This thesis concerns the construction of a detailed, compartmental population balance model for twin-screw granulation using the stochastic weighted particle method. A number of new particle mechanisms are introduced and existing mechanisms augmented including immersion nucleation, coagulation, breakage, consolidation, liquid penetration, primary particle layering and transport. The model’s predictive power is assessed over a range of liquid-solid mass feed ratios using existing experimental data and is demonstrated to qualitatively capture key experimental trends in the physical characteristic of the granular product. As part of the model development process, a number of numerical techniques for the stochastic weighed method are constructed in order to efficiently solve the population balance model. This includes a new stochastic implementation of the immersion nucleation mechanism and a variable weighted inception algorithm that dramatically reduces the number of computational particles (and hence computational power) required to solve the model. Optimum operating values for free numerical parameters and the general convergence properties of the complete simulation algorithm are investigated in depth. The model is further refined though the use of distinct primary particle and aggregate population balances, which are coupled to simulate the complete granular system. The nature of this coupling permits the inclusion of otherwise computational prohibitive mechanisms, such as primary particle layering, into the process description. A new methodology for assigning representative residence times to simulation compartments, based on screw geometry, is presented. This residence time methodology is used in conjunction with the coupled population balance framework to model twin-screw systems with a number of different screw configurations. The refined model is shown to capture key trends attributed to screw element geometry, in particular, the ability of kneading elements to distribute liquid across the granular mass.

In this thesis, a thorough study of the batch top-spray fluidized bed granulation was carried out including experimental study, population balance model (PBM), computational fluid dynamic (CFD) study and control strategy development. For the experimental study, the influence variables of pulsed frequency, binder spray rate and atomization pressure of a batch top-spray fluidized bed granulation process were studied using the Box-Behnken experimental design method. Different mathematical models were developed to predict the mean size of granules, yield, relative width of granule distribution, Hausner ratio and final granule moisture content. Validation experiments have shown the reliability and effectiveness of using the Box-Behnken experimental design method to study a fluidized bed granulation process. The one-dimensional population balance models (ODPBMs) have been developed to model a pulsed top-spray fluidized bed granulation, linking the operating factors of the pulsed frequency, the binder spray rate, and atomization air pressure with the granule properties to predict granule growth behavior at different operating conditions. A multi-stage open optimal control strategy based on the developed ODPBMs was proposed to reduce the model and process mismatch through adjusting the trajectory of the evolution of the granule size distribution at predefined sample intervals. The effectiveness of the proposed modeling and multi-stage open optimal control strategy has been validated by experimental and simulation tests. In addition, an Eulerian-Eulerian two-fluid model (EETFM) was developed to describe the gas-particle two-phase flow in the fluidized bed granulator. By computational fluid dynamic analysis, it has been proven that the fluidized bed granulation system is not homogeneous, based on which a two-compartmental population balance model (TCPBM) was developed to describe the particle growth in the fluidized bed granulation. Validation experiments have shown the effectiveness and superior accuracy of the TCPBM comparing with the ODPBM in predicting the final particle size distribution.

