Evaluation of multiple disinfection by-products in tap, filtered and bottled water in Barcelona, Spain

Abstract

BACKGROUND AND AIM: The formation of disinfection by-products (DBPs) as an unintended consequence of drinking water potabilization is a current global concern due to their potential relation to chronic diseases. With the aim of assessing human exposure to DBPs in drinking water, we quantified 4 trihalomethanes (THMs), 11 haloacetic acids (HAAs), 4 haloacetonitriles (HANs), chlorate, chlorite, 1,1-dichloropropanone (DCP), 1,1,1-trichloropropanone (TCP) and chloropicrin, in different drinking water types in Barcelona, Spain. METHODS: DBPs and physicochemical parameters such as conductivity and total organic carbon (TOC), were measured in tap water samples collected at 42 households, and in tap water filtered with activated carbon (AC) (N=6) and reverse osmosis (RO) (N=5). Selected DBPs were also analyzed in 10 popular bottled water brands. RESULTS:All tap water samples contained THMs and HANs, and 98% contained HAAs and chlorate above the limit of quantification (LOQ). Chlorite and TCP were found in 62 and 36% of the tap water samples, respectively, while DCP and chloropicrin were not detected. The median and interquartile range of total THMs, HAAs, HANs, and chlorate in tap water, calculated using values LOQ, were 42.0 (37.5-48.5), 18.0 (10.5-30.3), 3.2 (2.6-3.9), and 214 (159-296) µg/L, respectively. Conductivity was negatively correlated with bromodichloromethane (R=-0.91), TOC (R=-0.83), total HAAs (R=-0.82), dichloroacetonitrile (R=-0.78), TCP (R=-0.75), chlorite (R=-0.70), and positively correlated with the bromine incorporation factor (BIF) of THMs (R=0.82), DXAA (R=0.66), and TXAA (R=0.65) (X=Br/Cl). Both AC and RO filters reduced DBPs in tap water and none of the analyzed DBPs were detected in bottled water. CONCLUSIONS:THMs and HAAs were below the EU regulatory limits (Drinking Water Directive 2020/2184). A clear geographical pattern for most DBPs and for the conductivity was found, driven by the water source. With the generated data, we will develop predictive models based on routine monitoring parameters that could be applied in epidemiological research. KEYWORDS: Water quality, Chemical exposures, Exposure assessment, Multi-pollutant