Behavior Of Pharmaceuticals, Cosmetics And Hormones In A Sewage Treatment Plant.pdf

ARTICLE IN PRESS

Water Research 38 (2004) 2918–2926

Behavior of pharmaceuticals, cosmetics and hormones in a sewage treatment plant Marta Carballaa, Francisco Omila,*, Juan M. Lemaa, Mar!ıa Llompartb, Carmen Garc!ıa-Jaresb, Isaac Rodr!ıguezb, Mariano Go´mezc, Thomas Ternesd a

Department of Chemical Engineering, School of Engineering, University of Santiago de Compostela, E-15782 Santiago de Compostela, Spain b Department of Analytical Chemistry, Institute of Food Science, University of Santiago de Compostela, E-15706 Santiago de Compostela, Spain c Central Laboratory, Aquagest Galicia, Isidro Parga Pondal, 9, E-15702 Santiago de Compostela, Spain d Bundesanstalt fu¨r Gewa¨sserkunde, Am Mainzer Tor 1, D-56068 Koblenz, Germany Received 31 July 2003; received in revised form 22 January 2004; accepted 10 March 2004

Abstract Two cosmetic ingredients (galaxolide, tonalide), eight pharmaceuticals (carbamazepine, diazepam, diclofenac, ibuprofen, naproxen, roxithromycin, sulfamethoxazole and iopromide) and three hormones (estrone, 17b-estradiol and 17a-ethinylestradiol) have been surveyed along the different units of a municipal Sewage Treatment Plant (STP) in Galicia, NW Spain. Among all the substances considered, significant concentrations in the influent were only found for the two musks (galaxolide and tonalide), two anti-inflammatories (ibuprofen and naproxen), two natural estrogens (estrone, 17b-estradiol), one antibiotic (sulfamethoxazole) and the X-ray contrast medium (iopromide), where the other compounds studied were below the limit of quantification. In the primary treatment, only the fragrances (30–50%) and 17b-estradiol (20%) were partially removed. On the other hand, the aerobic treatment (activated sludges) caused an important reduction in all compounds detected, between 35% and 75%, with the exception of iopromide, which remained in the aqueous phase. The overall removal efficiencies within the STP ranged between 70–90% for the fragrances, 40–65% for the anti-inflammatories, around 65% for 17b-estradiol and 60% for sulfamethoxazole. However, the concentration of estrone increased along the treatment due to the partial oxidation of 17b-estradiol in the aeration tank. r 2004 Elsevier Ltd. All rights reserved. Keywords: Pharmaceuticals; Cosmetics; Hormones; PPCP; Wastewaters; Sewage treatment plant; Adsorption; Elimination; Antibiotics; Estrogens; Anti-inflammatories

1. Introduction Municipal wastewaters contain many organic compounds, among them, active ingredients of pharmaceuticals and personal care products, which are used in large quantities throughout the world. Both groups of *Corresponding author. Tel.: +34-981-56-31-00x16778; fax: +34-981-54-71-68. E-mail address: [email protected] (F. Omil).

chemicals will be collectively referred to as ‘‘Pharmaceutical & Personal Care Products’’ (PPCPs). Generally, drugs are absorbed by the organism after intake and are subject to metabolic reactions. However, a significant fraction of the original substances leave human or animal organisms unmetabolized via urine or feces being therefore emitted into raw sewage, sewage sludge or manure. For example, Høverstad et al. [1] determined several antibiotics in human feces during 6 days of regular application. Furthermore, some of the

