Chromophore-Dependent Intramolecular Exciton–Vibrational Coupling in the FMO Complex: Quantification and Importance for Exciton Dynamics

Abstract

In this paper, we adopt an approach suitable for monitoring the time evolution of the intramolecular contribution to the spectral density of a set of identical chromophores embedded in their respective environments. We apply the proposed method to the Fenna-Matthews-Olson (FMO) complex, with the objective to quantify the differences among site-dependent spectral densities and the impact of such differences on the exciton dynamics of the system. Our approach takes advantage of the vertical gradient approximation to reduce the computational demands of the normal modes analysis. We show that the region of the spectral density that is believed to strongly influence the exciton dynamics changes significantly in the timescale of tens of nanoseconds. We then studied the impact of the intramolecular vibrations on the exciton dynamics by considering a model of FMO in a vibronic basis and neglecting the interaction with the environment to isolate the role of the intramolecular exciton-vibration coupling. In agreement with the assumptions in the literature, we demonstrate that high frequency modes at energy much larger than the excitonic energy splitting have negligible influence on exciton dynamics despite the large exciton-vibration coupling. We also find that the impact of including the site-dependent spectral densities on exciton dynamics is not very significant, indicating that it may be acceptable to apply the same spectral density on all sites. However, care needs to be taken for the description of the exciton-vibrational coupling in the low frequency part of intramolecular modes because exciton dynamics is more susceptible to low frequency modes despite their small Huang-Rhys factors.

Multiscale QM/MM molecular dynamics simulations of the trimeric major light-harvesting complex II

“…Furthermore, Coker and co-workers suggested an alternative formalism to accurately describe the intramolecular part of spectral densities 38 . Although, both approaches provide an impressive agreement between the theoretical and experimental spectral densities taking into account the size of the studied system, they suffer from high numerical costs required to construct the PES 39 or to determine the normal mode analysis while calculating the intra-molecular contribution at a high level of accuracy 40 . A reliable ground state QM/MM MD dynamics followed by excitation calculations using a high level QM method would be an ideal choice to accurately describe the dynamics of the energy gap for pigment molecules.…”

Time-dependent atomistic simulations of the CP29 light-harvesting complex

“…In spite of providing impressive agreements compared to experimental results, these aforementioned schemes are computationally still demanding. 51,53 A sophisticated quantum-mechanics/molecular mechanics (QM/MM) MD with an accurate description of the vibrational properties of the pigments would be an alternative way; however, semi-empirical schemes have a limited accuracy, 54 while DFT-based calculations are numerically expensive for pigments in LH systems. 55 To this end, we recently proposed a scheme using QM/MM MD dynamics employing the numerically efficient density functional-based tight binding (DFTB) approach 56 and have shown to be able to obtain a good agreement between the spectral densities obtained in such a manner and their experimental counterparts.…”

Investigating ultrafast two-pulse experiments on single DNQDI fluorophores: a stochastic quantum approach

“…Although the simulated spectrum (in vacuo) is rigidly red-shifted by 0.13 eV with respect to the experimental one 70 (in toluene), the general features are properly maintained: the distance between the two peaks is around 1370 cm À1 in both spectra, which is a typical value of vibrational modes in aromatic compounds. 71 Generally speaking, the lower-energy peak may be related to the |0 g i À |0 e i transition (from the vibrational ground state of the electronic ground state to the vibrational ground state of the electronic excited state) and the other one to a |0 g i À |1 e i transition (from the vibrational ground state of the electronic ground state to the first vibrational state of the electronic excited state). We also calculated the absorption energy of DNQDI in toluene within the PCM model, red-shifted to 0.24 eV with respect to the maximum absorption in the experimental spectrum.…”