Determination of hymexazol in 26 foods of plant origin by modified QuEChERS method and liquid chromatography tandem-mass spectrometry

Food Chemistry 228 (2017) 411–419

Contents lists available at ScienceDirect

Food Chemistry journal homepage: www.elsevier.com/locate/foodchem

Analytical Methods

Determination of hymexazol in 26 foods of plant origin by modified QuEChERS method and liquid chromatography tandem-mass spectrometry Zejun Jiang, Hui Li, Xiaolin Cao, Pengfei Du, Hua Shao ⇑, Fen Jin, Maojun Jin, Jing Wang ⇑ Institute of Quality Standard and Testing Technology for Agro-Products, Key Laboratory of Agro-product Quality and Safety, Chinese Academy of Agricultural Sciences, Beijing 100081, PR China Research Center of Quality & Standards for Agro-products, Ministry of Agriculture, Beijing 100081, PR China Key Laboratory of Agro-product Quality and Safety, Ministry of Agriculture, Beijing 100081, PR China

a r t i c l e

i n f o

Article history: Received 27 May 2016 Received in revised form 28 January 2017 Accepted 5 February 2017 Available online 8 February 2017 Chemical compounds studied in this article: Hymexazol (PubChem CID: 24781) Magnesium sulfate (PubChem CID: 24083) Ammonium acetate (PubChem CID: 517165) Formic acid (PubChem CID: 284) Acetonitrile (PubChem CID: 6342) Sodium chloride (PubChem CID: 5234)

a b s t r a c t A rapid and sensitive method based on modified QuEChERS for hymexazol determination in 26 plantderived foods using liquid chromatography tandem-mass spectrometry (LC-MS/MS) was developed. Variables affecting the separation (LC column, mobile phase additives) and clean-up effects of various dispersive phases, such as PSA, C18, GCB, MWCNTs, PEP-2, Al2O3, Florisil, and PVPP were evaluated. The method was validated using 26 matrices at spiked levels of 0.01 or 0.02, 0.05, 0.1, and 0.5 mg/kg (0.05, 0.2, 0.5, and 1.0 mg/kg for green tea). Mean recoveries were between 71.2% and 113.8%, and intra and inter-day precisions were below 14.8%. The limit of quantitation for 26 matrices ranged from 10 to 50 lg/kg. Matrix-matched calibration was used. The method was subsequently applied for real sample analysis, and hymexazol was detected in a cucumber (below the LOQ) and was not detected in any other sample. The method is simple and effective, and meets the routine monitoring requirements for hymexazol residue in foods. Ó 2017 Elsevier Ltd. All rights reserved.

Keywords: Hymexazol residue Foods of plant origin QuEChERS LC–MS/MS Matrix-matched calibration

1. Introduction Pesticides play a key role in increasing the yield and quality of agro-products in modern agriculture (Jin, Wang, Shao, & Jin, 2010). However, the use of pesticides leads to residues in agroproducts, and thus has negative effects on the quality and safety

of agro-products (Chen, Dong, Xu, Liu, & Zheng, 2015; Jin et al., 2010). Therefore, it is necessary to develop methods to monitor pesticide residues in foods. Hymexazol, a broad-spectrum fungicide and soil disinfectant, has been used to control various diseases caused by fungi such as Fusaruim, Pythium, or Aphanomyces cochlioides; it also has a

Abbreviations: ACN, acetonitrile; Al2O3, neutral aluminium oxide; C18, octadecylsilyl; CAD, collision gas; CE, collision energy; CUR, curtain gas; CXP, collision cell exit potential; DP, declustering potential; EP, entrance potential; EU, European Union; FA, formic acid; GCB, graphitized carbon black; GC-FPD, gas chromatography with flame photometric detector; GC–MS, gas chromatography–mass spectrometry; GC-NPD, gas chromatography with nitrogen phosphorus detector; GS1, atomization air pressure; GS2, auxiliary gas; HILIC, hydrophilic interaction liquid chromatography; IS, ion spray voltage; LC–ESI-MS/MS, liquid chromatography tandem-mass spectrometry with electrospray ionization; LOQ, limit of quantitation; ME, matrix effects; MgSO4, magnesium sulfate; MRLs, Maximum residue limits; MRM, multiple reaction monitoring; MWCNTs, multi-walled carbon nanotubes; NaCl, sodium chloride; NH4AC, ammonium acetate; PEP-2, polar enhanced polymer-2; PSA, primary secondary amine; PVPP, polyvinylpolypyrrolidone; QuEChERS, Quick Easy Cheap Effective Rugged and Safe; RSD, relative standard deviation; TEM, source temperature; UPLC-DAD, ultra-high performance liquid chromatography with diode-array detector. ⇑ Corresponding authors at: No. 12, Southern Street of Zhongguancun, Haidian District, Beijing 100081, PR China. E-mail addresses: [email protected] (Z. Jiang), [email protected] (H. Li), [email protected] (X. Cao), [email protected] (P. Du), [email protected] (H. Shao), [email protected] (F. Jin), [email protected] (M. Jin), [email protected] (J. Wang). http://dx.doi.org/10.1016/j.foodchem.2017.02.014 0308-8146/Ó 2017 Elsevier Ltd. All rights reserved.