0043-1354/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2004.03.029

ARTICLE IN PRESS M. Carballa et al. / Water Research 38 (2004) 2918–2926

excreted metabolites can even be transformed back to the original active drug [2]. 1.1. Sewage treatment plants Some of the most representative PPCPs found in Sewage Treatment Plants (STP) are antibiotics, lipid regulators, anti-inflammatories, antiepileptics, tranquillizers, contrast media and contraceptives with very different chemical structures. Because of them, a considerable effort is being made in order to develop the analytical techniques needed to quantify their occurrence in effluents, but also to assess their chemical properties, their biodegradability potential, etc. Recent works have reported the presence of a large variety of PPCPs in STP effluents and surface waters, with concentrations up to several mg l 1 [23]. In fact, more than 50 PPCPs have been detected during the last years in different environmental samples, due to the continuous improvement of the analytical techniques. Many of these samples have been taken from wastewater [3–5], but also from surface or groundwaters [6,7]. Most of these compounds come either from domestic sewage or from hospital or industrial discharges and enter municipal STPs. Modern STPs can effectively accomplish carbon and nitrogen removal, as well as microbial pollution control. However, urban STPs normally receive streams that contain a lot of different trace polluting compounds (synthetic and natural), for which conventional treatment technologies have not been specifically designed. Their removal efficiencies are influenced, apart by the chemical properties of specific compounds, by microbial activity and environmental conditions [8–10]. Recent studies have clearly shown that the elimination of PPCPs in municipal STPs is often incomplete [5], with efficiencies ranging between 60% and 90% for a variety of polar compounds. Their removal can be attributed not only to biodegradation, but also to adsorption onto solid surfaces [11,12]. As a consequence, significant fractions of PPCPs are discharged with the final effluent of the STP into the aquatic environment. Besides, these substances can also imply an important pollution source for the soil if primary and secondary sludges (to which they are adsorbed) are spread on land. A major factor influencing the efficiency of pollutants removal from water is their ability to interact with solid particles, both natural (clay, sediments, microorganisms) or added to the medium (active carbon, coagulants), because this facilitates their removal by physical– chemical (settling, flotation) or biological processes (biodegradation). However, compounds with low adsorption coefficients tend to remain in the aqueous phase, which favors their mobility through the STP and

2919

the receiving environment [13]. In this way, many PPCPs remain in the aqueous phase, such as the antiinflammatories and the antibiotics, whereas some of them are adsorbed to the sludges, such as the musks and the estrogens [14]. The overall removal rates published in literature vary strongly. In Germany, reported efficiencies range from 10% to 90% depending on the nature of the compound [5]. In Brazil, removal efficiencies for pharmaceutical polar compounds vary from 12% to 90%, where the efficiencies obtained in activated sludge processes were higher than in biofilters [15]. Another study, carried out in the USA concluded that many PPCPs (around 80%) were removed [30]. In all these cases, removal includes both degradation and adsorption and the difference between both mechanisms has not been assessed yet. In the case of polar compounds, such as carboxylic acids, for which the adsorption effects are expected to be very low, the main mechanism of elimination is attributed to biodegradation. However, the studies carried out by Scha¨fer and Waite [16] indicate that less than 10% are effectively biodegraded. Significant differences in the concentrations found can be observed between different geographical areas as mentioned by Heberer [17] for fragrances and their occurrence in the environment. So far, most of the studies focused on PPCPs have been carried out in the USA and central and northern countries of the EU, both areas with moderate climates. On the other hand, data from treatment plants located in Southern Europe are scarce, a lack of information that should be dealt with to have a complete picture of the occurrence and fate of these compounds in the whole EU, as well as to compare the situation in areas with moderate and hot climates. The aim of this study was therefore to investigate the behavior of 13 cosmetic and pharmaceutical compounds belonging to different groups (musks, anti-inflammatories, antiepileptics, tranquillizers, antibiotics, natural and synthetic estrogens and contrast media) along the different units of a municipal STP located in Galicia (NW Spain). The removal efficiency from the water phase of each substance in each particular unit has been determined.

2. Materials and methods 2.1. Sewage treatment plant The sewage treatment plant studied in this work corresponds to a population of approximately 100,000 inhabitants located in Galicia (NW Spain). The plant includes three main sections: pre-treatment, primary treatment and secondary treatment (Fig. 1). After the reception and pumping of the inlet wastewaters, the pretreatment section comprises units for coarse screening

ARTICLE IN PRESS M. Carballa et al. / Water Research 38 (2004) 2918–2926

2920

SPS

SO Pretreatment

SB

SF

SSS

Primary Sedimentation

Biological reactor

Secondary Sedimentation

Effluent

Air

Air

Influent Reception of waste sludges

Solid waste

Treatment of sludges

Recirculation of liquid supernatant

Solid waste

Fig. 1. Diagram of the municipal sewage treatment plant and location of the sampling points.

