Investigation of full-scale ozonation at a municipal wastewater treatment plant using a toxicity-based evaluation concept.

Effluents from municipal wastewater treatment plants (WWTPs) are known to be point sources of micropollutants for surface waters. The aim of this study was to examine a reconstructed full-scale ozonation equipped with a pump-injector system for ozone (O) dosage and a fluidized moving-bed reactor as biological posttreatment at a municipal WWTP utilizing an effect-directed approach. This approach consists of chemical analysis in combination with toxicological tests for the assessment of treatment efficiency of the plant. Chemical analysis showed elimination rates > 80% for pharmaceuticals and industrial chemicals. Analysis of endocrine disruptors was limited due to substance concentrations below the limit of detection (LOD). Estrogenic activity was detected by the Arxula Adeninivorans yeast estrogen screen (A-YES) at low concentrations (pg to ng EEQ/l range). Estrogenic activity was reduced by more than 90% after ozonation. In contrast, androgenic activity (measured in the Adeninivorans yeast androgen screen, A-YAS) was still found after O treatment and after biological posttreatment, which is consistent with the data obtained by chemical analysis. Furthermore, no marked genotoxic or cytotoxic effects were observed after ozonation using the alkaline comet and 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromid (MTT) assays, respectively. Results suggest that the applied specific O dose of 0.4 mg/mg is a safe operation setup in terms of toxicologically relevant transformation products. In addition, no adverse effects on primary producers, as evidenced by algae growth inhibition tests, were detected. The monitored biofilm growth in the biological posttreatment exhibited a steady state after one month. Based on computational fluid dynamics (CFD) simulations and biomass, one might conclude that O did not apparently enter biological posttreatment to a great extent and that hydraulic retention time in the O reactor was sufficient. Our data demonstrate the effectiveness of a full-scale O treatment in combination with a fluidized moving-bed reactor as biological posttreatment for the reduction of a majority of micropollutants without the release of relevant toxic transformation products as assessed by a chemical and toxicity-based approach.