Catalytic C-C Bond Cleavage for the Production of Chemicals from Lignin

Abstract

Lignin is a major component of lignocellulosic biomass and could be an important renewable feedstock in industry for the production of (aromatic) bulk and fine chemicals. To this end, the development of new catalytic processes is required; both to depolymerise the biopolymer into small aromatic building blocks, as well as to further convert these to valuable products. A major challenge in lignin depolymerisation is the recondensation of reactive fragments back to larger, more recalcitrant, fragments. In this thesis, this was circumvented by employing a tandem catalytic depolymerisation procedure. One catalyst, a water-stable Lewis acid, such as scandium triflate, was shown to be responsible for cleaving ether bonds between the lignin monomers, while a second rhodium-based catalyst further converted the liberated small fragments to stable end-products. The rhodium catalyst was shown to have two possible modes of action and the interplay between both catalyst components was shown to be crucial for determining the ultimate product selectivity. Consequently the desired product streams could be chosen: when using a weak Lewis acid, the major products of depolymerisation were 4-(1-propenyl)phenols, whereas when using strong Lewis acids, the major products were 4-methylphenols. The selective oxidation of catechol was studied as a potential route for valorisation of small lignin-derived aromatics. Using an iron(III) complex with a tris(2-pyridylmethyl)amine ligand as catalyst and molecular oxygen as the reagent, catechol could be readily and selectively converted to derivatives of cis,cis-muconic acid, under mild conditions (50 °C, atmospheric pressure of air). Further two-stage conversion by hydrogenation and transesterification, using commercially available supported catalysts, allowed the isolation of dimethyl adipate in 62% yield. A systematic investigation of the reaction conditions furthermore allowed the activity of the oxidation step to be increased 35-fold under optimal conditions (80 °C, 25 bar 5% O2 in N2, 0.1 mol% catalyst). The catalytic oxidative cleavage reaction was also demonstrated to work directly on catechols obtained from catalytic depolymerisation of candlenut lignin, although the yield of cleavage products found was lower in that case. In addition, the propyl substituent of the lignin-derived catechols drives isomerisation of the cis,cis-muconic acid, leading to significant formation of g-muconolactones instead. The mechanism of this unique and highly selective oxidative carbon-carbon cleavage reaction was investigated using a combined experimental and theoretical study. Based on the combined results, a mechanism is proposed where iron is able to act as a ‘redox’ buffer by taking on an intermediate spin state, enabling the spin-forbidden electron transfer from the catecholate moiety to oxygen. This leads to formation of a bridging peroxide intermediate, which undergoes facile homolytic splitting, leading to oxygen insertion into the aromatic catechol ring. The rate-determining step was found to be the formation of the bridging peroxide. Crucially, the experimentally observed effect of both substrate and ligand substitution was accurately reproduced by the computational study, lending credibility to the mechanism. Importantly, results on other ligand systems reported in literature can now also be understood using this mechanism.