A significant percentage of computational time in compositional simulations is spent performing flash calculations to determine the equilibrium compositions of hydrocarbon phases in situ. Flash calculations must be done at each time step for each grid block; thus billions of such calculations are possible. It would be very important to reduce the computational time of flash calculations significantly so that more grid blocks or components may be used. In this dissertation, three different methods are developed that yield fast, robust and accurate phase behavior calculations useful for compositional simulation and other applications. The first approach is to express the mixing rule in equations-of-state (EOS) so that a flash calculation is at most a function of six variables, often referred to as reduced parameters, regardless of the number of pseudocomponents. This is done without sacrificing accuracy and with improved robustness compared with the conventional method. This approach is extended for flash calculations with three or more phases. The reduced method is also derived for use in stability analysis, yielding significant speedup. The second approach improves flash calculations when K-values are assumed constant. We developed a new continuous objective function with improved linearity and specified a small window in which the equilibrium compositions must lie. The calculation speed and robustness of the constant K-value flash are significantly improved. This new approach replaces the Rachford-Rice procedure that is embedded in the conventional flash calculations. In the last approach, a limited compositional model for ternary systems is developed using a novel transformation method. In this method, all tie lines in ternary systems are first transformed to a new compositional space where all tie lines are made parallel. The binodal curves in the transformed space are regressed with any accurate function. Equilibrium phase behavior calculations are then done in this transformed space non-iteratively. The compositions in the transformed space are translated back to the actual compositional space. The new method is very fast and robust because no iteration is required and thus always converges even at the critical point because it is a direct method. The implementation of some of these approaches into compositional simulators, for example UTCOMP or GPAS, shows that they are faster than conventional flash calculations, without sacrificing simulation accuracy. For example, the implementation of the transformation method into UTCOMP shows that the new method is more than ten times faster than conventional flash calculations.

An extension of the current acetone Planar Laser-Induced Fluorescence (PLIF) was formulated for mixing studies of fluids at subcritical and supercritical conditions. The new technique, called Planar Laser-Induced Fluorescence and Phosphorescence (PLIFP), employs the difference in the mass diffusivity of the denser (liquid) to the less dense (vapor/supercritical fluid) to delineate the interface where a phase change occurs. The vapor/supercritical acetone fluorescence signal is utilized to measure of the acetone vapor density, the mixture fractions and liquid acetone phosphorescence signal to determine the location of the phase interface. The application of the technique requires the photophysical properties of vapor and liquid acetone to be known. Therefore, a series of controlled experiments were done to determine their photophysics at elevated temperatures and pressures up to T/TC = 1.2 and p/pC =1.25. The demonstration of the techniques shows it was able to provide quantitative measurements of acetone number density and the overall mixture fraction within the test chamber. Also, the size and mass of droplets that have broken off from the main jet were determined as well, though the ability is limited to small droplets (d~100 μm). In addition, the technique was able to delineate the low diffusivity (subcritical)/high diffusivity (supercritical) interface very well.

