The last decade has been marked by a large increase of demand for green chemistry processes. Consequently, chemists have focused their efforts on the development of more direct routes toward different classes of targets. In that regard catalysis has played a crucial role at enabling key bond formations that were otherwise inaccessible or very energy and resources consuming. The central theme of this body of work concerns the formation of C–N bonds, either through transition metal catalysis or organocatalysis. These structural units being highly recurrent in biologically active molecules, the establishment of more efficient routes for their construction is indispensable. The first part of this thesis describes a new method for the synthesis of isoquinolines from the oxidative coupling/annulation of alkynes with N-tert-butyl benzaldimines via Rh(III) catalysis (Chapter 2). Preliminary mechanistic investigations of this system pointed to the involvement of Rh(III) in the C–H bond cleavage step as well as in the C–N bond reductive elimination that provides the desired heterocycle. Following this oxidative process, a Rh(III)-catalyzed redox-neutral approach to isoquinolones from the reaction of benzhydroxamic acids with alkynes is presented (Chapter 3). The discovery that an N–O bond contained in the substrate can act as an internal oxidant was found to be very enabling. Indeed, it allowed for milder reaction conditions, broader scope (terminal alkyne and alkene compatible) and low catalyst loadings (0.5 mol%). Mechanistic investigations on this system were also conducted to identify the nature of the C–N bond formation/N–O bond cleavage as well as the rate-determining step. The second part of this work presents mechanistic investigations performed on a recently developed intermolecular hydroamination reaction catalyzed through tethering organocatalysis (Chapter 4). This transformation operates via the reversible covalent attachment of two reactants, a hydroxylamine and an allylamine, to an aldehyde catalyst by the formation of a mixed aminal. This allows a difficult intermolecular Cope-type hydroamination to be performed intramolecularly. The main kinetic parameters associated with this reaction were determined and they allowed the generation of a more accurate catalytic cycle for this transformation. Attempts at developing new families of organocatalysts are also discussed.

A series of cationic Rh(I) diphosphine and phosphine-alkene complexes have been isolated and fully characterised. The reactivity of these species towards hydrogenation, dehydrogenation and dehydrocoupling reactions has been investigated. The use of potentially hemilabile ligands DPEphos and XANTphos in the intramolecular dehydrogenation chemistry of tricyclopentylphosphine is reported. The comparison in reactivity of these isolated diphosphine phosphine-alkene complexes towards hydrogenation and with acetonitrile is discussed along with their ability to dehydrocouple secondary silane, Ph₂SiH₂, and amine-borane H₃B·NMe₂H. The acceptorless dehydrogenation of a tethered cyclopentane with cationic Rh(I) diphosphine complexes has also been extended to include thioethers. Isolated cationic Rh(I) phosphine-alkene complexes with labile fluorobenzene ligands are found to act as a source of the reactive 12-electron [Rh{PR₂(ƞ²-C₅H₇)}]+ (R = cyclopentyl (Cyp)/ iPr) fragment in solution and can coordinate two amine-borane ligands (either H₃B·NMe₃, H₃B·NMe₂H or H₃B·NMeH₂) in a novel and unique bis-σ-binding mode. The catalytic activity of some of these isolated complexes in the dehydrocoupling of H₃B·NMe₂H and H₃B·NMeH₂ has been determined. With a view to further understanding the mechanism of catalytic transition metal assisted amine-borane dehydrogenation and dehydrocoupling, known B-N intermediates H₃B·NMe₂BH₂·NMe₂H and [H₂B·NMeH]₃ were also coordinated to the [Rh{PCyp₂(ƞ²-C₅H₇)}]+ fragment and investigated with regard to their role in the catalytic cycle. Structure activity relationships determined from stoichiometric reactions of cationic Rh(I) diphosphine fluorobenzene complexes with amine-boranes enabled the design of a highly efficient homogeneous catalyst capable of dehydrogenating H₃B·NMe₂H to [H₂BNMe₂]₂ at 0.2 mol% loading in 30 minutes at 298 K. Rapid dehydrogenation and dehydrocoupling of H₃B·NMeH₂ to form high molecular weight poly(N-methylaminoborane) with a low PDI has also been achieved. Investigations using model aminoborane H₂B=NiPr₂ and intermediate B-N species H₃B·NMe₂BH₂·NMe₂H and [H₂B·NMeH]₃ has helped establish an overall mechanistic rationale for this process.

