Difference in the matrix components by cleanup methods between the notified multiresidue pesticide analysis method in Japan and the QuEChERS method

Abstract

Although the interest of the QuEChERS method has been increasing in Japan due to its simplified method, there is no information on the difference in cleanup degree between the official method for multiresidue pesticide analysis in Japan and the QuEChERS method. The purpose of this study was to compare the matrix components remaining in the sample extracts prepared using two different methods. A metabolomics technique was used to investigate the matrix components. The result showed that sugars, flavonoids, and fatty acids remained in the sample extracts using the QuEChERS method. The lack of a buffer solution and insufficient dehydration were considered as reasons for the remaining sugars and flavonoids. In the case of fatty acids, the ion exchange interaction was insufficient using dispersive SPE. Whichever preparation was used, matrices specific to the sample, such as caffeine in powdered green tea, capsaicins in chili peppers, or gingerols in ginger, remained.

Introduction

With the introduction of the Japanese Positive List System (JPL) by the Ministry of Health, Labour and Welfare (MHLW) in Japan,¹⁾ the multiresidue analysis became a major part of pesticide analyses. The multiresidue pesticide analysis method was based on Fillion’s study,^2,3) and this was a significant change. In the classical method, acetone extraction, liquid–liquid extraction, and a filled column, such as the florisil or silica gel column, had been used for the single or group analysis, while acetonitrile extraction, C18 solid-phase extraction (SPE), and double-layered SPE with carbon black and amino-propyl silica gel are used for the multiresidue analyses.

On the other hand, the QuEChERS (Quick, Easy, Cheap, Effective, Rugged, Safe) method^4,5) has been developed by Anastassiades and Lehotay and standardized as the AOAC 2007.01.⁶⁾ and EN 15662 methods.⁷⁾ The significant features of this method are the nonuse of glassware, no filter suction, and no need to evaporate. Using dispersive SPE is also an important feature of the QuEChERS method. Recently, the interest of this simplified method has been increasing in Japan. Okihashi et al.^8,9) evaluated its rapid approach; however, concerned about the weak extraction potency and insufficient cleanup, they modified the QuEChERS method.

Incidentally, Sugitate et al.^10,11) have studied the matrix components in the pesticide analysis using a metabolomics technique. Metabolites are small molecular compounds existing in all living beings, that is to say, plants, animals and microbes. Many metabolites have polar functional groups, such as carboxyl, amino, or hydroxyl, to metabolize in living organisms. Some metabolites can be analyzed by gas chromatography-mass spectrometry (GC-MS) without derivatization, but most of them turn into highly volatile and less polar compounds with derivatization, and a highly sensitive analysis is possible. Metabolomics is a comprehensive study to analyze these compounds for various purposes, such as clarification of diseases, plant breeding, and development of biofuels.

The purpose of this study was to reveal the difference in the matrix components between the multiresidue pesticide analysis in Japan and the QuEChERS method using the metabolomics technique. Understanding the differences provides us with important information for further method improvement or development because matrices sometimes interfere with the pesticide peaks, shift the pesticide’s retention time (t_R) or cause the matrix effect. Many kinds of samples were chosen in order to investigate various components from agricultural products.

Materials and Methods

1. Sample

Spinach, ginger, avocados, tomato juice (from concentrated juice, additive-free), black sesame, raisins, toasted soybean flour, chili peppers and powdered green tea were selected as the different types of foods. Several of these samples are generally difficult to analyze because of their heavy matrices. Spinach and powdered green tea contain pigments; avocados have a lot of fats; raisins contain much sugar; toasted soybean flour has high protein; and ginger, chili pepper, sesame and powdered green tea have specific matrices.

2. Chemicals

Acetone and n-hexane for the pesticide and PCB analyses were obtained from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Acetonitrile for the LC-MS was obtained from Kanto Chemical Co., Inc. (Tokyo, Japan). Methoxyamine hydrochloride and pyridine hydride were also obtained from Kanto Chemical Co., Inc. (Tokyo, Japan). N-Methyl-N-trimethylsilyltrifluoroacetamide (MSTFA) with 1% trimethylchlorosilane (TMCS) was obtained from Thermo Fisher Scientific (Rockford, IL, USA). Myristic acid-d₂₇ was used as an internal standard and purchased from Sigma-Aldrich (St. Louis, MO, USA).

