In this thesis, the c-axis-oriented AlN films were deposited on piezoelectric substrates, lithium niobate (LiNbO3), ST-Quartz, and non-piezoelectric substrate, silicon (Si), by reactive rf magnetron sputtering. AlN films were deposited with the nitrogen concentration (N2/Ar+N2) of 20¡ã80%, the chamber pressure of 1¡ã15mTorr, the rf power of 200¡ã450W, the deposition time of 1~3 hours and the substrate temperature of 100¡ã400¢J. The correlation between growth parameters and piezoelectric coefficients will be investigated by XRD¡Bd33 and K2 analysis in this study. The experimental results showed that the values of d33 become larger as the intensity of X-ray is stronger. It can also be concluded that the smaller the FWHM of (002) XRD peak is, the larger the value of d33 is. With various sputtering parameters, the K2 values exhibit diversely. The multilayer structures of AlN/LiNbO3 and AlN/ST-Quartz both make lower values of K2. In general, by combining the higher K2 and d33 values of LiNbO3 and ST-Quartz with high wave velocity of AlN, the high-frequency with high performance SAW devices can be obtained.

In this thesis, we have analyzed the coupling gap dependent micro-ring loss in a single ring all-pass filter configuration using the two dimension (2D) finite difference time domain (FDTD) and EIM (effective index method). We utilized a new analysis scheme by calculating the transmission signal as a function of input wavelength and fitting the transmission spectrum with a phenomenological ring loss parameter. This novel scheme circumvents the complex waveguide mode analysis process, when the coupling gap is narrow and the all-pass coupling region becomes multi-mode. We first find that the radiation loss increases rapidly with decreasing coupling gap width. Our results show that the intrinsic bending losses of silicon micro-rings (on oxide) with the radius of 1.5 £gm, 2.5£gm, and 5£gm are about 20 dB/cm, 3 dB/cm, 1 dB/cm for TE polarization modes, respectively. For TM modes, the intrinsic bending losses with the radius of 1.5£gm, 2.5 £gm, and 5£gm are about 2573 dB/cm, 64 dB/cm, and 0.8 dB/cm, respectively. Next, we find that power coupling coefficients of the single ring all-pass filter configuration using the ring to ring couplers are much higher than the bus to ring couplers. The radiation losses of the ring to ring couplers for TM modes are improved significantly in all coupling gap widths. In a high-density integrated optics circuit, specially designed ring coupling region device structure is needed to address this serious optical loss issue.

