A Thermodynamic Cycle‐Based Electrochemical Windows Database of 308 Electrolyte Solvents for Rechargeable Batteries

Rational design of wide electrochemical window (ECW) electrolytes to pair with high‐voltage cathodes is an emerging trend to push the energy density limits of current rechargeable batteries. Traditional single‐electronic/gas‐phase approximation‐based methods (e.g., highest occupied molecular orbital/lowest unoccupied molecular orbital) are increasingly recognized to have large deviations from experiments when predicting ECWs of electrolytes involving complex solvent interactions. Specifically, by examining available experimental ECWs of 68 electrolyte solvents extracted from ≈140 000 literature sources, which are conventionally divided into five functional‐group categories (covering commonly used carbonate‐based ethylene carbonate (EC)/propylene carbonate (PC) and ether‐based tetrahydrofuran), it is found that mean‐absolute‐errors (MAE) of traditional methods reach up to 3.25 V. Herein, a thermodynamic cycle‐based ECW prediction approach is proposed including two long‐term overlooked reorganization‐energy and solvation‐energy corrections, each of which can be quantified by two geometric descriptors (λ and ΔGsol), reducing MAE below 0.68 V. Following this, a database containing ECWs for 308 electrolyte solvents, obtained by traversing single functional‐group substitutions, is established. Furthermore, two omitted solvents with ECWs over 6.00 V and excellent structural stabilities (bond‐length change < 0.10 Å during redox process) are retrieved by stepwise screening of structural/electronic parallel properties. This study demonstrates the benefits of improving ECW prediction accuracy and accumulating descriptors to accelerate rapid screening of superior battery electrolytes. A thermodynamic cycle approach including reorganization‐energy (λ) and solvation‐energy (ΔGsol) corrections is proposed, realizing a 34.31% decrease in mean absolute error (MAE) compared to the traditional highest occupied molecular orbital/lowest unoccupied molecular orbital method. On this basis, a database containing electrochemical windows (ECWs) for 308 solvents by traversing single‐functional‐group substitutions is established, achieving a retrieval of two omitted solvents with ECWs over 6.00 V and excellent structural stabilities.