No description available.

Le chauffage par micro-ondes (MO) est employé depuis plus de 20 ans dans nombreux laboratoires pour l'activation de réactions en synthèse chimique. Il existe un débat sur l'existence d'effets MO sur la vitesse des réactions puisque des augmentations des vitesses des réactions ont été parfois observées. Des caractéristiques du chauffage MO peuvent être intéressantes pour l'intensification des procédés mais cette technique a rarement fait l'objet de productions à grande échelle. Peu de méthodes d'extrapolation de ces applications ont été proposées et celles qui existent sont purement empiriques. Nous avons voulu établir une méthodologie d'extrapolation de réacteurs MO en nous servant d'observations expérimentales fiables. La méthodologie des travaux a d'abord consisté au choix d'une réaction cible pour nos études : l'estérification de l'acide acétique avec le pentanol catalysée par une résine acide d'échange ionique. Nous avons conçu et construit un réacteur agité avec application monomode des MO. Des études dans le pilote ont montré le comportement hydrodynamique parfaitement agité du réacteur, la reproductibilité et la haute efficacité du chauffage. De plus, aucun gradient de température dans le réacteur n'a été mis en évidence. Des tests chimiques sous MO en réacteur fermé et en continu, ainsi qu'en milieu peu polaire ont montré que l'application des MO n'a pas eu d'influence sur la cinétique de la réaction cible. Finalement, nous avons développé une procédure d'extrapolation de réacteurs micro-ondes de type cuve agitée, qui est basée sur l'absence d'effet MO sur la chimie et sur le contrôle des zones chaudes dans le réacteur

Les microARNs sont de petits ARNs qui participent à la régulation de l'expression génique dans les organismes vivants. Depuis l'apparition des nouvelles techniques de séquençage à haut débit, leur identification et caractérisation dans les espèces animales et végétales font partie des défis actuels de la bioinformatique. Dans ce mémoire nous abordons ce problème en recherchant les microARNs exprimés chez le blé sous une dizaine de conditions environnementales différentes. Contrairement aux études antérieures similaires dans d'autres organismes, des difficultés additionnelles se sont ajoutées, associées à la partialité du génome du blé disponible, la multitude de conditions expérimentales et le peu de microARNs connus retrouvés par prédiction. Une nouvelle approche, employant une double validation par deux algorithmes de prédiction, a permis d'identifier plus de 3862 microARNs potentiels chez le blé ainsi que leurs gènes cibles. Parmi eux, 206 sont différentiellement exprimés entre les conditions expérimentales. Ces microARNs ont été répartis en 1222 familles en fonction des paramètres de leur similarité intra et intergroupe déduite des microARNs connus de mirBase. Pour minimiser le nombre de faux positifs, nous avons développé une méthode d'apprentissage machine (miRdup) pour valider la position des microARNs séquencés sur son pré-microARN en estimant plusieurs caractéristiques associées à ces petits ARNs. Ce dernier nous a permis d'établir 1016 microARNs avec un haut score de prédiction (502 familles). Pour chacun des microARNs prédits, en exploitant les données de leurs niveaux d'expression et l'identification des gènes ciblés, nous avons décrypté leurs rôles dans les différentes conditions expérimentales imposées au blé. Plusieurs microARNs sont en cours de validation expérimentale. Par ailleurs, pour faciliter l'implémentation de cette plateforme, nous avons intégré le pipeline de recherche conçu au cours de ce mémoire dans le logiciel Armadillo. Ce dernier permettra à l'avenir de faciliter la reproduction d'une telle étude chez d'autres plantes. ______________________________________________________________________________ MOTS-CLÉS DE L’AUTEUR : microARNS, prédiction, séquençage haut débit, blé, apprentissage machine.

