Occurrence, fate and ecotoxicological assessment of pharmaceutically ...

Science of the Total Environment 470–471 (2014) 618–630

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Occurrence, fate and ecotoxicological assessment of pharmaceutically active compounds in wastewater and sludge from wastewater treatment plants in Chongqing, the Three Gorges Reservoir Area Qing Yan a,b,1, Xu Gao a,⁎, You-Peng Chen a,c, Xu-Ya Peng a, Yi-Xin Zhang a, Xiu-Mei Gan a, Cheng-Fang Zi a, Jin-Song Guo a,c a b c

Key Laboratory of the Three Gorges Reservoir Region's Eco-Environments of Ministry of Education, Chongqing University, Chongqing 400045, PR China College of Geography Science and Tourism, Chongqing Normal University, Chongqing 400047, PR China Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing 401122, PR China

H I G H L I G H T S • • • •

All the 21 analyzed PhACs were detected in wastewater and 18 in sludge. The removal of PhACs was insignificant in primary and disinfection processes. Contribution of sorption to sludge was only 1.5% for the target PhACs. Five antibiotics and a mixture of 21 detected PhACs may pose a risk to algae.

a r t i c l e

i n f o

Article history: Received 6 June 2013 Received in revised form 23 August 2013 Accepted 9 September 2013 Available online xxxx Editor: Damia Barcelo Keywords: Pharmaceutically active compound (PhACs) Sludge Apparent solid-water partition coefficient Treatment efficiency Mass balance analysis Ecotoxicological risk

a b s t r a c t The occurrence, removal and ecotoxicological assessment of 21 pharmaceutically active compounds (PhACs) including antibiotics, analgesics, antiepileptics, antilipidemics and antihypersensitives, were studied at four municipal wastewater treatment plants (WWTP) in Chongqing, the Three Gorges Reservoir Area. Individual treatment unit effluents, as well as primary and secondary sludge, were sampled and analyzed for the selected PhACs to evaluate their biodegradation, persistence and partitioning behaviors. PhACs were identified and quantified using high performance liquid chromatography/tandem mass spectrometry after solid-phase extraction. All the 21 analyzed PhACs were detected in wastewater and the target PhACs except acetaminophen, ibuprofen and gemfibrozil, were also found in sludge. The concentrations of the antibiotics and SVT were comparable to or even higher than those reported in developed countries, while the case of other target PhACs was opposite. The elimination of PhACs except acetaminophen was incomplete and a wide range of elimination efficiencies during the treatment were observed, i.e. from “negative removal” to 99.5%. The removal of PhACs was insignificant in primary and disinfection processes, and was mainly achieved during the biological treatment. Based on the mass balance analysis, biodegradation is believed to be the primary removal mechanism, whereas only about 1.5% of the total mass load of the target PhACs was removed by sorption. Experimentally estimated distribution coefficients (b 500 L/kg, with a few exceptions) also indicate that biodegradation/transformation was responsible for the removal of the target PhACs. Ecotoxicological assessment indicated that the environment concentrations of single compounds (including sulfadiazine, sulfamethoxazole, ofloxacin, azithromycin and erythromycin-H2O) in effluent and sludge, as well as the mixture of the 21 detected PhACs in effluent, sludge and receiving water had a significant ecotoxicological risk to algae. Therefore, further control of PhACs in effluent and sludge is required before their discharge and application to prevent their introduction into the environment. © 2013 Elsevier B.V. All rights reserved.

1. Introduction

⁎ Corresponding author. Tel.: +86 2365120768; fax: +86 2365128095. E-mail addresses: [email protected] (Q. Yan), [email protected] (X. Gao). 1 Tel. +86 13883570863. 0048-9697/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.scitotenv.2013.09.032

Pharmaceutically active compounds (PhACs), which are primarily designed to elicit a specific biological response for human and veterinary use, are regarded as “pseudopersistent” contaminants due to their continual input into the ecosystem and permanent presence (Daughton

Q. Yan et al. / Science of the Total Environment 470–471 (2014) 618–630

and Ternes, 1999; Hernando et al., 2006). The environmental pollution from PhACs has therefore provoked considerable scientific attention around the world over the past decade not only because of their potential adverse effects on various organisms (Pomati et al., 2006; Quinn et al., 2009; Santos et al., 2010; Wilson et al., 2003), but also due to their role in the development/maintenance/transfer/spread of antibiotic-resistant bacteria and antibiotic-resistant genes in the long term in the environment (P. Gao et al., 2012; Huang et al., 2011; Martinez, 2008; Zhu et al., 2013). Besides, the presence of PhACs could also endanger the reuse of treated wastewater, escalating human exposure to PhACs (Kim and Aga, 2007). Significant fractions of PhACs are excreted in the unmetabolized or metabolized form, via urine and feces of humans or animals. These PhACs are delivered into raw sewage systems and eventually reach the municipal wastewater treatment plants (WWTPs). WWTPs were not originally designed to deal with pharmaceutical contaminants, and they were built and upgraded with the principal aim of removing biodegradable carbon, nitrogen and phosphorus compounds and microbiological organisms. Therefore, WWTPs have limited capability of removing PhACs from wastewater. Even higher concentrations were found in effluent than in influent for some recalcitrant PhACs such as carbamazepine (CBZ), diclofenac (DCF) and metoprolol (MTP) (L. Gao et al., 2012; Jelic et al., 2011). As a result, treated wastewater containing PhACs is discharged into water bodies or reused for irrigation and recreation, and PhACs can end up onto soil when biosolids produced during wastewater treatment are reused as soil amendment. Given the poor elimination, WWTP discharges (sludge and effluent) are frequently identified as the principal pathway for the entry of PhAC residues into the environment (Jelic et al., 2012, 2011). Other sources may include the disposal of unwanted PhACs and waste from pharmaceutical manufacturing processes, as well as direct discharges from untreated domestic, agricultural and aquaculture wastewater. With the continual improvements of advanced analytical technology and methodologies, PhACs have been widely identified and their concentration can be detected at ng/L levels in various environmental matrices (Huerta-Fontela et al., 2011). Today, the occurrence and fate of PhACs in urban centers in developed countries, such as Europe, North American and Australia, have been extensively investigated and well documented (Jelic et al., 2011; Kasprzyk-Hordern et al., 2009; Rosal et al., 2010; Verlicchi et al., 2012). PhACs have been detected in WWTP effluents, groundwater, surface water, river water and even drinking water at concentrations ranging from ng/L to μg/L level (Huerta-Fontela et al., 2011). Their concentrations in biosolids can range from μg/kg to mg/kg level (Gobel et al., 2005; Golet et al., 2003; Jelic et al., 2011; Martin et al., 2012). Due to inappropriate uses, uncontrolled disposal of PhACs and lack of regulations, the risk of exposure to PhACs is probably greater in the developing world. For example, China, a developing country, has considerable PhAC production and consumption (approximately 1.9 million tons consumption in 2009). The extensive use of PhACs may imply that PhACs are present in the environment at higher concentrations, and possibly with wider distributions, than those found in western countries. However, only a few studies about the situation in China have been reported (Chang et al., 2010; Gulkowska et al., 2008; Li and Zhang, 2011; Lin et al., 2008; Xu et al., 2007). Furthermore, most of studies simply focused on the determination of PhACs in the aqueous phase, and considered WWTPs as “black boxes”, and calculated their overall removal efficiencies based on the concentrations of PhACs in raw influent and final effluent. The concentrations of PhACs in sludge were rarely determined mainly due to the demanding efforts required in the analysis of challenging matrices. Few studies considered the fate and distribution of PhACs in individual treatment processes, such as primary treatment, biological treatment and chlorination. To date, no studies in China have focused on the concentration of multiple therapeutic classes of pharmaceuticals except antibiotic in sludge and on the potential environmental risks from pharmaceutical residuals on ecosystems. Such information is

