Twin screw wet granulation has been proved as a feasible alternative for traditional batch granulation process due to its continuous processing feature; considered as a significant processing method especially in the pharmaceutical industry. This thesis will explore the processibility of twin screw wet granulation with various formulations. The first section of the thesis focused on examining the processibility of wet granulation in a twin screw while using various grades of hydroxypropyl methylcellulose (HPMC) as an extended-release excipient. The method to find the processibility was by modifying the liquid-to-solid ratio for each formulation. The process window was defined by examining the amount of granules that fall in a pre-determined acceptable size range. This part focused on three substitution types of HPMC (Type 2910, Type 2208, and Type 2906) that varied in molecular weights. It was found that only Type 2910 HPMC showed a shift in the process window (also known as granulation range) in relation to the molecular weight of the formulations. A higher demand for binder liquid was found for higher molecular weight Type 2910 HPMC in order to form granules with acceptable sizes. The second part of this thesis was focused on understanding the process variables that might influence the processibility of the HPMC formulations. This part examined the impact of feed rate on the granulation range of Type 2910 HPMC specifically was examined. Multiple feed rates were tested, and it was found that the granulation range for lower molecular weight Type 2910 HPMC was easier to shrink when higher feed rates were applied. A transition in the granule formation method from liquid-bridging to compaction with respect to the feed rate was found for low molecular weight HPMC, whereas the high molecular weight HPMC always formed granules through compaction at all feed rates due to strong water retainability. / Thesis / Master of Applied Science (MASc)

Complexity of production processes, high computing capabilities, and massive datasets characterize today’s manufacturing environments, such as those of continuous andbatch production industries. Continuous production has spread gradually acrossdifferent industries, covering a significant part of today’s production. Commonconsumer goods such as food, drugs, and cosmetics, and industrial goods such as iron,chemicals, oil, and ore come from continuous processes. To stay competitive intoday’s market requires constant process improvements in terms of both effectivenessand efficiency. Statistical process control (SPC) and design of experiments (DoE)techniques can play an important role in this improvement strategy. SPC attempts toreduce process variation by eliminating assignable causes, while DoE is used toimprove products and processes by systematic experimentation and analysis. However,special issues emerge when applying these methods in continuous process settings.Highly automated and computerized processes provide an exorbitant amount ofserially dependent and cross-correlated data, which may be difficult to analyzesimultaneously. Time series data, transition times, and closed-loop operation areexamples of additional challenges that the analyst faces.The overall objective of this thesis is to contribute to using of statisticalmethods, namely SPC and DoE methods, to improve continuous production.Specifically, this research serves two aims: [1] to explore, identify, and outlinepotential challenges when applying SPC and DoE in continuous processes, and [2] topropose simulation tools and new or adapted methods to overcome the identifiedchallenges.The results are summarized in three appended papers. Through a literaturereview, Paper A outlines SPC and DoE implementation challenges for managers,researchers, and practitioners. For example, problems due to process transitions, themultivariate nature of data, serial correlation, and the presence of engineering processcontrol (EPC) are discussed. Paper B further explores one of the DoE challengesidentified in Paper A. Specifically, Paper B describes issues and potential strategieswhen designing and analyzing experiments in processes operating under closed-loopcontrol. Two simulated examples in the Tennessee Eastman (TE) process simulatorshow the benefits of using DoE techniques to improve and optimize such industrialprocesses. Finally, Paper C provides guidelines, using flow charts, on how to use thecontinuous process simulator, “The revised TE process simulator,” run with adecentralized control strategy as a test bed for developing SPC and DoE methods incontinuous processes. Simulated SPC and DoE examples are also discussed.

