Machine learning-accelerated discovery and design of electrode materials and electrolytes for lithium ion batteries

Abstract

With the development of artificial intelligence and the intersection of machine learning (ML) and materials science, the reclamation of ML technology in the realm of lithium ion batteries (LIBs) has inspired more promising battery development approaches, especially in battery material design, performance prediction, and structural optimization. Data-driven ML approach displays the advantage of quickly capturing the complex structure-activity-process-performance relationship, and is promising to offer a new paradigm for the burgeoning of battery materials. This work provided a comprehensive review of material design research using ML as a framework in the field of LIBs. Specifically, the latest progress in the application of ML in the design, performance prediction, and composition optimization of cathode/anode and liquid/solid electrolyte materials for LIBs was summarized. Besides, the shortcomings of ML application in battery material researches were analyzed and the future development direction was prospected. This work provides a significant guidance for the screening and development of advanced LIBs materials via ML-assisted method.