The production of municipal solid waste has continued to grow in recent years. In Italy, municipal solid waste production reaches about 29 million tons per year. The organic fraction of municipal solid waste (OFMSW), which accounts for 30-40% of the total waste, usually undergoes biological treatments such as anaerobic digestion or composting, or is incinerated or landfilled. Biological treatments are considered not economically viable due to the long processing time (20-30 days), while incineration and landfilling are considered as low cost but polluting processes. In contrast, Hydrothermal Carbonization (HTC) is a cost-effective process to treat organic waste especially for the shorter processing time (0.5-8 h) and the possibility to treat directly wet heterogeneous materials. This thesis aims to investigate the potential use of HTC to upgrade OFMSW and other biomasses to biofuels or valuable byproducts. Chapter 1 gives an overview on the state of the art of HTC technology applied to organic wastes, focusing both on mechanisms and on the characteristics of reaction products. Chapter 2 investigates the potential use of HTC to upgrade the organic fraction of municipal solid waste (OFMSW) into biofuel and byproducts. The impact of process conditions (process time, temperature and solid load) on the formation, chemical and energy properties of hydrochar was deeply investigated. To analyze the behavior of hydrochar as a solid biofuel, the combustion (oxidation) of hydrochar and the co-firing (co-oxidation) of hydrochars and coals was also investigated. The results show that, especially at HTC harsher conditions, hydrochar is a “coal-like” material, that can be used as a valuable solid biofuel. The results evidenced that hydrochar is composed of primary char and secondary char. Primary char (non-extractable with organic solvents) has characteristics similar to coal and could be recovered for combustion, while secondary char is an amorphous and more volatile solid, easily extracted with organic solvents, which found possible application as a source of biochemicals and liquid bio-fuel. Chapter 3 investigates the use of HTC as a promising pre-treatment to enhance the biomethane potential during anaerobic digestion of OFMSW. Anaerobic digestion experiments were carried out using the HTC process liquid and the entire HTC reaction mixtures. Results proved that, when compared to the raw OFMSW, the use of HTC liquid and HTC mixture into AD lead to an increase of biomethane production of up to 37% and 363% by volume, respectively. Chapter 4 reports an HTC kinetics study and a kinetic model, which accounted for reactions leading to the production of primary and secondary char, as well as the liquid and gas phases. The model was optimized using experimental data performed on a lignocellulosic feedstock (olive trimmings) and validated on two other types of biomasses (grape marc and Opuntia Ficus Indica) and was used as a reliable tool to predict the carbon distribution among HTC products. In this chapter an in-depth analysis was also carried out to understand the evolution of feedstock characteristics during the heat-up transient phase before reaching the HTC set-point temperature. The results show that during heat up, the feedstock carbonized to a considerable extent at 220-250 °C. Tests clearly show evidence of the transition between thermal hydrolysis and HTC. Chapter 5 presents a study conducted to evaluate the economic feasibility and the detailed energy and cost analyses of a hypothetical HTC plant transforming wet biomass into pelletized dry hydrochar. To achieve these goals, a model was developed on the basis of experimental results obtained previously on two other organic materials (grape marc and off-specification compost). The results show that, when operating the HTC plant with grape marc at the optimal HTC conditions (T=220 °C, t=1 h, dry biomass to water ratio=0.19), the production cost of hydrochar were determined to be 157 €/ton, competitive with the price of wood pellets (150-200 €/ton). This makes HTC a promising process for a large development at the industrial scale.

Le travail réalisé en collaboration entre l’entreprise CNIM et le laboratoire IUSTI a permis la miseen place d’un outil de simulation numérique afin d’étudier à l’échelle locale différents phénomènescouplés qui se produisent dans un réacteur de gazéification en lit fluidisé dense. L’approche choisie,DEM (Discrete Element Method), est basée sur le suivi de paquets de particules. Les résultats desimulation ont été comparés à des mesures expérimentales réalisées dans les laboratoires IUSTIet LERMAB : mesure de pression dans un lit fluidisé peu profond et caractérisation de la réactionde pyrolyse par suivis temporels de la masse volumique par une méthode innovante et de latempérature de la particule de biomasse ainsi que la composition des gaz produits. À plus grandeéchelle, les prédictions numériques du code ont été comparées à des mesures expérimentales deprofil de vitesse des particules et de mélange et ségrégation issues de la littérature. Les résultatsnumériques du lit fluidisé bisolide sans et avec réactions chimiques, obtenus sont en bon accordavec les mesures expérimentales. Ils ont permis de mettre en évidence le fort couplage entre lesphénomènes hydrodynamiques et thermochimiques.Ces travaux ouvrent de nouvelles perspectives tant sur le plan de l’expérimentation que de la simulation numérique où l’approche DEM retenue a montré un fort potentiel. L’extrapolation dumodèle DEM pour la simulation de la gazéification à des réacteurs industriels reste encore un défiau regard des ressources informatiques. / Energy production from green and renewable resources, such as biomass, are currently experiencinga significant growth. Thermochemical conversion of this biomass by gasification is a process usedfor over a century but still requires significant developments in terms of rentability optimizationand quality improvement of products gases.The work carried out in collaboration between the CNIM company and the IUSTI laboratoryallowed the establishment of a numerical simulation tool to study locally different coupled phenomenaoccurring in a dense fluidized bed gasification reactor. The chosen approach, DEM (DiscreteElement Method), is based on the monitoring of particle packets. The simulation results werecompared to experimental measurements realised in IUSTI and LERMAB laboratories : measuringpressure in a shallow fluidized bed and characterization of pyrolysis reaction with the timetracking of particle density by an innovative method, temperature and the composition of theproducts gases. On a larger scale, the numerical code predictions were compared with velocity,particles mixing and segregation profiles from experimental measurements of the literature. Theobtained numerical results of bi-solid fluidized bed with and without chemical reactions are ingood agreement with the experimental measurements. It helped to highlight the strong couplingbetween hydrodynamic and thermochemical phenomena.This work opens up new perspectives on the experimental plan and numerical simulation whichDEM approach has shown great potential. The extrapolation of the DEM model for the simulationof gasification industrial reactors remains a challenge in terms of computer resources.