Nous étudions expérimentalement l'effondrement d'une colonne granulaire dans un liquide visqueux. Contrairement au cas sec, la morphologie des dépôts n'est principalement plus contrôlée par le rapport d'aspect initial du tas mais par la fraction volumique initiale de la masse granulaire. Deux régimes différents sont identifiés selon l'empilement initial. L'empilement lâche donne lieu à des dépôts minces et longs et la dynamique est rapide. Une surpression du liquide est mesurée sous de la colonne. Pour l'empilement dense, l'étalement final est deux fois moindre, le mouvement est lent et une dépression interstitielle est mesurée. Ces observations suggèrent que la dynamique de l'effondrement granulaire dans un fluide est fortement affectée par le comportement de la dilatance du milieu granulaire.Nous développons ensuite un modèle théorique basé sur des équations diphasiques moyennées dans l’épaisseur prenant en compte les mécanismes de dilatance. L’étude dimensionnelle de notre modèle permet de montrer que l’effondrement d’une colonne est contrôlé par trois paramètres sans dimension : le rapport d’aspect de la colonne, la fraction volumique initiale, et le nombre de grains dans l’épaisseur. On montre également que le temps caractéristique met en compétition le frottement visqueux et la gravité.De ce modèle, nous développons un algorithme de résolution lagrangien. Cette approche, grossière mais robuste, permet d’implanter s sans trop de difficulté. Le code est validé sur des configurations simples sur plan incliné avant de simuler l’effondrement de colonnes granulaires immergées dans la même gamme de paramètres que nos expériences. / Nous étudions expérimentalement l'effondrement d'une colonne granulaire dans un liquide visqueux. Contrairement au cas sec, la morphologie des dépôts n'est principalement plus contrôlée par le rapport d'aspect initial du tas mais par la fraction volumique initiale de la masse granulaire. Deux régimes différents sont identifiés selon l'empilement initial. L'empilement lâche donne lieu à des dépôts minces et longs et la dynamique est rapide. Une surpression du liquide est mesurée sous de la colonne. Pour l'empilement dense, l'étalement final est deux fois moindre, le mouvement est lent et une dépression interstitielle est mesurée. Ces observations suggèrent que la dynamique de l'effondrement granulaire dans un fluide est fortement affectée par le comportement de la dilatance du milieu granulaire.Nous développons ensuite un modèle théorique basé sur des équations diphasiques moyennées dans l’épaisseur prenant en compte les mécanismes de dilatance. L’étude dimensionnelle de notre modèle permet de montrer que l’effondrement d’une colonne est contrôlé par trois paramètres sans dimension : le rapport d’aspect de la colonne, la fraction volumique initiale, et le nombre de grains dans l’épaisseur. On montre également que le temps caractéristique met en compétition le frottement visqueux et la gravité.De ce modèle, nous développons un algorithme de résolution lagrangien. Cette approche, grossière mais robuste, permet d’implanter s sans trop de difficulté. Le code est validé sur des configurations simples sur plan incliné avant de simuler l’effondrement de colonnes granulaires immergées dans la même gamme de paramètres que nos expériences.

