Alchemical Predictions for Computational Catalysis: Potential and Limitations

Area of Science:

• Computational Chemistry
• Materials Science
• Catalysis

Background:

• Kohn-Sham density functional theory (DFT) is crucial for calculating adsorbate binding energies in catalysis.
• DFT's high computational cost limits its application in broad catalyst candidate screening.
• Developing faster methods is essential for efficient catalyst discovery.

Purpose of the Study:

• To assess the potential of computational alchemy as a rapid alternative to DFT for predicting binding energies.
• To benchmark computational alchemy's accuracy against DFT for oxygen reduction reaction intermediates on Pt, Pd, and Ni alloys.
• To identify limitations and areas for improvement in computational alchemy modeling.

Main Methods:

• Computational alchemy, a perturbation theory approach, was employed to predict binding energies.
• Binding energies of oxygen reduction reaction intermediates on Pt, Pd, and Ni alloys were calculated using both alchemy and DFT.
• Alchemical predictions were benchmarked against DFT results.

Main Results:

• Computational alchemy provided binding energy predictions within 0.1 eV of DFT values in many instances.
• Alchemical estimates were thousands of times faster than DFT calculations.
• Specific cases showed significant discrepancies, highlighting areas requiring modeling enhancements.

Conclusions:

• Computational alchemy is a highly promising tool for accelerating the prediction of binding energies in catalysis.
• The method demonstrates significant speed advantages over DFT for high-throughput screening.
• Further research is warranted to refine computational alchemy for broader and more accurate application in heterogeneous catalyst design.