Determination of emerging chlorinated byproducts of diazepam in drinking water

Diazepam (DZP) is often found in source water and drinking water at dozens of nanograms per liter levels. The transformation of DZP in water chlorination disinfection process has aroused new concern because the toxic disinfection byproducts (DBPs) might be produced. However, the DBPs of DZP have not been fully identified, and their occurrence levels in drinking water have not been reported. In our chlorination experiment, five emerging DBPs of diazepam: (5-chloro-2-(methylamino) phenyl) (phenyl)methanone (BP-246), 6-chloro-1-methyl-4-phenylquinazolin-2(1H)-one (BP-271), N-(2-benzoyl-4,6-dichlorophenyl)formamide (BP-294), methyl-(2-benzoyl-4-chlorophenyl) (methyl)carbamate (BP-304 (1)) and 6-chloro-4-methoxy-1-methyl-4-phenyl-1,4-dihydro2H -benzo[d][1,3]oxazin-2-one (BP-304 (2)), were tentatively identified by high-resolution mass spectrometry and further characterized by nuclear magnetic resonance spectroscopy. We developed a trace analytical method for the analysis of these five DBPs in drinking water based on solid-phase extraction (SPE) followed liquid chromatography coupled with electrospray ionization tandem mass spectrometric detection. Ultrahigh sensitivities were achieved with limits of detection as low as 7 pg per liter. The recoveries at different spiking levels were all higher than 80% except for that of BP-246. Four of the DBPs and DZP were detected in real drinking water samples at concentrations ranging from several to dozens of nanograms per liter with relatively high detection frequencies. This is the first report on the existence of DZP-DBPs in drinking water. The method and results will be useful for further studies on the occurrence, toxicity, human exposure and control measures of these DBPs.

Reaction of diazepam and related benzodiazepines with chlorine. Kinetics, transformation products and in-silico toxicological assessment

In this work, the reaction of four benzodiazepines (diazepam, oxazepam, nordazepam and temazepam) during water chlorination was studied by means of liquid chromatography-quadrupole-time of flight-mass spectrometry (LC-QTOF-MS). For those compounds that showed a significant degradation, i.e. diazepam, oxazepam and nordazepam, parameters affecting to the reaction kinetics (pH, chlorine and bromide level) were studied in detail and transformation products were tentatively identified. The oxidation reactions followed pseudofirst-order kinetics with rate constants in the range of 1.8-42.5 M^-1 s^-1, 0.13-1.16 M^-1 s^-1 and 0.04-20.4 M^-1 s^-1 corresponding to half-life values in the range of 1.9-146 min, 1.8-87 h and 2.5-637 h for oxazepam, nordazepam and diazepam, respectively, depending of the levels of studied parameters. Chlorine and pH affected significantly the reaction kinetics, where an increase of the pH resulted into a decrease of the reaction rate, whereas higher chlorine dosages led to faster kinetics, as expected in this case. The transformation of the studied benzodiazepines occurs mainly at the 1,4-diazepine 7-membered-ring, resulting in ring opening to form benzophenone derivatives or the formation of a 6-membered pyrimidine ring, leading to quinazoline derivatives. The formation of these by-products was also tested in real surface water samples observing kinetics of oxazepam degradation slower in river than in creek water, while the degradation of the two other benzodiazepines occurred only in the simpler sample (creek water). Finally, the acute and chronical toxicity and mutagenicity of precursors and transformation products were estimated using quantitative structure-activity relationship (QSAR) software tools: Ecological Structure Activity Relationships (ECOSAR) and Toxicity Estimation Software Tool (TEST), finding that some transformation products could be more toxic/mutagenic than the precursor drug, but additional test would be needed to confirm this fact.