833 © IWA Publishing 2012 Water Science & Technology | 65.5 | 2012

The removal and degradation of pharmaceutical

compounds during membrane bioreactor treatment

H. Fr. Schröder, J. L. Tambosi, R. F. Sena, R. F. P. M. Moreira, H. J. José

and J. Pinnekamp

ABSTRACT

Pharmaceutical compounds such as non-steroidal anti-inflammatory drugs (NSAIDs) and antibiotics

have been detected in sewage treatment plant (STP) effluents, surface and ground water and even in

drinking water all over the world, and therefore have developed as compounds of concern.

Membrane bioreactor (MBR) treatment has gained significant popularity as an advanced wastewater

treatment technology and might be effective for an advanced removal of these pollutants. This paper

evaluates the treatment of wastewater containing three NSAIDs (acetaminophen, ketoprofen and

naproxen) and three antibiotics (roxithromycin, sulfamethoxazole and trimethoprim) performed in

two MBRs with sludge retention times (SRTs) of 15 (MBR-15) and 30 (MBR-30) days over a period of

four weeks. It was observed that NSAIDs were removed with higher efficiencies than the antibiotics

for both MBRs, and the MBR-30 presented higher removal efficiencies for all the compounds than

obtained by MBR-15. Removal rates ranged from 55% (sulfamethoxazole) up to 100%

(acetaminophen, ketoprofen). Besides mineralisation biological transformation products of

ketoprofen and naproxen produced by wastewater biocoenosis were identified in both MBR

permeates using liquid chromatography coupled with mass spectrometry (LC-MS). The results

indicated the importance of investigating the environmental fate of pharmaceuticals and their

transformation products reaching the environment.

doi: 10.2166/wst.2012.828

H. Fr. Schröder (corresponding author)J. L. TambosiR. F. SenaJ. PinnekampInstitute of Environmental Engineering,RWTH Aachen University,Aachen,GermanyE-mail: hf.schroeder@post.rwth-aachen.de

J. L. TambosiR. F. SenaR. F. P. M. MoreiraH. J. JoséDepartment of Chemical Engineering and Food

Engineering,Federal University of Santa Catarina,Florianópolis,Brazil

Key words | antibiotics, high resolution mass spectrometry (HRMS), membrane bioreactor (MBR),

non-steroidal anti-inflammatory drugs (NSAID), transformation products, wastewater

treatment

INTRODUCTION

To reach the desired therapeutic effects of drugs normallyapplied under therapeutic use, on the one hand pharma-

ceutical compounds are designed to be stable afteringestion till the resorption in the digestive tract has takenplace. On the other hand, these drugs should be polar struc-tured to facilitate renal excretion after their successful

absorption and after the desired therapeutic effects hastaken place. Both properties, persistence against biochemicaldegradation and polar structure, however, may be respon-

sible for the incomplete removal during conventionalbiological sewage treatment. With the rapid developmentof analytical techniques, it has been reported that the aquatic

environment will become more and more polluted withpharmaceutically active compounds (drugs) at low concen-trations (Gebhardt & Schröder ; Göbel et al. ).

Different sources can be indicated to explain the appearanceof drugs in the aquatic environment. Nowadays, it is widely

accepted that the main source of pollution are sewage treat-ment plant (STP) effluents (Andreozzi et al. ). Theoccurrence of several drugs has been reported in STP efflu-ents as well as in surface and drinking water in Brazil

(Stumpf et al. ; Bieling et al. ; Favier et al. ),Canada (Miao et al. ), China (Xu et al. ), Germany(Kümmerer ), Italy (Andreozzi et al. ), Spain

(Carballa et al. ), Switzerland (Tauxe-Wuersch et al.) and the United States (Brown et al. ).