Le coefficient de couplage électromécanique (CCEM) est un paramètre essentiel pour la description des matériaux piézoélectriques, il traduit la conversion d’énergie électrique en énergie mécanique et vice versa. Ce coefficient de couplage est étudié et déterminé dans le cadre de cette thèse pour des céramiques piézoélectriques. Ces dernières sont utilisées pour trois structures différentes ; la première structure étudiée est constituée d’une poutre longue et mince avec des petits patchs collés symétriquement sur les deux faces de la poutre en Aluminium, la seconde structure se compose d’une poutre courte et épaisse avec deux grands patchs. La dernière structure étudiée est une plaque composite multicouche du type aéronautique avec un seul grand patch. Ces trois structures ont été étudiées afin de déterminer le coefficient de couplage électromécanique effectif qui est considéré comme un indicateur de performance de l’amortissement passif shunté. Ce coefficient de couplage a été évalué de différentes manières en utilisant différents paramètres dont les conditions limites électriques, les propriétés élastiques des patchs, les propriétés modales de la poutre seule ainsi que les facteurs de couplages piézoélectriques. Une première étude expérimentale a été menée sur la poutre longue pour deux types de configurations en court circuit et circuit ouvert pour identifier ses propriétés modales. La poutre longue a été simulée pour deux types de polarisations, identiques et opposées, et simulée dans les deux codes Ansys® et Abaqus®. L’influence de la condition d’équipotentielle sur le coefficient de couplage a été étudiée. Une seconde campagne expérimentale et numérique sur une autre structure a été nécessaire pour valider les résultats obtenus. Pour pouvoir atteindre cet objectif, il était nécessaire de travailler sur une structure plus courte et plus rigide. Ainsi, la poutre courte a été simulée dans Ansys® et les résultats obtenus ont confirmé la nécessité de prendre en compte l’équipotentialité sur les faces des patchs. Cette condition a pour effet de réduire le couplage électromécanique et parfois de découpler certains modes. L’écart résultant de la corrélation expérimentale / numérique des deux poutres instrumentées a incité à recaler les modèles numériques. Ce recalage peut se présenter sous trois formes : mécanique en remplaçant l’encastrement par des ressorts linéaires, électrique en remplaçant les capacités fournies par le fabricant par les valeurs mesurées expérimentalement et électromécanique en utilisant les deux recalages précédents simultanément. Les deux poutres ont ensuite été simulées en déformations planes et contraintes planes et recalées afin d’approcher les résultats expérimentaux. L’étude de ces deux structures a permis de confronter les différentes méthodes d’évaluation du CCEM effectif, d’évaluer l’influence de l’équipotentialité sur les faces des électrodes et de comparer les simulations bidimensionnelles aux tridimensionnelles. Une plaque composite multicouche du type aéronautique a été ensuite étudiée pour généraliser la méthode d’évaluation du CCEM effectif pour les structures minces composites. La plaque seule a d’abord été simulée dans Ansys® pour valider le modèle numérique. Des tests sur la structure adaptative ont ensuite été menés pour l’évaluation du CCEM expérimental. La position choisie du patch a été déterminée par une analyse de l’énergie de déformation de la plaque seule pour les modes d’intérêt. Cette méthode de placement du patch s’est avérée efficace dans le sens où elle a conduit à des CCEM effectifs élevés pour certains modes de la bande de fréquence retenue. / The electromechanical coupling coefficient (EMCC) is an important parameter for the description of piezoelectric materials; it measures the conversion of electrical energy into mechanical one and vice versa. The coupling coefficient is studied and determined in this dissertation for piezoelectric ceramics. The latter are used for different structures: the first studied one is a long and thin Aluminium beam with small patches bonded symmetrically on its faces, the second one is a short and thick Aluminium beam with symmetrically bonded two large patches, and the third structure is considered more complex because it is an aeronautic-type multilayer composite plate with a single large patch. These three structures were studied to determine the electromechanical coupling coefficient which is considered as a performance indicator for passive shunted damping. The coupling coefficient was evaluated in different ways using different parameters, including the electrical boundary conditions, the elastic properties of the patches, the modal properties of the base beam and the piezoelectric coupling factor. A first experimental study was conducted on the long beam for two configurations, short circuit and open circuit, to identify its modal properties. The long beam was simulated for two configurations of polarization, same and opposite, in Ansys® and Abaqus® commercial codes. The equipotential condition influence on the coupling coefficient has been studied. A second experimental and numerical campaign for a different structure was necessary to validate the obtained results. For this purpose, it was necessary to work on a shorter and more stiff structure. Thus, the short beam was simulated in Ansys® which results have confirmed the necessity to consider the equipotentiality of the patches faces. This condition was found to reduce the electromechanical coupling and to uncouple some modes. The difference between experimental and numerical results of both adaptive structures was reduced by updating the numerical models. This updating is made in three ways: mechanically, by replacing the theoretical clamp conditions by linear springs, electrically, by replacing the capacities provided by the supplier by the experimental measured values, and electromechanically by considering previous updatings simultaneously. Both beams were simulated in 2D plane-strain and plane-stress and updated in order to approximate the experimental results. The study of these two structures allowed to assess different methods for the evaluation of the EMCC, to evaluate the influence of the equipotentiality constraints on the electroded faces, and to compare two-dimensional simulations to three-dimensional ones. Finally, an aeronautic-type multilayer plate composite has been studied in order to generalize the evaluation method of the EMCC for thin composite structures. The base plate was first simulated in Ansys® in order to validate the numerical model, then tests of the adaptive plate were conducted in order to evaluate the experimental EMCC. The selected position of the patch results from a strain energy analysis of the base plate for the mode of interest. The patch placement method was efficient in the sense that it provided high EMCC for some modes in the retained frequency range.

