Metal-catalysed cross-coupling is a powerful transformation in chemistry due to its applicability in the preparation of a wide range of important products relevant in pharmaceutical, agrochemical, and other industries. Recently, there has been a growing interest towards the copper catalytic system which stems from its low cost and toxicity compared to traditional heavier transition metal catalysts. In fact, copper-mediated cross-coupling has been known for a century since Fritz Ullmann reported the first biaryl coupling using copper. However, this classical reaction suffered from harsh conditions, which impeded further development. A renaissance in copper-catalysed cross-coupling reactions can be witnessed in the last decades owing to the incorporation of auxiliary ligands which allows for milder conditions to be employed. However, there is still a lack of coherent mechanistic understanding of the general copper-catalysed cross-coupling, which hinders its wider applications.
The work discussed in this thesis focusses on the copper-catalysed cross-coupling for the formation of new carbon-carbon bonds. Specifically, the work concerns the Ullmann-Hurtley or simply Hurtley reaction, defined as the copper-catalysed cross-coupling between an aryl halide and an activated methylene compound. Copper-catalysed C-C couplings are much less reported in the literature compared to analogous C-N and C-O couplings, hence this work aims to expand the scope and improve the performance of C-C cross-coupling, in addition to elucidating the mechanism for a better understanding of the copper catalytic system.
The first part of the work explores catalytic studies of the Ullmann-Hurtley cross-coupling reaction based on a selected model reaction between diethyl malonate and iodobenzene. The CuI / phen system was found to successfully catalyse the model reaction to give 78% yield of the product under the optimised conditions. The scope of the ligand was explored, focussing on second-generation ligands made up of oxalic and tartaric diamides as they are known to activate less reactive aryl chlorides. However, the performance of these ligands is limited by their poor solubility. The scope of both aryl halide and methylene substrates were also expanded. Successful cross-coupling was obtained with a wide range of mono-substituted aryl iodides and symmetric malonic esters. However, the activation of aryl bromides and chlorides remains challenging under the present conditions. Additionally, the extent of decarboxylation of malonic esters in the current scope varied depending on the nature of the side group. Activation of diethyl malonate using organolithium is also reported.
The second part of the work aims to elucidate the mechanism of the Ullmann-Hurtley cross-coupling reaction. Firstly, attempts toward the synthesis of solid-state Cu(I) complexes were performed. [L]Cu-nucleophile structure is particularly of interest as it is proposed to be the active catalytic species. X-ray crystallography was used as the primary characterisation method to determine the explicit structural information of the copper complexes. A number of Cu(I) and Cu(II) complexes with varying structural geometries were successfully synthesized and characterised - eight of which are novel including two ligated, substrate-bound Cu(I) complexes. The reactivity of some of these complexes was then exploited to assess their potential intermediacy in the cross-coupling reaction. The observed reactivity of the substrate-bound Cu(I) complexes highlights their competency as an active catalyst in the copper-catalysed reactions. Pre-equilibrium and off-cycle processes of various copper states during the catalytic cycle was also proposed based on the structure-activity results. Separate attempts toward the synthesis of organolithium cuprate complexes were unsuccessful.
Finally, investigations into the kinetics of the Ullmann-Hurtley cross-coupling were conducted based on the model catalytic reaction under the optimised conditions. The progress of the reactions was monitored by regular sampling and the concentration data was derived via 1H NMR spectroscopy. The VTNA method was adopted for data analysis and manipulation to obtain useful kinetic information from a series of experiments. A general positive dependence of the rate in [malonate], [aryl iodide], [Cu] and [ligand] was observed as anticipated. Unexpectedly, a rate retardation occurs after an apparent initial rate acceleration for the reactions at high ligand concentrations. Furthermore, orders in catalyst of less than unity suggest that catalyst deactivation is present in this system. The same “excess” experiments revealed that the catalyst deactivation is not due to product inhibition, although the presence of water can inhibit the cross-coupling to some extent. Disproportionation of the copper catalyst was proposed to be a potential pathway towards catalyst deactivation, evident from the formation of Cu(II) species experimentally.