Avec la diminution perpétuelle des dimensions des circuits intégrés, la gravure de dispositifs à l’échelle nanométrique constitue un véritable défi pour les procédés plasma qui montrent des limites dans le cas d’empilement de couches minces notamment. L’anisotropie de gravure réside sur la formation de couches de passivations sur les flancs des motifs, qui agissent comme film protecteur bloquant la gravure latérale par les radicaux du plasma. Cependant, cette fine couche est responsable de l’apparition de pente dans les profils gravés et il est difficile de contrôler son épaisseur. De plus, une deuxième couche réactive est aussi formée en fond de motifs. Les produits de gravures sont formés au sein de cette couche permettant d’augmenter la vitesse de gravure du substrat. Il a récemment été admis que contrôler l’épaisseur de ces couches réactives constitue le paramètre clé pour obtenir une gravure à très haute sélectivité.Cependant, les couches réactives à analyser hautement réactives, en raison de leur forte concentration en halogènes, s’oxydent immédiatement lors de la remise à l’air.Cette étude se propose de développer une approche originale, simple et extrêmement puissante pour observer ces couches de passivation quasi in-situ (sans contact avec l’air ambiant) : Apres gravure, l’échantillon est transporté sous vide à l’intérieur d’une valise spécifique jusqu’à une enceinte de dépôt, où il est alors encapsulé par une couche métallique (PVD assisté par magnétron). L’échantillon ainsi encapsulé peut être observé ex-situ sans modification chimique grâce à des analyses STEM EDX. / As the size of integrated circuit continues to shrink, plasma processes are more and more challenged and show limitations to etch nanometer size features in complex stacks of thin layers. The achievement of anisotropic etching relies on the formation of passivation layers on the sidewalls of the etched features, which act like a protective film that prevents lateral etching by the plasma radicals. However, this layer also generate a slope in the etch profile and it’s difficult to control its thickness. Another thin layer called “reactive layer” is also formed at the bottom of the features. Etch products are formed in this layer allowing a high etch rate of the substrate. It starts to be realized that controlling the thickness of this reactive layer is the key to achieve very high selective processes.However, the layers to be analyzed are chemically highly reactive because they contain large concentrations of halogens and they get immediately modified (oxidized) when exposed to ambient atmosphere.In this work we develop an original, simple and extremely powerful approach to observe passivation layers quasi in-situ (i.e. without air exposure): After plasma etching, the wafer is transported under vacuum inside an adapted suitcase to a deposition chamber where it is encapsulated by a metallic layer (magnetron sputtering PVD). Then, the encapsulated features can be observed ex situ without chemical / thickness modification thanks to (S)TEM-EDX analysis.

Dans le contexte du stockage profond des déchets radioactifs, il est important d'identifier l’hystérésis de saturation de la roche hôte, l'argilite du Callovo-Oxfordien, et sa capacité de scellement (en particulier, la porosité et la distribution de taille des pores). Tout d'abord, six cycles différents d'humidité relative sont destinés à évaluer l’hystérésis de saturation, qui n'est pas observée dans les cycles de faible amplitude. D'autre part, une nouvelle méthode est proposée pour la mesure de la porosité, qui utilise l'injection d'un gaz pour évaluer le volume des pores. Par rapport à la porosité par adsorption d'eau, l’injection de gaz fournit des porosités supérieures d’environ 5%. L'injection de gaz est également utilisée pour quantifier les isothermes de sorption-désorption, qui sont sensiblement différentes de celles obtenues par la méthode gravimétrique, avec un volume poreux accessible au gaz plus élevé pour une humidité relative <43%. Enfin, par Microscopie Electronique à Balayage couplée à un Faisceau Ionique Focalisé (FIB/MEB), on reconstruit le réseau poreux 3D de l'argilite à partir de séries d'images 2D espacées de 10nm : la porosité et la distribution de taille des pores sont quantifiés jusqu’à 20nm, ainsi que l’orientation et l'anisotropie. Avec une résolution plus élevée (jusqu’à moins de 1nm), la Microscopie Electronique à Transmission (MET) montre une grande quantité de pores de l’ordre de quelques nm, situés entre les agrégats d'argile. / In the context of deep underground storage of radioactive nuclear waste, it is important to identify the saturation hysteresis of the host rock, i.e. of Callovo-Oxfordian (COx) claystone, and its porosity and pore size distribution. Firstly, six different cycles of relative humidity are applied for saturation hysteresis, which is not observed in the cycles with low magnitude. Secondly, a new method is proposed for measuring porosity, which uses injection of gas to evaluate the pore volume. In contrast to porosity given by water adsorption, the gas injection method provides larger porosity values of around 5%. The gas injection method is also used to quantify the sorption-desorption isotherms of COx claystone, which are significantly different from those obtained by the gravimetric method, with a bigger pore volume accessible to gas in relative humidities < 43%. Finally, by Focused Ion Beam/Scanning Electron Microscopy (FIB/SEM), we obtain 2D image stacks and 3D reconstructed pore volumes, by which porosity and pore size distribution are quantified down to 20nm, as well as pore orientation and anisotropy. At a higher resolution (below 1nm), Transmission Electron Microscopy (TEM) imaging reveals significant amounts of smaller pores (of a few nm) between clay aggregates.

