Journal of Chromatography A

Transcription

1 Journal of Chromatography A, 1216 (2009) Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: Determination of triclosan, triclocarban and methyl-triclosan in aqueous samples by dispersive liquid liquid microextraction combined with rapid liquid chromatography Jie-Hong Guo a,b, Xing-Hong Li a,, Xue-Li Cao b, Yan Li a,c, Xi-Zhi Wang a, Xiao-Bai Xu a a State Key Laboratory of Environmental Chemistry and Eco-toxicology, Research Center of Eco-Environment Sciences, Chinese Academy of Sciences, P.O. Box 2871, Beijing , China b Beijing Technology and Business University, Beijing , China c Hebei University, Shijiazhuang , China article info abstract Article history: Received 29 October 2008 Received in revised form 2 February 2009 Accepted 9 February 2009 Available online 13 February 2009 Keywords: DLLME UHPLC Pharmaceuticals and personal care products Aqueous sample In this study, dispersive liquid liquid microextraction (DLLME) combined with ultra-high-pressure liquid chromatography (UHPLC) tunable ultraviolet detection (TUV), has been developed for pre-concentration and determination of triclosan (TCS), triclocarban (TCC) and methyl-triclosan (M-TCS) in aqueous samples. The key factors, including the kind and volume of extraction solvent and dispersive solvent, extraction time, salt effect and ph, which probably affect the extraction efficiencies were examined and optimized. Under the optimum conditions, linearity of the method was observed in the range of gl 1 for TCS, gl 1 for TCC, and gl 1 for M-TCS, respectively, with correlation coefficients (r 2 ) > The limits of detection (LODs) ranged from 45.1 to 236 ng L 1. TCS in domestic waters was detected with the concentration of 2.08 gl 1. The spiked recoveries of three target compounds in river water, irrigating water, reclaimed water and domestic water samples were achieved in the range of %, %, % and %, respectively. As a result, this method can be successfully applied for the rapid and convenient determination of TCS, TCC and M-TCS in real water samples Elsevier B.V. All rights reserved. 1. Introduction Triclosan (5-chloro-2-(2,4-dichloro-phenoxy)-phenol, TCS) and triclocarban (N-(4-chlorophenyl)-N-(3,4-dichlorophenyl) urea, TCC) were widely used as antimicrobial agents in consumer products such as shampoos, soaps, creams, mouthwash and toothpaste [1]. However, TCS is found to be acutely toxic to some aquatic organisms, particularly certain algae species, at low gl 1 levels and was recently shown to modulate thyroid function in amphibians at concentrations as low as 0.15 gl 1 [2]. In addition, it has also been shown to photo-transform into members of the dioxin family, which is known as the most carcinogenic chemicals in the world [3,4]. Although relatively few data exist about the toxicity of triclocarban, it has been found to impair reproduction in laboratory rats and that some of its degradation products are carcinogenic. Methyl-triclosan (M-TCS), a metabolite of TCS, is more lipophilic and environmentally persistent [2,5], suggesting its relatively high bioaccumulation potential in aquatic organisms [5,6]. Corresponding author. Tel.: ; fax: address: lxhzpb@rcees.ac.cn (X.-H. Li). It has been reported that TCS is one of the most frequently detected organic pollutants in rivers and lakes [2,5,7], and both TCC and M-TCS are also co-contaminants with TCS in waste water and surface water [5,8]. Since the concentrations of these compounds are normally low ( gl 1 or less) in aquatic environment, extraction/pretreatment procedures and high sensitivity instruments are necessary for their final determination. Their general pretreatment methods reported include solid-phase extraction (SPE) [7], solid-phase microextraction (SPME) [9], hollow fiber assisted liquid-phase microextraction (HF-LPME) [10] and stir-bar sorptive