Determination of Benzene, Toluene, Ethyl Benzene, Xylenes in Water

7
Determination of benzene, toluene, ethylbenzene, xylenes in water at sub-ng l 1 levels by solid-phase microextraction coupled to cryo-trap gas chromatography–mass spectrometry Maw-Rong Lee a, * , Chia-Min Chang a , Jianpeng Dou a,b a Department of Chemistry, National Chung Hsing University, Taichung 40227, Taiwan, ROC b College of Biological and Agricultural Engineering, Jilin University, Changchun 130022, PR China Received 26 December 2006; received in revised form 1 May 2007; accepted 3 May 2007 Abstract A trace analytical method of benzene, toluene, ethylbenzene and xylenes (BTEX) in water has been developed by using headspace solid-phase microextraction (HS-SPME) coupled to cryo-trap gas chromatography–mass spectrometry (GC–MS). The chromatographic peak shape for BTEX was improved by using cryo-trap equipment. The HS-SPME experimental procedures to extract BTEX from water were optimized with a 75 lm carboxen/polydimethylsiloxane (CAR/PDMS)-coated fiber at a sodium chloride concentration of 267 g l 1 , extraction for 15 min at 25 °C and desorption at 290 °C for 2 min. Good linearity was verified in a range of 0.0001–50 lgl 1 for each analyte (r 2 = 0.996–0.999). The limits of detection (LODs) of BTEX in water reached at sub-ng l 1 levels. LODs of benzene, toluene, ethylbenzene, m/p-xylene and o-xylene were 0.04, 0.02, 0.05, 0.01 and 0.02 ng l 1 , respectively. The proposed analytical method was suc- cessfully used for the quantification of trace BTEX in ground water. The results indicate that HS-SPME coupled to cryo-trap GC–MS is an effective tool for analysis of BTEX in water samples at the sub-ng l 1 level. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Solid-phase microextraction; Benzene; Toluene; Ethylbenzene; Xylene; Cryo-trap 1. Introduction Benzene, toluene, ethylbenzene, m/p-xylene and o-xylene (BTEX) are ubiquitously environmental contaminants in air, water and soil and are widely used in industries, such as printing, paint, synthetic resin, and synthetic rubber (Holcomb and Seabrook, 1995; Nollet, 2001). BTEX are also abundant in petroleum products, such as fuel oil and gasoline. But BTEX has an effect on human health includ- ing neurological diseases or cancer (Chiou et al., 1982). The US Environmental Protection Agency (EPA) establishes the maximum contaminant level (MCL) for benzene 5 lgl 1 , for toluene 1000 lgl 1 , for ethylbenzene 700 lgl 1 and for xylenes 10 000 lgl 1 in drinking water (USEPA, 2004). The maximum contaminant level of 1 lgl 1 for benzene in drinking water is established by the European Union (EU, 1998). Therefore, in order to protect people’s health, it is necessary to establish an effec- tive and convenient quantification method for monitoring trace BTEX in water. BTEX in water were analyzed by direct aqueous injec- tion (DAI) and gas chromatography combined with flame ionization detector (GC–FID), and limits of detection (LOD) ranged from 0.6 lgl 1 for benzene to 1.1 lgl 1 for o-xylene (Kubinec et al., 2005). BTEX in water samples were also analyzed by headspace solid-phase microextrac- tion (SPME) and GC–FID (Flo ´rez Mene ´ndez et al., 2000; Almeida and Boas, 2004; Ezquerro et al., 2004), and LOD obtained with PDMS/DVB/CAR fibre ranged from 15 ng l 1 (benzene) to 160 ng l 1 (toluene) (Almeida and Boas, 2004). Substituted benzene was analyzed by liquid-phase microextraction (LPME) and GC–FID, and 0045-6535/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2007.05.004 * Corresponding author. Tel.: +886 4 2285 1716. E-mail address: [email protected] (M.-R. Lee). www.elsevier.com/locate/chemosphere Chemosphere xxx (2007) xxx–xxx ARTICLE IN PRESS Please cite this article in press as: Lee, M.-R. et al., Determination of benzene, toluene, ethylbenzene, xylenes in water ..., Chemo- sphere (2007), doi:10.1016/j.chemosphere.2007.05.004

description

Determination of benzene, toluene, ethylbenzene, xylenes in water at sub-ng/l levels by solid-phase microextraction coupled to\cryo-trap gas chromatography–mass spectrometry

