Thesis (Ph. D.)--Mechanical Engineering, Georgia Institute of Technology, 2009. / Committee Chair: Yoda, Minami; Committee Member: Aidun, Cyrus; Committee Member: Breedveld, Victor; Committee Member: Fedorov, Andrei; Committee Member: Zhu, Cheng. Part of the SMARTech Electronic Thesis and Dissertation Collection.

Thesis (Ph. D.)--University of Canterbury, 2009. / Typescript (photocopy). "January, 2009." Includes bibliographical references (p. 213-227). Also available via the World Wide Web.

Thesis (M.S.)--Brigham Young University. Dept. of Chemistry and Biochemistry, 2007. / Includes bibliographical references.

Thesis (Ph. D.)--Oregon State University, 2006. / Printout. Includes bibliographical references. Also available on the World Wide Web.

L’utilisation d’outils microfluidiques pour la mise en œuvre de procédés sous-pression tels que des réactions chimiques, des synthèses de matériaux nano-structurés, ou en tant qu’outils de détermination de grandeurs physico-chimiques est une thématique de recherche récente. Quelques travaux précédents ont démontré l'intérêt des procédés supercritiques en microcanal pour la chimie organique et la synthèse de nanocristaux. Le développement de ces procédés est concomitant à la mise au point de dispositifs capables de résister à des conditions de pression et température élevées tout en étant compatibles avec l’utilisation de fluides supercritiques. Les avantages de ces fluides pour ce type de procédé sont une faible viscosité et une diffusivité élevée, ce qui offre des conditions de mélange favorables. Cependant, dans le même temps, les propriétés de transport – comme la masse volumique – sont très sensibles aux variations de température et de pression, qui ne peuvent être évités dans ces systèmes où les fluides sont en écoulement. Dans des systèmes diphasiques où le CO2 supercritique (PC = 74 bar) est utilisé en tant que solvant, les transferts thermique et de matière sont fortement influencés par la nature des écoulements. Dans un souci de maîtrise de ces procédés, la compréhension du comportement hydrodynamique, à la fois locale et globale, des fluides supercritiques en microcanal devient fondamentale. Dans cet objectif, un dispositif de microfluidique transparent et résistant à des pressions supérieures à la pression critique du CO2 a été développé. En adaptant une méthodologie propre à la lithographie molle, permettant la fabrication de puces microfluidiques pour des applications à pression atmosphérique, nous sommes parvenus à établir un protocole de fabrication de puces en verre et résine photosensible, viables pour une utilisation à plus de 100 bar en conditions CO2 supercritique. Grâce à ces dispositifs, des expérimentations d’ombroscopie ont pu être réalisées pour observer des écoulements composés de CO2 et d’éthanol dans le microcanal de section carrée de 200 x 200 µm à des pressions comprises entre 40 et 90 bar. Pour identifier et comprendre les phénomènes qui entrent en jeu lors de la création de l’écoulement à haute pression, une approche thermodynamique relative aux équilibres de phase est indispensable. En effet, la connaissance du diagramme de phase permet d’ores et déjà de représenter les zones d’équilibres thermodynamiques (pression, température et composition) pour lesquelles le mélange créé est monophasique liquide ou diphasique liquide-vapeur. L’illustration expérimentale par les séquences d’écoulement obtenues justifie la modélisation thermodynamique du diagramme de phase du binaire d’étude. Le régime d’écoulement de Taylor, obtenu spécifiquement à l’intérieur de la zone d’équilibre diphasique liquide-vapeur, est étudié. Ce régime est caractérisé par des bulles allongées entourées par un film liquide et séparées les unes des autres par une poche liquide. L’évolution des caractéristiques hydrodynamiques de ce régime – longueur de bulle, longueur de slug et vitesse de bulle – est étudiée en fonction des conditions opératoires, des débits et propriétés des fluides. L’objectif étant de repérer les similitudes avec les caractérisations à pression ambiante de la littérature et les particularités résultantes d’une manipulation à haute pression. Ce travail a été à l’origine de plusieurs avancées pour les communautés microfluidique et supercritique. Un nouveau protocole de fabrication à moindre coût de puces microfluidiques compatibles avec l’utilisation de CO2 supercritique et des méthodes de visualisation avancées est présenté. Une modélisation thermodynamique et une étude hydrodynamique expérimentale permettent de construire une carte d’écoulement des régimes biphasiques observés à haute pression ainsi qu’une caractérisation hydrodynamique du régime de Taylor à haute pression en microcanal. / The use of microdevices to run high pressure processes for chemical reaction, nanomaterial synthesis, or as analysis tools for determining physical properties have become of increasing interest in recent years. Several works in the literature have demonstrated the advantages of supercritical microfluidics for organic chemistry and complex nanomaterial synthesis. The development of pressure-resistant microfluidic chips, which also are compatible with the properties of supercritical fluids, is a key step in order to increase knowledge about these processes. Supercritical fluids have low viscosity and high diffusivity, which are advantageous for microprocesses since they facilitate mixing between species. However, the properties of these fluids are also very sensitive with small changes in pressure, temperature and composition. In twophase applications where supercritical CO2 may be used as solvent or reactant, these varying properties can result in very different flow patterns and hydrodynamics with pressure change. Since the hydrodynamics of such systems largely influence heat and mass transfer, the study of flow behavior under supercritical conditions in microchannel is fundamental. In pursuit of this objective, a transparent microdevice, which is suitable for experiments at pressures higher than critical pressure of CO2 (PC = 74 bar), has been developed in this thesis. Using a soft lithography method that is currently used to fabricate microfluidic chips for applications under ambient pressure, a methodology for fabricating a highly resistant chip made from glass and UV-curable polymer was developed. These chips can resist more than 100 bar in supercritical CO2 conditions. The microchips were then used to observe the flow behavior of a CO2-ethanol mixture created in a T junction microchannel (cross section: 200 x 200 µm) for pressures ranging from 40 to 90 bar using high-speed imaging. To identify and interpret phenomena that occur during the flow formation at high pressure, a thermodynamic approach was essential. Depending on the pressure, temperature and composition of the CO2-ethanol mixture, the flow at equilibrium can either be in the single phase liquid domain or in the two-phase liquid vapor domain, according to the phase diagram. Imaging experiments were conducted over the boundaries between the two-phase liquid vapor domain and the single phase liquid and the observed two-phase flow patterns and transitions confirm the predictions of the phase diagram. High-pressure CO2-ethanol Taylor flow, which was obtained in the twophase domain, was then studied. This flow pattern, which is characterized by elongated bubbles surrounded by a liquid film and separated from each other by liquid slugs, is well-known at low pressure and has been widely described in the literature. The objective here was therefore to compare the flow characteristics such as bubble length, slug length and bubble velocity obtained under high pressure operation with the behavior at low pressure. Differences coming from fluid characteristics or operating at high pressure were pointed . This work provides a variety of new results on high pressure microfluidics that will be of interest to both the microfluidics and the supercritical fluids communities. It presents a new protocol to fabricate low cost pressure-resistant microfluidic chips suitable for supercritical CO2 and advanced visualization methods. It also presents new findings obtained with this technology on map flow pattern at high pressure in correlation with thermodynamics approach and characterization of Taylor flow hydrodynamics under high pressure in microchannel.

