<p>This thesis considers integration of antennas and active electronics manufactured on the same substrate. The main topic is on-chip antennas for commercial silicon processes, but hybrid integration using printed circuit board technology is also addressed.</p><p>The possible use of micromachining techniques as a means of reducing substrate losses of antennas manufactured on low resistivity silicon wafers is investigated. Compact dipole, loop, and inverted-F antennas for the 20-40 GHz frequency range are designed, implemented, and characterized. The results show significantly improved antenna efficiency when micromachining is used as a post-processing step for on-chip antennas manufactured in silicon technology.</p><p>High resistivity wafers are used in a commercial silicon germanium technology to improve the efficiency of dipole antennas realized using the available circuit metal layers in the process. Monolithically integrated 24 GHz receivers with on-chip antennas are designed and evaluated with regard to antenna and system performance. No noticeable degradation of the receiver performance caused by cross talk between the antenna and the integrated circuit is observed.</p><p>For low frequency antenna arrays, such as base station antennas, hybrid integration of active devices within the antenna aperture is treated. A compact varactor based phase shifter for traveling wave antenna applications is proposed and evaluated. Electrically steerable traveling wave patch antenna arrays, with the phase shifters implemented in the same conductor layer as the radiating elements, are designed and manufactured in microstrip technology. It is experimentally verified that the radiation from the feed network and phase shifters in the proposed antenna configuration is small.</p>

This thesis considers integration of antennas and active electronics manufactured on the same substrate. The main topic is on-chip antennas for commercial silicon processes, but hybrid integration using printed circuit board technology is also addressed. The possible use of micromachining techniques as a means of reducing substrate losses of antennas manufactured on low resistivity silicon wafers is investigated. Compact dipole, loop, and inverted-F antennas for the 20-40 GHz frequency range are designed, implemented, and characterized. The results show significantly improved antenna efficiency when micromachining is used as a post-processing step for on-chip antennas manufactured in silicon technology. High resistivity wafers are used in a commercial silicon germanium technology to improve the efficiency of dipole antennas realized using the available circuit metal layers in the process. Monolithically integrated 24 GHz receivers with on-chip antennas are designed and evaluated with regard to antenna and system performance. No noticeable degradation of the receiver performance caused by cross talk between the antenna and the integrated circuit is observed. For low frequency antenna arrays, such as base station antennas, hybrid integration of active devices within the antenna aperture is treated. A compact varactor based phase shifter for traveling wave antenna applications is proposed and evaluated. Electrically steerable traveling wave patch antenna arrays, with the phase shifters implemented in the same conductor layer as the radiating elements, are designed and manufactured in microstrip technology. It is experimentally verified that the radiation from the feed network and phase shifters in the proposed antenna configuration is small.

Algunas aplicaciones tales como la radio astronomía y las comunicaciones con el espacio profundo requieren receptores muy sensibles. Esta tesis trata sobre la tecnología criogénica aplicada a la ingeniería de microondas y se centra en el diseño de amplificadores de muy bajo ruido tanto en tecnología híbrida (MIC) como monolítica (MMIC). El trabajo cubre un ancho campo de conocimiento desde la fabricación mecánica y la configuración de los sistemas hasta el diseño y medida de las aplicaciones finales. Comenzando con pautas y consejos para diseñar sistemas criogénicos (criostatos) este documento profundiza en la medida de parámetros-S y ruido. El diseño de circuitos criogénicos se inicia con el estudio de los efectos de las bajas temperaturas sobre los transistores y componentes de microondas centrándose en los dispositivos de fosfuro de indio (InP). El conocimiento adquirido en este estudio se aplica al diseño de amplificadores de muy bajo ruido en banda Ka. / Some applications such as radio astronomy and deep space communications require very sensitive receivers. This dissertation deals with the cryogenic technology applied to the microwave engineering and focuses on the design of very low noise amplifiers both in hybrid (MIC) and monolithic (MMIC) technologies. The work covers a wide field of knowledge from hardware manufacture and system set up to final applications design and measurement. Starting from guidelines and advices to design cryogenic systems (cryostats) this document goes into S-parameters and noise measurements in deep. The design of cryogenic circuits is initialized with the study of the effect of low temperatures on microwave transistors and components focusing in indium-phosphide (InP) devices. The knowledge gained with this study is applied to the design of very low noise amplifiers in Ka-band.

Shape memory alloys (SMAs) provide exciting opportunities for miniature actuation systems. As SMA actuators are scaled down in size, cooling increases and bandwidth, improves. However, the inclusion of a bias element with which to cycle the SMA actuator becomes difficult at very small scales. One technique used to avoid the necessity of having to include a separate bias element is the use of local annealing to fabricate a monolithic device out of nickel titanium (NiTi). The actuator geometry is machined out of a single piece of non-annealed NiTi. After locally annealing a portion of the complete device, that section exhibits the shape memory effect while the remainder acts as structural support and provides the bias force required for cycling. This work proposes one such locally-annealed monolithic SMA actuator for future incorporation in a device that navigates the digestive tract. After detailing the derivation of lumped electro-mechanical models for the actuator, a description of the prototyping procedure, including fabrication and local annealing of the actuator, is provided. This thesis presents the experimental prototype actuator behaviour and compares it with simulations generated using the developed models.

Shape memory alloys (SMAs) provide exciting opportunities for miniature actuation systems. As SMA actuators are scaled down in size, cooling increases and bandwidth, improves. However, the inclusion of a bias element with which to cycle the SMA actuator becomes difficult at very small scales. One technique used to avoid the necessity of having to include a separate bias element is the use of local annealing to fabricate a monolithic device out of nickel titanium (NiTi). The actuator geometry is machined out of a single piece of non-annealed NiTi. After locally annealing a portion of the complete device, that section exhibits the shape memory effect while the remainder acts as structural support and provides the bias force required for cycling. This work proposes one such locally-annealed monolithic SMA actuator for future incorporation in a device that navigates the digestive tract. After detailing the derivation of lumped electro-mechanical models for the actuator, a description of the prototyping procedure, including fabrication and local annealing of the actuator, is provided. This thesis presents the experimental prototype actuator behaviour and compares it with simulations generated using the developed models.

Η μετατροπή του φυσικού αερίου προς αέριο σύνθεσης αποτελεί εξαιρετικά σημαντική διεργασία αφού το αέριο σύνθεσης είναι απαραίτητο για την παραγωγή μεθανόλης και τη σύνθεση Fischer-Tropsch. Η ευρύτερα διαδεδομένη διεργασία για τη μετατροπή του μεθανίου – το οποίο αποτελεί το κυριότερο συστατικό του φυσικού αερίου - προς αέριο σύνθεσης είναι η αναμόρφωσή του με ατμό. Τα τελευταία χρόνια όμως ερευνήθηκαν κι άλλες μέθοδοι παραγωγής αερίου σύνθεσης από μεθάνιο όπως η αναμόρφωση με CO2, η καταλυτική και μη μερική οξείδωση του μεθανίου κ.ά. Μεταξύ των διεργασιών που προαναφέρθηκαν η καταλυτική μερική οξείδωση του μεθανίου προς αέριο σύνθεσης κερδίζει έδαφος σε ερευνητικό επίπεδο. Η διεργασία παρουσιάζει σημαντικά πλεονεκτήματα όπως το γεγονός ότι δεν είναι τόσο ισχυρά εξώθερμη όσο η καύση του μεθανίου προς CO2 και Η2, καθώς και το γεγονός ότι η αναλογία του παραγόμενου αερίου σύνθεσης είναι CO/H2 : ½ αναλογία που είναι η απαραίτητη για την παραγωγή μεθανόλης και τη σύνθεση Fischer-Tropsch. Η μερική οξείδωση του μεθανίου προς αέριο σύνθεσης έχει μελετηθεί εκτενώς σε αντιδραστήρες σταθερής κλίνης. Λόγω όμως των μειονεκτημάτων που παρουσιάζουν κυριότερα από τα οποία είναι η ανάπτυξη θερμών κηλίδων λόγω της εξώθερμης καύσης που λαμβάνει χώρα στην καταλυτική κλίνη και η υψηλή πτώση πίεσης, η έρευνα στράφηκε στην αναζήτηση νέου τύπου αντιδραστήρων. Στην παρούσα εργασία η μερική οξείδωση του μεθανίου προς αέριο σύνθεσης μελετήθηκε σε αντιδραστήρα τοιχώματος θερμικής ολοκλήρωσης και σε μονολιθικό αντιδραστήρα. Ο αντιδραστήρας τοιχώματος θερμικής ολοκλήρωσης αποτελείται από ένα κεραμικό μη πορώδη σωλήνα υψηλής θερμικής αγωγιμότητας στον οποίο εναποτίθεται