Detection of mixtures of organic pollutants in water by polymer film receptors in fibre-optical sensors based on reflectometric interference spectrometry

A measurement system, which includes a polymer based sensor, a flow injection analysis set-up and a diode array spectrometer, was used to detect binary mixtures of organic pollutants in water. For our present examination, the organic pollutants are toluene, dichloromethane, 1,2-dichloroethane, and 1,1,1-trichloroethane. The response characteristics of the sensor to pure and mixed organic substances in water was investigated. By including the dynamics of the polymer/analyte interaction, i.e., the difference between transient signal profile shapes, the concentration of the components in binary mixtures was determined by multivariate techniques. The results of principal component regression, partial least-squares regression and multivariate linear regression were compared. The relative errors for the prediction of new samples were below 6%.