(bar racks), fine screening and aerated chambers for grit and fat removal. The primary treatment is carried out in circular sedimentation tanks. Finally, the secondary treatment is carried out in biological reactors using the conventional activated sludge process (mixed reactors followed by a sedimentation tank). The supernatant of the secondary sedimentation unit constitutes the final effluent of the plant. The excess of secondary sludges, together with the solids obtained from the primary sedimentation, are treated in a specific unit from which a solid waste and a liquid stream, recycled to the inlet of the plant, are obtained (Fig. 1). The sampling points for analysis are the following (Fig. 1): (i) inlet to the grit removal unit (So); (ii) inlet to the primary sedimentation tank (Sps); (iii) inlet to the biological reactor (Sb); (iv) inlet to the secondary sedimentation tank (Sss) and; (v) outlet of the secondary sedimentation tank (Sf). Three analytical campaigns, during 1 year, were carried out. Taking into account that the operating hydraulic retention time (HRT) in the STP is 24 h, the integrated samples were obtained by mixing the 24 liquid samples collected every hour by an automatic device at each sample point. All compounds were measured during the three integrated campaigns, with the exception of estrogens, antibiotics and contrast media, which were only analyzed for the samples obtained in the last campaign (April 2002). In order to avoid estrogens degradation during the samples transportation to Germany, the pH was adjusted to 2 after collecting the samples.

2.2. Analytical methods Total solids (TS), volatile solids (VS), total suspended solids (TSS), volatile suspended solids (VSS), pH and total and soluble chemical oxygen demand (CODt and CODs, respectively) were determined by Standard Methods [18]. Total organic carbon (TOC) was measured with a Shimazdu model TOC-5000 total organic carbon analyzer, TOC concentrations were calculated by the difference between total carbon (TC) and inorganic carbon (IC). NO2 , NO3 , Cl , PO34 and SO24 were analyzed by capillary electrophoresis (Waters Capilary Ion Analyzer, CIA model). Sodium chromate was used as electrolyte (0.005 mol l 1) as well as an electro-osmotic modifier CIA-Pakt OFM Anion BT (Waters) 0.46 mM [19]. The soluble content of the fragrances, anti-inflammatories, carbamazepine and diazepam was determined after solid-phase extraction (SPE) of 500 ml samples using 60 mg OASIS HLB cartridges (Waters, Milford, MA, USA). Meclofenamic acid and dihydrocarbamazepine were added to the samples as surrogate standards. All compounds were quantitatively eluted from the cartridge using 3 ml of ethyl acetate. This extract was then divided into two fractions: one of them was used for the direct determination of the soluble content of carbamazepine, diazepam and fragrances; the second one was employed for the determination of the antiinflammatory species. In the latter case compounds were silylated previously to their gas chromatographic separation according to a previously published method [20]. In both cases, GC/MS was used to determine the

ARTICLE IN PRESS M. Carballa et al. / Water Research 38 (2004) 2918–2926

concentration of the investigated compounds in the SPE extract. In the cases of galaxolide and tonalide, complementary methodology was used to determine the overall amount present in samples containing solids: the total load. A previously developed method [21,22], based on a solid phase micro-extraction (SPME) technique using PDMS/DVB fiber was used for this purpose. The whole sample, including the soluble fraction and the solid particles, was thermostatized and magnetically stirred during the extraction process. The SPME fiber was exposed to the headspace over the sample. After the sampling time (30 min), the fiber was desorbed into the GC injector and GC-MS analysis was performed. Due to their occurrence as ingredients in all kinds of cleansing products and cosmetics, the risk of sample contamination with musks during analyses is significant, so it is advisable to take extreme precautions to avoid sources of interference in the laboratory environment. To prevent sample contamination, appropriate steps should be taken. Blank samples of the whole process have been analyzed every set of samples to discard potential contamination. In addition, spiked water samples have been analyzed periodically to evaluate the performance of the method. Antibiotics, X-ray contrast media and estrogens were analyzed by the group of Dr. Ternes in Germany. Antibiotics and X-ray contrast media were analyzed by LC electrospray tandem MS after an enrichment step using an SPE method and elution with methanol [23]. Estrogens were analyzed by GC (ion trap) MS/MS after an enrichment step using an SPE method, elution with acetone and derivatization with MSTFA/DTE/TMSI for 1 h at 60 C [24]. Quantification limits and recoveries are given in Table 1. Values given for the different samples of the STP considered in this work correspond to the mean value of two aliquots of each composite sample. The

2921

chemical structures of the selected compounds are shown in Fig. 2. 2.3. Calculations Removal efficiencies from the aqueous phase for all PPCPs were calculated taking into account the measured concentration at the inlet of the plant (So), inlet of the biological reactor (Sb) and the final effluent (Sf). The percentage related to the primary treatment was then calculated as (So Sb)/So  100, the percentage related to the biological reactor was obtained using (Sf Sb)/So and the overall efficiency using (So Sf)/So. In all cases, So was used as the reference in order to be able to compare and to add the partial percentages and obtain the overall one.