Les métallacarboranes sont des clusters anioniques constitués d'atomes de bore, de carbone et d'hydrogène avec un cation métallique pris en « sandwich » au cœur de ce cluster (Co3+ en général) et hautement stable thermiquement que chimiquement. Ces entités sont représentées trivialement par la lettre grecque « thêta » où leurs pôles sont considérés comme hydrophobes et possèdent une charge négative délocalisée qui peut être contre balancée par un proton acide. Cette forte stabilité ainsi que ces diverses propriétés leur procurent un intérêt particulier dans des applications telles que la co-extraction du Cs et Sr dans les effluents nucléaires ou bien en médecine pour leur aptitude à inhiber la protéase du VIH de par ses liaisons hydrogènes. La chimie du bore étant relativement riche, une synthèse de quelques dérivés a été réalisée en laboratoire. Il est ainsi possible de substituer spécifiquement des atomes d'hydrogène par des atomes d'iode ou de chlore ou bien de changer le métal central par un cation Fe3+ formant respectivement les anions diiodo-COSAN (I2COSAN), dichloro-COSAN (Cl2COSAN) et ferrabisdicarbollide (FESAN). Ces métallacarboranes sont aussi considérés comme une nouvelle classe de tensioactif pour lesquels une première étude de leur auto-assemblage a été réalisée. Les résultats obtenus sur l'un d'entre eux, le cobaltabisdicarbollide ou COSAN, ont montré un effet sur la tension de surface ainsi qu'une structuration spontanée de ces clusters en forme de vésicule (à partir de 0,5 mmol/L) et par interaction Coulombienne, tendent à former des micelles à plus forte concentration (après 15 mmolL). Mais ces derniers offrent davantage de surprise de par la formation de phases lyotropes en solution aqueuse. En effet, ces phases, clairement identifiées par des techniques de diffusion au rayon-X et microscopie, montrent une dépendance en température mais aussi en concentration dont un diagramme de phase a pu être établi pour le I2COSAN en particulier. Cette thèse fait ainsi l'objet de la compréhension des phénomènes contrôlant cette agrégation de ces curieux composés. / Metallacarboranes are anionic clusters composed of boron, carbon and hydrogen with a metallic cation sandwiched at the heart of this cluster (Co3+ in general) and highly stable in thermic and chemical point of view. These entities are trivially represented by the Greek letter "theta" where the poles are considered hydrophobic and have a negative charge delocalized and counter-balanced by an acidic proton. This high stability as well as the various properties gives a particular interest in applications such as the co-extraction of Cs and Sr in nuclear waste or in medicine for their ability to inhibit HIV protease by its hydrogen bonds. The chemistry of boron is relatively rich; a synthesis of some derivatives was performed in laboratory. It is possible to substitute specifically hydrogen by iodine or chlorine atoms or change the metallic heart by other atom as Fe3+, respectively forming diiodo- COSAN (I2COSAN), dichloro-COSAN (Cl2COSAN) and ferrabisdicarbollide (FESAN) anions. Metallacarboranes are also considered as a new class of surfactant where a first study of their self-assembly has been performed. The results of one of them, cobaltabisdicarbollide or COSAN, have shown an effect at the surface tension and theses clusters formed spontaneously vesicles in dilute regime (from 0.5 mmol/L) and by Coulomb interactions, they form micelles at higher concentration (after 15 mmol/L). But they offer more surprise by the formation of lyotropic phases in aqueous solution. Indeed, these phases, clearly identified by X-ray scattering techniques and microscopy, showed temperature and concentration dependence where a phase diagram was established for the I2COSAN in particular. Therefore, the thesis is focused on the understanding of the phenomena controlling the aggregation of these curious compounds.

Les vésicules sont des gouttes immergées dans un fluide externe visqueux, dont le rayon vautquelques dizaines de micromètres et entourées par une membrane imperméable constituée de lipides, dont l’épaisseur est approximativement 4 nm. La membrane d’une vésicule est un systèmeoriginal du point de vue mécanique : celle-ci présente à la fois des propriétés fluides (les lipidespeuvent s’écouler librement le long de la membrane, mais la surface est incompressible locale-ment) et des propriétés solides (la membrane résiste à la flexion). Les propriétés spécifiques de lamembrane rendent ce système à la fois très déformable et très contraint.Ce manuscrit s’intéresse à la modélisation d’une vésicule soumise à des efforts extérieurs d’o-rigine hydrodynamique, dans le régime de Stokes. Une attention particulière est consacrée à lasituation d’une vésicule qui sédimente. Cette situation est étudiée analytiquement dans le régimedes faibles déformations. Il est montré que plusieurs familles de solutions stationnaires non triv-iales existent, grâce aux propriétés spécifiques de la membrane. L’étude de la sédimentation d’unevésicule est poursuivie par le développement d’un code numérique capable de simuler de grandesdéformations. Pour cela, des méthodes numériques originales de calcul de prise en compte de laflexion et de l’incompressibilité surfacique sont développées. Ce code permet d’étudier la forma-tion d’un tube à l’arrière d’une vésicule en sédimentation. Ces tubes sont de fins (rapport d’aspecttypique longueur/rayon ∼ 100) cylindres connectés à la vésicule d’origine. Il est montré que cesformes tubes sont des formes stationnaires. Un modèle théorique est proposé et comparé auxsimulations numériques. Ce modèle met en lumière l’importance particulière de la tension dansces formes. Une modélisation mécanique basée sur un milieu de Cosserat surfacique courbé estégalement présentée, et permet d’identifier la contribution de la flexion au tenseur des contraintes.Cette contribution est