Dec 2, 2024
3:45pm - 4:00pm
Hynes, Level 3, Ballroom C
Oskar Soulas1,Alex Squires1,Jeremy Lowen1,Josh Makepeace1,David Scanlon1
University of Birmingham1
Oskar Soulas1,Alex Squires1,Jeremy Lowen1,Josh Makepeace1,David Scanlon1
University of Birmingham1
Antifluorite lithium imide (Li<sub>2</sub>NH) has been extensively studied for use in hydrogen storage, but its high Li-ion conductivity has allowed it to be re-examined as a potential solid electrolyte.<sup>1,2</sup> This is particularly interesting as the constituent elements are earth abundant and lightweight. Synthesis and processing are also cheap and require low energy input compared to other leading lithium solid electrolytes such as the lithium garnets.<sup>1,2</sup> Previous studies have identified a wide electrochemical stability window further suggesting that Li<sub>2</sub>NH could be suitable for use as a solid electrolyte.<sup>2</sup> Working with experimental collaborators, we are looking to understand the diffusion processes and validate as-yet-unpublished promising experimental results.<br/><br/>In this study, we investigate the nature of the ion-conducting properties found in members of the lithium-amide lithium-imide lithium-nitride solid solution. This starts with the end members, Li<sub>2</sub>NH and LiNH<sub>2</sub>. We observe atom dynamics on a femtosecond scale using high-quality bulk Ab Initio Molecular Dynamics (AIMD), allowing us to compare the bulk conductivities of the stoichiometric, lithium stuffed, and lithium deficient structures. Benchmarking MLIPs against DFT allows us to run these simulations over a long period of time at a reduced cost whilst ensuring that accuracy is maintained. Our analysis provides unprecedented insight into the coupled nature of ion movement in these systems.<br/><br/><sup>1</sup> Chen et al., <i>Nat., </i> 2002,<b> 420</b>, 302– 304<br/><sup>2</sup> Paik et al., <i>J. Phys. Chem. C</i>, 2019, <b>123</b>, 3, 1619–1625