Observation of dendrite formation at Li metal-electrolyte interface by a machine-learning enhanced constant potential...

Area of Science:

• Materials Science
• Electrochemistry
• Computational Chemistry

Background:

• Uncontrolled lithium dendrite growth in lithium metal batteries causes low Coulombic efficiency and safety concerns.
• Understanding dendrite formation is crucial for improving lithium metal battery performance.

Purpose of the Study:

• To develop and apply a novel simulation method for investigating lithium dendrite nucleation at the atomic scale.
• To elucidate the mechanism of dendrite formation under constant potential conditions.

Main Methods:

• Utilized machine learning-accelerated molecular dynamics simulations.
• Implemented a constant potential approach combining a machine learning force field with the charge equilibration method.
• Simulated lithium deposition dynamics on lithium metal anode surfaces.

Main Results:

• Identified inhomogeneous lithium deposition and aggregation within solid electrolyte interphase amorphous inorganic components as key initiators of dendrite nucleation.
• Provided atomic-scale insights into the dynamic process of dendrite formation.
• Demonstrated the efficacy of the proposed constant potential simulation method.

Conclusions:

• The developed simulation method accurately models constant potential conditions for electrochemical interfaces.
• Microscopic insights into lithium dendrite formation offer pathways for enhancing lithium metal battery performance and safety.
• The simulation approach has broad potential for modeling complex electrochemical systems.