Uno de los mayores retos que debe ser afrontado por los diseñadores de micro-procesadores es el de mitigar la gran latencia de las instrucciones de carga de datos en registros. Esta tesis analiza una de las posibles alternativas para atacar dicho problema: predicción de direcciones y ejecución especulativa.Varios autores han comprobado que las direcciones efectivas calculadas por las instrucciones de carga son bastante predecibles. En primer lugar, hemos analizado a qué es debida dicha predictabilidad. Este estudio intenta establecer las estructuras típicas presentes en lenguajes de alto nivel que, al ser compiladas, generas instruciones de carga predecibles. También se analizan los predictores convencionales con el objetivo de determinar qué predictores son más adecuados para las típicas aplicaciones.El estudio continúa con la propuesta de nuevos predictores de direcciones que utilizan sus recursos de almacenamiento de forma más eficiente que los previos predictores. Los predictores alamacenan información respecto al comportamiento de las instrucciones de carga; sin embargo, los requisitos de las instrucciones predecibles son diferentes de los de las instrucciones no predecibles. Consecuentemente, se propone una organización de las tablas de predicción que considere la existencia de ambos tipos de instruciones. También se muestra que existe un cierto grado de redundnacia en las tablas de predicción de los predictores. Este estudio propoen organizar las tablas de predicción de forma que se reduzca dicha redundancia. Todas estas propuestas permiten reducir los requisitos de los predictores referentes a espacio de alamacenamiento, sin causar menoscabo en el rendimiento de los predictores.Posteriormente, se evalúa el impacto de la predicción de direcciones en el rendimiento de los processadores. Las evaluaciones asumen que las predicciones se utilizan para iniciar de forma especulativa accessos a memoria y para ejecutar de forma especulativa sus instrucciones dependientes. En caso de una predicción correcta, todo el trabajo realizado de forma especulativa puede considerarse como correcto; en caso de error de predicción, el tranajo realizado especulativamente debe ser descartado. El estudio se centra en diversos aspectos como la interacción entre predicción de direcciones y predicción de saltos, la implementación de mecanismods de verification, los mecanismos re recuperación en casos de errores de predicción y la influencia de varios parámetreos del procesador (el tamaño de la ventana de emisión de instrucciones, la latencia de la memora cache, y la anchura de emisión de instrucciones) en le impacto de la predicción de direcciones en el rendimiento de los procesadores.Finalmente, se han evaluado mechanismos de recuperación para el caso de errores de predicción de latencia. La predicción de latencia es una técnica de ejecución especulativa utilizada por los planificadores de alguncos procesadores superescalares para tratar las instrucciones de latencia variable (por ejemplo, las instrucciones de carga). Nuestras evaluaciones se centran en un mecanismo convencional de recuperación para errores de predicción de latencia y en una nueva propuesta. También se evalúan los mecanismos propuestos en el ámbito de predicción de direcciones. Se concluye con que éstos mecanismos representan una alternativa rentable a los mecanismos de recuperación convencionales utilizados para tratar los errores de predicción de direcciones. / Mitigating the effect of the large latency of load instructions is one of challenges of micro-processor designers. This thesis analyses one of the alternatives for tackling this problem: address prediction and speculative execution.Several authors have noticed that the effective addresses computed by the load instructions are quite predictable. First of all, we study why this predictability appears; our study tries to detect the high-level language structures that are compiled into predictable load instructions. We also analyse the conventional address predictors in order to determine which address predictors are most appropriate for the typical applications.Our study continues by proposing address predictors that use their storage structures more efficiently. Address predictors track history information of the load instructions; however, the requirements of the predictable instructions are different from the requirements of the unpredictable instructions. We then propose an organization of the prediction tables considering the existence of both kinds of instructions. We also show that there is a certain degree of redundancy in the prediction tables of the address predictors. We propose organizing the prediction tables in order to reduce this redundancy. These proposals allow us to reduce the area cost of the address predictors without impacting their performance.After that, we evaluate the impact of address prediction on processor performance. Our evaluations assume that address prediction is used to start speculatively some memory accesses and to execute speculatively their dependent instructions. On a correct prediction, all the speculative work is considered as correct; on a misprediction, the speculative work must be discarded. Our study is focused on several aspects such as the interaction of address prediction and branch prediction, the implementation of verification mechanisms, the recovery mechanism on address mispredictions, and the influence of several processor parameters (the issue-queue size, the cache latency and the issue width) on the performance impact of address prediction. Finally, we evaluate several recovery mechanisms for latency mispredictions. Latency prediction is a speculative technique used by the schedulers of some superscalar processors to deal with variable-latency instructions (for instance, load instructions). Our evaluations are focused on a conventional recovery mechanism for latency mispredictions and a new proposal. We also evaluate the proposed recovery mechanism in the scope of address prediction; we conclude that it represents a cost-effective alternative to the conventional recovery mechanisms used for address mispredictions.

