684 Geng Leng1 Wenjin Chen1 Mingfang Zhang1 Fang Huang1 Qiming Cao1,2 1 School
of Resources and Environment, University of Electronic Science and Technology of China, Chengdu, China 2 Griffith School of Engineering, Nathan Campus, Griffith University, Nathan, Australia Received September 18, 2013 Revised November 20, 2013 Accepted December 18, 2013
J. Sep. Sci. 2014, 37, 684–690
Research Article
Determination of phthalate esters in liquor samples by vortex-assisted surfactant-enhanced-emulsification liquid–liquid microextraction followed by GC–MS A novel method using vortex-assisted surfactant-enhanced-emulsification liquid–liquid microextraction has been developed for the extraction of phthalate esters (PAEs) in Chinese liquor samples prior to analysis by GC–MS. In the proposed method, a high-density extraction solvent (carbon tetrachloride) was dispersed into samples with the aid of a surfactant (Triton X-100) and vortex agitation, resulting in a short extraction equilibrium (30 s). After centrifugation, a single microdrop of solvent was easily collected for GC–MS analysis. Key factors that affected the extraction efficiency were optimized. Under the optimum conditions, linearity was found in the range from 0.05 to 50 g/L. Coefficients of determination varied from 0.9938 to 0.9971. LODs, based on an S/N of 3, ranged from 4.9 to 13 ng/L. Enrichment factors varied from 140 to 184. Reproducibility and recoveries were assessed by testing a series of three liquor samples that were spiked with different concentration levels. Finally, the proposed method was successfully applied to the determination of PAEs in 16 Chinese liquor samples. In this work, high-density-solvent vortex-assisted surfactantenhanced-emulsification liquid–liquid microextraction was applied for the first time for the extraction of PAEs in Chinese liquor samples and was proved to be simple, rapid, and sensitive. Keywords: GC-MS / Liquid-phase microextraction / Liquor / Phthalate esters / Surfactants DOI 10.1002/jssc.201301033
1 Introduction Phthalate esters (PAEs) are used primarily as plasticizers in polymeric materials to increase their flexibility and moldability. Some studies indicated that PAEs as well as their metabolites and degradation products can cause damage to the kidney, lung, endocrine, and reproductive system [1–4]. Because of their risks and abundant presence in many plastic mateCorrespondence: Prof. Qiming Cao, School of Resources and Environment, University of Electronic Science and Technology of China, 4 Jianshe North Road 2nd Section, Chenghua, Chengdu, Sichuan 610051, China E-mail: [email protected] Fax: +86–028–61831571
Abbreviations: BBP, butyl benzyl ester; CCl4 , carbon tetrachloride; CTAB, cethyltrimethyl ammonium bromide; DBP, dibutyl phthalate; DEHP, di-2-ethyl hexyl phthalate; DNOP, dioctyl phthalate; DLLME, dispersive liquid–liquid microextraction; EF, enrichment factor; ER, extraction recovery; LPME, liquid-phase microextraction; NaCl, sodium chloride; PAE, phthalate ester; SDME, single-drop microextraction; VALLME, vortex-assisted liquid–liquid microextraction; VSLLME, vortex-assisted surfactant-enhancedemulsification liquid–liquid microextraction C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
rials, there is a potential risk of PAEs contamination during liquor making. This may arise both from the raw materials and the use of plastic tools during processing. Moreover, additives and technological coadjuvants may also contribute to the health risk of PAEs. Consequently, PAEs contamination in liquor has now become a major public health concern in China, and the development of analytical methods for their determination is of considerable interest. However, due to the trace concentration levels of PAEs in liquor samples and the complexity of the liquor matrices, a sensitive and reliable pretreatment method for analysis of PAEs in liquor samples is imperative. Liquid-phase microextraction (LPME) represents an important development in the field of sample preparation for its simplicity, miniaturization, and time efficiency [5]. The first methodology evolved from LPME was single-drop microextraction (SDME), which was developed by Liu [6]. Since the introduction of SDME, different analytical modes, including liquid–liquid–liquid microextraction [7], continuous-flow microextraction [8], headspace SDME [9], hollow-fiber LPME [10] and solid-drop LPME [11] have been developed for various analytical applications. Colour Online: See the article online to veiw Figs. 1–5 in colour. www.jss-journal.com
Sample Preparation
J. Sep. Sci. 2014, 37, 684–690
Among all these analytical modes, a new type of LPME method termed as dispersive liquid–liquid microextraction (DLLME) has attracted more and more attention since its introduction by Rezaee et al. in 2006 [12]. In DLLME, an appropriate mixture of extraction and disperser solvent is rapidly injected into the aqueous sample. The extraction solvent is dispersed into the aqueous sample as very fine droplets and a cloudy solution is formed. The extension of the contact surface between the extraction solvent and aqueous phase can greatly enhance the extraction efficiency. However, the main drawback of DLLME is the necessity of using the dispersion solvent, which may decrease the partition coefficient of analytes into the extraction solvent [13]. In order to avoid the use of dispersion solvent, ultrasound was used to assist the dispersion of the extraction solvent into aqueous samples [14, 15]. However, degradation of the analyte occurred under ultrasound conditions has been reported [16]. More recently, Yiantzi et al. [17] introduced a novel LPME technique, the vortex-assisted liquid–liquid microextraction (VALLME). In VALLME, microvolumes of organic extraction solvent were dispersed into an aqueous sample by using vortex mixing. The fine droplets could rapidly extract analytes from the aqueous sample because of the shorter diffusion distance and larger specific surface area. After centrifugation, the extractant solvent restores its initial single microdrop shape and is ready for instrumental analysis. VALLME overcomes the main disadvantage of DLLME, and has been successfully applied for the extraction of different compounds since its introduction [17–23]. It is well known that surfactants are amphiphilic compounds containing both hydrophobic and hydrophilic groups, thus surfactants can reduce the interfacial tension between oil and water by adsorbing at the liquid–liquid interface to increase the dispersion [24]. So surfactants can be used to enhance the dispersion of extraction solvents into aqueous samples. Based on these considerations, using surfactants to assist extraction solvents to better disperse into the sample in VALLME was found to be a very advantageous idea [25]. To the best of our knowledge, only one application of vortex-assisted surfactant-enhanced-emulsification liquid– liquid microextraction (VSLLME) for the extraction of PAEs was reported [26]. In this study, low-density organic solvent was used as extractant solvent. However, it is very difficult to collect the single microdrop that floats on the upper surface of the aqueous solution after centrifugation. To overcome this drawback, other researchers have recently attempted to use some special homemade extraction devices to assist in the collection of the floating microdrop [27–30]. However, such methods are complicated and the homemade devices are not readily available for wider use. The aim of this study was to develop a simple and efficient analytical method for the rapid determination of trace level of PAEs in Chinese liquor samples using VSLLME followed by GC–MS. In this method, a high-density solvent was employed for the first time as extraction solvent for the pretreatment of four PAEs in Chinese liquor samples prior to quantification by GC–MS. Various experimental factors, such as the type and C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
685
volume of the extraction solvent, the type and concentration of the surfactant, extraction time, sample pH, and salt addition were investigated and optimized. Finally, the proposed method has been successfully applied to the determination of PAEs in 16 real Chinese liquor samples from different regions in Sichuan Province, China.