Nesse trabalho é apresentada uma estratégia de controle para manter a potência entregue a carga do link indutivo de um sistema de transferência de potência sem fios maximizada continuamente. Inicialmente, foi desenvolvido o modelo do link indutivo do sistema, de forma a encontrar uma expressão para a potência na carga. Em seguida, essa expressão foi analisada, buscando os pontos de potência máxima correspondentes às frequências ressonantes quando o fator de acoplamento magnético do link indutivo era variado. Assim, foi estabelecido um método para estimar esse fator de acoplamento k e consequentemente da indutância mútua M a partir do monitoramento direto da fase e do módulo da corrente de entrada do link e indireto da impedância de entrada. Portanto, a estratégia de controle proposta consiste em monitorar a corrente da fonte de entrada do link para estimar o fator de acoplamento magnético k e finalmente, sintonizar a fonte na frequência de ressonância associada ao coeficiente de acoplamento magnético pela função obtida fres(k). Os resultados experimentais mostraram que a estratégia proposta aumenta significativamente a potência entregue a carga em comparação a uma fonte de alimentação sintonizada em uma frequência fixa. / A control strategy is presented in this work to maintain the power delivered to a load continuously maximized at the output of a wireless power transfer system. Initially, the inductive link model of the system was developed in order to find an expression for the power delivered to the load. Thus, the maximum power delivered to the load points were obtained experimentally while the magnetic coupling factor of the inductive link was varied and associated with the frequency of the source. The frequency and magnetic coupling variables were related to the maximum power on the load by a simple analytic function. Thus, a method has been established to estimate the coupling factor k and consequently the mutual inductance M by direct monitoring of the phase and the module of the input current of the link and indirect monitoring of the input impedance. Therefore, the proposed control strategy is to monitor the current of the input source of the link to estimate the magnetic coupling factor k and finally to tune the source at the resonance frequencies associated to the magnetic coupling coefficient by the obtained function fres(k). The experimental results showed that the proposed strategy increases significantly the power delivered to the load compared to a power source tuned at a fixed frequency.

Abstract Compared with the other micro acoustic wave devices, the flexural plate-wave (FPW) device is more suitable for being used in liquid-sensing applications due to its higher mass sensitivity, lower phase velocity and lower operation frequency. However, conventional FPW devices usually present a high insertion loss and low fabrication yield. To reduce the insertion loss and enhance the fabrication yield of FPW device, a 1.5 £gm-thick silicon-trench reflective grating structure (RGS), a high electromechanical coupling coefficient ZnO thin-film and a 5 £gm-thick silicon oxide membrane substrate are adopted in this research. The influences of the amount of silicon trench and the distance between inter-digital transducer (IDT) and RGS on the insertion loss and quality factor of FPW device are investigated. The main fabrication technology adopted in the study is bulk micromachining technology and the main fabrication steps include six thin-film deposition and five photolithography processes. Under the optimized conditions of the sputtering deposition processes (200¢J substrate temperature, 200 W radio-frequency power and 75% gas flow ratio), a high C-axis (002) orientation ZnO piezoelectric thin-film with 31.33% electromechanical coupling coefficient can be demonstrated. The peak of XRD intensity of the standard ZnO film occurs at diffraction angle 2£c = 34.422¢X, which matches well with our results (2£c = 34.282¢X). By controlling the thickness of ZnO/Au/Cr/SiO2/Si3N4 sensing membrane less than 6.5 £gm-thick, the fabrication yield of FPW device can be improved and a low operation frequency (6.286 MHz) and high mass sensitivity (-113.63 cm2 / g) can be achieved. In addition, as the implemented FPW device with four silicon trenches RGS and 37.5 £gm distance between IDT and RGS, a low insertion loss (-40.854 dB) and very high quality factor (Q=206) can be obtained. Keywords¡Gflexural plate-wave; silicon-trench reflective grating structure; electromechanical coupling coefficient; ZnO; bulk micromachining technology

Nesse trabalho é apresentada uma estratégia de controle para manter a potência entregue a carga do link indutivo de um sistema de transferência de potência sem fios maximizada continuamente. Inicialmente, foi desenvolvido o modelo do link indutivo do sistema, de forma a encontrar uma expressão para a potência na carga. Em seguida, essa expressão foi analisada, buscando os pontos de potência máxima correspondentes às frequências ressonantes quando o fator de acoplamento magnético do link indutivo era variado. Assim, foi estabelecido um método para estimar esse fator de acoplamento k e consequentemente da indutância mútua M a partir do monitoramento direto da fase e do módulo da corrente de entrada do link e indireto da impedância de entrada. Portanto, a estratégia de controle proposta consiste em monitorar a corrente da fonte de entrada do link para estimar o fator de acoplamento magnético k e finalmente, sintonizar a fonte na frequência de ressonância associada ao coeficiente de acoplamento magnético pela função obtida fres(k). Os resultados experimentais mostraram que a estratégia proposta aumenta significativamente a potência entregue a carga em comparação a uma fonte de alimentação sintonizada em uma frequência fixa. / A control strategy is presented in this work to maintain the power delivered to a load continuously maximized at the output of a wireless power transfer system. Initially, the inductive link model of the system was developed in order to find an expression for the power delivered to the load. Thus, the maximum power delivered to the load points were obtained experimentally while the magnetic coupling factor of the inductive link was varied and associated with the frequency of the source. The frequency and magnetic coupling variables were related to the maximum power on the load by a simple analytic function. Thus, a method has been established to estimate the coupling factor k and consequently the mutual inductance M by direct monitoring of the phase and the module of the input current of the link and indirect monitoring of the input impedance. Therefore, the proposed control strategy is to monitor the current of the input source of the link to estimate the magnetic coupling factor k and finally to tune the source at the resonance frequencies associated to the magnetic coupling coefficient by the obtained function fres(k). The experimental results showed that the proposed strategy increases significantly the power delivered to the load compared to a power source tuned at a fixed frequency.