extraction (SBSE) [11]. However, manual operation of SPE is tedious and time consuming, although it requires small volumes of organic solvents. SPME has drawbacks of high cost, sample carry-over, and a decline in performance with time. LPME produces some obvious disadvantages: solvent drops are ready to break, air bubble is tending to form, time is long and sometimes equilibrium is not achieved easily in short time [12]. Recently, Assadi [13] has introduced a simple and rapid pre-concentration method, named dispersive liquid liquid microextraction (DLLME), in which analytes in aquatic samples were extracted by a cloudy solution formed by an appropriate mixture of extraction and dispersive solvents, then extraction solvent was transferred after centrifugation and compounds in the /$ see front matter 2009 Elsevier B.V. All rights reserved. doi: /j.chroma

2 J.-H. Guo et al. / J. Chromatogr. A 1216 (2009) extraction solvent were determined by gas chromatography (GC) or liquid chromatography (LC). Due to such advantages as rapidity, simplicity of operation, little time consumption, low cost and high recovery, DLLME has been successfully applied for the determination of clenbuterol [12], polycyclic aromatic hydrocarbons [13], polar organic compounds [14], chlorophenols [15], polybrominated diphenyl ethers [16], phthalate esters [17,18], triazine herbicides [19], plastic additives [20], organophosphorus pesticides [21] and organophosphate esters [22] in water samples in recent years. In DLLME, dispersive solvent plays a key role which helps extraction solvent form fine droplets in aqueous samples, representing about 97 99% of the total volume of the extraction mixture [22]. Comparing to other methods, abundant contact surface of fine droplets and analyte speeds up mass transferring processes of analytes from aquatic phase to organic phase, which not only greatly enhances the extraction efficiency but also overcomes the time-consuming problem [14,22]. Acetone, methanol, ethanol and acetonitrile have generally been used as dispersive solvents for low toxicity and cost [11,18,22,23]. However, they could apply to limited extraction solvents to form constant and large volume of sedimented phase when the dosage of extraction solvent is low [20]. Though more costly and noxious than other dispersive solvents, tetrahydrofuran (THF) could constitute larger settled volume, which could provide convenient operation and reduce the volume requirement of toxic chlorinated extraction solvents [14,20,24]. It seemed that THF as dispersive solvent has more advantages in DLLME. In general, analytical instruments for the determination of TCS, TCC or M-TCS in water samples have been used, including GC/MS (mass spectrum) [5,7,11], GC/MS/MS [25], LC/MS [5,7] and LC/MS/MS [2,26], which required long time to run and high-cost detectors such as MS and MS/MS. Recently, ultra-high-pressure liquid chromatography (UHPLC) has proven to be one of the most promising developments in the area of high-speed chromatographic separations with quite short time, sensitivity three times of ordinary HPLC (see more of your sample) and resolution with peak capacities twice as high as ordinary HPLC (see all of your samples) and separations could be nine times faster with equal resolution [27]. Therefore, UHPLC tunable ultraviolet detection (TUV) combined with DLLME was expected to be a highly sensitive and fast method to evaluate the potential risks posed by TCS, TCC and M-TCS in aquatic environment. The aim of this study is to examine the DLLME UHPLC TUV suitability for the determination of TCS, TCC and M-TCS simultaneously in water samples. The effect of various experiment conditions on the extraction of TCC, TCS and M-TCS is described and discussed in detail. In the end, the recommended method was employed to investigate the levels of the target compounds in real water samples. 