The high temperature BCC phase (b) of titanium undergoes a martensitic transformation to HCP phase (a) upon cooling, but can be stabilized at room temperature by alloying with BCC transition metals such as Mo. There exists a metastable composition range within which the alloyed b phase separates into a + b upon equilibrium cooling but not when rapidly quenched. Compositional partitioning of the stabilizing element in as-quenched b microstructure creates nanoscale precipitates of a new simple hexagonal w phase, which considerably reduces ductility. These phase transformation reactions have been extensively studied experimentally, yet several significant questions remain: (i) The mechanism by which the alloying element stabilizes the b phase, thwarts its transformation to w, and how these processes vary as a function of the concentration of the stabilizing element is unclear. (ii) What is the atomistic mechanism responsible for the non-Arrhenius, anomalous diffusion widely observed in experiments, and how does it extend to low temperatures? How does the concentration of the stabilizing elements alter this behavior? There are many other w forming alloys that such exhibit anomalous diffusion behavior. (iii) A lack of clarity remains on whether w can transform to a -phase in the crystal bulk or if it occurs only at high-energy regions such as grain boundaries. Furthermore, what is the nature of the a phase embryo? (iv) Although previous computational results discovered a new wa transformation mechanism in pure Ti with activation energy lower than the classical Silcock pathway, it is at odds with the a / b / w orientation relationship seen in experiments. First principles calculations based on density functional theory provide an accurate approach to study such nanoscale behavior with full atomistic resolution, allowing investigation of the complex structural and chemical effects inherent in the alloyed state. In the present work, a model Ti-Mo system is investigated to resolve these fundamental questions. Particular attention is paid to how Mo- (i) influences the bonding in Ti, (ii) distorts the local structure in the Ti lattice, (iii) impacts the point and interfacial defect formation and migration energies, and (iv) affects the mechanism and energetics of b w and wa transformations. Our results are correlated with appropriate experimental results of our collaborators and those in open literature. The modification of Ti bonding by Mo solutes and the attendant distortion of the lattice hold the key to answering the diverse questions listed above. The solutes enhance electron charge density in the <111> directions and, consequently, stiffen the lattice against the displacements necessary for b w transformation. However, Ti atoms uncoordinated by Mo remain relatively mobile, and locally displace towards w lattice positions. This effect was further studied in a metastable Ti-8.3 at.% Mo system with an alternate cell geometry which allows for either b w or $\betaa transformation, and it was found that after minimization Ti atoms possessed either a or w coordination environments. The creation of this microstructure is attributed to both the disruption of uniform b w transformation by the Mo atoms and the overlap of Ti-Mo bond contractions facilitating atomic displacements to the relatively stable a or w structures in Mo-free regions. The vacancy migration behavior in such a microstructure was then explored. Additionally, several minimized configurations were created with planar interfaces between Mo-stabilized b region and its adjacent a- or w- phases, and it was found that the positioning of Mo at the interface strongly dictates the structure of the adjacent Mo depleted region.

Vlasov-Maxwell equilibria are characterised by the self-consistent descriptions of the steady-states of collisionless plasmas in particle phase-space, and balanced macroscopic forces. We study the theory of Vlasov-Maxwell equilibria in one spatial dimension, as well as its application to current sheet and flux tube models. The ‘inverse problem' is that of determining a Vlasov-Maxwell equilibrium distribution function self-consistent with a given magnetic field. We develop the theory of inversion using expansions in Hermite polynomial functions of the canonical momenta. Sufficient conditions for the convergence of a Hermite expansion are found, given a pressure tensor. For large classes of DFs, we prove that non-negativity of the distribution function is contingent on the magnetisation of the plasma, and make conjectures for all classes. The inverse problem is considered for nonlinear ‘force-free Harris sheets'. By applying the Hermite method, we construct new models that can describe sub-unity values of the plasma beta (βpl) for the first