A

tb2wdoisla©

K

1

uipBeraddtost

eH

0d

Available online at www.sciencedirect.com

Journal of Chromatography A, 1189 (2008) 278–284

Electrokinetic supercharging for on-line preconcentration of sevennon-steroidal anti-inflammatory drugs in water samples

Mohamed Dawod, Michael C. Breadmore, Rosanne M. Guijt ∗, Paul R. HaddadAustralian Centre for Research on Separation Science (ACROSS), School of Chemistry, University of Tasmania, Hobart, Tasmania, Australia

Available online 26 December 2007

bstract

The development of new sensitive methods for the analysis of non-steroidal anti-inflammatory drugs (NSAIDs) in water samples is of great impor-ance. In this work, seven NSAIDs were separated within 9 min using 15 mM sodium tetraborate (pH 9.2) containing 0.1% (w/v) hexadimethrineromide (HDMB) and 10% (v/v) methanol. Field-amplified sample injection (FASI) was examined and found to improve the detection limits by00-fold providing detection limits of 0.6–2.0 �g/L, but these are insufficient for the determination of NSAIDs as environmental pollutants inater samples. To improve the sensitivity further, electrokinetic supercharging (EKS) was examined. The optimum EKS method involved hydro-ynamic injection leading electrolyte (100 mM NaCl, 30 s, 50 mbar), electrokinetic injection of the sample (200 s, −10 kV) and finally injectionf the terminating electrolyte (100 mM 2-(cyclohexylamino) ethanesulphonic acid, CHES, 40 s, 50 mbar). With this method, the sensitivity wasmproved by 2400-fold giving detection limits of 50–180 ng/L. The developed method was validated and then applied to the analysis of wastewater

amples from a local sewage treatment plant. The detection limits were found to increase by approximately 10-fold, however, this is still lower thanevels previously found in wastewater samples from European and Mediterranean cities. The proposed method has the advantage of simplicity andchieving sensitivity through high-preconcentration power without the use of off-line chromatographic sample cleanup. 2007 Elsevier B.V. All rights reserved.

dal an

goirsga[

fpmC

eywords: Capillary electrophoresis; Electrokinetic supercharging; Non-steroi

. Introduction

Non-steroidal anti-inflammatory drugs (NSAIDs) have beensed widely for several decades for the treatment of differentnflammatory disorders, pain relief, and also for their anti-yretic effect; some NSAIDs are available without prescription.ecause of their high solubility and poor degradability in water,limination of NSAIDs in sewage treatment plants (STPs) isather low and consequently they are able to penetrate throughll natural filtration steps and enter groundwater as well asrinking water [1]. The continuous environmental input of suchrugs may lead to a relatively high long-term concentration andhereby to promote continuous, but unnoticed adverse effects

n aquatic and terrestrial organisms [2]. Different toxicologicaltudies have been performed showing the possible environmen-al hazards of NSAIDs, for example ibuprofen stimulates the

∗ Corresponding author at: Australian Centre for Research on Separation Sci-nce (ACROSS), School of Chemistry, University of Tasmania, Private Bag 75,obart, Tasmania, Australia. Tel.: +61 3 6226 2171; fax: +61 3 6226 2858.

E-mail address: Rosanne.Guijt@utas.edu.au (R.M. Guijt).

iicecoivt

021-9673/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.chroma.2007.12.056

ti-inflammatory drugs; Field-amplified sample injection; Water samples

rowth of the cyanobacterium Synechocystis, inhibits the growthf duckweed Lemna minor [3] and results in steroidogenesisn rainbow trout [4]. Chronic exposure to diclofenac leads toenal damage in brown trout [5] and rainbow trout [6]. Moreignificantly, it has been found that a mixture of NSAID anal-esics was toxic for certain aquatic organisms at concentrationst which the single compounds showed no or only little effects7].

