Seasonal occurrence and removal of pharmaceutical products in municipal wastewaters

1

2

3

4 Q1

567

89

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

Journal of Environmental Chemical Engineering xxx (2014) xxx–xxx

Q2

G Model

JECE 272 1–8

Seasonal occurrence and removal of pharmaceutical products inmunicipal wastewaters

Manuel Fernandez a, Monica Fernandez b, Amanda Laca c, Adriana Laca a, Mario Dıaz a,*a Department of Chemical Engineering and Environmental Technology, University of Oviedo, C/Julian Claverıa s/n. 33071 Oviedo, Spainb Anes Innovacion S.L. Avenida de Galicia, 31. 33005 Oviedo, Spainc Scientist-Technical Services, University of Oviedo, C/Fernando Bonguera s/n. 33071 Oviedo, Spain

A R T I C L E I N F O

Article history:

Received 11 November 2013

Accepted 30 January 2014

Keywords:

Seasonal

Wastewater

PPCPs

Treatment

WWTP

A B S T R A C T

The occurrence in municipal wastewaters of six pharmaceutical products, paracetamol, ibuprofen,

naproxen, diclofenac, caffeine and carbamazepine, which belong to different therapeutic classes (analgesic

drugs, anti-inflammatory, antiepileptic and stimulant compounds), have been investigated. Influent and

effluent water samples from two conventional wastewater treatment plants (WWTPs) of the North of Spain

were collected at different seasons and analyzed. Ranges of PPCPs concentrations were similar to levels

reported in other studies worldwide. Influent concentrations ranges were 2.3–42 mg/L for ibuprofen and

naproxen, 0.04–7.8 mg/L for caffeine and paracetamol, and 0.03–0.4 mg/L for carbamazepine and

diclofenac. The highest concentrations were found for ibuprofen in the untreated municipal wastewaters.

Effluent concentrations were always below 5.7 mg/L. Diclofenac and carbamazepine persisted in WWTP

effluents, whereas paracetamol, ibuprofen, naproxen and caffeine showed removal efficiencies between

75% and 99%. Considering first-order kinetics for the biodegradation of these compounds, apparent kinetic

constants were calculated and similar values were obtained for both WWTPs, although one of them

resulted to be more sensitive to temperature changes.

� 2014 Published by Elsevier Ltd.

Contents lists available at ScienceDirect

Journal of Environmental Chemical Engineering

jou r n al h o mep ag e: w ww .e lsev ier . co m / loc ate / jec e

26272829303132333435363738394041

42

Introduction

Pharmaceutical and personal care products (PPCPs) and theirmetabolites are contaminants extensively found in the aquaticenvironment [1]. These emerging environmental pollutants deservespecial attention due to the fact that some of them may causeecological and health harm [2–4]. Increasing numbers of watersamples obtained from lakes, streams, aquifers and municipalsupplies across the world have been found to be contaminated bytrace quantities of such residues [5]. These compounds might beexcreted by patients or be improperly disposed by users and end upin municipal wastewaters. One of the major sources of PPCPs in theaquatic environment is the effluent discharge from wastewatertreatment plants (WWTPs) [1,6]. Current municipal wastewatertreatment processes are insufficient at degrading many PPCPs andremoval rates vary depending on the treatment technology used andthe compound considered. Hence, variable amounts of PPCPs arecontinuously released into surface, ground and coastal waters [7].

43

4445

* Corresponding author. Tel.: +34 985103439; fax: +34 985103434.

E-mail addresses: [email protected], [email protected] (M. Dıaz).

Please cite this article in press as: M. Fernandez, et al., Seasonal ocwastewaters, J. Environ. Chem. Eng. (2014), http://dx.doi.org/10.101

http://dx.doi.org/10.1016/j.jece.2014.01.023

2213-3437/� 2014 Published by Elsevier Ltd.

The concentration of pollutants in influents and effluents ofWWTPs are routinely monitored in many countries [5]. Despite ofthe fact that little attention has been paid to seasonal variation ofPPCPs, results of different studies showed that the concentrationsof PPCPs in municipal wastewater and their treated effluents mayvary along the year [8–11]. Furthermore, diurnal variation pattersin specific PPCPs that correlates with daily drug administrationhave also been identified in some cases [12].

The goal of this work was to assess the occurrence and removalof selected pharmaceutical products from municipal wastewatersin the North of Spain. As far as we know, this is the first study of thiskind carried out in this region. With this aim, samples from twoWWTPs were collected and analyzed along the four seasons in oneyear. Additionally, local hospital effluents were also analyzed.Moreover the seasonal variability in PPCPs occurrence and removalwas also investigated.

Materials and methods

Selected PPCPs

The PPCPs considered in this study include: paracetamol,ibuprofen, naproxen, diclofenac, caffeine and carbamazepine.

currence and removal of pharmaceutical products in municipal6/j.jece.2014.01.023

46 Th47 ph48 sy49 an50 de51 di52 ph53 to

54 Sa

55

56 tw57 th58 ha59 pu60 re61 au62 de63 co64 tr65 w66 st

67 De

68

69 re70 ac71 co72 se73 pl74 zo75 w76 af77

78 re79 an80 bi81 w82 th83 re84 cl85

86 2

87 th88 W89 m90 w

919293949596979899100101102103104105106107

108109110111112113114

115116117118

119

120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150

Table 1Sampling details.

