L’augmentation du nombre d’appareils communicants et du débit de données a pour conséquence une montée en fréquence des dispositifs micro-ondes, notamment dans le secteur du spatial. L’optimisation des modules existants n’est pas toujours suffisante. Il faut donc synthétiser de nouveaux circuits. Cependant, la plupart des méthodes de synthèse existantes, inclues dans les logiciels de CAO, ne prennent pas en compte les modèles à pertes des fondeurs. Or, plus la fréquence de fonctionnement est élevée, plus leurs prises en compte est indispensable. Cette thèse propose une nouvelle méthode de synthèse et de conception pour les circuits faible bruit intégrés (amplificateur faible bruit et mélangeur). Elle prend en compte les modèles à pertes des composants passifs des fondeurs, les lignes de connexion, les jonctions et elle combine plusieurs fonctions comme l’amplification et le filtrage ainsi que le mélange et le filtrage. Elle a été validée en simulation et en mesure. / Due to the evolution of wireless systems and data rate, it is necessary to increase microwave operating frequencies, especially in space industry. Optimization of existing circuit topologies are always not enough and therefore, we need to synthetize new circuits. Unfortunately, most of the existing synthesis methods, including in CAD softwares, are only based on lossless passive component models. With the increase of operating frequency, we need to take the effect of losses in the passive component models during synthesis. This thesis introduces a new synthesis and design method for low noise integrated circuits(low noise amplifier and mixer). Lossy passive component models from foundries, connecting wires, junctions and co-design (amplification and filtering or mixing and filtering)are included. The design procedure was validated by simulations and measurements.

The recent advancement of superconducting nano devices has allowed for making a Superconducting Nanowire Single Photon Detector (SNSPD), whose extraordinary features have strongly motivated the research community to exploit it in many practical applications. In this thesis, an experimental setup for testing the SNSPD has been established. It contains an in-house packaging that meets the requirements of RF/microwave and optoelectronic characterizations. The quantum efficiency and detection efficiency measurements have confirmed that our approach is satisfactory. The dark count performance has reached the anticipated level. The factors affecting rise and fall times of the photoresponses are addressed. Based on the successful setup, the characterizations including dc, small signal ac measurements have been undertaken. The measurements are aimed at quantitatively investigating Cooper pair density in the superconducting nanowire. The experimental method involves a two-step, small signal S-parameter measurement either in the presence or absence of optical powers. The subsequent measurements by varying the temperature and dc bias current have achieved remarkable understanding on the physical properties of SNSPD nanowires. Then, the electrically induced nonlinearity is studied via the large signal RF and Microwave measurements. The experiments are a set of one-tone and two-tone measurements, in which either the RF driving power is varied at a fixed frequency, or vice versa. Two major nonlinear microwave circuit analysis methods, i.e. time-domain transient and hybrid-domain harmonic balance analysis, are employed. The simulation result reveals the optimized conditions of reaching the desired nonlinearity. Finally, we have successfully measured the optoelectronic mixing products in an electrically pumped optoelectronic mixer, which has identical structures as that of the SNSPD. The experiments confirm that this mixer is not only sensitive to the classical light intensities, but also to that of the single photon level. Meanwhile, the quantum conversion matrices is derived to interpret the quantum optoelectronic mixing effects.

The recent advancement of superconducting nano devices has allowed for making a Superconducting Nanowire Single Photon Detector (SNSPD), whose extraordinary features have strongly motivated the research community to exploit it in many practical applications. In this thesis, an experimental setup for testing the SNSPD has been established. It contains an in-house packaging that meets the requirements of RF/microwave and optoelectronic characterizations. The quantum efficiency and detection efficiency measurements have confirmed that our approach is satisfactory. The dark count performance has reached the anticipated level. The factors affecting rise and fall times of the photoresponses are addressed. Based on the successful setup, the characterizations including dc, small signal ac measurements have been undertaken. The measurements are aimed at quantitatively investigating Cooper pair density in the superconducting nanowire. The experimental method involves a two-step, small signal S-parameter measurement either in the presence or absence of optical powers. The subsequent measurements by varying the temperature and dc bias current have achieved remarkable understanding on the physical properties of SNSPD nanowires. Then, the electrically induced nonlinearity is studied via the large signal RF and Microwave measurements. The experiments are a set of one-tone and two-tone measurements, in which either the RF driving power is varied at a fixed frequency, or vice versa. Two major nonlinear microwave circuit analysis methods, i.e. time-domain transient and hybrid-domain harmonic balance analysis, are employed. The simulation result reveals the optimized conditions of reaching the desired nonlinearity. Finally, we have successfully measured the optoelectronic mixing products in an electrically pumped optoelectronic mixer, which has identical structures as that of the SNSPD. The experiments confirm that this mixer is not only sensitive to the classical light intensities, but also to that of the single photon level. Meanwhile, the quantum conversion matrices is derived to interpret the quantum optoelectronic mixing effects.

