Multi‐Objective Optimization of Ionic Polymer Electrolytes for High‐Voltage Fast‐Charging and Versatile Lithium Batteries

Abstract Designing ionic polymer electrolytes (IPEs) for high‐voltage and fast‐charging lithium batteries involves searching in a highly complex and discrete chemical space. Traditional material discovery processes struggle with this complexity due to high costs and long evaluation time. A kernel‐based Bayesian optimization is described to complete the multi‐objective optimization by considering ionic conductivity, electrochemical stability, and discharge capacity simultaneously. According to a recommender based on a union set of acquisition functions, promising IPEs through three iterations with only 2.8% of the chemical space is targeted. The achieved lithium metal batteries exhibit promising performance with ultrahigh cutoff voltage with NCM811 (LiNi 0.8 Co 0.1 Mn 0.1 O 2 , 4.8 V) and LNMO (LiNi 0.5 Mn 1.5 O 4 , 4.92 V). To further extend the versatility of IPEs and diminish the high cost associated with the glove‐box environment, an aqueous and high‐voltage lithium‐ion battery is developed by introducing water molecules in IPEs coupled with Li 4 Ti 5 O 12 ||LiMn 2 O 4 , a strong hydrogen bonding network formed between the rigid‐rod polyelectrolyte and the embedded water molecules, which effectively suppresses the water reactivity, meanwhile boosting the ionic conductivity. This work reveals an innovative multi‐objective optimization that effectively handles multi‐targets and discontinuous parameter space, offering critical insights to address complex challenges in material discovery and property optimization for advanced and versatile lithium batteries.