Hydrazine detection limits in the cigarette smoke matrix using infrared tunable diode laser absorption spectroscopy

• Published in: Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy Vol. 58; no. 11; pp. 2505 - 2517
• Main Authors: Plunkett, Susan, Parrish, Milton E, Shafer, Kenneth H, Shorter, Joanne H, Nelson, David D, Zahniser, Mark S
• Format: Journal Article
• Language: English
• Published: England Elsevier B.V 01.09.2002
• Subjects: Cigarette smoke; Hydrazine; Hydrazines - analysis; Infrared tunable diode laser absorption spectroscopy; Nicotiana - chemistry; Smoke - analysis; Spectrophotometry, Infrared - instrumentation; Spectrophotometry, Infrared - methods; Hydrazine; Infrared tunable diode laser absorption spectroscopy; Cigarette smoke
• ISSN: 1386-1425
• DOI: 10.1016/S1386-1425(02)00068-9

Infrared absorption lines of hydrazine are broad and typically not baseline resolved, with line strengths ∼100 times weaker than the more widely studied compound ammonia. Hardware and software improvements have been made to a two-color infrared tunable diode laser (IR-TDL) spectrometer in order to improve the limit of detection (LOD) of hydrazine (N 2H 4) in the cigarette smoke matrix. The detection limit in the smoke matrix was improved from 25 parts-per-million-by-volume (ppmv) to 4.2 ppmv using a 100 m pathlength cell with acquisition of background spectra immediately prior to each sample and 100 ms temporal resolution. This study did not detect hydrazine in cigarette smoke in the 964.4–964.9 cm −1 spectral region, after mathematically subtracting the spectral contributions of ethylene, ammonia, carbon dioxide, methanol, acrolein, and acetaldehyde. These compounds are found in cigarette smoke and absorb in this spectral region. The LOD is limited by remaining spectral structure from unidentified smoke species. The pseudo random noise (root mean square) in the improved instrument was 2×10 −4 absorbance units (base e) which is equivalent to a 0.09 ppmv hydrazine gas sample in the multipass cell. This would correspond to a detection limit of 0.44 ppmv of hydrazine, given the dilution of the smoke by a factor of 5 by the sampling system. This is a factor of 10 less than the 4.2 ppmv detection limit for hydrazine in the smoke matrix, and indicates that the detection limit is primarily a result of the complexity of the matrix rather than the random noise of the TDL instrument.