845
J. Sep. Sci. 2014, 37, 845–852
Jiaying Xue Huichen Li Fengmao Liu Wenqing Jiang Xiaochu Chen Department of Applied Chemistry, College of Science, China Agricultural University, Beijing, P. R. China Received November 14, 2013 Revised January 18, 2014 Accepted January 19, 2014
Research Article
Determination of strobilurin fungicides in cotton seed by combination of acetonitrile extraction and dispersive liquid−liquid microextraction coupled with gas chromatography The simultaneous determination of four strobilurin fungicides (picoxystrobin, kresoximmethyl, trifloxystrobin, and azoxystrobin) in cotton seed by combining acetonitrile extraction and dispersive liquid−liquid microextraction was developed prior to GC with electron capture detection. Several factors, including the type and volume of the extraction and dispersive solvents, extraction condition and time, and salt addition, were optimized. The analytes were extracted with acetonitrile from cotton seed and the clean-up was carried out by primary secondary amine. Afterwards, 60 L of n-hexane/toluene (1:1, v/v) with a lower density than water was mixed with 1 mL of the acetonitrile extract, then the mixture was injected into 7 mL of distilled water. A 0.1 mL pipette was used to collect a few microliters of n-hexane/toluene from the top of the aqueous solution. The enrichment factors of the analytes ranged from 36 to 67. The LODs were in the range of 0.1 × 10−3 −2 × 10−3 mg/kg. The relative recoveries varied from 87.7 to 95.2% with RSDs of 4.1−8.5% for the four fungicides. The good performance of the method, compared with the conventional pretreatments, has demonstrated it is suitable for determining low concentrations of strobilurin fungicide residues in cotton seed. Keywords: Acetonitrile extraction / Cotton seed / Dispersive liquid−liquid microextraction / Strobilurin fungicides DOI 10.1002/jssc.201301223
Additional supporting information may be found in the online version of this article at the publisher’s web-site
1 Introduction Cotton is one of the most important fiber and economic crops [1], and the seed of the cotton plant can provide cotton seed oil, an important type of edible oil [2]. Strobilurin fungicides are a new class of synthetic pesticides, which have been widely used to control powdery mildew and scab in vegetables, fruits, and cotton crops because of their advantages of high bactericidal activity, high selectivity, and low dosage rates. However, the application of strobilurin fungicides on cotton crops can leave residues which may increase the potential risks to humans. For health protection, the European commission [EU pesCorrespondence: Professor Fengmao Liu, Department of Applied Chemistry, College of Science, China Agricultural University, Beijing 100193, P. R. China E-mail: [email protected] Fax: 0086-10-62733620
Abbreviations: DLLME, dispersive liquid−liquid microextraction; DSPE, dispersive solid-phase extraction; ECD, electron capture detection; PSA, primary secondary amine; EF, enrichment factor; ER, extraction recovery; RR, relative recovery; QuEChERS, quick, easy, cheap, effective, rugged, and safe C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
ticides database, 2013. http://ec.europa.eu/sancopesticides/ public/index.cfm?event=commodity.selection] has established maximum residue limits in cotton seed for health protection (Supporting Information Table S1). Maintaining the strobilurin fungicide residues as low as possible in commodities is an important quality criterion in market monitoring and international trade. Cotton seed comprises lipids, proteins, fibers, and fatty acids, which may interfere with the determination of the analytes of interest, thus it is considered to be a complex matrix. Generally, isolation of pesticides from matrices containing high-content fat requires relatively complicated sample pretreatments, including liquid−liquid extraction [2, 3], gel permeation chromatography [4], low-temperature precipitation [5–7], and SPE [8], which are time consuming and need large amounts of solvents. To make the pretreatments simple, several QuEChERS-based (quick, easy, cheap, effective, rugged, and safe) methods have been developed for the analysis of pesticide residues in oily matrices [9, 10]. However, these efforts presented relatively high limits by direct injection without preconcentration of the analytes from extracts, Colour Online: See the article online to view Fig. 1 in colour. www.jss-journal.com
