Machine Learning-Assisted Discovery of Co-free and Low Strain Cathode Materials for Sodium-ion Batteries KIM, MIN-SEON Department of mechanical engineering Graduate School of Soongsil University This study underscores the promise of sodium-ion batteri...
Machine Learning-Assisted Discovery of Co-free and Low Strain Cathode Materials for Sodium-ion Batteries KIM, MIN-SEON Department of mechanical engineering Graduate School of Soongsil University This study underscores the promise of sodium-ion batteries(SIBs) as viable alternatives to lithium-ion counterparts, emphasizing their cost efficiency and similar insertion mechanisms. Layered transition metal oxides(LTMOs) are highlighted as SIB cathodes, attributed to their high voltages and ease of synthesis. However, O3-type LTMOs encounter phase transitions and performance degradation during the insertion/extraction processes. This study innovates by creating multi-element compounds to enhance performance, utilizing machine learning(ML) and density functional theory(DFT) for evaluating cathode stability and performance. Through data sampling and feature engineering, it enhances model accuracy and introduces a database for identifying stable Co-free cathodes. The research offers solutions for high energy density and reduced structural deformation in SIBs, supported by a ML-based platform for phase transition detection.