A multi-resolution systematically improvable quantum embedding scheme for large-scale surface chemistry calculations

Area of Science:

• Computational Chemistry
• Surface Science
• Quantum Many-Body Methods

Background:

• Predictive simulation of surface chemistry is crucial for catalysis, electrochemistry, and clean energy.
• Ab-initio quantum many-body methods offer electronic-level insights but face computational cost limitations.

Purpose of the Study:

• To develop accurate and computationally efficient quantum chemistry methods for large-scale surface chemistry simulations.
• To benchmark water-graphene interactions and study carbonaceous molecule adsorption on complex surfaces.

Main Methods:

• Utilizing state-of-the-art correlated wavefunctions for 'gold standard' accuracy.
• Employing graphics processing unit (GPU) acceleration and multi-resolution techniques.
• Achieving linear computational scaling for systems up to 392 atoms.

Main Results:

• Demonstrated consistency in large-scale simulations, validating results across different boundary conditions.
• Provided a benchmark for water-graphene interaction, clarifying water orientation preferences.
• Achieved chemical accuracy for adsorption of carbonaceous molecules on metal oxides and metal-organic frameworks.

Conclusions:

• Advanced ab-initio quantum many-body methods enable reliable and improvable first-principles modeling of molecular adsorption on surfaces.
• The developed methods are critical for understanding and designing systems in catalysis, electrochemistry, and clean energy.