un ingrédient indispensable pour comprendre les formes tubes. / Vesicles are drops of radius of a few tens micrometers, bounded by an impermeable lipidmembrane of approximately 4 nm thickness, and embedded in an external viscous fluid. Thevesicle membrane is an original system from the mechanical point of view : it presents bothincompressible fluid properties (the lipids can flow freely along the membrane, but membraneis incompressible locally) and solid properties (the membrane resists to bending). The specificproperties of the membrane make the system very deformable and very constrained at the sametime.This manuscript deals with the modelisation of a vesicle subjected to external stresses of hydrodynamical origin, in the Stokes regime. A particular attention is paid to the situation of asettling vesicle. This situation is studied analytically in the small deformation regime. It is foundthat several families of non-trivial stationnary shapes exist, owing to the specific properties ofthe membrane. The study of a settling vesicle is pursued by the development of a numerical codeable to deal with large deformations. Original numerical methods are developped to deal with thecomputation of the bending and with the surface incompressibility constraint. This code permitsto study the formation of tether at the rear of a settling vesicle. These tethers are thin (typicalaspect ratio : length/radius ∼ 100) cylinders of membrane connected to the original vesicle. Itis shown that these tethered shapes are stationary shapes. A theoretical model is proposed andcompared to numerical simulations. This model shows the particular importance of tension inthese shapes. A mechanical modelling based on a curved Cosserat surface is also presented, andpermits to identify the bending contribution to the stress tensor. This contribution is a salientingredient to understand tethered shapes.

Com base em resultados de DSC, análise estrutural por difração de raios X e observações de textura, analisamos os efeitos da mistura do solvente apolar hexadecano com cristais líquidos formados pelas moléculas banana 1,3-fenilenobis[4-(4-tetradecilfeniliminometil)benzoato] (MB14) e 4-cloro-1,3-fenilenobis[4-(4-tetradecilfeniliminometil) benzoato] (MB14Cl). Propusemos um modelo estrutural para explicar as modificações causadas no arranjo molecular da fase B2 pelo acréscimo gradual do solvente. Observamos a diminuição da temperatura de transição entre esta fase e a fase isotrópica, porém a transição entre a fase B2 e a fase a temperaturas mais baixas não sofre alteração significativa. Para concentrações em massa de hexadecano de 45 % no MB14 e de 55 % no MB14Cl, a fase B2 já não é mais observada. Para o MB14Cl, resultados de difração de raios X revelaram que as moléculas de hexadecano penetram entre as camadas esméticas, aumentando a distância intercamada em torno de 3 Å. Acima de 5 % de concentração do solvente, o aumento da distância intercamada satura e ocorre segregação de fases em escala nanométrica. O comportamento da fase B2 sob a ação de um campo elétrico variável também foi analisado para o MB14 puro. Apresentamos um modelo para a linha de base do sinal de corrente de polarização, que leva em conta a não linearidade da condutividade para valores altos de campo aplicado, devido à movimentação iônica na amostra. Para o cálculo da viscosidade, consideramos a não linearidade da constante dielétrica com o campo aplicado, e adaptamos outro modelo, inicialmente utilizado para cristais líquidos ferroelétricos sob a ação de um campo quadrado, para o caso de um cristal líquido antiferroelétrico sob um campo triangular. Quanto aos dois tipos de arranjo molecular da fase B2, o arranjo homoquiral se mostrou bem mais estável que o racêmico, mesmo sob aplicação de campo triangular, quando este último é inicialmente favorecido. O arranjo racêmico se apresentou mais viscoso que o arranjo homoquiral, contrariando nossas previsões. / Based on DSC results, structural analysis by X-ray diffraction and texture observations, we observed the effects of mixing the nonpolar solvent hexadecane with the banana molecules liquid crystals ,3-phenilenebis[4-(4-tetradecilpheniliminometil)benzoate] (MB14) and 4-chloro-1,3-phenilenebis[4-(4-tetradecilpheniliminometil) benzoate] (MB14Cl). We propose a structural model to explain the changes in the molecular ordering of the B2 phase caused by the gradual increase of the solvent. We observed a decreasing of the transition temperature between B2 and isotropic phases, however the transition between B2 and lower temperature phases did not change significantly. For hexadecane concentrations above 45 wt% in MB14 and 55 wt% in MB14Cl, the B2 phase is no longer present. In MB14Cl, X-ray diffraction results showed that the hexadecane molecules penetrate between the smectic layers, increasing the interlayer spacing by about 3 Å. Above 5 wt% of solvent concentration, the increasing of the interlayer spacing saturates, and a phase segregation in nanometric scale occurs. The behavior of the B2 phase under variable electric field was also analysed for the pure MB14. We present a model for the baseline of the polarization current signal, which considers the non-linearity of the conductivity for high values of applied field, due to the presence of ions in the sample. In order to calculate the viscosity, we considered the non-linearity of the dielectric constant with the applied field, and adapted another model, initially used in ferroelectric liquid crystals under rectangular field, for the case of an antiferroelectric liquid crystal under triangular field. Concerning the two kind of molecular ordering in the B2 phase, the homoquiral ordering proved to be far more stable than the racemic, even under triangular field, when the latest is favored. Our measurements resulted in a racemic ordering more viscous than the homoquiral, going against our predictions.