Future wireless communication devices must support multiple standards and features on a single-chip. The trend towards software-defined radio requires flexible and efficient RF building blocks which justifies the adoption of broadband receiver front-ends in modern and future communication systems. The broadband receiver front-end significantly reduces cost, area, pins, and power, and can process several signal channels simultaneously. This research is mainly focused on the analysis and realization of the broadband receiver architecture and its various building blocks (LNA, Active Balun-LNA, Mixer, and trans-impedance amplifier) for multi-standard applications. In the design of the mobile DTV tuner, a direct-conversion receiver architecture is adopted achieving low power, low cost, and high dynamic-range for DVB-H standard. The tuner integrates a single-ended RF variable gain amplifier (RFVGA), a current-mode passive mixer, and a combination of continuous and discrete-time baseband filter with built-in anti-aliasing. The proposed RFVGA achieves high dynamic-range and gain-insensitive input impedance matching performance. The current-mode passive mixer achieves high gain, low noise, and high linearity with low power supplies. A wideband common-gate LNA is presented that overcomes the fundamental trade-off between power and noise match without compromising its stability. The proposed architecture can achieve the minimum noise figure over the previously reported feedback amplifiers in common-gate configuration. The proposed architecture achieves broadband impedance matching, low noise, large gain, enhanced linearity, and wide bandwidth concurrently by employing an efficient and reliable dual negative-feedback. For the wideband Inductorless Balun-LNA, active single-to-differential architecture has been proposed without using any passive inductor on-chip which occupies a lot of silicon area. The proposed Balun-LNA features lower power, wider bandwidth, and better gain and phase balance than previously reported architectures of the same kind. A surface acoustic wave (SAW)-less direct conversion receiver targeted for multistandard applications is proposed and fabricated with TSMC 0.13?m complementary metal-oxide-semiconductor (CMOS) technology. The target is to design a wideband SAW-less direct coversion receiver with a single low noise transconductor and current-mode passive mixer with trans-impedance amplifier utilizing feed-forward compensation. The innovations in the circuit and architecture improves the receiver dynamic range enabling highly linear direct-conversion CMOS front-end for a multi-standard receiver.

Au-dessus d’une certaine température appelée température de trouble (Tc), les solutions aqueuses de la majorité des tensioactifs non ioniques polyéthoxylés se séparent en deux phases liquides en équilibre : la phase diluée et le coacervat. Grâce à la solubilisation micellaire de composés hydrophobes, amphiphiles ou même ioniques et à leur concentration dans le (faible) volume de coacervat, une extraction à deux phases aqueuses (extraction par point de trouble ou par coacervat)peut être réalisée et appliquée à des opérations de dépollution d'effluents industriels ou à la concentration ou encore à la séparation de substances à haute valeur ajoutée. L’extraction par point de trouble (CPE) est une technique relativement simple et écologique pour l'élimination des matières toxiques de l'environnement, qui s’est avérée efficace dans le traitement de divers contaminants (organiques et inorganiques dissous ou dispersés) de l’eau. Une première partie de cette thèse consiste à rappeler quelques notions sur la pollution industrielle des eaux et quelques généralités sur les tensioactifs (TA) et sur l’extraction liquide-liquide. Elle est suivie d’une description des réactifs, du matériel et des méthodes utilisées au cours de ce travail, comme préliminaire à la mise au point d'un procédé d'extraction par coacervat. Quelques propriétés thermodynamiques superficielles (adsorption) et d’association (micellisation) de deux tensioactifs non ioniques industriels (le Simulsol NW342 et le Tergitol 15-S-7) ont été déterminées. Les courbes de démixtion des systèmes binaires (eau/TA) ainsi que l’effet de divers additifs (sel, composés organiques, tensioactifs ioniques) sur le point de trouble sont étudiés. Le diagramme isotherme du système ternaire (eau/TA/ phénol) est tracé. Une application du modèle de Flory-Huggins-Rupert pour la prédiction des courbes de démixtion des tensioactifs non ioniques a été expérimentée. L’extraction à un seul contact, à partir de solutions modèles, utilise des alcools oxo éthoxylés biodégradables (Simulsol NW342 et Tergitol 15-S-7) pour des polluants organiques dissous (phénol, 1-phényléthanol et alcool benzylique) et un mélange de tensioactifs non ionique (Simulsol NW342) et ionique (SDS ou CTAB) pour des polluants métalliques solubles (plomb(II), molybdène(VI)). On cherche le meilleur compromis entre le pourcentage de soluté extrait (E%),la fraction volumique du coacervat (фv), et les pourcentages de soluté et de tensioactif restants dans la phase diluée (Xs,d et XTA, d), en utilisant un plan d'expériences de type Scheffé et un lissage empirique des courbes. Les résultats sont très prometteurs car les pourcentages d’extraction varient de 60 à 95% pour les solutés organiques et de 40 à 85% pour les solutés métalliques, les meilleures performances étant obtenues pour le phénol et le plomb. D’autre part, il est possible, en jouant sur le pH, d’améliorer les séparations et de recycler le tensioactif après désextraction des solutés. Les cinétiques d’extraction, de séparation et de clarification ont été aussi étudiées pour une meilleure compréhension de ces systèmes. Enfin, l’extraction continue du phénol à partir du mélange eau/4%Simulsol NW342/0,2%phénol (mass.) a été testée sur deux appareillages (extracteur centrifuge et mélangeurdécanteur)thermostatés. Dans un procédé multi-étagé à courants croisés sur un mélangeur-décanteur, on arrive à réduire la concentration du phénol à moins de 0,3ppm (concentration limite selon la législation) après six étages. / Above a certain temperature called cloud point (Tc), aqueous solutions of most nonionic polyethoxylated surfactants separate into two liquid phases in equilibrium: the dilute phase and the coacervate. Thanks to the micellar solubilization of hydrophobic, amphiphilic or even ionic compounds and their concentration in the low volume of coacervate, two-aqueous phase extraction (cloud-point or coacervate extraction) can be performed and applied to the removal of pollutants from aqueous industrial effluents or to the concentration or even separation of high added-value chemicals. Cloud point extraction (CPE) is a relatively simple and ecologically-safe technique for the removal of toxic materials from the environment; this process has proved efficient in treating water for various contaminants including dissolved or dispersed organic and inorganic chemicals. The first part of this thesis consists of recalling some notions on industrial wastewater, some generalities about surfactants (TA) and liquid-liquid extraction, followed by a description of all the reagents, materials and methods used in this work as a preliminary study of a coacervate extraction process. Some surface thermodynamic (adsorption) and association (micellization) properties of two industrial nonionic surfactants (Simulsol NW342 and Tergitol 15-S-7) were determined. The cloud point curves of water /TA binary systems are drawn and the effect of various additives (salt, organic compounds, ionic surfactants) on the cloud point is studied. The isothermal diagram of a water/TA/phenol ternary system is drawn. An application of the Flory-Huggins-Rupert model for the prediction of cloud point curves of nonionic surfactants is discussed. Single contact extraction, from model solutions, uses biodegradable polyethoxylated nonionic surfactants (Simulsol NW342 and Tergitol 15-S-7) for dissolved organic pollutants (phenol, 1-phenylethanol and benzyl alcohol) and mixed micelles of nonionic (Simulsol NW342) and ionic (SDS, CTAB) surfactants for soluble electolytes (lead (II), molybdenum (VI)). We search for the best compromise between the percentage of solute extracted (E%), the coacervate volume fraction (фc) and the percentages of solute and surfactant remaining in the dilute phase (Xs,d and XTA,d), These experimental results are subject to an empirical smoothing through a Scheffé-type experimental design, and an empirical curve fitting procedure. The results are very promising, due to the percentage of solute extracted, which varies between 60 and 95% for organic solutes and from 40 to 85% for electrolytes, the best performances being obtained for phenol and lead. On the other hand, it is possible, by adjusting pH, to improve the separation and recycle the surfactant after back-extraction. The kinetics of extraction yield and phase separation and clarification were also investigated for a better understanding of these systems. Finally, the continuous extraction of phenol from a model solution (water/4wt.% Simulsol NW342/0.2wt.% phenol) on two thermostated equipments (a centrifugal extractor and a mixer-settler) was attempted. In a multi-stage process on a cross-current mixer-settler, the concentration of residual phenol in the dilute phase could be reduced to less than 0.3 ppm (concentration allowed by standard European regulations) after six stages.

Submitted by HUMBERTO FABRIZZI DE FIGUEIREDO PUPO null (betopupo@gmail.com) on 2017-08-31T20:25:11Z No. of bitstreams: 1 Tese_Humberto_Pupo.pdf: 4727522 bytes, checksum: bc9418ef91885cfe96ab6372394b80b6 (MD5) / Approved for entry into archive by Luiz Galeffi (luizgaleffi@gmail.com) on 2017-09-01T14:51:09Z (GMT) No. of bitstreams: 1 pupo_hff_dr_bot.pdf: 4727522 bytes, checksum: bc9418ef91885cfe96ab6372394b80b6 (MD5) / Made available in DSpace on 2017-09-01T14:51:09Z (GMT). No. of bitstreams: 1 pupo_hff_dr_bot.pdf: 4727522 bytes, checksum: bc9418ef91885cfe96ab6372394b80b6 (MD5) Previous issue date: 2017-03-03 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / A utilização dos resíduos sólidos como matéria-prima pode diminuir consideravelmente a extração de recursos naturais, diminuindo assim o impacto sobre o meio ambiente. Dentre os resíduos, os plásticos estão entre os que mais apresentam problemas, pois, mesmo quando descartados corretamente nos aterros sanitários, causam danos ao formarem uma camada impermeável que dificulta a compactação do lixo, afetando as trocas de líquidos e gases gerados e reduzindo a eficiência do processo de biodegradação da matéria orgânica. Diante dos problemas, este trabalho teve por objetivos avaliar a viabilidade técnica de utilização de dois tipos de processos: a extrusora dupla rosca e o misturador termocinético (Drais), na produção de compósitos de BOPP (Polipropileno Biorientado metalizado), reforçados com resíduos de cascas, de eucalipto (CE) e de arroz (CA). Das misturas entre esses resíduos, foram confeccionados 10 tratamentos, de onde se extraíram os corpos de provas, moldados por injeção. Foram avaliadas as propriedades mecânicas (flexão – módulo de ruptura-MOR e módulo de elasticidade-MOE, resistência à tração, resistência ao impacto e dureza Shore D), térmicas (avaliação da condutividade térmica e isolamento térmico), termomecânica (temperatura de deflexão sob aquecimento-HDT) e morfológica, com microscópio estéreo e aumento de 50 vezes para analisar a estrutura interna do corpo de prova, depois de fraturado pelo ensaio de impacto. Para análise das propriedades mecânicas dos corpos de provas, foram utilizadas como referência as normas ASTM D638-14, ASTM D790-10, ASTM D256-10 e a ASTM D-2240-15, para os ensaios de tração, flexão, impacto e dureza Shore D, respectivamente. Para as análises térmicas foi utilizada a norma ASTM E1530-11. Para o HDT, foi utilizada a norma ASTM D648-16. Os resultados mostraram influência de todos os parâmetros analisados nas propriedades dos compósitos. O uso de reforço (CE e CA) nos dois tipos de processos melhorou as propriedades de flexão e de resistência à tração em relação à matriz sem reforço e piorou as propriedades de resistência ao impacto e de dureza Shore D. Com relação à matriz sem reforço, a extrusora mostrou influência no ensaio de resistência à tração. Para adição de casca de eucalipto, o misturador termocinético (Drais) mostrou influência no ensaio de MOE. Já para adição de casca de arroz, a drais mostrou influência nos ensaios de flexão – MOR e MOE e na resistência à tração. Houve melhor adesão entre fibra e matriz nos compósitos confeccionados por extrusão em relação aos confeccionados na drais. As CE apresentaram melhor adesão com a matriz, em relação às CA. Os vazios – bolhas de ar foram observados em todos os tratamentos, com exceção dos tratamentos com ausência de reforço (BOPP 100). Esses vazios foram observados em maior proporção nos compósitos com CE em relação aos compósitos com CA, e também nos compósitos processados na extrusora em relação à drais. Os melhores resultados foram obtidos utilizando o misturador termocinético (Drais) e para as maiores proporções de cascas usadas ou na ausência delas. Porém, pelo fato de o processo utilizado na Drais ser por batelada, a extrusora se torna mais eficiente, levando em consideração o rendimento da produção. Os melhores resultados foram com a utilização de casca de eucalipto. Para os ensaios de flexão MOR, MOE e de resistência à tração, apresentaram interação entre os processos e os resíduos testados. Confirmou-se a viabilidade técnica de utilização de resíduos sólidos industriais (BOPP e CE) e agroindustrial (CA) na produção de compósitos fibra-polímero, sendo que o controle dos processos utilizados é fundamental para a qualidade do produto final. / The use of solid waste as raw material can considerably reduce the extraction of natural resources, thus reducing the impact on the environment. Among residues, plastics are among the most problematic ones