Probabilistic graphical models like Bayesian Networks (BNs) are powerful artificial-intelligence formalisms, with similarities to cognition and higher order reasoning in the human brain. These models have been, to great success, applied to several challenging real-world applications. Use of these formalisms to a greater set of applications is impeded by the limitations of the currently used software-based implementations. New emerging-technology based circuit paradigms which leverage physical equivalence, i.e., operating directly on probabilities vs. introducing layers of abstraction, promise orders of magnitude increase in performance and efficiency of BN implementations, enabling networks with millions of random variables. While majority of applications with small network size (100s of nodes) require only single digit precision for accurate results, applications with larger size (1000s to millions of nodes) require higher precision computation. We introduce a new BN integrated circuit fabric based on mixed-signal magneto-electric circuits which perform probabilistic computations based on the principle of approximate computation. Precision scaling in this fabric is logarithmic in area vs. linear in prior directions. Results show 33x area benefit for a 0.001 precision compared to prior direction, while maintaining three orders of magnitude performance benefits vs. 100-core processor implementations.

The first thermal cycling experiments of ionic self-assembled multilayer (ISAM) films have been reported examining their survivability through repeated thermal cycles from -20° C to 120° C in ambient atmospheric conditions. The films were constructed from alternating layers of Nile Blue A and gold nanoparticles which provided a strong absorbance in the optical wavelength range. No degradation of the optical characteristics of the ISAM films was observed [1]. Techniques for measuring the capacitance and resistivity of various ISAM films have also been developed allowing for a more complete electrical characterization of ISAM films. Capacitance measurements enabled a calculation of the dielectric function and breakdown field strength of the ISAM films. The capacitance measurement technique was verified by measuring the dielectric function of a spin-coated thin film PMMA, which has a well characterized dielectric function [2]. Surface-enhanced Raman spectroscopy (SERS) has been studied as a possible detection method for malignant melanoma revealing spectral differences in blood sera from healthy horses and horses with malignant melanoma. A SERS microscope system was constructed with the capability of resolving the Raman signal from biologically important molecules such as beta-carotene and blood sera. The resulting Raman signals from sera collected from horses with malignant melanoma were found to have additional peaks not found in the Raman signals obtained from sera collected from healthy horses. A systematic analysis of the combination of absorbance and fluorescence signals of blood sera collected from populations of healthy dogs and dogs with cancer has resulted in a rapid and cost-effective method for monitoring protein concentrations that could possibly be used as part of a cancer screening process. This method was developed using the absorbance and fluorescence signals from known serum proteins, the combinations of which were used to match the absorbance and fluorescence signals of blood sera allowing for an accurate determination of protein concentrations in blood sera [3]. Finally, a novel method for measuring the melting point of DNA in solution using capacitance measurements is presented. This method allows for the determination of the melting temperature as well as the melting entropy and melting enthalpy of DNA strands. Two different short strands of DNA, 5'-CAAAATAGACGCTTACGCAACGAAAAC-3' along with its complement and 5'-GGAAGAGACGGAGGA-3' along with its complement were used to validate the technique as the characteristics of these strands could be modeled using theoretical methods. This experimental technique allows for the precise determination of the melting characteristics of DNA strands and can be used to evaluate the usefulness of theoretical models in calculating the melting point for particular strands of DNA. Additionally, a micro-fluidic device has been proposed that will allow for a rapid and cost-effective determination of the melting characteristics of DNA [4]. / Ph. D.

This research has developed a lead-free semiconductor device interconnect technology by studying the processing-microstructure-property relationships of low-temperature sintering of nanoscale silver pastes. The nanoscale silver pastes have been formulated by adding organic components (dispersant, binder and thinner) into nano-silver particles. The selected organic components have the nano-particle polymeric stabilization, paste processing quality adjustment, and non-densifying diffusion retarding functions and thus help the pastes sinter to ~80% bulk density at temperatures no more than 300°C. It has been found that the low-temperature sintered silver has better electrical, thermal and overall thermomechanical properties compared with the existing semiconductor device interconnecting materials such as solder alloys and conductive epoxies. After solving the organic burnout problems associated with the covered sintering, a lead-free semiconductor device interconnect technology has been designed to be compatible with the existing surface-mounting techniques with potentially low-cost. It has been found that the low-temperature sintered silver joints have high electrical, thermal, and mechanical performance. The reliability of the silver joints has also been studied by the 50-250°C thermal cycling experiment. Finally, the bonging strength drop of the silver joints has been suggested to be ductile fracture in the silver joints as micro-voids nucleated at microscale grain boundaries during the temperature cycling. The low-temperature silver sintering technology has enabled some benchmark packaging concepts and substantial advantages in future applications. / Ph. D.