Bond Elut C18 (1,000 mg), carbon/NH₂ (500 mg/500 mg) and the QuEChERS kits were obtained from Agilent Technologies (Folsom, CA, USA).

3. Sample preparation

The sample preparations were performed according to the Multiresidue Method for Agricultural Chemicals by GC-MS (Agricultural Products) in the “Analytical Method for Residual Compositional Substances of Agricultural Chemicals, Feed Additives, and Veterinary Drugs in Food” under the Japanese Positive List System of the MHLW (hereafter, called the JPL method) and the QuEChERS method. The difference between the JPL method and the QuEChERS method was the buffer solution. A phosphate buffer was added apart from the water addition for swelling of the samples in the JPL method; otherwise, water was added just after sampling, and the buffering function worked by just adding salt in the QuEChERS method. Due to the fact that the amount of water during the salting out was different between the two methods, the relationship of the sample amount, acetonitrile amount, and water amount during the salting out is shown in Table 1. As for the QuEChERS method, there are two types of methods, AOAC and EN. The main difference between these two methods is the type of salt used during the salting out. Sodium acetate is used in the AOAC method, and sodium citrate is used in the EN method. There are some dispersive SPE kits based on the types of samples, and the combinations of the QuEChERS dispersive SPE kits and samples are shown in Table 2. For the JPL method, Bond Elut C18 SPE was used before the carbon/NH₂ cleanup for the avocado, sesame, toasted soybean flour and chili pepper in order to remove the fats.

Table 1. Relationship of sample amount, acetonitrile amount and water amount during salting out

Sample	JPL method				QuEChERS EN methodc)
	Sample amounta)	Watera)	ACNb)	Buffer solution	Sample amount	Water	ACNb)
Tomato juice	4 g (20 g)	none	20 mL	20 mL	10 g	none	10 mL
Spinach	4 g (20 g)	none	20 mL	20 mL	10 g	none	10 mL
Avocado	4 g (20 g)	none	20 mL	20 mL	10 g	none	10 mL
Ginger	4 g (20 g)	none	20 mL	20 mL	10 g	none	10 mL
Chili pepper	2 g (10 g)	4 mL (20 mL)	20 mL	20 mL	2 g	10 mL	10 mL
Sesame (black)	2 g (10 g)	4 mL (20 mL)	20 mL	20 mL	5 g	10 mL	10 mL
Toasted soybean flour	2 g (10 g)	4 mL (20 mL)	20 mL	20 mL	5 g	10 mL	10 mL
Raisin	2 g (10 g)	4 mL (20 mL)	20 mL	20 mL	5 g	7.5 mL	10 mL
Powdered green tea	1 g (5 g)	4 mL (20 mL)	20 mL	20 mL	2 g	10 mL	10 mL

^a) Sample was divided into five before salting out. ( ) indicates the original amount. ^b) Acetonitrile. ^c) In the case of AOAC method, sample amount, water volume and acetonitrile volume were 1.5 times than those of EN method.

Table 2. Combination of QuEChERS dispersive SPE kits and type of samples

Category	AOAC		EN
	Contents	Sample	Contents	Sample
General fruits and vegetables	PSA 400 mg	Ginger	PSA 150 mg	Ginger
	MgSO4 1,200 mg		MgSO4 900 mg
Fruits and vegetables with fats and waxes	PSA 400 mg	Sesame	PSA 150 mg	Avocado
	C18EC 400 mg	Toasted soybean flour	C18EC 150 mg	Chili pepper
	MgSO4 1,200 mg		MgSO4 900 mg	Sesame
				Toasted soybean flour
Pigmented fruits and vegetables	PSA 400 mg	Powdered green tea	PSA 150 mg	Raisin
	GCB 400 mg	Spinach	GCB 15 mg	Tomato juice
	MgSO4 1,200 mg	Raisin	MgSO4 900 mg
		Tomato juice
Highly pigmented fruits and vegetables			PSA 150 mg	Powdered green tea
			GCB 45 mg	Spinach
			MgSO4 900 mg
Fruits and vegetables with pigment and fats	PSA 400 mg	Avocado
	C18EC 400 mg	Chili pepper
	GCB 400 mg
	MgSO4 1,200 mg

4. Analytical method

4.1. Relationship between sample amount and corresponding volume for pesticide analysis and matrix evaluation

The sample solution was adjusted to 1 mL containing 1 g of the sample to compare the matrices (in the case of the powdered green tea, a 5 mL sample solution contained a 1 g sample) because the sample amount in 1 mL of the final solutions for the pesticide analysis was different depending on which method and which sample were used. The sample solution was analyzed with and without derivatization. The advantages of the derivatization are vaporization and highly sensitive analysis of the polar components and utilization of the metabolomics library, which contains the derivatized spectra and the retention time (t_R) information, while the relationship between the eluted t_R of pesticides and matrix components can be obtained without derivatization. The relationship between the sample amount and corresponding volume for the pesticide analysis and matrix evaluation is shown in Table 3.

