Search

The fourth industrial revolution and the Internet of things introduce new challenges to reliable circuit design. In this context, new transports circuits must be able to reliably work at an extensive temperature range. Transistor characteristics have been known to be highly temperature-dependent. Zero Temperature Coefficients (ZTC) bias points have already been proposed to minimize transistor's temperature drift effects. However, this ZTC approach considers the influence of a single transistor on the different circuit specifications' temperature behavior. Moreover, it does not allow and does not allow a more general and methodological approach. On the other hand, design frameworks such as the gm/ID allow a methodological approach for analog design, with no possibility of temperature analysis. This work proposes an extension of the gm/ID methodology by introducing the concept of temperature normalized gm/ID parameters. This concept allows the designer to take temperature effects in a very early design stage in most analog circuits, allowing a unified methodology for temperature-aware design. Those results are validated from simulation and measurement data from -40 °C to 200 °C on the XT018 technology node of X-FAB. Three different design examples are also presented in this thesis, being: a Band Gap voltage reference, an operational transconductance amplifier (OTA), and a voltage-controlled oscillator (VCO). Those circuits present a better presenting a better temperature performance comparable to the state-of-the-art specifications., La quatrième révolution industrielle et l'arrivée de l'Internet des Objets introduisent de nouveaux défis pour la conception de circuits fiables. Dans les nouveaux véhicules, les circuits doivent pouvoir fonctionner de manière fiable dans une plage de température étendue. Les caractéristiques des transistors sont connues pour être fortement dépendantes de la température. Pour minimiser ces effets de dérive thermique, l'utilisation de points de polarisation ZTC (Zero Temperature Coefficients) a déjà été proposée. Cependant, cette approche ZTC considère l'influence d'un seul transistor sur le comportement en température des différentes spécifications du circuit et ne permet pas une approche plus générale et méthodologique. D'autre part, les techniques de conception tels que la gm/ID permettent une approche méthodologique pour la conception analogique, sans possibilité d'analyse en température. Ce travail propose une extension de la méthodologie gm/ID en introduisant le concept de paramètres gm/ID normalisés en température. Ces paramètres permettent au concepteur de prendre en compte les effets de la température à un stade très précoce de la conception de la plupart des circuits analogiques, permettant ainsi une méthodologie unifiée pour la conception sensible à la température. Ces résultats sont validés à partir de données de simulation et de mesures de -40 °C à 200 °C avec la technologie XT018 de X-FAB. Trois exemples de conception différents sont également présentés dans cette thèse: circuit de référence de tension (Bandgap), Amplificateur à Transconductance (OTA), et oscillateur contrôlé par tension (VCO). Ces circuits présentent une meilleure performance en température comparable aux spécifications de l'état de l'art.

