A novel approach to automation of dynamic ... - Wiley Online Library

16 downloads 76393 Views 259KB Size Report
A novel approach to automation of dynamic hollow fiber liquid-phase microextraction. An automated dynamic two-phase hollow fiber microextraction apparatus ...
957

J. Sep. Sci. 2011, 34, 957–964

Ali Esrafili Yadollah Yamini Mahnaz Ghambarian Morteza Moradi Shahram Seidi Department of Chemistry, Faculty of Sciences, Tarbiat Modares University, Tehran, Iran

Received December 20, 2010 Revised January 17, 2011 Accepted January 22, 2011

Research Article

A novel approach to automation of dynamic hollow fiber liquid-phase microextraction An automated dynamic two-phase hollow fiber microextraction apparatus combined with high-performance liquid chromatography was developed for extraction and determination of chlorophenoxy acid (CPA) herbicides from environmental samples. The extraction device, called TT-extractor, consists of a polypropylene hollow fiber mounted inside a stainless steel tube by means of two tee-connectors in flow system. An organic solvent, which fills the lumen and the pores of the hydrophobic fiber, is pumped through the fiber repeatedly and the sample is pumped along the outer side of the fiber. The factors affecting the dynamic hollow fiber liquid-phase microextraction (DHF-LPME) of target analytes were investigated and the optimal extraction conditions were established. To test the applicability of the designed instrument, CPAs were extracted from environmental aqueous samples. The limits of detection (LODs) as low as 0.5 mg/L, linear dynamic range in the range of 1–100 mg/L and the relative standard deviations of o7% were obtained. The developed method can provide perconcentration factors as large as 230. A hollow fiber membrane can be used at least 20 times with neither loss in the efficiency nor carryover of the analytes between runs. The system is cheap and convenient and requires minimal manual handling. Keywords: Dynamic hollow fiber liquid-phase microextraction / Chlorophenoxy acids / High-performance liquid chromatography DOI 10.1002/jssc.201000913

1 Introduction Human population is constantly exposed to numerous chemical species present in the environment. Among these compounds, chlorophenoxy acid (CPA) herbicides are of significant importance because of their wide distribution and extensive use in agriculture and forestry for control of unwanted weeds and other vegetation. When being applied, they are easily transferred to surface and ground waters due to their polar nature and relatively good solubility in water [1–10]. Although their decomposition in the presence of oxygen is relatively fast, these herbicides are persistent under reducing conditions and their extended use can lead to pollution of surface and ground waters. These compounds are hazardous to living organisms because of their mutagenic, teratogenic, and carcinogenic properties.

Correspondence: Dr. Yadollah Yamini, Department of Chemistry, Faculty of Sciences, Tarbiat Modares University, P. O. Box 14115-175, Tehran, Iran E-mail: [email protected] Fax:198-21-88006544

Abbreviations: CPA, chlorophenoxy acid; MCPA, 2-methyl-4chlorophenoxyacetic acid; PF, preconcentration factor; SPME, solid-phase microextraction; 2,4-D, 2,4-dichlorophenoxyacetic acid

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Therefore, it is important that they be monitored in the environment, especially in aqueous samples. Determination of trace CPAs in aqueous samples requires sensitive and selective analytical techniques such as gas chromatography (GC) [11–17] and high-performance liquid chromatography (HPLC) [18–26]. Currently, GC remains the main analytical method used for herbicide residue analysis because of its excellent separation and detection potential, especially when combined with mass spectrometric detection [27]. However, it suffers from some drawbacks such as the necessity of pretreatment procedures in the case of weakly volatile and/ or thermally labile compounds. The second most utilized method is HPLC, whose procedures have been developed mainly with UV detection [28–30]. To achieve low detection limits (LODs), a preconcentration step prior to analysis is essential. Conventional method such as liquid–liquid extraction (LLE) [31, 32] or solid-phase extraction (SPE) [5, 11, 13, 15] has been applied for this purpose. Nevertheless, these types of extraction methods require large amount of sample and a large volume of organic solvent, and are time-consuming. These problems can be overcome by miniaturized techniques such as solid-phase microextraction (SPME), which has been successfully used for the extraction of acidic herbicides [8, 10, 16, 17]. This simple and solventless extraction technique has proved to be a powerful alternative to traditional extraction techniques. SPME fibers are, however, fragile and www.jss-journal.com

