Ion Transport Mechanisms via Time-Dependent Local Structure and Dynamics in Highly Concentrated Electrolytes

Artikel i vetenskaplig tidskrift, 2020

Highly concentrated electrolytes (HCEs) are attracting interest as safer and more stable alternatives to current lithium-ion battery electrolytes, but their structure, solvation dynamics and ion transport mechanisms are arguably more complex. We here present a novel general method for analyzing both the structure and the dynamics, and ultimately the ion transport mechanism(s), of electrolytes including HCEs. This is based on automated detection of bonds, both covalent and coordination bonds, including how they dynamically change, in molecular dynamics (MD) simulation trajectories. We thereafter classify distinct local structures by their bond topology and characterize their physicochemical properties by statistical mechanics, giving both a qualitative and quantitative description of the structure, solvation and coordination dynamics, and ion transport mechanism(s). We demonstrate the method by in detail analyzing an ab initio MD simulation trajectory of an HCE consisting of the LiTFSI salt dissolved in acetonitrile at a 1:2 molar ratio. We find this electrolyte to form a flexible percolating network which limits vehicular ion transport but enables the Li+ ions to move between different TFSI coordination sites along with their first solvation shells. In contrast, the TFSI anions are immobilized in the network, but often free to rotate which further facilitates the Li+ hopping mechanism.Highly concentrated electrolytes (HCEs) are attracting interest as safer and more stable alternatives to current lithium-ion battery electrolytes, but their structure, solvation dynamics and ion transport mechanisms are arguably more complex. We here present a novel general method for analyzing both the structure and the dynamics, and ultimately the ion transport mechanism(s), of electrolytes including HCEs. This is based on automated detection of bonds, both covalent and coordination bonds, including how they dynamically change, in molecular dynamics (MD) simulation trajectories. We thereafter classify distinct local structures by their bond topology and characterize their physicochemical properties by statistical mechanics, giving both a qualitative and quantitative description of the structure, solvation and coordination dynamics, and ion transport mechanism(s). We demonstrate the method by in detail analyzing an ab initio MD simulation trajectory of an HCE consisting of the LiTFSI salt dissolved in acetonitrile at a 1:2 molar ratio. We find this electrolyte to form a flexible percolating network which limits vehicular ion transport but enables the Li+ ions to move between different TFSI coordination sites along with their first solvation shells. In contrast, the TFSI anions are immobilized in the network, but often free to rotate which further facilitates the Li+ hopping mechanism.