The possibility of simultaneous quantification of dilute aqueous solutions of five polycyclic aromatic hydrocarbons (PAHs) with substantial overlap of fluorescence spectra without preseparation was examined by using total synchronous fluorescence spectroscopy (TSFS) with multivariate methods like N-way partial least square (N-PLS), unfolded-PLS and multivariate curve resolution-alternating least square (MCR-ALS) analysis. The PAHs chosen were anthracene, benzo[a]pyrene, chrysene, perylene and pyrene. Two calibration sets were made using different approaches. Even with significant spectral overlap, the N-PLS, unfolded-PLS and MCR-ALS models were found to be robust for the quantification of all five PAHs in the calibration set where the amounts of PAHs were changed with the fixed relative ratio. The second calibration set, which is relatively difficult to analyse, consisted of samples where amounts of PAHs were changed randomly. It was found that even for this set, predictions were satisfactory. The three calibration models based on N-PLS, unfolded-PLS and MCR-ALS in the second set were relatively more robust for chrysene, benzo[a]pyrene and perylene compared to other PAHs. Root mean square error of calibration (RMSEC) and root mean square error of prediction (RMSEP) and regression parameters were calculated for all the three calibration models. The various parameters show that the combination of TSFS and N-PLS, unfolded-PLS and MCR-ALS analysis can be used for the simultaneous quantification of PAHs in water without any preseparation even when there is substantial overlap of fluorescence.
You have access to this article
Please wait while we load your content...
Something went wrong. Try again?