Assessing Emissions from Pharmaceutical Manufacturing Based on Temporal High-Resolution Mass Spectrometry Data

This study presents a nontarget approach to detect discharges from pharmaceutical production in municipal wastewater treatment plant (WWTP) effluents and to estimate their relevance on the total emissions. Daily composite samples were collected for 3 months at two WWTPs in Switzerland, measured using liquid chromatography high-resolution mass spectrometry, and time series were generated for all features detected. The extent of intensity variation in the time series was used to differentiate relatively constant domestic inputs from highly fluctuating industrial emissions. We show that an intensity variation threshold of 10 correctly classifies compounds of known origin and reveals clear differences between the two WWTPs. At the WWTP receiving wastewater from a pharmaceutical manufacturing site, (i) 10 times as many potential industrial emissions were detected as compared to the WWTP receiving purely domestic wastewater; (ii) for 11 pharmaceuticals peak concentrations, >10 μg/L and up to 214 μg/L were quantified, which are clearly above typical municipal wastewater concentrations; and (iii) a pharmaceutical not authorized in Switzerland was identified. Signatures of potential industrial emissions were even traceable at the downstream Rhine monitoring station at a >4000-fold dilution. Several of them occurred repeatedly, suggesting that they were linked to regular production, not to accidents. Our results demonstrate that small wastewater volumes from a single industry not only left a clear signature in the effluents of the respective WWTP but also influenced the water quality of one of Europe's most important river systems. Overall, these findings indicate that pharmaceutical production is a relevant emission source even in highly developed countries with a strong focus on water quality, such as Switzerland.