Table 3. Relationship between sample amount and corresponding volume for pesticide analysis and matrix evaluation

Sample	JPL method			QuEChERS method
	Sample amount	Corresponding volumea) for pesticide analysis	Corresponding volumea) for matrices evaluation	Sample amount	Corresponding volumea) for pesticide analysis	Corresponding volumea) for matrices evaluation
Tomato juice	4 g	2 mL	4 mL	10 g	10 mL	10 mL
Spinach	4 g	2 mL	4 mL	10 g	10 mL	10 mL
Avocado	4 g	2 mL	4 mL	10 g	10 mL	10 mL
Ginger	4 g	2 mL	4 mL	10 g	10 mL	10 mL
Chili pepper	2 g	1 mL	2 mL	2 g	10 mL	2 mL
Sesame (black)	2 g	1 mL	2 mL	5 g	10 mL	5 mL
Toasted soybean flour	2 g	1 mL	2 mL	5 g	10 mL	5 mL
Raisin	2 g	1 mL	2 mL	5 g	10 mL	5 mL
Powdered green tea	1 g	1 mL	5 mL	2 g	10 mL	10 mL

^a) Corresponding volume means the final sample solution volume which corresponded to sample amount.

4.2. Derivatization procedure

The derivatization procedure was followed using our previous report.^10,11) The sample solutions were dried by a centrifugal concentrator before derivatization. The residues were then methoxyaminated with methoxyamine hydrochloride in pyridine (20 mg/mL) for 90 min at 30°C. For the trimethylsilylation, MSTFA+1% TMCS was added to the methoxyaminated samples, and the mixtures were reacted for 30 min at 37°C. We also tested the sample solutions without derivatization. To identify the matrix components, the Fiehn metabolomics library with t_R information (Agilent Technologies)¹²⁾ and the NIST Mass Library (National Institute of Standards and Technology, USA) were used. An Agilent 7890B gas chromatograph equipped with a 5977 single quadrupole mass spectrometer and 7693A autoinjector was used to analyze the matrix components (Agilent Technologies, DE, USA). The analytical system and conditions are described in the following section.

4.3. Analytical system and condition

4.3.1. With derivatization

Injection liner, Ultra Inert liner, split, single taper, with wool (Agilent Technologies, Middleburg, Netherlands); column, DB-5msDG, 30 m×0.25 mm i.d., 0.25 µm thickness (nonpolar deactivated precolumn connected to DB-5ms, 10m), (Agilent Technologies, Middleburg, Netherlands); oven temperature program, 60°C (1 min hold) to 325°C at 10°C/min (10 min hold); injector temperature, 250°C; injection mode, split (split ratio, 10 : 1); carrier gas, helium 1.1 mL/min constant flow; transferline, 290°C; ion source temperature, 250°C; scan range, m/z 50–600.

4.3.2. Without derivatization

Injection liner, Ultra inert liner, splitless, single taper, with wool, (Agilent Technologies, Middleburg, Netherlands); column, VF-5 msec, 30 m×0.25 mm i.d., 0.25 µm thickness (Agilent Technologies, Middleburg, Netherlands); oven temperature program, 70°C (2 min hold) to 150°C at 25°C/min, to 200°C at 3°C/min, then to 310°C at 8°C/min (5 min hold); injector temperature, 250°C; injection mode, pulsed splitless (25 psi, 1 min); carrier gas, helium 1.1 mL/min constant flow; transferline, 280°C; ion source temperature, 300°C; scan range, m/z 50–550.