Because of its advantageous high separation efficiency andast analysis time, capillary electrophoresis (CE) has beenroven to be a useful technique for the separation and deter-ination of NSAIDs in a range of sample matrices. DifferentE modes have been described for the analysis of NSAIDs,

ncluding; capillary zone electrophoresis (CZE) [8–23], cap-llary electrochromatography [24,25], micellar electrokineticapillary chromatography (MEKC) [8,26–31], microemulsionlectrokinetic chromatography (MEEKC) [32,33], and isota-hophoresis (ITP) [34,35]. While excellent separations can be

btained by these approaches, the major disadvantage of CEs its low concentration detection limit. This is due to theery small optical path length for spectrophotometric detec-ion (typically 50–100 �m) and the limited amount of sample

atogr

tTatcaoasmssre(spiltAest

gTaatEffiaetaEltn

ostspo(testsll

2

2

lipwhfMMAA

pNmbs

ctTf

2

3w2ilt

11wf

2

c4tia

2

M. Dawod et al. / J. Chrom

hat can be introduced into the capillary (typically <1 �L).his drawback is very important when low limits of detectionre required, such as for the analysis of many environmen-al and biological samples. This can be overcome by on-lineoncentration and various electrophoretic and chromatographicpproaches have been developed to improve the sensitivityf CE [36,37]. A number of these approaches have beenpplied to the analysis of NSAIDs, including field-amplifiedample injection with sample matrix removal using electroos-otic flow (EOF) pumping (FAEP), large volume sample

tacking with EOF pumping (LVSEP), LVSEP with anionelective exhaustive injection (ASEI) [21,38,39], stacking witheversed migrating micelles (SRMM), SRMM-ASEI, and field-nhanced sample injection with reverse migrating micellesFESI-RMM) [31]. Stacking with these approaches has beenhown to improve the sensitivity of the method (based oneak area) by factors ranging between 100 and 1800-fold, giv-ng detection limits as low as 100 ng/L. However, even thisevel of sensitivity requires an additional solid phase extrac-ion and enrichment step when dealing with real water samples.

system that could achieve the detection limits required fornvironmental analysis of NSAIDs without the need for off-lineample processing would obviously have significant advan-ages.

A recent on-line preconcentration method for CE that hasreat potential is that of electrokinetic supercharging (EKS).his is the combination of electrokinetic injection under field-mplified conditions (field-amplified sample injection, FASI)nd transient isotachophoresis (tITP) and was first described forhe analysis of rare-earth ions [40,41] by the group of Hirokawa.KS was developed to extend the range of FASI and is per-

ormed by hydrodynamic injection of a leading electrolyte,ollowed by EKI of the analytes, and finally hydrodynamicnjection of a terminating electrolyte. Upon applying the sep-ration voltage the diffuse band of analytes introduced duringlectrokinetic injection is stacked between the leading and theerminating electrolytes by tITP until the ITP stage destacksnd the analytes are allowed to separate by conventional CE.KS is an exceptionally simple but powerful approach to on-

ine sample preconcentration and has been shown to improvehe sensitivity of analytical response by several orders of mag-itude.

In the current work, the separation and preconcentrationf seven NSAIDs by CE has been examined, with empha-is on the development of a simple and rapid CE method forhe determination of these NSAIDs in environmental wateramples. Towards this end, co-EOF separations have beenerformed using an EOF reversal agent and the separationptimized by variation of the composition of the electrolyteelectrolyte concentration and methanol content). After selec-ivity optimization, the potential of FASI and EKS for on-linenrichment of NSAIDs has been examined. The results pre-ented in this work provide the lowest detection limits for

hese NSAIDs without using off-line solid phase extraction forample enrichment, and the preconcentration procedure takesess time than previously published methods for these ana-ytes.

cit

. A 1189 (2008) 278–284 279

. Experimental

.1. Standards and reagents

Naproxen was purchased from Fluka (Buchs, Switzer-and), while diclofenac, diflunisal, fenoprofen, ibuprofen,ndomethacin, ketoprofen, 2-(cyclohexylamino) ethanesul-honic acid (CHES), and hexadimethrine bromide (HDMB)ere from Sigma–Aldrich (St. Louis, MO, USA). Sodiumydroxide (98%) and disodium tetraborate decahydrate wererom BDH (Kilsyth, Australia). Sodium chloride was from