Energy harvesting from the surrounding environment has become a hot topic in research as an alternative powering solution. The concept deals with scavenging, as well as, harvesting energy from the surrounding energy sources. Harvesting vibrations, through Micro-Power Generators (MPGs) , has drawn a lot of attention recently due to the reduction in the power requirement of the current sensors and integrated ciruits, and the abundance of ambient vibrations in many environments. Vibration Micro-Power generators (VMPGs) use one of three transduction mechanisms: piezoelectric, electromagnetic or electrostatic. Although electrostatic MPGs are the most compatible mechanism with ICs technology, many challenges face their optimal operation including low efficiency due to power electronics switching losses, the need for pre-charge, and the inability to operate in vibration environments with low frequencies and amplitudes. The objective of this thesis is to develop novel electrostatic micro-power generators using switchless architecture to achieve low cost, small footprint, self-sustained and optimal power generation in different vibration environments including low frequencies and amplitudes. The first electrostatic MPG uses an out-of-plane capacitive transducer. The new generator is sensitive enough to extract output power at very low base excitations. It is designed to use ready-made electret as a charging source and is therefore portable and self-sustained. Moreover, the new MPG can be configured as a wideband MPG in its impact mode of operation. A bandwidth of up to 9 Hz has been realized in this mode of operation. An improved version of the MPG is also presented that produces almost 1mW output power at a base excitation amplitude and frequency of 0.08g (RMS) and 86 Hz. Two nonlinear models developed for the free-flight and impact modes of operation of the MPG are presented to allow future analysis and optimization of the system. The second electrostatic MPG uses a novel interdigitated in-plane parallel plate electrostatic transducer. The new implementation can achieve 78% more output power than the original cited implementation. The MPG is fabricated using MEMS surface micromachining. The MPG introduces a new beam suspension system in which the source voltage is unlimited by the pull-in instability and low MPG center frequency can be realized. The MPG uses charged silicon nitride as a charging source. The MPG produces 65 mV at a base acceleration amplitude and frequency of 2g and 1.1 kHz. The prototype achieves 27% less resonance frequency with only one eight the size of the previous implementation. A third electrostatic MPG architecture is introduced. The new architecture eliminates the need for restoring force elements (springs) in the MPG. The architecture can realize arbitrarily low MPG center frequency. It is suitable for both rectilinear and cylindrical structures and can be used with different vibration energy transduction methods. A prototype is fabricated and tested to demonstrate the feasibility of this architecture. The center frequency of the prototype is found to be 2 Hz demonstrating low frequency operation. The nonlinear behavior of switchless (continuous) electrostatic MPGs is further studied for optimal power operation. A consistent approximate analytical solution is developed to describe the nonlinear behavior of switchless comb-finger electrostatic MPGs. The method of multiple scales is used to develop such model. The model was found to be valid for MPGs operating under tight electromechanical coupling conditions and for moderately-large base excitations.

