微萃取-離子液體-分散液液

1、 Analytical MethodsA rapid shaking-based ionic liquid dispersive liquid phase microextraction for the simultaneous determination of six synthetic food colourants in soft drinks, sugar-and gelatin-based confectionery by high-performance liquidchromatographyHao Wu a , Jing-bo Guo b , Li-ming Du a , ,

2、Hong Tian a , Cheng-xuan Hao a , Zhi-feng Wang b , Jie-yan Wang aa Analytical and Testing Center, Shanxi Normal University, Shanxi Linfen 041004, PR China bDepartment of Engineering, Shanxi Normal University, Shanxi Linfen 041004, PR Chinaa r t i c l e i n f o Article history:Received 8February 2012

3、Received in revised form 2March 2013Accepted 5March 2013Available online 14March 2013Keywords:Rapid shaking-based ionic liquid dispersive liquid phase microextraction Synthetic food colourantsHigh-performance liquid chromatographya b s t r a c tA novel and simple rapid shaking-based method of ionic

4、liquid dispersive liquid phase microextraction for the determination of six synthetic food colourants (Tartrazine,Amaranth, Sunset Yellow, Allura Red, Ponceau 4R, and Erythrosine in soft drinks, sugar-and gelatin-based confectionery was established. High-performance liquid chromatography coupled wit

5、h an ultraviolet detector was used for the determi-nations. The extraction procedure did not require a dispersive solvent, heat, ultrasonication, or additional chemical reagents. 1-Octyl-3-methylimidazolium tetrauoroborate (C8MIMBF4was dispersed in an aqueous sample solution as ne droplets by manual

6、 shaking, enabling the easier migration of analytes into the ionic liquid phase. Factors such as the C8MIMBF4volume, sample pH, extraction time, and centrifugation time were investigated. Under the optimum experimental conditions, the proposed method showed excellent detection sensitivity with limit

7、s of detection (signal-to-noiseratio =3 within 0.0150.32ng/mL.The method was also successfully used in analysing real food samples. Good spiked recoveries from 95.8%104.5%were obtained.Ó2013Elsevier Ltd. All rights reserved.1. IntroductionColour is a main feature of foods. Its affect on people

8、is not only visual; it is also associated with food variety, quality, and fresh-ness. Food colourants have been used to replace natural food col-our, which can be lost during preparation processes. Colourants are also used to prevent colour changes in the nal product (Berzas, Flores, Llerena, &F

9、arinas, 1999 and provide attractiveness to consumers, particularly children (Hofer &Jenewein, 1997. In recent years, natural food colourants isolated from suitable plants, fungi, or insects have been increasingly used. However, many natural colourants become unstable under processing conditions

10、such as, light, oxygen, and pH. Natural colourants are also more expensive than synthetic ones. The use of synthetic organic dyes has been recognised as the most reliable and economical method of restoring or providing colour to a processed product. However, some of these substances pose potential r

11、isks to human health, especially when consumed in excess.To prevent indiscriminate use, laws and regulations based in toxicological studies on experimental animals and human clinical studies have been developed in many countries. The policies limit the types, purities, uses, and amounts of food colo

12、urants permitted in food and drinks. Consequently, sensitive, accurate, and reliable methods for determining synthetic colourants are required to en-sure food safety.Several analytical techniques have been developed to facilitate the simultaneous determination of various synthetic food colourants. S

13、uch techniques include derivative spectrometry and other spectrophotometric methods related with chemometrics (Al-Degs, 2009; Berzas et al., 1999; Ni &Gong, 1997; Sayar &Özdemir,1998, adsorptive voltammetry (Ni, Bai, &Jin, 1997, differential pulse polarography (Chanlon, Joly-Pottuz,

14、 Chatelut, Vittori, &Cretier, 2005; Combeau, Chatelut, &Vittori, 2002, thin-layer chromatography (Morlock &Oellig, 2009, capillary electro-phoresis (Dossi et al., 2007; Ryvolova, Taborsky, Vrabel, Krasensky, &Preisler, 2007, high-performance liquid chromatography (HPLC(Minioti, Sakel

15、lariou, &Thomaidis, 2007; Pereira Alves, Brum, Branco de Andrade, &Pereira Netto, 2008; Vidotti, Costa, &Oli-veira, 2006; Yoshioka &Ichihashi, 2008, as well as ion chromatog-raphy (Chen, Mou, Hou, Riviello, &Ni, 1998.A recently proposed method, dispersive liquidliquid microex-tra