time. Whilst analytical convergence is proven for all βpl, numerical convergence is attained for βpl = 0.85, and then βpl = 0.05 after a ‘re-gauging' process. We consider the properties that a pressure tensor must satisfy to be consistent with ‘asymmetric Harris sheets', and construct new examples. It is possible to analytically solve the inverse problem in some cases, but others must be tackled numerically. We present new exact Vlasov-Maxwell equilibria for asymmetric current sheets, which can be written as a sum of shifted Maxwellian distributions. This is ideal for implementations in particle-in-cell simulations. We study the correspondence between the microscopic and macroscopic descriptions of equilibrium in cylindrical geometry, and then attempt to find Vlasov-Maxwell equilibria for the nonlinear force-free ‘Gold-Hoyle' model. However, it is necessary to include a background field, which can be arbitrarily weak if desired. The equilibrium can be electrically non-neutral, depending on the bulk flows.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O objetivo do presente trabalho é apresentar um estudo sobre o equilíbrio de fases sólido-líquido-vapor para hidratos de metano. A análise do equilíbrio trifásico sólido-líquido-vapor tem encontrado diversas aplicações para sistemas hidrocarboneto-água, uma vez que permite, por exemplo, a determinação da região de estabilidade de hidratos de metano e hidratos de gás natural. Inicialmente foi feita uma pesquisa sobre o estado da arte no que diz respeito ao comportamento termodinâmico e equilíbrio de fases de hidratos. Foram implementados os modelos apresentados por Ballard (2002) e Zhang et al. (2005). A proposta de Zhang et al. (2005) é aplicável para equilíbrios de fases a temperaturas abaixo de 300 K. Sua abordagem combinou a teoria de van der Waals e Platteeuw para a fase hidrato com a equação do estado de Peng-Robinson (1976) modificada por Stryjek e Vera (1986) para ambas as fases fluidas (fase vapor e fase aquosa). A abordagem de Ballard (2000) considerou a distorção do hidrato do seu estado padrão, o que fornece uma exata composição do hidrato e melhora a previsão da formação dos hidratos a altas pressões. Ao esclarecer a mudança de volume no hidrato, o raio da gaiola do hidrato é uma função do seu volume. Com isso, Ballard propôs uma nova abordagem considerando tal variação de volume e gerou um equilíbrio de fases em uma rotina de flash multifásico através da minimização da energia livre de Gibbs. Assim, o presente trabalho apresenta as abordagens de Zhang et al. (2005) e Ballard (2002) para o comportamento termodinâmico de hidratos e faz uma análise e comparação entre eles. Para resolver o problema do flash computacionalmente, foi utilizada a ferramenta lsqnonlin (built-in do software MATLAB). O lsqnonlin é um algoritmo baseado no método de Levenberg-Marquadt. / The objective of the present work is to present a study of solid-vapor-liquid three-phase equilibrium for methane hydrates. The analysis of three-phase equilibrium has several applications for water-hydrocarbon systems, since it permits, for example, determination of stability region for methane hydrates and natural gas hydrates. We have started seeking in literature about the state-of-art for thermodynamic behaviour and phase equilibrium for hydrates. And then the models proposed by Ballard (2002) and Zhang et al. (2005) were implemented. Zhang et al. (2005) have proposed a phase equilibrium for single-guest gas hydrates at temperatures below 300 K. Their approach has combined the van der WaalsPlatteeuw theory for the hydrate phase and the PengRobinson equation of state for both fluid phases (vapor and aqueous phase) (1976) modified by Stryjek and Vera (1986). Ballards (2000) approach has allowed the hydrate distortion from its standard state and has gave a more accurate composition of the hydrate and has improved hydrate formation predictions at high pressures. As a direct result of accounting for a changing hydrate volume, the cage radii were functions of the hydrate volume. Thus, Ballard have proposed the hydrate phase equilibrium by Gibbs energy minimization in a multi-phase flash routine. Thus, this work presents the Zhang et al. (2005) and Ballards (2002) approaches for hydrate thermodynamic behavior and makes an analysis and comparison of them. To compute the flash problem, we use the tool lsqnonlin (built-in of MATLAB software). The algorithm lsqnonlin is based on the Levenberg-Marquadt method.