Centre de recherche sur les Ions, les MAtériaux et la Photonique (UMR 6252)

The activity of SIMUL has moved progressively toward physico-chemistry with the introduction of a new theme at the frontier of water radiolysis and our work on solutes in inert matrices. A complete simulation reaction path between two reactants in a solvent is probably a dream for a physico-chemist. We are moving one step ahead in this direction with the study of charge transfer to solvent states (CTTS) for anions in water. We focus on hydroxide anion embedded in water clusters. It is a generic system for closed shell molecular excitation in a polar solvent and several experiments dedicated to the relaxation of the CTTS by means of time-resolved spectroscopy experiments are now available. Moreover the recombination reaction of the electron with its parent OH is one of the most important reactions for radical recombination in liquid water radiolysis.
This work was initiated as a collaborative project (NANOBIODOS) between the IPN Lyon, the LOA in Palaiseau and the LPC in Orsay and founded by the physicancer action. The aim of the project was to investigate the dynamics of electron solvation in liquid water, in conjunction with experiment. Such a solvation process is present in CTTS formation, where it competes with geminate recombination.
The choice of the hydroxide anion is also motivated by our development of an accurate potential for water hydroxide and water hydronium interaction. This work was founded by the ANR grant PIBALE dedicated to the investigation of small water clusters under irradiation. It provides a remarkable description of the twin surface of the degenerated ground state of the OH-H₂O molecular complex, with good transferability to larger systems.
We investigated first the structure of small OH-(H₂O)n=1-7 clusters in their ground and first excited states. We obtain thus the knowledge of geometries, absorption spectrum and PES minima necessary to build a synthetic model of electronic structure designed to perform molecular dynamics simulation (see scientific achievements 1). The development of the model is the main task of the C. Dubosq thesis. Unfortunately, existing model potentials for water are not transferable and give poor results for hydroxide. We are thus working at the parameterization of the solute solvent interaction. Our preliminary results using a Hartree-Fock density of water to represent the solute-water interaction shows that such a parameterization can be achieved with enough accuracy to model hydroxide anion, and that such a model should be transferable to other solutes. Though cumbersome, such a model allows us to obtain analytical derivatives and thus to perform molecular dynamics simulations of a solute in its ground or excited state.