&B Pronalys Analytical Reagents (West Footscray, Australia).ethanol (HPLC-Grade) was from Ajax Finechem (Seven Hills,ustralia). Water was treated with a Millipore (North Ryde,ustralia) Milli-Q water purification system.A stock standard solution of 1 mg/ml of each drug was pre-

ared in methanol. A mixed standard solution of the sevenSAIDs was prepared at a concentration of 0.1 mg/ml inethanol. The working standard solutions were prepared daily

y diluting the stock standard solution with Milli-Q water. Allolutions were stored in dark containers at 4 ◦C.

The working background electrolyte (BGE) solution had aoncentration of 15 mM of disodium tetraborate (pH 9.2) con-aining 10% methanol and 0.1% HDMB unless otherwise stated.he buffer solutions were prepared freshly each day, sonicated

or 5 min and filtered through a 0.45 �m membrane filter.

.2. Instrumentation

Electrophoretic separations were performed using an AgilentD CE (Agilent Technologies, Waldbronn, Germany) equippedith a UV diode-array detection (DAD) system operating at14 nm. Separations were carried out using fused silica capillar-es (Polymicro Technologies, Phoenix, AZ, USA) of 85 cm totalength (76.5 cm effective length) and 50 �m i.d. The capillaryemperature was set at 25 ◦C.

New capillaries were flushed with 1 M sodium hydroxide for20 min, with Milli-Q water for 20 min, and with the BGE for0 min. Each day the capillaries were equilibrated by rinsingith 1 M sodium hydroxide for 10 min, with HDMB (1%, w/v)

or 10 min, and with the BGE for 5 min.

.3. Field-amplified sample injection

The analytes dissolved in Milli-Q water were injected into theapillary electrokinetically with a negative voltage (−3 kV) for0 s. All NSAIDs are weakly acidic (Fig. 1) and under the condi-ions used these analytes were negatively charged and migratednto the capillary by a combination of electrophoretic migrationnd EOF.

.4. Electrokinetic supercharging

A small volume of the leading electrolyte (100 mM sodiumhloride) was introduced into the capillary by hydrodynamicnjection at 50 mbar for 30 s, then the sample was injected elec-rokinetically by a negative voltage (−10 kV) for 200 s, and

280 M. Dawod et al. / J. Chromatogr. A 1189 (2008) 278–284

pKa values of the studied NSAIDs.

fiCvts

2

(tAs

3

heuliisw

Fig. 1. Chemical structures and

nally a small volume of the terminating electrolyte (100 mMHES) was injected hydrodynamically at 50 mbar for 40 s. Aoltage of −28 kV was applied for both the on-line focusing andhe separation of the analytes. The EKS procedure is representedchematically in Fig. 2.

.5. Preparation of wastewater samples

Wastewater was obtained from the effluent of Selfs Point STPHobart, Australia). Prior to analysis, the sample was filteredhrough a 0.45 �m nylon membrane syringe filter (Phenomenex,ustralia) to eliminate particulate matter. The samples were

tored in the refrigerator in dark glass containers at −4 ◦C.

. Results and discussion

While a number of CE methods for the analysis of NSAIDsave been previously developed [8–29,31–35,38,39], theseither involve long separation times (15–30 min) or involve these of short-end injection which is incompatible with most on-ine enrichment strategies capable of providing improvements

n sensitivity greater than 1000. We were interested in develop-ng a highly efficient and rapid method for the separation of theelected NSAIDs using conditions that would also be compatibleith EKS.

