MEMS (Micro-Electro-Mechanical Systems) é um microsistema invasivo, intermediador e interativo que se desenvolve de forma inteligente, versátil e eficiente. Entretanto, a interatividade, característica que o torna altamente atrativo e suas qualidades de leveza, invisibilidade, economia quanto a consumo de energia, robustez e alta confiabilidade são foco de investigação. Através da obtenção dos parâmetros característicos muitos desses aspectos podem ser otimizados. Conseqüentemente, este trabalho se propõe obter os parâmetros característicos necessários ao modelo matemático de atuadores MEMS, baseados em deformação elástica e dinâmica combdrive, de forma a descrever com precisão o comportamento linear destes em nível de sistema. Os parâmetros característicos de MEMS podem ser extraídos no próprio simulador, ou identificados através da manipulação dos dados dos sinais de entrada e saída obtidos na execução de testes modais sobre o protótipo. Quando a identificação é determinística, utiliza sinais de excitação que obedecem a uma boa relação sinal-ruído (SNR-Signal-Noise Rate). Quando a identificação é estocástica utiliza sinais de excitação misturados com ruído. Essas duas formas de identificação podem ser interpretadas como os dois extremos de identificação. A rigor, qualquer procedimento que não esteja em nenhum desses extremos pode ser denominado de identificação “caixa-cinza”. Dessa forma, a proposta deste trabalho investigativo consiste em utilizar a identificação “caixa cinza” para obter os parâmetros característicos dos atuadores eletromecânicos MEMS combinando as vantagens dos procedimentos determinísticos e estocásticos. Sob este propósito, foi feita revisão das propriedades da matéria, conceituação de atuadores, compreensão da visão bottom-up e, finalmente, estudo de modelos estocásticos com entradas exógenas ARX (Autoregressive with Exogenous Inputs) e uso de estimadores recursivos, Mínimos Quadrado e Variável Instrumental. A comparação dos resultados do modelo determinístico produzido através de FEM/BEM permite testar o desempenho entre dois modelos de índoles diferentes. Os resultados obtidos após a coleta de dados, a escolha da representação matemática, a determinação da estrutura do modelo, a estimação dos parâmetros e a validação do modelo das três tipologias de atuadores desenvolvidos: pontes simples, ponte dupla e dobradiça dupla permitem identificar os parâmetros característicos com erro quadrático médio menor a 1% e validar esses parâmetros num período não maior a 0,5s. Os resultados se mostram altamente satisfatórios, tornando este trabalho uma contribuição científica à síntese de MEMS em nível de sistemas. / MEMS (Micro Electro-Mechanical Systems) is an invasive, intermediator and interactive small size system that develops in an intelligent, versatile and efficient way. However, the interactivity, feature that makes the system highly attractive and its qualities of lightness, invisibility, economy with regard to power consumption, robustness and high reliability are the focus of research. By obtaining the characteristic parameters many of these aspects can be optimized. Therefore, this study proposes to obtain the characteristic parameters necessary for the mathematical model for MEMS actuators, based on elastic deformation and dynamic comb-drive in order to accurately describe the linear behavior in level system. The characteristic parameters of MEMS can be extracted in the own simulator or identified through the manipulation of input and output data signals obtained in the execution of modal tests on the prototype. When the identification is deterministic, it uses the excitation signals that follow a good signal noise rate (SNR). When the identification is stochastic it uses excitation signals mixed with noise. These two forms of identification can be interpreted as the two extremes of the identification. Strictly speaking, any procedure that is not in any of these extremes may be called the "gray-box" identification. Thus, the propose of this research work consists of using the “gray-box” identification to obtain the characteristic parameters of the MEMS electro-mechanical actuators combining the advantages of the deterministic and stochastic procedures. Under this purpose, it was made revision of the matter features, conceptualization of the actuators, comprehension of the bottom-up vision and, finally, study of the stochastic models with autoregressive exogenous inputs (ARX) and the use of recursive estimators, Least Square and Instrumental Variable. The comparison of results of the deterministic model generated by FEM / BEM, allows testing the performance between two models of different kinds. The results obtained after the data collection, the choice of mathematical representation, the determination of the structure of the model, the estimation of the parameters and validation of the model of three actuators topologies developed (simple bridges, double bridge and double hinge) that permit to identify the parameters with a average quadratic error minor than 1% and to validate these parameters in a period not more than 0.5s. The results show highly satisfactory, becoming this work a scientific contribution to MEMS synthesis at system levels.