This study focused on the development and evaluation of die-attach material and substrate technology for high-temperature applications. For the die-attach material, a low-temperature sintering technique enabled by a nanoscale silver paste was developed for attaching large-area (>100 mm2) semiconductor chips. The nanoscale silver paste can be sintered at a much lower temperature (<300 oC) than in the conventional sintering process (>800 oC), and at the same time reached about 80 vol% bulk density. Analyses of the sintered joints by scanning acoustic imaging and electron microscopy showed that the attachment layer had a uniform microstructure with micron-sized porosity with the potential for high reliability under high temperature applications. We also investigated the effects of a large temperature cycling range on the reliability of direct bonded aluminum (DBA) substrate. DBA substrates with different metallization were thermally cycled between -55 oC and 250 oC. Unlike with the DBC substrate, no delamination of aluminum from the aluminum nitride ceramic base-plate was observed for the DBA substrates. However, aluminum surface became roughened during the thermal cycling test. It was believed that in the high-temperature regime, the significant amount of thermomechanical stress and grain-scale deformation would cause recrystallization and grain-boundary sliding in the aluminum layer, which would further lead to the observed increase in surface roughness. The influence of metallization over the aluminum surface on the extent of surface roughness was also characterized. In addition to evaluating the reliability of nanoscale silver paste and DBA substrate individually, this work also conducted experiments that characterize the compatibility of nanoscale silver paste on DBA substrate in terms of reliability in a high-temperature environment. In the large-area attachment, the sintered silver was found to be very compliant with the deformed aluminum. The device-to-silver and silver-to-substrate interfaces remain intact after up to 800 cycles. No large scale delamination and horizontal cracks were observed. However, some vertical crack lines began to show after certain number of cycles. It was believed that these vertical cracks were caused by the thermomechanical stresses in the sintered silver layer. In addition, with regard to the thermal performance, since most of the heat was generated from the semiconductor devices and were transferred vertically through the die-attach material to substrate, these vertical cracks were also considered more advantageous than horizontal cracks. / Ph. D.

<p dir="ltr">The technological drive of humanity to explore the cosmos, travel at hypersonic speeds, and pursue clean energy solutions requires ceramic scientists and engineers to constantly push materials to their functional, behavioral, and chemical extremes. Ultra-high temperature ceramics, and particularly transition metal carbides, are promising materials to meet the demands of extreme environment materials with their >4000 °C melting temperature and impressive thermomechanical behaviors in extreme conditions. The advent of the 2D version of these transition metal carbides, known as MXenes, added a new direction to design transition metal carbides for energy, catalysis, flexible electronics, and other applications. Toward extreme conditions, although MXenes remain yet unexplored, we believe that the ~1 nm flakes of MXenes gives ceramics scientists and engineers the ability to truly engineer transition metal carbides layer-by-layer at the nanoscale to endure the extreme conditions required by future harsh environment technology. Although MXenes have this inherent promise, fundamental study of their behavior in high-temperature environments is necessary to understand how their chemistry and 2D nature affects the high-temperature stability and phase behavior of MXenes toward application in extreme environments.</p><p dir="ltr">In this dissertation, we investigate the high-temperature phase behavior of 2D MXenes in high temperature inert environments to understand the stability and phase transition behavior of MXenes. In this work, we demonstrate that 1) MXenes’ transition at high-temperatures is to highly textured transition metal carbides is due to the homoepitaxial growth of these phases onto ~1-nm-thick MXenes’ highly exposed basal plane, 2) the MXene to MXene interface plays a major role in the phase behavior of MXenes, particularly toward building layered transition metal carbides using MXenes as ~1-nm-thick building blocks, and 3) Defects are the primary site at which atomic migration begins during phase transition of MXenes into these highly textured transition metal carbides, and these defects can be engineered for different phase stability of MXenes. To do so, we investigate the phase behavior of Ti₃C₂T_<em>x</em>, Ta₄C₃T_<em>x</em>, Mo₂TiC₂T_<em>x</em>, and other MXenes using a combination of <i>in situ</i> x-ray diffraction and scanning transmission electron microscopy and other <i>ex situ</i> methods, such as secondary ion mass spectrometry and x-ray photoelectron spectroscopy, with other methods. By investigating the fundamentals of the high-temperature phase behavior of MXenes, we hope to establish the basic principles behind use of MXenes as the ideal material for application in future extreme environments.</p>