Results and Discussion

1. Remaining sugars and flavonoids with the QuEChERS method

All sample solutions except that for the ginger, which were treated by the QuEChERS method, contained high amounts of sugars and flavonoids. These components were not observed until they were derivatized. The comparison chromatograms of the powdered green tea with/without derivatization are shown in Fig. 1. The detected matrix components and their intensities from each extract are shown in Table 4. Comparative example chromatograms of the powdered green tea, which were extracted by the JPL method and the QuEChERS AOAC method, are also shown in Fig. 2. A high amount of fructose was observed in the chromatograms of the tomato juice, raisin and chili pepper extracts. Flavonoids were extracted from the toasted soybean flour (e.g., daidzein and genistein) and powdered green tea (i.e., catechins). Since both the sugars and flavonoids have many hydroxyl groups, one probable reason was the lack of a buffer solution. For example, 15 g of tomato juice was extracted with 15 mL of acetonitrile, then salted out into water and acetonitrile layers with 15 mL of acetate buffer solution in the QuEChERS AOAC method, while 20 g of tomato juice was extracted with acetonitrile and diluted with acetonitrile to 100 mL, then 20 mL of the diluted solution (4 g for tomato juice) was salted out with 20 mL of phosphate buffer solution using the JPL method. Briefly, 1 mL of buffer solution acted on a 1 g sample in the QuEChERS method, while 5 mL of buffer solution acted on a 1 g sample in the JPL method. The other probable reason was poor anhydration. Magnesium sulfate was used for the salting-out in the QuEChERS method and its amount was insufficient for dehydration. Therefore, some of the sugars and flavonoids moved into the acetonitrile layer along with a small amount of water during the salting out. Ribonucleosides (e.g., uridine, cytidine) were also extracted from most of the samples as unique components when using the QuEChERS method. However, when pesticides are analyzed by GC-MS, the sugars and ribonucleosides do not enter the column because they do not vaporize without derivatization. There is a possibility that sugars and ribonucleosides remain in the injector liner and therefore contaminate the liner after many sample injections. As for the flavonoids, some of them enter the column, because they sometimes appeared in the chromatogram without derivatization if their amount was high.

Fig. 1. Total current ion chromatograms of powdered green tea extracted with QuEChERS AOAC method. (a) without derivatization; (b) with derivatization, 1. caffeine, 2. phytol, 3. fatty acids, 4. monoacylglycerols, 5. tocopherols, 6. sterols, 7. organic acid (unknown), 8. sugars, 9. catechins.

Table 4. Total intensity of each matrix type by derivatization

Sample	JPL method		QuEChERS AOAC mehod
	Compounds	Total intensity	Compounds	Total intensity
Tomato juice	monoacylglycerols	4.3×107	monoacylglycerols	7.2×106
			sugars	2.5×108
			fatty acids	1.3×108
			ribonucleosides	4.3×107
Spinach	monoacylglycerols	5.8×107	monoacylglycerols	4.2×107
	phytol	7.2×106	phytol	2.2×107
			fatty acids	3.5×108
			ribonucleosides	1.3×108
			glycerol	3.8×107
Avocado	long-chain alcohols	3.4×108	long-chain alcohols	1.3×108
	monoacylglycerols	8.0×106	monoacylglycerols	9.6×107
			fatty acids	5.2×108
			ribonucleosides	5.1×107
			nucleic acid	4.0×107
			avocadyne	3.6×107
			sterols	2.4×107
			phytol	2.2×107
			ferulic acid	1.7×107
			sugars	1.2×107
Ginger	gingerols	1.6×109 S	gingerols	3.2×109 S
	terupenes	3.3×108	terupenes	3.2×109
			fatty acids	2.3×108
			sterols	1.2×108
Chili pepper	capsaisins	2.5×108	capsaisins	5.2×108
	monoacylglycerols	1.2×108	monoacylglycerols	3.8×108
			fatty acids	3.4×109 S
			sugars	7.3×108
			glycerol	3.1×108
			organic acids	3.1×108
			ribonucleosides	4.5×107
			nucleic acid	2.2×107
Sesame (black)	sesamins	7.1×108	sesamins	7.6×108
	monoacylglycerols	8.4×107	monoacylglycerols	1.1×108
			fatty acids	1.5×109
			glycerol	8.9×107
			ribonucleosides	7.2×107
			tocopherols	6.1×107
			sterols	2.7×107
Toasted soybean flour	monoacylglycerols	3.4×107	monoacylglycerols	5.3×107
			fatty acids	4.6×108
			ribonucleosides	1.8×108
			organic acids	1.5×108
			flavonoids (daizein, genistein)	1.3×108
			tocopherols	1.2×108
			sugars	1.1×108
			glycerol	5.6×107
			sterols	2.5×107
Raisin	monoacylglycerols	1.2×108	monoacylglycerols	1.4×107
			sugars	2.2×109 S
			organic acids	9.4×108
			fatty acids	3.2×108
			furfural	1.6×108
			glycerol	1.1×108
Powdered green tea	caffeine	9.8×108 S	caffeine	2.5×109 S
	phytol	5.6×107	phytol	1.1×108
	monoacylglycerols	3.7×107	monoacylglycerols	5.1×105
	sterols	2.5×107	sterols	2.1×107
	tocopherols	2.4×107	tocopherols	1.6×107
			flavonoids (catechins)	9.7×108
			sugars	2.3×108
			fatty acids	2.0×108
			ribonucleosides	1.3×107

