Chemically specific dynamic bond percolation model for computational screening of polymer electrolytes

Abstract

Polymer electrolytes have significant promise for many lithium-ion battery applications because they are non-flammable, electrochem. stable, and easily manufd. However, significant improvements in the ionic cond. of even state- of-the-art polymer electrolytes is needed for technol. viability. The computationally- guided design of more conductive polymer electrolytes would save significant time and monetary resources, but these efforts require both a better understanding of ion- transport mechanisms in polymers and the development of tools for screening candidate polymers prior to synthesis and characterization. Using atomistic mol. dynamics (MD) simulations, we provide new insights into the mechanisms of lithium-ion diffusion in polymer electrolyte materials, including that the spatial distribution of lithium- ion solvation sites, in addn. to polymer segmental motion, is a crucial factor for detg. polymer electrolyte performance. In addn., we have leveraged these insights to develop a powerful new coarse-grained simulation protocol that enables the rapid screening of large nos. of polymer electrolyte materials using short-timescale MD trajectories. In particular, we extend the well-known dynamic bond percolation model [J. Phys. 1983, 79, 3133- 3142] to utilize short MD trajectories to predict the distribution of ion solvation sites and the rate of hopping among the ion solvation sites. Among the most striking results from our initial studies is the unexpectedly good lithium-ion diffusivity of poly(trimethylene oxide-alt-ethylene oxide) by comparison to PEO, which is widely used. Efforts to screen upwards of 500 candidate polymer structures utilizing the new model are currently underway.