Occurrences and removals of pharmaceuticals and endocrine disruptors in water treatment

Abstract

Micropollutants have been discharged to surface waters by the untreated effluents in sewage treatment plants (STPs) and wastewater treatment plants (WWTPs). The contaminated waters are naturally utilized as a source of drinking water in water treatment plants (WTPs). Many micropollutants resist conventional WTP systems and survive in tap water. In particular, pharmaceuticals and endocrine disruptors (ECDs) are examples of frequently detected micropollutants in drinking water. Although chronic exposure to micropollutants in drinking water has unclear adverse effects for humans, peer reviews have argued that the continuous accumulation of these substances in water environments and inappropriate removal of them in WTP systems may potentially affect human health, in future. Therefore, WTP monitoring and effective elimination process studies for pharmaceuticals and ECDs are required to control micropollutant contamination in drinking water. In this study, 14 micropollutants, which are frequently used in daily life and whose detection in domestic surface waters have been reported, were selected as the target compounds: acetaminophen, caffeine, carbamazepine, diclofenac, metoprolol, naproxen, ibuprofen, sulfamethoxzole, sulfamethazine, atrazine, 2,4-dichlophenoxyacetic acid, bisphenol A, nonylphenol, and triclocarban. The main objectives of this study were to visualize the fate and removal of the selected micropollutants in WTP systems and to contribute to efficient elimination of the pollutants by modification of the WTPs. The specific aims of this study were to (1) investigate the occurrence and behavior of the selected micropollutants in a WTP, and the dominant removal mechanisms in a conventional system, using liquid chromatography tandem mass spectrometer (LC-MS/MS) (Study 1)

(2) elucidate the effects of natural contents (dissolved organic matters (DOM), water temperature, and pH) and operational conditions (adsorbent dosage and contact time) of the adsorption for the selected micropollutants, and estimate their sorption coefficients based on the hydrophobicity using Freundlich isotherm and linear isotherm (Study 2)

(3) evaluate the removal efficiencies for metoprolol in chlorination (Cl2), ultraviolet (UV-C) radiation, and a Cl2/UV system, obtain the optimal conditions using two-level factorial design of experiments for metoprolol, and identify its byproducts in the suggested process (Study 3)

and (4) evaluate the suitability of micropollution indicators for use in the monitoring of river water and various drinking waters, analyze the correlations of their co-occurrences with each other, and estimate their persistence in conventional water treatments (coagulation, adsorption, and chlorination) and some advanced treatments (O3, UV, UV/peroxide (UV/H2O2), and Cl2/UV) (Study 4). Study 1 monitored the presence and behavior of the micropollutants in a WTP system. The levels of 14 micropollutants were measured in a water treatment plant in Seoul, Korea with LC-MS/MS. Among the measured micropollutants, 12 (all except atrazine and triclocarban) were found in the influent and effluent from the WTP, at levels ranging from 2 ng/L to 482 ng/L. Concentrations of acetaminophen, metoprolol, ibuprofen, and naproxen were higher in winter, while levels of the herbicide 2,4-dichloro-phenoxyacetic acid (2,4-D) were higher in summer. Only a small amount of metoprolol was removed in the water treatment process. The average removal efficiencies of the detected micropollutants in the WTP diversely ranged from 6% to 100%. The laboratory experiment showed that micropollutants with log Kow higher than 2.5 (especially bisphenol-A, 2,4-D, carbamazepine, triclocarban, and nonylphenol) were effectively removed by the coagulation process. Sunlight photodegradation also effectively removed sulfamethoxazole, sulfamethazine, caffeine, diclofenac, ibuprofen, and acetaminophen. This study implied that the micropollutants were mainly removed during the coagulation stage in the WTP by the combination of several mechanisms, such as adsorption to the particles and sunlight photodegradation. Study 2 investigated the effect of carbon dosage, contact time, sorption coefficients, pH, DOM, and temperature on the removal of nine micropollutants (acetaminophen, caffeine, diclofenac, naproxen, sulfamethoxazole, sulfamethazine, atrazine, 2,4-D, and triclocarban) by adsorption with activated carbon (AC). Increasing the carbon dosage and the contact time enhanced the removal of the micropollutants. The sorption coefficients of the hydrophilic compounds (caffeine, acetaminophen, sulfamethoxazole, and sulfamethazine) fit a linear isotherm, and the hydrophobic compounds (naproxen, diclofenac, 2, 4-D, triclocarban, and atrazine) fit a Freundlich isotherm. The removal of the hydrophobic pollutants and the caffeine were independent of pH changes, but acetaminophen, sulfamethazine, and sulfamethoxazole were adsorbed by mainly electrostatic interaction with AC