abstract: This dissertation presents my work on development of deformable electronics using microelectromechanical systems (MEMS) based fabrication technologies. In recent years, deformable electronics are coming to revolutionize the functionality of microelectronics seamlessly with their application environment, ranging from various consumer electronics to bio-medical applications. Many researchers have studied this area, and a wide variety of devices have been fabricated. One traditional way is to directly fabricate electronic devices on flexible substrate through low-temperature processes. These devices suffered from constrained functionality due to the temperature limit. Another transfer printing approach has been developed recently. The general idea is to fabricate functional devices on hard and planar substrates using standard processes then transferred by elastomeric stamps and printed on desired flexible and stretchable substrates. The main disadvantages are that the transfer printing step may limit the yield. The third method is "flexible skins" which silicon substrates are thinned down and structured into islands and sandwiched by two layers of polymer. The main advantage of this method is post CMOS compatible. Based on this technology, we successfully fabricated a 3-D flexible thermal sensor for intravascular flow monitoring. The final product of the 3-D sensor has three independent sensing elements equally distributed around the wall of catheter (1.2 mm in diameter) with 120° spacing. This structure introduces three independent information channels, and cross-comparisons among all readings were utilized to eliminate experimental error and provide better measurement results. The novel fabrication and assembly technology can also be applied to other catheter based biomedical devices. A step forward inspired by the ancient art of folding, origami, which creating three-dimensional (3-D) structures from two-dimensional (2-D) sheets through a high degree of folding along the creases. Based on this idea, we developed a novel method to enable better deformability. One example is origami-enabled silicon solar cells. The solar panel can reach up to 644% areal compactness while maintain reasonable good performance (less than 30% output power density drop) upon 40 times cyclic folding/unfolding. This approach can be readily applied to other functional devices, ranging from sensors, displays, antenna, to energy storage devices. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2014

abstract: Intracranial pressure is an important parameter to monitor, and elevated intracranial pressure can be life threatening. Elevated intracranial pressure is indicative of distress in the brain attributed by conditions such as aneurysm, traumatic brain injury, brain tumor, hydrocephalus, stroke, or meningitis. Electrocorticography (ECoG) recordings are invaluable in understanding epilepsy and detecting seizure zones. However, ECoG electrodes cause a foreign body mass effect, swelling, and pneumocephaly, which results in elevation of intracranial pressure (ICP). Thus, the aim of this work is to design an intracranial pressure monitoring system that could augment ECoG electrodes. A minimally invasive, low-cost epidural intracranial pressure monitoring system is developed for this purpose, using a commercial pressure transducer available for biomedical applications. The system is composed of a pressure transducer, sensing cup, electronics, and data acquisition system. The pressure transducer is a microelectromechanical system (MEMS)-based die that works on piezoresistive phenomenon with dielectric isolation for direct contact with fluids. The developed system was bench tested and verified in an animal model to confirm the efficacy of the system for intracranial pressure monitoring. The system has a 0.1 mmHg accuracy and a 2% error for the 0-10 mmHg range, with resolution of 0.01 mmHg. This system serves as a minimally invasive (2 mm burr hole) epidural ICP monitor, which could augment existing ECoG electrode arrays, to simultaneously measure intracranial pressure along with the neural signals. This device could also be employed with brain implants that causes elevation in ICP due to tissue - implant interaction often leading to edema. This research explores the concept and feasibility for integrating the sensing component directly on to the ECoG electrode arrays. / Dissertation/Thesis / Masters Thesis Bioengineering 2015