Microtubules play a major role in intracellular trafficking of vesicles in endocrine cells. Detailed knowledge of microtubule organization and their relation to other cell constituents is crucial for understanding cell function. However, their role in insulin transport and secretion is currently under debate. Here, we use Fib-Sem to image islet beta cells in their entirety with unprecedented resolution. We reconstruct mitochondria, Golgi apparati, centrioles, insulin secretory granules and micro-tubules of seven beta cells, and generate a comprehensive spatial map of microtubule-organelle interactions. We find that micro-tubules form non-radial networks that are predominantly not connected to either centrioles or endomembranes. Microtubule number and length, but not microtubule polymer density, vary with glucose stimulation. Furthermore, insulin secretory granules are enriched near the plasma membrane where they associate with microtubules. In summary, we provide the first 3D reconstructions of complete microtubule networks in primary mammalian cells together with evidence regarding their importance for insulin secretory granule positioning and thus supportive role in insulin secretion.

No description available.

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第19730号 / 工博第4185号 / 新制||工||1645(附属図書館) / 32766 / 京都大学大学院工学研究科物質エネルギー化学専攻 / (主査)教授 江口 浩一, 教授 安部 武志, 教授 陰山 洋 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DGAM

The goal of this work is to further understand the relationships between porous electrode microstructure and mass transport properties. This understanding allows us to predict and improve cell performance from fundamental principles. The investigated battery systems are the widely used rechargeable Li-ion battery and the non-rechargeable alkaline battery. This work includes three main contributions in the battery field listed below. Direct Measurement of Effective Electronic Transport in Porous Li-ion Electrodes. An accurate assessment of the electronic conductivity of electrodes is necessary for understanding and optimizing battery performance. The bulk electronic conductivity of porous LiCoO2-based cathodes was measured as a function of porosity, pressure, carbon fraction, and the presence of an electrolyte. The measurements were performed by delamination of thin-film electrodes from their aluminum current collectors and by use of a four-line probe. Imaging and Correlating Microstructure To Conductivity. Transport properties of porous electrodes are strongly related to microstructure. An experimental 3D microstructure is needed not only for computation of direct transport properties, but also for a detailed electrode microstructure characterization. This work utilized X-ray tomography and focused ion beam (FIB)/scanning electron microscopy (SEM) to obtain the 3D structures of alkaline battery cathodes. FIB/SEM has the advantage of detecting carbon additives; thus, it was the main tomography tool employed. Additionally, protocols and techniques for acquiring, processing and segmenting series of FIB/SEM images were developed as part of this work. FIB/SEM images were also used to correlate electrodes' microstructure to their respective conductivities for both Li-ion and alkaline batteries. Electrode Microstructure Metrics and the 3D Stochastic Grid Model. A detailed characterization of microstructure was conducted in this work, including characterization of the volume fraction, nearest neighbor probability, domain size distribution, shape factor, and Fourier transform coefficient. These metrics are compared between 2D FIB/SEM, 3D FIB/SEM and X-ray structures. Among those metrics, the first three metrics are used as a basis for SG model parameterization. The 3D stochastic grid (SG) model is based on Monte Carlo techniques, in which a small set of fundamental inter-domain parameters are used to generate structures. This allows us to predict electrode microstructure and its effects on both electronic and ionic properties.

