Preferable single-atom catalysts enabled by natural language processing for high energy density Na-S batteries

Employing appropriate single-atom (SA) catalysts in room-temperature sodium-sulfur (Na-S) batteries is propitious to promote the performance, whereas a universal designing strategy for the highly-efficient single-atom catalysts is absent. In this work, we adopt natural language processing techniques to screen the potential single-atom catalysts, then a binary descriptor is constructed to optimize the catalyst candidates. Atomically dispersed cobalt anchored to both nitrogen and sulfur atoms (SA Co-N/S) is selected as an ideal catalyst to significantly facilitate sulfur reduction reaction. The sulfur cathode catalyzed with SA Co-N/S almost realizes complete transformation, and the corresponding pouch cell exhibits satisfactory performance with high mass loading. In-situ X-ray absorption spectroscopy reveals the dynamical interactions between SA Co-N/S and sulfur species in the sulfur reduction reaction. Our work provides a method to select the preferable SA catalyst and to understand the interfacial catalysis dynamics in the sustainable Na-S systems. Employing appropriate catalysts in room-temperature sodium-sulfur batteries can significantly enhance performance. Here, authors utilize natural language processing techniques in conjunction with a binary descriptor to screen preferrable single-atom catalysts to achieve high specific capacity.