An end‐to‐end artificial intelligence platform enables real‐time assessment of superionic conductors

• Published in: SmartMat (Beijing, China) China), 2023-12, Vol.4 (6), p.n/a
• Main Authors: Wang, Zhilong, Han, Yanqiang, Cai, Junfei, Chen, An, Li, Jinjin
• Format: Article
• Language: English
• Subjects: artificial intelligence; cathode material; solid‐state battery; solid‐state electrolyte; superionic conductor
• ISSN: 2688-819X; 2766-8525; 2688-819X
• DOI: 10.1002/smm2.1183

Superionic conductors (SCs) exhibiting low ion migration activation energy (Ea) are critical to the performance of electrochemical energy storage devices such as solid‐state batteries and fuel cells. However, it is challenging to obtain Ea experimentally and theoretically, and the artificial intelligence (AI) method is expected to bring a breakthrough in predicting Ea. Here, we proposed an AI platform (named AI‐IMAE) to predict the Ea of cation and anion conductors, including Li+, Na+, Ag+, Al3+, Mg2+, Zn2+, Cu(2)+, F−, and O2−, which is ~105 times faster than traditional methods. The proposed AI‐IMAE is based on crystal graph neural network models and achieves a holistic average absolute error of 0.19 eV, a median absolute error of 0.09 eV, and a Pearson coefficient of 0.92. Using AI‐IMAE, we rapidly discovered 316 promising SCs as solid‐state electrolytes and 129 SCs as cathode materials from 144,595 inorganic compounds. AI‐IMAE is expected to completely solve the challenge of time‐consuming Ea prediction and blaze a new trail for large‐scale studies of SCs with excellent performance. As more experimental and high‐precision theoretical data become available, AI‐IMAE can train custom models and transfer the existing models to new models through transfer learning to constantly meet more demands. An artificial intelligence platform (AI‐IMAE) to make accurate predictions of ion migration activation energies is proposed. AI‐IMAE is expected to completely solve the challenge of time‐consuming activation energy prediction and blaze a new trail for large‐scale studies of superionic conductors. As more high‐fidelity data become available, AI‐IMAE can train custom models and transfer the existing models to new models through transfer learning to constantly meet more demands.