Extraction of Hemicellulose Produced Furfural using Green Designer Solvents

Abstract

The rapid growth in demand for materials worldwide urges scientific research to find novel and renewable sources for traditional fossil fuels. Organic wastes from various sources (household, agricultural, municipal, and food wastes) can be exploited as raw materials for the synthesis of value-added products. Amongst the various valorization technologies utilized, catalytic treatment of hemicellulosic biowaste for furfural production is promising. Furfural is a key platform chemical utilized as a building block to synthesize a wide range of important products, such as pharmaceutical products, phenol resins, lubricants, nylons, plastics, and other organic solvents.

In the current state-of-art, hemicellulose is first hydrolyzed into monosaccharides (xylose and arabinose) that are then converted to furfural through acidic dehydration with aqueous HCl, H2SO4, or formic acid. Consequently, a separation step is needed to extract the furfural to avoid promoting the competing side reactions. The conventional approaches include reactive distillation or steam stripping, leading to high energy consumptions, or require volatile organic compounds (VOCs) such as methyl isobutyl ketone (MIBK) or toluene as extraction agents. However, the utilization of these VOCs poses risks to safety, health, and the environment due to their volatile and toxic nature. Therefore, the research of alternative separation techniques for the sustainable production of furfural from biowaste is essential.

In line with the global direction toward the application of “green solvents” in processing, ionic liquids (ILs) and deep eutectic solvents (DESs) have been widely proposed as alternative solvents capable of overcoming the disadvantages of VOCs. ILs and DESs are primarily characterized by their negligible vapor pressure. Additionally, depending on the choice of the molecular structure constituting the ILs or the DESs, they can also be considered biodegradable, non-toxic, inexpensive, easily synthesized, and highly tunable. This dissertation focuses on the design of novel hydrophobic ILs and DESs to improve the production of furfural in the biorefinery process by replacing VOCs, MIBK, and toluene with more sustainable and renewable solvents.

The research methodology combines high-throughput computational screening and rigorous experimental validation to identify and implement sustainable alternatives. The investigation begins with a quantum chemical analysis using the COSMO-RS method to screen hydrophobic deep eutectic solvents (HPDESs) and ionic liquids (ILs), enhanced by machine learning techniques for molecular design insights. This computational phase examines sigma profiles and thermodynamic properties to predict extraction performance. The study then progresses through systematic experimental stages, where selected promising candidates undergo characterization and testing for efficiency, followed by optimization studies using one-factor-at-a-time and design of experiments methodologies. The research then reports their usage in an integrated biphasic reaction-extraction system, investigating critical process parameters, including reaction time, temperature, phase ratios, and initial concentrations of both acid and xylose. The study also includes analyses of solvent stability, reusability, and regeneration across multiple cycles, addressing key industrial implementation concerns. This systematic investigation encompasses quantum chemical modeling, thermodynamic analysis, reaction kinetics, and process optimization, providing both theoretical insights and practical solutions for sustainable furfural production. The findings contribute to the broader green chemistry and sustainable biorefining fields, offering a framework for designing and implementing environmentally friendly solvents in biomass valorization processes.

Date of Award	27 May 2025
Original language	American English
Supervisor	Fawzi Banat (Supervisor)