Amidst depleting crude oil stocks and growing environmental concerns, eco-friendly manufacturing and sustainable energy alternatives are increasingly in demand. Biomass offers a promising and renewable energy source for the production of valuable fuels and chemicals such as 5-HMF, LA and their derivatives.
In recent decades, Sn-Beta has been extensively studied as a Lewis acid catalyst for biomass applications. Despite its high catalytic activity in glucose isomerisation (31% fructose yields in 30 minutes at 110 °C), Sn-Beta faces challenges for industrial feasibility due to a complex and time-consuming synthesis process, even at low Sn loading concentrations (< 2 wt.%). The use of fluoride anions as mineralizers in hydrothermal synthesis has environmental drawbacks and leads to the formation of micron-sized crystals, limiting intracrystalline diffusion. In this regard, this PhD project focused on a thorough investigation of active zeolite-based catalysts for glucose isomerisation and fructose dehydration. This involved the study of various metal-doped zeolite catalysts, using commercially available zeolites, particularly zeolite Y in its acidic form and doped with metals such Sn, Ga, Nb and Fe. The primary emphasis was placed on understanding the roles of various factors, such as Lewis and Brønsted acid sites, porosity, and solvent nature in these processes.
Prior to catalytic tests, the solubility of sugars was verified using an accurate HPLC method. By doing so, we aimed to identify any potential implications of this chemical property on catalytic activity and provide a reliable basis for accurate catalytic testing carried out afterwards. Our solubility measurements showed no detrimental impact on the catalytic activity obtained from glucose isomerisation and fructose dehydration under the reaction conditions studied.
Through a meticulous mass spectroscopic analysis, we detected two well-defined peaks of fragmentation pattern at m/z = 217 within the reaction mixture of our glucose isomerisation. These peaks directly corresponded to an anomeric mixture of methyl glucoside and fructoside, providing valuable insights into the composition of our glucose isomerisation reaction mixture.
In glucose isomerisation, the use of Sn- and Ga-doped zeolite Y, prepared via wetness impregnation protocol, in water as a solvent proved to be inactive due to strong solvent adsorption within the zeolite pores. However, these materials exhibited significant activity when methanol was used as a solvent. A remarkable glucose conversion of 90% and a high fructose selectivity of up to 55% were accomplished at 100 °C, under endogenous pressure, for 1-2 hours. Our investigation revealed a reaction pathway involving a hydride shift, facilitating the conversion of glucose into fructose and mannose and a Brønsted acid pathway, initiating the formation of methyl fructoside intermediate and its subsequent hydrolysis to fructose if water was added afterwards. While Brønsted acidity and porosity showed significant influence on the observed catalytic results, it was difficult to establish precise correlations between the catalytic performance and individual parameters. Therefore, it is imperative that various catalyst properties be explored in order to uncover and validate potential trends in their catalytic activity.
We conducted a screening of zeolite Y catalysts doped with various metals (such as Sn, Ga, Nb, and Fe) and metal oxides (such as CeO2, Nb2O5, and TiO2) for the selective dehydration of fructose in water. Our objective was to achieve enhanced production of 5-HMF, surpassing the reported literature yield values (≤ 50%) under similar conditions. Among the tested catalysts, CeO2/Y(80) and Nb2O5/Y(80) exhibited the highest 5-HMF yields, of approximately 40% under optimal reaction conditions (140 °C, 2 hours, endogenous pressure, and a molar ratio of substrate to catalyst of 1:300). The lower 5-HMF yield was attributed to the formation of unwanted by-products such as levulinic acid, formic acid, and/or undesired insoluble humins.
