The work in the dissertation expands the applications of DIEF and describes the development of incorporating DIEF in a microfluidic chip to create a comprehensive proteomics tool. Proof-of-concept DIEF experiments have been done previously, so the focus of this work is to explore the capabilities of DIEF. Dynamic isoelectric focusing (DIEF) is a separation technique invented by Dr. Luke Tolley. It is similar to capillary isoelectric focusing except it uses four high voltage electrodes to form a pH gradient instead of only two. The additional two electrodes are able to manipulate the pH gradient resulting in selection of the region and of the range of pH within a pre-defined sampling or extraction point. One of the first applications described for DIEF was to isolate a single protein from a complex mixture. The protein isolated was a cellulase enzyme capable of degrading multiple cellulose materials over a wide range of environmental conditions. DIEF did isolate the protein in a pH span of 0.005 which is equivalent to 0.075% of the total pH range. Fractions were collected for sequencing analysis, but the fractions were contaminated with keratin both times. DIEF was also successfully performed in an open air channel. Though other electromigration techniques have been successfully done in open air channels, these techniques were severely time and pH limited. In contrast, DIEF in an open air channel is capable of using the entire 3-10 pH range and can perform isolations until the proteins are completely separated. The device developed was also an improvement on increasing sample capacity. The channel was significantly bigger than the traditional glass capillaries used. Since the channel was open, fraction collection was made simpler by collecting using a pipette. This work also demonstrated that DIEF can be made through the use of silicone molding compounds and polyurethane. The amount of milling needed is reduced, the pieces are produced quickly, and a single mold can produce several pieces. Machining pieces with fragile bits is not needed to be done as much since only one acrylic piece is required produce a mold. The mold can produce several polyurethane pieces. This fabrication method has proven useful for making DIEF holders. The next step was to make DIEF a truly comprehensive proteomic tool by incorporating it into a microfluidic chip. Multiple sample fractions are rapidly generated on chip through the use of multiple bubbles simultaneously injected into the separation channel. This stops the separation and, since each droplet is isolated from others by a bubble on each side, the protein peaks are not able to broaden. This novel use of digital microfluidics is still a work in progress, but the fundamentals have been demonstrated. The fabrication protocol for making molds and PDMS casts was developed using materials and procedures that can be done in a common laboratory environment. DIEF is a separation technique still in its infancy, with a wide variety of available applications. DIEF will continue to be tested in other areas and developed into a comprehensive proteomic tool.

The present study introduces an analysis of the computer science resources as a tool of teaching-learning and its contributions on the construction of knowledge in students who have special necessities. It was objectified, with this intention, to verify the occurrence of changes on the behavior of these students in cognitive, affective and social aspects through pedagogical observations in classroom and in the computing environment, as well as, to detect its contribution to the development of new pedagogical practices. The universe of the research embraces a state public school, focusing the students in the first grades of elementary school who participate in computing activities. As an instrument to collect data, we searched for the semi-structured interviews that were applied to the students, their parents, schoolteachers and special teachers of the resources room and to the student teachers of the Special Education Course. Documents were used as points of view, students works and computer science projects with the purpose of ratifying or comparing data and information. After the research, the study concluded that the use of educational computer science applied to the Special Education supported the learning of the students, contributing to their growth. It was observed, through the work on the computer that this resource helped the four students who participated in this research, being decisive on their motor progress, on their self-esteem, self-confidence, progression and challenges, involving the teaching-learning process, offering them more motivation, creativity, development of the co-operative spirit, of the lectowritten and intensification in the socio affective relations, above all in the process of communication and interaction, promoting the social and scholar inclusion. / O estudo realizado apresenta uma análise dos recursos da informática como ferramenta de ensino-aprendizagem e suas contribuições na construção do conhecimento dos alunos com necessidades especiais. Neste intuito, objetivou-se verificar a ocorrência de mudanças de comportamento nesses educandos nos aspectos cognitivo, afetivo e social, através de observações pedagógicas em sala de aula e no ambiente informatizado, bem como, detectar a sua contribuição para o desenvolvimento de novas práticas pedagógicas. O universo da pesquisa abarca