because, even when they are properly disposed in landfills, they cause damage by forming a barrier layer which makes it difficult to compact the garbage, affecting the exchange of generated liquids and gases, reducing the efficiency of the biodegradation process of the organic matter. Before such problems, the objective of this work was to evaluate the technical feasibility of two types of processes, the double screw extruder and the thermokinetic mixer - dryser, in the production of BOPP (Bioriented metallized Polypropylene) composites, strengthened by residues of barks eucalyptus (CE) and rice (CA). From the mixtures among these residues 10 treatments have been prepared, from where the test bodies were extracted, molded by injection. Mechanical properties (flexurerupture modulus-MOR and elasticity modulus-MOE, tensile strength and impact strength towards Shore D hardness), as well as thermal (thermal conductivity and thermal insulation evaluation), thermo-mechanical (heat deflexion temperature-HDT) and morphological characteristics have been evaluated by using a stereo microscope and 50x magnification to analyze the internal structure of the test body, after being fractured by the impact test. For analyzing the mechanical properties of the proof bodies the standards ASTM D638-14, ASTM D790-10, ASTM D256-10 and ASTM D-2240-15 have been used as a reference for tensile, bending, impact and Shore D hardness tests, respectively. For the thermal analyzes, the ASTM E1530-11. For the HDT the ASTM D648-16 standard was used. The results showed the influence from all the analyzed parameters on the composites properties. The use of reinforcement (CE and CA) in both types of processes have improved the flexure and tensile strength properties in relation to the non-reinforced matrix and has worsened the impact strength properties as well as the Shore D hardness. Concerning the non-reinforced matrix, the extruder showed influence on the tensile strength test. For the addition of CE, the thermokinetic mixer - dryser has showed influence in the flexure modulus test. As for rice husk addition, dryser has showed influence in flexure (stress tests on maximum strengthMOR and elasticity modulus-MOE and tensile strength). There was a better adhesion between fiber and matrix, in the composites made by extrusion, compared to those made in the dryser. The CE has shown better adhesion with the matrix, compared to CA. The voids - air bubbles have been observed in all treatments with no reinforcement (BOPP 100). These voids have been more often observed in composites with CE, compared to the composites with CA and also in composites processed in the extruder, in relation to the dryser. The best results were obtained by using the thermokinetic mixer - dryser and for the largest proportions of the barks or in their absence. However, because the process used in Drais is by batch, the extruder becomes more efficient, taking into account the yield of the production. The best results have been from the use of CE. For flexure - MOR and MOE and tensile strength, there was an interaction among the processes and the tested residues. The technical feasibility of the use of industrial solid wastes (BOPP and CE) and agroindustrial (CA) in the production of fiberpolymer composites has been confirmed, and the control of the processes used is fundamental for the quality of the final product.

Made available in DSpace on 2014-12-17T15:01:42Z (GMT). No. of bitstreams: 1 WillaciEFJ.pdf: 1499759 bytes, checksum: 03aff567e782194c97d56cdcac63888c (MD5) Previous issue date: 2006-12-19 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / The aim of this work is to use a new technology in the treatment of produced wastewaters from oil industry. An unit for treat produced waters called UTMDIF, was designed, installed and operated in an industrial plant for treatment of effluents from oil industry. This unit operates by means of the method of separation of phase inversion and can become a promising alternative to solve the problem of oil/water separation. This method constitutes the basis of the working of a new design of mixersettler of vertical configuration which occupies small surface area. The last characteristic becomes specially important when there is limitation on the lay-out of the plant, for example, over maritime platforms to explore oil. This equipment in a semi-industrial scale treats produced wastewaters contaminated with oil at low concentrations (ranging from 30 to 150 mg/L) and throughputs of 320 m3/d (47,4 m3 m-2 h-1). Good results were