Dans cette thèse nous avons eu pour but d’étudier la réactivité des complexes à bas degré d’oxydation du fer stabilisés par des ligands phosphines vis-à-vis des substrats insaturés. Cet objectif s’inscrit dans une démarche plus large d’accès à des systèmes catalytiques au fer capables de transformer de manière sélective l’éthylène, par la réaction d’oligomérisation, vers des alpha-oléfines linéaires courtes (butène-1, héxène-1, octène-1), ce qui n’a jamais été rapporté dans la littérature. Pour se faire, le passage par le mécanisme métallacyclique de transformation de l’éthylène est la voie privilégiée. Cependant, les propriétés électroniques et géométriques de ligand requises pour suivre un tel mécanisme, et en particulier son étape clé de couplage oxydant de deux molécules d’éthylène sur le centre métallique, ne sont pas clairement identifiées. Nous rapportons ici dans un premier temps la synthèse d’une bibliothèque de complexes de fer(II) et fer(III) à base de ligands phosphines présentant des propriétés électroniques et géométriques variées. Ces nouveaux complexes sont opportunément testés en tant que catalyseurs, d’une part pour la réaction d’oligomérisation de l’éthylène; mais également en tant qu’espèces réduites in situ en association avec des diènes (isoprène, butadiène) conduisant à leur polymérisation. Afin de nous rapprocher davantage de notre objectif d’accès aux espèces à bas degré d’oxydation du fer, l’emploi d’une voie de réduction en une étape et en conditions douces est rapportée; conduisant à l’isolement et la caractérisation de neuf complexes phosphorés de fer(0) à 18 et à 16 électrons. Bien qu’a priori inactifs vis-à-vis de l’éthylène ou des oléfines plus longues, nous montrons la capacité de ces complexes à promouvoir des réactions de couplage oxydant de substrats insaturés (alcynes) le couplage catalytique de l’éthylène et du butadiène et l’activation de divers composés tels que les silanes ou les halogénures d’alkyles, posant ainsi un premier jalon vers leur optimisation et application en tant que catalyseurs d’autres réactions, et potentiellement dans l’avenir, pour la transformation de l’éthylène. / In this thesis our aim was to study the reactivity of phosphine-based low-valent iron complexes towards unsaturated substrates. This goal is part of a wider approach of access to an iron catalytic system able to transform ethylene in a selective manner, by the reaction of oligomerization, towards short linear alpha-olefins (1-butene, 1-hexene, 1-octene), which is unreported in the literature. To achieve this, going through the metallacyclic mechanism of ethylene transformation is the most likely way. However, the electronic and geometrical features of the ligand required to follow this mechanism, and in particular the key step of the oxidative coupling of two molecules of ethylene to the metal center, are ill-defined. We thus report here in a first part the synthesis of a library of P-based iron(II) and iron(III) complexes bearing varied electronic and geometrical features. These novel complexes are opportunistically screened as catalysts for ethylene oligomerization on one hand, and as in situ reduced species in association with dienes (isoprene, butadiene) leading to the polymerization of the latter substrates, on the other hand. In order to get closer to our goal of access to low-valent iron complexes, the use of a one-pot reduction methodology under mild conditions is reported, leading to the isolation and characterization of nine 18- and 16-electron iron(0) complexes. Even though they do not show reactivity towards ethylene or longer olefins, we demonstrate the ability of these complexes to promote reactions involving the oxidative coupling elementary step of unsaturated substrates (alkynes), the catalytic coupling of ethylene and butadiene and the activation of various compounds such as silanes and organic halides, representing a first milestone towards their optimization and application as catalysts for other reactions, including potentially in the near future, for ethylene transformation