Fig. 2. Schematic representation of the steps used in EKS: (1) filling the capil-lary with background electrolyte (BGE), (2) hydrodynamic injection of leadingelectrolyte (L), (3) elecktrokinetic injection of sample (S), (4) hydrodynamicinjection of terminating electrolyte (T) and (5) starting tITP-CZE.

atogr. A 1189 (2008) 278–284 281

3

otbwratbtimsasotr

e6dsiOtTaniint

Fsdv(

Fc

tewwc

awEateNM

M. Dawod et al. / J. Chrom

.1. Optimization of separation selectivity

Highly efficient and rapid separations in CE are typicallybtained by performing separations in which analyte migra-ion is in the same direction as the EOF. This approach haseen used by Macia et al. [19] and Ahrer and Buchberger [20]ho performed co-EOF separations of NSAIDs using HDMB to

everse the EOF. Separation times of 15 min were obtained usingmmonium acetate as BGE and electrospray mass spectrome-ry detection. A similar system was utilized in the present studyy employing a low concentration of HDMB (0.01%, w/v) inhe electrolyte to reverse the EOF but using sodium tetraboratenstead of ammonium acetate due to the superior UV trans-

ission of borate electrolytes at low UV wavelengths. Initialeparations using 10 mM sodium tetraborate, pH 9.2 with theddition of 0.1% HDMB to the electrolyte provided very rapideparations (less than 7 min), but baseline resolution was notbtained due to co-migration of some of the analytes. As such,he separation electrolyte was further optimized to improve theesolution of the NSAIDs.

First, the effect of concentration of the separation buffer wasxamined. Borate buffer concentrations ranging between 10 and0 mM were examined, with concentrations above 60 mM pro-ucing high and unstable currents leading to very irreproducibleeparations. Fig. 3 shows the change in electrophoretic mobil-ty of the NSAIDs over the concentrations of borate examined.ne of the most interesting findings was the increase in elec-

rophoretic mobility for all drugs with increasing ionic strength.his behaviour is contrary to that expected in CZE but may bettributable to decreased electrostatic interactions between theegatively charged analytes and the HDMB in the BGE due toncreased ion-exchange competition effects from anions present

n the high ionic strength BGE. Further support for this expla-ation was obtained from the noticeable increase in migrationime for all analytes with increasing HDMB concentration in

ig. 3. Effect of buffer concentration on the electrophoretic mobility of theelected NSAIDs. CE conditions: fused silica capillary 80 cm × 50 �m i.d.; BGEisodium tetraborate containing 10% methanol and 0.1% HDMB; separationoltage: −30 kV; detection, UV at 214 nm; hydrodynamic injection of NSAIDs10 ppm) at 50 mbar for 10 s.

actnItmhubt

3

esssFSilsn

ig. 4. Effect of concentration of methanol on the mobility of the drugs. CEonditions as in Fig. 3 with a concentration of 15 mM disodium tetraborate.

he BGE (data not shown). Based on consideration of speed andfficiency of separation, the optimal concentration of the bufferas found to be 15 mM, although it is important to note that itas still difficult to resolve ketoprofen and naproxen under these

onditions.To try to improve the separation, the influence of the percent-

ge of MeOH added to the BGE on the selectivity of separationas studied. Organic modifiers have been used often in CE andKC to optimize selectivity via changes in solvation and/orcid–base equilibria. In this case, MeOH can also impact onhe interaction between the NSAIDs and the HDMB. The influ-nce of adding up to 50% (v/v) MeOH on the migration of theSAIDs can be seen from Fig. 4. The addition of more than 50%eOH was found to prolong the migration time and adversely

ffect the peak shape of most of the analytes. From the figure, itan be seen that as the concentration of MeOH was increased,he migration time decreased for all analytes (except for diflu-isal), which is typical behaviour for MeOH additions in CZE.n contrast, diflunisal showed a slight increase in mobility withhe addition of more than 20% (v/v) MeOH which we speculate

ay be due to a change in the ionization state of the phenolicydroxyl group on diflunisal. The best separations were obtainedsing 10–20% MeOH, however, 10% MeOH was judged aseing optimal because it gave a similar resolution but in a shorterime.

.2. Optimization of detection sensitivity

Having optimized the separation selectivity, it was then nec-ssary to improve the sensitivity to provide detection limitsuitable for the detection of NSAIDs in environmental wateramples. Macia et al. [31,38] have employed a number oftrategies to improve the sensitivity of NSAIDs, includingAEP, LVSEP, LVSEP-ASEI for CZE separations and SRMM,RMM-ASEI, FESI-RMM for MEKC separations. The best

mprovement was 1800 by FAEP, providing detection limits asow as 100 ng/L using electrokinetic injection for 170 s, whichuggests that the best approach is one based on long electroki-etic injections.