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O objetivo do presente trabalho é apresentar um estudo sobre o equilíbrio de fases sólido-líquido-vapor para hidratos de metano. A análise do equilíbrio trifásico sólido-líquido-vapor tem encontrado diversas aplicações para sistemas hidrocarboneto-água, uma vez que permite, por exemplo, a determinação da região de estabilidade de hidratos de metano e hidratos de gás natural. Inicialmente foi feita uma pesquisa sobre o estado da arte no que diz respeito ao comportamento termodinâmico e equilíbrio de fases de hidratos. Foram implementados os modelos apresentados por Ballard (2002) e Zhang et al. (2005). A proposta de Zhang et al. (2005) é aplicável para equilíbrios de fases a temperaturas abaixo de 300 K. Sua abordagem combinou a teoria de van der Waals e Platteeuw para a fase hidrato com a equação do estado de Peng-Robinson (1976) modificada por Stryjek e Vera (1986) para ambas as fases fluidas (fase vapor e fase aquosa). A abordagem de Ballard (2000) considerou a distorção do hidrato do seu estado padrão, o que fornece uma exata composição do hidrato e melhora a previsão da formação dos hidratos a altas pressões. Ao esclarecer a mudança de volume no hidrato, o raio da gaiola do hidrato é uma função do seu volume. Com isso, Ballard propôs uma nova abordagem considerando tal variação de volume e gerou um equilíbrio de fases em uma rotina de flash multifásico através da minimização da energia livre de Gibbs. Assim, o presente trabalho apresenta as abordagens de Zhang et al. (2005) e Ballard (2002) para o comportamento termodinâmico de hidratos e faz uma análise e comparação entre eles. Para resolver o problema do flash computacionalmente, foi utilizada a ferramenta lsqnonlin (built-in do software MATLAB). O lsqnonlin é um algoritmo baseado no método de Levenberg-Marquadt. / The objective of the present work is to present a study of solid-vapor-liquid three-phase equilibrium for methane hydrates. The analysis of three-phase equilibrium has several applications for water-hydrocarbon systems, since it permits, for example, determination of stability region for methane hydrates and natural gas hydrates. We have started seeking in literature about the state-of-art for thermodynamic behaviour and phase equilibrium for hydrates. And then the models proposed by Ballard (2002) and Zhang et al. (2005) were implemented. Zhang et al. (2005) have proposed a phase equilibrium for single-guest gas hydrates at temperatures below 300 K. Their approach has combined the van der WaalsPlatteeuw theory for the hydrate phase and the PengRobinson equation of state for both fluid phases (vapor and aqueous phase) (1976) modified by Stryjek and Vera (1986). Ballards (2000) approach has allowed the hydrate distortion from its standard state and has gave a more accurate composition of the hydrate and has improved hydrate formation predictions at high pressures. As a direct result of accounting for a changing hydrate volume, the cage radii were functions of the hydrate volume. Thus, Ballard have proposed the hydrate phase equilibrium by Gibbs energy minimization in a multi-phase flash routine. Thus, this work presents the Zhang et al. (2005) and Ballards (2002) approaches for hydrate thermodynamic behavior and makes an analysis and comparison of them. To compute the flash problem, we use the tool lsqnonlin (built-in of MATLAB software). The algorithm lsqnonlin is based on the Levenberg-Marquadt method.

Le bicarbonate de soude raffiné, produit industriellement par la société Solvay, est fabriqué dans des colonnes à bulles de grande taille, appelées les colonnes BIR.<p>Dans ces colonnes, une phase gazeuse contenant un mélange d’air et dioxyde de carbone (CO2) est dispersée sous forme de bulles dans une solution aqueuse de carbonate et de bicarbonate de sodium (respectivement Na2CO3 et NaHCO3). Cette dispersion donne lieu à un transfert de CO2 des bulles vers la phase liquide. Au sein des colonnes, la phase gazeuse se répartit dans deux populations de bulles :des petites bulles (diamètre de quelques mm) et des grandes bulles (diamètre de quelques cm). Le transfert bulle-liquide de CO2 est couplé à des réactions chimiques prenant place en phase liquide, qui conduisent à la conversion du Na2CO3 en NaHCO3. Une fois la concentration de saturation dépassée le NaHCO3 précipite sous forme de cristaux et un mélange liquide-solide est recueilli à la sortie de ces colonnes.<p>Ce travail, réalisé en collaboration avec la société Solvay, porte sur l’étude et la modélisation mathématique des phénomènes de transfert de matière entre phases, couplés à des réactions chimiques, prenant place au sein d’une colonne BIR. L’association d’études sur des colonnes à bulles à l’échelle industrielle ou réduite (pilote) et d’études plus fondamentales sur des dispositifs de laboratoire permet de développer une meilleure compréhension du fonctionnement des colonnes BIR et d’en construire un modèle mathématique détaillé.