2. Experimental 2.1. Materials and reagents Triclosan (purity, 99.5%), triclocarban (purity, 99.5%) and methyl-triclosan (purity, 99%) were purchased from Dr. Ehrenstorfer (Augsburg, Germany). Acetonitrile (HPLC grade) and tetrahydrofuran (THF, HPLC grade) were purchased from Tedia (Fairfield, OH, USA) while methanol (HPLC grade) was bought from Fisher (Pittsburgh, PA, USA) and acetone (Beijing Chemical Reagents Company, Beijing, China) was of analytical grade but redistilled with rectifying tower. Tetrachloroethylene (C 2 Cl 4 ), 1,4-dichlorobutane (C 4 H 8 Cl 2 ), trichloroethylene (C 2 HCl 3 ), 1,3- dichlorobenzene (C 6 H 4 Cl 2 ), 1,3-dichloropropane (C 3 H 6 Cl 2 ) were of chromatographic grade and produced by Shanghai Chemical Reagent Factory One (Shanghai, China). Water was purified by Milli- Q water purification system (Millipore, Bedford, MA, USA). Buffer solution (ph 9) was composed of boracic acid (Beijing Chemical Reagents Company, Beijing, China) g, potassium chloride (Sinopharm Chemical Reagent Co., Ltd, Shanghai, China) g and sodium hydroxide (Beijing Chemical Reagents Company, Beijing, China) g, which were dissolved in 140 ml purified water. Before used as mobile phase, solution above was diluted 20 times Standard solution The individual stock solution (1.00 g L 1 ) of TCS, TCC and M- TCS was prepared by dissolving the solid standard substance in methanol, respectively. Then, six levels of mixed standards ( mg L 1 TCS and M-TCS and mg L 1 TCC, mg L 1 TCS and M-TCS and mg L 1 TCC, mg L 1 TCS and M-TCS and mg L 1 TCC, 1.00 mg L 1 TCS and M-TCS and mg L 1 TCC, 10.0 mg L 1 TCS and M-TCS and 5.00 mg L 1 TCC, 100mgL 1 TCS and M-TCS and 50.0 mg L 1 TCC) were prepared further by diluting the stock solution with methanol Instrumentation TCS, TCC and M-TCS were analyzed by Aquity Ultra Performance Liquid Chromatography (Waters, Milford, MA, USA), consisting of a binary solvent manager and a sample manager, was coupled to a tunable UV detector. UHPLC analyses were performed on a bridged ethylene hybrid (BEH) C 18 analytical column (50 mm 2.1 mm, 1.7 m). Empower software was used for chromatographic data gathering and integration of chromatograms. A model CR22GII High-speed Refrigerated Centrifuge (Hitachi, Tokyo, Japan) was used for centrifugation and the super-mixer (Melrose, Park, IL, USA) was used to disperse the samples after the mixture of dispersive and extracting solvents was injected into the aqueous samples UHPLC conditions The mobile phase consisted of buffer solution (A), and acetonitrile (B) with the gradient elution as follows: 0 min, 50% A/50% B with a flow rate of 0.4 ml min 1, then 40% A/60% B with a flow rate of 0.5 ml min 1 at 1.0 min (Waters curve type 6), finally, reconditioning the column with 50% A/50% B after washing column with 90% B at the rate of 0.3 ml min 1 for 1.5 min. The total run time for analysis was 5 min. The peaks were detected at wavelength of 283 nm with TUV. 25 and 30 C were adopted as sample temperature and column temperature, respectively. The injection volume was 5.0 L Dispersive liquid liquid microextraction procedure The 5.00 ml water solution was placed into a 10-mL glass test tube with conical bottom. Mixture of THF (1.00 ml) as dispersive solvent and C 6 H 4 Cl 2 (15.0 L) as extraction solvent was rapidly injected into the aqueous solution to form a milky cloudy solution (water/thf/c 6 H 4 Cl 2 ). The milky cloudy mixture was centrifuged for 5.0 min at 3000 rpm, then, the dispersed fine particles were sedimented in the bottom of conical test tube, about 20.0 ± 1.0 L. The sedimented phase was then transferred to sample bottle by a microsyringe and blew until dryness L of methanol was used to dissolve the mixture left, 5.0 L of which was injected automatically for UHPLC analysis. All the steps were operated under ambient temperature.