καταλύτης σε μορφή συνεχούς φιλμ. Ο αντιδραστήρας εσωκλείεται σε ένα δεύτερο σωλήνα που μπορεί να είναι από κεραμικό ή χαλαζία. Λόγω της κατασκευής του αντιδραστήρα υπάρχει δυνατότητα εναπόθεσης του καταλύτη στο εσωτερικό και στο εξωτερικό τοίχωμα του κεραμικού σωλήνα. Ο καταλύτης που χρησιμοποιήθηκε ήταν Rh/Al2O3. Μελετήθηκε η επίδραση του τρόπου εναπόθεσης του καταλύτη, η μεταβολή της παροχής εισόδου, η μεταβολή της θερμοκρασίας του φούρνου και η επίδραση της ποσότητας του καταλύτη που εναποτίθεται στον αντιδραστήρα. Τα αποτελέσματα έδειξαν ότι η εναπόθεση κατάλληλης ποσότητας καταλύτη στην εξωτερική και εσωτερική επιφάνεια του αντιδραστήρα μπορεί να οδηγήσει σε ομοιόμορφο θερμοκρασιακό προφίλ κατά μήκος του αντιδραστήρα αφού η θερμότητα που παράγεται στην εσωτερική επιφάνεια του αντιδραστήρα κατά την εξώθερμη καύση, μεταφέρεται μέσω του τοιχώματος στην εξωτερική επιφάνεια όπου πραγματοποιούνται οι ενδόθερμές αντιδράσεις αναμόρφωσης. Η απουσία θερμών κηλίδων δίνει τη δυνατότητα χρήσης υψηλών παροχών εισόδου. Από τα πειραματικά αποτελέσματα υπήρχαν ενδείξεις ότι μέρος του παραγόμενου CO παράγεται μέσω του άμεσου σχήματος. Η μερική οξείδωση του μεθανίου προς αέριο σύνθεσης μελετήθηκε και σε μονολιθικό αντιδραστήρα. Το κυριότερο πλεονέκτημα των μονολιθικών αντιδραστήρων είναι η χαμηλή πτώση πίεσης και η μεγάλη ενεργός επιφάνεια. Ο καταλύτης που εναποτέθηκε στο εσωτερικό των καναλιών του μονόλιθου ήταν 0.5% Ru/TiO2 (2%CaO) ο οποίος είναι ενεργός και σταθερός για την μερική οξείδωση του μεθανίου προς αέριο σύνθεσης. Στα πειράματα που πραγματοποιήθηκαν σε μονολιθικό αντιδραστήρα μελετήθηκε η επίδραση της μεταβολής της παροχής, της μεταβολής της θερμοκρασίας του φούρνου, καθώς και η σύσταση του μίγματος εισόδου. Τα πειραματικά αποτελέσματα έδειξαν ότι αρχικά λαμβάνει χώρα η εξώθερμη καύση του μεθανίου προς CO2 και Η2Ο και ακολουθούν οι αντιδράσεις αναμόρφωσης. Η καύση του μεθανίου ευνοείται σε χαμηλή θερμοκρασία ενώ οι αντιδράσεις αναμόρφωσης ευνοούνται σε υψηλή θερμοκρασία φούρνου. Η μετατροπή μεθανίου και η εκλεκτικότητα προς αέριο σύνθεσης επηρεάζονται τόσο από τη παροχή εισόδου όσο και από τη θερμοκρασία του φούρνου. Σημαντικό παράγοντα για τη διατήρηση της σταθερότητας του καταλύτη αποτελεί το μέγεθος των κόκκων του καταλύτη η διάμετρος των οποίων πρέπει να είναι d<0.09mm. Προκειμένου να γίνει πρόβλεψη των φαινομένων που λαμβάνουν χώρα σε ένα μονολιθικό αντιδραστήρα καταστρώθηκε μοντέλο το οποίο προσομοίαζε το μονολιθικό αντιδραστήρα κατά τη διάρκεια της αντίδρασης. Θεωρήθηκε ότι ο καταλύτης που έχει εναποτεθεί στο μονόλιθο ήταν 0.5% Ru/TiO2 (2%CaO). Μελετήθηκε η επίδραση της αρχικής θερμοκρασίας, της αρχικής πίεσης, της σύστασης και της παροχής εισόδου. Τα αποτελέσματα έδειξαν ότι μπορεί να προβλεφθεί ποιοτικά η ανάπτυξη του θερμοκρασιακού προφίλ κατά μήκος του μονόλιθου, καθώς και οι τιμές της μετατροπής μεθανίου και της εκλεκτικότητας προς CO και Η2. Η παράλειψη κάποιων παραδοχών ώστε να προσεγγίζονται καλύτερα οι πραγματικές συνθήκες λειτουργίας του αντιδραστήρα θα οδηγούσε στην προσέγγιση των πειραματικών αποτελεσμάτων με εκείνα που προβλέπονται από τους υπολογισμούς όχι μόνο ποιοτικά αλλά και ποσοτικά. / The conversion of natural gas to synthesis gas is a very important process because synthesis gas is necessary for the Fischer-Tropsch synthesis and also for the production of methanol. The most common process for producing synthesis gas from natural gas is the steam reforming of methane, since methane is the main component of natural gas. Recently, a significant amount of effort has been directed to processes alternative to steam reforming of methane such as CO2 reforming, catalytic and not catalytic partial oxidation of methane, e.t.c. Among the above mentioned processes partial oxidation of methane seems to be the most attractive alternative to steam reforming of methane for converting natural gas to synthesis gas, owing to a number of advantages. The most important among them are, the small amount of heat that is produced during the reaction compared to the heat that is produced during the combustion of methane to CO2 and H2O, and the ratio of the produced