Rozprawy doktorskie na temat „Thick Film Sensors”

The development of thick film sensors is described from their early stages, where thick film technology was employed primarily for the fabrication of interface circuits and as a sensor support medium, through to their complete fabrication as screen printed sensor arrays. The concept of processing signal data from sensor arrays as a means of improving sensor performance is also introduced and examples given. These examples include the reduction of the effects of 1/f noise in thick film conductimetric sensors and the use of pattern recognition techniques applied to arrays of low specificity gas sensors to improve their selectivity to particular analyte gases. Examples of various thick film gas sensors are given and the fabrication methods and associated techniques for their use in the detection of a variety of toxic, flammable and environmental gases are described. Various sensors for liquids are described and, in particular, arrays of sensors for water quality determination. The investigation of useful lifetimes and operational performance of these devices is reported with particular reference to sensors for the measurement of temperature, conductivity, redox potential, pH and dissolved oxygen concentration in water.

Engine oil condition monitoring has attracted considerable interests from industries and general public over the years due to its critical role in maintaining the performance and longevity of cars and industrial engines. Lubricants degrade during the course of operation and can be costly or detrimental to the engine if oil change intervals are not optimised. However, on-line robust monitoring for oils has been very challenging since oil degradation process is often complicated and influenced by a number of parameters, such as the operating temperature and contamination. Due to the complexity of the oil chemistry and their degradation processes, there have not been any commercially available on-line sensors for oil chemical property monitoring reported. Oil acidity is traditionally measured through potentiometric or photometric/colorimetric titration methods. Recent attempts at miniaturising infrared (IR) and chronopotentiometric (CP) sensors based on solid-state devices for acidity/alkalinity determination of oil have not been very successful since the CP technology suffers from sensitivity and stability issues and IR sensors are still bulky and adversely susceptible to the engine‟s harsh environment. This project, sponsored by Shell Global Solutions, aims to develop robust chemical sensors that can detect oil acidity due to oil degradation. The initial comprehensive literature review has identified that thick film (TF) sensor technology offers compact and low cost mass production solutions and have been proved to be robust with good reproducibility for aqueous solutions acidity measurements. Their feasibility in detecting oil acidity was thus investigated in this study and experimental work has been carried out to fabricate TF electrodes and evaluate them in a range of oils to explain their performance in detecting acid content. Based on their performance in aqueous solutions in previous studies, this study has investigated the performance of one type of TF working electrode (Ruthenium Oxide (RuO2)) combined with various TF reference electrodes in order to develop the most suitable electrodes for oils. To simulate oil ageing, a fully formulated engine oil and a base oil were oxidised under controlled conditions. Also, different amount of nitric acid was added to a fully formulated oil to simulate the oil acidity changes. Acid number (AN) of the oil samples was obtained using conventional titration methods and viscosity and conductivity of the oil samples were measured using laboratory-based equipment in order to validate TF sensor measurements and establish a relationship between different properties of oil samples during their degradation process. Temperature effects on thick film electrodes as well as their long-term stability and repeatability were also investigated. The results show that, for the first time, TF sensors respond to the acidity changes in all oil types tested and linear correlation between the TF responses and the AN was found in the oxidised oils within certain ranges at the tested temperatures (50 °C and 80 °C). TF sensors can detect oil acidity up to AN of 28 mgKOH/g. Although oil conductivity and viscosity were affected by the oil oxidation process, but no direct relationship was found between them and the TF responses. Based on the experimental results, sensing mechanisms of the TF electrodes in oils are proposed.

DE LA TESIS DOCTORAL
Título: Diseño, fabricación y caracterización de sensores de capa gruesa
Doctorando: Peter Tsolov Ivanov
Director: Xavier Correig Blanchar
Los sensores de gases de estado sólido han demostrado ser muy prometedores para
supervisar la emisión de los agentes contaminadores en el aire, porque son una opción
de bajo coste para la construcción de analizadores de gases. Algunos problemas se
relacionados con este tipo de dispositivos, especialmente su baja selectividad y el alto
consumo de energía, siguen sin resolver. El objetivo de esta tesis doctoral es el
desarrollo de nuevos sensores y matrices de sensores con mejorada selectividad y
reducido consumo de energía.
La metodología usada en esta tesis consiste en fabricar matrices de sensores hechas de
sensores con distintas selectividades. Como la respuesta del sensor es diferente en
distintas temperaturas de trabajo y como los distintos dopantes o los filtros catalíticos
aceleran o reducen la respuesta del sensor, los diferentes sensores dan diferentes
reacciones. Combinando estas reacciones y con la ayuda de técnicas del reconocimiento
de patrones, se pueden crear grupos de sensores capaces de distinguir entre distintos
agentes contaminantes.
La tesis comienza repasando los métodos usados para la fabricación de los sensores de
gases y discutiendo los problemas relacionados con la baja selectividad de los óxidos
metálicos. Se especifican también los diferentes métodos para aumentar la selectividad.
Se introduce y se describe detalladamente la técnica de screen-printing. Los
experimentos se realizaron con cuatro tipos de substratos de sensores (cerámica, silicio,
microhotplate y silicon-on-insulator) y con más de 15 capas activas basadas en dióxido
de estaño y trióxido de tungsteno (puras y dopadas con oro, platino, plata, titanio y
paladio). Una amplia variedad de compuestos volátiles (amoníaco, etanol, acetona y
benceno), de gases (monóxido de carbono, dióxido de nitrógeno, metano y sulfuro de
PhD thesis of Peter Tsolov Ivanov Resumen de la tesis doctoral
hidrógeno) y de algunas mezclas binarias ha sido medida. Los resultados obtenidos por
los análisis cuantitativos y cualitativos de los gases estudiados con una matriz de
sensores basada en cuatro sensores simples nos han llevado a descubrir el óptimo
sensor/matriz para los distintos gases/mezclas binarias.
Los resultados demostraron que, con la ayuda de redes neuronales Fuzzy ARTMAP, es
posible identificar y cuantificar simultáneamente los gases analizados usando solamente
una matriz de microhotplates (cuatro sensores) con la misma capa activa. Los sensores
de SnO2 y de WO3 dopados demostraron diversa respuesta a los agentes contaminantes
probados. Componiendo cuidadosamente la matriz de sensores y definiendo bien la
temperatura de trabajo podemos discriminar, con un alto grado de éxito, los diversos
gases probados sin la necesidad de técnicas de reconocimiento de patrones.
La conclusión principal que se puede sacar de esta tesis es que las matrices de sensores,
junto con las técnicas de reconocimiento de patrones, se pueden utilizar para aumentar
perceptiblemente la selectividad de los sensores de óxidos metálicos. La simplicidad de
los métodos propuestos permite su uso en el desarrollo de analizadores de gases más
baratos y narices electrónicas portátiles.
A partir de la investigación realizada durante esta tesis doctoral se han elaborado 15
artículos publicados en revistas internacionales, 10 comunicaciones en las conferencias
internacionales y 3 comunicaciones en conferencias españolas.



PhD thesis of Peter Tsolov Ivanov Resume of the doctoral thesis
OF THE DOCTORAL THESIS
Title: Design, Fabrication and Characterization of Thick-Film Gas Sensors
Doctorate: Peter Tsolov Ivanov
Director: Xavier Correig Blanchar
Solid-state gas sensors have proved to be very promising for monitoring the emission of
air pollutants because they are a low cost option for constructing gas analysers. Some
problems associated to this approach, especially their deficient selectivity and high
power consumption, remain unsolved. The aim of this doctoral thesis is to develop new
sensors and sensor matrices that can improve the selectivity of metal oxide gas sensors
and decrease their power consumption.
The methodology used here consists of creating sensor matrices made from sensors with
different selectivities. As the sensor response is different at different working
temperatures and as the different dopants or catalytic filters accelerate or reduce the
sensor response, the different sensors give different reactions. If these reactions are
combined, sensor groups capable of discriminating between different pollutants can be
obtained with the help of pattern recognition techniques.
The thesis begins by reviewing the methods used for fabricating gas sensors and
discussing the problems caused by the poor selectivity of metal oxide gas sensors and
the methods for increasing their selectivity. Then, the screen-printing technique is
introduced and described. The experiments were performed with four different types of
gas sensor substrates (ceramic, silicon, microhotplate and silicon-on-insulator) and more
than 15 active layers (undoped and doped with gold, platinum, silver, titanium and
paladium tin dioxide and tungsten trioxide sensitive layers). A wide variety of volatile
compounds (ammonia, ethanol, acetone and benzene), gases (carbon monoxide,
nitrogen dioxide, methane and hydrogen sulphide) and some binary mixtures were
measured. The results obtained from quantitative and qualitative gas analysis using the
PhD thesis of Peter Tsolov Ivanov Resume of the doctoral thesis
sensor response from a simple 4 sensor based matrix led to the optimal sensor/sensor
matrix for gas/binary mixtures.
The results showed that, with the help of fuzzy ARTMAP neural networks, it is possible
to identify and simultaneously quantify the gases analysed by using only one MHP-chip
(four sensors) with the same active layer. The doped SnO2 and WO3 sensors showed
different response to the tested pollutants. Composing carefully the sensor matrix and
defining well the working temperature we were able to discriminate, with a high success
rate, between the different test gases with no need for pattern recognition techniques.
The main conclusion that can be drawn from this thesis is that sensor matrices can be
used, coupled to dynamic pattern recognition techniques, to significantly increase the
selectivity of metal oxide sensors. The simplicity of the methods implemented makes
them suitable for developing low-cost gas analysers and hand-held e-noses.
The research carried out during this doctoral thesis resulted in 15 articles being
published in international journals, 10 communications at international conferences and
3 communications at a Spanish national conference.
PhD thesis of Peter Tsolov Ivanov Resumen de la tesis doctoral

Aspects relating to and including the development of thick film amperometric zirconia oxygen sensors were investigated. These devices, which were operated in the range 550-950°C, had a laminated structure in which a cathode, an electrolyte and an anode were printed, in that order, onto a planar alumina substrate. The anode and electrolyte were porous and during sensor Operation also acted as a diffusion barrier, restricting the rate of oxygen diffusion to the cathode. A thick film platinum heater was also developed to maintain the sensor at its operating temperature while acting simultaneously as a résistance thermometer; it was screen-printed onto the substrate on the reverse side to the sensor. The individual components were characterised and optimised prior to assembly of complete sensors. Zirconia films were deposited by screen-printing onto alumina substrates. Careful attention was paid to formulation of zirconia inks, drying and firing procedures. Temperatures above 1350°C were necessary to sinter the zirconia to a low (<0.1%) though not zero porosity. The high sintering temperatures were found to result in the diffusion of impurities from the 96% alumina Substrate into the zirconia film which accelerated grain growth. X-ray diffraction showed that the grain growth resulted in transformation of the metastable tetragonal zirconia to the monoclinic form: where this occurred frequency response analysis of the films showed the expected decrease in ionic conductivity. These effects were absent on high purity (99.6%) alumina substrates. Platinum-zirconia cermets were investigated as possible electrodes. When screen-printed and fired at 1000°C for 1 hour and operated in the range 500-700°C, electrode activity was orders of magnitude greater than for pure porous platinum electrodes and increased substantially with increasing zirconia fractions provided electronic continuity was maintained within the film. High firing temperatures (> 1000°C), which were necessary for preparing a sensor with co-fired electrolyte and electrodes, decreased electrode activities although cermets remained greatly superior to pure platinum. Planar amperometric zirconia oxygen sensors were prepared using thick-film technology exclusively. When a voltage (0.5-1.4 V) was applied between the electrodes, a current flowed which was directly proportional to the oxygen concentration in the range up to 21%; this has not previously been achieved with such sensors. Characteristics were shown to be dependent upon firing temperature and substrate purity. Interestingly, temperature coefficients of the output were positive and negative for sensors fired at temperatures up to 1400 and above 1450°C respectively. Operation in the combustion products of a gas-burning flue demonstrated linear dependence upon calculated oxygen concentration. Heaters, printed using either fritted or unfritted platinum inks, were given extended treatments in a furnace at elevated temperatures (1000-1300°C) to accelerate ageing effects. Measurements were made of résistance (at 20°C), platinum evaporation rate and film cross-sectional area and these were correlated with the microstructure. The variation of résistance (at 20°C) of the films was analysed using effective medium theory invoked in order to quantify the blocking effect of the non-metallic fractions. During the initial phase (résistance decreasing) the governing factor was probably the high resistance of necks between contacting platinum particles. During the subsequent phase (resistance increasing) the resistance was controlled principally by the formation and growth of voids.

