Tackling Electronic Excited States in Complex Surroundings.


Theory-Modeling-Simulation, Université de Lorraine, France


Most of time hybrid Quantum Mechanics/Molecular Mechanics (QM/MM) calculations are carried out with the help of non-polarizable force field. This is generally a well-accepted approximation if one recall that force field point charges are often defined to reproduce condensed phase properties. Thus, one can say that the MM point charges are implicitly already polarized in average. This is true and accurate only if the quantum part does not vary too much during the studied chemical process. Electronic excited states can have very different nature than the electronic ground state. One can think for example of excited states with a marked charge transfer character. In these situations, the approximation defined above does not hold any longer. Hence, to correctly treat excited states with any QM/MM method, the polarization of the surrounding is mandatory.

In this lecture, I will present a recent development that can be considered as an “universal” force field that is shortly described in the following. For the UV/Visible absorption of molecules in the gas phase, the Franck-Condon principle is very often invoked. It states that, the electronic reorganization during an electronic transition being so fast compared to the nuclear motion, the molecular geometry can be considered fixed. In solution, the geometry of the chromophore is still considered unchanged during the absorption process, but the internal geometry of solvent molecule and their relative orientations also. A contrario, the electrons of the solvent molecules can react instantaneously to the modification of the electronic cloud of the chromophore. This is called the electronic response of the surrounding (ERS). A solvent is generally an isotropic media, in average, that can be modeled by a polarizable continuum characterized by the relative dielectric constant with separated electronic and nuclear contributions. Hence, extracting the electronic contribution is trivial in so called self-consistent reaction field approaches (SCRF). Oppositely, macromolecules need an atomic description using QM/MM methods. Instead of using polarizable force fields, we propose a simple method which combines hybrid QM/MM techniques with SCRF approaches to evaluate the ERS in macromolecules. This method will be detailed and applied to the interpretation of the photophysics of some biological systems. Special attention will be devoted to the light switch effect and to DNA photosensibilization.