Date Week day Sampling time Average day

temperature

(8C)

Average day

precipitations

(mm)

20/12/2010 Monday �8:30 (WWTP1)

�10:00 (WWTP2)

�11:00 (Hospital)

�9 0.2

22/02/2011 Tuesday �8:30 (WWTP1)

�10:00 (WWTP2)

�11:00 (Hospital)

�12 3.6

19/05/2011 Thursday �8:30 (WWTP1)

�10:00 (WWTP2)

�11:00 (Hospital)

�15 0.6

25/07/2011 Monday �8:30 (WWTP1)

�10:00 (WWTP2)

�11:00 (Hospital)

�19 0.4

M. Fernandez et al. / Journal of Environmental Chemical Engineering xxx (2014) xxx–xxx2

G Model

JECE 272 1–8

ese compounds were chosen to represent different groups ofarmaceutical products widely reported to occur in aquaticstems, specifically: analgesic, anti-inflammatory, antiepilepticd stimulant drugs. Caffeine is among the 30 most frequentlytected organic wastewater pollutants and carbamazepine,clofenac and ibuprofen are among the top 10 high priorityarmaceuticals identified in a European assessment of PPCPs due

their high consumption [4,7].

mple collection

Wastewater was sampled from influent and effluent flows ofo local water utilities (WWTP1 and WWTP2) and effluents from

e University Central Hospital of Asturias (HUCA). This hospitals 1324 beds and the wastewater is directly discharged into theblic sewage system. All facilities were located in Asturias, a

gion sited in the North of Spain. Grab samples were collected intumn, winter, spring and summer (see Table 1) using a samplevice consisting in a plastic bottle attached to a stick. Afterllection, samples were transferred to 2.5 L glass bottles and

ansported to laboratory. The same day of collection, samplesere adjusted to pH 2.00 � 0.10 using hydrochloric acid 3.5 M andored at 4 8C in the dark until extraction (maximum 12 h).

scription of treatment plants

The treatment in WWTP1 consists of screening, grit and greasemoval, primary clarification, activated sludge treatment tohieve removal of biochemical oxygen demanding organicmpounds (BOD), nitrogen and phosphorus and, finally, acondary clarification (Fig. 1). The biological degradation takesace in a ‘‘channel type’’ bioreactor with anaerobic/anoxic/aerobicnes and an average retention time of 8 h. The influent samplesere taken after screening and the effluent samples were takenter secondary clarification.

The treatment in WWTP2 consists of screening, grit and greasemoval, activated sludge treatment to achieve removal of BODd nitrogen and, finally, secondary clarification (Fig. 1). Theological degradation takes place in a ‘‘carrousel type’’ bioreactorith anoxic/aerobic zones and an average retention time of 10 h. Inis case, the influent samples were taken after sand and greasemoval and the effluent samples were taken after secondaryarification.

Both facilities receive a day contaminant charge between 1 andkgCOD/m3d, being the BOD5/COD relationship upon 0.4–0.9, soese are middle or easily biodegradable wastewaters. However,WTP1 receives a 25% of industrial wastewater and 75%unicipal wastewater, whereas WWTP2 receives only municipalastewater that includes several hospital effluents (around 3% of

Please cite this article in press as: M. Fernandez, et al., Seasonal owastewaters, J. Environ. Chem. Eng. (2014), http://dx.doi.org/10.10

the total wastewater that arrives to WWTP2 comes fromhospitals). WWTP1 and WWTP2 serve a population equivalentof 260,000 and 20,000 respectively.

Removal of micropollutants within activated sludge systemscan be associated to three main mechanisms: volatilization to air,sorption to the sludge and biological conversion. Models referringto pharmaceutical compounds usually did not include volatiliza-tion because it is not considered a significant removal mechanismfor this family. Additionally, sorption mechanism is complex andstill remains not sufficiently documented [13]. In this work, onlydegradation in the biological reactors was considered to determineapparent kinetic constants. Large municipal wastewater treatmentplants could be represented as plug flow or ideally mixed tank inseries [14]. Equations for biodegradation modeling usuallyconsider the degradation of dissolved micropollutant concentra-tion following a first-order kinetic [13]. Plug flow and first-ordertransformation kinetics were assumed in this case:

lnCi

Ce¼ ku (1)

where Ci is the pollutant influent concentration, Ce the pollutanteffluent concentration, k is the apparent kinetic constant forcontaminant removal and u is the hydraulic retention time of thebiological reactor.

In order to calculate apparent activation energy, an Arrheniustype equation was employed:

k ¼ k0e�Ea=RT (2)

where k0 is the pre-exponential factor, Ea is the apparent activationenergy, T is the absolute temperature and R is the universal gasconstant.

Analytical methodology

As first step, wastewater samples underwent vacuum filtrationtwice (20–25 mm Whatman filter paper and 0.45 mm AlbetLabscience nitrocellulose filter). Solid-phase extraction (SPE)method was employed to concentrate the analytes from theaqueous samples; MCX 3cc/60 mg, 60 mm (Waters Oasis) car-tridges were used and 0.5 L of influent samples and 1 L of effluentsamples were loaded. Recovery values for MCX extraction arereported in Table 2. The volumes of sample to be filtered wereselected considering previous works [15,16] and the cartridgemanufacturer’s instructions. After SPE, cartridges were dried for1 h, the analytes were eluted (3 mL of ethyl acetate, 3 mL of 50/50ethyl acetate/acetone and 3 mL of 48/48/2 ethyl acetate/acetone/ammonium hydroxide) and extracts were evaporated to drynessunder a nitrogen stream. Ethyl acetate (�99.8%, Sigma–Aldrich)(1.5 mL) was used for reconstitution and the reconstituted sampleswere filtered (0.20 mm Whatman nylon filter) [15]. All compounds,except caffeine, were analyzed after a derivatization step with N-Methyl-N-(trimethylsilyl)trifluoroacetamide (MSTFA) (Sigma–Aldrich). For this purpose, 100 mL of MSTFA were added to 100 mLof the reconstituted sample and this mixture was kept for 35 minin an oven at 65 8C. Amber autosampler vials were employed.