2 Materials and methods 2.1 Reagents and standards All standards of PAEs including dibutyl phthalate (DBP), butyl benzyl ester (BBP), di-2-ethyl hexyl phthalate (DEHP), and dioctyl phthalate (DNOP) were purchased from Merck (Darmstadt, Germany). The stock standard solutions of 1 g/L of each compound were prepared in methanol. The standard working solutions were prepared daily by mixing an appropriate amount of the stock standard solution with ultrapure water to the required concentrations. The stock and working standard solutions were stored at 4⬚C. HPLC grade dichloromethane, chloroform, carbon tetrachloride (CCl4 ), and sodium chloride (NaCl) were purchased from Rionlon (Tianjing, China). Cethyltrimethyl ammonium bromide (CTAB), SDS, Triton X-100, and Triton X-114 were purchased from Solarbio (Shanghai, China). All solutions were prepared in ultra-pure water, with a resistivity of 18.2 M⍀·cm obtained from a Milli-Q Integral 3 system from Millipore (MA, USA). To avoid PAEs contamination, all glassware used in the study was soaked in acetone for at least 6 h, then washed with acetone and dried at 140⬚C for at least 4 h. Furthermore, all reagents and glassware were checked for potential phthalate contamination using GC–MS analysis.
2.2 Instrumentation Separation and quantification were performed with an Agilent (Agilent Technologies, CA, USA) 7890 series gas chromatograph equipped with an Agilent 5975 mass selective detector system. Chromatographic separation was accomplished on an HP-5MS (5% phenyl, 95% methylpolysiloxane, 30 m × 0.25 mm id × 0.25 m) capillary column, obtained from J&W Scientific (CA, USA). Helium (>99.999%) was employed as carrier gas at a flow rate of 1.0 mL/min. The injector temperature was set at 250⬚C. The column temperature was programmed as: 60⬚C, held for 1 min, 20⬚C/min up to 220⬚C, held for 1 min, 5⬚C/min up to 280⬚C, held for 4 min. The GC–MS interface was maintained at 280⬚C, and all injections were in splitless mode. The mass spectrometer was operated in the electron ionization mode (70 eV) and the analytes were recorded in selected ion monitoring mode. The start scan times and masses monitored for each compound were set as follows: DBP, 10.0 min, m/z 149, 223, 205; BBP 14.5 min, 149, 206, 238; DEHP, 16.6 min, 149, 167, 279; DONP, 19.0 min, 149, 279, 167. www.jss-journal.com
686
J. Sep. Sci. 2014, 37, 684–690
G. Leng et al.
2.3 Samples and VSLLME procedures Real Chinese liquor samples with an alcohol content between 38–53 alc/vol were purchased from the local market to prepare the liquor samples. After being filtered through a 0.22 m membrane to eliminate particulate matter, the liquor samples were stored in amber glass bottle and kept at 4⬚C. In VSLLME, aliquots of 5.0 mL liquor samples were placed in a 15 mL screw cap glass tube with conical bottom. Two hundred and fifty microliters of carbon tetrachloride as extraction solvent and 5.0 L of 0.2 mmol/L Triton X-100 as emulsifier were added into the sample solution. The tube was capped immediately and the mixture was then vigorously shaken using a XW-80A vortex agitator (Huxi Instruments, Shanghai, China) at 2800 rpm for 30 s. As a result fine droplets were formed, facilitating mass transfer of the target analytes into the CCl4 phase. The two phases were separated after centrifuging the mixture using a Multifuge X3 centrifuge (Thermo Scientific, MA, USA) at 6000 rpm at 2⬚C for 5 min. After that, the CCl4 phase was sedimented at the bottom of the glass tube and could easily be withdrawn by using a 1.0 L microsyringe and directly injected into GC–MS for analysis.
3 Results and discussion In order to achieve the optimum extraction conditions, the effects of several different experimental factors, such as the type and volume of the extraction solvent, the type and concentration of the surfactant, extraction time, sample pH, and salt addition, on the performance of VSLLME were investigated. All the experiments for optimizing the proposed VSLLME procedure were performed with a spiked real liquor sample with the appropriate PAEs concentration (50 g/L) using GC–MS analysis. Furthermore, blank tests were performed in each set of the experiments, the results of which indicated that the analytical processes of the proposed method were free from PAEs contamination. Enrichment factors (EFs) and extraction recoveries (ERs) were employed to evaluate the extraction efficiency. The EF is defined as the ratio between the concentration of analyte in the sediment phase (Csed ) and the initial concentration of analyte (C0 ) in the sample: EF = Csed /C0
(1)
The ER is defined as the percentage of the total analyte (n0 ) that was extracted to the sediment phase (nsed ): ER = nsed /n0 × 100 = ((Csed × Vsed )/(C0 × Vaq )) × 100
(2)
where Vsed and Vaq are the volume of sediment phase and sample solution, respectively.
Figure 1. Effect of the type of extraction solvent on extraction.
in this method should possess a higher density than liquor samples, a low solubility in liquor samples, high extraction efficiency, and a good gas chromatographic behavior [12]. Based on the above considerations, we investigated the performance of dichloromethane, chloroform, and CCl4 as potential extraction solvents for the proposed method. As shown in Fig. 1, compared with other extraction solvents, the highest EFs could be obtained when carbon tetrachloride was selected as the extraction solvent. Consequently, carbon tetrachloride was selected as the optimum extraction solvent and used in subsequent studies. 3.2 Effect of the volume of extraction solvent In VSLLME, the volume of extraction solvent also plays a very important role, which could affect the EF. To study the effect of extraction solvent volume, different volumes of carbon tetrachloride between 250 and 450 L were subjected to the same VSLLME procedures. As can be seen in Fig. 2, by increasing the volume of CCl4 from 250 to 450 L, the EFs decreased from 139–183 to 41–50. Theoretically, a higher EF could be obtained at smaller volume of the extraction solvent. However, due to the complexity of the matrix effect of the real liquor sample (52 alc/vol), extraction solvent volumes 0.4 mmol/L was used. Based on the above considerations, an optimum concentration of 0.2 mmol/L was selected for Triton X-100. Figure 3. Effect of the type of surfactant on extraction.
anionic surfactant (SDS), and nonionic surfactant (Triton X100 and Triton X-114) were investigated. As shown in Fig. 3, the highest EFs can be obtained when Triton X-100 was used. This can be understood by the fact that the effect of different surfactants on the extraction efficiency was related to the hydrophobicity and polarity of the analytes as well as the hydrophile–lipophile balance of the surfactants [31], and Triton X-100 showed a suitable hydrophobicity for the PAEs, resulting in better and comparative extraction efficiency. Therefore, Triton X-100 was selected as the optimum surfactant for the proposed method. 3.4 Effect of the concentration of surfactant Different concentrations of Triton X-100 ranging from 0 to 0.5 mmol/L were investigated to study its effect. As can be seen in Fig. 4, the EF increased when the concentration of Triton X-100 was increased from 0 to 0.2 mmol/L, after that, the EF began to decrease. This can be explained by the fact that when the Triton X-100 concentration was in C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3.5 Effect of the vortex time In the proposed method, the extraction process was achieved by vigorously shaking with a vortex. The dispersion of the extraction solvent into samples is dependent on the rotational speed and vortex. Due to the limitation of vortex agitator, the vortex rotational speed was set at a constant speed of 2800 rpm. The effects of different vortex times (0.5, 1, 3, 5 and 10 min) on the extraction efficiency were studied. There was no significant effect on the extraction efficiency when the vortex time increased from 0.5 to 5 min, indicating that the addition of surfactant could greatly enhance the mass transfer and the equilibrium state could be achieved within 30 s. Therefore, 30 s of vortex time was used in the proposed method. 3.6 Effect of salt addition and sample pH Salt was often added into the sample solution to enhance the extraction efficiency. Different concentrations of NaCl (0–5%, m/v) were added into samples to study the effect of salt addition on the performance of the proposed method. www.jss-journal.com