abstract: Long-term monitoring of deep brain structures using microelectrode implants is critical for the success of emerging clinical applications including cortical neural prostheses, deep brain stimulation and other neurobiology studies such as progression of disease states, learning and memory, brain mapping etc. However, current microelectrode technologies are not capable enough of reaching those clinical milestones given their inconsistency in performance and reliability in long-term studies. In all the aforementioned applications, it is important to understand the limitations & demands posed by technology as well as biological processes. Recent advances in implantable Micro Electro Mechanical Systems (MEMS) technology have tremendous potential and opens a plethora of opportunities for long term studies which were not possible before. The overall goal of the project is to develop large scale autonomous, movable, micro-scale interfaces which can seek and monitor/stimulate large ensembles of precisely targeted neurons and neuronal networks that can be applied for brain mapping in behaving animals. However, there are serious technical (fabrication) challenges related to packaging and interconnects, examples of which include: lack of current industry standards in chip-scale packaging techniques for silicon chips with movable microstructures, incompatible micro-bonding techniques to elongate current micro-electrode length to reach deep brain structures, inability to achieve hermetic isolation of implantable devices from biological tissue and fluids (i.e. cerebrospinal fluid (CSF), blood, etc.). The specific aims are to: 1) optimize & automate chip scale packaging of MEMS devices with unique requirements not amenable to conventional industry standards with respect to bonding, process temperature and pressure in order to achieve scalability 2) develop a novel micro-bonding technique to extend the length of current polysilicon micro-electrodes to reach and monitor deep brain structures 3) design & develop high throughput packaging mechanism for constructing a dense array of movable microelectrodes. Using a combination of unique micro-bonding technique which involves conductive thermosetting epoxy’s with hermetically sealed support structures and a highly optimized, semi-automated, 90-minute flip-chip packaging process, I have now extended the repertoire of previously reported movable microelectrode arrays to bond conventional stainless steel and Pt/Ir microelectrode arrays of desired lengths to steerable polysilicon shafts. I tested scalable prototypes in rigorous bench top tests including Impedance measurements, accelerated aging and non-destructive testing to assess electrical and mechanical stability of micro-bonds under long-term implantation. I propose a 3D printed packaging method allows a wide variety of electrode configurations to be realized such as a rectangular or circular array configuration or other arbitrary geometries optimal for specific regions of the brain with inter-electrode distance as low as 25 um with an unprecedented capability of seeking and recording/stimulating targeted single neurons in deep brain structures up to 10 mm deep (with 6 μm displacement resolution). The advantage of this computer controlled moveable deep brain electrodes facilitates potential capabilities of moving past glial sheath surrounding microelectrodes to restore neural connection, counter the variabilities in signal amplitudes, and enable simultaneous recording/stimulation at precisely targeted layers of brain. / Dissertation/Thesis / Masters Thesis Bioengineering 2016

abstract: Mechanical behavior of metallic thin films at room temperature (RT) is relatively well characterized. However, measuring the high temperature mechanical properties of thin films poses several challenges. These include ensuring uniformity in sample temperature and minimizing temporal fluctuations due to ambient heat loss, in addition to difficulties involved in mechanical testing of microscale samples. To address these issues, we designed and analyzed a MEMS-based high temperature tensile testing stage made from single crystal silicon. The freestanding thin film specimens were co-fabricated with the stage to ensure uniaxial loading. Multi-physics simulations of Joule heating, incorporating both radiation and convection heat transfer, were carried out using COMSOL to map the temperature distribution across the stage and the specimen. The simulations were validated using temperature measurements from a thermoreflectance microscope. / Dissertation/Thesis / Masters Thesis Materials Science and Engineering 2016

abstract: The instrumentational measurement of seismic motion is important for a wide range of research fields and applications, such as seismology, geology, physics, civil engineering and harsh environment exploration. This report presents series approaches to develop Micro-Electro-Mechanical System (MEMS) enhanced inertial motion sensors including accelerometers, seismometers and inclinometers based on Molecular Electronic Transducers (MET) techniques. Seismometers based on MET technology are attractive for planetary applications due to their high sensitivity, low noise floor, small size, absence of fragile mechanical moving parts and independence on the direction of sensitivity axis. By using MEMS techniques, a micro MET seismometer is developed with inter-electrode spacing close to 5 μm. The employment of MEMS improves the sensitivity of fabricated device to above 2500 V/(m/s2) under operating bias of 300 mV and input velocity of 8.4μm/s from 0.08Hz to 80Hz. The lowered hydrodynamic resistance by increasing the number of channels improves the self-noise to -135 dB equivalent to 18nG/√Hz (G=9.8m/s2) around 1.2 Hz. Inspired by the advantages of combining MET and MEMS technologies on the development of seismometer, a feasibility study of development of a low frequency accelerometer utilizing MET technology with post-CMOS-compatible fabrication processes is performed. In the fabricated accelerometer, the complicated fabrication of mass-spring system in solid-state MEMS accelerometer is replaced with a much simpler post-CMOS-compatible process containing only deposition of a four-electrode MET structure on a planar substrate, and a liquid inertia mass of an electrolyte droplet. With a specific design of 3D printing based package and replace water based iodide solution by room temperature ionic liquid based electrolyte, the sensitivity relative to the ground motion can reach 103.69V/g, with the resolution of 5.25μG/√Hz at 1Hz. By combining MET techniques and Zn-Cu electrochemical cell (Galvanic cell), this letter demonstrates a passive motion sensor powered by self-electrochemistry energy, named “Battery Accelerometer”. The experimental results indicated the peak sensitivity of battery accelerometer at its resonant frequency 18Hz is 10.4V/G with the resolution of 1.71μG without power consumption. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2016

