Machine learning-assisted design and prediction of materials for batteries based on alkali metals

Abstract

Since the commercialization of lithium ion batteries in 1990s, batteries based on alkali metals have been the promising candidates for energy storage. The performances of the batteries, in term of cost efficiency, energy density, safety, and cycle life, are requiring continous improvement. The performances are highly determined by the electrode materials, yet the bottleneck of the improvement of electrode material results from the long experialmental period, numberous reseachers, and high cost. Machine learning (ML), which is being increasingly integrated into materials science, offers transformative potential by reducing R&D period and cost. ML also demonstrates significant perspective in the performance prediction of various materials, and it can also help reveal the structure-performance relationship of materials. ML-assisted material design and performance prediction enable thus the advanced material inventions. The implemention of ML for researching alkali metal-based batteries is outlined with highlighting the ML algorithms as well as electrode reaction mechanisms.