Design, fabrication and characterization of thick-film gas sensors

Identificador:  TDX:1902

Handle: https://hdl.handle.net/20.500.11797/TDX1902

Autors:  Tsolov, Peter

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