La manipulation d'objets de tailles microscopiques, dont le comportement est régi par des effets de surfaces, nécessite des modèles d'interaction fiables entre les micro-objets et les organes terminaux des robots. De nombreux paramètres entrent en considération dans la modélisation des forces d'adhésion. Il apparaît donc nécessaire de déterminer l'influence de chacun afin de maîtriser en particulier la force à exercer pour séparer deux objets en contact dite force de pull-off. De plus, à l'échelle microscopique, la mesure des efforts d'interaction au cours de la manipulation est complexe et n'est possible que dans certains cas particuliers. L'approche générale proposée dans ce manuscrit repose sur trois étapes. Après une présentation du contexte de ces travaux, la première étape consiste à identifier les modèles et les moyens de mesure de la littérature (chapitre 1). Deux approches sont alors mises en avant pour modéliser les forces de pull-off. Les forces et les faiblesses des modèles sont étudiées plus en détail au chapitre 2 afin de développer, au cours de la seconde étape, un nouveau modèle de force d'adhésion (chapitre 3). Ce dernier met en avant le couplage entre les déformations et les forces d'adhésion et révèle des différences majeures entre les échelles microscopique et nanoscopique. Enfin, l'objectif de la troisième étape est de corréler ce modèle avec des observations expérimentales. L'analyse des résultats expérimentaux effectuée au chapitre 4 montre la difficulté de modéliser des interactions entre des micro-objets réels et met en avant un effet d'échelle s'appliquant aux forces électrostatiques.

A l'échelle nanoscopique, un problème scientifique fondamental réside dans la difficulté de manipuler de façon interactive et répétable un nano-objet. Cette difficulté est un frein majeur pour des applications comme les nanotransistors, les nanosystèmes ou les futurs NEMS (Nano Electro Mechanical System). Ces dispositifs émergents sont ainsi ralentis dans leur cadre expérimental. Cette thèse s'inscrit dans la continuité des recherches développées au sein de l'équipe de microrobotique de l'ISIR. Elle se focalise sur l'exploitation de capteurs d'effort pour la manipulation contrôlée à plusieurs doigts actifs. Le microscope à force atomique est utilisé pour ses propriétés de capteur d'effort. Dans un premier temps, un préhenseur composé de deux doigts indépendants avec mesures des forces d'intercation a été conçue. Avec ce système original, des micromanipulations en trois dimensions de microsphères ont été réalisée avec succès dans l'air, en mesurant de façon continue les efforts d'interaction. Ce système a aussi été utilisé pour saisir et déposer des nanofils afin de former des nanocroix, ces dernières étant des nanostructures émergentes pour la fabrication, par jonctions, de nanotransistors. Par la suite, des oscillateurs en quartz ont été utilisés pour la caractérisation de nanostructures, avec retour d'effort dy- namique. Le comportement non-linéaire en raideur de nanohélices lors de l'élongation a été caractérisé pour la première fois sur la totalité de la plage. Enfin, des sondes en quartz de haute fréquence ont été exploitées pour augmenter la vitesse d'acquisition d'images de l'AFM. De cette manière, la tâche de manipulation et d'imagerie en parallèle sous AFM a été optimisée et de nombreuses applications sont maintenant envisagées.

L'intégration de diélectriques High- k dans les empilements de grille des transistors a fait naître des problèmes de fiabilité complexes. A cela vient s'ajouter, en vue des technologies sub-32nm planaires, de nouvelles problématiques liées à l'utilisation de substrats silicium sur isolant complètement désertés FDSOI. En effet, l'intégration d'un oxyde enterré sous le film de silicium va modifier électrostatique de la structure et faire apparaître une nouvelle interface Si/SiO2 sujette à d'éventuelles dégradations. Ce manuscrit présente différentes méthodes de caractérisation électrique ainsi que différentes études de fiabilité des dispositifs FDSOI intégrants des empilements High- /grille métallique. Dans un premier temps, une étude complète du couplage électrostatique dans des structures FDSOI est réalisée, permettant de mieux appréhender l'effet d'une tension en face arrière sur les caractéristiques électriques des dispositifs. Différentes méthodes de caractérisation des pièges d'interface sont ensuite présentées et adaptées, lorsque possible, au cas spécifique du FDSOI, où les défauts entre le film de silicium et l'oxyde enterré doivent être pris en compte. Enfin, différentes études de fiabilité sont présentées, des phénomènes de PBTI et de NBTI sur des dispositifs à canaux longs aux phénomènes propres aux dispositifs de petite dimension, tels que l'impact des porteurs chauds dans des structures FDSOI à film ultra fins et les effets d'augmentation de tension de seuil lorsque les largeurs de grille diminuent.

