Vorkommen naturlicher und synthetischer ostrogener Steroide in Wassern des sud- und mitteldeutschen Raumes

Natural and synthetic hormones can reach surface waters via domestic sewage effluents. For drinking water production, bank filtration of river waters is a common procedure and hormone contaminations can potentially reach groundwater levels and drinking water sources. In order to analyse steroid hormones in the different aquatic compartments (raw sewage and effluent, surface water, groundwater, raw and drinking water) of South and Middle Germany, a sensitive analytical method was developed and employed to detect the natural steroid hormones estradiol (E2) and estrone and the synthetic estrogen ethinylestradiol (EE2). Samples which were taken in two subsequent series were subjected to clean-up and enrichment procedure and subsequently analysed by HPLC-MS. The limit of quantitation for the method was determined to be 0.05 to 0.5 ng/L, depending on the matrix. By treating the samples with glucuronidases/arylsulfatases, conjugates were amenable to analysis and the sum of conjugates and unconjugated steroids was calculated. In raw sewage, the median of the concentrations of the unconjugated steroids was 7 ng/L for EE2, 1.5 ng/L for E2, and 5.5 ng/L for estrone. After cleavage of conjugates, the medians of total steroids were 9.5 ng/L (EE2), 3 ng/L (E2), and 13 ng/l (estrone). Conjugates therefore contributed up to 50 % of the total steroid concentration in raw sewage. In treated effluents, the concentrations of steroids were much lower than in the raw sewage. The medians of free steroids were determined to be 0.3 ng/L for EE2, 0.2 ng/L for E2, and 2.5 ng/L for estrone Overall the medians in the effluent were thus less than 10 % of those in the influent. Conjugates still contributed significantly (40 % and more) to the steroid concentrations (medians: EE2: 0.5 ng/L, E2: 0.8 ng/L, and estrone: 8 ng/L). In surface waters, the median concentrations of the unconjugated analytes were generally below the limit of quantitation. After cleavage of conjugates, only the medians of E2 and estrone were above the limit of quantitation (E2: 0.1 ng/L and estrone: 0.4 ng/L). In groundwater only the median for EE2 was above the limit of quantitation. After conjugate cleavage, the median of total estrone was also above the limit of quantitation with 0.1 ng/L. In raw water, only the median of estrone was above the limit of quantitation after cleavage (0.2 ng/L). In the drinking water samples, the medians of all steroids were below the limit of quantitation even after cleavage of conjugates. In some compartments the steroid concentrations differed significantly between the two series. The contribution of conjugates was markedly lower in surface, ground-, raw, and drinking water most probably because cleavage occurred more effectively in this oligothrophic environment which is also characterised by long hydraulic retentions. The measured concentrations are at least for the compartments sewage and surface water well in the range of published data of German or foreign samples. The measured ranges do not support the assumption of harmful effects to aquatic organisms or human health.