3. Results and discussion Table 2 shows the values obtained in the different integrated campaigns for the main characteristics of the wastewaters, such as solids content (total and suspended), chemical oxygen demand, total organic carbon, nitrogen, chloride, sulfate and phosphate are shown. The overall efficiencies achieved for COD and TSS along the entire STP were 80–94% and 92–94%, respectively. 3.1. Occurrence of drugs in wastewaters Among all PPCPs considered in this work, the following have been detected in the wastewaters investigated: galaxolide and tonalide (fragrances), ibuprofen and naproxen (anti-inflammatories), sulfamethoxazole (antibiotic), estrone and 17b-estradiol (natural estrogens) and iopromide (contrast medium). However, diazepam, carbamazepine, diclofenac, roxi-

Table 1 Selected PPCPs, limits of detection (LOD) and quantification (LOQ) in ng l 1, and recovery rates (%) of the analytical methods Name

Application

CAS

Formula

LOD

LOQ

Recovery (%)

Galaxolide Tonalide Diazepam Carbamazepine Diclofenac Ibuprofen Naproxen Roxithromycin Sulfamethoxazole Iopromide Estrone 17b-estradiol 17a-ethinylestradiol

Fragrance Fragrance Tranquillizer Antiepileptic Anti-inflammatory Anti-inflammatory Anti-inflammatory Antibiotic Antibiotic Contrast medium Natural estrogen Natural estrogen Synthetic estrogen

1222-05-5 1506-02-1 439-14-5 298-46-4 15307-86-5 15687-27-1 22204-53-1 80214-83-1 723-46-6 73334-07-3 53-16-7 50-28-2 57-63-6

C18H26O C18H26O C16H13ClN2O C15H12N2O C14H11Cl2NO2 C13H18O2 C14H14O3 C41H76N2O15 C10H11N3O3S C18H24I3N3O8 C18H22O2 C18H24O2 C20H24O2

1.2 1.8 18.9 22.2 16.7 6.7 6.7 6.7 6.7 6.7 0.5 0.5 0.5

4 6 63 74 50 20 20 20 20 20 1 1 1

88 90 99 67 105 90 88 75 75 75 84 80 82

ARTICLE IN PRESS M. Carballa et al. / Water Research 38 (2004) 2918–2926

2922

Musks

Pharmaceuticals O N

O

N

H3C

O

N C NH 2 O

Cl Galaxolide

Tonalide

Diazepam

Carbamazepine

HOOC

H

COOH

Cl

N

COOH O

Cl Ibuprofen

Naproxen

Diclofenac

R1

O H 2N

S

OR 2

HO

H N

N

OH HO O

O

I

O

O

N

O

H3C

O CH 3

H

O

N N H

sulfamethoxazole

Hormones

O

Iopromide

HO

OH

HO

Estrone

OH OH

Roxithromycin

HO

I

O

OH

O

OH

I

O

O O

CH3 OH N

O

HO

17β-estradiol

17α-ethinylestradiol

Fig. 2. Chemical structures of the PPCPs selected.

thromycin and 17a-ethinylestradiol were below the quantification limits (Table 1). Table 3 shows the concentrations of the PPCPs detected during the three sampling campaigns at the different sampling points. 3.2. Concentration in raw wastewaters Apart from the usual variation between samples at the inlet of the STP (point So), it can be seen that all these compounds are present in the range of 0.6–6.6 mg l 1. The two polycyclic musks, galaxolide and tonalide, were detected in the ranges 2.1–3.4 and 0.9–1.7 mg l 1, respectively. These values are lower than those reported by Heberer et al. [25] in surface waters in Berlin, which

had high percentages of treated sewage (maximum concentrations of 10 mg l 1). The acidic compounds, ibuprofen and naproxen, were detected in the ranges 2.6–5.7 and 1.8–4.6 mg l 1, significantly higher than the ones previously reported by Stumpf et al. [26] in a Brazilian STP influent, with concentrations around 0.3 and 0.6 mg l 1, respectively. In the cases of selected antibiotics, sulfamethoxazole was quantified with concentrations of around 0.6 mg l 1 whereas roxithromycin was below the LOQ. According to the results reported by Hirsch et al. [23] these values are in the same range as those reported for German wastewaters. Iopromide was found in the range of 6–7 mg l 1, quite a high value comparing it with other studies [27]. Finally, the natural estrogens detected in

