Défense de thèse de doctorat en sciences chimiques : "proline-catalyzed aldol reactions"
Investigation of proline-catalyzed aldol reactions using a reactive force field
Date : 26/10/2016 15:00 - 26/10/2016 17:00
Lieu : Amphithéâtre CH01, rue Grafé, 5000 Namur
Orateur(s) : Pierre HUBIN
Organisateur(s) : Daniel VERCAUTEREN
Jury
Jérémy HARVEY (KULeuven), Jérôme HENIN (Institut de biologie physico-chimique, Paris), Denis JACQUEMIN, promoteur (Université de Nantes), Steve LANNERS (UNamur), Benoît CHAMPAGNE, président (UNamur), Daniel VERCAUTEREN, promoteur (UNamur)
Résumé
Organocatalysis is defined as the use of metal-free molecules with relatively low molecular weight to catalyze organic reactions. In the last two decades the field has blossomed considerably since it was rationalized that chiral organocatalysts like proline can be used to efficiently drive asymmetric synthesis. Such catalysts present several practical advantages that explain their popularity. They are often relatively cheap (or derived from cheap compounds), not sensitive to moisture, and most of the time non-toxic. The mechanism of organocatalyzed reactions has been a fertile subject of discussions in the literature.
In the present work, theoretical chemistry tools are used to study the mechanisms of proline-catalyzed aldol reactions, one of the pivotal example of asymmetric organocatalysis. Molecular Dynamics (MD) simulations performed with the ReaxFF reactive force field allow to model several key steps of the reaction taking into account an explicit description of its chemical environment, e.g., solvent molecules. The MD simulations are combined with free energy methods, notably the adaptive biasing force approach, in order to trigger the chemical reactions to be observed and to obtain free energy profiles of the reactions. ReaxFF is a force field based approach that allows to handle breaking and formation of chemical bonds. It relies on a dynamic atomic connectivity derived from bond orders computed from interatomic distances. Relying on ReaxFF requires specific parameters for the system under study. As the available parameters were not accurate enough for the reactions considered; a parameterization scheme based on a Monte Carlo simulated annealing procedure was built during the thesis. The optimization is based on density functional theory calculations with the M06-2X functional which turned out to provide accurate results for the considered reactions according to benchmarks performed with highly reliable quantum mechanics approaches.
La défense est publique
Télecharger :
vCal