Dynamical Simulations of Carotenoid Photoexcited States Using Density Matrix Renormalization Group Techniques

We present a dynamical simulation scheme to model the highly correlated excited state dynamics of linear polyenes. We apply it to investigate the internal conversion processes of carotenoids following their photoexcitation. We use the extended Hubbard-Peierls model, $\hat{H}_{\textrm{UVP}}$, to describe the $\pi$-electronic system coupled to nuclear degrees of freedom supplemented by a Hamiltonian, $\hat{H}_{\epsilon}$, that explicitly breaks both the particle-hole and two-fold rotation symmetries of idealized carotenoids. The electronic degrees of freedom are treated quantum mechanically by solving the time-dependent Schr\"odinger equation using the adaptive time-dependent DMRG (tDMRG) method, while nuclear dynamics are treated via the Ehrenfest equations of motion. By defining adiabatic excited states as the eigenstates of the full Hamiltonian, $\hat{H}=\hat{H}_{\textrm{UVP}}+\hat{H}_{\epsilon}$, and diabatic excited states as eigenstates of $\hat{H}_{\textrm{UVP}}$, we present a computational framework to monitor the internal conversion process from the initial photoexcited state to the singlet triplet-pair states of carotenoids. We further incorporate Lanczos-DMRG to the tDMRG-Ehrenfest method to calculate transient absorption spectra from the evolving photoexcited state. We describe the accuracy and convergence criteria for DMRG, and show that this method accurately describes the dynamics of carotenoid excited states. We also discuss the effect of $\hat{H}_{\epsilon}$ on the internal conversion process, and show that its effect on the extent of internal conversion can be described by a Landau-Zener-type transition. This methodological paper is a companion to our more explanatory discussion of carotenoid excited state dynamics in, $\textit{Photoexcited state dynamics and singlet fission in carotenoids}$, D. Manawadu, T. N. Georges and W. Barford, $\textit{J. Phys. Chem. A}$ (2023).
Comment: 47 pages, 11 figures