Database Driven Solid-State Electrolyte Material Search for Li and Na-Metal

Li and Na metal anode-based batteries are currently some of the most desirable technologies for high-energy density storage systems. However, internal dendrite formation is considered one of the biggest challenges in battery manufacturing. Suppressing the dendrite formation of Li or Na metal might be achievable with solid-state electrolytes as they have demonstrated a higher chemical stability in the presence of the anode. The development of a solid-state electrolyte for Li and Na metal anode involves several challenges, including the need for the material to have high ionic conductivity, good mechanical properties, and good compatibility with the anode material. To address these challenges, there is a need for a systematic approach to discover new solid-state electrolytes that are specifically tailored for Li or Na metal anodes. The objective of this work is to screen out optimal and compatible solid-state electrolytes for both Li and Na metal batteries from the Materials Project (MP) database [1]. In total, five criteria were considered in the screening process [2] – the formation energy, energy above hull, band gap, reaction energy with anode material and mechanical property – which are all available within the MP database. After screening out the potential candidates, we assessed their dynamic/transport properties. By performing molecular dynamics simulations using the universal graph deep learning interatomic potential M3Gnet [3], the ionic diffusivity for all solid-state electrolyte candidates was analyzed. Through this work, we provide the Pareto front for the most optimal solid-state electrolytes for both Li and Na metal anode batteries.<br/><br/>[1] A. Jain, S. P. Ong, G. Hautier, W. Chen, W. D. Richards, S. Dacek, S. Cholia, D. Gunter, D. Skinner, G. Ceder, and K. A. Persson, <i>APL Mater</i><i>.</i> <b>1</b>, 011002 (2013)<br/>[2] L. Kahle, A. Marcolongo, and N. Marzari, <i>Energy Environ. Sci.</i> <b>13</b>, 928 (2020)<br/>[3] C. Chen and S. P. Ong, <i>Nat. Comput. Sci.</i> <b>2</b>, 718 (2022)