Improving Molecule-Metal Surface Reaction Networks Using the Meta-Generalized Gradient Approximation: CO₂...

Area of Science:

• Computational chemistry
• Materials science
• Catalysis research

Background:

• Density functional theory (DFT) is crucial for understanding catalysis, particularly molecule-metal surface reactions.
• Generalized gradient approximation (GGA) density functionals (DFs) often used in catalysis studies exhibit limitations in accurately describing both gas-phase molecules and metal surfaces.
• GGA DFs tend to underestimate reaction barriers due to self-interaction error, limiting their predictive power.

Purpose of the Study:

• To investigate the efficacy of a meta-generalized gradient approximation (mGGA) density functional combined with nonlocal correlation for studying catalytic reaction networks.
• To compare the performance of the chosen mGGA functional against traditional GGA functionals for molecule-metal surface interactions.
• To provide new insights into the CO2 hydrogenation reaction network on copper (Cu) using an improved DFT approach.

Main Methods:

• Utilized a meta-generalized gradient approximation (mGGA) density functional (rMS-RPBEl-rVV10) in conjunction with a nonlocal correlation functional.
• Applied the chosen DFT method to study the CO2 hydrogenation reaction network on a Cu surface.
• Evaluated the accuracy of the mGGA functional for predicting adsorption and activation energies of molecules on metal surfaces.

Main Results:

• The rMS-RPBEl-rVV10 mGGA functional demonstrated superior performance compared to typical GGA DFs.
• The mGGA functional provided comparable or improved accuracy for describing metals, gas-phase molecules, and molecule-metal surface interactions.
• Accurate prediction of adsorption and activation energies was achieved, outperforming standard GGA functionals.

Conclusions:

• The tested mGGA density functional (rMS-RPBEl-rVV10) is a more suitable choice for modeling molecule-metal surface reaction networks in catalysis.
• This approach offers a more accurate and affordable alternative to screened hybrid functionals for catalysis research.
• The findings pave the way for more reliable computational studies of catalytic processes.