Dissertations / Theses on the topic 'Continuous pharmaceutical manufacturing'

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, February 2010.
Cataloged from PDF version of thesis.
Includes bibliographical references (p. 290-299).
Continuous manufacturing offers a number of operational and financial benefits to pharmaceutical companies. This research examines the critical blending step for continuous pharmaceutical manufacturing and the characteristics of continuous downstream pharmaceutical manufacturing systems. Discrete element method (DEM) simulations were used to develop novel insights into the mechanism of mixing for continuous blending of cohesive pharmaceutical powders and to examine the effects of particle properties, blender design and operating conditions on blend homogeneity. To place continuous blending into the context of pharmaceutical manufacturing, the scope of the analysis was expanded to process system models of continuous downstream pharmaceutical manufacturing. DEM simulations were used to study the mechanisms of mixing in the continuous blending of pharmaceutical powders. Diffusive mixing from the avalanching particles appears to be the dominant mechanism of mixing in both the axial and radial direction for the double helical ribbon blender. This result can guide the development of future continuous pharmaceutical powder blenders by optimizing the mixing elements to increase the amount of particles transported to a position where they can avalanche/flow and diffusively mix. A range of particle properties, blender designs and operating conditions were examined for their effects on flow behavior and blend homogeneity. Three particle properties were examined: particle size, polydispersity and cohesive force.
(cont.) Particle size was observed to be positively correlated to both flow rates and blend homogeneity. Polydispersity had no effect on flow rate and was negatively correlated to homogeneity. Cohesive force was negatively correlated to flow rate and had little to no effect on homogeneity. Two modifications of blender design were analyzed: changes in blender size and changes in shaft design. Blender size was observed to be positively correlated to flow rate and negatively correlated to homogeneity. The paddle shaft designs created a more homogeneous powder blend than the double helical ribbon shaft. Two operating parameters were also studied: rotation rate and fill fraction. Rotation rate was positively correlated to both flow rate and homogeneity. Fill fraction had the interesting result of being positively correlated to the absolute flow rate, but negatively correlated to the fill mass normalized flow rate. In addition, fill fraction has a clear negative correlation to homogeneity above fill fractions of 0.55, but is inconsistent for fill fractions lower than this. This research on particle properties, blender designs and operating conditions will help to guide the operation of continuous pharmaceutical blenders and the design of continuous pharmaceutical manufacturing systems. Process simulations comparing model batch and continuous downstream pharmaceutical manufacturing systems have quantified some of the potential size, cost and performance benefits of continuous processes. The models showed significant reductions in process equipment sizes for continuous manufacturing particularly in the blending step.
(cont.) This reduction in equipment size translates to capital cost (CAPEX) savings for both the continuous process equipment and manufacturing facilities. The steady state operation of continuous processing also reduces the labor requirements and gives the continuous processes an operating cost (OPEX) advantage over batch processes. This research has contributed to the understanding of continuous pharmaceutical powder blending and quantified some of the benefits of continuous downstream pharmaceutical manufacturing. This work is being continued by the Novartis-MIT Center for Continuous Manufacturing whose work is providing the foundation for future industrial scale pharmaceutical continuous manufacturing systems.
by Matthew J. Abel.
Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 2011.
Cataloged from PDF version of thesis.
Includes bibliographical references.
Melt extrusion is an alternative processing technique that operates continuously, reduces the total number of unit operations, allows for incorporation of difficult-to-process drug substances, and has the potential to achieve tablets of better quality and consistency compared to traditional methods. Thus, our goal was to evaluate melt extrusion as a viable processing alternative and expand our scientific knowledge such that we gain predictive capabilities of tablet characteristics, i.e., quality by design. This new knowledge will aid future process design thereby helping to reduce time and costs associated with pharmaceutical solid dosage form production. The residence time distribution for melt extrusion has been characterized using a single parameter model. When combined with assumed first-order reaction rate kinetics and an Arrhenius reaction rate constant, the model can accurately predict the amount of drug product lost to temperature driven degradation. The model prediction agreed well with experimentally determined fractional conversion. The physical stability of amorphous Molecule A was characterized using enthalpy of relaxation measurements. Molecular level rearrangements are the source of physical instability for the fragile glass forming Molecule A. The instability can be modified by introducing a second component, which contributes to the overall enthalpy change. Coating amorphous Molecule A tablets with a polyvinyl alcohol based coating material reduces moisture uptake during storage. The coating material preferentially uptakes water from the atmosphere, restricting moisture from entering the tablet core and causing premature dissolution or degradation. The dissolution behavior of Molecule A tablets can be tailored with the addition of water soluble materials. Dissolution rate constants for Molecule A tablets have been calculated for different formulations and can be used as a resource when designing new solid dosage forms with desired dissolution characteristics. A novel 100% Molecule A melt extrusion process has been created, reducing the number of overall unit operations and eliminating troublesome blending inconsistencies. An additional formulation that maintains the crystallinity of Molecule A by processing with polyethylene glycol below Molecule A's melting temperature is physically and chemically stable and ready for implementation in a continuous production line. The mixing achieved within the extruder for this formulation is sufficient to eliminate a pre-mixing unit operation.
by Erin R. Bell.
Ph.D.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Chemical Engineering, 2015.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 92-105).