La catalyse homogène à l’or a longtemps été sous-estimée. Cela a changé au début du 21ème lorsque la communauté des chimistes a reconsidéré les nombreuses singularités que ce métal peut apporter. De nos jours, plus d’une centaine de groupe de recherche mondialement reconnus a permis d’élargir les domaines d’applications de ce précieux métal. De nombreuses avancées ont été réalisées, mais à ce jour très peu sont utilisées à grande échelle et la fréquence de turnover est souvent faible. Ces problèmes sont souvent liés à la rapide dégradation des espèces actives d’or cationiques. Ce manuscrit expose les avancées qui ont été proposée jusque là et les alternatives que nous avons offert. Nous avons ainsi décrit une méthode de cationisation lente et réversible de l’espèce d’or cationique en substituant les sels d’argent traditionnellement utilisés au profit d’autres acides de Lewis. La découvert de cette lente métathèse d’anion a permis : de retarder la décomposition de l’espèce active dans le milieu, de diminuer la charge catalytique d’or et de monter en charge les réactions à l’échelle du gramme. Cette méthode a ensuite été appliquée avec succès à des réactions énantiosélectives d’hydroalkylation d’alcènes non activés, mais également à permis de considérer l’indium et le bismuth comme une alternative aux métaux de transition en π-catalyse. / Homogeneous gold catalysis has been underestimated for nearly a century. After reconsideration the by chemists’ community, it rapidly became a hot topic in chemistry. To date, more than a 100 worldwide groups have embraced the golden opportunity. Considerable breakthroughs have been made, nevertheless, the classical experimental procedures still suffer from low turnover numbers (TON) and scalability. These issues can be attributed to the decay of cationic gold catalysts and several solutions have been proposed in the literature. In this thesis, we will present our way to circumvent this thorny issue. Specifically, we have used Lewis acids other than silver salts for that purpose and found that some of them could indeed generate an active gold species through slow anion metathesis. The slow anion metathesis could retard the decomposition of the active gold species. Thus, gold could be used in low amounts in scalable reactions by simply avoiding the use of silver. These silver-free two-component mixtures could address the difficult issue of C-C bond forming reaction. Besides, we have developed catalytic enantioselective hydroalkylations of unactivated alkenes. This study led us to consider indium and bismuth catalysis in addition to gold.

Transition-metal-catalyzed cross-coupling reactions are reliable tools for forging C–C bonds. The Engle Lab has previously pioneered the intermolecular difunctionalization of unactivated alkenes facilitated by nickel catalysis, where regioselectivity is controlled through the use of a bidentate directing group. A limitation of existing methods is that allyl groups have not yet been successfully incorporated, as the electrophile scope has been limited to alkyl and aryl species. Fundamentally, C–C p-bonds have served as key building blocks for the assembly of complex molecules, and the ability to introduce allyl moieties in a controlled manner enables diverse, downstream functionalization in multi-step synthesis. This work focuses on the use of diverse azaheterocycle directing groups connected to non-conjugated alkenes. Through the use of nickel catalysis, we have been able to successfully introduce and preserve allyl and cinnamyl species at the g-position and alkyl zinc nucleophiles at the b-position with high yield under mild conditions. This novel, 1,2-allylalkylation can accommodate a vast array of substituents with different electronic and steric properties (>20 examples). Our efforts have shifted to exploring different monodentate directing groups and to conduct mechanistic studies to shed light on the catalytic cycle. Interestingly, electron-rich electrophiles provide nearly quantitative NMR and isolated yields, whereas electron-poor electrophiles lead to lower yields. We report a competition experiment to further elucidate this mechanism. While isolated yields were generally higher for electron-rich groups, a competition between p-OMe and p-CF3 electrophiles led to preferential incorporation of the trifluoromethyl-substituted coupling partner, which supports oxidative addition as the product-determining step.

