Answers are grounded in this paper and cite their sources.
Superionic conductors (SICs) are essential for solid-state Li-ion batteries, yet their practical application is hindered by high Li-ion concentrations that increase inactive Li and battery cost. Here, we develop an artificial intelligence-driven discovery framework integrating a large language model and graph neural network to accelerate SIC design. The large language model, GPT-4o, extracts key transport parameters from 465 peer-reviewed studies, identifying oxyhalides as promising candidates. A materials space of 69,920 oxyhalides spanning 230 space groups is constructed, and ionic conductivity is predicted using a graph neural network. Compounds with gravimetric and volumetric Li-ion concentrations below 3.05 mol kg −1 and 11.49 mol L −1 , respectively, as well as ionic conductivities above 10 mS cm −1 are further validated via ab initio molecular dynamics simulations. Notably, LiMoOBr 4 exhibits a low Li-ion concentration (2.28 mol kg −1 ) and high ionic conductivity (13.53 mS cm −1 ). This work accelerates the design of advanced SICs for battery utilization.
No takes yet. Share an insight, caveat, or question.
Lan et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: