Dans ce mémoire, une étude des conditions de contrôles d’écoulements aérodynamiques par des réseaux de générateurs de tourbillons fluidiques pulsés est menée pour établir un cahier des charges des micro-actionneurs instrumentés de faible coût, indispensables à la réalisation de ce type de contrôle actif à l’échelle industrielle. Une discussion des problématiques rencontrées dans la mise en place de ces dispositifs permet de définir des solutions techniques pertinentes. Une micro-valve encapsulée, constituée d’un canal micro-fluidique en silicium dont l’ouverture est contrôlée par un résonateur annulaire en PDMS, actionnée par différents dispositifs macroscopiques magnétiques, est alors modélisée, fabriquée et caractérisée. Le dispositif permet de générer des jets d’air pulsés complètement contrôlés jusqu’à des vitesses de 150m/s sur la gamme de fréquence [0 ; 500 Hz]. Des réseaux de ces micro-actionneurs polyvalents sont ensuite utilisés en soufflerie pour démontrer sur différents profils aérodynamiques classiques l’intérêt du contrôle par jet pulsé. Le recollement du flux d’air décollé est obtenu sur chacune de ces maquettes pour des conditions réalistes et avec un rendement fluidique supérieur à celui des jets continus.Un débitmètre massique composé d’un capteur de température, d’un capteur de frottement pariétal et d’un capteur de pression de type Pirani, réalisés dans le même procédé de fabrication, est intégré au micro-actionneur. Il permet de caractériser in-situ les jets d’air produits.Enfin un prototype répondant complètement au cahier des charges industriel est obtenu. Sa taille est minimisée par l’optimisation de l’actionneur grâce à un algorithme génétique / This thesis starts with a study of aerodynamic flow control conditions by arrays of pulsed fluidic vortex generators. Detailed specifications are synthesised for the conception of low cost, sensors equipped, micro-actuators required to manage industrial scale active flow control experiments. Devices implementation is discussed to define relevant technical solutions.A packaged micro valve is modelled, fabricated and characterized. It is composed of a micro fluidic channel modulated by an annular membrane resonator made of PDMS. The membrane is controlled by different kinds of magnetic actuation. Fully controlled pulsed air jets are obtained in the frequency range [0; 500 Hz] with velocities up to 150 m/s. Arrays of these micro actuators are used in wind tunnel experiments to demonstrate the ability of pulsed jet to manage control on a wide range of classical separated flows. Reattachment is achieved under industrial flow conditions with improved fluidic yield compared to continuous jets.A mass flow meter constituted of a thermal sensor, a shear flow sensor and a Pirani pressure sensor, all of them built in a single fabrication step, is integrated in the micro-actuator. It allows in-situ characterization of the produced air jets.Finally, a fully satisfying demonstrator is obtained. Its sized is minimized through the use of a genetic algorithm

Now-a-days sensors are not limited only to industry or research laboratories but have come to common man’s usage. From kids toys to house hold equipment like washing machine, microwave oven as well as in automobiles, a wide variety of sensors and actuators can be easily seen. The aim of the present thesis work is to discuss the design, development, fabrication and testing of miniaturized piezoresistive, absolute type, low pressure sensor and flow sensor. Detailed performance study of these sensors in different ambient conditions (including harsh environment such as radiation, temperature etc.) has been reported. Extensive study on designing of thin silicon diaphragms and optimization of piezoresistor parameters is presented. Various experiments have been performed to optimize the fabrication and packaging processes. In the present work, two low range absolute type pressure sensors (0-0.5 bar and 0-1 bar) and a novel flow sensor (0-0.1 L min-1) for gas flow rate measurement