3 3040 J.-H. Guo et al. / J. Chromatogr. A 1216 (2009) Results and discussion 3.1. Optimization of DLLME For DLLME, several parameters affecting the extraction performance, such as the kind and volume of extraction and dispersive solvents, extraction time, salt addition and ph, were tested, and discussed in detail in the following sections. Extraction recovery was used to assess the method optimized parameters as described by Li et al. [16]. The spiked concentrations of compounds in the water samples were 10.0 gl 1 for TCS and M-TCS, and 5.00 gl 1 for TCC Selection of dispersive solvent To investigate the significance of dispersive solvent, four composition modes of extraction solution were done including addition of only extraction solvent C 6 H 4 Cl 2, addition of extraction solvent C 6 H 4 Cl 2 followed by dispersive solvent THF, addition of dispersive solvent THF followed by extraction solvent C 6 H 4 Cl 2 and addition of mixture of extraction solvent C 6 H 4 Cl 2 and dispersive solvent THF. It was found that the best recoveries could be achieved when the mixed composition of C 6 H 4 Cl 2 and THF was spiked, while results were relatively worse without THF or sequentially spiking of C 6 H 4 Cl 2 and THF (see Fig. 1). This can be accounted for by the fact that the dispersive solvent THF in the mixed composition helps to disperse the extraction solvent C 6 H 4 Cl 2 to fine drops, which greatly enlarge the interface of extraction solvent with aqueous sample, resulting in high recoveries and the short equilibrium extraction time (discussed in Section 3.1.4). When the extraction solution was added with C 6 H 4 Cl 2 alone, the extraction solvent sedimented down fast and the interface with water sample was limited, so was the efficiency. And when extraction solvent and dispersive solvent were added separately, fine drops were not observed, but shaking rendered dispersive solvent play certain role to disperse the extraction solvent. It summed that dispersive solvent played a key role in the aspect of improving extraction efficiency, suggesting that it was important experimental process to select appropriate dispersive solvent in DLLME. As for the choice of dispersive solvent in DLLME, the miscibility in organic phase (extraction solvent) and aqueous phase (sample Table 1 Extraction recovery and standard deviation of different dispersive solvents evaluated for extraction of TCC, TCS and M-TCS by DLLME a. Extraction recovery ± SD (%) Acetonitrile Methanol Acetone THF TCS 88.5 ± 2.11 b 75.3 ± ± ± 1.65 TCC 97.3 ± ± ± ± 2.45 M-TCS 81.8 ± ± ± ± 7.11 a Extraction conditions: water sample volume, 5.00 ml; disperser solvent (methanol, tetrahydrofuran, acetonitrile and acetone) volume, 1.00 ml; extraction solvent (C 6H 4Cl 2) volume, 25.0 L; sedimented phase volume, acetonitrile 23.0 L, methanol 22.0 L, acetone 22.5 L and THF 35.0 L; concentration of target compounds, 10.0 gl 1 for TCS and M-TCS, and 5.00 gl 1 for TCC. b Standard deviation, n =3. solution) is a key factor, which can disperse extraction solvent into very fine droplets in aqueous phase [28]. Based on the consideration, acetonitrile, methanol, acetone and THF were selected due to their property. The experimental condition was to inject the extraction solution containing 1.00 ml different dispersive solvents and 25.0 L C 6 H 4 Cl 2 into water sample (other parameters seen in Table 1). Of the four solvents examined, all showed satisfactory recoveries and standard deviations as shown in Table 1. However, due to the different solubilities in two phases, acetonitrile, methanol, acetone and THF achieved 23.0, 22.0, 22.5 and 35.0 L volume of sedimented phase, respectively. Since relative large volume obtained was more convenient for operation and had less chance for accidental error [14], THF was selected as dispersive solvent for the following study Kind and volume of extraction solvent It is very important to select an appropriate extraction solvent for obtaining good extraction recoveries in DLLME method. As a proper extraction solvent, some primary requirements must be met: higher density than water, low water solubility and high extraction capability of interested compounds [29]. In accordance with the above requirements, tetrachloroethylene (C 2 Cl 4, = 1.63), 1,4-dichlorobutane (C 4 H 8 Cl 2, = 1.16), trichloroethylene (C 2 HCl 3, = 1.46), 1,3-dichlorobenzene (C 6 H 4 Cl 2, = 1.29), 1,3- dichloropropane (C 3 H 6 Cl 2, = 1.19) were selected and tested. Each solvent was evaluated through comparing the extraction recoveries of these compounds in the extraction of a 5.00 ml water sample with mixture of 25.0 L different extraction