Dispersões coloidais estão presentes em muitas aplicações industriais e biológicas, tais como indústria de alimentos, cosméticos, produtos farmacêuticos e nanoestruturas. Devido a efeitos entrópicos, macromoléculas, quando imersas em uma solução de partículas pequenas, tendem a se aglomerar. Para evitar este fenômeno, cargas são adicionadas à superfície do colóide. Conseqüentemente, para manter o sistema neutro, a solução contém uma série de pequenos contraíons que neutralizam a carga do colóide. A inclusão de cargas pode evitar a aglomeração dos colóides, mas incorpora ao problema uma série de novos efeitos de natureza eletrostática. Todos estes efeitos, bem como o comportamento termodinâmico do sistema, no entanto, podem ser compreendidos analisando-se a distribuição de contraíons ao redor do colóide. Uma teoria muito simples que fornece esta distribuição de contraíons é a teoria de Poisson Boltzmann, na qual os contraíons são partículas pontuais e as interações eletrostáticas entre os íons são obtidas via potencial médio criado por todos os íons sobre um determinado contraíon, desconsiderando, desta forma, as correlações. Nesta tese respondemos à pergunta: quando correlações eletrostáticas e de tamanho são relevantes? Propusemos que correlações eletrostáticas são relevantes quando as interações eletrostáticas entre íos superam os efeitos entrópicos, ou seja, quando o parâmetro de plasma está acima de um certo limiar, G2d > 2. Sugerimos, também, que o tamanho dos íons torna-se relevante quando a fração de volume dos contraíons na superfície do colóide está acima de um limiar, ou seja, fs > 0,2. As duas propostas são testadas comparando-se os resultados obtidos via teoria de Poisson Boltzmann com os resultados provenientes de simulações. Em seguida, empregamos a teoria de Debye-Hückel Buraco Cavidade para incorporar correlações eletrostáticas. Comparamos os resultados obtidos via esta teoria com os resultados de simulações. Mostramos que a incorporação dos efeitos de correlações eletrostáticas resultam em um maior número de contraíos próximos à superfície do macroíon do que o observado via teoria de Poisson Boltzmann. Depois, introduzimos duas teorias de funcionais de densidade ponderada que incluem efeitos de correlações de tamanho: funcional de densidade ponderada com um peso baseado na teoria Debye-Hückel Buraco Cavidade e uma funcional de densidade ponderada com um peso constante. Comparamos os resultados obtidos através destas duas teorias com simulações, e observamos que a segunda teoria apresenta uma melhor concordância com as simulações. De maneira geral, as correlações de tamanho deixam os íons mais afastados do colóide do que o predito via teoria de Poisson Boltzmann. Finalmente, propusemos uma combinação da teoria Debye-Hückel-Buraco-Cavidade e funcional de densidade ponderada com um peso constante para tratar de problemas onde tanto correlações eletrostáticas como de tamanho se façam presentes. Observamos que para G2d < 2 e fs < 0,2 a teoria mista fornece os mesmos resultados que a teoria de Poisson Boltzmann; para G2d > 2 e fs < 0,2 a teoria mista fornece os mesmos resultados que a teoria Debye-Hückel-Buraco-Cavidade, pois somente correlações eletrostáticas são relevantes; para G2d < 2 e fs > 0,2 a teoria mista oferece o mesmo resultado que a teoria funcional de densidade ponderada com um peso constante; para G2d ≈ 2 e fs ≈ 0,2 ocorre uma compensação entre efeitos e a teoria mista fornece o mesmo resultado que a teoria de Poisson Boltzmann. Para G2d > 2 and fs > 0,2 efeitos de tamanho dominam e a teoria mista oferece os mesmos resultados que a teoria da funcional de densidade ponderada com umpeso constante. / Colloidal dispersions are present in many industrial and biological applications going from food, cosmetics, pharmaceutical and nanostructures. Due to entropic effects the large macromolecules in a solvent made of small particles, agglomerate. In order to avoid this effect, charged groups are added to the colloidal surface. Consequently in order to keep the charge neutrality the solution is full of counterions. The addition of charges might stop agglomeration but adds a number of new phenomena that are deeply related to the distribution of the counterions around the macroion. One simple theory that describes this distribution is the Poisson Boltzmann approach in which the counterions are assumed to be point ions and where the electrostatic interactions between the counterions are taken into account as the average field on a singel ions, ignoring correlations. In this thesis we address the question; when does electrostatic correlations and the size of the counterions are relevant? We propose that electrostatic correlations are relevant when the electrostatic interactions between the ions are bigger than the entropic effects. This assumption can be expressed by the plasma parameter being above a certain threshold, G2d > 2. We also propose that the size of the counterions become relevant when the volume fraction of ions at the surface of the colloid is above a certain threshold fs > 0.2. This two propositions are tested comparing results obtained with the PB theory with simulations. We then propose a theory to take into account the electrostatic correlations, the Debye-Hückel-Hole-Cavity and test this approach with simulations. The electrostatic correlation leads to more ions close to the colloid than the Poisson Boltzmann predicts. Next, we present two different approaches to account for size effects, the Weight Density Approximation based in the Debye-Hückel-Hole-Cavity theory and the Weight Density Approximation based in a constant weight. Comparison with simulations show that the second approach gives a better agreement. The size correlations leads to less ions close to the colloid than the Poisson Boltzmann approach predicts. Finally we propose a combination of the Debye-Hückel-Hole-Cavity and the Weight Density Approximation based in a constant weight to be the theory able to take into account both electrostatic and size correlations. Our result shows that for G2d <2 and fs <0.2 electrostatic and size correlations are irrelevant so Poisson Boltzmann is a good approach; for G2d > 2 and fs < 0.2 electrostatic correlations dominates and Debye-Hückel-Hole-Cavity gives a good approach; G2d < 2 and fs > 0.2 the Weight Density Approximation based in a constant weight gives the correct behavior; for G2d ≈ 2 and fs ≈ 0.2 the electrostatic correlation effects cancel the size effects and Poisson Boltzmann gives a good approximation. For G2d > 2 and fs > 0.2 size effects dominate.