846
J. Sep. Sci. 2014, 37, 845–852
J. Xue et al.
especially for the targets that had high LODs in chromatography process. Dispersive liquid−liquid microextraction (DLLME) is a novel mode of liquid-phase microextraction technique [11] with the outstanding merits of simplicity of operation, rapidity, less time and cost, high recovery, and enrichment factor (EF), making it widely acceptable for extraction and preconcentration of different target compounds in aqueous samples [11–17]. Despite the well-demonstrated quality of state-of-the-art DLLME, several studies used DLLME to analyze pesticides in a range of solid samples, including apples [18–20], grapes [21,22], bananas [23,24], tomatoes [25,26], watermelons, and cucumbers [27–29], few reports have developed DLLME methods to cope with the determination of targets in soils [30], sediments [31, 32], pharmaceutical samples [33], tea [34, 35], and herbal medicines [36]. The capacity for conventional DLLME, however, is limited in the analysis of pesticide residues in cotton seed because of unique chemical or nutritional compositions of fatty commodities in seeds [USDA, 2007. http://ndb.nal.usda.gov/ ndb/foods/list], such as high lipid and protein contents which may interfere with the extraction and detection of the target compounds. An additional drawback of DLLME is the necessity of using an extraction solvent having a density greater than that of water [37], since the number of organic solvents meeting this requirement is relatively small, and the majority of these extraction solvents are halogenated hydrocarbons that are environmentally hazardous. To solve these problems, we attempted to (i) combine one or more techniques to achieve higher sensitivity, and (ii) use solvents with lower density than that of water for the application in the analysis of cotton seed. In this study, the simultaneous determination of four strobilurin fungicides (picoxystrobin, kresoxim-methyl, trifloxystrobin, and azoxystrobin) in cotton seed, a popular oily commodity, by combining acetonitrile extraction with DLLME was developed prior to GC with electron capture detection (ECD) analysis. The effects of various experimental parameters including the type and volume of the extraction solvent and dispersive solvent, extraction condition and time, and salt addition were optimized. Under the optimum conditions, acetonitrile was used to extract the pesticides from cotton seed samples and the clean-up was carried out by primary secondary amine (PSA). In the DLLME procedure, n-hexane/toluene (1:1, v/v) with a lower density than water was considered to be an appropriate extraction solvent. It is noted that collecting n-hexane/toluene from the top of the mixed n-hexane/toluene/acetonitrile/water system was difficult since the n-hexane/toluene mixture floated on the surface of the aqueous phase and could not solidify in an ice bath. One possible way of enabling the application of low-density solvents in DLLME is the use of special extraction devices [38]. A novelty of the proposed approach highlighted the use of a 0.1 mL pipette for collecting n-hexane/toluene from the top of the aqueous solution instead of designing a special device, making the method simple, rapid, and cheap. The performance of the proposed method was evaluated by comparing C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
the results with those obtained from the QuEChERS-based and SPE techniques. This combination method with high sensitivity, good accuracy, precision, and separation shows that it is a reliable technique for the analysis of low residue levels of the selected strobilurin fungicides in cotton seed samples.