Advancements in electronic performance lead to a decrease in device size and an increase in power density. Because of these changes, current cooling mechanisms for electronic devices are beginning to be ineffective. Microchannels, with their large heat transfer surface area to volume ratio, cooled with either gas or liquid coolant, have shown some potential in adequately maintaining a safe surface temperature. By modifying the walls of the microchannel with fins, the cooling performance can be improved. Using computational fluid dynamics software, microfins placed in a staggered array on the bottom surface of a rectangular minichannel are modeled in order to optimize microstructure geometry and maximize heat transfer dissipation through convection from a heated surface. Fin geometry, dimensions, spacing, height, and material are analyzed. Correlations describing the Nusselt number and the Darcy friction factor are obtained and compared to recent studies. These correlations only apply to short fins in the laminar regime. Triangular fins with larger fin height, smaller fin width, and spacing double the fin width maximizes the number of fins in each row and yields better thermal performance. Once the effects of microfins were found, an experiment with multi-walled carbon nanotubes (MWNTs) grown on the surface were tested using both water and Al2O3/H2O nanofluid as the working medium. Minichannel devices containing two different MWNT structures – one fully coated surface of MWNTs and the other with a circular staggered fin array of MWNTs - were tested and compared to a minichannel device with no MWNTs. It was observed that the sedimentation of Al2O3 nanoparticles on a channel surface with no MWNTs increases the surface roughness and the thermal performance. Finally, using the lattice Boltzmann method, a two dimensional channel with suspended particles is modeled in order to get an accurate characterization of the fluid/particle motion in nanofluid. Using the analysis based on an ideal fin, approximate results for nanofluids with increase surface roughness was obtained. Microchannels have proven to be effective cooling systems and understanding how to achieve the maximum performance is vital for the innovation of electronics. Implementation of these modified channel devices can allow for longer lasting electronic systems.

No description available.

Energy harvesting from the surrounding environment has become a hot topic in research as an alternative powering solution. The concept deals with scavenging, as well as, harvesting energy from the surrounding energy sources. Harvesting vibrations, through Micro-Power Generators (MPGs) , has drawn a lot of attention recently due to the reduction in the power requirement of the current sensors and integrated ciruits, and the abundance of ambient vibrations in many environments. Vibration Micro-Power generators (VMPGs) use one of three transduction mechanisms: piezoelectric, electromagnetic or electrostatic. Although electrostatic MPGs are the most compatible mechanism with ICs technology, many challenges face their optimal operation including low efficiency due to power electronics switching losses, the need for pre-charge, and the inability to operate in vibration environments with low frequencies and amplitudes. The objective of this thesis is to develop novel electrostatic micro-power generators using switchless architecture to achieve low cost, small footprint, self-sustained and optimal power generation in different vibration environments including low frequencies and amplitudes. The first electrostatic MPG uses an out-of-plane capacitive transducer. The new generator is sensitive enough to extract output power at very low base excitations. It is designed to use ready-made electret as a charging source and is therefore portable and self-sustained. Moreover, the new MPG can be configured as a wideband MPG in its impact mode of operation. A bandwidth of up to 9 Hz has been realized in this mode of operation. An improved version of the MPG is also presented that produces almost 1mW output power at a base excitation amplitude and frequency of 0.08g (RMS) and 86 Hz. Two nonlinear models developed for the free-flight and impact modes of operation of the MPG are presented to allow future analysis and optimization of the system. The second electrostatic MPG uses a novel interdigitated in-plane parallel plate electrostatic transducer. The new implementation can achieve 78% more output power than the original cited implementation. The MPG is fabricated using MEMS surface micromachining. The MPG introduces a new beam suspension system in which the source voltage is unlimited by the pull-in instability and low MPG center frequency can be realized. The MPG uses charged silicon nitride as a charging source. The MPG produces 65 mV at a base acceleration amplitude and frequency of 2g and 1.1 kHz. The prototype achieves 27% less resonance frequency with only one eight the size of the previous implementation. A third electrostatic MPG architecture is introduced. The new architecture eliminates the need for restoring force elements (springs) in the MPG. The architecture can realize arbitrarily low MPG center frequency. It is suitable for both rectilinear and cylindrical structures and can be used with different vibration energy transduction methods. A prototype is fabricated and tested to demonstrate the feasibility of this architecture. The center frequency of the prototype is found to be 2 Hz demonstrating low frequency operation. The nonlinear behavior of switchless (continuous) electrostatic MPGs is further studied for optimal power operation. A consistent approximate analytical solution is developed to describe the nonlinear behavior of switchless comb-finger electrostatic MPGs. The method of multiple scales is used to develop such model. The model was found to be valid for MPGs operating under tight electromechanical coupling conditions and for moderately-large base excitations.

