A Rapid and\nSimplified Approach to Correct Atmospheric\nAbsorptions in Infrared Spectra

Infrared (IR) spectroscopy is a powerful analytical technique\nused\nto identify and quantify different components within a sample. However,\nspectral interference from fluctuating concentrations of water vapor\nand CO<sub>2</sub> in the measurement chamber can significantly impede\nthe extraction of quantitative information. These temporal fluctuations\ncause absorption variations that interfere with the sample’s\nspectrum, making accurate analysis challenging. While several techniques\nto overcome this problem exist in the literature, many are time-consuming\nor ineffective. We present a simple method utilizing just two sample\nspectra taken sequentially. The difference of these spectra, multiplied\nby a scaling factor, determined by minimization of the point-to-point\nspectral length, provides a correction spectrum. Subtracting this\nfrom the spectrum to be corrected results in a fully corrected spectrum.\nWe demonstrate the effectiveness of this method via the improved ability\nto determine analyte concentration from corrected spectra over uncorrected\nspectra using a partial least square regression (PLSR) model. This\ntechnique therefore offers rapid, effective, and automated spectral\ncorrection, which is ideal for a nonexpert user in a clinical or industrial\nsetting.