2 Materials and methods 2.1 Reagents and materials All HPLC-grade solvents including isooctane, cyclohexane, n-hexane, toluene, xylene, acetonitrile, acetone, ethyl acetate, and methanol were purchased from Dikma, USA. Distilled water was obtained using a Milli-Q water purification system. Sodium chloride (analytical-reagent grade, Merck) was used to adjust the ionic strength of aqueous solutions. PSA was obtained from Agela Technologies, China. An RJ-TDL-40B low-speed desktop centrifuge was purchased from Jiangsu Ruijiang, China. The shaker (HZQ-C) used was from Harbin Donglian Electron Technology Exploiter, China. The vortex meter and hand centrifuge were from Jiangsu Qilinbeier, China. A 0.1 mL pipette was supplied by Beijing Guangdahengyi, China. Four strobilurin pesticides (picoxystrobin, kresoximmethyl, trifloxystrobin, and azoxystrobin) of analytical standard grade had a purity of >95% and were purchased from Standard Technology Development (Beijing, China). The whole cotton samples provided by Institute of plant protection in Anhui (China), were shelled into cotton seeds, and the seed samples were pulverized and filtered through a 250 m sieve before the analysis. 2.2 Preparation of standards Individual 1000 mg/L stock solutions were prepared by dissolving each pesticide in acetonitrile. For daily preparation of working solutions, two composite stock standard solutions were prepared in acetonitrile at a concentration of 10 mg/L for picoxystrobin and azoxystrobin, and 50 mg/L for kresoximmethyl and trifloxystrobin. All solutions were stored at −20⬚C in the dark. 2.3 Sample preparation The samples were homogenized for 10 s. For recovery determination, the portion samples (2 g, weighed to a precision of 0.01 g) were spiked by the addition of the standard stock solutions at three levels. The spiked samples were allowed to stand for a few minutes for the spiking solutions to penetrate the matrix. For the acetonitrile extraction procedure: The homogenized sample (2 g, weighed to a precision of 0.01 g) was placed in a 15 mL centrifuge tube. A total of 5 mL of acetonitrile was added and the mixture was shaken for 20 min. After that, the solution was centrifuged at 3800 rpm for 5 min to separate the www.jss-journal.com
Sample Preparation
J. Sep. Sci. 2014, 37, 845–852
fine solid particles of cotton seed from the solution. Then 1.2 mL of the extract was transferred for clean-up by mixing with 50 mg PSA for the dispersive solid-phase extraction (DSPE) procedure. The supernatant was collected after vortexing for 1 min and high-speed centrifugation at 10 000 rpm; the sample was then filtered through a 0.22 m organic system filter prior to the DLLME step. For the DLLME procedure: 1 mL of the supernatant acetonitrile solution that acted as the dispersive solvent for the DLLME, was mixed with 60 L of n-hexane/toluene (1:1, v/v) (as the extraction solvent). The acetonitrile/n-hexane/toluene was transferred rapidly into a 15 mL screw-cap plastic tube with conical bottom containing 7 mL of distilled water, where a cloudy solution (n-hexane/toluene/acetonitrile/water) was formed. Next, the tube was vigorously shaken on a vortex meter immediately for 2 min. After centrifugation at 3800 rpm for 5 min, there was a dispersed fine droplet of n-hexane/toluene floating on the upper of the aqueous phase due to its lower density than that of water. Then, a 0.1 mL pipette was used to collect a few microliters of n-hexane/toluene for overcoming the indistinct band formed in a relatively wide plastic tube. Approximately 0.1 mL of the upper of the mixed solution was removed into the narrow portion of the pipette, then the partition between nhexane/toluene and the aqueous phase was observed in this narrow portion. Lastly, the lower aqueous layer could be readily released and the upper n-hexane/toluene layer was transferred into a 250 L conical vial prior to the GC−ECD analysis. The microextraction procedure is shown as a schematic in Supporting Information Fig. S1. 2.4 GC analysis Chromatographic analysis was performed on an Agilent Technologies 6890N network GC system (Agilent Technologies, CA, USA) equipped with an ECD detector, an Agilent 7683 series autosampler and a 30 m × 0.25 mm DB-5 fusedsilica capillary column (0.25 m film thickness). The injector and detector were held at 260 and 290⬚C, respectively. The carrier gas was helium at a flow rate of 1.0 mL/min. The oven temperature was initially at 180⬚C for 1 min, increased to 230⬚C at a rate of 8⬚C/min and held for 2 min, then increased to 250⬚C at a rate of 3⬚C/min and held for 1 min, and finally increased at a rate of 20⬚C/min to 270⬚C and kept for 9 min. The sample injection volume was 2 L and the injection mode was splitless. 2.5 Calculation of EF, extraction recovery (ER), and relative recovery (RR) EF is expressed as the ratio of the analyte concentration in the floated phase (Cflo ) to the initial concentration of analyte in the acetonitrile solution (C0 ): EF =
Cflo C0
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
(1)
847
ER is defined as the percentage of total analyte amount extracted to the extraction phase and can be calculated according to the following equations: ER(%) =
Cflo × Vflo × 100% C0 × V0
(2)
where Vflo and V0 are volumes of the extraction solvent (the floated phase) and the acetonitrile phase in DLLME, respectively. The RR can be calculated from the following equation: RR(%) =
Cmeasured − Creal × 100% Cadded
(3)
where Cmeasured is the concentration of analyte measured after fortification of the pesticides, Creal is the concentration of analyte measured before the fortification, and Cadded is the level of fortification of the pesticides.