ABSTRACT This work applies the concepts of the attainable region for process synthesis in comminution. The attainable region analysis has been successfully applied for process synthesis of reactor networks. The Attainable Region is defined as the set of all possible output states for a constrained or unconstrained system of fundamental processes (Horn, 1964). A basic procedure for constructing the attainable region for the fundamental processes of reaction and mixing has been postulated in reaction engineering (Glasser et al., 1987). This procedure has been followed in this work to construct the candidate attainable region for size reduction processes as found in a size reduction environment. A population balance model has been used to characterise the evolution of particle size distributions from a comminution event. Herbst and Fuerstenau (1973) postulated the dependency of grinding on the specific energy. A specific energy dependent population balance model was used for the theoretical simulations and for the fitting of experimental data. A new method of presenting particle size distributions as points in the Euclidian space was postulated in place of the traditional cumulative distribution. This allows successive product particle size distributions to be connected forming a trajectory over which the objective function can be evaluated. The curve connects products from successive batch grinding stages forming a pseudo-continuous process. Breakage, mixing and classification were identified as the fundamental processes of interest for comminution. Agglomeration was not considered in any of the examples. Mathematical models were used to describe each fundamental process, i.e. breakage, mixing and classification, and an The application of the attainable region analysis in comminution Abstract algorithm developed that could calculate the evolution of product particle size distributions. A convex candidate attainable region was found from which process synthesis and optimisation solutions could be drawn in two dimensional Euclidian space. As required from Attainable Region Theory, the interior of the bounded region is filled by trajectories of higher energy requirements or mixing between two boundary optimal points. Experimental validation of the proposed application of the attainable region analysis results in comminution was performed. Mono-sized feed particles were broken in a laboratory ball mill and the products were successfully fitted using a population balance model. It was shown that the breakage process trajectories were convex and they follow first order grinding kinetics at long grind times. The candidate attainable region was determined for an objective function to maximise the mass fraction in the median size class 2. It was proved that the same specific energy input produces identical products. The kinematic and loading conditions are supposed to be chosen as a subsequent event after the required specific energy is identified. Finally the fundamental process of classification was added to the system of breakage and mixing. The attainable regions analysis affords the opportunity to quantify exactly the reduction in energy consumption due to classification in a comminution circuit, thus giving optimal targets. Classification showed the potential to extend the candidate attainable region for a fixed specific energy input. The boundary of the attainable region is interpreted as pieces of equipment and optimum process conditions. This solves both the original process synthesis and successive optimisation problems.

Orientador: Florencia Cecília Menegalli / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia de Alimentos / Made available in DSpace on 2018-08-20T07:44:45Z (GMT). No. of bitstreams: 1 Souza_DiogoOtaviodeCastro_D.pdf: 2512415 bytes, checksum: cc03ff39105bf1fed4d6dae52c95bbce (MD5) Previous issue date: 2012 / Resumo: A granulação é um processo de aumento de tamanho de pós finos, que pode ser utilizado na indústria de alimentos para aumentar a qualidade de alimentos em pó, pelo aumento da sua dispersibilidade e solubilidade em líquidos. O modelo de balanço populacional (MBP) é normalmente utilizado para modelar este processo. Entretanto, nos trabalhos existentes, o MBP não leva em consideração as perdas de partículas no processo por elutriação e incrustação, muito relevantes para granulação de sucos em pó. Além disso, não existe na literatura correlações para estimar estas constantes com base nas condições operacionais do equipamento e nas características das partículas. Diante disso, o objetivo deste trabalho foi o estudo do impacto das condições operacionais no modelo de balanço populacional, durante a granulação de suco de goiaba em pó, em leito fluidizado. Diversos experimentos foram realizados de forma a adquirir os dados que alimentaram os modelos matemáticos desenvolvidos. Estes dados foram obtidos fora das zonas de máximo rendimento a fim de se obter um modelo mais abrangente, que pudesses