Le cytosquelette d’actine régule la forme de la cellule au cours du temps. Pour lecomprendre, il faut étudier les mécanismes moléculaires qui le constituent. In vivo,ces mécanismes sont masqués par la complexité du vivant. Si nous parvenons àreproduire pièce par pièce le cytosquelette d’actine in vitro et si nous pouvons lecontrôler, aussi bien dans l’espace et dans le temps, alors nous pourrons élucider lessecrets de son fonctionnement. Cette thèse montre que nous avons maintenantdéveloppé la technologie qui nous permet de le faire pour certaines architectures ducytosquelette d’actine.Nous avons développé de nouvelles méthodes de « micropatterning » pour contrôlerla nucléation des monomères d’actine dans l’espace en deux dimensions. Ceci nous aamenés à la reconstitution et au guidage de réseaux de filaments d’actine parallèles àpolarité identique et à la reconstitution de l’anneau de cytocinèse.J’ai crée une puce microfluidique innovante pour contrôler la compositionbiochimique des systèmes d’actine reconstitués au cours du temps. Ceci nous a permisde contrôler la cinétique de polymérisation du filament d’actine individuel libre, et decontrôler la séquence d’intervention des protéines sur les réseaux de filamentsd’actine parallèles à polarité identique.Enfin, nous avons utilisé cette même puce microfluidique pour étudier ladifférentiation des cellules souches hématopoïétiques. / Actin cytoskeleton regulate cell shape over time. To understand that, we have to studymolecular mechanisms that constitute actin cytoskeleton. In vivo, those mechanismsare hidden by cellular complexity. If we achieve to reproduce piece by piece actincytoskeleton in vitro and if we can control it in space and time, then we are able toelucidate the secrets of it operate. This thesis show that we have developed thetechnology that allow us to do it for a few actin cytoskeleton architectures.We have developed new micro-patterning methods to control actin monomersnucleation into two-dimensional space. This led us to the reconstitution and guidanceof parallel actin filament networks with same polarity and allowed us to reconstituteactin contractil ring.I created an innovating microfluidic chip to control biochemical composition ofreconstituted actin systems over time. This allowed us to control kinetics of freeindividual actin filament polymerisation and to control the intervention sequence ofproteins on parallel actin filament networks.Finaly, we used the microfluidic chip to study hematopoïetic stem cellsdifferentiation.

