L'objectif de cette étude est de comprendre et de modéliser la phénoménologie du transport vertical dense descendant de particules de la classe A de la classification de Geldart. Dans un premier temps, une étude expérimentale est réalisée sur une maquette en statique (absence de circulation de solide) dans le but de déterminer expérimentalement l'effet des paramètres opératoires sur les grandeurs caractéristiques de la défluidisation des particules de FCC : vitesses de sédimentation, porosité de la phase dense, temps caractéristiques…. Ces données sont nécessaires pour l'étude de l'écoulement gaz/solide dense vertical descendant. La simulation numérique en 2D de la défluidisation est effectuée et les prédictions sont comparées aux données expérimentales. Dans un deuxième temps, des essais sur une maquette permettant de reproduire les phénomènes observés industriellement dans les écoulements denses verticaux descendants de particules sont entrepris. Les observations visuelles complètent les mesures de pressions locales obtenues le long de l'écoulement à différentes conditions avec et sans injection de gaz d'aération. L'étude expérimentale consiste à : - déterminer les limites des différents régimes en termes de débit surfacique de solide et de débit d'aération ; - établir les propriétés de l'écoulement dans les différents régimes. Dans un troisième temps, les propriétés des écoulements de différents régimes sont étudiées et modélisées par une approche monodirectionnelle du type bulle-émulsion. / The objective of this study is to understand and model the phenomenology of the vertical downward dense transport of class A particles of the Geldart classification. Initially, an experimental study is conducted on a static fluidized bed (no flow of solid) in order to determine experimentally the effect of operating parameters on the defluidization properties of FCC particles, such as, sedimentation rates, dense phase porosity, characteristic times ... These data are needed to study gas/solid dense downward flow. 2D numerical simulations of defluidization are performed and the predictions are compared with experimental data. In a second step, experiments are undertaken in a pilot unit able to reproduce the gas/solid dense downward flow phenomena observed in industrial units. Visual observations complement the local pressure measurements profile obtained for the different flow conditions with and without external injection of gas. The experimental study is conducted to: - determine the boundaries of different flow patterns in terms of solid mass flux and gas flowrate; - establish flow properties in different flow patterns. Finally, flow properties of the different patterns are studied and modelled by a monodimensional bubble/emulsion approach.

Many particulate processes are preferably implemented in circulating fluidized beds (CFB) over traditional low-velocity fluidization to take advantage of the many benefits of circulating systems. Fluid catalytic cracking (FCC) is one of the most successfully applied processes in CFB technology, with more than 350 FCC units in operation worldwide. Despite its extensive use, an understanding of the complex behaviour of these units is incomplete.<p> A theoretical and experimental evaluation of the fluidization behaviour was conducted in the CFB riser, standpipe, and stripper. Initially, an extension of the existing CFB in the Fluidization Laboratory of Saskatchewan was designed. The experimental program conducted in this study included an examination of the solids flow behaviour in the riser, interstitial gas velocity in the downcomer, and stripping efficiency measurements. The hydrodynamic behaviour of the stripper was modeled using Multiphase Flow with Interphase eXchanges (MFIX) CFD code.<p> The solids flow behaviour in the bottom zone of a high-density riser was investigated by measuring the local upwards and downwards solids flux. Solids circulation rates between 125 and 243 kg/(m2⋅s) were evaluated at a constant riser superficial gas velocity of 5.3 m/s. The effect of the riser superficial gas velocity of the local upflow at the riser centerline was also conducted at a solids circulation rate of 187 kg/(m2⋅s). The results show that there is little variation in the local net solids flux at radial locations between 0.00 ¡Ü r/R ¡Ü 0.87. The results indicate that a sharp regime change from a typical parabolic solids flux profile to this more radially uniform solids flux profile occurs at a gas velocity between 4.8 and 4.9 m/s.<p> To quantify stripping efficiency, the underflow of an injected tracer into the standpipe must be known. Quantification of the underflow into the standpipe requires knowledge of two main variables: the interstitial gas velocity and the tracer gas concentration profiles in the standpipe. Stripping efficiency was determined for stripper solids circulation rates of 44, 60, and 74 kg/(m2⋅s) and gas velocities of 0.1, 0.2, and 0.3 m/s. For most conditions studied, the interstitial gas velocity profile was found to be flat for both fluidized and packed bed flow. The stripping efficiency was found to be sensitive to the operating conditions. The highest efficiency is attained at low solids circulation rates and high stripping gas velocities.<p> In the numeric study, stripper hydrodynamics were examined for similar operating conditions as those used in the experimental program. Due to an improved radial distribution of gas and decreasing bubble rise velocity, mass transfer is deemed most intense as bubbles crest above the baffles into the interspace between disc and donut baffles. Stripping efficiency is thought to improve with increasing gas velocity due to an increased bubbling frequency. Stripping efficiency is thought to decrease with increasing solids circulation rates due to a lower emulsion-cloud gas interchange coefficient and a decreased residence time of the emulsion in the stripper.

Many particulate processes are preferably implemented in circulating fluidized beds (CFB) over traditional low-velocity fluidization to take advantage of the many benefits of circulating systems. Fluid catalytic cracking (FCC) is one of the most successfully applied processes in CFB technology, with more than 350 FCC units in operation worldwide. Despite its extensive use, an understanding of the complex behaviour of these units is incomplete.<p> A theoretical and experimental evaluation of the fluidization behaviour was conducted in the CFB riser, standpipe, and stripper. Initially, an extension of the existing CFB in the Fluidization Laboratory of Saskatchewan was designed. The experimental program conducted in this study included an examination of the solids flow behaviour in the riser, interstitial gas velocity in the downcomer, and stripping efficiency measurements. The hydrodynamic behaviour of the stripper was modeled using Multiphase Flow with Interphase eXchanges (MFIX) CFD code.<p> The solids flow behaviour in the bottom zone of a high-density riser was investigated by measuring the local upwards and downwards solids flux. Solids circulation rates between 125 and 243 kg/(m2⋅s) were evaluated at a constant riser superficial gas velocity of 5.3 m/s. The effect of the riser superficial gas velocity of the local upflow at the riser centerline was also conducted at a solids circulation rate of 187 kg/(m2⋅s). The results show that there is little variation in the local net solids flux at radial locations between 0.00 ¡Ü r/R ¡Ü 0.87. The results indicate that a sharp regime change from a typical parabolic solids flux profile to this more radially uniform solids flux profile occurs at a gas velocity between 4.8 and 4.9 m/s.<p> To quantify stripping efficiency, the underflow of an injected tracer into the standpipe must be known. Quantification of the underflow into the standpipe requires knowledge of two main variables: the interstitial gas velocity and the tracer gas concentration profiles in the standpipe. Stripping efficiency was determined for stripper solids circulation rates of 44, 60, and 74 kg/(m2⋅s) and gas velocities of 0.1, 0.2, and 0.3 m/s. For most conditions studied, the interstitial gas velocity profile was found to be flat for both fluidized and packed bed flow. The stripping efficiency was found to be sensitive to the operating conditions. The highest efficiency is attained at low solids circulation rates and high stripping gas velocities.<p> In the numeric study, stripper hydrodynamics were examined for similar operating conditions as those used in the experimental program. Due to an improved radial distribution of gas and decreasing bubble rise velocity, mass transfer is deemed most intense as bubbles crest above the baffles into the interspace between disc and donut baffles. Stripping efficiency is thought to improve with increasing gas velocity due to an increased bubbling frequency. Stripping efficiency is thought to decrease with increasing solids circulation rates due to a lower emulsion-cloud gas interchange coefficient and a decreased residence time of the emulsion in the stripper.