Development of a new simple fluorometric method for total N-nitrosamines analysis in water

N-nitrosamines are known as carcinogenic compounds and arouse today more and more attention as micropollutants and disinfection by-products. Indeed, they were newly found in several drinking waters, wastewaters and greywaters. It was proven that they can form during oxidative water treatment, such as chloramination. Among numerous complex and expensive methods to detect total N-nitrosamines in water samples, this project positions itself as an exploratory empirical study, which introduces a new promising, simple, rapid, lowcost, sensitive and selective method. After optimisations, it successfully combines a pre-UVC-photolysis with the subsequent formation of fluorophores, namely the N-hydroxydiazene tautomers. The photo-denitrosation produces NO2 – as common photo-product for all N-nitrosamines and the fluorophores are formed by the photoreaction between NO2 – and the added fluorescence probe (folic acid). This induces a fluorescence enhancement (compared to the folic acid intrinsic fluorescence), which was measured by a fluorometer. After optimizations, the feasibility of the method was tested individually for five different N-nitrosamines and for a mix of nine Nnitrosamines (EPA mix). Low limits of detection (LOD), namely 0.057, 0.06 and 0.08 ¹M, were reached for Nnitrosodimethylamine (NDMA), N-nitrosodiphenylamine (NDPhA) and the EPA mix, respectively. For the EPA mix, a good second order polynomial relationship (R2 È 0.99) was observed between the fluorescence enhancements and a wide range of initial concentrations (from 0.06 and 50 ¹M). For NDMA, a good linear correlation was observed (R2 = 0.99, range 0.06-50 ¹M). Coupled with a previously optimized SPE method, the LOD could even be lowered down to 8 and 12 ng(NDMA eq.)/l for NDMA and the EPA mix, respectively, i.e. close to the Californian NDMA notification limit for drinking water (10 ng/l). To ensure a reliable quantification and not only a detection of total N-nitrosamines, further developments should be done. This assay was finally applied to detect the total N-nitrosamines content of an effluent wastewater sample. To assess the reliability of the results, they were compared with a reliable chemiluminescence-based method. The corresponding results were unsatisfactory. Principal limitations were highlighted to be (1) the complexity of the UV-denitrosation of the N-nitrosamines; (2) the errors due to laboratory manipulations; (3) the complexity of the water matrix; (4) the unexpected behaviours of some N-nitrosamines. Suggestions for future developments are provided