<p>L’objectif appliqué de ce travail est la mise au point d’un modèle mathématique complet et opérationnel d’une colonne BIR. Cet objectif est supporté par trois blocs de travail, dans lesquels différents outils sont développés et exploités.<p><p>Le premier bloc est consacré à la modélisation mathématique du transfert bulle-liquide de CO2 dans une solution aqueuse de NaHCO3 et de Na2CO3. Ce transfert est couplé à des réactions chimiques en phase liquide qui influencent sa vitesse. Dans un premier temps, des modèles sont développés selon des approches unidimensionnelles classiquement rencontrées dans la littérature. Ces approches passent par une idéalisation de l’écoulement du liquide autour des bulles. Une expression simplifiée de la vitesse du transfert bulle-liquide de CO2, est également développée et validée pour le modèle de colonne BIR.<p>Dans un second temps, une modélisation complète des phénomènes de transport (convection et diffusion), couplés à des réactions chimiques, est réalisée en suivant une approche bidimensionnelle axisymétrique. L’influence de la vitesse de réactions sur la vitesse de transfert est étudiée et les résultats des deux approches sont également comparés.<p><p>Le deuxième bloc est consacré à l’étude expérimentale du transfert gaz-liquide de CO2 dans des solutions aqueuses de NaHCO3 et de Na2CO3. A cette fin, un dispositif expérimental est développé et présenté. Du CO2 est mis en contact avec des solutions aqueuses de NaHCO3 et de Na2CO3 dans une cellule transparente. Les phénomènes provoqués en phase liquide par le transfert de CO2 sont observés à l’aide d’un interféromètre de Mach-Zehnder.<p>Les résultats expérimentaux sont comparés à des résultats de simulation obtenus avec un des modèles unidimensionnels développés dans le premier bloc. De cette comparaison, il apparaît qu’une mauvaise estimation de la valeur de certains paramètres physico-chimiques apparaissant dans les équations de ce modèle conduit à des écarts significatifs entre les grandeurs observées expérimentalement et les grandeurs estimées par simulation des équations du modèle.<p>C’est pourquoi une méthode d’estimation paramétrique est également développée afin d’identifier les valeurs numériques de ces paramètres physico-chimiques sur base des résultats expérimentaux. Ces dernières sont également discutées.<p><p>Dans le troisième bloc, nous apportons une contribution à l’étude des cinétiques de précipitation du NaHCO3 dans un cristallisoir à cuve agitée. Cette partie du travail est réalisée en collaboration avec Vanessa Gutierrez (du service Matières et Matériaux de l’ULB).<p>Nous contribuons à cette étude par le développement de trois outils :une table de calcul Excel permettant de synthétiser les résultats expérimentaux, un ensemble de simulations de l’écoulement au sein du cristallisoir par mécanique des fluides numérique et une nouvelle méthode d’extraction des cinétiques de précipitation du NaHCO3 à partir des résultats expérimentaux. Ces trois outils sont également utilisés de façon combinée pour estimer les influences de la fraction massique de solide et de l’agitation sur la cinétique de germination secondaire du NaHCO3.<p><p>Enfin, la synthèse de l’ensemble des résultats de ces études est réalisée. Le résultat final est le développement d’un modèle mathématique complet et opérationnel des colonnes BIR. Ce modèle est développé en suivant l’approche de modélisation en compartiments, développée au cours du travail de Benoît Haut. Ce modèle synthétise les trois blocs d’études réalisées dans ce travail, ainsi que les travaux d’Aurélie Larcy (du service Transferts, Interfaces et Procédés de l’ULB) et de Vanessa Gutierrez. Les équations modélisant les différents phénomènes sont présentées, ainsi que la méthode utilisée pour résoudre ces équations. Des simulations des équations du modèle sont réalisées et discutées. Les résultats de simulation sont également comparés à des mesures effectuées sur une colonne BIR. Un accord raisonnable est observé.<p>A l’issue de ce travail, nous disposons donc d’un modèle opérationnel de colonne BIR. Bien que ce modèle doive encore être optimisé et validé, il peut déjà être utilisé pour étudier l’effet des caractéristiques géométriques des colonnes BIR et des conditions appliquées à ces colonnes sur le comportement des simulations des équations du modèle et pour identifier des tendances.<p>//<p>The refined sodium bicarbonate is produced by the Solvay company using large size bubble columns, called the BIR columns.<p>In these columns, a gaseous phase containing an air-carbon dioxyde mixture (CO2) is dispersed under the form of bubbles in an aqueous solution of sodium carbonate and sodium bicarbonate (Na2CO3 and NaHCO3, respectively). This dispersion leads to a CO2 transfer from the bubbles to the liquid phase. Inside these columns, the gaseous phase is distributed in two bubbles populations :small bubbles (a few mm of diameter) and large bubbles (a few cm of diameter).