Machine Learning Assisted Screening of MXene with Superior Anchoring Effect in Al–S Batteries

Dissolution of polysulfide intermediates into electrolytes has been a major bottleneck in the development of the Al–S battery. MXenes can be promising anchoring materials, even though finding the most suitable candidates from a vast search space in a short span of time is challenging. Herein, a combined density functional theory and machine learning (ML) approach has been implemented to predict suitable M1M2XT₂-type MXene materials that can optimally anchor the polysulfide intermediates. By employing various ML algorithms, the trained XGBR model is found to exhibit remarkable precision in predicting the anchoring effect of MXenes. 42 promising candidates have been identified to show optimum anchoring across the Al–S intermediates. The F and O terminal groups are found to majorly exhibit the optimum anchoring effect toward the possible polysulfide intermediates. This work provides crucial insights into the development of next-generation Al–S batteries accelerated by the ML approach, contributing to the advancement of energy storage technologies.