CFD simulations were used to study single phase and multiphase flows through structured packing. Simulations utilizing a high fidelity, digital copy of a packing element were validated against experimental results for both single phase and multiphase flows. Single phase simulations were carried out on a variety of periodic packing elements to examine the impact of packing channel geometry on pressure drop. Multiphase simulations on periodic elements were used to examine the effect of hydrodynamic properties and boundary conditions. Single-phase simulations of nitrogen flow through the high fidelity geometry produced via X-ray CT scans showed average deviations less than 15% when compared to experimental measurements. This error was reduced to 7% when a mesh utilizing prism layers to accurately resolve the boundary layer was used. With a validated model for single phase flow, the application of CFD to packing design was investigated on periodic geometries with varied packing parameters (e.g. channel corrugation angle and channel side length). It was found that current industrial packings have channel geometries maximizing pressure drop, indicating some degree of optimization around channel geometry is possible depending on separation needs. Multiphase simulations using the Volume of Fluid model examined the effects of liquid density, viscosity, surface tension, and contact angle on small-scale packing geometries. Contact angle had the most pronounced influence on predicted wetting, and simulations demonstrated that using experimentally determined static contact angles was not an appropriate choice for the simulation contact angle. The predicted influence of surface tension qualitatively matched experimental data for wetted area. Liquid viscosity and density also demonstrated qualitative agreement with semi-empirical models derived from experimental data. Experimental data collected via absorption of CO2 into 0.1 mol/L NaOH were compared to simulation predictions using a geometry generated via X-ray CT scans. Wetted area predictions matched experimental data best when a fully wetting static contact angle (0°) was used, yielding simulated results 3.4% lower than experimental data on average. Irrigated pressure drop and holdup predictions were significantly higher than experimental data. / text

This research explored the use of CFD simulations to study single phase flows through structured packing. Flow rates were chosen to approximate those used in the vapor phase of industrial distillation columns. The results were evaluated against experimental results obtained with the same packing model and packed height. Several novel methods were employed to quickly obtain high validity results. A high-fidelity, digital copy of an actual packing element was created in seven hours through CT scanning. The meshing strategy employed adaptive, polyhedral meshing algorithms which resulted in high quality volume meshes with 80 percent less mesh elements than would be required with traditional tetrahedral meshing. Meshing and computation were performed on the TACC clusters. The permitted meshing with up to 57 million volume cells in less than 30 hours while simulations employing a realizable k-[epsilon] model converged in approximately two days using up to 544 processors. Nitrogen simulation predictions were found to be, on average, 7 percent below experimental measurements with water simulations showing considerably more error (~40%). The error is likely attributable a discrepancy between the simulation and experimental geometries. This discrepancy is due to an oversight in sample preparation and not a flaw in the CT scanning process of geometry creation. The volume of data generated in CFD simulation was found to be very valuable for understanding and benchmarking packing performance. Streamlines and contour plots were used to analyze the variation in performance both locally and throughout the packing stack. Significant variation was observed in flow pattern, velocity distribution, and pressure profiles throughout the column. However, the joint regions were found to be most adverse to column energy efficiency. / text

The mass transfer area of nine structured packings was measured as a function of liquid load, surface tension, liquid viscosity, and gas rate in a 0.427 m (16.8 in) ID column via absorption of CO₂ from air into 0.1 mol/L NaOH. Surface tension was decreased from 72 to 30 mN/m via the addition of a surfactant (TERGITOL[trademark] NP-7). Viscosity was varied from 1 to 15 mPa·s using poly(ethylene oxide) (POLYOX[trademark] WSR N750). A wetted-wall column was used to verify the kinetics of these systems. Literature model predictions matched the wetted-wall column data within 10%. These models were applied in the interpretation of the packing results. The packing mass transfer area was most strongly dictated by geometric area (125 to 500 m²/m³) and liquid load (2.5 to 75 m³/m²·h or 1 to 30 gpm/ft²). A reduction in surface tension enhanced the effective area. The difference was more pronounced for the finer (higher surface area) packings (15 to 20%) than for the coarser ones (10%). Gas velocity (0.6 to 2.3 m/s), liquid viscosity, and channel configuration (45° vs. 60° or smoothed element interfaces) had no appreciable impact on the area. Surface texture (embossing) increased the area by 10% at most. The ratio of effective area to specific area (a[subscript e]/a[subscript p]) was correlated within limits of ±13% for the experimental database: [mathematical formula]. This area model is believed to offer better predictive accuracy than the alternatives in the literature, particularly under aqueous conditions. Supplementary hydraulic measurements were obtained. The channel configuration significantly impacted the pressure drop. For a 45°-to-60° inclination change, pressure drop decreased by more than a factor of two and capacity expanded by 20%. Upwards of a two-fold increase in hold-up was observed from 1 to 15 mPa·s. Liquid load strongly affected both pressure drop and hold-up, increasing them by several-fold over the operational range. An economic analysis of an absorber in a CO₂ capture process was performed. Mellapak[trademark] 250X yielded the most favorable economics of the investigated packings. The minimum cost for a 7 m MEA system was around $5-7/tonne CO₂ removed for capacities in the 100 to 800 MW range. / text

The key to success in separation of liquid mixtures is the efficient creation and utilization of vapour-liquid contact area. By packing the column with gas-liquid contact devices such as structured packing, the vapour-liquid contact area can be increased. However, the efficiency of these packed columns depends strongly on the local flow behaviour of the liquid and vapour phase inside the packing. The aim of this work was to develop three-dimensional CFD models to study the hydrodynamic behaviour on the corrugated sheets of packing. Different approaches are possible to simplify the problem and to extend it for more complex flow scenarios. In this work, three-dimensional CFD simulations were performed to study the complete fluid-dynamic behaviour. This was performed in two steps. As a first step, the developed model was validated with experimental studies using a simplified geometry i.e., an inclined plate. The three-dimensional Volume-of-Fluid (VOF) model was utilized to study the flow behaviour of the gas-liquid countercurrent flow. The influence of the liquid surface tension was taken into consideration using the Continuum Surface Force (CSF) model. The wetting characteristics of liquids with different viscosity (1 and 5 mPas) and contact angle (70° and 7°) were studied for different flow rates. Three different mixtures (water, water-glycerol (45 wt. %) and silicon-oil (DC5)) were considered. Initially, the rivulet width of experiments and simulations were compared and an error of 5 % maximum was determined. The results were also in good agreement with earlier studies. The percentage of wetting due to changes in flow rate, viscosity and contact angle was compared and discussed. For all tested systems, excellent agreement between the experiments and simulation studies was found. In addition, profiles of the velocity in the film at film flow conditions over a smooth inclined plate obtained from simulations were compared with experimental profiles obtained using a μPIV technique. A detailed sensitivity study was also performed in order to understand the changes in the velocity profiles due to small change in liquid flow rate, temperature and inclination angle. As a next step, the developed model was extended to geometries resembling real corrugated sheets of packing used in industrial applications. In earlier numerical studies of structured packing, geometries were simplified to enable easy meshing and faster computation. In this work, the geometries of corrugated sheets of packing were developed without any simplification and the flow behaviour was studied using the model validated in the first step. The flow behaviour on sheets with different geometrical modifications such as smooth and triangular crimp surfaces as well as perforations on the sheets were numerically studied and quantitatively compared with experimental studies for the three different fluid test systems. The agreement between the simulations and experiments was within an acceptable range for all system. The difference in the interfacial area between the corrugated sheets of a packing with and without perforation was analyzed and the prediction ability of different empirical correlations for the interfacial area available in literature was also compared and discussed. Furthermore, the numerical study was extended to understand the influence of the second corrugated sheet. Studying the flow behaviour between two sheets experimentally is very challenging, especially inside opaque packing. The model proved to be a very suitable tool to study the hold-up of the liquid between two sheets, the change in wetting behaviour due to small change in liquid inlet position. The results are also in good agreement with the earlier experimental studies, where researchers measured the liquid hold-up mainly in the region where two corrugated sheets touch each other. The three-dimensional CFD model was validated to study the flow behaviour on corrugated sheets of packing. The results from the simulations agree very well with findings from the experimental studies in terms of wetting and hold-up.

The scope of the present thesis is the development of a Computational Fluid Dynamics model to describe the multiphase flow inside a structured packing absorber for postcombustion carbon capture. The work focuses mainly on two flow characteristics: the interface tracking and the reactive mass transfer between the gas and the liquid. The interface tracking brings the possibility of studying the liquid maldistribution phenomenon, which strongly affects the mass transfer performance. The development of a user-defined function to account for the reactive mass transfer between phases constitutes the second major concept considered in this thesis. Numerical models found in the literature are divided into three scales due to the current computational capacity: small-, meso- and large-scale. Small-scale has usually dealt with interface tracking in 2D computational domains. Meso-scale has usually been considered to assess the dry pressure drop performance of the packing (considering only the gas phase). Large-scale studies the liquid distribution over the whole column assuming that the structured packing behaves as a porous medium. This thesis focuses on small- and meso-scale. The novelty of this work lies in expanding the capabilities of the aforementioned scales. At small-scale, the interfacial tracking is implemented in a 3D domain, instead of 2D. The user-defined function that describes the reactive mass transfer of CO2 into the aqueous MEA solution is also included to assess the influence of the liquid maldistribution on the mass transfer performance. At the meso-scale, the Volume of Fluid method for interface tracking is included (instead of only the gas phase) to describe flow characteristics such as the liquid hold-up, the interfacial area and the mass transfer. At the theoretical level, this model presents the particularity of including both a mass and a momentum source term in the conservation equations. A comprehensive mathematical development shows the influence of the mass source terms on the momentum equation.

Une colonne de séparation d'air réalise un écoulement liquide-gaz à contre courant dans une structure complexe, le garnissage. Au sein de ce garnissage, l'écoulement du liquide est du type film drainé par gravité, alors que l'écoulement du gaz est turbulent. La fonction de ces contacteurs est de développer une surface d'échange interfaciale aussi grande que possible pour favoriser le transfert d'un composé chimique de la phase liquide vers la phase vapeur (et inversement) tout en offrant des pertes de charge raisonnables. Ces dispositifs sont constitués par l'assemblage de plaques métalliques ondulées, avec ou sans perforations, où deux plaques adjacentes sont respectivement inclinées d'un angle et son opposé par rapport à l'axe de la colonne. Ce type de contacteur peut être considéré comme un milieu poreux bi-structuré avec un taux de porosité élevé. Les écoulements peuvent être décrits à deux échelles : une échelle du pore et une échelle macroscopique. A cause de cette double structuration, la modélisation macroscopique des écoulements dans ce type de structure reste un problème difficile. En particulier, les mécanismes macroscopiques qui entraînent l'étalement d'un jet dans les garnissages sont incompris. Par ailleurs, une difficulté de modélisation supplémentaire est due aux effets liés à la turbulence. Au cours de cette thèse, nous avons développé, à partir d'une méthode de changement d'échelle, un modèle complet pour simuler les écoulements et le transfert de matière dans les colonnes équipées de garnissages structurés. Notre étude se focalise sur les trois points suivants. Premièrement, nous avons obtenu, à l'aide d'une prise de moyenne volumique, une loi de Darcy-Forchheimer qui inclue les effets de la turbulence. Ensuite, pour modéliser la dispersion radiale du liquide dans la colonne, nous avons trouvé pratique de séparer la phase liquide en deux films distincts, qui s'écoulent sur chaque plaque ondulée selon des directions préférentielles différentes. Ces phases fictives ne sont pas indépendantes puisque de la matière peut passer de l'une à l'autre au niveau des points de contact entre les feuilles ondulées. Finalement, nous avons proposé un modèle macroscopique pour simuler le transport d'espèces chimiques dans un système diphasique, multiconstituants. Tous les paramètres effectifs qui apparaissent dans ce modèle sont évalués à partir de solutions analytiques ou numériques de l'écoulement à la petite échelle. Les résultats de simulation ont été comparés avec succès à des mesures expérimentales obtenues en laboratoire ou sur pilote industriel. / Structured packings play a large role in chemical engineering processes involving gasliquid separation such as air distillation unit or CO2 absorption columns. Such structures maximize the exchange surface between gas and liquid while pressure drops remain low enough. Generally, the columns are operated in the counter-current flow mode : a liquid gravity film is sheared by the turbulent flow of a gas phase. The packings are made of an assembly of corrugated sheets where two adjacent sheets are respectively inclined by an angle and the opposite of this angle from the vertical axis. We can apprehend such a device as a bi-structured porous medium with high porosity defining two scales of description : a pore-scale and a macro-scale assimilated to the packing scale. Due to this peculiar structured geometry, the flow modeling from a macroscopic point of view, remains a challenging problem that has to be overcome to design enhanced devices. In particular, the macroscopic phenomena that leads to the spreading of a liquid point source at the top of a packing are still unknown, and the classical two-phase flow models in porous media failed to properly catch the liquid distribution within the column. Moreover, turbulence effects lead to additional difficulties. We developed a comprehensive mathematical model based on a multi-scale analysis to simulate gas-liquid flow through the distillation columns. We investigate three main points. First, we derived a Darcy-Forchheimer law that includes turbulence effects using the method of volume averaging. Then, to model the liquid spreading, we found convenient to split the liquid phase into two fictitious phases flowing along each sheet with a preferential direction. Moreover, these phases are not (except perhaps at very low saturation) completely independent since adjacent sheets are in contact and the liquid can flow from one sheet to the other. Finally, we proposed a macro-scale dispersion model to simulate two-phase, multicomponent transport in structured packing. All the effective properties that appear in this model are evaluated from either simulations or analytical solutions of the flow at the pore-scale. Simulation results have been successfully compared to laboratory-scale experiments and industrial-scale measurements.

The effective mass transfer area (a [subscript e]), liquid film mass transfer coefficient (k [subscript L]), and gas film mass transfer coefficient (k [subscript G]) of eleven structured packings and three random packings were measured consistently in a 0.428 m packed column. Absorption of CO₂ with 0.1 gmol/L NaOH with 3.05 m packing was used to measure a [subscript e], while air stripping of toluene from water with 1.83 m packing was used to measure k [subscript L], and absorption of SO₂ with 0.1 gmol/L NaOH with 0.51 m packing was used to measure k [subscript G]. The experiments were conducted with liquid load changing from 2.5 to 75 m³/(m²*h) and gas flow rate from 0.6 to 2.3 m/s. Packings with surface area from 125 to 500 m²/m³ and corrugation angle from 45 to 70 degree were tested to explore the effect of packing geometries on mass transfer. The effective area increases with packing surface area and liquid flow rate, and is independent of gas velocity. The packing corrugation angle has an insignificant effect on mass transfer area. The ratio of effective area to surface area decreases as surface area increases due to the limit of packing wettability. A correlation has been developed to predict the mass transfer area with an average deviation of 11%. [Mathematical equation]. The liquid film mass transfer coefficient is only a function of liquid velocity with a power of 0.74, while the gas film mass transfer coefficient is only a function of gas velocity with a power of 0.58. Both k [subscript L] and k [subscript G] increase with packing surface area, and decrease with corrugation angle. A new concept, Mixing Point Density, was introduced to account for effect of the packing geometry on k[subscript L] and k [subscript G]. Mixing Point Density represents the frequency at which liquid film is refreshed and gas is mixed. The mixing point density can be calculated by either packing characteristic length or by surface area and corrugation angle: [mathematical equation]. The dimensionless k [subscript L] and k [subscript G] models can then be developed based on the effects of liquid/gas velocity, mixing point density, packing surface area: [mathematical equation] [mathematical equation]. Mi is the dimensionless form of Mixing Point Density (M), which is M divided by a [subscript P]³. Because Mi is only a function of corrugation angle (θ), it is a convenient transformation to represent the effect of θ on mass transfer parameters. An economic analysis of the absorber was conducted for a 250 MW coal-fired power plant. The optimum operating condition is between 50 to 80 % of flooding, and the optimum design is to use packing with 200 to 250 m²/m³ surface area and high corrugation angle (60 to 70 degree). The minimum total cost ranges from $4.04 to $5.83 per tonne CO₂ removed with 8 m PZ.

