Lignocellulosic biomass is a promising alternative to non-renewable fossil fuel resources for the sustainable production of fuels and chemicals. Levulinic acid (LA) is considered to be a versatile platform chemical which can be derived from biomass and further upgraded to high-value products like levulinate esters. In this work, the direct chemical conversion of lignocellulosic biomass to ethyl levulinate (EL) through LA was studied in collaboration with Gascon Biomass Research Inc. in an effort to valorize their surplus of lignocellulosic feedstock obtained from biomass recycling activities. A novel one-pot, biphasic, acid-catalyzed hydrolysis-esterification process developed by Prof. Tom Baker's lab (uOttawa Chemistry) was used. Kinetic analyses of the overall reaction pathway and major reaction steps were performed to investigate the effect of reaction conditions such as time, temperature, alcohol concentration, and catalyst concentration and evaluate the kinetic and thermodynamic parameters. Reactions were well modeled by power law rate equations with applicability of the Arrhenius expression and exhibited nearly first-order dependence with respect to each of the corresponding reactants and catalyst, as expected. Kinetic studies of LA esterification to various levulinate esters showed that the reaction is endothermic, endergonic, non-spontaneous, and non-rate-limiting. Increased alcohol alkyl chain length and branching reduced the thermodynamic favourability of LA esterification, likely due to steric effects. Process considerations such as reactor pressure, LA and EL partitioning, solvent:feedstock ratio, solvent recycling, and alternative feedstocks were investigated through various lab-scale studies and simulation case studies with UniSim® Design. The results of experimental studies informed the development of a preliminary potential process flow diagram consisting of eleven batch slurry reactors under agitation in parallel. The product streams feed into a continuous separation process composed of flash vessels and distillation columns to separate and purify the various products, intermediates, unconsumed reactants, and recoverable catalyst and solvent. This work ultimately informs decisions regarding future pilot-scale experiments, process design, and optimization for the envisioned large-scale biorefinery to produce EL.

L'élaboration de méthodes de synthèse, basées sur l’utilisation d’espèces métalliques a été un sujet de tous les instances en chimie organique. Malgré l’efficacité des métaux utilisés en catalyse homogènes, leurs procédures de recyclage restent limitées. Ce pourquoi, une contrainte supplémentaire a été placée dans la conception de catalyseurs, pouvant offrir à la fois l'efficacité de la catalyse homogène et le recyclage de l’hétérogène. Dans ce contexte, les nanoparticules métalliques sont apparues comme objet phare, en raison de leurs propriétés physico-chimiques inégalées. On a découvert que les nanoparticules de métaux nobles présentaient des propriétés catalytiques similaires dans certains cas, aux complexes monoatomiques. De plus, les Au NPs ont montré une activité catalytique remarquable dans l'oxydation d’alcools activés sous O2. Nous avons donc envisagé des procédures multicatalytiques, basées sur les NPs d’Au. Notre choix d'utiliser des catalyseurs solides était pertinent, puisque les nano-catalyseurs, pour lesquels la fraction de sites actifs se trouve en surface, limitent les risques de cross-quenching. Ici, nous présentons trois nouveaux procédés bicatalytiques permettant l’accès, à des chromenes/quinoléines (53-93%) via une oxydation / Michael Addition/ aldolisation, combinant nanocatalyse et catalyse basique, l’accès à des ortho-THC (50-81%) via oxydation / arylation / cyclisation, combinant nanocatalyse et catalyse supportée, ainsi qu’une une oxydation / hydrolyse en cascade, pour accéder à l’HMLA (86%, sel 93%), un grand panel de produits d'activité biologique reconnue, utilisé en parfumerie ou visant une pré-industrialisation via la chimie en flux continu. / Elaboration of synthetic methods based on metal-catalyzed reactions has been a hot topic in organic chemistry. Despite good efficiency, catalysis proceeding homogeneously, are limited in the operation of recovering/recycling of the catalysts. An important stress was placed to design catalysis, offering both the efficiency of homogeneous catalysts and the recyclability of heterogeneous catalysts. In this context, metal nanoparticles merged as a key tool, due to their unique physical and chemical properties. Notably, Au NPs have shown remarkable catalytic activity in the oxidation of activated alcohols under O2 atmosphere. Since now, the access to more complex molecules is the next step forward for this field, we envisioned multicatalytic roads, based on the oxidation of activated alcohols via supported Au NPs. Our choice of using solid catalysts was relevant, since nanostructured catalysts for which the fraction of active sites are located on the surface, limit the risk of cross-quenching. The latter carbonyl formed, could be further converted in situ, via tandem protocol. Herein, we developed novel, atom- and step-economical bicatalytic one-pot processes, to access substituted chromenes/quinolines (53-93%) by tandem oxidation/hetero-Michael addition/aldolisation combining nanocatalysis and base catalysis, ortho-THCs (50-81%) via tandem oxidation/arylation/cyclisation combining nanocatalysis and supported catalysts and a tandem cascade oxidation/hydrolysis to access HMLA (86%, sel 93%). A large panel of products of biological activity relevance, pertaining to the fragrance chemistry or aiming in some cases, pre-industrial scalability via continuous flow applications.