Dispersões coloidais estão presentes em muitas aplicações industriais e biológicas, tais como indústria de alimentos, cosméticos, produtos farmacêuticos e nanoestruturas. Devido a efeitos entrópicos, macromoléculas, quando imersas em uma solução de partículas pequenas, tendem a se aglomerar. Para evitar este fenômeno, cargas são adicionadas à superfície do colóide. Conseqüentemente, para manter o sistema neutro, a solução contém uma série de pequenos contraíons que neutralizam a carga do colóide. A inclusão de cargas pode evitar a aglomeração dos colóides, mas incorpora ao problema uma série de novos efeitos de natureza eletrostática. Todos estes efeitos, bem como o comportamento termodinâmico do sistema, no entanto, podem ser compreendidos analisando-se a distribuição de contraíons ao redor do colóide. Uma teoria muito simples que fornece esta distribuição de contraíons é a teoria de Poisson Boltzmann, na qual os contraíons são partículas pontuais e as interações eletrostáticas entre os íons são obtidas via potencial médio criado por todos os íons sobre um determinado contraíon, desconsiderando, desta forma, as correlações. Nesta tese respondemos à pergunta: quando correlações eletrostáticas e de tamanho são relevantes? Propusemos que correlações eletrostáticas são relevantes quando as interações eletrostáticas entre íos superam os efeitos entrópicos, ou seja, quando o parâmetro de plasma está acima de um certo limiar, G2d > 2. Sugerimos, também, que o tamanho dos íons torna-se relevante quando a fração de volume dos contraíons na superfície do colóide está acima de um limiar, ou seja, fs > 0,2. As duas propostas são testadas comparando-se os resultados obtidos via teoria de Poisson Boltzmann com os resultados provenientes de simulações. Em seguida, empregamos a teoria de Debye-Hückel Buraco Cavidade para incorporar correlações eletrostáticas. Comparamos os resultados obtidos via esta teoria com os resultados de simulações. Mostramos que a incorporação dos efeitos de correlações eletrostáticas resultam em um maior número de contraíos próximos à superfície do macroíon do que o observado via teoria de Poisson Boltzmann. Depois, introduzimos duas teorias de funcionais de densidade ponderada que incluem efeitos de correlações de tamanho: funcional de densidade ponderada com um peso baseado na teoria Debye-Hückel Buraco Cavidade e uma funcional de densidade ponderada com um peso constante. Comparamos os resultados obtidos através destas duas teorias com simulações, e observamos que a segunda teoria apresenta uma melhor concordância com as simulações. De maneira geral, as correlações de tamanho deixam os íons mais afastados do colóide do que o predito via teoria de Poisson Boltzmann. Finalmente, propusemos uma combinação da teoria Debye-Hückel-Buraco-Cavidade e funcional de densidade ponderada com um peso constante para tratar de problemas onde tanto correlações eletrostáticas como de tamanho se façam presentes. Observamos que para G2d < 2 e fs < 0,2 a teoria mista fornece os mesmos resultados que a teoria de Poisson Boltzmann; para G2d > 2 e fs < 0,2 a teoria mista fornece os mesmos resultados que a teoria Debye-Hückel-Buraco-Cavidade, pois somente correlações eletrostáticas são relevantes; para G2d < 2 e fs > 0,2 a teoria mista oferece o mesmo resultado que a teoria funcional de densidade ponderada com um peso constante; para G2d ≈ 2 e fs ≈ 0,2 ocorre uma compensação entre efeitos e a teoria mista fornece o mesmo resultado que a teoria de Poisson Boltzmann. Para G2d > 2 and fs > 0,2 efeitos de tamanho dominam e a teoria mista oferece os mesmos resultados que a teoria da funcional de densidade ponderada com umpeso constante. / Colloidal dispersions are present in many industrial and biological applications going from food, cosmetics, pharmaceutical and nanostructures. Due to entropic effects the large macromolecules in a solvent made of small particles, agglomerate. In order to avoid this effect, charged groups are added to the colloidal surface. Consequently in order to keep the charge neutrality the solution is full of counterions. The addition of charges might stop agglomeration but adds a number of new phenomena that are deeply related to the distribution of the counterions around the macroion. One simple theory that describes this distribution is the Poisson Boltzmann approach in which the counterions are assumed to be point ions and where the electrostatic interactions between the counterions are taken into account as the average field on a singel ions, ignoring correlations. In this thesis we address the question; when does electrostatic correlations and the size of the counterions are relevant? We propose that electrostatic correlations are relevant when the electrostatic interactions between the ions are bigger than the entropic effects. This assumption can be expressed by the plasma parameter being above a certain threshold, G2d > 2. We also propose that the size of the counterions become relevant when the volume fraction of ions at the surface of the colloid is above a certain threshold fs > 0.2. This two propositions are tested comparing results obtained with the PB theory with simulations. We then propose a theory to take into account the electrostatic correlations, the Debye-Hückel-Hole-Cavity and test this approach with simulations. The electrostatic correlation leads to more ions close to the colloid than the Poisson Boltzmann predicts. Next, we present two different approaches to account for size effects, the Weight Density Approximation based in the Debye-Hückel-Hole-Cavity theory and the Weight Density Approximation based in a constant weight. Comparison with simulations show that the second approach gives a better agreement. The size correlations leads to less ions close to the colloid than the Poisson Boltzmann approach predicts. Finally we propose a combination of the Debye-Hückel-Hole-Cavity and the Weight Density Approximation based in a constant weight to be the theory able to take into account both electrostatic and size correlations. Our result shows that for G2d <2 and fs <0.2 electrostatic and size correlations are irrelevant so Poisson Boltzmann is a good approach; for G2d > 2 and fs < 0.2 electrostatic correlations dominates and Debye-Hückel-Hole-Cavity gives a good approach; G2d < 2 and fs > 0.2 the Weight Density Approximation based in a constant weight gives the correct behavior; for G2d ≈ 2 and fs ≈ 0.2 the electrostatic correlation effects cancel the size effects and Poisson Boltzmann gives a good approximation. For G2d > 2 and fs > 0.2 size effects dominate.