In this thesis work, we aimed to explore crystallization processes for small molecule API compounds based on engineered polymer surfaces that could be used in continuous manufacturing. First, we identified a library of polymers that can be used and selected PVA as the model polymer based on its solution and film properties. We also illustrated a rational approach for designing and fabricating PVA film surfaces for increasing heterogeneous nucleation rate of different compounds and enable polymorph selection. The design philosophy was to select prevalent angles between major faces of crystals according to a selection of compounds, and to create substrate surfaces with indentations that include these angles. Nucleation induction time trends showed that heterogeneous nucleation rates were accelerated by at least an order of magnitude in the presence of PVA due to the favorable interactions between the model compounds and the polymer. Nucleation rates were further increased for patterned substrates with matching geometries. Surface indentations with non-matching angles resulted in faster nucleation rates than flat films but slower than matching geometries because they only increased the effective area of the films and their roughness. X-ray diffraction was used to reveal faces that preferentially interacted with the PVA side chains and to deduce possible arrangement of solute molecules at the corners of the indentations. Combining X-ray data and morphology of the crystal product, we suggest that matching geometries on the substrate enhanced nucleation of compounds. In addition to enhancing nucleation rate, polymorph selection was possible in the presence of the polymer substrate to yield a higher percentage of thermodynamically stable gamma indomethacin. Offline Raman experiments and in-line morphology determination confirmed that polymorph control of the final crystal product via kinetic control of the nucleation process was viable. For the aspirin system, the 85 degree angle lead to the highest rate of nucleation; for the polymorphic indomethacin system, XRPD results showed that gamma form preferentially formed on the PVA films with 65 and 80 degree angles leading to the largest reduction in nucleation induction time. Kinetic Monte Carlo simulation showed that a crystallizer incorporating both nucleation and crystal growth in the absence of active mass transfer would have too small a throughput and too large a footprint to be useful. The main reasons were long average nucleation induction times and slow crystal growth in the absence of convection. A set of batch desupersaturation experiments showed that mass transfer limited growth dominate the crystal growth kinetics at low supersaturations when nucleation events were suppressed. An increase in the bulk fluid velocity increased the effective growth kinetics in the system when mass transfer kinetics dominated. Steady state modeling based on the first principle approach was performed using a combination of Navier Stokes Equations and diffusion-convection mass transport equations. The modeling result demonstrated that for mass transfer from a moving fluid to a stationary surface, a thin momentum and concentration boundary layer existed at the leading edge, which resulted in much higher local mass transfer rates. In the absence of momentum boundary layers, mass transfer could only occur via diffusion, which resulted in slow growth kinetics. The first principle model was used to derive dimensionless number correlations for the continuous crystallizer.
by Li Tan.
Ph. D.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Materials Science and Engineering, 2012.
Page 43 blank. Cataloged from PDF version of thesis.
Includes bibliographical references (p. 41-42).
The development of manufacturing tablets in a continuous way has been possible greatly to the fabrication of polymer based thin-films. It is estimated that the pharmaceutical industry loses as much as a 25% on revenues based on the currently employed batch manufacturing method. Here we studied a continuous way of manufacturing tablets based on API/based polymer formulations that are cast and subsequently rolled into a tablet. Selections of two active pharmaceutical ingredients (SPP-100 and Acetaminophen) were studied into how well it forms mechanical robust, chemical and physical compatible HPMC polymer based films. As well, HPMC polymer based films with no drug loading were compared to measure out the dispersion of the drug on the film. Physiochemical studies were performed by DSC, XRD, FT-IR, and SEM. Moisture content was measured out by Karl Fischer Titration and mechanical properties such as tensile strength were measured for all API/HPMC and placebo films. It was found that the mechanical and physiochemical properties of SPP-100/HPMC films were regarded as the most promising thin film tablet candidate and it is further being tested for other mechanical properties such as bonding, friction, and compression.
by Jose R. Barcena.
S.B.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2013.
Cataloged from PDF version of thesis.
Includes bibliographical references (p. 51-53).
In this thesis, a continuous rollforming process for the folding of thin-films was proposed and studied as a key step in the continuous manufacturing of pharmaceutical tablets. HPMC and PEG based polymeric thin-films were considered for this application. An experimental apparatus was designed and developed to test the folding of thin-films. The experimental apparatus was designed in a modular fashion to facilitate testing of various process parameters. Analysis was carried out for the folding operations, based on which two folding strategies were proposed - (i) without scoring and (ii) with scoring. The first strategy relies on elastic deformation of the thin-films, whereas the later depends on localized, plastic deformation caused by the scoring geometry. From the experiments on folding we identified three regimes of process operation namely: insufficient scoring, appropriate scoring, and excessive scoring. The implications of different levels of scoring were observed and understood carefully for the scoring and folding operation. Practical guidelines were developed for carrying out folding successfully and the scope of future work was discussed.
by Ryan Slaughter.
S.M.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Chemical Engineering, 2016.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 121-126).
In the pharmaceutical industry there has recently been much interest in design spaces: sets of critical process parameters (CPPs) which guarantee that critical quality attributes (CQAs) of a manufacturing process are within specifications. For continuous pharmaceutical processes, design spaces are usually calculated by assuming steady state operation and approximating the mapping between CPPs and CQAs using a Taylor series. The full design space can then be calculated using a plantwide approach or a unit-by-unit approach. Common inner approximations of the design space (e.g. hyper-rectangles) can result in significant conservatism, especially when a unit-by-unit approach is employed. Because control loops tend to have a linearizing effect on processes, design spaces for closed-loop processes can often be calculated using low-order Taylor series approximations, resulting in simpler expressions for the full design space (e.g. polytopes). Control loops also tend to enlarge design spaces, sometimes by more than an order of magnitude. Unfortunately, disturbances, noise, and uncertainties will prevent real processes from ever reaching "steady state". Therefore, design spaces calculated at steady state cannot be used to guarantee quality specifications. In fact, because design spaces fail to take into account any process dynamics, constraining a controller to work within a design space may result in failure to meet quality specifications, significant degradation of controller performance, and input jitter. As a substitute for design space, robust model predictive control (RMPC) is a promising technology for dynamically guaranteeing constraint satisfaction on process outputs. Although many RMPC algorithms have been proposed in the literature, the computational cost of these algorithms tends to be a strong function of the state vector size. This is problematic for continuous pharmaceutical processes, which are typically high- or infinite-dimensional. However, input-output models (e.g. finite step response models) can integrated with traditional RMPC strategies to robustly control high-dimensional systems. Although RMPC can be used to counteract the presence of disturbances, uncertainty, and measurement noise, faults also present a threat to quality constraint satisfaction of continuous pharmaceutical processes. Active fault diagnosis of hybrid systems is particularly difficult due to the explosion of mode combinations with prediction horizon. Fortunately, the set of input sequences which do not guarantee diagnosis can be outer bounded offline as a function of a parameterized initial condition set. This enables an algorithm for guaranteed active fault diagnosis of hybrid systems which can be implemented quickly online.
by Lucas Charles Foguth.
Ph. D.

