A new approach to removing interference of moisture from FTIR spectrum.

An approach is developed to remove the interference of moisture from FTIR spectra. The interference arises from two aspects: the fluctuation on the temperature of the HeNe laser and the fluctuation on the transient concentration of moisture in the light - path of an FTIR spectrometer. The temperature fluctuation on the HeNe laser produces a systematic spectral shift between single-beam sample and background spectra, which often makes spectral subtraction method invalid in removing the interference of moisture. Herein, the Carbo similarity metric (the C value) is used to reflect the subtle spectral shift. A database of single-beam background spectra is established based on the concept of big-data and the pigeon-hole theory. The spectral shift is corrected by selecting suitable single-beam background spectra from the database to match with the given single-beam sample spectrum according to the C value. The interference caused by the fluctuation of the transient concentration of moisture is removed using a comprehensive 2D-COS method. We apply the approach on two polymeric samples to retrieve high-quality spectra and reliable second derivative spectra without the interference of moisture. The present work provides a new opportunity of obtaining the reliable second derivative spectra in the spectral region masked by moisture.