Buildings, roads, bridges and structures in general suffer many kinds of damages due to overstress caused by settlements of foundations, high winds, dynamic forces, passing traffic, vibration and unexpected external loads beyond the safe design forces. The damages manifest itself by cracks, falling of plaster and render uneven roads and some time complete collapse. The cost of maintaining and fixing damages caused by the above is quite high for the building and construction industry. The same phenomenon is common to many other structures like airplanes, wind turbine and machinery in general. Structural Health Monitoring (SHM) is the engineering branch, which aims to give, at every moment during the life of a structure, a diagnosis of the "state" of the constituent materials, of the different parts of a structure. The state of the structure must remain in the domain specified in the design, although this can be altered due to usage or due to normal aging by the action of the environment, and by accidental events. By using special electronic sensors to monitor the unexpected high concentration of stresses or changes of these stresses throughout the life of the structure and pavement, reduces the cost of maintenance and repair. Historic buildings would also benefit from using such sensors to monitor the overstress in the old and frugally stones and bricks. The sensors can be embedded in the lime mortar joints and an electronic meter is used periodically to check for any unusual overstress during the life of the building. The main aim of the proposed research project is to investigate the possibility of using thick-film technology stress sensors in masonry, concrete and building materials in general to monitor overstress and instability throughout the life of the structures. The sensors could be used in brick, block, stone, and concrete and they could be mounted on the surface or embedded in the materials. There are many research studies on strain gauge devices in structural monitoring; Thick Film (TF) piezo-resistive sensors are proposed as a direct alternative to the widely used metal Foil Strain Gauges (FSG). Due to the low cost of TF sensors, their ease of use, suitability to integrate electronics on board, and to have different geometrical shapes, they could be deployed at different locations in a building, road or be distributed in arrays. This offers the continuous monitoring of stresses at any time by using a data logger on two points on the surface or by using wireless electronic transmission. In this research, new thick film screen-printed ceramic piezo-resistive sensor has been developed and characterized as discrete device for deployment on surface of a structure and embedded into the structure during building material curing or after structure erection. The sensor response on different building materials has been experimented and compared. Mechanical and electronic simulation tools were used to characterise the sensor and to choose an adequate interface electronic circuit. The experimental results of the simulated sensor and circuitry, showed the suitability of the sensor to be embedded in building materials during curing period and on erected structures. Materials used were wood, concrete, brick and plaster. In addition, the overall linearity of response of the sensors applied on building material surface was asserted which makes the technology a candidate for a more wide deployment in SHM field.

The measurement of dissolved oxygen in natural waters is vital as an indicator of pollution incidents, water quality and for the general maintenance of water courses and reservoirs. Commercially available instruments used in such applications tend to be bulky items which require frequent calibration and are prone to fouling problems. Consequently, they are incapable of providing continuous accurate in situ measurements. One of the most predominant problems though is the extreme cost of these instruments, resulting from the employment of expensive fabrication techniques. Thick film technology is a bulk batch automatic fabrication process which affords reproducible, uniform and miniature surfaces at low cost, making it an ideal technique for sensor production. This thesis describes the application of thick film technology to the production of miniature amperometric dissolved oxygen sensors based on a three electrode potentiostatic mode of operation, for initial use hi the water industry. The devices have been designed and fabricated with gold electrodes covered by a gel electrolyte held intact by an oxygen permeable cellulose acetate membrane. Novel screen-printable membrane and gel thick film pastes have been realised in the process. An external silver/silver chloride reference electrode completes the electrochemical cell. When powered, the working electrode of a device is held at an applied potential of-0.6V versus the reference electrode. A full investigation of device performance has been carried out including sensitivity, linearity and response times. The oxygen permeable membrane has been shown to act as a protective layer as well as providing a diffusion barrier to the electrodes. Methods of reducing instability effects and characteristics caused by variations hi physical parameters have been studied. As a consequence an in depth knowledge of the behaviour of membrane covered dissolved oxygen sensors has been acquired. Results show that the use of a more negative applied potential, alternative membranes (such as PVC and PTFE) and a microelectrode configuration reduce instability and fouling problems, maintaining oxygen sensitivity and minimising secondary operating characteristics caused by the effects of parameters such as flow and temperature. This in turn improves the lifetime of thick film dissolved oxygen devices as they are continuously powered.

La tecnologia thick film pot definir-se com el procés on es du a terme la creació de dipòsits de circuits impresos en un substrat de ceràmica rígida mitjançant la tècnica de la serigrafia. Les pastes que s’utilitzen per aquest propòsit es formulen amb l’addició de vidres i diferents òxids que promouen l’adhesió dels dipòsits generats al substrat a temperatures entre 600 i 950ºC. Per altra banda, la tecnologia ceràmica multicapa, permet una disposició densa de circuits impresos incorporant components interns en un únic dispositiu amb estructura multicapa monolítica. Els substrats ceràmics pels sistemes multicapa presenten una baixa constant dielèctrica similar als substrats tradicionals emprats en la tecnologia thick film i les pastes per dur a terme les metal•litzacions s’ha de dissenyar per tal de co-sinteritzar amb el substrat ceràmic. La primera mostra d’aquesta tecnologia la trobem en els sistemes HTCC (High Temperature Co-fired Ceramics) que es basen tradicionalment en l’ús de materials basats en alúmina. L’elevada temperatura de cocció que requereixen aquests substrats (aprox. 1600ºC) limita el nombre de materials conductors amb els quals poden co-sinteritzar. Aquests conductors solen ser tals com el W, Mo, Mn i Pt. L’evolució d’aquesta tecnologia la trobem en l’ús de compostos vitro ceràmics que cristal•litzen i reaccionen total o parcialment amb diferents òxids addicionats a la mescla de vidres. L’ús de vidres de baix punt de fusió que envolten els additius ceràmics és una altra aproximació a aquesta evolució de la tecnologia. Es tracta del que s’ha anomenat LTCC (Low Temperature Co-fired Ceramics) i que permet unes temperatures de sinterització per sota dels 950ºC fent viable l’ús de materials d’elevada conductivitat (Ag, Au, Cu...) que co-sinteritzen amb el substrat. Tant la tecnologia thick film com la tecnologia ceràmica multicapa són el nexe que uneix tots els estudis que s’han realitzat en aquest treball. Els camps sobre els quals ha girat l’aplicació d’aquesta tecnologia són per una banda, els sensors d’oxigen dels gasos d’escapament del vehicle (exhaust gas oxygen sensors), concretament els basats en un òxid semiconductor, el diòxid de titani, per ser usat com a sonda lambda i per altra banda, els sensors ceràmics capacitius i la viabilitat de fabricar-los amb materials LTCC. La primera part del treball és una introducció a les tecnologies emprades, les característiques de les quals en ajudaran a entendre certs detalls en l’evolució d’aquest estudi que es relacionen íntimament amb el tipus de tecnologia utilitzada. Pel que fa el sensor lambda basat en diòxid de titani, la primera part de l’estudi ha consistit en la preparació d’un material catalític amb una elevada àrea superficial dispersat en el suport porós de TiO2. El procediment ha consistit en realitzar mètodes d’impregnació i l’objectiu ha estat establir els principals paràmetres que afecten al procés d’addició del catalític i determinar la mínima quantitat de material necessària per tal d’obtenir un dispositiu sensor basat en Pt-TiO2 adequat per ser usat com sensor lambda. En aquesta primera part, el dipòsit del material sensor es realitzà amb tècniques thick film sobre un substrat d’alúmina HTCC ja sinteritzat. El següent pas pel que fa el sensor de TiO2 va ser l’estudi de la millora en el seu procés de fabricació centrant els esforços en aconseguir una millor adhesió entre el material sensor i el substrat d’alúmina. L’objectiu fou eliminar el major nombre de processos thick film després del sinteritzat del substrat (post-firing) amb diferents propostes per dipositar el material sensor sobre la ceràmica en verd i co-sinteritzar els dos elements. Els treballs van consistir en el disseny de pastes adequades per poder dipositar el material sensor, l’estudi del paper dels compostos de titanat d’alumini que es formen a la temperatura de sinterització de l’alúmina, la caracterització funcional dels dispositius proposats i finalment, com aquests mètodes proposats afecten a l’addició del material catalític. Els resultats obtinguts han permès establir una quantitat de platí del 1.8% en pes, respecte la quantitat de TiO2 en el dispositiu. Amb mètodes d’impregnació s’ha obtingut una pols de partida amb la qual s’ha preparat una pasta que aplicada amb tècniques thick film ha permès la obtenció de dispositius aptes per treballar com a sensor lambda. En quant a la co-sinterització del material sensor els millors resultats s’han obtingut controlant el grau de formació de titanat d’alumini i controlant la seva estabilitat tèrmica amb l’ús de l’MgO com additiu en la formulació de la pasta del material sensor. La porositat del material sensor juga un paper crucial en aquest mètode degut al fet que l’addició del material catalític s’ha dut a terme mitjançant processos post-firing per tal d’evitar les elevades temperatures de sinteritzat. Pel que fa els sensors de pressió ceràmics capacitius, el treball es focalitzà en l’estudi de la viabilitat dels materials LTCC per tal de fabricar aquest tipus de dispositius. Es van dur a terme comparacions entre les característiques funcionals de dispositius fabricats en tecnologia thick film sobre alúmina i dispositius amb membranes flectores de diferents materials LTCC i es caracteritzà la seva sensibilitat així com la seva estabilitat en la resposta. Finalment es proposà un dispositiu miniaturitzat plenament integrat en tecnologia LTCC en el qual es caracteritzà la seva resposta. Els resultats obtinguts avalen la possibilitat d’utilitzar els materials LTCC per fabricar sensors de pressió ceràmics capacitius. S’han establert els criteris de planitud i distància entre elèctrodes en funció de la mida de l’elèctrode de mesura del dispositiu. La caracterització funcional ha mostrat la dependència del disseny amb el tipus de fluid de pressió, la sensibilitat requerida i el rang de pressió de treball. La capacitat paràsita generada per la interacció entre el fluid incident i el dispositiu i les condicions de segellat són els aspectes principals que afecten a l’estabilitat de la resposta.
The thick film technology can be defined as the process that involves the deposition of metal circuitry on an already fired ceramic substrate using screen-printing technology. Pastes for this purpose are formulated with glass frit and oxides to promote the adhesion to the substrate firing the parts at low temperatures (600-950ºC). By the other hand, the multilayer technology allows a dense circuitry layout incorporating buried components in a single, monolithic, hermetic package. Ceramic substrates for the multilayer systems use low dielectric constant materials similar to traditional ceramic substrates for thick film technology and pastes for metallization that must be designed in order to co-fire with ceramic substrate. First approach of this technology was the high temperature co-fired ceramics (HTCC) traditionally based on alumina material. The relatively high temperature of alumina-based ceramics (approx. 1600ºC) limits the number of conductor materials able to co-fire with ceramic substrate. Such conductor materials are typically based on W, Mo, Mn and Pt. Next approach of this technology, involves the use of glass-ceramic based materials that undergo devitrification to a crystalline phase during firing or containing glass frit with low temperature melting point and different crystalline fillers. Development of Low Temperature Co-fired Ceramics (LTCC) allow to decrease the firing temperature down to 950ºC and conductor pastes based on high conductivity materials (Ag, Au, Cu…) can be used to co-fire with these substrates. Thick film and multilayer ceramic technology are both the link of the work carried out. The fields in which both technologies have been deployed are by one side, the heated exhaust gas oxygen sensor (HEGO), specifically semiconductor gas sensor based on TiO2 suitable to use as lambda probe and by the other side, the ceramic capacitive pressure sensors and its availability to be manufactured using LTCC materials. First part of the work is an introduction to the technologies and main features that will help us later to understand details on the work development, that are strongly influenced by the kind of technology we are working with. Regarding lambda sensor based on TiO2, first part of the study have consisted of a process for preparing a catalyst which has a high surface area, finely divided and catalytically active on a porous carrier structure based on titania. Process was based on impregnation methods and the aim of this work was to establish main parameters that affect deposition control and to determine minimum amount of catalyst which is needed in order to get suitable Pt-TiO2 based sensor to be used as lambda probe. In this approach sensor deposit is generated using thick film technologies over an alumina HTCC substrate. Next step regarding titania sensor was to study the improvement of manufacturing process in order to improve the adhesion between sensor material and alumina based substrate. This approach was focused in to avoid thick film post-firing processes proposing and single sensor material deposition in green state and co-firing both substrate and sensor material. The aim of this work was to design sensor pastes in order to get suitable sensor deposit, to study the role of aluminum titanate compounds generated at firing temperatures, to study the functional features of proposed devices and finally how this method affects the catalytic material addition. Results lead us to establish a minimum amount of 1.8wt% of Pt/TiO2-nominally using impregnation methods to obtain impregnated titania powder which can be applied to the substrate using thick film technology (prior paste preparation). Response of the devices is suitable to be used as lambda sensors. The sensor material co-firing approach gave us better results by controlling the amount of aluminum titanate formation and controlling its thermal stability by using MgO as additive in sensor material paste formulation. Porosity of the sensor material plays a key role in this approach due to the fact that catalyst addition must be done in post-firing process in order to avoid high sintering temperatures. Regarding ceramic capacitive pressure sensors, the work was focused in the study of suitability of LTCC as materials for manufacturing such devices. Comparison between functional features of thick film over alumina devices and LTCC membrane devices were carried out and sensibility and response stability was characterized as well. Finally, a miniaturized, fully integrated LTCC device was proposed and the sensor response was characterized. Results showed us the feasibility of LTCC materials to be used in capacitive pressure sensors. Flatness criteria was established regarding measuring electrode size and the functional characterization gave us as result the sensor design dependence on the kind of working fluid, accuracy and pressure range. Parasitic capacitance generated by device interaction with working fluid and sealing conditions was established as main features that can affect the stability in the response.

