Process design for removal of pharmaceuticals in wastewater treatment plants based on predicted no effect concentration (PNEC)

Increasing concern on water contamination by micropollutants like pharmaceuticals fuels the development and implementation of technologies to remove micropollutants from municipal wastewater treatment plants (WWTP). However, often the targets and criteria for process design of such technologies are not clarified. This study was conducted to test whether predicted no-effect concentrations (PNEC) can be used as a design parameter for advanced treatment technologies to achieve pharmaceutical levels in WWTP effluents. This goal is consistent with environmental requirements, currently being discussed both by the Danish authorities and the European goals on zero emissions as documented in the draft of the Urban Wastewater Directive. The effluent of a conventional activated sludge WWTP was treated by ozonation and granular activated carbon (GAC) and monitored for 50 pharmaceuticals and iodinated X-ray contrast media, as well as 23 transformation products. Treatment with GAC alone initially achieved concentrations below PNEC for all targeted compounds, but after treating 3,000 - 5,000 bed volumes, the removal for several compounds decreased and the effluent concentrations for clarithromycin and venlafaxine were no longer below PNEC. Ozonation alone effectively reduced the concentrations of most of the compounds with standard ozone dosing of 0.5 mg O3/mg DOC. However, ozonation was unable to remove bicalutamide and oxazepam to reach target concentrations. The operation of both technologies in combination achieved concentrations of all measured pharmaceuticals below the PNEC (even with ozone concentrations of below 0.5 mg O3/mg DOC). Nonetheless, this study suggests that proper steering of WWTP design via the PNEC values alone is obstructed by lack of reliable primal PNEC data and absence of PNEC references for emerging pollutants and potential biologically active transformation products.