SurFF: a foundation model for surface exposure and morphology across intermetallic crystals

Area of Science:

• Materials Science
• Catalysis
• Computational Chemistry

Background:

• Heterogeneous catalysis is crucial for ~90% of industrial reactions.
• Accurate prediction of surface properties is vital for designing effective heterogeneous catalysts.
• Current experimental and computational methods for surface property prediction are costly and time-consuming.

Purpose of the Study:

• To develop a rapid and accurate model for predicting surface exposure and synthesizability of intermetallic crystals.
• To accelerate the design and discovery of novel heterogeneous catalysts.

Main Methods:

• Developed a foundation force-field-based model named SurFF.
• Created a comprehensive intermetallic surface database using active learning and high-throughput density functional theory (DFT) calculations.
• The database includes 12,553 unique surfaces and 344,200 single points.

Main Results:

• SurFF achieves DFT-level precision with a prediction error of 3 meV/Ų.
• Enables a 10⁵-fold acceleration in large-scale surface exposure prediction.
• Validation against computational and experimental data shows strong alignment.

Conclusions:

• SurFF provides a highly accurate and efficient method for predicting surface properties of intermetallic crystals.
• The model facilitates large-scale predictions of surface energy and Wulff shapes for over 6,000 intermetallic crystals.
• This work offers valuable data resources for the catalysis research community.