Le travail présenté dans ce mémoire de thèse s’est inscrit dans le cadre du développement de la technologie flamme oxyacétylénique pour la synthèse de nanotubes de carbone (NTC) au Laboratoire de Physique et Mécanique Textile. La simplicité et l’originalité de ce procédé en font un candidat très sérieux pour envisager la mise en œuvre d’un pilote industriel pour la production de fibres de renforcement multidimensionnelles, notamment composées de fibres de carbone sur lesquelles ont cru de nanotubes de carbone. Ce travail a consisté à :- Réaliser une étude bibliographique sur les procédés de croissance de NTC,- Etablir la preuve de concept de la croissance des nanotubes sur des fibres,- Concevoir et réaliser un dispositif assurant une bonne maîtrise de la croissance,- Mettre au point le procédé de croissance sur des fibres,- Identifier les principaux paramètres influençant la qualité et la quantité des nanotubes,- Caractériser les nanotubes obtenus,- Faire croître ces nanotubes sur des fibres de carbone,- Intégrer des nouveaux matériaux multidimensionnels dans des matrices afin de réaliser des matériaux composites structurels,- Caractériser ces matériaux,- Décrire les mécanismes de croissance dans la flamme.Notre effort a porté sur le traitement des fibres avant exposition à la flamme et à évaluer les conditions de croissance des NTC en faisant varier notamment, la température d’exposition des fibres et la qualité des catalyseurs de croissance. Après synthèse des NTC sur les fibres de carbone et leur caractérisation nous avons réalisé des matériaux composites. Nous avons mesuré que les NTC améliorent significativement le module d’Young des composites mais altèrent sensiblement la contrainte à la rupture. Les propriétés électriques longitudinales et transversales sont améliorées d’un facteur 8 et 5 respectivement. Nous avons proposé des mécanismes de croissance des NTC. Ces mécanismes sont directement en relation avec propriétés physiques et chimiques des particules de catalyseur. / The research presented in this work aims to develop the oxyacetylene flame method for the Carbon Nanotubes (CNT) synthesis at the Laboratory of Physics and Mechanics of Textiles. The simplicity and the degree of innovation of this process make of it a serious candidate for manufacturing a pilot in order to produce new kind of tridimensional material made of CNT having grew on carbon fibres. This work consisted of:- Make a bibliographic study,- Establish a proof of concept of the growth of CNT,- Design and manufacture a device allowing process control,- Setup the process of growth on the fibres,- Identify the main parameters influencing CNT quality and quantity,- Characterize CNT,- Assume the CNT growth on carbon fibers,- Integrate these multidimensional materials into an organic matrix to realize structural composite materials,- Characterize these materials,- Describe and explain the growth mechanism in the flame.First, we focused our work on the fibres chemical treatment before flame exposition to evaluate the NTC growth conditions by varying notably, the fibres exposition temperature and the quality of the catalysts. After, the NTC syntheses on carbon fibres (CF) was done. The multidimensional product was characterized par various examinations and analyses. Composite materials were molded with epoxy resin to evaluate mechanical properties of NTC-FC. Young’s modulus was increased and tensile strength at break decreased. Transverse and longitudinal electrical properties were increased by 500 to 800 % respectively. Finally, we proposed NTC growth mechanisms. They are directly linked to the chemical and physical catalyst particles properties.

Esta dissertação considera a atuação das empresas responsáveis pelas redes de distribuição de água, energia elétrica e gás natural denominadas Redes de Distribuição de Utilidades (RDUs) como sistemas produtivos, com o propósito de conhecer e avaliar a maneira como os esforços de melhoria da produtividade são conduzidos em busca de metas como maior qualidade, menor custo e menor prazo de entrega ao cliente. Este estudo é desenvolvido tomando-se como referência o modelo da produção enxuta, desenvolvido pela Toyota Motor Company do Japão, fundamentado no princípio de buscar o aumento da produtividade por meio da identificação e eliminação de desperdícios. Embora originalmente disseminada na cadeia da indústria automobilística, a adoção e adaptação da abordagem de melhoria de processos deste modelo têm sido consideradas em diferentes setores. Assim, o presente trabalho tem como motivação explorar as