synthesis gas which is CO/H2 : ½ which is exactly the necessary ratio for the Fischer Tropsch and methanol synthesis. The catalytic partial oxidation of methane to synthesis gas has been studied using fixed bed reactors, mainly. Due το the disadvantages of this kind of reactors such as the development of hot spots during the combustion of methane and the high pressure drop, the study of partial oxidation of methane has been made in other kind of reactors such as fluidized bed reactors, honeycomb reactors, auto-thermal reactors, e.t.c. In the present study the catalytic partial oxidation of methane has been studied in a Heat Integrated Wall Reactor (HIWAR) and also in a monolithic reactor. The Heat Integrated Wall Reactor consists of a non-porous ceramic tube of high thermal conductivity in the inner and outer surface of which the catalytic film is deposited. The tube is enclosed in another tube, which can be ceramic or quartz. Because of the way of construction of the reactor the catalyst that is deposited in the inner surface of the ceramic tube can be different from the one that is deposited on the outer surface of the tube. The catalyst that was used for the study of the partial oxidation to synthesis gas in the Heat Integrated Wall Reactor was Rh/Al2O3. The factors the influence of which was studied here were the type of deposition of the catalyst film on the tube, the furnace temperature, the flow rate and the amount of catalyst that was deposited in the inner and outer surface of the tube. The experimental results indicated that there is an optimum amount of catalyst that can be deposited on the inner and outer surface of the reactor in order to obtain a flat temperature profile without the development of hot spots. The range of flow rates that can be used in the HIWAR is bigger than the range that can be used in a conventional reactor, due to the absence of hot spots. Some experimental results also indicated that part of CO is produced via the direct scheme. Partial oxidation of methane to synthesis gas had been studied in a monolithic reactor as well. The main advantages of monolithic reactors are the very low pressure drop and the high surface area. The catalyst that was used during this study was 0.5% Ru/TiO2 (1%CaO), which is a very active and stable catalyst for this reaction. During the experiments that took place in the monolithic reactor the factors the influence of which was studied here were the flow rate, the furnace temperature and the feed composition. The experimental results indicated that the combustion of methane takes place initially which is followed by the endothermic reforming reactions. The combustion of methane to CO2 and H2O is favored in low temperatures while the endothermic reforming reaction is favored in high temperatures. Methane conversion as well as CO and H2 selectivity change as the flow rate and furnace temperatures change. Experimental results indicated that a catalyst grain diameter less than 0,09 mm is necessary for the catalyst to be stable during the reaction. In order to predict the behavior of factors such as temperature, methane conversion and CO and H2 selectivity during the catalytic partial oxidation of methane in a monolithic reactor, a one-dimensional model was developed in order to simulate the monolithic reactor during the reaction. The calculations were based on the rate equations that were developed using experimental results for the 0,5%Ru/TiO2 (1% CaO). The influence of the initial temperature, initial pressure, flow rate and feed composition were studied. Comparison of the results of the simulations against those of the experimental studies indicated that calculations qualitatively predict the behavior of the reactor but not quantitatively, probably due to the assumptions that were introduced during the development of the model, i.e. one dimensionality of the model, uncertainties in the rate expressions etc. Relaxing some of those assumptions in order to approach the real operating conditions of the reactor is expected to enhance quantitative agreement between the experimental and numerical results.