Finally, samples were injected onto a GC/MS (Agilent Technol-ogies; 6890 N Network GC System, 5975 inert Mass SelectiveDetector, 7683B Series injector, 7683 series Autosampler) fittedwith a column HP-5MS (30 m � 0.25 mm id � 0.25 mm, 19091S-433, Agilent Technologies). The carrier gas was ultrapure helium ata constant flow of 1.3 mL/min. The oven temperature was held at50 8C for 30 s, and then programmed at 10 8C/min to 250 8C withthe final temperature being held for 5 min. A sample volume of1 mL was injected in the splitless mode. The transfer line and ionsource were set at 280 8C and 230 8C, respectively. Each compound

ccurrence and removal of pharmaceutical products in municipal16/j.jece.2014.01.023

Fig. 1. Schematic representation of studied WWTPs indicating points of sampling.

Table 2LOD, LOQ, RSD and recovery values for each compound.

PPCP WWTPs influents and hospital effluent WWTPs effluents MCX extraction recovery valuesb

LOD (ng/L)a LOQ (ng/L)a RSD (%) LOD (ng/L)a LOQ (ng/L)a RSD (%)

Caffeine 2.6 8.7 <16 1.3 4.4 <8 68

Paracetamol 2.3 7.6 <25 1.1 3.8 <18 76

Naproxen 2.0 6.8 <25 1.0 3.4 <25 90

Carbamazepine 1.7 5.5 <21 0.8 2.8 <25 120

Ibuprofen 3.5 11.6 <10 1.7 5.8 <12 80

Diclofenac 3.4 11.2 <25 1.6 5.6 <22 76

a Values obtained in this work.b Values reported by Togola and Budzinski [15].

M. Fernandez et al. / Journal of Environmental Chemical Engineering xxx (2014) xxx–xxx 3

G Model

JECE 272 1–8

Please cite this article in press as: M. Fernandez, et al., Seasonal occurrence and removal of pharmaceutical products in municipalwastewaters, J. Environ. Chem. Eng. (2014), http://dx.doi.org/10.1016/j.jece.2014.01.023

151 w152 id153 an154

155 Al156 an157 co158 of

159 Re

160 PP

161

162 hi163 th164 an165 pa166 ov167 al168 pr169 as170

171 av172 ca173 W174 to175 fo176 w177 va178 on179 re180 co181 co182 co

183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217

Table 3Levels of PPCPs in the WWTPs 1 and 2 (average concentrations � SD and minimal and maximum values) compared with values reported by other authors. Average concentrations

were obtained by employing all the concentrations measured during the year.

PPCP Average concentrations (mg/L) � SD Concentration range (minimum–maximum) (mg/L)

WWTP1 WWTP2 This work Literature

Influent Effluent Influent Effluent Influent Effluent Influent References Effluent References

Caffeine 2.3 � 1.7 0.02 � 0.01 5.4 � 2.6 0.05 � 0.05 0.8–7.8 n.d.-0.1 0.2–63 [17,21,29] 0.002–9.9 [15,17,21,29]

Paracetamol 0.35 � 0.60 0.01 � 0.00 2.3 � 1.2 0.03 � 0.04 0.04–4.0 0.004–0.1 0.07–26 [26,30] n.d.-5.9 [26,30]

Naproxen 3.5 � 1.1 0.71 � 0.33 11.6 � 1.9 0.57 � 0.68 2.3–14 0.1–1.6 1.1–53 [3,17,31] 0.2–6.2 [3,15,17,31]

Carbamazepine 0.10 � 0.10 0.16 � 0.13 0.13 � 0.07 0.19 � 0.10 0.03–0.3 0.08–0.4 n.d.-3.7 [3,5,17,21,29] n.d.-6.2 [3,5,17,21,29]

Ibuprofen 11.2 � 4.0 1.6 � 2.8 30.5 � 10.0 0.91 � 1.04 6.5–42 0.03–5.7 8.8–168 [31,33] n.d.-3.8 [15,31–33]

Diclofenac 0.16 � 0.06 0.55 � 0.41 0.30 � 0.11 0.39 � 0.18 0.08–0.4 0.2–1.1 n.d.-5.0 [3,17,31,33] n.d.-2.5 [3,15,17,31–33]

n.d.: not detected.

TaLe

an

co

the

P

C

P

N

C

I

D

n.d

M. Fernandez et al. / Journal of Environmental Chemical Engineering xxx (2014) xxx–xxx4

G Model

JECE 272 1–8

as firstly characterized individually in scan mode in order toentify the main ions (m/z ratio) constituting the mass spectrumd to choose the ions for quantification.PPCP GC grade reference standards were supplied by Sigma–

drich. The recovery values, the RSD (Relative Standard Deviation)d the limits of detection (LOD) and quantification (LOQ) werempound dependent (see Table 2). The correlation coefficient (r2)

the calibration curves was always higher than 0.990.

sults and discussion

CP occurrence

As can be seen in Tables 3 and 4, the compounds with theghest concentrations in the influents in both WWTPs and also ine hospital wastewaters were ibuprofen, paracetamol, naproxend caffeine. This is not surprising given that ibuprofen,racetamol and naproxen are painkillers found in numerouser-the-counter medications commonly used in hospitals but

so in households. Additionally to its use in pharmaceuticaloducts, caffeine is frequently employed in beverages and foods

a stimulant.It should be also pointed out that, as it is shown in Table 3,

erage parameter values are much higher (excepting forrbamazepine, in two-fold concentration or even more) inWTP2 than in WWTP1. When the amount of PPCPs that arrived

the WWTPs per capita was calculated, similar values were foundr both facilities in the case of carbamazepine and diclofenac,hereas four-fold values were found for paracetamol and two-foldlues were found for the rest of the compounds. WWTP2 receivesly municipal and hospital wastewaters, whereas WWTP1ceives municipal wastewaters but also wastewaters frommmercial and industrial parks. Obviously, the wastewatersming from commercial and industrial facilities would notntribute much to the pharmaceutical loads. Surprisingly, and

218219220221222223224225

226

227228229230231

ble 4vels of PPCPs in hospital wastewaters (average concentrations � SD and minimal

d maximum values) compared with values reported by other authors. Average

ncentrations were obtained by employing all the concentrations measured during

year.

PCP Concentrations

(mg/L) � SD

Concentration range

(minimum–maximum) (mg/L)

This work Literature References

affeine 24.0 � 22.7 8.5–50.0 12.3–42.0 [22]

aracetamol 0.92 � 1.01 0.1–2.1 3.1–21.2 [22]

aproxen 1.3 � 0.4 0.9–1.6 n.d.–21.8 [22,34]

arbamazepine 0.06 � 0.11 n.d.–0.19 n.d.–2.0 [15,22,34]

buprofen 13.1 � 8.6 4.5–21.7 1.5–151 [15,22,34]

iclofenac 0.12 � 0.07 0.04–0.2 n.d.–6.9 [15,22,34]

.: not detected.

Please cite this article in press as: M. Fernandez, et al., Seasonal owastewaters, J. Environ. Chem. Eng. (2014), http://dx.doi.org/10.10

excepting for caffeine, in general, hospital values were on the sameorder of magnitude as those obtained for both WWTPs (Table 4).Therefore, in this case hospital effluents could not be considered asresponsible for increasing the concentration of these pharmaceu-ticals in the untreated wastewaters.

The ranges of PPCPs concentrations detected in the influents ofboth WWTPs were comparable to concentrations reported in otherstudies worldwide (see Table 3). Particularly, caffeine, carbamaze-pine, and naproxen concentrations were similar to average valuesobserved by Santos et al. [17] (4.9–7.4 mg/L, 0.4–0.5 mg/L and 4.3–8.1 mg/L respectively) in different WWTPs located in the south ofSpain, whereas ibuprofen and diclofenac influent concentrationswere similar to those reported by Hijosa-Valsero et al. [18] (8.4–24.2 mg/L and 0.4–0.8 mg/L respectively) in WWTPs located in thecenter of Spain. Regarding WWTPs effluents, caffeine, carbamaze-pine, ibuprofen and naproxen, concentration levels were in generallower than those reported for several WWTPs located in theSpanish south (0.2–2.4 mg/L, 0.2–0.7 mg/L, 0.06–10.2 mg/L and0.8–4.4 mg/L respectively) [17,19], whereas most of diclofenacconcentrations were slightly higher than average values reportedby these authors (n.d.–0.24 mg/L).

Table 4 compares the PPCPs levels observed here in the hospitalwastewater with those found in the literature. With the exceptionof caffeine, our results were slightly lower. However, ibuprofen andcarbamazepine mean value concentrations were comparable tothose reported by Gomez et al. [20] (19.8 mg/L and 0.04 mg/Lrespectively) in wastewaters from a private healthcare centerlocated in the southeast of Spain.

PPCPs concentrations in municipal effluents depend on differ-ent factors, such as influent composition, removal efficacies,WWTP performance, season or localization. In general, theoutcoming concentrations measured here were within the rangefound in literature for WWTPs worldwide (see Table 3). As it can beseen, maximum values were lower than the upper values reportedby other authors, excepting for ibuprofen. Minimum values wereslightly higher, with the exception of naproxen. Considering onlystudies carried out in Spanish WWTPs, concentrations rangesfound in this work were included in the range described by Santoset al. [17] and Hijosa-Valsero et al. [18] for caffeine (0.17–12.8 mg/L), naproxen (0.05–5.1 mg/L), carbamazepine (<0.02–1.29 mg/L)and ibuprofen (0.002–55.0 mg/L), whereas diclofenac maximumvalue was higher than those reported by these authors (0.14–0.66 mg/L).

Removal efficiencies

Percentages of PPCPs removal in the aqueous phase duringwastewater treatment were calculated for both facilities ([influ-ent] � [effluent]/[influent] � 100). Removal efficiencies were cal-culated for the four seasons and values shown in Fig. 2 wereevaluated on the basis of the average of these four values.

ccurrence and removal of pharmaceutical products in municipal16/j.jece.2014.01.023

232

233

234

235

236237238239240241242243244245246247248249250251252253254255256257

Fig. 2. Average removal efficiency for paracetamol, ibuprofen, naproxen and

caffeine in both WWTPs. Bars indicate SD.

M. Fernandez et al. / Journal of Environmental Chemical Engineering xxx (2014) xxx–xxx 5

G Model

JECE 272 1–8

Therefore, it was considered the removal of PPCPs by degradation,but also by absorption phenomena.