O presente projeto de trabalho buscou avaliar e estabelecer relações de causa e efeito que justificassem a ocorrência de singularidades no comportamento reológico de formulações complexas contendo elevada carga de sólidos em suspensão incorporada. O mecanismo de estabilização coloidal empregado para preparação das formulações e posterior análise foi o mecanismo de estabilização eletroestérico. O caso particular estudado correspondeu à dispersão de cargas inorgânicas (óxido de alumínio) em meio aquoso auxiliada pela incorporação de conteúdo orgânico com ação dispersante (poliacrilato de amônio) e ligante (emulsão acrílica) respectivamente. A implementação de técnicas analíticas complementares permitiu o estudo detalhado de uma singularidade associada ao comportamento reológico do sistema, exclusivamente evidenciada em função da elevada carga de sólidos, oferecendo, portanto, subsídios para elaboração de um modelo supramolecular fundamentado em possíveis alterações de configuração das cadeias poliméricas para justificativa do fenômeno. A contribuição referente ao pleno entendimento do fenômeno, considerando-se suas causas e efeitos, reside na possibilidade de se estabelecer diretrizes de formulação que propiciem a otimização das quantidades de aditivos incorporados e o respectivo controle do estado de dispersão e estabilidade do sistema. Considerou-se, para efeitos práticos de engenharia de formulação, que o mínimo secundário de viscosidade corresponda ao ponto ótimo de estabilidade do sistema mesmo que a condição de máxima dispersão já tenha sido alcançada para uma concentração de dispersante equivalente ao mínimo primário de viscosidade. A contribuição do efeito estérico associada à reduzida distância de separação entre as partículas (IPS) mostrou-se de vital importância para assegurar a viabilidade de uma dada formulação. A proposta aqui apresentada fundamentou-se no mecanismo de depleção como principal força motriz para desencadear possíveis mudanças de conformação espacial das cadeias poliméricas implicando, portanto, em alterações do raio hidrodinâmico, da fração volumétrica efetiva ocupada e da intensidade de interação (repulsão) entre as partículas. O efeito da dinâmica configuracional sofrida pelo sistema refletiu-se diretamente no comportamento reológico do mesmo. O caráter de inovação do presente trabalho reside na utilização não convencional da técnica de espectroscopia eletroacústica, empregada para avaliação de mudanças do raio hidrodinâmico de partículas em suspensão (sistemas concentrados), integrada com a técnica de turbidimetria dinâmica de varredura para avaliação de desempenho (cinética de sedimentação) de formulações complexas (suspensões). / The present research project pretended to evaluate and establish cause and effect relationships able to justify the singularities evidenced when exploring the rheological behaviour of complex formulations containing elevated solids loading. The stabilization mechanism used for the preparation of the dispersed systems and subsequent analyses corresponded to the electrosteric mechanism. The particular case studied herein corresponded to the dispersion of inorganic colloidal particles (aluminun oxide) in aqueous media assisted by the incorporation of organic components acting respectively as dispersant (ammonium polyacrylate) and binder agent (acrylic emulsion). The implementation of complementary analytical techniques enabled the detailed study of a singularity related to the rheological behaviour of the system that could only be evidenced for high solids loading condition. The results obtained provided the fundamental concepts necessary to propose a supramolecular model based on possible conformational changes associated with the adsorbed polymeric chains and sufficient to justify the phenomenon. Considering the practical aspects related to formulation engineering, it was determined that the secondary minimum viscosity corresponds to the optimum stablity condition although the maximum dispersion condition was achieved for dispersant concentrations equal to the primary minimum viscosity. The contribution of the steric effect associated with the reduced interparticle separation distance (IPS) showed itself to be of vital importance to ensure the viability of a stable formulation. The proposal here presented suggests that the depletion mechanism corresponds to the major cause that imparts possible configurational changes related to the adsorbed polymer chains. As a result of, significant alterations of the hydrodynamic radius, effective volume fraction occupied by and interaction intensity (repulsion) between the particles are expected. The effect of the configurational dynamics experienced by the system reflected directly on the associated rheological behaviour.The innovation character of the present work relies on the employment of the electrosacoustic spectroscopy technique to evaluate possible changes of the hydrodynamic radius of the suspended particles (concentrated systems) integrated with the dynamic turbidimetry technique to evaluate the performance (sedimentation kinetics) of the complex formulations.

