Aqueous magnesium-air (Mg-air) batteries possess a high theoretical voltage and specific energy density, extended shelf life, and safety features. They consist of a magnesium anode and an air electrode. However, their performance remains suboptimal due to several factors, the main ones being the overvoltage resulting from insoluble discharge products and rapid self-corrosion of magnesium anodes in aqueous electrolytes. Addressing these challenges is crucial for unlocking the full potential of Mg-air batteries.
There are two main approaches to address these limitations – Mg alloy development and electrolyte engineering. Electrolyte additives have the potential to be efficient regulators of interfacial processes. Here, we present insights into the mechanisms of action of several Mg2+ complexing agents as electrolyte additives for aqueous Mg-air batteries. The results suggest that these agents increase the battery utilization efficiency by inhibiting self-corrosion and decreasing the non-uniform dissolution of the Mg anode. Selection of superior electrolyte additives was facilitated by data-driven models that allow rapid and efficient screening of large numbers of organic compounds as potential electrolyte additives. Here, we describe two data-driven quantitative structure-property relationship (QSPR) machine learning models that were trained and used for in silico searches for promising battery performance-boosting candidates.