Le sujet de thèse a pour objectif de revisiter le paradigme d'acquisition d'images dans les circuits intégrés pour le rendre robuste et scalable en technologies nanométriques (telles que le 28nm FDSOI). Ceci, afin de répondre aux contraintes d’imagerie imposée par des applications de type internet des objets. Dans ce cas, un système sur puce (SoC) hétérogène conçut en technologie avancée permettrait de répondre aux contraintes de consommation d’énergie. L’utilisation des imageurs standard actuels n’est alors pas compatible avec cette exigence à cause de leur consommation excessive et leur non compatibilité avec les technologies FDSOI 28nm. De plus, les ressources importantes de calcul numérique disponibles dans ces types de SoC couplées avec de nouveaux modes de captures d'images permettraient d’atteindre des niveaux de consommation d’énergie extrêmement bas tout en offrant la possibilité d’implémenter des algorithmes de traitement d’image complexes. Après une étude bibliographique sur les différentes méthodes d’acquisition d’image ainsi qu’une étude bibliographique sur les imageurs en technologies dites avancées pour l’imagerie et pour des applications basse consommation, il a été montré qu’il était nécessaire de numériser au plus tôt l’information lumineuse reçue par le capteur. C’est pourquoi le sujet a été orienté vers une architecture de type événementielle. L’architecture d’un capteur d’image événementiel avec traitement intelligent associé a été développée, en prenant en considération les contraintes liées à la technologie. Afin de définir ces contraintes, un circuit de test de pixel en FDSOI 28nm a été réalisé permettant d’évaluer la réponse électro-optique. Les pixels ont chacun des types et des tailles de photodiodes différentes afin de valider le type et la taille les plus efficaces. Deux architectures événementielles ont été étudiées durant cette thèse afin de répondre aux contraintes d’une implémentation en technologies FDSOI 28nm : une architecture de type « Time-to-first-Spike » (TTFS) avec un système d’inhibition et une architecture dite « multi-bus » utilisant les possibilités d’interconnections denses offertes par la technologie. Ces deux architectures visent à réduire le flot de données sortant ainsi que la consommation d’énergie. Les traitements associés à l’acquisition ont été validés par des simulations MATLAB émulant l’acquisition événementielle et les prétraitements. Ce système de vision extrait donc une carte binaire correspondant aux contrastes locaux en utilisant un principe d’inhibition par bloc. Cette architecture de traitement est basée sur le pixel TTFS (et son principe d’inhibition) en adaptant son implémentation. La carte binaire est extraite de manière synchrone ce qui permet d’éviter l’ajout de matériel lié à une implémentation purement événementielle. Cette carte binaire peut servir dans des applications telles que de la détection de mouvement, ou de la classification telles que la méthode des histogrammes des gradients (HoG) le permet. La carte binaire extraite se rapproche des motifs binaires locaux (LBP) qui sont des outils fréquemment utilisés dans la détection et la reconnaissance de visage. Une partie de la thèse a également été consacrée à l’exploitation des possibilités qu’offre la technologie FDSOI 28nm. Notamment des architectures pixels utilisant une photodiode sous le transistor ont été étudiées. Il a également été développé dix matrices de 3 par 3 pixels en intégration 3D séquentielle utilisant la technologie CoolCube™ du LETI. / The goal of this thesis is to study new image acquisition paradigm in integrated vision circuits to enhance their robustness and scalability using nanometric technologies (such as the 28nm FDSOI) in order to satisfy the imaging constraints imposed by applications such as Internet of Things. In this case, a heterogeneous system-on-chip (SoC) designed in advanced technology would meet the energy consumption constraints. Using standard imagers is not compatible with this requirement because of their excessive power consumption and their architectures non-compatible with 28nm FDSOI technologies. In addition, in these SoC, significant available digital computational resources coupled with new image acquisition modes would allow ultra-low power consumption while providing the ability to implement complex image processing.After a bibliographic study on the state of the art on image acquisition methods and a study on imagers designed with advanced technologies and on low-power applications, it has been shown that it is necessary to quickly digitize light information received by the sensor (i.e. in the pixel). This is why the subject has been oriented towards an event-based vision sensor architecture.The architecture of an event-based image sensor with its associated smart processing has been developed, taking into account technology constraints. In order to define these constraints, a 28nm FDSOI pixel test circuit has been carried out to evaluate the electro-optical response. Each pixel has a different type and size of photodiodes in order to validate the most effective type and size.Two event-based architectures were studied during this thesis in order to fit with the constraints of an implementation in 28nm FDSOI technologies: a "Time-to-first-Spike" (TTFS) architecture with an inhibition system and an architecture called "multi-bus "using the dense interconnections possibilities offered by the technology. These two architectures aim to reduce the data throughput as well as energy consumption.The processing associated to the acquisition have been validated by MATLAB simulations emulating the event acquisition and pre-processing. This vision system therefore extracts a binary map corresponding to the local contrasts using block inhibition mechanism. This processing architecture is based on TTFS pixel (and its inhibition mechanism) with a dedicated pixel schematic. The binary map is extracted in a synchronous manner, thus avoiding hardware addition inherent to an AER (Adress Event Representation) implementation. This binary map can be used for applications such as motion detection, or classification such as histogram of gradient method (HoG). This extracted binary map approaches local binary patterns (LBP), which are frequently used tools in face detection and recognition.A part of this thesis has been dedicated also to the exploration of FDSOI 28nm capabilities in terms of pixel implementation. Notably, by studying pixels using a photodiode under the FDSOI transistor. It has also been developed ten 3 by 3 pixels matrices using 3D integration with LETI technology CoolCube™.