uma escola da rede pública estadual, centrando-se nos alunos incluídos nas séries iniciais do ensino comum que participam de atividades em ambiente informatizado. Como instrumento para coleta de dados, recorreu-se às entrevistas semiestruturadas aplicadas aos alunos, aos seus pais, as suas professoras do ensino comum, às educadoras especiais da sala de recursos e às estagiárias do curso de Educação Especial. Utilizaram-se documentos como Pareceres, produções das alunas e projetos de Informática, com a finalidade de ratificar, confrontar dados e informações. Após a pesquisa, o estudo concluiu que o uso da informática educativa aplicada à Educação Especial favoreceu o aprendizado das alunas, contribuindo para o seu crescimento. Observou-se, através do trabalho realizado no computador, que este recurso auxiliou as quatro alunas, participantes desta pesquisa, sendo decisivo no seu progresso motor, na sua auto-estima, segurança e desafios, envolvendo o processo ensino-aprendizagem, oferecendo-lhes maior motivação, criatividade, desenvolvimento do espírito cooperativo, da lecto-escrita e intensificação nas relações sócio-afetivas, sobretudo no processo de comunicação e interação, promovendo a inclusão social e escolar.

Submitted by Automação e Estatística (sst@bczm.ufrn.br) on 2018-07-02T21:58:30Z No. of bitstreams: 1 WashingtonCavalcanteDaSilva_DISSERT.pdf: 19392258 bytes, checksum: a56a3a92ea29a494f3f687951927abd8 (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2018-07-09T12:29:25Z (GMT) No. of bitstreams: 1 WashingtonCavalcanteDaSilva_DISSERT.pdf: 19392258 bytes, checksum: a56a3a92ea29a494f3f687951927abd8 (MD5) / Made available in DSpace on 2018-07-09T12:29:25Z (GMT). No. of bitstreams: 1 WashingtonCavalcanteDaSilva_DISSERT.pdf: 19392258 bytes, checksum: a56a3a92ea29a494f3f687951927abd8 (MD5) Previous issue date: 2018-05-28 / A lógica linear (LL) tem se consolidado como um bom arcabouço para especificar sistemas computacionais, uma vez que fórmulas podem ser interpretadas como recursos que podem ser consumidos e/ou produzidos. Além disso, do ponto de vista da Teoria da Prova, a LL reconcilia o aspecto construtivo da lógica intuicionista com a simetria da lógica clássica. Desta maneira, é possível, de uma forma mais flexível, modelar estados de um sistema como fórmulas lógicas, e as transições entre esses estados como etapas na construção de uma prova. No entanto, a busca por provas, em geral, não é determinística, pois existem diferentes maneiras de se provar a mesma proposição. Visando reduzir esse problema, um sistema de provas focado pode ser utilizado nesse processo de busca. A grosso modo, um sistema focado considera provas em forma normal, reduzindo assim as ocorrências de provas não essenciais que são sintaticamente diferentes, mas equivalentes no final do processo. Neste trabalho, é apresentada a prova da completude de um sistema focado para a LL, bem como outras propriedades essenciais desse sistema. Além disso, o sistema focado será utilizado para especificar e verificar protocolos de votação, que definem como deve ser escolhido um candidato, feita uma contagem de votos, e divulgado o resultado de uma eleição. Para que isso seja possível, dois protocolos de votação serão especificados formalmente utilizando sistemas de transição de estados, que modelam de forma natural os estados e comportamento de tais protocolos. Junto a isso, para cada sistema, uma especificação em LL será definida e demonstrada que é correta, ou seja, que um passo focado representa uma determinada transição no sistema de estados modelado. Por fim, propriedades inerentes aos protocolos de votação, como a garantia de que o resultado da eleição reflete o desejo dos eleitores, serão apresentadas e demonstradas através de derivações no sistema focado. / Linear logic (LL) has been consolidated as a good framework for specifying computational systems, since formulas can be interpreted as resources that can be consumed and / or produced during a proof. From the point of view of Proof Theory, LL reconciles the constructive aspect of intuitionistic logic with the symmetry of rules in classical logic. Hence, LL offers a more flexible way to model states of a system as well as transitions among those states. We note that the proof search procedure is inherently non-deterministic since there are different ways of proving the same proposition. In order to tame this problem, focused systems have been proposed aiming at reducing the number of choices during the proof search procedure. Roughly, a focused system considers normal form proofs, thus eliminating the occurrences of nonessential proofs that are syntactically different but equivalent. In this work, we present the proof of the completeness of a focused system for LL as well as some other essential properties of this system. In addition, we use the focused system for specifying and verifying voting protocols. Such systems define how a candidate is chosen by tallying votes and computing the final result of an election. We formally specify two voting protocols using transition systems, which naturally model the states and behavior of such protocols. For each system, an encoding in LL is defined and we show that our specification is correct: a focused step corresponds exactly to a transition in the modeled system. Finally, we verify important properties of the voting protocols such as the fact that the system ensures that the election result reflects the voters’ desire.