It was recently shown that V-doped acid-prepared mesoporous silica (APMS) nanoparticles are active catalysts for the oxidation of the mustard gas analogue 2-chloroethyl ethyl sulfide (CEES) under ambient conditions in the presence of aldehydes, using O2 from air as the oxidation source. However, the vanadium ion leached from the surface when water was present, leading to decreased catalytic activity. Therefore, in this work, the environment around the vanadium is changed, using diethylenetriamine pentaacetic acid (dtpa) as a ligand and anchoring it to the surface of a mesoporous silica nanoparticle, to investigate its effect on vanadium’s ability to perform oxidation reactions. VO(dtpa)-APMS was synthesized by covalently linking the multi-dentate chelator dtpa onto the surface through peptide coupling of one of the acetate groups to aminopropyltriethoxysilane (APTES), condensing the dtpa-APTES molecule onto the mesoporous silica surface, and then exchanging a vanadyl salt into the resulting solid. Physical characterization of the material confirmed that the substrate retained its porosity after modification, and that the vanadium did not leach from the solid, in contrast to samples that did not contain dtpa. Solid-state EPR spectroscopy, combined with ongoing computational modeling, indicated that the vanadium was in a distorted five-coordinate environment. Various vanadium catalysts have been shown to oxidize alkanes, alkenes, alcohols and aromatic compounds. To further understand the catalyst’s ability to perform oxidation reactions, mechanisms of sulfides and alkenes were studied. Two model substrates were chosen for the investigation: CEES and cis-cyclooctene. The catalytic system effectively oxidizes CEES at room temperature in less than 15 minutes and cis-cyclooctene at 47 °C within 3 hours, using a peroxyacid generated in situ as the oxidant source. Kinetic experiments demonstrated that the mechanism of the sulfide reaction changed at higher temperatures, while the alkene reaction did not. In each reaction, a partial negative charge on the peroxyacid during the oxidation process was indicated. The confirmation of radical formation in the mechanism was experimentally shown by the appearance of an induction period when diphenylamine, a radical trap, was introduced into the reaction. VO(dtpa)-APMS performs two catalytic oxidations: the oxidation of propionaldehyde to make the peroxyacid and the oxidation of alkenes or sulfides. In the first reaction, O2 binds to the vanadium complex to form a superoxo eta-1-bound O2 radical. This species leads to the formation of peroxyacid through a radical process. The peroxyacid produced in this manner can then react with a sulfide or an alkene in a process also catalyzed by the VO(dtpa) complex. The peroxyacid coordinates with the vanadium center. Upon coordination, the sulfide or alkene directly reacts with the oxygen of the peroxyacid while the peroxyacid is being deprotonated. A 6-coordinate catalyst intermediate is formed prior to the release of the oxidation product and propionic acid to regenerate the VO(dtpa) complex.

Optimized methods for the regioselective and stereospecific synthesis of both trisubstituted and tetrasubstituted olefins as single isomers from E-β-chloro-α-iodo-α,β-unsaturated esters have been developed from previous work done in the Ogilvie lab. These optimized methods have led to the synthesis of trans isomeric enediynes that can be photoisomerized to their respective cis isomers and subsequently undergo microwave-assisted Bergman cycloaromatizations. Furthermore, both cis and trans isomeric enediynes that have propargyl ether substituents have been found to be able to undergo photoactivated Bergman cyclizations without the need for an intermolecular hydrogen donor. A mechanism study has confirmed that the Bergman cyclization products that form without the presence of an intermolecular hydrogen donor undergo a series of 1,5-hydrogen shifts as intermediates. A series of optimizations to these reactions were carried out, in part by utilizing electron-donating or electron-withdrawing functional groups to help stabilize the resulting radicals that form on the intermediates, and thus increase the yield of the associated Bergman cyclization products.