412

Z. Jiang et al. / Food Chemistry 228 (2017) 411–419

synergistic effect on endogenous auxin, which is widely used on sugar beets, rice, and vegetables (Harveson et al., 2007; Myresiotis, Karaoglanidis, Vryzas, & Papadopoulou-Mourkidou, 2012; Vera et al., 2011). As a systemic fungicide, it can be absorbed by roots and leaves and translocated to other plant tissues; thus, it may pose a potential threat to consumer health if it is present at harvest (Qian et al., 2011). Maximum residue limits (MRLs) for hymexazol in many foods have been established by the European Union (EU Pesticides Database., 2016) and Japan (Japan Pesticides Database., 2016), while provisional MRLs in a few foods have been set by China (China Pesticides Database., 2016). Thus, it is necessary to develop a sensitive analytical technique to quantify hymexazol residues. However, owing to its special physical–chemical characteristics, such as low mass, high polarity, and lack of chromophores, it is difficult to identify and quantify hymexazol at trace levels. To date, the number of reports available regarding hymexazol determination in foods is limited. Tamura et al. (2008) analysed hymexazol in agricultural products by gas chromatography with a nitrogen phosphorus detector (GC-NPD), and Tan and Guo (2011) reported a method for detecting hymexazol in rice by GCNPD. Sun et al. (2011) developed a method for monitoring hymexazol in cucumbers using GC with a flame photometric detector (GC-FPD), which included a time-consuming pre-column derivatization step. Moreover, the pretreatments required in GC-based techniques involve multi-stage procedures, large volumes of solvents, and are complicated, time-consuming, and laborious. Viñas, Aguinaga, Campillo, and Hernández-Córdoba (2008a) reported a method for analysing oxazole fungicides including hymexazol in wine and juice by ultra-high performance liquid chromatography with a diode-array detector (UPLC-DAD); however, this method lacks sensitivity and selectivity. Some reports (Viñas, Campillo, Aguinaga, Martínez-Castillo, & Hernández-Córdo ba, 2008b; Viñas, Martínez-Castillo, Campillo, & Hernández-Córdo ba, 2010) also described the use of GC-mass spectrometry (GC– MS) for the analysis of hymexazol in malt beverages, juices, and fruits. Furthermore, Martínez-Domínguez, Romero-González, and Garrido Frenich (2015) reported a LC-Orbitrap-MS method for the determination of many compounds including hymexazol in Ginkgo biloba nutraceuticals. All of the aforementioned methodologies are available for pesticide determination in one or a few kinds of food matrices; however, this does not meet the current needs in pesticide monitoring. Hence, it is imperative to develop a simple, efficient, and reliable method for the detection of hymexazol in foods. Fortunately, liquid chromatography tandem-mass spectrometry with electrospray ionization (LC-ESI-MS/MS) in multiple reaction monitoring (MRM) mode is an effective and powerful tool for monitoring pesticides in various food matrices due to its high selectivity and sensitivity (Stachniuk & Fornal, 2016; Wong et al., 2010). The QuEChERS methodology, developed by Anastassiades, Lehotay, Stajnbaher, and Schenck (2003), has many merits over classical methods such as flexibility (modification of sorbents depending on analyte properties and matrix composition), simplicity, rapidity, low-solvent consumption, and wide analytical scope (González-Curbelo et al., 2015; Payá et al., 2007; Wilkowska & Biziuk, 2011), thus making it an attractive alternative sample preparation procedure for monitoring pesticides in various food matrices (Choi, Kim, Shin, Kim, & Kim, 2015; Sinha, Vasudev, & Vishnu.Vardhana.Rao, 2012). Here, we describe a sensitive and reliable LC-ESI-MS/MS method using a modified QuEChERS approach for monitoring hymexazol residues in 26 foods of plant origin. In the present study, various experimental parameters such as MS/MS conditions, LC column, mobile phase additives, and clean-up effects of different sorbents were compared and optimized in order to obtain sensitive and reliable results. The matrix effects and performance of

the developed method were also evaluated. The method was successfully applied in the analysis of real samples. 2. Materials and methods 2.1. Reagents and chemicals Hymexazol (3-hydroxy-5-methylisoxazole, 90%) was purchased from Sigma (St. Louis, MO, USA). LC-MS grade formic acid (FA) and ammonium acetate (NH4AC) were purchased from Sigma-Aldrich (Steinheim, Germany), and LC grade acetonitrile (ACN) was purchased from Fisher Scientific (Pittsburgh, PA, USA). Graphitized carbon black (GCB, 120–400 mesh), primary secondary amine (PSA, 40–60 lm), octadecylsilyl (C18, 40–60 lm), florisil (100–120 mesh), multi-walled carbon nanotubes (MWCNTs, 10–20 nm), and polar enhanced polymer-2 (PEP-2, 40–60 lm) sorbents were obtained from Agela (Tianjin, China). Neutral aluminium oxide (Al2O3, 100–200 mesh) was obtained from J&K (Beijing, China) and polyvinylpolypyrrolidone (PVPP) was obtained from Solarbio (Beijing, China). Analytical grade anhydrous magnesium sulfate (MgSO4) and sodium chloride (NaCl) were obtained from Sinopharm Chemical Reagent Co. (Shanghai, China). Anhydrous MgSO4 was dried at 500 °C for at least 5 h, cooled naturally, and stored in a desiccator. Ultra-pure water was obtained from a Milli-Q water purification system (Millipore, Bedford, MA, USA). 2.2. Sample preparation Six fruit samples (apple, grape, orange, peach, pear, watermelon), 11 vegetable samples (beetroot, cabbage, celery, rape, cucumber, eggplant, leek, kidney bean, potato, tomato, turnip), 3 cereal samples (maize, rice, wheat), 2 oilseed samples (peanut, soybean), 1 nut sample (almond), 1 tea sample (green tea), 1 edible fungi sample (mushroom), and 1 plant oil sample (soybean oil) were purchased from local super markets. The blank samples for the calibration and spiked experiments were prescreened to confirm that they were residue-free. Only the edible parts were used for analysis. All samples were chopped into small pieces and homogenized using a Multiquick-3 food processor (Braun, NeuIsenburg, Germany), and then stored in a deep freezer at 20 °C until analysis. The QuEChERS methodology developed by Anastassiades et al. (2003) was used with slight modifications. Representative portions (10 g for fruits, vegetables, edible fungi, and plant oil; 5 g for cereals, oilseeds and nuts; 2 g for tea) of well-homogenized samples were weighed into a 50-mL plastic centrifuge tube. Recovery assays were carried out by adding appropriate volumes of the working standard solution to blank samples. Then, the spiked samples were vortexed for 30 s and equilibrated for at least 30 min at room temperature for even hymexazol distribution and interaction with the sample matrix. Afterwards, water (according to the indications in Table 1) was added and the sample was soaked for 15 min before 10 mL ACN was added. The tube was shaken vigorously for 3 min. Then, 4 g MgSO4 and 1 g NaCl were added to the tube and the samples were shaken for 2 min. After centrifuging for 5 min at 5000 rpm, 1.0 mL of the extract was transferred into a 1.5-mL centrifuge tube containing 150 mg MgSO4 and the sorbents (see Table 1). The tube was shaken vigorously for 1 min and centrifuged for 5 min at 6000 rpm. The resulting supernatant was filtered through a 0.22-lm membrane into a glass autosampler vial for LC-MS/MS analysis. 2.3. LC-MS/MS analysis Chromatographic analyses were carried out using an Agilent 1200SL Series HPLC system (Agilent, Waldbronn, Germany)

413

Z. Jiang et al. / Food Chemistry 228 (2017) 411–419 Table 1 Amount of water to be added and amount of cleaning sorbents used to sample. Category

Matrix

Typical water content (%)

Fruit

Apple Grape Orange Peach Pear Watermelon Beetroot Cabbage Celery Cucumber Eggplant Kidney bean Potato Tomato Turnip Rape Leek

Vegetable

a

mL of water to be added to 10 g sample

Amount of sorbents used to 1.0 mL extract

85 80 85 90 85 95 90 90 95 95 90 75 80 95 90 90 85

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

20 mg GCB

50 mg GCB

Edible fungi

Mushroom

90

0

20 mg GCB

Plant oil Nut Oilseed

Soybean oil Almond Peanut Soybean Maize Rice Wheat

<10 <20 <20 <20 10 10 10

5 8.5–5 g sample 8.5–5 g sample 8.5–5 g sample 10–5 g sample 10–5 g sample 10–5 g sample

20 mg GCB and 50 mg Al2O3

Green tea

<10

10–2 g sample

50 mg GCB and 100 mg PVPP

Cereal

Tea

Reference: Anastassiades, M., Kolberg, D. I., Eichhorn, E., Benkenstein, A., Lukacˇevic´, S., Mack, D., Wildgrube, C., Sigalov, I., D. Dörk, D., & Barth, A. (2015). Quick method for the analysis of numerous highly polar pesticides in foods of plant origin via LC-MS/MS involving simultaneous extraction with methanol (QuPPe-Method, Version 8.1). Stuttgart: EU Reference Laboratory for pesticides requiring Single Residue Methods (EURL. SRM). a data based on the QuPPe-Method (Version 8.1).

