Multiclass Pesticide Analysis of Soy Milk Using a Matrix-Compatible SPME Fiber

Emanuela Gionfriddo^{1, 2}, Dominika Gruszecka, Xiujuan Li^{1, 3}, Janusz Pawliszyn¹

¹Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada, ²Department of Chemistry and Biochemistry, The University of Toledo, Toledo OH, 43606, USA, ³Key Laboratory of Environment Correlative Dietology (Ministry of Education), College of Food Science & Technology, Huazhong Agricultural University, Wuhan, China

Article from Analytix Reporter - Issue 11

• Abstract
• Introduction
• Experimental Procedure
• Result and Discussion
• Analysis of Real Samples
• Conclusions

Abstract

A matrix-compatible direct-immersion solid-phase microextraction (SPME) fiber, named PDMS/DVB/PDMS or SPME-OC Fiber, was used for the determination of pesticides in soy milk via direct immersion. Combined with gas chromatography-mass spectrometry, it eliminated the need for extensive sample pre-treatment procedures. To extend the lifetime of the SPME device, rapid pre- and post-desorption cleaning steps were implemented. This method allowed to achieve limits of quantitation (1–2.5 μg/kg) for the targeted analytes that were below the Maximum Residue Levels (MRLs) mandated for soy-based products.

Introduction

Soy-based products are a category of nutraceuticals extensively used worldwide for their health benefits and also as a more sustainable alternative to dairy products. Raw soy grains are the starting material for all soy- based products and are often exposed to agrochemicals from agricultural and post-harvesting practices. It is important to monitor the level of pesticide residues in soy derivatives to ensure their compliance with tolerance limits set by various regulatory agencies across the world.^1,2 Soy milk, being a stable emulsion of oils, water, and proteins, is a challenging sample to treat for the extraction of pesticides residues at ultra-trace levels. To propose an automated and sensitive method, solid phase microextraction (SPME) was considered as an extraction technique in this work. This is because of SPME’s ability to provide an automated analytical workflow and pre-concentration to achieve limits of quantitation for the targeted pesticides at low part-per-billion levels.³ Moreover, the use of a matrix- compatible SPME fiber enabled direct immersion extraction from soymilk, improving the recovery of pesticides with good water solubility.

Experimental Procedure

The final optimized direct immersion (DI)SPME-GC-MS method is described in Table 1.

Calibration was performed via matrix-matched calibration, spiking the analytes of interest and three deuterated internal standards: diazinon-D₁₀, malathion-D₆, and thiabendazole-D₄. The soymilk samples were purchased at local grocery stores and were refrigerated until analyzed.

Table 1. Optimized DI-SPME-GC-MS Method

Sample/matrix:	4.5 g of soy milk + 45 µL of I.S. mix at 25 mg/kg + 4.5 mL of an acetone:water solution (3:7 v:v)
SPME fiber:	SPME-OC Fiber (57439-U)
Incubation:	1 min, 35 °C, 500 rpm
Extraction:	40 min, 35 °C, 500 rpm
Post-extraction rinsing	10 s, 500 rpm, in acetone: water (1:9 v:v)
Desorption:	15 min at 270 °C
Post-desorption washing:	1 min, 500 rpm, acetone: water solution (1:1 v:v)
Column:	5% Phenyl MS capillary column (30 m × 0.25 mm x 0.25 μm)
Oven:	80 °C (2 min), 6 °C/min to 280 °C (4 min)
Carrier gas:	helium, 1.5 mL/min, constant flow
Detector:	MSD, full scan, m/z= 35-450
MSD transfer line	250°C
Injection:	splitless
Liner:	0.75 mm I.D., SPME

Result and Discussion

 

Optimization of the DI-SPME Procedure

The SPME procedure necessitated the optimization of fiber washing after the extraction (rinsing) and desorption (washing), in order to prolong its lifetime. And previous studies demonstrated that this optimization needed to be performed based on the type of food matrix analyzed and the targeted analytes.^4-7 Several rinsing and washing solutions were tested (Table 2).

Table 2. Fiber rinsing and washing methods tested in this work, after extraction of pure soy milk, unless noted.