688
J. Sep. Sci. 2014, 37, 684–690
G. Leng et al.
The results indicated that the EFs of the analytes decreased when the NaCl concentration increased from 0 to 5% m/v. Although the increase of ionic strength could decrease the solubility of analytes in aqueous sample and enhances their partitioning into the organic phase [24], this may also lead to decrease of CCl4 solubility in aqueous phase, which further increased the volume of sedimented phase and decreased the EF. Furthermore, the increase of the salt content in aqueous sample may also increase the viscosity and density of the solution, forming a physical barrier across samples and extraction solvent interface, which inhibited the mass transfer between the two phases [17]. Therefore, no salt was added in this study. The pH of the sample is another important factor that potentially affects the extraction. The effects of pH on VSLLME were studied in the range of 2–10. However, negligible effects were observed on the extraction efficiency within the studied pH range. This is due to the fact that PAEs are very stable in liquor matrix and their chemical structures would barely change at different pH. Thus, no pH adjustment is needed in the proposed method. 3.7 Analytical figures of merit The chromatograms obtained after VSLLME of liquor sample spiked with 10 ng of each PAEs was shown in Fig. 5. The main analytical figures of merit of the method are summarized in Table 1. The LODs, based on S/N = 3, ranged from 4.9 to 13 ng/L. Linearity was found in the concentration range from 0.05 to 50 g/L for DBP, DEHP, and DNOP, and 0.1– 50 g/Lfor BBP. Coefficients of determination (R2 ) ranged from 0.9938 to 0.9971. Reproducibility and recoveries were assessed by extracting a series of three independent liquor samples, which were spiked with different concentration levels. A comparison of the LODs and the extraction time obtained by other analytical techniques for the determination of PAEs in alcoholic samples are summarized in Table 2. As shown, the LODs of the proposed method are comparable with other analytical techniques. In addition, the proposed method has the advantage of having a much shorter extrac-
Figure 5. Chromatograms of DBP, BBP, DEHP, and DNOP before and after spiking with 10 ng of each PAE. Experiments were carried out under the optimum conditions.
tion time than other methods. Such comparison indicated that the proposed VSLLME is a fast, simple, inexpensive, and environmentally friendly technique that can be used for the trace analysis of PAEs in liquor samples. 3.8 Application to real liquor samples In order to investigate the applicability of the developed method to real liquor samples, the proposed method has been
Table 1. Figures of merit of the VSLLME–GC–MS method for the determination of PAEs in Chinese liquor samples.
Analytes
DBP BBP DEHP DNOP
Limit of detection (ng/L)
Linearity (g/L)
4.9 13 9.2 8.2
0.05–50 0.1–50 0.05–50 0.05–50
Coefficients of determinations (R2 )
Enrichment factors
0.9949 0.9971 0.9938 0.9962
184 140 166 152
Recoveries (%)a)
Reproducibility (RSD,%) High levelb)
Mid levelc)
Low leveld)
High levelb)
Mid levelc)
Low leveld)
7.2 6.9 10.4 6.2
8.3 8.8 7.5 6.4
6.8 11.2 9.5 8.5
92.1 ± 3.4 80.8 ± 4.2 89.1 ± 4.8 77.8 ± 3.7
90.5 ± 4.0 79.4 ± 3.8 87.6 ± 3.2 78.3 ± 5.1
93.7 ± 3.1 81.2 ± 4.0 90.2 ± 3.5 80.5 ± 6.2
a) n = 3 replicates. b) Liquor sample spiked with 10 ng PAEs. c) Liquor sample spiked with 20 ng PAEs. d) Liquor sample spiked with 50 ng PAEs.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jss-journal.com
Sample Preparation
J. Sep. Sci. 2014, 37, 684–690
689
Table 2. Comparison of proposed VSLLME–GC–MS method with previously reported methods for extraction and determination of PAEs in alcoholic samples
Method
Instruments
Matrix
Extraction time (min)
LODs (g/L)
Ref.
MSPEa) SPMEb) HS-SPMEc) SPEd) USVADLLMEe) VSLLME
GC-MS GC-FID GC-MS GC-MS GC-IT/MS GC-MS
Beverage Beer Wine Wine Wine Chinese liquor
30 >50 10 50 0.5 + 10 0.5
0.003–0.04 0.003–3.4 0.0048–0.0288 25–35 0.022–0.1 0.0049–0.013
[32] [33] [34] [35] [36] This work
a) Magnetic solid-phase extraction. b) Solid-phase microextraction. c) Headspace solid-phase microextraction. d) Solid-phase extraction. e) Ultrasound-vortex-assisted dispersive liquid-liquid microextraction. Table 3. Determination of PAEs in commercial Chinese liquor samples by the VSLLME–GC–MS method
Sample Alcohol DBP content (alc/vol) Found (g/L)a) Recoveries (%)b) S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S15 S16
52 52 38 38 52 43 39 39 39 53 53 53 45 49 42 42
52.4 91.7 133.7 60.9 221.5 184.6 214.6 170.8 113.9 465 56.4 725 633 332.4 207.8 24.1
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
3.03 5.29 6.50 4.04 11.12 9.74 13.72 10.97 6.12 13.3 2.28 30.61 28.59 22.26 10.86 1.30
90.9 ± 5.47 91.8 ± 5.25 90.0 ± 7.72 90.2 ± 5.47 91.0 ± 7.29 92.0 ± 6.49 91.0 ± 5.43 91.8 ± 4.81 91.9 ± 5.02 92.9 ± 6.67 91.7 ± 5.04 92.1 ± 4.83 92.3 ± 7.64 91.2 ± 5.65 91.0 ± 4.81 90.3 ± 6.67
BBP
DEHP
DNOP
Found (g/L)a)
Recoveries (%)b)
Found (g/L)a)
–c) – 24.7 ± 0.77 – 8.6 ± 0.64 – – – – 31.7 ± 2.04 – 38.3 ± 2.29 – – 6.5 ± 0.27 –
79.5 ± 4.09 80.9 ± 5.55 78.3 ± 5.54 78.2 ± 6.22 79.3 ± 5.48 79.4 ± 5.31 78.4 ± 6.60 78.0 ± 5.99 79.5 ± 5.63 78.8 ± 4.79 80.5 ± 4.28 80.7 ± 4.80 80.2 ± 5.02 80.9 ± 6.87 78.6 ± 6.71 78.1 ± 5.30
15.0 46.7 24.7 30.4 43.8 28.7 18.4 42.4 39.2 26.5 12.2 35.4 42.2 36.1 42.7 7.1
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.79 2.80 1.83 2.38 3.06 1.46 0.97 2.13 2.61 1.79 0.63 2.77 2.38 2.80 2.62 0.56
Recoveries (%)b)
Found (g/L)a)
Recoveries (%)b)
88.6 ± 4.27 90.1 ± 4.27 90.7 ± 5.53 88.9 ± 4.22 89.7 ± 3.88 90.9 ± 3.91 90.2 ± 4.36 88.1 ± 5.26 90.7 ± 5.83 90.6 ± 6.16 88.0 ± 4.25 89.7 ± 3.78 89.7 ± 5.44 89.2 ± 5.29 89.5 ± 6.24 89.1 ± 3.85
– – 3.8 ± 0.08 – – – – – – 1.7 ± 0.04 – 2.8 ± 0.08 – – – –
76.3 ± 5.16 75.2 ± 4.31 79.7 ± 3.51 77.9 ± 4.79 78.4 ± 4.97 77.4 ± 3.77 79.5 ± 4.76 76.8 ± 5.21 78.2 ± 3.70 79.6 ± 3.46 77.0 ± 3.54 76.5 ± 3.61 77.4 ± 3.45 78.0 ± 3.28 80.4 ± 4.94 76.5 ± 3.65
a) n = 3 replicates. b) Samples were spiked with 20 ng PAEs. c) Not detected.
applied to the determination of PAEs in 16 commercial Chinese liquor samples from the local market. The results are shown in Table 3. The results indicate that DBP and DEHP were always present in all samples with concentrations ranging from 24.1 ± 1.30 to 725 ± 30.61 and 7.1 ± 0.56 to 46.7 ± 2.80 g/L, respectively, whereas BBP and DNOP were not systematically present in all samples. The results indicated that DEHP, BBP, and DNOP in all the samples met Food Safety National Standards in China. However, the health risk of PAEs for humans depends on the exposure levels and bioaccumulation [2, 24, 37–39]. Furthermore, in the recovery experiments, samples were spiked with 20 g/L PAEs standard solution before VSLLME and following the procedures of the proposed method. The recoveries of PAEs in all the liquor samples were in the range C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
of 75.2 ± 4.3 to 92.9 ± 6.7%, indicating that no significant matrix effect existed.
4 Conclusion In this work, high-density-solvent VSLLME was developed and applied for the first time for the extraction of PAEs in Chinese liquor samples prior to analysis by GC–MS. The proposed method not only produced good linearity, extraction efficiency and repeatability, but also gave a comparable detection limit to those reported in previous studies. More importantly, the high-density extraction solvent could better disperse with the aid of surfactant and vortex agitation, resulting in short extraction equilibrium within only 30 s. The www.jss-journal.com
690
G. Leng et al.
single microdrop that sedimented at the bottom of the tube could be easily collected after centrifugation. Finally, the proposed method was successfully applied to the determination of PAEs in 16 commercial Chinese liquor samples at different alcohol content. The proposed method is a simple, low-cost, and rapid method for the determination of PAEs in liquor samples.
J. Sep. Sci. 2014, 37, 684–690
[16] Sanchez, P. L., Barro, R., Garcia, J. C., Llompart, M., Lores, M., Petrakis, C., Kalogerakis, N., Mantzavinos, D., Psillakis, E., Ultrason. Sonochem. 2008, 15, 689–694. [17] Yiantzi, E., Psillakis, E., Tyrovola, K., Kalogerakis, N., Talanta 2010, 80, 2057–2062. [18] Yang, Z. H., Liu, D. H., Zhao, W. T., Wu, T., Zhou, Z. Q., Wang, P., J. Sep. Sci. 2012, 36, 916–922 [19] Ozcan, S., J. Sep. Sci. 2011, 34, 574–584.
This work was supported in part by a research fund from The National Natural Science Foundation of China (NSFC). The authors have declared no conflict of interest.
5 References [1] Hoppin, J. A., Ulmer, R., London, S. J., Environ. Health Perspect. 2004, 112, 571–574. [2] Doull, J., Cattley, R., Elcombe, C., Lake, B. G., Swenberg, J., Wilkinson, C., Williams, G., van Gemert, M., Regul. Toxicol. Phar macol. 1999, 29, 327–357. [3] Overturf, M. L., Druilhet, R. E., Liehr, J. G., Kirkendall, W. M., Caprioli, R. M., Bull. Environ. Contam. Toxicol. 1979, 22, 536–542. [4] Latini, G., Verrotti, A., De Felice, C., Curr. Drug Targets Immune Endocr. Metabol. Disord. 2004, 4, 37–40. [5] Pawliszyn, J., Pedersen, B. S., J. Chromatogr. Sci. 2006, 44, 291–307.
[20] Leng, G., Yin, H., Li, S., Chen, Y., Dan, D., Talanta 2012, 99, 631–636. [21] Zacharis, C. K., Christophoridis, C., Fytianos, K., J. Sep. Sci. 2012, 35, 2422–2429 [22] Qin, H., Li, B., Liu, M. S., Yang, Y. L., J. Sep. Sci. 2013, 36, 1463–1469 [23] Zhang, L. J., Chen, F., Liu, W. S., Chen, B. Y., Pan C. P., J. Sep. Sci. 2012, 35, 2514–2519 [24] Moradi, M., Yamini, Y., Esrafili, A., Seidi, S., Talanta 2010, 82, 1864–1869. [25] Wu, C., Liu, N., Wu, Q., Wang, C., Wang, Z., Anal. Chim. Acta 2010, 679, 56–62. [26] Zhang, Y., Lee, H. K., J. Chromatogr. A 2013, 1274, 28–35. [27] Farajzadeh, M. A., Seyedi, S. E., Shalamzari, M. S., Bamorowat, M., J. Sep. Sci. 2009, 32, 3191–3200. [28] Hashemi, P., Beyranvand, S., Mansur, R. S., Ghiasvand, A. R., Anal. Chim. Acta 2009, 655, 60–65. [29] Saleh, A., Yamini, Y., Faraji, M., Rezaee, M., Ghambarian, M., J. Chromatogr. A 2009, 1216, 6673–6679.
[6] Liu, H., Dasgupta, P. K., Anal. Chem. 1996, 68, 1817–1821.
[30] Yang, Z., Wang, P., Zhao, W., Zhou, Z., Liu, D., J. Chromatogr. A 2013, 1300, 58–63.
[7] Pedersen, B. S., Rasmussen, K. E., Anal. Chem. 1999, 71, 2650–2656.
[31] Wu, Q., Chang, Q., Wu, C., Rao, H., Zeng, X., Wang, C., Wang, Z., J. Chromatogr. A 2010, 1217, 1773–1778.
[8] Liu, W., Lee, H. K., Anal. Chem. 2000, 72, 4462–4467.
[32] Luo, Y. B., Yu, Q. W., Yuan, B. F., Feng, Y. Q., Talanta 2012, 90, 123–131.
[9] Theis, A. L., Waldack, A. J., Hansen, S. M., Jeannot, M. A., Anal. Chem. 2001, 73, 5651–5654. [10] Rasmussen, K. E., Pedersen, B. S., Krogh, M., Ugland, H. G., Gronhaug, T., J. Chromatogr. A 2000, 873, 3–11. ¨ [11] Khalili Z. M. R., Yamini, Y., Shariati, S., Jonsson, J. A., Anal. Chim. Acta 2007, 585, 286–293. [12] Rezaee, M., Assadi, Y., Milani H. M. R., Aghaee, E., Ahmadi, F., Berijani, S., J. Chromatogr. A 2006, 1116, 1–9. [13] Leng, G., Lui, G., Chen, Y., Yin, H., Dan, D., J. Sep. Sc. 2012, 35, 2796–2804.
[33] Ye, C. W., Gao, J., Yang, C., Liu, X. J., Li, X. J., Pan, S. Y., Anal. Chim. Acta 2009, 641, 64–74. [34] Carrillo, J. D., Martinez, M. P., Tena, M. T., J. Chromatogr. A 2008, 1181, 125–130. [35] Michele, D. C., Alesssia, P., Giampiero, S., Dario, C., Dino, M., Angelo, C., Food Chem. 2008, 111, 771–777. [36] Cinelli, G., Avino, P., Notardonato, I., Centola, A., Russo, M. V., Anal. Chim. Acta 2013, 769, 72–78. [37] Latini, G., Clin. Chim. Acta, 2005, 361, 20–29.
[14] Regueiro, J., Llompart, M., Psillakis, E., Garcia, M. J. C., Garcia, J. C., Talanta 2009, 79, 1387–1397.
[38] Martino, A. A. J., Chahoud, I., Mol. Nutr. Food Res. 2010, 54, 148–157.
[15] Zhou, Q., Zhang, X., Xiao, J., J. Chromatogr. A 2009, 1216, 4361–4365.
