerating Li-based battery design by computationally engineering materials

Satisfying renewable energy markets impels engineering of highly energy-dense, temperature-adaptive, sustainable, safe, and cost-effective batteries. Identifying which parameters are critical to this endeavor entails connecting technological advancements to battery components—including cathodes, anodes, and electrolytes—that are fundamentally characterized correctly. Data-driven battery design reinforces overarching technological improvements through multiscale investigations of fundamental material properties and phenomena. This encompasses computational simulations, machine learning, and economics. Li-ion, Li-metal, Li-S, and anode-free Li cell materials are selected to favorably tune properties for battery applications. This review first develops a fundamental computational approach to materials selection and property tuning, merging precise atomistic simulation, machine learning, and data-driven techniques. Subsequently, it reconciles that approach with accelerating anodic, cathodic, and electrolytic design in Li-based battery applications. Beyond extending discussion to generalized battery performance metrics and all-solid-state battery development, this review ultimately provides recommendations on how future research can be improved by implementing the described methodologies.