<p>The increase in shale gas exploitation has motivated the studies towards new processes for converting light alkanes into higher valuable chemicals, including fuels. The works in this dissertation focuses on two processes: propane dehydrogenation and ethylene oligomerization. The former involves the conversion of propane into propylene and hydrogen, while the latter converts light alkenes into higher molecular weight products, such as butylene and hexene. </p> <p>The thesis project focuses on understanding the effect of geometric effects of Pt alloy catalysts for propane dehydrogenation and the methodologies for their characterization. Pt-Co bimetallic catalysts were synthesized with increasing Co loadings, characterized and evaluated for its propane dehydrogenation performance. In-situ synchrotron X-Ray Powder Diffraction (XRD) and X-Ray Absorption (XAS) were used to identify and differentiate between the intermetallic compound phases in the nanoparticle surface and core. Difference spectra between oxidized and reduced catalysts suggested that, despite the increase in Co loading, the catalytic surface remained the same, Pt₃Co in a Au₃Cu structure, while the core became richer in Co, changing from a monometallic Pt fcc core at the lowest Co loading to a PtCo phase in a AuCu structure at the highest loading. Co^II single sites were also observed on the surface, due to non-reduced Co species. The catalytic performance towards propane dehydrogenation reinforced this structure, as propylene selectivity was around 96% for all catalysts, albeit the difference in composition. The Turnover Rate (TOR) of these catalysts was also similar to that of monometallic Pt catalysts, around 0.9 s^-1, suggesting Pt was the active site, while Co atoms behaved as non-active, despite both atoms being active in their monometallic counterparts.</p> <p>In the second project, a single site Co^II catalyst supported on SiO₂ was evaluated for ethylene oligomerization activity. The catalyst was synthesized, evaluated for propane dehydrogenation, propylene hydrogenation and ethylene oligomerization activities and characterized <i>in-situ</i> by XAS and EXAFS and H₂/D₂ exchange experiments. The catalysts have shown negligible conversion at 250^oC for ethylene oligomerization, while a benchmark Ni/SiO₂ catalyst had about 20% conversion and TOR of 2.3x10^-1 s^-1. However, as the temperature increased to above 300^oC, ethylene conversion increased significantly, reaching about 98% above 425^oC. <i>In-situ</i> XANES and EXAFS characterization suggested that H₂ uptake under pure H₂ increased in about two-fold from 200^oC to 500^oC, due to the loss of coordination of Co-O bonds and formation of Co-H bonds. This was further confirmed by H₂/D₂ experiments with a two-fold increase in HD formation per mole of Co. <i>In-situ</i> XAS characterization was also performed with pure C­₂H₄ at 200^oC showed a similar trend in Co-O bond loss, suggesting the formation of Co-alkyl, similarly to that of Co-H. The <i>in-situ</i> XANES spectra showed that the oxidation state remained stable as a Co²⁺ despite the change in the coordination environment, suggesting that the reactions occurs through a non-redox mechanism. These combined results allowed the proposition of a reaction pathway for dehydrogenation and oligomerization reactions, which undergo a similar reaction intermediate, a Metal-alkyl or Metal-Hydride intermediates, activating C-H bonds at high temperatures.</p>

There is an increasing trend towards advancing the understanding and development of ethylene oligomerization catalysts, both in academia and industry. The metal of choice in this chemistry is invariably chromium, which has shown great versatility in selective trimerization/tetramerization, non-selective oligomerization and polymerization of ethylene. While much success has been achieved in ethylene trimerization, the same con not be said about tetramerization catalysis. Aminophosphine based ligands have demonstrated their ability towards selective 1-octene production, however, the popular PNP catalyst is able to achieve only 70% selectivity. In order to explore the possibility of developing and enhancing the selectivity of chromium based ethylene tetramerization catalyst, this thesis work was undertaken. The ligand systems we chose for our work were bidentate aminophosphine based (PN(CH2)nNP), which has yielded interesting selective oligomerization. Subtle modifications were found to result in drastic changes in selectivity, from tetramerization (PN(CH2)3NP) to trimerization (PN(CH2)2NP). We managed to successfully develop the first truly selective (over 90%) 1-octene catalyst with polymer-free behavior. Further modifications on the ligand framework, where one atom of Si was used to link the two NP units, resulted in non-selective oligomerization, in which case we determined that the oxidation-state of chromium is a key player. We explored other modifications on our selective ligands in which one of the arms on the bidentate ligand was replaced with a base-donor amine, phosphine or pyridine, and resulted in interesting selectivity changes. The final modification that we tested was a novel N(CH2)2P ligand and found it to be a highly active, non-selective oligomerization catalyst.