[39] Matsumoto, M., Hirata, K. M., Ema, M., Regul. Toxicol. Pharm. 2008, 50, 37–49.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jss-journal.com
of Resources and Environment, University of Electronic Science and Technology of China, Chengdu, China 2 Griffith School of Engineering, Nathan Campus, Griffith University, Nathan, Australia Received September 18, 2013 Revised November 20, 2013 Accepted December 18, 2013
J. Sep. Sci. 2014, 37, 684–690
Research Article
Determination of phthalate esters in liquor samples by vortex-assisted surfactant-enhanced-emulsification liquid–liquid microextraction followed by GC–MS A novel method using vortex-assisted surfactant-enhanced-emulsification liquid–liquid microextraction has been developed for the extraction of phthalate esters (PAEs) in Chinese liquor samples prior to analysis by GC–MS. In the proposed method, a high-density extraction solvent (carbon tetrachloride) was dispersed into samples with the aid of a surfactant (Triton X-100) and vortex agitation, resulting in a short extraction equilibrium (30 s). After centrifugation, a single microdrop of solvent was easily collected for GC–MS analysis. Key factors that affected the extraction efficiency were optimized. Under the optimum conditions, linearity was found in the range from 0.05 to 50 g/L. Coefficients of determination varied from 0.9938 to 0.9971. LODs, based on an S/N of 3, ranged from 4.9 to 13 ng/L. Enrichment factors varied from 140 to 184. Reproducibility and recoveries were assessed by testing a series of three liquor samples that were spiked with different concentration levels. Finally, the proposed method was successfully applied to the determination of PAEs in 16 Chinese liquor samples. In this work, high-density-solvent vortex-assisted surfactantenhanced-emulsification liquid–liquid microextraction was applied for the first time for the extraction of PAEs in Chinese liquor samples and was proved to be simple, rapid, and sensitive. Keywords: GC-MS / Liquid-phase microextraction / Liquor / Phthalate esters / Surfactants DOI 10.1002/jssc.201301033
1 Introduction Phthalate esters (PAEs) are used primarily as plasticizers in polymeric materials to increase their flexibility and moldability. Some studies indicated that PAEs as well as their metabolites and degradation products can cause damage to the kidney, lung, endocrine, and reproductive system [1–4]. Because of their risks and abundant presence in many plastic mateCorrespondence: Prof. Qiming Cao, School of Resources and Environment, University of Electronic Science and Technology of China, 4 Jianshe North Road 2nd Section, Chenghua, Chengdu, Sichuan 610051, China E-mail: [email protected] Fax: +86–028–61831571
Abbreviations: BBP, butyl benzyl ester; CCl4 , carbon tetrachloride; CTAB, cethyltrimethyl ammonium bromide; DBP, dibutyl phthalate; DEHP, di-2-ethyl hexyl phthalate; DNOP, dioctyl phthalate; DLLME, dispersive liquid–liquid microextraction; EF, enrichment factor; ER, extraction recovery; LPME, liquid-phase microextraction; NaCl, sodium chloride; PAE, phthalate ester; SDME, single-drop microextraction; VALLME, vortex-assisted liquid–liquid microextraction; VSLLME, vortex-assisted surfactant-enhancedemulsification liquid–liquid microextraction C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
rials, there is a potential risk of PAEs contamination during liquor making. This may arise both from the raw materials and the use of plastic tools during processing. Moreover, additives and technological coadjuvants may also contribute to the health risk of PAEs. Consequently, PAEs contamination in liquor has now become a major public health concern in China, and the development of analytical methods for their determination is of considerable interest. However, due to the trace concentration levels of PAEs in liquor samples and the complexity of the liquor matrices, a sensitive and reliable pretreatment method for analysis of PAEs in liquor samples is imperative. Liquid-phase microextraction (LPME) represents an important development in the field of sample preparation for its simplicity, miniaturization, and time efficiency [5]. The first methodology evolved from LPME was single-drop microextraction (SDME), which was developed by Liu [6]. Since the introduction of SDME, different analytical modes, including liquid–liquid–liquid microextraction [7], continuous-flow microextraction [8], headspace SDME [9], hollow-fiber LPME [10] and solid-drop LPME [11] have been developed for various analytical applications. Colour Online: See the article online to veiw Figs. 1–5 in colour. www.jss-journal.com
Sample Preparation
J. Sep. Sci. 2014, 37, 684–690
Among all these analytical modes, a new type of LPME method termed as dispersive liquid–liquid microextraction (DLLME) has attracted more and more attention since its introduction by Rezaee et al. in 2006 [12]. In DLLME, an appropriate mixture of extraction and disperser solvent is rapidly injected into the aqueous sample. The extraction solvent is dispersed into the aqueous sample as very fine droplets and a cloudy solution is formed. The extension of the contact surface between the extraction solvent and aqueous phase can greatly enhance the extraction efficiency. However, the main drawback of DLLME is the necessity of using the dispersion solvent, which may decrease the partition coefficient of analytes into the extraction solvent [13]. In order to avoid the use of dispersion solvent, ultrasound was used to assist the dispersion of the extraction solvent into aqueous samples [14, 15]. However, degradation of the analyte occurred under ultrasound conditions has been reported [16]. More recently, Yiantzi et al. [17] introduced a novel LPME technique, the vortex-assisted liquid–liquid microextraction (VALLME). In VALLME, microvolumes of organic extraction solvent were dispersed into an aqueous sample by using vortex mixing. The fine droplets could rapidly extract analytes from the aqueous sample because of the shorter diffusion distance and larger specific surface area. After centrifugation, the extractant solvent restores its initial single microdrop shape and is ready for instrumental analysis. VALLME overcomes the main disadvantage of DLLME, and has been successfully applied for the extraction of different compounds since its introduction [17–23]. It is well known that surfactants are amphiphilic compounds containing both hydrophobic and hydrophilic groups, thus surfactants can reduce the interfacial tension between oil and water by adsorbing at the liquid–liquid interface to increase the dispersion [24]. So surfactants can be used to enhance the dispersion of extraction solvents into aqueous samples. Based on these considerations, using surfactants to assist extraction solvents to better disperse into the sample in VALLME was found to be a very advantageous idea [25]. To the best of our knowledge, only one application of vortex-assisted surfactant-enhanced-emulsification liquid– liquid microextraction (VSLLME) for the extraction of PAEs was reported [26]. In this study, low-density organic solvent was used as extractant solvent. However, it is very difficult to collect the single microdrop that floats on the upper surface of the aqueous solution after centrifugation. To overcome this drawback, other researchers have recently attempted to use some special homemade extraction devices to assist in the collection of the floating microdrop [27–30]. However, such methods are complicated and the homemade devices are not readily available for wider use. The aim of this study was to develop a simple and efficient analytical method for the rapid determination of trace level of PAEs in Chinese liquor samples using VSLLME followed by GC–MS. In this method, a high-density solvent was employed for the first time as extraction solvent for the pretreatment of four PAEs in Chinese liquor samples prior to quantification by GC–MS. Various experimental factors, such as the type and C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
685
volume of the extraction solvent, the type and concentration of the surfactant, extraction time, sample pH, and salt addition were investigated and optimized. Finally, the proposed method has been successfully applied to the determination of PAEs in 16 real Chinese liquor samples from different regions in Sichuan Province, China.