S=saturation.

Fig. 2. Comparative chromatograms of powdered green tea extracted by the JPL method and QuEChERS AOAC method. (a) JPL method; (b) QuEChERS AOAC method, 1. caffeine, 2. phytol, 3. monoacylglycerols, 4. tocopherols, 5. sterols, 6. organic acid (unknown), 7. fatty acids, 8. sugars, 9. catechins.

2. Remaining fatty acids and sterols with the QuEChERS method

The fatty acids were almost completely removed by the JPL method, while many fatty acids remained when using the QuEChERS method. The role of the NH₂ (sometimes PSA is used instead of NH₂) column in the JPL method and PSA in the QuEChERS method was ion-exchange chromatography. In the case of the JPL method, the ion exchange was efficiently performed because a filled column was used. However, the ion-exchange ability was insufficient in the QuEChERS method since the dispersive SPE was used and just shaken with an organic solvent. If it was an adsorption-type SPE, a dispersive SPE might have worked efficiently. The chili pepper was mentioned as an example. The flowchart of the analytical method for the chili pepper is shown in Fig. 3. A 500 mg amount of NH₂ was used for 2 g of chili pepper in the JPL method, while 400 mg of PSA was used for 1.8 g of chili pepper in the QuEChERS method. There was little difference in the efficiency of the ion-exchange sorbent in both methods. In fact, the amount of working sorbent in the JPL method was lower than that in the QuEChERS method because the ion-exchange ability of NH₂ is lower than that of PSA. The sample concentration in the final solutions for the pesticide residue analysis as indicated in the Analytical method was different depending on which method and which sample were used. In the case of the chili pepper, the corresponding volume to the sample amount using the QuEChERS method was ten times as much as that using the JPL method, that is, 1 mL of acetone/n-hexane (1 : 1, v/v) contained 2 g of chili pepper using the JPL method, while 1 mL of acetonitrile contained 0.2 g of it using the QuEChERS method. The remaining high amounts of fatty acids caused the peak shape deformation. The example chromatograms are shown in Fig. 4.

Fig. 3. Flowchart of sample preparation of chili pepper. (a) analytical method for multiresidue pesticide in Japan; (b) QuEChERS method.

Fig. 4. Influence of remaining fatty acids. (a) diclobutrazole, upper: JPL method, lower: QuEChERS method; (b) flusilazole, upper: JPL method, lower: QuEChERS method.

A similar result was obtained from the avocado extract. Monoacylglycerols were observed in the chromatogram using the JPL method, while many fatty acids and long-chain alcohols were observed using the QuEChERS method.

Sterols were also observed from the avocado, sesame and toasted soybean flour which were treated by the QuEChERS method despite using the C18. This is because the amount of C18 was different between the JPL method (1,000 mg) and QuEChERS method (400 mg). The sample amount loading the C18 was also different, as shown in Table 3.

3. Components specific to samples

Caffeine in the powdered green tea extract was not removed by either method. Sesamin, capsaicins, and gingerols also remained in the extracted solutions from the black sesame, chili pepper, and ginger. Many kinds of sesquiterpenes were also observed from the ginger extract. Caffeine and sesamin were single compounds, while the capsaicins and gingerols had many isomers and analogues. Specifically, the gingerols were eluted over a wide t_R range in the chromatograms. Since these components were significantly contained in the extract, these caused the t_R shift of the target pesticides or deformed peak shapes. The t_R and the peak shape are very important information to identify the pesticides, so these matrix components should be removed by an additional cleanup.

