Gas Chromatography with Capillary Flow Technology

May 1, 2012

Special Issues

An effective analytical method for detecting pesticide residues in olive oil

ABSTRACT: The detection of residual organophosphorous (OP) pesticides in processed olive oil is complicated by the chromatographically active nature of these compounds, which compromises chromatographic resolution. This study demonstrates a quick and effective analytical method for the determination of low ppm and trace-level OP pesticide residues in an olive oil extract. A J&W DB-35ms Ultra Inert (UI) 30 m × 0.25 mm, 0.25 µm column resolved the pesticides of interest in less than 16 minutes, yielding excellent peak shape for even the more problematic OP pesticides. The detection limits for most of the pesticides were 10–15 ng/mL. A simplified QuEChERS (Quick, Easy, Cheap, Effective, Rugged, and Safe) method provided sufficient sample matrix clean-up while preserving low-level analyte detection. A capillary flow technology (CFT) device was installed post-column to split the effluent between the MSD and FPD and implement an automated backflush to diminish residual sample carryover and reduce instrument cycle times.

INTRODUCTION

The health benefits of a Mediterranean diet, and of olive oil in particular, are widely acknowledged (1, 2). However, as 4 kg of olives are needed to produce 1 kg of olive oil, residual pesticides can be concentrated in the final product and must be monitored to ensure toxic residues do not exceed safe levels (3). Many common insecticides used in olive tree pest protection belong to the organophosphorous (OP) class, and human toxicities for OP pesticides have shown acute as well as chronic effects from pesticide poisoning (4). OP pesticides present a challenge for analysis as they are chromatographically active compounds that can adsorb onto active sites in the sample flow path, particularly at trace levels, compromising the analytes' response.

Here, we report a sample preparation extraction to detect 16 different pesticides in olive oil samples, using a procedure based on the evaluation of the QuEChERS (Quick, Easy, Cheap, Effective, Rugged, and Safe) approach for the analysis of pesticide residues in the high-lipid olive oil matrix (5). This approach simplifies the traditional, labor-intensive extraction and clean-up procedure, while providing just enough sample matrix clean-up for pesticide residues analysis. A gas chromatographic system capable of multisignal detection can provide complementary data for identification, confirmation, and quantitation of target analytes from a single injection. This method enables simultaneous detection of OP pesticides by gas chromatography with electron ionization mass spectrometry in selective ion monitoring mode (GC/ MS-SIM) and flame photometric detection (FPD) in phosphorus mode by splitting the column effluent 1:1 between the mass selective detector (MSD) and FPD. The approach chosen here uses a GC/MSD/FPD system to identify and confirm the order of elution for peaks of interest. The GC/MS system was also equipped with backflush capability. This capability enables faster instrument cycle time by backflushing late-eluting matrix components back through the inlet purge valve.

An analyte protectant (AP) was included in the study methodology to help minimize the errors caused by matrix-induced signal enhancements-L-glulonic acid γ-lactone (gulonolactone), was chosen based on the results of a previous study examining APs (6).

Table 1. Chromatographic Conditions

EXPERIMENTAL

An Agilent 7890 GC/5975C MSD equipped with an FPD and 7683B autosampler was used for this series of experiments. A purged two-way capillary flow technology (CFT) device was used to split the effluent 1:1 to the MSD:FPD. The CFT device also enabled post-column backflush. Table 1 lists the chromatographic conditions used for these analyses. Table 2 lists flow path consumable supplies used in these experiments.

Table 2. Flow Path Supplies

REAGENTS AND CHEMICALS

All reagents and solvents were HPLC or Ultra Resi grade. Acetonitrile (ACN) from Honeywell (Muskegon, MI, USA), toluene from Burdick & Jackson, and acetone from JT Baker were purchased through VWR International (West Chester, PA, USA). The neat pesticide standards were purchased from Chem Service, Inc. (West Chester, PA, USA), gulonolactone from Aldrich (St. Louis, MO), and triphenyl phosphate from Alfa Aesar (Ward Hill, MA).

1 µg/mL and 5 µg/mL spiking solutions were prepared of each of the test pesticides. TPP was prepared at concentrations of 1, 15, and 100 µg/mL in toluene. An analyte protectant solution was prepared by dissolving the neat gulonolactone in a minimum amount of water and appropriate amount of ACN to yield a 50 mg/mL concentration. The appropriate amount of gulonolactone solution was added to the calibration standards to yield a 0.5 mg/mL concentration in each standard.

SAMPLE PREPARATION

A sample of extra virgin olive oil was purchased from a local grocery store. The sample extraction method used a modified QuEChERS approach, as illustrated in Figure 1. Once the samples were prepared in this way, the extract was analyzed by GC/MS/FPD using the chromatographic conditions in Table 1. Extractions of water and acetonitrile aliquots were prepared in the same manner as the samples and served as reagent blanks.

Figure 1. Flow chart for the QuEChERS sample preparation procedure for pesticides in olive oil.

RESULTS AND DISCUSION

The 16 targeted OP pesticides were resolved on the Agilent J&W DB-35ms UI 30 m × 0.25 mm, 0.25 µm analysis column in less than 16 minutes.

The pesticide matrix-matched standard in the Figure 2 chromatogram exhibits good separation and peak shape for all of the pesticides.