619

important and urgently needed as many urban areas of China have increasingly large populations, huge annual pharmaceutical consumption and inadequacy of sewage treatment facilities. The major objective of this study was to verify the occurrence and behavior of 21 target PhACs, including several antibiotics, analgesics, antiepileptics, antilipidemics and antihypersensitives, at four WWTPs with different treatment technologies in Chongqing, the Three Gorges Reservoir Area. Individual treatment unit effluents along activated sludge treatment process, as well as primary and secondary sludge, were sampled and studied in order to evaluate the biodegradation, persistence and partitioning behaviors of the target PhACs in both aqueous and solid phases. PhACs were chosen to represent a wide range of physicochemical properties. Mass balance analysis was performed to estimate fate of PhACs in the WWTPs and to explore their potential removal mechanisms. Furthermore, Kd values were calculated, from the measured concentrations of PhACs in the collected wastewater and sludge samples, to further identify the contribution of sorption onto sludge during the removal. We also assessed the potential ecological risks caused by the target PhACs on aquatic species based on calculated risk quotients (RQs). To the best of our knowledge, this is the first report on the occurrence of PhACs from multiple therapeutic classes except antibiotics in sludge in China to estimate adsorption of the target PhACs onto sludge. 2. Materials and methods 2.1. Chemicals and reagents All the analytical standards for the studied PhACs were of high purity grade (N 90%). Amlodipine (as besilate, ALP) and moxifloxacin (as hydrochloride, MOX) were obtained from European Pharmacopeia. Ibuprofen (IBP), DCF, clofibric acid (CA), bezafibrate (BZB), simvastatin (SVT), atorvastatin (as calcium salt, ATT), CBZ, erythromycin (as hydrate, ERY), roxithromycin (ROX) and azithromycin (as dehydrate, AZM) were purchased from Sigma-Aldrich, USA. Acetaminophen (ACM), gemfibrozil (GFB), MTP (as tartrate), sulfamethoxazole (SMZ), sulfadiazine (SDZ), sulfamethazine (SM1), trimethoprim (TMP), ofloxacin (OFX) and norfloxacin (NOR) were provided by Dr. Ehrenstorfer from Germany. Internal standards, namely, simatone (SMT), dihydrocarbamazepine (DCBZ), caffeine-13C3 (CF-13C) and mecoprop-D3 were purchased from Accustandard (New Haven, CT), Sigma-Aldrich, C/D/N Isotopes (Quebec, Canada) and Dr. Ehrenstorfer (Augsburg, Germany), respectively. Oasis hydrophilic–lipophilic balanced (HLB, 6 cc3, 200 mg) cartridges were purchased from Waters Corporation (Milford, MA, USA). Syringe filters with 0.45 mm pore size were purchased from Pall Corp (USA). Milli-Q water was used throughout the study. High-performance liquid chromatography (HPLC)-grade methanol was provided by Merck (Germany). The individual standard solutions as well as internal standard solutions were prepared at concentrations of 500 mg/L by dissolving appropriate amounts of PhACs in methanol. The dehydrated form of ERY (ERY-H2O) is most frequently detected in the environment and therefore ERY-H2O was measured in this study. A standard solution of ERY-H2O was obtained by adjusting the pH of an ERY solution to 3.0 using 3 M H2SO4. After 4 h of stirring at room temperature complete conversion to ERY-H2O was achieved. The completeness of the reaction was verified by checking for leftover ERY and possible metabolites using mass spectrometry. All stock solutions were stored in the dark at –20 °C, and new stock solutions were prepared every three months. A mixture of all PhACs was prepared by appropriate dilution of individual stock solutions in methanol–water (30:70, v/v) and it was renewed before each analytical run. A separate mixture of internal standards, used for internal standard quantification, was prepared in methanol and subsequently diluted in methanol–water (30:70, v/v) mixture.

620


2.2. Selected PhACs The PhACs analyzed were selected according to the following criteria: (i) the most consumed active principles with medical prescription in China; and (ii) previous information reported in scientific literature about occurrences in surface and wastewater. Compounds were chosen to represent different groups of PhACs, such as antibiotics, antiepileptics, analgesic drugs, antilipidemics and antihypersensitive. The antibiotics, fluoroquinolone, macrolide and sulfonamide (including TMP, because TMP is always prescribed with SMZ) groups, selected in our study were the most frequently used antibiotics in China, contributing to approximating 15%, 20% and 12% of the total amount of antibiotics used for human and livestock purposes, respectively. As noted in Table S1, SMZ, ERY and ACM were among the 30 most frequently detected organic wastewater contaminants according to the US Geological Survey. Five of the selected PhACs – ERY, SMZ, CBZ, IBP and DCF – were among the top 10 high priority PhACs identified in a European assessment of Pharmaceutical and Personal Care Products. The physicochemical, biological properties and the structures of the target PhACs are listed in Table S1 in the Supplementary information.

The samples from primary clarifiers (primary sludge) and secondary sludge storage tanks (secondary sludge) of the Tangjiatuo and Jiguanshi WWTPs, secondary sludge storage tanks of Jingkou WWTP, and sludge thickening tanks of Lijiatuo WWTP were collected five times with same quantity, with equal time intervals at each sampling site during every sampling campaign. The samples were immediately put into ice-packed cooler, and mixed to give a single sample (around 500 mL) in brown glass bottles, and transported to laboratory. During sampling, sodium azide was added into each sample (0.5 g/L sample) to suppress potential biodegradation, and these sludge samples were centrifuged immediately at a speed of 10,000 rpm at 4 °C for 10 min upon the arrival to the laboratory, and freeze-dried at −50 °C. The sludge was ground and sieved to smaller than 0.5 mm and then stored at −20 °C until the analysis. Besides the solid phase for each sludge sample, 100 mL of corresponding supernatant samples were analyzed as described for the aqueous samples, in order to determine Kd coefficients. Receiving water samples were also collected from downstream (1–2 km) of WWTP discharge points. Receiving water samples were obtained from the river bank at a depth of b1 m, at least 4–5 grab samples were collected from different sites within a radius of about 20–50 m at each sampling location, and combined as one composite sample.

2.3. Samples and sampling sites 2.4. Analytical methods Four full-scale municipal WWTPs were selected in the present study (Fig. S1). The sampling schedule is listed in Table S2. The weather was stable and without precipitation, and the environmental temperatures ranged from 5 to 18 °C during the sampling periods. Samples of raw influent (after the screen), primary effluent, settled sewage after secondary clarifier (secondary effluent) and disinfection effluent were collected from Jiguanshi and Tangjiatuo WWTPs. Jiguanshi WWTP is the fourth largest WWTP in China and the largest WWTP in southwestern China, and Tangjiatuo is the second largest WWTP in Chongqing. The two WWTPs are on both banks of the Yangtze River and employ similar treatment processes: primary treatment to remove particles coupled with secondary biological treatment. For the secondary biological treatment, both WWTPs employ the anaerobic/anoxic/oxic (A2/O) activated sludge process that assists the simultaneous removal of nitrogen and phosphorus, the secondary effluent is discharged after chlorination. Other detailed information such as inhabitant served, service area, daily flow, hydraulic retention time (HRT) and sludge retention time (SRT) are shown in Table S3. Almost 80% of the total municipal sewage in Chongqing main urban region is treated by the two WWTPs, and their effluents are discharged into the Yangtze River, which is the largest river in Asia and serves as a drinking water supply for downstream communities. It raises the Tibetan Plateau and subsequently it flows through the urbanized areas of Chongqing, Wuhan, Nanjing, and Shanghai where it flows into the East China Sea, the Pacific. The raw influents (after the screen), the secondary biological treatment effluents, and disinfection effluents of another two WWTPs, namely, Jingkou WWTP and Lijiatuo WWTP, were also sampled. For the secondary biological treatment process, Lijiatuo WWTP uses cyclic activated sludge technology (CAST); whereas Jingkou WWTP employs an oxidation ditch (OD). The latter discharges its treated wastewater into the Jialing River; whereas the former into the Huaxi River which is a small, shallow river flowing through rural lowlands and villages, and reaching the Yangtze River. The effluent discharge outlet of the Lijiatuo WWTP is only about 3 km distance from the Yangtze River. The description of the involved WWTPs locations is shown in the Supplementary information (Fig. S1). Prior to this study, no information was available on the occurrence and fate of PhACs at the four WWTPs in Chongqing, the Three Gorges Reservoir Area. The wastewater samples from WWTPs were collected according to HRT in each WWTP, ensuring that the collected effluent originated from the collected influent. Schematic diagrams of treatment processes in the four WWTPs with the sampling locations are presented in Fig. 1.

