Instigated by environmental concerns and resource scarcity, a shift towards a circular, bio-based economy is on the horizon. This inevitable transition will hinge on the deployment of alternative technologies that are able to fulfil the needs of our society in a sustainable way. Within this context, plant biomass in the form of lignocellulose represents a promising renewable resource. Lignocellulose is a heterogeneous substance, mainly composed of carbohydrates (cellulose and hemicellulose) and a phenolic biopolymer, named lignin. These constituents form a recalcitrant composite material, engineered by Nature to withstand (bio)chemical conversion. In analogy to the petrorefinery, a biorefinery aims to transform biomass into a range of useful products, including chemicals, materials, and fuels.Historically, lignocellulosic biorefineries have been geared towards optimal carbohydrate valorisation (e.g. paper, cellulosic ethanol), whereas lignin has been considered as an inferior and inconvenient biopolymer. Lignin though embodies the largest source of bio-aromatics, and the utilisation of this feedstock for the production of chemicals is gaining increasing interest. However, primarily focusing on carbohydrate valorisation often evokes irreversible lignin condensation, impeding the selective conversion of lignin into bio-aromatics. To cope with the problem of lignin degradation, this dissertation promotes a shift towards an alternative mind-set, stating that lignin valorisation should already be taken into account in the early stages of the biorefinery, while dealing with (hemi)cellulose later (i.e. lignin-first concept). By doing so, next generation biorefineries will be able to obtain more value from lignin, and in extension, from the entire biomass.One of the few biorefinery schemes that elegantly circumvents lignin condensation, is named Reductive Catalytic Fractionation (RCF). RCF targets solvolytic extraction of lignin from raw biomass, combined with instantaneous tandem depolymerisation-stabilisation. The latter step is facilitated by a heterogeneous redox catalyst (e.g. Ru/C, Pd/C) under hydrogen atmosphere, and yields a low molecular weight lignin oil containing phenolic monomers, dimers and small oligomers. The carbohydrate fraction (cellulose and hemicellulose) on the other hand is retrieved as a delignified pulp, ready for further valorisation. Despite its promising potential, the RCF technology is still in its infancy. The relationship between process conditions and their effect on the main biomass constituents has not yet been profoundly investigated. A better understanding would allow to modify and steer the RCF biorefinery towards a desired (alternative) outcome, which is the ultimate goal of this PhD dissertation.One of the hurdles that obstructs industrial implementation is the requirement for relatively high temperatures to achieve effective biomass delignification. RCF is typically performed with pure low boiling alcohols at 250 °C, resulting in an unfavourably high operating pressure (e.g. 110 bar in case of methanol). To facilitate RCF at milder conditions, two strategies were investigated, being (i) the implementation of acidic or alkaline additives and (ii) the utilisation of alcohol/water mixtures instead of pure alcohols. Mildly acidic media were found to effectuate lignin extraction at lower temperature and pressure (200 °C, 58 bar), and are furthermore compatible with reductive depolymerisation-stabilisation. Alkaline media on the other hand cause substantial lignin repolymerisation, rendering low lignin monomer yields and a larger fraction of oligomers. Besides applying additives, tuning the composition of the solvent mixture forms another strategy since alcohol/water mixtures (e.g. methanol/water, ethanol/water) were found to extract more lignin than either the pure alcohol or pure water. The degree of delignification correlates with the polarity of the solvent mixture.The utilisation of acids and alcohol/water mixtures not only enhances biomass delignification, it also impacts the hemicellulose fraction. Mildly acidic and/or relatively polar media promote extraction of hemicellulose and lignin from the biomass, while the cellulose fraction remains intact. This phenomenon provides a tool to tune the composition of the carbohydrate pulp (holocellulose vs. cellulose) and thus increases the versatility of RCF. In case hemicellulose solubilisation is anticipated, conversion to stable carbohydrate products should be targeted in order to mitigate degradation. Two strategies were put forward that are compatible with RCF. The first one targets hemicellulose conversion by means of acid-catalysed alcoholysis into alkyl pyranosides (e.g. methyl xyloside). The presence of water should be avoided to prevent hydrolysis towards xylose or other sugars, which are relatively unstable. Secondly, the hydrolytic hydrogenation of hemicellulose towards C5 and C6 polyols was demonstrated. This approach uses the same redox catalyst for carbohydrate hydrogenation as for lignin stabilisation. Fine-tuning the various reaction parameters is key to enable effective conversion of both biopolymers.Simultaneous depolymerisation of lignin and hemicellulose obligates down-stream separation of the obtained products. Therefore, RCF was demonstrated using a solvent mixture comprising n‑butanol and water. This solvent system enables conversion of lignin and hemicellulose during RCF, and provides an integrated separation approach for the solubilised products (i.e. phenolics and polyols, respectively) once the reaction is completed. The limited miscibility of n‑butanol and water can be used to separate lignin-derived phenolics (preferentially in n‑butanol phase) from the more polar carbohydrate products (preferentially in aqueous phase) via liquid-liquid extraction. Noteworthy, the n‑butanol/water mixture is biphasic below the upper-critical solution temperature (circa 125 °C), but monophasic at typical RCF temperatures (≥160 °C). Because of this dual behaviour, the intrinsic complexity of a truly biphasic catalytic system is omitted, while still exploiting the benefits of an integrated product separation after reaction. A proof-of-concept on 2 L scale demonstrates the scalability potential of this biorefinery concept.