Post-extraction-rinsing	Post-desorption-washing
30 sec in ultra-pure H2O	2 min MeOH: H2O (1:1 v/v)
1 sec in acetone	2.5 min in acetone
1 sec in H2O: acetone (1:9 v/v) (static)	30 sec in acetone
10 sec in H2O: acetone (9:1 v/v)	1 min in H2O: acetone (1:1 v/v)
10 sec in H2O: acetone (9:1 v/v)*	1 min in H2O: acetone (1:1 v/v)
*After extraction of soy milk diluted 1:1 (w/w) with ultra-pure H2O, further used for analysis

The best cleaning method involved a rinsing step in water:acetone (9:1 v/v) for 10 s and 1 min washing in water:acetone (1:1 v/v), in combination with a 1:1 dilution of the soy milk sample with ultra pure water prior to SPME. This method allowed for 120 consecutive extractions with an average signal variation of +/- 25% and % RSD of less than 15%. Furthermore, the matrix modifiers were optimized for enhanced extraction of hydrophobic analytes. Salting out effects were investigated by varying the ionic strength of the solution, by adding sodium chloride within a range from 5 to 20% to the soy milk/water mixture (1:1, w:w). However, due to no significant improvement noticed in the recovery of the analytes, the addition of salt was discarded for further optimization. An alternative strategy to improve recovery is the addition of organic modifiers. For aqueous samples, optimal recoveries are obtained keeping the content of the organic solvent below 1%. But for complex samples containing matrix constituents that can bind the analytes, the addition of organic modifiers is useful to shift the binding equilibrium toward the free, unbound form thus improving recovery by SPME. In this work, four organic solvents were considered, namely, acetonitrile, acetone, methanol, and ethanol. Each solvent was added at concentrations of 10%, 20%, 30%, and 50% (v:v) to the samples. Solvent concentrations above 50% induced congealing of the soy milk, thus were not further tested. The results showed that the addition of a solution containing 30:70 acetone:water (v:v) to the soy milk sample (dilution ratio 1:1) allowed the best recovery of the targeted analytes. Further, other parameters were finely tuned to optimize both the extraction and desorption process (Table 1).

Method validation

A matrix-matched calibration approach was used by spiking pesticide-free soymilk samples with all analytes in a concentration range of 1-1000 µg/kg; with the exception of phosalone which was spiked at 2.5-1000 µg/kg. Calculations were performed using linear regression for each of the targeted analytes, except phosalone, which required a 1/x2 weight. The accuracy and precision of the method were assessed at three concentration levels of 15, 75, and 200 µg/kg in quadruplicate measurements over three days. Limits of quantitation (LOQs) were determined at the lowest concentration level with an RSD of below 20%, and accuracy within 30% of the nominal concentration. LOQs ranged between 1 and 2.5 µg/kg. The LOQs achieved by this method allowed the detection of the targeted pesticides below the recommended limits set for soy products by the European Commission⁶ and Office of the Federal Register⁷ for the USA Market. A summary of the figures of merit for this work is provided in Table 3.

Table 3. Figures of merit of the DI-SPME-GC-MS method, reproduced from ref. 3 with permission from Elsevier, Elsevier Copyright 2020.

Analytes	Linearity (µg/kg)	LOQ (µg/kg)	r2	Concentration Level (µg/kg)	Day 1 Accuracy (%)	%RSD (n=4)	Day 2 Accuracy (%)	%RSD (n=4)	Day 3 Accuracy (%)	%RSD (n=4)
Trifuralin	1-1000	1	0.9998	15	88	2	80	2	74	8
				75	94	4	91	2	58	4
				200	103	5	92	5	77	12
Dimethoate	1-1000	1	0.9958	15	94	14	102	6	128	1
				75	72	6	89	15	120	11
				200	96	3	114	16	118	23
Diazinon	1-1000	1	0.9996	15	99	6	107	8	102	9
				75	102	1	102	1	119	3
				200	109	2	102	2	130	3
Malathion	1-1000	1	0.9988	15	111	6	117	6	119	5
				75	78	1	85	2	100	4
				200	81	2	82	2	106	3
Chlorpyrifos	1-1000	1	0.9956	15	87	2	96	6	85	11
				75	94	3	94	4	93	5
				200	102	2	91	6	99	5
Thiabendazole	1-1000	1	0.9972	15	106	13	81	4	82	11
				75	98	16	90	16	121	12
				200	81	16	119	15	124	3
Phosalone	2.5-1000	2.5	0.9851	15	118	18	116	12	123	13
				75	116	4	121	4	122	11
				200	123	2	116	8	124	8
Cyhalothrin	1-1000	1	0.9989	15	111	4	101	13	87	11
				75	99	14	74	7	80	21
				200	80	4	77	5	71	9
Cyfluthrin	1-1000	1	0.9947	15	114	2	89	15	89	11
				75	97	4	77	8	89	19
				200	88	5	94	10	104	20
Esfenvalerate	1-1000	1	0.9971	15	90	14	105	16	97	13
				75	78	6	74	9	99	10
				200	79	3	124	10	112	11

