Etude théorique de gros systèmes : analyse de liaison et modélisation

L'objectif de cette thèse était de travailler sur les groupements phosphines afin de modéliser ces groupements par des pseudopotentiels de groupe (EGP). Après un court exposé des principales théories et approximations de la chimie quantique, nous exposons la méthode des pseudopotentiels de g...

Full description

Saved in:
Bibliographic Details
Main Author: Bessac, Fabienne
Contributors: Frenking, Gernot (Prof. Dr.) (Thesis advisor)
Format: Dissertation
Published: Philipps-Universität Marburg 2004
Online Access:PDF Full Text
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents: The aim of this Phd work was to get an insight into phosphine groups in order to model them by Effective Group Potentials (EGP). After a concise report of the main approximations and theories of quantum chemistry, the design of a pseudopotential for the carbonyl group is carried out. This particular case helps to explain the EGP method. Following this a preliminary overview of the hybrid methods is shown. Subsequently the possibility of using pseudopotentials to treat the frontier between QM and MM regions working with hybrid methods is investigated. The ONIOM method and the pseudopotential replacing SiH3 are chosen to carry out the first tests. Next there is an introduction for energy partition analysis. The usefulness of this kind of method in attempting to find out why the trihalide boron compound BF3 is a weaker Lewis acid than BCl3 with respect to strong Lewis bases as NH3 is illustrated. Afterwards the results using energy decompositions on various compounds containing phosphorus is detailed. The amine and phosphine groups in different donor-acceptor complexes (metallic complexes or complexes containing boron and aluminum) are detailed. The final chapter deals with the molecules in which the elements P, As, Sb and Bi of group 15 are bonded to chromium metallic fragments in an unusual way. The conclusion details the original results of this work before revealing possible developments.