Transcript of Determination of Benzene, Toluene, Ethyl Benzene, Xylenes in Water

Page 1: Determination of Benzene, Toluene, Ethyl Benzene, Xylenes in Water

ARTICLE IN PRESS

www.elsevier.com/locate/chemosphere

Chemosphere xxx (2007) xxx–xxx

Determination of benzene, toluene, ethylbenzene, xylenes in waterat sub-ng l�1 levels by solid-phase microextraction coupled to

cryo-trap gas chromatography–mass spectrometry

Maw-Rong Lee a,*, Chia-Min Chang a, Jianpeng Dou a,b

a Department of Chemistry, National Chung Hsing University, Taichung 40227, Taiwan, ROCb College of Biological and Agricultural Engineering, Jilin University, Changchun 130022, PR China

Received 26 December 2006; received in revised form 1 May 2007; accepted 3 May 2007

Abstract

A trace analytical method of benzene, toluene, ethylbenzene and xylenes (BTEX) in water has been developed by using headspacesolid-phase microextraction (HS-SPME) coupled to cryo-trap gas chromatography–mass spectrometry (GC–MS). The chromatographicpeak shape for BTEX was improved by using cryo-trap equipment. The HS-SPME experimental procedures to extract BTEX from waterwere optimized with a 75 lm carboxen/polydimethylsiloxane (CAR/PDMS)-coated fiber at a sodium chloride concentration of 267 g l�1,extraction for 15 min at 25 �C and desorption at 290 �C for 2 min. Good linearity was verified in a range of 0.0001–50 lg l�1 for eachanalyte (r2 = 0.996–0.999). The limits of detection (LODs) of BTEX in water reached at sub-ng l�1 levels. LODs of benzene, toluene,ethylbenzene, m/p-xylene and o-xylene were 0.04, 0.02, 0.05, 0.01 and 0.02 ng l�1, respectively. The proposed analytical method was suc-cessfully used for the quantification of trace BTEX in ground water. The results indicate that HS-SPME coupled to cryo-trap GC–MS isan effective tool for analysis of BTEX in water samples at the sub-ng l�1 level.� 2007 Elsevier Ltd. All rights reserved.

Keywords: Solid-phase microextraction; Benzene; Toluene; Ethylbenzene; Xylene; Cryo-trap

1. Introduction

Benzene, toluene, ethylbenzene, m/p-xylene and o-xylene(BTEX) are ubiquitously environmental contaminants inair, water and soil and are widely used in industries, suchas printing, paint, synthetic resin, and synthetic rubber(Holcomb and Seabrook, 1995; Nollet, 2001). BTEX arealso abundant in petroleum products, such as fuel oil andgasoline. But BTEX has an effect on human health includ-ing neurological diseases or cancer (Chiou et al., 1982). TheUS Environmental Protection Agency (EPA) establishesthe maximum contaminant level (MCL) for benzene5 lg l�1, for toluene 1000 lg l�1, for ethylbenzene 700lg l�1 and for xylenes 10000 lg l�1 in drinking water

0045-6535/$ - see front matter � 2007 Elsevier Ltd. All rights reserved.

doi:10.1016/j.chemosphere.2007.05.004

* Corresponding author. Tel.: +886 4 2285 1716.E-mail address: [email protected] (M.-R. Lee).

Please cite this article in press as: Lee, M.-R. et al., Determinationsphere (2007), doi:10.1016/j.chemosphere.2007.05.004

(USEPA, 2004). The maximum contaminant level of1 lg l�1 for benzene in drinking water is established bythe European Union (EU, 1998). Therefore, in order toprotect people’s health, it is necessary to establish an effec-tive and convenient quantification method for monitoringtrace BTEX in water.

