Understanding Ionic Diffusion Mechanisms in Li2S Coatings for Solid-State Batteries: Development of a Tailored Reactive Force Field for Multiscale Simulations

In order to investigate Li2S as a potential protective coating for lithium anode batteries using superionic electrolytes, we need to describe reactions and transport for systems at scales of >10,000 atoms for time scales beyond nanoseconds, which is most impractical for quantum mechanics (QM) calculations. To overcome this issue, here, we first report the development of the reactive analytical force field (ReaxFF) based on density functional theory (DFT) calculations on model systems at the PBE0/TZVP and M062X/TZVP levels. Then, we carry out reactive molecular dynamics simulations (RMD) for up to 20 ns to investigate the diffusion mechanisms in bulk Li2S as a function of vacancy density, determining the activation barrier for diffusion and conductivity. We show that RMD predictions for diffusion and conductivity are comparable to experiments, while results on model systems are consistent with and validated by short (10-100 ps) ab initio molecular dynamics (AIMD). This new ReaxFF for Li2S systems enables practical RMD on spatial scales of 10-100 nm (10,000 to 10 million atoms) for the time scales of 20 ns required to investigate predictively the interfaces between electrodes and electrolytes, electrodes and coatings, and coatings and electrolytes during the charging and discharging processes.

Automated summary

This study develops a reactive analytical force field (ReaxFF) method based on density functional theory (DFT) for investigating Li2S as a protective coating for lithium anode batteries. Through reactive molecular dynamics simulations, the diffusion and conductivity of Li2S crystals are found to be consistent with experimental results, validating the reliability of the model. This method provides a practical tool for studying interfaces during the charging and discharging processes of batteries.