Development of an Early Warning Fire Detection System using Correlation Spectroscopy

Combustion byproducts are numerous. A few examples of the gaseous byproducts include carbon dioxide, carbon monoxide, hydrogen chloride, hydrogen cyanide and ammonia. For detecting these chemical species, classic absorption spectroscopy has been used for many decades, but the sensitivity of steady-state methods is often unsuitable for the detection of trace compounds at the low levels (parts per million to parts per billion) appropriate for scientific purposes. This is particularly so for monitoring equipment, which must be compact and cost-effective, and which is often subjected to shock, vibration, and other environmental effects that can severely degrade the performance of high-sensitivity spectrometers in an aircraft. Steady-state techniques also suffer from a lack of specificity; the deconvolution of the spectra of complex mixtures is a laborious and error prone process. These problems are exacerbated in remote fiber-optic monitoring where, for practical reasons, the fundamental absorbance region of the spectrum (often between 3 and 8 microns) is inaccessible, and the low-strength, closely spaced, near-infrared overtone absorbance bands must be used. We circumvented these challenges by employing correlation spectroscopy, a variation of modulation spectroscopy.