The gallium nitride (GaN) µLED is a prime candidate to meet the specifications of new augmented reality applications and could also find its place in a wider range of applications in the longer term. However, the high perimeter-to-surface ratio of µLEDs makes them very sensitive to edge defects generated by pixelation etching. Therefore, passivation post-treatment steps of the µLEDs sidewalls must be implemented and optimized in order to minimize the decrease in efficiency that inorganic µLEDs are facing more generally. This constitutes the problematic of this thesis, dealing more specifically with GaN/InGaN blue µLEDs.First, this thesis has attempted to identify the most promising path for the passivation of GaN µLEDs from the literature. Then, physicochemical characterizations allowed to study the effect of chemical treatments or alumina deposition by ALD (Atomic Layer Deposition) methods on the surface properties of InGaN from µLEDs active region. Some specific behaviors have been observed especially after treatments with hydrochloric acid or ammonium sulfide, rather effective in reducing the amount of indium on the surface, or even oxidation of the latter. However, an ammonia treatment would be more favorable to enhance ALD passivation layer nucleation, which is just as important for the quality of the interface. This could be supported by C-V (capacitance-voltage) electrical characterizations of MIS (metal-insulator-semiconductor) capacitors, which highlighted a strong reduction in the density of shallow interface states between an alumina ALD layer and n-GaN by adding a simple pretreatment with ammonia prior to deposition (over 98% reduction). Finally, using a short cycle study methodology, cathodoluminescence characterizations were used to compare the effect of more than twenty different passivation variants on the luminescence properties of 4µm µLED mesas. The study illustrated the effectiveness of chemical etching with potassium hydroxide combined with ALD alumina deposition, but also of the very promising aluminum nitride/alumina bilayer passivation alternative. In both cases, the emission was significantly more homogeneous across mesas surfaces : the "critical" area with attenuated luminescence at the µLED periphery was reduced by nearly 50% and 65%, respectively., La µLED à base de nitrure de gallium (GaN) constitue un candidat de choix pour répondre aux spécifications des nouvelles applications de réalité augmentée, et pourrait également trouver sa place dans une gamme plus large d’applications à plus grande échéance. Cependant, le ratio périmètre-surface élevé des µLED les rend très sensibles aux défauts de bords générés, en particulier, par la gravure de pixellisation. Des étapes de post-traitement des flancs des µLED, relevant du domaine de la passivation de surface, doivent donc être implémentées et optimisées, afin de minimiser la chute d’efficacité à laquelle les µLED inorganiques sont confrontées plus généralement. C’est ce qui constitue la problématique de cette thèse, traitant plus spécifiquement de µLED bleues à base de GaN/InGaN.En premier lieu, cette thèse s’est efforcée d’identifier les pistes les plus prometteuses pour la passivation des µLED GaN à partir de la littérature. Ensuite, des caractérisations physico-chimiques ont permis d’étudier l’effet de traitements chimiques ou encore de dépôts d’alumine par méthodes ALD (Atomic Layer Deposition) sur les propriétés de surface de l’InGaN, le matériau constituant la zone active des µLED. Certains comportements spécifiques ont pu être observés, notamment après des traitements avec de l’acide chlorhydrique ou du sulfure d’ammonium, plutôt efficaces pour réduire la quantité d’indium en surface, ou encore l’oxydation de ce dernier. Cependant, un traitement avec de l’ammoniaque serait plus favorable à une bonne nucléation de la couche ALD de passivation, toute aussi importante pour la qualité de l’interface. Cela a pu être étayé par des caractérisations électriques C-V (capacité-tension) de condensateurs MIS (métal-isolant-semiconducteur), qui ont permis de souligner une forte réduction de la densité d’états d’interface peu profonds entre une couche ALD d’alumine et du n-GaN, par l’ajout d’un simple prétraitement avec l’ammoniaque avant dépôt (plus de 98% de réduction). Enfin, à l’aide d’une méthodologie d’étude en cycles courts, des caractérisations de cathodoluminescence ont permis de comparer l’effet de plus d’une vingtaine de variantes différentes de passivation sur les propriétés de luminescence de mésas µLED de 4µm. L’étude a permis d’illustrer l’efficacité d’une gravure chimique avec de l’hydroxyde de potassium combinée à un dépôt ALD d’alumine, mais également celle de l’alternative très prometteuse que constitue la passivation en bicouche nitrure d’aluminium/alumine. Dans les deux cas, l’émission était nettement plus homogène sur la surface des mésas : la zone « critique » à luminescence atténuée en périphérie de µLED était réduite de près de 50% et 65%, respectivement.

International audience; The simulation of nanoscale 2D DG-MOSFETs and similar semiconductor devices through a de-terministic and accurate model can be very useful for the industry but is particularly costly from the computational point of view. In this paper, we develop a hybrid parallel solver: the computing phases which corresponds to the simulation of the Boltzmann transport equation in the longitudinal dimension are performed on the GPU, while that phases related to the modelling of the electrons as waves in the transversal dimension are computed on the multi-core CPUs by using OpenMP. We have adapted the most costly computing phases to GPU in an efficient manner, achieving high performance and reducing drastically the simulation time. We give details about the parallel-design strategy and show the performance results.

Microwave cavities have been widely used to investigate the behavior of closed few-level systems. Here, we show that they also represent a powerful probe for the dynamics of charge transfer between a discrete electronic level and fermionic continua. We have combined experiment and theory for a carbon nanotube quantum dot coupled to normal metal and superconducting contacts. In equilibrium conditions, where our device behaves as an effective quantum dot-normal metal junction, we approach a universal photon dissipation regime governed by a quantum charge relaxation effect. We observe how photon dissipation is modified when the dot admittance turns from capacitive to inductive. When the fermionic reservoirs are voltage biased, the dot can even cause photon emission due to inelastic tunneling to/from a Bardeen-Cooper-Schrieffer peak in the density of states of the superconducting contact. We can model these numerous effects quantitatively in terms of the charge susceptibility of the quantum dot circuit. This validates an approach that could be used to study a wide class of mesoscopic QED devices., Comment: 15 pages, 8 figures, minor differences with published version

International audience; We use coupled micropillars etched out of a semiconductor microcavity to engineer a spin-orbit Hamiltonian for photons and polaritons in a microstructure. The coupling between the spin and orbital momentum arises from the polarization-dependent confinement and tunneling of photons between adjacent micropillars arranged in the form of a hexagonal photonic molecule. It results in polariton eigenstates with distinct polarization patterns, which are revealed in photoluminescence experiments in the regime of polariton condensation. Thanks to the strong polariton nonlinearities, our system provides a photonic workbench for the quantum simulation of the interplay between interactions and spin-orbit effects, particularly when extended to two-dimensional lattices.

International audience; We describe the first implementation of a coupled atom transistor where two shallow donors (P or As) are implanted in a nanoscale silicon nanowire and their electronic levels are controlled with three gate voltages. Transport spectroscopy through these donors placed in series is performed both at zero and microwave frequencies. The coherence of the charge transfer between the two donors is probed by Landau-Zener-Stuckelberg interferometry. Single-charge transfer at zero bias (electron pumping) has been performed and the crossover between the adiabatic and non-adiabatic regimes is studied.