Orientadores: Daniela Zanchet, Jose Antonio Brum / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Fisica Gleb Wataghin / Made available in DSpace on 2018-08-10T04:28:03Z (GMT). No. of bitstreams: 1 Rocha_TulioCostaRizutida_D.pdf: 8952935 bytes, checksum: 2283ed573c4cf94d5cba5aa42d7b2113 (MD5) Previous issue date: 2008 / Resumo: Nanopartículas de metais nobres têm atraído uma renovada atenção nos últimos anos devido às novas aplicações científicas e tecnológicas explorando suas propriedades ópticas únicas. No regime nanométrico, é bem conhecido que a resposta óptica de metais, associada aos plásmons de superfície, depende fortemente do tamanho e também da forma. De fato, grande parte das aplicações ópticas de nanopartículas de ouro e prata baseia-se na exploração dos efeitos de forma. Porém, apesar dos esforços realizados, os processos que levam à formação de morfologias anisotrópicas ainda não são bem compreendidos e a formulação de um mecanismo geral ainda é um desafio. Nesse trabalho, foram abordados os mecanismos de formação e crescimento de nanoprismas triangulares de prata produzidos por métodos de síntese coloidal. Uma combinação de diferentes técnicas experimentais foi utilizada para estudar diversos aspectos da síntese fotoquímica, dentre eles, a evolução morfológica, a cinética da reação e a estrutura cristalina das nanopartículas. As sólidas evidências experimentais obtidas associadas a outras observações da literatura foram utilizadas na formulação de um modelo fenomenológico para explicar a formação e crescimento dos nanoprismas de prata em métodos fotoquímicos. Esse modelo baseia-se na influência dos defeitos cristalográficos, que induzem a formação dos nanoprismas nos momentos iniciais da síntese, e na excitação de plásmons de superfície, que ocorre em estágios avançados, sendo responsável pela definição do tamanho final dos nanoprismas. Adicionalmente, cálculos teóricos indicaram que aspectos energéticos podem ter um papel ativo nesse sistema, favorecendo o crescimento dos nanoprismas em relação às nanopartículas esféricas durante os estágios iniciais da síntese. Finalmente, os nanoprismas triangulares de prata produzidos foram aplicados ao estudo de efeitos de intensificação do espalhamento Raman de moléculas. Medidas espectroscópicas de moléculas depositadas na superfície de nanoprismas com diferentes tamanhos foram realizadas e a comparação quantitativa dos resultados indicou a presença de um tamanho ótimo, que é determinado por processos de perda de energia dos plásmons de superfície / Abstract: Noble metal nanoparticles have attracted a recent renewed interest due to the new scientific and technological applications exploiting their unique optical properties. At nanometric scale, it is well known that the optical response of metals, related to the excitation of surface plasmons, strongly depends not only on the size of the particles but also on their shape. Several methodologies to produce silver and gold nanoparticles with different shapes are available in the literature. However, notwithstanding the efforts that have been made, the process that lead to the formation of anisotropic morphologies has not been fully understood yet and a general mechanism is still a challenge. In this work, we address the formation and growth mechanisms of silver triangular nanoprisms produced by photochemical methods. A set of characterization tools was used to study different aspects of the photochemical synthesis, namely, the morphological evolution, the reaction kinetics and the crystalline structure of the nanoprisms. The solid experimental evidences obtained here were used to build a phenomenological model that explains the formation and growth of silver triangular nanoplates in photochemical methods. This model was based on the influence of crystallographic defects, which induce the formation of the nanoprismas in the initial stages of the synthesis, and on the excitation of surface plasmons, which occurs in advanced stages and it is responsible for the definition of the final size of the nanoprismas. Additionally, theoretical calculations indicate that energetics might play an important role in this system, favoring the growth of nanoprismas relative to spheres. Finally, the silver triangular nanoprisms were used to study enhancement effects in the Raman scattering of molecules. We performed spectroscopic measurements for nanoplates with different sizes and the quantitative comparison of the curves indicated the existence of an optimum size that is dictated by surface plasmon energy losses / Doutorado / Física da Matéria Condensada / Doutor em Ciências

Dispersões coloidais estão presentes em muitas aplicações industriais e biológicas, tais como indústria de alimentos, cosméticos, produtos farmacêuticos e nanoestruturas. Devido a efeitos entrópicos, macromoléculas, quando imersas em uma solução de partículas pequenas, tendem a se aglomerar. Para evitar este fenômeno, cargas são adicionadas à superfície do colóide. Conseqüentemente, para manter o sistema neutro, a solução contém uma série de pequenos contraíons que neutralizam a carga do colóide. A inclusão de cargas pode evitar a aglomeração dos colóides, mas incorpora ao problema uma série de novos efeitos de natureza eletrostática. Todos estes efeitos, bem como o comportamento termodinâmico do sistema, no entanto, podem ser compreendidos analisando-se a distribuição de contraíons ao redor do colóide. Uma teoria muito simples que fornece esta distribuição de contraíons é a teoria de Poisson Boltzmann, na qual os contraíons são partículas pontuais e as interações