Excitonic States in Photosystem II Reaction Center

Abstract

The excited states of a structurally well-determined photosystem II (PSII) reaction center are obtained using an effective Hamiltonian for the interaction between the Q(y) states. The latter are calculated using the time-dependent density functional theory (DFT) method in DFT-optimized geometries, but with conserved side group orientations. Of particular importance is the orientation of the vinyl group of ring I. Couplings are calculated using actual transition charge distributions via the INDO/S model. Good agreement with experimental spectra is obtained. The lowest excited state is mainly located on the inactive B-side, but with a large component on P(A) too, making charge separation to H(A) possible at low temperature. The "trap state" and triplet state are localized on the inactive B-side. Since the spin singlet Q(y) states of the reaction center are all within a rather small energy range, the state with the highest component of B(A)*, on the blue side of the Q(y) absorption, has a rather high Boltzmann population at room temperature. The charge-transfer states, however, have a rather large spread and cannot be calculated accurately at present. The orientation of the phytyl chains is important and has as a consequence that the energy for the charge-separated B(A)+ H(A)- state is significantly lower than the corresponding state on the B-side. It follows that the B(A)* and P(A)* states are both possible origins for a fast charge separation in PSII.

Quantum Chemical Description of Absorption Properties and Excited‐State Processes in Photosynthetic Systems

“…This was mainly rationalized by the orientation of the vinyl chain. [477] Consequently, it was suggested that the major contribution to the lowest excited state is located at the accessory Chl pigment in branch A. [471,474,478] According to these studies, the second lowest excited state is dominated by the pheophytin residue at branch A.…”

A TDDFT investigation of the Photosystem II reaction center: Insights into the precursors to charge separation

“…Complementing the vast body of experimental work on the RC are various computational studies of the system ( 2 , 5 , 6 , 10 , 13 , 14 , 25 – 35 ). Early theoretical studies of the PSII RC revealed that there must be some key structural differences between this and the purple bacterial RC.…”

“…This was proven first by the Multimer model ( 2 ), which considered weak excitonic interactions between the majority of RC cofactors and could produce a good approximation of the experimental absorption spectrum despite there being no crystal structure data available at the time. As improvements in the resolution of crystal structures of PSII were published ( 3 , 36 , 37 ), it enabled new models to be built, based upon these parameters ( 13 , 26 – 28 , 32 , 33 ). These new models incorporated the coupling between pigments ( 32 ) and/or the atomic positions of the pigments themselves ( 13 , 26 , 33 ) as determined from the crystal structure.…”

“…As improvements in the resolution of crystal structures of PSII were published ( 3 , 36 , 37 ), it enabled new models to be built, based upon these parameters ( 13 , 26 – 28 , 32 , 33 ). These new models incorporated the coupling between pigments ( 32 ) and/or the atomic positions of the pigments themselves ( 13 , 26 , 33 ) as determined from the crystal structure. Among the latter, the size of the model became restrictive, and care had to be taken in choosing how much of the protein environment could be included and at what level of theory ( 26 – 28 ).…”

“…Among the latter, the size of the model became restrictive, and care had to be taken in choosing how much of the protein environment could be included and at what level of theory ( 26 – 28 ). Several authors opted for an idealized model to reduce computational cost ( 26 , 28 ); however, in one example, it was shown that inclusion of the phytol chains in single residue QM calculations emphasized the preference for the active branch CT pathway over the inactive ( 33 ). Additionally, there is some debate in the literature as to whether geometry optimization ( 13 , 28 , 33 ), or relaxation using molecular dynamics ( 27 , 34 , 38 ), of the crystal structure atomic positions is necessary to produce an accurate model of the PS II RC or whether the crystal structure geometries are sufficient ( 25 , 26 ).…”