Cette thèse s'inscrit dans le cadre de l'étude de la stabilisation de modules d'un futur collisionneur linéaire, CLIC (Compact Linear Collider). Afin d'assurer le fonctionnement et la collision des particules dans ce futur grand instrument de physique, il faut garantir l'alignement de modules guidant des faisceaux de dimensions nanométriques. Les travaux développés ont pour support expérimental deux dispositifs : un dispositif de micropositionnement, avec une résolution inférieur au 1µm, où les perturbations peuvent être simulées et un prototype de nanostabilisation active pour charges importantes (>50kg @300Hz), avec une résolution validée expérimentalement inférieur à 0,15 nm, permettant de démontrer la faisabilité du contrôle de la stabilisation subnanométrique en s'intéressant particulièrement au rejet des mouvements du sol. Les problématiques traitées lors de ces travaux portent sur la méthodologie de conception de tels systèmes, ce qui inclue la conception électromécanique et l'instrumentation, ; la mise en œuvre et la modélisation du comportement des prototypes ,; le contrôle avec notamment les aspects de non linéarité des actionneurs. Les performances obtenues de ces différents travaux et validées expérimentalement incluent notamment les points suivants: La bande passante de fonctionnement pour du micro-positionnement à l'aide d'actionneurs piézoélectriques a été augmentée grâce à la compensation d'hystérésis : Rejet de perturbation issue du support jusqu'à 100 Hz et positionnement jusqu'à 190 Hz. Il a été démontré la faisabilité du contrôle à l'aide de capteurs sismiques. L'isolation active réalisée présente une atténuation des nano-mouvement du sol dans une bande passante comprise entre 12 et 100Hz. Expérimentalement, cela conduit à une réduction des mouvements du sol de 0,6nm [rms] à 0,25nm [rms] à 50Hz et de 3,7nm [rms] à 0,9nm [rms] à 20Hz. / This thesis takes place in the framework of a general study about the stabilization of the mechanical modules of a future linear collider, CLIC (Compact Linear Collider). In order to guarantee the good operation and the particle collision, the nanometer sized beams need to be stabilized. The proposed approach was developed on two mock-ups: one dedicated to micropositioning with disturbances generation capabilities, and an active isolation system operating heavy load (up to 50kg at 300Hz) at the nanometer scale with an experimentally validated resolution of 0.15 nm. This work studies the electromechanical design and the instrumentation, the implementation of the two set-ups and their modeling,; the control scheme that takes into account the nonlinearities of the actuators. The experimental achievements include the increase of the bandwidth for piezoelectric micro-positioning thanks to an inverse hysteresis operator: the perturbation rejection is efficient until 100 Hz and the tracking control until 190 Hz. A control scheme using seismic sensors is developed to attenuate ground motion and to isolate a platform in a 12 Hz to 100 Hz frequency range. The experimental displacement is reduced from 0.6 nm to 0.25 nm at 50 Hz and from 3.7 to 0.9 at 20 Hz.