eletrostáticas entre os íons são obtidas via potencial médio criado por todos os íons sobre um determinado contraíon, desconsiderando, desta forma, as correlações. Nesta tese respondemos à pergunta: quando correlações eletrostáticas e de tamanho são relevantes? Propusemos que correlações eletrostáticas são relevantes quando as interações eletrostáticas entre íos superam os efeitos entrópicos, ou seja, quando o parâmetro de plasma está acima de um certo limiar, G2d > 2. Sugerimos, também, que o tamanho dos íons torna-se relevante quando a fração de volume dos contraíons na superfície do colóide está acima de um limiar, ou seja, fs > 0,2. As duas propostas são testadas comparando-se os resultados obtidos via teoria de Poisson Boltzmann com os resultados provenientes de simulações. Em seguida, empregamos a teoria de Debye-Hückel Buraco Cavidade para incorporar correlações eletrostáticas. Comparamos os resultados obtidos via esta teoria com os resultados de simulações. Mostramos que a incorporação dos efeitos de correlações eletrostáticas resultam em um maior número de contraíos próximos à superfície do macroíon do que o observado via teoria de Poisson Boltzmann. Depois, introduzimos duas teorias de funcionais de densidade ponderada que incluem efeitos de correlações de tamanho: funcional de densidade ponderada com um peso baseado na teoria Debye-Hückel Buraco Cavidade e uma funcional de densidade ponderada com um peso constante. Comparamos os resultados obtidos através destas duas teorias com simulações, e observamos que a segunda teoria apresenta uma melhor concordância com as simulações. De maneira geral, as correlações de tamanho deixam os íons mais afastados do colóide do que o predito via teoria de Poisson Boltzmann. Finalmente, propusemos uma combinação da teoria Debye-Hückel-Buraco-Cavidade e funcional de densidade ponderada com um peso constante para tratar de problemas onde tanto correlações eletrostáticas como de tamanho se façam presentes. Observamos que para G2d < 2 e fs < 0,2 a teoria mista fornece os mesmos resultados que a teoria de Poisson Boltzmann; para G2d > 2 e fs < 0,2 a teoria mista fornece os mesmos resultados que a teoria Debye-Hückel-Buraco-Cavidade, pois somente correlações eletrostáticas são relevantes; para G2d < 2 e fs > 0,2 a teoria mista oferece o mesmo resultado que a teoria funcional de densidade ponderada com um peso constante; para G2d ≈ 2 e fs ≈ 0,2 ocorre uma compensação entre efeitos e a teoria mista fornece o mesmo resultado que a teoria de Poisson Boltzmann. Para G2d > 2 and fs > 0,2 efeitos de tamanho dominam e a teoria mista oferece os mesmos resultados que a teoria da funcional de densidade ponderada com umpeso constante. / Colloidal dispersions are present in many industrial and biological applications going from food, cosmetics, pharmaceutical and nanostructures. Due to entropic effects the large macromolecules in a solvent made of small particles, agglomerate. In order to avoid this effect, charged groups are added to the colloidal surface. Consequently in order to keep the charge neutrality the solution is full of counterions. The addition of charges might stop agglomeration but adds a number of new phenomena that are deeply related to the distribution of the counterions around the macroion. One simple theory that describes this distribution is the Poisson Boltzmann approach in which the counterions are assumed to be point ions and where the electrostatic interactions between the counterions are taken into account as the average field on a singel ions, ignoring correlations. In this thesis we address the question; when does electrostatic correlations and the size of the counterions are relevant? We propose that electrostatic correlations are relevant when the electrostatic interactions between the ions are bigger than the entropic effects. This assumption can be expressed by the plasma parameter being above a certain threshold, G2d > 2. We also propose that the size of the counterions become relevant when the volume fraction of ions at the surface of the colloid is above a certain threshold fs > 0.2. This two propositions are tested comparing results obtained with the PB theory with simulations. We then propose a theory to take into account the electrostatic correlations, the Debye-Hückel-Hole-Cavity and test this approach with simulations. The electrostatic correlation leads to more ions close to the colloid than the Poisson Boltzmann predicts. Next, we present two different approaches to account for size effects, the Weight Density Approximation based in the Debye-Hückel-Hole-Cavity theory and the Weight Density Approximation based in a constant weight. Comparison with simulations show that the second approach gives a better agreement. The size correlations leads to less ions close to the colloid than the Poisson Boltzmann approach predicts. Finally we propose a combination of the Debye-Hückel-Hole-Cavity and the Weight Density Approximation based in a constant weight to be the theory able to take into account both electrostatic and size correlations. Our result shows that for G2d <2 and fs <0.2 electrostatic and size correlations are irrelevant so Poisson Boltzmann is a good approach; for G2d > 2 and fs < 0.2 electrostatic correlations dominates and Debye-Hückel-Hole-Cavity gives a good approach; G2d < 2 and fs > 0.2 the Weight Density Approximation based in a constant weight gives the correct behavior; for G2d ≈ 2 and fs ≈ 0.2 the electrostatic correlation effects cancel the size effects and Poisson Boltzmann gives a good approximation. For G2d > 2 and fs > 0.2 size effects dominate.

