Design of next-generation polymer electrolytes: A site network model for ion transport from mechanistic insight

Abstract

We employ atomistic mol. dynamics (MD), in tandem with exptl. studies, to elucidate the mechanisms of lithiumion transport in new carbonyl-, allyl-, and ether-contg. polymer electrolytes. We find that that changes in the lithium-ion coordination environment play a crucial role in the lithium-ion hopping dynamics in all polymers, and this motivates a more general view of ion-transport in which the ion diffusion occurs via transitions between viable solvation sites in the polymer. This view is incorporated into the development of a general model for ion transport in polymer electrolyte solvents. In this model, ion diffusion depends on two essential characteristics: the distribution of sites and the transition rates between them. The proposed model extends previously developed approaches for ion transport in polymers by establishing how site distributions and transition rates can be derived from polymer-specific properties. Importantly, all properties needed for the model are readily detd. from relatively short (10-20 ns) MD simulations. The transport model is used to assess the relative lithium-ion transport characteristics of six polyethers that differ by percent oxygen content and backbone stiffness. Good agreement is obsd. between the results of long timescale (>100 ns) MD simulations and those of kinetic Monte Carlo (KMC) simulations based on the model, suggesting that the essential physics of ion transport in polymers is included. The model provides a new and unexpected prediction that poly(trimethylene oxide-alt-ethylene oxide) exhibits a higher lithium-ion cond. than poly(ethylene oxide) in the dil.-ion limit; the conduction properties of the former have not been studied to our knowledge, whereas the latter is among the most conductive homopolymer electrolyte solvents. These results illustrate the potential of the model as a tool for screening candidate polymer electrolytes at reduced computational cost.