obtained in oil/water separation which leads to the necessary specification to discharge those wastewaters. Besides, the non dependence of the efficiency of separation in spite of the salinity of the medium becomes the equipment an attractive new technology to treat wastewaters containing oil at low concentrations / Este trabalho tem por objetivo utilizar uma nova tecnologia no tratamento de ?guas produzidas da ind?stria do petr?leo. Uma Unidade de Tratamento de ?guas Produzidas, denominada UT-MDIF, foi projetada, instalada e operou dentro de uma planta industrial de tratamento de efluentes de petr?leo. Esta Unidade de Tratamento piloto utiliza o m?todo de separa??o ? invers?o de fases e pode se tornar numa alternativa promissora para resolver o problema de separa??o ?leo/?gua. Este m?todo constitui a base do funcionamento de um novo modelo de misturador-decantador de configura??o vertical e que ocupa pequena ?rea superficial. Esta ?ltima caracter?stica se torna especialmente importante, quando existe limita??o de espa?o no lay-out da planta industrial, por exemplo, uma plataforma mar?tima para explora??o de petr?leo. Este equipamento em escala semi-industrial trata ?guas produzidas contaminadas com petr?leo em baixas concentra??es da ordem de 30 a 150 mg/L e vaz?es efetivas de 320 m3/d (47,4m3m-2h-1). Os bons resultados na separa??o ?leo/?gua, obtendo-se a especifica??o necess?ria para descarte, associado ? n?o depend?ncia da efici?ncia de separa??o face ? salinidade do meio, podem tornar o equipamento uma nova tecnologia no tratamento de ?guas contaminadas com ?leo a baixas concentra??es

Made available in DSpace on 2014-12-17T15:01:52Z (GMT). No. of bitstreams: 1 NorbertoAM_TESE.pdf: 2336588 bytes, checksum: 940ad1b0befe8c431d2e9fb731a885b6 (MD5) Previous issue date: 2011-09-30 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / This work aims to determine a better methodology to help predicting some operational parameters to a new design of mixer-settler on treating wastewater produced by petroleum industry, called MDIF (Misturador-Decantador ? Invers?o de Fases/ Mixer-Settler based on Phase Inversion MSPI). The data from this research were obtained from the wastewater treatment unit, called MSPI-TU, installed on a wastewater treatment plant (WTP) of PETROBRAS/UO-RNCE. The importance in determining the better methodology to predict the results of separation and extraction efficiency of the equipment, contributes significantly to determine the optimum operating variables for the control of the unit. The study was based on a comparison among the experimental efficiency (E) obtained by operating MSPI-TU, the efficiency obtained by experimental design equation (Eplan) from the software Statistica Experimental Design? (version 7.0), and the other obtained from a modeling equation based on a dimensional analysis (Ecalc). The results shows that the experimental design equation gives a good prediction of the unit efficiencies with better data reliability, regarding to the condition before a run operation. The average deviation between the proposed by statistic planning model equation and experimental data was 0.13%. On the other hand, the efficiency calculated by the equation which represents the dimensional analysis, may result on important relative deviations (up 70%). Thus, the experimental design is confirmed as a reliable tool, with regard the experimental data processing of the MSPI-TU / Este trabalho objetiva determinar a melhor metodologia para auxiliar na predi??o de vari?veis operacionais de um novo design de misturador-decantador aplicado ao tratamento de ?guas produzidas da ind?stria do petr?leo, que vem sendo denominado de MDIF (Misturador-Decantador ? Invers?o de Fases). Os dados utilizados neste estudo foram obtidos da Unidade de Tratamento de ?guas produzidas, denominada de UT-MDIF, que se encontra instalada em uma das esta??es de tratamento de efluentes (ETE) da PETROBRAS/UORNCE. A import?ncia em determinar a melhor metodologia que permita a predi??o dos resultados das efici?ncias de separa??o e extra??o do equipamento, contribui de forma significativa para otimiza??o das vari?veis operacionais para o controle da unidade. O estudo se baseou na compara??o dos resultados das efici?ncias experimentais (E) obtidas pela UTMDIF, com os resultados das efici?ncias obtidas pela equa??o do modelo do planejamento estat?stico (Eplan), gerado pelo programa Statistica Experimental Design? (vers?o 7.0), e com a utiliza??o de uma equa??o utilizando modelagem com base em an?lise dimensional (Ecalc). Os resultados mostram que a equa??o decorrente do planejamento estat?stico ? capaz de predizer, com boa aproxima??o, os resultados de efici?ncia a serem obtidos com a unidade, possibilitando uma melhor confiabilidade dos dados, no que se refere ? condi??o que antecede uma corrida operacional. O desvio m?dio entre a equa??o proposta e os dados experimentais ? de 0,13%. Por outro lado, a efici?ncia calculada pela equa??o da an?lise dimensional pode apresentar desvios relativos importantes (at? 70%). Desta forma, o planejamento estat?stico experimental se confirma como sendo uma ferramenta confi?vel, no que se refere ao tratamento dos dados experimentais da UT-MDIF