equipped with a binary pump (G1312B), a column oven (G1316A), and an autosampler (G1367C). The chromatographic separation was performed on an XBridge HILIC column (150 mm  2.1 mm, 3.5 lm) (Waters, Ireland). The mobile phase was composed of 0.1% FA aqueous containing 7.5 mM NH4AC (A) and ACN (B), and was pumped at a flow rate of 0.2 mL/min. Separation was completed using a gradient elution of 0.0 min/5% A, 2.5 min/90% A, 3.0 min/90% A, 3.1 min/5% A, and 5.0 min/5% A. The elution program was achieved in 5.0 min. The column compartment was maintained at 30 °C and the injection volume was 5 lL. Mass spectrometric detection was conducted using an API 5000 tandem quadrupole mass spectrometer (AB SCIEX, Chromos, Singapore) in the positive ionization mode with multiple-reaction monitoring (MRM); the monitoring conditions of the electrospray ionization source (ESI+) were optimized to obtain the highest sensitivity and resolution. Typical parameters were as follows: ion spray voltage (IS), 5500 V; source temperature (TEM), 550 °C; collision gas (CAD), 5 V; curtain gas (CUR), 35 psi; atomization air pressure (GS1), 55 psi; auxiliary gas (GS2), 45 psi; declustering potential (DP), 68 V; entrance potential (EP), 8 V; collision cell exit potential (CXP), 12 V; dwell time, 150 ms. The following MRM transitions were recorded: m/z = 100.1 > 54.1 and m/z = 100.1 > 44.2 for hymexazol with respective collision energies (CE) of 20 and 26 eV. The most abundant MS/MS ion transition m/z 100.1 > 54.1 was used for quantitative analysis.

absence of interferences around the retention time of hymexazol. To assess the linearity, seven-point calibration curves were prepared using matrix-matched standard samples (apple, grape, orange, peach, pear, watermelon, beetroot, cabbage, celery, rape, cucumber, eggplant, leek, kidney bean, potato, tomato, turnip, maize, rice, wheat, peanut, soybean, almond, Sichuan pepper, green tea, shiitake mushroom, and soybean oil) in the concentration range of 0.01–1.0 mg/L. Matrix effects (MEs) for different matrices were evaluated using the ratio between the calibration curve slopes of matrix-matched calibration standards and solvent-based standards. Recovery experiments were carried out by spiking blank samples at four spiking levels of 0.01 or 0.02, 0.05, 0.1, and 0.5 mg/kg (0.05, 0.2, 0.5, and 1.0 mg/kg for green tea) in five replicates for each matrix to estimate the accuracy and precision of the method. The accuracy was calculated using the recovered spiked samples. The precision, expressed as relative standard deviation (RSD), was determined by the intra-day (the same day) and inter-day (three different days) repeatability. The LOQ was set as the lowest validated spike level meeting the acceptable accuracy (70–120%) and precision (RSD 6 20%).

2.4. Method validation

To optimize the mass spectrometry, 1 mg/L standard solution of hymexazol was injected into a mass spectrometer through direct infusion via a syringe pump at a flow rate of 10 lL/min. Under positive ion mode (ESI+), the protonated molecule ion [M+H]+ at m/z 100.1 was acquired from the full-scan MS spectrum. Three abundant fragments at m/z 54.1, 82.0, and 44.2 were obtained from the MS/MS spectrum. Although the intensity of the signal at m/z 82.0 was higher than that at m/z 44.2 in the spiked cucumber

In order to ensure the applicability and reliability of the method for routine analysis, the method was validated for specificity, linearity, matrix effects, accuracy, precision, and limit of quantitation (LOQ) according to the SANCO document (European Commission., 2013). The specificity of the method was investigated by the analysis of spiked and non-spiked blank samples to confirm the

3. Results and discussion 3.1. Mass spectrometry Optimization

414

Z. Jiang et al. / Food Chemistry 228 (2017) 411–419

sample (see Fig. S1 in Supplementary material), the baseline noise of the transition at 100.1 > 82.0 was very high (in other words, S/N ratio was quite low). The reason may be that the signal at m/z 82.0 corresponded to [M H2O+H]+ (non-specific loss), which was more likely subject to matrix interference in real sample analysis. Thus, the transition at 100.1 > 54.1 was selected as the quantitation ion, while the transition at 100.1 > 44.2 was selected as the confirmation ion, which was consistent with a recent publication (Kiljanek et al., 2016).

3.2. Liquid chromatography optimization 3.2.1. Chromatogram column Previous studies (Viñas et al., 2008a; Xu, Cui, Wang, Wang, & Gao, 2015) revealed that satisfactory results could be obtained by using C18 columns for hymexazol analysis. Therefore, the MS C18 column (150 mm  2.1 mm  3.5 lm) was initially selected for LC analysis in this study. However, the C18 column exhibited poor retention (see Fig. S2a in Supplementary material), which was

Fig. 1. Effect of different sorbents on recovery of hymexazol in different matrices spiked at 0.05 mg/kg level (n = 3).

415

Z. Jiang et al. / Food Chemistry 228 (2017) 411–419

inconsistent with previous studies (Viñas et al., 2008a; Xu, Cui, Wang, Wang, & Gao, 2015). Kiljanek et al. (2016) also tried to use several C18 columns for hymexazol analysis and failed to obtain satisfactory results. These different results could possibly be attributed to different brands of C18 columns, which have different column packing types and characteristics. Consequently, the C18 column was discarded for further studies. Then, four types of LC columns used for polar pesticides analysis, namely HYPERCARB (150 mm  2.1 mm  3 lm), PAK ADME (150 mm  2.1 mm  3 lm), HSS T3 (150 mm  2.1 mm  3.5 lm), and XBridge HILIC (150 mm  2.1 mm  3.5 lm), were tested for the separation of hymexazol due to its relatively high polarity and low mass. The results (see Fig. S2b-d in Supplementary material) illustrate that the HYPERCARB, ADME, and T3 columns were not suitable for the separation of hymexazol as broad peaks with long tailing were observed. Fortunately, the HILIC column overcame the early elution, peak broadening, and peak tailing, and provided the best peak shape and signal response. Hence, the XBridge HILIC column was chosen as the optimal column for this study (see Fig. S2e in Supplementary material). 3.2.2. Mobile phase additives In LC-MS/MS analysis, the mobile phase composition (pH value, salt concentration) has a significant impact on the signal response,

Intensity, cps

(a1)

2.16

500

5000

400

4000

300

3000

200

2000

100

1000

0

1

2 3 Time, min

0

4

(b1)