BTEX in water were analyzed by direct aqueous injec-tion (DAI) and gas chromatography combined with flameionization detector (GC–FID), and limits of detection(LOD) ranged from 0.6 lg l�1 for benzene to 1.1 lg l�1

for o-xylene (Kubinec et al., 2005). BTEX in water sampleswere also analyzed by headspace solid-phase microextrac-tion (SPME) and GC–FID (Florez Menendez et al.,2000; Almeida and Boas, 2004; Ezquerro et al., 2004),and LOD obtained with PDMS/DVB/CAR fibre rangedfrom 15 ng l�1 (benzene) to 160 ng l�1 (toluene) (Almeidaand Boas, 2004). Substituted benzene was analyzed byliquid-phase microextraction (LPME) and GC–FID, and

of benzene, toluene, ethylbenzene, xylenes in water ..., Chemo-

Page 2: Determination of Benzene, Toluene, Ethyl Benzene, Xylenes in Water

2 M.-R. Lee et al. / Chemosphere xxx (2007) xxx–xxx

ARTICLE IN PRESS

LODs of BTEX ranged from 0.3 lg l�1 for o-xylene to1.6 lg l�1 for toluene in a water matrix (Wang et al.,2006). At present, gas chromatography–mass spectrometry(GC–MS) is widely used to determine BTEX in differentwater samples. Generally, sample preparation of BTEXfrom real samples is an important procedure before analy-sis. Liquid–liquid extraction, solid-phase extraction andpurge-and-trap techniques were used for determination ofBTEX in the past (Stan and Kirsch, 1995; Beketov et al.,1996; Miermans et al., 2000; Bianchi et al., 2002; Rosellet al., 2003). BTEX in water were analyzed by purge-and-trap and GC–MS, and LOD ranged from 2.8 ng l�1 foro-xylene to 22.6 ng l�1 for ethylbenzene (Bianchi et al.,2002). However, solid-phase microextraction (SPME), asrapid, selective and solvent-free techniques, are more andmore widely used for analysis of trace BTEX (Potter andPawliszyn, 1992; Elke et al., 1998; Fuoco et al., 1999; Koz-iel et al., 2000; Li et al., 2001; Matisova et al., 2002;Almeida and Boas, 2004; Arambarri et al., 2004). BTEXwere analyzed by SPME–GC–MS, and LODs ranged from15 ng l�1 for benzene to 50 ng l�1 for o-xylene in a watermatrix (Potter and Pawliszyn, 1992).

The aim of this study was to develop a higher sensitivitymethod for determination of trace BTEX in water samples.In this work, BTEX were adsorbed on the fiber coating ofthe SPME. A headspace (HS) SPME method was used toextract trace BTEX in water. The cryo-trap could curtailchromatographic peak breadth of volatile compoundsand improve chromatographic shape. Therefore, BTEXextract was analyzed by cryo-trap-GC–MS. The conditionsfor extracting BTEX from water samples are optimized.The optimized HS-SPME–GC–MS method was used todetermine BTEX in ground water.

2. Materials and methods

2.1. Chemicals and materials

Benzene (99.99%), toluene (99.5%), ethylbenzene(99.97%), o-xylene (99.3%), m-xylene (99.8%), and p-xylene(99.9%) were purchased from TEDIA Company (Fairfield,OH, USA). Sodium chloride (99.8%) was obtained fromRiedel-deHaen Company (Seelze, Germany). One standardmixture of BTEX were prepared at individual concentra-tion of 100 mg l�1 in acetone and diluted with water toyield the required concentration. All solutions were storedat 4 �C in a refrigerator. All chemicals and reagents used in

Table 1Analytical conditions of BTEX by GC–MS with SIM mode

Compound Molecular weight Retention time (min

Benzene 78 4.95Toluene 92 5.99Ethylbenzene 106 7.78m/p-Xylene 106 7.98o-Xylene 106 8.45

Please cite this article in press as: Lee, M.-R. et al., Determinationsphere (2007), doi:10.1016/j.chemosphere.2007.05.004

this work were analytical grade without further purifica-tion. The purified water was obtained by an SG Ultra clearwater purification system (SG Water Company, Barsbuttel,Germany). All glassware was silanized before it was usedby soaking the glassware overnight in toluene solution with10% dichlorodimethylsilane. The glassware was rinsed withtoluene and methanol and then thoroughly dried for 4 h.

2.2. GC–MS

GC–MS and some parameters set were in accordancewith the description of a literature (Lee et al., 2000). TheGC oven temperature program was as follows: 40 �C held2 min, rate 25 �C min�1 to 70 �C, rate 2 �C min�1 to80 �C, rate 25 �C min�1 to 250 �C, held for 1 min. Electronimpact ionization (EI) was used as ionization mode forBTEX analysis. An injector temperature was maintainedat 290 �C. The mass spectra were obtained at a mass-to-charge ratio (m/z) scan range from 45 to 300 u. Cryo-trapwas produced by Scientific Instrument Services Inc. (Mode961, Ringoes, NJ, USA). BTEX are trapped in the cryo-trap accumulation cell, which allows all of BTEX werequantitatively transferred from the cryo-trap cell to thechromatographic column by the helium flow. The temper-ature of cryo-trap was initially set at �100 �C (held 2 min),and then reached sharply to 250 �C. To increase sensitivity,the selected ion monitoring (SIM) mode of EI was alsoapplied in quantitative analysis.

