Detection, occurrence and risk assessment of N-nitrosamines in drinking water

One of the most successful public health measures ever implemented has been the disinfection of drinking water. Reactions between oxidants ( e.g. chlorine or chloramines) and source water natural organic matter produce chemical disinfection by-products (DBPs). Since detecting trihalomethanes in 1974, including chloroform---the first recognized DBP in drinking water---regulatory and public health concerns have focused on halogenated DBPs. Identification of specific halogenated DBP species that could reasonably explain observed epidemiological correlations between urinary bladder cancer and chlorinated drinking water exposure have yet to be made. N-Nitrosamines are a class of non-halogenated DBPs, which warrant consideration because the "probable" human carcinogen N-nitrosodimethylamine (NDMA) is a DBP that has been found in drinking waters throughout North America.; Extraction of ultra-trace concentrations of N-nitrosamines from water is an analytical challenge because these compounds are hydrophilic, polar compounds. Successful development of a selective and sensitive analytical method capable of detecting eight N-nitrosamines (detection limits: 0.4-1.6 ng/L) was achieved using solid-phase extraction (SPE) coupled with GC/MS, using ammonia positive chemical ionization. NDMA concentrations in drinking waters collected throughout Alberta ranged from non-detectable to 180 ng/L, representing some of the highest reported values in the absence of anthropogenic contamination. Additionally, N-nitrosopyrrolidine (up to 4 ng/L) and N-nitrosomorpholine (up to 3 ng/L) were reported for the first time as drinking water DBPs.; A survey of utilities serving a majority of Alberta's population prompted modifications to the SPE method, allowing simultaneous extraction of 10 samples. NDMA was detected in 30% of the 20 utilities surveyed, with two locations exceeding Ontario's Drinking Water Quality Objective of 9 ng/L. Most frequently NDMA occurrence was associated with chloraminating facilities.; Bench-scale disinfection experiments suggested maximum NDMA production occurs near the theoretical monochloramine maximum (1:1 Cl2:NH 3-N, M:M) in the sub-breakpoint region of the disinfection curve. Treatment conditions that included free-chlorine contact (2 hours) before ammoniation resulted in significant reductions in NDMA formation (up to 93%) compared to no free-chlorine contact time.; NDMA as a DBP does not represent a major route of exposure compared to dietary and commercial product sources. Drinking water risk assessments need to consider comprehensive NDMA exposure scenarios in order to truly protect public health.