Thesis (M.B.A.)--Massachusetts Institute of Technology, Sloan School of Management; and, (S.M.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering; in conjunction with the Leaders for Global Operations Program at MIT, June 2011.
"June 2011." Cataloged from PDF version of thesis.
Includes bibliographical references (p. 72-73).
Previous studies of continuous manufacturing processes have used equipment-factored cost estimation methods to predict savings in initial plant investment costs. In order to challenge and validate the existing methods of cost estimation, feature-based cost estimates were constructed based on a modular process design model. Synthesis of an existing chemical intermediate was selected as the model continuous process. A continuous process was designed that was a literal, step by step, translation of the batch process. Supporting design work included process flow diagrams and basic piping and instrumentation diagrams. Design parameters from the process model were combined with feature-based costs to develop a series of segmented cost estimates for the model continuous plant at several production scales. Based on this analysis, the continuous facility seems to be intrinsically less expensive only at a relatively high production scale. Additionally, the distribution of cost areas for the continuous facility differs significantly from the distribution previous assumed for batch plants. This finding suggests that current models may not be appropriate for generating cost estimates for continuous plants. These results should not have a significant negative impact on the value proposition for the continuous manufacturing platform. The continuous process designed for this project was not optimized. Therefore, this work reiterates that the switch to continuous must be accompanied with optimization and innovation in the underlying continuous chemistry.
by Donovan Collins.
S.M.
M.B.A.

Traditionally pharmaceutical manufacture is conducted on a batch basis but significant resources are being invested into the use of intensified continuous processes. This dissertation evaluates the use of a combined twin screw and segmented fluid bed drying process to produce granules on a continuous basis. The experimental program was conducted using structured Design of Experiments in three stages. • Wet granulation only: Investigated the initial relationships between liquid/solid ratio and power required for wet granulation, as well as granule structure using SEM Imaging. • Wet granulation and fluid bed drying: Concluded that the biggest control over, the measured mean granule size (d50) produced from the combined system was still the ratio of water to dry powder in the wet granulation. • Wet granulation through to compression: The effects of changes in the granulation process were not statically relevant on the final tablet for the process set up. The study also used PEPT data to assess motion within the TSG. The studies showed: • The time spent in the kneading zone directly after the liquid addition in relation to the overall time spent in the granulation process appears independent of the process j . set up at 32% ± 2%. • As the barrel speed of the granulator increases the relative time spent in the final ' breakage zone' of the TSG increases, therefore increasing breakage. Using the findings from the literature, the results of the experimental program were used to define the mechanisms occurring within the TSG. The experimental findings were input into a model to predict the outcome of collisions between particles. The model predicts agglomeration of the smaller particles to the larger ones and by calculating changes in the viscosity of the binder the subsequent secondary agglomeration of these granules can also be shown using this model The model is limited due to assumptions in deriving it. The model excludes capillary forces that if given sufficient time to form could have the same order of magnitude strength as other forces.