500 Intensity, cps

peak shape, and retention of the target analyte in the LC column, especially in HILIC (Hemström & Irgum, 2006; Jandera, 2011; Kawachi et al., 2011). In our study, the effect of different percentages (0, 0.05, 0.1, 0.2, 0.5%) of FA was investigated in the aqueous phase when acetonitrile used as the organic solvent. As listed in Fig. S3 (in Supplementary material), the retention time of hymexazol shifted slightly, and the MS signal response was the highest when the percentage of FA was 0.1%. Considering the sensitivity, 0.1% FA was chosen. However, baseline drift and irregular peak shape (tailing) were observed when 0.1% FA was used in the aqueous phase. The poor peak shape and shift in retention time may be attributed to electrostatic interactions between the charged stationary phase and analyte. Previous publications (Hemström & Irgum, 2006; Jandera, 2011) revealed that the use of buffered eluents can reduce electrostatic interactions. Therefore, NH4AC was also added to the aqueous phase and the effect of five different amounts (5, 7.5, 10, 20, 50 mM) of NH4AC was also investigated. The baseline stability and peak shape were improved when the concentration of NH4AC was increased from 5 to 50 mM (see Fig. S4 in Supplementary material). However, as the concentration of NH4AC increased from 7.5 to 50 mM, the signal intensity decreased. As a result, 7.5 mM NH4AC was selected. Ultimately, an aqueous solution containing 0.1% FA and 7.5 mM NH4AC (A) and acetonitrile (B) was selected as the optimum mobile phase. A

400

4

2.15

(b2)

8000 6000

200

4000

100

2000 1

2

3

0

4

1

Time, min

2

3

4

Time, min

(c1) Intensity, cps

2 3 Time, min

1.00e4

300

0

1

(a2)

300

2.16

(c2)

1000 800 600

200

400 100 0

200 1

2

3

Time, min

4

0

1

2

3

4

Time, min

Fig. 2. Representative chromatograms of hymexazol in selected matrices: beetroot (a1: blank, a2: spiked at 0.01 mg/kg); soybean oil (b1: blank, b2: spiked at 0.01 mg/kg); green tea (c1: blank, c2: spiked at 0.05 mg/kg).

416

Z. Jiang et al. / Food Chemistry 228 (2017) 411–419

stable and reproducible separation was obtained with high sensitivity and excellent peak shape under the optimized gradient elution program. 3.3. Optimization of clean-up Foods are very complex matrices and contain a variety of interfering components, such as sugars, pigments, organic acids, fats, sterols, and alkaloids. Therefore, the clean-up procedure to remove these co-extractants is necessary and important. The QuEChERS method can be used to remove matrix interferences through a variety of dispersive cleaning sorbents, such as PSA, C18, GCB, MWCNTs, and florisil (González-Curbelo et al., 2015; Wilkowska & Biziuk, 2011). PSA is typically used to remove polar impurities owing to its strong interaction with some sugars, polar pigments, polar organic acids, and fatty acids (Anastassiades et al., 2003). C18 is used for the adsorption nonpolar compounds (Hou et al., 2015; Wang et al., 2015). GCB is used to eliminate planar matrix constituents, such as pigments and sterols (Anastassiades et al., 2003; Hou et al., 2015). MWCNTs are used to reduce pigments (Zhao et al., 2012). PEP-2 is a new polymer sorbent that is used to absorb polar and non-polar co-extracts. First, the clean-up effect of different sorbents (PSA, C18, GCB, MWCNTs, and PEP-2) in spiked blank samples (0.05 mg/kg) was evaluated. As shown in Fig. 1, the recoveries of hymexazol from all samples were only 4.56–33% when 50 mg of PSA was used for 1 mL extract. The poor recovery was possibly attributed to strong electrostatic interactions between PSA and hymexazol, which contains ANHA, @NA, and AOH groups (Anastassiades et al., 2003). Interestingly, Kiljanek et al. (2016) used a mixture of sorbents (PSA and Z-Sep+) for 200 pesticides and their metabolites (including hymexazol) analysis in honeybee samples and Martínez-Domín guez et al. (2015) used a mixture of PSA, GCB, C18, and Z-Sep+ for determination of 250 toxic compounds (including hymexazol) in Ginkgo biloba nutraceuticals, and excellent recoveries of hymexazol were obtained. These findings are ascribed to the fact that the Z-Sep+ and PSA are cooperative relations that contribute to recovering more acidic pesticides (Kiljanek et al., 2016). As shown in

Fig. 1, the recoveries from some matrices were lower than 70% when 50 mg C18 and 50 mg PEP-2 were used for 1 mL extract, illustrating that they were not suitable for clean-up in some food samples because of the poor pigment removal. Acceptable recoveries from all samples (except green tea and leeks) were obtained when 20 mg GCB (73.2–113.3%) and 10 mg MWCNTs (73–116.2%) were used for 1 mL extract, indicating that they were useful to remove matrix interferences. Considering the cost (MWCNTs are relatively expensive) and purification, 20 mg GCB was selected as the optimal sorbent for subsequent studies (Fig. 2). Since leeks, rapes, and green tea are rich in chlorophyll, the clean-up effect of different amounts (30, 40, 50 mg) of GCB for 1 mL extract was optimized. When 50 mg GCB was used, the transparent and nearly colourless extracts (good pigment removal) and satisfactory recoveries (94.4%, 87.9%) were obtained for leeks and rapes; however, the clean-up performance (the final green tea extract was intensely coloured) and recovery (67.2%) were not good for green tea because it is rich in polyphenols and other interfering substances. PVPP is used to eliminate polar interfering constituents in tea, such as polyphenols (Hou et al., 2013, 2015). Therefore, the effect of the combination of GCB (50 mg) and PVPP was tested, and the amount of PVPP was optimized. The colour of the green tea extracts became less intense and the recoveries were acceptable (72.3–94.8%) with increasing amounts of PVPP from 50 to 100 mg. Consequently, a combination of 50 mg GCB and 100 mg PVPP was used as the cleaning sorbent for green tea (1 mL extract). In addition, cereals (maize, rice, and wheat), oilseeds (peanut, soybean), nuts (almond), and plant oil (soybean oil) contain some fats. Florisil and Al2O3 are used to remove fats (Wang et al., 2015). Although the use of GCB (20 mg) alone can achieve desirable recovery, further studies using a co-sorbent, such as Al2O3 and florisil, were carried out to obtain cleaner extracts and to avoid the clogging of the LC system tubing after repeated injections of the extracts. As shown in Fig. 1b, satisfactory recoveries (74.5– 103.7%) were obtained for cereals, oilseeds, nuts, and plant oil when a combination of GCB (20 mg) and Al2O3 (50 mg) was used for 1 mL extract, while poor recoveries (8.8–75.1%) were obtained

Table 2 Calibration equation, correlation coefficient (r2), matrix effect (ME), LOQ, and MRLs of hymexazol for different matrices. Matrix

Calibration equation

r2

ME (%)

LOQ (lg/kg)

MRLs (mg/kg)