Made available in DSpace on 2014-10-09T12:27:10Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T13:57:19Z (GMT). No. of bitstreams: 0 / A Radiocirurgia Estereotática é a técnica de tratamento que usa feixes estreitos de radiação focalizados com grande exatidão em uma lesão pequena. A introdução dos colimadores micro multi-lâminas (mMLC) permite que essa técnica alcance um maior grau de conformação de dose na lesão alvo, possibilitando uma menor irradiação das estruturas críticas e dos tecidos normais. Este trabalho visa apresentar os resultados dos testes de aceite e do comissionamento de um acelerador linear Varian ® 6EX dedicado à Radiocirurgia, associado com o colimador multi-lâminas da BrainLab ® instalado no Hospital das Clínicas da Faculdade de Medicina da USP (HC-FMUSP) e estabelecer um Programa de Garantia de Qualidade que seja exequível para os serviços que pretendem implantar essa técnica especial. Os resultados dos testes de aceitação mostraram-se coerentes e satisfatórios com as especificações definidas pelo fabricante e os testes de comissionamento ficaram dentro das recomendações internacionais. Os testes e as medidas que compõem o processo de controle de qualidade devem ser específicos para cada unidade de tratamento, assim como a necessidade, a frequência e os níveis de tolerância. / Dissertacao (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN-CNEN/SP

O presente trabalho tem como objetivo principal a eletrofiação de nanofibras poliméricas de poliacrilonitrila (PAN) e polifluoreto de vinilideno (PVDF), incorporadas com negro de fumo (NF) e ftalocianina de cobre (CuPc), visando aplicações em dispositivos sensores. Inicialmente foram preparadas soluções de PAN puro a 6 % em peso e PVDF puro a 20% em peso e foram misturadas a essas soluções partículas de negro de fumo e ftalocianina de cobre, obtendo soluções de PAN/NF, PVDF/NF, PAN/CuPc e PVDF/CuPc. Foi determinada a viscosidade absoluta das soluções. Realizou-se a eletrofiação para obtenção de nanofibras que foram caracterizadas segundo o diâmetro e morfologia, usando microscópio óptico e microscópio eletrônico de varredura. Para avaliar as interações polímero-polímero, polímero-partícula foram analisadas por espectroscopia FITR e Raman. Com as fibras de PAN/NF foi analisada a resistência e condutância elétrica das membranas usando um picoamperímetro digital, visando aplicação como filtro eletrostático. Foi construído canal na lâmina de silício usando um feixe de laser visando a deposição de fibras dentro do canal usando a técnica de focagem eletrodinâmica com tensão aplicada em máscaras de cobre. Foi usada a técnica da microbalança de cristal de quartzo para determinar a variação de massa adsorvida por membranas de PAN/CuPc e PVDF/CuPc por meio da medida da variação de frequência usando um frequencímetro digital, onde se observou que essas membranas são promissoras para atuar como sensores de vapor de amônia. / The main objective of this work is the incorporation of different particles in order to electrospun polymeric nanofibers of polyacrylonitrile (PAN) and polyvinylidene fluoride (PVDF), aiming at applications in sensor devices. Initially, solutions of PAN pure 6 wt% and PVDF pure 20 wt% were prepared and these solutions were mixed with carbon black (NF) particles and copper phthalocyanine (CuPc), obtaining solutions of PAN/NF, PVDF/NF, PAN/CuPc and PVDF/CuPc. The absolute viscosity of the solutions was determined. The electrospinning was performed to obtain nanofibers that were characterized according to the diameter and morphology, using optical microscope and scanning electron microscopy. To evaluate the polymer-polymer and polymer-particle interactions, FITR and Raman spectroscopy were performed. The resistance and conductance of the membranes electrospun from PAN/NF solution were analyzed using a digital picoammeter, and an increase in the resistance was measured. This result shows that the membrane is suitable to be applied as electrostatic filter. A channel was constructed on the silicon wafer using a laser beam for the deposition of fibers inside the channel using the electrodynamic focusing technique. The quartz crystal microbalance technique was used to determine the applicability of the membranes as sensor layer. The results of PAN/CuPc and PVDF/CuPc membranes suggests that these membranes are promising to act such as ammonia vapor sensors.