La conversion de fréquence est une fonction clef présente dans divers contextes, particulièrement dans les systèmes mixtes photoniques-hyperfréquences. Aujourd’hui, la suprématie des réseaux optiques pour le transport de données à haut débit sur de grandes distances incite à l’intégration de telles fonctions dans le domaine optique afin de bénéficier des faibles pertes, larges bandes passantes, faibles poids et tailles propres aux technologies optiques. Dans ce travail, nous étudions un mélangeur tout-optique utilisant un composant SOA-MZI (Semiconductor Optical Amplifier Mach-Zehnder Interferometer) et une technique d’échantillonnage permettant la conversion vers les hautes et les basses fréquences. Le principe du mélange exploite les caractéristiques spectrales d’un signal échantillonné pour lequel des répliques du signal d’origine existent à différentes autres fréquences. Utiliser une telle technique pour la conversion de fréquences offre deux avantages : la conversion vers les hautes et les basses fréquences utilise la même configuration du mélangeur et la fréquence de l’oscillateur local peut être inférieure à la gamme des fréquences visées.L’implémentation d’une telle technique d’échantillonnage nécessite un interrupteur optique contrôlé optiquement.Comme cela est montré dans ce travail, un SOA-MZI peut jouer ce rôle. Selon la phase relative entre ses bras, un interféromètre Mach-Zehnder (MZI) peut transmettre ou non un signal optique d’entrée. En plaçant un SOA dans chaque bras de la structure MZI, la modulation croisée de la phase qui existe au sein d’un SOA est mise à profit pour contrôler l’état de l’interféromètre. Contrôlé par une source impulsionnelle optique, cet interrupteur optique permet d’échantillonner un signal optique porteur de données à modulation complexe. La conversion de fréquence de signaux mono et multi-porteuses dans le domaine 0,5-39,5 GHz a été obtenue avec succès. Par utilisation d’une configuration différentielle du SOA-MZI, des conversions vers les hautes et les basses fréquences jusqu’à un débit de 1 Gb/s ont pu être réalisées. / Frequency mixing is a key function existing in different systems, especially in mixed photonic-microwave ones. Today, the supremacy of optical networks to carry high bitrate data over large distances motivates the optical implementation of such functions to benefit from the low loss, high bandwidth, low size and weight of optical technologies. In this work, we study a photonic mixer based on a SOA-MZI (Semiconductor Optical Amplifier Mach-Zehnder Interferometer) device and a sampling technique allowing both conversion towards high and low frequencies.The involved mixing principle exploits the spectral characteristics of a sampled signal in which replicas of the original spectrum exist at different other frequencies. Basing the frequency conversion on a sampling technique gives two advantages: the photonic mixer configuration is the same for up and down conversions, and the frequency of the local oscillator can be less than the addressed frequency range.The implementation of such a sampling technique needs an optically-controlled high-frequency optical switch. As shown in this work, a SOA-MZI can play this role. Depending on the relative phase between its arms, an interferometric structure (MZI) can transmit or cancel an optical input signal. By locating one SOA in each arm of the MZI structure, the cross-phase modulation that exists inside an SOA is exploited to optically control the optical switch state of the MZI.Controlled by an optical pulse source, this optical switch is able to sample an optical input signal carrying complexmodulated data. Frequency conversions of mono and multi-carrier signals in the range 0.5-39.5 GHz have been successfully achieved. By using a differential configuration of the SOA-MZI, both up and down conversions at bitrates up to 1 Gb/s are reached.

