Abstract
M.Sc.
This study initially focused on the synthesis of a set of triaryl phosphine ligands,
encompassing a broad range of electron withdrawing functionalities on the ortho-position
of one of the aryl rings. These varying moieties were readily incorporated into
diphenylphosphino benzaldehyde as starting material through both Wittig and Knoevenagel
chemistry. The ligands produced were tested in the Pd-catalysed Suzuki reaction. The
electronic as well as the steric nature of the alkene ligands largely dictated the activities
observed: the more electron poor or the bulkier the ligand, the higher the activity observed
in the Suzuki reaction. This is in contrast to much work in the literature stating that highly
active Suzuki catalysts require a very electron rich system. At the same time, the literature
indicates that co-ordinatively unsaturated Pd-complexes are also active catalysts. The
activities observed were ascribed to the ability of these electron poor bulky phosphine
alkene ligands to stabilise or promote the reductive elimination step of the Suzuki
mechanism in preference to the oxidative addition step, which is the typical rate
determining step.
The study then investigated carbonylation reactions, specifically the methoxycarbonylation
and hydrocarboxylation reactions, which are typically Brønsted acid co-catalysed. The
alternative was the first time use of metal-triflate based Lewis acids as co-catalysts in these
types of reactions. Thus, a systematic study was performed. It was found that metal
trifluoromethane sulfonate (hereafter referred to as triflate) based Lewis acid co-catalysts
outperformed the typical Brønsted acid co-catalysts by between one and a half to two and a
half times on the rate of the methoxycarbonylation reaction, depending on the substrate
used. The system was tested with Pd loadings in the region 2–0.03 mol%. A competing
heat-induced styrene polymerisation reaction ultimately affected the results at such low Pd
loadings. A low level kinetic analysis was performed indicating zero order kinetics on the
alkene concentration of the reaction, with a fractional order dependence on the Lewis acid
concentration. There was little to no effect on the linear/branched ratio of the product in
response to the use of the Lewis acid. The nature of the metal within the metal triflate
based co-catalyst also seemed to be critical to the reaction, with the 4+ based Zr and Hf
ultimately providing the highest obtainable turn over frequencies, the 1+ and 2+ based
metals providing no conversion, and reactivity in the presence of the 3+ metals depended on
the specifics of the reaction. In addition to a great deal of work being performed on styrene
and ethylene as substrates, reactions using phenylacetylene were also optimised. Here, it
was found that the bidentate BINAP ligand and the literature preferred ligand, PyPPh2
afforded good catalyst activity. Both of these ligands offered much faster catalyst systems
than PPh3 and various other bidentate ligands tested.