Using CO Adsorption on Cu Surfaces to Benchmark DFT Parameters for Catalytic Modeling

ABSTRACT Accurate modeling of CO adsorption on copper surfaces is essential for catalytic processes such as syngas conversion and methanol synthesis. However, standard generalized gradient approximation (GGA) functionals cannot reproduce physical site preferences due to the “CO adsorption puzzle.” Nonetheless, resolving site preference alone does not ensure a computationally feasible methodology for studying catalytic COCu systems. A systematic approach is required to identify a balanced framework that is both accurate and feasible for large‐scale catalytic studies. Thus, this work systematically evaluates the effects of basis sets, exchange functionals, dispersion and Hubbard‐U adjustments for CO adsorption on Cu(100), Cu(110), Cu(111), and Cu(211) surfaces using periodic DFT computations. While experimentally observed on‐top adsorption can be achieved using the Hubbard‐U correction, dispersion corrections only improve adsorption energy agreement with experimental data, not the site preference. Among the tested setups, projector augmented‐wave method with the Perdew–Burke–Ernzerhof exchange‐correlation (PAW‐PBE) with U = 8 eV and Grimme‐D3 dispersion correction emerged as a reliable and efficient combination, consistently reproducing adsorption energies and vibrational properties in agreement with experimental reports. To validate this setting in catalytic CO–Cu systems, the approach was extended to methanol adsorption. This work provides a unified, computationally efficient DFT + U + D3 protocol that reproduces experimental CO adsorption behavior and can be readily applied to other catalytic Cu systems.