Les récents progrès en matière de nanotechnologies ont permis d'élaborer de nouveaux matériaux aux propriétés physico-chimiques uniques tels que les nanoparticules de fer zéro valent (NPFe0). Ces nanoparticules ont prouvé leur efficacité pour dégrader les composés organiques chlorés comme le trichloroéthylène (TCE), cependant leur transport dans les milieux poreux est souvent limité. Une solution pour pallier à ce problème est de modifier leur surface par adsorption de molécules organiques. Toutefois, cet enrobage modifie la réactivité des NPFe0 vis-à-vis du TCE, ce qui peut potentiellement affecter l'efficacité du traitement. Ainsi, le premier volet de cette thèse concerne l'étude de la réactivité de NPFe0 brutes ou modifiées par des polyélectrolytes anioniques vis-à-vis du TCE. Cette réactivité a été évaluée au travers des cinétiques de dégradation du TCE et de ses produits de transformation.D'autre part, les eaux souterraines contiennent souvent des espèces dissoutes réductibles pouvant réagir avec Fe0. Ces dernières peuvent affecter la réactivité des NPFe0 vis-à-vis du polluant ciblé et donc l'efficacité du traitement de dépollution. Le deuxième volet de cette thèse porte sur l'effet de deux accepteurs d'électrons (CrVI et NO3-) sur la réactivité des NPFe0 brutes et modifiées. Enfin, le troisième volet de cette thèse concerne l'évaluation de la réactivité des NPFe0 vis-à-vis du TCE dans un cas « réel », afin de valider la technique de dépollution. Une étude pilote et une application in situ ont ainsi été menées sur un site industriel dont l'eau souterraine est contaminée par le TCE (polluant ciblé) avec de fortes teneurs en CrVI et NO3-. / Recent advances in nanotechnology have led to the development of new materials with unique physicochemical properties such as nanoscale zero valent iron particles (nZVI). These nanoparticles proved their efficiency to degrade chlorinated organic compounds such as trichlorethylene (TCE), but their migration in porous media is often limited. To overcome this problem, a solution is to modify their surface by adsorption of organic molecules. However, this coating modifies the reactivity towards TCE, which can potentially affect the treatment efficiency.Thus, the first part of this PhD focuses on the reactivity of nZVI (bare or modified by anionic polyelectrolytes) towards TCE. This reactivity was evaluated through the TCE degradation kinetics rates and its transformation products.In addition, groundwaters often contain reducible species that can react with dissolved Fe0. These last species may affect the reactivity of nZVI towards the target pollutant and therefore the remediation efficiency. The second part of this PhD focuses on the effect of two electron acceptors (CrVI and NO3-) on the reactivity of bare and modified nZVI.Finally, the third part of this PhD presents the assessment of the reactivity of nZVI towards TCE in a "real" case, in order to validate the remediation process. A pilot study and in situ application have been carried out on an industrial site which groundwater is contaminated with TCE (targeted pollutant) with high levels of CrVI and NO3-.