Solvent Apple Grape Orange Peach Pear Watermelon Beetroot Cabbage Celery Rape Cucumber Eggplant Leek Kidney bean Potato Tomato Turnip Mushroom Soybean oil Almond Peanut Soybean Maize Rice Wheat Green tea

y = 1.67e6x 1.8e3 y = 1.8e6x 3.7e3 y = 1.54e6x 857 y = 1.31e6x 1.13e3 y = 1.21e6x 2.81e3 y = 1.49e6x 603 y = 1.41e6x 1.3e3 y = 1.12e6x 1.07e3 y = 1.19e6x 1.77e3 y = 1.14e6x+105 y = 8.44e5x+6.6e4 y = 1.45e6x 363 y = 1.22e6x 1.59e3 y = 6.35e5x+2.18e3 y = 8.38e5x 1.46e3 y = 1.12e6x 1.37e3 y = 1.82e6x 2.82e3 y = 1.1e6x+2.32e3 y = 5.87e5x 2.17e3 y = 1.91e6x+1.26e3 y = 9.31e5x 2.05e3 y = 6.57e5x 1.59e3 y = 1.54e6x 2.38e3 y = 7.13e5x 2.13e3 y = 8.33e5x 2.39e3 y = 5.06e5x 769 y = 1.71e5x 334

0.9995 0.9989 0.9993 0.9988 0.9978 0.9990 0.9998 0.9998 0.9995 0.9997 0.9990 0.9994 0.9993 0.9995 0.9988 0.9989 0.9996 0.9993 0.9994 0.9997 0.9982 0.9992 0.9995 0.9982 0.9970 0.9977 0.9991

– 7.78 7.78 21.56 27.54 10.78 15.57 32.93 28.74 31.74 49.46 13.17 26.95 61.98 49.82 32.93 8.98 34.13 64.85 14.37 44.25 60.66 7.78 57.31 50.12 69.70 89.76

– 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 20 20 20 20 20 20 50

– 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(CN, JP) 0.05(EU), 0.1(CN), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(CN, JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 1(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.5(JP) 0.05(EU), 0.02(JP) 0.05(EU), 0.1(CN), 0.5(JP) 0.05(EU), 0.02(JP) 0.05(EU), 0.1(CN), 0.02(JP)

ME (Matrix effect, %) = ((slope obtained from matrix matched standard/slope obtained from solvent-based standard)

1)  100. EU: European Union; JP: Japan; CN: China.

417

Z. Jiang et al. / Food Chemistry 228 (2017) 411–419

when a mixture of GCB (20 mg) and florisil (50 mg) was used for 1 mL extract. Thus, a combination of GCB (20 mg) and Al2O3 (50 mg) was used for maize, rice, wheat, peanut, soybean, almonds, and soybean oil. 3.4. Method validation 3.4.1. Specificity, linearity, and LOQ In the present study, the specificity was found to be satisfactory, as interference peaks were not observed around the retention time (2.15 min) of hymexazol. Furthermore, all matrix-matched calibration curves showed good linearity with correlation coefficients (r2) > 0.997 (Table 2) in the concentration range of 0.01–1.0 mg/L. The LOQ for hymexazol (Table 2) in different matrices ranged from 10 to 20 lg/kg (green tea: 50 lg/kg), and was lower than those of previous reports (Kiljanek et al., 2016;

Sun et al., 2011; Tamura et al., 2008; Tan & Guo, 2011; Viñas et al., 2010). Notably, the LOQ was lower than or equal to the MRLs established for hymexazol by the EU, Japan (except green tea), and China (China Pesticides Database, 2016; EU Pesticides Database, 2016; Japan Pesticides Database, 2016), demonstrating that the proposed method can meet the sensitivity requirements for routine analysis of all food matrices (except green tea in Japan). 3.4.2. Matrix effect Matrix effects (MEs) are the primary shortcoming of LC-ESI-MS/ MS, as they adversely affect the performance in terms of detection capability, selectivity, sensitivity, linearity, accuracy, and precision, and cause quantification errors (Gosetti, Mazzucco, Zampieri, & Gennaro, 2010). In this study, the MEs for different matrices were evaluated by comparing the ratio of the slope of the matrixmatched calibration standard to that of the solvent; the percentage

Table 3 Mean recoveries and relative standard deviation (RSD) of hymexazol from different matrices at four spiked levels. Matrix

Spiked level (mg/kg)

Intra-day mean recoveries (RSD)(%) (n = 5) Day 1

Day 2

Day 3

Apple

0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5

105.2(2.3) 73.4(5.4) 80.8(3.9) 71.5(2.5) 103.3(4.3) 90.3(3.1) 94.6(3.8) 93.4(3.9) 82.5(5.7) 88.0(3.0) 86.5(2.0) 92.4(3.5) 107.2(4.7) 84.2(4.1) 87.2(2.7) 86.4(1.5) 91.1(7.6) 74.9(4.4) 77.8(4.5) 77.1(2.6) 99.1(2.9) 76.7(6.5) 77.6(8.9) 85.4(1.9) 103.8(6.4) 79.5(2.3) 75.7(1.7) 77.4(1.9) 107.4(3.5) 94.6(1.6) 95.9(5.0) 93.1(3.3) 99.5(11.3) 79.0(2.4) 82.0(4.1) 76.3(2.6) 96.1(7.6) 83.8(6.6) 93.1(5.3) 91.3(4.6) 101.1(1.2) 106.6(2.9) 107.3(6.2) 97.09(4.5) 101.4(5.5) 89.6(3.3) 82.0(7.4) 85.8(9.2) 85.2(3.1) 77.4(13.8) 78.2(1.8) 78.0(5.9)

100.7(5.8) 72.7(4.4) 95.0(6.4) 72.6(9.5) 102.0(5.4) 84.9(3.9) 91.2(5.5) 80.2(8.5) 85.1(7.4) 87.4(4.2) 93.7(3.3) 92.0(1.7) 103.7(5.8) 84.2(9.3) 88.4(8.3) 74.2(4.2) 91.8(11.5) 80.0(3.8) 83.2(1.2) 71.6(0.8) 98.9(4.7) 91.7(4.1) 90.8(4.0) 85.7(3.0) 100.6(9.4) 71.2(3.6) 74.0(2.2) 76.3(5.1) 104.0(4.6) 94.3(2.8) 95.4(3.7) 96.4(1.4) 105.5(7.7) 78.4(5.2) 80.4(5.0) 74.5(1.2) 89.0(9.1) 87.7(5.4) 98.4(8.7) 88.6(1.6) 113.8(4.9) 113.6(4.6) 112.3(6.9) 102.7(2.5) 101.8(4.9) 80.4(3.6) 77.7(5.9) 87.1(0.8) 83.6(7.8) 86.2(3.4) 87.4(4.3) 84.0(4.6)