3 Results and discussion In the experiment, 2 g of blank cotton seed samples spiked at 0.1 mg/kg each of the four strobilurin fungicides was used to study the performance under different experimental parameters. All the experiments were performed in four replicates and the average of the results was used for evaluation of performance of different factors.
3.1 Selection of the extraction solvent (as the dispersive solvent in DLLME) for cotton seed samples QuEChERS, is an emerging sample preparation method that is becoming increasingly popular in the area of multiresidue analysis in food and agricultural products [39]. In QuEChERS, matrices with low water content need the addition of water to the sample prior to extraction with acetonitrile [40]. However, the lipid fraction in cotton seed diffused to the solvent with the presence of water, resulting in a significant reduction of supernatant, and the addition of NaCl could not achieve the partitioning. Thus, the cotton seed sample was extracted with organic solvent alone. In this procedure, the solvent plays two important roles: (i) as the extraction solvent to dissolve the pesticides in cotton seed samples, and (ii) as the dispersive solvent in DLLME. Therefore, it should have extraction capability of the analytes of interest and miscibility with both organic and aqueous phases in DLLME. The selected four strobilurin pesticides are suggested to be partially fat soluble with octanol/water partition coefficient values of 2.5−4.5 (Supporting Information Table S2), which may be contained in lipid or fat of cotton seed, contributing to the low recoveries. To get high extraction efficiency of the targets from cotton seed sample as well as a good dispersive solvent, three organic solvents, namely methanol, acetone, and acetonitrile, were tested. Four spiked replications were extracted with 5 mL of the three extraction solvents, and the average www.jss-journal.com
848
J. Xue et al.
recoveries were in the range of 54.3−63.8 and 86.9−96.2% obtained from methanol and acetonitrile, respectively. However, attempts to use acetone with real cotton seed samples were unsuccessful as serious emulsification occurred when acetone was used as the solvent, resulting in
J. Sep. Sci. 2014, 37, 845–852
Jiaying Xue Huichen Li Fengmao Liu Wenqing Jiang Xiaochu Chen Department of Applied Chemistry, College of Science, China Agricultural University, Beijing, P. R. China Received November 14, 2013 Revised January 18, 2014 Accepted January 19, 2014
Research Article
Determination of strobilurin fungicides in cotton seed by combination of acetonitrile extraction and dispersive liquid−liquid microextraction coupled with gas chromatography The simultaneous determination of four strobilurin fungicides (picoxystrobin, kresoximmethyl, trifloxystrobin, and azoxystrobin) in cotton seed by combining acetonitrile extraction and dispersive liquid−liquid microextraction was developed prior to GC with electron capture detection. Several factors, including the type and volume of the extraction and dispersive solvents, extraction condition and time, and salt addition, were optimized. The analytes were extracted with acetonitrile from cotton seed and the clean-up was carried out by primary secondary amine. Afterwards, 60 L of n-hexane/toluene (1:1, v/v) with a lower density than water was mixed with 1 mL of the acetonitrile extract, then the mixture was injected into 7 mL of distilled water. A 0.1 mL pipette was used to collect a few microliters of n-hexane/toluene from the top of the aqueous solution. The enrichment factors of the analytes ranged from 36 to 67. The LODs were in the range of 0.1 × 10−3 −2 × 10−3 mg/kg. The relative recoveries varied from 87.7 to 95.2% with RSDs of 4.1−8.5% for the four fungicides. The good performance of the method, compared with the conventional pretreatments, has demonstrated it is suitable for determining low concentrations of strobilurin fungicide residues in cotton seed. Keywords: Acetonitrile extraction / Cotton seed / Dispersive liquid−liquid microextraction / Strobilurin fungicides DOI 10.1002/jssc.201301223
Additional supporting information may be found in the online version of this article at the publisher’s web-site
1 Introduction Cotton is one of the most important fiber and economic crops [1], and the seed of the cotton plant can provide cotton seed oil, an important type of edible oil [2]. Strobilurin fungicides are a new class of synthetic pesticides, which have been widely used to control powdery mildew and scab in vegetables, fruits, and cotton crops because of their advantages of high bactericidal activity, high selectivity, and low dosage rates. However, the application of strobilurin fungicides on cotton crops can leave residues which may increase the potential risks to humans. For health protection, the European commission [EU pesCorrespondence: Professor Fengmao Liu, Department of Applied Chemistry, College of Science, China Agricultural University, Beijing 100193, P. R. China E-mail: [email protected] Fax: 0086-10-62733620
Abbreviations: DLLME, dispersive liquid−liquid microextraction; DSPE, dispersive solid-phase extraction; ECD, electron capture detection; PSA, primary secondary amine; EF, enrichment factor; ER, extraction recovery; RR, relative recovery; QuEChERS, quick, easy, cheap, effective, rugged, and safe C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
ticides database, 2013. http://ec.europa.eu/sancopesticides/ public/index.cfm?event=commodity.selection] has established maximum residue limits in cotton seed for health protection (Supporting Information Table S1). Maintaining the strobilurin fungicide residues as low as possible in commodities is an important quality criterion in market monitoring and international trade. Cotton seed comprises lipids, proteins, fibers, and fatty acids, which may interfere with the determination of the analytes of interest, thus it is considered to be a complex matrix. Generally, isolation of pesticides from matrices containing high-content fat requires relatively complicated sample pretreatments, including liquid−liquid extraction [2, 3], gel permeation chromatography [4], low-temperature precipitation [5–7], and SPE [8], which are time consuming and need large amounts of solvents. To make the pretreatments simple, several QuEChERS-based (quick, easy, cheap, effective, rugged, and safe) methods have been developed for the analysis of pesticide residues in oily matrices [9, 10]. However, these efforts presented relatively high limits by direct injection without preconcentration of the analytes from extracts, Colour Online: See the article online to view Fig. 1 in colour. www.jss-journal.com
846
J. Sep. Sci. 2014, 37, 845–852
J. Xue et al.
especially for the targets that had high LODs in chromatography process. Dispersive liquid−liquid microextraction (DLLME) is a novel mode of liquid-phase microextraction technique [11] with the outstanding merits of simplicity of operation, rapidity, less time and cost, high recovery, and enrichment factor (EF), making it widely acceptable for extraction and preconcentration of different target compounds in aqueous samples [11–17]. Despite the well-demonstrated quality of state-of-the-art DLLME, several studies used DLLME to analyze pesticides in a range of solid samples, including apples [18–20], grapes [21,22], bananas [23,24], tomatoes [25,26], watermelons, and cucumbers [27–29], few reports have developed DLLME methods to cope with the determination of targets in soils [30], sediments [31, 32], pharmaceutical samples [33], tea [34, 35], and herbal medicines [36]. The capacity for conventional DLLME, however, is limited in the analysis of pesticide residues in cotton seed because of unique chemical or nutritional compositions of fatty commodities in seeds [USDA, 2007. http://ndb.nal.usda.gov/ ndb/foods/list], such as high lipid and protein contents which may interfere with the extraction and detection of the target compounds. An additional drawback of DLLME is the necessity of using an extraction solvent having a density greater than that of water [37], since the number of organic solvents meeting this requirement is relatively small, and the majority of these extraction solvents are halogenated hydrocarbons that are environmentally hazardous. To solve these problems, we attempted to (i) combine one or more techniques to achieve higher sensitivity, and (ii) use solvents with lower density than that of water for the application in the analysis of cotton seed. In this study, the simultaneous determination of four strobilurin fungicides (picoxystrobin, kresoxim-methyl, trifloxystrobin, and azoxystrobin) in cotton seed, a popular oily commodity, by combining acetonitrile extraction with DLLME was developed prior to GC with electron capture detection (ECD) analysis. The effects of various experimental parameters including the type and volume of the extraction solvent and dispersive solvent, extraction condition and time, and salt addition were optimized. Under the optimum conditions, acetonitrile was used to extract the pesticides from cotton seed samples and the clean-up was carried out by primary secondary amine (PSA). In the DLLME procedure, n-hexane/toluene (1:1, v/v) with a lower density than water was considered to be an appropriate extraction solvent. It is noted that collecting n-hexane/toluene from the top of the mixed n-hexane/toluene/acetonitrile/water system was difficult since the n-hexane/toluene mixture floated on the surface of the aqueous phase and could not solidify in an ice bath. One possible way of enabling the application of low-density solvents in DLLME is the use of special extraction devices [38]. A novelty of the proposed approach highlighted the use of a 0.1 mL pipette for collecting n-hexane/toluene from the top of the aqueous solution instead of designing a special device, making the method simple, rapid, and cheap. The performance of the proposed method was evaluated by comparing C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
the results with those obtained from the QuEChERS-based and SPE techniques. This combination method with high sensitivity, good accuracy, precision, and separation shows that it is a reliable technique for the analysis of low residue levels of the selected strobilurin fungicides in cotton seed samples.