Les vésicules extracellulaires nanométriques (nEVs) issues de cellules souches mésenchymateuses (CSMs) et les nanovésicules synthétiques sont au centre de nombreuses recherches pour le développement de nouvelles stratégies thérapeutiques en médecine régénérative. La mise en place d’une méthode standardisée pour isoler les nEVs à partir de milieu conditionné de CSMs et de pouvoir les caractériser a été nécessaire. Nous nous sommes concentrés sur leur taille qui se situe entre 30 et 150 nm ainsi que la présence de certains de leur marqueurs membranaires (CD9, CD63 et CD81). Durant ce travail, deux méthodes d’isolement ont été testées. Les résultats obtenus par les analyses physiques (Nanosight®, microscopie électronique à transmission) et biologiques (cytométrie en flux) des différents échantillons ont permis de standardiser la méthode d’isolement des nEVs par centrifugations et ultracentrifugations successives. Ensuite, nous nous sommes intéressés à l’utilisation de ces nEVs sécrétées par les CSMs en culture cellulaire. Il a été mis en évidence que des interactions existent entre ces nEVs et des cellules endothéliales (CEs) in vitro. Ces interactions vont entraîner des modifications dans le comportement cellulaire des CEs en augmentant leur potentiel de formation de réseaux vasculaires. En parallèle de ces travaux sur les nEVs, une étude a été réalisée sur l’utilisation de nanovésicules synthétiques, des nanoliposomes (NLPs), élaborées à partir de lécithine d’agro-ressource (saumon) comme transporteur de TGF-ß1 pour une application en médecine régénérative. Après leur caractérisation physico-chimique, cette étude préliminaire a montré que ces NLPs ne présentent pas de cytotoxicité pour les CSMs in vitro. Il existe un potentiel important d’utilisation des nEVs de CSMs ainsi des NLPs pour développer de nouvelles stratégies innovantes en thérapie « cell-free » dans le domaine de la médecine régénérative / Nanoscale extracellular vesicles (nEVs) derived from mesenchymal stem cells (MSCs) and synthetic nanovesicles are at the centre of many research studies for the development of new therapeutic strategies in regenerative medicine. A standardized method was used to isolate nEVs from conditioned media of CSMs and to characterize them. We focused on their size with a range of 30 to 150 nm and the presence of some of their membrane markers (CD9, CD63 and CD81). During this work, two isolation methods were tested. The results obtained by the physical (Nanosight®, transmission electron microscopy) and biological (flow cytometry) analyses of the different samples allowed to standardize the method of isolation of the nEVs by successive centrifugation and ultracentrifugation. Then, we studied the use of these nEVs derived from MSCs in cell culture. Interactions between these nEVs and endothelial cells (ECs) have been demonstrated in vitro. These interactions lead to changes in the cellular behaviour of ECs by increasing their potential to form vascular networks. In parallel of this work on nEVs, we studied the use of synthetic nanovesicles, called nanoliposomes (NLPs) prepared from agro-resource derived lecithin (salmon) as TGF-β1 transporters for applications in regenerative medicine. After their physicochemical characterization, this preliminary study showed that these NLPs do not exhibit cytotoxicity for MSCs in vitro. There is an important potential for the use of nEVs derived from MSCs as well as NLPs to develop new cell-free therapy innovative strategies in the field of regenerative medicine

Made available in DSpace on 2017-07-21T20:42:41Z (GMT). No. of bitstreams: 1 Graciela Aparecida Santos Silva.pdf: 4725206 bytes, checksum: 1b5275fc3c55802f9ed0465ffbc48d75 (MD5) Previous issue date: 2013-02-22 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The purpose of this work was the synthesis of nanoscaled BaO-Al2O3-MnO2 system and its characterization. The nanoparticulate system was prepared by mixing salts of Ba, Al and Mn in a chamber of low temperature, lyophilizing the solutions of their salts, and followed by calcination in the temperature range from 230 °C to 1190° C.This synthesis was compared with two other systems, prepared using a chemical mixture of oxides, with a particle size of conventional size and also with the phase of nanoscale oxides, followed by calcination in the same temperature conditions. The colorimetric properties of ceramic powders were evaluated by UV-visible spectroscopy and coordinates CIE-L a b and characterized by X-ray diffraction, Xray fluorescence, simultaneous thermal analyisis, scanning electron microscopy and laser scattering technique. This nanoscaled synthesis provides a possible application in the production on ceramic pigments because the resulting powder showed different color in comparison to the other systems. The nanoparticulate system showed the advantage of needing lower calcination temperatures, and, at 1135ºC,showed a different color that was not presented by powders synthetized by other methods. / A proposta deste trabalho consiste na síntese em nanoescala do sistema BaO-Al2O3-MnO2, bem como sua caracterização. O sistema nanoparticulado foi preparado misturando sais de Ba, Al e Mn em câmara de baixa temperatura, liofilizando-se assim as soluções dos respectivos sais, seguidos de calcinação na faixa de temperatura de 230 °C a 1190 °C. Comparou-se esta síntese com duas metodologias, preparadas através de uma mistura química de óxidos, com tamanho de partícula convencional e também com tamanho de fase dos óxidos em nanoescala, seguidas de calcinação nas mesmas condições de temperatura. Os pós cerâmicos foram avaliadas por espectroscopia no UV-visível e coordenadas CIE-L a b e caracterizado por difração raios X, fluorescência de raios X, termogravimetria e análise termodiferencial simultânea, microscopia eletrônica de varredura e espectroscopia de fotocorrelação. Essa síntese por nanoescala traz a possível aplicação na produção de pigmentos cerâmicos, pois trouxe alterações colorimétricas diferentes comparando com os outros sistemas. Esse método teve vantagem, pois trouxe temperaturas de calcinação mais baixa, e em 1135 ºC, o sistema nanoparticulado apresentou uma coloração que não foi observado para os outros métodos de síntese.