The last decade has been marked by a large increase of demand for green chemistry processes. Consequently, chemists have focused their efforts on the development of more direct routes toward different classes of targets. In that regard catalysis has played a crucial role at enabling key bond formations that were otherwise inaccessible or very energy and resources consuming. The central theme of this body of work concerns the formation of C–N bonds, either through transition metal catalysis or organocatalysis. These structural units being highly recurrent in biologically active molecules, the establishment of more efficient routes for their construction is indispensable. The first part of this thesis describes a new method for the synthesis of isoquinolines from the oxidative coupling/annulation of alkynes with N-tert-butyl benzaldimines via Rh(III) catalysis (Chapter 2). Preliminary mechanistic investigations of this system pointed to the involvement of Rh(III) in the C–H bond cleavage step as well as in the C–N bond reductive elimination that provides the desired heterocycle. Following this oxidative process, a Rh(III)-catalyzed redox-neutral approach to isoquinolones from the reaction of benzhydroxamic acids with alkynes is presented (Chapter 3). The discovery that an N–O bond contained in the substrate can act as an internal oxidant was found to be very enabling. Indeed, it allowed for milder reaction conditions, broader scope (terminal alkyne and alkene compatible) and low catalyst loadings (0.5 mol%). Mechanistic investigations on this system were also conducted to identify the nature of the C–N bond formation/N–O bond cleavage as well as the rate-determining step. The second part of this work presents mechanistic investigations performed on a recently developed intermolecular hydroamination reaction catalyzed through tethering organocatalysis (Chapter 4). This transformation operates via the reversible covalent attachment of two reactants, a hydroxylamine and an allylamine, to an aldehyde catalyst by the formation of a mixed aminal. This allows a difficult intermolecular Cope-type hydroamination to be performed intramolecularly. The main kinetic parameters associated with this reaction were determined and they allowed the generation of a more accurate catalytic cycle for this transformation. Attempts at developing new families of organocatalysts are also discussed.

Optimized methods for the regioselective and stereospecific synthesis of both trisubstituted and tetrasubstituted olefins as single isomers from E-β-chloro-α-iodo-α,β-unsaturated esters have been developed from previous work done in the Ogilvie lab. These optimized methods have led to the synthesis of trans isomeric enediynes that can be photoisomerized to their respective cis isomers and subsequently undergo microwave-assisted Bergman cycloaromatizations. Furthermore, both cis and trans isomeric enediynes that have propargyl ether substituents have been found to be able to undergo photoactivated Bergman cyclizations without the need for an intermolecular hydrogen donor. A mechanism study has confirmed that the Bergman cyclization products that form without the presence of an intermolecular hydrogen donor undergo a series of 1,5-hydrogen shifts as intermediates. A series of optimizations to these reactions were carried out, in part by utilizing electron-donating or electron-withdrawing functional groups to help stabilize the resulting radicals that form on the intermediates, and thus increase the yield of the associated Bergman cyclization products.