Over the past decade vehicle judder caused by Disc Thickness Variation (DTV) has become of major concern to automobile manufacturers worldwide. Judder is usually perceived by the driver as minor to severe vibrations transferred through the chassis during braking [1-9]. In this research, DTV is investigated via the use of a Smart Brake Pad (SBP). The SBP is a tool that will enable engineers to better understand the processes which occur in the harsh and confined environment that exists between the brake pad and disc whilst braking. It is also a tool that will enable engineers to better understand the causes of DTV and stick-slip the initiators of low and high frequency vibration in motor vehicle brakes. Furthermore, the technology can equally be used to solve many other still remaining mysteries in automotive, aerospace, rail or anywhere where two surfaces may come in contact. The SBP consists of sensors embedded into an automotive brake pad enabling it to measure pressure between the brake pad and disc whilst braking. The two sensor technologies investigated were Thick Film (TF) and Fibre Optic (FO) technologies. Each type was tested individually using a Material Testing System (MTS) at room and elevated temperatures. The chosen SBP was then successfully tested in simulated driving conditions. A preliminary mathematical model was developed and tested for the TF sensor and a novel Finite Element Analysis (FEA) model for the FO sensor. A new method called the Total Expected Error (TEE) method was also developed to simplify the sensor specification process to ensure consistent comparisons are made between sensors. Most importantly, our achievement will lead to improved comfort levels for the motorist.

The measurement of the partial pressure of oxygen in arterial blood is essential for the analysis of a patient's respiratory condition. There are several commercially available methods and systems to measure this parameter transcutaneously. However, they tend to be cumbersome and costly. To overcome the disadvantages presented, a new type of sensor for transcutaneous blood gas measurement was investigated, employing thick film technology, which is an excellent technique to produce sensors in bulk, as it is cost effective and easy for reproducible fabrication. This thesis describes the application of thick film technology to the investigation and production of these sensors, which are small in size and readily disposable. Such advantages are greatly welcomed from the medical point of view. The devices under investigation were based on amperometry. Gold electrodes were printed on an alumina substrate and covered with a layer of electrolyte gel and then finally, with a membrane. An external silver/silver chloride reference electrode was also employed in this electrochemical cell set-up. The project also involved several electronic circuits to support the transcutaneous oxygen measurement. The main studies were concentrated on the materials employed as the electrolyte and membrane. Investigations were carried out to evaluate the performance of these devices in atmospheric and hydrated environment as well as under the influences of different temperatures. Detailed discussions of the results were presented and future work for the project is identified. The novel contributions towards this research work were categorized into two major modules. Firstly, in the heating module, a single element taking both the roles of a heating element and a feedback temperature sensor was employed for the wheatstone bridge circuit configuration to regulate the transcutaneous temperature. In addition, the oxygen sensing module included studies on the effectiveness of using Nafion polyelectrolytes to achieve amperometric measurements for transcutaneous oxygen monitoring. From the experimental results, the most promising choice for the thick film transcutaneous oxygen sensor was the prototype with Nafion as the electrolyte and PTFE as the membrane. The disposable prototype produced results achieving low manufacturing cost of approximately £1 and was able to make continuous measurement of up to 46 hours. It proved to be compact, non-biohazardous and portable with a good degree of user-friendliness. It also provided accurate and reproducible measurements of not more than 3% error. Thorough intensive research activities were carried out to overcome all existing problems in order to achieve the objectives of the research project. With more than 70% of the specifications being met, the positive results had presented a successful design for the fundamental version of the disposable transcutaneous oxygen sensor employing thick film fabrication processes.