simular o processo de granulação em todas as condições de processo. Foram desenvolvidas e validadas correlações para estimativa da taxa de elutriação, para estimativa das constantes de aglomeração e de quebra do MBP. Além disso, foi desenvolvido um modelo para estimativa da massa incrustada de partículas. Os modelos se mostraram capazes de estimar o rendimento do processo, os valores de diâmetro médio e da distribuição de tamanho de partículas. Foram feitas simulações alterando a velocidade do ar de fluidização, a vazão de ligante, o diâmetro médio inicial da distribuição, o formato da distribuição e a massa total de partículas. A análise das simulações realizadas sugere que as condições ótimas precisam ser avaliadas para cada distribuição de tamanho e para cada quantidade de partículas no equipamento, não somente para cada tipo de produto. De modo geral, os modelos desenvolvidos se mostraram adequados para aumentar a compreensão do processo de granulação e para obtenção de ótimos de processamento. Cabe ressaltar a necessidade de futuras investigações para verificar a aplicação destes modelos com outros tipos de material particulado e em outras faixas de operação / Abstract: Granulation is a size-enlargement process, in which small particles are agglomerated together. It is used in the food industry to increase the quality of food powder, by increasing its solubility. The population balance model (PBM) is a widely used model for this process. However, it is usually applied without account losses in the process by elutriation and wall deposition, which are very important for granulation of powdered juices. In addition, there is no correlation in the literature able to estimate the constants of the model from the operational conditions and from the particles¿ characteristics. The objective of this work was to study the impact of operational conditions into the population balance model for the granulation of guava juice powder, in a fluidized bed. Several experiments were performed in order to acquire data that fed the mathematical models developed in this work. These data weren¿t obtained from the zones of maximum process yield in order to achieve a more comprehensive model, which could simulate the granulation process in all process¿ conditions. It was developed and validated a correlation for the estimation of the rate of elutriation, a model for the estimation of the particles¿ wall deposition, a correlation for estimation the constants of agglomeration and for estimation of the breakage constant of the PBM. The models predicted, with good accuracy, the process yield, the mean diameter and particle size distribution. Simulations were made by changing the fluidizing air temperature, the liquid binder flow rate, the fluidizing air velocity, the initial mean diameter, the shape of the particles size distribution and the total mass of particles. The analysis of the simulations suggests that the optimal conditions must be evaluated for each kind of particle size distribution and for each mass of particles into the equipment, not only for each type of product. The models developed showed to be suitable to increase the understanding of the granulation process and to obtain process¿ optimal conditions. It is worth emphasizing the need of further research to verify the application of these models for others types of particles and others operational ranges / Doutorado / Engenharia de Alimentos / Doutor em Engenharia de Alimentos

Abstract : The industrial partner of this project uses a slurry bubble reactor for the production of biogenic methanol. In the latter syngas is dispersed into the slurry continuous phase containing both liquid and solid phases. The rising bubbles containing a wide spectrum of the bubbles sizes, interact with the continuous phase due to the interface momentum transfer. The latter includes the drag, lift, wall lubrication and turbulent dispersion terms that require average bubble size, which needs to be calculated. One way to predict this average bubble size is by using population balance model (PBM), which can be coupled with the Eulerian framework. PBM also needs closure kernels for the bubble coalescence and bubble breakup. In this study, the influence of bubble coalescence and bubble breakup kernels have been studied in two- and three-phase system using eulerian approach, which solves momentum equation for each phase. The influence of the mesh sizes, number of bubble classes, numerical schemes, wall lubrication force and turbulent dispersion force are also included. In the two-phase system, results show that the Luo coalescence model needs to be tuned when used in combination with the Luo breakup kernel. The combination of the Luo coalescence and the Lehr breakup kernels (Luo-Lehr) show promising time-averaged radial profiles of gas holdup and axial liquid velocity as compared to empirical values. In the three-phase system, the combination of the Luo coalescence and the Lehr breakup kernels (Luo-Lehr) and