2 Materials and methods 2.1 Reagents and materials All HPLC-grade solvents including isooctane, cyclohexane, n-hexane, toluene, xylene, acetonitrile, acetone, ethyl acetate, and methanol were purchased from Dikma, USA. Distilled water was obtained using a Milli-Q water purification system. Sodium chloride (analytical-reagent grade, Merck) was used to adjust the ionic strength of aqueous solutions. PSA was obtained from Agela Technologies, China. An RJ-TDL-40B low-speed desktop centrifuge was purchased from Jiangsu Ruijiang, China. The shaker (HZQ-C) used was from Harbin Donglian Electron Technology Exploiter, China. The vortex meter and hand centrifuge were from Jiangsu Qilinbeier, China. A 0.1 mL pipette was supplied by Beijing Guangdahengyi, China. Four strobilurin pesticides (picoxystrobin, kresoximmethyl, trifloxystrobin, and azoxystrobin) of analytical standard grade had a purity of >95% and were purchased from Standard Technology Development (Beijing, China). The whole cotton samples provided by Institute of plant protection in Anhui (China), were shelled into cotton seeds, and the seed samples were pulverized and filtered through a 250 m sieve before the analysis. 2.2 Preparation of standards Individual 1000 mg/L stock solutions were prepared by dissolving each pesticide in acetonitrile. For daily preparation of working solutions, two composite stock standard solutions were prepared in acetonitrile at a concentration of 10 mg/L for picoxystrobin and azoxystrobin, and 50 mg/L for kresoximmethyl and trifloxystrobin. All solutions were stored at −20⬚C in the dark. 2.3 Sample preparation The samples were homogenized for 10 s. For recovery determination, the portion samples (2 g, weighed to a precision of 0.01 g) were spiked by the addition of the standard stock solutions at three levels. The spiked samples were allowed to stand for a few minutes for the spiking solutions to penetrate the matrix. For the acetonitrile extraction procedure: The homogenized sample (2 g, weighed to a precision of 0.01 g) was placed in a 15 mL centrifuge tube. A total of 5 mL of acetonitrile was added and the mixture was shaken for 20 min. After that, the solution was centrifuged at 3800 rpm for 5 min to separate the www.jss-journal.com
Sample Preparation
J. Sep. Sci. 2014, 37, 845–852
fine solid particles of cotton seed from the solution. Then 1.2 mL of the extract was transferred for clean-up by mixing with 50 mg PSA for the dispersive solid-phase extraction (DSPE) procedure. The supernatant was collected after vortexing for 1 min and high-speed centrifugation at 10 000 rpm; the sample was then filtered through a 0.22 m organic system filter prior to the DLLME step. For the DLLME procedure: 1 mL of the supernatant acetonitrile solution that acted as the dispersive solvent for the DLLME, was mixed with 60 L of n-hexane/toluene (1:1, v/v) (as the extraction solvent). The acetonitrile/n-hexane/toluene was transferred rapidly into a 15 mL screw-cap plastic tube with conical bottom containing 7 mL of distilled water, where a cloudy solution (n-hexane/toluene/acetonitrile/water) was formed. Next, the tube was vigorously shaken on a vortex meter immediately for 2 min. After centrifugation at 3800 rpm for 5 min, there was a dispersed fine droplet of n-hexane/toluene floating on the upper of the aqueous phase due to its lower density than that of water. Then, a 0.1 mL pipette was used to collect a few microliters of n-hexane/toluene for overcoming the indistinct band formed in a relatively wide plastic tube. Approximately 0.1 mL of the upper of the mixed solution was removed into the narrow portion of the pipette, then the partition between nhexane/toluene and the aqueous phase was observed in this narrow portion. Lastly, the lower aqueous layer could be readily released and the upper n-hexane/toluene layer was transferred into a 250 L conical vial prior to the GC−ECD analysis. The microextraction procedure is shown as a schematic in Supporting Information Fig. S1. 