Volume 137, Issue 30 e202503105
Forschungsartikel

Uni-Electrolyte: An Artificial Intelligence Platform for Designing Electrolyte Molecules for Rechargeable Batteries

Xiang Chen

Corresponding Author

Xiang Chen

Tsinghua Center for Green Chemical Engineering Electrification & Beijing Key Laboratory of Complex Solid State Batteries, Department of Chemical Engineering, Tsinghua University, Beijing, 100084 China

AI for Science Institute, Beijing, 100080 China

Both authors contributed equally to this work.

Email: [email protected]; [email protected]

Search for more papers by this author
Mingkang Liu

Mingkang Liu

AI for Science Institute, Beijing, 100080 China

Both authors contributed equally to this work.

Search for more papers by this author
Shiqiu Yin

Shiqiu Yin

AI for Science Institute, Beijing, 100080 China

Search for more papers by this author
Yu-Chen Gao

Yu-Chen Gao

Tsinghua Center for Green Chemical Engineering Electrification & Beijing Key Laboratory of Complex Solid State Batteries, Department of Chemical Engineering, Tsinghua University, Beijing, 100084 China

Search for more papers by this author
Nan Yao

Nan Yao

Tsinghua Center for Green Chemical Engineering Electrification & Beijing Key Laboratory of Complex Solid State Batteries, Department of Chemical Engineering, Tsinghua University, Beijing, 100084 China

Search for more papers by this author
Qiang Zhang

Corresponding Author

Qiang Zhang

Tsinghua Center for Green Chemical Engineering Electrification & Beijing Key Laboratory of Complex Solid State Batteries, Department of Chemical Engineering, Tsinghua University, Beijing, 100084 China

Email: [email protected]; [email protected]

Search for more papers by this author
First published: 16 April 2025

Abstract

Electrolytes are an essential part of rechargeable batteries, such as lithium batteries. However, electrolyte innovation is facing grand challenges due to the complicated solution chemistry and infinite molecular space (>1060 for small molecules). This work reported an artificial intelligence (AI) platform, namely Uni-Electrolyte, for designing advanced electrolyte molecules, which mainly includes three parts, i.e., EMolCurator, EMolForger, and EMolNetKnittor. New molecules can be designed by combining high-throughput screening and generative AI models from more than 100 million alternative molecules in the EMolCurator module. The molecular properties, including frontier molecular orbital information, formation energy, binding energy with a Li ion, viscosity, and dielectric constant, can be adopted as the screening parameters. The EMolForger and EMolNetKnittor modules can predict the retrosynthesis pathway and solid electrolyte interphase (SEI) formation mechanism for a given molecule, respectively. With the assistance of advanced AI methods, the Uni-Electrolyte is strongly supposed to discover new electrolyte molecules and chemical principles, promoting the practical application of next-generation rechargeable batteries.

Conflict of Interests

The authors declare no conflict of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

The full text of this article hosted at iucr.org is unavailable due to technical difficulties.