2 Materials and methods 2.1 Reagents and standards All standards of PAEs including dibutyl phthalate (DBP), butyl benzyl ester (BBP), di-2-ethyl hexyl phthalate (DEHP), and dioctyl phthalate (DNOP) were purchased from Merck (Darmstadt, Germany). The stock standard solutions of 1 g/L of each compound were prepared in methanol. The standard working solutions were prepared daily by mixing an appropriate amount of the stock standard solution with ultrapure water to the required concentrations. The stock and working standard solutions were stored at 4⬚C. HPLC grade dichloromethane, chloroform, carbon tetrachloride (CCl4 ), and sodium chloride (NaCl) were purchased from Rionlon (Tianjing, China). Cethyltrimethyl ammonium bromide (CTAB), SDS, Triton X-100, and Triton X-114 were purchased from Solarbio (Shanghai, China). All solutions were prepared in ultra-pure water, with a resistivity of 18.2 M⍀·cm obtained from a Milli-Q Integral 3 system from Millipore (MA, USA). To avoid PAEs contamination, all glassware used in the study was soaked in acetone for at least 6 h, then washed with acetone and dried at 140⬚C for at least 4 h. Furthermore, all reagents and glassware were checked for potential phthalate contamination using GC–MS analysis.
2.2 Instrumentation Separation and quantification were performed with an Agilent (Agilent Technologies, CA, USA) 7890 series gas chromatograph equipped with an Agilent 5975 mass selective detector system. Chromatographic separation was accomplished on an HP-5MS (5% phenyl, 95% methylpolysiloxane, 30 m × 0.25 mm id × 0.25 m) capillary column, obtained from J&W Scientific (CA, USA). Helium (>99.999%) was employed as carrier gas at a flow rate of 1.0 mL/min. The injector temperature was set at 250⬚C. The column temperature was programmed as: 60⬚C, held for 1 min, 20⬚C/min up to 220⬚C, held for 1 min, 5⬚C/min up to 280⬚C, held for 4 min. The GC–MS interface was maintained at 280⬚C, and all injections were in splitless mode. The mass spectrometer was operated in the electron ionization mode (70 eV) and the analytes were recorded in selected ion monitoring mode. The start scan times and masses monitored for each compound were set as follows: DBP, 10.0 min, m/z 149, 223, 205; BBP 14.5 min, 149, 206, 238; DEHP, 16.6 min, 149, 167, 279; DONP, 19.0 min, 149, 279, 167. www.jss-journal.com
686
J. Sep. Sci. 2014, 37, 684–690
G. Leng et al.
2.3 Samples and VSLLME procedures Real Chinese liquor samples with an alcohol content between 38–53 alc/vol were purchased from the local market to prepare the liquor samples. After being filtered through a 0.22 m membrane to eliminate particulate matter, the liquor samples were stored in amber glass bottle and kept at 4⬚C. In VSLLME, aliquots of 5.0 mL liquor samples were placed in a 15 mL screw cap glass tube with conical bottom. Two hundred and fifty microliters of carbon tetrachloride as extraction solvent and 5.0 L of 0.2 mmol/L Triton X-100 as emulsifier were added into the sample solution. The tube was capped immediately and the mixture was then vigorously shaken using a XW-80A vortex agitator (Huxi Instruments, Shanghai, China) at 2800 rpm for 30 s. As a result fine droplets were formed, facilitating mass transfer of the target analytes into the CCl4 phase. The two phases were separated after centrifuging the mixture using a Multifuge X3 centrifuge (Thermo Scientific, MA, USA) at 6000 rpm at 2⬚C for 5 min. After that, the CCl4 phase was sedimented at the bottom of the glass tube and could easily be withdrawn by using a 1.0 L microsyringe and directly injected into GC–MS for analysis.
3 Results and discussion In order to achieve the optimum extraction conditions, the effects of several different experimental factors, such as the type and volume of the extraction solvent, the type and concentration of the surfactant, extraction time, sample pH, and salt addition, on the performance of VSLLME were investigated. All the experiments for optimizing the proposed VSLLME procedure were performed with a spiked real liquor sample with the appropriate PAEs concentration (50 g/L) using GC–MS analysis. Furthermore, blank tests were performed in each set of the experiments, the results of which indicated that the analytical processes of the proposed method were free from PAEs contamination. Enrichment factors (EFs) and extraction recoveries (ERs) were employed to evaluate the extraction efficiency. The EF is defined as the ratio between the concentration of analyte in the sediment phase (Csed ) and the initial concentration of analyte (C0 ) in the sample: EF = Csed /C0
(1)
The ER is defined as the percentage of the total analyte (n0 ) that was extracted to the sediment phase (nsed ): ER = nsed /n0 × 100 = ((Csed × Vsed )/(C0 × Vaq )) × 100
(2)
where Vsed and Vaq are the volume of sediment phase and sample solution, respectively.
Figure 1. Effect of the type of extraction solvent on extraction.
in this method should possess a higher density than liquor samples, a low solubility in liquor samples, high extraction efficiency, and a good gas chromatographic behavior [12]. Based on the above considerations, we investigated the performance of dichloromethane, chloroform, and CCl4 as potential extraction solvents for the proposed method. As shown in Fig. 1, compared with other extraction solvents, the highest EFs could be obtained when carbon tetrachloride was selected as the extraction solvent. Consequently, carbon tetrachloride was selected as the optimum extraction solvent and used in subsequent studies. 3.2 Effect of the volume of extraction solvent In VSLLME, the volume of extraction solvent also plays a very important role, which could affect the EF. To study the effect of extraction solvent volume, different volumes of carbon tetrachloride between 250 and 450 L were subjected to the same VSLLME procedures. As can be seen in Fig. 2, by increasing the volume of CCl4 from 250 to 450 L, the EFs decreased from 139–183 to 41–50. Theoretically, a higher EF could be obtained at smaller volume of the extraction solvent. However, due to the complexity of the matrix effect of the real liquor sample (52 alc/vol), extraction solvent volumes 0.4 mmol/L was used. Based on the above considerations, an optimum concentration of 0.2 mmol/L was selected for Triton X-100. Figure 3. Effect of the type of surfactant on extraction.
anionic surfactant (SDS), and nonionic surfactant (Triton X100 and Triton X-114) were investigated. As shown in Fig. 3, the highest EFs can be obtained when Triton X-100 was used. This can be understood by the fact that the effect of different surfactants on the extraction efficiency was related to the hydrophobicity and polarity of the analytes as well as the hydrophile–lipophile balance of the surfactants [31], and Triton X-100 showed a suitable hydrophobicity for the PAEs, resulting in better and comparative extraction efficiency. Therefore, Triton X-100 was selected as the optimum surfactant for the proposed method. 3.4 Effect of the concentration of surfactant Different concentrations of Triton X-100 ranging from 0 to 0.5 mmol/L were investigated to study its effect. As can be seen in Fig. 4, the EF increased when the concentration of Triton X-100 was increased from 0 to 0.2 mmol/L, after that, the EF began to decrease. This can be explained by the fact that when the Triton X-100 concentration was in C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3.5 Effect of the vortex time In the proposed method, the extraction process was achieved by vigorously shaking with a vortex. The dispersion of the extraction solvent into samples is dependent on the rotational speed and vortex. Due to the limitation of vortex agitator, the vortex rotational speed was set at a constant speed of 2800 rpm. The effects of different vortex times (0.5, 1, 3, 5 and 10 min) on the extraction efficiency were studied. There was no significant effect on the extraction efficiency when the vortex time increased from 0.5 to 5 min, indicating that the addition of surfactant could greatly enhance the mass transfer and the equilibrium state could be achieved within 30 s. Therefore, 30 s of vortex time was used in the proposed method. 3.6 Effect of salt addition and sample pH Salt was often added into the sample solution to enhance the extraction efficiency. Different concentrations of NaCl (0–5%, m/v) were added into samples to study the effect of salt addition on the performance of the proposed method. www.jss-journal.com
