Effect of different precursors and disinfection processes on the formation of nitrosamines

Abstract

In drinking water treatment, it is important to focus on public health and ensure the quality of the treated water. Disinfection is a must-process which is applied before supply of drinking water to the water supply system. Disinfection byproducts such as Trihalomethanes (THM), Haloacetic acids (HAA), and Nitrosodimethylamine (NDMA) are present in different water media, i.e., drinking water in treatment facilities, rivers, and wastewaters. The effect of NDMA on human health has become an important issue in recent years, and investigations are planned to define NDMA precursors and the conditions that lead to the formation of NDMA. Nitrosamines, the group of compounds NDMA belongs to, are carcinogenic and toxic compounds with negative effects even at low concentrations. Peter Magee's discovery led to investigations to determine their carcinogenic properties. Nitrosamines are found in industrial products and food industry, and can also form during disinfection of potable water. Although the MCL in drinking water are usually set as the concentration that increases cancer risk one in a million, the concentration of NDMA at which the public has to be warned is set to be 10 ng/L by California Department of Public Health (CDPH, 2022) since the concentration of NDMA causing a one in a million increase in cancer risk 0.7 ng/L (EPA IRIS 1993) is hard to measure properly. NDMA is a disinfection by-product of chlorination, chloramination and ozonation processes, with chloramination producing the highest amount. Precursors are organic nitrogen containing compounds that are converted to nitrosamines during these processes. This thesis aims to reveal the formation potential of NDMA at the end of disinfection processes from selected precursors in different types of water media. It is also aimed to compare the disinfection processes based on the NDMA concentration formed from the potential of precursors in different water matrices. The main steps of this thesis are NDMA method optimization, precursors potential determination, swimming pool monitoring, and surface water monitoring. Optimization of method used for NDMA includes the change of parameters such as mobile phase content and the gradient of them, and LC-MS/MS parameters such as tube lens offset, collision energy. Pharmaceuticals, swimming pool, and surface water were investigated to determine their potential to form NDMA. In addition to target analysis measuring NDMA, non-target analysis was also performed. Transformation products that formed during chloramination of Doxylamine were investigated and the compounds that lead to formation of other products and not NDMA were determined using LC-Q Exactive Hybrid Quadropole-Orbitrap MS and Compound Discoverer Programme version 2.1. NDMA formation in presence of pharmaceuticals which are known precursors resulted with a concentration of 1380 ng/L for Ranitidine, 344 ng/L for Sumatriptan, 156 ng/L for Doxylamine and 4 ng/L for Metformin. The mixture of all four pharmaceuticals resulted in formation of 1896 ng/L NDMA. The highest amount of NDMA was formed by Ranitidine as it was expected from the previous studies in the literature. In order to research the possibility of synergistic or antagonistic effects of the pharmaceuticals in terms of NDMA formation, binary combinations were also examined under formation potential test conditions and 1841 ng/L, 1623 ng/L, 1548 ng/L, 513 ng/L and 322 ng/L NDMA was formed respectively for the couples of Ranitidine-Sumatriptan, Ranitidine-Doxylamine, Ranitidine-Metformin, Sumatriptan-Doxylamine and Metformin-Doxylamine. Ranitidine dominated the formation of NDMA as it was the same with the single formation potential test. To simulate real conditions in the water treatment plants and water supply networks, chloramination was performed in lower concentrations of chlorine than formation potential test. As expected, the NDMA concentration formed during chloramination (Cl2/N ratio of 4:1; Cl2: 2 mg/L) was lower than the NDMA concentration obtained during the formation potential test. The time dependent change of NDMA concentration for each single pharmaceutical was observed and 3 different time intervals were chosen; 30 min, 2 h and 6 h. At the end of the 6h- period the highest amount of NDMA was formed for each of every pharmaceutical except Metformin. Metformin behaved differently and NDMA formed at the second hour decreased at the end of 6 h. In ozonation, the same behavioural trend for Metformin was observed and the concentration of the NDMA formed during ozonation was decreased after a while. Also for the ozonation process the other pharmaceuticals behaved same with Metformin and NDMA was degraded in the second period of application time. The possible effects between pharmaceuticals during formation of NDMA was also investigated under formation potential test conditions. The results from the tests conducted with binary combinations of pharmaceuticals suggested that the relationship between Doxylamine and Metformin was synergistic as they formed higher amount of NDMA when they were together. Other binary combinations of pharmaceuticals did not have synergistic or antogonistic effect when they were together during formation potential test. Sakarya river was investigated as an example of a surface water for the existence of NDMA and its precursors.