Nous étudions la génération de rayonnement TeraHertz (THz) dans des microplasmas produits par des lasers femtosecondes. Cette technique est prometteuse pour créer efficacement des sources THz compactes et étendue spectralement (0.3-30 THz), qui intéressent de nombreuses applications, comme l’identification spectroscopique de substances dangereuses ou encore l’imagerie en biologie et médecine. Contrairement aux sources conventionnelles, comme les interrupteurs photo-conducteur, les sources THz basées sur des plasmas ne sont pas limitées par la tenue au flux et couvrent l’ensemble du spectre THz. Afin de modéliser des microplasmas générés par des faisceaux laser fortement focalisés, nous présentons un nouvel algorithme qui permet d’injecter tout type de laser dans des codes électromagnétiques. Nous dérivons aussi un modèle compatible avec les équations de Maxwell qui inclut les deux mécanismes générateurs de THz: le courant d’ionisation (IC) et le mécanisme “Transition-Cherenkov” (TC). Ce dernier mécanisme domine la production de THz pour des lasers à plusieurs cycles optiques, où l’émission est produite par les courants d’électron longitudinaux. Dans le cas des microplasmas où un champ électrostatique externe est ajouté, le taux de conversion énergétique laser/THz peut être augmenté de deux ordres de grandeur via le mécanisme IC lorsque le champs statique ou la pression du gaz sont accrus. De plus, les simulations 3D montrent que pour un faisceau laser à deux couleurs et dans des conditions optimales de focalisation, une énergie laser de 10 micro-Joule est suffisante pour atteindre des taux de conversion bien au-dessus de 10−4. Dans ce cas, la nature transverse du courant IC est cruciale pour accroitre l’efficacité avec la longueur du plasma. En considérant un faisceau laser à deux couleurs de forme elliptique, nous proposons de contrôler les spectres d’émission en exploitant les effets plasmoniques résonants. / We investigate terahertz (THz) generation in fs-laser-induced microplasmas, which are promising candidates for compact and efficient broadband THz sources (0.3-30 THz). Such sources have various applications as spectroscopic identification of hazardous substances or THz imaging in biology and medicine. Unlike conventional THz sources as photoconductive switches, gas-plasma-based THz sources do not suffer from irreversible material damage and can cover the whole THz range at once. To simulate tightly-focused-laser-induced microplasmas, we propose an efficient numerical algorithm that can introduce any arbitrarily shaped laser pulses into electromagnetic codes. We derive a Maxwell-consistent model that includes two major THz generation mechanisms, the ionization current (IC) and transition-Cherenkov mechanisms (TC). The latter mechanism is shown to dominate for single-color multi-cycle lasers pulses where the emission is driven by longitudinal electron currents. For microplasmas a constant electric field can boost the laser-to-THz converison efficiency by two orders of magnitude via the IC mechanism when increasing the gas-pressure and bias-voltage. Moreover for two-color-driving laser pulses, Maxwell-consistent 3D simulations show, that only 10 μJ laser pulse energy are sufficient to reach conversion efficiencies well above 10−4 when optimizing the focusing conditions. Here, the transverse nature of the IC currents is crucial for the up-scaling of the efficiency with the plasma length. By using elliptically-shaped two-color-driving laser beams, we propose to control the emission spectra by exploiting resonant plasmonic effects.