The most abundant ion was generally monitored as wellas quantified and the specific ion was used as the confirmedion. The compounds of m-xylene and p-xylene were com-bined and analyzed as sum of the corresponding peak areasbecause they could not be separated during the GC analysisand have the same fragmentation pathways. In Table 1, themost abundant ion of benzene is m/z 78 ([M]+) ion and themost abundant ion of the other BTEX is m/z 91 ([M�H]+

or [M�CH3]+) ion. Hence, the fragment ions m/z 78 and 91were used for quantification of benzene and other BTEX,respectively.

2.3. Sample preparation

The SPME fibers were purchased from Supelco Com-pany (Bellefonte, PA, USA). SPME sampling was similarto the description of a literature (Lee et al., 2007). Theground water samples (nearby gas station, paint factory,river and university in Taichung, Taiwan) were stored at

) Quantification ions (m/z) Confirmed ions (m/z)

78 77, 52, 5191 92, 65, 5191 106, 77, 6591 106, 77, 6591 106, 77, 65

of benzene, toluene, ethylbenzene, xylenes in water ..., Chemo-

Page 3: Determination of Benzene, Toluene, Ethyl Benzene, Xylenes in Water

M.-R. Lee et al. / Chemosphere xxx (2007) xxx–xxx 3

ARTICLE IN PRESS

4 �C in a refrigerator and analyzed within 48 h to avoidstorage losses. A 15 ml of sample solution was undertakenin a 40 ml sample vials and closed with a PFTE-coated sep-tum. A 75 lm CAR/PDMS fiber was used to extract BTEXfrom water samples. Triplicate analyses were performed.Extraction temperature was at 25 �C. During extraction,the sample solution was continuously agitated at a constantvelocity of 1000 rpm with a Teflon-coated stir bar (0.8 cm ·2.0 cm) on a magnetic stirrer.

3. Results and discussion

3.1. Optimization of SPME conditions

The extraction efficiency of the SPME experiment couldbe affected by type of fiber, desorption temperature,desorption time, absorption time, sample matrix and vari-ous other factors (Lagalante and Felter, 2004). Differenttypes of coatings provide different absorption propertiesfor different kinds of analytes. The choice of an appropriatecoating is crucial for the SPME method. Various coatingswere tested under the conditions: standard solution con-centration of BTEX, 40 lg l�1; extraction temperature,25 �C; extraction time, 15 min. Triplicate analyses wereperformed, relative standard deviations of all fibers wereless than 10%. The result in Fig. 1 reveals that the 75 lmCAR/PDMS fiber is the best for simultaneous extractionof BTEX. CAR/PDMS coating (Elke et al., 1998; Kozielet al., 2000; Li et al., 2001; Arambarri et al., 2004) was suit-able for analysis of small molecular and nonpolar com-pounds. Although Cho et al. proposed that CAR/PDMSshould be carefully used for determination of BTEX ingroundwater samples, there is hardly interference amongBTEX components when the concentration of the analyteis at a low level (Cho et al., 2003). Therefore, 75 lmCAR/PDMS was used for the analysis of low-concentra-tion BTEX in this study.

The principle of SPME is based on partition equilibriumbetween the concentration of analytes in a sample and that

Fig. 1. Extraction efficiencies of 40 lg l�1 BTEX in water with v

Please cite this article in press as: Lee, M.-R. et al., Determinationsphere (2007), doi:10.1016/j.chemosphere.2007.05.004