the Luo coalescence and the Luo breakup kernels (Luo-Luo) predict convincing time-averaged radial profile of axial solid velocity as compared to experiments. However, at an elevated superficial gas velocity, a non-realistic behavior was predicted when compared to empirical observations. The sensitivity analysis results show that the 3 mm mesh size depicts a trend similar to the empirical values of the radial profiles of the gas holdup, axial liquid velocity, and solid axial velocity. The number of bubble classes influence the predicted bubble size distribution in the three-phase system while the numerical discretizing schemes have no influence on the results. The bench simulation results show that the inclusion of the turbulent dispersion term using a single porous tubular sparger influences the hydrodynamic behavior of the bubble column. / Le partenaire industriel de ce projet utilise un réacteur à suspension à trois phases pour la production de méthanol biogénique. Dans celui-ci, le gaz de synthèse est diffusé par barbotement dans la phase à suspension qui contient à la fois les phases liquide et solide. Les bulles en ascension présentent un large spectre de tailles et interagissent avec la phase à suspension en échangeant de la quantité de mouvement via leurs surfaces. Cet échange comprend les forces de trainé, de portance, de lubrification en proche parois et de dispersion par turbulence; lesquelles requièrent notamment le calcul de la taille moyenne des bulles. Une façon de prédire numériquement cette taille moyenne est de recourir à un modèle de bilan de population (PBM, de l’anglais Population Balance Model), qui peut être couplé avec un model multiphasique eulérien. Un tel PBM a requière des modèles de fermetures pour la coalescence et la rupture des bulles. Dans la présente étude, l'influence des modèles noyaux de coalescence et de rupture des bulles a été étudiée pour des systèmes à deux et à trois phases en utilisant l’approche eulérienne. L'influence de la taille du maillage, du nombre de classes de bulles, du schéma numérique, de la force de lubrification en proche parois et de la force de dispersion par turbulence sont également incluses. Dans un système bi-phasique, les résultats montrent que le modèle de coalescence Luo doit être ajusté lorsqu'il est utilisé en combinaison avec le noyau de rupture Luo. La combinaison des noyaux de coalescence Luo et de rupture Lehr (Luo-Lehr) montrent des profils radiaux moyennés dans le temps qui sont valides pour la concentration de gaz et la vitesse axiale du liquide par rapport aux mesures expérimentales. Dans le système triphasé, la combinaison des modèles noyaux de coalescence de Luo et de rupture de Lehr (Luo-Lehr) et de la coalescence de Luo et de rupture de Luo (Luo-Luo) prédisent des profils radiaux moyennés dans le temps qui sont valides pour la vitesse axiale moyenné dans le temps par rapport aux expériences. Cependant, à une vitesse de gaz superficielle élevée, ces profils prédisent un comportement non réaliste par rapport aux observations empiriques. Les résultats de l'analyse de sensibilité du maillage montrent qu’avec des cellules de 3 mm, le model prédit une tendance similaire aux valeurs empiriques pour les profils radiaux de concentration du gaz, de vitesse axiale du liquide et de vitesse axiale solide. Le nombre de classes de bulles influe sur les distributions prédites de taille de bulle dans le système triphasé alors que les schémas de discrétisation numériques n'ont aucune influence sur les résultats. Les résultats des simulations d’un banc d’essai avec diffuseur à bulles poreux montrent que tenir compte du terme de dispersion influence le comportement hydrodynamique de la colonne à bulles.

The early steps of activation are crucial in deciding the fate of T-cells leading to the proliferation. These steps strongly depend on the initial conditions, especially the avidity of the T-cell receptor for the specific ligand and the concentration of this ligand. The recognition induces a rapid decrease of membrane TCR-CD3 complexes inside the T-cell, then the up-regulation of CD25 and then CD25-IL2 binding which down-regulates into the T-cell. This process can be monitored by flow cytometry technique. We propose several models based on the level of complexity by using population balance modeling technique to study the dynamics of T-cells population density during the activation process. These models provide us a relation between the population of T-cells with their intracellular and extracellular components. Moreover, the hypotheses are proposed for the activation process of daughter T-cells after proliferation. The corresponding population balance equations (PBEs) include reaction term (i.e. assimilated as growth term) and activation term (i.e. assimilated as nucleation term). Further the PBEs are solved by newly developed method that is validated against analytical method wherever possible and various approximate techniques available in the literature.