possibilidades de estender a aplicação de conceitos e práticas da produção enxuta em empresas de RDU. Primeiramente, a cadeia de valor das RDUs é analisada, e são escolhidos dois processos construção e manutenção para o desenvolvimento do estudo. Em seguida, é apresentada uma revisão bibliográfica, a partir da qual um conjunto de dezenove práticas enxutas possivelmente aplicáveis aos processos escolhidos é identificado. Por meio de três estudos de caso, analisa-se em que medida as empresas responsáveis pelas RDUs aplicam tais práticas enxutas em seus processos, e em que medida a aplicação atual ou uma possível aplicação futura de cada prática enxuta é motivada pelo pensamento enxuto. Com base nos dados levantados na pesquisa de campo, as práticas consideradas são avaliadas quanto ao potencial de efetiva aplicação em cada processo, levando-se em conta o objetivo de eliminação de desperdício. O estudo possibilitou verificar que todas as dezenove práticas são aplicáveis a pelo menos um dos processos considerados, sendo a maioria aplicável a ambos. / This dissertation considers the performance of the companies that are responsible for water, electricity and natural gas networks named Utility Distribution Networks (UDN) as productive systems, in order to understand and evaluate the way that the efforts for continuous productivity improvement of their processes are conducted in search of goals such as more quality, less cost and less lead time to customer. The model that is used as a reference for this study is the lean production model, which was developed by the Toyota Motor Company in Japan, founded on the principle of search for productivity increase by the elimination on waste. Although it was originally disseminated in the auto industry value chain, the process improvement approach from this model has been considered in different industries. Therefore, the present work is motivated by the chance to explore the possibility of extending the lean production application concepts and practices to UDN companies. Firstly, UDNs value stream is analyzed, and two processes are chosen construction and maintenance for the development of study. Then, a bibliographic review is presented, from which a set of nineteen lean practices that maybe possibly applicable on the chosen processes is identified. By the development of three case studies, an evaluation is performed with the objective to analyze to what extent the companies that are responsible for UDN apply those lean practices on their processes, and to what extent the current application or a possible future application of each lean practice is motivated by lean thinking. Based on the data from the field research, the potential for effective application of the considered practices is evaluated for each process, taking into consideration the objective of elimination of waste. The study made it possible to verify that all the nineteen practices are applicable at least on one of the considered processes, and the majority is applicable on both processes.

S'han creat les eines necessàries per a estudiar la viabilitat de dur a terme la separació de mescles de dos i tres components basats en un mètode de separació en continu tipus SMB. Amb aquesta finalitat, s'ha creat un sistema per al disseny ràpid i econòmic de separadors SMB que permet simular el comportament d'una columna qualsevol integrada en un sistema SMB. Aquest sistema, està constituït per una sola columna cromatogràfica que pot treballar en qualsevol de les quatre zones en què es pot dividir un SMB.El sistema de control de l'instrument permet simular cicles complets de separadors SMB. A més, el programa també permet treballar amb separacions formalment binàries o ternàries, en conformacions de SMB de 4, 6 i 8 columnes i combinació d'elles agrupades en 4 zones de treball. El sistema també integra un programa de simulació numèrica i optimització de separadors SMB. La combinació del programa i el sistema de disseny de SMB permet establir la variació de la concentració de cada substància amb el temps podent així determinar: el nombre de columnes per zona, el temps de canvi de cada columna, el cabal d'operació a cada zona i la concentració de cada substància per cada sortida de l'SMB.Per tal de verificar el correcte funcionament de l'equip complet s'ha dut a terme la separació de dos components, fenol i triclorofenol, determinant-ne les condicions òptimes de separació. Tot i que la retenció entre aquestes dues substàncies és molt similar, el prototipus ha permès estudiar el comportament de la unitat SMB complerta. També s'ha estudiat la separació del p-aminofenol i l'o-aminofenol com a part integrant del procés de fabricació del paracetamol i se n'han trobat les condicions òptimes d'operació. En treballar amb aquest procés s'ha buscat la màxima puresa del p-aminofenol en detriment de la puresa de l'o-aminofenol aconseguint una puresa de p-aminofenol a la sortida refinat del 97%. Finalment, i des del punt de vista del canvi d'escala, s'han determinat els valors de la longitud no utilitzada de rebliment i del coeficient global de transferència de matèria per tal de poder extrapolar els resultats del prototipus a escala industrial. / Se han creado las herramientas necesarias para estudiar la viabilidad de desarrollar separaciones de mezclas de 2 y 3 componentes basados en un método de separación en continuo tipo SMB. Con esta finalidad se ha creado un sistema para el diseño rápido y económico de separadores SMB que permite simular el comportamiento de una columna cualquiera integrada en un sistema SMB. Este sistema está constituido por una única columna cromatográfica que puede trabajar en cualquiera de las cuatro zonas en que se puede dividir un SMB.El sistema de control del instrumento permite simular ciclos completos de separadores SMB. Además, el programa también permite trabajar con separaciones formalmente binarias o ternarias, en conformaciones de SMB de 4, 6 y 8 columnas y combinación de ellas agrupadas en 4 zonas de trabajo. El sistema también integra un programa de simulación numérica y optimización de separadores SMB. La combinación del programa y el sistema de diseño de SMB permite establecer la variación de la concentración de cada sustancia con el tiempo pudiendo así determinar: el número de columnas por zona, el tiempo de cambio de cada columna, el caudal de operación en cada zona y la concentración de cada sustancia por cada salida del SMB. Para verificar el correcto funcionamiento del equipo completo se ha separado una mezcla de dos compuestos, fenol y triclorofenol, determinando las condiciones óptimas de separación. Aunque la retención entre estas dos sustancias es muy similar, el prototipo ha permitido estudiar el comportamiento de la unidad SMB completa. También se ha estudiado la separación del p-aminofenol y el o-aminofenol como parte integrante del proceso de fabricación del paracetamol y se han determinado las condiciones óptimas de operación. Al trabajar con este proceso se ha buscado la máxima pureza del p-aminofenol con menoscabo de la pureza del o-aminofenol consiguiendo una pureza de p-aminofenol a la salida de refinado del 97%. Finalmente se han determinado los valores de la longitud de lecho no utilizada y del coeficiente global de transferencia de materia para extrapolar los resultados del prototipo a escala industrial. / The necessary tools to study the viability of developing separations of mixtures of 2 and 3 components using a continuous method like the SMB have been developed. A fast and non-expensive system to design such separations has been developed. This system can simulate the behaviour of a single column integrated in the SMB. This system is composed by a single column capable to work in every one of the four zones into which an SMB can be divided. The control system gives the possibility of simulate complete cycles of an SMB separator. In addition the program also can work with binary or ternary separations, in SMB conformations of 4, 6 and 8 columns and its combinations grouped together in 4 working zones. The system also integrates a numerical simulation and optimization program of SMB separators. The combination of program and SMB design system establishes the concentration of each substance within the time. It gives the following parameters: the number of columns per zone, the column switch time, the flow in each zone and the concentration at the SMB outlet. In order to check the correct operation of the whole unit the separation of phenol and trichlorophenol have been done determining the best operation conditions. However the retention of this substances is very similar, the prototype has studied the complete separation. It also has been studied the separation of p-aminophenol and o-aminophenol as a part of the sysntesis process os paracetamole and the best separation condition have been stablished. While working with this process the maximum purity of p-aminophenol has been searched. The purity of p-aminophenol at the raffinate outlet is about 97%. Finally the values of unused bed lenght and overall mass transfer coefficient have been calculated to extrapolate the results to the industrial scale.

Esta dissertação considera a atuação das empresas responsáveis pelas redes de distribuição de água, energia elétrica e gás natural denominadas Redes de Distribuição de Utilidades (RDUs) como sistemas produtivos, com o propósito de conhecer e avaliar a maneira como os esforços de melhoria da produtividade são conduzidos em busca de metas como maior qualidade, menor custo e menor prazo de entrega ao cliente. Este estudo é desenvolvido tomando-se como referência o modelo da produção enxuta, desenvolvido pela Toyota Motor Company do Japão, fundamentado no princípio de buscar o aumento da produtividade por meio da identificação e eliminação de desperdícios. Embora originalmente disseminada na cadeia da indústria automobilística, a adoção e adaptação da abordagem de melhoria de processos deste modelo têm sido consideradas em diferentes setores. Assim, o presente trabalho tem como motivação explorar as possibilidades de estender a aplicação de conceitos e práticas da produção enxuta em empresas de RDU. Primeiramente, a cadeia de valor das RDUs é analisada, e são escolhidos dois processos construção e manutenção para o desenvolvimento do estudo. Em seguida, é apresentada uma revisão bibliográfica, a partir da qual um conjunto de dezenove práticas enxutas possivelmente aplicáveis aos processos escolhidos é identificado. Por meio de três estudos de caso, analisa-se em que medida as empresas responsáveis pelas RDUs aplicam tais práticas enxutas em seus processos, e em que medida a aplicação atual ou uma possível aplicação futura de cada prática enxuta é motivada pelo pensamento enxuto. Com base nos dados levantados na pesquisa de campo, as práticas consideradas são avaliadas quanto ao potencial de efetiva aplicação em cada processo, levando-se em conta o objetivo de eliminação de desperdício. O estudo possibilitou verificar que todas as dezenove práticas são aplicáveis a pelo menos um dos processos considerados, sendo a maioria aplicável a ambos. / This dissertation considers the performance of the companies that are responsible for water, electricity and natural gas networks named Utility Distribution Networks (UDN) as productive systems, in order to understand and evaluate the way that the efforts for continuous productivity improvement of their processes are conducted in search of goals such as more quality, less cost and less lead time to customer. The model that is used as a reference for this study is the lean production model, which was developed by the Toyota Motor Company in Japan, founded on the principle of search for productivity increase by the elimination on waste. Although it was originally disseminated in the auto industry value chain, the process improvement approach from this model has been considered in different industries. Therefore, the present work is motivated by the chance to explore the possibility of extending the lean production application concepts and practices to UDN companies. Firstly, UDNs value stream is analyzed, and two processes are chosen construction and maintenance for the development of study. Then, a bibliographic review is presented, from which a set of nineteen lean practices that maybe possibly applicable on the chosen processes is identified. By the development of three case studies, an evaluation is performed with the objective to analyze to what extent the companies that are responsible for UDN apply those lean practices on their processes, and to what extent the current application or a possible future application of each lean practice is motivated by lean thinking. Based on the data from the field research, the potential for effective application of the considered practices is evaluated for each process, taking into consideration the objective of elimination of waste. The study made it possible to verify that all the nineteen practices are applicable at least on one of the considered processes, and the majority is applicable on both processes.

Made available in DSpace on 2017-07-10T17:59:27Z (GMT). No. of bitstreams: 1 Ana Carolina de Araujo Abdala.pdf: 1086977 bytes, checksum: 3669a77a33625e3d6a3660105a7eb19d (MD5) Previous issue date: 2013-02-15 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The current work aimed to investigate the production of ethyl esters which result from the esterification reactions of oleic acids (AO) and transesterifiction of residual fried oils (ORF), both in supercritical ethanol throught non-catalic method. The experiments were carried out in a packed bead reactor operated in continuous mode. The esterification reactions were carried out under a temperature from 200 to 325ºC, pressure of 15 and 20 MPa and molar reason AO: ethanol in the range of 1:3 to 1:12. From the results on, was possible to note, that the temperature and the pressure influenced positively the conversion on the esterification reaction of oleic acid, with an income of 93, 5%, observed under 325°C and 20Mpa, in a reaction that last 40 minutes. The excess of alcohol in the reactional system didn´t provide significant differences in the conversions, with the optimum molar reason, defined in 1:6 (AO: ethanol). The transesterification reactions, were carried out with pauses of temperatures from 275 to 325ºC, pressure fixed of 20MPa, adopting