Diclofenac and carbamazepine persisted in WWTP effluentswith similar or even slightly higher concentrations than those

0

10

20

30

40

50

60

Autumn Winter Spring Summer

Con

cent

ratio

n (µ

g/L

)

Caffeine

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

4,0

4,5

Autumn Winter Spring Summer

Con

cent

ratio

n (µ

g/L

)

Paracetamol

0

2

4

6

8

10

12

14

16

Autumn Winter Spring Summer

Con

cent

ratio

n (µ

g/L

)

Naproxen

Fig. 3. Seasonal influent concentrations of studied PPCPs in both WWTPs and hospital wa

carbamazepine was not detected in winter and spring hospital samples.

Please cite this article in press as: M. Fernandez, et al., Seasonal ocwastewaters, J. Environ. Chem. Eng. (2014), http://dx.doi.org/10.101

measured in the untreated sewages (see Table 3). This phenome-non was also described by other authors [21–23]. The persistenceof these compounds in the water would indicate that conventionalwastewater treatment process is not effective in breaking downthese compounds. The higher concentration levels found in theeffluent could be explained by the formation of products of humanmetabolism and/or transformation products (e.g. hydroxyl- andepoxy-derivatives or glucuronides) which may act as a reservoirfrom which a later yield of the parent substance can occur [10,22–24]. Additionally, it is necessary to take into account that themeasured concentrations for these compounds are near to thedetection limit, thus relative SDs are quite high.

The removal values calculated for diclofenac and carbamaze-pine were very low or even negative, so they were omitted in Fig. 2,which shows the treatment removal efficiency for caffeine,paracetamol, ibuprofen and naproxen (calculated with averagevalues). The average removal efficiencies for these compoundswere higher than 95% in WWTP2. These values were similar tothose reported for WWTPs sited in Europe and USA that includedprimary and biological treatment [17,25,26].

Concerning WWTP1, the average removal efficiencies werearound 80% for paracetamol, naproxen and ibuprofen and 99% for

0

5

10

15

20

25

30

35

40

45

Autumn Winter Spring Summer

Con

cent

ratio

n (µ

g/L

)

Ibuprofen

0

0,1

0,2

0,3

0,4

0,5

Autumn Winter Spring Summer

Con

cent

ratio

n (µ

g/L

)

Diclofenac

0,0

0,1

0,2

0,3

Autumn Winter Spring Summer

Con

cent

ratio

n (µ

g/L

)

Carbamazepine

stewater (~ WWTP1; & WWTP2; � Hospital). Hospital autumn sample is missing,

currence and removal of pharmaceutical products in municipal6/j.jece.2014.01.023

258 ca259 ib260 by261 W262 tr263

264 av265 an266 re267 hy268 W269 pr270 in271 ra272 pr273 an274 re

275 Se

276

277 co278 (F279 w280 w281 ob282 an283 Th

284285286287288289290291292293294295296297298299300301302303304305306307308309310311

FigW

M. Fernandez et al. / Journal of Environmental Chemical Engineering xxx (2014) xxx–xxx6

G Model

JECE 272 1–8

ffeine. Despite of being less efficient than WWTP2, naproxen anduprofen removal values were still slightly higher than those cited

Blair et al. [27] (73% and 88% respectively) for North AmericanWTPs that included preliminary, primary and biological

eatment and chlorine disinfection.Regarding removal efficiencies for BOD5 and nitrogen, again the

erage values are slightly higher in WWTP2 than in WWTP1 (99%d 80%, 98% and 68%, respectively). The higher efficiencymovals observed in WWTP2 could be explained because thedraulic retention time of the biologic reactor was higher inWTP2 than in WWTP1 (10 and 8 h, respectively). Besides, theesence of pollutants coming from industrial wastewaters mayterfere in the elimination process, decreasing biodegradationtes in WWTP1. Finally, it is important to point out that theimary clarification previous to biological treatment and theaerobic step in WWTP1 did not contribute to improve the PPCPsmoval efficiencies.

asonality

Samples taken in autumn, winter, spring and summer werempared to examine seasonal trends in PPCPs concentrationig. 3). Excepting for the increase detected in spring in hospitalastewater, the seasonal variation of caffeine in WWTP influentsas minimal. Regarding carbamazepine, the highest value wasserved in winter in WWTP1 influent and in summer in WWTP2d hospital wastewater (0.25, 022 and 0.19 mg/L, respectively).e increase of the concentration observed in WWTP2 coincided

312313

0

100

200

300

400

500

600

SummerSpringWinterAutumn

Con

cent

ratio

n (m

g/L

)

BOD 5

0

10

20

30

40

50

60

SummerSpringWinterAutumn

Con

cent

ratio

n (m

g/L

)

Total N

. 4. Seasonal influent concentrations of water parameters in both WWTPs (~WTP1; & WWTP2) (data from the ‘‘Consorcio de Aguas de Asturias’’).