Ultracold neutron (UCN) projection imaging is demonstrated using a boron-coated back-illuminated CCD camera and the Los Alamos UCN source. Each neutron is recorded through the capture reactions with10B. By direct detection at least one of the byproducts α, 7Li and γ (electron recoils) derived from the neutron capture and reduction of thermal noise of the scientific CCD camera, a signal-to-noise improvement on the order of 104 over the indirect detection has been achieved. Sub-pixel position resolution of a few microns is confirmed for individual UCN events. Projection imaging of test objects shows a spatial resolution less than 100μm by an integrated UCN flux one the order of 106 cm−2. The bCCD can be used to build UCN detectors with an area on the order of 1 m2. The combination of micrometer scale spatial resolution, low readout noise of a few electrons, and large area makes bCCD suitable for quantum science of UCN.

京都大学 / 新制・課程博士 / 博士(工学) / 甲第23221号 / 工博第4865号 / 新制||工||1759(附属図書館) / 京都大学大学院工学研究科分子工学専攻 / (主査)教授 水落 憲和, 教授 浜地 格, 教授 SIVANIAH Easan / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DGAM

In this paper, a nanometer-sized gold electrode with an effective radius around several tens of nanometers has been modified with a monolayer of alkanethiols. There are pinholes in the monolayer, and the pinholes can be used as very small electrode. Our evaluation shows that it is possible to have only one pinhole on the monolayer covered electrode. The single pinhole electrode has been used to study the electrochemical behaviors of fast and slow electrochemical reactions. Our results show that the electrochemical response of a slow electrochemical reaction is not a Butler-Volmer response if the electrode is small enough.

This research reports on the fabrication and characterization of integrated dual nanometer-sized electrodes. The electrodes are made of closely spaced nanometer-sized platinum and gold achieved by inserting and pulling platinum wire in cylindrical glass pipette plated with gold. Cyclic voltammetry has been used to characterize the electrodes. Our results show that both electrodes can work individually and can accomplish generation/collection experiments. Factors that may affect the performance of the electrodes as well as formation mechanism of the gold film by electroless plating are discussed.

Hydrogen evolution via proton reduction occurs at a high rate at the surface of Pt than at Au electrodes. Using cyclic voltammetry, chemically modified nanometer-sized Au electrodes, prepared by the Laser-Assisted Puller Method, were employed to examine current amplification by electrocalysis at Pt nanoparticles adsorbed on the modified Au electrode surfaces. The electrodes were modified with Self-Assembled Monolayers (SAMs) of cysteamine and soaked in Pt colloid solutions overnight. Monitoring the decrements of the characteristic steady-state catalytic current for proton reduction indicated that aggregates of Pt nanoparticles are adsorbed on the cysteamine monolayers and desorb from them particle by particle. The results also indicate that some particles are strongly attached to the modified electrode surface and do not deplete even after thorough rinsing.