958

J. Sep. Sci. 2011, 34, 957–964

A. Esrafili et al.

relatively expensive and tend to degrade with repeated use. Liquid-phase microextraction (LPME) [33–35] is an emerging technique based on the use of a small amount of organic solvent to extract analytes from aqueous matrixes. Hollow fiber-protected LPME [36, 37] is an improved type of LPME in which the extraction solvent is protected and stabilized in the hollow fiber. Hollow fiber LPME has mostly been performed in static mode, but several studies in dynamic mode [38, 39] and also some automated techniques [40, 41] have been published. The present study describes an automated dynamic extractor, which is well suited for automated liquid– liquid microextraction. It is called the TT-tube extractor, since two tee (T)-connectors and stainless steel tubing were used to mount a hollow fiber membrane in a flow system. In addition to two-phase extraction, this extractor has the potential to be used for three-phase extraction of pharmaceuticals in biological samples. The extraction method requires a minimum of manual handling with a reusable hollow fiber and the design makes it straightforward to couple with analytical instruments. In this study, the dynamic extractor was applied to the extraction of two CPA herbicides in water samples. The effects of various experimental parameters on the extraction

of CPAs from the water samples were investigated and optimized.

2 Materials and methods 2.1 Chemicals and materials The Accurel Q3/2 polypropylene hollow fiber membrane (600-mm id, 200 mm wall thickness, 0.2 mm pore size) was supplied by Membrana (Wuppertal, Germany). 2-Methyl-4chlorophenoxyacetic acid (MCPA) and 2,4-dichlorophenoxyacetic acid (2,4-D) were purchased from Aldrich (Milwaukee, WI, USA). 1-Octanol, dihexyl ether, ethyl acetate, hexanol, MIBK, methanol, and sodium chloride were of the highest purity available from Merck (Darmstadt, Germany). Stock standard solution of each analyte with concentration of 1000 mg/L was prepared separately in methanol and stored at 41C. Mixtures of standard working solutions with different concentrations were prepared daily by dilution of stock solutions with ultrapure water. This ultrapure water was purified by a Milli-Q water purification system from Millpore (Bedford, MA, USA). Analytes at spiking levels were used for the optimization study. Standard

Figure 1. Schematic diagram of the proposed dynamic HF-LPME apparatus.

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.jss-journal.com

Sample Preparation

J. Sep. Sci. 2011, 34, 957–964

solution containing 100 mg/L of the analytes was used in the optimization experiments.

2.2 Instrumentation Chromatographic separations were carried out on a Varian HPLC equipped with a 9012 HPLC pump (Mulgrave, Victoria, Australia) as well as a 9010 autosampler having a 20-mL sample loop and a Varian 9050 UV/Vis detector. Chromatographic data were recorded and analyzed using a computerized software designed in-house. Separations were carried out on a PerfectSil Target ODS column (250 mm  4.6 mm, with particle size of 3 mm) from MZAnalysentechnik (Germany). A mixture of 50 mM phosphate (NaH2PO4) buffer solution (pH 3.0) and acetonitrile (52:48) at a flow rate of 1 mL/min was used as a mobile phase in isocratic elution mode. The injection volume was 20 mL for all of the samples and the detection was performed at a wavelength of 280 nm. Also, 25-mL (model 702N) and 50-mL (model 1705N) Hamilton microsyringes (Bonaduz, Switzerland) were employed for injection and extraction, respectively.