<p>The bubble-liquid CO2 transfer is coupled with chemical reactions taking places in the liquid phase that leads to the conversion of Na2CO3 to NaHCO3. When the solution is supersaturated in NaHCO3, the NaHCO3 precipitates under the form of crystals and a liquid-solid mixture is extracted at the outlet of the BIR columns.<p>This work, realized in collaboration with Solvay, aims to study and to model mathematically the mass transport phenomena between the phases, coupled with chemical reactions, taking places inside a BIR column. Study of bubble columns at the industrial and the pilot scale is combined to a more fundamental study at laboratory scale to improve the understanding of the BIR columns functioning and to develop a detailed mathematical modeling.<p>The applied objective of this work is to develop a complete and operational mathematical modeling of a BIR column. This objective is supported by three blocks of work. In each block, several tools are developed and used.<p><p>The first block is devoted to the mathematical modeling of the bubble-liquid CO2 transfer in an NaHCO3 and Na2CO3 aqueous solution. This transfer is coupled with chemical reactions in liquid phase, which affect the transfer rate.<p>In a first time, mathematical models are developed following the classical one-dimensional approaches of the literature. These approaches idealize the liquid flow around the bubbles. A simplified expression of the bubble-liquid CO2 transfer rate is equally developed and validated for the BIR column model.<p>In a second time, a complete modeling of the transport phenomena (convection and diffusion) coupled with chemical reactions is developed, following an axisymmetrical twodimensional approach. The chemical reaction rate influence on the bubble-liquid transfer rate is studied and the results of the two approaches are then compared.<p><p>The second block is devoted to the experimental study of the gas-liquid CO2 transfer to NaHCO3 and Na2CO3 aqueous solutions. An experimental set-up is developed and presented. CO2 is put in contact with NaHCO3 and Na2CO3 aqueous solutions in a transparent cell. The phenomena induced in liquid phase by the CO2 transfer are observed using a Mach-Zehnder interferometer.<p>The experimental results are compared to simulation results that are obtained using one of the one-dimensional model developed in the first block. From this comparison, it appears that a wrong estimation of some physico-chemical parameter values leads to significative differences between the experimentally observed quantities and those estimated by simulation of the model equations. Therefore, a parametric estimation method is developed in order to estimate those parameters numerical values from the experimental results. The found values are then discussed.<p><p>In the third block is presented a contribution to the NaHCO3 precipitation kinetic study in a stirred-tank crystallizer. This part of the work is realized in collaboration with Vanessa Gutierrez (Chemicals and Materials Department of ULB).<p>Three tools are developed :tables in Excel sheet to synthetize the experimental results, a set of simulations of the flow inside the crystallizer by Computational Fluid Dynamic (CFD) and a new method to extract the NaHCO3 precipitation kinetics from the experimental measurements. These three tools are combined to estimate the influences of the solid mass fraction and the flow on the NaHCO3 secondary nucleation rate.<p><p>Finally, the synthesis of all these results is realized. The final result is the development of a complete and operational mathematical model of BIR columns. This model is developed following the compartmental modeling approach, developed in the PhD thesis of Benoît Haut. This model synthetizes the three block of study realized in this work and the studies of Aurélie Larcy (Transfers, Interfaces and Processes Department of ULB) and those of Vanessa Gutierrez. The equations modeling the phenomena taking place in a BIR column are presented as the used method to solve these equations. The equations of the model are simulated and the results are discussed. The results are equally compared to experimental measurement realized on a BIR column. A reasonable agreement is observed.<p>At the end of this work, an operational model of a BIR column is thus developed. Although this model have to be optimized and validated, it can already be used to study the influences of the geometrical characteristics of the BIR columns and of the conditions applied to these columns on the behaviour of the model equation simulations and to identity tendencies. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished