Dose—response of Escherichia coli in ozone demand-free phosphate buffer

Abstract

The intention of this investigation was to determine the mean response of E. coli at predetermined ozone dose levels. By using replicated, randomized, independent batch experiments, precise estimates of the response could be obtained. Particular care was taken in preparing the bacteria and glassware to eliminate effects caused by extraneous ozone demand. A total of 57 batch experiments were conducted at bench-scale using ozone doses of 4.4, 11.45, 200 and 800 μg l −1 at contact times of 30, 60 and 120 s. The logarithm of the E. coli survival ratio was plotted as a function of applied ozone dose, utilized ozone and contact time. Contact time was found to have some effect for intermediate doses of ozone, but the effects were small relative to those associated with the utilized ozone dose. It was also found that the disinfection reaction was not first-order with respect to the surviving bacteria concentration and two distinct stages were observed; an initial rapid inactivation stage followed by a slower inactivation stage. Regression analysis of the E. coli response on the logarithm of the dose parameter is typical of dose-response models for water and wastewater disinfection. However, this model form was inadequate when fitted to the data of this study as a result of a significant tailing effect in the log-log dose-response plots. It was postulated that the observed dose-response could be explained by chemical kinetic theory and that the dose-response was affected by the disinfection reaction rates which were dependent upon the surviving number of bacteria and the residual ozone concentration. For practical application, it was found that an adequate dose-response model for 0–99.99% reduction of E. coli could be developed using the logarithm of utilized ozone as the dose function.