In this study, several fluorescence methods and traditional precursor surrogates were compared for predicting the formation of trihalomethanes (THMs) and haloacetic acids (HAAs) for four different sources waters (two rivers and two lakes).  Precursors were quantified using three common measures: dissolved organic carbon, ultraviolet absorbance at 254 nm, and specific ultraviolet absorbance as well as by principal component analysis (PCA), peak picking, and parallel factors (PARAFAC) analysis of fluorescence spectra.  PARAFAC and PCA are multivariate methods which allow for representation of the entire fluorescence spectra in a reduced dimensionality.  PARAFAC is most commonly used for decomposition of three-way fluorescence spectra; however, it was hypothesized that the higher variance explained and orthogonality constraints of PCA would improve subsequent DBP modelling.  Based on multi-linear modelling of THMs and HAAs, principle component (PC) scores resulted in the lowest mean squared prediction error of cross-folded test sets.  Inclusion of PCs representative of protein-like material significantly decreased prediction error for both THMs and HAAs.  PARAFAC analysis did not identify a protein-like component and resulted in prediction errors similar to traditional NOM surrogates as well as fluorescence peak picking.  In contrast to PARAFAC, which identified highly interpretable components of the spectra, PCA resulted in a superior mathematical representation of the fluorescence spectra which decreased DBP modelling errors.  These results support the value of FEEM-PCA as a suitable method for DBP precursor quantification for use in predicting the formation of THMs and HAAs.

Comparison of fluorescence analysis methods for predicting disinfection by-product formation

248th National Meeting (2014)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.