4. Main residue components from the JPL method

Although the amount of matrices, except for the specific components, using the JPL method was extremely low compared to the QuEChERS method, monoacylglycerols remained in all the sample extracts. This result coincided with our previous study.^10,11) Since the monoacylglycerols are influential components of the matrix enhancement effect, these components have to be removed for accurate measurement. In the case of the QuEChERS method, other matrix components remained extremely high compared to the monoacylglycerols, therefore, the influence of the monoacylglycerols should be low.

Sterols also remained in the sample extract when C18 was not used. Although C18 is used for grains, beans, nuts and seeds in the JPL method, C18 could be used for other agricultural products.

5. Others

Although the difference between the AOAC and EN methods of the QuEChERS method was not mentioned above, the same kinds of matrix components remained in the sample extract. However, the amount of matrices using the AOAC method was less than that using the EN method. Specifically, the color of the extracts from the powdered green tea and spinach was dark because the amount of GCB was insufficient.

Removing sugars and flavonoids in the QuEChERS method should be examined in future studies. Using MgSO₄ or Na₂SO₄ before the dispersive SPE is one possibility. We also must consider how to react PSA or NH₂ with the fatty acids. Since the combination of the QuEChERS extraction and filled SPE cleanup like Okihashi’s^8,9) method has been adopted by many laboratories in Japan, the matrix components using that kind of method need to be investigated in the future. It is also necessary to develop a method to remove the monoacylglycerols in the JPL method. Using LC-(Q)TOF for the high molecular weight components, such as pigments and flavonoid glycoside, is open to further discussion.

We concluded that knowing the differences in the matrix components between the JPL method and the QuEChERS method is very helpful for method improvement or development. In addition, after understanding the advantage and disadvantage of each method, not only the residual matrices, but also the maintenance schedule and cost, we can choose the most suitable method.

References

• 1) Ministry of Health, Labour and Welfare: Introduction of the Positive List System for Agriculutural Chemical Residues in Foods (http://www.mhlw.go.jp/english/topics/foodsafety/positivelist060228/). (Accessed 07 Feb., 2015)
• 2) Ministry of Health, Labour and Welfare: Analytical Methods for Residual Compositional Substances of Agricultural Chemicals, Feed Additives, and Veterinary Drugs in Food (http://www.mhlw.go.jp/english/topics/foodsafety/positivelist060228/dl/060526-1a.pdf). (Accessed 07 Feb., 2015)
• 3) J. Fillion, F. Sauvé and J. Selwyn: J. AOAC Int. 83, 698–713 (2000).
• 4) M. Anastassiades, S. J. Lehotay, D. Štajnbaher and F. J. Schenck: J. AOAC Int. 86, 412–431 (2003).
• 5) S. J. Lehotay, K. Maśtvská and A. R. Lightfield: J. AOAC Int. 88, 615–629 (2005).
• 6) AOAC Official Method 2007.01: http://www.weber.hu/PDFs/QuEChERS/AOAC_2007_01.pdf (Accessed 07 Feb., 2015)
• 7) EN15662 Version 2007-10_24, Foods of Plant Origin-Determination of Pesticide Residues Using GC-MS and/or LC/MS/MS Following Acetonitrile Extraction/Partitioning and Clean-up by Dispersive SPE (QuEChERS metod). European Committee for Standardization (2008).
• 8) M. Okihashi, Y. Kitagawa, K. Akutsu, H. Obana and Y. Tanaka: J. Pestic. Sci. 30, 368–377 (2005).
• 9) M. Okihashi, Y. Kitagawa, H. Obana, Y. Tanaka, Y. Yamagishi, K. Sugitate, K. Saito, M. Kubota, M. Kanai, T. Ueda, S. Harada and Y. Kimura: Food 1, 101–110 (2007).
• 10) K. Sugitate, S. Nakamura, N. Orikata, K. Mizukoshi, M. Nakamura, A. Toriba and K. Hayakawa: J. Pestic. Sci. 37, 156–163 (2012).
• 11) K. Sugitate, M. Saka, T. Serino, S. Nakamura, A. Toriba and K. Hayakawa: J. Agric. Food Chem. 60, 10226–10234 (2012).
• 12) Agilent G1676AA Agilent Fiehn GC/MS Metabolomics RTL Library, User Guide, Agilent Technologies (2013).