Predictive power of long-range corrected functionals on the spectroscopic properties of tetrapyrrole derivatives for photodynamic therapy

Porphyrin and chlorin based compounds possess promising properties to be utilized as photosensitizers in photodynamic therapy (PDT). However, the photosensitizers available on the market today are not ideal for use in PDT, which has emphasized the need for new photosensitizers with improved photodynamic properties to be developed. Computational drug-design can be utilized in the search for improved pharmaceutical compounds, provided that the methods used are able to reproduce experimental data. In the present study we investigated, by the use of time-dependent density functional theory (TD-DFT), the performance of the long-range corrected functionals ωB97, ωB97X and ωB97XD on their ability to predict low-lying singlet excitations (>600 nm) of a set of well-known photosensitizing compounds. It was found that ωB97X reproduced the experimental red-most absorption band most satisfactorily. The use of either B3LYP, ωB97XD or M06 in geometry optimizations has a minor effect on the spectra in most cases. Calculated energy differences between the optimized singlet ground states and optimized first excited triplet states show consistent and overall higher triplet state energies for B3LYP, M06, and PBE0 compared with ωB97, ωB97X, and ωB97XD. The calculated triplet state energies are, however, sufficient to generate singlet oxygen in most cases.