AI Framework Proposed to Accelerate Solid-Electrolyte Discovery for Batteries
An international team, led by Associate Professor Eric Jianfeng Cheng and Professor Hao Li of Tohoku University's Advanced Institute for Materials Research (WPI-AIMR), has proposed a new AI-driven framework to accelerate the discovery of solid electrolytes for safer and longer-lasting batteries. Published in Advanced Materials, this perspective addresses the limitations of current 'trial and error' methods, which often focus solely on ion conductivity while neglecting other critical properties like chemical stability, mechanical stress resistance, and manufacturability. The proposed system integrates living databases, AI-powered simulations for ion movement, and large language models (LLMs) to convert research data into organized knowledge. This 'closed loop' approach involves AI suggesting candidates, physics-based models checking them, experiments confirming results, and new data feeding back into the AI for continuous improvement. The framework emphasizes supporting scientists rather than replacing them...