2 romat

obltbatrtstswlcaombb

tFadtfttiipigt

Fwdv

tCatnts

iddai1jqFb−eiuc−swtlslid

82 M. Dawod et al. / J. Ch

One of the methods that has not been examined so far is thatf FASI, which is much simpler than the other methods and haseen shown by several authors to improve the sensitivity forow conductivity samples by at least 1000 [42,43]. To examinehis approach, the analytes were introduced into the capillaryy injection at different voltages, ranging from −2 to −30 kVnd at different times ranging from 20 to 200 s. It was foundhat injection at lower voltages and longer times provided moreeproducible results than those that at higher voltages and shorterimes. Injection at−3 kV for 40 s gave the highest response whiletill maintaining acceptable resolution between the peaks. Usinghis approach and the electrolyte conditions developed above, theelected NSAIDs were successfully resolved in less than 10 minith a LOD of 600 ng/L, as shown in Fig. 5. This detection

imit represents an improvement in sensitivity of about 200 timesompared with the use of conventional hydrodynamic injectionnd CZE separation, but is still considerably higher than thatbtained by Macia et al. [38]. While the FASI approach is auch simpler system than that previously developed, it cannot

e applied for analysis of the NSAIDs in wastewater samplesecause it is insufficiently sensitive.

To further improve the sensitivity, the use of EKS was inves-igated. As shown in the schematic representation of EKS inig. 2, the capillary is filled initially with the BGE, after whichplug of leading electrolyte (100 mM sodium chloride) is intro-uced into the capillary by hydrodynamic injection. Sample ishen introduced into the capillary by electrokinetic injection,ollowed by a final hydrodynamic injection of terminating elec-rolyte (100 mM CHES). When the separation voltage is applied,he diffuse band of NSAIDs introduced during the electrokineticnjection is stacked between the leading ion and the terminat-ng ion by tITP, after which it destacks and the separation is

erformed by CZE. The choice of both leading and terminat-ng ions is very important for EKS to function properly as theseovern the range of ions that can be stacked and the extent ofITP stacking that occurs. In this work, we selected chloride as

ig. 5. FASI of a standard mixture of the NSAIDs (50 ppb) dissolved in Milli-Qater. CE conditions: fused silica capillary 85 cm × 50 �m i.d.; BGE 15mMisodium tetraborate containing 10% methanol and 0.1% HDMB; separationoltage: −28 kV; detection, UV at 214 nm. EKI of sample at −3 kV for 40 s.

t

3

EdsFa(wsfFnct

3

fwt

ogr. A 1189 (2008) 278–284

he leading ion, which has a high electrophoretic mobility, whileHES was selected as a terminating electrolyte because it haslow mobility at the pH used in this work. Also important is

he hydrodynamic injection times of both the leader and termi-ator, and the electrokinetic injection time and voltage used forhe sample. These parameters were examined using a univariateequential approach.

First, different volumes of 100 mM NaCl were introducednto the capillary by hydrodynamic injection at 50 mbar forifferent times ranging between 10 and 100 s. Second, the hydro-ynamic injection time of the terminator was varied between 10nd 100 s at 50 mbar. In both cases, the maximum hydrodynamicnjection of 100 s at 50 mbar was used as this corresponded to3.5% of the total capillary length and longer injections wereudged to leave too little of the capillary available for subse-uent separation of the sample after destacking of the tITP step.inally, we investigated the electrokinetic injection of analytesy injecting the sample at different voltages ranging between2 and −30 kV for 5–300 s. The system that provided the high-

st peaks and still allowed baseline resolution was obtained bynjection of the leader for 30 s at 50 mbar (2% of capillary vol-me), injection of the terminator for 40 s at 50 mbar (2.7% ofapillary volume) and electrokinetic injection of the sample at10 kV for 200 s. Using this approach, all the analytes were

uccessfully concentrated and separated in less than 10 min,ith a limit of detection of 50 ng/L. A representative separa-

ion obtained under these conditions is shown in Fig. 6. Thisimit of detection corresponds to 2400 times improvement inensitivity when compared to hydrodynamic injection and isower than that obtained by Macia et al. [38]. This improvementn sensitivity is the best reported to date for analysis of stan-ard solutions of NSAIDs without the use of off-line samplereatment.