Analytical procedures for the 21 PhACs in the wastewater and sludge samples were developed based on EPA Method (1694 USEPA, 2007) with some modifications. Water samples were extracted and cleaned by solid phase extraction (SPE) with an Oasis HLB cartridge (6 cc3, 200 mg; Waters Corp., Milford, USA). Sludge samples were extracted by ultrasonic technology, and then cleaned by SPE with an HLB cartridge. Detailed pretreatment information is listed in the Supplementary information (Text S1). The target PhACs were separated and quantified using an Agilent 1200 liquid chromatography (LC) system coupled to an Agilent 6410 triple quadrupole mass spectrometer equipped with an electrospray ion source (ESI) (Agilent, USA). Chromatographic separation of analytes was achieved using ZORBAX Eclipse Plus C18 column (4.6 × 150 mm, 3.5 μm) and C-18 guard column (3 × 2 mm, 3 μm) supplied by Agilent at a flow rate of 0.25 mL/min. Identification of the target PhACs was accomplished by comparing the retention time (within 2%) and the ratio (within 20%) of the two selected multiple-reaction monitoring (MRM) ion transitions with those of standards. The physicochemical characteristics of the target PhACs were different, and were classified into two groups (group A: detected by positive electrospray ionization mode, group B: detected by negative electrospray ionization mode) by considering their analytical conditions (polarity of produced ion, and mobile phase solvent). The conditions of LC and electrospray ionization tandem mass spectrometry for the determination of 21 target PhACs and 4 internal standards are summarized in Tables S4, S5 and S6 in the Supplementary information. 2.5. Method validation Quantification of the PhACs was performed with internal standard method to eliminate the influence of matrix effect. The calibration samples in triplicate at concentrations ranging from 0.1 to 1000 μg/L were prepared by spiking the working solutions in methanol–water (30:70, v/v). The correlation coefficients (r2) of the calibration curves were more than 0.99, except for ATT (r2 = 0.986). To identify the recovery, accuracy and precision of the analytical method, standard solutions were spiked into tap water in triplicate. The samples were extracted and analyzed. The recoveries of the target PhACs ranged from 43% to 104%, and the accuracy and precision ranged from 96% to 118% and from 2% to 10%, respectively. To confirm the matrix effects, the analytes and the internal standards were spiked into three effluents, three influent wastewater, and three sludge samples (using ultrasonic-assisted


621

Fig. 1. Schematic diagram of the treatment processes in the studied WWTPs and sampling sites along the process flow of WWTPs.

extraction), and were then analyzed using the same method. The recoveries achieved for the target PhACs except ATT and SVT, ranged from 41% to 140% for the influent, from 65% to 158% for the effluent, and from 54% to 139% for the sludge and their relative standard deviation was below 13%. Only ATT and SVT had lower recoveries mainly because the conditions chosen were not the most appropriate for the compounds. However, low recoveries were not considered an obstacle to their reliable determination in samples, as their sensitivity was fairly good. In multi-residue methods, obtained recoveries were low for SVT (42.85% in influent, 51.49% in effluent and 43.64% in sludge) and for ATT (21.43% in influent, 28.44% in effluent and 26.6% in sludge), but high for MTP (157.91% in effluent), for this reason, SVT, ATT and MTP were quantified by dilution of sample extracts when detected in the challenging matrix—sludge and influent. The accuracy and precision of the instrumental analysis over the experimental period were monitored

by replicate injections of standard solutions at 10 ng/ml and 100 ng/ml. The instrumental intra-day and inter-day precisions ranged from 3% to 15%, and from 5% to 18%, respectively, for the target PhACs. Limits of quantification (LOQs) (signal-to-noise ratio 10) ranged from 0.03 ng/L to 3.4 ng/L for the surface water, from 0.2 ng/L to 17.5 ng/L for influent wastewater, from 0.2 ng/L to 5.6 ng/L for effluent wastewater, and from 0.17 μg/kg to 5.83 μg/kg for sludge, respectively (in Table S6 in the Supplementary information). 2.6. Calculation of the removal rate The removal efficiencies of the selected PhACs during wastewater treatment were assessed quantitatively by mass balance approach, and we can assume that the WWTP behaves as a black box with only one entrance (i.e., influent water including total suspended solids)

622


and two outlets (i.e., effluent water and treated sludge), and the WWTP operates steadily during the sampling periods. Hence, from the measured aqueous and sludge phase concentrations and the operation parameters (i.e., flow rates of influent, effluent and sludge production), the overall removal (Roverall) was calculated as follows:

Roverall ð% Þ ¼

Cinfluent Q− Ceffluent Q þ Csludge Psludge Cinfluent Q

n

:

ð1Þ

The removal from aqueous phase (Raqueous-phase) was calculated as follows: Raqueous‐phase ð% Þ ¼

Cinfluent Q −Ceffluent Q Cinfluent Q

The ecological risk posed by the studied PhACs on aquatic organisms was conducted based on the European Commission's Technical Guidance Document (European Commission, 2003) and on previous studies (de Souza et al., 2009; Jelic et al., 2012; Leung et al., 2012; Lin et al., 2008; Lopez-Serna et al., 2012; Santos et al., 2010; Zhao et al., 2010). A risk quotient (RQ) for each compound was calculated using the following formula:

PNECwater

To quantify the sorption of PhACs, solid-water partition coefficient (Kd, L/kg) was experimentally estimated for the selected PhACs. Assuming that the system was at equilibrium, Kd was calculated as the ratio of the measured concentrations of a compound in the collected primary and secondary sludge samples, Csorbed, to the concentration in corresponding supernatants, Cdissolved: Kd ¼

Csorbed : Cdissolved

ð7Þ

Nevertheless, it should be stressed out that these Kd values were taken as rough estimates, considering the following reasons: The solid–liquid phase equilibrium state can be easily disturbed in the samples (Gobel et al., 2005; Ort et al., 2010; Radjenovic et al., 2009), and sewage and sludge streams are very inhomogeneous mediums that vary globally on both a spatial and a temporal scale in sewer and WWTPs. Also, neither for the aqueous nor for the solid phase conjugated forms or metabolites were included in the analysis. 3. Results and discussion

ð4Þ

where EC50 or LC50 (i.e., concentration of a drug that produces 50% of the drug's maximal effect) was obtained from the literature focusing on aquatic organisms at different trophic levels: Daphnia magna, algae and fish, and AF is the safety factor of 1000 as it was recommended by the Water Framework Directive (Directive 2000/60/EC). In order to calculate values for the worst-case scenario, maximum concentrations detected as well as the lowest EC50 values were applied. On the other hand, as experimental toxicity values are not available for each of the detected PhACs for all trophic levels, PNECs estimated using USEPA's ECOSAR (Ecological Structure Activity Relationships) model (ECOSAR v1.02), were also included to assess risk to D. magna, algae and fish. The PNEC for the sewage sludge (PNECsludge) may be estimated from the PNECwater using the following equation (Yu et al., 2013): PNECsludge ¼ Kd PNECwater

It is to be noted that ERY-H2O is structurally similar to the active parent form and may exert similar effects on non-target organisms, so ERY-H2O was assessed based on the toxicity of the parent compound rather than that of metabolite as relevant information was scarce (Leung et al., 2012).

ð3Þ

where, MEC is the measured environmental concentration, and PNEC is the predicted no effect concentration. PNECwater was calculated using the following formula: LC or EC50 ¼ 50 AF

ð6Þ

2.8. The calculation of solid-water partition coefficient (Kd)

2.7. Derivation of predicted no effect concentrations (PNECs) and ecotoxicological assessment

MEC PNEC

RQ iðsumÞ ¼ ∑i¼1 RQ i :

ð2Þ

where Cinfluent (ng/L), Ceffluent (ng/L) and Csludge (ng/g) are the measured concentrations in the influent (including dissolved and the total suspended solids), effluent and sludge, respectively; Q (m3/d) and Psludge (kg/d) are the flow rate of wastewater and the production rate of sludge, respectively; and the concentration of the total suspended solids in effluent is negligible.

RQ ¼

The combined risks of the selected PhACs were assessed based on the simple concentration addition model (Backhaus and Faust, 2012; Santos et al., 2010), which is the sum of all individual pharmaceutical RQs for each trophic level (Zhao et al., 2010). The combined risks were expressed by an overall RQi (sum) as follows:

ð5Þ

where Kd is the solid–water distribution coefficient of the corresponding pharmaceutical. The PNEC values used for the calculation are provided in Table S7. A common ranking criterion was applied according to de Souza et al. (2009) and Hernando et al. (2006) in the present study: RQ b 0.1, low risk to aquatic organisms; 0.1 ≤ RQ b 1, medium risk; and RQ ≥ 1, high risk.

3.1. Occurrence of PhACs in influents Concentrations of the target PhACs varied in different sampling periods in a given WWTP (Table S8; Fig. 2). This phenomenon is possibly due to changes of the composition of the influent waters, weather conditions and operation conditions of the WWTP, as well as the amount of drug consumption. Besides, the sampling protocol itself may have significant influence on concentration values obtained (Ort et al., 2010). Given that the PhACs arrive at the influent of WWTP in a small number of wastewater packets, in unpredictable amounts and time intervals, the influent loads are easily underestimated (Jelic et al., 2011). The concentrations and the frequencies of detection of individual PhACs in influents from the WWTPs included in this study are shown in Fig. 2 (full details in Supplementary information, Table S8). All the 21 target PhACs were detected, and 14 of them were detected in all the collected influent samples, whereas other PhACs except MOX (43.75%) and ALP (31.25%) were quantified with a detection frequency above 50%. ACM predominated in all analyzed influent samples at concentrations ranging between 1.2 and 7.7 μg/L, accounting for more than 50% of the PhACs analyzed (Fig. 3), followed by SMZ (approximately from 4 to 8 to 3180 ng/L), AZM (from 80.86 to 661.93 ng/L), ROX (from 28.83 to 655.98 ng/L) and OFX (from 42.69 to 595.6 ng/L). Extremely low concentrations (some approached the corresponding LOQs) were detected for DCF, GFB, ALP, ATT and MOX or they were not detected at all. Apart from CA and ALP, the mean concentrations of the studied PhACs in the Jiguanshi and Tangjiatuo WWTPs were statistically greater than those detected in Jingkou and Lijiatuo WWTPs (using Duncan's multiple range test, p b 0.05). Meanwhile, in most cases, the


623

Fig. 2. Box plots of concentration ranges (Min, P 0.25, Median, P 0.75 and Max) and detection frequency (%) of the target PhACs in wastewater influent, effluent and sewage sludge from the studied WWTPs. Circles in the box plots refer to the mean values.

