Spectroscopic properties of chlorophylls and bacteriochlorophylls studied by molecular orbital CI methods

Abstract

Systematic computational methods are needed for prediction of spectroscopic properties of photosynthetic chromophores in order to understand their function in specialised photosynthetic complexes. We have used quantum chemical MO/CI methods to study spectroscopic properties of chlorophylls a, b, c1, c2, c3 and d and bacteriochlorophylls a, b, c, d, e, f, g, h. Full energy minimisation of the chromophores were performed at semi-empirical PM3, ab-initio 6-31G* and density functional (DFT) level of calculation. Several CI-expansions were tested for their predictive power. All methods gave linear correlations between the experimentally observed and calculated spectroscopic Qy, Qx and Soret transition energies within a group of chromophores. The ZINDO/S CIS (15,15) method gave best results in the overall simulation of absorption spectra, both intensities and wavelengths. Calculations for 1:1 choromophore ¿ solvent complexes predict solvent induced spectroscopic shifts of the Qx and Soret bands, while the Qy transitions remain practically unshifted, in accord with experimental findings. The shifts originate from energy level perturbations due to solvent interaction with the charge distribution of the magnesium atom in the excited states. Such weak interactions have been shown to be are important in the B800 ring of the LH2 of purple bacteria and are expected to be important in all light harvesting protein complexes, where chromophores are far apart and interact with the nearby protein environment, e.g. in algae and plants. This work is dedicated to late Prof. Jan Amesz as he personally helped us to get the experimental solution spectra of Bchl b and Bchl g for this work.