There is an increasing trend towards advancing the understanding and development of ethylene oligomerization catalysts, both in academia and industry. The metal of choice in this chemistry is invariably chromium, which has shown great versatility in selective trimerization/tetramerization, non-selective oligomerization and polymerization of ethylene. While much success has been achieved in ethylene trimerization, the same con not be said about tetramerization catalysis. Aminophosphine based ligands have demonstrated their ability towards selective 1-octene production, however, the popular PNP catalyst is able to achieve only 70% selectivity. In order to explore the possibility of developing and enhancing the selectivity of chromium based ethylene tetramerization catalyst, this thesis work was undertaken. The ligand systems we chose for our work were bidentate aminophosphine based (PN(CH2)nNP), which has yielded interesting selective oligomerization. Subtle modifications were found to result in drastic changes in selectivity, from tetramerization (PN(CH2)3NP) to trimerization (PN(CH2)2NP). We managed to successfully develop the first truly selective (over 90%) 1-octene catalyst with polymer-free behavior. Further modifications on the ligand framework, where one atom of Si was used to link the two NP units, resulted in non-selective oligomerization, in which case we determined that the oxidation-state of chromium is a key player. We explored other modifications on our selective ligands in which one of the arms on the bidentate ligand was replaced with a base-donor amine, phosphine or pyridine, and resulted in interesting selectivity changes. The final modification that we tested was a novel N(CH2)2P ligand and found it to be a highly active, non-selective oligomerization catalyst.

L’objectif de ce travail fut la synthèse de ligands fonctionnels de type N,N'-diphosphanyl-NHC (NHC = carbènes N-hétérocycliques) et l’étude de leur chimie de coordination. La synthèse du nouveau ligand tridentate, stable et rigide, N,N'-diphosphanyl-imidazol-2-ylidene a permis des études expérimentales et théoriques et l’accès à des complexes mono-, di-, tri-, penta-, et hexanucléaires des métaux du groupe 11 (Cu, Ag et Au) originaux et aux propriétés structurales uniques. Les complexes mono- et dinucléaires avec un ou deux atomes de phosphore libres ont permis d’accéder à des complexes hétérotrinucléaires à interactions d10-d10 qui sont luminescents. La transmétallation partielle ou totale des complexes homotrinucléaires de Cu ou d’Ag avec des réactifs contenant du Pd(0) ont conduit à des complexes hétérotrinucléaires à interactions d10-d10. En plus de son comportement pontant, ce ligand peut se agir en chélate dans des complexes du palladium et du chrome. Dans le cas du Cr(III), ils montrent une activité catalytique en oligomérisation de l’éthylène supérieure à celle des complexes du Cr(II) et conduisent principalement à des oligomères. / The purpose of this work was the synthesis of N,N'-diphosphanyl-functionalized NHC ligands andtheir coordination chemistry. The novel stable and rigid tridentate N,N'-diphosphanyl-imidazol-2-ylidene was synthesized and experimental and computational information on its stability weregained. It served as a unique platform for the synthesis of novel mono-, di-, tri-, penta-, hexanuclear complexes with the coinage metals (Cu, Ag and Au), exhibiting rare structural features. The mono- and dinuclear complexes with one or two dangling P-donors provided rational access to heterotrinuclear complexes. All these coinage metal complexes have short metal-metalseparations, indicating the presence of d10-d10 interactions, and display excellent luminescentproperties. Partial or complete transmetallation of the homotrinuclear Cu or Ag complexes withPd(0) precursors led to hetero-trinuclear complexes with d10-d10 interactions. In addition to itsbridging behavior, this ligand also showed its chelating behavior in Pd or Cr(III) complexes. Thelatter displayed superior performance in ethylene oligomerization than the Cr(II) complexes andgave mostly oligomers.

There is an increasing trend towards advancing the understanding and development of ethylene oligomerization catalysts, both in academia and industry. The metal of choice in this chemistry is invariably chromium, which has shown great versatility in selective trimerization/tetramerization, non-selective oligomerization and polymerization of ethylene. While much success has been achieved in ethylene trimerization, the same con not be said about tetramerization catalysis. Aminophosphine based ligands have demonstrated their ability towards selective 1-octene production, however, the popular PNP catalyst is able to achieve only 70% selectivity. In order to explore the possibility of developing and enhancing the selectivity of chromium based ethylene tetramerization catalyst, this thesis work was undertaken. The ligand systems we chose for our work were bidentate aminophosphine based (PN(CH2)nNP), which has yielded interesting selective oligomerization. Subtle modifications were found to result in drastic changes in selectivity, from tetramerization (PN(CH2)3NP) to trimerization (PN(CH2)2NP). We managed to successfully develop the first truly selective (over 90%) 1-octene catalyst with polymer-free behavior. Further modifications on the ligand framework, where one atom of Si was used to link the two NP units, resulted in non-selective oligomerization, in which case we determined that the oxidation-state of chromium is a key player. We explored other modifications on our selective ligands in which one of the arms on the bidentate ligand was replaced with a base-donor amine, phosphine or pyridine, and resulted in interesting selectivity changes. The final modification that we tested was a novel N(CH2)2P ligand and found it to be a highly active, non-selective oligomerization catalyst.