Figure 2. GC/FPD chromatogram of a 100 ng/mL matrix-matched OP pesticide standard with analyte protectant analyzed on an J&W DB-35ms UI 30 m × 0.25 mm, 0.25 µm capillary GC column. Chromatographic conditions are listed in Table 1.

Chromatography of OP pesticides can be problematic, especially for polar pesticides, often yielding broad peak shapes or excessive tailing, making reliable quantitation at low levels difficult. The high level of inertness of the DB-35ms UI results in better peak shape and decreased sample adsorption on active sites within the column, enabling lower detection limits. Figure 3 depicts the excellent peak shape at 15 ppb for the four polar OP pesticides with the DB-35ms UI column.

Figure 3. Enlarged section of the GC/FPD chromatogram of a 15 ng/mL matrixmatched pesticide standard with analyte protectant analyzed on an J&W DB-35ms UI capillary column. Chromatographic conditions are listed in Table 1.

The analyte protectant used in this analysis, gulonolactone, effectively reduced matrix-related effects and improved the analyte response. Since FPD in phosphorus mode is selective only to analytes containing phosphorus, it is able to detect low levels of OP pesticides in complex matrices such as olive oil with minimal matrix interferences. Excellent signal-to-noise ratios were seen at trace levels, indicating a high level of sensitivity.

The FPD was able to detect OP pesticides down to 10 ng/mL with the exception of omethoate, diazinon, azinphos-methyl, and azinphos-ethyl, which were detected at a slightly higher limit of detection of 15 ng/mL. The detection levels for the targeted OP pesticides were within the maximum residue levels (MRLs) range of 0.01–2 mg/kg established by the US, EU, and Codex Alimentarius for pesticide residues in olives (7-9).

Sample preparation using the QuEChERS approach was effective in retaining the OP pesticides in the spiked oil sample and providing sufficient clean-up of the sample matrix for GC analysis. Figure 4 shows an olive oil sample which was fortified with the OP pesticide mix and prepared using QuEChERS. A blank matrix trace is shown below the analyte trace to indicate the level of potential matrix interference with the analytes of interest. Peak shapes for the organophosphorus pesticides are still quite sharp and well-resolved, indicating excellent performance on the DB-35ms UI column in an olive oil matrix. The performance of the DB-35ms UI column yielded excellent linearity over the calibration range of this study. The linearity of the column as defined by the R2 values of the calibration standard curve was ≥ 0.999 for all the pesticides studied.

Figure 4. GC/FPD chromatogram of the olive oil extract blank and a 100 ng/mL fortified olive oil extract both with analyte protectant analyzed on an J&W DB-35ms UI capillary column. Chromatographic conditions are listed in Table 1.

Recoveries were determined by GC/FPD at the 20, 100, and 500 ng/mL levels. The recoveries of the pesticides were greater than 70 percent with RSDs below 10 percent except in the case of acephate, which was slightly lower with an average recovery of 66 percent.

CONCLUSIONS

The Agilent J&W DB-35ms UI capillary column resolved the targeted OP pesticides and provided excellent peak shapes for the polar pesticides, allowing for more reliable quantitation at low levels. Detection levels for the OP pesticides in olive oil were at or below the EU, Codex, and US maximum residue levels for olives. Matrix-matched calibration standards yielded regression coefficients R2 ≥ 0.999 and recoveries of fortification studies were 63 percent to 107 percent with an average RSD < 9 percent, further demonstrating the effectiveness of using the J&W DB-35ms UI columns for residual pesticide determination.

By splitting the column effluent between the MSD and FPD, selectivity, identification, and confirmation of OP pesticides from a single injection are achieved, thereby increasing laboratory productivity. GC/MS-SIM provides selectivity and confirmation, while further specificity and quantitation is achieved by FPD in phosphorus mode. The QuEChERS approach was successful at providing just enough sample clean-up to minimize matrix interferences while still maintaining low-level analyte detection. The simple QuEChERS extraction method allows for faster sample prep facilitating higher sample throughput. Residual sample matrix carryover is removed through use of backflush, which eliminates the need for a bakeout cycle, significantly reducing analytical run times.

This trial successfully demonstrates a quick and efficient analytical method to monitor low- and trace-level OP pesticide residues in olive oil samples.

REFERENCES

(1) J. Brill Bond, Am J Lifestyle Med.. 3, p. 44 (2009).

(2) T. Psaltopoulou, A. Naska et al., Am J Clin Nutr.. 80, pp. 1012-1018 (2004).

(3) E.G. Amvrazi, T.A. Albanis, JAgric Food Chem., 54, pp. 9642-9651 (2006).

(4) L.G. Sultatos, 43, pp. 271 -289 (1994).

(5) S.C. Cunha, S.J. Lehotay et al., JSep Sci. 30, pp. 620-632 (2007).

(6) M. Anastassiades, K. Mastovska et al. J Chromatogr A., 1015, pp. 163-184 (2003).

(7) The Maximum Residue Limit Database. Available at: http://www.mrldatabase. com. Accessed August 2011.

(8) EU Pesticides Database. Available at: http://ec.europa.eu/sanco_pesticides/ public/index.cfm. Accessed August 2011.

(9) Codex Maximum Residue Limits. Codex Alimentarius Commission. Available at: http://www.codexalimentarius.net/pestres/data. Accessed August 2011.

www.agilent.com/chem

Agilent shall not be liable for errors contained herein or for incidental or consequential damages in connection with the furnishing, performance, or use of this material. Information, descriptions, and specifications in this publication are subject to change without notice.