Les premières étapes d’activation des lymphocytes T sont cruciales pour déterminer leur comportement, ainsi que leur prolifération. Ces étapes dépendent fortement des conditions initiales, particulièrement de l’avidité du récepteur du lymphocyte (TCR) pour le ligand spécifique provenant de l’antigène. La reconnaissance du virus entraine une séquence des réactions biochimiques mettant en œuvre de protéines membranaires et cellulaires. Le processus peut être mesuré par cytométrie en flux. On propose ici plusieurs modèles de différents niveaux de complexité. Ces modèles décrivent une relation entre la population de lymphocytes T et leurs composants intracellulaires et extracellulaires. Ils conduisent à des systèmes d’EDO et d’EDP dont la résolution permet d’étudier la dynamique de la densité de population des lymphocytes au cours du processus d'activation. En outre, différentes hypothèses sont proposées pour le processus d'activation des cellules filles après prolifération. Les équations de bilan de population (EBPs) sont résolues par une nouvelle méthode validée par une solution analytique quand elle existe, ou par comparaison à différentes méthodes numériques disponibles dans la littérature. L’avantage de cette nouvelle méthode est d’être utilisable dans certains cas où les méthodes classiques ne le sont pas. / The early steps of activation are crucial in deciding the fate of T-cells leading to the proliferation. These steps strongly depend on the initial conditions, especially the avidity of the T-cell receptor for the specific ligand and the concentration of this ligand. The recognition induces a rapid decrease of membrane TCR-CD3 complexes inside the T-cell, then the up-regulation of CD25 and then CD25–IL2 binding which down-regulates into the T-cell. This process can be monitored by flow cytometry technique. We propose several models based on the level of complexity by using population balance modeling technique to study the dynamics of T-cells population density during the activation process. These models provide us a relation between the population of T-cells with their intracellular and extracellular components. Moreover, the hypotheses are proposed for the activation process of daughter T-cells after proliferation. The corresponding population balance equations (PBEs) include reaction term (i.e. assimilated as growth term) and activation term (i.e. assimilated as nucleation term). Further the PBEs are solved by newly developed method that is validated against analytical method wherever possible and various approximate techniques available in the literature.

Submitted by Rafael Santana (rafael.silvasantana@ufpe.br) on 2017-07-10T18:48:53Z No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) Dissertação completa.pdf: 8076647 bytes, checksum: 5dc16e9ca5f10026afed3fda08fda16b (MD5) / Made available in DSpace on 2017-07-10T18:48:53Z (GMT). No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) Dissertação completa.pdf: 8076647 bytes, checksum: 5dc16e9ca5f10026afed3fda08fda16b (MD5) Previous issue date: 2016-05-12 / Minerais industriais em faixas ultrafinas (< 10 μm) têm suas propriedades potencializadas em relação ao mesmo mineral com maior granulometria. Os materiais ultrafinos são utilizados em diversos seguimentos industriais; por exemplo: materiais cerâmicos, papel e celulose, fármacos, polímeros e tintas. A produção de ultrafinos ocorre em moinhos de alta energia. Dentre estes, o moinho planetário de bolas destaca-se pelas altas taxas de redução de tamanho e pelo fato de poderem ser alimentados via seco ou via úmido em regime contínuo (escala industrial) ou por batelada. A produção de ultrafinos é limitada pelo consumo de energia e pela necessidade de controle das condições operacionais, tamanho, morfologia e composição das partículas. Para otimização das variáveis do processo, usam-se ferramentas computacionais embasadas em modelos matemáticos, tais como os modelos de balanço populacional (MBP), dada pela equação da moagem por batelada, e modelos energéticos. Este trabalho teve por objetivo estudar modelos cinético e energéticos, sendo o primeiro uma solução analítica da equação da moagem por batelada utilizado para descrever as distribuições de tamanhos de partículas, e o segundo dado pela relação energia-tamanho, que prevê uma taxa de redução de tamanho ilimitada, e pela relação tempo-tamanho, que está fundamentada na taxa de moagem () e no limite de moagem. Os modelos foram aplicados em duas centenas de curvas granulométricas resultantes de ensaios de moagem executados anteriormente em alíquotas de calcário e quartzo (duas procedências) com massa e granulometria controladas. Os tempos de moagem variaram de 2 a 960 minutos com velocidades de revolução de 100 a 300 rpm. Os ajustes dos modelos cinético e energéticos foram avaliados considerando os seguintes fatores: coeficiente de determinação (R2), erro padrão (EP), erro de ajuste () e índice de dependência (ID). O modelo cinético apresentou, para a maioria das condições de moagem testadas, grande incerteza associada a alguns de seus