Molecular Dynamics and Emerging Network Graphs of Interactions in Dinitrile-Based Li-Ion Battery Electrolytes

Advancements in battery research have shown interesting formulations of battery electrolytes that have helped improve the efficiency of Li-ion batteries over the decades. However, the quest for a safer and affordable battery electrolyte still proceeds with more unique formulations reported in the literature regularly. The dinitriles, especially adiponitrile and glutaronitrile, have caught the attention of the research community as part of this quest. In this work, we performed molecular dynamics simulations of dinitrile electrolytes with lithium bistrifluorosulfonimide (LiTFSI) as the electrolyte salt at varying concentrations and temperatures. On analysis of our simulations, we find that the densities of the mixtures follow the same trend as that of experimental values. The solvation properties were explored using the radial distribution functions. The connectivity of the Li⁺ with the dinitrile molecules and anions is established for all of the electrolyte concentrations using network graphs. We observe that the electrolytes form highly networked structures as the concentration increases without being affected by the rise in temperature. The networking of ionic interactions was quantified by calculating the average degree of each graph. Ionic conductivity calculations were computed using three methods: Nernst-Einstein relation, correlated method, and current autocorrelation function. We report the importance of accounting for the correlated motion of ions while estimating the ionic conductivity. The correlated conductivity and current autocorrelation function calculations provide a satisfactory estimation of the ionic conductivity compared to the experimental values.