Monitoring and controlling pH is of utmost importance in bioprocessing as it directly affects product yield and quality. Multiplexed experiments can be performed in nanobioreactors for optimisation of yield and cell heterogeneity in a relatively quick and inexpensive manner. In this thesis, a pH holographic sensor (holosensor) is miniaturised to 3.11 nL in volume and integrated into a PDMS-glass microfluidic chip for monitoring the growth of Lactobacillus casei Shirota. Although other established methods for monitoring cell cultures can be utilised, miniaturised holosensors enable real-time and non-consumptive monitoring of the bacterial cell culture growth medium. The 2-hydroxyethylmethacrylate (HEMA)-co-2-(trifluoromethyl) propenoic acid (TFMPA) holosensor was fabricated using an adapted technique from photolithography, coupled with the use of a polymerisation inhibitor to control the gel polymerisation with diameters not exceeding a standard deviation of 0.067. The hologram brightness was optimised to 1.05 ms integration time with 36X magnification using a low power (0.290 mW) 532 nm green continuous wave (CW) laser with a devised beam-offset technique. The holosensor was characterised with ionic strength balanced (9.50 mS/cm) McIIvaine pH buffers and a calibration curve plotted together with measured ionic strength, optical density at 600 nm (OD600) and pH. Correspondingly, RGB-xyY transformed values were plotted in the CIE 1931 chromaticity diagram. Later, a miniaturised 0.4φ HEMA-co-TFMPA holosensor and array was also demonstrated. Together with the 3.0φ holosensor, an accuracy parameter for the 0.4φ spot and array holosensors were calculated to be 99.08%, 99.38% and 97.77% respectively. Further work involved studying the issues associated with fabricating gels with unusually flat gel profiles. Other preliminary results suggested the alternative of utilising polymers as a holosensor substrate, together with a dye-free method for hologram fabrication, outlined the prospective possibility of a miniaturised holosensor integrated into a polymer microfluidic chip with the flexibility of hologram colour customisation for cell culture monitoring.

abstract: Filtration for microfluidic sample-collection devices is desirable for sample selection, concentration, preprocessing, and downstream manipulation, but microfabricating the required sub-micrometer filtration structure is an elaborate process. This thesis presents a simple method to fabricate polydimethylsiloxane (PDMS) devices with an integrated membrane filter that will sample, lyse, and extract the DNA from microorganisms in aqueous environments. An off-the-shelf membrane filter disc was embedded in a PDMS layer and sequentially bound with other PDMS channel layers. No leakage was observed during filtration. This device was validated by concentrating a large amount of cyanobacterium Synechocystis in simulated sample water with consistent performance across devices. After accumulating sufficient biomass on the filter, a sequential electrochemical lysing process was performed by applying 5VDC across the filter. This device was further evaluated by delivering several samples of differing concentrations of cyanobacterium Synechocystis then quantifying the DNA using real-time PCR. Lastly, an environmental sample was run through the device and the amount of photosynthetic microorganisms present in the water was determined. The major breakthroughs in this design are low energy demand, cheap materials, simple design, straightforward fabrication, and robust performance, together enabling wide-utility of similar chip-based devices for field-deployable operations in environmental micro-biotechnology. / Dissertation/Thesis / Additional Paper / M.S. Civil and Environmental Engineering 2011