parâmetros ( > 10%), tornando os ajustes insatisfatórios segundo os critérios utilizados. Os fatores de avaliação para o modelo cinético só foram adequados para o quartzo de uma procedência, na faixa de 38x75 μm, satisfazendo a condição de compensação estabelecida. No caso dos modelos energéticos, os ajustes obtidos para a relação energiatamanho foram melhores para aquelas situações em que os diâmetros característicos não apresentaram uma estabilização em seu decrescimento. Por sua vez, a relação tempo-tamanho mostrou ajustes compatíveis com as situações em que foi observado um estado estático de decrescimento dos diâmetros característicos, atingindo o limite de moagem. A partir dos ajustes da relação tempo-tamanho foi possível determinar uma constante k’ que caracterizasse a resistência à fragmentação do material em função das condições de moagem estudadas. Os valores dessa constante mostraram que materiais mais resistentes à fragmentação possuem os menores valores de k’, que variaram entre 0,96 e 2,6 g/J para o calcário e entre 0,06 a 0,53 g/J para o quartzo. Concluiu-se, que o modelo cinético foi incompatível com a moagem ultrafina, devida a presença de eventos significativos de aglomeração e interações mecânicas multipartículas, confirmados pela variação do índice de uniformidade () com o tempo de moagem. Os modelos energéticos se complementam na descrição dos resultados experimentais. Logo um modelo intermediário que considere uma taxa de redução de tamanho como uma função potência, com um expoente e um parâmetro (l) que represente o limite de moagem, seja o mais recomendado para a representação dos processos de moagem ultrafina de minerais industriais. / Industrial minerals in ultrafine ranges (< 10 μm) have their properties potentiated compared to the same mineral with larger particle size. The ultrafine materials are used in several industries; for example, ceramics, paper and cellulose, pharmaceuticals, polymers and paints. The production of ultrafine occurs in high energy mills. Among these mills, the planetary ball mill stand out by high rates and can be fed dry or wet in continuous (industrial scale) or batch operation. The production of ultrafine is limited by the energy consumption and the need to control of the operating conditions, size, morphology and composition of the particles. Computational tools based on mathematical models are used in the optimization and control of process variables, such as the population balance models (MBP), given by equation milling batch, and energetic models. This work has as objective to study kinetic and energetic models, the first is a analytical solution for the batch grinding equation used to describe particle size distributions; the second given by the energy-size relations, which predict a size reduction rate unlimited, and by the time-size relations, which are based on the milling rate () and grinding limit. The models were applied to two hundred of particle sizes distributions curves resulting from grinding tests performed previously in aliquots of limestone and quartz (two origins) with control of mass and particle size. The milling times range 2 - 960 min with revolution speeds of 100 to 300 rpm. The fits of the kinetic and energetic models were evaluated considering the following factors: coefficient of determination (R2), standard error (SE), fit error (ε) and dependency index (ID). The kinetic model showed uncertainty associated with some of its parameters (ε > 10%) for most of the grinding conditions tested, resulting in unsatisfactory fits to the criteria used. The evaluation factors for the kinetic model were only suitable for one quartz, in the range of 38x75 μm satisfying the compensation condition. In the case of energetic models, the fits to the energy-size relation were better for those situations where the characteristic diameters did not show a stabilization in its decrease. On the other hand, the sizetime relation shown compatible fits with the situation where was observed a decrease static state of the characteristic diameters reaching the grinding limit. From the fits of the time-size relation was possible to determine a constant ′ that characterizes the resistance to fragmentation of the material depending on the grinding conditions studied. The values of this constant showed that materials more resistant to fragmentation have the smaller ′ values, ranging between 0.96 and 2.6 g/J for the limestone and from 0,06 to 0.53 g/J to quartz. It was concluded that the kinetic model was incompatible with ultrafine grinding, due to the presence of significant events of multi-particle interactions and agglomeration, which was confirmed by variation in the uniformity index (′) in milling time function. Energetic models complemented each other for description of the experimental results. Ready an intermediate model which consider a size reduction rate as a power function with an exponent η and a parameter (l) representing the grinding limit is the most recommended for the representation of the ultrafine grinding processes of industrial minerals.