Nesse trabalho é apresentada uma estratégia de controle para manter a potência entregue a carga do link indutivo de um sistema de transferência de potência sem fios maximizada continuamente. Inicialmente, foi desenvolvido o modelo do link indutivo do sistema, de forma a encontrar uma expressão para a potência na carga. Em seguida, essa expressão foi analisada, buscando os pontos de potência máxima correspondentes às frequências ressonantes quando o fator de acoplamento magnético do link indutivo era variado. Assim, foi estabelecido um método para estimar esse fator de acoplamento k e consequentemente da indutância mútua M a partir do monitoramento direto da fase e do módulo da corrente de entrada do link e indireto da impedância de entrada. Portanto, a estratégia de controle proposta consiste em monitorar a corrente da fonte de entrada do link para estimar o fator de acoplamento magnético k e finalmente, sintonizar a fonte na frequência de ressonância associada ao coeficiente de acoplamento magnético pela função obtida fres(k). Os resultados experimentais mostraram que a estratégia proposta aumenta significativamente a potência entregue a carga em comparação a uma fonte de alimentação sintonizada em uma frequência fixa. / A control strategy is presented in this work to maintain the power delivered to a load continuously maximized at the output of a wireless power transfer system. Initially, the inductive link model of the system was developed in order to find an expression for the power delivered to the load. Thus, the maximum power delivered to the load points were obtained experimentally while the magnetic coupling factor of the inductive link was varied and associated with the frequency of the source. The frequency and magnetic coupling variables were related to the maximum power on the load by a simple analytic function. Thus, a method has been established to estimate the coupling factor k and consequently the mutual inductance M by direct monitoring of the phase and the module of the input current of the link and indirect monitoring of the input impedance. Therefore, the proposed control strategy is to monitor the current of the input source of the link to estimate the magnetic coupling factor k and finally to tune the source at the resonance frequencies associated to the magnetic coupling coefficient by the obtained function fres(k). The experimental results showed that the proposed strategy increases significantly the power delivered to the load compared to a power source tuned at a fixed frequency.

Une recherche bibliographique a permis de faire un état de l'art, d'une part des différents principespermettant de détecter le givre et d'autre part de l'application de matériaux magnétostrictifs à la mesure defaibles masses. La faisabilité et l’applicabilité de certains alliages amorphes et nanocristallins à la réalisationde capteurs de givre ont été démontrées.Suite à cet état de l'art, le principe de détection par résonateur magnétostrictif qui a été retenu, a faitl'objet d'une étude approfondie aboutissant à l'établissement d'un nouveau modèle analytique complet desrésonateurs magnétostrictifs sous forme de ruban. L'intérêt principal de ce modèle est la prise en compte depertes mécaniques et l'expression finale contenant la fonction de transfert complète entre les entrées et sortiesélectriques. Ainsi, la réponse en fréquence du modèle est vraiment similaire aux réponses expérimentales. Deplus, il apporte une explication aux retournements observés, pour certains harmoniques entre les fréquencesde résonance et d'antirésonance, pour lesquels l’origine n’avait pu être clairement identifiée.La mise en œuvre de ce nouveau modèle pour la caractérisation de rubans amorphes a montré qu'ilest possible de déterminer le coefficient de couplage à partir de la réponse en fréquence ainsi que le moduled'Young ou l'amortissement. On a donc pu tracer pour des rubans de 2605SC et 2826MB les évolutions avecle champ de polarisation des paramètres du résonateur. De l'évolution de ces paramètres, il est possible dereconstruire les courbes de magnétostriction d'un ruban, ce qui constitue actuellement une méthode decaractérisation.La mise en œuvre d'un prototype a permis de détecter du givre et ainsi de confirmer la faisabilité dela détection de givrage par ce moyen, mais les essais préliminaires qui sont très prometteurs n’ont puapporter une réponse quantitative du capteur soumis à un dépôt de givre. Des mesures utilisant des enceintesde givrage contrôlées en température et en hygrométrie constituent une première perspective. Par ailleurs, lesrésultats présentés dans ce manuscrit s’avèrent une bonne base pour la valorisation de ces travaux,notamment pour la réalisation d’un prototype de capteur, son conditionnement et sa mise en œuvre dans devraies conditions de givrage. / A review of literature allowed first to a state of the art of the different routes to detect the ice on theone hand and of the use of magnetostrictive materials to measure small masses. Both the feasibility andapplicability of metglasses and nanocrystalline alloys to frost resonating sensors have been clearlyestablished.The principle of magnetostrictive sensing resonator was developed and a new comprehensiveanalytical model of ribbonshaped magnetostrictive resonators was successfully achieved. The present modelincludes mechanical losses and the final expression contains the complete transfer function between the inputand output powers. Thus, the predicted frequency response of the model fairly fits to the experimental one.In addition, it clearly explains henceforth the reversals of some harmonic frequencies which occur from theresonance and antiresonance phenomena.The implementation of this new model for the characterization of amorphous ribbons showed thusthat it is now possible to well estimate the coupling coefficient from the frequency response and the Young'smodulus or damping. So we could draw developments with the bias field parameters of the resonator for2826MB and 2605SC amorphous ribbons. The evolution of these parameters allows the magnetostrictioncurves to be well established, that consists in a new characterization method.The implementation of a first prototype permits to detect frost, confirming thus the feasibility ofdetecting icing this way, but these promising preliminary tests could not bring quantitative data to model thefrost deposition on a resonating sensor. Measurements using pregnant icing controlled temperature andhumidity are a first perspective. Finally, it is important to emphasize that the present results appear as a goodsupport for the valorization of this study, especially to elaborate of a prototype sensor based onmagnetostrictive ribbons, its packaging and its implementation in in situ icing conditions.