Neste trabalho caracterizamos dispersões aquosas de vesículas lipídicas por espalhamento de luz. Utilizamos os diferentes métodos da técnica: espalhamento de luz estático (SLS) e espalhamento de luz dinâmico (DLS) no estudo desses sistemas. No SLS a intensidade da luz espalhada foi coletada em vários ângulos de espalhamento para diferentes concentrações, e obtivemos os seguintes parâmetros: peso molecular, Mw, raio de giração, Rg, que esta relacionado _a forma do centro espalhador e segundo coeficiente do virial, A2, que está relacionado a interações entre partículas. No DLS a correlação da luz espalhada foi obtida em função do tempo, para vários ângulos e concentrações. As medidas de DLS fornecem o coeficiente de difusão dos centros espalhadores, que estão relacionados a seus diâmetros efetivos. Em especial a técnica de SLS foi amplamente estudada para desenvolvimento de uma apostila coordenada pela Profa. Vera B. Henriques. Nesta apostila a teoria de espalhamento de luz estático está apresentada através de uma abordagem diferente da convencional, sob o ponto de vista da mecânica estatística. O principal objeto de estudo neste trabalho foram as dispersões aquosas de dimiristoil fosfatidil glicerol (DMPG). O DMPG, lipídio aniônico, saturado, com 14 carbonos nas cadeias hidrocarbônicas, em baixa força iônica, exibe um comportamento termo-estrutural muito peculiar, apresentando uma larga região de transição de fase, entre as fases gel e fluida. O lipídio dimiristoil fosfatidil colina (DMPC) com as mesmas cadeias carbônicas mas cabeça polar neutra, foi utilizado como controle, pois não apresenta esta anomalia. Os tamanhos efetivos obtidos pelos diferentes métodos (SLS) e (DLS) são semelhantes nas fases gel e fluída de ambos os lipídios, mas isso não ocorre na região de transição do DMPG. Ambos os métodos indicam um aumento na dimensão das vesículas de DMPG, nesta região de temperatura. No entanto, por SLS temos que o Rg triplica na região de transição, levando a um raio efetivo muito maior que o observado por DLS. Como é sabido que não há fusão entre as vesículas, considerando os dados de espalhamento de luz e a literatura, discutimos um modelo em que o agregado lipídico apresente grandes deformações isotrópicas e grandes flutuações de forma. / Light scattering used to structurally characterize lipid dispersions. In the present work light scattering was used to characterize aqueous dispersions of lipid vesicles. Two different methods were applied: static light scattering (SLS) and dynamic light scattering (DLS). With SLS, the intensity of the scattered light was measured at several scattering angles and different lipid concentrations. The following parameters were obtained: molecular weight (Mw) and radius of gyration (Rg), related to the mass and form of the scattering particle, and the second Virial coefficient (A2), related to interactions among particles. With DLS, the autocorrelation of the intensity of the scattered light was obtained, also at several scattering angles and lipid concentrations. Measurements of DLS provide the difusion coefficient of the particles, which are related to their hydrodynamic radii. In particular, the theory of SLS was largely studied to the elaboration of a manuscript, under the supervision of Dr. Vera B. Henriques, which presents an unconventional approach to the technique. Here, the main object of investigation was the aqueous dispersion of the anionic lipid dimyristoyl phosphatidylglycerol (DMPG), a saturated anionic lipid, with 14 C-atoms in the hydrophobic chains. At low ionic strength, this lipid exhibits a very peculiar thermo-structural behavior, presenting a large region of transition, between the gel and fluid phases. The lipid dimyristoyl phosphatidylcholine (DMPC), with the same hydrocarbon chains but zwitterionic polar headgroup, was used as a control, since it does not present such anomaly. Lipid dispersions were extruded through 100 nm filters (pore diameter). In the gel and fluid phases of DMPC and DMPG, both SLS and DLS indicate similar dimensions for the lipid vesicles, assuming them spherical. For DMPC, the obtained effective radius (Reff ) values were around 53 and 62nm for the gel and fluid phases, respectively, indicating the expected expansion of the uid bilayer, and radii compatible with the extrusion process. For DMPG, the fluid vesicle (Reff _ 30nm) was also found to be larger than the gel vesicle (Reff _ 27nm), but much smaller than the dimensions of the filter pore (R = 50nm). More interesting, over the DMPG transition region, SLS indicated a three times increase for the vesicle radius of gyration, whereas the hydrodynamic radius, measured by DLS, increased by 30% only. Besides, no light depolarization could be detected, indicating that the DMPG aggregates are isotropic particles, in average, at all temperatures. Considering the results presented here, and those in the literature, the presence of large bilayer fluctuations over the DMPG transition region will be discussed, possibly including the existence of bilayer pores.

