Strategies to Improve the Quantum Computation Accuracy for Electrochemical Windows of Ionic Liquids

Abstract

Rational design of ionic liquids (ILs) with wide electrochemical windows (ECWs) for high-voltage cathodes is essential to elevating the energy density of current rechargeable batteries. It is significant to determine appropriate strategies for accurately predicting the ECWs of ILs. In this study, we compare the calculated ECWs based on three quantum methods, including the highest occupied molecular orbital−lowest unoccupied molecular orbital (HOMO−LUMO) method, the ionization potential−electron affinity (IP−EA) method, and the thermodynamic method, under four unique combinations of simulation environments, and assess the discrepancies between the calculated and the experimental results of ECWs. For the three quantum methods, although HOMO−LUMO and IP−EA methods show limited prediction accuracy compared to the experimental values, they can qualitatively rank the anodic limits of anions and the cathodic limits of cations. For the thermodynamic method, we demonstrate that the highest accuracy can be achieved by considering the most rational redox reaction process. By varying the calculation environments, the calculated ECWs tend to be underestimated by considering separate cations and anions of ILs under gas phase, whereas they always exhibit overestimated results when calculating based on the cation−anion pairs. When the computation considers isolated ions with the solvent model plus proper thermodynamic corrections, we observe improved consistency with the experimental results. Though all methods have limitations to achieving perfect predictions of ECWs, we believe rational selection of calculation methods based on application-oriented scenarios can balance the efficiency and accuracy of the results for the development of a high-throughput and accurate database of ECWs for ILs.

Integrated Supervised and Unsupervised Machine Learning Approach to Map the Electrochemical Windows Over 4500 Solvents for Battery Applications

“…Thus, the thermodynamic cycle-based approach offers substantial advantages over the HOMO/LUMO method by accounting for the reorganization energy and solvation effects, leading to more accurate and practical predictions of red–ox potentials. Recently Wang and his co-worker also demonstrated that the highest accuracy in determining the ECW can be achieved by the red–ox reaction process (thermodynamic method) compared to other methods . The traditional nonaqueous electrolytes primarily rely on organic carbonates, which exhibit ECWs of less than 5.0 V. These carbonates include both linear and cyclic solvent electrolytes such as dimethyl carbonate, diethyl carbonate, propylene carbonate, and ethylene carbonate.…”