Review of AI-driven discovery of cathode anode and electrolyte materials for advancing sodium-ion batteries

Abstract Sodium-ion batteries (SIBs) are emerging as a sustainable, cost-effective alternative to lithium-ion batteries (LIBs) for grid storage, electric vehicles, and electronics. However, commercialization depends on overcoming the performance bottlenecks in energy density, kinetic rates, and safety. Artificial Intelligence (AI) and machine learning (ML) are accelerating this shift by enabling a transition from empirical “trial-and-error” approach to data-driven, predictive “closed-loop autonomous” material discovery. This review covers AI-driven discovery advancements in SIB materials, including layered oxides/polyanionic/Prussian blue analogues cathodes, hard carbon/alloy-type/anode-free anodes, and advanced electrolyte formulations. The key AI techniques are discussed, including graph neural networks (GNNs), generative AI (GenAI) models, and deep neural networks (DNNs) coupled with multi-objective optimization to identify Pareto-optimal materials by linking atomic-scale design to macroscale performance. Some promising materials for enhancing electrode performance are discussed. AI-driven workflows are explored spanning inverse design, high-throughput screening, property prediction, and autonomous discovery, targeting high-performance materials for next-generation applications. The current challenges and knowledge gaps are discussed, including data scarcity, interfacial modeling complexities, and model interpretability, while outlining future perspectives for fully autonomous, closed-loop material discovery platform.