Machine learning for data-driven design of high-safety lithium metal anode

Area of Science:

• Materials Science
• Electrochemistry
• Computational Chemistry

Background:

• Lithium metal anodes are crucial for high-energy-density batteries.
• Developing stable interphase layers is key to improving lithium metal anode performance and safety.
• Current methods for identifying suitable interphase materials can be time-consuming and expensive.

Purpose of the Study:

• To present a protocol for developing inorganic-organic hybrid interphase layers for lithium metal anodes.
• To enable a cost-effective workflow for identifying promising self-assembled monolayers (SAMs).
• To enhance the surface of lithium metal anodes for improved performance.

Main Methods:

• Utilizing the self-assembled monolayers (SAMs) technique.
• Extracting organic molecules from open-sourced databases.
• Calculating microscopic properties of organic molecules.
• Developing a machine learning model to predict ionic diffusion barriers.
• Preparing input data for machine learning predictions.

Main Results:

• A protocol for creating hybrid interphase layers on lithium metal anodes.
• A method for cost-effectively identifying SAMs with exceptional performance.
• The protocol facilitates the prediction of ionic diffusion barriers.
• Enables efficient screening of potential SAM candidates.

Conclusions:

• The presented protocol offers a streamlined and economical approach to designing advanced interphase layers.
• This method accelerates the discovery of high-performance SAMs for lithium metal anodes.
• The developed workflow contributes to the advancement of battery technology through materials innovation.