Please cite this article in press as: M. Fernandez, et al., Seasonal owastewaters, J. Environ. Chem. Eng. (2014), http://dx.doi.org/10.10

with an increase in hospital wastewater which is understandableas this hospital wastewater is treated in WWTP2. Paracetamolincreased notably in all effluents in summer, while ibuprofen had amaximum in winter in WWTP2 and hospital wastewaters.Variations in naproxen were minimal for hospital wastewaters,whereas in WWTP1, the concentrations measured in spring andsummer were double than those obtained in autumn and winter.On the opposite, the highest values for WWTP2 were found inautumn and winter. Minimum values of diclofenac were found inautumn and summer for both WWTPs and hospital wastewaters,observing a maximum in spring for WWTP2. Considering all thesedata, tendencies in the PPCPs concentrations of the untreatedwastewaters could not be established, depending the followedpattern on the compound and specific water. It is necessary to takein mind that several factors are involved to determine thecomposition of these wastewaters, mainly the amount of PPCPsreleased and the abundance of rainfall that dilute the contamina-tion (there is no separation between rainwater and sewage).

Seasonal influent concentrations of BOD5 and total nitrogenwere higher in WWTP2 than in WWTP1, with the exception ofsummer total nitrogen data (see Fig. 4). This can be due to the factthat, as it was mentioned above, the service area of WWTP2 ismunicipal, while WWTP1 serves a mix of municipal and industrialwastewaters that are likely to have lower concentrations of thesecontaminants.

Fig. 5 shows the seasonal variation of removal efficiencies forPPCPs (caffeine, paracetamol, carbamazepine and ibuprofen), BOD5

and total nitrogen. PPCPs removal efficiencies in WWTP2 werehigher than 95% with the exception of ibuprofen and naproxenvalues in autumn, which were lower that 90%. Values higher than

50556065707580859095

100

5 8 11 14 17 20 23

% R

emov

al E

ffic

ienc

y

WWTP1

T (ºC)Autum Wi nter Sp ring Summ er

70

75

80

85

90

95

100

5 8 11 14 17 20 23

% R

emov

al E

ffic

ienc

y

WWTP2

T (ºC)Autum Wi nter Sp ring Summ er

Fig. 5. Seasonal variation of removal efficiency in both WWTPs for: ^ Caffeine, &Paracetamol, ~ Naproxen, � Ibuprofen, * BOD5 and + Total N. Average day

temperatures indicated by bars.

ccurrence and removal of pharmaceutical products in municipal16/j.jece.2014.01.023

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356357358359360

361

362363364365366367368369370371372373374375376377378

379

380381382383

384

385386387388389390391392393394395396397398399400401402403404405406407408

Table 5Seasonal values of apparent kinetic constant (k) (h�1) for different contaminant removal in both WWTPs.

Pollutants WWTP1 WWTP2

Autumn Winter Spring Summer Autumn Winter Spring Summer

BOD5 0.41 0.43 0.51 0.53 0.44 0.46 0.50 0.45

Total N 0.10 0.13 0.18 0.19 0.13 0.18 0.20 0.14

Caffeine 0.46 0.58 0.65 0.69 0.52 0.43 – 0.46

Paracetamol 0.20 0.22 0.24 0.71 0.31 0.41 0.52 0.64

Naproxen 0.19 0.22 0.23 0.12 0.21 0.38 0.46 0.35

Ibuprofen 0.38 0.06 0.54 0.63 0.22 0.34 0.71 0.58

Table 6Values of apparent activation energy (kJ/mol) for different contaminants in both

WWTPs.

Pollutant WWTP1 WWTP2

BOD5 18.4 15.6

Total N 46.4 52.5

Caffeine 26.5 –

Paracetamol 21.4 50.0

Naproxen 21.0 86.1

Ibuprofen 34.1 63.6

M. Fernandez et al. / Journal of Environmental Chemical Engineering xxx (2014) xxx–xxx 7

G Model

JECE 272 1–8

95% were also obtained in WWTP1 for caffeine and ibuprofen.However, paracetamol removal efficiency in the summer samplewas above 90%. In the case of naproxen the values in WWTP1 werebelow 85% during all the year. Considering the exact date ofsampling, average day temperatures were �9 8C, �12 8C, �15 8Cand �19 8C for autumn, winter, spring and summer respectively.Therefore, as it was reported in other works [18,26], the seasonaldifferences of PCPP removal levels reflected that in generaltreatment processes were more effectively for warmer tempera-tures. This trend was clearly observed in WWTP1, with theexception of naproxen whose lowest value was reached insummer. Regarding removal efficiencies for BOD5 and totalnitrogen, WWTP2 values for these parameters did not show aclear tendency, whereas in case of WWTP1 achieved a maximum insummer.

The apparent kinetic constants (k) for the removal of thestudied contaminants were calculated for both WWTPs (Table 5).The expected increase in k values with temperature was clearlyobserved in WWTP1 with only two anomalous values (naproxen insummer and ibuprofen in winter). In the case of WWTP2, almost allthe summer values were lower than expected. This behavior islikely to be related with some kind of modification in microbialcommunity running, as a consequence of a drastic change intemperature or composition of the water suffered in previous days.A lack of tendency was also observed for caffeine in WWTP2, butthis can be explained because concentrations measured in theeffluent were very low, near LOD and therefore, these measure-ments have larger uncertainty. It is important to remark that k

values are within the same order in both WWTPs (between 0.06and 0.71 h�1 in all cases), observing values slightly higher inWWTP2 for naproxen and paracetamol.

Ferguson et al. [28] observed a negative correlation betweentemperature and total pharmaceutical abundance in an USA lake.This indicated that lower temperatures reduce the rate ofbiodegradation of these PPCPs in surface water and sewagetreatment plants, finding that temperature is more important todegradation of some pollutants than another.

