Method Development in the Use of Solid Phase Microextraction for the GC/MS Analysis of Pesticide Residues in Baby Food

By: Klaus Buckendahl1, Katherine K. Stenerson2, Tyler Young (Summer Intern)2, Robert Shirey2, and Yong Chen2,
1Sigma-Aldrich Chemie GmbH, Taufkirchen, Germany
2MilliporeSigma, Bellefonte, PA, USA

Introduction

Immersion Solid Phase Microextraction (SPME) has not been a common approach for pesticide residue analysis of foods. The high background in many foods poses a challenge to immersion SPME due to the presence of fats, sugars, pigments and other macromolecules. These can stick to the fiber and reduce its usable life and/or be transferred to the GC, where they may interfere with chromatographic analysis. A new development in SPME fiber technology incorporates a protective overcoating on the fiber. This overcoating, which consists of polydimethylsiloxane (PDMS) protects the fiber, making it more physically robust and less prone to chemical fouling.

In this work, an SPME method was developed using an overcoated (OC) version of a PDMS/DVB fiber for the immersion extraction of specific pesticides from pureed prune baby food. The pesticides studied were selected due to their inclusion in EU directive 2006/125/E for baby food.

The optimized SPME method was then used with a standard, non-overcoated (non-OC) version of the PDMS/DVB fiber. The OC fiber showed advantage over the non-OC version with regards to data quality and physical durability.

Experimental

  • Pureed prune baby food was obtained from a local store and spiked at 10 ng/g. After spiking, samples were allowed to
    equilibrate for a minimum of 3 hours prior to extraction. This time was to allow for any matrix binding to occur
    between sample and analytes.
  • The final optimized SPME method is listed in Table 1.
  • Analysis was done by GC/MS-SIM.
  • Samples were quantitated against a calibration curve from 1-20 ng/g prepared in blank pureed prune baby food. No
    internal standards were used.


Table 1. Final, Optimized SPME Method
 

sample:    10 mL vial containing 4 g prune baby food + 4 mL 0.1 M
sodium phosphate buffer at pH 7 containing 25% NaCl
fiber:    65 μm PDMS/DVB-OC fiber and standard non-OC PDMS/DVB fiber
incubation:    50 °C 6 min with agitation at 600 rpm
extraction:    immersion 30 min at 50 °C  with agitation at 250 rpm;
vial penetration set to 30 mm
post extraction wash:    30 sec. in deionized water with agitation
desorb:    250 °C 3 min splitless; injection penetration set to 45 mm
postbake: 
260 °C 2 min  split 10:1 (done in second GC inlet)


SPME Method Optimization

  • Sample Dilution
    - Decreased sample viscosity allowed for sufficient mixing during incubation and extraction steps.
    - Phosphate buffer at pH 7 was used as the diluent to maintain a consistent sample pH.
  • Addition of Salt (Figure 1)
    - Improved reproducibility for many of the pesticides.
  • Mixing During Incubation
    - Created a homogenous mixture for the extraction step (in which agitation was slowed to 250 rpm).
  • Post Extraction Wash
    - Increased fiber life by removing residual sample.
    - No loss in pesticide response seen with 30 sec wash time.
  • Increased Equil./Extraction Temp. from 30 °C to 50 °C (Table 2)
    - Improved linearity; increased r2 values for most pesticides.
    - Improved accuracy; >80% for most pesticides.
    - Improved reproducibility; lower %RSDs for most pesticides.

 

Effect of Salt on SPME Method Reproducibility, Pesticides Spiked at 10 ng/g in Prune Baby Food

Figure 1. Effect of Salt on SPME Method Reproducibility, Pesticides Spiked at 10 ng/g in Prune Baby Food

 

Table 2. Effect of Increased Extraction Temperature on SPME Method Linearity, Accuracy and Precision
 

  r2 values; 1-20 ng/g Accuracy; 10 ng/g spks %RSD; n=5
Equil./Extraction temp.: 30 oC 50 oC 30 oC 50 oC 30 oC 50 oC
Demeton-S-Methyl NC 0.993 NC 58% NC 28%
Ethoprophos 0.992 0.996 72% 100% 13% 6%
Cadasufos 0.992 0.990 76% 108% 10% 6%
Hexachlorobenzene 0.971 0.997 74% 106% 10% 7%
Terbufos 0.994 0.990 65% 82% 10% 6%
Heptachlor 0.970 0.991 66% 105% 14% 9%
Aldrin 0.978 0.989 84% 92% 7% 13%
Fipronil 0.988 0.996 69% 117% 10% 4%
Dieldrin 0.975 0.994 72% 107% 12% 6%
Endrin 0.992 0.991 87% 112% 6% 6%
Nitrofen 0.984 0.995 77% 121% 9% 6%
Fensulfothion 0.994 0.999 65% 103% 15% 4%

NC: not calculated

Comparison of OC and non-OC Fibers

  • Replicate sets of spiked prune baby food were extracted with the optimized SPME method using OC and non-OC
    PDMS/DVB fibers.
  • Accuracy and reproducibility were better using the OC fiber (Figure 2).
    - Accuracies for 10 of 12 pesticides were in the range of 80-120% (indicated in yellow) using the OC fiber.
       6 of 12 were within this range using the non-OC fiber.
    - The two most hydrophobic pesticides, aldrin & dieldrin, showed significantly better results using the OC fiber.
    - Demeton-S, the most polar pesticide in the list, was difficult to extract using SPME, regardless of fiber type.
  • GC/MS background (Figure 3) was lower using the OC fiber; especially in the pesticide elution range.

Comparison of Results for Extraction of Pesticides from Spiked Prune Baby Food; OC vs non-OC PDMS/DVB Fibers

Figure 2. Comparison of Results for Extraction of Pesticides from Spiked Prune Baby Food; OC vs non-OC
PDMS/DVB Fibers

GC/MS-SIM Background of Pureed Prune Baby Food Extracted by SPME (same Y-scale)

Figure 3. GC/MS-SIM Background of Pureed Prune Baby Food Extracted by SPME (same Y-scale)

GC Method Ruggedness and Fiber Durability

  • 25 extractions of spiked prune baby food were performed with the OC and non-OC fibers and pesticide response was
    monitored.
  • Decline in response was much more rapid with the non-OC fiber. Examples for two of the pesticides are shown in
    Figure 4.
  • After the test sequence, the OC fiber showed some patches of discoloration, however the white color of the DVB layer
    was still in good condition (Figure 5).
  • The non-OC fiber became physically detached from its assembly after the 25th extraction and was lost.

Pesticide Response from OC and non-OC Fibers over Repeated Extractions of Spiked Prune Baby Food

Figure 4. Pesticide Response from OC and non-OC Fibers over Repeated Extractions of Spiked Prune
Baby Food

 

OC PDMS/DVB SPME Fiber after Repeated Extractions of Pureed Prune Baby Food

Figure 5. OC PDMS/DVB SPME Fiber after Repeated Extractions of Pureed Prune Baby Food

Conclusions

  • A direct immersion SPME method using an overcoated PDMS/DVB fiber was developed for extraction of pesticides
    from baby food. The final method yielded accuracies of >80% and % RSD values of <15% for all but one pesticide.
  • Optimization of the SPME method included sample dilution, vigorous mixing, increased incubation/extraction
    temperature, and a post-extraction fiber wash step.
  • The overcoated PDMS/DVB fiber showed advantages over the standard, non-overcoated version of the same fiber
    for method accuracy, reproducibility and ruggedness.

Materials

     
Related Links