16、ction (DLLME(Rezaee et al., 2006, is based on the formation of a turbid solution by the rapid injection of a mixture containing0308-8146/$-see front matter Ó2013Elsevier Ltd. All rights reserved. /10.1016/j.foodchem.2013.03.015Corresponding author. Tel./fax:+863572057969.E-mail

17、address:lmd(L.-m.Du.extraction and disperser solvents into an aqueous solution. The extraction solvent is dispersed into the aqueous sample as very ne droplets, enabling the analytes to transfer easily to the extrac-tion solvent. When extraction equilibrium is achieved, phase separation is performed

18、 by centrifugation and the enriched ana-lytes in the sediment phase can be determined. Compared with other microextraction methods, this technique is more convenient, simple, and requires less expensive devices. More importantly, DLLME can be applied under batch conditions and extraction can be comp

19、leted in several seconds, resulting in faster extraction and shorter analytical time.Room temperature ionic liquids (RTILsare a group of new or-ganic salts consisting of organic cations and various anions that are liquid at room temperature. RTILs have been used as extraction solvents in place of or

20、ganic solvents because of their unique phys-icochemical properties, such as negligible vapour pressure, misci-bility with water and organic solvents, good solubility in organic and inorganic compounds, and high thermal stability as well as being environmentally benign (Pandey, 2006; Poole &Poole

21、, 2010. DLLME based on ion liquids (ILs(IL-DLLMEwas introduced by Zhou et al. in 2008(Zhou, Bai, Xie, &Xiao, 2008. This approach needs an organic solvent as the dispersive solvent and heat to aid the complete dispersion of a water-immiscible IL into the aqueous phase, then sedimentation by cooli

22、ng with ice water. Extraction re-quires a specic heating time and cooling process, which is rela-tively time and energy consuming. To improve the extraction performance of temperature-controlled DLLME, ultrasound is used to disperse the IL extraction solvent (Zhou, Zhang, &Xiao, 2009, but the co

23、oling process and dispersive organic solvent are still needed to obtain a turbid solution. Then Yao and Anderson (2009reported a method for in situ IL formation DLLME, wherein the hydrophilic IL is completely dissolved in the aqueous phase and an ion-exchange reagent is added to form a water-immisci

24、ble IL. Although this method overcomes the weaknesses described above, the addition of excess ion-exchange reagent is required, which complicates the method.In the current work, a simple and efcient manual shaking-based method of IL-DLLME was developed. The procedure does not require a dispersive so

25、lvent, heat, ultrasonication, or additional chemical reagents, in contrast to conventional IL-DLLME. IL (C8MIMBF4was dispersed in an aqueous solution as ne drop-lets by manual shaking, promoting migration of the analytes to the ionic liquid phase, then coupled with HPLC-ultraviolet (UVspec-trophotom

26、etry determination. The effects of various experimental parameters, including the C8MIMBF4volume, sample pH, extraction time, and centrifugation time, have been investigated and optimised for the extraction of six synthetic food colourants. 2. Materials and methods2.1. Reagents and standardsThe stan

27、dard stock solutions of the colourants Tartrazine (TAR; C.I. Food Yellow 4; 0.5mg/mL,Amaranth (AMA;C.I. Food Red 9; 0.5mg/mL,Sunset Yellow (SUN;C.I. Food Yellow 3; 0.5mg/mL, Allura Red (ALL;C.I. Food Red 17; 1.0mg/mL,Ponceau 4R (PON; C.I. Food Red 7; 0.5mg/mL,and Erythrosine (ERY;C.I. Food Red 14; 0

28、.1mg/mLwere obtained from the National Research Center for Certied Reference Materials (Beijing,China. The mixed stan-dard solutions containing all colourants at 0.05mg/mLwas prepared by mixing and dilution of appropriate aliquots from standard stock solution of each substance. Working solutions wer

29、e prepared by appropriate dilutions of the mixed standard solutions with water. HPLC-grade methanol and acetonitrile were purchased from Tianjin Guangfu Fine Chemical Research Institute (Tianjin,China. 1-Octyl-3-methylimidazolium tetrauoroborate (C8MIMBF4, 1-hexyl-3-methylimidazolium chloride (C6MIM

30、Cl,and 1-oc-tyl-3-methylimidazolium chloride (C8MIMClwere obtained from Shanghai Cheng Jie Chemical Co., Ltd. (Shanghai,China. Milli-Q water (Millipore,Bedford, MA, USA was used throughout the study. All other reagents were analytical grade and were pur-chased from Tianjin Kemiou Chemical Reagent Co

