Tailored Production of Sustainable Low-Cost Lignocellulosic Advanced Biofuel Blends as Diesel and Petrol Substitutes

Abstract

This thesis advances the production and understanding of alkyl levulinates from lignocellulosic biomass via ethanolysis and butanolysis, delivering four key contributions. It established steady-state operation for alcoholysis reactions, providing the first systematic demonstration of steady-state conditions for alcoholysis reactions under defined conditions. Earlier studies typically reported values at arbitrary time points without distinguishing between kinetic and steady-state phases. By defining conditions under which steady-state is achieved, this work created a rigorous benchmark for yield determination and enabled robust comparison across conditions and substrates, including the model carbohydrates and the lignocellulosic feedstock investigated here, corn cob. Building on this, the second contribution introduced a new framework for yield representation. The alcoholysis studies revealed for the first time that alkyl levulinates originate from both cellulose and hemicellulose, requiring a redefinition of molar yield normalisation to reflect both contributors accurately. Alongside this, the work quantified previously neglected co-products, most notably dialkyl ethers, which, together with levulinates and residual alcohols, form the three-component blends that can be used as blending components (demonstrated stable up to ~25 vol% in diesel). Additional co-products such as n-butyl acetate and n-butyl formate were also identified, expanding the known product spectrum and strengthening the basis for techno-economic evaluation. The third contribution is the determination, under clearly defined one-pot conditions, of the minimum acid charge required to catalyse the alcoholysis of a real lignocellulosic feedstock (corn cob). Because one-pot alcoholysis of whole biomass remains comparatively understudied, this threshold had not been established. Our results show that a non-trivial fraction of the acid is irreversibly consumed by interactions with the biomass matrix, so a baseline acidity must be supplied before productive conversion can proceed. Quantifying this requirement provides actionable guidance for catalyst utilisation and delivers data directly relevant to process design, technoeconomic modelling, and sustainability assessment. Finally, the thesis developed and validated a surrogate kinetic modelling approach based on molecular group additivity and hierarchical representation. Calibrated to glucose, cellulose, xylan, and corn cob, the model reproduced yields and product distributions for both ethanolysis and butanolysis processes and linked biochemical composition to predicted outcomes. While Chapter 4 characterised a broader set of Annex IX feedstocks, these remain to be tested experimentally, providing a basis for future validation and extension of the modelling framework. Together, these advances move alcoholysis from laboratory feasibility towards industrial relevance. They clarify what can be achieved under steady-state operation, establish a comprehensive and accurate yield framework, quantify neglected co-products of practical importance, and introduce predictive modelling tools that bridge laboratory data with process optimisation. More broadly, this thesis demonstrates how rigorous experimentation and modelling can transform alcoholysis from an empirical observation into a predictable, scalable, and strategically relevant pathway for advanced biofuel production.