Orientadores: Watson Loh, Antonio Carlos Bannwart / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Quimica / Made available in DSpace on 2018-08-09T07:38:37Z (GMT). No. of bitstreams: 1 Santos_RonaldoGoncalvesdos_D.pdf: 3351599 bytes, checksum: e5f17ac5f9d0db6db39884f2b5c5ec78 (MD5) Previous issue date: 2007 / Resumo: Emulsões óleo em água (o/a) têm sido propostas como alternativa aos métodos convencionais para redução dos problemas associados ao escoamento, causados pela alta viscosidade destes óleos. Nestas emulsões, o óleo se encontra disperso na fase aquosa sob a forma de gotículas e pode ser bombeado sem necessariamente entrar em contato com as paredes do duto, reduzindo os efeitos viscosos e ampliando a capacidade de transferência. Neste trabalho, realizou-se um estudo experimental sobre a aplicação de emulsões de óleo em água (o/a) no transporte de óleos pesados brasileiros. Emulsões contendo 50-70% de óleo disperso foram preparadas sob diferentes parâmetros da formulação fisico-química, permitindo adequação de suas propriedades às exigência da aplicação. Eficaz protocolo de preparação produziu gotas com diâmetro médio de Sauter D3,2da ordem de 10-50 micrômetros e com distribuição de diâmetros obedecendo a uma distribuição lognormal. A reologia e a estabilidade das emulsões foram dependentes do diâmetro médio de gota e da distribuição de tamanhos. Emulsões mostraram alta estabilidade sob condições severas de avaliação. A emulsificação reduziu a viscosidade de 1.000 cP do óleo original para 30-150 cP da emulsão o/a, produzindo decréscimo de cerca de 30 vezes na viscosidade do óleo. As emulsões mostraram comportamento pseudoplástico descrito pela lei de potência (modelo de Ostwald-de Waele), apresentando influência do envelhecimento. O óleo cru pode.ser recuperado pela quebra da emulsão utilizando desemulsificantes comerciais, produzindo BS&W de cerca de 5%. Um sistema' de fluxo experimental em escala piloto com comprimento reto de teste de 3 m foi constiuído com tubulação de aço API com % in d.i. para ensaios de escoamento das emulsões. A viscosidade foi calculada a partir de dados de queda de pressão. A viscosidade de emulsões óleo em água com 50%, 60% e 70% de óleo disperso foi adequadamente prevista pela lei de potência, mostrando valores até 100 vezes menor que a viscosidade do óleo em escoamento a 500 S.I. As perdas de carga observadas durante o escoamento de emulsões são até 20 vezes inferior às do óleo em fluxo monofásico, o que também se verifica através da estimativa da potência requerida de bombeio. As emulsões óleo em água produzidas neste estudo mostram a viabilidade técnica da aplicação de emulsões óleo em água para o escoamento de óleos pesados, incluindo operações de elevação e transporte em oleodutos convencionais. / Abstract: Use of oil-in-water (o/w) emulsions for transportation of heavy crude oils has been proposed as altemative means of reducing problems associated with difficulties in transport of these fluids for their high viscosity, with some advantages over the currently used methodologies. In these emulsions, oil is dispersed as stabilized droplets allowing its transportation with low pressure drops and avoiding oil contact with the pipeline walls. This study involved the experimental evaluation of o/w emulsions for the transportation of Brazilian crude oils. Emulsions containing 50-70% oi! were prepared with different formulations. An efficient methodology of preparation produced emulsions with droplet diameters (Sauter diameter, D3,2) in the range of 10-50 micrometers and a log-normal size distribution. Rheology and stability of these emulsions depend on their average diameters and size distributions. These emulsions displayed high stability under severe conditions (centrifugation). Using this approach, oil viscosity of around 1 000 cP was reduced to 30-150 cP. These emulsions display a pseudoplastic rheological behavior following the power law described by the Ostwald-de Waele model, which changed with their aging. Crude oil could be recovered after emulsification using commercial additives, producing BS&W values of 5 %, which could be further reduced with more severe treatments (down to 0.5 %). An experimental set-up was built for flow assays in pilot scale with 3 m length for measurements made of API steel with :;4 in (i.d.). Viscosity was obtained from drop pressure data. The power law adequately adjusted emulsions containing several oil contents. Emulsions viscosity was found to. be up to 100 times lower tIutn the oil viscosityat 500 S-I and drop pressure in the emulsion flow was 20 times lower than q,il monophasic flow. In summary, all of these measurements confirm the viability of these emulsions for use in transportation of heavy crude oils, induding their lift and transportation in conventional pipelines. / Doutorado / Engenharia de Processos / Doutor em Engenharia Química

Le travail présenté dans ce mémoire s'inscrit dans le cadre des études théoriques dedispersions colloïdales, ou suspensions de particules dont la taille varie du nanomètre aumicromètre. Dans ces milieux, les interactions entre les particules peuvent être moduléesen jouant par exemple sur leur composition superficielle, de même qu'il est possible demodifier l'environnement des colloïdes comme le solvant, le confinement du mélange et/ouéventuellement un champ extérieur pour influer sur leurs propriétés thermodynamiques.La modélisation-simulation permet alors de tester sur ordinateurcertains jeux de paramètres pouvant produire le phénomène souhaité,avant son éventuelle réalisation expérimentale.Ce travail se concentre sur cette étape préliminaire en considérant un mélange desphères dures dipolaires et apolaires, placé dans milieu confiné etsoumis à un champ électrique (magnétique pour des ferrocolloïdes).Dans une première étape, nous nous intéressons aux états d'équilibres du mélange,en étudiant par simulations Monte-Carlo un mélange symétrique en composition,non-additif et confiné entre deux murs éloignés.En comparant les résultats pour différentes densités et directions du champ extérieur,nous retrouvons certaines propriétés déjà observées pour des systèmes similaires.Nous commençons par la situation de référence sans champ où à faible densité,le mélange est monophasique et l'espèce dipolaire fuit les murs.L'augmentation de la densité favorise