the mass reason of ORF to ethanol of 1:0.5 and 1:1, varying the residence time, leading the determination of great conditions to the ethyl esters synthesis. Notable results in the alcoholysis reactions were observed at 325ºC, mass ratio of ORF to ethanol at 1:1 and residence time of 40 minutes. Using the molar ratio ORF: ethanol selected as the best suitable, was also evaluated the effect of the water addition to the reactional medium (0, 5 and10% of water relative to the ethanol mass), in order to investigate the technical viability of the use of low quality and costs of the raw materials containing water in the composition. The effect of addition of deco- hexane solvent (0, 5, 10 and 20% of hexane compared to the substrates mass), as well the effect of addition of soy bio-fuel (0, 20 and 40% of soy bio-fuel compared to substrates mass), were also investigated, in order to improve the performance of the reaction. Notable incomes were observed in the reaction with 5% of water addition and 20% of hexane. The best results related to the soy bio-fuel addition to the reactional medium were observed in the reactions carried out with 40% of soy bio-fuel, however, taking into consideration the amount of esters added to the process, the condition considered great was with 20% soy bio-fuel.. From the best conditions of the process, the kinetic daalcoholise super-critical was investigated with the reactions conducted with the residence time from 0 to 70 minutes and the notable results were: 83% in ethyl esters, obtained in 300ºC, 70 minuts of reaction, using 5% of water, 87% in ethyl esters observed in 300ºC, 70minutes of reaction, using 20% of hexane and 82% in ethyl esters obtained in 325ºC, 50 minutes of reaction, using 20% of soy bio-fuel. Lastly, the stability of the daalcoholise reactional medium was investigated. Temperatures of reaction till to 300ºC resulted in decomposition <10% with residence time of 20 minutes. However, the reactions with longer residence time, showed a higher ratio of decomposition of the constituents of the reactional medium. / O presente trabalho teve como objetivo investigar a produção de ésteres etílicos nas reações de esterificação do ácido oleico (AO) e transesterificação de óleo residual de fritura (ORF), ambas em etanol supercrítico pelo método não catalítico. Os experimentos foram realizados em um reator tubular com leito empacotado operado em modo contínuo. As reações de esterificação foram conduzidas com temperatura na faixa de 200 a 325ºC, pressão de 15 e 20 MPa e razão molar AO:etanol na faixa de 1:3 a 1:12. A partir dos resultados, foi possível observar que a temperatura e a pressão influenciaram positivamente a conversão nas reações de esterificação do ácido oleico, com rendimentos de 93,5%, observados a 325ºC e 20 MPa, em 40 minutos de reação. Já o excesso de álcool no sistema reacional não proporcionou diferenças significativas nas conversões, com a razão molar ótima, definida em 1:6 (AO:etanol). As reações de transesterificação foram conduzidas no intervalo de temperatura de 275 a 325ºC, pressão fixa de 20 MPa, adotando razão mássica de ORF para etanol de 1:0,5 e 1:1, variando-se o tempo de residência, possibilitando a determinação das condições ótimas para a síntese de ésteres etílicos. Resultados apreciáveis nas reações de álcoolise foram verificados em 325ºC, razão mássica de ORF para etanol de 1:1 e tempo de residência de 40 minutos. Utilizando a razão mássica ORF:etanol selecionada como a mais adequada, ainda foi avaliado o efeito da adição de água ao meio reacional (0, 5 e 10% de água em relação à massa do etanol), visando investigar a viabilidade técnica da utilização de matérias-primas de menor qualidade e menor custo que contenham água em sua composição. O efeito da adição de co-solvente hexano (0, 5, 10 e 20% de hexano em relação à massa dos substratos) bem como o efeito da adição de biodiesel de soja (0, 20 e 40% de biodiesel de soja em relação à massa dos substratos) também foram investigados, buscando melhorar o desempenho da reação. Rendimentos apreciáveis foram observados nas reações com 5% de adição água e 20% de hexano. Os melhores resultados referentes à adição de biodiesel de soja ao meio reacional foram observados nas reações conduzidas com 40% de biodiesel de soja, porém, levando-se em conta a quantidade de ésteres adicionada ao processo, a condição considerada ótima foi de 20% de adição de biodiesel de soja. A partir das melhores condições de processo, a cinética da álcoolise supercrítica foi investigada com reações conduzidas com tempo de residência na faixa de 0 a 70 minutos e os resultados apreciáveis foram de: 83% em ésteres etílicos, obtidos em 300°C, 70 minutos de reação, utilizando 5% de água, 87% em ésteres etílicos observados em 300°C, 70 minutos de reação, utilizando 20% de hexano e 82% em ésteres etílicos alcançados em 325°C, 50 minutos de reação, utilizando 20% de biodiesel de soja. Por fim, a estabilidade dos componentes do meio reacional da alcoólise foi investigada. Temperaturas de reação de até 300ºC resultam em decomposições <10% com tempo de residência de 20 minutos. Porém, as reações com tempos de residência mais longos, apresentaram taxas de decomposição dos constituintes do meio reacional mais elevadas.