Thesis (M.B.A.)--Massachusetts Institute of Technology, Sloan School of Management; and, (S.M.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering; in conjunction with the Leaders for Global Operations Program at MIT, 2010.
Cataloged from PDF version of thesis.
Includes bibliographical references (p. 52-53).
Manufacturing in the pharmaceutical industry is presently characterized as a batch production system, which has existed in its current form for decades. This structure is the result of historical regulatory policy as well as the conservative nature of the industry. Recent clarification by US and European regulatory bodies has opened the possibility to new approaches to the manufacturing process. This combined with changes in the market for the pharmaceutical industry has accelerated the rate at which new manufacturing technologies are explored. Continuous manufacturing is a paradigm shift in the pharmaceutical industry manufacturing structure, encompassing several new technologies and systems. The business impact of continuous manufacturing has not been well defined. This assessment aims to compare a continuous manufacturing process to a batch manufacturing process for a particular Novartis product. The product has an established batch production process. Cost estimates and the continuous process cost is estimated using a four-step process: defining the process flow, performing the material balance, estimating the capital costs, and estimating the operating costs. This analysis shows that for the particular Novartis product considered, a continuous process is an improvement over the batch process in four performance characteristics: capital investment, operating cost, throughput time, and working capital requirement.
by Kristopher Ray Wilburn.
S.M.
M.B.A.

Thesis: S.B., Massachusetts Institute of Technology, Department of Mechanical Engineering, May, 2020
Cataloged from the official PDF of thesis.
Includes bibliographical references (pages 65-66).
Lyophilization, or freeze drying, of unit-dose pharmaceuticals eliminates the need to preserve quality through regulated temperature storage during transport. Typical lyophilization is performed through batch processing, but implementing continuous manufacturing instead can lead to improved process control, higher quality product, and increased flexibility. Lyophilization involves four processing chambers at specific pressures and transferring vials between the chambers requires a load lock system. Load lock systems, commonly used in the semiconductor industry, act as a transfer chamber in order to reduce pump-down and venting time of its adjacent processing chambers. This thesis documents the research and design of a load lock system for a continuous manufacturing lyophilization machine.
by Serena Le.
S.B.
S.B. Massachusetts Institute of Technology, Department of Mechanical Engineering

Robust process monitoring in real-time is a challenge for Continuous Pharmaceutical Manufacturing. Sensors and models have been developed to help to make process monitoring more robust, but they still need to be integrated in real-time to produce reliable estimates of the true state of the process. Dealing with random and gross errors in the process measurements in a systematic way is a potential solution. In this work, we present such a systematic framework, which for a given sensor network and measurement uncertainties will predict the most likely state of the process. As a result, real-time process decisions, whether for process control, exceptional events management or process optimization can be based on the most reliable estimate of the process state.

Data reconciliation (DR) and gross error detection (GED) have been developed to accomplish robust process monitoring. DR and GED mitigate the effects of random measurement errors and non-random sensor malfunctions. This methodology has been used for decades in other industries (i.e., Oil and Gas), but it has yet to be applied to the Pharmaceutical Industry. Steady-state data reconciliation (SSDR) is the simplest forms of DR but offers the benefits of short computational times. However, it requires the sensor network to be redundant (i.e., the number of measurements has to be greater than the degrees of freedom).

In this dissertation, the SSDR framework is defined and implemented it in two different continuous tableting lines: direct compression and dry granulation. The results for two pilot plant scales via continuous direct compression tableting line are reported in this work. The two pilot plants had different equipment and sensor configurations. The results for the dry granulation continuous tableting line studies were also reported on a pilot-plant scale in an end-to-end operation. New measurements for the dry granulation continuous tableting line are also proposed in this work.

A comparison is made for the model-based DR approach (SSDR-M) and the purely data-driven approach (SSDR-D) based on the use of principal component constructions. If the process is linear or mildly nonlinear, SSDR-M and SSDR-D give comparable results for the variables estimation and GED. The reconciled measurement values generate using SSDR-M satisfy the model equations and can be used together with the model to estimate unmeasured variables. However, in the presence of nonlinearities, the SSDR-M and SSDR-D will differ. SSDR successfully estimates the real state of the process in the presence of gross errors, as long as steady-state is maintained and the redundancy requirement is met. Gross errors are also detected whether using SSDR-M or SSDR-D.