Consumption of drugs, such as non-steroidal anti-inflam-

matory drugs (NSAIDs) and antibiotics belonging to classesof pharmaceuticals that are extensively used worldwideand consumed predominantly in developed countries, is

834 H. Fr. Schröder et al. | Removal and degradation of pharmaceutical compounds Water Science & Technology | 65.5 | 2012

assumed to be higher than several hundreds of tons per year

(Daughton & Ternes ; Kümmerer ). It is supposedthat in 1995 in Germany alone more than 100 t of thesedrugs were prescribed by doctors to their patients. Recent

estimations indicate that in Europe, which holds about26% of the international pharmaceutical market, morethan 2000 different pharmaceutical products are used. Sothe annual consumption of antibiotic-type substances is

comparable in quantity with that of some pesticides appliedin agriculture (Stackelberg et al. ; Molinari et al. ).

Some of the adverse effects caused by drug pollution

include aquatic toxicity, resistance development in patho-genic bacteria, genotoxicity, and endocrine disruption(Le-Clech et al. ). The presence of trace pharmaceutical

and other xenobiotic compounds in drinking water is ofpublic health concern, as little is known about potentialchronic health effects associated with long-term ingestionof mixtures of these compounds with the drinking water

(Quintana et al. ).As conventional water and wastewater treatment pro-

cesses are unable to act as a reliable barrier towards some

pharmaceuticals, it is necessary to introduce and applyadditional advanced treatment technologies. Membranebioreactor (MBR) technology combines the biological

degradation process using activated sludge with a directsolid–liquid separation by membrane filtration (Terneset al. ). MBR treatment has gained significant popularity

recently and can be useful for the above mentioned pur-poses. Drug removal by MBRs has been reported in theliterature (Gebhardt & Schröder ; Göbel et al. ).

In this study, six pharmaceuticals of large consumption

worldwide, three NSAIDs (acetaminophen, ketoprofen andnaproxen) and three antibiotics (roxithromycin, sulfa-methoxazole and trimethoprim) were selected to monitor

their fate during MBR pilot plant treatment, as well astheir total removal efficiencies. The qualitative and quanti-tative monitoring of the compounds along the MBR

treatment was performed using liquid chromatographycoupled with mass spectrometry (LC-MS).

METHODS

Materials

Ultra-pure water used during sample treatment or as LCmobile phase component was prepared using a Milli-Q

system (Millipore, Milford, MA, USA). All solvents used asmobile phases, for desorption of the pharmaceuticals and

their potential degradation products extracted by solid

phase extraction (SPE), were nanograde solvents purchasedfrom LGC Promochem (Wesel, Germany). All other chemi-cals used were of ‘analytical reagent’ or ‘residue analysis’

purity grade. Gases applied were products of Linde,Germany, and were of 99.99% purity.

The pharmaceutical compounds used in this studywere purchased from Sigma-Aldrich. Relevant information

about these pharmaceuticals is given in Table 1. Concen-trated stock solutions were used for the analyticaldetermination and for spiking purposes. To avoid degra-

dation during the test period these methanolic solutionswere kept at �18 WC.

Membrane bioreactor (MBR) pilot plants

Wastewater used as feed for the MBR pilot treatment plant

was taken continuously from the effluent of the pre-settlingtank of the municipal STP of Aachen, Germany. MBR treat-ment was performed over a period of four weeks. The sludgeretention time (SRT), sludge concentration (SC) and the

hydraulic retention time (HRT) of the pilot plants were15 d, 12 g/L and 9 h for MBR-15, and 30 d, 12 g/L and13 h for MBR-30, respectively. Both MBRs used in this

study were equipped with hollow-fibre ultrafiltration (UF)membranes (PURON, KMS Germany), 1.43 m2 in dimen-sion. The nominal pore size and material of the membrane

were 0.04 μm and polyethersulfone (PES), respectively.The steady-state in the MBR was reached over six monthsbefore the final monitoring tests started. The ideal mixingin the MBRs was monitored and confirmed by using a con-

ductivity probe. This was accomplished by a sodiumchloride (NaCl) addition applied as tracer compound in par-allel to the daily drug dosages.

