Development of Stable Layered Oxide Cathode Materials Assisted By Machine Learning for K-Ion Batteries

Abstract

Recently, the emergence of an artificial intelligence method, especially a machine-learning (ML) approach, has shown its potential to accelerate research and development for materials science. On the other hand, density functional theory (DFT) calculations have the ability to predict materials properties with relatively high accuracy.1 In the field of battery researches, using traditional methods, such as the empirical trial and error methods, hinders the development of new advanced electrode materials, particularly in potassium-ion batteries (KIB).2-4 To address this issue, here we develop for the first time a protocol using a combined ML and DFT calculations for materials screening followed by experimental verification. Notably, layered oxide cathode materials, such as KxMnO2, which have been widely investigated for KIB due to their high energy and power densities, suffer from structural instability and highly hygroscopic nature. Therefore, this proposed protocol is used for designing structurally stable KxMnO2 with enhanced environmental stabilities and high electrochemical performances. Among the numbers of candidates, notably, the ML and DFT-assisted strategies identify P′3-type K0.3Mn0.9Cu0.1O2 (KMCO) as a promising cathode for a high-performance KIB application. Finally, the experimental results further prove that the KMCO cathode has substantially improved K-storage properties with high-power density and cycling stability even after 4-week air-exposure period. We believe that this study opens a new avenue for identifying and developing suitable electrode materials for future battery applications. References B. Scrosati, M. Bottini and T. Mustelin, Nanotechnol., 2007, 2, 598 —599.N. Yabuuchi, K. Kubota, M. Dahbi and S. Komaba, Chem. Rev., 2014, 114, 11636 —11682. A. Eftekhari, Z. Jian and X. Ji, ACS Appl. Mater. Interfaces, 2017, 9, 4404 —4419.N. Xiao, W. D. McCulloch and Y. Wu, J. Am. Chem. Soc., 2017, 139, 9475 —9478