AI-assisted quantitative deciphering of molecular configuration and steric effects for high-performance zinc battery electrolytes

Zinc-ion batteries depend on molecular design of electrolyte additives for performance optimization. However, the influence of molecular structure and steric effects on solvation behavior has not been systematically quantified. This research combines Retrieval-Augmented Generation (RAG) technology with large language models, establishing a system from intelligent literature analysis to molecular screening. Through prompt engineering optimization, screening precision under defined criteria improved from 30.0% to 100%, identifying two structurally distinct additives from over twenty thousand molecules—rigid cyclic 2-methylimidazole (MI) and flexible chain-like 3-aminopropanol (AP). Through theoretical calculations and experimental validation, we demonstrate that MI’s rigid structure restricted solvation shell adaptability, whereas AP’s conformational flexibility enhanced zinc-ion migration efficiency, suppressed hydrogen evolution and dendrite formation, and extended battery cycle life. This study quantitatively elucidates molecular structure and steric effects in zinc battery performance, establishes principles for electrolyte design, and develops an AI-driven precision screening methodology. Zinc‑ion battery performance depends on the molecular design of electrolyte additives, but the effects of molecular structure and steric factors on solvation behavior remain unquantified. Here, Retrieval‑Augmented Generation screening identifies rigid cyclic 2‑methylimidazole and flexible chain-like 3‑aminopropanol: 2‑methylimidazole limits solvation, whereas 3‑aminopropanol improves ion transport and extends battery life.