Integrating Molecular Dynamics and Machine Learning for Solvation-Guided Electrolyte Optimization in Lithium Metal...

Area of Science:

• Materials Science
• Electrochemistry
• Computational Chemistry

Background:

• Liquid electrolytes are crucial for lithium metal battery performance, but side reactions and dendrite formation limit stability and lifespan.
• Low Coulombic efficiency (CE) and reduced cycle life are persistent challenges in current lithium metal battery technology.

Purpose of the Study:

• To develop a rapid, cost-effective strategy for optimizing liquid electrolytes for lithium metal batteries.
• To enhance long-term cycling stability and Coulombic efficiency through integrated computational and experimental approaches.

Main Methods:

• Integrated high-throughput molecular dynamics simulations with machine learning predictions for electrolyte screening.
• Experimental validation of promising electrolyte formulations identified through computational modeling.
• Identification of key molecular descriptors influencing electrolyte performance.

Main Results:

• A mixed electrolyte (LiFSI salt, DEE solvent, LiNFS additive) achieved a significantly improved CE of 98.32%.
• Identified optimal electrolyte compositions favoring specific salt concentrations and elemental content in salts and solvents.
• Established a reusable modeling framework for accelerated electrolyte design.

Conclusions:

• The integrated simulation and machine learning approach effectively accelerates the discovery of high-performance electrolytes.
• The optimized electrolyte composition demonstrates superior cycling stability and Coulombic efficiency for lithium metal batteries.
• This framework offers a resource-efficient pathway for targeted electrolyte design and optimization.