Evaluation of the role of fixed beam open path Fourier transform infrared spectroscopy in air monitoring strategies.

Abstract

Industrial hygienists and other health and safety professionals rely on air monitoring to assess occupational exposures to gases, vapors and aerosols. As remote sensing instruments become more available, there is an increasing interest in the evaluation of advantages and possible applications of open path techniques in workplace and environmental monitoring. The goal of this dissertation was to elucidate the role of fixed beam OP-FTIR spectroscopy in monitoring gases and vapors in air, specifically as it applied to the assessment of personal exposure and the rapid detection of leaks or upsets. The utility of OP-FTIR data, which is spatially and temporally averaged, in comparison to traditional general area point or multipoint and personal monitoring was evaluated. In addition, issues of obtaining background spectra and performance of the OP-FTIR spectrometer in industrial environments were investigated. Time series modeling and process control techniques were used to determine if data collected in a controlled ventilation chamber and in an acrylonitrile, 1,3-butadiene and styrene process building were in-control. Time series were created by plotting consecutive air concentrations measured by the OP-FTIR spectrometer. Autocorrelation in time series data was removed using autoregressive moving average (ARMA) models. Once data were uncorrelated, control charts were used to rapidly detect a change from pre-defined normal operating conditions. A shift in conditions might indicate a leak, upset, fugitive emission or operator error. The models introduced could be used to sound an alarm, to increase ventilation in a monitored area, or to determine needs for respiratory protection. A model was also developed to use beam path averaged data to determine an upper bound on personal exposure. Worst case exposure conditions were assumed. A factor, $\gamma$, was introduced to account for mixing and source characteristics in a monitored area. Upper bound estimates were compared to breathing zone measurements and point samples. Data were collected in a controlled ventilation chamber and at a pharmaceutical manufacturing facility. Worst case assumptions were found to be too conservative in all monitoring situations, indicating that a better estimate of $\gamma$ was necessary. A method of creating a valid background spectrum to use with sample spectra collected during field studies was also addressed. Recommendations for further research included using multiple/moveable beams to increase spatial resolution, and development of guidance documents for utilization of beam path averaged data.