solvents and 1.00 ml THF as dispersive solvent at ambient temperature. The milky cloudy solution was centrifuged with 3000 rpm for 5 min in 4 C. Average recovery (n = 3) and standard deviation (SD) obtained for different extraction solvents were shown in Table 2, which revealed that all the five extract solvents achieved satisfactory extraction recoveries (77 106%) and standard deviation (below 16%). However, volumes of sedimented phase of these extraction solvents differed dramatically, which ranged from 27.0, 30.0, 22.0, 35.0 to Table 2 Extraction recovery and standard deviation of different extraction solvents evaluated for extraction of TCC, TCS and M-TCS by DLLME a. Extraction recovery ± SD (%) C 2Cl 4 C 4H 8Cl 2 C 2HCl 3 C 6H 4Cl 2 C 3H 6Cl 2 TCS 83.0 ± 9.10 b 77.2 ± ± ± ± 12.3 TCC 85.1 ± ± ± ± ± 8.05 M-TCS 96.3 ± ± ± ± ± 10.2 Fig. 1. Effect of the spiked modes of dispersive solvent on the recovery of TCS, TCC and M-TCS by DLLME. Extraction conditions: water sample volume, 5.00 ml; dispersive solvent (THF) volume, 1.00 ml; volume of C 6H 4Cl 2, 15.0 L; room temperature; concentration of each compounds, 10.0 gl 1 for TCS and M-TCS, and 5.00 gl 1 for TCC. a Extraction conditions: water sample volume, 5.00 ml; disperser solvent (THF) volume, 1.00 ml; extraction solvent volumes, 25.0 L; sedimented phase volume, C 2Cl L, C 4H 8Cl L, C 2HCl L, C 6H 4Cl L, C 3H 6Cl L; concentration of target compounds, 10.0 gl 1 for TCS and M-TCS, and 5.00 gl 1 for TCC. b Standard deviation, n =3.

4 J.-H. Guo et al. / J. Chromatogr. A 1216 (2009) Fig. 2. Effect of the volume of C 6H 4Cl 2 on the recovery of TCS, TCC and M-TCS obtained from DLLME. Extraction conditions: water sample volume, 5.00 ml; dispersive solvent (THF) volume, 1.00 ml; room temperature; concentration of each compound, 10.0 gl 1 for TCS and M-TCS, and 5.00 gl 1 for TCC. Fig. 3. Effect of the volume of THF on the recovery of TCS, TCC and M-TCS by DLLME. Extraction conditions: water sample volume, 5.00 ml; sedimented phase volume, 20.0 ± 1.0 L; room temperature; concentration of each compound, 10.0 gl 1 for TCS and M-TCS, and 5.00 gl 1 for TCC L in accordance with C 2 Cl 4,C 4 H 8 Cl 2,C 2 HCl 3,C 6 H 4 Cl 2 and C 3 H 6 Cl 2. Due to its relatively low volatility, C 6 H 4 Cl 2 could maintain a relative stable sedimented phase. Besides, its achieved high sedimented phase renders reducing the dosage of C 6 H 4 Cl 2 feasible which was examined as below. Thus, C 6 H 4 Cl 2 was selected as extraction solvent. In our study, 10.0, 15.0, 20.0, 25.0, 30.0 and 35.0 LC 6 H 4 Cl 2 were investigated while keeping other conditions mentioned above constant. Fig. 2 presents the curve of recovery of target compounds while using different volumes of extraction solvent. Though the volumes of sedimented phase increased from 13.5 to 63.0 L with the ascending volume of C 6 H 4 Cl 2, the recoveries kept satisfactory when the volume of C 6 H 4 Cl 2 was higher than 15.0 L. On this basis, 15.0 LC 6 H 4 Cl 2 was used to study the performance of DLLME because it enabled the good recoveries for three target compounds with the least extraction organic solvent consumption Volume of dispersive solvent Dispersive solvent volume is important to make extraction solvent form very fine droplets, which has direct effect on the extraction efficiencies. However, since the variation of dispersive solvent volume would result in the change of the volume of sedimented phase, it is impossible to assess the effect of the volume of dispersive solvent on the extraction efficiency [13]. Hence, it is necessary to spontaneously adjust the volume of dispersive solvent and extraction solvent in order to obtain constant volume of sedimented phase. In our study, when the volume of THF was changed from 0.5, 1.0, 1.5 to 2.0 ml, sedimented phase (20.0 ± 1.0 L) was kept constant by adjusting the volume of C 6 H 4 Cl 2 from 19.0, 15.0, 13.0 to 13.0 L. According to the result (Fig. 3), the recoveries of three compounds were satisfactory with the exception of the combination of 0.5 ml THF and 19.0 LC 6 H 4 Cl 2. That is because 0.5 ml THF cannot disperse extraction solvent properly, so cloudy solution is not formed completely, which resulted in the relative low recoveries. Accordingly, 1.0 ml was selected in our experiment due to less volume use of THF. and extraction solvent) and before centrifugation. This experiment investigated the time influence from 1.00 to 60.0 min with the optimized operation parameters described above. As shown in Fig. 4, extraction time has little influence on the extraction recoveries in DLLME. This could be explained that the formation of cloudy solution causes the