El control dels nivells d'oxigen és una etapa crítica en molts processos industrials.
En alguns d'aquests, els nivells d'oxigen han de ser detectats i controlats fins i tot en el rang de les ppm. Encara que ja es coneguin molts mètodes, ja provats per a la detecció d'oxigen, la majoria d'ells són costosos i complexes. Altres mètodes més accessibles com els sensors Lambda i les cel·les electroquímiques també presenten alguns problemes. Els primers requereixen d'elevades concentracions d'oxigen o bé de controls de temperatura força precisos amb la finalitat de treballar sense interferència d'altres gasos. Els segons es poden veure afectats per temps d'exposició massa perllongats a gasos "àcids" com el diòxid de carboni i no es recomana el seu ús en atmosferes amb un contingut de més del 25% d'aquests gas.
Són moltes les avantatges dels sensors de gasos: baix cost, mida reduïda i solidesa entre d'altres. Això fa que aquests dispositius despuntin com a solució per a la detecció d'oxigen. Molts autors han fet esment a la detecció d'oxigen a nivells de ppm emprant aquests tipus de sensors. De totes formes, la majoria s'han desenvolupat mitjançant tecnologia de capa fina. En aplicacions industrials, la tecnologia més usada és la de capa gruixuda, ja que aquests sensors són més fàcils de fabricar i de dopar que els de capa fina. En el cas dels sensors de capa gruixuda la detecció de traces d'oxigen és una tasca encara difícil i es necessita treballar a elevades temperatures (> 700 ºC).
El mecanisme de detecció d'oxigen en els sensors basats en òxids semiconductors es basa en la interdependència de la conductivitat elèctrica d'aquests òxids i la pressió parcial d'oxigen en l'ambient. El diòxid de titani és l'òxid semiconductor que més s'usa en la detecció d'oxigen. Els sensors basats en òxid de titani (en fase cristal·lina rutil) necessiten d'elevades temperatures per a un correcte funcionament (700 ºC-1000 ºC), ja que la detecció d'oxigen en la fase rutil es deu principalment a la difusió dels ion d'aquests gas en el volum del material. Per a que es produeixi la reacció en el volum o propi cos del material es necessiten elevades temperatures. Això comporta un consum també elevat de potència, que a la llarga serà un problema en determinades aplicacions industrials.
Al tenir l'òxid de titani en fase anatasa més electrons lliures, la reacció de l'oxigen amb aquests material es pot associar a una reacció superficial. Aquesta reacció es pot donar a temperatures més baixes, al voltant dels 400 ºC-500 ºC i per tant, mantenir la fase anatasa en l'òxid permet disminuir la temperatura de treball, la qual cosa és desitjable per al disseny del sensor.
Tal i com es reflecteix en algunes referències bibliogràfiques, en dopar l'òxid de titani amb ions pentavalents com el Nb5+, aquests ions s'introdueixen en l'estructura cristallina de l'òxid de titani en fase anatasa, creant certa tensió interna que provoca una resistència al canvi de fase anatasa a rutil, inhibint el creixement del gra.
Per altra banda, també s'ha posat de manifest en moltes fonts que el dopatge d'òxid de titani amb niobi augmenta la sensibilitat d'aquests material vers l'oxigen. Aquests dopatge fa que l'òxid presenti una impedància menor, per la qual cosa es facilita el disseny de l'electrònica associada al sensor.
Tenint en compte aquestes dues premisses, un dels objectius d'aquesta tesi ha estat la síntesi, mitjançant la tècnica del sol-gel, de diferents tipus d'òxid de titani per a la fabricació d'un sensor d'oxigen que operi en una marge de temperatura relativament moderat (300 ºC-600 ºC). Es treballà amb òxids sense dopar i amb òxids dopats amb un 3 % de niobi.
Cadascun es calcinà a diferents temperatures: 600 ºC, 700 ºC, 800 ºC i 900 ºC.
Amb l'objectiu de correlacionar l'estructura i la sensibilitat i selectivitat dels òxids sintetitzats, aquests es van sotmetre a diferents tècniques de caracterització. En primer lloc l'espectrometria de plasma acoblada inductivament (ICP) s'utilitzà amb el fi de determinarne la composició química i quantificar la proporció de cada component. En segon lloc també es va fer servir la difracció de raigs X (XRD) per a determinar les fases cristallines de l'òxid i la mida dels cristalls. També es va analitzar la porositat i es van fer mesures de l'àrea superficial (BET) dels nanopols . En darrer lloc, la microscopia d'escàner d'electrons (SEM) fou aplicada amb la finalitat d'obtenir detalls de l'estructura de la capa activa i de la grandària dels grans. Per a l'anàlisi quantitatiu i també qualitatiu de les capes s'utilitzà l'espectroscopia de dispersió d'energia per raig X (EDS).
Com a substrat, es va desenvolupar un substrat d'alúmina que pot albergar quatre capes actives treballant a una mateixa temperatura, conformant així una matriu de sensors.
Treballant amb aquests substrat aparegueren certs inconvenients (relacionats amb l'encapsulat del sensor) operant a temperatures superiors als 450 ºC. Això comportà un retard en l'obtenció de resultats amb aquests substrat i es decidí introduir-ne un de nou, resistent a altes temperatures i adquirit en l'Institut Kurchatov (Moscou, Rússia). Tots els resultats presentats en aquesta tesi es van obtenir amb l'ús d'aquests últim substrat. En l'actualitat s'està treballant amb el primer d'ells.
Un cop fabricat els sensors, es provà la sensibilitat d'aquests en un sistema de flux continu. Es provaren tres diferents concentracions: 20 ppm d'O2 amb balanç de N2, 30 ppm d' O2 amb balanç CO2 i 15 ppm d'O2 amb balanç CO2. També es testaren les respostes dels sensors cap a altres gasos interferents (SO2, CH4, H2S i C2H4) per a determinar la selectivitat d'aquests sensor vers l'O2 en presència d'interferents.
La millor resposta vers l'oxigen s'aconseguí amb els sensors dopats i calcinats a 700 ºC. Aquesta millora es pot atribuir al dopatge amb niobi. Aquets metall inhibeix el creixement del gra i per tant s'aconsegueixen òxids amb major àrea superficial. Un altre motiu pot ser el retard del pas de la fase rutil a la fase anatasa induït també pel dopatge de Niobi.
La sensibilitat denotada pels òxids dopats amb niobi i calcinats a 600 ºC fou molt baixa tot i que en principi aquests òxids presentaven característiques físiques potencialment millors que els calcinats a 700 ºC. Aquests fet es pot explicar degut a la presència d' alguns depòsits de carboni que no es pogueren eliminar durant la calcinació. La presència d'aquests dipòsits fou confirmada per l'anàlisi Raman d'aquests materials. Les estructures de carboni presents en aquests residus cobreixen gran part de la superfície de l'òxid i contribueixen a la desactivació del procés catalític que té lloc en aquesta superfície.
Concernint a les mesures realitzades sota atmosfera de CO2, els òxids dopats amb niobi i calcinats a 700 ºC també respongueren a l'oxigen. La resposta en el cas de 15 ppm d'O2 s'invertí del tipus oxidant a tipus reductor. A baixes concentracions d'oxigen els ion CO- provinents de la dissociació del CO2 s'adsorbeixen a la superfície del material actiu.
L'oxígen, enlloc de deplexionar-se, interactua amb aquests ions per a formar CO2, alliberant electrons a la capa activa i això fa que la natura de la resposta sigui reductora. Les respostes vers altres gasos contaminants com SO2, CH4, H2S foren tipus reductor com era esperat, la qual cosa indicava que el canvi de resposta no era atribuïble al canvi de l'òxid conductor de tipus n a tipus p, si no més aviat en un canvi en la naturalesa de la reacció.
Amb l'objectiu de millorar la sensibilitat de l'òxid dopat es provà d'incrementar la seva àrea superficial i porositat utilitzant un surfactant com a motlle durant el procés de síntesi. El surfactant usat fou la dodecilamina, que forma una estructura miscel·lar que fa de motlle en el procés de nucleació de l'òxid, generant així grans menors amb major àrea superficial i major porositat. De tres diferents temptatives, els millors resultats es varen obtenir quan s'addicionaven 8 ml de dodecilamina a la solució del sol-gel immediatament desprès de la hidròlisi dels alcòxids. Les proves XRD mostraren que l'addició del surfactant retarda encara més la transició de les fases rutil a anatasa i també inhibeix el creixement dels cristalls. Aquests resultats es recolzen sobre les micrografies del SEM i l'anàlisi BET, que revelaren un creixement en la porositat del material i un gra menor. Malgrat aquests resultats prometedors en la determinació estructural, les mesures de l'oxigen amb aquests sensors revelaren una poca sensibilitat dels sensors modificats amb el surfactant. Els espectres RAMAN mostraren alguns pics corresponents a diferents morfologies de carboni.
Els dipòsits d'aquests material en la superfície, tal i que com s'ha mencionat anteriorment, inhibeixen la resposta de la capa activa vers l'oxigen.
El control de los niveles de oxígeno es una etapa crítica en muchos procesos industriales. En algunos de estos procesos, los niveles de oxígeno deben ser detectados y controlados incluso en el rango de las ppm. Aunque existen varios métodos ya probados para la detección de oxígeno en estos sistemas de control, la mayoría de ellos son costosos y complejos. Otros métodos de detección más accesible como los sensores Lambda o las celdas electoquímicas también presentan problemas: los primeros requieren de altas concentraciones de oxígeno, o de controles de temperatura bastante precisos para poder trabajar sin interferencias de otros gases Los segundos pueden verse afectados por una prolongada exposición a gases "ácidos" como el dióxido de carbono y no se recomienda para uso continuo en atmósferas con un contenido de mas del 25 % de dicho gas.
Debido a muchas ventajas tales como su bajo costo, tamaño reducido y solidez, los sensores basados en semiconductores aparecen como una solución para la detección de oxígeno. Algunos autores han reportado la detección de oxígeno a niveles ppm empleando este tipo de sensores. Sin embargo, la mayoría de ellos han sido desarrollados mediante tecnología de capa fina. En aplicaciones industriales, la tecnología más usada es la de capa gruesa, ya que estos sensores son más fáciles de fabricar y de dopar que los sensores de capa fina. En los sensores de capa gruesa, la detección de trazas de oxígeno es aun una tarea difícil de alcanzar, siendo normalmente necesarias altas temperaturas (> 700 ºC) para lograrlo.
El mecanismo de detección de oxígeno en los sensores basados en óxidos semiconductores esta basado en la fuerte dependencia de la conductividad eléctrica de estos materiales a la presión parcial de oxígeno en el ambiente. El dióxido de titanio es el óxido semiconductor más ampliamente usado en la detección de oxígeno. Los sensores basados en TiO2 (usualmente en la fase cristalina rutilo) necesitan trabajar a elevadas temperaturas (700 ºC - 1000 ºC), ya que la detección de oxígeno en la fase rutile se debe principalmente a la difusión de los iones de oxígeno en el volumen del material. Para que se produzca la reacción en el volumen hacen falta altas temperaturas, lo que conlleva un alto consumo de potencia que puede ser un handicap en determinadas aplicaciones industriales.
Por otro lado, el dióxido de titanio en fase anatase posee más electrones libres, así que la detección de oxígeno en este material puede asociarse a una reacción de superficie, la cual tiene lugar a no tan altas temperaturas (400 ºC - 500 ºC).Por lo tanto, puede concluirse que mantener la fase anatase permitiría la detección de oxígeno a temperaturas moderadas, lo cual es deseable para el diseño del sensor.
Se ha reportado que cuando el TiO2 es dopado con iones pentavalentes, i.e. Nb5+, tales iones se introducen en la estructura cristalina de dicho óxido en estado anatase, obstruyendo la transformación de dicha fase a rutile, inhibiendo el crecimiento del grano.
También se ha reportado que el dopado con niobio aumenta la sensibilidad del dióxido de titanio hacia el oxígeno. El óxido dopado también presenta una impedancia más baja a temperaturas de trabajo menores, por lo que se facilita el diseño de la electrónica asociada al sensor.
Basándonos en estos puntos, uno de los objetivos de esta tesis fue la síntesis, mediante un proceso de sol-gel, de diferentes tipos de dióxido de titanio para la fabricación de un sensor de oxígeno que opere en un margen de temperaturas moderado (300 ºC - 600 ºC). Para ello se desarrollaron óxidos dopados con un 3 % de niobio y óxidos sin dopar.
Cada uno de ellos fue calcinado a diferentes temperaturas: 600 ºC, 700 ºC, 800 ºC y 900 ºC.
Con el objetivo de correlacionar la estructura y la sensibilidad y selectividad de los óxidos sintetizados, estos se sometieron a diferentes técnicas de caracterización. La espectroscopia de Plasma Acoplado Inductivamente (ICP) fue empleada para determinar la composición química de las muestras y cuantificar la porción de cada componente. La Difracción de Rayos-X (XRD) fue usada para establecer las fases presentes en la estructura cristalina del material y para determinar el tamaño el tamaño de los cristales en cada material. Se realizaron medidas de área BET con los nanopolvos para conocer el área superficial y la porosidad de cada material. La Microscopia de Escáner de Electrones (SEM) fue aplicado para obtener detalles de la estructura de la capa y del tamaño de las partículas.
Para hacer un análisis cuantitativo y cualitativo de las capas se utilizó la Espectroscopia de Dispersión de Energía por Rayos -X (EDS).
Para realizar las medidas, se desarrolló un substrato de alúmina para ser usado en el sensor de oxígeno. Este substrato puede soportar cuatro capas activas trabajando a la misma temperatura formando una matriz de sensores. Sin embargo, debido a algunos problemas relacionados con el encapsulado del substrato cuando las temperaturas de trabajo sobrepasaban los 450 ºC, su empleo para el sensor de oxígeno se retrasó y los resultados obtenidos con el mismo no estaban lo suficientemente completos para ser presentados en este trabajo. Para solventar la necesidad de un substrato que pudiese resistir altas temperaturas para aplicaciones de detección de oxígeno, se introdujo un nuevo substrato adquirido en el Instituto Kurchatov (Moscú - Rusia). Los resultados expuestos en este trabajo se obtuvieron con este último substrato.
Una vez fabricados los sensores, las capacidades de sensado de los materiales fueron probadas. La sensibilidad hacia el oxígeno fue medida en tres situaciones diferentes: 20 ppm de O2 en balance de N2, 30 ppm y 15 ppm de O2 en balance de CO2. Las respuestas de los sensores hacia otros gases contaminantes (SO2, CH4, H2S y C2H4) también fueron probadas para observar la influencia de dichos gases en el proceso de detección de oxígeno.
La mejor respuesta hacia el oxígeno se consiguió con los sensores basados en materiales dopados calcinados a 700 ºC. Esto puede atribuirse a los iones de niobio que inhiben el crecimiento del grano y por lo tanto producen un aumento del área superficial de dichos materiales, que beneficia al mecanismo de detección, y a su fase cristalina, mayoritariamente anatase, la cual permite la detección a las temperaturas deseadas (300 ºC - 600 ºC). Las especies de niobio también contribuyen al proceso de catálisis.
Otro motivo puede ser la fase cristalina, mayoritariamente anatase, la cual permite la detección a las temperaturas deseadas (300 ºC - 600 ºC).
Por otro lado, los materiales dopados calcinados a 600 ºC tuvieron una pobre respuesta, a pesar de que estos tienen mejores características físicas que los calcinados a 700 ºC. Este hecho puede explicarse por la presencia de algunos depósitos de carbono, residuales del proceso de síntesis, que no pudieron ser eliminados durante la calcinación. La presencia de estos depósitos fue confirmada mediante los análisis Raman de estos materiales.
Ya que estas estructuras de carbono cubren gran parte de la superficie del material, y que además son poco catalíticas, el resultado es una desactivación de la catálisis.
Concerniente a las medidas realizadas en atmósfera de CO2, los óxidos dopados con niobio y calcinados a 700 ºC también respondieron al oxígeno. Sin embargo, la respuesta hacia 15 ppm de oxígeno presentó una inversión de tipo oxidante a tipo reductora. A bajas concentraciones de oxígeno, los iones de CO-, provenientes de la disociación del CO2, se adsorben en la superficie del material activo. El oxígeno, en lugar de deplexionarse, interactúa con estos iones para formar CO2, liberando electrones a la capa activa, dando lugar así a una respuesta reductora. Por otra parte, las respuestas hacia otros gases contaminantes tales como H2S, SO2 y CH4 fueron de tipo reductora, como era esperado, lo cual indica que el cambio de respuesta no puede atribuirse al cambio del óxido conductor de tipo n a tipo p, sino más bien a un cambio en la naturaleza de la reacción.
Para mejorar la sensibilidad de los óxidos dopados, se intentó incrementar su área superficial y porosidad usando un surfactante como plantilla o molde durante el proceso de síntesis. El surfactante empleado fue dodecylamina, la cual forma una estructura micelar que hace de molde durante el proceso de nucleación del óxido, generando así granos menores con mayor área superficial y mayor porosidad. Entre tres diferentes intentos, los mejores resultados se obtuvieron cuando 8ml de dodecylamina fueron agregados a la solución del sol-gel inmediatamente después de la hidrólisis de los alkóxidos. Los análisis de XRD de este material mostraron que la adición de surfactante retarda aun más la transición de fase de anatase a rutile y también evita el crecimiento de los cristalitos. Estos resultados son soportados por las microfotografías de SEM y los análisis BET, los cuales mostraron un retardo en el crecimiento del grano y un incremento del área superficial. El área BET también evidenció un incremento de la porosidad del óxido. Sin embargo, los resultados de las medidas de oxígeno revelaron una pobre respuesta de los sensores basados en el óxido en cuestión. Los espectros Raman de este material mostraron algunos picos correspondientes a carbono con diferentes morfologías. Como fue explicado previamente, estos depósitos de carbono retardan la respuesta de este óxido al oxígeno.
Control of oxygen levels is a critical step in many industrial processes. In some of these processes, levels of oxygen must be detected and controlled even in ppm range.
Although there are several probed methods for oxygen detection in these control systems, most of them are expensive and complex. More accessible methods as Lambda sensors or electrochemical cells present also problems: first ones need high concentration of oxygen, or an extremely accurate temperature control, to work without interference from other gases.
Second ones may be affected by prolonged exposure to "acid" gases such as carbon dioxide and it is not recommended for continuous use in atmospheres which contain more than 25% of CO2.
Due to many advantages as low cost, small size and robustness, semiconductor sensors appear as a good solution for oxygen detection. Some authors had reported detection of oxygen at ppm levels employing this kind of sensors. However, most of them were made through thin film technology. For industrial applications, the most usual technology is thick film, because it is easier to fabricate and to dope than thin film sensors. For thick film sensors, detection of traces of oxygen is still a very difficult goal to reach and usually high temperatures (>700 ºC) are needed.
The basic oxygen sensitivity mechanism of oxygen sensors based on semiconductor oxides is their strong dependencies of electrical conductivity on oxygen partial pressures.
Titanium dioxide is the semiconductor material most widely used for oxygen detection.
Titania (usually rutile crystalline phase) based sensors need to work at high temperatures (700 ºC - 1000 ºC), since oxygen detection in rutile state is mainly due to diffusion of oxygen ions in the bulk of the material. For bulk reaction it is necessary to work at these high temperatures, leading to high power consumption, which is not desirable for electronic applications.
On the other hand, anatase state Titania has more free electrons. So, oxygen detection can be associated to surface reactions, which take place at not so high temperatures (400 ºC - 500 ºC). Then, it can be derived that maintaining an anatase structure would allow the detection of oxygen at medium temperatures, which is desirable for sensor design.
It had been reported that when Titania is doped with pentavalent impurity ions, i.e. Nb5+, such ions get into the anatase Titania crystalline structure, giving rise to a hindering in the phase transformation to rutile and an inhibition in grain growth. It has been also reported that Nb-doped Titania shows higher sensitivity towards oxygen than pure TiO2. The doped material also shows lower impedance at low operating temperatures and hence, it is easier to design associated electronic circuitry.
In this work pure Titania and Niobium doped Titania nanopowders were synthesized by a simplified sol-gel route. Based on the literature, the doping concentration in doped materials was set to Nb/Ti = 3 at%. In order to set the crystalline structure of the active materials, they were calcined at four different temperatures: 600 ºC, 700 ºC, 800 ºC and 900 ºC.
The obtained materials were characterized by different techniques. The objective of these characterizations was to obtain information about the material structure that could be related to its detection properties. Inductively Coupled Plasma (ICP) spectroscopy was employed to determine the chemical composition of the samples and quantify the amount of each component. X-ray Diffraction (XRD) was used to establish the phases present in the crystalline structure of the material and to determine the size of the crystallites in each material. Area BET measurements were done to nanopowders to know the surface area and the porosity of each material. Scanning Electron Microscopy (SEM) was used to obtain details on the film structure and the grain size. To make quantitative and qualitative analysis, Energy-Dispersive X-ray Spectroscopy (EDS) was also applied.
For the measurements, an alumina substrate was developed to be used in the oxygen sensor. This substrate can support four active layers working at the same temperature forming a sensor array. However, due some problems related with the substrate package at working temperatures above 450 ºC, the use of this substrate for oxygen sensor application was delayed and the results obtained whit it were not complete enough to be presented in this work. In order to resolve the necessity of a substrate that can resist high working temperatures for oxygen sensing applications, it was introduced a new substrate acquired from the Kurchatov Institute (Moscow-Russia). The results exposed in this work were obtained by using this last substrate.
Using these substrates, the sensing capabilities of the materials were also tested. The sensitivity toward oxygen was measured under three different conditions: 20 ppm of O2 in N2 balance, 30 ppm and 15 pmm of O2 in CO2 balance. The sensors responses toward other pollutant gases (SO2, CH4, H2S and C2H4) were also tested in order to see the influence of such gases in the oxygen detection process.
The best response toward oxygen was achieved in those sensors based on Nb-doped materials calcined at 700 ºC. This may be attributed to the niobium ions which hinder the grain growth and hence give rise to a high surface area that benefits the detection mechanism. The good response may be also attributed to the crystalline phase, mostly anatase, which allows the detection at desire temperatures (300 ºC - 600 ºC).
On the other hand, the doped materials calcined at 600 ºC had a poor response, in spite of they have better physical characteristics than doped materials calcined at 700 ºC.
This problem was mainly related to some carbon deposits detected in these materials trough Raman analyses. Such carbon deposits may be residual of the synthesis process, which could not be eliminated during the calcination. Since these carbon structures cover a great part of the material surface, and they are also poorly catalytic, the result is a deactivation of the catalysis.
Concerning to the measurements carried out under CO2 atmosphere, Nb-doped Titania calcined at 700 ºC also responded to oxygen. However, the response toward 15 ppm of oxygen presented an inversion from oxidative type to reductive type. At low oxygen concentrations, the CO- ions from CO2 disassociation are adsorbed on the active material surface. The oxygen, instead deplexing, interacts with these ions forming CO2, liberating electrons to the active layer giving rise to a reductive type response. On the other hand, the responses toward other pollutant gases such H2S, SO2, C2H4 and CH4 were of reduction type, as they were expected, which support the idea that the change in the response type is not due to the change in the physics of the semiconductor oxide from n type to p type, but to a change in the reaction nature.
In order to improve the sensing capabilities of doped materials, it was attempted to increase their surface area and porosity by using a surfactant as a template during the synthesis process. The surfactant employed was dodecylamine, which forms a micellar structure that works as template in the nucleation process of the oxide, generating small grains with higher surface area and porosity. Among three different attempts, the best results were obtained when 8 ml of surfactant were added to the sol-gel solution just after the hydrolysis of the metal alkoxides. XRD analyses of this material showed that the addition of surfactant retards even more the phase transition from anatase to rutile and also hinder the crystallites growth. These results were supported by SEM micrographs and BET analysis, which show a hinder in grain growth and an increase of the surface area. Area BET also evidences an increment of the material porosity. However, the results of the oxygen measurement reveal a poor response of the considered material. The Raman spectroscopy of this oxide shows some peaks that correspond to carbon with different morphologies. As it was explained before, these deposits of carbon retard the response of this material toward oxygen.