alors la séparation de phase et dans la phase richeen dipôles, l'espèce dipolaire mouille les murs.L'application d'un champ perpendiculaire aux murs favorise la stabilité du mélange malgrésa densité élevée et la non-additivité entre les deux espèces.En faisant croître ce champ, nous observons une structuration de l'espèce dipolaire,notamment près des murs ainsi que la formation de <<colonnes>> de dipôles dansla direction du champ.Enfin un champ parallèle aux murs provoque la démixtion du mélange dèsla plus faible densité considérée. Les dipôles fuient à nouveau les murs etnous observons de longues chaînes intriquées de dipôles.Dans une seconde étape, nous nous intéressons à la dynamique d'un mélange asymétrique encomposition et soumis à un champ. Nous combinons dans cette étudedes simulations Monte-Carlo et de dynamique moléculaire (Langevin).Le mélange est placé dans une boîte présentant un goulot d'étranglement afinde simuler un pore ouvert en contact avec un réservoir de particules,à travers une interface explicite. Le champ, perpendiculaire aux murs, sera appliquéau niveau du goulot d'étranglement afin d'y attirer les dipôles.Nous considérons d'abord un mélange peu dense afin que le cycle remplissage / vidagedu milieu confiné soit réversible. Dans le but d'accélérer ces cycles,l'intensité du champ est progressivement augmentée.Le remplissage en dipôles est effectivement plus rapide mais sa composition satureprématurément.Nous lançons ensuite une série de cycles avec des coefficients de frottement deLangevin croissants mais relativement petits afin de limiterla durée des simulations. Nous notons alors que les temps de remplissage oude vidage du pore varient linéairement en fonction du coefficient de frottementce qui nous permet d'estimer par extrapolation la durée d'un cycle pour les colloïdes.En jouant sur la non-additivité et la densité,nous parvenons à rendre les cycles irréversibles : selon l'application envisagée,l’irréversibilité pourra s'avérer utile ou devra être évitée.Nous terminons ce chapitre en estimant la variation de la durée des cycles avecla taille des colloïdes. Un modèle d'interaction entre colloïdes constitués pardes centres répulsifs en loi de puissance, uniformément répartis dans une sphèrenous permet de prévoir, moyennant des hypothèses sur les lois d'échelle,la variation des durées de remplissage ou de vidage pour des tailles allantde petits colloïdes aux dimensions quasi-moléculaires / The work presented in this dissertation is in the framework of the theoretical study ofcolloidal dispersions, i.e. suspensions of particles whose size varies from nanometers tomicrometers. In such a medium, the interactions between particles can be tuned through their surfacecomposition for instance. One may also modify the environment of the colloids:a specific solvent can be combined with confinement of the mixture andan external can field applied on it in order to tune its thermodynamic properties.Once a model of a physical system is defined, computer simulation can be used to explorea range of parameters to check if the sought phenomenon occurs, before carrying outany real experiment. This work focuses on this preliminary step: our model consists ofa mixture of dipolar and apolar hard spheres in a confined medium and subjected to anelectric field (or a magnetic one for ferrocolloids).In a first step, we use Monte Carlo simulation to study equilibrium states ofa binary mixture confined between distant walls,with symmetric composition of the two species having non additive interactions.By comparing the results of different densities and field directions,we recover some properties already observed for similar systems.In the reference state where the field is turned off, the mixture at low density is stableand we notice that the dipoles stay away from the walls.A denser mixture separates into two phases and in the dipoles rich one,the dipolar particles now wet the walls.When the mixture is subjected to a field perpendicular to the walls,it remains stable in spite of its high density and non additivity between unlike particles.Increasing the field induces a structuring of the dipolar component near the wallsand we observe column shaped clusters of dipoles along the direction of the field.Finally, the application of a field parallel to the walls separates the mixture,even at the lowest density we chose. Dipoles stay away from the walls and we observeentangled dipoles chains.In a second step we explore the dynamics of a mixture with asymmetric composition andsubjected to a field. We combine Monte Carlo and molecular dynamic (Langevin) simulationsin this study. The mixture is confined in a box with a bottleneck channel in order tosimulate an open pore exchanging particles with a reservoir through an explicit interface.The field which is perpendicular to the walls is applied in the bottleneck regionto attract dipoles there.We first consider a low density mixture such that the filling / emptying cycleof the pore is reversible.The intensity of the field is then increased to speed up the cycles.As expected, the dipoles fill the pore faster then. However their composition saturatesunder the maximum value found for a lower field.A series of cycles was performed with increasing Langevin damping coefficients but stilllow enough to reduced the computation time.We then notice that the filling or emptying duration is a linear function ofthe damping coefficient. The duration of a cycle for colloids is then obtained fromextrapolation.Combining non additivity and high enough density, we are able to make an irreversible cycle:depending on the application sought for, this irreversibility can be useful ormust be avoided.This chapter ends with the assessment of the duration of a cycle with respect tothe size of colloids. We use an interaction model between colloidal particles wherea colloid is uniformly made of repulsive centers following a power law.With some scaling law hypotheses, the duration of a filling or an emptying is estimated forsmall colloids down to nearly molecular dimensions

