Relation between structure and functionality in photosynthetic antenna complex of green sulfur bacteria: efficiency under natural sunlight pumping

TL;DR

Large-scale simulations of light-matter interaction in natural photosynthetic antenna complexes of the Chlorobium Tepidum green sulfur bacteria containing more than one hundred thousand chlorophyll molecules have been performed, showing the high sensitivity of light harvesting to the specific orientation of the dipole moments.

Abstract

Large-scale simulations of light-matter interaction in natural photosynthetic antenna complexes of the Chlorobium Tepidum green sulfur bacteria (GSB) containing more than one hundred thousand chlorophyll molecules, comparable with natural size, have been performed. Here we have modeled the entire process of the exciton energy transfer, from sunlight absorption to exciton trapping in the reaction centers (RCs) in presence of a thermal bath. The energy transfer has been analyzed using the radiative non-Hermitian Hamiltonian and solving the rate equations for the populations. Sunlight pumping has been modeled as black-body radiation with an attenuation factor that takes the Sun-Earth distance into account. Cylindrical structures typical of GSB antenna complexes, and the dimeric baseplate comparable to natural size have been considered. Our analysis shows that under natural sunlight, in photosynthetic antennae of GSB the number of excitations reaching the RC per unit time matches the RC closure rate and the internal efficiency shows values close to 80%. We also considered cylindrical structures where the orientation of the dipoles does not reflect the natural one. Specifically, we vary continuously the angle of the transition dipole with respect to the cylinder main axis, focusing on the case where all dipoles are parallel to the cylinder axis. We also consider the important case where the dipoles are randomly oriented. In all cases the light-harvesting efficiency is lower than in the natural structure, showing the high sensitivity of light harvesting to the specific orientation of the dipole moments. Our results allow for a better understanding of the relationship between structure and functionality in photosynthetic antennae of GSB and could drive the design of efficient light-harvesting devices.

Figures (17)

• Figure 1: Architecture of GSB light-harvesting unit under natural sunlight. The chlorosome of GSB comprising three adjacent concentric cylinders and the dimeric baseplate (BPL) have been represented. Each aggregate is made of four concentric MT model cylinders with radii of $30$, $51$, $72$ and $93 \ \hbox{\AA}$ and containing, respectively, $30$, $51$, $72$ and $93$ dipoles per ring. The entire chlorosome contains $132840$ BChl c with a length $L=1485.7 \ \hbox{\AA}$ and the distance between two adjacent concentric cylinders is $d_1 = 30 \ \hbox{\AA}$. Under the chlorosome a dimeric baseplate with dimesnsions $1147.5 \times 3075.3 \ \hbox{\AA}^2$ and comprising $3350$ BChl a has been represented. The distance between the cylinder and the baseplate is set to $d_2=20 \ \hbox{\AA}$, according to Refs. guziklinnanto. Finally the energy transfer from the baseplate to the RCs is mediated by the $k_{FMO}$ rate, represented by red arrows connecting the baseplate to the RCs.
• Figure 2: Section of the different cylindrical models. Panel (A) shows the MT model, the structure derived by genetic modification of the natural wild-type model with a radius $R=60 \ \hbox{\AA}$. In panels (B-C) PD and RD cylinders made of a stack of rings with the same radius $R=60 \ \hbox{\AA}$ are presented. For the sake of clarity we show only 30 dipoles per ring instead of 60 as we consider in this paper. For more details about the geometry see Refs. macroscopicvalzelli2024large.
• Figure 3: Representation of the dimeric baseplate (BPL) in GSB light-harvesting unit. Panel (A): top view of a portion of the dimeric baseplate, a planar structure in the xz-plane containing BChl a molecules. The blue and red arrows represent the TDM associated to each BChl a and belonging to the bottom and top layer respectively. Panel (B): zoomed-in view of the dimeric unit formed by a red arrow ($\mu_{t}$) and a blue one ($\mu_{b}$). Dipoles orientations have been found in Ref. guzik and the corresponding unit vectors are given by $\hat{\mu}_{t}=(0.2795,0.7484,0.5982)$ and $\hat{\mu}_{b} =(0.2533,0.1607,-0.9533)$. The distance between two dipoles in the same dimeric unit cell is set to $12.8 \ \hbox{\AA}$ , while the distance along consecutive BChl a on x and z axis are $45.9 \ \hbox{\AA}$ and $30.1 \ \hbox{\AA}$ respectively. For sake of clarity in panel B the dipole length is multiplied by a factor of $16$.
• Figure 4: Dipole strength $|D_n|^2$ (DH and HH models) and radiative decay rate $\Gamma_n/\gamma$ (NHH model) in cylindrical aggregates and baseplate. Comparison between the dipole strength (DH and HH models) and radiative decay rate (NHH model) for (A) a single MT cylinder, (B) the chlorosome and (C) the dimeric baseplate. Note that in Ref. valzelli2024large for a single MT cylinder (A) the ratio $\Gamma_{n}/\delta$ is always less than $1$, while for the chlorosome (B) the maximum value of $(\Gamma_{n}/\delta)_{\text{max}}$ is almost $10^2$, proving that the perturbative regime for this aggregate fails and only the NHH model can be used to describe superradiance. The mean level spacing $\delta$ has been computed as the ratio between the energy spectral width and the total number of eigenmodes for each complex. For the geometry of the system, refer to Tab. \ref{['table_size']}. Panels (A-B) show only the lowest part of the energy spectrum, while panel C represents the entire energy spectrum.
• Figure 5: Rate equation scheme. Scheme of the rate equations, Eqs. \ref{['ILE']}. Here, $\Gamma_n/\hbar$ is the radiative recombination rate, $\kappa_{NR}$ is the non-radiative recombination rate (equal for all levels), $R_n=f_Sn_S(\omega_n)\frac{\Gamma_n}{\hbar}$ are the sunlight absorption and stimulated emission rates, $T_{m,n}$ are the thermalization rates within each aggregate, see Eq. \ref{['th-rates']}, $K_{m,n}$ are the transfer rates between eigenstates of each aggregates, see Eq. \ref{['forster']}, and $\kappa$ is the trapping rate from the baseplate, through the FMOs, to the RC.