31、., Ltd. (Tianjin, China. All solutions prepared for HPLC were ltered through 0.45l m membranes before use.2.2. InstrumentsThe chromatography equipment was a 1525binary HPLC pump and a 2489dual k UV detector from Waters (WatersCorporation, USA. The Waters Breeze software was used to control the instr

32、u-ments and acquire data. The chromatographic separation of the analytes was carried out on a Gemini C18column (5l m; 4.6mm Â250mm; Phenomenex, Torrance, CA, USA. A pH meter (ModelpHS-3C, Shanghai Tianda Apparatus Co., Ltd., China was used for pH adjustment. A centrifuge Model TDZ4-WS (XiangYi

33、Centrifuge Instrument Co., Ltd., China was employed to accelerate the phase-separation process.2.3. Preparation of the sample solutionAll samples, including soft drink, sugar-based and gelatin-based confectionery, were obtained from a local market. Appropriate amounts (0.32.5g of the samples were di

34、ssolved in 25mL of water. The carbonated drinks were degassed by ultrasonication for 5min. A warming process (50°C, 30min was used for the complete dissolution of the sugar-based and gelatin-based confec-tionery. Samples were diluted to 50mL in a volumetric ask with an acetate buffer solution (

35、0.2mol/L,pH 5.0. These solutions were ltered through a folded Xinhua paper lter (No.102, and the l-trate was collected after discarding the rst 15mL.2.4. Extraction procedureA homogeneous sample solution (10.0mL containing the ana-lytes was placed in a 15mL screw-cap conical-bottom graduated plastic

36、 centrifugal tube. Using a 500l L syringe, 350l L of RTIL was injected into the sample solution. Manual shaking (30times in 20s resulted in the formation of a turbid solution, which was centrifuged for 8min at a rate of 3500rpm (1685g . The upper aqueous solution was removed using a pipette, and the

37、 volume of residual IL was almost 180l L. Methanol was added to the IL res-idue enriched with analytes to obtain a volume of 300l L. Using a 25l L HPLC microsyringe, 10l L of the enriched solution was in-jected directly into the HPLC system. All experiments were per-formed in triplicate. The syringe

38、 was rinsed with methanol and acetonitrile multiple times to remove residual analytes and IL. 2.5. Interference experimentsThe interferences were studied by analysing 10mL solution containing 100ng mL À1colourants and other chemical species at different concentrations (0.1100l g mL À1, acc

39、ording to the rec-ommended extraction procedure. Tolerance limit of each species was taken as the largest amount yielding an error in the determi-nation of the analyte not exceeding 5%.2.6. Recovery and data handlingRecovery evaluations were performed by spiking known amounts of the colours into the

40、 samples before processing and comparing the results with those from the same samples priorH. Wu et al. /Food Chemistry 141(2013182186183 spiking. Recoveries were estimatedtions and expressed as percentages.Final treatment of data and and recovery were performed using soft Excel.2.7. Chromatographic

41、 conditionsThe ow rate of the mobile min. The sample injection volume of the column was controlled at 30tained 0.1mol/Lammonium acetate justed by 10mol/Lsodium methanolacetonitrile (30:70,v/v.5%50%B (020min followed by detection wavelength was set at 430TAR and 510nm for the other grams of the post-

42、extraction mixed tions are shown in Fig. 1. 3. Results and discussion 3.1. Comparison of ionic liquidIn the current study, ve MIMBF4,thanesulfonylimide(C6MIMNTfzolium hexauorophosphate (C6dazolium and 1-octyl-3-methylimidazolium MIMPF6were investigated. A by manual shaking (Fig. 2a after The other c

43、ommonly used ILs were DLLME method (Yao &Anderson, C6MIMCland C8MIMClwere also used. The ion-exchange re-agents were lithium bis(triuoromethanesulfonyimide(LiNTf2 and sodium hexauorophosphate (NaPF6. When 350l L of water-miscible IL and 0.35g of salt were added to the aqueous phase, immiscible I

44、Ls C6MIMNTf2,C6MIMPF6,C8-MIMNTf2,and C8MIMPF6were formed. At the same time, a turbid solution with ne microdroplets was also formed, as shown in Fig. 2a. After 8min of centrifugation at 3500rpm, the IL phase was well separated from the aqueous phase, as shown in Fig. 2b.C8MIMBF4was found to have the

45、 best extraction efciency of the ve ILs for the articial colours. 3.2. Effect of the IL volumeThe amount of C8MIMBF4used in the preconcentration pro-cedure is a critical factor for obtaining a high extraction perfor-mance. Therefore, the extraction system was carefully studied to determine the lowes