624


Mass flows composition (%)

Influent

Effluent

Sludge

100% Antiepileptic

80%

statin drugs Antihypersensitive

60%

Lipid regulators Analgesic

40%

Macrolide

20%

Quinolone Sulfonamide

Lijiatuo

Jingkou

Tangjiatuo

Jiguanshi

Lijiatuo

Jingkou

Tangjiatuo

Jiguanshi

Lijiatuo

Jingkou

Tangjiatuo

Jiguanshi

0%

Fig. 3. Composition profiles of the target PhACs in influents, effluents and sludge from the studied WWTPs.

concentrations of the target PhACs in Lijiatuo WWTP were statistically significantly different from those detected in Jingkou WWTP. Therefore, even in different districts of a big city, such as Chongqing, the consumption patterns of PhACs may vary greatly. The relative concentration profiles of the eight groups of analyzed PhACs detected in the influents varied among WWTPs (Fig. 3). Influents from the WWTPs at Jiguanshi, Tangjiatuo and Lijiatuo exhibited similar relative composition profiles of the eight groups of PhACs. Analgesic (ACM, contributing almost all to the relative analgesic concentration profiles) was the most abundant, accounting for 58.09–69.07% of the total, followed by macrolide antibiotics (MAs) (8.03%–19.98%), sulfonamide antibiotics (SAs) (4.71%–26.69%) and quinolone antibiotics (QAs) (4.44%–7.53%) at Jiguanshi, Tangjiatuo and Lijiatuo WWTPs. A similar composition pattern was also observed in the influent samples from Jingkou WWTP except that lipid regulators contributed a more amount (4.54%) as compared with other WWTPs (Fig. 3). The concentrations of the target PhACs, except antibiotics and SVT, in the present study were much lower than those reported in Europe and North America. Our results were in agreement with those obtained by Sui et al. (2010). The concentrations of GFB in wastewater influents were recorded to be between 2 μg/L and 5.9 μg/L (Radjenovic et al., 2009) or 0.41 and 17.1 μg/L (Rosal et al., 2010) in Spain, 1.5 μg/L in Australia (Khan and Ongerth, 2005), and 0.45 μg/L in Canada (Lishman et al., 2006), while the concentration of GFB in the WWTPs in this study was below LOQ or only several ng/L levels. The concentrations of ACM, BZB and IBP in the present study were one or two order of magnitude lower than those reported in developed countries (Clara et al., 2005; Gomez et al., 2007; Khan and Ongerth, 2005; Radjenovic et al., 2009). This is probably due to the lower per capita consumption in the Three Gorges Reservoir Area than in the countries having higher socioeconomic status, and where medical care is more advanced. The wide use of traditional medical care may have reduced the application of Western medicine. By contrast, the concentrations of the antibiotics found in the present study were comparable to those reported in developed countries, with the concentrations of ROX and SMZ being even higher than those reported in developed countries (Clara et al., 2005; Gobel et al., 2005; Martin et al., 2012). This may be due to the abuse of antibiotics in China. SDZ was also detected in the current study; however, its occurrence in other countries has not been reported. Considering the aforementioned findings, it is urgent to restrict the use of antibiotics, because they were regarded as “priority pharmaceutics” due to the greater potential adverse effects on biota and on the ecosystem. The average daily mass load per capita, in μg/day/capita, of individual PhACs was obtained by multiplying the concentrations in sewage and the average treated flow rate during the sampling period and normalizing this value to the population served by the corresponding WWTP. The mass loads of NOR (ranging from 5.82 to 92.58 μg/day/capita), OFX (ranging from 8.89 to 203.8 μg/day/capita), MOX (ranging from

bLOQ to 7.02 μg/day/capita), TMP (ranging from 2.68 to 84.58 μg/day/ capita) and ERY-H2O (ranging from 14.72 to 102.9 μg/day/capita) in the studied WWTPs were lower than those detected in Shatin, Hong Kong (NOR: 187; OFX: 280; TMP: 130; ERY-H2O: 300 μg/day/capita) (Li and Zhang, 2011) and in Beijing (NOR: 381; OFX: 632; MOX: 35 μg/day/capita) (Jia et al., 2012). However, the mass loads of SMZ (300–1144 μg/day/capita), SDZ (59.02–89.57 μg/day/capita) and ROX (157.39–211.45 μg/day/capita) in Jiguanshi and Tangjiatuo WWTPs were greater than those detected in Guangzhou (SMZ: 45; SDZ: 27; ROX: 62.5 μg/day/capita) (Xu et al., 2007) and in Shatin, Hong Kong (SMZ: 31.75; SDZ: 18.25; ROX: 25 μg/day/capita) (Li and Zhang, 2011). The average daily mass load per capita of the target PhACs varied in different cities in China (Chang et al., 2010; L. Gao et al., 2012; Gulkowska et al., 2008; Jia et al., 2012; Leung et al., 2012; Xu et al., 2007), reflected different pharmaceutical usage consumed by human within the same country, assuming that attenuation of a compound during transportation from toilets to WWTPs was the same in different areas. 3.2. Occurrence of individual PhACs in effluents The amount of PhACs found in the effluents depended on the removal efficiency of WWTP, the physicochemical properties of the compounds and the influent concentrations, thereby resulting in different composition profiles from those in the influents. The concentrations of target PhACs varied among WWTPs, as shown in Table S8. In particular, ACM, a widely used over-the-counter analgesic (pain reliever) and antipyretic (fever reducer), accounted for more than 50% of the total in the influents of the studied WWTPs. However, its concentration was below or very close to the LOQ in most of the effluent samples, mainly because ACM is easily biodegradable in water (Behera et al., 2011). Comparatively greater levels of antibiotics were measured in the four studied WWTP effluents (Figs. 2 and 3). Similar to the influent samples, the concentrations of DCF, GFB, ALP and ATT were extremely low or below the corresponding LOQs in the effluent samples. The relative composition profiles for the eight groups of PhACs in the effluents were similar in three of WWTPs except Jingkou WWTP, in which lipid regulators contributed a greater amount compared with the other three WWTPs (Fig. 3). Evidently, of the eight-grouped analyzed PhACs, the antibiotics were the predominant compounds in effluent. In addition, MAs had the highest percentages in the WWTPs except Jiguanshi WWTP, where SAs were the highest. The concentrations of most PhACs except antibiotics measured in the secondary effluent were also lower than those reported in developed countries. In some cases, they exceeded 1 μg/L in the wastewater effluents in European and North American countries (Clara et al., 2005; Muñoz et al., 2009; Rosal et al., 2010). By contrast, the concentrations of the selected PhACs except antibiotics were less than 100 ng/L or below the LOQs, and all the analyzed PhACs except SMZ and ROX did not exceed 400 ng/L in any effluent samples in the present study (Fig. 2).