Le sujet de thèse porte sur des micro/nanorésonateurs ainsi que leurs électroniques de lecture. Les composants mécaniques sont utilisés pour mesurer des masses inférieures à l'attogramme (10-18 g) ou de très faibles concentrations de gaz. Ces composants peuvent ensuite être mis en réseau afin de réaliser des spectromètres de masse ou des détecteurs de gaz. Afin d'atteindre les résolutions nécessaires, il a été choisi d'utiliser une détection harmonique de résonance détectant les variations de la fréquence de résonance d'une nanostructure mécanique. Les dimensions du résonateur sont réduites afin d'augmenter sensibilité en masse, cependant le niveau du signal électrique en sortie du composant est également réduit. Ce faible signal nécessite donc de concevoir de nouvelles transductions électromécaniques ainsi que des architectures électroniques qui minimisent le bruit, les couplages parasites et qui peuvent être mise en réseau.

Ultrasonic wire bonding is the most utilized technique in forming electrical interconnections in microelectronics. However, there is a lacking in the fundamental understanding of the process. In order for there to be improvements in the process a better understanding of the process is required. The mechanism of the bond formation in ultrasonic wire bonding is not known. Although there have been theories proposed, inconsistencies have been shown to exist in them. One of the main inconsistencies is the contribution of ultrasound to the bonding process. A series of experiments to investigate the mechanism of bond formation are performed on a semi automatic wire bonder at room temperature. 25 µm diameter Au wire is ball bonded and also 25 µm diameter Al wire is wedge-wedge bonded onto polished Cu sheets of thickness 2 mm. It is found that a modified microslip theory can describe the evolution of bonding. With increasing ultrasonic power the bond contact transitions from microslip into gross sliding. The reciprocating tangential relative motion at the bond interface results in wear of surface contaminants which leads to clean metal/metal contact and bonding. The effect of superimposed ultrasound during deformation on the residual hardness of a bonded ball is systematically studied for the first time. An innovative bonding procedure with in-situ ball deformation and hardness measurement is developed using an ESEC WB3100 automatic ball bonder. 50 µm diameter Au wire is bonded at various ultrasound levels onto Au metallized PCB substrate at room temperature. It is found that sufficient ultrasound which is applied during the deformation leads to a bonded ball which is softer than a ball with a similar amount of deformation without ultrasound. No hardening of the 100 µm diameter Au ball is observed even with the maximum ultrasonic power capable of the equipment of 900 mW. In summary, the fundamental effect of ultrasound in the wire bonding process is the reciprocating tangential displacement at the bond interface resulting in contaminant dispersal and bonding. A second effect of ultrasound is the softening of the bonded material when compared to a similarly non-ultrasound deformed ball.

Ultrasonic wire bonding is the most utilized technique in forming electrical interconnections in microelectronics. However, there is a lacking in the fundamental understanding of the process. In order for there to be improvements in the process a better understanding of the process is required. The mechanism of the bond formation in ultrasonic wire bonding is not known. Although there have been theories proposed, inconsistencies have been shown to exist in them. One of the main inconsistencies is the contribution of ultrasound to the bonding process. A series of experiments to investigate the mechanism of bond formation are performed on a semi automatic wire bonder at room temperature. 25 µm diameter Au wire is ball bonded and also 25 µm diameter Al wire is wedge-wedge bonded onto polished Cu sheets of thickness 2 mm. It is found that a modified microslip theory can describe the evolution of bonding. With increasing ultrasonic power the bond contact transitions from microslip into gross sliding. The reciprocating tangential relative motion at the bond interface results in wear of surface contaminants which leads to clean metal/metal contact and bonding. The effect of superimposed ultrasound during deformation on the residual hardness of a bonded ball is systematically studied for the first time. An innovative bonding procedure with in-situ ball deformation and hardness measurement is developed using an ESEC WB3100 automatic ball bonder. 50 µm diameter Au wire is bonded at various ultrasound levels onto Au metallized PCB substrate at room temperature. It is found that sufficient ultrasound which is applied during the deformation leads to a bonded ball which is softer than a ball with a similar amount of deformation without ultrasound. No hardening of the 100 µm diameter Au ball is observed even with the maximum ultrasonic power capable of the equipment of 900 mW. In summary, the fundamental effect of ultrasound in the wire bonding process is the reciprocating tangential displacement at the bond interface resulting in contaminant dispersal and bonding. A second effect of ultrasound is the softening of the bonded material when compared to a similarly non-ultrasound deformed ball.