Orientador: Antonio Riul Júnior / Banca: Cleber Renato Mendonça / Banca: Nilson Cristino da Cruz / O Programa de Pós-Graduação em Ciência e Tecnologia de Materiais, PosMat, tem caráter institucional e integra as atividades de pesquisa em materiais de diversos campi da Unesp / Resumo: Fabricamos neste trabalho microcanais em uma matriz de PDMS para otimização de uma configuração que permita, futuramente, a inserção de eletrodos interdigitados no interior dos mesmos para a integração da "língua eletrônica" que estamos trabalhando ao longo dos últimos anos com dispositivos "lab-on-a-chip". O objetivo final é a fabricação de um sensor "tongue-on-a-chip", não havendo nada similar na literatura até o presente momento, tendo-se em vista, principalmente, o potencial de aplicação de ambos dispositivos ("língua eletrônica" e "lab-on-a-chip"). Neste sentido, esta dissertação torna-se uma chave importante para o desenvolvimento de uma tecnologia nova e com forte apelo comercial. Como a fabricação dos microcanais envolve técnicas e equipamentos de litografia que não dispomos em nossos laboratórios, estendemos colaborações com o laboratório Nacional de Luz Síncroton (LNLS - através de um projeto específico nessa linha de atuação), onde fabricamos os microcanais e eletrodos interdigitados envolvidos neste trabalho. Como é a primeira vez que a microfluídica está sendo aplicada em dispositivos do tipo "língua eletrônica", tivemos um trabalho minucioso de verificação das melhores condições envolvidas na fabricação dos dispositivos. As medidas em fluxo no interior dos microcanais mostraram-se mais rápidas e eficientes que as estáticas utilizadas anteriormente, e, adicionalmente à redução no volume das amostras analisadas com os microcanais, necessitamos ainda de um refinamento para aplicações futuras (análises clínicas e biológicas, controle ambiental, análise de bebidas...), pois o maior empecilho encontrado até o momento tem sio a selagem do dispositivo devido à deposição de filmes ultrafinos sobre os eletrodos metálicos, que esperamos resolver em trabalhos futuros. / Abstract: In this work microchannels were fabricated in a PDMS matrix to optimize a configuration that allows, in future works, the insertion of interdigitated electrodes into the microchannel for the integration of the electronic tongue that we have being working in the last couple of years with lab-on-a-chip devices. The final goal is the fabrication of a tongue-on-a-chip sensor, having nothing similar in the literature up to date, bearing in mind the high potential of application of both devices (electronic tongue and lab-on-a-chip). In that sense, this Msc work becomes an important key to the development of a new technology with strong commercial appeal. As the microchannel fabrication needs equipments and techniques not available in our laboratory, we extend the collaboration with the Brazilian Synchrotron Light Laboratory (LNLS - throughout a specific project in this area) where the microchannels and interdigitated electrodes were fabricated. As it is the first time microfluidic is applied in e-tongue sensors, we did a detailed work verifying the best conditions involved in the device fabrication. Flow measurements inside the microchannels shown to be faster and more efficient than the static ones previously made, and, besides the volume reduction of the samples analysed with the microchannels, we still need a refinement for future applications (clinical and biological analysis, environmental control, beverage analysis, ...), as the major problem has being the sealing of the device due to the deposition of ultra-thin films onto the interdigitated electrodes, which we hope to solve in future works. / Mestre

Proteomics is an increasingly important area of biological research and has gathered much attention over recent years. Major challenges that make a proteomic analysis difficult are sample complexity, diversity and dynamic range. Progress in the area of proteomics relies heavily on new analytical tools for the sensitive, selective, and high-throughput studies of target analytes. It is estimated that there are several hundred thousand proteins in a human cell. In order to be able to analyze such a complex sample, an analytical method must be capable of separating and detecting many different sample peaks. The complexity of such samples indicates that a single separation method will not be able to provide the needed resolution. If two methods that are orthogonal are combined, then the peak capacity of the combined system is the product of the two individual peak capacities. Development of such systems would cater to the current demands of proteomics studies. Matrix assisted laser desorption/ionization (MALDI) mass spectrometry has evolved into a primary analytical tool for proteomics research. MALDI is fast and efficient and has a high tolerance to non-volatile buffers and impurities. The samples for MALDI are typically applied to solid supports after having been subjected to off-line liquid or gel separations. Several methods have been reported involving various chromatographic or electrophoretic separation methods. However, the current methods often require highly sophisticated sample handling systems, which are often expensive and in need of skilled human resources. The current demands of proteomic analyses require fast, efficient and inexpensive methods for separation to fully harness the capability of MALDI mass spectrometry. In this work a microfluidic device has been designed to perform dynamic isoelectric focusing (DIEF) based protein separation with digital sample deposition directly on a MALDI target for offline analysis. DIEF is related to capillary isoelectric focusing which and can facilitate the interface without the loss of the separation resolution. Compared to traditional capillary isoelectric focusing (cIEF) DIEF uses additional high-voltage power supplies to control the pH gradient by manipulating the electric field. The proteins can be focused at a desired sampling position according to their isoelectric point, to be collected for further analysis by MALDI mass spectrometry. DIEF has a peak capacity of over a thousand and offers an ease of interfacing to other techniques making it a preferred separation method for the interface with mass spectrometric techniques such as MALDI. The design of the microfluidic device is based on a digital droplet fractionation. Multiple fractions of the sample solution from DIEF are generated to retain the resolution and to act as an additional separation mode. The microfluidic device is controlled by actuating pneumatic valves built into the device. The DIEF operational parameters were optimized according to the surface functionality and the design of the microfluidic device. A suitable MALDI sample preparation method was found by studying different existing methods. The methods were studied using test proteins prepared in solutions having the additives used in the experiment. A simple mixture of three proteins was used to demonstrate the application of the developed method. The separation between the proteins insulin, hemoglobin and the myoglobin was demonstrated by varying the separation resolution in three experiments.