Constructing Mixed Density Functionals for Describing Dissociative Chemisorption on Metal Surfaces: Basic Principles

Area of Science:

• Computational Chemistry
• Materials Science
• Chemical Physics

Background:

• Heterogeneous catalysis is vital in chemical production, with rates often limited by dissociative chemisorption transition states on metals.
• Accurate barrier heights for these processes are essential for benchmarking electronic structure theories, particularly density functionals.
• The semiempirical specific reaction parameter (SRP) approach to density functional theory (DFT) can yield accurate barriers but requires better methods for parameter determination.

Purpose of the Study:

• To investigate and improve heuristic approaches for determining mixing parameters in parametrized density functional expressions for SRP-DFT.
• To assess the performance of mixed density functionals combining generalized gradient approximation (GGA) exchange with GGA or nonlocal correlation in reproducing dissociative chemisorption barrier heights.

Main Methods:

• Tested seven mixed, parametrized density functional expressions on a database of 16 dissociative chemisorption systems (SBH17).
• Focused on functionals combining GGA exchange with GGA or nonlocal correlation, including those approximating van der Waals interactions.
• Analyzed the correlation between optimal exchange mixing parameters and the charge-transfer parameter for functionals with GGA correlation.

Main Results:

• Derived three highly tunable mixed density functionals using PBE GGA or nonlocal correlation forms.
• Found a generally weak correlation between the optimal fraction of RPBE GGA exchange and the charge-transfer parameter for GGA-correlated functionals.
• This weak correlation suggests the charge-transfer parameter is not a sufficiently effective strategy for predicting mixing parameters.

Conclusions:

• Developed improved mixed density functionals for accurate calculation of dissociative chemisorption barrier heights.
• Demonstrated that while charge-transfer parameter shows some correlation, it's insufficient for reliably predicting optimal mixing parameters in SRP-DFT.
• Highlights the need for further development of heuristic strategies for parameter determination in DFT for catalysis research.