Search

The generalized quantum master equation (GQME) provides a general and formally exact framework for simulating the reduced dynamics of open quantum systems. The recently introduced modified approach to the GQME (M-GQME) corresponds to a specific implementation of the GQME that is geared toward simulating the dynamics of the electronic reduced density matrix in systems governed by an excitonic Hamiltonian. Such a Hamiltonian, which is often used for describing energy and charge transfer dynamics in complex molecular systems, is given in terms of diabatic electronic states that are coupled to each other and correspond to different nuclear Hamiltonians. Within the M-GQME approach, the effect of the nuclear degrees of freedom on the time evolution of the electronic density matrix is fully captured by a memory kernel superoperator, which can be obtained from short-lived (compared to the time scale of energy/charge transfer) projection-free inputs. In this paper, we test the ability of the M-GQME to predict the energy transfer dynamics within a seven-state benchmark model of the Fenna–Matthews–Olson (FMO) complex, with the short-lived projection-free inputs obtained via the Ehrenfest method. The M-GQME with Ehrenfest-based inputs is shown to yield accurate results across a wide parameter range. It is also found to dramatically outperform the direct application of the Ehrenfest method and to provide better-behaved convergence with respect to memory time in comparison to an alternative implementation of the GQME approach previously applied to the same FMO model. [ABSTRACT FROM AUTHOR]

This work explores the possibility of simulating an electron transfer process between a donor and an acceptor in real time using time-dependent density functional theory electron dynamics. To achieve this objective, a central issue to resolve is the definition of the initial state. This must be a non-equilibrium electronic state able to trigger the charge transfer dynamics; here, two schemes are proposed to prepare such states. One is based on the combination of the density matrices of the donor and acceptor converged separately with appropriate charges (for example, −1 for the donor and +1 for the acceptor). The second approach relied on constrained DFT to localize the charge on each fragment. With these schemes, electron transfer processes are simulated in different model systems of increasing complexity: an atomic hydrogen dimer, a polyacetylene chain, and the active site of the T. cruzi hybrid type A heme peroxidase, for which two possible electron transfer paths have been postulated. For the latter system, the present methodology applied in a hybrid Quantum Mechanics - Molecular Mechanics framework allows us to establish the relative probabilities of each path and provides insight into the inhibition of the electron transfer provoked by the substitution of tryptophan by phenylalanine in the W233F mutant. [ABSTRACT FROM AUTHOR]