Les ondes de choc de cisaillement dans les solides quasi-incompressibles sont un type d'onde peu exploré car accessible à la mesure seulement depuis une douzaine d'années. Elles se distinguent des ondes de choc de compression par leur non-linéarité cubique au lieu de quadratique. Les faibles vitesses des ondes de cisaillement, de l'ordre du mètre par seconde, permettent notamment d'atteindre des régimes assez fortement non linéaires sur de faibles distances, et de contrebalancer l'absorption relativement importante dans ces milieux. A notre connaissance, ces ondes ont été étudiées presque exclusivement en ondes planes. Or, la focalisation est un phénomène fondamental dans la physique des ondes et ses applications. C'est cette focalisation que l'on s'est attaché à étudier spécifiquement ici, tant du point de vue théorique et numérique, qu'expérimental. L'objectif était de mettre en évidence la focalisation des ondes de choc de cisaillement. Ce travail est en particulier motivé par son application potentielle en biomécanique. On a ici fait l'hypothèse que les ondes de choc de cisaillement pourraient jouer un rôle dans la formation de certaines lésions cérébrales suite à un traumatisme crânien, la géométrie du crâne induisant leur focalisation. Ces considérations nous ont conduits à dimensionner les phénomènes étudiés en conséquence, en termes de fréquence, d'amplitude et de géométrie. La focalisation a ainsi été montrée dans le cas d'un gel simulant un tissu biologique. Les comparaisons avec le modèle théorique et numérique se sont avérées favorables. Pour approfondir l'analyse de l'hypothèse, il resterait à prouver que ces ondes de choc peuvent induire des lésions. / Shear shock waves in quasi-incompressible soft solids have been observed experimentally only twelve years ago. They differ from compression shock waves because their nonlinearity is cubic instead of quadratic. Velocities of shear waves are small, of the order of the meter per second, and thus induce a strong nonlinear behavior over small distances. This allows to counter-balance the quite strong absorption in such media. To our knowledge, these shear shock waves have been studied only for plane waves. However, focusing is a fundamental phenomenon in the wave physics and its applications. This is why our objective is to investigate the focusing of shear waves theoretically, numerically, and experimentally. This work was in particular motivated by a potential application in biomechanics. We have hypothesized that shear shock waves could play a role in the formation of some traumatic brain injuries, the geometry of the skull producing the focusing effect. These hypotheses have governed our studies for the choice of the different parameters such as the frequency, the amplitude and the geometry. The focusing has been demonstrated in a biological tissue-mimicking gel. Comparisons with the theoretical and numerical model have shown a good agreement. Further studies should concern the possible formation of lesions by shear shock waves.

There is a strong push worldwide to develop multi-Joule femtosecond duration laser pulses at wavelengths around 3.5-4 and 9-11 mu m within important atmospheric transmission windows. We have shown that pulses with a 4 mu m central wavelength are capable of delivering multi-TW powers at km range. This is in stark contrast to pulses at near-IR wavelengths which break up into hundreds of filaments with each carrying around 5 GW of power per filament over meter distances. We will show that nonlinear envelope propagators fail to capture the true physics. Instead a new optical carrier shock singularity emerges that can act to limit peak intensities below the ionization threshold leading to low loss long range propagation. At LWIR wavelengths many-body correlations of weakly-ionized electrons further suppress the Kerr focusing nonlinearity around 10 mu m and enable whole beam self-trapping without filaments.