95.1(4.2) 77.2(8.0) 84.2(3.7) 74.0(2.6) 97.3(3.1) 88.5(5.7) 90.4(2.7) 87.2(6.5) 87.2(5.6) 96.8(5.8) 95.2(3.2) 80.9(7.3) 102.1(3.0) 88.7(3.0) 91.6(2.5) 82.2(5.3) 88.0(8.2) 81.8(2.1) 78.1(3.4) 71.8(2.3) 89.4(5.1) 96.4(1.5) 88.8(5.1) 82.2(2.5) 98.3(6.0) 73.5(2.4) 71.6(3.8) 75.1(4.3) 104.4(3.7) 88.7(5.4) 92.8(3.9) 95.2(4.6) 82.6(5.4) 78.5(5.4) 77.1(3.4) 73.1(3.5) 95.2(3.4) 75.9(4.9) 95.7(3.6) 88.5(3.7) 100.0(4.4) 105.8(4.3) 110.8(6.7) 88.6(3.8) 103.0(3.6) 80.0(5.5) 74.2(6.8) 80.8(5.9) 82.4(8.5) 82.2(9.5) 87.2(9.0) 87.2(5.0)

Grape

Orange

Peach

Pear

Watermelon

Beetroot

Cabbage

Celery

Rape

Cucumber

Eggplant

Leek

Inter-day RSD (%) x(n = 15)

Matrix

Spiked level (mg/kg)

Intra-day mean recoveries (RSD)(%) (n = 5) Day 1

Day 2

Day 3

5.9 6.3 8.6 5.6 4.9 4.8 4.4 8.7 6.3 6.5 5.1 7.5 4.8 6.1 5.3 7.4 8.8 5.1 4.4 4.0 6.3 10.6 9.0 3.0 7.3 5.5 3.4 3.9 3.9 4.5 4.2 3.5 13.3 4.2 4.7 3.0 7.4 8.1 6.3 3.6 7.2 5.0 6.4 7.1 4.4 6.7 7.6 6.6 6.5 9.9 7.6 6.7

Kidney bean

0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.01 0.05 0.1 0.5 0.02 0.05 0.1 0.5 0.02 0.05 0.1 0.5 0.02 0.05 0.1 0.5 0.02 0.05 0.1 0.5 0.02 0.05 0.1 0.5 0.02 0.05 0.1 0.5 0.05 0.2 0.5 1.0

103.1(4.5) 85.3(8.7) 93.8(6.1) 103.3(1.3) 105.8(3.7) 99.2(2.4) 97.2(1.1) 89.6(1.5) 98.7(7.3) 100.6(2.0) 82.3(2.1) 88.7(4.5) 91.9(11.8) 92.7(4.2) 89.8(1.9) 93.5(1.0) 97.1(5.4) 93.9(7.7) 107.0(4.8) 92.6(3.4) 94.4(9.3) 97.5(3.7) 100.2(3.8) 101.7(4.3) 106.2(3.7) 102.5(4.6) 87.6(9.7) 91.8(3.2) 98.7(4.9) 93.8(4.2) 90.3(2.3) 93.8(1.4) 82.8(7.4) 102.9(2.5) 109.1(2.2) 103.8(1.5) 99.1(8.1) 89.3(2.4) 77.8(2.1) 86.2(1.5) 105.4(4.7) 93.0(2.6) 80.4(7.4) 74.7(2.2) 101.4(5.2) 95.3(2.3) 79.9(1.7) 95.3(1.1) 95.4(9.7) 89.2(4.4) 80.9(5.8) 84.7(2.7)

106.8(5.8) 92.5(3.4) 89.9(5.2) 96.7(5.5) 106.4(7.5) 98.0(4.3) 101.1(3.3) 94.9(7.4) 90.3(6.6) 99.2(3.1) 73.1(8.1) 87.8(8.9) 94.1(9.2) 85.8(2.8) 82.1(2.7) 79.0(5.2) 95.8(9.8) 96.5(4.9) 101.5(5.0) 87.2(2.4) 94.9(9.3) 93.4(4.4) 79.0(4.2) 84.6(3.7) 104.0(4.2) 107.0(6.0) 97.2(5.5) 90.0(2.7) 100.3(5.9) 99.6(4.7) 86.0(4.1) 85.0(3.6) 90.8(10.3) 97.3(6.5) 106.9(2.0) 107.6(2.9) 96.9(8.3) 90.9(4.9) 91.7(4.0) 88.5(1.6) 103.4(6.4) 111.6(4.2) 101.0(4.9) 86.1(3.5) 90.4(8.2) 99.4(3.8) 75.0(4.1) 84.4(3.2) 83.2(3.9) 91.4(11.2) 77.2(2.6) 80.5(6.2)

106.4(4.1) 89.9(9.8) 93.9(2.8) 95.8(6.2) 105.6(5.0) 94.7(6.8) 96.4(3.4) 89.5(2.8) 87.2(5.4) 96.5(4.8) 77.1(4.3) 80.7(4.6) 86.3(6.3) 95.2(1.8) 89.0(3.7) 97.6(1.7) 92.1(6.4) 92.1(7.3) 103.3(6.6) 92.0(3.4) 93.4(7.9) 86.8(3.1) 89.8(3.0) 87.8(1.8) 95.4(7.2) 113.8(3.9) 105.9(0.9) 97.4(4.9) 98.7(6.6) 96.6(6.1) 92.0(6.6) 93.5(1.4) 84.6(7.8) 101.8(5.8) 106.9(4.2) 106.2(1.6) 100.6(3.8) 103.8(7.0) 96.5(3.5) 81.1(4.7) 97.2(7.4) 104.7(1.9) 94.1(2.0) 77.9(3.8) 88.5(10.1) 94.6(1.1) 76.6(3.9) 83.8(2.4) 86.7(7.7) 82.9(4.0) 80.4(2.8) 79.6(4.3)

Potato

Tomato

Turnip

Mushroom

Soybean oil

Almond

Peanut

Soybean

Maize

Rice

Wheat

Green tea

Inter-day RSD (%) (n = 15) 4.8 7.9 5.0 5.6 5.2 4.9 3.4 5.3 8.2 3.7 6.9 7.4 9.5 5.4 4.9 9.5 7.3 6.5 5.6 4.0 8.2 6.0 10.5 9.0 6.7 6.3 9.7 5.0 5.5 5.3 5.3 5.1 9.0 5.4 2.9 2.5 6.7 8.7 9.8 4.6 6.7 8.2 10.7 6.9 9.6 3.4 4.2 6.6 9.4 14.8 4.3 5.1