Tese de mestrado, Engenharia Farmacêutica, Universidade de Lisboa, Faculdade de Farmácia, 2018
In the last decade, there have been significant advances in the areas of engineering and science, allowing the implementation of pharmaceutical continuous manufacturing (CM). These advances are coupled with the adoption of the quality by design paradigm for pharmaceutical development and the advances on process analytical technology to improve design, analysis and control. These advances have contributed significantly to advances on the design and manufacturing of pharmaceutical, namely the adoption of continuous processing. Continuous manufacturing can be used for the production of medicines in multiple pharmaceutical forms. If advances are being operated in the field of continuous manufacturing, it is also true that substantial efforts are still required to fully understand how this manufacturing paradigm can be efficiently integrated within industry, for full advantages to be achieved. This work had the objective of pursuing the goal to better understand how materials behave under continuous processing. It aimed at evaluating the performance of a direct compression process for tablets production in continuous mode. Direct compression, a unit operation especially interesting for the manufacturing of tablets, has the ability to be efficiently integrated within a continuous manufacturing framework. The investigated continuous manufacturing process resourced to the continuous table production line located at the PROMIS Center at the School of Pharmacy of the University of Eastern Finland (Kuopio, Finland). The line was configured for direct compression purposes, encompassing multiple gravimetric feeders, a continuous mixer and a tableting machine. Tablets monitoring was accomplished with near-infrared spectroscopy. A formulation containing simultaneously caffeine (2.6%) and paracetamol (20%) was selected for this study. Selected process variables were varied according to an experimental design in order to understand the effects on tablets' properties. Mixer speed (350-1200 rpm), feed rate (5-10 kg/h) and the existence or not of premixture were the selected process variables. Tablets were evaluated according to the weight, hardness and thickness. Feed rate demonstrated as was expected to be fundamental for the stability of the direct compression process. For instance, experiments carried out at lower feed rates (lower than 5kg/h) revealed poor fluidity and tablets were not acceptable. The influence of process variables on tablets properties was modelled by partial least squares regression. Tablets mass is significantly affected, in a positive way, by the speed of the mixer and negatively affected by the feed rate. The range of coefficients of determination for the calibration (R2) and test (Q2) for the three responses were 0.78-0.94 for R2 and 0.56-0.88 for Q2. Near-infrared spectra collected from tablets allowed the development of PLS models for the caffeine and paracetamol content. Validation experiments reveal that the root mean square errors of prediction for caffeine and paracetamol were respectively 11.96% and 10.48%.
Historicamente, a produção em descontínuo de formas de dosagem sólidas teve grande sucesso e dominou a indústria farmacêutica. Durante muito tempo a indústria entendeu não haver motivação para inovar no sentido do desenvolvimento de novas tecnologias de fabrico, dada a rentabilidade desta forma de produção. No entanto, atualmente, na era pós-blockbuster, tendo em conta que os custos dos materiais, durante o desenvolvimento de medicamentos, são significativos, que novos medicamentos, provavelmente, serão fabricados em quantidades muito menores e que, para novos tratamentos, o desenvolvimento de um processo de produção comercial não é garantido, é cada vez mais reconhecida a necessidade de novos paradigmas de produção. A produção em contínuo surge como uma alternativa à produção em (semi-) descontínuo e tem por objetivo aumentar a eficácia e a eficiência na produção farmacêutica. Esta nova abordagem exige que a indústria farmacêutica, primária e secundária, aborde de maneira diferente a forma como desenvolve e otimiza os processos de fabrico para produção de substâncias ativas e formulações farmacêuticas. É fundamental a compreensão do processo como um todo, bem como inovação ao nível empresarial. É crucial entender e minimizar a variabilidade das matérias-primas, executar medições contínuas durante o processo, definir uma amostragem representativa e caracterizar a propagação de alterações e distúrbios através do sistema. Ao contrário da produção em descontínuo, em que o controlo local de cada equipamento é considerado suficiente, na produção em contínuo, o controlo local é não só obrigatório, como todo o fluxo do processo deve ser coordenado e equipado com sistemas de controlo de segundo nível, supervisionando e controlando todas as operações unitárias. Monitorizar e controlar a composição de um produto durante todo o seu processo de fabrico, com a finalidade de alcançar a qualidade e robustez pretendida, é um importante objetivo a ser alcançado numa produção em modo contínuo. Um passo importante para a implementação da produção em contínuo foi dado em 2004, quando a Food and Drug Administration publicou uma diretriz de tecnologias analíticas de processo (PAT), que promove a adoção de tecnologias inovadoras para realizar medições oportunas em atributos críticos de qualidade de materiais brutos e em processo, permitindo alcançar uma melhor compreensão e controlo do processo. O conceito PAT está intimamente ligado à ideia do desenho pela qualidade a qual considera não apenas a avaliação de risco para a qualidade e o conhecimento sobre o processo, mas também a forma como as operações unitárias afetam a qualidade e estabilidade do produto. A espectroscopia de infravermelho próximo tem sido utilizada como uma ferramenta de controlo PAT. Esta ferramenta de controlo combinada com a análise de dados multivariados tornou-se uma ferramenta interessante na análise farmacêutica, tanto a nível qualitativo como quantitativo. Esta tese foi realizada utilizando a linha de produção contínua de comprimidos do PROMIS Centre (Escola de Farmácia, Universidade do Leste da Finlândia) em Kuopio, Finlândia. A monitorização do processo foi realizada em tempo real, por um sistema de infravermelho próximo com uma câmara espectral SPECIM e um sensor ImSpector (SPECIM, Finlândia). Com o propósito de avaliar o processo de compressão direta em modo contínuo foram definidas um conjunto de experiências variando a velocidade do misturador (350 a 1200 rpm), fluxo (5 a 10 kg/h). Foi ainda avaliada importância de existência de etapa de pré-mistura. Como respostas foram avaliados a massa, dureza e espessura dos comprimidos. A análise de componentes principais foi usada como método de análise exploratória dos espetros obtidos, assim como para identificar medições atípicas. A monitorização da taxa de alimentação, elemento fundamental na compressão direta, decorreu sem grandes variações em relação aos set points definidos para cada matéria prima. Deste modo, os dados analisados, relativamente à taxa de alimentação de cafeína e paracetamol, permitiram perspetivar que as concentrações dos mesmos, nos comprimidos seriam as esperadas. Durante a realização das experiências ficou ainda visível que as que tinham uma taxa de alimentação de 5 kg/h, evidenciavam uma fraca fluidez, já com valores de 10 kg/h a fluidez melhorava significativamente. De acordo com os modelos de regressão múltipla (PLS) para avaliar a influência das variáveis alteradas nas respostas selecionadas, a massa de cada comprimido é significativamente afetada, de uma forma positiva, pela velocidade do misturador e negativamente afetada pela taxa de fluxo. Para além disso, este modelo prevê que a dureza seja afetada pela existência ou não de pré-mistura. Para os três modelos o R2 não se verificou muito elevado, variando entre 0.78 e 0.94. Relativamente ao valor de Q2, este variou entre 0.56 e 0.88, valores um pouco a baixo dos valores ideais. Assim, pode-se concluir que este não será um modelo com uma capacidade preditiva muito elevada. Relativamente aos modelos PLS, baseados nos espetros NIR para estimar a concentração de paracetamol e cafeina, foi possível concluir que a capacidade preditiva foi boa com erros quadrados médios (RMSEP) de 12 e 10% para o paracetamol e cafeína respetivamente.
The experimental work was performed in PROMIS continuous tablet manufacturing line (University of Eastern Finland, School of Pharmacy, Kuopio, Finland). All the facilities, equipments, materials and support were gently provided by University of Eastern Finland.