Unspecific peroxygenase (EC 1.11.2.1) represents a group of secreted hemethiolate proteins that are capable of catalyzing the selective mono-oxygenation of diverse organic compounds using only H2O2 as a cosubstrate. In this study, the peroxygenase from Agrocybe aegerita (AaeUPO) was found to catalyze the hydroxylation of various linear (e.g n-hexane), branched (e.g. 2,3-dimethylbutane) and cyclic alkanes (e.g. cyclohexane). The size of n-alkane substrates converted by AaeUPO ranged from gaseous propane (C3) to n-hexadecane (C16). They were mono-hydroxylated mainly at the C2 and C3 position, rather than at the terminal carbon, and the corresponding ketones were formed as a result of overoxidation. In addition, a number of alkenes were epoxidized by AaeUPO, including linear terminal (e.g. 1-heptene), branched (2-methyl-2-butene) and cyclic alkenes (e.g. cyclopentene), as well as linear and cyclic dienes (buta-1,3-diene, cyclohexa-1,4-diene). Furthermore, the conversion of terminal alkynes (e.g. 1- octyne) gave the corresponding 1-alkyn-3-ol in low yield. Some of the reactions proceeded with complete regioselectivity and - in the case of linear alkanes, terminal linear alkenes and alkynes - with moderate to high stereoselectivity. The conversion of n-octane gave (R)-3-octanol with 99% enantiomeric excess (ee) and the preponderance of the (S)-enantiomer reached up to 72% ee of the epoxide product for the conversion of 1-heptene. Catalytic efficiencies (kcat/ Km) determined for the hydroxylation and respectively epoxidation of the model compounds cyclohexane and 2-methyl-2-butene were 2.0 × 103 M-1 s-1 and 2.5 × 105 M−1 s−1. The results obtained in the deuterium isotope effect experiment with semideuterated n-hexane and the radical clock experiment with norcarane clearly demonstrated that the hydroxylation of alkanes proceeds via hydrogen abstraction, the formation of a substrate radical and a subsequent oxygen rebound mechanism. Moreover, stopped-flow experiments and substrate kinetics proved the involvement of a porphyrin radical cation species (compound I; AaeUPO-I) as reactive intermediate in the catalytic cycle of AaeUPO, similar to other hemethiolate enzymes (e.g. cytochrome P450 monooxygenases, P450s). / Die Gruppe der Unspezifischen Peroxygenasen (EC 1.11.2.1) umfasst extrazelluläre Häm-Thiolat-Enzyme, die mittels H2O2 als Cosubstrat die selektive Monooxygenierung unterschiedlicher organischer Verbindungen katalysieren. In der vorliegenden Arbeit konnte gezeigt werden, dass die von Agrocybe aegerita sekretierte Peroxygenase (AaeUPO) verschiedene lineare (z. B. n-Hexan), verzweigte (z. B. 2,3-Dimethylbutan) und zyklische Alkane (z. B. Cyclohexan) hydroxyliert. Die Größe der von der AaeUPO umgesetzten Substrate reichte vom gasförmigen Propan (C3) bis hin zu n-Hexadekan (C16). Die Alkane wurden bevorzugt am zweiten und dritten Kohlenstoffatom (C2 und C3) hydroxyliert; eine Hydroxylierung am terminalen Kohlenstoff konnte nur vereinzelt und in geringem Umfang beobachtet werden. Die Überoxidationen der primär gebildeten, sekundären Alkohole führte außerdem zur Entstehung der entsprechenden Ketonderivate. Darüber hinaus wurde eine Vielzahl linearer terminaler (z. B. 1-Hepten), verzweigter (z. B. 2-Methyl-2-Buten) und zyklischer Alkene (z. B. Cyclopenten) sowie linearer und zyklischer Diene (1,3-Butadien, 1,4-Cyclohexadien) durch die AaeUPO epoxidiert. Die Umsetzung terminaler Alkine (z. B. 1-Octin) führte zur Entstehung der jeweiligen 1-Alkin-3-ole. Manche dieser Reaktionen verliefen ausgeprägt regioselektiv und, im Falle der linearen Alkane sowie der linearen terminalen Alkene und Alkine, mit mittlerer bis hoher Stereoselektivität. So ergab beispielsweise die Umsetzung von n-Octan einen Enantiomerenüberschuss größer 99% für (R)-3-Octanol; die Epoxidierung von 1-Hepten lieferte einen Enatiomeerenüberschuss (ee) von bis zu 72% für das (S)-Enantiomer. Die katalytischen Effizienzen, die für die Hydroxylierung bzw. Epoxidierung der Modellverbindungen Cyclohexan und 2-Methyl-2-Buten ermittelt wurden, betragen 2.0 × 103 M-1 s-1 und 2.5 × 105 M−1 s−1. Der ausgeprägte Deuterium-Isotopen-Effekt, der im Zuge der Umsetzung von semideuteriertem n-Hexan beobachtet wurde sowie die Ergebnisse des Radical-Clock-Experiments mit Norcarane als Substrat bestätigten, dass die Hydroxylierung von Alkanen über Wasserstoffabstraktion, die Bildung eines Substratradikals und anschließende direkte Sauerstoffrückbindung verläuft. Die Stopped-Flow-Experimente belegen zudem das Auftreten eines Porphyrin-Kationradikal-Intermediates (Compound I; AaeUPO-I) im katalytischen Zyklus der AaeUPO (vergleichbar mit dem reaktiven Intermediat der P450-Monooxygenasen).

None.