Time-interleaved analog-to-digital converters make use of parallelization to increase the rate at which an analog signal can be digitized. Using M channels at their maximum sampling frequency allows for an overall sampling frequency of M times the individual converters' sampling rate. However, the performance of interleaved systems suffers from mismatches between the sub-converters. Offset mismatches, gain mismatches, and timing mismatches all contribute to the degradation of the resolution of the ADC system. Offset and gain mismatches can be corrected for in the digital domain with minimal extra processing. However, the effects of timing mismatches (specifically, the magnitude of the spurious tones that are introduced) are dependent on the frequency of the input, so digital correction is not a trivial task. This makes a circuit-based correction mechanism a much more desirable solution to the problem. This work explores the effect of timing mismatches on interleaved analog-to-digital converter performance. A set of requirements is derived to specify the performance of a variable-delay circuit for the tuning of sample clocks. Since the mismatches can be composed of both fixed and random components, several candidate architectures are modeled for their delay and jitter performance. One candidate is selected for design, based on its jitter performance and on practical considerations. A practical implementation of the clock-adjustment circuit is designed, featuring low-noise differential clock paths with high precision delay adjustment. A means of testing the circuit and verifying the precision of adjustment is presented. The design is implemented for fabrication, and post-layout simulations are shown to demonstrate the feasibility and functionality of the design.

Thermal management is a key challenge to improving the performance of microelectronic devices. For many high performance applications, the thermal resistance between chip and heat sink may account for half of the total thermal budget. Chip-level heat dissipation is therefore a critical bottleneck to the development of advanced microelectronics with high junction temperatures. Recently aligned carbon nanotube arrays have been developed as possible next generation thermal interface materials to overcome this thermal limitation, however the thermal physics of these nanoscale interfaces remains unclear. In this thesis, the thermal interface resistance between a carbon nanotube and adjoining carbon, silicon, or copper substrate is investigated through non-equilibrium molecular dynamics simulation. Phonon transmission is calculated using a simplified form of the diffuse mismatch model with direct simulation of the phonon density of states. The results of theory and simulation are reported as a function of temperature in order to estimate the importance of anharmonicity and inelastic scattering. The results of this work provide a better understand of the mechanisms of thermal transport to assist future CNT TIM research and development.

A variety of techniques are available to enable the fabrication of photonic crystal structures. Multi-beam-interference lithography (MBIL) is a relatively new technique which offers many advantages over more traditional means of fabrication. Unlike the more common fabrication methods such as optical and electron-beam lithography, MBIL is a method that can produce both two- and three-dimensional large-area photonic crystal structures for use in the infrared and visible light regimes. While multi-beam-interference lithography represents a promising methodology for the fabrication of PC structures, there has been an incomplete understanding of MBIL itself. The research in this thesis focuses on providing a more complete, systematic description of MBIL in order to demonstrate its full capabilities. Analysis of both three- and four-beam interference is investigated and described in terms of contrast and crystallography. The concept of a condition for primitive-lattice-vector-direction equal contrasts} is introduced in this thesis. These conditions are developed as nonlinear constraints when optimizing absolute contrast for producing lithographically useful interference patterns (meaning high contrast and localized intensity extrema). By understanding the richness of possibilities within MBIL, a number of useful interference patterns are found that can be created in a straightforward manner. These patterns can be both lithographically useful and structurally useful (providing interference contours that can define wide-bandgap photonic crystals). Included within this investigation are theoretical calculations of band structures for photonic crystals that are fabricatable through MBIL. The resulting calculations show that not only do most MBIL-defined structures exhibit similar performance characteristics compared to conventionally designed photonic crystal structures, but in some cases MBIL-defined structures show a significant increase in bandgap size. Using the results from this analysis, a number of hexagonal photonic crystals are fabricated using a variety of process conditions. It is shown that both rod- and hole-type photonic crystal structures can be fabricated using processes based on both positive and negative photoresist. The "light-field" and "dark-field" interference patterns used to define the hexagonal photonic crystal structures are quickly interchanged by the proper adjustment of each beam's intensity and polarization. The resulting structures, including a large area (~1 cm², 1 x 10⁹ lattice points) photonic crystal are imaged using a scanning electron microscope. Multi-beam-interference lithography provides an enabling initial step for the wafer-scale, cost-effective integration of the impressive PC-based devices into manufacturable DIPCS. While multi-beam-interference lithography represents a promising methodology for the fabrication of PC structures, it lacks in the ability to produce PC-based integrated photonic circuits. Future research will target the lack of a large-scale, cost-effective fabrication methodology for photonic crystal devices. By utilizing diffractive elements, a photo-mask will be able to combine both MBIL and conventional lithography techniques into a single fabrication technology while taking advantage of the inherent positive attributes of both.

In this work, a 3.5 GHz RF Transmitter and Successive Approximation ADC design has been presented. The transmitter serves as an intermediate block which translates 350 MHz signal into 3.5 GHz signal. This signal is applied to 6-40 GHz wideband transmitter. The emphasis is on the design of Up conversion Mixer with high linearity, low noise and moderate image rejection performance. The successive approximation analog to digital converter was designed as a part of feedback loop control, which consists of a sensor circuit to detect the temperature changes in a power amplifier and the ADC to convert the sensor output to digital data. The data is used to determine the necessary control signals to restore the performance of the power amplifier. The circuits have been designed and implemented in ST Microelectronics CMOS 90nm process.