in the solid-phase fiber coating (Arthur and Pawliszyn,1990). Although a stir bar could accelerate the mass trans-fer of analytes through the aqueous matrix, the time ofextraction, the temperature and headspace of sample vialhave an effect on the partition equilibrium. Extractionswere performed from 5 to 30 min to determine the effectof extraction time (Fig. 2). Fresh samples were used foreach extraction time studied. BTEX have all reached parti-tion equilibrium in 15 min. The extraction temperaturedetermines the mass transfer rate of BTEX from water intofiber. Extraction temperatures from 25 to 75 �C were inves-tigated. The maximum amount extracted was at 25 �C, andthen decreased gradually as the temperature increased fur-ther. The reason probably is the partition coefficient ofBTEX on fiber coating decreased as the temperatureincreased. The more BTEX volatilized from water matrixto gas phase as extraction temperature rose, the more ana-lytes were adsorbed on the fiber, but at the same time themuch more analytes were desorbed from fiber for higherextraction temperature, therefore the total abundance ofBTEX gradually decreased as the extraction temperatureincreased. The headspace of vial would also affect extrac-tion efficiencies for sample by SPME. The absolute amountof BTEX on headspace of sample vial was relevant to head-space and solution volume of vial. A solution of 1, 2, 5, 10,15 and 20 ml was added in 40 ml of vial, respectively. Theresults in Fig. 3 show the peak areas of BTEX increasedwith the volume of solution to a maximum at 15 ml excepto-xylene. Therefore, 15 ml of BTEX water solution in 40 mlvial are extracted for 15 min by HS-SPME at 25 �C.

The desorption temperature and desorption time deter-mine the amount of analytes desorbed from the fiber coat-ing, as determined by the SPME method. The desorptiontemperature was investigated with a range of 260–300 �C.The results indicate that the peak areas of BTEX increasedwith the desorption temperature, reaching a maximumdesorption amount at 290 �C. Desorption time was investi-gated within a range of 0.5–4 min, by leaving the fiber inthe injector for an increasing period of time and maintain-

arious fiber coatings adsorption for 15 min at 25 �C (n = 3).

of benzene, toluene, ethylbenzene, xylenes in water ..., Chemo-

Page 4: Determination of Benzene, Toluene, Ethyl Benzene, Xylenes in Water

Fig. 2. Effect of extraction time on peak areas of 40 lg l�1 BTEX in water produced by HS-SPME at 25 �C.

Fig. 3. Effect of solution volume in 40 ml vial on peak areas of 40 lg l�1 BTEX in water produced by HS-SPME for 15 min at 25 �C.

4 M.-R. Lee et al. / Chemosphere xxx (2007) xxx–xxx

ARTICLE IN PRESS

ing the temperature of the injector at 290 �C. The amountof BTEX desorption increased with desorption time andreached a maximum after 2 min. Therefore, a 290 �Cdesorption temperature and 2 min desorption time wereused in the experiment. The SPME fiber can be continu-ously used during the experiment without any carry overafter desorbed 290 �C for 2 min.

Changes in the sample matrix had a significant effect onthe signal intensities of analytes obtained by SPME (Leeet al., 2000, 2007). The analyte of neutral molecular formis extracted easily by SPME. In order to investigate theeffect of salting out on the signal intensities of analytes,sodium chloride was added to the water samples to yieldfinal concentrations of sodium chloride of 0.067, 0.133,0.200, 0.267 and 0.333 g ml�1. A blank solution (40 lg l�1

of BTEX standard solution) with no added sodium chlo-ride was also tested. Sodium chloride is able to increase

Please cite this article in press as: Lee, M.-R. et al., Determinationsphere (2007), doi:10.1016/j.chemosphere.2007.05.004

the ionic strength and improve the amount of analytesextracted by the SPME fiber. The results indicate that thepeak areas of BTEX increase with the amount of NaClto a maximum at 0.267 g ml�1. Therefore, the extractionis pursued at additional sodium chloride concentration of0.267 g ml�1.

3.2. Method validation

The linearity, limits of detection and precision were cal-culated when the optimum conditions for the HS-SPME–cryo-trap-GC–MS procedure were established. The linear-ity of the HS-SPME method was examined by extractingstandard solutions spiked 0.0001, 0.001, 0.01, 1, 10 and50 lg l�1 BTEX, respectively. Triplicate injections wereperformed. The r2 were above 0.996. Lack-of-fit and Man-del’s fitting test were performed to check the goodness of fit

of benzene, toluene, ethylbenzene, xylenes in water ..., Chemo-

Page 5: Determination of Benzene, Toluene, Ethyl Benzene, Xylenes in Water

M.-R. Lee et al. / Chemosphere xxx (2007) xxx–xxx 5

ARTICLE IN PRESS

and linearity (Bianchi et al., 2002; Prikryl et al., 2006).Lack-of-fit tests demonstrated that the linear models wereadequate because the whole p values were more than 0.05at significance level of 95%. Mandel’s fitting tests were alsoperformed for the mathematical verification of linearity. F

values calculated lower than the tabulated F-value at theconfidence level of 95% indicating that a quadratic regres-sion would not provide a significantly better fit than a lin-ear one (Table 2). The linear range experiments provided