.3. Method validation

Within-day and between-day precision values of FASI andKS were investigated to assess method validation and repro-ucibility. Within-day precision was evaluated by five replicateeparations of the seven NSAIDs in a concentration of 5 �g/L forASI and 0.5 �g/L for EKS. Between-day precision was evalu-ted by doing the same separations for 5 different days. Precisionexpressed as percentage relative standard deviation (R.S.D.%))as calculated for both migration times and peak areas. Table 1

hows the R.S.D.% values for FASI, while Table 2 shows dataor EKS. While EKS showed much lower detection limits thanASI, it also exhibited much lower precision due to the electroki-etic injection of the sample, and also temperature-associatedhanges in viscosity of the leader and terminator associated withemperature fluctuations in an unregulated laboratory.

.4. Application to real samples

To demonstrate the potential of the developed EKS methodor the analysis of NSAIDs in environmental samples, waste-ater was collected from Selfs Point STP effluent (Hobart, Aus-

ralia). Fig. 7 shows results from the direct injection of STP

M. Dawod et al. / J. Chromatogr. A 1189 (2008) 278–284 283

Fig. 6. EKS injection of a standard mixture of the NSAIDs (5 ppb) in Milli-Qwater (A) full injection and (B) close up of analyte separation. CE conditions:fused silica capillary, 85 cm × 50 �m I.D; BGE 15 mM disodium tetraboratecontaining 10% methanol and 0.1% HDMB. Voltage −28 kV, hydrodynamici2U

eoPgt

Table 2Within-day and between-day reproducibilities (R.S.D.%) of migration times andpeak areas, and LODs for EKS

Compound Within-dayR.S.D.% (n = 5)

Between-dayR.S.D.% (n = 5)

LOD(ng/L)

Migrationtime

Peakarea

Migrationtime

Peakarea

Diclofenac 1.06 11.41 10.49 10.85 124Diflunisal 1.36 13.90 10.21 14.23 68Fenoprofen 0.90 14.04 9.79 11.89 84Ibuprofen 0.85 13.51 9.77 14.46 180Indomethacin 1.16 13.55 10.22 13.16 149Ketoprofen 0.93 13.15 9.86 14.37 100Naproxen 0.95 14.69 9.80 12.09 50

Conditions as in Fig. 6.

Fwi

womue

TW

C

DDFIIKN

C

njection of 100 mM NaCl at 50 mbar for 30 s, EKI of sample at −10 kV for00 s, hydrodynamic injection of 100 mM CHES at 50 mbar for 40 s; detection,V at 214 nm.

ffluent and also the same effluent spiked with 20 �g/L of each

f the drugs. No significant peaks were observed in the Selfsoint STP effluent (Hobart, Australia), which is unsurprisingiven the relatively small population density. Interestingly a sys-em peak was observed for this sample that partially co-migrated

tstw

able 1ithin-day and between-day reproducibilities (R.S.D.%) of migration times and pea

ompound Within-day R.S.D.% (n = 5)

Migration time Peak area

iclofenac 0.63 8.43iflunisal 0.68 5.06enoprofen 0.55 4.91buprofen 0.59 3.98ndomethacin 0.63 6.19etoprofen 0.60 5.94aproxen 0.61 5.57

onditions as in Fig. 5.

ig. 7. Electropherogram obtained from EKS of (A) wastewater sample spikedith 20 ppb of the NSAIDs and (B) blank wastewater sample. CE conditions as

n Fig. 6.