To date, very few studies on the pharmaceutical contamination in aquatic environment in China have been reported, and one specific therapeutic class, antibiotics, has been investigated by limited previous studies. The concentrations of antibiotics in effluents varied among cities in China. For instance, the concentrations of SAs (SDZ, SMZ and SM1) detected in the present study were higher than those reported in Hong Kong, Beijing and Guangzhou (L. Gao et al., 2012; Leung et al., 2012; Li and Zhang, 2011). The mean concentrations of two QAs (NOR and OFX) in the effluents of four WWTPs in Guangzhou were 27 ng/L and 48 ng/L (Xu et al., 2007), respectively, which were lower than the data of the current study (NOR, 71 ng/L; OFX 120 ng/L), while the concentrations of three QAs (NOR, 256 ng/L; OFX, 528 ng/L; and MOX, 40 ng/L) in a particular WWTP in Beijing (Jia et al., 2012) were greater than those obtained in this study. The concentration of ERY-H2O (0.02–0.227 μg/L) detected in the present study was lower than those recorded in other cities in China (L. Gao et al., 2012; Gulkowska et al., 2008; Leung et al., 2012; Li and Zhang, 2011; Xu et al., 2007), and the level of ROX (0.0244–0.36 μg/L) was higher than those reported in other parts of the country (L. Gao et al., 2012; Gulkowska et al., 2008; Leung et al., 2012; Li and Zhang, 2011; Xu et al., 2007). The present study quantified the concentration of AZM (87.5–609.0 ng/L in influents; 9.7–340.2 ng/L in effluents); however, there were no recorded data on the concentration of AZM in aquatic environments elsewhere in China. 3.3. Concentrations of PhACs in sludge Analyses of the sludge samples showed the presence of 18 out of the 21 PhACs targeted in this study. The concentrations of the PhACs detected in sludge are summarized in Fig. 2. GFB, ACM and IBP were excluded from Fig. 2, because these compounds appeared in only a few samples or at a concentration below the LOQs in sludge. The composition profile of the target PhACs in sludge showed that the mass loads of MAs, QAs and cholesterol lowering statin drugs accounted for more than 90% of the analyzed PhACs (Fig. 3). SMZ, with a detection frequency of 100% in wastewater, was also found in the sludge samples but at low concentrations ranging from 0.47 to 21.14 μg/kg dry weight (dw). Similarly, other studies found SMZ in sludge with low concentrations (5.7 and 27 μg/kg dw) (L. Gao et al., 2012; Gobel et al., 2005). SDZ was also detected in all the collected sludge samples with concentrations ranging from 0.66 to 20.39 μg/kg dw. SM1 was found in sludge but at extremely low concentrations ranging from 0.36 to 1.93 μg/kg dw. SAs are generally present as neutral and anionic species in wastewater at pH 7 (calculated by ACD/logPow ver. 1.0, Advanced Chemistry Development, Inc.), and have low Log Kow values. Therefore, the sorption to sludge is expected to be weak due to the electrostatic repulsion from the negatively charged functional groups in the activated sludge. Conversely, because all the target QAs contain nitrogen as positively-charged moiety and AZM possesses positively charged dimethylamino group, QAs and AZM would have high sorption potentials through electrostatic interactions between their positively charged locations and the negatively charged sludge. AZM manifested the strongest sorption to sludge in the present study. The concentrations of acidic compounds such as IBP, CA, GFB, BZB and DCF were low or below LOQs in sludge, because at the current wastewater pH they are mainly in their ionized form and, consequently, mainly dissolved in the aqueous phase. So far there are few reports for acidic drugs' concentrations in sludge. Recently, Samaras et al. (2013) also reported relatively low concentrations of DCF and IBP, while, in several cases, their concentrations were below the LOQs. The present study, for the first time, reported the presence of ALP in sludge. 3.4. Overall removal of PhACs during wastewater treatment The daily mass loads of the eight groups of the target PhACs in wastewater influent and effluent, as well as in sludge (g/day), were calculated by multiplying the concentrations C (ng/L) in the sewage

625

and the average treated flow rate Q (m3/d) during the sampling period. The results are shown in Fig. 4. As shown in Figs. 3 and 4, the composition profiles of each group of PhACs were different among the influent, effluent and sludge, suggesting that the WWTPs can remove a certain amount of the target PhACs. The mass loads of the PhACs discharged from the WWTP with effluent, treated sludge and removed during treatment were normalized on the influent mass loads and are shown in Fig. 5. For most of the selected pharmaceutical residues, the portion wasted with the treated sludge into the environment was negligible compared with the aqueous fraction. However, the sorption of QAs, MAs and cholesterol lowering statin drugs contributed to their elimination from the aqueous. The results indicated the importance of analyzing the sludge when researching wastewater treatment performance for PhACs, given that the treated sludge from WWTPs is used as agricultural fertilizer, thus, PhACs find their way into the terrestrial environment and become available to soil organisms when conditions for desorption are created. The overall removal rate was calculated using Eq. (1). During wastewater treatment, the different PhACs investigated in this study exhibited very different removal efficiencies, ranging from −32% to 100%, and the efficiency of the removal of the PhACs varied from one WWTP to another, as can be seen in Fig. 4. DCF, GFB, ALP and ATT were either detected in concentrations close to their corresponding LOQs, or were not detected in the wastewater at all, so no reliable conclusion could be made on their behavior. ACM, which was present at the highest concentration in the primary effluent, was almost completely removed and did not accumulate in sludge regardless of the type of treatment applied. The consistently excellent removal efficiency of ACM has been mainly attributed to microbial degradation (Roberts and Thomas, 2006; Rosal et al., 2010). The removal efficiency for the other PhACs varied largely in different WWTPs studied. For instance, the removal rate of TMP varied from 29% to 86%; OFX, from 22% to 73%; SMZ, from −7.5% to 61%; and ERY, from 26% to 66%. Gulkowska et al. (2008) and Choi et al. (2008) also reported a large variation in the removal rate of TMP, with ranges from −17% to 62.5% and from −11% to −79% in the studied WWTPs, respectively, and the values were lower than the result in the present study. Li and Zhang (2011) reported a variation in the removal rate of OFX, which ranged from 26% to 59%. Karthikeyan and Meyer (2006) presented a large variation in the removal rate of SMZ and ERY, which ranged from −24% to 96%, and from 44% to 75%, respectively. Gobel et al. (2007) reported a large variation in the removal of ERY, which ranged from 26% to 87%. It is very difficult to give a final conclusion on the removal of majority of the detected PhACs but it seems that the variations in the removal efficiencies of the selected PhACs in WWTPs may have results from many factors, such as the specific treatment process implemented, the SRT and the HRT in different WWTPs, physicochemical properties of pollutants (Clara et al., 2005; Gobel et al., 2007), and the sampling procedures and climatic conditions (e.g., temperature, rainwater input and level of sunlight) (Ort et al., 2010). Obviously, the type of the wastewater treatment technology implemented in WWTPs is the main factor. For instance, the Jingkou WWTP, which employs the oxidation ditch biological treatment process, showed the highest removal rates for TMP, SM1 and MAs which were statistically significantly different from the data measured in other WWTPs (using Duncan's multiple range test, p b 0.05), while the Jiguanshi WWTP demonstrated statistically better removal rates for other selected SAs and QAs compared with other WWTPs. An increase in the concentration of an analyzed parent compound, namely “negative removal” during treatment was observed in this study (Fig. 6). Concentrations of CBZ and MTP were higher in effluents than in influents from the Jiguanshi and Tangjiatuo WWTPs, the same pattern was found for SDZ, SMZ and CBZ from the Lijiatuo WWTP, and for CA from the Tangjiatuo WWTP. This phenomenon of “negative removal” for some PhACs has also been reported in previous studies (Clara et al., 2005; Karthikeyan and Meyer, 2006; Lindberg et al., 2005; Xu et al., 2007). The phenomenon of “negative removal”

626


8,000

Sludge

140

Wasterwater 7,000

statin drugs 250

120

200

100

Antihypersensitive

6,000

Daily Mass Loads (g/d)

Antiepileptic

Lipid regulators

6

5,000 150 80 4,000

100 60

3,000

50

2,000

0

40

5

Analgesic

4

Macrolide

3

Quinolone

2

Sulfonamide

1 0

20

1,000 0

0

IF

EF

IF

EF

IF

EF

Jiguanshi Tangjiatuo Jingkou

IF

EF

Lijiatuo

Jiguanshi Tangjiatuo Jingkou

Lijiatuo

Sewage sludge

Fig. 4. Daily mass loads (g/d) of eight groups of analyzed PhACs in the influent (IF) and effluent (EF), and in the sludge from the studied WWTPs.

can be explained in several ways. The first probable explanation involves the sampling protocols (Ort et al., 2010). In the present study, the calculations of all the removal efficiencies were based on grab samples collected with a hydraulic lag; therefore, plausible errors due to diurnal variations in pharmaceutical concentrations cannot be ruled out. Future studies must focus on the analysis of 24 h composite samples lagged by HRT. Secondly, they can be explained by the formation of unmeasured products of human metabolism and/or transformation products (glucuronide conjugate, methylates, glycinates etc.) that convert back to the parent compounds during wastewater treatment (Gobel et al., 2005; Miao et al., 2005; Radjenovic et al., 2009). Thirdly, the negative removal may be caused by a change in the adsorption behavior of the analytes onto particles during treatment, influencing the ratio between influent and effluent waters (Lindberg et al., 2005; Lindsey et al., 2001). Furthermore, this may be due to the signal suppression of the MS/MS detector in raw effluent samples due to high concentrations of organic matter (Miao et al., 2005). The distribution of the detected PhACs along the treatment units is shown in Table S8. The results showed that the concentrations of most PhACs during the primary step had no significant reduction, with even a slight increase in some cases. Significant reduction (N 40%) was observed for many PhACs during the biological treatments, though elimination was not apparent or negative for MTP, SDZ and CBZ in all the WWTPs and low removal efficiencies were found for MAs and TMP in most cases. The removal of most PhACs in the disinfection processes was negligible, except in a few cases (Table S8). This could be due to the trace levels (b1 μg/L) of most PhACs flowing into the disinfection treatment. The removal of PhACs has been reported to be ineffective or significantly slow during the disinfection processes designed to reduce pathogenic microorganisms in treated wastewater (Pinkston and Sedlak, 2004). The Jingkou WWTP, which employs ClO2 disinfection, had statistically better removal efficiencies for MAs compared with the other WWTPs with chlorine disinfection (using Duncan's multiple range test, p b 0.05). However, most of the PhACs could not be mineralized during disinfection, and oxidation products are formed from the parent pharmaceutical compounds. Thus, further research is required to identify the oxidation products and their potential toxicity (Buth et al., 2007; Dodd and Huang, 2004). 3.5. Ecotoxicological assessment Fig. 6 shows the RQ values of each pharmaceutical compound in three different scenarios: effluent wastewater, receiving water and

sludge. Detailed toxicological information is also provided in Table S7. As seen in Fig. 6, each organism showed different susceptibility for different PhACs. In most of individual cases, algae were much more sensitive to the individual PhACs than Daphnia and fish. The overall relative order of susceptibility to PhACs was estimated to be green algae ≫ Daphnia N fish. In what PhACs are concerned, SDZ, SMZ, OFX and ERY had an individual RQ above 1 in effluent and sludge, and in all cases that toxicity corresponded to algae, and AZM could induce a high risk to algae and Daphnia in sludge. All RQ values for the target PhACs were lower than 1 in the receiving water because of the dilution effect. However, single antibiotics including SMZ, OFX and ERY could pose a medium risk to algae in the receiving water. Of these eightgrouped PhACs investigated in this study, the groups of antibiotics were shown to be the most harmful for organisms, and contributed the most to the overall RQ (sum) for the three organisms. This observation can be attributable to the low EC50 and high concentration of the foresaid antibiotics in the environment. The persistence and continual input of antibiotics in aquatic environments have been shown to impose selective pressure on bacterial populations, resulting in the prevalence of antimicrobial resistance. Thus, we should avoid the inappropriate use of antibiotics by following strict antibiotic policies and taking precautionary measures to reduce the risks. Using a huge mass of literature data, Verlicchi et al. (2012) also concluded that antibiotics can lead to high risks using a huge mass of literature data. During wastewater treatment, pharmaceutical residues may be absorbed by the mixed liquor suspended solids, and subsequently removed from water stream through sedimentation (Jelic et al., 2011). Therefore, municipal sewage sludge, the solid fractions separated from the wastewater stream, is potentially a sink for the wastewaterborne PhACs. The publicly-owned WWTPs in China generate over 4.2 million tons (dry weight) of sewage sludge annually, with only 20% harmlessly utilized and more than 60% applied to land or landfilled. Land application and landfill in the disposition of sewage sludge are potential sources of PhACs in the terrestrial environment and in groundwater (EPA 832-R-06-005, 2006). However, few studies have assessed the potential ecological risks from PhACs in sewage sludge (Martin et al., 2012; Yu et al., 2013). The current research shows that most of the PhACs present in the aqueous phase were also found in the sludge samples, and the individual RQs for several antibiotics obtained in the sewage sludge (Fig. 6) were more than 1, which indicated potential ecotoxicological risks unless these antibiotics would be degraded shortly after land application of the sewage sludge.


627

dissimilar mechanisms of action have been studied, and the ecotoxicological data have shown that mixtures of different PhACs can induce additive adverse effects at environmentally relevant concentrations where single compounds show less or no effect (Backhaus et al., 2000; Cleuvers, 2004; Quinn et al., 2009; Yang et al., 2011). However, mixture effects are complex, and studies on the toxicity of mixtures of active substances, particularly of those with dissimilar mechanisms are still sparse. Thus, in view of the precautionary principle, the concentration addition model was used in the present study to estimate the toxicity of the mixture of 21 PhACs by adding the RQ values for the worst case scenario using the maximum concentration of each compound (Zhao et al., 2010). This method resulted in the RQs increasing in each scenario and the mixture RQ (sum) for algae was much more than 1 in three different scenarios, and the mixture could pose high risks to Daphnia and medium risks to fish in sludge and effluent. Nevertheless, it is important to note that the concentration addition approach may overestimate the overall toxicity (Backhaus et al., 2000). In addition to the possible direct effect on aquatic organisms, a series of cascade effects along the food chain should not be excluded. Additional ecotoxicological studies for more compounds and more aquatic species are needed to fully assess the risks of different mixtures of PhACs, given that most published studies focused on the mixture of analgesics, antibiotics and blood lipid lowering agents (Cleuvers, 2004). Therefore, risk control could be achieved through the application of appropriate sewage treatment technologies and municipal solid waste final treatment methods. 3.6. Calculation of Kd coefficients and literature comparison

Fig. 5. Mass loads of the selected PhACs normalized to the influent mass load: fraction discharged with effluent, sorbed to sludge, and removed during treatment.

Previous research and the present study have shown that PhACs do not occur alone in the environment, where organisms are exposed to the combined effects of pharmaceutical contaminants (Leung et al., 2012; Verlicchi et al., 2012). Mixtures of chemicals with similar or

The Kd values for the primary and secondary activated sludge are shown in Table S9. The calculated Kd values for the selected PhACs in the different WWTP varied. Generally all PhACs encountered in the sewage sludge exhibited the same tendency of greater enrichment in the secondary sludge when compared to the primary sludge. The variation in the composition and pH of the sludge (Horsing et al., 2011), as well as the concentration and physicochemical properties (Kow and pKa values) of the selected PhACs in the aqueous phase, are probably the reason for such results. It has been reported that the adsorption of chemicals was correlated with the cation exchange capacity (CEC) and the organic matter (OM) of sludge samples (Hang and Brindley, 1970; Jelic et al., 2012). Therefore, the CEC and OM values of the sludge samples from the four WWTPs were measured to assess the variation of Kd. As shown in Table S10, the adsorption of the detected PhACs was not correlated with CEC and OM. From the identified Kd values (i.e., Kd b 500 L/kg) in Table S9, it can be concluded that for the target PhACs except QAs, AZM and SVT, removal by sorption plays a minor role in the overall removal of PhACs during wastewater treatment. The relatively low log Dow values (b2.5) of the detected QAs (i.e., NOR, MOX and OFX) and AZM indicated that these compounds have a weak sorption potential through hydrophobic interactions (Golet et al., 2003). Given that the sludge biomass (i.e., both microorganisms and their extracellular polymers) carries an overall negative charge within the pH range typical for wastewater treatment (pH 5–9) (Busch and Stumm, 1968), electrostatic interaction with the positively charged dimethylamino group in the AZM molecules (pKa 8.7) (Jelic et al., 2012) and nitrogen atoms in the NOR, MOX and OFX molecules (pKa 10.56, 10.68 and 8.87, respectively) may be responsible for the high Kd values of these PhACs (Jia et al., 2012). SVT is a neutral compound with the highest log Dow among the studied PhACs, and hydrophobic interactions (logDow ~4.46 for pH 6–8) (calculated by ACD/logPow ver. 1.0, Advanced Chemistry Development, Inc.) resulted in its high Kd values. Jia et al. (2012) found the Kd values of the detected QAs (12,300–37,700) to be one or two orders of magnitude higher compared with that in the present study. Similarly, Gobel et al. (2005) and Jelic et al. (2012) reported higher Kd values for AZM. By contrast, significantly lower or non-detectable Kd values were found for IBP, CA, GFB (pKa 4.91, 4.15 and 2.9, respectively) and SAs, because in wastewater

628


1.0E+03 1.0E+02

Algae

Fish

Daphnia High Risk

1.0E+01

RQ

1.0E+00

Medium Risk

1.0E-01

Low Risk

1.0E-02 1.0E-03 1.0E-04

1.0E-06

Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge Receiving water Effluent Sludge

1.0E-05

RQ(sum) SDZ

SM1

SMZ

TMP

OFX

NOR

MOX

ERY

ROX

AZM

IBP

DCF

ACM

BZB

CA

GFB

MTP

ALP

ATT

SVT

CBZ

Fig. 6. Receiving water, effluent and sludge RQ values for the target PhACs estimated from the maximum measured concentrations and literature data on acute toxicity for algae, fish and Daphnia magna.

with average pH at 7.5, these PhACs are primarily in their ionized forms, and consequently mainly dissolved in the aqueous phase. A low Kd value for CBZ was found in the present study, which is in line with the results of the EU POSEIDON project (http://poseidon.bafg.de) that CBZ does not adsorb onto the sludge. In the present study, Kd values were calculated from the measured concentrations of PhACs in wastewater and sewage sludge. However, in many studies, Kd values are derived from the Kow values and the fraction of the organic carbon (ƒoc) in the sludge, as shown in the following equation (Jones et al., 2002; Yu et al., 2013).