Table 2Linear range, limit of detection (LOD) and precision of BTEX in water by SP

Compound Linear range (lg l�1) r2 LOD (ng l�1) L

Benzene 0.0001–50 0.998 0.04 1Toluene 0.0001–50 0.998 0.02 0Ethylbenzene 0.0001–50 0.996 0.05 0m/p-Xyleneb 0.0001–50 0.999 0.01 0o-Xylene 0.0001–50 0.998 0.02 0

a Confidence interval, 95%.b The value of m/p-xylene expressed the sum of m-xylene and p-xylene.

Table 3Concentration (lg l�1) of BTEX in water samples by SPME–cryo-trap-GC–M

Real water sample Benzene Toluene

S1b 5.9 ± 0.5 6.2 ± 0.8

S2 0.72 ± 0.08 0.95 ± 0.04S3 <c <S4 0.18 ± 0.05 36 ± 4.20S5 < 24.2 ± 2.2S6 15.2 ± 2.5 35.2 ± 6.0S7 < 1.21 ± 0.09

a The value of m/p-xylene expressed the sum of m-xylene and p-xylene.b S1–S3: ground water nearby different gas station in Taichung; S4–S5: ground

some university in Taichung; S7: river water in Taichung.c < : <LOQ value.

Fig. 4. Mass ion chromatogram of ground water

Please cite this article in press as: Lee, M.-R. et al., Determinationsphere (2007), doi:10.1016/j.chemosphere.2007.05.004

the necessary information to estimate LODs, based onthe lowest detectable peak with a signal-to-noise ratio ofthree. LODs of SPME used to determine BTEX in waterrelies on the amount of analytes adsorbed by coating onthe fiber and the sensitivity of the GC–MS. Under theexperimental conditions, LODs were 0.01–0.05 ng l�1 inwater. The precision of the HS-SPME method was evalu-ated by analyzing BTEX at two concentration levels foreach analyte. The results in Table 2 showed that the RSDs

ME–cryo-trap-GC–MS

ack-of-fit, pa Mandel’s fitting test, Fa RSD (%, n = 9)

0.1 lg l�1 40 lg l�1

.000 0.01 11.2 5.2

.862 2.27 8.9 4.5

.999 0.18 11.6 6.8

.606 4.16 8.4 3.1

.661 4.17 7.8 4.8

S

Ethylbenzene m/p-Xylenea o-Xylene

0.62 ± 0.05 0.85 ± 0.04 0.35 ± 0.050.15 ± 0.02 0.10 ± 0.03 0.12 ± 0.02< < <ND 1.60 ± 0.05 0.56 ± 0.040.28 ± 0.06 0.52 ± 0.02 0.98 ± 0.057.4 ± 0.9 12.1 ± 0.4 4.0 ± 0.4< < <

water nearby different paint factory in Taichung; S6: ground water nearby

S6 analyzed by HS-SPME–cryo-trap-GC–MS.

of benzene, toluene, ethylbenzene, xylenes in water ..., Chemo-

Page 6: Determination of Benzene, Toluene, Ethyl Benzene, Xylenes in Water

6 M.-R. Lee et al. / Chemosphere xxx (2007) xxx–xxx

ARTICLE IN PRESS

were lower than 12% in aqueous matrix. Therefore, BTEXin sorption on a CAR/PDMS-coated fiber in HS-SPMEwere deemed acceptable for determining at sub-ng l�1 lev-els in water solution.

3.3. HS-SPME–cryo-trap-GC–MS of real samples

The proposed method was used to quantify BTEX inground water. The HS-SPME was operated under the opti-mum conditions. Triplicate analyses were performed. Theresults in Table 3 shows that BTEX was present in groundwater. Fig. 4 shows the mass ion chromatogram of S6,which the contents of BTEX were 4.0 lg l�1 (o-xylene) to35.2 lg l�1 (toluene). Recoveries in the range 102–106%were obtained for all samples. The results indicate the suit-ability of the HS-SPME–cryo-trap-GC–MS method foranalyzing trace BTEX in water.

