The use of pesticides in food production is regulated by the Codex Alimentarius Commission, which specifies 2,930 maximum residue levels (MRLs) for a total of 218 pesticides. These MRLs are formulated in accordance with defined good agricultural or veterinary practices and based on scientific advice concerning the safety of the residues remaining after the substances are used.[1] The Codex Alimentarius 198-1995[2] for rice states that rice must comply with the MRLs established by the Commission for this commodity. This standard is applicable to rice intended for direct human consumption, including husked rice, milled rice and parboiled rice, either in packaged form or sold directly to consumers loose from the package. It is not applicable to other products derived from rice or to glutinous rice.

There are several directives in the EU which specify MRLs for pesticide residues in food, including Directive 76/895/EEC[3] for selected fruit and vegetables, Directive 86/362/EEC[4] for cereals, Directive 86/363/EEC[5] for foodstuffs of animal origin and Directive 90/642/EEC[6] for certain products of plant origin. All MRLs for pesticides within the EU are then harmonized in Directive EC 396/2005. The regulation states a default limit of 0.01 mg/kg for all pesticide/commodity combinations for which no MRLs have been set, unless MRLs are not required or different defaults have been set.[7]

Pesticide residues must be accurately and reliably monitored in food to comply with stringent regulations. A key part of this process is sample preparation.

Sample Preparation
The QuEChERS (Quick, Easy, Cheap, Effective, Rugged and Safe) sample preparation method, published recently as AOAC Method 2007.01,[8] has simplified the determination of pesticides in fruits, vegetables, grains and herbs. Sample preparation is simplified by using a single-step buffered acetonitrile (MeCN) extraction and liquid-liquid partitioning from water in the sample by salting out with sodium acetate and magnesium sulfate. QuEChERS prepares fruit, vegetable, grain and herb samples before analysis by gas chromatography/tandem mass spectrometry (GC/MSn) is performed.

There are four steps to the QuEChERS sample preparation procedure. These are extraction, initial clean-up, solvent exchange and final clean-up (Figure 1).

Figure 1: Flow Diagram of QuEChERS Sample Preparation Procedure

Click here to download Figure 1

The solvent exchange step is necessary to provide a final solvent that is more amenable to splitless injection. It is important that extraction solvents are buffered and that the powdered reagents are measured at appropriate amounts for the size of the sample prepared. Care must be taken to ensure the sample is homogenized. When analyzing grains such as rice, water must be added during the homogenization step to prepare the sample for liquid-liquid extraction. This must be considered in the final calculations of spikes and standards. As some reagents can cause an exothermic reaction when mixed with water, which can adversely affect the recoveries of target compounds, it is important that they are added carefully. The following experimental shows the use of ion trap GC/MSn and modified QuEChERS extraction in the analysis of rice.

Experimental Methods
Analysis was performed by GC/MSn on the Thermo Scientific ITQ 700 GC-ion trap MS. The aim was to determine the linear ranges, quantitation limits and detection limits for a list of pesticides that are commonly used on rice crops, addressing the various functional groups and different physical properties of most pesticides. A surge splitless injection of 33 pesticides was made into a Thermo Scientific TRACE TR-527 35% diphenyl/65% dimethyl polysiloxane column, (0.25 mm × 30 meter, and a film thickness of 0.25 μm with a 5-m guard column) with detection in electron ionization (EI) MS/MS. The recommended consumables required for sample preparation and analysis were rigorously tested.

Sample Extraction and Clean-Up
A thoroughly homogenized 15 g sample of rice was weighed into the extraction tube. Then, 15 mL of 1% acetic acid-MeCN extraction solvent was poured into the tube on top of the sample. The surrogate and the pesticide solutions were spiked into this MeCN layer for the method validation (MVD) and method detection limit (MDL) samples.

The tube was capped and vortexed for 30 seconds. The cap was removed and the powder reagents were poured slowly into the MeCN layer. The cap was tightened securely on the 50-mL extraction tube and was vortexed for 30 seconds until all of the powder reagents were mixed with the liquid layers. The tube was placed on a mechanical shaker for 5 min and then centrifuged for 5 min at 3,000 rpm. Following this, 11 mL of the top MeCN layer were removed and transferred to a 15-mL clean-up tube. This tube was capped and vortexed for 30 seconds and centrifuged for 5 min at 3,000 rpm. A 5-mL aliquot of the top layer was transferred into a clean test tube for solvent exchange.

Solvent Exchange
The 5-mL aliquot of the cleaned-up extract was blown down to dryness with a gentle stream of nitrogen at 40 °C in approximately 1 hour. Care was taken to remove the tube immediately when dried. A 900-μL aliquot of hexane/acetone (9:1) was added and 100 μL of the internal standard (d10-parathion) was spiked into the organic solution. The tube was capped and vortexed for 15 seconds. Next, 1 mL of extract was transferred to a 2-mL clean-up tube, capped tightly and vortexed for 30 seconds. After centrifuging for 5 min at 3,000 rpm, 200 μL of the clear extract were transferred to an autosampler vial with a small glass insert for analysis on the GC-ion trap MS. The individual calibration levels were spiked into each extract for the calibration curve in matrix before the final clean-up step.