688
J. Sep. Sci. 2014, 37, 684–690
G. Leng et al.
The results indicated that the EFs of the analytes decreased when the NaCl concentration increased from 0 to 5% m/v. Although the increase of ionic strength could decrease the solubility of analytes in aqueous sample and enhances their partitioning into the organic phase [24], this may also lead to decrease of CCl4 solubility in aqueous phase, which further increased the volume of sedimented phase and decreased the EF. Furthermore, the increase of the salt content in aqueous sample may also increase the viscosity and density of the solution, forming a physical barrier across samples and extraction solvent interface, which inhibited the mass transfer between the two phases [17]. Therefore, no salt was added in this study. The pH of the sample is another important factor that potentially affects the extraction. The effects of pH on VSLLME were studied in the range of 2–10. However, negligible effects were observed on the extraction efficiency within the studied pH range. This is due to the fact that PAEs are very stable in liquor matrix and their chemical structures would barely change at different pH. Thus, no pH adjustment is needed in the proposed method. 3.7 Analytical figures of merit The chromatograms obtained after VSLLME of liquor sample spiked with 10 ng of each PAEs was shown in Fig. 5. The main analytical figures of merit of the method are summarized in Table 1. The LODs, based on S/N = 3, ranged from 4.9 to 13 ng/L. Linearity was found in the concentration range from 0.05 to 50 g/L for DBP, DEHP, and DNOP, and 0.1– 50 g/Lfor BBP. Coefficients of determination (R2 ) ranged from 0.9938 to 0.9971. Reproducibility and recoveries were assessed by extracting a series of three independent liquor samples, which were spiked with different concentration levels. A comparison of the LODs and the extraction time obtained by other analytical techniques for the determination of PAEs in alcoholic samples are summarized in Table 2. As shown, the LODs of the proposed method are comparable with other analytical techniques. In addition, the proposed method has the advantage of having a much shorter extrac-
Figure 5. Chromatograms of DBP, BBP, DEHP, and DNOP before and after spiking with 10 ng of each PAE. Experiments were carried out under the optimum conditions.
tion time than other methods. Such comparison indicated that the proposed VSLLME is a fast, simple, inexpensive, and environmentally friendly technique that can be used for the trace analysis of PAEs in liquor samples. 3.8 Application to real liquor samples In order to investigate the applicability of the developed method to real liquor samples, the proposed method has been
Table 1. Figures of merit of the VSLLME–GC–MS method for the determination of PAEs in Chinese liquor samples.
Analytes
DBP BBP DEHP DNOP
Limit of detection (ng/L)
Linearity (g/L)
4.9 13 9.2 8.2
0.05–50 0.1–50 0.05–50 0.05–50
Coefficients of determinations (R2 )
Enrichment factors
0.9949 0.9971 0.9938 0.9962
184 140 166 152
Recoveries (%)a)
Reproducibility (RSD,%) High levelb)
Mid levelc)
Low leveld)
High levelb)
Mid levelc)
Low leveld)
7.2 6.9 10.4 6.2
8.3 8.8 7.5 6.4
6.8 11.2 9.5 8.5
92.1 ± 3.4 80.8 ± 4.2 89.1 ± 4.8 77.8 ± 3.7
90.5 ± 4.0 79.4 ± 3.8 87.6 ± 3.2 78.3 ± 5.1
93.7 ± 3.1 81.2 ± 4.0 90.2 ± 3.5 80.5 ± 6.2
a) n = 3 replicates. b) Liquor sample spiked with 10 ng PAEs. c) Liquor sample spiked with 20 ng PAEs. d) Liquor sample spiked with 50 ng PAEs.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jss-journal.com
Sample Preparation
J. Sep. Sci. 2014, 37, 684–690
689
Table 2. Comparison of proposed VSLLME–GC–MS method with previously reported methods for extraction and determination of PAEs in alcoholic samples
Method
Instruments
Matrix
Extraction time (min)
LODs (g/L)
Ref.
MSPEa) SPMEb) HS-SPMEc) SPEd) USVADLLMEe) VSLLME
GC-MS GC-FID GC-MS GC-MS GC-IT/MS GC-MS
Beverage Beer Wine Wine Wine Chinese liquor
30 >50 10 50 0.5 + 10 0.5
0.003–0.04 0.003–3.4 0.0048–0.0288 25–35 0.022–0.1 0.0049–0.013
[32] [33] [34] [35] [36] This work
a) Magnetic solid-phase extraction. b) Solid-phase microextraction. c) Headspace solid-phase microextraction. d) Solid-phase extraction. e) Ultrasound-vortex-assisted dispersive liquid-liquid microextraction. Table 3. Determination of PAEs in commercial Chinese liquor samples by the VSLLME–GC–MS method
Sample Alcohol DBP content (alc/vol) Found (g/L)a) Recoveries (%)b) S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S15 S16
52 52 38 38 52 43 39 39 39 53 53 53 45 49 42 42
52.4 91.7 133.7 60.9 221.5 184.6 214.6 170.8 113.9 465 56.4 725 633 332.4 207.8 24.1
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
3.03 5.29 6.50 4.04 11.12 9.74 13.72 10.97 6.12 13.3 2.28 30.61 28.59 22.26 10.86 1.30
90.9 ± 5.47 91.8 ± 5.25 90.0 ± 7.72 90.2 ± 5.47 91.0 ± 7.29 92.0 ± 6.49 91.0 ± 5.43 91.8 ± 4.81 91.9 ± 5.02 92.9 ± 6.67 91.7 ± 5.04 92.1 ± 4.83 92.3 ± 7.64 91.2 ± 5.65 91.0 ± 4.81 90.3 ± 6.67
BBP
DEHP
DNOP
Found (g/L)a)
Recoveries (%)b)
Found (g/L)a)
–c) – 24.7 ± 0.77 – 8.6 ± 0.64 – – – – 31.7 ± 2.04 – 38.3 ± 2.29 – – 6.5 ± 0.27 –
79.5 ± 4.09 80.9 ± 5.55 78.3 ± 5.54 78.2 ± 6.22 79.3 ± 5.48 79.4 ± 5.31 78.4 ± 6.60 78.0 ± 5.99 79.5 ± 5.63 78.8 ± 4.79 80.5 ± 4.28 80.7 ± 4.80 80.2 ± 5.02 80.9 ± 6.87 78.6 ± 6.71 78.1 ± 5.30
15.0 46.7 24.7 30.4 43.8 28.7 18.4 42.4 39.2 26.5 12.2 35.4 42.2 36.1 42.7 7.1
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.79 2.80 1.83 2.38 3.06 1.46 0.97 2.13 2.61 1.79 0.63 2.77 2.38 2.80 2.62 0.56
Recoveries (%)b)
Found (g/L)a)
Recoveries (%)b)
88.6 ± 4.27 90.1 ± 4.27 90.7 ± 5.53 88.9 ± 4.22 89.7 ± 3.88 90.9 ± 3.91 90.2 ± 4.36 88.1 ± 5.26 90.7 ± 5.83 90.6 ± 6.16 88.0 ± 4.25 89.7 ± 3.78 89.7 ± 5.44 89.2 ± 5.29 89.5 ± 6.24 89.1 ± 3.85
– – 3.8 ± 0.08 – – – – – – 1.7 ± 0.04 – 2.8 ± 0.08 – – – –
76.3 ± 5.16 75.2 ± 4.31 79.7 ± 3.51 77.9 ± 4.79 78.4 ± 4.97 77.4 ± 3.77 79.5 ± 4.76 76.8 ± 5.21 78.2 ± 3.70 79.6 ± 3.46 77.0 ± 3.54 76.5 ± 3.61 77.4 ± 3.45 78.0 ± 3.28 80.4 ± 4.94 76.5 ± 3.65
a) n = 3 replicates. b) Samples were spiked with 20 ng PAEs. c) Not detected.
applied to the determination of PAEs in 16 commercial Chinese liquor samples from the local market. The results are shown in Table 3. The results indicate that DBP and DEHP were always present in all samples with concentrations ranging from 24.1 ± 1.30 to 725 ± 30.61 and 7.1 ± 0.56 to 46.7 ± 2.80 g/L, respectively, whereas BBP and DNOP were not systematically present in all samples. The results indicated that DEHP, BBP, and DNOP in all the samples met Food Safety National Standards in China. However, the health risk of PAEs for humans depends on the exposure levels and bioaccumulation [2, 24, 37–39]. Furthermore, in the recovery experiments, samples were spiked with 20 g/L PAEs standard solution before VSLLME and following the procedures of the proposed method. The recoveries of PAEs in all the liquor samples were in the range C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
of 75.2 ± 4.3 to 92.9 ± 6.7%, indicating that no significant matrix effect existed.
