Nitrogenous disinfection by-product occurrence, formation, and stability in drinking water

Abstract

This research had three main objectives – to investigate the typical occurrence concentrations of nitrogenous disinfection by-products (N-DBPs) in drinking water in England, understand the organic precursors and water treatment conditions that are responsible for N-DBP formation, and investigate the factors which influence the stability of one class of N-DBPs, the halonitromethanes, in drinking water. For the England occurrence survey, 21 N-DBP compounds from four groups, haloacetamides (HAcAms), haloacetonitriles (HANs), halonitromethanes (HNMs) and cyanogen chloride, were included. Sampling was conducted quarterly in 20 water supply systems over a one-year period to investigate potential relationships between N-DBP concentrations and water quality parameter. The survey was the first of its kind to gather N-DBP concentration data in the England. The occurrence of N-DBP compounds was broadly similar to those reported in the US and occurred well below existing World Health Organisation guidelines for two of the HANs. The survey allowed new correlations to be observed between some N-DBP groups, but there were poor links with other individual water quality parameters or operational parameters (e.g. water age). Bromine substitution factors (BSFs) were also calculated for HANs and HAcAms, the first such report of BSFs for HAcAms; BSFs were similar for the two groups. In the laboratory, seven model amine precursor compounds were tested at three pH levels to investigate the effects of chlorine and chloramines on the formation of HANs and other DBPs. The formation of chloroform and HANs was reduced significantly by applying pre-formed chloramines compared to free chlorine, however the opposite was true for HNMs. The formation of DBPs from L-aspartic acid was particularly interesting, since chlorination and chloramination promoted DCAN or chloroform formation, respectively, suggesting different reaction pathways being favoured by the two disinfectants. The decay of two HNM compounds was also examined in ultrapure water, under three pH conditions, and with/without chlorine or chloramine residuals. Although stable in ultrapure water, the studied HNMs were shown to decay rapidly in the presence of disinfectant residuals, which may explain the low occurrence concentrations of HNMs that was observed in the occurrence survey. Mechanisms for the decay of HNMs in drinking water are proposed. An important practical finding of this research is that switching from chlorination to chloramination, a strategy that is widely proposed for the reduction of regulated DBPs such as THMs, might not be effective in controlling N-DBPs, and may perhaps even increase the formation of certain N-DBPs in some water types. While N-DBPs occur at much lower concentrations than THMs, previous research suggests that they are of higher geno- and cytotoxicity, therefore strategies to minimise currently regulated DBPs (e.g. trihalomethanes) should take into account the impacts of these strategies on N-DBP formation as well.