Historical Inputs of N-Nitrosodimethylamine to the Public Drinking Water Supply in Wilmington, Massachusetts

It has been hypothesized that N-nitrosodimethylamine (NDMA), a potent carcinogen that is present at high levels in the ground water beneath the Olin Chemical Company site in Wilmington (Massachusetts), has contaminated the town's water supply wells. NDMA was not manufactured or used at the site, nor is it known to be a byproduct of chemical manufacturing activities at the site; thus it is suspected that NDMA precursor compounds were present in waste materials released on the site, and that the precursors then reacted in the subsurface to form NDMA. We undertook this study to establish a timeline for the release of NDMA precursor compounds into the aquifer, investigate the likely reaction mechanisms and kinetics of NDMA formation, and estimate the travel time of NDMA from source areas to the town of Wilmington's water supply wells. Based on the history of chemical waste disposal at the site, there is strong evidence for nitrosation of organic nitrogen compounds as being a major pathway leading to NDMA formation at the site. In particular, the manufacture Nitropore 5 PT, a blowing agent, required the use of sodium nitrite as one of the ingredients and produced dimethyl formamide as a waste material. Dimethyl formamide can hydrolyze to form dimethylamine, which reacts with nitrite by the nitrosation pathway to form NDMA. The kinetics of NDMA formation by nitrosation are relatively rapid (timescales of hours-to-days), suggesting that NDMA was formed onsite soon after the precursors were dumped there. A ground-water model was constructed using MODFLOW to estimate NDMA travel times between the site and the town wells. The trajectories of individual particles introduced into the model at the NDMA source area were simulated assuming steady-state hydraulic conditions. Two transport simulations were carried out based on 1965 and 1973 as being alternative estimates of the earliest production dates of Nitropore 5 PT, the suspected major source of NDMA precursors. For the 1965 simulation, NDMA started to arrive in the wells in 1971, and reached a peak concentration at the Butters Row well 1 in 1980. The 1973 simulation resulted in a 1977 arrival and a peak concentration at the Butters Row well 1 in 1983. In both cases, the wellhead concentrations of NDMA decreased from their peak concentrations, but were not eliminated entirely, due to the capture of NDMA by nearby industrial wells. Although the wells were shut off in 2003, it is suspected that chemicals in the water supply may be linked to high rates of childhood cancer. Results from this study could be useful in determining whether temporal correlations exist between NDMA in drinking water supplies and disease incidence among Wilmington residents.