yes
Solid state shear milling (S3M) is reported as a scalable, continuous, polymer-assisted cocrystallization technique. A specially designed milling pan was employed to provide high levels of applied shear, and the addition of a polymeric processing aid enabled generation of high stress fields. Carbamazepine–salicylic acid cocrystals were produced with 5–25 wt % of poly(ethylene oxide) (PEO). A systematic study was carried out to understand the effect of process variables on properties and performance of the cocrystals. S3M offers an important new route for continuous manufacturing of pharmaceutical cocrystals.

Trabalho Final de Mestrado Integrado, Ciências Farmacêuticas, Universidade de Lisboa, Faculdade de Farmácia, 2016
Pharmaceutical manufacturing has been changing over the last years, especially, due to significant advances in science and engineering. The lack of flexibility, quality consistency, robustness of manufacturing, the need to respond faster to market shortage and the necessity of better process control demanded the establishment of new technologies, such as Process Analytical Technology (PAT), Quality by Design (QbD) and, more recently, Continuous Pharmaceutical Manufacturing (CPM). Through these approaches became possible to design manufacturing processes using principles of engineering, material science and quality control to ensure acceptable and reproducible product quality and performance throughout a product's shelf life. Fundamentally, they allow an understanding of the product and manufacturing process, starting with product development and basically building quality in, not testing it. This review focuses on these three approaches, but specially on the continuous manufacturing process, that is starting to be incorporated and developed in some breakthrough pharmaceuticals. Companies, especially the ones with new biopharmaceuticals, have begun utilizing continuous manufacturing as an alternative to batch manufacturing in various parts of the production process. Continuous manufacturing allows that materials are sent directly and continuously to the next step for further processing, after reliably produce an intermediate material or product with acceptable characteristics, guaranteeing their consistent quality. Unfortunately, in Portugal, the development of these techniques still needs to be encouraged. This could happen because companies have the perception that the existing regulatory system is rigid and unfavorable to the introduction of innovative systems. On the contrary, regulatory entities support de introduction of these technologies, providing guidelines with models for an effective quality management system for the pharmaceutical and biotech industry.
A indústria farmacêutica tem vindo a sofrer mudanças nos últimos anos, especialmente, devido aos avanços na ciência e na engenharia. A falta de flexibilidade, consistência na qualidade dos produtos, robustez do processo de produção e a necessidade de responder, com brevidade, à falta de medicamentos e de ter um melhor controlo do processo exigiu o estabelecimento de novas tecnologias como o Process Analytical Technology (PAT), o Quality by Design (QbD) e, mais recentemente, a produção contínua. Através destas estratégias torna-se possível desenhar o processo de produção usando fundamentos de engenharia, ciência dos materiais e de controlo de qualidade, assegurando uma qualidade do produto aceitável e reprodutível e também a sua performace durante o ciclo de vida do medicamento. Fundamentalmente, estas permitem um conhecimento sobre o produto e o processo de produção, começando no desenvolvimento do produto e, desta forma, construindo qualidade e não apenas testando-a. Esta monografia tem como foco estas três abordagens, mas especialmente o processo de produção contínua que está a começar a ser incorporado e desenvolvido em alguns medicamentos de ponta. As indústrias, especialmente aquelas a produzir novos biossimilares já começaram a utilizar a produção contínua, como uma alternativa aos processos com lotes, em várias secções do processo de produção. A produção contínua permite que os materiais sejam enviados, directa e continuamente, para o próximo passo, depois de produzirem materiais intermédios ou o produto final com fiabilidade e características aceitáveis, garantindo uma qualidade consistente. Infelizmente, em Portugal, o desenvolvimento destas técnicas ainda precisa de ser encorajado. Isto poderá acontecer, já que as indústrias têm a percepção que o sistema regulatório é rígido e não vê de forma favorável a introdução de sistemas inovadores. Contrariamente, as entidades regulatórias incentivam a introdução das mesmas, providenciando guidelines com modelos para um sistema de controlo de qualidade efectivo para a indústria farmacêutica e biotecnológica.