Ces travaux de thèse s’inscrivent dans le cadre d’une collaboration entre les laboratoires G2ELAB et IMEP-LAHC en lien avec le projet BQR WiSiTUDe (Grenoble-INP). Le but de cette thèse concerne la conception, modélisation et caractérisation du gate driver intégré pour transistors de puissance à base d’un transformateur sans noyau pour le transfert isolé d'ordres de commutation. La thèse est composée de deux grandes parties : - Une partie de la conception, la modélisation et la caractérisation du transformateur intégré dans deux technologies CMOS 0.35 µm bulk et CMOS 0.18 µm SOI. - Une partie de la conception, la simulation et la mise en œuvre de deux circuits de commande intégrée dans ces deux technologies. Ainsi, l’aspect du système du convertisseur de puissance sera étudié en proposant une nouvelle conception couplée commande/puissance à faible charge. Les résultats de ce travail de thèse ont permis de valider les approches proposées. Deux modèles fiables (électrique 2D et électromagnétique 3D) du transformateur ont été établis et validés via une réalisation CMOS 0.35 µm standard. De plus, un driver CMOS bulk, intégrant l’ensemble du transformateur sans noyau avec plusieurs fonctions de pilotage de la commande rapprochée a été caractérisé et validé. Finalement, un gate driver générique a été conçu en technologie CMOS SOI, intégrant dans une seule puce les étages de commande éloignée, l’isolation galvanique et la commande rapprochée pour transistors de puissance. Ce gate driver présente nombre d’avantages en termes d’interconnexion, de la consommation de la surface de silicium, de la consommation énergétique du driver et de CEM. Les perspectives du travail de thèse sont multiples, à savoir d’une part l’assemblage 3D entre le gate driver et le composant de puissance et d’autre part les convertisseurs de multi-transistors. / This thesis work focuses on the design, modelling and the implementation of integrated gate drivers for power transistors based on CMOS coreless transformer. The main objectives of thesis are the design, modeling and characterization of coreless transformer in two technologies CMOS 0.35 µm bulk and CMOS 0.18 µm SOI, as well as the design and the characterization of two integrated gate drivers in these two technologies. The results of thesis allow us to validate our proposal models for coreless transformer: 2D electrical model and 3D electromagnetic model. Moreover, one CMOS bulk isolated gate driver which monolithically integrates the coreless transformer, the secondary side control circuit for power transistors has been fabricated and validated for both high side and low side configuration in a Buck converter. Finally, a CMOS SOI isolated gate driver is designed; integrates in one single chip the external control, the coreless transformer and the close gate driver circuit for power transistors. This one-chip solution presents a numerous advantages in term of interconnect parasitic, energy consumption, silicon surface consumption, and EMI with a high level of galvanic isolation. The perspectives of this SOI gate driver are multiple, on the one hand, are the 3D assemblies between gate driver/power transistors and on the other hand, are the multiple-switch converter.

