Thesis presented October 19, 2018

Abstract:
Electronic noses are promising candidates for low-cost field monitoring of volatile organic compounds (VOCs). However, these techniques are still challenged with limitations. The use of a limited number of receptors prevents selective VOCs differentiation. A lack of reliability caused by pollution of receptors hinders their use for many applications. In this context, an electronic nose based on surface plasmon resonance imaging (SPRI) is developed for the detection of VOCs in the gas phase. This system consists of a broad array of receptors obtained by the self-assembly of organic molecules and peptides. SPRI optical transduction provides multiplexed monitoring of a large number of interactions in real time. The objective is to ensure the best possible chemical resolution to reliably differentiate VOCs. To this end, different strategies are explored to improve the sensitivity, selectivity and stability of the measurements. Thanks to these optimizations, our system offers a high selectivity for a large number of VOCs. VOCs are differentiated according to their chemical nature and structure. A good repeatability and a stability of at least two months are obtained. Preliminary tests show that our instrument is also effective for the analysis of more complex samples.

Keywords:
VOCs, cross-reactive receptor array, multivariate statistics, pattern recognition, SPRI