Neste trabalho realizamos dois estudos independentes sobre a termodinâmica de modelos de água e o dobramento de proteínas em rede. / On this work we develop two independent studies on lattice models for water thermodynamics and protein folding.

Les variations de température à la surface externe d'une conduite peuvent provoquer la solidification du fluide en écoulement interne, induisant la formation de dépôts solides sur les parois en contact avec le fluide. Ces dépôts peuvent s'avérer néfastes au bon fonctionnement du système intégrant cette conduite ce qui motive l'étude de ces phénomènes physiques afin de comprendre les mécanismes qui les gouvernent et de permettre leur prévention. Ce genre de situations peuvent être rencontrées dans le domaine pétrolier, par exemple lors du transport en pipeline d'huiles chargées de paraffines, d'hydrates ou d'asphaltènes ; dans le domaine agroalimentaire pour la circulation de produits liquides dans des tubes refroidis ou bien tout simplement pour des canalisations d'eau subissant un refroidissement. Dans la majorité des cas le fluide en écoulement dans les conduites pétrolières est un matériau à changement de phase (MCP) complexe et sa composition multiple avec des interactions entre la rhéologie, la constitution et la température impactant la thermo-hydrodynamique de l'écoulement et la cinétique de transition de phase. Dans le cadre de cette thèse nous proposons de nous pencher sur ces aspects par voie de modélisation. Dans la première partie de ce travail, nous avons développé un modèle physique basé sur la méthode enthalpique pour étudier l’impact de la convection naturelle sur des écoulements et des transferts thermiques avec transition de phase solide liquide dans différentes géométries (cavité carré, cylindre, sphère, autour d’un faisceau de cylindre, …). L'effet du maillage et de l'ordre de précision des schémas numérique est aussi présenté. Le même modèle a été utilisé pour analyser la cristallisation d’un mélange de deux paraffines dans des cellules calorimétriques (Differential Scanning Calorimetry). Pour cette étude, le modèle a été validé à l’aide des résultats expérimentaux de la technique DSC. Le problème de l’écoulement d’un mélange de deux paraffines dans une conduite cylindrique a été aussi abordé afin de prédire la cristallisation du mélange sur la paroi interne de la conduite. Une étude paramétrique a été effectuée pour analyser l’influence de différents paramètres (nombre de Reynolds, concentration initiale de la solution binaire, température de la paroi, …) sur la cinétique de la cristallisation du mélange. La deuxième partie de la thèse a été consacrée à l’étude de l’écoulement avec changement de phase liquide-solide d’un brut paraffinique dans des conduites pétrolières. En effet, la cristallisation des paraffines dans les bruts engendre un équilibre phase thermodynamique liquide/solide. Un modèle thermodynamique rendant compte de cet équilibre a été développé (et validé à l’aide de résultats expérimentaux disponible dans la littérature). Ce modèle nous a permis de prédire la température de cristallisation commençante (Wax Appearance Temperature), ainsi que la quantité de paraffines cristallisées. Ce modèle thermodynamique a été couplé avec un modèle physique reposant sur les équations de conservation de la masse, de la quantité de mouvement, de l'énergie et des espèces pour décrire l’écoulement d’un brut paraffinique avec changement de phase liquide-solide dans des conduites pétrolières. Le modèle physique a été validé à l’aide de résultats expérimentaux disponibles dans la littérature. / Temperature change at the outer surface of a pipe can cause solidification of the fluid flowing internally, inducing the formation of solid deposits on the walls in contact with the fluid. This kind of phenomena can be encountered in many situations in the industrial context such as petroleum industries, for example when transporting pipeline oils with paraffins, hydrates or asphaltenes; food industry for the flow of liquids in cooled tubes or simply for water pipes undergoing cooling.The rheology of the transported fluid, its composition and temperature are among the key factors that influence the thermo-hydrodynamics of the flow and the kinetics of the formation of the deposits. The analysis of their impact is required to enhance the understanding of these mechanisms in order to avoid their undesirable effects as fouling or complete blockage of pipes.In this thesis we propose to study these mechanisms by numerical modelling. We will focus on the modelling of phase change and especially on the description of liquid-solid interfaces by models able to integrate the thermodynamics of physical phenomena such as diffuse interface methods. We seek the two-dimensional numerical resolution of the equations of flow, heat and mass transfer while taking into account the coupling between rheology, composition and temperature in the liquid phase as well as at the interface. The work will be undertaken in the basis of the in-house computational fluid dynamics (CFD) code developed in the laboratory.