4 Conclusion In this work, high-density-solvent VSLLME was developed and applied for the first time for the extraction of PAEs in Chinese liquor samples prior to analysis by GC–MS. The proposed method not only produced good linearity, extraction efficiency and repeatability, but also gave a comparable detection limit to those reported in previous studies. More importantly, the high-density extraction solvent could better disperse with the aid of surfactant and vortex agitation, resulting in short extraction equilibrium within only 30 s. The www.jss-journal.com
690
G. Leng et al.
single microdrop that sedimented at the bottom of the tube could be easily collected after centrifugation. Finally, the proposed method was successfully applied to the determination of PAEs in 16 commercial Chinese liquor samples at different alcohol content. The proposed method is a simple, low-cost, and rapid method for the determination of PAEs in liquor samples.
J. Sep. Sci. 2014, 37, 684–690
[16] Sanchez, P. L., Barro, R., Garcia, J. C., Llompart, M., Lores, M., Petrakis, C., Kalogerakis, N., Mantzavinos, D., Psillakis, E., Ultrason. Sonochem. 2008, 15, 689–694. [17] Yiantzi, E., Psillakis, E., Tyrovola, K., Kalogerakis, N., Talanta 2010, 80, 2057–2062. [18] Yang, Z. H., Liu, D. H., Zhao, W. T., Wu, T., Zhou, Z. Q., Wang, P., J. Sep. Sci. 2012, 36, 916–922 [19] Ozcan, S., J. Sep. Sci. 2011, 34, 574–584.
This work was supported in part by a research fund from The National Natural Science Foundation of China (NSFC). The authors have declared no conflict of interest.
5 References [1] Hoppin, J. A., Ulmer, R., London, S. J., Environ. Health Perspect. 2004, 112, 571–574. [2] Doull, J., Cattley, R., Elcombe, C., Lake, B. G., Swenberg, J., Wilkinson, C., Williams, G., van Gemert, M., Regul. Toxicol. Phar macol. 1999, 29, 327–357. [3] Overturf, M. L., Druilhet, R. E., Liehr, J. G., Kirkendall, W. M., Caprioli, R. M., Bull. Environ. Contam. Toxicol. 1979, 22, 536–542. [4] Latini, G., Verrotti, A., De Felice, C., Curr. Drug Targets Immune Endocr. Metabol. Disord. 2004, 4, 37–40. [5] Pawliszyn, J., Pedersen, B. S., J. Chromatogr. Sci. 2006, 44, 291–307.
[20] Leng, G., Yin, H., Li, S., Chen, Y., Dan, D., Talanta 2012, 99, 631–636. [21] Zacharis, C. K., Christophoridis, C., Fytianos, K., J. Sep. Sci. 2012, 35, 2422–2429 [22] Qin, H., Li, B., Liu, M. S., Yang, Y. L., J. Sep. Sci. 2013, 36, 1463–1469 [23] Zhang, L. J., Chen, F., Liu, W. S., Chen, B. Y., Pan C. P., J. Sep. Sci. 2012, 35, 2514–2519 [24] Moradi, M., Yamini, Y., Esrafili, A., Seidi, S., Talanta 2010, 82, 1864–1869. [25] Wu, C., Liu, N., Wu, Q., Wang, C., Wang, Z., Anal. Chim. Acta 2010, 679, 56–62. [26] Zhang, Y., Lee, H. K., J. Chromatogr. A 2013, 1274, 28–35. [27] Farajzadeh, M. A., Seyedi, S. E., Shalamzari, M. S., Bamorowat, M., J. Sep. Sci. 2009, 32, 3191–3200. [28] Hashemi, P., Beyranvand, S., Mansur, R. S., Ghiasvand, A. R., Anal. Chim. Acta 2009, 655, 60–65. [29] Saleh, A., Yamini, Y., Faraji, M., Rezaee, M., Ghambarian, M., J. Chromatogr. A 2009, 1216, 6673–6679.
[6] Liu, H., Dasgupta, P. K., Anal. Chem. 1996, 68, 1817–1821.
[30] Yang, Z., Wang, P., Zhao, W., Zhou, Z., Liu, D., J. Chromatogr. A 2013, 1300, 58–63.
[7] Pedersen, B. S., Rasmussen, K. E., Anal. Chem. 1999, 71, 2650–2656.
[31] Wu, Q., Chang, Q., Wu, C., Rao, H., Zeng, X., Wang, C., Wang, Z., J. Chromatogr. A 2010, 1217, 1773–1778.
[8] Liu, W., Lee, H. K., Anal. Chem. 2000, 72, 4462–4467.
[32] Luo, Y. B., Yu, Q. W., Yuan, B. F., Feng, Y. Q., Talanta 2012, 90, 123–131.
[9] Theis, A. L., Waldack, A. J., Hansen, S. M., Jeannot, M. A., Anal. Chem. 2001, 73, 5651–5654. [10] Rasmussen, K. E., Pedersen, B. S., Krogh, M., Ugland, H. G., Gronhaug, T., J. Chromatogr. A 2000, 873, 3–11. ¨ [11] Khalili Z. M. R., Yamini, Y., Shariati, S., Jonsson, J. A., Anal. Chim. Acta 2007, 585, 286–293. [12] Rezaee, M., Assadi, Y., Milani H. M. R., Aghaee, E., Ahmadi, F., Berijani, S., J. Chromatogr. A 2006, 1116, 1–9. [13] Leng, G., Lui, G., Chen, Y., Yin, H., Dan, D., J. Sep. Sc. 2012, 35, 2796–2804.
[33] Ye, C. W., Gao, J., Yang, C., Liu, X. J., Li, X. J., Pan, S. Y., Anal. Chim. Acta 2009, 641, 64–74. [34] Carrillo, J. D., Martinez, M. P., Tena, M. T., J. Chromatogr. A 2008, 1181, 125–130. [35] Michele, D. C., Alesssia, P., Giampiero, S., Dario, C., Dino, M., Angelo, C., Food Chem. 2008, 111, 771–777. [36] Cinelli, G., Avino, P., Notardonato, I., Centola, A., Russo, M. V., Anal. Chim. Acta 2013, 769, 72–78. [37] Latini, G., Clin. Chim. Acta, 2005, 361, 20–29.
[14] Regueiro, J., Llompart, M., Psillakis, E., Garcia, M. J. C., Garcia, J. C., Talanta 2009, 79, 1387–1397.
[38] Martino, A. A. J., Chahoud, I., Mol. Nutr. Food Res. 2010, 54, 148–157.
[15] Zhou, Q., Zhang, X., Xiao, J., J. Chromatogr. A 2009, 1216, 4361–4365.
[39] Matsumoto, M., Hirata, K. M., Ema, M., Regul. Toxicol. Pharm. 2008, 50, 37–49.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jss-journal.com