The aim of this project was to develop a thick film sensor made of zinc oxide that would accurately respond to various fluxes of atomic oxygen (AO), allowing it to be regenerated after saturation. The sensors were manufactured using a thick film technique of screen printing over a substrate (alumina) that is inert to the action of AO; some of sensors were developed with pure ZnO while others used a binder to the substrate. The expectation has been that impinging atomic oxygen is captured upon the zinc oxide which will consequently increase its resistance; this change is an indicator of the AO flux. By suitable heating, the adsorbed atomic oxygen atoms are released from the sensor and its original properties are then restored. This thesis describes the manufacturing of the sensors as well as all the tests conducted so as to characterize and determine the performance of the thick film ZnO sensors. It also includes the relevant conclusions to this work. It has been concluded that the sensors respond to AO by increasing the overall DC resistance of the sensors; impedance spectroscopy reveals that an increase of the resistance of the grain boundaries of the crystallites that make-up the sensors to be the dominant process. AO flux can be measured by both resistance and impedance spectroscopy measurements. Sensors can be regenerated by suitable heating. However, each sensor needs to be independently calibrated before AO measurements can be achieved. A model to relate the change of DC conductance with AO flux is described, as well as an adaptation to use with impedance spectroscopy. UV radiation does not seem to have an effect on the resistance of the sensors, despite evidence provided by other researchers. This is the first time that thick film ZnO sensors have been used to measure AO flux, and also the first time that impedance spectroscopy has been used for this purpose and to characterize thick film ZnO sensors for outer space applications. Screen-printing of pure ZnO has not produced a sensor robust enough for outer space applications; the use of a glass binder shows promise to overcome this limitation.