ARTICLE IN PRESS M. Carballa et al. / Water Research 38 (2004) 2918–2926

2923

Table 2 Characterization of the wastewaters along the different units of the STP (in mg l 1) TOC

TN

Cl

SO24

HPO24

51.2 70.2 52.6 79.2 28.1

22.6 34.3 20.5 55.6 13.0

16.5 21.7 19.5 9.1 11.3

42.6 56.3 54.5 59.0 51.6

— 42.4 32.5 43.6 40.8

— 4.0 — 12.6 —

149 136 84 259 14

81.1 72.6 66.8 103.0 41.2

40.8 36.2 37.4 74.3 17.2

12.7 26.0 21.2 19.7 13.9

51.3 85.6 53.0 57.0 50.3

— 92.7 59.8 63.9 44.9

— 1.9 4.0 24.1 —

265 172 145 811 52

53.8 45.5 61.3 234.7 28.5

23.0 17.5 30.0 218.0 11.8

18.1 15.9 15.6 6.2 9.8

45.1 41.9 51.1 50.5 48.4

— 48.2 76.0 76.2 73.3

— 3.9 1.7 5.9 1.1

Month

Sample

TS

VS

TSS

VSS

CODt

CODs

October 2001

So Sps Sb Sss Sf

581 553 368 2573 323

330 308 195 1843 105

258 223 65 2234 20

191 175 55 1787 18

331 299 107 1432 49

137 134 99 436 40

January 2002

So Sps Sb Sss Sf

863 835 500 2510 335

500 418 240 1878 118

298 268 85 2123 23

235 220 78 1718 18

503 497 242 3196 30

April 2002

So Sps Sb Sss Sf

530 515 500 1110 345

305 295 195 695 110

258 243 170 860 15

207 197 88 697 15

— 275 272 2017 56

TC

Table 3 Profiles of galaxolide (GLX), tonalide (TON), ibuprofen (IBU), naproxen (NPX), sulfamethoxazole (SFMT), estrone (E1), estradiol (E2) and iopromide (IOP) along the different units of the selected municipal STP Month

Sample

GLX (mg l 1)

October 2001

So Sps Sb Sss Sf

2.10 4.40 1.40 45.40 0.60

January 2002

So Sps Sb Sss Sf

April 2002

So Sps Sb Sss Sf

TON (mg l 1)

IBU (mg l 1)

NPX (mg l 1)

SFMT (mg l 1)

IOP (mg l 1)

E1 (ng l 1)

E2 (ng l 1)

0.90 1.50 0.60 3.25 0.20

2.75 2.83 2.84 0.20 0.91

3.45 3.75 3.48 1.40 1.85

nm nm nm nm nm

nm nm nm nm nm

nm nm nm nm nm

nm nm nm nm nm

3.40 3.10 1.60 28.70 0.50

1.69 1.63 0.97 14.78 0.15

5.70 5.80 5.80 0.60 2.10

4.60 4.10 4.80 2.10 2.60

nm nm nm nm nm

nm nm nm nm nm

nm nm nm nm nm

nm nm nm nm nm

3.18 2.30 1.82 17.72 0.49

1.53 1.14 0.94 7.82 0.16

2.64 2.81 2.95 0.52 0.97

1.79 1.78 1.59 0.65 0.80

0.58 0.47 0.64 0.25 0.25

6.60 7.50 7.20 8.80 9.30

2.40 2.40 3.40 nm 4.40

nm 3.00 2.40 oLOQ oLOQ

nm: not measured.

these wastewaters were in the range of 2–3 ng l 1 whereas 17a-ethinylestradiol was below the LOQ. These values are low, even in the case of natural estrogens, since previous works have given 15 and 27 ng l 1 for 17b-estradiol and estrone, respectively, in municipal German STPs; or 21 and 40 ng l 1, respectively, in Brazilian STPs [24].

3.3. Behavior of PPCPs along the STP Fig. 3 shows the removal efficiencies calculated during both the primary and the secondary treatments. It can be seen (Fig. 3a) that fragrances are well removed during the primary treatment, with values of around 40% in most cases, as well as 17b-estradiol

ARTICLE IN PRESS M. Carballa et al. / Water Research 38 (2004) 2918–2926

2924

100

100 80

80

60

60

Removal (%)

Removal (%)

40 20 0 -20 -40

40 20

-60

0

-80 -20

-100

(A)

GLX

TON

E1

E2

(B)

IBU

NPX

SFMT

Fig. 3. Removal efficiencies (%) obtained for the PPCPs detected in the STP during primary, biological and overall treatment.