Monolithic Active Pixel Sensors (MAPS) are good candidates to be used in High Energy Physics (HEP) experiments for charged particle detection. In the HEP applications, MAPS chips are placed very close to the interaction point and are directly exposed to harsh environmental radiation. This thesis focuses on the study and improvement of the MAPS radiation hardness. The main radiation effects and the research progress of MAPS are studied firstly. During the study, the SRAM IP cores built in MAPS are found limiting the radiation hardness of the whole MAPS chips. Consequently, in order to improve the radiation hardness of MAPS, three radiation hard memories are designed and evaluated for the HEP experiments. In order to replace the SRAM IP cores, a radiation hard SRAM is developed on a very limited area. For smaller feature size processes, in which the single event upset (SEU) effects get significant, a radiation hard SRAM with enhanced SEU tolerance is implemented by an error detection and correction algorithm and a bit-interleaving storage. In order to obtain higher radiation tolerance and higher circuitry density, a dual-port memory with an original 2-transistor cell is developed and evaluated for future MAPS chips. Finally, the radiation hardness of the MAPS chips using new available processes is studied, and the future works are prospected.

In recent years, scientific studies are carried out to find new refractory material. Having good mechanical properties under very high temperatures, refractories are widely used in industries like iron, steel, glass, cement and pottery. Researches are focused on monolithic refractory making because of their superior properties comparing to conventional firebrick refractories. Providing a mono-block body, having no joints makes the monolithic refractories more durable at elevated temperatures. Easier production and installation are two main points that people are choosing monolithic refractories, thus an economy is made. In this study, for monolithic refractory production, high alumina cement was used as binding material. It is known that the increase in alumina (Al2O3) content increases the high temperature resistance, so that crushed firebrick, having 85% Al2O3 was used as aggregate. Pozzolanic materials, which are silica fume, fly ash, ground granulated firebrick and ground granulated blast furnace slag, were added to improve physical and mechanical properties of mortar. With the addition of steel fibres, change in compressive strength and flexural strength was observed.Superplasticizer was used to understand its behaviour under high temperatures. Portland cement containing mortars were also prepared to make comparison with high alumina cement containing specimens. Specimens were prepared in 5x5x5 cm and 4x4x16 cm prisms. They were cured for one day at curing room, then heated to 105&deg / C and then heated to 1100&deg / C. Weight, size and ultrasound velocity change, compressive strength and flexural strength tests were done to determine physical and mechanical properties of the monolithic refractories, before and after heating. Heated and non-heated specimens were pulverized for microstructural investigation with X-Ray diffraction (XRD) method. Using high alumina cement with 50 &ndash / 60 % granulated blast furnace slag or granulated firebrick, by the weight of cement, and crushed firebrick as aggregate, a satisfactory monolithic refractory material was made. It was observed that, mechanical properties were decreased at the Portland cement used mortars after several times of heating and cooling cycles. Also, it was determined that the microstructure of the high alumina cement containing mortars did not deteriorate much at 1100&deg / C, as long as there was no change observed from the results.

Doppelt gekrümmte Schalen und Gitterschalen sind Tragsysteme, die, obwohl in Ihrer Wirkungsweise grundsätzlich verschieden, besonders für den Einsatz von Textilbeton geeignet sind. Aufgrund ihrer doppelten Krümmung weisen derartige Tragsysteme zahlreiche Besonderheiten hinsichtlich Formfindung, Lastabtrag und Herstellung auf. Anhand eines Ausführungsbeispiels von monolithischen Schirmschalen und Konstruktionsstudien zu Gitterschalen werden die Strukturkonzepte und ihre Auswirkungen auf das Tragverhalten, das Bewehrungskonzept und die Auswirkungen auf die Herstellung sowie die Bauteilfügung erläutert. / Structural systems based on double curved monolithic shells or gridshells are both characterised by their good load bearing behaviour depending on the chosen geometry. Although both systems are different regarding their functionality they are suitable applications for TRC while taking advantage of the main constructive and design characteristics of the material. In describing an execution example of an umbrella-shaped shell-structure and design studies on gridshells the relation between structural concept and load bearing behaviour and the relation between the reinforcement concept and production as well as joining are discussed.