Trihalomethane formation during water disinfection in four water supplies in the Somes river basin in Romania

Abstract

After the discovery of chloroform in drinking water, an extensive amount of work has been dedicated to the factors influencing the formation of halogenated disinfections by-products (DBPs). The disinfection practice can vary significantly from one country to another. Whereas no disinfectant is added to many water supplies in Switzerland or no disinfectant residual is maintained in the distribution system, high disinfectant doses are applied together with high residual concentrations in the distribution system in other countries such as the USA or some southern European countries and Romania. In the present study, several treatment plants in the Somes river basin in Romania were investigated with regard to chlorine practice and DBP formation (trihalomethanes (THMs)). Laboratory kinetic studies were also performed to investigate whether there is a relationship between raw water dissolved organic matter, residual chlorine, water temperature and THM formation. Drinking water samples were collected from different sampling points in the water treatment plant (WTP) from Gilau and the corresponding distribution system in Cluj-Napoca and also from Beclean, Dej and Jibou WTPs. The water samples were collected once a month from July 2006 to November 2007 and stored in 40-mL vials closed with Teflon lined screw caps. Water samples were preserved at 4 degrees C until analysis after sodium thiosulfate (Na(2)S(2)O(3)) had been added to quench residual chlorine. All samples were analysed for THMs using headspace GC-ECD between 1 and 7 days after sampling. The sample (10 mL) was filled into 20-mL headspace vials and closed with a Teflon-lined screw cap. Thereafter, the samples were equilibrated in an oven at 60 degrees C for 45 min. The headspace (1 mL) was then injected into the GC (Cyanopropylphenyl Polysiloxane column, 30 m x 53 mm, 3 microm film thickness, Thermo Finnigan, USA). The MDLs for THMs were determined from the standard deviation of eight standards at 1 microg/L. The MDLs for CHCl(3), CHBrCl(2), CHBr(2)Cl and CHBr(3) were 0.3, 0.2, 0.3 and 0.6 microg/L, respectively. All kinetic laboratory studies were carried out only with water from the WTP Gilau. The experiments were conducted under two conditions: baseline conditions (pH 7, 21 degrees C, 2.5 mg/L Cl(2)) to gain information about the change of the organic matter in the raw water and seasonally variable conditions to simulate the actual process at the treatment plant and the distribution system. This study shows that the current chlorination practice in the investigated plants complies with the THM drinking water standards of the EU. The THM concentrations in all samples taken in the four treatment plants and distributions systems were below the EU drinking water standard for TTHMs of 100 microg/L. Due to the low bromide levels in the raw waters, the main THM formed in the investigated plants is chloroform. It could also be seen that the THM levels were typically lower in water supplies with groundwater as their water resource. In one plant (Dej) with a pre-ozonation step, a significantly lower (50%) THM formation during post-chlorination was observed. Laboratory chlorination experiments revealed a good correlation between chloroform formation and the consumed chlorine dose. Also, these experiments allowed a semi-quantative prediction of the chloroform formation in the distribution system of Cluj-Napoca. CHCl(3) was the most important trihalomethane species observed after the chlorination of water in all of the sampled months. However, TTHM concentrations did not exceed the maximum permissible value of 100 microg/L (EU). The THM formation rates in the distribution system of Cluj-Napoca have a high seasonal variability. Kinetic laboratory experiments could be used to predict chloroform formation in the Cluj-Napoca distribution system. Furthermore, an empirical model allowed an estimation of the chloroform formation in the Gilau water treatment plant.