A questão da produtividade em investigação e desenvolvimento (I&D) é um assunto de grande debate, principalmente porque um importante resultado destas actividades é intangível. Algo que é transversal a todo o I&D é a incerteza e o risco de falhar, que forçam a que a produtividade seja medida como a média entre o falhanço e o sucesso. Na industria farmacêutica, para ambos os ramos de negócio (inovadores e genéricos), o tempo necessário para chegar ao mercado e o custo em I&D são factores-chave de sucesso, impulsionados grandemente pela produtividade da I&D. Quality by Design (QbD) e produção em contínuo são dois conceitos sobre os quais a FDA expressou o desejo que as empresas construam e apliquem nos seus processos e produtos para melhorar a qualidade, conhecimentos de processo, flexibilidade e robustez na manufactura de produtos farmacêuticos. Nesta tese, a implementação de química em fluxo e a metodologia de desenvolvimento em contínuo são apresentados através de casos de estudo baseados na pipeline da Hovione em fase de desenvolvimento e comercial. Estes casos de estudo incluem diferentes transformações químicas, tais como N-alquilações, ciclizações intramoleculares, reacções organometálicas e formação de sais para a preparação de ingredientes farmacêuticos activos (trifenatato de vilanterol e brometo de umeclidínio) que envolvem desvantagens, como longo tempo de reacção, reacção em condições criogénicas, reagentes perigosos, intermediários sensíveis, formação de impurezas ou longos tempos de separação e purificação do produto. De modo a entender a mudança de mentalidade entre desenvolvimento de processos em descontinuo e em contínuo, os casos de estudo foram desenvolvimentos e optimizados começando com o procedimento descrito em patentes e considerando as vantagens de operar em modo contínuo. A implementação de química de micro-ondas, tecnologia analítica de processo (PAT), desenho de experiências (DoE) e modelação cinética suportaram a redução do tempo, custo, risco de falha do desenvolvimento, aumentando consequentemente a produtividade e o conhecimento do processo. Os casos de estudo resultaram em maior rendimento, menor tempo de reação, menor preocupação com segurança, em alguns casos maior pureza e menor teor de impurezas. As melhorias no processo através dos casos de estudo mostraram que a produção em contínuo pode ser uma alternativa à produção em descontínuo. A metodologia desenvolvida demonstrou ser uma nova estratégia que fornece um fluxo de trabalho para desenvolvimento de processos químicos que explora oportunidade de produção em contínuo seguindo o conceito de QbD.
The question of research and development (R&D) productivity is a matter of a great debate, mainly because an important outcome of these activities is intangible. Transversal to all R&D is the uncertainty and risk of failure, which forces the productivity to be measured as an average between failure and success. In the pharmaceutical industry, for both innovator and generic branches of the business, time to market and R&D cost are key success factors, largely driven by R&D productivity. Quality by Design (QbD) and continuous manufacturing are two concepts that FDA expressed desire that companies build in and apply for their processes and products to improve quality, process understanding, flexibility and robustness in the manufacture of pharmaceuticals. In this thesis the implementation of flow chemistry and the continuous development methodology are presented through case studies based on the development and commercial pipeline of Hovione. These case studies include different chemical transformations, such as N-alkylations, intramolecular cyclization, organometallic reaction and salt formation for the preparation of active pharmaceutical ingredients (vilanterol trifenatate and umeclidinium bromide), that involve drawbacks, such as long reaction time, cryogenic reaction conditions, hazardous reagents, sensitive intermediates, impurity formation or time-consuming downstream processing. In order to understand the mindset difference between batch and continuous process development, the case studies were developed and optimized starting from previously described procedure of patents and considering the drivers of operating in continuous mode. The implementation of microwave chemistry, process analytical technology (PAT), Design of Experiments (DoE) and kinetic modelling supported the reduction of development time, cost and risk of failure, therefore increased productivity and process understanding. The case studies resulted in higher yield, shorter reaction time, reduced safety concerns, in some cases higher purity and lower impurity content. The process improvements showcased that continuous manufacturing could be the alternative of traditional batch production. The developed methodology demonstrated to be a novel strategy that provided a workflow for chemical process development exploring the opportunities for continuous manufacturing following the QbD concept.