are developed. The thesis is divided into following six chapters. Chapter 1: It gives a general introduction about miniaturization, MEMS technology and its applications in sensors area. A brief overview of different micromachining techniques is presented, giving their relative advantages and limitations. Literature survey of various types of MEMS based pressure sensors along with recent developments is presented. At the end, the motivation for the present work and organization of the thesis is discussed. Chapter 2: In this chapter, various design aspects of low, absolute type pressure sensors (0-0.5 bar and 0-1 bar) are discussed in detail. Static analysis of the silicon diaphragms has been carried out both analytically as well as through finite element simulations. Piezoresistive analysis is carried out to optimize the piezoresistor dimensions and locations for maximum sensitivity and minimum nonlinearity. All the Finite Element Analyses (FEA) were carried out using Coventorware software. A novel approach for the selection of resistor parameters (sheet resistance, length to width ratio) is reported . Finally, the expected performance of the designed sensors is summarized. Chapter 3: This chapter is divided into two parts. The first part presents the fabrication process flow adopted to develop these low range absolute pressure sensors. Two fabrication process approaches (wet etching and dry etching) which are used to fabricate the thin diaphragms are discussed in detail. Following an overall description, various aspects of the fabrication are elaborated on, like mask design, photolithography process, ion-implantation, bulk micromachining and wafer bonding. The required parameters for implantation doses, annealing cycles, low stress nitride deposition and anodic bonding are optimized through extensive experimental trials. The second part of this chapter discusses about the different levels of packaging involved in the realization of pressure sensors. Finite Element Analyses (FEA) of Level -0 and Level-1 packages has been carried out using ANSYS software to optimize the packaging materials. Exhaustive experimental studies on the selection of die attach materials and their characterization is carried out. Based upon these studies, the glass thickness and die-attach materials are selected. Chapter 4: The chapter discusses the measurement of the fabricated devices. The wafer level characterization which includes I-V characterization, measurement of offset and full scale output is discussed first. And then the temperature coefficient of resistance and offset is measured at wafer level itself. The performance characteristics like sensitivity, nonlinearity, hysteresis and offset of packaged pressure sensors is presented for all the variants (0.5 bar and 1 bar sensors fabricated by KOH and DRIE process) and their comparison with simulated values shows a close match. The measurement of dynamic characteristics using in-house developed test set-up are presented. The next section discussed detailed study about the stability of the developed sensors. The last part of this chapter reports the harsh environment characterization of the sensors viz. high temperature, humidity exposure, radiation testing etc. Chapter 5: The development of a novel micro-orifice based flow sensor for the flow rate measurement in the range of L min-1 is presented in this chapter. The sensing element is a thin silicon diaphragm having four piezoresistors at the edges. A detailed theoretical analysis showing the relationship between output voltage generated and flow rate has been discussed. The flow sensor is calibrated using an in-house developed testing set-up. Novelty of the design is that the differential pressure is measured at the orifice plate itself without the need of two pressure sensors or u-tube which is required otherwise. Chapter 6: This chapter summarizes the salient features of the work presented in this thesis with the conclusion. And then the scope for carrying out the further work is discussed.