Analysis of Real Samples

The validated method was further used for the analysis of different brands of soy milk samples obtained from local grocery stores. The results in Table 4, show the occurrence of several targeted pesticides, up to 118.9 µg/kg, in two different commercial brands of soy milk.

Table 4. Quantitative analysis of commercial soy milk samples, reproduced from ref. 3 with permission from Elsevier, Elsevier Copyright 2020

	Brand #1	Brand #2
Compound	Concentration detected (µg/kg)	Concentration detected (µg/kg)
Trifluralin	N. D.	N. D.
Dimethoate	118.90	6.50
Diazinon	N. D.	N. D.
Malathion	27.40	28.20
Chlorpyrifos	7.40	7.70
Thiabendazole	N. D.	N. D.
Phosalone	40.10	33.60
Cyhalothrin	N. D.	N. D.
Cyfluthrin	20.50	N. D.
Esfenvalerate	N. D.	N. D.
N.D= Not detected

Conclusions

A new method for the analysis of pesticides in soymilk was optimized and validated using a matrix-compatible SPME fiber. This DI-SPME-GC-MS method was able to quantitatively monitor the presence of pesticides with LOQs of 1-2.5 µg/kg, with a completely automated workflow including rinsing and washing of the SPME fiber. The excellent robustness of the SPME matrix compatible fiber enabled its use of up to 120 extraction/desorption cycles.

For our complete SPME portfolio or to download the guide "SPME for GC - Getting Started with Solid Phase Microextraction" visit us at SigmaAldrich.com/SPME

Related Applications

Chemical Analysis Food & Beverage Testing
Solid Phase Microextraction (SPME)
Gas Chromatography (GC)

References

1. Council of the E.U. European Parliament, Regulation (EC) No 396/2005 of the European Parliament and of the Council of 23 February 2005 on Maximum Residue Levels of Pesticides in or on Food and Feed of Plant and Animal Origin and Amending Council Directive 91/414/EEC, Official Journal of the European Union, Brussels, 2008. http://data.europa.eu/eli/reg/2005/396/oj

2. Office of the Federal Register, Part 180—Tolerances and Exemptions for Pesticide Chemical Residues in Food, 36 FR 22540, United States of America, 2019.

3. Gionfriddo E, Gruszecka D, Li X, Pawliszyn J. Direct-immersion SPME in soy milk for pesticide analysis at trace levels by means of a matrix-compatible coating. Talanta. 2020;211:120746 https://doi.org/10.1016/j.talanta.2020.120746

4. Souza-Silva ÉA, Pawliszyn J. Direct Immersion Solid-Phase Microextraction with Matrix-Compatible Fiber Coating for Multiresidue Pesticide Analysis of Grapes by Gas Chromatography-Time-of-Flight Mass Spectrometry (DI-SPME-GC-ToFMS). J. Agric. Food Chem.. 2015;63:4464–4477 https://doi.org/10.1021/jf506212j

5. De Grazia S, Gionfriddo E, Pawliszyn J. A new and efficient Solid Phase Microextraction approach for analysis of high fat content food samples using a matrix-compatible coating. Talanta. 2017;167:754–760 https://doi.org/10.1016/j.talanta.2017.01.064

6. Souza Silva ÉA, Pawliszyn J. Optimization of Fiber Coating Structure Enables Direct Immersion Solid Phase Microextraction and High-Throughput Determination of Complex Samples. Anal. Chem. 2012;84:6933–6938 https://doi.org/10.1021/ac301305u

7. Silva ÉAS, Lopez-Avila V, Pawliszyn J. Fast and robust direct immersion solid phase microextraction coupled with gas chromatography-time-of-flight mass spectrometry method employing a matrix compatible fiber for determination of triazole fungicides in fruits. Journal of Chromatography A. 2013;1313:139–146 https://doi.org/10.1016/j.chroma.2013.07.071