4. Conclusions

Results from this study indicate that SPME coupled toGC–MS is a precise method for reproducibly analyzingtrace BTEX in water. Better sensitivity were obtained byheadspace SPME combined with GC–MS, and the chro-matographic shape is improved further by cryo-trap.Detection limits of BTEX in water at sub-ng l�1 concentra-tion levels were achieved and linear ranges were over fiveorders of magnitude for all the analytes. Earlier studieson determination of BTEX were at from lg l�1 to ng l�1

level and linear ranges were less than four orders of magni-tude. BTEX were indeed present in ground water samplesprobably contaminated by gasoline, paint and thinners,and chemical solvents.

Acknowledgements

The authors thank the National Science Council of theRepublic of China for financially supporting this researchunder contract No. NSC92-2113-M-005-024.

References

Almeida, C.M.M., Boas, L.V., 2004. Analysis of BTEX and othersubstituted benzenes in water using headspace SPME-GC–FID:method validation. J. Environ. Monitor. 6, 80–88.

Arambarri, I., Lasa, M., Garcia, R., Millan, E., 2004. Determination offuel dialkyl ethers and BTEX in water using headspace solid-phasemicroextraction and gas chromatography–flame ionization detection.J. Chromatogr. A 1033, 193–203.

Arthur, C.L., Pawliszyn, J., 1990. Solid phase microextraction withthermal desorption using fused silica optical fibers. Anal. Chem. 62,2145–2148.

Beketov, V.I., Parchinski, V.Z., Zorov, N.B., 1996. Effects of high-frequency electromagnetic treatment on the solid-phase extraction ofaqueous benzene, naphthalene and phenol. J. Chromatogr. A 731, 65–73.

Bianchi, F., Careri, M., Marengo, E., Musci, M., 2002. Use ofexperimental design for the purge-and-trap-gas chromatography–massspectrometry determination of methyl tert.-butyl ether, tert.-butyl

Please cite this article in press as: Lee, M.-R. et al., Determinationsphere (2007), doi:10.1016/j.chemosphere.2007.05.004

alcohol and BTEX in groundwater at trace level. J. Chromatogr. A975, 113–121.

Chiou, C.T., Schmedding, D.W., Manes, M., 1982. Partition of organiccompounds in octanol–water systems. Environ. Sci. Technol. 16, 4–10.

Cho, H.-J., Baek, K., Lee, H.-H., Lee, S.-H., Yang, J.-W., 2003.Competitive extraction of multi-component contaminants in waterby carboxen–polydimethylsiloxane fiber during solid-phase micro-extraction. J. Chromatogr. A 988, 177–184.

Elke, K., Jermann, E., Begerow, J., Dunemann, L., 1998. Determinationof benzene, toluene, ethylbenzene and xylenes in indoor air atenvironmental levels using diffusive samplers in combination withheadspace solid-phase microextraction and high-resolution gas chro-matography–flame ionization detection. J. Chromatogr. A 826, 191–200.

EU, 1998. Council Directive 98/83/EC of 3 November 1998 on quality ofwater intended for human consumption, Official Journal of theEuropean Communities, L 330, 5/12/1998, 0032–0054.

Ezquerro, O., Ortiz, G., Pons, B., Tena, M.T., 2004. Determination ofbenzene, toluene, ethylbenzene and xylenes in soils by multipleheadspace solid-phase microextraction. J. Chromatogr. A 1035, 17–22.

Florez Menendez, J.C., Fernandez Sanchez, M.L., Sanchez Urıa, J.E.,Fernandez Martınez, E., Sanz-Medel, A., 2000. Static headspace,solid-phase microextraction and headspace solid-phase microextrac-tion for BTEX determination in aqueous samples by gas chromato-graphy. Anal. Chim. Acta 415, 9–20.

Fuoco, R., Ceccarini, A., Onor, M., Marrara, L., 1999. Analysis ofpriority pollutants in environmental samples by on-line supercriticalfluid chromatography cleanup–cryo-trap-gas chromatography–massspectrometry. J. Chromatogr. A 846, 387–393.

Holcomb, L.C., Seabrook, B.S., 1995. Indoor concentrations of volatileorganic compounds: implications for comfort, health and regulation.Indoor Environ. 4, 7–26.

