An application of the multivariate calibration technique of partial least-squares (PLS) regression to near-infrared spectra of a fiber-optic sensor based on the evanescent wave principle is presented. The sensing element consists of a quartz glass fiber with a silicone cladding which enriches nonpolar water contaminants. Due to the interaction of the extracted molecules with the part of the light which is transmitted in the evanescent wave zone of the cladding, absorbance spectra of the contaminants can be collected. In view of a sensor application for in-situ environmental analysis, aqueous solutions of chlorinated hydrocarbon solvents (CHS), which often can be found as major water contaminants, have been measured. PLS regression was applied to three sets of CHS samples, representing typical features of NIR evanescent wave spectral data. These are, e.g., strong overlapping of the absorption bands of different CHS components, peak distortions due to temperature variations between reference and sample measurement and noisy data at analyte concentrations near to the limit of detection, respectively. For trichloroethene and 1,1-dichloroethene, where the calibration model was built for samples within a small concentration range of 1–9 mg l−1, satisfactory prediction results could be obtained with a relatively small root-mean-square error of 0.3 mg l−1 compared to analytical reference measurements. In contrast to this, for a three component system of dichloromethane, trichloromethane and trichloroethene with strongly overlapping absorption bands, where samples over a very broad concentration range from 3–4940 mg l−1 were included in the PLS model, the prediction accuracy decreased enormously and for some samples strong deviations between real and predicted data occurred. Nevertheless, applying multivariate calibration to this difficult system with similar spectral features and huge differences in the concentration of the species allowed an acceptable spectral distinction and at least a semi-quantitative determination of the CHS species.
Read full abstract