No description available.

Les nouvelles techniques de fabrication nanométriques comme l’auto-assemblage ou la nanoimpression permettent de réaliser des matrices régulières (crossbars) atteignant des densités extrêmes (jusqu’à 1012 nanocomposants/cm2) tout en limitant leur coût de fabrication. Cependant, il est attendu que ces technologies s’accompagnent d’une augmentation significative du nombre de défauts et de dispersions de caractéristiques. La capacité à exploiter ces crossbars est alors conditionnée par le développement de nouvelles techniques de calcul capables de les spécialiser et de tolérer une grande densité de défauts. Dans ce contexte, l’approche neuromimétique qui permet tout à la fois de configurer les nanodispositifs et de tolérer leurs défauts et dispersions de caractéristiques apparaît spécialement pertinente. L’objectif de cette thèse est de démontrer l’efficacité d’une telle approche et de quantifier la fiabilité obtenue avec une architecture neuromimétique à base de crossbar de memristors, ou neurocrossbar (NC). Tout d’abord la thèse introduit des algorithmes permettant l’apprentissage de fonctions logiques sur un NC. Par la suite, la thèse caractérise la tolérance du modèle NC aux défauts et aux variations de caractéristiques des memristors. Des modèles analytiques probabilistes de prédiction de la convergence de NC ont été proposés et confrontés à des simulations Monte-Carlo. Ils prennent en compte l’impact de chaque type de défaut et de dispersion. Grâce à ces modèles analytiques il devient possible d’extrapoler cette étude à des circuits NC de très grande taille. Finalement, l’efficacité des méthodes proposées est expérimentalement démontrée à travers l’apprentissage de fonctions logiques par un NC composé de transistors à nanotube de carbone à commande optique (OG-CNTFET). / Novel manufacturing techniques, such as nanoscale self-assembly or nanoimprint, allow a cost-efficient way to fabricate high-density crossbar matrices (1012 nanodevices/cm2). However, it is expected that these technologies will be accompanied by a significant increase of defects and dispersion in device characteristics. Thus, programming these crossbars require new computational techniques that possess high tolerance for such variations. In this context, approaches based on neural networks are promising for configuring nanodevices, since they provide a natural way for tolerating low yields and device variations. The main objective of this thesis is to explore such a neural-network approach, by examining factors such as efficiency and reliability, using the memristor crossbar architecture or neurocrossbar (NC). We introduce algorithms for learning the logic functions on the NC, and the tolerance of NC against static defects (stuck-defect) and dispersion of device properties is discussed. Probabilistic analytical models for predicting the convergence of NC are proposed and compared with Monte Carlo simulations, which take into account the impact of each type of defect and dispersion. These analytical models can be extrapolated to study large-sized NCs. Finally, the effectiveness of the proposed methods is experimentally demonstrated through the learning of logic functions by a real NC made of Optically Gated Carbon Nanotube Field Effect Transistor (OG-CNTFET).