Screening and Understanding Li Adsorption on Two-Dimensional Metallic Materials by Learning Physics and...

Area of Science:

• Materials Science
• Computational Chemistry
• Surface Science

Background:

• Lithium (Li) interaction energetics with surfaces are crucial for developing advanced materials.
• Applications include Li-based electrochemical capacitors, Li sensors, Li separation membranes, and Li-ion batteries.

Purpose of the Study:

• To establish a high-throughput screening scheme for evaluating Li adsorption energetics on 2D metallic materials.
• To accelerate the discovery of novel 2D materials for Li-related technologies.

Main Methods:

• Utilized density functional theory (DFT) and graph convolution networks (GCNs) to calculate minimum Li adsorption energies.
• Developed a predictive model based on ionization potential, work function, and Li+ coupling energy.
• Employed physics-simplified machine learning by decomposing material properties.

Main Results:

• Successfully screened Li adsorption energetics across a wide range of 2D metallic materials.
• Identified key descriptors governing Li adsorption, enabling accurate predictions.
• Demonstrated higher accuracy and transferability of the physics-simplified machine learning approach.

Conclusions:

• The developed high-throughput screening scheme significantly accelerates the identification of suitable 2D materials for Li applications.
• Physics-simplified machine learning offers a more accurate and transferable method for predicting complex material properties.
• This work provides a foundation for designing next-generation Li-based energy storage and separation technologies.