AbstractWe have applied our practice‐oriented TDDFT benchmark strategy to assess the performance of pure, hybrid, range‐separated hybrid and double‐hybrid functionals for reproducing the electronic spectra of a set of 137 transition‐metal complexes. Our results enable simple pre‐screening of new transition‐metal based photocatalysts. The reference data is based on recently published measurements, from which we have established a new database. The database is named TMPHOTCAT‐137 to reflect the number of complexes (with Cu, Ru, Ir, Fe, Au, Mo and W centres) included and the fact that all of them are either proven or potential photocatalysts. We have found that the M06 functional is the best performer both in terms of practical accuracy and consistency for all metals except Fe. While B3LYP is similarly accurate for Ru and Ir complexes, for the rest of the metals significant wavelength scaling is necessary. Compared to our previous benchmark for organic photocatalysts, double‐hybrid functionals exhibit considerably poorer results while the range separated hybrids CAM−B3LYP and ωB97X−D offer the most consistent performance for both datasets. Among the metals, iron proved to be most difficult for TDDFT: even M06 predicts singlet‐singlet absorption spectra and quintet ground states erroneously, especially for weak ligand fields (i. e., only Fe−N bonds). The functionals that do not exhibit this behaviour are the pure (GGA, meta‐GGA) functionals, but the role of HF exchange in the spectral prediction is equivocal; while functionals with high amounts of HF exchange perform insufficiently, smaller amount of exact exchange yields overall better performance (M06, B3LYP). We have also shown that inclusion of spin‐orbit coupling for the heavier metals does not improve the results. The updated spectrum optimizer code, the TMPHOTCAT‐137 database and the Jupyter Notebook used for analysis are available at github.com/PeterF1234/spectrum_optimizer.
Read full abstract