Evaluation of an evanescent fiber optic chemical sensor for monitoring aqueous volatile organic compounds

Linear chemometric algorithms were used to model the quantitative response of an evanescent fiber optic chemical sensor in aqueous mixtures with concentrations between 20 and 300 ppm. Four data sets were examined with two different experimental arrangements. Two data sets contained trichloroethene, 1,1,2 trichloroethane, and toluene. Partial Least Squares, PLS, and Principal Component Regression, PCR, algorithms performed comparably on these calibration sets with cross-validated root mean squared errors of prediction (RMSEP) for trichloroethene, 1,1,1 trichloroethane, and toluene of approximately 26, 29 and 22 ppm, respectively. The third data set contained trichloroethene, 1,1,2 trichloroethane, toluene, and chloroform and the fourth contained these four analytes as well as tetrachloroethene. Again, both chemometric algorithms performed comparably on a given data set with RMSEP for trichloroethene, 1,1,2 trichloroethane, toluene, and chloroform of approximately 6, 6, 9, and 16 ppm from the first set, and 7, 11, 13, and 31 ppm from the second set with tetrachloroethene RMSEP of 31 ppm. The decrease in the quantitative performance of the sensor for modeling toluene and chloroform upon addition of tetrachloroethene to the sample solutions is due to increased cladding absorption features in the spectral response matrix. These features overlap with the analyte absorption features of toluenemore » and chloroform. These results suggest one of the limitations with this type of sensing format.« less