959

ends. The device was placed in a vertical position and connected to programmable syringe and peristaltic pumps; the former supplied the movement of organic phase inside the fiber, whereas the latter supplied the recycling of the aqueous sample solution. The organic solvent was placed in a 300-mL insert-vial from Agilent (Palo Alto, CA, USA) at the end of the extractor as shown in Fig. 1.

2.4 Extraction procedure After mounting a new fiber inside the TT-extractor, the organic extraction solvent which was placed in the microvial was pumped through the fiber channel at upward direction with a flow rate of 3 mL/s. After filling the fiber channel and saturating its pores with organic solvent, the aqueous sample was pumped around the fiber with adjusted flow rate. After each extraction, the sample-phase channel was washed with water and the inside of the fiber flushed with organic phase at a flow rate of 3 mL/s for 5 min. After this step, the system was ready for subsequent experiments. A single hollow fiber membrane could be used at least 20 times with no loss of the efficiency and also without any carryover of the analytes between runs.

2.3 The TT-tube extractor Schematic diagram of TT-tube extractor is shown in Fig. 1. The extractor is easily assembled by hand without any tools. The extraction device contains two stainless steel tee (T)connectors which are connected by stainless steel tubing (3 mm id) housing a 10-cm-long piece of hollow fiber membrane. The hollow fiber at one end was connected to a 50-mL HPLC syringe needle and its other end was fixed with a medical syringe needle. These two syringe needles were connected to the tube by suitable ferrules at their unglued

3 Results and discussion 3.1 Optimization of dynamic HF-LPME conditions 3.1.1 Selection of organic extraction solvent An important step in hollow-fiber microextraction is the choice of the most suitable extraction solvent. The criteria for the selection of a suitable organic solvent are high

Figure 2. Effect of the type of organic extraction solvent on the extraction efficiency. Conditions: sample pH 5 1; volume of sample 5 20 mL; analyte’s concentration 5 100.0 mg/mL; sample flow rate 5 20 mL/min; time 5 30 min; temperature 5 251C; dwell time 5 2 min; no salt added.

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.jss-journal.com

960

A. Esrafili et al.

solubility of the analytes in the organic solvent, low volatility to prevent solvent loss, the solvent’s compatibility with hollow fiber, and no or less toxicity. Considering the above issues, some organic solvents, namely n-octanol, ethyl acetate, dihexyl ether, MIBK, and hexanol were evaluated as the extraction solvents for HF-LPME. Figure 2 shows the effect of organic solvent on the extraction efficiency of two CPAs. As shown in the figure, n-octanol is the most suitable extraction solvent as it resulted in an increased response of the analytical instrument. Thus, n-octanol was selected as the most suitable solvent for subsequent studies. 3.1.2 Effect of sample pH To extract the CPAs as weak acids into the organic phase, acidification of the sample solution is preferred to keep the targeted compounds in the molecular form. The effect of pH in the range of 1.0–5.0 on the extraction efficiency was investigated. The variations in extraction efficiency of CPAs versus pH are shown in Fig. 3. The results indicated that the extraction recovery decreased by increasing the pH from 1.0 to 5.0. The pKa values of the studied CPAs were 2.98 and 3.14. Theoretically, a pH value of 1.0 for the donor phase (equal to pKa 5 2) would be sufficiently acidic. Therefore, the pH value of 1.0 was chosen as the optimum value of pH for extraction. 3.1.3 Effect of temperature In liquid–liquid extraction, temperature has an influence on both equilibrium and mass transfer. In the present study, dynamic processes facilitate mass transfer of the analytes from the sample to the organic solvent and thus increase the extraction efficiency. Effect of temperature on extraction