This thesis studies how the mechanical design of a surveillance camera affects the acoustic performance, mainly in terms of the frequency response within the human hearing range. During the project, the mechanical characteristics that affect frequency response were investigated by measuring the camera’s audio behavior in an anechoic chamber. A theoretical and adaptable acoustic model was built in COMSOL to simulate the frequency response of the sound path. Measurement and simulation results were compared to identify critical aspects of the mechanical design and adjust accordingly for better acoustic performance.

Ce travail de thèse porte sur l'étude de composites magnétoélectriques laminaires dans le but de réaliser des dispositifs innovants intégrés sur silicium tel que l'inductance RF variable. Grâce au couplage mécanique entre des couches adjacentes magnétostrictive ultra douce et piézoélectrique, il est possible d'obtenir un couplage magnétoélectrique indirect qui est supérieur de plusieurs ordres de grandeur à celui des matériaux multiferroïques naturels. Dans un premier temps, nous avons utilisé l'approche phénoménologique basée sur les énergies pour décrire le panorama des effets attendus dans des composites magnétoélectriques laminaires (multicouches). Ensuite, des composites magnétoélectriques macroscopiques à base de substrats piézoélectriques de type MFC et de couches minces de FeCoB ont été réalisés. L'étude du couplage magnétoélectrique en fonction de la composition de FeCoB a permis de déterminer les propriétés clés des matériaux, notamment le rapport λs/Ms, qui sont essentielles pour obtenir un effet magnétoélectrique élevé. Un coefficient magnétoélectrique record de 250 V∙cm‐1Oe‐1 a été obtenu. Par ailleurs, un microscope à effet Kerr a été spécialement développé pour pouvoir observer de manière quasi-instantanée la modification de la structure en domaines sous l'effet de la tension électrique dans ces composites. Pour la première fois, l'observation directe de la rotation de l'axe facile d'aimantation sous commande électrique a été réalisée. La deuxième partie de ce manuscrit est consacrée à la conception, simulation, fabrication et caractérisation d'un dispositif MEMS hybride d'inductance variable intégrée. Ce dispositif exploite l'effet magnétoélectrique indirect entre un élément moteur en PZT (sol gel) et un élément inductif à base de FeCoB. Etant donné le caractère multiphysique hors norme de ce dispositif, un ensemble de tests électriques, mécaniques, optiques et magnétiques a été déployé tout au long de la fabrication. Les résultats concluent à une preuve de concept partiellement fonctionnelle en raison principalement d'une mauvaise gestion des contraintes internes liées à la fabrication. Les pistes d'amélioration aux niveaux du design, des matériaux et des procédés sont identifiées. / This work is focused on the study of laminated magnetoelectric composites aiming at the realization of novel components integrated on silicon, such as variable inductors. Thanks to the mechanical coupling between two adjacent layers of ultra-soft magnetostrictive and piezoelectric materials it is possible to obtain an indirect magnetoelectric effect which is several orders of magnitude higher than in natural multiferroics. Firstly, we used an energy-based phenomenological approach to describe a range of expected effects in such laminated magnetoelectric composites. Thereupon, macroscopic magnetoelectric composites based on piezoelectric MFC substrates and magnetostrictive thin films of FeCoB were realized. The study of the magnetoelectric coupling vs. FeCoB composition leads to the identification of the key material parameters, such as λs/Ms, that are essential for high magnetoelectric effect. A record magnetoelectric coefficient of 250 V∙cm‐1Oe‐1 is obtained. In parallel, a specific Kerr effect microscope devoted to live observation of the magnetic domains change vs. applied electrical field was developed. For the first time, direct observation of the magnetic easy-axis rotation with voltage in such composites is reported. The second part of this work concerns the design, simulation, fabrication and characterization of a hybrid MEMS variable inductor. This device exploits the indirect magnetoelectric effect between a PZT sol gel driving element and a FeCoB-based inductive element. The unusual multi-physics nature of the device prompted us to deploy a set of electrical, mechanical, optical and magnetic tests throughout the manufacturing. The results conclude with partially functional proof of concept, mainly due to the lack of management of internal stress during the fabrication. Areas for improvement of design, materials and process are identified.