infinitely large interface of extraction solvent and aqueous phase, which results in the rapid transfer of analytes from aqueous phase to organic phase and instantaneous establishment of extraction equilibrium. Thus, centrifugation procedure could be done immediately after forming milky solution, an important element of DLLME for rapid determination of pollutants Ionic strength The presence of salt probably has different influences on extraction efficiency and enrichment factor of target compounds in the DMLLE [13,15]. In our study, the effect of sodium chloride with the percentage of 0%, 2.0%, 4.0%, 6.0%, 8.0% and 10% (w/v), was investigated. Plot of extraction recovery versus ionic strength is Effect of extraction time In DLLME, extraction time is defined as the interval time after injecting the extraction solution (mixture of dispersive solvent Fig. 4. Effect of extraction time on the recovery of TCS, TCC and M-TCS by DLLME. Extraction conditions: asinfig. 1.

5 3042 J.-H. Guo et al. / J. Chromatogr. A 1216 (2009) Table 3 Main method parameters of DLLME a. Compounds Linear range ( gl 1 ) r 2 RSD (%) (n = 7) LOD (ng L 1 ) TCS TCC M-TCS a Extraction conditions: water sample volume, 5.00 ml; disperser solvent (THF) volume, 1.00 ml; extraction solvent (C 6H 4Cl 2) volume, 15.0 L; sedimented phase volume, 20.0 ± 1.0 L; room temperature. Fig. 5. Effect of salt addition on the recovery of TCS, TCC and M-TCS by DLLME. Extraction conditions: asinfig. 1. shown in Fig. 5, which clearly indicates that the addition of NaCl has no significant effect on the extraction recoveries of target compounds. However, the enrichment factors decreased from 210, 210 and 234 to 104, 103 and 111 when the concentration of NaCl increased from 0%, 2.0%, 4.0%, 6.0%, 8.0% to 10%. In addition, addition of salt could increase the density of aqueous sample and renders the organic phase to suspend in the water phase, difficult to be drawn into microsyringe [30]. Thus, no addition of salt was selected for further discussion Effect of ph To examine the effect of ph, concentrated hydrochloric acid was used to regulate acidity (2 6) and sodium hydroxide to regulate alkalinity (8 14) [23]. Fig. 6 shows the curve of recoveries of TCS, TCC and M-TCS in response to ph 2 ph 14. The result indicates that the extraction efficiency of these three compounds maintains high from ph 2 to ph 8. However, under the condition of strong alkalinity, the recovery of TCC and M-TCS kept satisfactory while that of TCS dropped dramatically, which may relate with the structure characteristics of these compounds. For TCC and M-TCS, whose pk a are [1] and low than 13, the ionized forms of TCC and M-TCS changed to molecule forms when ph value was above 13, which are readily extracted to organic phase [20,31]. For TCS, above ph 8, the phenolate form predominates and can be rapidly photo-degraded in the presence of sunlight [2,32], which decreases the recovery of TCS. The phenomenon became more obvious as the increase of alkalinity. Based on the results above, neutrality was selected as extraction condition in our experiment. Finally, the optimized conditions were as follows: analytes were extracted by mixed solution of 1.00 ml THF and 15.0 L C 6 H 4 Cl 2 in a 5.00 ml water solution, and then solution was centrifuged at 3000 rpm for 5 min, and finally 20.0 L sedimented phase was drawn out for the following UHPLC analysis Evaluation of method performance Main method parameters The corresponding parameters for these analytes with DLLME UHPLC method were tested using spiked water samples under the optimum conditions as described above. Calibration curve was made at six different concentration levels and all yielded a good linearity (Table 3). As can be seen, linearity was observed in the range of gl 1 for TCS, gl 1 for TCC and gl 1 for M-TCS, respectively. The correlation coefficient (r 2 ) ranged from , to The limits of detection (LODs), based on signal-to-noise ratio (S/N) of 3, ranged from 45.1 to 236 ng L 1. The relative standard deviations varied between 4.60% and 6.01% with the levels of 100 gl 1 for TCS and M-TCS, and 50.0 gl 1 for TCC (n = 7) Real water samples analysis For the real environmental sample determination, several water samples including river water, irrigating water, reclaimed water and Fig. 6. Effect of ph on the recovery of TCS, TCC and M-TCS by DLLME. Extraction conditions: asinfig. 1. Fig. 7. Chromatograms of the blank domestic water (A) and domestic water spiked (B) at concentration levels of 2.00 gl 1 for TCS and M-TCS and 1.00 gl 1 for TCC obtained using DLLME combined with UHPLC TUV. Extraction conditions: as in Fig. 1.

6 J.-H. Guo et al. / J. Chromatogr. A 1216 (2009) Table 4 Amount of TCC, TCS and M-TCS detected originally, spiked recoveries (%) and relative standard deviations (RSD, %) in real aqueous samples a. Reclaimed water Irrigating water River water Domestical water Found ( gl 1 ) Recovery (%) RSD c (%) Found ( gl 1 ) Recovery (%) RSD (%) Found ( gl 1 ) Recovery (%) RSD (%) Found ( gl 1 ) Recovery (%) RSD (%) TCS ND b ND ND TCC ND ND ND ND m-tcs ND ND ND ND a Extraction conditions: water sample volume, 5.00 ml; disperser solvent (THF) volume, 1.00 ml; extraction solvent (C 6H 4Cl 2) volume, 15.0 L; sedimented phase volume, 20.0 ± 1.0 L; room temperature; added amounts of TCS, TCC and M-TCS were 2.00, 1.00 and 2.00 gl 1, respectively. b Not detected. c n =3. domestic water were selected for the determination of the three target compounds. River water was sampled from Qinghe, Beijing, China. Irrigating water was sampled from Research Center for Eco- Environment Science, Chinese Academy of Sciences, Beijing, China. Reclaimed water was sampled from middle water treatment plant of North China University of Technology, Beijing, China. Domestic water was sampled from the drainage in our lab after washing hands with hand lotion. All samples were filtered through 0.22 m filter membrane and stored at 4 C before being used. As a result, TCS was detected with a concentration of 2.08 gl 1 in domestic water, and no target compounds were found in the other water samples (see in Fig. 7 and Table 4). To evaluate the matrix effect, 2.00 gl 1 TCS and M-TCS, and 1.00 gl 1 TCC were spiked to the media above. As shown in Table 4, the spiked recoveries and relative standard deviation (RSD) were in the range of % and %, respectively. The amount of analytes extracted by DMLLE accorded well with the known values spiked in the medium. The good linearity and satisfactory recoveries showed that DLLME with UHPLC TUV was feasible for the determination of TCS, TCC and M-TCS in aqueous samples. 4. Conclusion In this study, a novel mode DLLME UHPLC TUV has been developed for the determination of trace amounts of TCS, TCC and M-TCS in aqueous samples, which showed wide linearity, good repeatability and satisfactory accuracy. The proposed method has many practical advantages, including small sample volume (5.00 ml), minimized consumption of toxic organic solvents (15.0 L) and sample preparation time (<1 min), high sensitivity and repeatability and convenient extraction procedure, all of which suggested that DLLME UHPLC TUV is an attractive technique for the preconcentration and rapid determination of TCS, TCC and M-TCS in aqueous samples. In addition to this application, DLLME is expected to extend to the pretreatment of other pharmaceuticals and personal care products