Les polymères de spécialité en solution dans l’eau sont typiquement produits en réacteur agité dans des ateliers multi-produit, offrant la possibilité de produire une large gamme de produits. Cependant, les limites inhérentes au réacteur batch se concrétisent dans ce type de production par une variation de la qualité de polymère d’une synthèse à l’autre, d’une phase de développement difficile de l’échelle laboratoire à la production industrielle, et d’une technologie offrant des échanges de chaleur limité. Des conditions diluées sont requises ce qui implique des cycles de production longs. Le passage vers un procédé continu pourrait être une bonne alternative s’il permet d’accélérer les temps de réaction mais aussi de garantir une qualité de polymère constante dans le temps. Cela signifie dans notre cas un polymère avec une masse molaire élevée et un indice de polymolécularité faible. Inscrit dans le cadre du projet européen F3Factory, l’étude a consisté à la conception d’un pilote de laboratoire destiné à la polymérisation en continu puis à l’étude d’un procédé industriel batch dans ce pilote. Il s’agit de la polymérisation radicalaire d’un copolymère de l’acide acrylique qui comporte deux problématiques principales, une réaction exothermique couplée à une viscosité élevée. Le pilote de laboratoire est équipé d’un réacteur tubulaire contenant des mélangeurs statiques de marque Fluitec avec un débit de 1 à 2kg.hr-1 et offrant un excellent transfert de chaleur, un bon mélange avec un écoulement à caractère piston. Une méthodologie de passage au continu a été développée. Une étude préliminaire a notamment été menée en couplant rhéologie et spectroscopie Raman. Cela nous a permis d'acquérir des données pertinentes à petite échelle (étude rhéocinétique), données clés pour le passage au continu. Ce travail démontre la faisabilité de la production en continu de polymère d’acide acrylique. L’impact des différents paramètres tel que les concentrations en monomères et amorceur et la température a été étudié. Finalement, notre travail s’est attaché à caractériser les polymères produits en batch et en continu et à déterminer les avantages et les limitations offerts par le procédé continu / Water soluble polymers are typically produced batchwise in multi-product plants, providing the ability to produce a wide range of products. However, some limitations due to the use of batch reactor remain which are the variation of the quality of a synthetic polymer from one run to another run, a non-obvious scale-up step from the laboratory scale to the industrial production, and a technology with limited heat exchange . Dilute conditions are required which implies long cycle production times. A transition from discontinuous to continuous production could be a good alternative if it can accelerate the reaction time but also to ensure a consistent quality of polymer over time. This means in our case a polymer with a high molecular weight and a low polydispersity index. In the frame of the European project F3Factory, a continuous laboratory-scale pilot of 1 to 2kg.hr-1 flowrate for the production of water soluble polymers was designed. The chosen intensified reactor was a tubular reactor containing static mixers. Providing excellent heat transfer, a good mixing with a piston flow character, this technology was also flexible thanks to the presence of standardized flanges and well instrumented for data acquisition in terms of temperature, pressure but also conversion with in-line Raman analysis. Free radical polymerization of acrylic acid was carried out in a continuous way. Two main challenges were faced: an exothermic reaction coupled to a high viscosity. A methodology relative to the transposition from bath to continuous process was developed. A preliminary study coupling rheology and Raman spectroscopy allowed us to acquire relevant data at small-scale (rheokinetic study). This work demonstrates the feasibility of the continuous production of acrylic acid polymer. The impact of various parameters such as the concentrations of monomer and initiator and the temperature was investigated. Finally, our work has focused on characterizing the polymers produced in batch and continuous and to determine the advantages and limitations offered by the continuous process