Thesis(Ph.D.)--Case Western Reserve University, 2010
Title from PDF (viewed on 2009-12-22) Department of Chemical Engineering Includes abstract Includes bibliographical references and appendices Available online via the OhioLINK ETD Center

We have investigated the Faraday effect of bismuth-doped rare-earth iron-garnets with varying doping levels of gallium from z = 1.0 to 1.35. We used lutetium to control the film's in-plane magnetic properties and found that gallium doping levels above the compensation point caused a loss of anisotropy control, a canted out-of-plane magnetization in the film, and an extremely weak but linear coercivity above 10 micro-Tesla fields. Using these results we focused on in-plane films to create 8 layer stacks of 500 um thick films to achieve a minimum detectable field of 50 pT at 1 kHz. Unlike previous Magneto-Optic (MO) studies that typically used thin films of approximately 1um thickness, we used approximately 400um thick films to allow experimentation with the final, robust, ideal form the MO sensor would take. We measured what most other MO studies with garnets neglected: the magnetic anisotropy axis or structure within the film. Knowledge of this structure is essential in improving the sensitivity of a stacked MO probe. Studying thick films proved to be key to understanding the magnetic anisotropy and domain properties that can degrade or enhance the sensitivity of the Faraday rotation in bismuth doped rare-earth iron-garnets to an applied magnetic field and to pointing the direction of future research to develop the conditions for rugged magnetometer sensors.

Physics
Ph.D.
We have investigated the Faraday effect of bismuth-doped rare-earth iron-garnets with varying doping levels of gallium from z = 1.0 to 1.35. We used lutetium to control the film's in-plane magnetic properties and found that gallium doping levels above the compensation point caused a loss of anisotropy control, a canted out-of-plane magnetization in the film, and an extremely weak but linear coercivity above 10 micro-Tesla fields. Using these results we focused on in-plane films to create 8 layer stacks of 500 um thick films to achieve a minimum detectable field of 50 pT at 1 kHz. Unlike previous Magneto-Optic (MO) studies that typically used thin films of approximately 1um thickness, we used approximately 400um thick films to allow experimentation with the final, robust, ideal form the MO sensor would take. We measured what most other MO studies with garnets neglected: the magnetic anisotropy axis or structure within the film. Knowledge of this structure is essential in improving the sensitivity of a stacked MO probe. Studying thick films proved to be key to understanding the magnetic anisotropy and domain properties that can degrade or enhance the sensitivity of the Faraday rotation in bismuth doped rare-earth iron-garnets to an applied magnetic field and to pointing the direction of future research to develop the conditions for rugged magnetometer sensors.
Temple University--Theses

This work deals with thick-film sensor’s electrodes optimization. The main goal of this work is to find suitable design of the electrodes for the best and reproducible current response of thick-film sensors.

Master’s thesis deals with thick/film sensor’s electrodes optimization. The aim of this work is to determine how the geometric size and shape of the electrodes affect the output current response. Technology and electro-analytical method is described in the theoretical part. The practical part is aimed at thick-films electrodes and sensors optimalization. It is here each optimization TLV electrodes in terms of their impact on the size of the output current response three-electrode electrochemical system using standard electrodes. Similarly, there is a study of the effect size performed on one common electrode free-electrodes sensor. Finally, there are studied various forms of electrode area three-electrode thick-film sensor in terms of their effect on the output current response. Summarized results are described in the final part of the work.

Os materiais de estrutura perovskita compreendem um vasto grupo de compostos cuja estrutura na sua forma mais simples pode ser representada pela fórmula química ABO3. Uma das principais vantagens que a estrutura perovskita apresenta é o alto grau de flexibilidade em acomodar uma grande variedade de átomos nos sítios A e B, permitindo um maior controle de suas propriedades físicas e químicas bem como a manutenção de sua estrutura básica, mesmo para altas concentrações de átomos substituintes. Devido estas propriedades, estes materiais têm sido aplicados com sucesso como capacitores, varistores, fotoeletrodos, memóriasferroelétricas e sensores de gases. Nas últimas décadas, tem sido reportada a utilização do titanato de estrôncio (SrTiO3, ST) na forma de filmes finos e espessos como sensor de gás oxigênio em altas temperaturas (>500 °C). Recentemente, foi mostrado que o titanato de estrôncio dopado com Fe apresentava uma excelente sensibilidade ao gás ozônio. Entretanto, neste trabalho, somente uma composição foi caracterizada e nem todas as propriedades importantes em relação a esta aplicação foram completamente exploradas. Neste contexto, esta tese de doutorado teve como objetivo verificar o efeito da substituição do átomo de estrôncio pelo lantânio e do titânio pelo Ferro no composto SrTiO3 na forma de pó e filmes finos nanoestruturados com a finalidade de verificar o efeito destas substituições nas propriedades sensoras do material. Inicialmente, com o objetivo de avaliar a influência da adição de La na estrutura do composto ST, amostras na forma de pó do sistema Sr1-xLaxTiO3.(SLT) foram preparadas através do método dos precursores poliméricos. Posteriormente, pelo mesmo método, foram sintetizadas amostras do sistema SrTi1-xFexO3 (STF) e Sr1-XLaxTi1-yFeyO3 (SLTF). A partir das amostras na forma de pó e na forma de solução, filmes finos e espessos foram respectivamente obtidos através das técnicas de deposição por feixe de elétrons (EBD) e spin-coating (SC). As amostras nanoestruturadas na forma de pó e na forma de filmes foram caracterizadas por difração de raios X, espectroscopia de absorção de raios X (XANES) na borda K do Ti e do Fe e espectroscopia de fotoelétrons excitados por raios X (XPS). A análise morfológica foi realizada através das técnicas de microscopia eletrônica de varredura (MEV) e microscopia de força atômica (MFA). As amostras do sistema STF e SLTF na forma de filmes finos foram avaliadas em relação à sensibilidade aos gases O3, NO2, NH3 e CO. Os resultados indicaram que os filmes do sistema SLTF depositados pela técnica de deposição por feixe de elétrons apresentam uma maior sensibilidade ao gás ozônio, enquanto o filme de mesma composição depositado pelo método de spin-coating apresentou uma melhor estabilidade e tempo de recuperação em relação a este mesmo gás.
The perovskite structure materials comprise a large group of compounds with the structure in simple form can be represented by the ABO3. chemical formula. The main advantage that the perovskite structure presents is the high degree of flexibility to accommodate a wide variety of atoms in sites A and B, allowing a greater control of physical and chemical properties of the material, maintaining its structure even for a high concentrations of substituent\'s. Due to these properties, these materials have been successfully applied as capacitors, varistors, photoelectrodes, ferroelectric memories and gas sensors. In the last decade, strontium titanate (SrTiO3, ST) in the form of thin and thick films have been reported as oxygen gas sensor at higher temperatures (> 500 °C). Recently, strontium titanate doped with Fe was used as the first ozone sensor. However, the work was carried out only with a certain composition and some important properties for this application have not been fully exploited. In this context, this PhD thesis aimed to the synthesis of strontium titanate system in powder form and nanostructured thin films with the substitution of Sr for La and Ti for Fe. In order to verify the effect of these substitutions in material properties initially Sr1-xLaxTiO3 (SLT) samples were prepared in powder form by the polymeric precursor method in order to evaluate the influence of the addition of La in the structure of the compound ST. Subsequently, samples were synthesized from SrTi1-x Fex O3 (STF) and Sr1-XLaxTi1-yFeyO3 (SLTF) systems through the polymeric precursors, which were used for the deposition of thin and thick films, which were respectively obtained through electron beam deposition techniques (EBD) and spin-coating (SC). Samples in the form of nanostructured powder and thin films were characterized by X-ray diffraction, X-ray absorption spectroscopy (XANES) at Ti and Fe K-edges and by X-ray photoelectron spectroscopy (XPS). Morphological analysis was performed using the scanning electron microscopy (SEM) and atomic force microscopy (AFM) techniques. The STF and SLTF samples in a thin film form were evaluated towards their sensitivity to O3, NO2, NH3 and CO gases. The results indicated that SLTF films deposited by electron beam deposition technique exhibit higher sensitivity to ozone gas. However the same composition deposited by spin-coating showed a better stability and recovery time relative to the same gas.

A new torque transfer standard using metallic TBTF resonant sensor was developed to overcome the overload capability problem which occurs with conventional metallic resistance strain gauges. Previous research work, however, has shown that the first prototype of the metallic TBTF resonant sensor was not suitable for use in a torque transfer standard due to its size and subsequent sensitivity to parasitic lateral forces. To maximize the benefits from this sensor, particularly overload capability and long-term stability, in the high accuracy torque measurement application area, there is a need to develop significantly smaller devices. The aim of this thesis is to research through FEA modelling and experimental characterisation the key performance parameters required to produce a miniaturised metallic TBTF resonant sensor that provides better performance when applied in a torque measurement system. For high accuracy any torque transducer using these sensors ought to have low sensitivity to parasitic influences such as bending moments and lateral forces, which can only be achieved with reduced size. The problems with the existing design, key design issues, possible configuration and packaging solutions of the metallic TBTF resonant sensor that could be used for achieving a higher accuracy torque transfer standard are considered. Two designs of miniaturised metallic TBTF resonant sensors, SL20 and SL12, are considered and experimentally investigated. The lateral forces are reduced by 52% for SL20 design and by 80% for SL12 design when compared to the original SL40 design. A torque transducer using the SL20 design was calibrated falling into the Torque Transfer Standard class of accuracy 1 category, uncertainty 0.8%. A torque transducer using the SL12 design was made and calibration showed a class of accuracy 0.5 category, uncertainty 0.2%. The results from this research indicate that the SL12 design is suitable for use in a torque transfer standard. The SL12 design is optimal and the smallest size possible based on the overload capability design criteria requiring the tine cross sectional area to remain constant.