<p>Crystallization is a key solid-liquid separation and purification technique used in pharmaceutical industry. Some of the critical quality attributes (CQAs) of a product from crystallization process include crystal size distribution (CSD), purity, polymorphic form, morphology, etc. Different size and polymorphs of a drug substance may have different dissolution profiles and different bioavailability, which can have adverse effect on human health. Therefore, it is important to design and control crystallization process to meet product CQAs. In recent years, drug substances are becoming more complex, often being heat sensitive, which may limit the temperature that can be used in the crystallization step. Consequently, the traditional cooling only crystallization may not be well suited to recover the high value drug substances. For these systems, antisolvent crystallization is typically employed to improve the yield. On the other hand, the solvent composition can significantly impact the polymorphic outcome. Therefore, designing combined cooling and antisolvent crystallization (CCAC) processes to solve the challenges of active pharmaceutical ingredient (API) crystallization in a highly regulated environment is a complex engineering problem. </p> <p>With rising energy costs and intense price competition from generic pharmaceutical companies, the pharmaceutical industry is looking for ways to reduce the cost of manufacturing via process intensification (PI). This thesis focuses on different PI techniques for CCAC of drug substances. Continuous or smart manufacturing is gaining popularity due to its potential to lower the cost of manufacturing while maintaining consistent quality. Continuous crystallization is an important link in the continuous manufacturing process. The first part of the thesis shows PI of a commercial drug substance, Atorvastatin calcium (ASC) for target polymorph development via continuous CCAC using an oscillatory baffled crystallizer (OBC). An existing batch CCAC process for ASC was compared with the continuous CCAC in OBC and it was found the continuous process 30-fold more productive compared to the batch process. An array of process analytical technology (PAT) tools was used in this work to assess key process parameters that affect the polymorphic outcome and CSD. The desired narrower CSD product was obtained in the OBC compared to that from a batch crystallizer.</p> <p>The next part of the thesis focused on model-based PI technique for efficient determination of crystallization kinetics of a polymorphic system in CCAC. A novel experimental design was proposed which significantly reduced the number of experiments required to determine crystallization kinetics in a CCAC process. The kinetic parameters were validated, and a validated polymorphic model was used to perform an in-silico design of experiment (DoE) to develop a design space that can be used to identify operating conditions to achieve a desired crystal size and polymorphic form. </p> <p>The final part of the thesis combines the experimental and model-based approach for designing a continuous CCAC process for ASC in a cascade of Coflore agitated cell reactor (ACR) and three-stage mixed suspension mixed product removal (MSMPR). A combined artificial neural network (ANN) and principal component analysis (PCA) method was used to calibrate an ultraviolet (UV) probe which was used to monitor ASC solute concentration in the cascade process. The crystallization kinetic parameters were estimated in ACR and MSMPR which was used to build a digital model of the cascade process. The digital model was then used to obtain a design space with different temperature profile in the three-stage MSMPR that yielded narrow CSD of ASC form I. Overall, this thesis demonstrates the benefits of applying PI in the CCAC of drug substances using a holistic approach including novel equipment, application of an array of PAT tools, and model-based digital design to achieve desired CQAs of the product.</p>

This study investigated the influence of the attainable region technique to ball milling as applied in reactor technology. Flow rate, ball filling, mill speed, ball size and mill density were varied. When each was varied, the rest of the parameters were kept constant in-order to determine the influence of each parameter on the process of milling. Selection function and breakage function parameters were selected for the mill model. These were kept constant for all four circuit configurations: open milling circuit, normal closed circuit, reverse closed circuit, and combined closed circuit. Flow rate was varied from 10 tph to 150 tph. It was observed that in all circuit configurations the optimum results were obtained from 90 tph upwards. When ball filling was varied, the optimum results were obtained between 30 % and 40 % of ball filling. At this range the mill is neither experiencing under-filling nor over-filling. When the mill speed was varied, at 60 – 80 % of critical speed the product specification was achieved and for grinding balls, sizes of between 60 mm and 90 mm yielded the optimum results. Varying the mill density resulted in insignificant changes. From the results, the combined closed circuit produced more of the product specification. / School of Engineering / M. Tech. (Engineering: Chemical)