ith diflunisal, the origin of which is currently unknown. Limitsf detection achieved with EKS for STP effluent were approxi-ately 10 times higher than when compared to those obtained

sing standards prepared in pure water, which is due to theffect of the salt content in wastewater samples on the elec-

rokinetic injection step. Although the LODs for real wateramples were lower than for standards, the LODs achieved inhe real samples were below levels reported for these drugs inastewater samples in Greece and Germany [44,45] and the

k areas, and LODs for FASI

Between-day R.S.D.% (n = 5) LOD (ng/L)

Migration time Peak area

1.19 14.48 16401.26 13.76 19001.13 11.71 11501.04 10.90 21401.20 14.60 19101.12 9.32 12401.07 7.33 600

2 romat

ds

4

Nis(wiclw1LEttip

A

epoa(c

R

[[

[

[[

[

[[[[[[[[[[[

[

[

[[

[

[

[[

[[[[[[[[

84 M. Dawod et al. / J. Ch

eveloped method would therefore have applicability to theseamples.

. Conclusions

A new simple and sensitive method for the separation ofSAIDs has been developed. A co-EOF system was used allow-

ng the separation of seven NSAIDs within 9 min using 15 mModium tetraborate (pH 9.2), 0.01% (w/v) HDMB and 10%v/v) MeOH. On-line preconcentration using FASI and EKSas also examined. The sensitivity for standard solutions was

ncreased by 200-fold using FASI, and by 2400-fold in thease of EKS due to the ability to inject at higher voltage foronger time. When applied to the analysis of STP effluentater, the LODs achieved for wastewater were approximately0 times higher than for standard solutions. However theseODs are below the concentrations expected to be found in mosturopean and Mediterranean cities and are below the concen-

rations of NSAIDs expected to cause environmental hazardso aquatic and terrestrial organisms. The developed methods therefore suitable for the analysis of environmental sam-les.

cknowledgements

M.D. would like to thank the Egyptian ministry of higherducation for provision of a scholarship. This work was sup-orted by the Australian Research Council through the awardf Australian Postdoctoral Fellowships to M.C.B. (DP0453223)nd R.M.G. (DP0557083) and a Federation Fellowship to P.R.H.FF0668673). The Authors would also like to thank the Hobartity council for supplying the STP effluent samples.

eferences

[1] J. Dorne, A.M.J. RaGas, G.K. Frampton, D.S. Spurgeon, D.F. Lewis, Anal.Bioanal. Chem. 387 (2007) 1167.

[2] M. Petrovic, M.D. Hernando, M.S. Dıaz-Cruz, D. Barcelo, J. Chromatogr.A 1067 (2005) 1.

[3] F. Pomati, A.G. Netting, D. Calamari, B.A. Neilan, Aquat. Toxicol. 67(2004) 387.

[4] A. Hontela, Toxicol. Sci. 93 (2006) 1.[5] B. Hoeger, B. Kollner, D.R. Dietrich, B. Hitzfeld, Aquat. Toxicol. 75 (2005)

53.[6] J. Schwaiger, H. Ferling, U. Mallow, H. Wintermayr, R.D. Negele, Aquat.

Toxicol. 68 (2004) 141.

[[

[

ogr. A 1189 (2008) 278–284

[7] M. Cleuvers, Ecotoxicol. Environ. Safe. 59 (2004) 309.[8] M.G. Donato, E. Vandeneeckhout, W. Vandenbossche, P. Sandra, J. Pharm.

Biomed. Anal. 11 (1993) 197.[9] Z.K. Shihabi, M.E. Hinsdale, J. Chromatogr. B 683 (1996) 115.10] W. Ahrer, E. Scherwenk, W. Buchberger, J. Chromatogr. A 910 (2001) 69.11] S.H. Kang, S.Y. Chang, K.C. Do, S.C. Chi, D.S. Chung, J. Chromatogr. B

712 (1998) 153.12] S. Toasaksiri, D.L. Massart, Y. Vander Heyden, Anal. Chim. Acta 416