Submitted by CAMILA PAIXÃO SANTOS (camila.paixao@yahoo.com.br) on 2016-11-03T08:48:02Z No. of bitstreams: 1 DEFESA_FINAL.pdf: 1918396 bytes, checksum: 8ba642a529d4534993d8d024baa1c739 (MD5)
Rejected by Felipe Augusto Arakaki (arakaki@reitoria.unesp.br), reason: Solicitamos que realize uma nova submissão seguindo a orientação abaixo: O arquivo submetido está sem a ficha catalográfica. A versão submetida por você é considerada a versão final da dissertação/tese, portanto não poderá ocorrer qualquer alteração em seu conteúdo após a aprovação. Corrija esta informação e realize uma nova submissão com o arquivo correto. Agradecemos a compreensão. on 2016-11-10T12:39:58Z (GMT)
Submitted by CAMILA PAIXÃO SANTOS (camila.paixao@yahoo.com.br) on 2016-11-11T18:37:20Z No. of bitstreams: 1 DEFESA_FINAL.pdf: 2077990 bytes, checksum: 7c050271b8596995c2ed6deeef0aca4e (MD5)
Approved for entry into archive by LUIZA DE MENEZES ROMANETTO null (luizaromanetto@hotmail.com) on 2016-11-16T19:53:18Z (GMT) No. of bitstreams: 1 santos_cp_me_guara.pdf: 2077990 bytes, checksum: 7c050271b8596995c2ed6deeef0aca4e (MD5)
Made available in DSpace on 2016-11-16T19:53:18Z (GMT). No. of bitstreams: 1 santos_cp_me_guara.pdf: 2077990 bytes, checksum: 7c050271b8596995c2ed6deeef0aca4e (MD5) Previous issue date: 2016-09-03
Não recebi financiamento
Este trabalho apresenta e discute o uso do óxido de cério na fabricação de filmes espessos por “screen printing” para aplicações em sensores de gás. Nesse estudo o CeO2 puro foi obtido pelo método dos precursores poliméricos utilizando como resina precursora o citrato de céria. O “puff” – espuma resultante da primeira fase do tratamento térmico da resina- foi calcinado a 550, 600, 700 e 750°C. O pó foi caracterizado por termogravimetria (TG) e as propriedades estruturais, morfológicas foram avaliadas por difratometria de raios X (DRX), espectroscopia Raman, área de superfície por isotermas Brunauer, Emmett e Taller (BET) e microscopia eletrônica de varredura (MEV). A resposta sensora foi estudada em uma câmara de teste construída no Laboratório de Catálise e Superfícies do Instituto de Ciência de Tecnologia de Materiais (INTEMA) da Universidade de Mar del Plata. A TG mostrou a formação de óxido de cério a 550°C, temperatura relativamente baixa quando comparada com outros métodos. Mediante DRX todas as amostras mostraram picos correspondentes à fase pura de CeO2 o qual cristaliza em uma estrutura cúbica do tipo fluorita, entretanto, maiores temperaturas de calcinação mostraram aumento da cristalinidade e tamanho do cristalito. No espectro Raman, um forte pico em torno do 461 cm-1 foi detectado, atribuído às vibrações simétricas do Ce-O. A área de superfície BET dos pós foi de 301, 77 m2/g o que evidencia a formação de partículas muito pequenas e altamente reativas. As micrografias obtidas por MEV mostram a presença de diferentes tamanhos na forma de aglomerados. A caracterização da resposta sensora mostrou que o sensor fabricado a partir de pós de CeO2 puro apresenta um bom tempo de resposta, alcançando a melhor performance com temperatura de trabalho de 400 °C, tanto em atmosferas redutoras e oxidantes. A característica principal observada foi que os resultados são dependentes dos ciclos anteriores, a reprodutibilidade do sistema é garantida quando se apaga a “memória” do sistema, expondo-o ao vácuo.
This paper presents and discusses the use of cerium oxide in the production of thick films for "screen printing" for applications in gas sensors. In this study the pure CeO2 was obtained by the polymeric precursor method using as a precursor resin citrate ceria. The "puff" - resulting foam from the first stage of thermal treatment of the resin-calcined at 550, 600, 700 and 750 °C. The powder was characterized by thermogravimetry (TG) and structural, morphological were evaluated by X-ray diffraction (XRD), Raman spectroscopy, isothermal Brunauer surface area, Emmett and Taller (BET) and scanning electron microscopy (SEM) . The sensor response was studied in a test chamber built in the Laboratory of Catalysis and Surface Materials Technology Institute of Science (INTEMA), University of Mar del Plata. The thermogravimetry showed the formation of cerium oxide at 550° C, relatively low temperature compared with other methods. Upon XRD all samples showed peaks corresponding to pure CeO2 phase which crystallizes in a cubic fluorite type structure, however, higher calcination temperatures showed increased crystallinity and crystallite size. In the Raman spectrum, a strong peak around 461 cm-1 was detected, assigned to symmetric vibrations of the Ce-O. The BET surface area of the powders was 301, 77 m2 /g which shows the formation of very small and highly reactive particles. The SEM micrographs show the presence of different sizes in the form of agglomerates. The characterization of the sensor response showed that the sensors manufactured from pure CeO2 powder has a good response time, achieving better performance at 400 °C working temperature in both reducing and oxidizing atmospheres. The main feature observed was that the results are dependent on previous cycles, the system reproducibility is guaranteed when it deletes the "memory" of the system, exposing it to vacuum.

Miniature electrochemical sensors can be produced by thick-film technology. The optimal properties of thick-film electrode system, the production technology optimization of thick-film sensors and the adjustment of optimal technological properties are main problems of sensor design. An area of working electrode and an electrode topology design is the important parts in design of electrode system optimization. The rules for optimization of electrode systems in classical electrochemistry exist. The design of standard thick-film electrode construction is limited (2-D construction, substrate dimensions, resolution, paste parameters …). Therefore rules for classical electrode systems cannot be used fully in this case and larger electrode area can be making by 3D structure only. One of resolutions is unconventional method. The increase of working electrode surface by unconventional method – ball and wire bonding is discussed in this work.

Biosensors have widespread applications in many areas. Currently the Surface Plasmon Resonance (SPR) biosensor is one of the most prevalent types of biosensor. However, it has several disadvantages such as being delicate, expensive, and non-portable. Ionic Self-Assembled Multilayers (ISAMs) adsorbed on Long Period Fiber Gratings (LPGs) provides an attractive platform for building optical sensors, which could potentially overcome the disadvantages of SPR biosensors. The ISAM technique is a type of layer-by-layer deposition technique for building nanoscale thin films. An LPG is a type of fiber device that is sensitive to physical property changes of the ambient environment. LPGs have been extensively investigated for use as optical sensors. We have carried out a study on combining these two techniques to build efficient biosensors. In this thesis, we demonstrate ultra-sensitive LPGs whose attenuation can be changed by 25 dB (~99.7%) over a 48-nm spectral band, with ambient-index changes of only 2.7E-4. The device schematic allows arbitrarily high index sensitivities to be achieved, which makes it an attractive platform for realizing sensors and modulators that respond to small index changes. For a thin-film coated LPG, we find theoretically that the resonant wavelength shift of the LPG can result from either the variation of the thickness of the film and/or the variation of its refractive index. Furthermore, results illustrate that the sensitivity of the sensor could be enhanced using a nm-thick thin-film (e.g. ISAM films) whose refractive index is greater than silica. Experimentally, we demonstrate the fabrication of nm-thick ISAM films deposited on LPGs, which induces dramatic shifts in the resonant wavelength. The refractive index and the thickness of the ISAM film was precisely controlled by altering the relative fraction of the anionic and cationic materials combined with layer-by-layer deposition. Finally, we demonstrate that ISAM-coated LPGs can function effectively as biosensors by using the biotin-streptavidin system. These demonstrations confirm that the ISAM-LPG scheme provides a thermally-stable, reusable, and robust platform for building efficient optical sensors.
Ph. D.

Biosensors have a multitude of important applications in basic research, environmental monitoring, biodefense, and medicine. This research aims to show that Ionic Self-Assembled Multilayers (ISAMs) adsorbed on Long Period Gratings (LPGs) can serve as a highly sensitive, robust, inexpensive optical-based biosensor platform. The ISAM technique is a layer-by-layer deposition method that builds nanometer-thick films based on the principle of Coulomb attraction between oppositely charged polyelectrolyte solutions while LPGs cause strong attenuation bands that enable an optical fiber to be extremely sensitive to changes in the surrounding environment. LPGs have been shown to be highly sensitive to the adsorption of nanoscale self-assembled films on the optical fiber cladding surface. In this work, we utilize Turnaround Point (TAP) LPGs, which possess even greater sensitivity than standard LPGs. This thesis focuses on evaluation of approaches to increasing the sensitivity of the sensor platfom, implementation of a biosensor for detection of several biomolecules, and preliminary evaluation of the potential for pH sensing. For a thin-film coated TAP LPG, we have demonstrated that shifts in the transmitted light intensity at the resonant wavelength of the LPG can result from the variation in film thickness and/or refractive index. We have observed decreases in intensity as large a 7 dB for one bilayer of ISAM film (~1 nm), which corresponds to an 80% decrease in the transmitted light intensity at the resonant wavelength. We have also shown that the sensitivity of the TAP LPG sensor can be increased by implementing nm-thick ISAM films that have a refractive index greater than silica. Furthermore, it is shown that incorporation of silica nanoparticles into the ISAM films significantly increases sensitivity through increased surface area and thickness. The biotin-streptavidin system was used as a model for implementaion and optimization of the ISAM-coated TAP LPG biosensor platform. Through evaluation of various biotin derivatives to maximize the amount functionalized onto the ISAM film, optimization of the ISAM film properties, and use of LPGs designed for higher sensitivity, the minimum detectable concentration of streptavidin was decreased from 0.0125 mg/ml to 12.0 ng/ml. The biosensor platform was then tested on prostate specific antigen (PSA), which is used as a clinical marker for early diagnosis of potential prostate cancer. Using a direct crosslinking approach of the monoclonal antibody to PSA into the ISAM film, a sensitivity level of 11.64 ng/ml PSA was obtained through combined optimization of the ISAM film and antibody surface coverage. Finally, the potential of ISAM TAP LPGs as pH sensors was examined based on the pH dependent swelling of ISAM films.
Ph. D.

The research on electrochemical sensor applications in combination with application-defined electronic circuits is a very fast developing field of modern sensorics. Device designing of unique integrated systems with specific properties and parameters is now becoming the everyday practice for producers of devices based on the latest technology progress in microelectronics. In the process of realization of a new device type, the main objective consists in increasing the possibility of detecting very low concentrations of heavy metals in the given chemical solution, extending the signal-to-noise ratio, and measuring the potential of a sensor. The general aim of this Master’s thesis is to design the topology of a sensor based on the thick film technology, to materialize the sensor, and to make experimental measurement using the sensor prototype. The practical reason for basing the measurement on the application of this thick film sensor is to find out whether the operational integration of the sensor and the circuit has been achieved according to the original hypothesis. Importantly, it is also necessary to determine whether the reproducibility of measured results has been reached. The innovative aspect of the created device consists in the improvement of output characteristics given by the integration of ASIC (IMAM CHIP). The integration of ASIC on the surface of the sensor eliminates the huge parasitic influence that affects the accuracy of measurement.

In recent years (Of late year)microelectronic technology significantly affected in the field of sensor technology, especially in the field of electrochemical sensors. This thesis deals with design and construction of thick-film voltammetric sensors array. The reason for proposal of voltammetric sensors array is to increase of speed, increase the accuracy and extension possibilities measurement compared with common systems.By reason of increase of sensitivity of these sensors it is efficient use the potentiostat chip for measurement output signal. Created sensors must fulfill the specified parameters and also requirements of thick-film technology. Next step is design the basic blocks of electronic circuit used to evaluate the measured values.

Tato diplomová práce se zabývá technologií tlustých vrstev a jejím možným použitím pro nekonvenční aplikace. První část práce je rešeršní a zaměřuje se především na aplikace tlustovrstvé technologie v různých oblastech použití. V teoretické části práce je dále popsána technologie LTCC, včetně technologických postupů výroby. V praktické části jsou řešeny dvě aplikace v tlustovrstvé technologii. Jedná se návrh a výrobu vysokofrekvenčního filtru a optického senzoru tlaku technologií LTCC. V závěru práce jsou provedena měření a jsou zhodnoceny dosažené výsledky.

This work covers an area of heavy metals detection by miniaturized thick film sensors. Working electrode of these sensors was fabricated using thick film paste that was prepared as a nanocomposition of carbon nanotubes powder, suitable vehicle and a thinner. The response of fabricated electrodes to cadmium and lead ions was measured using differential pulse voltammetry in an acetate buffer solution using a three electrode sytem. The results were compared with the carbon electrode made from special commercial thick film polymer paste. Finally the possibility of working electrode surface activation with nitric acid and heavy metals on real biological sample detection is described here.

