Molecular Mechanics Simulations and Improved Tight-binding Hamiltonians for Artificial Light Harvesting Systems: Predicting Geometric Distributions, Disorder, and Spectroscopy of Chromophores in a Protein Environment

Abstract

We present molecular mechanics {and spectroscopic} calculations on prototype artificial light harvesting systems consisting of chromophores attached to a tobacco mosaic virus (TMV) protein scaffold. These systems have been synthesized and characterized spectroscopically, but information about the microscopic configurations and geometry of these TMV-templated chromophore assemblies is largely unknown. We use a Monte Carlo conformational search algorithm to determine the preferred positions and orientations of two chromophores, Coumarin 343 together with its linker, and Oregon Green 488, when these are attached at two different sites (104 and 123) on the TMV protein. The resulting geometric information shows that the extent of disorder and aggregation properties, and therefore the optical properties of the TMV-templated chromophore assembly, are highly dependent on the choice of chromophores and protein site to which they are bound. We used the results of the conformational search as geometric parameters together with an improved tight-binding Hamiltonian to simulate the linear absorption spectra and compare with experimental spectral measurements. The ideal dipole approximation to the Hamiltonian is not valid since the distance between chromophores can be very small. We found that using the geometries from the conformational search is necessary to reproduce the features of the experimental spectral peaks.