Cette thèse a pour objectif de mieux comprendre les mécanismes physiques qui contrôlent les trajectoires de particules anisotropes dans des écoulements confinés, afin d’en améliorer la prédiction. Nous avons dans un premier temps développé des outils expérimentaux basés sur la microscopie et le traitement d’images afin d’analyser les positions de particules en écoulement confiné dans des microcanaux de section carrée. Ces outils ont ensuite permis l’obtention de résultats originaux sur la migration latérale de particules sphériques dans des écoulements faiblement inertiels. Nous avons montré en particulier que les particules migrent au centre du canal à faible nombre de Reynolds et à proximité du centre de chaque face à Reynolds plus élevé et que ces deux régimes co-existent pour des Reynolds intermédiaires. Parallèlement à leur migration latérale, les particules en écoulement confiné peuvent s’espacer régulièrement sous certaines conditions pour former des trains. Ce travail a donc consisté à mener une étude statistique pour quantifier et localiser la formation des trains. Il a été montré que la formation des trains était contrôlée par la configuration de l’écoulement dans le sillage des particules et que leurs caractéristiques, i.e., le pourcentage de particules en trains et la distance interparticulaire, étaient fonction du nombre de Reynolds particulaire. Enfin, des résultats préliminaires sur le cas d’écoulements bi-disperses ont été obtenus. Pour terminer, les perspectives et développements futurs de ce travail sont dégagés. / This thesis aims to better understand the physical mechanism controlling the trajectories of particles in confined flows in order to improve predictive models. In the first step we have developed experimental tools based on microscopy and image analysis in order to identify the particles positions in confined flows in square section microchannels. These tools have then permitted to obtain original results on the lateral migration of spherical particles in flows at low inertia. In particular we have shown that neutrally buoyant particles in square channels are focused at channel center at low Reynolds number and at four channel faces at high Reynolds, and that there is a co-existence of the two regimes for intermediate Reynolds. In addition to their lateral migration, under certain conditions, particles can also align in the flow direction to form trains of evenly spaced particles. This work has thus been devoted to the statistical study on the quantification and localization of the train formation and configuration. It has been shown that the formation of trains is controlled by the flow configuration in the wake of the particles, and that the train characteristics, i.e., the fraction of particles in trains and the interparticle distance, are functions of the particle Reynolds number. Finally, preliminary results on flows of bidisperse suspensions have been obtained. To conclude, the perspectives and future developments of this work are presented.

Gekrümmte Röntgen-Multilagen-Spiegel fokussieren Synchrotron-Strahlung auf wenige zehn Nanometer große Strahlflecke. Die Propagation zweier Lichtwellen durch elliptisch geformte fokussierende Multilagen-Spiegel wird durch eine wellenoptische Theorie beschrieben; dieser analytische Ansatz wird numerisch integriert. Mit diesem Verfahren kann die Schichtstruktur für hohe Reflektivität und geringe Strahl-Aberrationen optimiert werden. Wir zeigen Simulationen aktueller und neu entwickelter Synchrotron-Strahlrohre. Die Grenzen der Theorie werden anhand von Modell-Parametern untersucht.Ein weiterer Teil dieser Arbeit beschäftigt sich mit partieller räumlicher Kohärenz, modelliert mit der Methode der stochastischen Superpositionen. Propagation und Filterung der Kohärenzeigenschaften eines Röntgenstrahls in einem Wellenleiter werden analytisch und numerisch untersucht. Das vorgestellte, anschauliche Modell soll die Entwicklung und Überprüfung neuer Algorithmen im Bereich der kohärenten Abbildung unterstützen. Wir vergleichen die Ergebnisse von Simulationen, die auch Realstruktureffekte berücksichtigen, mit experimentellen daten aus dem GINIX-Instrument am Kohärenz-Strahlrohr P10 an der PETRA III-Quelle am DESY.Diese Arbeit zeigt Ergebnisse aus einer Kollaboration zwischen der Georg-August-Universität Göttingen und der European Synchrotron Radiation Facility (ESRF) in Grenoble.