Quantum Coherence in Photosynthesis for Efficient Solar Energy Conversion

Abstract

The crucial step in the conversion of solar to chemical energy in Photosynthesis takes place in the reaction center where the absorbed excitation energy is converted into a stable charge separated state by ultrafast electron transfer events. However, the fundamental mechanism responsible for the near unity quantum efficiency of this process is unknown. Here we elucidate the role of coherence in determining the efficiency of charge separation in the plant photosystem II reaction centre (PSII RC) by comprehensively combining experiment (two-dimensional electronic spectroscopy) and theory (Redfield theory). We reveal the presence of electronic coherence between excitons as well as between exciton and charge transfer states which we argue to be maintained by vibrational modes. Furthermore, we present evidence for the strong correlation between the degree of electronic coherence and efficient and ultrafast charge separation. We propose that this coherent mechanism will inspire the development of new energy technologies.

Fig. 1. The electronic structure of the PSII RC

(a) Exciton-CT states configuration and coherent energy levels scheme. The energy of the exciton-CT states in the different realizations of the disorder (central wavelength) is represented as vertical lines on top of the 80 K absorption spectrum. The spatial distribution of the exciton-CT states is shown on top of the X-ray crystal structure of the PSII RC (cofactor arrangement adapted from): stars and rectangles represent exciton and CT character, respectively. Line colour code: (P_D2P_D1)₊*_660nm (orange), (P_D2^δ+P_D1^δ−)₋*_673nm (red), (P_D2⁺P_D1⁻)^δ*_684nm (dark red) and (Chl_D1^δ+Phe_D1^δ−)*_681nm (blue). Cofactors colour code: P_D1 (red), P_D2 (blue), Chl_D1 (purple), Chl_D2 (cyan), Phe_D1 (orange) and Phe_D2 (green). The horizontal wiggled arrows represent the coherences between electronic states observed as cross-peaks in the 120, 340 and 730 cm⁻¹ 2D frequency maps (Fig. 1d,f-g and Table 1). (b) Absorption spectra and laser spectral profile. (c) Representative 2D spectral traces (left and central panel: cross peaks; bottom right panel: diagonal peaks) and their location in the real rephasing 2D spectrum at T = 200 fs (top right frame). Some traces have been translated vertically for better visualization. (d-f) 120, 440, 340 and 730 cm⁻¹ 2D frequency maps.

Fig. 2. Experimental vs. calculated real rephasing 2D PSII RC spectra at 80K

(a) Experimental PSII RC real rephasing 2D spectra at T = 20 fs, 100 fs, 1 ps and 50 ps. (b) Calculated PSII RC real rephasing 2D spectra at T = 20 fs, 100 fs, 1 ps and 50 ps. These 2D spectra are calculated with Standard Redfield theory according to the parameters of our disordered exciton-CT model (Supplementary Equations and Tables 2-4).

Fig. 3. Calculated dynamics of the site populations of the cofactors excited states and the primary CT state

(a) Dynamics averaged over all realizations of the disorder. (b-c) Dynamics corresponding to specific realizations of the disorder: (b) the P_D1 charge separation pathway dominates. (c) the Chl_D1 path dominates (note that the formation of the Chl_D1⁺Phe_D1⁻ is not shown in the figure). The legend is common for all panels. The scale is linear before the break and logarithmic after the break. The populations are created by the first two pulses with zero coherence time (τ = 0). The initial site populations of the cofactors correspond to their transition dipole strength, 16 D for Chl and 10 D for Phe.