Natural organic matter removal in a full-scale drinking water treatment plant using ClO2 oxidation: Performance of two virgin granular activated carbons

The removal of different natural organic matter (NOM) components from boreal lake water was studied at ambient temperature (0.4-18.5 degrees C) in a full-scale drinking water treatment plant (DWTP). The specific aim was to compare the performance of granular activated carbon (GAC) filters containing different commercial carbons. Liquid chromatography with organic carbon detection (LC-OCD), fluorescence excitation and emission matrices (FEEM), and biodegradable dissolved organic carbon (BDOC) analyses showed that surface water NOM was dominated by humic substances (HS), had the highest fluorescence intensity at the region associated with fulvic acid-like NOM and contained, on average, 0.2 mg/L of BDOC. Coagulation-flocculation and flotation removed 61-82 % of HS and biopolymers and, on average, 50 % of the BDOC. Chlorine dioxide oxidation increased the proportion of HS with 18-22 % and the concentration of BDOC with, on average, 0.05 mg/L, while it decreased the proportion of low-molecular-weight neutrals (LMWN). Sand and GAC filtration removed BDOC to a level below 0.15 mg/L. The two studied GACs showed a 25 % difference in their cumulative total organic carbon (TOC) removals. The distributions of NOM fractions were similar in the effluents of both filters, with LMWN being the most efficiently removed fraction. NOM removal was due to adsorption rather than biodegradation in the filters. In conclusion, DWTPs benefit from complementary analyses, such as LC-OCD, FEEM and BDOC, for understanding and optimising NOM removal. Comparison of different GACs in full-scale studies is also important due to differences in the TOC removals of GACs with similar product specifications.

Automated summary

The study investigated the removal of different natural organic matter components from boreal lake water in a full-scale drinking water treatment plant using granular activated carbon filters containing different commercial carbons. Results showed that coagulation-flocculation, flotation, and chlorine dioxide oxidation were effective in reducing organic matter content, while sand and GAC filtration were efficient in removing BDOC. It was concluded that complementary analyses such as LC-OCD, FEEM, and BDOC are beneficial for optimizing NOM removal in DWTPs, and comparing different GACs in full-scale studies is important due to variations in TOC removal efficiencies.