Spiking and sampling

For spiking a mixture of a defined stock solution of thetarget pharmaceuticals in combination with a NaCl solutionwas spiked into the feed of each MBR to reach a mean

steady-state concentration of 50 μg/L of the target pharma-ceuticals and 0.8 g/L NaCl. Within the week, spiking wasperformed by the addition of 2/3 (33.33 μg/L) of the totaldaily spiking quantity in the morning (10:00 a.m.) and 1/3

(16.67 μg/L) in the afternoon (04:00 p.m.). Spiking duringthe weekend was performed by the addition of the totalamount of 50 μg/L at 04:00 p.m. in the afternoon. Every

day at 03:00 p.m. in the afternoon samples of 500 mL ofeach MBR and its permeate were taken. In parallel, a

Table 1 | Name, sum formula, CAS NW

, structure, pKa and log KOW of pharmaceuticals and transformation products under research (Kimura et al. 2005; Kim et al. 2007)

Name (Sum formula) CAS Structure Acidic, base or neutral pKa log Kow

Acetaminophen (C8H9O2N) 103-90-2 N 9.39 0.46

Ketoprofen (C16H14O3) 22071-15-4 A 4.45 3.12

Naproxen (C14H14O3) 22204-53-1 A 4.15 3.18

Roxithromycin (C41H76N2O15) 80214-83-1 B 8.8–9.2 2.75

Sulfamethoxazole (C10H11N3O3S) 723-46-6 A 1.8–5.7 0.89

Trimethoprim (C14H18N4O3) 738-70-5 B 6.6–7.2 0.91

Ketoprofen transformation product (C11H12O5) – A – –

Naproxen transformation product (C13H12O3) – A – –

835 H. Fr. Schröder et al. | Removal and degradation of pharmaceutical compounds Water Science & Technology | 65.5 | 2012

sludge quantity of 16L (MBR-15) or 8L (MBR-30) of excesssludge was discharged in order to keep the sludge concen-

tration (SC) constant in each MBR.

Sample preparation

The pharmaceuticals present in the MBR permeates wereconcentrated using commercially available solid phase extrac-

tion (SPE) cartridges (1 mL) filled with 100 mg of IsoluteENVþmaterial from IST (Mid Glamorgan, UK). Prior to

use SPE cartridges were handled as prescribed by the manu-facturer. After the SPE procedure, the cartridges were

rinsed with three bed volumes of ultra-pure water to removesalts prior to being dried in a gentle stream of nitrogen at30 WC. The pharmaceuticals adsorbed at the SPE material

were desorbed by the in-series addition of 6 × 1 mL of metha-nol. STP eluates were brought to dryness in a gentle streamof nitrogen at 60 WC. The dry residues containing the pharma-ceuticals were reconstituted in 1 mL of methanol/water (1:1)

and were used for injection during LC-MS analysis.

836 H. Fr. Schröder et al. | Removal and degradation of pharmaceutical compounds Water Science & Technology | 65.5 | 2012

Liquid chromatographic and MS data

LC-separations were carried out with a Hypersil GOLD aQcolumn (RP5, 5 μm, spherical; 150 × 2.1 mm I.D.) equipped

with a Hypersil GOLD aQ pre-column (10 × 2.1 mm I.D.),also filled with 5 μm spherical material (Thermo Electron,USA). Gradient elution by means of (A) methanol/water90:10 (v:v) in combination with (B) Milli-Q-purified water/

methanol 90:10 (v:v) was applied, both containing 2 mMammonium acetate. The gradient was programmed as fol-lows. Starting with 20% A/80% B the concentration was

increased linearly to 90% A/10% B within 12 min. The com-position was kept constant for 20 min. The overall flow ratewas 0.2 mL/min. Instrument control data acquisition and