therefore, they were affected by pH. The decrease in adsorption removal in the surface water samples was observed and this decrease was significant for the hydrophobic compounds. The decline in the adsorption capacity in the surface water samples is caused by the competitive inhibition of DOM with micropollutants onto activated carbon. Low temperature (5C) also decreased the adsorption removal of the micropollutants, and that affected the hydrophobic compounds more than the hydrophilic compounds. The results obtained in this study could be applied to optimize the adsorption capacities of micropollutants using AC in the water treatment process. Study 3 assessed the degradation of metoprolol based on the type of treatment system, using Cl2, UV-C (λ=254 nm) and Cl2/UV, in order to investigate whether the disinfection for metoprolol removal was effective, and to predict the disinfection application at different conditions of UV light intensity (1.1-4.4 mW/cm2), chlorine dose (1-5 mg/L as Cl2), pH (2-9), and DOM (1-4 mgC/L), using a two-level factorial design for the experiments. The results showed that, among the three treatment options, the Cl2/UV process was the most effective for metoprolol removal. For the optimization of metoprolol removal during Cl2/UV treatment, 16 experiments, combining the four factors, were conducted to investigate the amount of metoprolol in distilled water after 1 hour of Cl2/UV treatment. Among the factors assessed, DOM inhibition of the OH radical was the most dominant in the terms of metoprolol degradation. The established model fitted well with the actual results in the experiments using surface water and tap water. The optimized conditions (UV-C = 4.4 mW/cm2

[Cl2] = 5 mg/L, pH = 7, and [DOM] = 0.8-1.1 mgC/L) of the model transformed more than 78.9% of the metoprolol in the Cl2/UV process. Five byproducts of metoprolol (molecular weights: 171, 211, 309, 313, and 341) were identified with LC-MS/MS during the Cl2/UV process. The major degradation mechanisms of metoprolol were constructed, and these results may provide valuable information concerning advanced oxidation processes. metoprolol was effectively degraded in the Cl2/UV system, and the optimal conditions and unidentified byproducts were elucidated during the treatment process. Study 4 estimated the suitability of caffeine, metoprolol, sulfamethoxazole, and carbamazepine as indicator micropollutants of with detection in various types of water, such as river water, spring water, tap water, and bottled water. Moreover, the lab-scale experiments of coagulation, adsorption, chlorination, O3, UV, UV/H2O2, and Cl2/UV were conducted to confirm the persistence of the four nominated pollutants in the WTPs. All indicator candidates were detected > 80% in the river sample, and caffeine had the highest concentration (158.6 ng/L) and a positive correlation in the co-occurrences. However, the tap water contained a higher level of metoprolol (14.2-40.4 ng/L) than caffeine, and metoprolol showed the lowest removal (10-18%) in the conventional treatments. Considering the detection and resistance in drinking water, metoprolol was recommended as an indicator micropollutant. Among the advanced oxidation processes, the Cl2/UV process was the most effective for removing the selected contaminants (90-100%). In conclusion, this thesis determined the fate and dominant removal mechanisms for micropollutants in WTPs, the effects of the control factors and the adsorption coefficients of the micropollutants in adsorption with AC, the optimized operational conditions of the Cl2/UV process for metoprolol as a recalcitrant micropollutant, and whether metoprolol could be a micropollution indicator in drinking water. According to the results of this thesis, many kinds of micropollutants can exist in WTPs at the level of parts-per-trillion (ppt) and parts-per-billion (ppb), and various technologies and related research are needed to examine the degradation of persistent micropollutants in order to supply the population with healthy drinking water.