Over the last years the pharmaceutical industry has been gradually starting to adhere to the continuous production instead of the traditional batch production. For the production of tablets through continuous direct compression (CDC) it is essential to ensure the control of all the equipment involved in the process, with special attention to the Loss-in-Weight (LIW) Feeders. This is due to the fact that even if there are small oscillations in the initial phase of the process, these will have big repercussions on the quality of the final product, making the drug getting out of the specification. This work explore different materials with the goal of creating a methodology for the selection of the components (the tools) to be assembled on the feeder in order to optimise the process. Furthermore, the minimum and maximum limits of the feeders were tested to determine a stable range of operation for each analysed material. Afterwards, a database with the different properties of the materials was developed with the goal of relating these properties with the components that would be necessary to integrate on the feeder and optimise its performance. To simplify the identification of similarities between the analysed materials, a principal component analysis (PCA) model was developed. Following this, some alternatives to reduce the time and costs of the production process were identified. Thus, the development of partial least squares (PLS) models was essential to predict the feed factor (FF) values. Lastly, the impact that the usage of the feeder had over the materials was analysed in order to find out if this equipment would cause any alterations in the properties of the materials to be used in the formulation.
Nos últimos anos a indústria farmacêutica tem vindo gradualmente a aderir à produção em contínuo em detrimento da tradicional produção em batch. Para a produção de comprimidos através de continuous direct compression (CDC) é essencial garantir o controlo de todos os equipamentos do processo, com especial atenção para os Loss-in-Weight (LIW) Feeders. Tal deve-se ao facto de que mesmo que sejam pequenas as oscilações na fase inicial do processo, estas terão grandes repercussões na qualidade do produto final, dando origem a um medicamento fora de especificação. Durante este trabalho procurou-se testar diversos materiais com o objetivo de criar uma metodologia para a seleção dos componentes (peças) a serem integrados no feeder de forma a otimizar o seu funcionamento. Para além disso, foram também testados os limites mínimos e máximos de funcionamento do feeder, de forma a se determinar um intervalo de operação estável para cada material analisado. Posteriormente, foi desenvolvida uma base de dados com as diversas propriedades dos materiais, cujo objetivo seria relacionar essas propriedades aos componentes que viriam a ser necessários integrar no feeder de forma a otimizar o seu funcionamento. Com o objetivo de facilitar a deteção das semelhanças entre os materiais analisados foi desenvolvido um modelo de principal component analysis (PCA). Em seguida, foram identificadas alternativas para reduzir o tempo e o custo dos processos de produção, para o que foi essencial o desenvolvimento de modelos de partial least squares (PLS) para prever os valores dos feed factor (FF). Por último, foi analisado o impacto que a utilização do feeder teve sobre os materiais, no sentido de se averiguar se a utilização deste equipamento provocaria alterações nas propriedades dos materiais a serem utilizados na formulação.

This work aims at exploring two applications of Raman spectroscopy in the development and manufacturing of pharmaceutical tablets made from amorphous solid dispersions (ASDs), in terms of both physical and chemical properties. The first application consisted on applying Raman spectroscopy to predict the tablets’ hard-ness. In contrast to typical destructive hardness testing, this strategy can be implemented as at-line and real-time process control via process analytical technology and thus provide real-time release during continuous manufacturing. This is investigated here for the first time for tablets composed of ASDs rather than crystalline-based formulations using a Confocal Raman Micro-scope. The generated data show an apparent relationship between the measured hardness and the Raman spectra baseline, consistent with previous findings in the literature. However, a more stringent analysis leads to the conclusion that the trend line is not strictly monotonic and with a slope across the measured range comparable to the variability of the Raman imaging analysis. The second application consisted on exploring the potential of confocal Raman microscopy to provide spatially resolved chemical images of the tablets’ surface, which allows to characterize the components spatial distribution. The goal was to evaluate the impact of two process param-eters: blending shear rate and sieving mesh size. These parameters were investigated using a structured 2-level design of experiments where Confocal Raman microscopy allowed the evalu-ation of the spatial distribution and the detection of API agglomerates and its relationship with the process parameters. However, the analysis was hindered by a lack of a perfectly flat surface of the tablets and to correct the influence of the change in focus on the Raman spectra, post-pro-cessing scripts were implemented. It is shown that there is an impact of both process parameters on the heterogeneity of tablets’ surface and, consequentially, a significant impact in disintegration of the tablets.