The aim of this study was to investigate the removal of eleven selected organic micropollutants (OMPs) occurring in municipal wastewater in a 1–100 μg L⁻¹ concentration range during the treatment in two (planted and unplanted) pilot-scale horizontal subsurface flow constructed wetlands in order to elucidate (i) the role of plants on the mass removal and (ii) to understand the removal behavior of the micropollutants at different sampling locations along the flow path. The research was conducted in Langenreichenbach (Saxony), Germany, over a period of one year. The high organic load was chosen for both constructed wetlands (CWs) to study the removal of OMPs, because it is often occurring in the “real” world, but seldom investigated as a research objective.

The concentration of the OMPs in the treated inflow wastewater varied more strongly (by a factor of 10) than the total organic and nitrogen loads (by a factor of 4). A wide range of removal efficiencies were found. Higher removals were observed in the pilot-scale CW planted with Phragmites australis than in the unplanted one. The stimulant caffeine was the investigated OMP which showed the highest removal with an average efficiency of 66%. The average removal efficiency of the other OMPs was below 30%. The plants in the CWs supported not only the removal of the fragrances galaxolide and tonalide but also the removal of the pharmaceuticals carbamazepine, diclofenac, ibuprofen, ketoprofen and naproxen, even under the conditions of high organic load. The endocrine disruptors bisphenol A and nonylphenol, and the antibacterial triclosan showed on average the lowest removal efficiencies. Under the studied operating conditions of high load, a positive correlation between the internal removal behavior of naproxen and diclofenac and concentration of sulfide was found.