(20%). These efficiencies calculated for both types of PPCPs are closely related to those obtained for suspended solids which indicates that adsorption onto solid particles is the key mechanism involved. In fact, among all the substances studied in this work, these compounds are more likely to have a high partition coefficient between the solid and liquid phase. However, no significant reduction was observed in the pre-treatment and sedimentation steps for ibuprofen and naproxen (Fig. 3b). This is concordant with their acidic structures, with very low solid–liquid partition coefficients, which results in their presence mainly in the aqueous phase. Previous works carried out in a Brazilian STP report removal efficiencies of around 75% and 78% for both compounds, respectively [26]. The same behavior was observed for sulfamethoxazol, iopromide and estrone. Furthermore, their concentration is higher after their passage through the primary sedimentation tank. These results are likely to be due to the analytical deviation caused by the different characteristics of the waters. However, in the particular case of the estrone, a higher concentration is always detected at the end of this step very likely due to the oxidation of the 17b-estradiol present, which explains the high negative removal efficiencies obtained for the estrone and the positive reduction of 17b-estradiol. Furthermore, during the first steps of the treatment, the concentrations of both natural estrogens increase, which can be explained by the cleavage of the glucuronides (Fig. 3a). All the PPCPs detected, except iopromide, are removed during biological treatment with efficiencies between 30% and 75%. The elimination of musks in this case (30–40% for galaxolide, 45–50% for tonalide) once again comprises adsorption and biodegradation. It is important to mention the high concentration peaks for galaxolide and tonalide in the samples containing solids (total load), especially in the outlet of the biological reactor (point Sss, Table 3). When these samples were filtered (to obtain the soluble load), the concentrations in the liquid were extremely low (around 2 and 0.5 mg l 1 for galaxolide and tonalide, respectively), a

clear indication of the important adsorption of these neutral compounds to solids, especially on the sludges from the biological reactor. The overall removal efficiencies from the water phase along the plant were 70–85% for galaxolide and 75–90% for tonalide. In the cases of the anti-inflammatories detected, significant overall removal efficiencies were also achieved (60–70% for ibuprofen, 40–55% for naproxen), although the elimination only took place during the biological treatment. Sulfamethoxazole was quantified in these wastewaters, which is removed, around 67%, during the biological step. Similar average concentrations of this substance (below 1.0 mg l 1) were reported in the effluent of other STP [23]. Concerning the hormones, 17b-estradiol was removed during the biological treatment (47%), resulting in concentrations below the LOQ in both the effluent of this unit and of the overall plant. On the contrary, estrone concentrations increased over the course of treatment, illustrating the fact that under oxidizing conditions, 17b-estradiol is quickly converted into estrone, which is much more slowly degraded [15]. Taking into account the initial concentration of the 17bestradiol (3 ng l 1) and the limit of quantification (LOQ) of 1 ng l 1, it can be assumed that at least 2 ng l 1 were removed, which agrees with the concentration detected for the estrone. As mentioned before for the primary treatment, this hypothesis would explain the negative removal efficiencies obtained for the estrone (Fig. 3a). The results obtained for the investigated contrast medium, iopromide, indicate that there is no significant removal of this compound throughout the plant. In fact, these compounds are designed to be highly stable so they are not readily biodegradable [27], although some batch experiments carried out with activated sludges yielded biotransformation into two metabolites but no mineralization [28]. With the exception of the fragrances, all the other compounds analyzed in this work have been measured in filtered samples only. However, as it is evidenced in

ARTICLE IN PRESS M. Carballa et al. / Water Research 38 (2004) 2918–2926

this work for tonalide and galaxolide, previous studies have already shown that unfiltered samples can give more complete information about the real presence of these substances in the environment [14,29]. It is important to take into account that even the PPCPs with medium or low partition coefficients could be present mainly in the solid phase in streams containing a high suspended solids concentration, such as those obtained from purges or recycling sludges from biological reactors.