J. Sep. Sci. 2011, 34, 957–964

efficiency was studied in the range of 25–601C. A water batch was applied to adjust the temperature of sample solution. The experimental results showed that extraction efficiency increases by raising the sample’s temperature. By increasing the temperature, viscosity of organic phase decreases, whereas the rate of mass transfer and extraction efficiency increases. Thus, the temperature of solution was adjusted at 601C for further experiments. 3.1.4 Effect of sample flow rate A peristaltic pump was applied to pump the water sample into the TT-extractor. Different sample flow rates in the range of 20–50 mL/min were examined. Extraction efficiency was increased by increasing the extraction contact area. For the hollow fiber and the extractor tube used in this study, outside volume of the fiber was 78.5 mm3 and the inner volume of extractor tube was 706.5 mm3, giving a sample volume of 0.628 mL between fiber and inner wall of extractor. The extractive contact area, which is the contact area between sample and outside wall of fiber, was 314 mm2. The quotient between the sample volume passing through the extractor in extraction time of 30 min at different donor flow rates and sample volume of the inner extractor (0.628 mL) multiplied by extractive contact area (314 mm2) determined the total extractive area. As the flow rate increased from 20 to 50 mL/min, the extractive area raised from about 300 000 to 750 000 mm2 (Fig. 4A). The experimental results supported these explanations. As shown in Fig. 4B, extraction recovery was increased by increasing the sample flow rate. This in turn led to an increase in pressure of the generated donor phase on hollow fiber, and thus the aqueous phase penetrated into the hollow fiber. Therefore, 50 mL/min was selected as the optimum flow rate.

Figure 3. Effect of pH of donor phase on the extraction efficiency of target analytes. The conditions are the same as described in Fig. 2, except extraction solvent 5 n-octanol and variable pH.

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.jss-journal.com

J. Sep. Sci. 2011, 34, 957–964

Sample Preparation

961

Figure 4. Effect of the flow rate of sample solution on the extraction efficiency of CPAs. The conditions are the same as described in Fig. 2, except temperature 5 601C and variable sample flow rate.

3.1.5 Effect of salt addition Addition of NaCl to the sample solution may result in several effects on LPME. It can decrease the solubility of analytes in the aqueous sample and enhance their partitioning into the organic phase. On the other hand, salt can change the physical properties of the Nernst diffusion film and thus reduce the rate of diffusion of analytes into the organic phase, and therefore decreasing the extraction efficiency of the organic phase within a prescribed time. In the present study, as shown in Supporting Information Fig. S1, extraction efficiency was decreased by addition of sodium chloride. Therefore, further experiments were performed in the absence of sodium chloride.

in the range of 10–60 min. The amounts of extracted analytes were increased dramatically by increasing exposure time from 10 to 30 min; no further substantial increase was obtained by increasing the extraction time (Supporting Information Fig. S2). HF-LPME is an equilibrium process rather than an exhaustive extraction. It requires a period of time for equilibrium to be established. Nonetheless, it is not normally practical to use extraction times as long as equilibrium to be established. On the other hand, if the extraction time is long, solvent loss and formation of air bubbles may occur, which would compromise the extraction efficiency. On the basis of these results, 30 min was selected as the extraction time. 3.1.7 Effect of sample volume

3.1.6 Effect of extraction time Mass transfer is a time-dependent process. Hence, the effect of extraction time on extraction efficiency was investigated

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

The influence of sample volume on the extraction efficiency was studied in the range of 20–80 mL, where flow rate and extraction time were kept constant. The amount of analytes

www.jss-journal.com

962

J. Sep. Sci. 2011, 34, 957–964

A. Esrafili et al.

increases by increasing the sample volume. In addition, due to constant extractive area for different sample volumes at constant flow rate and extraction time, the extraction recoveries changed less than expected. Therefore, a volume of 20 mL was chosen as the optimum sample volume. 3.1.8 Effect of dwell time In the dynamic process, extraction is performed by automatically manipulating the plunger in and out of the microsyringe barrel. Dwelling time is described as the time between refilling and infusing of the organic extractant solvent within the lumen of hollow fiber, which is an important factor for the repeated plunger movement. The shorter the dwelling time, the higher the frequency of the plunger movement, and this in turn allows a greater number of extraction cycles. Higher frequency of the plunger movement was beneficial to mass transfer; however, short dwelling time may result in less contact time between the organic phase in lumen and the organic extractant layer in pores of hollow fiber. To investigate the effect of dwelling time on extraction efficiency, the plunger speed was kept at 3 mL/s and the dwelling time was varied in the range of 1–5 min. As shown in Fig. 5, the results indicated that when the dwelling time varied from 0 to 2 min, extraction efficiency increased, and longer dwelling times resulted in a decrease in the extraction efficiency. For practical reasons, 2 min was chosen as the dwelling time for the rest of the study.

of the analytes in water. Performance of this method under the optimum conditions is summarized in Table 1. The calibration curve was linear in the range of 1–100 mg/L with coefficient of determination (r2) >0.994. The relative standard deviation (RSD) for extraction and determination of the analytes was less than 7.2% based on five replicate measurements. LODs of 0.5 mg/L were obtained, based on S/N 5 3 for both of the analytes. The preconcentration factor (PF), defined as the ratio of peak areas after extraction and those before extraction, was used to evaluate the extraction efficiency. It should be noted that PFs of 219and 230-fold were obtained for the extraction of MCPA and 2,4-D at a concentration of 50 mg/L from the aqueous samples, respectively. This new dynamic technique gave good repeatability due to automated design for regular pumping of donor phase in accurate flow rate by a peristaltic pump and also the syringe plunger movement by the programmable syringe pump which provided sufficient accuracy in controlling the plunger movement and dwelling time. The comparison between the figures of merit of the proposed method and some of the published methods for the extraction and determination of CPAs are summarized in Table 2. Clearly, the proposed method has a good sensitivity and precision with a suitable dynamic linear range. Also, the obtained LODs for the analytes by the present method at an extraction time of 30 min are Table 1. Figures of merit of the proposed dynamic HF-LPME method

3.2 Method evaluation

Analyte

Linear dynamic range (mg/L)

r2

LOD (mg/L)

RSD % (n 5 5)

PF

To evaluate the practical applicability of the proposed method, analytical quality parameters (i.e. linearity, repeatability, and LOD) were investigated using standard solution

2,4-D MCPA

1–100 1–100

0.994 0.995

0.5 0.5

7.2 6.4

229 219

Figure 5. Effect of dwell time on the extraction efficiency of CPAs. The conditions are the same as described in Supporting Information Fig. S2, except time 5 30 min and variable dwell time.

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.jss-journal.com

Sample Preparation

J. Sep. Sci. 2011, 34, 957–964

963

Table 2. Comparison of the proposed method with other methods applied for the extraction and determination of CPAs Method

DLR (mg/L)

LOD (mg/L)

RSD %

PF

Extraction time

Ref.

DLLME-HPLC HF-LLLME-HPLC HF-LPME-HPLC Dyanmic HF-LPME-HPLC

0.1–400 1–200 1–125 1–100

0.2 0.5 0.3–0.4 0.5

4.2 4.75 o12 o7.2

689 400 219–230

60 min 60 min 4h 30 min

[42] [43] [44] Proposed study

Table 3. Results obtained for the analysis of real samples Sample

Tap water Well water River waterb) River waterc)

2,4-D

MCPA

Added

Found7SDa)

Recovery (%)

Added

Found7SD

Recovery (%)

5.0 5.0 5.0 5.0

4.7970.3 4.6370.4 4.7870.5 4.5470.4

95.8 92.6 95.6 90.8

5.0 5.0 5.0 5.0

4.8070.2 4.6770.4 4.7470.4 4.6170.3

96.0 93.4 94.8 92.2

a) SD based on three replicates. b) Sampling from Tajan River (Mazandaran, Iran). c) Sampling from Zardi River (Mazandaran, Iran).

relative recoveries of the CPAs from waters were in the range of 92.3–96.1% (Table 3). The results demonstrated that this method is applicable to real aqueous samples. Figure 6 shows the typical chromatograms of the extracted CPAs from river water sample 1 before and after spiking with 5 mg/L of CPAs.

4 Concluding remarks Figure 6. HPLC chromatograms of the (A) nonspiked and (B) 5 mg/L of MCPA and 2,4-D spiked in river water sample.

comparable with those obtained by other methods. The main advantages of the proposed method are decreasing sample handling due to its high automation degree and the potential of coupling with autoinjection system of HPLC. By coupling the proposed technique with autoinjection system of HPLC, a full automatic technique for the extraction and determination of different pollutants from water samples will be obtained.

3.3 Application of the proposed method to real samples In order to demonstrate the applicability of the proposed method to real samples, the procedure was applied to preconcentrate the CPAs in different water samples. Two aqueous samples were collected from tap and a well in Tarbiat Modares University (Tehran, Iran). Two river water samples were obtained from Tajan and Zardi (Mazandaran, Iran) rivers. CPAs were not found in water samples. Thus, the water samples were spiked with 5 mg/L of CPAs. The

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

In the present study, the dynamic HF-LPME procedure combined with HPLC-UV was developed for the analysis of trace levels of CPAs in water samples. The proposed method has analytical characteristics similar to or better than the previous static extraction methods (it has suitable linearity and LODs, typical PFs up to 230). Also, dynamic hollow fiber liquid-phase microextraction has provided better extraction efficiency and improved reproducibility as compared with the static modes. The main advantages of the proposed method are decreasing sample handling due to its high automatization degree, as well as drastically reducing the extraction time. The dynamic HF-LPME as a simple, economic, sensitive, and promising sample preparation technique provided an alternative method for the determination of trace levels of CPAs in water samples. The authors have declared no conflict of interest.

5 References [1] Wells, M. J. M., Yu, L. Z., J. Chromatogr. A 2000, 885, 237–250.

www.jss-journal.com

964

A. Esrafili et al.

J. Sep. Sci. 2011, 34, 957–964

[2] Macutkiewicz, E., Rompa, M., Zygmunt, B., Crit. Rev. Anal. Chem. 2003, 33, 1–17.

[22] Hogendoorn, E. A., Dijkman, E., Baumann, B., Anal. Chem. 1999, 71, 1111–1118.

[3] Rompa, M., Kremer, E., Zygmunt, B., Anal. Bioanal. Chem. 2003, 377, 590–599.

[23] Zhu, L., Ee, K. H., Zhao, L., Lee, H. K., J. Chromatogr. A 2002, 963, 335–343.

[4] Boucharat, C., Desauziers, V., Cloirec, P. L., Talanta 1998, 47, 311–323.

[24] Takino, M., Daishima, S., Nakahara, T., Analyst 2001, 126, 602–608.

[5] Li, N., Lee, H. K., Anal. Chem. 2000, 72, 3077–3084.

[25] Balinova, A., J. Chromatogr. A 1996, 728, 319–324.

[6] David, M. D., Campbell, S., Li, Q. X., Anal. Chem. 2000, 72, 3665–3670.

[26] Wu, J., Ee, K. H., Lee, H. K., J. Chromatogr. A 2005, 1082, 121–127.

[7] Rompa, M., Kremer, E., Zygmunt, B., Anal. Lett. 2004, 37, 3299–3312.

[27] Geerdink, R. B., Niessen, W. M. A., Brinkman, U. A. Th., J. Chromatogr. A 2002, 970, 65–93.

[8] Rodrı´guez, I., Rubı´, E., Gonza´lez, R., Quintana, J. B., Cela, R., Anal. Chim. Acta 2005, 537, 259–266.

[28] Cserha´ti, T., Forga´s, E., J. Chromatogr. B 1998, 717, 157–178.

[9] Campbell, S., Li, Q. X., Anal. Chim. Acta 2001, 434, 283–289.

[29] Aprea, C., Colosio, C., Mammone, T., Minoia, C., Maroni, M., J. Chromatogr. B 2002, 769, 191–219.

