Excited state properties of the astaxanthin radical cation: A quantum chemical study

Vibrational relaxation as the driving force for wavelength conversion in the peridinin-chlorophyll a-protein

We present a computationally derived energy transfer model for the peridinin-chlorophyll a-protein (PCP), which invokes vibrational relaxation in the two lowest singlet excited states rather than internal conversion between them. The model allows an understanding of the photoinduced processes without assuming further electronic states or a dependence of the 2Ag state character on the vibrational sub-state. We report molecular dynamics simulations (CHARMM22 force field) and quantum mechanics/molecular mechanics (QM/MM) calculations on PCP. In the latter, the QM region containing a single peridinin (Per) chromophore or a Per-Chl a (chlorophyll a) pair is treated by density functional theory (DFT, CAM-B3LYP) for geometries and by DFT-based multireference configuration interaction (DFT/MRCI) for excitation energies. The calculations show that Per has a bright, green light absorbing 2Ag state, in addition to the blue light absorbing 1Bu state found in other carotenoids. Both states undergo a strong energy lowering upon relaxation, leading to emission in the red, while absorbing in the blue or green. The orientation of their transition dipole moments indicates that both states are capable of excited-state energy transfer to Chl a, without preference for either 1Bu or 2Ag as donor state. We propose that the commonly postulated partial intramolecular charge transfer (ICT) character of a donating Per state can be assigned to the relaxed 1Bu state, which takes on ICT character. By assuming that both 1Bu and 2Ag are able to donate to the Chl a Q band, one can explain why different chlorophyll species in PCP exhibit different acceptor capabilities.

Density functional theory studies of the extent of hole delocalization in one-electron oxidized adenine and guanine base stacks

This study investigates the extent of hole delocalization in one-electron oxidized adenine (A) and guanine (G) stacks and shows that new IR vibrational bands are predicted that are characteristic of hole delocalization within A-stacks. The geometries of A-stacks (A(i); i = 2-8) and G-stacks (GG and GGG) in their neutral and one-electron oxidized states were optimized with the bases in a B-DNA conformation using the M06-2X/6-31G* method. The highest occupied molecular orbital (HOMO) is localized on a single adenine in A-stacks and on a single guanine in GG and GGG stacks located at the 5'-site of the stack. On one-electron oxidation (removal of an electron from the HOMO of the neutral A- and G-stacks) a "hole" is created. Mulliken charge analysis shows that these "holes" are delocalized over two to three adenine bases in the A-stack. The calculated spin density distribution of A(i)(•+) (i = 2-8) also showed delocalization of the hole predominantly on two adenine bases, with some delocalization on a neighboring base. For GG and GGG radical cations, the hole was found to be localized on a single G in the stack. The calculated HFCCs of GG and GGG are in good agreement with the experiment. Further, from the vibrational frequency analysis, it was found that IR spectra of neutral and the corresponding one-electron oxidized adenine stacks are quite different. The IR spectra of A(2)(•+) has intense IR peaks between 900 and 1500 cm(-1) that are not present in the neutral A(2) stack. The presence of A(2)(•+) in the adenine stack has a characteristic intense peak at ~1100 cm(-1). Thus, IR and Raman spectroscopy has potential for monitoring the extent of hole delocalization in A stacks.