This thesis is divided into two parts; the first part presents results on the effect of the flow of fluids and electrochemical bias on single walled carbon nanotubes (SWNT). Issues pertaining to the entry of water into the cylindrical pores of the SWNT and its freezing dynamics have also been addressed in the first part of the thesis. The second part of the thesis deals with Raman scattering studies of iron perovskite namely CaFeO3 and La0.33Sr0.67FeO3 across their charge-disproportionation transition temperatures. PART 1 Chapter 1: Introduction This chapter presents an overview of the systems studied in this thesis, i.e., (i) SWNT and (ii) iron perovskite’s containing iron in Fe4+(d4) state, namely CaFeO3 and La0.33Sr0.67FeO3. It also contains an introduction to the two spectroscopic techniques used in the present thesis, namely Raman scattering and Nuclear Magnetic Resonance. A quantum mechanical picture of Raman scattering, in general, and resonance Raman scattering in particular along with a brief introduction to the apparatus used both for the micro Raman and the low temperature experiments is presented in this chapter. A general introduction to Nuclear Magnetic Resonance (NMR) is also given with an emphasis on various interactions leading to the broadening of the NMR absorption linewidths. Chapter 2: Carbon nanotube liquid flow sensors This chapter presents experimental results and theoretical understandings of the generation of electrical signals by flowing polar/ionic liquids over a mat of SWNT. We first present experimental findings that the flow of a variety of liquids on SWNT bundles induces an electrical signal (voltage/current) in the sample along the direction of the flow. The electrical response is found to be a logarithmic function of the flow speed over a wide range. The magnitude of the signal generated depends sensitively on the ionic conductivity and the polar nature of the liquid and weakly on the viscosity of the liquid. Furthermore its direction can be controlled by electrochemical biasing of the nanotubes. The ratio of the open circuit voltage to the short circuit current is found to be governed by the nanotube resistance. These experimental findings are inconsistent with the conventional idea of a streaming potential as the possible cause. Our measurements suggest that the dominant mechanism responsible for this highly sub-linear response should involve a direct forcing of the free charge carriers in the nanotubes by the fluctuating Coulombic field of the liquid flowing past it. Two alternative understandings of the experimental findings are also presented in this chapter. The first mechanism invokes the idea of a “pulsating ratchet” whereby the charge carriers in the nanotubes experience an asymmetric spatial bias because of the shear-induced deformation of the ion-plus-polar atmosphere at the liquid-solid interface temporally modulated by the liquid flow. In addition, we also propose that experimental findings can be understood qualitatively in terms of three interrelated ideas: (a) Induced friction: The fluctuating charge density of the ions close to the nanotube couples coulombically to the charge carriers in the nanotube and, therefore, offers a friction to the motion of these charge carriers (in addition to the Ohmic friction intrinsic to the carbon nanotubes); (b) Flow-induced drag: In virtue of the above frictional coupling, an imposed liquid flow drags the charge carriers along through the nanotube; (c) Reduction of induced friction at high flow speeds: The space-time correlated Coulombic fluctuations, inherent to the liquid electrolyte, are advected by the liquid flow, and thus get Galilean boosted (Doppler shifted) as seen in the mean rest frame of the drifting carriers in the nanotube. This would cause a reduction of the frictional grip to the motion of the charge carriers in the nanotube with increasing flow speed resulting in a sublinear dependence of the charge drift-velocity (electrical response) on the liquid flow speed. With the above in mind, a quantitative derivation of these frictional effects, first from a heuristic argument, and then analytically from a Langevin-equation treatment have been presented. Chapter 3: Direct generation of voltage and current by gas flow over carbon nanotubes and semiconductors Having obtained experimental evidence of the generation of liquid flow induced electrical signals over single-walled carbon nanotubes, it was only natural to look for the same effect by flowing gases over nanotubes. We show here a direct generation of measurable voltages and currents when gas flows at modest speeds of a few meters per second over single-walled carbon nanotubes . Interestingly, unlike the previous effect (generation of voltages by flow of liquids over single-walled carbon nanotubes), this effect is not specific to single-walled carbon nanotubes and occurs for a wide variety of solids, including single and multi-walled carbon nanotubes, doped semiconductors and metals. Moreover, the gas flow induced signals depend quadratically on the gas flow velocities. This is in sharp contrast to the results obtained by flowing liquids over single-walled carbon nanotubes where the liquid flow generated signal was found to be logarithmically dependent on the flow velocities. In this chapter we provide evidence that the underlying