O presente projeto de trabalho buscou avaliar e estabelecer relações de causa e efeito que justificassem a ocorrência de singularidades no comportamento reológico de formulações complexas contendo elevada carga de sólidos em suspensão incorporada. O mecanismo de estabilização coloidal empregado para preparação das formulações e posterior análise foi o mecanismo de estabilização eletroestérico. O caso particular estudado correspondeu à dispersão de cargas inorgânicas (óxido de alumínio) em meio aquoso auxiliada pela incorporação de conteúdo orgânico com ação dispersante (poliacrilato de amônio) e ligante (emulsão acrílica) respectivamente. A implementação de técnicas analíticas complementares permitiu o estudo detalhado de uma singularidade associada ao comportamento reológico do sistema, exclusivamente evidenciada em função da elevada carga de sólidos, oferecendo, portanto, subsídios para elaboração de um modelo supramolecular fundamentado em possíveis alterações de configuração das cadeias poliméricas para justificativa do fenômeno. A contribuição referente ao pleno entendimento do fenômeno, considerando-se suas causas e efeitos, reside na possibilidade de se estabelecer diretrizes de formulação que propiciem a otimização das quantidades de aditivos incorporados e o respectivo controle do estado de dispersão e estabilidade do sistema. Considerou-se, para efeitos práticos de engenharia de formulação, que o mínimo secundário de viscosidade corresponda ao ponto ótimo de estabilidade do sistema mesmo que a condição de máxima dispersão já tenha sido alcançada para uma concentração de dispersante equivalente ao mínimo primário de viscosidade. A contribuição do efeito estérico associada à reduzida distância de separação entre as partículas (IPS) mostrou-se de vital importância para assegurar a viabilidade de uma dada formulação. A proposta aqui apresentada fundamentou-se no mecanismo de depleção como principal força motriz para desencadear possíveis mudanças de conformação espacial das cadeias poliméricas implicando, portanto, em alterações do raio hidrodinâmico, da fração volumétrica efetiva ocupada e da intensidade de interação (repulsão) entre as partículas. O efeito da dinâmica configuracional sofrida pelo sistema refletiu-se diretamente no comportamento reológico do mesmo. O caráter de inovação do presente trabalho reside na utilização não convencional da técnica de espectroscopia eletroacústica, empregada para avaliação de mudanças do raio hidrodinâmico de partículas em suspensão (sistemas concentrados), integrada com a técnica de turbidimetria dinâmica de varredura para avaliação de desempenho (cinética de sedimentação) de formulações complexas (suspensões). / The present research project pretended to evaluate and establish cause and effect relationships able to justify the singularities evidenced when exploring the rheological behaviour of complex formulations containing elevated solids loading. The stabilization mechanism used for the preparation of the dispersed systems and subsequent analyses corresponded to the electrosteric mechanism. The particular case studied herein corresponded to the dispersion of inorganic colloidal particles (aluminun oxide) in aqueous media assisted by the incorporation of organic components acting respectively as dispersant (ammonium polyacrylate) and binder agent (acrylic emulsion). The implementation of complementary analytical techniques enabled the detailed study of a singularity related to the rheological behaviour of the system that could only be evidenced for high solids loading condition. The results obtained provided the fundamental concepts necessary to propose a supramolecular model based on possible conformational changes associated with the adsorbed polymeric chains and sufficient to justify the phenomenon. Considering the practical aspects related to formulation engineering, it was determined that the secondary minimum viscosity corresponds to the optimum stablity condition although the maximum dispersion condition was achieved for dispersant concentrations equal to the primary minimum viscosity. The contribution of the steric effect associated with the reduced interparticle separation distance (IPS) showed itself to be of vital importance to ensure the viability of a stable formulation. The proposal here presented suggests that the depletion mechanism corresponds to the major cause that imparts possible