4. Conclusions A group of 13 PPCPs corresponding to different kinds of substances (musks, pharmaceuticals and hormones) has been used as an indicator of the presence of this type of pollution in the municipal wastewaters generated by a city of around 100,000 inhabitants in Galicia (NW Spain). Most of the few works found in literature on this issue study the overall removal of PPCPs along STPs (difference between the influent and effluent loads), being the main goal of this work, the assessment of their fate along the different units in order to evaluate the possible improvements for enhancing the removal efficiency. The eight compounds which were detected in raw wastewaters (galaxolide, tonalide, ibuprofen, naproxen, sulfamethoxazole, estrone, 17b-estradiol and iopromide) had a different behavior along the units of STP: *

*

*

During the primary treatment, the lipophilic properties of fragrances and 17b-estradiol facilitate their removal within fat separation. Besides, their good adsorption onto solid surfaces allows an important elimination in the primary settler to be obtained. During the secondary treatment (conventional activated sludges) all compounds detected have been partially removed, with the exception of iopromide, which remained in the water phase. Most of 17bestradiol was partially oxidized in the aeration tank, which explains the increase of estrone concentration in the effluent, apart from the fraction coming from the cleavage of the conjugates. The overall removal efficiencies of the STP ranged between 70% and 90% for the fragrances, 40–65% for the anti-inflammatories, around 60% for the antibiotic sulfamethoxazole and 65% for 17b-estradiol. The term overall removal means disappearance from the water phase. Two mechanisms can be used to explain this elimination: the degradation of the compound and the adsorption onto primary and secondary sludges. This implies that a complete management of the pollution associated with these particular compounds must necessarily take into

2925

consideration the treatment of the excess of sludges, a topic which requires a further study.

According to the conclusions stated before, some modifications or improvements can be implemented in existing STPs. In the primary treatment (coagulation– flocculation and flotation units) the use of some additives as well as the proper adjustment of the operating conditions could be a tool to remove PPCPs from the water phase prior to the biological treatment. During biological treatment, the variation of operational parameters, such as solids retention time (SRT), or the combination of anoxic/aerobic steps could improve the efficiency. Finally, the concentrations of some PPCPs in sludges are expected to be quite high, and therefore an effort should be made to quantify this load, either by developing analytical techniques to measure PPCPs concentrations in sludge samples or by the determination of adsorption coefficients (Kd) to calculate indirectly those concentrations.

Acknowledgements This work was supported by the EU Project Poseidon (EVK1-CT-2000-00047).

References [1] Høverstad T, Carlstedt-Duke B, Lingaas E, Norin E, Saxerholt H, Stein-bakk M, Midtvedt T. Influence of oral intake of seven different antibiotics on faecal short-chain fatty acid excretion in healthy subjects. Scand J Gastroenterol 1986;21:997–1003. [2] Berger K, Petersen B, Buening-Pfaue H. Persistence of drugs occurring in liquid manure in the food chain. Arch Lebensmittelhygiene 1986;37(4):99–102 (in German). [3] Buser HR, Poiger T, Mu¨ller MD. Occurrence and environmental behaviour of the chiral pharmaceutical drug ibuprofen in surface waters and in wastewater. Environ Sci Technol 1999;33:2529–35. [4] Huppert N, Wu¨rtele M, Hahn HH. Determination of the plasticizer N butylbenzenesulfonamide, the pharmaceutical Ibuprofen in wastewater using solid phase microextraction (SPME). Fres J Anal Chem 1998;362:529–36. [5] Ternes TA. Occurrence of drugs in German sewage treatment plants and rivers. Water Res 1998;32(11): 3245–60. [6] Buser HR, Mu¨ller MD. Occurrence of the pharmaceutical drug clofibric acid and the herbicide mecoprop in various Swiss Lakes and in the North Sea. Environ Sci Technol 1998;32:188–97. [7] Halling-Sørensen B, Nielsen SN, Lanzky PF, Ingerslev F, Lu¨tzhoft HCH, Jorgensen SE. Occurrence, fate and effects of pharmaceutical substances in the environment - a review. Chemosphere 1998;36:357–93.