<p>  </p> <p>Transition to a sustainable power grid entails the use of all net-zero carbon emission technology that is currently available. Liquid metal-cooled fast nuclear reactors (LMFRs) are technologies capable of competitively providing power while attaining sustainability and reliability. Uranium-zirconium metallic alloys have been proposed as LMFRs fuels based on the performance of the fuel in experimental scale reactors, achieving up to 20 at.% burn-up. The following phenomena affects the irradiation performance of U-Zr fuels: constituent redistribution, swelling, fuel-cladding mechanical interaction (FCMI), and fuel-cladding chemical interaction (FCCI). Further understanding of these phenomena, and development of predictive models requires data collection of variables such as composition, morphology of the redistribution regions, porosity distribution, porosity morphology, fission gas release, and the relation between local composition and porosity evolution. </p> <p>To achieve this, focused ion beam-scanning electron microscopy (FIB-SEM) serial sectioning was applied to specimens from the different compositional regions developed during constituent redistribution of a U-10 wt.% Zr fuel neutron irradiated to 5.7 at.% burn-up. High-resolution backscattered electron (BSE) micrographs, and energy dispersive spectroscopy (EDS) spectra were obtained for several sections of each specimen. Each section was analyzed to identify the microstructural and compositional evolution in the specimen volume. Three-dimensional porosity and phase volume distribution was obtained using image processing and three-dimensional object classification. The study revealed local segregation of phases within each of the regions, porosity distribution dependency on temperature and local composition, preferential porosity nucleation sites, porosity evolution trends, interconnectivity, possible sinks/nucleation sites for porosity and precipitates, as well as possible mechanisms for fission gas release.</p>

<p> Monolithic uranium molybdenum alloys (U-Mo) are considered a candidate for converting high-performance research and test reactors from high-enriched uranium to low enrichment alternatives. The metallic fuel is capable of conversion due to the high U loading and favorable radiation performance. During irradiation, the fuel undergoes a three-part swelling behavior, with an initial linear swelling rate followed by an increase in the swelling rate represented by an increase in the nucleation of fission-gas bubbles, and ending with stabilization at the highest fission densities. Understanding the high burnup regime is critical to extending the life of the reactor and creating accurate fuel performance models. To accurately inform swelling models, it is necessary to experimentally characterize the pore evolution as a function of burnup and the influence of diffusion barrier-fuel interaction on the morphology. Therefore, a systematic approach was conducted to experimentally characterize the influence of irradiation and fuel-diffusion barrier interaction on the pore morphology and then empirically model the porosity evolution. Initially, three separate locations in a monolithic U-Mo fuel plate with burnups ranging from 8.9-9.4x1021 fissions/cm3 were investigated using scanning electron micrography (SEM) to characterize the morphological porosity dependence on fission density. To investigate the impact of the Zr-fuel interface on the pore morphology, two specimens were imaged using synchrotron microcomputed tomography (Sr-µCT) from a U-Mo monolithic miniplate irradiated to 9.8x1021 fissions/cm3, one at the diffusion barrier and one in the bulk fuel. Synchrotron microcomputed tomography allows for the characterization of the influence of fuel-Zr diffusion barrier interaction on the pore morphology in three dimensions; however, due to the novelty of this technique applied to nuclear fuels the results were verified with SEM serial sectioning. The multimodal comparison between the Sr-µCT and SEM serial sectioning allows for a direct assessment of the capabilities of each technique for nuclear fuel applications. Due to the complex microstructure and imaging challenges in analyzing these samples, several model-based image processing and characterization tools were developed to aid in the analysis. An empirical model for porosity evolution in high-burnup U-Mo was developed and accurately modeled the porosity behavior. The experimental results from the current work and the empirical model developed can be used to inform mechanistic modeling efforts in the community. </p>

