RESUMO
The current study compared the impact of three different unit processes, coagulation, granular activated carbon (GAC), and a novel suspended ion exchange (SIX) technology, on disinfection by-product formation potential (DBPFP) from two UK lowland water sources with medium to high bromide content. Specific attention was given to the influence of the organic molecular weight (MW) fraction on DBPFP as well as the impact of bromide concentration. Whilst few studies have investigated the impact of MW fractions from Liquid Chromatography with Organic Carbon Detection (LC-OCD) analysis on dissolved organic carbon (DOC) removal by different processes, none have studied the influence of DOC MW fractions from this analysis on DBP formation. The impact of higher bromide concentration was to decrease the total trihalomethane (THM) and haloacetic acid (HAA) mass concentration, in contrast to previously reported studies. Results indicated that for a moderate bromide concentration source (135 µg/L), the THM formation potential was reduced by 22% or 64% after coagulation or SIX treatment, respectively. For a high bromide content source (210 µg/L), the THM formation potential removal was 47% or 69% following GAC or SIX treatment, respectively. The trend was the same for HAAs, albeit with greater differences between the two processes/feedwaters with reference to overall removal. A statistical analysis indicated that organic matter of MW > 350 g/mol had a significant impact on DBPFP. A multiple linear regression of the MW fractions against DBPFP showed a strong correlation (R2 between 0.90 and 0.93), indicating that LC-OCD analysis alone could be used to predict DBP formation with reasonable accuracy, and offering the potential for rapid risk assessment of water sources.
RESUMO
The UV/Chlorine process has gained attention in recent years due to the high quantum yield and absorbance of the chlorine species. However, there are still many unknowns around its application as a treatment for drinking water. The potential for the formation of disinfection by-products (DBPs) is one of them. There are no studies reporting on the formation of trihalomethanes (THMs) or haloacetic acids (HAAs) in complex matrices, such as real source waters, at UV wavelengths tailored to the UV/Chlorine process, which has been possible thanks to the development of light emitting diodes (LEDs). In addition, consideration of mitigation measures that might be needed after UV/Chlorine treatment for full scale application have not been previously reported. Specifically, the novelty of this work resides in the use of an innovative reactor using UV-LEDs emitting at 285 nm for the removal of three pesticides (metaldehyde, carbetamide and mecoprop), the evaluation of THM, HAA and bromate formation in real water sources by UV/Chlorine treatment and the mitigation effect of subsequent GAC treatment. A new parameter, the applied optical dose (AOD), has been defined for UV reactors, such as the one in the present study, where the irradiated volume is non-uniform. The results showed the feasibility of using the UV/Chlorine process with LEDs, although a compromise is needed between pH and chlorine concentration to remove pesticides while minimising DBP formation. Overall, the UV/Chlorine process did not significantly increase THM or HAA formation at pH 7.9-8.2 at the studied wavelength. At acidic pH, however, THM formation potential increased up to 30% after UV/Chlorine treatment with concentrations up to 60 µg/L. HAA formation potential increased between 100 and 180%, although concentrations never exceeded 35 µg/L. In all cases, GAC treatment mitigated DBP formation, reducing THM formation potential to concentrations between 3 and 16 µg/L, and HAA formation potential between 4 and 30 µg/L.