Research of conditions when sulphure dioxide is absorbed from flue gas, is the main objective of this master's thesis. Efficiency of this flue gas cleaning was measured on an experimental device, which consist of an absorption column filled with structured packing’s, and another component so-called "O-element". Further, a possibility of upgrade this "O-element" to tread more flue gas is presented.

The main theme of this master’s thesis is to acquaint with flue gas cleaning problems. Also modernization of experimental device called “O-element”, installed in laboratory at Institute of process and environmental engineering, is involved in this master’s thesis. “Oelement is used for cleaning acid elements from flue gas. Further there are results of experimental measuring on this device, and their estimation.

The key to success in separation of liquid mixtures is the efficient creation and utilization of vapour-liquid contact area. By packing the column with gas-liquid contact devices such as structured packing, the vapour-liquid contact area can be increased. However, the efficiency of these packed columns depends strongly on the local flow behaviour of the liquid and vapour phase inside the packing. The aim of this work was to develop three-dimensional CFD models to study the hydrodynamic behaviour on the corrugated sheets of packing. Different approaches are possible to simplify the problem and to extend it for more complex flow scenarios. In this work, three-dimensional CFD simulations were performed to study the complete fluid-dynamic behaviour. This was performed in two steps. As a first step, the developed model was validated with experimental studies using a simplified geometry i.e., an inclined plate. The three-dimensional Volume-of-Fluid (VOF) model was utilized to study the flow behaviour of the gas-liquid countercurrent flow. The influence of the liquid surface tension was taken into consideration using the Continuum Surface Force (CSF) model. The wetting characteristics of liquids with different viscosity (1 and 5 mPas) and contact angle (70° and 7°) were studied for different flow rates. Three different mixtures (water, water-glycerol (45 wt. %) and silicon-oil (DC5)) were considered. Initially, the rivulet width of experiments and simulations were compared and an error of 5 % maximum was determined. The results were also in good agreement with earlier studies. The percentage of wetting due to changes in flow rate, viscosity and contact angle was compared and discussed. For all tested systems, excellent agreement between the experiments and simulation studies was found. In addition, profiles of the velocity in the film at film flow conditions over a smooth inclined plate obtained from simulations were compared with experimental profiles obtained using a μPIV technique. A detailed sensitivity study was also performed in order to understand the changes in the velocity profiles due to small change in liquid flow rate, temperature and inclination angle. As a next step, the developed model was extended to geometries resembling real corrugated sheets of packing used in industrial applications. In earlier numerical studies of structured packing, geometries were simplified to enable easy meshing and faster computation. In this work, the geometries of corrugated sheets of packing were developed without any simplification and the flow behaviour was studied using the model validated in the first step. The flow behaviour on sheets with different geometrical modifications such as smooth and triangular crimp surfaces as well as perforations on the sheets were numerically studied and quantitatively compared with experimental studies for the three different fluid test systems. The agreement between the simulations and experiments was within an acceptable range for all system. The difference in the interfacial area between the corrugated sheets of a packing with and without perforation was analyzed and the prediction ability of different empirical correlations for the interfacial area available in literature was also compared and discussed. Furthermore, the numerical study was extended to understand the influence of the second corrugated sheet. Studying the flow behaviour between two sheets experimentally is very challenging, especially inside opaque packing. The model proved to be a very suitable tool to study the hold-up of the liquid between two sheets, the change in wetting behaviour due to small change in liquid inlet position. The results are also in good agreement with the earlier experimental studies, where researchers measured the liquid hold-up mainly in the region where two corrugated sheets touch each other. The three-dimensional CFD model was validated to study the flow behaviour on corrugated sheets of packing. The results from the simulations agree very well with findings from the experimental studies in terms of wetting and hold-up.