The colloquially named leopard rock of the Holder and Laborcita formations (Late Pennsylvanian-Early Permian) is an algal/foraminiferal boundstone that occurs within a cyclic, interfingered, carbonate siliciclastic system in the Sacramento Mountains of southeastern New Mexico) and often accompanies phylloid algal mounds. This project is the first to fully characterize and evaluate the evidence that leopard rock is microbial in origin and assess the potential influence of methane seeps and deltaic organics on its genesis. Characterization of the algal-foraminiferal boundstone revealed a highly variable expression in outcrop based on geolocated photo imagery, hand samples, optical microscopy, and scanning electron microscopy (SEM) data. Leopard rock is interpreted as microbial in origin based upon all features observed in aggregate, particularly upward-oriented concentric gradational laminae and striking clusters of segmented curvilinear cylinders (~1000 nm long). Leopard rock is best described as thrombolytic. A comprehensive categorization into thrombolite types was conducted and field and analytical data were used in creating a geospatial data base. The data was evaluated spatially in ArcMap for co-occurrence, trends, and possible associations within, and between, categorizations and formations. Distribution and associations of dome (1-3 m), small coniform (< 1 m), and planar outcrop structures and thrombolite types reflect a hierarchy of complexity and prevalence that would be expected from a microbial system. An extensive multi-scale feature comparison of potential modern analogues from Australia, the Bahamas, and Canada, in conjunction with contiguous paleo-analogues, support the interpretation of a highly adaptable complex microbial ecosystem. Results were also consistent within the global chemical, biologic, and physical context at the time of deposition. Evidence for methane seep contribution support the plausibility of supplemental microbial energy sources based on modern examples and limited paleo-analogues. Stratigraphic position and a paucity of data do not support a significant role for deltaic bathymetry or organic influx in leopard rock genesis. The results of this study provide robust evidence that leopard rock is a multi-faceted complex microbial thrombolite that displays a continuum of expression not represented by one all-encompassing term and illustrates the value of multi-parameter analyses augmented across time and space using analogues and geospatial software.

No description available.

The iron ore mining industry makes use of various processes that result in the release of airborne dust into the surrounding atmosphere where workers are exposed, to produce a final product. The deposition in the lung and toxicological influences of airborne dust can be determined by their physical- and chemical characteristics. The Occupational Health and Safety Act (OHSA) regulations for hazardous chemical substances have no current system of how the physical- and chemical properties of particulates originating from specific areas will influence a worker‘s exposure and health, especially for ultrafine particles (UFP). It is therefore imperative to characterise airborne dust containing micrometer and UFP size particles originating from specific areas to determine if there are physical- and chemical characteristics that may or may not have an influence on the workers‘ health. Aim: This pilot study is aimed at the physical- and chemical characterisation of the airborne iron ore dust generated at the process areas of an opencast iron ore mine. Method: Sampled areas included the Primary-secondary crusher, Tertiary crusher, Quaternary crusher and Sifting house. Gravimetric sampling was conducted through the use of static inhalable- and respirable samplers in conjunction with optical- and condensation particle counters that were placed near airborne dust- emitting sources. Physical- and chemical characterisation was done with the use of scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDS). Results: The results found in the study indicate high mass concentration levels of inhalable dust at all four process areas, as well as high levels of respirable dust found at the primary- secondary crusher area. Particle size distribution optical particle counter (OPC) results indicate that the majority of particles at all four process areas are in the region of 0.3 μm in size. Condensation particle counter (CPC) results integrated with OPC results indicate that at the primarysecondary and Tertiary crushers the majority of particles are found to be in the size fraction <0.3 μm. SEM analysis indicates that particle agglomeration largely occurs in the airborne iron ore dust. Particle splinters originating from larger particle collisions and breakages are present in the airborne dust. EDS analysis indicates that the elemental majority of the airborne iron ore dust consists of iron, oxygen, carbon, aluminium, silicon, potassium and calcium. The elemental percentages differ from each process area where an increase in iron and decrease in impurities can be seen as the ore moves through the beneficiation process from the Primary-secondary crusher to the Sifting house. Conclusion: The results obtained from the physical- and chemical properties of the airborne iron ore dust indicate high risk of over-exposure to the respiratory system, as well as possible ultrafine particle systemic exposure, that may overwhelm the physiological defense mechanisms of the human body and lead to reactive oxygen species (ROS) formation and the development of pathologies such as siderosis, silicasiderosis and lung cancer. / MSc (Occupational Hygiene), North-West University, Potchefstroom Campus, 2014