2.4 GC analysis Chromatographic analysis was performed on an Agilent Technologies 6890N network GC system (Agilent Technologies, CA, USA) equipped with an ECD detector, an Agilent 7683 series autosampler and a 30 m × 0.25 mm DB-5 fusedsilica capillary column (0.25 m film thickness). The injector and detector were held at 260 and 290⬚C, respectively. The carrier gas was helium at a flow rate of 1.0 mL/min. The oven temperature was initially at 180⬚C for 1 min, increased to 230⬚C at a rate of 8⬚C/min and held for 2 min, then increased to 250⬚C at a rate of 3⬚C/min and held for 1 min, and finally increased at a rate of 20⬚C/min to 270⬚C and kept for 9 min. The sample injection volume was 2 L and the injection mode was splitless. 2.5 Calculation of EF, extraction recovery (ER), and relative recovery (RR) EF is expressed as the ratio of the analyte concentration in the floated phase (Cflo ) to the initial concentration of analyte in the acetonitrile solution (C0 ): EF =
Cflo C0
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
(1)
847
ER is defined as the percentage of total analyte amount extracted to the extraction phase and can be calculated according to the following equations: ER(%) =
Cflo × Vflo × 100% C0 × V0
(2)
where Vflo and V0 are volumes of the extraction solvent (the floated phase) and the acetonitrile phase in DLLME, respectively. The RR can be calculated from the following equation: RR(%) =
Cmeasured − Creal × 100% Cadded
(3)
where Cmeasured is the concentration of analyte measured after fortification of the pesticides, Creal is the concentration of analyte measured before the fortification, and Cadded is the level of fortification of the pesticides.
3 Results and discussion In the experiment, 2 g of blank cotton seed samples spiked at 0.1 mg/kg each of the four strobilurin fungicides was used to study the performance under different experimental parameters. All the experiments were performed in four replicates and the average of the results was used for evaluation of performance of different factors.
3.1 Selection of the extraction solvent (as the dispersive solvent in DLLME) for cotton seed samples QuEChERS, is an emerging sample preparation method that is becoming increasingly popular in the area of multiresidue analysis in food and agricultural products [39]. In QuEChERS, matrices with low water content need the addition of water to the sample prior to extraction with acetonitrile [40]. However, the lipid fraction in cotton seed diffused to the solvent with the presence of water, resulting in a significant reduction of supernatant, and the addition of NaCl could not achieve the partitioning. Thus, the cotton seed sample was extracted with organic solvent alone. In this procedure, the solvent plays two important roles: (i) as the extraction solvent to dissolve the pesticides in cotton seed samples, and (ii) as the dispersive solvent in DLLME. Therefore, it should have extraction capability of the analytes of interest and miscibility with both organic and aqueous phases in DLLME. The selected four strobilurin pesticides are suggested to be partially fat soluble with octanol/water partition coefficient values of 2.5−4.5 (Supporting Information Table S2), which may be contained in lipid or fat of cotton seed, contributing to the low recoveries. To get high extraction efficiency of the targets from cotton seed sample as well as a good dispersive solvent, three organic solvents, namely methanol, acetone, and acetonitrile, were tested. Four spiked replications were extracted with 5 mL of the three extraction solvents, and the average www.jss-journal.com
848
J. Xue et al.
recoveries were in the range of 54.3−63.8 and 86.9−96.2% obtained from methanol and acetonitrile, respectively. However, attempts to use acetone with real cotton seed samples were unsuccessful as serious emulsification occurred when acetone was used as the solvent, resulting in