46、t IL-phase volume necessary for achieving the best extraction. The effect of C8MIMBF4was studied within the range of 250450l L. Fig. 3shows that with increased amount of C8MIMBF4,the peak area increased and reached a constant184 value when C8MIMBF4exceeded 350l L, except for TAR. Further increases i

47、n IL reduced the enrichment factor achieved. Therefore, 350l L of C8MIMBF4was used in the subsequent experiments. 3.3. Effect of sample pHThe effects of pH on the extraction were studied within the pH range of 0.712using hydrochloric acid and sodium hydroxide, and the results are shown in Fig. 4. Th

48、e extraction efciency of all col-ourants remained relatively constant over the pH range of 211. However, the extraction recovery decreased with further de-creased or increased pH. Considering most foods are weak acids or neutral, pH 5.0was used in all subsequent experiments.3.4. Effect of extraction

49、 timeA turbid solution was easily formed at room temperature (25±1°C; thus, the equilibration temperature in the extraction process was set at room temperature. Liquid-phase microextrac-tion is a time-dependent process. Consequently, the effect of the extraction time was examined within th

50、e range of 020min at room temperature. In this experiment, extraction time was from the IL dispersion into the solution after manual shaking to before the initiation of centrifugation. The extraction time had no signi-cant effect on the extraction efciency. Therefore, to keep the anal-ysis time as s

51、hort as possible, the cloudy solution was centrifuged immediately after the preparation at room temperature.In comparison with the other reported IL-DLLME methods, such as temperature-controlled (30min (Zhou et al., 2008, ultrasound-assisted (35min (Zhou et al., 2009 and in situ solvent formation (3

52、0s (Yao &Anderson, 2009 IL-DLLME, the rapid shaking-based IL-DLLME (20s has a signicantly shorter operation time for extraction.3.5. Effect of centrifuge timeCentrifugation, which controls phase separation, is a crucial step in the proposed method. The nal performance benets from a full phase se

53、paration. To achieve the best extraction efciency, centrifugation times within the range of 614min were examined at a rate of 3500rpm. At 8min, extraction recovery became con-stant, indicating the complete transfer of the IL phase. Therefore, the optimum centrifugation time was determined as 8min. 3

54、.6. Inuence of interfering substancesThe selectivity of the proposed method was studied using vari-ous chemical species that commonly interfere in the determination of colourants. The tolerance limit was dened as the concentration of a substance causing less than ±5%relative error for the six c

55、ol-ourants using the proposed method. The samples contained a xed amount of colourants (100ng ml À1 and their effect was deter-mined using the proposed method, as described in Section 2.5. The results indicated that Na +, K +, NH 4+,Ca 2+, Zn 2+, Mg 2+, Mn 2+, Table 1Calibration equations, line

56、ar range, limits of detection, limits of quantication and coefcients of determination (R 2 of all colourants (n =11.Colourant Calibration equation (ng/mLLinear range (ng/mLDetection limit (ng/mLQuantication limit (ng/mLR 2Tartrazine y =400x +21400.52000995 Amaranth y =5816x +3630.053000.01

57、70.050.9999 Sunset Yellow y =6268x +2130.053000.0150.050.9999 Allura Red y =3953x +18450.053000.0160.050.9998 Ponceau 4R y =7170x +930.053000.0150.050.9999 Erythrosine y =540x +12241.020000.321.00.9997Table 2Determination of colourants in food samples from the local market.Sample Colourant Concentra

58、tion before extraction (ng/mLSpiked (ng/mLRecovery a (%Content of colourant (mg/kgCarbonated drink Tartrazine 170.310095.8±2.78.5Sunset Yellow 11.410.098.3±1.60.57Fruit-avored drink Amaranth 25.420.0101.2±1.20.50Ponceau 4R 54.150.099.5±1.81.08Fruit-avored candy Tartrazine 123.710

59、0100.8±2.19.7Allura Red 8.010.098.5±1.90.63Lollipop 1Tartrazine 158.610096.7±1.812.4Sunset Yellow 1.72.0102.7±1.50.13Erythrosine 147.4100101.7±1.611.5Lollipop 2Allura Red 3.94.097.0±1.30.65Ponceau 4R 1.21.0104.5±1.80.20Lactic acid jelly Allura Red 9.21099.8±1.50.92Fruit-avored jelly Tartrazine 83.610096.4±1.61.67Allura Red 0.531.0103.1±2.50.01Ponceau 4R 8.010.099.4±1.70.17a Mean ±standard deviation (n =5.H. Wu et al. /Food Chemistry 141(201318218