Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of GraphSearch.
Many diseases including sick building syndrome, respiratory problems and cancers can be caused by exposure to high concentrations of toxins commonly used in modern building materials, paints, glues and furniture. To minimize exposure to these toxins, it is important to have air exchange with the outside environment. However, this is usually not energy efficient because the air must be heated or cooled. The purpose of the Intasense project was to resolve this problem by creating an air quality monitor which could determine the current quality of air within the building. This information would then be used to control the ventilation system within an energy efficient building. The purpose of this Ph.D. thesis was both to build a system which could be used to test and calibrate sensors used in the air quality monitor as well as to build its final fluidic platform. During this thesis, three fluidic platforms which include a humidity buffering system were designed, tested and constructed. The first platform was used to test an electrochemical sensor or metal oxide sensor in the laboratory. The second and third platforms held three electrochemical or metal oxide sensors, respectively. They are able to deliver equal quantities of gas to the sensors simultaneously. To date, the multisensor metal oxide platform has been used to create two versions of the Intasense Demonstrator and used to calibrate novel and commercial gas sensors. In both of these applications simultaneous delivery of equal quantities of the sample gas are required for optimal gas sensor signal processing: equal quantities are required so that the amount of analyte in contact with each sensor is the same and simultaneous delivery allows all three sensors to respond concurrently. This was achieved by creating a computational model of gas flow through the system, conducting a parameter sweep and then selecting the optimal set. To confirm that the model and machine error were not significant, a Latin square design was used to test the platform. This test showed that gas delivery to each channel was equal within statistical boundaries. Another major challenge was buffering humidity fluctuations without loss of analyte gas. This was necessary because metal oxide gas sensors are also sensitive to humidity fluctuations. To prevent a false positive caused by a rapid humidity increase, a reversible adsorbent was placed upstream of the gas sensors. However, there was concern that the adsorbent may remove analyte gasses. To optimize the tradeoff between humidity buffering and loss or delay of analyte signal, a computational model of adsorption and desorption of gasses within the filtration system was developed. It was experimentally tested using benzene, carbon monoxide, formaldehyde and nitrogen dioxide. To date ten platforms have been delivered to laboratories throughout Europe. Three are being used to test and calibrate gas sensors. The rest have been used to create demonstrators of the Intasense Air Quality Monitor.
Carlo Peruzzo, Claude-Alain Roulet, Lee Ann Nicol