mechanism for the gas flow generated electrical signal is an interplay of Bernoulli’s principle and the Seebeck effect: Pressure difference along streamlines gives rise to temperature difference across the sample which, in turn, produces the measured voltage. Chapter 4: Water at nanoscale confined in single-walled carbon nanotubes studied by NMR In this chapter, we seek experimental evidence of the occupancy of water in the cylindrical pores of the nanotubes. Proton NMR studies have been carried out as a function of temperature from 210 K to 300 K of water confined within SWNT. The NMR lineshape at and below the freezing point of bulk water is asymmetric which can be decomposed into a sum of two Lorentzians. The intensities of both the components decrease with lowering of temperature below 273 K, one component L1 vanishing below 242 K and the other component L2 below 217 K. Following the simulations of Koga et al. (Nature, 412, 802, 2001) showing that the radial density profile of confined water in SWNT has a distribution peak at the center which disappears below the freezing temperature, the L1 component is associated with the protons of the water molecules at the center and the L2 component is associated with protons of water molecules associated at a distance ~ 3Å away from the walls of the nanotubes. In this scenario the complete freezing of the water at ~ 212 K is preceded by the withdrawal of the water molecules from the center of the nanotubes. Chapter 5: Electrochemical tuning of band structure of single walled carbon nanotubes probed by in-situ resonance Raman scattering The work presented in this chapter is motivated by the experimental observation that SWNT have excellent actuating properties, i.e, porous sheets of carbon nanotubes were shown to suffer length changes when subjected to electrochemical bias, with action observed up to 1 KHz. The fast response of the nanotube actuator rules out any mechanism related to the intercalation of ions as is applicable to the case of the polymer actuators. This chapter presents results of in-situ resonance Raman scattering of SWNT investigated under electrochemical biasing. The experimental results show that the intensity of the radial breathing mode varies significantly in a non-monotonic manner as a function of the cathodic bias voltage, but does not change appreciably under anodic bias. The tangential mode is, however, not affected. These results can be quantitatively understood in terms of the changes in the energy gaps between the one-dimensional van Hove singularities in the electron density of states arising possibly due to the alterations in the overlap integral of π-bonds between the p-orbitals of adjacent carbon atoms. This chapter also contains results from ab-initio restricted Hartree Fock calculations in a simplistic geometry where a nanotube is surrounded by two concentric rings of ions. The ab-initio calculation results suggest that the dominant contribution to the strain developed in the nanotubes originates from the electrostatic interactions between the ions and the delocalized π electrons as compared to the doping effects. PART 2 Chapter 6: Raman scattering in CaFeO3 and La0.33Sr0.67FeO3 across the charge disproportionation phase transition Temperature dependent micro-Raman studies of orthorhombic CaFeO3 and rhombohedral La0.33Sr0.67FeO3 were carried out with an aim to study the role of phonons in the formation of the charge disproportioned state (Fe4+ → Fe5++Fe3+) below the transition temperature (Tco) of 290 K and 200 K, respectively. Shell model lattice dynamics calculations were performed for CaFeO3 to assign the Raman modes and determine their vibrational pattern. Temperature dependence of the peak positions and peak widths of various modes for both the systems show distinct changes across their respective transition temperatures. In CaFeO3 the symmetric stretching mode at 707 cm−1 splits into two modes, 707 cm−1 and 684 cm−1 . Interestingly, the 707 (684) cm−1 mode appears only in HH(HV) polarization. In comparison, the Raman band at 704 cm−1 in La0.33Sr0.67FeO3 which has been assigned to the Raman forbidden symmetric stretching mode, disappears below Tco. In addition, two modes at 307 cm−1 and 380 cm−1 of La0.33Sr0.67FeO3 approach each other at Tco. Our experiments show that for both the systems, CaFeO3 and La0.33Sr0.67FeO3, the lattice distortion changes across Tco. Chapter 7: Summary and future outlook The last chapter summarizes our main findings reported in the thesis. It also contains possible future studies which are worth pursuing to add further insights in the issues addressed.

This diploma thesis is divided into two parts - theoretical and practical. In its first, theoretical part, deals with the theory of fluid and gas flow, heat transfer and diversification of sensors for flow measurement working on the electrical principle. It also deals with thermodynamic principle, which can be used for measuring very small flow and low-temperature ceramics that is used to implement microcanals for sensing very low flows. The practical part of the thesis deals with the very simulation of the entire structure in the program “COMSOL Multiphysics” - both in 2D and 3D views. Then there is shown the implementation and measurement of the flow sensor in a low-temperature ceramics, working on a thermodynamic principle.