418

Z. Jiang et al. / Food Chemistry 228 (2017) 411–419

of ME was then calculated according to the following equation: ME (%) = ((slope obtained from matrix matched standard/slope obtained from solvent-based standard) 1)  100. MEs were classified according to percentage as follows: mild signal suppression or enhancement effect when ME was between 20% and +20%; medium effect when ME was between 50% and 20% or +20% and +50%; and strong signal suppression or enhancement when ME was below 50% or above +50% (Li et al., 2013). As indicated in Table 2, a strong signal suppression ( 89.76% 6 ME 6 50.12%) was observed in 7 matrices (leek, mushroom, maize, peanut, rice, wheat, green tea), a medium effect ( 49.82% 6 ME 6 21.56%) was observed in 11 matrices (orange, peach, beetroot, cabbage, celery, rape, eggplant, kidney bean, potato, turnip, almond), and a mild effect ( 15.17% 6 ME 6 7.78%, 7.78% 6 ME 6 14.37%) was observed in 8 matrices (apple, grape, pear, watermelon, cucumber, tomato, soybean, soybean oil). Thus, matrix-matched calibration standards were used to compensate for the MEs and obtain precise results in all matrices in the present study. 3.4.3. Accuracy and precision Recovery experiments were performed to validate the method by spiking samples in five replicates at four concentration levels of 0.01 or 0.02, 0.05, 0.1, and 0.5 mg/kg (0.05, 0.2, 0.5, and 1.0 mg/kg for green tea). The accuracy was evaluated using the recoveries and the precision was calculated using the RSD of the recoveries of the spiked samples. The intra-day and inter-day precisions were investigated by analysing the spiked samples on the same day and on three different days, respectively. As presented in Table 3, the mean recoveries of hymexazol ranged from 71.2% to 113.8% with intra-day RSDs of 0.8%-13.8% (n = 5) and inter-day RSDs of 2.5%-14.8% (n = 15), which were in the acceptable range for pesticide residue analysis in foods. 3.4.4. Application to actual samples To further demonstrate the applicability, the proposed method was applied to actual samples, which were obtained from local markets and supermarkets in Beijing and Hangzhou. A total of 104 samples were analysed (four samples for each matrix). Hymexazol was detected at 0.0085 mg/kg in a cucumber obtained from a local market in Beijing (Fig. 3), which was lower than the LOQ. Hymexazol residues were not detected in any other samples. The result that all fungicides, including hymexazol, were detected below their LOD in different kinds of fruits from Spain was obtained by Viñas et al. (2010), which is similar to our study. Sun et al. (2011) analysed cucumber samples from a field trial (7, 14, and 21 days after hymexazol application), and found hymexazol residues were lower than 0.5 mg/kg. Another study (Martí nez-Domínguez et al., 2015) revealed that hymexazol was detected

in one sample of 8 Ginkgo biloba nutraceuticals from Spain, and the concentration of hymexazol was at 0.01 mg/kg, which was lower than the MRL (0.05 mg/kg) in ginkgo leaves set by the EU. The presence of hymexazol at low levels in foods of plant origin does not pose a threat to the consumer because they are below the MRLs set by EU, Japan or China (China Pesticides Database, 2016; EU Pesticides Database, 2016; Japan Pesticides Database, 2016). 4. Conclusion In the present study, a rapid, simple, sensitive, and dependable method was developed and validated for the determination of hymexazol residues in 26 foods of plant origin using LC-MS/MS and a modified QuEChERS procedure. Various experimental parameters such as the MS/MS conditions, LC column, mobile phase additives, and clean-up effects of different sorbents were compared and optimized to obtain a method with high sensitivity and satisfactory results. Good analytical results, including specificity, linearity, LOQ, accuracy, and precision, were obtained for all matrices. The mean recoveries ranged from 71.2% to 113.8% with intra-day RSDs of 0.8–13.8% (n = 5) and inter-day RSDs of 2.5–14.8% (n = 15), which satisfied the requirements for pesticide residue analysis. Decent linearities (r2 > 0.997) were obtained for all matrices. Matrixmatched calibration standard was applied to compensate for the MEs observed in most matrices. The LOQ for 26 matrices ranged from 10 to 50 lg/kg, which was lower than or equal to the MRLs established for hymexazol in EU, Japan (except green tea), and China. The proposed method was successfully applied in the analysis of real samples. Hymexazol was detected in a cucumber (0.0085 mg/kg, below the LOQ), and hymexazol residues were not detected in any other samples. In conclusion, this method was sensitive and reliable for the routine monitoring of hymexazol in foods of plant origin. In addition, this study may be useful for setting up MRLs for more foods in China, and may facilitate registration of hymexazol in other crops (including minor crops). However, the limitations of the current study should also be taken into account. More food samples and continuous monitoring of hymexazol residues are required for better characterization of potential human exposure. Other pesticides that are typically used with hymexazol, such as metalaxyl, metalaxyl-M, thiram, and isoprothiolane, should also be monitored to ensure the quality and safety of foods of plant origin. Last but not least, this method was not applicable for all types of matrices; specifically, it could not be applied with spices or coffee beans. However, it could be applied with 26 different types of foods of plant origin, which is a significant advance over previous reports. Conflict of interest statement

2.15 The authors declare that they have no conflicts of interest.

Intensity, cps

5000

cucumber-BJ-M-01

Acknowledgments

4000 This work was supported by the Special Program for Basic Work of the Ministry of Science and Technology of China (2013FY110100), and the Agricultural Science and Technology Innovation Program of the Chinese Academy of Agricultural Sciences (CAAS-ASTIP-2015-IQSTAP, CAAS-ASTIP-2016-IQSTAP).

3000 2000 1000 0

1

2

3

4

Time, min Fig. 3. Chromatogram of hymexazol in a cucumber sample from a local market in Beijing, China.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foodchem.2017. 02.014.

Z. Jiang et al. / Food Chemistry 228 (2017) 411–419

References Anastassiades, M., Lehotay, S. J., Stajnbaher, D., & Schenck, F. J. (2003). Fast and easy multiresidue method employing acetonitrile extraction/partitioning and ‘‘dispersive solid-phase extraction” for the determination of pesticide residues in produce. Journal of AOAC International, 86(2), 412–431. Chen, Z., Dong, F., Xu, J., Liu, X., & Zheng, Y. (2015). Management of pesticide residues in China. Journal of Integrative Agriculture, 14(11), 2319–2327. China Pesticides Database. (2016). Accessed 10.10.16. Choi, S., Kim, S., Shin, J. Y., Kim, M., & Kim, J. (2015). Development and verification for analysis of pesticides in eggs and egg products using QuEChERS and LC–MS/ MS. Food Chemistry, 173, 1236–1242. EU Pesticides Database. (2016). Accessed 10.10.16. European Commission. (2013). Analytical quality control and method validation procedures for pesticide residues analysis in food and feed. Document N° SANCO/12571/2013. González-Curbelo, M. Á., Socas-Rodríguez, B., Herrera-Herrera, A. V., GonzálezSálamo, J., Hernández-Borges, J., & Rodríguez-Delgado, M. Á. (2015). Evolution and applications of the QuEChERS method. TrAC Trends in Analytical Chemistry, 71, 169–185. Gosetti, F., Mazzucco, E., Zampieri, D., & Gennaro, M. C. (2010). Signal suppression/ enhancement in high-performance liquid chromatography tandem mass spectrometry. Journal of Chromatography A, 1217(25), 3929–3937. Harveson, R. M., Windels, C. E., Smith, J. A., Brantner, J. R., Cattanach, A. W., Giles, J. F., ... Cattanach, N. R. (2007). Fungicide registration and a small niche market: A case history of hymexazol seed treatment and the U.S. sugar beet industry. Plant Disease, 91(7), 780–790. Hemström, P., & Irgum, K. (2006). Hydrophilic interaction chromatography. Journal of Separation Science, 29(12), 1784–1821. Hou, R., Jiao, W., Qian, X., Wang, X., Xiao, Y., & Wan, X. (2013). Effective extraction method for determination of neonicotinoid residues in tea. Journal of Agricultural and Food Chemistry, 61(51), 12565–12571. Hou, R., Jiao, W., Xiao, Y., Guo, J., LV, Y., Tan, H., Hu, J., & Wan, X. (2015). Novel use of PVPP in a modified QuEChERS extraction method for UPLC-MS/MS analysis of neonicotinoid insecticides in tea matrices. Analytical Methods, 7(13), 5521–5529. Jandera, P. (2011). Stationary and mobile phases in hydrophilic interaction chromatography: A review. Analytica Chimica Acta, 692(1–2), 1–25. Japan Pesticides Database. (2016). Accessed 10.10.16. Jin, F., Wang, J., Shao, H., & Jin, M. (2010). Pesticide use and residue control in China. Journal of Pesticide Science, 35(2), 138–142. Kawachi, Y., Ikegami, T., Takubo, H., Ikegami, Y., Miyamoto, M., & Tanaka, N. (2011). Chromatographic characterization of hydrophilic interaction liquid chromatography stationary phases: Hydrophilicity, charge effects, structural selectivity, and separation efficiency. Journal of Chromatography A, 1218(35), 5903–5919. Kiljanek, T., Niewiadowska, A., Semeniuk, S., Gaweł, M., Borze˛cka, M., & Posyniak, A. (2016). Multi-residue method for the determination of pesticides and pesticide metabolites in honeybees by liquid and gas chromatography coupled with tandem mass spectrometry—Honeybee poisoning incidents. Journal of Chromatography A, 1435, 100–114. Li, M., Liu, X., Dong, F., Xu, J., Kong, Z., Li, Y., & Zheng, Y. (2013). Simultaneous determination of cyflumetofen and its main metabolite residues in samples of plant and animal origin using multi-walled carbon nanotubes in dispersive solid-phase extraction and ultrahigh performance liquid chromatography– tandem mass spectrometry. Journal o Chromatography A, 1300, 95–103. Martínez-Domínguez, G., Romero-González, R., & Garrido Frenich, A. (2015). Determination of toxic substances, pesticides and mycotoxins, in Ginkgo

