Open Collections

Development of microfluidic-based olfaction system for nuisance sewer gas monitoring

University of British Columbia

Nuisance compounds such as hydrogen sulfide (H₂S) are the main cause of unpleasant odors and corrosion in sewer pipes. The available chemical treatment methods are expensive and have negative side effects. Hotspot monitoring, identifying specific locations for targeted treatment, minimizes the use of chemicals and their negative impacts. This thesis presents microfluidic-based artificial olfaction detectors for hotspot monitoring of nuisance sewer compounds. The detector is fabricated using a highly-selective microchannel coupled with a sensitive metal oxide semiconductor (MOS) sensor. To study the role of the detector’s dimension and the behavior of different analytes in the detector, a comprehensive 3D numerical model is developed. In essence the model simulates the effects of diffusion, surface adsorption/desorption, chemical reactions on the MOS sensor, and heat and momentum transfer due to the MOS heating element. After verification/validation of the numerical model, it is used to compare the effect of the above-mentioned phenomena on the detector response. The results show that diffusion is the main parameter affecting the response, followed by surface adsorption/desorption; heat and momentum transfer has a minimum effect on the response. The model is also used to investigate the effect of the detector’s dimension (including microchannel length, microchannel height, and MOS sensor housing volume) on its performance in terms of sensitivity, selectivity and the recovery time. The results show that there is a trade-off between selectivity and sensitivity; thus, a sum indicator is defined to investigate the overall performance for various conditions. To obtain the experimental responses for different concentrations of H₂S, a setup (consisting of a custom-built syringe pump, automated sample delivery, vaporization chamber, and detector) is developed. The detector responses are calibrated using gas chromatography and analyzed by feature extraction methods. To study the effect of humidity on the response, a gaseous-based setup is developed. The results of these two setups reveal that humidity damps the responses, indicating the importance of calibrating the detector against liquid samples. The model and experimental setup developed in this thesis can be used for generating a prodigious number of responses in the most cost-effective manner for distinguishing H₂S from mixture of several sewer gases.

Text

2018-09-20

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/67222

Mechanical Engineering

University of British Columbia

UBCO

http://creativecommons.org/licenses/by-nc-nd/4.0/