Supervisor: Prof. Persico Maurizio

Title: Development of a new method for molecular dynamics simulations for photochemical and photophysical processes.

Abstract: The aim of my research work is the development of a new approach in the field of excited state dynamics methods that will be called “Parameterized Non-Adiabatic Dynamics” (PNAD). The PNAD strategy is based on the observation that the essential information about the excited state dynamics of a given chromophore is partially transferable among different environments. The procedure consists of three steps. First, we run a ”model dynamics” with the usual nonadiabatic trajectory methods such as TSH, in conditions that allow to reduce the computational burden by at least two orders of magnitude (one chromophore, normally in vacuo, and few trajectories, stopped shortly after reaching the ground state). Second, we extract from the results of the model dynamics the electronic energy differences, the nonadiabatic transition probabilities and possibly other dynamical quantities, and we fit them as functions of the most important nuclear space phase variables (internal coordinates, kinetic energies, etc). Third, we use a ground state force--field and the information obtained in the second step to compute also the excited state energies and the transition rates in order to run the PNAD simulations. The latter will have a computational cost close to standard ground state Molecular Dynamics, so it will be possible to treat more complex problems (many interacting chromophores, long simulation times covering (hot) ground state processes, many trajectories if needed to obtain statistically significant results, etc). As a test case, we choose the photoisomerization of trans-azobenzene excited in the n->pi* band, because of our previous results with classical molecular dynamics methods that can be good benchmark to check PNAD results.