Pharmaceuticals in wastewater: Occurrence, short-term and seasonal variability, sampling strategies, biodegradation rates and elimination efficiencies.

摘要

This study was performed to investigate the occurrence, short-term and seasonal variability, sampling strategies, biodegradation rates, and elimination efficiencies of pharmaceuticals in an activated sludge wastewater treatment plant (WWTP). Selected compounds were four over-the-counter (acetaminophen, caffeine, naproxen and ibuprofen) and eight prescribed (trimethoprim, sulfamethoxazole, meprobamate, erythromycin, dilantin, carbamazepine, carisoprodol and diclofenac) pharmaceuticals.;First, the occurrence, short-term and seasonal variability influent and effluent of an activated sludge WWTP were investigated based on two sampling campaigns carried out in the warm month (summer) and cold month (winter) using grab samples every 3 hours for 72 hours. All selected pharmaceuticals except diclofenac were detected in the influent and effluent. Acetaminophen and caffeine exhibited the largest influent concentrations in both the summer and winter compared to the other measured pharmaceuticals. The influent and effluent short-term variability (expressed as relative standard deviation, RSD, and max/min concentrations ratio) generally was higher for over-the-counter than prescription pharmaceuticals in both campaigns (e.g. summer influent RSD values were 78-159% and 27-78%, respectively). In general, the short-term variability in the effluent was lower than the short-term variability in the influent especially in the winter. The influent and effluent concentrations in winter were generally higher than those recorded in the summer. Moreover, most pharmaceuticals exhibited higher short-term variability in winter than in summer.;Second, a general, efficient, and easy-to-implement approach using transfer functions was developed to characterize the hydrodynamics in a full-scale WWTP. The resulting model was calibrated with electrical conductivity as a natural tracer and applied to estimate first-order removal rate constants and removal efficiencies for selected pharmaceuticals. Hydrodynamics in the study WWTP were described by eight reactors in series. The transfer function approached revealed that at least 54 hours of sampling is needed before the effluent volume can be nearly accounted for by influent sampling. Estimated removal rate constants for selected pharmaceuticals were ranged from 0.009 to 0.65 hr-1 in the summer campaign and were generally smaller in the winter. The highest removal rates were estimated for acetaminophen and caffeine. The transfer function approach showed a large variability of removal for pharmaceuticals with smaller rate constants.;Third, this study demonstrates that the variations in pharmaceutical daily mass load obtained by different sampling strategies in influent and effluent of wastewater treatment plan on any given day can range widely. Over-the-counter pharmaceuticals showed higher variability than other pharmaceuticals. The mean, variances, and RSDs of mass loading across days were not statistically significant between the sampling strategies at 95% confidence level. The current datasets suggest that, although the compositing methods often give different results, day-to-day variability is even greater in both the influent and effluent except for some cases.