With the aim of determining the relation between the k

constants and temperature, an Arrhenius type equation wasemployed. Values with anomalous behavior (showed in gray inTable 5) were not considered in the calculation of the apparentactivation energy. Apparent activation energy values for BOD5 and

Please cite this article in press as: M. Fernandez, et al., Seasonal ocwastewaters, J. Environ. Chem. Eng. (2014), http://dx.doi.org/10.101

total N were similar in both WWTPs, whereas WWTP2 showedvalues twice or three times higher for paracetamol, naproxen andibuprofen (see Table 6). This indicated that, the PPCP removalprocess is much more sensible to temperature in case of WWTP2than in WWTP1.

Conclusions

Among the six PPCPs considered in this study, ibuprofen,paracetamol, naproxen and caffeine were the compounds detectedin higher levels in the analyzed wastewaters. Tendenciesthroughout the year could not be established for PPCPs concen-trations found at the entrance of the WWTPs. Results showed thattraditional treatment for municipal wastewaters were not effectivein removing diclofenac and carbamazepine, whereas paracetamol,ibuprofen, naproxen and caffeine concentrations decreasedbetween 75% and 99% during the process. Differences of PCPPremoval levels found in samples taken at different seasonsreflected that, in general, the treatment throughout the WWTPswere more effectively for warmer temperatures. Additionally, oneof the studied WWTP resulted to be more sensible to temperaturechanges, as it was showed by kinetic parameter values. Furtherresearch would be interesting in order to complete the conclusionshere achieved and establish the convenience of controlling certainof these pollutants in municipal WWTPs in a systematic way.

Acknowledgments

The authors thank the ‘‘Consorcio de Aguas de Asturias’’(Government of Asturias, Spain) and the ‘‘Hospital UniversitarioCentral de Asturias’’ (HUCA) for supplying the samples andMEDYCSA for funding the project.

References

[1] C. Reyes-Contreras, V. Matamoros, I. Ruiz, M. Soto, J.M. Bayona, Evaluation ofPPCPs removal in a combined anaerobic digester-constructed wetland pilot planttreating urban wastewater, Chemosphere 84 (2011) 1200–1207.

[2] V.L. Cunningham, M. Buzby, T. Hutchinson, F. Mastrocco, N. Parke, N. Roden,Effects of human pharmaceuticals on aquatic life: next steps, Environ. Sci.Technol. 40 (2006) 3456–3462.

[3] M. Al Aukidy, P. Verlicchi, A. Jelic, M. Petrovic, D. Barcelo, Monitoring release ofpharmaceutical compounds: occurrence and environmental risk assessment oftwo WWTP effluents and their receiving bodies in the Po Valley, Italy, Sci. TotalEnviron. 438 (2012) 15–25.

[4] X. Yang, R.C. Flowers, H.S. Weinberg, P.C. Singer, Occurrence and removal ofpharmaceuticals and personal care products (PPCPs) in an advanced wastewaterreclamation plan, Water Res. 45 (2011) 5218–5228.

[5] Y. He, W. Chen, X. Zheng, X. Wang, X. Huang, Fate and removal of typicalpharmaceuticals and personal care products by three different treatment pro-cesses, Sci. Total Environ. 447 (2013) 248–254.

[6] T. Heberer, D. Feldmann, Contribution of effluents from hospitals and privatehouseholds to the total loads of diclofenac and carbamazepine in municipalsewage effluents-modeling versus measurements, J. Hazard. Mater. 122 (2005)211–218.

[7] D.W. Kolpin, E.T. Furlong, M.T. Meyer, E.M. Thurman, S.D. Zaugg, L.B. Barber, H.T.Buxton, Pharmaceuticals, hormones, and other organic wastewater contaminantsin US streams, 1999–2000: a national reconnaissance, Environ. Sci. Technol. 36(2002) 1202–1211.

currence and removal of pharmaceutical products in municipal6/j.jece.2014.01.023

409 [8410

411

412 [9413

414

415 [10416

417

418 [11419

420

421 [12422

423

424 [13425

426

427 [14428

429

430 [15431

432 [16433

434

435 [17436

437

438 [18439

440

441 [19442

443

444 [20445

446

447 Q3

448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486

M. Fernandez et al. / Journal of Environmental Chemical Engineering xxx (2014) xxx–xxx8

G Model

JECE 272 1–8

] N.M. Vieno, T. Tuhkanen, L. Kronberg, Seasonal variation in the occurrence ofpharmaceutical sin effluents from a sewage treatment plant and in the recipientwater, Environ. Sci. Technol. 39 (2005) 8220–8226.

] G.A. Loraine, M.E. Pettigrove, Seasonal variations in concentrations of pharma-ceuticals and personal care products in drinking water and reclaimed wastewaterin southern California, Environ. Sci. Technol. 40 (2006) 687–695.

] Q. Sui, J. Huang, S. Deng, G. Yu, Q. Fan, Occurrence and removal of pharmaceu-ticals, caffeine and DEET in wastewater treatment plants of Beijing, China, WaterRes. 44 (2010) 417–426.

] P. Verlicchi, M. Al Aukidy, A. Galletti, M. Petrovic, D. Barcelo, Hospital effluent:investigation of the concentrations and distribution of pharmaceuticals andenvironmental risk assessment, Sci. Total Environ. 430 (2012) 109–118.

] B. Gy Plosz, H. Leknes, H. Liltved, K.V. Thomas, Diurnal variations in the occur-rence and the fate of hormones and antibiotics in activated sludge wastewatertreatment in Oslo, Norway, Sci. Total Environ. 408 (2010) 1915–1924.

] M. Pomies, J.M. Choubert, C. Wisniewski, M. Coquery, Modelling of micropollutantremoval in biological wastewater treatments: a review, Sci. Total Environ. 443(2013) 733–748.