The understanding of the deformation mechanics within porous structures is an important field of study as these materials exist in nature as well as can be manufactured industrially influencing our lives daily. The motivation of the research contained within this manuscript was inspired by the desire to understand the mechanics of an elastomeric closed–cell porous material. This type of porous material is often used in load–bearing applications such as sport helmet liners and packing material which can be subjected to large deformations at high rates. Additionally, short term transient effects were explored. In order to investigate the deformation mechanics of a closed cell elastomeric foam, a polychloroprene (neoprene) material was chosen as it was available in both rubber form and a foam with relatively consistent cell size. Compression tests were conducted on the polychloroprene rubber at strain rates ranging from 0.001/s to 2700/s which identified that it had a hyper–viscoelastic behaviour with a significant strain rate dependence. A newly developed constitutive model was created to capture the response of the polychloroprene rubber. A coupled finite element model of the polychloroprene foam was created and compared to experimental tests for validation. The model slightly over predicted the stress level response of the experimental tests. The model was used to identify momentum dissipation mechanisms that contributed to the low wave speed measurement of approximately 70 m/s determined from the model. The investigation of wave transit times through use of the model was key to interpreting experimental data. Of the morphological factors investigated, it was determined that wall thickness had the most significant impact on the stress response of the foam. The pore–scale model was useful for visualizing wavepropagation effects and deformation mechanics which was not feasible experimentally.

Effects of biomimetic micro-patterning of polymeric materials on their interfacial properties were studied experimentally. Micropillars of PDMS and SU-8 epoxy were fabricated through soft lithography and UV lithography techniques, respectively. PDMS pillars were topped by thin terminal films of the same material through dipping method with different thicknesses and viscosities. Adhesion and frictional properties of biomimetic microstructures were examined in two modes of contact, i.e. laid and conformal contact. In the first mode of contact, i.e. laid contact, the contact between adhesive and adherent is laid on top of the micro-protrusions or is in contact with side wall of micropillars. Adhesion properties of the smooth and patterned PDMS were characterized through micro-indentation test. Moreover, the friction properties of the smooth PDMS sample and PDMS micropillars with different aspect ratios were examined in unidirectional friction testing. JKR theory of continuum contact mechanics was utilized to interpret the obtained data. To study the effect of second mode of contact, peeling behaviour of a conformal contact between solidified liquid PDMS and SU-8 micropillars was monitored. Kendall’s model of elastic peeling was used to interpret the peeling data. It was found that patterning of the materials would decrease the real area of contact and accordingly adhesion and friction to the mating surface. Termination of the micropillars with a thin layer of the same material result in increment of adhesion as reduction of the real contact area could be compensated and the compliance of the near surface increases. Elastic energy dissipation as a result of enhanced compliance and crack trapping and crack propagation instabilities are the main reasons behind increment of adhesion of thin film terminated structures. Viscoelasticity of the terminal thin film remarkably increased the adhesion as a result of coupling mentioned mechanisms and viscoelastic loss on the surface. Decline of the overall friction could be tailored through use of different aspect ratios. Higher aspect ratios pillars show higher friction comparing to lower aspect ratio pillars. 550 folds enhancement of adhesion was observed for peeling of the PDMS tape from rigid micropillars with aspect ratio ranging from 0 to 6. It is concluded that for the lower aspect ratio micropillars, the elastic energy dissipation is playing the key role in adhesion enhancement. This role shifts toward side-wall friction during separation by increase in aspect ratio. These all give in hand a versatile tool to control and fine tune the interfacial properties of materials, whether they are concerned with adhesion or friction.