Qualitative and quantitative analysis of volatile organic compounds using passive Fourier transform infrared emission measurements.

摘要

Analytical instrumental analysis has played a fundamental role in monitoring and enforcing environmental standards. The increase in industrial growth, among other factors, has increased the number and quantity of effluents that are released into the environment by various avenues. Instrumental analysis has been key in performing qualitative and quantitative determination of these pollutants. However the advent of better technology, especially computational technology, and more know-how has been an impetus to try and find better and more efficient ways to monitor these pollutants to satisfy today's demands of automated, real time and robust analytical techniques.; Classical methods have used techniques that require separation of the sample from the sample matrix before the analysis was performed. Fourier transform infrared (FTIR) spectrometry has gained ground as a viable technique for monitoring of various air pollutants. Attractive features about using FTIR spectrometry are its inherent selectivity and sensitivity for compounds that exhibit IR signatures. This has made the technique overcome the limitation of separation of the sample from the sample matrix consistent with classical methods. Another key feature of the technique is the ability to analyze pollutants remotely. Thus one does not have to directly sample the analyte.; High resolution FTIR data have been used at factory sites and in outdoor environmental monitoring of pollutants using relatively complex instrumentation. For ideal use for remote sensing the instrument needs to be rugged and transportable which is not consistent with high resolution data. The work that is reported in this dissertation explores data analysis techniques that use low resolution data collected in a passive mode that can be used in a more rugged instrument. Some of the problems that are encountered are the variation of the background, variations due to the background and signals from interfering species. This work tries to address some of these problems and some solutions have been shown. The use of information about interfering species to make a more robust model for qualitative and quantitative analysis of these air pollutants is also investigated.