] M. Novak, P. Horvat, Mathematical modelling and optimisation of a waste watertreatment plant by combined oxygen electrode and biological waste watertreatment model, Appl. Math. Model. 36 (2012) 3813–3825.

] A. Togola, H. Budzinski, Multi-residue analysis of pharmaceutical compounds inaqueous samples, J. Chromatogr. A 1177 (2008) 150–158.

] S.S. Verenitch, C.J. Lowe, A. Mazumder, Determination of acidic drugs and caffeinein municipal wastewaters and receiving waters by gas chromatography-ion traptandem mass spectrometry, J. Chromatogr. A 1116 (2006) 193–203.

] J. Santos, I. Aparicio, M. Callejon, E. Alonso, Occurrence of pharmaceutically activecompounds during 1-year period in wastewaters from four wastewater treatmentplants in Seville (Spain), J. Hazard. Mater. 164 (2009) 1509–1516.

] M. Hijosa-Valsero, V. Matamoros, J. Martın-Villacorta, E. Becares, J. Bayona,Assessment of full-scale natural systems for the removal of PPCPs from waste-water in small communities, Water Res. 44 (2010) 1429–1439.

] M.D. Camacho-Munoz, J.L. Santos, I. Aparicio, E. Alonso, Presence of pharmaceu-tically active compounds in Donana Park (Spain) main watersheds, J. Hazard.Mater. 177 (2010) 1159–1162.

] M. Gomez, M. Patrovic, A. Fernandez-Alba, D. Barcelo, Determination of pharma-ceuticals of various therapeutic classes by solid-phase extraction and liquidchromatography-tandem mass spectrometry analysis in hospital effluent waste-waters, J. Chromatogr. 1114 (2006) 224–233.

Please cite this article in press as: M. Fernandez, et al., Seasonal owastewaters, J. Environ. Chem. Eng. (2014), http://dx.doi.org/10.10

[21] A. Spongberg, J. Witter, Pharmaceutical compounds in the wastewater processstream in Northwest Ohio, Sci. Total Environ. 397 (2008) 148–157.

[22] C. Kosma, D. Lambropoulou, T. Albanis, Occurrence and removal of PPCPs in munici-pal and hospital wastewaters in Greece, J. Hazard. Mater. 179 (2010) 804–817.

[23] A. Jelic, M. Gros, A. Ginebreda, R. Cespedes-Sanchez, F. Ventura, M. Petrovic, D.Barcelo, Occurrence, partition and removal of pharmaceuticals in sewage waterand sludge during wastewater treatment, Water Res. 45 (2011) 1165–1176.

[24] T. Rauch-Williams, C. Hoppe-Jones, J. Drewes, The role of organic matter in theremoval of emerging trace organic chemicals during managed aquifer recharge,Water Res. 44 (2010) 449–460.

[25] S. Zorita, L. Martensson, L. Mathiasson, Occurrence and removal of pharmaceu-ticals in a municipal sewage treatment system in the south of Sweden, Sci. TotalEnviron. 407 (2009) 2760–2770.

[26] J.T. Yu, E.J. Bouwer, M. Coelthan, Occurrence and biodegradability studies ofselected pharmaceuticals and personal care products in sewage effluent, Agric.Water Manage. 86 (2006) 72–80.

[27] B.D. Blair, J.P. Crago, C.J. Hedman, R.J.F. Treguer, C. Magruder, L.S. Royer, R.D. Klaper,Evaluation of a model for the removal of pharmaceuticals, personal care products,and hormones from wastewater, Sci. Total Environ. 444 (2013) 515–521.

[28] P.J. Ferguson, M.J. Bernot, J.C. Doll, T.E. Lauer, Detection of pharmaceuticals andpersonal care products (PPCPs) in near-shore habitats of southern Lake Michigan,Sci. Total Environ. 458–460 (458) (2013) 187–190.

[29] A.Y.C. Lin, T.H. Yu, S.K. Lateef, Removal of pharmaceuticals in secondary waste-water treatment processes in Taiwan, J. Hazard. Mater. 167 (2009) 1163–1169.

[30] P.H. Roberts, K.V. Thomas, The occurrence of selected pharmaceuticals in waste-water effluent and surface waters of the lower Tyne catchment, Sci. Total Environ.356 (2006) 143–153.

[31] L. Lishman, S.A. Smyth, K. Sarafin, S. Kleywegt, J. Toito, T. Peart, B. Lee, M. Servos,M. Beland, P. Seto, Occurrence and reductions of pharmaceuticals and personalcare products and estrogens by municipal wastewater treatment plants inOntario, Canada, Sci. Total Environ. 367 (2006) 544–558.

[32] T. Ternes, Occurrence of drugs in German sewage treatment plants and rivers,Water Res. 32 (1998) 3245–3260.

[33] K. Fent, A.A. Weston, D. Caminada, Ecotoxicology of human pharmaceuticals,Aquat. Toxicol. 76 (2006) 122–159.

[34] L.H.M.L.M. Santos, M. Gros, S. Rodriguez-Mozaz, C. Delerue-Matos, A. Pena, D.Barcelo, M.C.B.S.M. Montenegro, Contribution of hospital effluents to the load ofpharmaceuticals in urban wastewaters: identification of ecologically relevantpharmaceuticals, Sci. Total Environ. 461–462 (461) (2013) 302–310.

ccurrence and removal of pharmaceutical products in municipal16/j.jece.2014.01.023