in aqueous environment. Acknowledgements We acknowledge financial support of this work by the National High Technology Research and Development Program ( 863 Program) of China (122007AA061601) and the National Natural Science Foundation of China and the National Basic Research Program of China ( , ). References [1] G.G. Ying, R.S. Kookana, Environ. Int. 33 (2007) 199. [2] S.G. Chu, C.D. Metcalfe, J. Chromatogr. A 1164 (2007) 212. [3] R. Renner, Environ. Sci. Technol. 36 (2002) 230A. [4] E. Engelhaupt, Environ. Sci. Technol. 42 (2008) [5] M.A. Coogan, R.E. Edziyie, T.W. La Point, B.J. Venables, Chemosphere 67 (2007) [6] G.G. Ying, X.Y. Yu, R.S. Kookana, Environ. Pollut. 150 (2007) 300. [7] A. Agüera, A.R. Fernández-Alba, L. Piedra, M. Mézcua, M.J. Gómez, Anal. Chim. Acta 480 (2003) 193. [8] R.U. Halden, D.H. Paull, Environ. Sci. Technol. 39 (2005) [9] P. Canosa, I. Rodriguez, E. Rubí, R. Cela, J. Chromatogr. A 1072 (2005) 107. [10] R.S. Zhao, J.P. Yuan, H.F. Li, X. Wang, T. Jiang, J.M. Lin, Anal. Bioanal. Chem. 387 (2007) [11] M. Kawaguchi, R. Ito, H. Honda, N. Endo, N. Okanouchi, K. Saito, Y. Seto, H. Nakazawa, J. Chromatogr. A 1206 (2008) 196. [12] M.B. Melwanki, M.R. Fuh, J. Chromatogr. A 1198 (2008) 1. [13] M. Rezaee, Y. Assadi, M.R. Milani Hosseini, E. Aghaee, F. Ahmadi, S. Berijani, J. Chromatogr. A 1116 (2006) 1. [14] M.B. Melwanki, M.R. Fuh, J. Chromatogr. A 1207 (2008) 24. [15] N. Fattahi, Y. Assadi, M.R.M. Hosseini, E.Z. Jahromi, J. Chromatogr. A 1157 (2007) 23. [16] Y.Y. Li, G.H. Wei, J. Hu, X.J. Liu, X.N. Zhao, X.D. Wang, Anal. Chim. Acta 615 (2008) 96. [17] H. Farahani, P. Norouzi, R. Dinarvand, M.R. Ganjali, J. Chromatogr. A 1172 (2007) 105. [18] P. Liang, J. Xu, Q. Li, Anal. Chim. Acta 609 (2008) 53. [19] D. Nagaraju, S.D. Huang, J. Chromatogr. A 1161 (2007) 89. [20] M. Ali Farajzadeh, M. Bahrama, J.Å. Jönsson, Anal. Chim. Acta 591 (2007) 69. [21] S. Berijani, Y. Assadi, M. Anbia, M.R.M. Hosseini, E. Aghaee, J. Chromatogr. A 1123 (2006) 1. [22] M. García-López, I. Rodríguez, R. Cela, J. Chromatogr. A 1166 (2007) 9. [23] P. Liang, H.B. Sang, Anal. Biochem. 380 (2008) 21. [24] M.I. Leong, S.D. Huang, J. Chromatogr. A 1211 (2008) 8. [25] W. Boehmer, H. Ruedel, A. Wenzel, C. Schroeter-Kermani, Organohalogen Compd (2004) 66. [26] W.Y. Hua, E.R. Bennett, R.J. Letcher, Environ. Int. 31 (2005) 621. [27] J.,J.D. Sherma, F.H. Larkin, J. AOAC Int. 88 (2005) 63A. [28] H. Sereshti, M. Karimi, S. Samadi, J. Chromatogr. A 1216 (2009) 198. [29] H.X. Chen, H. Chen, J. Ying, J.L. Huang, L. Liao, Anal. Chim. Acta 632 (2009) 186. [30] M. Baghdadi, F. Shemirani, Anal. Chim. Acta 634 (2009) 186. [31] Y.C. Fan, Z.L. Hu, M.L. Chen, C.S. Tu, Y. Zhu, Chin. Chem. Lett. 19 (2008) 985. [32] B.E. Erickson, Environ. Sci. Technol. 36 (2002) 228A.