Raman system for sensitive and selective identification of volatile organic compounds

Abstract

Analysis of volatile organic compounds (VOCs) for medical and industrial applications typically requires complex and expensive mass spectrometric systems to achieve the desired sensitivity and specificity. Raman spectroscopy enables specific compound identification based on distinct spectral fingerprints but traditionally has low sensitivity. We have developed a novel Raman system that provides VOC detection in the low ppm range. VOCs from the gas phase are absorbed and enriched in a thin polymer film coated on the surface of an optical sensor element. The VOC enriched polymer film is probed with evanescent wave excitation and optimized Raman signal collection. The herein described Raman sensor is low-cost, robust, readily manufacturable, and capable of providing high sensitivity without the nanostructured sensors required for surface enhanced Raman spectroscopy. To suppress background noise and interfering spectral features while improving our ability to resolve VOC mixture spectra, we applied two-dimensional (2D) correlation analysis to time series spectra acquired during VOC absorption into the polymer film. We established proof of principle through analysis of known VOC biomarkers for selected bacterial pathogens. The system enabled VOC detection with significantly higher sensitivity than commercial Raman probes. We observed good agreement between spectra of VOCs in the thin film obtained after 2D correlation analysis, and of pure VOCs using a commercial probe. Mixture spectra were further de-convoluted based on the off-diagonal peaks observable in the 2D Raman spectrum in conjunction with segmented data analysis. We anticipate that this system can be applied to a variety of medical, industrial, and biodefense applications.