Cette thèse porte sur les différentes voies de fonctionnalisation et d'auto-organisation d'un système modèle dans le domaine de la matière condensée : le virus fd et ses mutants. Alors que son diagramme de phase cristal-liquide a été établi et sa correspondance qualitative avec les prédictions théoriques montrée, une des perspectives majeure consiste en son utilisation comme brique élémentaire dans la construction de nouveaux auto-assemblages. De telles avancées passent nécessairement par l'ajout de fonctions de manière régio-sélective sur le corps de la particule. Nous proposons dans ces travaux l'étude de plusieurs voies de fonctionnalisation menant à l'ajout d'espèces moléculaires ou macromoléculaires soit sur l'ensemble du virus ou bien uniquement à son extrémité.En réalisant le greffage de polymères thermosensibles, il est alors possible d'explorer les possibilités d'induire une transition de phase par variation du diamètre effectif du bâtonnet. En utilisant des diblocs d'élastine, ce principe est montré sur la transition entre le liquide isotrope et la phase nématique. L'utilisation de mutants particuliers, conçus par phage display, permet de s'intéresser alors uniquement à la fonctionnalisation de la protéine p3 située à une des extrémités du phage. L'ajout de chromophores permet alors une visualisation unique de la phase smectique et de ses défauts et crée également un effet patchy perturbant le diagramme de phase cristal-liquide. La biotine quant à elle permet la création d'auto-assemblages du fait de son interaction spécifique avec les dérivés d'avidine et un tel système est alors comparé avec un mutant dont l'ADN modifié permet l'expression directe d'une étiquette biologique complémentaire de la streptavidine. Les résultats prometteurs obtenus sont également complétés par une étude encourageante pour l'utilisation des systèmes cristal-liquides colloïdaux dans le domaine de l'électro-optique. / This thesis deals with the different paths of functionalization and self-organization of a model system of colloidal rod-like particles: the fd virus and its mutants. While its liquid-crystalline phase diagram is well established and proven to be in qualitative agreement with theory and numerical simulations, one of the most trending perspectives is its use as building-block in new self-assemblies. For such purposes, it is mandatory to add functions regio-specifically on the particle. We show in this work the study of several ways of functionalization leading to the grafting of molecular or macromolecular compounds onto the whole virus or only onto its tip.When grafting thermoresponsive polymers, we can then explore the possibilities to induce phase transitions by a variation of the effective diameter of the rod. Using diblocs of elastin-like peptides, this principle is shown to work on the isotropic-to-nematic phase transition. The use of particular mutants, engineered by phage display, allows us to functionalize only the tip of the virus. The addition of dyes provides unique features on the smectic phase and its defects and creates a patchy effect which is modifying the liquid-crystalline phase diagram. The functionalization with biotin leads towards the creation of new self-assemblies thanks to its specific interaction with avidine and such a system is then compared with a mutant displaying a biological tag interacting with streptavidin. The results obtained are promising and are completed by a whole study of the use of colloidal liquid-crystalline system in electro-optics.

Ce travail est consacré à l'étude, par des techniques originales de diffusion dynamique de la lumière, de la dynamique lente de deux systèmes colloïdaux modèles. On a utilisée une technique de Corrélation Résolue dans le Temps (TRC) et des simulations numériques pour sonder, sur une grande plage de concentrations et sur sept décades en temps de relaxation, la dynamique d'un système de sphères dures colloïdales approchant de la transition vitreuse. A faible fraction volumique les temps de relaxation peuvent être représentés par une loi de puissance critique comme prévu par la Théorie de Couplage des Modes (MCT). A des concentrations élevées le temps de relaxation augmente selon une inattendue loi exponentielle critique avec une fraction volumique critique supérieure à celle prévue par MCT. L'étude des fluctuations de la dynamique confirme la présence d'un régime de dynamique activée: ce scénario est le même que l'on retrouve pour les systèmes vitreux moléculaires. Le deuxième sujet étudié est le comportement d'une suspension de colloïdes attractifs soumis à la contrainte gravitationnelle. Initialement, des agrégats de particules sédimentent en formant un gel sur le fond de la cellule. Les techniques spéciales employées nous ont permis de mesurer l'évolution des profils de concentration, des profils de vitesse et la dynamique locale dans la phase gel: la sédimentation peut être complètement décrite par le taux de déformation locale qui permet aussi une remise à l'échelle de la dynamique locale. Le rôle de la friction aux parois des cellules a été sondé par des expériences de polarimétrie et par la résolution numérique d'un modèle poroélastique pour la sédimentation du gel. / We investigated, using original light scattering techniques, the slow dynamics of two different model colloidal systems. We used a Time Resolved Correlation (TRC) technique and numerical simulations to study the dynamics of systems of colloidal hard spheres over a broad range of densities and seven decades in relaxation time. At low volume fraction the equilibrium relaxation time is described by the algebraic divergence predicted by mode-coupling theory (MCT). At higher density the relaxation time increases with an unexpected critical exponential form, whose the critical volume fraction is much larger then the MCT singularity. The fluctuations of the dynamics confirm the presence of an activated regime, as founded in molecular glass formers close to the glass transition. The second research subject concerns the behaviour of suspensions of attractive colloidal particles under gravitational stress. Initially, clusters fall forming a gelled deposit at the cell bottom. Our apparatus allows us to follow over time the concentration profile, the velocity profile and the local dynamics in the gel phase: the settling may be fully described by the local strain rate and dynamics exhibits remarkable scaling properties when time is normalized by strain rate. The role of the solid friction at the cell walls has been investigated by polarimetry experie nces and solving numerically a poroelastic model of gel sedimentation.