data processing were performed using Xcalibur 2.0 software(Thermo Electron). Identification and quantification ofspiked pharmaceuticals as well as recognition and identifi-cation of transformation products, respectively, in MBR

permeates were performed by means of a LTQ Orbitrapmass spectrometer (Thermo Electron, Germany). Electro-spray ionization (ESI) was applied in the positive (for

roxithromycin, sulfamethoxazole, trimethroprim and aceta-minophen) or in the negative (for ketoprofen andnaproxen and their transformation products) ionization

mode as described by Gebhardt & Schröder (). To cal-culate the efficiency of drug removal from the spikedwastewater, the concentrations of the precursor drugs in

the MBR inflow were determined in parallel quantitativelyusing LC-MS after SPE.

Recovery, determination and quantification of spikedpharmaceuticals in feed and permeates were performed by

means of LC-MS applying ESI ionization in positive or

Figure 1 | Mean elimination rates of pharmaceutical compounds during a four week MBR trea

days (MBR-30).

negative mode (Bieling et al. ) under high resolution

mass spectrometric conditions (HRMS). The limits of detec-tion (LODs) and the limits of quantification (LOQs) inLC-MS mode were calculated by a signal-to-noise ratio of

3 (S/N 3:1).

RESULTS AND DISCUSSION

Removal of pharmaceutical compounds

The mean removal rates of target pharmaceuticals spiked

into two MBRs with different SRTs and treated over aperiod of four weeks are shown in Figure 1. During the spik-ing period we observed the behaviour of the compounds

differing one from another. This will be discussed below.A complete removal of the hydrophilic compound acet-

aminophen (log Kow< 1) in both MBRs was observed (cf.Figure 1) due to its less complex chemical structure. Elimin-

ation could be cleared up by LC-MS monitoring asbiodegradation. These results are in agreement with theresults reported by Kim et al. (), who obtained a 99%

removal of acetaminophen during the treatment of munici-pal wastewater in a MBR pilot plant.

Concerning the ketoprofen behaviour, an almost com-

plete removal of the precursor drug can also be observedfor this compound in both MBRs. The compound ketopro-fen has a moderate hydrophobic nature (log Kow> 3) andacidic character. According to Quintana et al. (), forpolar compounds like acidic pharmaceuticals, microbialdegradation obviously is the most important removal pro-cess in activated sludge wastewater treatment while

tment period applying in parallel two different sludge ages. □: 15 days (MBR-15) or ▪: 30

837 H. Fr. Schröder et al. | Removal and degradation of pharmaceutical compounds Water Science & Technology | 65.5 | 2012

neither retention by adsorption at the sewage sludge nor

membrane material will take place. Kimura et al. ()compared the elimination of ketoprofen treating municipalwastewater in parallel using either an MBR pilot plant or a

conventional activated sludge treatment (CAST) plant.Wastewater treatment in a CAST plant led to a concen-tration of 300 ng/L of ketoprofen in the effluent while inthe permeate after MBR treatment only about 10 ng/L was

found.The compound naproxen has comparable physicochem-

ical properties as ketoprofen, however, naproxen removal

in both MBRs was lower than ketoprofen removal. This be-haviour of naproxen can be partially explained by its moresteric complex chemical structure stabilized by an aromatic

naphthalene ring system. According to the study of Quintanaet al. () on microbial degradation of pharmaceuticals, thedegradation of naproxen was found to be low with anapproximately 60% transformation in 28 days. In parallel

only one transformation product could be detected.The compound roxithromycin has a low hydrophobic

nature (log Kow¼ 2.75) and a basic character. Roxithromy-

cin equipped with the most complex chemical structurecompared to the target compounds acts as an antibacterialagent. Similar results for roxithromycin removal for MBR

were reported by Göbel et al. (), where the eliminationvaried between 39% for an SRT of 16 days and 60% forhigher SRTs (33 and 60 days).

The compound sulfamethoxazole is a hydrophilic com-pound (log Kow< 1) with two ionizable amine groups. AtpH values between the pKa values of the compound (pH1.8 and 5.7), sulfamethoxazole is present predominantly as

a neutral species, while above the second pKa value of thecompound (pH 5.7), it becomes a negatively chargedspecies. These physicochemical properties give an indi-

cation that in the MBR system we studied, the sludgeadsorption mechanism at pH 7.2 played a negligible role,due to electrostatic repulsion between the negatively

charged groups of the compound and the negatively chargedsurfaces of the sludge. Therefore, biodegradation has to bethe main mechanism responsible for the removal, however,

will be diminished by its antibacterial property. Göbel et al.() who also studied the elimination of sulfamethoxazolereported an elimination efficiency of around 80%, indepen-dent of adjusted SRTs.

For the compound trimethoprim the highest removalefficiency among the antibiotics examined could beobserved. This can be partially explained by its hydrophilic

nature (log Kow< 1), the basic character and its reducedantibacterial potency compared to sulfamethoxazole and

roxithromycin. As also observable for the other drugs the tri-

methoprim removal observed for MBR-30 was 10% higherthan with the shorter SRT in the MBR-15 treatment.Göbel et al. () also studied the elimination of trimetho-

prim by MBR, reporting comparable elimination rates forSRT of 16 and 33 days (30%), while 87% of removal wasobtained for SRT of 60–80 days.

Mechanisms influencing the removal of pharmaceuticalcompounds

The elimination of pharmaceutical compounds can occurthrough various mechanisms during MBR treatment pro-cess. Sorption onto sludge is one of the mechanisms and

therefore the absorption and adsorption factors have to betaken into account. According to Carballa et al. (),absorption refers to the hydrophobic interactions of the ali-

phatic and aromatic groups of a compound with fats presentin the sludge or with the lipophilic cell membranes of themicroorganisms, depending on the target Kow value. Adsorp-tion refers to the electrostatic interactions of positively

charged groups of dissolved chemicals with the negativelycharged surfaces of the microorganisms (characterized bythe dissociation constant pKa). Göbel et al. () studied

the elimination of pharmaceuticals by MBRs as well asCAST and concluded that the contribution of activatedsludge adsorption in the case of pharmaceutical compounds

was less than 6%.The physical retention of the pharmaceutical targets we

have to take into account, was a result of adsorption by anincreased sludge concentration (MLSS) while no retardation

will happen by membrane separation. Membranes used inthe ultrafiltration MBR process have molecular weightcut-offs (MWCO) of approx. 100,000 Da while target phar-

maceuticals had molecular weights below 1,000 Da. Targetcompounds which are unpolar will adsorb onto the biomassand therefore will be removed together with the excess

sludge, while polar drugs, with a low tendency to adsorb atthe lipophilic sludge surface, will neither be eliminated byadsorption nor by biodegradation. The reason is that the

interaction with wastewater biocoenosis essential for the bio-degradation process may be too short for a degradation.

Nevertheless, biochemical degradation is the mostimportant elimination mechanism for the target compounds

in wastewater. This degradation often has transformationproducts which are more stable than their precursor drugs.In our examinations part of the transformation products

observed by Quintana et al. () for ketoprofen andnaproxen were also observed, with their structure as

838 H. Fr. Schröder et al. | Removal and degradation of pharmaceutical compounds Water Science & Technology | 65.5 | 2012

shown in Table 1. Their recognition in the wastewater was

only possible by extraction of their mass traces from thenegatively generated TICs recorded in HRMS mode. There-fore, the presence of the ketoprofen transformation product

3-(hydroxy-carboxymethyl)hydratropic acid as an intermedi-ate compound and the naproxen transformation productO-desmethyl-naproxen could easily be confirmed.

CONCLUSIONS

The performance of two MBR pilot plants with submergedmembranes was examined in this study. The results obtainedproved that the MBR-30 presented higher removal efficien-

cies for all the compounds than obtained by MBR-15. Thecompounds acetaminophen and ketoprofen had the highestremoval efficiencies, while roxithromycin and sulfamethox-

azole as bacteriostatics exhibited persistence to microbialattack and were removed to a less extent in both MBRs.Concerning the potential mechanisms responsible for theremoval of the target pharmaceuticals in MBRs (sludge sorp-

tionþ biodegradationþmembrane retention), it was notpossible to determine exactly to what extent each mechan-ism contributed to the removal efficiency because excess

sludge analysis was not performed in this examination.Retention by membranes using microfiltration or ultrafiltra-tion membranes with MWCO of 100,000 Da can be

neglected. Biodegradation, however, played an importantrole, since higher removal efficiencies was obtained forhigher SRTs. Nevertheless, the elimination by MBR treat-ment using ultrafiltration was only partially successful and

therefore, persistent pharmaceuticals in small concen-trations were discharged with the wastewater into theenvironment. This discharge could be reduced with the

application of additional post treatment steps usingadvanced treatment techniques, e.g., activated carbonadsorption, ozone oxidation or advanced oxidation pro-

cesses (AOP). In conclusion, the results indicated theimportance of investigating the generation and environ-mental fate of transformation products of pharmaceuticals,

especially their whereabouts during wastewater treatmentprocess. Thus, additional research is overdue to assess theimpact of these compounds, hitherto hardly known butmore polar than their precursors and therefore quite

mobile in the aquatic environment. Discharged with treat-ment plant effluents into surface waters the compoundswill reach ground water predetermined to reach drinking

water treatment and finally drinking water (Stackelberget al. ; Kim et al. ).

ACKNOWLEDGEMENTS

The authors would like to thank the Coordenação de Aper-

feiçoamento de Pessoal de Nível Superior (CAPES) and theDeutscher Akademischer Austauschdienst (DAAD, GermanAcademic Exchange Service) for financial support. Analyti-cal support from the staff of the Environmental Analytical

Laboratory of the Institute of Environmental Engineeringof RWTH Aachen University is also kindly acknowledged.

REFERENCES

Andreozzi, R., Marotta, R. & Paxéus, N. Pharmaceuticals inSTP effluents and their solar photodegradation in aquaticenvironment. Chemosphere 50, 1319–1330.

Bieling, U., Sena, R. F., Tambosi, J. L., Gebhardt, W., Moreira,R. F. P. M., José, H. J. & Schröder, H. Fr. LC-MS-and -MSn -determination of anthropogenic and biogenicendocrine disruptor compounds (EDC) with an estrogenicpotential in Brazilian wastewater and surface waters. In:International Conference on Sustainable Sanitation, ‘Foodand Water Security for Latin America’ (A. B. dos Santos, ed.).Proceedings, Fortaleza, Ceará, Brazil, pp. 24/1–4.

Brown,K.D.,Kulis, J., Thomson,B.,Chapman,T.H.&Mawhinney,D. B. Occurrence of antibiotics in hospital, residential,and dairy effluent, municipal wastewater, and the Rio Grandein New Mexico. Sci. Total Environ. 366, 772–783.

Carballa, M., Omil, F. & Lema, J. M. Removal of cosmeticingredients and pharmaceuticals in sewage primarytreatment. Water Res. 39 (19), 4790–4796.

Daughton, C. G. & Ternes, T. A. Pharmaceuticals andpersonal care products on the environment: agents of subtlechange? Environ. Health Persp. 107, 907–938.

Favier, M., Sena, R. F., José, H. J. & Schröder, H. Fr. Liquidchromatography-tandem mass spectroscopy for the screeningof pharmaceuticals, their metabolites and transformationproducts in various water bodies in Florianópolis, SantaCatarina, Brazil. In: International Conference on SustainableSanitation, ‘Food and Water Security for Latin America’(A. B. dos Santos, ed.). Proceedings, Fortaleza, Ceará, Brazil,pp. 59/1–9.

Gebhardt, W. & Schröder, H. Fr. Liquid chromatography-(tandem) mass spectrometry for the follow-up of theelimination of persistent pharmaceuticals during wastewatertreatment applying biological wastewater treatment andadvanced oxidation. J. Chromatogr. A 1160, 34–43.

Göbel, A., McArdell, C. S., Joss, A., Siegrist, H. & Giger, W. Fate of sulfonamides, macrolides, and trimethoprim indifferent wastewater treatment technologies. Sci. TotalEnviron. 372, 361–371.

Kim, S. D., Cho, J., Kim, I. S., Vanderford, B. J. & Snyder, S. A. Occurrence and removal of pharmaceuticals andendocrine disruptors in South Korean surface, drinking, andwaste waters. Water Res. 41 (5), 1013–1021.

839 H. Fr. Schröder et al. | Removal and degradation of pharmaceutical compounds Water Science & Technology | 65.5 | 2012

Kimura, K., Hara, H. & Watanabe, Y. Removal ofpharmaceutical compounds by submerged membranebioreactors (MBRs). Desalination 178, 135–140.

Kümmerer, K. Drugs in environment: emission of drugs,diagnostic aids and disinfectants into wastewater by hospitalsin relation to other sources – a review. Chemosphere 45,957–969.

Kümmerer, K. Resistance in the environment. J. Antimicrob.Chemoth. 54, 311–320.

Le-Clech, P., Chen, V. & Fane, T. A. G. Fouling in membranebioreactors used in wastewater treatment. J. Membr. Sci. 284,17–53.

Miao, X. S., Bishay, F., Chen, M. & Metcalfe, C. D. Occurrence of antimicrobials in the final effluents ofwastewater treatment plants in Canada. Environ. Sci.Technol. 38, 3533–3541.

Molinari, R., Pirillo, F., Loddo, V. & Palmisano, L. Heterogeneous photocatalytic degradation of pharmaceuticalsin water by using polycrystalline TiO2 and a nanofiltrationmembrane reactor. Catal. Today 118, 205–213.

Quintana, J. B., Weiss, S. & Reemtsma, T. Pathways andmetabolites of microbial degradation of selected acidicpharmaceutical and their occurrence in municipal

wastewater treated by a membrane bioreactor. Water Res.39 (12), 2654–2664.

Stackelberg, P. E., Furlong, E. T., Meyer, M. T., Zaugg, S. D.,Henderson, A. K. & Reissman, D. B. Persistence ofpharmaceutical compounds and other organic wastewatercontaminants in a conventional drinking-water treatmentplant. Sci. Total Environ. 329, 99–113.

Stumpf, M., Ternes, T. A., Wilken, R. D., Rodrigues, S. V. &Baumann, W. Polar drug residues in sewage and naturalwaters in the state of Rio de Janeiro, Brazil. Sci. TotalEnviron. 225, 135–141.

Tauxe-Wuersch, A., De Alencastro, L. F., Grandjean, D. &Tarradellas, J. Occurrence of several acidic drugs insewage treatment plants in Switzerland and risk assessment.Water Res. 39 (9), 1761–1772.

Ternes, T. A., Bonerz, M., Hermann, N., Teiser, B. & Andersen,H. R. Irrigation of treated wastewater in Braunschweig,Germany: an option to remove pharmaceuticals and muskfragrances. Chemosphere 66, 894–904.

Xu, W., Zhang, G., Li, X., Zou, S., Li, P., Hu, Z. & Li, J. Occurrence and elimination of antibiotics at four sewagetreatment plants in the Pearl River Delta (PRD), SouthChina. Water Res. 41 (19), 4526–4534.

First received 23 May 2011; accepted in revised form 18 August 2011