ARTICLE IN PRESS 2926

M. Carballa et al. / Water Research 38 (2004) 2918–2926

[8] Tyler CR, Routledge EJ. Natural and anthropogenic environmental estrogens: the scientific basis for risk assessment. Estrogenic effects in fish in English rivers with evidence of their causation. Pure Appl Chem 1998;70: 1795–804. [9] Johnson AC, Belfroid A, Di Corcia A. Estimating steroid estrogens inputs into activated sludge treatment works and observations on their removal from the effluent. Sci Total Environ 2000;256:163–73. [10] Johnson AC, Sumpter JP. Removal of endocrine disrupting chemicals in activated sludge treatment works. Environ Sci Technol 2002;35:4697–703. [11] Stumpf M, Ternes TA, Haberer K, Seel P, Baumann W. Nachweis von Arzneimittelru¨cksta¨nden in Kla¨ranlagen und Fliebgewa¨ssern [Determination of drugs in sewage treatment plants and river water]. Vom Wasser 1996;86: 291–303. [12] Hirsch R, Ternes TA, Haberer K, Kratz KL. Determination of betablockers and b-sympathomimetics in the aquatic environment. Vom Wasser 1996;87:263–74. [13] Ohlenbusch G, Kumke MU, Frimmel FH. Sorption of phenols to dissolved organic matter investigated by solid phase microextraction. Sci Total Environ 2000;253: 63–74. [14] Jensen RL, Scha¨fer AI. Adsorption of estrone and 17b bestradiol by particulates—activated sludge, bentonite, hematite and cellulose. Recent Advances in Water Recycling Technologies Workshop, Brisbane, 26 November 2001, 93–102, ISBN 0 7334 1858 9. [15] Ternes TA, Kreckel P, Mueller J. Behaviour and occurrence of estrogens in municipal sewage treatment plants—II. Aerobic batch experiments with activated sludge. Sci Total Environ 1999;225:91–9. [16] Scha¨fer AI, Waite TD, Removal of endocrine disrupters in advanced treatment. IWA World Water Congress, Melbourne, 7–12 April 2002. [17] Heberer T. Occurrence, fate, and assessment of polycyclic musk residues in the aquatic environment of urban areas— a review. Acta Hydrochim Hydrobiol 2002;30(5–6): 227–43. [18] APHA-AWWA-WEF. In: Clesceri LS, Greenberg AE, Eaton AD, editors. Standard methods for the examination of water and wastewater. 20th ed. Washington, DC: American Public Health Association, American Water Work Association, Water Environment Federation; 1999. [19] Ewing AG, Wallingford RA, Olefirowicz TM. Capillary electrophoresis. Anal Chem 1989;61:292A–303A.

[20] Rodriguez I, Quintana JB, Carpinteiro J, Carro AM, Lorenzo RA, Cela R. Determination of acidic drugs in sewage water by gas chromatography-mass spectrometry as tert.-butyldimethylsilyl derivates. J Chromatogr A 2003; 985:265–74. [21] Garc!ıa-Jares C, Llompart M, Polo M, Salgado C, Mac!ıas S, Cela R. Optimisation of a solid-phase microextraction method for synthetic musk compounds in water. J Chromatogr A 2002;963:277–85. [22] Llompart M, Garc!ıa-Jares C, Salgado C, Polo M, Cela R. Determination of musk compounds in sewage treatment plant sludge samples by solid-phase microextraction. J Chromatogr A 2003;999:185–93. [23] Hirsch R, Ternes TA, Haberer K, Ludwig KK. Occurrence of antibiotics in the aquatic environment. Sci Total Environ 1999;225:109–18. [24] Ternes TA, Stumpf M, Mueller J, Haberer K, Wilken RD, Servos M. Behaviour and occurrence of estrogens in municipal sewage treatment plants—I. Investigations in Germany, Canada and Brazil. Sci Total Environ 1999; 225:81–90. [25] Heberer T, Gramer S, Stan H-J. Occurrence, distribution of organic contaminants in the aquatic system in Berlin Part. III: Determination of synthetic musks in Berlin surface water applying solid-phase microextraction (SPME), gas chromatography-mass spectometry (GC/ MS). Acta Hydrochim Hydrobiol 1999;27:150–6. [26] Stumpf M, Ternes TA, Wilken R, Rodrigues SV, Baumann W. Polar drugs residues in sewage and natural waters in the state of Rio de Janeiro, Brazil. Sci Total Environ 1999;225:135–41. [27] Ternes TA, Hirsch R. Occurrence and Behaviour of X-ray contrast media in sewage facilities and the aquatic environment. Environ Sci Technol 2000;34(13):2741–8. [28] Kalsch W. Biodegradation of the iodinated X-ray contrast media diatrizoate and iopromide. Sci Total Environ 1999;225:143–53. [29] Kolpin DW, Furlong ET, Meyer MT, Thurman EM, Zaugg SD, Barber LB, Buxton HT. Pharmaceuticals, hormones, and other organic wastewater contaminants in US streams, 1999–2000: a national reconnaissance. Environ Sci Technol 2002;36(6):1202–11. [30] US Environmental Protection Agency, 2003. Pharmaceuticals and Personal Care Products (PPCPs) as Environmental Pollutants: Pollution from Personal Actions, Activities and Behaviors. Available in Internet in: o http://www.epa.gov/nerlesd1/chemistry/pharma>.