Abstract: Layered carbon materials (LCMs) are composed of basic carbon layer units, such as graphite, soft carbon, hard carbon, and graphene. While they have been widely applied in the anode of potassium-ion batteries, the potassium storage mechanisms and performances of various LCMs are isolated and difficult to relate to each other. More importantly, there is a lack of a systematic understanding of the correlation between the basic microstructural unit (crystallinity and defects) and the potassium storage behavior. In this review, we explored the key structural factors affecting the potassium storage in LCMs, namely, the crystallinity and defects of carbon layers, and the key parameters (La, Lc, d002, ID/IG) that characterize the crystallinity and defects of different carbon materials were extracted from various databases and literature sources. A structure-property database of LCMs was thus built, and the effects of these key structural parameters on the potassium storage properties, including the capacity, the rate and the working voltage plateau, were systematically analyzed. Based on the structure-property database analysis and the guidance of thermodynamics and kinetics, a relationship between various LCMs and potassium storage properties was established. Finally, with the help of machine learning, the key structural parameters of layered carbon anodes were used for the first time to predict the potassium storage performance so that the large amount of research data in the database could more effectively guide the scientific research and engineering application of LCMs in the future.

Key words: Layered carbon materials, Crystallinity and defects, Potassium storage properties, Machine learning