L’enrobage à sec par action mécanique permet de formuler des particules composites dont les fonctionnalités et les propriétés physiques, comme l’écoulement, sont améliorées et cela sans l’ajout de solvant ou de liant. Actuellement la plupart des procédés d’enrobage à sec sont conduits en mode discontinu. L’objectif de ce travail est donc de développer et d’étudier une opération d’enrobage à sec par un procédé continu innovant, une extrudeuse bi-vis corotative sans filière, constituant une rupture technologique dans son domaine. Dans le cadre de ce travail, des sphères de cellulose microcristalline, appelées particules hôtes sont enrobées avec soit des talcs de différentes granulométries soit du stéarate de magnésium, nommées particules invitées. Dans un premier temps, le comportement des particules hôtes est étudié dans le procédé. Les particules sorties du procédé sont analysées par microscopie électronique à balayage, par granulométrie laser et par voluménométrie. Pour la configuration de vis présentant que des éléments de transport, la vitesse de rotation des vis (25 à 200 rpm) et le débit d’alimentation (0,5 à 2 kg/h) choisis influencent le taux de remplissage dans le fourreau, et pour un taux supérieur à environ 14 % : la quantité de particules endommagées est négligeable. Un taux de remplissage faible favorise les frictions particules-particules ou particules-métal au niveau de l’entrefer. L’ajout de zones de mélanges dans la configuration des vis accentue ce phénomène de brisure par l’augmentation des contraintes de cisaillement. Par ailleurs, des masses retenues de la poudre dans le fourreau évoluent linéairement en fonction de la masse de poudre transportée en un tour de vis pour différentes configurations de vis. Ces relations laissent penser à l’existence d’un volume mort. Les mesures de distribution de temps de séjour mettent en évidence par l’application d’un modèle d’association de réacteurs idéaux, la présence d’écoulement piston et d’un volume mort. Dans un second temps, la faisabilité d’une opération d’enrobage à sec par le procédé étudié est analysée. Une couche d’enrobage continu de talc modifiant le comportement hydrophile des particules hôtes, est obtenue pour une vitesse de rotation de vis de 50 rpm et une configuration de vis cisaillante. Le type de particules invitées influence la morphologie de la couche d’enrobage : de type film avec le stéarate de magnésium et de type discret avec le talc micronisé. Néanmoins les propriétés de taille, d’écoulement, de compressibilité des particules composites sont semblables à celle des particules hôtes, seule la propriété de mouillabilité est modifiée. La présence de stéarate de magnésium diminue le temps de séjour des particules dans le procédé grâce à son caractère lubrifiant. / Dry particle coating with mechanical action allows the production of composite particles whose functionalities and physical properties, such as flowability, are improved thanks to the absence of solvent or binder. Currently, most of dry particle coating processes are carried out in a discontinous mode. The objective of this work is thus to develop and study a dry particle coating operation with a continuous innovative process, twin screw co-rotating extruder without die, constituting a technological rupture in this application field. For this, spherical particles of microcristalline cellulose, as host particles, are coated with either talc particles of two sizes or magnesium stearate particles (MgSt), as guest particles. Firstly, the host particles behaviour is studied in the process. The exiting particles are analyzed by scanning electronic microscopy, laser granulometry and tapping test. Regarding the transport configuration, the chosen screw speed (between 25 an 200 rpm) and the feed rate (between 0.5 and 2 kg/h) have an important influence on the filling level in the barrel. For a filling level superior to approximatly 14 %, the quantity of broken particles is not significant. A low filling level facilitates the particle-particle or particle-metal friction in the barrel gap. When more mixing zones are added to the screw configuration, the shear stresses increase and, therefore, the breakage particle phenomenon becomes more prominent. Futhermore, the retained powder mass increases linearly when the transported mass in one screw turn increases for all the screw configurations. These evolutions suggest the existence of a dead volume. The residence time distribution mesurements highlight, thanks to the application of flowing model, the presence of plug-flow and a dead volume. Secondly, the feasibility of using the process for dry particle coating is analyzed. A continuous coating layer of talc modifiying the hydrophilic behaviour of the host particles is obtained for a low screw rotation speed (50 rpm) and a screw configuration with one mixing zone. The type of guest particles influence the morphology of the coating layer: film layer with the MgSt particles and discontinous layer with the micronized talc. However, size, flow and compressibility of the coated particles are similar to the ones of the initial host particles, only the wettability properties are modified. The presence of MgSt decreases the particle mean residence time in the process thanks to its lubricant action.

No description available.