Le carbure de silicium est pressenti comme un matériau prometteur dans plusieurs domaines de l’électroniquetels que la nano-électronique, l’électronique de puissance ou les capteurs travaillant en milieuxhostiles (hautes températures, milieux corrosifs, milieux biologiques) du fait de ses propriétés physicochimiquessupérieures à celles du silicium, notamment. Cependant, parmi les différentes méthodesd’élaboration par voie descendante ou ascendante permettant de fabriquer des nano-objets à 1D enSiC, aucune n’a pour l’instant permis d’obtenir du SiC d’excellente qualité cristalline.Le travail de cette thèse a porté sur la démonstration de l’élaboration de nanostructures 1D àbase de SiC, à savoir nanofils coeur-coquille Si-SiC, nanofils de SiC et nanotubes de SiC, par unprocédé original de carburation de nanofils de silicium, eux-mêmes élaborés par gravure plasma. Cettedémonstration a été possible grâce au contrôle de la pression de carburation, ce qui permet la maîtrisede l’exodiffusion des atomes de silicium à travers le carbure de silicium.À pression atmosphérique l’exodiffusion des atomes de silicium est restreinte ce qui permet d’élaborerdes nanofils coeur-coquille Si-SiC avec une coquille de SiC monocristalline et entièrement recouvrante.En se servant de la biocompatibilité du SiC et du bon contrôle électronique dans le silicium, ilest possible d’envisager l’utilisation de ces nanofils coeur-coquille Si-SiC pour des bio-nano-capteurs.En diminuant la pression au cours de la carburation, il est possible d’augmenter l’exodiffusion etainsi d’obtenir des nanotubes de SiC cubique de très bonne qualité cristalline avec des parois denses.Ces nanotubes de SiC sont largement modulables en termes de dimensions, et la faisabilité de leurouverture a été démontrée, permettant ainsi l’utilisation du fort rapport surface sur volume de telsnano-objets pour des capteurs électroniques notamment.Un premier pas a été franchi vers les applications des nanofils coeur-coquille Si-SiC et des nanotubesde SiC, puisque les mesures électriques réalisées sur des nano-transistors à effet de champ utilisant cesdeux types de nano-objets comme canal sont prometteurs. / Due to their superior physical and chemical properties —such as high breakdown field, high thermalconductivity and biocompatibility— compared to other semiconductors, silicon carbide is forseento be a promising materials for power electronics, bio-nano-sensors and nano-electronics in harsh environments.However, among the numerous top-down or bottom-up methods used to synthesise siliconcarbide 1D nano-objects, none has been able yet to produce SiC with a high cristalline quality.The aim of this project is to demonstrate the synthesis of silicon carbide- based 1D nanostructures—e.g. core-shell Si-SiC nanowires, SiC nanowires and SiC nanotubes— through an original processbased on the carburization of plasma-etched silicon nanowires. This demonstration is based on thecontrol of the pressure during the carburization process, which leads to the monitoring of the outdiffusionof silicon atoms through silicon carbide.Thus if the pressure is kept at the atmospheric pressure, the out-diffusion of silicon is limited andSi-SiC core-shell nanowires can be synthesized with a single-crystalline cubic SiC shell. Thanks to thebiocompatibility of the SiC shell and the good electronic transport into the Si core, bio-nano-sensorscan be considered.If the pressure is decreased during the carburization process, the outdiffusion of silicon atomsthrough SiC is enhanced, and leads to SiC nanotubes synthesis. SiC nanotubes sidewalls are dense,with an excellent crystalline quality. These original SiC nanotubes have a high surface to volume ratioand thus can be used for sensors or storage devices.The first step for direct applications has also been demonstrated since first results on electricalperformances of nano-field effect transistors, with these nano-objects as channel, are promising.

Dans le cadre du développement des réacteurs de génération IV, de nouvelles études sont menées dans le domaine des matériaux. L'objectif de ce travail est d'améliorer la compréhension du comportement mécanique et de l'endommagement de composites à matrice métallique Mo(TiC)x% contenant une forte fraction volumique de particules, et cela, dans le domaine de températures [25 – 700 °C].<br /><br />La caractérisation microstructurale a permis de comprendre l'histoire du matériau et d'identifier la nature d'une troisième phase (Mo,Ti)C fomée par diffusion du molybdène dans le carbure de titane. L'étude expérimentale a aussi révélé la percolation des particules céramiques au sein de la structure.<br />Les essais mécaniques ont mis en évidence les principales caractéristiques du matériau : le comportement macroscopique dépend à la fois de la vitesse de déformation et de la température. Ces mécanismes sont attribués au comportement thermiquement activé du molybdène.<br /><br />Nous proposons alors diverses simulations de microstructures comprenant des inclusions élastiques-fragiles dans une matrice viscoélastique. L'évolution du comportement mécanique du composite a été modélisée à l'aide d'une approche cristalline sur un agrégat 3D réel. L'agrégat numérique utilisé pour modéliser le comportement mécanique fait appel à une technique de reconstruction 3D via une acquisition par FIB/SEM/EBSD.<br /><br />Ainsi, la réponse du modèle est en bon accord avec les résultats expérimentaux et permet de décrire en fonction de la température :<br />- les mécanismes de plasticité du molybdène, en tenant compte de la transition basse/haute température ;<br />- l'endommagement du carbure de titane et les effets de la percolation.