AUTOMATION OF LONG-TERM CHLORINE DEMAND M EASUREMENT OF TREATED WATERS

In drinking water treatment plants, process evaluation and water quali ty control are based on standard parameters such as DOG, BDOC, short o r long term chlorine demand, oxidation by-products. Chlorine demand is one of the most important parameters because it allows the adjustment of the chlorination rate to maintain a residual concentration of acti ve chlorine and to limit chlorination by-products. According to the re sults previously observed, the total reaction of chlorine with dissolv ed organic carbon requires prolonged contact times (several days or se veral weeks). In order to determine the chlorine demand in the water t anks and in the networks, the laboratory technicians are obliged to us e a high chlorine dosage and to follow the decrease of residual chlori ne over a very long time. The kinetic model described by Jadas-Hecart et al. (1992) (Water Research 26, 1073) leads to the determination of the long term chlorine demand of water by determining its chlorine con sumption as a function of the reaction time and then calculating the k inetic parameters. The use of this kinetic model requires the strict o bservance of analytical rules in order to avoid any pollution of the s amples. In addition to that, these experiments are time-consuming, whi ch greatly limits the number of determinations. According to that, the objective of the present work was to develop an automatic analysis of the chlorine demand in order to reduce the risks of sample contaminat ion while increasing the analytical possibilities. For this purpose, a utomation of long term chlorine demand measurements was achieved by co upling a Gilson sample changer with a Secomam spectrophotometer. (C) 1 997 Elsevier Science Ltd.