Injection
The GC-ion trap MS was paired with the Thermo Scientific FOCUS GC, a single-channel GC with a standard split/splitless (SSL) injection port. The SSL inlet temperature was set to 250 °C. A 5-mm i.d. splitless liner with a volume of 1.6 mL was selected for the surged pressure injection. For the surge splitless injection, the inlet pressure was held at an elevated level of 250 kPa for the 0.5-min injection time. This technique reduced the vapor cloud of a 2-μL injection from 0.37 mL to 0.19 mL. At an elevated injection flow rate of 4.6 mL/min., the liner was swept several times during injection. The target compounds moved through the inlet rapidly, reducing the interaction time with the inside walls of the liner. This minimized the amount of breakdown of the more fragile pesticides.

A performance solution consisting of endrin and 4,4'-DDT was analyzed as a daily check to determine system activity. The analysis of endrin, DDT and their breakdown products as part of daily quality control can alert analysts that the system has developed active sites and maintenance is needed. Without performing a breakdown analysis, the laboratory may need to continually maintain the equipment and replace consumables, even when it may not be needed. Monitoring breakdown can decrease the cost of running the analysis and save significant amounts of time. Endrin breakdown was determined by adding up the response for the two breakdown products, endrin aldehyde and endrin ketone, and dividing by the total response for the breakdown products and endrin in percent. The breakdown products of DDT are DDE and DDD and were calculated similarly. The breakdown check results showed less than 5% breakdown for both compounds on a daily basis. For routine use, the liner would be changed when the breakdown of either compound reached greater than 20%. The injection port liner was tested and generated very good results over a long period without the need for maintenance.

Separation
Chromatographic separation was achieved by using the 35% diphenyl/65% dimethyl polysiloxane column. This column was chosen to provide sufficient resolution of the more polar compounds. The oven was programmed as follows: initial temperature 40 °C, 1.5 min, 25 °C/min to 150 °C, 0.0 min, 5 °C/min to 200 °C, 7.5 min, 25 °C/min to 290 °C with a final hold time of 12 min and a constant column flow rate of 1 mL/min.

Detection
The detection of the pesticides was performed using the GC-ion trap MS with optional MSn mode. This scanning mode offered enhanced selectivity over either full scan or selected ion monitoring (SIM). In SIM at the elution time of each pesticide, the ratio of the intensity of matrix ions increased exponentially versus that of the pesticide ions as the concentration of the pesticide approached the detection limit, decreasing the accuracy at lower levels. The GC-ion trap MS was operated in the MSn mode and performed tandem MS functions by injecting ions into the ion trap and destabilizing matrix ions, isolating only the pesticide ions.

The pesticide ions were given sufficient energy to further fragment and were then scanned. This process provided the product ion spectrum. This was done by setting up a stable field for the pesticide precursor ion. Once the precursor ion was isolated from the matrix ions, collision induced dissociation (CID) energy was applied to fragment it into its respective product ions. Finally, these unique product ions were scanned out to generate the product ion spectrum. Because of the elimination of matrix interferences, this process produced more accurate results at the lower levels. Figures 2 and 3 show a comparison between a full scan and an MSn total ion current (TIC) plot.

Figure 2: MS/MS Scan of 160 ng/g in Rice Matrix

Figure 3: Full Scan Chromatogram of 160 ng/g of Pesticides in Rice

Results and Discussion
The calibration curve was spiked into the rice matrix. Levels ranged from 1 ng/g to 1,200 ng/g, depending on the compound and its MRL in rice. The linearity for all compounds was R² > 0.995. The results of the linearity are shown in Table 1.

Table 1: Calibration Curve Results

The actual limit of detection (LOD) and limit of quantitation (LOQ) were determined by preparing matrix spikes at a level near or below the MRL. Concentrations of 16, 32, 40, 80 and 120 ng/g were analyzed in seven matrix samples and the LOD and LOQ were calculated from these results by multiplying the standard deviation of the calculated amounts by 3 and 10, respectively.

The MVD calculations were performed using five matrix samples spiked at concentrations of 160, 320 or 480 ng/g per pesticide. Samples had an average of 98% recovery with an average %RSD of 5.9%.

Conclusions
As recently formulated pesticides are smaller in molecular weight than their predecessors and designed to break down rapidly in the environment, the use of a reliable sample preparation method is necessary for efficient pesticide analysis in foods. The QuEChERS sample preparation method simplifies the analysis of pesticides in foods and can be used to prepare fruit, vegetable, grain and herb samples before GC/MSn analysis. Following the use of QuEChERS to prepare samples, GC-ion trap MS delivers high accuracy at low concentrations of pesticide residues analyzed in rice. Due to the method’s MSn functionality, only one analytical run is needed to identify, confirm and quantify all pesticide residues. Low endrin and DDT breakdown are achieved on a daily basis, demonstrating that GC-ion trap mass spectrometry can analyze active compounds without requiring continual, time-consuming and expensive maintenance. ♦

David Steiniger, Jessie Butler and Eric Phillips are with Thermo Fisher Scientific, Austin, TX, USA.

References

1. ftp.fao.org/codex/Publications/understanding/Understanding_EN.pdf
2. www.codexalimentarius.net/search/advancedsearch.do
3. eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:31976L0895:EN:HTML
4. eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:31986L0362:EN:HTML
5. eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:31986L0363:EN:HTML
6. www.observatoire-pesticides.fr/upload/bibliotheque/683347745667126555067255243995/directive_90_642_CE.pdf
7. www.food.gov.uk/safereating/chemsafe/pesticides/pesticidesmainqa/ecregulation
8. Lehotay, S. 2007. AOAC Official Method 2007.01 Pesticide residues in foods by acetonitrile extraction and partitioning with magnesium sulfate. Journal of AOAC International 90:485–520.

For more information, visit www.thermofisher.com.