La modélisation des phénomènes de brisure de gouttes lors d’opérations d’émulsification par bilan de population (PBE), a pour but de suivre l’évolution de la distribution des tailles de gouttes (DSD). Ceci a fait l’objet d’un grand nombre d’études au cours des deux dernières décennies. Une approche multiéchelle, couplant la modélisation des phénomènes de brisure à l’échelle d’une goutte avec les phénomènes agissant à l’échelle du champ d’écoulement est nécessaire pour simuler correctement les procédés d’ émulsification dans des géométries complexes tels que des mélangeurs statiques ou des homogénéisateurs à haute pression. Une telle approche est présentée dans cette thèse par l’emploi d’un couplage entre PBE et mécanique des fluides numériques (CFD). Trois types de procédés d’émulsification huile dans l’eau ont été étudiés : une cuve agitée de deux litres,équipée avec d’une hélice Mixel-TT générant un écoulement axial de la phase continue, pour deux systèmes modèles : Di-Stereate d’éthylène Glycol (EGDS) dans l’eau d’une part, huiles silicones de différentes viscosités d’autre part. Un montage expérimental sur mesure a été conçu pour l’émulsification d’huiles silicones dans eau basé sur l’emploi de mélangeurs statiques de type SMX+. Des expériences d’émulsification des huiles végétales de qualité alimentaire dans un homogénéisateur à haute pression (HPH) ont été réalisées dans le laboratoires d’UNILEVER R&D à Vlaardingen, Pays-Bas. Deux techniques d’analyse granulométrique in-situ ont été comparées aux résultats obtenus par la technique ex situ de diffraction laser : une sonde vidéo avec traitement automatisé d’images (basé sur la transforméecirculaire de Hough), et une sonde de réflectance laser « Focused Beam Reflectance Measurement »(FBRM), qui mesure la distribution de cordes (CLD). Les sondes ont été introduites dans la cuve agitée et une cellule de mesure a été conçue spécialement pour implanter en-ligne la sonde video et mesurer ainsi la DSD en amont et en aval des mélangeurs statiques. La technique FBRM n’a pas permis de détecter les plus grosses gouttes et la transformation de la CLD en DSD donnait une sous-estimation de la taille des gouttes. Cette méthode n’est par conséquent pas adaptée à l’analyse granulométrique des gouttes transparentes, telles que les huiles silicones. Par contre, la détection des gouttes sur les images prises par la sonde vidéo, permet de produire des mesures fiables de la DSD pour des concentration dephase dispersée faible (≤ 10 %). L’algorithme de détection a été amélioré pour être capable de mesurerla DSD des émulsions avec 10 – 20 % de phase dispersée. La partie modélisation de cette thèse se compose premièrement de l’élaboration d’un nouveau modèlede brisure qui est capables de représenter l’effet de la viscosité de la phase dispersée. Ce modèle est une amélioration du modèle phénoménologique proposé par Luo & Svendsen (1996). Deuxièmement le couplage entre des PBE discrétisées par volumes finis avec la CFD en régime turbulent a été réalisé, dans le code open-source OpenFOAM (OpenCFD). / The modelling of breakage phenomena with the goal to simulate the evolution of drop size distributions(DSDs) in turbulent emulsification by Population Balance Equation (PBE) modeling has been an activearea of research over the last decade. A multi-scale approach, combining the breakage phenomena on the droplet scale with the larger scale flow-field characteristics is necessary to accurately simulate emulsification in complex geometries such as High-Pressure homogenizers and static mixers. Emulsifications were performed for Ethylene Glycol Di-Stearate-in-water and Silicone oil-in-water systemsin a stirred tank reactor, using an axial-flow Mixel-TT impeller, as well as SMX+ (Sulzer) static mixers for the silicone oil system at the LAGEP, Lyon, France. Emulsifications of food-grade vegetableoils in a High-pressure Homogenizer were performed at UNILEVER R&D, Vlaardingen, Netherlands. Two in-situ DSD measurement techniques were compared to results obtained form laser diffraction measurements of samples. Focused Beam Reflectance Measurement (FBRM), which generates achord length distribution was found give an under-prediction of the DSD and failed to detect the largerdroplets of the transparent silicone oils. This technique, while providing a continuous, in situ measurement of the DSD is not reliable for measuring transparent droplets. An in situ video probe with off-line droplet detection via Hough transform, developed at LAGEP, was found to give reliable and traceable DSD measurements for dilute emulsions. The image detection algorithm was improved to be capable of measuring droplets in emulsions with 10 – 20 % dispersed phase hold-up fraction.The modelling part of this thesis consists of the development of a framework for the coupling of PBEand CFD modelling, as well as a significant improvement to the well-known Luo & Svendsen (1996)breakage model. Different breakage models were compared for their applicability to emulsification ofEthylene Glycol di-Stereate (EGDS) and silicone oil in water emulsions in a 2-L stirred tank reactor. This analysis revealed the need for a phenomenological breakage model which does not rely on system dependent parameters and is able to accurately take the dispersed phase viscosity effects into account.Such a model was proposed, based on the Han et al. (2011) to the Luo & Svendsen (1996) framework. A dispersed phase viscosity term was added to the surface energy opposing breakage. This was validated by experimental data from emulsification of silicone oils with varying viscosities (20, 50, 100, 350 mPa.s). The new model was found to provide better predictions than the Alopaeus et al. (2002) and Vankova etal. ( 2007) breakage models, without the need for empirically determined parameters. The one-way coupling of CFD and PBE modelling was implemented in the open-source finite volumes software package OpenFOAM. This was applied to emulsification of vegetable oils with varying viscosities(25, 50, 100 mPa.s) in a Niro-soave bench-scale HPH. The new model was found to deliver good predictions for the drop size distribution after three consecutive passes through the HPH valve at the three different viscosities and varying pressure drops (200, 400, 600 bar).