Kd ¼ ƒoc 0:41 Kow :

ð8Þ

A comparison suggested that the modeled values (calculated by Eq. (8)) are one to several orders of magnitude greater than the measured values for hydrophobic pollutants (in Table S9), although the Kd values for GFB are not available. On the contrary, the predicted Kd values derived from Eq. (8) were lower than the measurements for polar or ionic PhACs (such as FQ antibiotics) in an aquatic environment (shown in Table S9). Thus, Eq. (8) can lead to unreasonable Kd values, and a new approach that considers the properties and the polarity of the functional groups of a molecule, and the structural and compositional difference in the matrices (i.e., the sludge) is needed for modeling the sorption constants of PhACs. 4. Conclusions Twenty-one target PhACs from 8 therapeutic classes covering a wide range of physicochemical properties and biological activities were analyzed at four full-scale WWTPs in Chongqing, the Three Gorges Reservoir Area. All the 21 analyzed PhACs were detected in wastewater at concentrations ranging from low ng/L to a few μg/L. Of all 21 target PhACs, 18 were present in sludge and most PhACs were found at ng/g dw levels. The concentrations of the target PhACs except antibiotics in the influent and effluent were lower than those reported in developed countries. In China, strengthening antibiotic management and sensible applications is of utmost importance, given that antibiotics could pose more potential hazardous effects on biota and the ecosystem, and a few ng/L levels of antibiotic residues in the environment could

contribute to the widespread resistance of bacterial pathogens. What is clear is that the elimination of PhACs except ACM is incomplete. The measured overall removal during the treatment varied widely among individual PhACs, from “negative removal” (e.g. CBZ, CA and MTP) to 99.5% (ACM), mainly depending on their physicochemical properties and the wastewater treatment process utilized at each WWTP. The calculation may have underestimated the removal efficiencies due to several factors: the removal efficiencies were calculated from the mean concentrations based on grab samples; the metabolism and/or transformation products of the PhACs were not defined; a change in the adsorption behavior of the PhACs onto particles; and signal suppression by the MS/MS detector. The total amount of the selected PhACs discharged into the receiving waters and sludge from the four WWTPs reached up to 2069 g and 171 g per day, respectively, with SMZ having the substantial effluent load of 447 μg/d/person and AZM having the highest sludge load with 39 μg/d/person in Jiguanshi WWTP. Based on the result of the risk assessment for the effluent, receiving water and sludge, the environment concentrations of single compounds (including sulfadiazine, sulfamethoxazole, ofloxacin, azithromycin and erythromycin-H2O) in effluent and sludge, as well as the mixture of the 21 detected PhACs in effluent, sludge and receiving water had a significant ecotoxicological risk to algae. The concentration addition model was used to estimate the toxicity of the mixture of the 21 PhACs. The overall RQ (sum) for algae was more than unity in the effluent, receiving water and sludge. Thus, the mixture can lead to a high risk in the Chongqing water environment, which must not be ignored. From the perspective of environment risk, further removal of PhACs could be achieved by adopting advanced treatment methods, such as activated carbon adsorption, ozonation or advanced oxidation, membrane separation and constructed wetlands. However, further research is required to assess the economic feasibility of such measures. The sorption of PhACs to sludge was low, with Kd below 500 L/kg for the studied PhACs except QAs, AZM and SVT. The Kd values of the selected PhACs were generally higher for the secondary sludge than for the primary sludge. For most of the selected pharmaceutical residues, the portion wasted with the treated sludge into the environment was negligible. However, in some cases, the sorption of QAs, MAs and cholesterol lowering statin drugs contributed to their elimination from the aqueous phase with a certain percentage, indicating the importance of sludge analysis. Only about 1.5% of the total daily mass load of the selected PhACs, determined from mass balance analysis, was retained


by sludge. These results indicate that biotransformation is the primary removal mechanism of the target PhACs during biological wastewater treatment. Conflict of interest statement We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript entitled. Acknowledgment We are grateful to Professor Baozhong Meng at the University of Guelph for the manuscript revise. The authors would like to thank Chongqing Water Group for their assistance during the sample collection. This study was funded by the National Science Foundation of China (Grant No. 51178484), the Chongqing Science and Technology Commission (Grant Nos. CSTC2012ggB20001 and CSTC2012jjA0775) and the National Science and Technology Support Program (Grant Nos. 2012BAJ25B09 and 2012BAJ25B06). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2013.09.032. References Backhaus T, Faust M. Predictive environmental risk assessment of chemical mixtures: a conceptual framework. Environ Sci Technol 2012;46:2564–73. Backhaus T, Scholze M, Grimme LH. The single substance and mixture toxicity of quinolones to the bioluminescent bacterium Vibrio fischeri. Aquat Toxicol 2000;49:49–61. Behera SK, Kim HW, Oh JE, Park HS. Occurrence and removal of antibiotics, hormones and several other pharmaceuticals in wastewater treatment plants of the largest industrial city of Korea. Sci Total Environ 2011;409:4351–60. Busch PL, Stumm W. Chemical interactions in the aggregation of bacteria bioflocculation in waste treatment. Environ Sci Tech 1968;2:49–53. Buth JM, Arnold WA, McNeill K. Unexpected products and reaction mechanisms of the aqueous chlorination of cimetidine. Environ Sci Technol 2007;41:6228–33. Chang X, Meyer MT, Liu X, Zhao Q, Chen H, Chen JA, et al. Determination of antibiotics in sewage from hospitals, nursery and slaughter house, wastewater treatment plant and source water in Chongqing region of Three Gorge Reservoir in China. Environ Pollut 2010;158:1444–50. Choi K, Kim Y, Park J, Park CK, Kim M, Kim HS, et al. Seasonal variations of several pharmaceutical residues in surface water and sewage treatment plants of Han River, Korea. Sci Total Environ 2008;405:120–8. Clara M, Strenn B, Gans O, Martinez E, Kreuzinger N, Kroiss H. Removal of selected pharmaceuticals, fragrances and endocrine disrupting compounds in a membrane bioreactor and conventional wastewater treatment plants. Water Res 2005;39:4797–807. Cleuvers M. Mixture toxicity of the anti-inflammatory drugs diclofenac, ibuprofen, naproxen, and acetylsalicylic acid. Ecotoxicol Environ Saf 2004;59:309–15. Daughton CG, Ternes TA. Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ Health Perspect 1999;107(Suppl. 6):907–38. de Souza SM, Vasconcelos EC, Dziedzic M, de Oliveira CM. Environmental risk assessment of antibiotics: an intensive care unit analysis. Chemosphere 2009;77:962–7. Dodd MC, Huang CH. Transformation of the antibacterial agent sulfamethoxazole in reactions with chlorine: kinetics, mechanisms, and pathways. Environ Sci Technol 2004;38:5607–15. EPA 832-R-06-005. Emerging technologies for biosolids management. Washington, DC: Office of Wastewater Management, US Environmental Protection Agency; 2006 [September]. European Commission. Technical guidance document in support of commission directive 93/67/EEC on risk assessment for new notified substances and commission regulation (EC) no 1488/94 on risk assessment for existing substances, part II; 2003 [Brussels, Belgium]. Gao L, Shi Y, Li W, Niu H, Liu J, Cai Y. Occurrence of antibiotics in eight sewage treatment plants in Beijing, China. Chemosphere 2012a;86:665–71. Gao P, Mao D, Luo Y, Wang L, Xu B, Xu L. Occurrence of sulfonamide and tetracyclineresistant bacteria and resistance genes in aquaculture environment. Water Res 2012b;46:2355–64. Gobel A, Thomsen A, McArdell CS, Joss A, Giger W. Occurrence and sorption behavior of sulfonamides, macrolides, and trimethoprim in activated sludge treatment. Environ Sci Technol 2005;39:3981–9.

629

Gobel A, McArdell CS, Joss A, Siegrist H, Giger W. Fate of sulfonamides, macrolides, and trimethoprim in different wastewater treatment technologies. Sci Total Environ 2007;372:361–71. Golet EM, Xifra I, Siegrist H, Alder AC, Giger W. Environmental exposure assessment of fluoroquinolone antibacterial agents from sewage to soil. Environ Sci Technol 2003;37:3243–9. Gomez MJ, Martinez Bueno MJ, Lacorte S, Fernandez-Alba AR, Aguera A. Pilot survey monitoring pharmaceuticals and related compounds in a sewage treatment plant located on the Mediterranean coast. Chemosphere 2007;66:993–1002. Gulkowska A, Leung HW, So MK, Taniyasu S, Yamashita N, Yeung LW, et al. Removal of antibiotics from wastewater by sewage treatment facilities in Hong Kong and Shenzhen, China. Water Res 2008;42:395–403. Hang PT, Brindley GW. Methylene blue absorption by clay minerals. Determination of surface areas and cation exchange capacities (clay-organic studies XVIII). Clays Clay Miner 1970;18:203–12. Hernando MD, Mezcua M, Fernandez-Alba AR, Barcelo D. Environmental risk assessment of pharmaceutical residues in wastewater effluents, surface waters and sediments. Talanta 2006;69:334–42. Horsing M, Ledin A, Grabic R, Fick J, Tysklind M, la Cour Jansen J, et al. Determination of sorption of seventy-five pharmaceuticals in sewage sludge. Water Res 2011;45: 4470–82. Huang JJ, Hu HY, Tang F, Li Y, Lu SQ, Lu Y. Inactivation and reactivation of antibioticresistant bacteria by chlorination in secondary effluents of a municipal wastewater treatment plant. Water Res 2011;45:2775–81. Huerta-Fontela M, Galceran MT, Ventura F. Occurrence and removal of pharmaceuticals and hormones through drinking water treatment. Water Res 2011;45:1432–42. Jelic A, Gros M, Ginebreda A, Cespedes-Sanchez R, Ventura F, Petrovic M, et al. Occurrence, partition and removal of pharmaceuticals in sewage water and sludge during wastewater treatment. Water Res 2011;45:1165–76. Jelic A, Fatone F, Di Fabio S, Petrovic M, Cecchi F, Barcelo D. Tracing pharmaceuticals in a municipal plant for integrated wastewater and organic solid waste treatment. Sci Total Environ 2012;433:352–61. Jia A, Wan Y, Xiao Y, Hu J. Occurrence and fate of quinolone and fluoroquinolone antibiotics in a municipal sewage treatment plant. Water Res 2012;46:387–94. Jones OA, Voulvoulis N, Lester JN. Aquatic environmental assessment of the top 25 English prescription pharmaceuticals. Water Res 2002;36:5013–22. Karthikeyan KG, Meyer MT. Occurrence of antibiotics in wastewater treatment facilities in Wisconsin, USA. Sci Total Environ 2006;361:196–207. Kasprzyk-Hordern B, Dinsdale RM, Guwy AJ. The removal of pharmaceuticals, personal care products, endocrine disruptors and illicit drugs during wastewater treatment and its impact on the quality of receiving waters. Water Res 2009;43:363–80. Khan SJ, Ongerth JE. Occurrence and removal of pharmaceuticals at an Australian sewage treatment plant; 200535–9. Kim S, Aga DS. Potential ecological and human health impacts of antibiotics and antibiotic-resistant bacteria from wastewater treatment plants. J Toxicol Environ Health B Crit Rev 2007;10:559–73. Leung HW, Minh TB, Murphy MB, Lam JC, So MK, Martin M, et al. Distribution, fate and risk assessment of antibiotics in sewage treatment plants in Hong Kong, South China. Environ Int 2012;42:1–9. Li B, Zhang T. Mass flows and removal of antibiotics in two municipal wastewater treatment plants. Chemosphere 2011;83:1284–9. Lin AY, Yu TH, Lin CF. Pharmaceutical contamination in residential, industrial, and agricultural waste streams: risk to aqueous environments in Taiwan. Chemosphere 2008;74:131–41. Lindberg RH, Wennberg P, Johansson MI, Tysklind M, Andersson BA. Screening of human antibiotic substances and determination of weekly mass flows in five sewage treatment plants in Sweden. Environ Sci Technol 2005;39:3421–9. Lindsey ME, Meyer TM, Thurman EM. Analysis of trace levels of sulfonamide and tetracycline antimicrobials in groundwater and surface water using solid-phase extraction and liquid chromatography/mass spectrometry. Anal Chem 2001;73:4640–6. Lishman L, Smyth SA, Sarafin K, Kleywegt S, Toito J, Peart T, et al. Occurrence and reductions of pharmaceuticals and personal care products and estrogens by municipal wastewater treatment plants in Ontario, Canada. Sci Total Environ 2006;367:544–58. Lopez-Serna R, Postigo C, Blanco J, Perez S, Ginebreda A, de Alda ML, et al. Assessing the effects of tertiary treated wastewater reuse on the presence emerging contaminants in a Mediterranean river (Llobregat, NE Spain). Environ Sci Pollut Res Int 2012;19: 1000–12. Martin J, Camacho-Munoz D, Santos JL, Aparicio I, Alonso E. Occurrence of pharmaceutical compounds in wastewater and sludge from wastewater treatment plants: removal and ecotoxicological impact of wastewater discharges and sludge disposal. J Hazard Mater 2012;239–240:40–7. Martinez JL. Antibiotics and antibiotic resistance genes in natural environments. Science 2008;321:365–7. Miao XS, Yang JJ, Metcalfe CD. Carbamazepine and its metabolites in wastewater and in biosolids in a municipal wastewater treatment plant. Environ Sci Technol 2005;39: 7469–75. Muñoz I, Gómez-Ramos MJ, Agüera A, García-Reyes JF, Molina-Díaz A, Fernández-Alba AR. Chemical evaluation of contaminants in wastewater effluents and the environmental risk of reusing effluents in agriculture. TrAC Trends Anal Chem 2009;28: 676–94. Ort C, Lawrence MG, Reungoat J, Mueller JF. Sampling for PPCPs in wastewater systems: comparison of different sampling modes and optimization strategies. Environ Sci Technol 2010;44:6289–96. Pinkston KE, Sedlak DL. Transformation of aromatic ether- and amine-containing pharmaceuticals during chlorine disinfection. Environ Sci Technol 2004;38:4019–25.

630


Pomati F, Castiglioni S, Zuccato E, Fanelli R, Vigetti D, Rossetti C, et al. Effects of a complex mixture of therapeutic drugs at environmental levels on human embryonic cells. Environ Sci Technol 2006;40:2442–7. Quinn B, Gagne F, Blaise C. Evaluation of the acute, chronic and teratogenic effects of a mixture of eleven pharmaceuticals on the cnidarian, Hydra attenuata. Sci Total Environ 2009;407:1072–9. Radjenovic J, Petrovic M, Barcelo D. Fate and distribution of pharmaceuticals in wastewater and sewage sludge of the conventional activated sludge (CAS) and advanced membrane bioreactor (MBR) treatment. Water Res 2009;43:831–41. Roberts PH, Thomas KV. The occurrence of selected pharmaceuticals in wastewater effluent and surface waters of the lower Tyne catchment. Sci Total Environ 2006;356: 143–53. Rosal R, Rodriguez A, Perdigon-Melon JA, Petre A, Garcia-Calvo E, Gomez MJ, et al. Occurrence of emerging pollutants in urban wastewater and their removal through biological treatment followed by ozonation. Water Res 2010;44:578–88. Samaras VG, Stasinakis AS, Mamais D, Thomaidis NS, Lekkas TD. Fate of selected pharmaceuticals and synthetic endocrine disrupting compounds during wastewater treatment and sludge anaerobic digestion. J Hazard Mater 2013;244–245:259–67. Santos LH, Araujo AN, Fachini A, Pena A, Delerue-Matos C, Montenegro MC. Ecotoxicological aspects related to the presence of pharmaceuticals in the aquatic environment. J Hazard Mater 2010;175:45–95. Sui Q, Huang J, Deng S, Yu G, Fan Q. Occurrence and removal of pharmaceuticals, caffeine and DEET in wastewater treatment plants of Beijing, China. Water Res 2010;44:417–26.

USEPA. Method 1694: Pharmaceuticals and Personal Care Products in Water, Soil, Sediment, and Biosolids by HPLC/MS/MS. Washington, D.C.: EPA-821-R-08-002; 2007b Verlicchi P, Al Aukidy M, Zambello E. Occurrence of pharmaceutical compounds in urban wastewater: removal, mass load and environmental risk after a secondary treatment —a review. Sci Total Environ 2012;429:123–55. Wilson BA, Smith VH, deNoyelles Jr F, Larive CK. Effects of three pharmaceutical and personal care products on natural freshwater algal assemblages. Environ Sci Technol 2003;37:1713–9. Xu W, Zhang G, Li X, Zou S, Li P, Hu Z, et al. Occurrence and elimination of antibiotics at four sewage treatment plants in the Pearl River Delta (PRD), South China. Water Res 2007;41:4526–34. Yang X, Flowers RC, Weinberg HS, Singer PC. Occurrence and removal of pharmaceuticals and personal care products (PPCPs) in an advanced wastewater reclamation plant. Water Res 2011;45:5218–28. Yu Y, Wu L, Chang AC. Seasonal variation of endocrine disrupting compounds, pharmaceuticals and personal care products in wastewater treatment plants. Sci Total Environ 2013;442:310–6. Zhao JL, Ying GG, Liu YS, Chen F, Yang JF, Wang L, et al. Occurrence and a screening-level risk assessment of human pharmaceuticals in the Pearl River system, South China. Environ Toxicol Chem 2010;29:1377–84. Zhu YG, Johnson TA, Su JQ, Qiao M, Guo GX, Stedtfeld RD, et al. Diverse and abundant antibiotic resistance genes in Chinese swine farms. Proc Natl Acad Sci U S A 2013;110: 3435–40.