CFD modelling of spouted beds for biomass thermochemical valorisation

Spouted bed reactors are a kind of fluidised bed reactors, particularly suitable to process coarse and irregular particles. For this reason, they can successfully be applied for the thermochemical conversion of agricultural residues. This is particularly valuable, as it not only leads to the generation of renewable energy, but also valorises a waste material, that would otherwise have to be disposed of, possibly with associated impacts. However, spouted bed reactors are still not an established technology: some aspects of their fluid dynamics are poorly understood, and no established criterion for their scale-up exists. In this framework, modelling activities can help tackling these problems, but researchers do not agree on many procedures of the different modelling methodologies. The features and applications of the main modelling techniques for spouted beds are reviewed at the beginning of the thesis, and the knowledge gaps are highlighted. The activities detailed in this thesis contributed to closing some of these knowledge gaps, fostering the development of the spouted bed technology. Initially, we performed an experimental analysis that assessed the extent of particle segregation in a lab-scale square-based spouted bed, containing several binary mixtures of particles. The focus was then mainly put on computational fluid dynamics (CFD) simulations, which can provide unparalleled insights on the behaviour of chemical reactors. Preliminary simulations provided a good reproduction of the segregation phenomena and operational parameters, but also showed that the approach is still not mature to reproduce severely non-ideal particles and must be tuned with care. Hence, we moved onto more fundamental studies, based on literature data. These allowed us to better understand how to apply the main CFD approaches, tune them more accurately, and provide answers to uncertainties related to the role of some sub-models, with a particular focus on the choice of the drag model. The thus optimised simulation strategy was finally applied to a scaled-up pilot spouted bed to reproduce the solids residence time distribution. Two side activities were also undertaken. The first one regards the inclusion of the reaction kinetics in an Aspen Plus model of the gasification process in a spouted bed. This approach provided enhanced results compared to a simpler equilibrium model. The second activity consisted in an experimental analysis of biomass gasification char samples, and a campaign to assess their suitability to adsorb hydrogen sulphide. Char currently represents a drawback in gasification, as it is a waste that must be disposed of, so establishing alternative pathways to valorise it is particularly noteworthy.