abstract: Advances in implantable MEMS technology has made possible adaptive micro-robotic implants that can track and record from single neurons in the brain. Development of autonomous neural interfaces opens up exciting possibilities of micro-robots performing standard electrophysiological techniques that would previously take researchers several hundred hours to train and achieve the desired skill level. It would result in more reliable and adaptive neural interfaces that could record optimal neural activity 24/7 with high fidelity signals, high yield and increased throughput. The main contribution here is validating adaptive strategies to overcome challenges in autonomous navigation of microelectrodes inside the brain. The following issues pose significant challenges as brain tissue is both functionally and structurally dynamic: a) time varying mechanical properties of the brain tissue-microelectrode interface due to the hyperelastic, viscoelastic nature of brain tissue b) non-stationarities in the neural signal caused by mechanical and physiological events in the interface and c) the lack of visual feedback of microelectrode position in brain tissue. A closed loop control algorithm is proposed here for autonomous navigation of microelectrodes in brain tissue while optimizing the signal-to-noise ratio of multi-unit neural recordings. The algorithm incorporates a quantitative understanding of constitutive mechanical properties of soft viscoelastic tissue like the brain and is guided by models that predict stresses developed in brain tissue during movement of the microelectrode. An optimal movement strategy is developed that achieves precise positioning of microelectrodes in the brain by minimizing the stresses developed in the surrounding tissue during navigation and maximizing the speed of movement. Results of testing the closed-loop control paradigm in short-term rodent experiments validated that it was possible to achieve a consistently high quality SNR throughout the duration of the experiment. At the systems level, new generation of MEMS actuators for movable microelectrode array are characterized and the MEMS device operation parameters are optimized for improved performance and reliability. Further, recommendations for packaging to minimize the form factor of the implant; design of device mounting and implantation techniques of MEMS microelectrode array to enhance the longevity of the implant are also included in a top-down approach to achieve a reliable brain interface. / Dissertation/Thesis / Ph.D. Bioengineering 2013

Neste trabalho é proposto um sensor capacitivo do tipo eletrodos interdigitados para avaliação da qualidade de combustíveis automotivos. Os eletrodos interdigitados apresentam algumas características adequadas ao sensor em questão. Entre elas o fato de elevar significativamente a capacitância por apresentar grande quantidade de capacitores em paralelo e de ser uma estrutura possível de se fabricar por processos convencionais de microfabricação. Além disso, esses eletrodos permitem que o combustível preencha seus espaçamentos funcionando como seu dielétrico. Foram feitas modelagens e simulações do sensor para verificação da influência de diversos parâmetros de projeto. Protótipos foram fabricados em substratos de alumina com eletrodos de níquel eletrodepositado. Os eletrodos têm larguras entre 50m e 100m, com espaçamento entre eletrodos tendo valores dessa mesma ordem. O comprimento dos eletrodos é de 800m. A altura dos eletrodos varia entre 20m e 40m. O sensor como um todo tem área em torno de 4cm². Foram realizadas medições com misturas álcool e água, gasolina e álcool, gasolina e querosene entre outras. As caracterizações mostraram bons resultados comprovando a validade do princípio proposto. O sensor se mostrou capaz de detectar os tipos de adulteração mais comuns no Brasil, adição de água ao álcool combustível e adição solventes orgânicos ou de álcool além do permitido à gasolina. / This work proposes a capacitive sensor with interdigitated electrodes in order to evaluate the quality of automotive fuel. Interdigitated electrodes have some interesting features for this type of sensor. Among them, they increase the capacitance by having several capacitors in parallel, and by having a structure feasible to be manufactured by conventional microfabrication processes. In addition, automotive fuel, serving as the dielectric material, fills the gaps of the electrodes. Modeling and computational simulations of the sensor were carried out in order to realize the influence of several design parameters. Samples were manufactured using alumina substrates with electroplated nickel electrodes. The width of the electrodes was chosen to be between 50m and 100m, with gaps of similar size. The paired length of the electrodes was 800m. The height of the electrodes varied between 20m and 40m. The whole sensor was around 4cm² in area. Several measurements were carried out using mixtures of alcohol and water, gasoline and alcohol, gasoline and kerosene, and others. Characterizations showed good results, validating the method. The sensor was capable of detecting the main types of fuel adulteration used in Brazil: addition of water to alcohol, and addition of organic solvents or alcohol to gasoline beyond the acceptable limit.