This thesis treats of nonlinear behaviors in two different types of bidimensional systems: semiconductor heterostructures as well as a monolayer material, graphene. It consists into two main parts: the theoritical study of Wannier-Stark quantification in electrically biased quantum well superlattices, and the modelling of new effects for electromagnetic wave generation in the Terahertz range. In quantum well superlattices under an external voltage, the electric field induces bidimensional confinement of the charge carriers, this effect is known as Wannier-Stark quantification. We examine two interesting consequences of this confinement: the strong photocurrent nonlinearities induced when the superlattice is placed between thick tunnel barriers, and the possibility to control light-matter coupling as well as Terahertz gain in superlattices coupled to a semiconductor microcavity. In a second part of this work, we study quantitatively two new nonlinear effects for Terahertz generation. The first one is Terahertz emission exaltation in a polaritonic system reaching the polariton lasing regime. We model precisely this effect and suggest a new scheme using a double microcavity and providing very significant reduction of the diffusion losses. The second effect is photon drag in graphene under pulsed excitation. We build a microscopic and predictive model for this phenomenon which provides a comprehensive insight on the relevant parameters for the optimisation of the Terahertz generation. These theoretical study is illustrated with several experimental results.; Cette thèse traite de comportements non-linéaires dans deux types de systèmes bidimensionnels différents: les hétérostructures semiconductrices ainsi qu'un matériau monocouche, le graphène. Elle comporte deux axes principaux: l'étude théorique de la quantification de Wannier-Stark dans les super-réseaux de puits quantiques biaisés électriquement, et la modélisation d'effets nouveaux pour la génération de rayonnement électromagnétique dans le domaine Terahertz.Dans les super-réseaux de puits quantiques soumis à une tension externe, le champ électrique induit un confinement bidimensionnel des porteurs de charge nommé quantification de Wannier-Stark. On modélise deux conséquences originales de cette quantification: d'une part, les fortes non-linéarités de photocourant dans un super-réseau placé entre deux barrières tunnel épaisses, et d'autre part, la possibilité de contrôler électriquement le couplage lumière-matière et le gain dans la gamme Terahertz dans un super-réseau biaisé couplé à une microcavité planaire. Dans un second temps, on étudie quantitativement deux effets non-linéaires nouveaux pour la génération Terahertz. Le premier est l'exaltation de l'émission Terahertz dans un système polaritonique en régime de laser à polaritons. On modélise précisément cet effet et on propose un nouveau dispositif utilisant une microcavité double et permettant de réduire très significativement les pertes par diffusion. Le second effet étudié est le transfert d'impulsion photonique dans le graphène sous excitation impulsionnelle. On construit un modèle microscopique prédictif de ce phénomène qui permet de déterminer les paramètres importants pour l'optimisation de l'impulsion Terahertz générée. Cette étude théorique est illustrée par plusieurs résultats expérimentaux.

We study theoretically two-dimensional single-crystalline sheets of semiconductors that form a honeycomb lattice with a period below 10 nm. These systems could combine the usual semiconductor properties with Dirac bands. Using atomistic tight-binding calculations, we show that both the atomic lattice and the overall geometry influence the band structure, revealing materials with unusual electronic properties. In rocksalt Pb chalcogenides, the expected Dirac-type features are clouded by a complex band structure. However, in the case of zinc-blende Cd-chalcogenide semiconductors, the honeycomb nanogeometry leads to rich band structures, including, in the conduction band, Dirac cones at two distinct energies and nontrivial flat bands and, in the valence band, topological edge states. These edge states are present in several electronic gaps opened in the valence band by the spin-orbit coupling and the quantum confinement in the honeycomb geometry. The lowest Dirac conduction band has S-orbital character and is equivalent to the pi-pi* band of graphene but with renormalized couplings. The conduction bands higher in energy have no counterpart in graphene; they combine a Dirac cone and flat bands because of their P-orbital character. We show that the width of the Dirac bands varies between tens and hundreds of meV. These systems emerge as remarkable platforms for studying complex electronic phases starting from conventional semiconductors. Recent advancements in colloidal chemistry indicate that these materials can be synthesized from semiconductor nanocrystals., 12 pages, 12 figures

The origin of superconductivity in bulk SrTiO$_{3}$ is a mystery, since the non-monotonous variation of the critical transition with carrier concentration defies the expectations of the crudest version of the BCS theory. Here, employing the Nernst effect, an extremely sensitive probe of tiny bulk Fermi surfaces, we show that down to concentrations as low as $5.5 \times 10^{17}cm^{-3}$, the system has both a sharp Fermi surface and a superconducting ground state. The most dilute superconductor currently known has therefore a metallic normal state with a Fermi energy as little as 1.1 meV on top of a band gap as large as 3 eV. Occurrence of a superconducting instability in an extremely small, single-component and barely anisotropic Fermi surface implies strong constraints for the identification of the pairing mechanism., Final version