configurational changes related to the adsorbed polymer chains. As a result of, significant alterations of the hydrodynamic radius, effective volume fraction occupied by and interaction intensity (repulsion) between the particles are expected. The effect of the configurational dynamics experienced by the system reflected directly on the associated rheological behaviour.The innovation character of the present work relies on the employment of the electrosacoustic spectroscopy technique to evaluate possible changes of the hydrodynamic radius of the suspended particles (concentrated systems) integrated with the dynamic turbidimetry technique to evaluate the performance (sedimentation kinetics) of the complex formulations.

Measurements of the small-scale dissipation statistics of turbulent boundary layer flows with and without free-stream turbulence are reported for Reτ ≈ 1000 (Reθ ≈ 2000). The scaling of the dissipation scale distribution is examined in these two boundary conditions of external wall-bounded flow. Results demonstrated that the local large-scale Reynolds number based on the measured longitudinal integral length-scale fails to properly normalize the dissipation scale distribution near the wall in these two free-stream conditions, due to the imperfect characterization of the upper bound of the inertial cascade by the integral length-scale. When a length-scale based on Townsend's attached-eddy hypothesis is utilized to describe the local large-scale Reynolds number near the wall, the description of the Reynolds number scaling was determined to be significantly improved and agreed with that found in homogeneous, isotropic turbulence. However, the scaling based on Townsend's attached-eddy hypothesis agreed best for the lowest 40% of the boundary layer thickness and then it degraded due to the loss of the validity of the attached eddy-hypothesis and the onset of external intermittency. A surrogate large-scale found from turbulent kinetic energy and mean dissipation rate improved the scaling of the dissipation scales, relative to the measured integral length-scale. The probability density functions of the local dissipation scales were calculated. When the three local large-scale Reynolds numbers are used for normalization, the one based on the longitudinal integral length-scale and the one based on the length-scale of attached-eddy hypothesis provide support for the existence of a universal distribution of the local dissipation scales up to the edge of the outer region of the turbulent boundary layer, which scales differently for inner and outer regions. However, the probability density functions of the local dissipation scales normalized by these two large-scale Reynolds numbers are deviated in interface locations for the flow without free-stream turbulence due to external intermittency. The surrogate large-scale provided the best agreement throughout the entire depth of the boundary layer. However, in the outer part of the boundary layer, a significantly reduced collapse in the scaled probability density functions was shown due to bias in the calculation introduced by the intermittent presence of laminar flow in the time series. To support that intermittency argument, injection of the free-stream turbulence was determined to improve the distribution of these normalized probability density functions in the intermittency locations for the flow regime without free-stream turbulence. In addition, unlike in channel flow, in the outer part of the turbulent boundary layer, the normalized distributions of the local dissipation scales were observed to be dependent on wall-normal position. This was found to be attributable to the presence of external intermittency in this outer part as the presence of free-stream turbulence was found to restore the scaling behavior by replacing the intermittent laminar flow with turbulent flow. Thus, the influence of external intermittency on the scaling of the dissipation scale distribution was examined in greater detail for the laminar free-stream condition. Probability density functions of the dissipative scales were compared with, and without, accounting for the external intermittency using an intermittency detection function. Results showed that accounting for the external intermittency produces restores universality in the shapes of the probability density functions at the same wall-normal location at different instances in time. In addition, properly scaling the dissipation-scale-distribution collapses the probability density functions calculated at different wall-normal locations. This improvement in the scaling of the dissipation-scale-distribution supports prior observations of universality of the small-scale description of the turbulence for wall-bounded flow.