Koziel, J., Jia, M.Y., Pawliszyn, J., 2000. Air sampling with porous solid-phase microextraction fibers. Anal. Chem. 72, 5178–5186.

Kubinec, R., Adamuscin, J., Jurdakova, H., Foltin, M., Ostrovsky, I.,Kraus, A., Sojak, L., 2005. Gas chromatographic determination ofbenzene, toluene, ethylbenzene and xylenes using flame ionizationdetector in water samples with direct aqueous injection up to 250 ll. J.Chromatogr. A 1084, 90–94.

Lagalante, A.F., Felter, M.A., 2004. Silylation of acrylamide for analysisby solid-phase microextraction/gas chromatography/ion-trap massspectrometry. J. Agric. Food Chem. 52, 3744–3748.

Lee, M.R., Song, Y.S., Hwang, B.H., Chou, C.C., 2000. Determination ofamphetamine and methamphetamine in serum via headspace deriva-tization solid-phase microextraction–gas chromatography–mass spec-trometry. J. Chromatogr. A 896, 265–273.

Lee, M.R., Chang, L.Y., Dou, J., 2007. Determination of acrylamide infood by solid-phase microextraction coupled to gas chromatography–positive chemical ionization tandem mass spectrometry. Anal. Chim.Acta 582, 19–23.

Li, K., Santilli, A., Goldthorp, M., Whiticar, S., Lambert, P., Fingas, M.,2001. Solvent vapour monitoring in work space by solid phase microextraction. J. Hazard. Mater. 83, 83–91.

Matisova, E., Medved’ova, M., Vraniakova, J., Simon, P., 2002. Optimi-sation of solid-phase microextraction of volatiles. J. Chromatogr. A960, 159–164.

Miermans, C.J.H., van der Velde, L.E., Frintrop, P.C.M., 2000. Analysisof volatile organic compounds, using the purge and trap injectorcoupled to a gas chromatograph/ion-trap mass spectrometer: review ofthe results in Dutch surface water of the Rhine, Meuse, NorthernDelta Area and Westerscheldt, over the period 1992–1997. Chemo-sphere 40, 39–48.

Nollet, L.M.L. (Ed.), 2001. Handbook of Water Analysis. Marcel Dekker,New York (Chapter 33).

Potter, D.W., Pawliszyn, J., 1992. Detection of substituted benzenes inwater at the pg/ml level using solid-phase microextraction and gaschromatography–ion trap mass spectrometry. J. Chromatogr. A 625,247–255.

of benzene, toluene, ethylbenzene, xylenes in water ..., Chemo-

Page 7: Determination of Benzene, Toluene, Ethyl Benzene, Xylenes in Water

M.-R. Lee et al. / Chemosphere xxx (2007) xxx–xxx 7

ARTICLE IN PRESS

Prikryl, P., Kubinec, R., Jurdakova, H., Sevcıky, J., Ostrovsk, I., Sojak,L., Berezkin, V., 2006. Comparison of needle concentrator with SPMEfor GC determination of benzene, toluene, ethylbenzene, and xylenesin aqueous samples. Chromatographia 64, 65–70.

Rosell, M., Lacorte, S., Ginebreda, A., Barcelo, D., 2003. Simultaneousdetermination of methyl tert.-butyl ether and its degradation products,other gasoline oxygenates and benzene, toluene, ethylbenzene andxylenes in Catalonian groundwater by purge-and-trap-gas chromato-graphy–mass spectrometry. J. Chromatogr. A 995, 171–184.

Please cite this article in press as: Lee, M.-R. et al., Determinationsphere (2007), doi:10.1016/j.chemosphere.2007.05.004

Stan, H.J., Kirsch, N.H., 1995. GC–FID determination of chloroben-.zene isomers in methanogenic batch-cultures from river sediments. Int.J. Environ. Anal. Chem. 60, 33–40.

USEPA, 2004. List of Contaminants & their Maximum ContaminantLevel (MCLs) in drinking water. http://www.epa.gov/safewater/mcl.html#organic.

Wang, J.-X., Jiang, D.-Q., Yan, X.-P., 2006. Determination of substitutedbenzenes in water samples by fiber-in-tube liquid phase microextrac-tion coupled with gas chromatography. Talanta 68, 945–950.

of benzene, toluene, ethylbenzene, xylenes in water ..., Chemo-