419

biloba nutraceutical products by liquid chromatography Orbitrap-mass spectrometry. Microchemical Journal, 118, 124–130. Myresiotis, C. K., Karaoglanidis, G. S., Vryzas, Z., & Papadopoulou-Mourkidou, E. (2012). Evaluation of plant-growth-promoting rhizobacteria, acibenzolar-Smethyl and hymexazol for integrated control of Fusarium crown and root rot on tomato. Pest Management Science, 68(3), 404–411. Payá, P., Anastassiades, M., Mack, D., Sigalova, I., Tasdelen, B., Oliva, J., & Barba, A. (2007). Analysis of pesticide residues using the Quick Easy Cheap Effective Rugged and Safe (QuEChERS) pesticide multiresidue method in combination with gas and liquid chromatography and tandem mass spectrometric detection. Analytical and Bioanalytical Chemistry, 389(6), 1697–1714. Qian, K., He, S., Tang, T., Shi, T., Li, J., & Cao, Y. (2011). A rapid liquid chromatography method for determination of glufosinate residue in maize after derivatisation. Food Chemistry, 127(2), 722–726. Sinha, S. N., Vasudev, K., & Vishnu.Vardhana.Rao, M. (2012). Quantification of organophosphate insecticides and herbicides in vegetable samples using the ‘‘Quick Easy Cheap Effective Rugged and Safe” (QuEChERS) method and a highperformance liquid chromatography–electrospray ionisation–mass spectrometry (LC–MS/MS) technique. Food Chemistry, 132(3), 1574–1584. Stachniuk, A., & Fornal, E. (2016). Liquid chromatography-mass spectrometry in the analysis of pesticide residues in food. Food Analytical Methods, 9(6), 1654–1665. Sun, D., Li, L., Ji, R., Li, W., Ye, H., Wu, Y., & Liu, C. (2011). Determination of hymexazol in cucumber and soil samples by derivatization using GC-FPD. Bulletin of Environmental Contamination and Toxicology, 87(6), 653–656. Tamura, Y., Takano, I., Kobayashi, M., Tomizawa, S., Tateishi, Y., Sakai, N., ... Nagayama, T. (2008). Determination of hymexazol in agricultural products by GC-NPD. Shokuhin Eiseigaku Zasshi, 49(3), 223–227. Tan, Z., & Guo, Z. (2011). Determination of hymexazol residue in rice by GC. Pesticide Science and Administration, 32(4), 29–31 (in Chinese). Vera, T., Muñoz, A., Ródenas, M., Vázquez, M., Borrás, E., Marqués, M., ... Sidebottom, H. (2011). Atmospheric fate of hymexazol (5-methylisoxazol-3-ol): Simulation chamber studies. Atmospheric Environment, 45(22), 3704–3710. Viñas, P., Aguinaga, N., Campillo, N., & Hernández-Córdoba, M. (2008a). Comparison of stir bar sorptive extraction and membrane-assisted solvent extraction for the ultra-performance liquid chromatographic determination of oxazole fungicide residues in wines and juices. Journal of Chromatography A, 1194(2), 178–183. Viñas, P., Campillo, N., Aguinaga, N., Martínez-Castillo, N., & Hernández-Córdoba, M. (2008b). Solid-phase microextraction for the gas chromatography mass spectrometric determination of oxazole fungicides in malt beverages. Analytical and Bioanalytical Chemistry, 391(4), 1425–1431. Viñas, P., Martínez-Castillo, N., Campillo, N., & Hernández-Córdoba, M. (2010). Liquid–liquid microextraction methods based on ultrasound-assisted emulsification and single-drop coupled to gas chromatography–mass spectrometry for determining strobilurin and oxazole fungicides in juices and fruits. Journal of Chromatography A, 1217(42), 6569–6577. Wang, P., Tian, F., Xu, J., Dong, F., Li, S., Zheng, Y., & Liu, X. (2015). Determination of flumetsulam residues in 20 kinds of plant-derived foods by ultra-performance liquid chromatography coupled with tandem mass spectrometry. Analytical Methods, 7(14), 5772–5779. Wilkowska, A., & Biziuk, M. (2011). Determination of pesticide residues in food matrices using the QuEChERS methodology. Food Chemistry, 125(3), 803–812. Wong, J., Hao, C., Zhang, K., Yang, P., Banerjee, K., Hayward, D., ... Oulkar, D. P. (2010). Development and interlaboratory validation of a QuEChERS-based liquid chromatography tandem mass spectrometry method for multiresidue pesticide analysis. Journal of Agricultural and Food Chemistry, 58(10), 5897–5903. Xu, Y., Cui, L., Wang, Y., Wang, C., & Gao, J. (2015). Study on residual decline and final residues of hymexazol in Panax ginseng and soil. Modern Agrochemicals, 14 (1), 29–31 (in Chinese). Zhao, P., Wang, L., Zhou, L., Zhang, F., Kang, S., & Pan, C. (2012). Multi-walled carbon nanotubes as alternative reversed-dispersive solid phase extraction materials in pesticide multi-residue analysis with QuEChERS method. Journal of Chromatography A, 1225, 17–25.