Active Learning Guided Optimization of Multi-Component Electrolytes for Fast-Charging Lithium-Ion Batteries

With the rapid development of electric vehicles and portable electronics, there is an urgent demand for lithium-ion batteries (LIBs) with excellent fast-charging performance, which heavily depends on electrolyte solvation environment and its derived anode/cathode interphases. However, the complex electrochemical nature of LIBs (dynamic component-interphase interactions, sensitive performance-response to component adjustments) makes traditional trial-and-error optimization of multi-component electrolytes inefficient. Herein, an active learning strategy has been successfully employed to optimize multi-component electrolyte formulas by coordinating key component proportions, targeting fast-charging capability improvement for LIBs with commercial graphite anodes and NCM811 cathodes. As a result, a well-optimized formula is screened out after only two iterations. The balanced solvation environment rich in contact ion pairs (CIPs) and the fine-tuned additives synergistically optimize Li⁺ interfacial transport behaviors and effectively improve the fast-charging performance. The pouch battery could retain over 90% capacity at 4C charging rate, while the coin battery achieves a long cycle life of 600 cycles. This work offers a reference for adjusting multi-component electrolyte formulas to enhance the fast-charging performance of LIBs.