Cette thèse concerne le développement de nanocristaux (NCs) cœur/coquille d'InP/ZnS émettant dans le proche infrarouge pour l'imagerie biologique. Dans la synthèse chimique des NCs cœur d'InP, nous avons utilisé la phosphine générée in situ comme précurseur de phosphore en combinaison avec le myristate d'indium comme précuseur d'indium et l'1-octadécène comme solvant. Les NCs obtenus sont hautement cristallins et présentent une fluorescence dans la gamme 720-750 nm, selon leur taille. La croissance d'une ou deux monocouches (coquille) de ZnS sur la surface des NCs d'InP a considérablement amélioré leur rendement quantique de fluorescence. Nous avons de plus étudié le transfert de phase de ces NCs InP/ZnS du milieu organique au milieu aqueux en utilisant diverses molécules hydrophiles contenant un groupe thiol. En particulier, nous nous sommes intéressés au transfert de phase avec des molécules zwitterioniques tels que la penicillamine et la cystéine afin d'obtenir une taille hydrodynamique compacte, et de réduire les interactions non-spécifiques en milieu biologique. Dans l'étude du transfert de phase, l'accent a été mis sur la stabilité colloïdale des NCs et sur la préservation de leur efficacité de fluorescence en milieu aqueux. La cytotoxicité des NCs InP/ZnS fonctionnalisés avec la pencillamine a été évaluée en culture cellulaire. Puis la bio-distribution de ces NCs a été étudiée dans des souris vivantes par imagerie de fluorescence grâce à leur émission dans le proche infrarouge. Pour finir, les fonctionnalisations de NCs InP/ZnS d'une part avec un peptide de pénétration cellulaire, d'autre part avec des agents de contraste IRM (complexes de gadolinium) et enfin avec un nombre contrôlé de molécules streptavidine ont été explorées, démontrant le grand intérêt de ces NCs pour l'imagerie biologique. Mots clés: phosphure d'indium, boîtes quantiques, nanocristaux, imagerie biologique de fluorescence, infrarouge, transfert de phase, fonctionnalisation de surface / This thesis concerns the development of near infrared (NIR) emitting InP/ZnS core/shell nanocrystals for biological imaging. In situ generated phosphine gas was used as the phosphorous precursor, indium myristate as the In precursor and 1-Ocadecene as the solvent to produce InP NCs with emission in the range of 720-750 nm. Growth of 1-2 monolayers of a ZnS shell on the surface of the InP NCs strongly improves their quantum yield. Next, we studied the phase transfer of the obtained InP/ZnS NCs to aqueous medium with various thiol group containing ligands. Emphasis is put on the colloidal stability and the retention of fluorescence quantum efficiency during the transfer. Zwitterionic ligands such as pencillaime and cysteine have been studied in view of their biological interest in providing compact size and reduced non-specific interaction with cells. The cytotoxicity of pencillamine capped InP/ZnS NCs has been evaluated in cell culture. The NIR emitting properties of the QDs have been exploited to their study bio-distribution in mice by fluorescence imaging. In addition, functionalisation of the InP/ZnS NCs with a cell penetrating peptide, with a MRI contrast agent (gadolinium complex) and with a controlled number of streptavidin molecules have been explored to demonstrate the large interest of InP/ZnS NCs in biology. Key words: Indium phosphide, quantum dots, near infrared fluorescence imaging, phase transfer, surface functionalisation

Anion recognition is a topical area of research warranted by the potential applications of anion receptors in environmental and biological monitoring. Anions are indeed widespread in nature, are involved in many biochemical processes and are also major aqueous pollutants. Molecular receptors able to recognize anions with high affinity and selectivity in organic solvents are well documented in the literature but only few systems are efficient in aqueous media. Water is undeniably a particularly challenging solvent to work with due to the competition of water in the recognition process. Moreover, most of the receptors that are known to bind anions with high affinity and selectivity in organic solvents are not soluble in water. One strategy used to make hydrophobic molecular receptors “water-compatible” is micellar incorporation. This strategy is straightforward as no synthetic modifications of the receptor are required and has furthermore been seen to enhance the apparent properties, in particular binding properties, of the receptors. Following previous work undertaken in the laboratory, this thesis was devoted to the study of the micellar incorporation of different anion receptors. The first part of this thesis focused on the potential of Nuclear Magnetic Resonance (NMR) and more precisely Paramagnetic Relaxation Enhancement (PRE) experiments to provide robust information on the localisation of receptors within micelles. We studied the effect of various parameters and were able to rationalize the effect of the nature and concentration of the counterion and of the surfactant concentration on the PRE values obtained with cationic cetyltrimetylammonium (CTAX) micelles. By applying a normalization procedure we were then able to compare different receptor/micelle systems. This work has been reported in the Journal of Physical Chemistry (“Paramagnetic Relaxation Enhancement Experiments: a Valuable Tool for the Characterization of Micellar Nanodevices”, F. Keymeulen, P. De Bernardin, A. Dalla Cort, and K. Bartik, J. Phys. Chem. B, 2013, 117, 11654–11659).The second part of our work consisted in the study of the role played by the surfactant on the efficiency of the supramolecular system formed. Using UV-vis and/or NMR titrations, we studied the impact of the nature of the surfactant (cationic, zwitterionic or neutral) as well as its concentration on the apparent binding affinity for fluoride of two uranyl-salophen receptors. We showed that the supramolecular systems formed with cationic micelles are the most efficient, due to the favourable electrostatic interaction between the positively charged micelle and the fluoride despite the competition of the surfactant counter-ion. The concentration of the cationic surfactant does however have an impact as the apparent affinity decreases with surfactant concentration as a consequence of this non-specific interaction of the guest with the micelles. PRE and DLS (Dynamic Light Scattering) experiments allowed us to better understand the differences between the different types of micelles. This work has been reported in Organic & Biomolecular Chemistry (“Fluoride binding in water with the use of micellar nanodevices based on salophen complexes”, F. Keymeulen, P. De Bernardin, I. Giannicchi, L. Galantini, K. Bartik, and A. Dalla Cort, Org. Biomol. Chem. 2015, 13, 2437–2443).The final part of our study was devoted to the investigation of the applicability of micellar incorporation to other receptors. Two other uranyl-based receptors were studied in cationic CTAX and neutral Triton X-100 micelles. One suffered from chemical stability issues and the other receptor did not perform any better than the ones previously studied. We also studied trimesitylborane which can bind fluoride via Lewis acid-base interactions. This system, which is highly efficient in organic solvents, was shown to be ineffective once incorporated into micelles, probably because the change in the hybridization of the boron atom upon fluoride binding is not favourable in the confined micellar environment. Indolocarbazole-based anion receptors, which recognize acetate and benzoate via hydrogen bonding, were successfully incorporated into DPC micelles, albeit at low concentrations, and were observed to be efficient as the apparent binding affinity measured in water is of the same order of magnitude or higher as the one observed in organic solvents. / Doctorat en Sciences de l'ingénieur et technologie / info:eu-repo/semantics/nonPublished