Revealing the optimal cycling intervals and atomic-scale mechanisms of alkaline alloy anodes for solid-state batteries

Abstract

Alkaline alloy anodes present a promising alternative to pure Li or Na anodes for solid-state batteries. However, identifying optimal cycling intervals (e.g., Li_xAl → Li_yAl) and atomic-scale mechanisms for driving multi-performance synergy remains challenging. This study employed high-throughput calculations and machine learning to screen Li/Na alloys, focusing on dendrite free and high reversibility. The cycling intervals for alloys (Li-Al, Li-Ga, and Li-In) containing Group 13 elements are cycled in the Li/Na-poor region, those for alloys (Li-Si, Li-Ge, and Na-Sn) containing Group 14 elements in the Li/Na-moderate region, and for alloys (Li-Mg) containing Group 2 element in the Li/Na-rich region. Through machine learning analysis, the atomic-scale mechanisms for achieving multi-performance synergy are a decrease in BCM (Bader charge of alloying element M) for Li alloys and an increase in VPA (volume per atom) for Na alloys. Experimental validation confirms that LiAl is optimal phase for Li-Al alloy, with the best cycling interval in the Li/Na-poor region, both aligning with computational results. These findings highlight the optimization of cycling intervals and atomic-scale mechanisms for improved Li/Na alloy performance, providing valuable guidance for their application in commercial solid-state batteries.