Efficient workflow for the investigation of the catalytic cycle of water oxidation catalysts: Combining GFN-xTB and density functional theory.

Photocatalytic water oxidation remains the bottleneck in many artificial photosynthesis devices. The efficiency of this challenging process is inherently linked to the thermodynamic and electronic properties of the chromophore and the water oxidation catalyst (WOC). Computational investigations can facilitate the search for favorable chromophore-catalyst combinations. However, this remains a demanding task due to the requirements on the computational method that should be able to correctly describe different spin and oxidation states of the transition metal, the influence of solvation and the different rates of the charge transfer and water oxidation processes. To determine a suitable method with favorable cost/accuracy ratios, the full catalytic cycle of a molecular ruthenium based WOC is investigated using different computational methods, including density functional theory (DFT) with different functionals (GGA, Hybrid, Double Hybrid) as well as the semi-empirical tight binding approach GFN-xTB. A workflow with low computational cost is proposed that combines GFN-xTB and DFT and provides reliable results. GFN-xTB geometries and frequencies combined with single-point DFT energies give free energy changes along the catalytic cycle that closely follow the full DFT results and show satisfactory agreement with experiment, while significantly decreasing the computational cost. This workflow allows for cost efficient determination of energetic, thermodynamic and dynamic properties of WOCs.