The doctoral thesis is aimed at research of application possibilities of low-temperature co-fired ceramics, especially its non-conventional usage. It deals with particular topics ensue from electronic chips package design. The thesis also touches optoelectronic sensor application of this ceramics.

Mestrado em Engenharia de Materiais
O presente trabalho propôs-se comparar sensores semicondutores de gases baseados em diferentes morfologias do óxido de SnO2 para a deteção de monóxido de carbono, recorrendo aos métodos de síntese hidrotermal e deposição eletroforética. Este estudo visa entender a importância da morfologia na resposta elétrica e sensibilidade de um sensor semicondutor. Por alteração do precursor de Sn obtiveram-se partículas de SnO2 de diferentes morfologias (nanopartículas e nanobastonetes). Para além da morfologia estas nanoestruturas comportam-se diferentemente em termos de grau de cristalinidade, de distribuição, de tamanho de partícula e de estabilidade coloidal. Estas diferença a nível de suspensão afetam marcadamente a deposição eletroforética e a qualidade dos filmes obtidos. A deposição dos filmes é facilitada para as nanopartículas. Sensores de SnO2 que combinam síntese hidrotermal das nanopartículas com deposição eletroforética do óxido sensor apresentam valores de sensibilidade de 11 e 3 para concentrações de 2000 e 5000 ppm respetivamente.
The main goal of this work is the comparison of semiconductor gas sensors based on different morphologies of SnO2 for the carbon monoxide detection, using techniques such hydrothermal synthesis and electrophorectic deposition. This work compromises to study the importance of morphology in the electric response and sensibility of a semiconductor sensor. By changing the precursor Sn, SnO2 particles were obtained with differents morphologies (nanoparticles and nanorods). This different structures present different cristalinity, particle size distribution and colloidal stability. For each suspension, the results of electrophorectic deposition and the quality of the obtained films were very distinct. Nanoparticles Deposition is easier when compared to nanorods. SnO2 sensors that combine hydrothermal synthesis and electrophorectic deposition presente values of sensibility of 11 and 3 to 2000 and 5000 ppm of carbon monoxide concentration.

La détection de gaz, qui utilise aujourd'hui principalement des capteurs optiques, des capteurs électrochimiques à base de plomb et des capteurs catalytiques est un marché très porteur (estimé à 3 milliards d'euros) et doté d'une forte croissance (10% par an). La nécessité de développer de nouveaux systèmes d'instrumentation dédiés à la surveillance de la qualité de l'air intérieur et à la détection de substances dangereuses implique l'étude et le développement de nouveaux capteurs élaborés à partir de produits compatibles avec les enjeux environnementaux (RoHs, REACH), et économiques (matériaux à faible coût, techniques de réalisation fiables, durée de vie élevée). Le projet ANR OSPÉGAZ (Oxydes sans plomb pour la détection de gaz) vise à développer des systèmes d'instrumentations intégrés innovants dédiés à la caractérisation des différentes expositions environnementales en lien notamment avec les actions recommandées dans le cadre du PNSE2 pour les impacts sanitaires avérés. Le travail de thèse présenté dans ce manuscrit fait partie du projet OSPÉGAZ. L'objectif du travail a été, d'une part, de mettre au point un nouveau procédé d'élaboration d'encre au sein du laboratoire UDSMM pour l'élaboration et la caractérisation de films épais poreux, et d'autre part, de réaliser des capteurs de gaz à base de ces films. Nous avons choisi d'utiliser le matériau BaTiO₃, bien connu de la littérature, comme matériau de départ afin de mettre au point le procédé d'élaboration de couches épaisses. Par la suite, nous avons choisi le BaSrTiFeO₃ comme matériau sensible au gaz, et nous avons étudié deux compositions correspondant à deux taux de fer : Ba₀.₈₅Sr₀.₁₅Ti₀.₉Fe₀.₁O₃ (BSTF 10%) et Ba₀.₈₅Sr₀.₁₅Ti₀.₉₈Fe₀.₀₂O₃ (BSTF 2%). Ces matériaux ont été caractérisés dans une large gamme de fréquence (100 Hz à 1 MHz) et de température (25°C à 500°C). Les propriétés diélectriques en fonction de la fréquence et de la température ont été étudiées sur deux structures différentes d'électrodes : capacité parallèle (MIM) et capacité interdigitée (CID). Enfin, des démonstrateurs de capteurs de gas basés sur le principe des capteurs semi-conducteurs, ont été réalisés à partir de films épais poreux (BT, BST, BSFT 10% et 2%). Ces démonstrateurs ont été testés dans les locaux de la société SIMTRONICS sous différents gaz comme le monoxyde de carbone CO (200ppm), le sulfure d'hydrogène H₂S (50ppm) et le dioxyde de soufre SO₂ (20ppm) à 400°C et 450°C. Sous H₂S (50ppm), ils ont montré une plus grande sensibilité relative du BSTF (10%) (55.4%) par rapport au BSTF (2%) (48%) à 450°C. La bonne sensibilité relative et la réponse dynamique très intéressante montrent que le matériau BSTF dispose d'un potentiel très intéressant pour la détection de gaz. L'optimisation de la géométrie des capteurs, du taux de fer et de la température de fonctionnement devrait nous permettre d'améliorer les performances de nos démonstrateurs
Today gas detection, which now mainly uses optical sensors, electrochemical sensors based on lead, and catalytic sensors, is a very promising market (estimated at 3 billion euros) with a strong growth (10% per year). The need for new instrumentation systems dedicated to the monitoring of the air quality and to the detection of hazardous substances, requires the study and development of new sensors compatible with the European environmental standards : Restriction of the use of Hazardous Substances (RoHS) ; Registration, Evaluation and Authorization of Chemicals (REACh). The OSPÉGAZ project aims to develop innovative integrated instrumentations systems for the characterization of different environmental exposures linked to the actions recommended by the PNSE2 for proven health impacts. Our research project aims to develop innovating and cost-effective gas sensors containing lead-free oxides and dedicated to the detection of flammable gases and protection against toxic risks. The works of the thesis presented in this manuscript is a part of this project. The objectives were, firstly, to develop a new process for ink preparation in UDSMM laboratory, for the elaboration, electrical and physicochemical characterizations, of thick porous film, and secondly to make gas sensors based on these films. We chose to use the BaTiO3 (well-known material in literature) material as a first material in order to develop the process of thick film elaboration. After that, we chose the BaSrTiFeO₃ as gas-sensitive material, and we studied two compositions of Ba₁₋ ₓ Sr ₓ Ti₁₋yFeyO₃ with two different concentrations or iron : Ba₀.₈₅Sr₀.₁₅Ti₀.₉Fe₀.₁O₃ (BSTF 10%) and Ba₀.₈₅Sr₀.₁₅Ti₀.₉₈Fe₀.₀₂O₃ (BSTF 2%). Electrical characterizations were made in a wide range of frequency (100 HZ to 1 MHz) and temperature (25° C to 500° C). The dielectric properties as a function of temperature and frequency were studied using two different structures of capacitance : metal-insulator-metal (MIM) and interdigital electrodes (CID). Finally we have developed semi-conductor gas sensors based on BT, BST and BSTF (10% ; 2%) thick films. All our sensors were tested under different gases such as carbon monoxide CO (200ppm), hydrogen sulphide H₂S (50ppm) and sulfur dioxide SO₂ (20ppm), at various temperature, in the laboratory of SIMTRONICS SAS. We have measured the greatest relative sensitivity under H₂S (50ppm) gas ; 55.4% and 48% respectively for BSTF (10%) and BSTF (2%), at 450°C. Good relative sensitivity and very interesting dynamic responsesof BSTF show that the material has a great potential for the detection of gas. The optimization of the sensor geometry, iron rate and operating temperature should allow us to improve the performance of our demonstrators

This thesis deals with non-conventional applications in the area of thick film technology, especially with indication of the temperature by thermodynamic sensors. The work summarizes thermodynamic sensors theory and also contains production data of the universal thermodynamic sensor realized on ceramic substrate, including measurement of its parameters and heat radiation.

This thesis is focused on the effect of plasmatic oxidation of carbon working electrode on its electrochemical response. In the theoretical part, the thesis describes thick film technology, working carbon electrodes, cyclic voltammetry, contact angle measurement and plasma. Experimental part is dedicated to characterization of plasma modified electrode. Results are compare with unmodified electrodes.

This thesis describes optimization and modification of standard thick-film electrochemical sensor to be able to be used for detection in microvolumes. In the theoretical part, the thick film technology is described. The work is then focused on the wettability of surfaces followed by introduction to electrochemistry and electro-analytical methods. In the experimental part, screen-printing of various types of thick film pastes on ceramic substrates for determination of their wettability and the following modification of the sensors with thick-film paste and with Parylene vaporization can be found, as well as the basic electrochemical measurements in microvolume using the modified sensor. Finally the results are summarized in conclusion.

碩士
國立成功大學
材料科學及工程學系
102
The zirconia oxygen sensors (or lambda sensors) for emission control of automobiles commonly use a conical thimble design. However, the thimble-type lambda sensors are approximately 1 cm wide in diameter and 5 cm long. It takes 30 sec for heating the zirconia solid electrolyte to the desired operation temperature(~400℃). With new processing technologies, namely the co-firing of the laminate between the green sheets by tape casting, the thickness of (planar) sensors may be reduced to 2mm. The size of planar oxygen sensor in the study is 50mm x 6mm x 2mm. The planar-type oxygen sensor provides the advantages of less weight, shorter warm-up time, smaller size and faster response time. In this study, the thick film technology, including tape casting, screen printing electrode for heater patterns, laminating and co-sintering processes, is used to fabricate planar lambda oxygen sensors. The lamination of green tapes was conducted at 75℃with pressure 〈50MPa. La0.8Sr0.2MnO3 has been considered for new cathode materials to replace Pt since it has adequate electrical property and nearly the same coefficient of thermal expansion as YSZ. However, the results of this work show that oxygen sensor with LSM cathode exhibits longer response time than standard oxygen sensor using Pt electrode.

碩士
華梵大學
機電工程研究所
90
The biosensors are very unfavorable in measurement because the working electrode will be polluted after detecting chemical solution. In order to avoid this problem, the disposable sensors must be developed in the future. The screen-printed carbon electrode can support the requirement because of its cheap price and mass production. Besides, if the screen-printed carbon electrode can be suitable to the extended gate ion sensitive field effect transistor measurement system which can not be affected by nature light and easily package, the low cost biosensors will be expected. In this thesis, first the pH value was measured by carbon electrode. We use two electrodes — carbon and Ag/AgCl to measure the response. The response of pH value was obtained and the sensitivity is between 30mV/pH and 45mV/pH. Next, we detect the urea concentration variation. We change the measurement structure to detect the urea concentration. The electrodes are three carbons and we employ differential pair structure of instrument amplifier to obtain the output voltage response. The linear range of urea is from 5mg/dl to 80mg/dl. The limitation of output voltage is about 93mV. This value is suitable to the pH value variation system. So we can prove that the carbon electrode detected urea concentration also can be used in the potentiometric circuit. At last we study the biosensor readout circuit. In the article, the readout circuit type will be conferred and we make use of the VLSI technology to fabricate its readout circuit chip. Then we hope the chip can match with the screen-printed carbon electrode.