[10] Lee, M. R., Lee, R. J., Lin, Y. W., Chen, C. M., Hwang, B. H., Anal. Chem. 1998, 70, 1963–1968.

[30] Krutz, L. J., Senseman, S. A., Sciumbato, A. S., J. Chromatogr. A 2003, 999, 103–121.

[11] Rodrı´guez Pereiro, I., Gonza´lez Iromia, R., Rubı´ Cano, E., Cela Torrijos, R., Anal. Chim. Acta 2004, 524, 249–256.

[31] Catalina, M. I., Dalluge, J., Vreuls, J. J., Brinkman, U. A. Th., J. Chromatogr. A 2000, 877, 153–166.

[12] Santos-Delgado, M. J., Crespo-Corral, E., Polo-Dı´ez, L. M., Talanta 2000, 53, 367–377.

[32] Suffet, I. H., J. Agric. Food Chem. 1973, 21, 591–598.

[13] Field, J. A., Monohan, K., Anal. Chem. 1995, 67, 3357–3362. [14] Chatfield, S. N., Croft, M. Y., Dang, T., Murby, E. J., Yu, G. Y. F., Wells, R. J., Anal. Chem. 1995, 67, 945–951. [15] Vink, M., van der Poll, J. M., J. Chromatogr. A 1996, 733, 361–366.

[33] Jeannot, M. A., Cantwell, F. F., Anal. Chem. 1996, 68, 2236–2240. [34] He, Y., Lee, H. K., Anal. Chem. 1997, 69, 4634–4640. [35] Wang, Y., Kwok, Y. C., He, Y., Lee, H. K., Anal. Chem. 1998, 70, 4610–4614. [36] Shen, G., Lee, H. K., Anal. Chem. 2002, 74, 648–654. [37] Zhao, L., Lee, H. K., Anal. Chem. 2002, 74, 2486–2492.

[16] Nilsson, T., Baglio, D., Galdo-Miguez, I., Madsen, J. Ø., Facchetti, S., J. Chromatogr. A 1998, 826, 211–216.

[38] Ouyang, G., Zhao, W., Pawliszyn, J., J. Chromatogr. A 2007, 1138, 47–54.

[17] Henriksen, T., Svensmark, B., Lindhardt, B., Juhler, R. K., Chemosphere 2001, 44, 1531–1539.

[39] Li, G., Zhang, L., Zhang, Z., J. Chromatogr. A 2008, 1204, 119–122.

[18] Dijkman, E., Mooibroek, D., Hoogerbrugge, R., Hogendoorn, E., Sancho, J. V., Pozo, O., Herna´ndez, F., J. Chromatogr. A 2001, 926, 113–125. [19] Bogialli, S., Curini, R., Di Corcia, A., Lagana´, A., Stabile, A., Sturchio, E., J. Chromatogr. A 2006, 1102, 1–10. [20] Freitas, L. G., Go¨tz, C. W., Ruff, M., Singer, H. P., Mu¨ller, S. R., J. Chromatogr. A 2004, 1028, 277–286. [21] Crescenzi, C., D’Ascenzo, G., Di Corcia, A., Nazzari, M., Marchese, S., Samperi, R., Anal. Chem. 1999, 71, 2157–2163.

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

[40] Ouyang, G., Pawliszyn, J., Anal. Chem. 2006, 78, 5783–5788. [41] Pezo, D., Salafranca, J., Nerin, C., J. Chromatogr. A 2007, 1174, 85–94. [42] Tsai, W. C., Huang, S. D., J. Chromatogr. A 2009, 1216, 7846–7850. [43] Zhu, L., Ee, K. H., Zhao, L., Lee, H. K., J. Chromatogr. A 2002, 963, 335–343. [44] Liu, J. F., Tora¨ng, L., Mayer, P., Jo¨nsson, J. A., J. Chromatogr. A 2007, 1160, 56–63.

www.jss-journal.com