(2000) 29.13] S. Pedersen-Bjergaard, K.E. Rasmussen, Electrophoresis 21 (2000) 579.14] J.R. Veraart, C. Gooijer, H. Lingeman, N.H. Velthorst, U.A.T. Brinkman,

J. Chromatogr. B 719 (1998) 199.15] J.R. Veraart, M.C.E. Groot, C. Gooijer, H. Lingeman, N.H. Velthorst,

U.A.T. Brinkman, Analyst 124 (1999) 115.16] Y.L. Chen, S.M. Wu, Anal. Bioanal. Chem. 381 (2005) 907.17] Y.L. Chen, Y.J. Jong, S.M. Wu, J. Chromatogr. A 1119 (2006) 176.18] S. Heitmeier, G. Blaschke, J. Chromatogr. B 721 (1999) 109.19] A. Macia, F. Borrull, M. Calull, C. Aguilar, Electrophoresis 25 (2004) 3441.20] W. Ahrer, W. Buchberger, Montash. Chem. 132 (2001) 329.21] A. Macia, F. Borrull, C. Aguilar, M. Calull, Electrophoresis 24 (2003) 2779.22] S. Cherkaoui, J.L. Veuthey, J. Chromatogr. A 874 (2000) 121.23] M. Fillet, I. Bechet, V. Piette, J. Crommen, Electrophoresis 20 (1999) 1907.24] Y.F. Pai, C.Y. Liu, J. Chromatogr. A 982 (2002) 293.25] C. Desiderio, S. Fanali, J. Chromatogr. A 895 (2000) 123.26] M.G. Donato, W. Baeyens, W. Vandenbossche, P. Sandra, J. Pharm.

Biomed. Anal. 12 (1994) 21.27] C.W. Maboundou, G. Paintaud, M. Berard, P.R. Bechtel, J. Chromatogr. B

657 (1994) 173.28] S. Orlandini, S. Fanali, S. Furlanetto, A.M. Marras, S. Pinzauti, J. Chro-

matogr. A 1032 (2004) 253.29] C. Mardones, A. Rıos, M. Valcarcel, Electrophoresis 22 (2001) 484.30] P. Zhang, G.W. Xu, J.H. Xiong, Y.F. Zheng, X.Z. Shi, Q. Yang, F.S. Wei,

J. Sep. Sci. 26 (2003) 1527.31] A. Macia, F. Borrull, M. Calull, C. Aguilar, J. Chromatogr. A 1117 (2006)

234.32] A. Macia, F. Borrull, M. Calull, C. Aguilar, Chromatographia 63 (2006)

149.33] A. Macia, F. Borrull, M. Calull, C. Aguilar, Electrophoresis 26 (2005) 970.34] J. Sadecka, M. Cakrt, A. Hercegova, J. Polonsky, I. Skacani, J. Pharm.

Biomed. Anal. 25 (2001) 881.35] A. Hercegova, J. Sadecka, J. Polonsky, Electrophoresis 21 (2000) 2842.36] M.C. Breadmore, Electrophoresis 28 (2007) 254.37] M.C. Breadmore, P.R. Haddad, Electrophoresis 22 (2001) 2464.38] A. Macia, F. Borrull, C. Aguilar, M. Calull, Electrophoresis 25 (2004) 428.39] B. Kim, D.S. Chung, Electrophoresis 23 (2002) 49.40] H. Okamoto, T. Hirokawa, J. Chromatogr. A 990 (2003) 335.41] T. Hirokawa, H. Okamoto, B. Gas, Electrophoresis 24 (2003) 498.42] C.X. Zhang, W. Thormann, Anal. Chem. 70 (1998) 540.

43] C.X. Zhang, W. Thormann, Anal. Chem. 68 (1996) 2523.44] M. Rabiet, A. Togola, F. Brissaud, J.L. Seidel, H. Budzinski, F. Elbaz-

Poulichet, Environ. Sci. Technol. 40 (2006) 5282.45] V. Koutsouba, T. Heberer, B. Fuhrmann, K. Schmidt-Baumler, D. Tsipi, A.

Hiskia, Chemosphere 51 (2003) 69.