Acoustic micro-sensors have already been applied in mass sensing including surface acoustic wave (SAW), flexural plate wave (FPW), thickness shear mode (TSM) and shear horizontal acoustic plate mode (SH-APM). The FPW micro-sensor is very suitable for liquid-sensing and bio-sensing applications due to the high mass-sensitivity and low phase-velocity in liquid. However, the conventional FPW micro-sensors presented a high insertion-loss (IL) and a low signal-to-noise ratio so it is difficult to combine with IC into a micro-system. To overcome these drawbacks, this study combine the Microelectromechanical System (MEMS) technology and the high C-axis orientation ZnO piezoelectric thin-film to develop a low insertion loss, low operation frequency, and high electromechanical coupling coefficient FPW device. In this study, a high C-axis orientation ZnO piezoelectric thin-film with a 20944A.U. X-Ray diffraction intensity at 34.200 degree and a 0.573 degree full width at half maximum (FWHM) was deposited by a commercial magnetic radio-frequency (RF) sputter system. The total processes of the FPW micro-sensor included five photolithography and seven thin-film depositions. In this study a low operation frequency (0.1MHz), low insertion loss (11dB to 14dB) and high electromechanical coupling coefficient (11%) FPW sensor was developed and fabricated.

Nesta dissertação é apresentado o estudo dos transistores de porta flutuante (Floating Gate Transistor - FG Transistor), sua modelagem, e a análise do efeito da dose de ionização total (Total Ionizing Dose- TID) sobre os transistores FG. Para isto foi procurado e implementado um modelo de simulação elétrica do transistor FG em condições de leitura (análise DC), baseado no cálculo quantitativo da tensão na porta flutuante em função das tensões nos terminais do transistor, no valor de carga armazenado na porta flutuante e nos coeficientes de acoplamento capacitivo que apresentam este tipo de dispositivos. Para a análise do efeito TID, a tensão limiar do transistor MOS foi variada usando o método de simulação Monte Carlo, tendo em conta as variações da tensão limiar que apresentam os transistores FG submetidos na radiação ionizante. O estudo obteve como resultado a confirmação da perda de carga do FG à medida que é incrementada a dose de radiação, o que implica uma alteração na característica de retenção de carga que caracteriza as células de memórias não voláteis (Non Volatile Memory - NVM). / In this dissertation work, a study of the the floating gate Transistor (FG transistor) performed. The focus in the electrical modeling, and the analysis of the impact of the Total Ionizing Dose (TID) on the electrical performance of the device. Aiming electrical level simulation, different electric simulation models for the FG transistor in read conditions (DC analysis) were evaluated and the model best suited for implementation into the simulation tool was selected. The selected model is based on Floating Gate voltage calculation as a function of polarization voltage of the FG transistor terminals, the stored charge value in the Floating Gate and the capacitive coupling coefficient presented by this device. For the TID analysis the threshold voltage of the MOS transistor was shifted by means of a Monte Carlo simulation method, considering the threshold voltage variations when the FG transistor is subjected to the ionizing radiation.The analysis lead to the confirmation that the loss charge stored in the FG increases with the radiation dose, affecting the retention characteristics of the memory cells.