Analysis of Pesticides in Oranges: Exploring QuEChERS for Peel, Pulp, and Whole Fruit

When someone is evaluating dissimilar things, you might hear them say “it’s like comparing apples and oranges.” In the world of pesticide residue analysis, where scientists evaluate real apples and oranges, there are big differences between them, but the variation doesn’t stop there. Even within a commodity, there are potentially significant differences that can affect results. In this article, we will discuss the analysis of pesticides in oranges using QuEChERS and look at the effect of extracting whole fruit versus separate peel and pulp samples.

From an analytical perspective, an orange’s peel has characteristics that differ from those of the pulp within it. The principal differences between the peel and pulp considered here are the relative water, oil, and sugar content. Pulp contains high amounts of water and sugar compared to peel, whereas peel contains less water and more oil than pulp.

When choosing sample preparation procedures, it is common to tailor the approach to the characteristics of the commodity. But, despite the very different characteristics of the orange’s component parts, the whole fruit is often processed for analysis. While preparing the whole orange certainly can work for many pesticides, there may be instances where poor recoveries are observed. Poor recoveries have many potential causes, but in the case of heterogeneous samples, such as oranges, differences in the nature of the sample components could be a contributing factor. Here we present two examples that explore the analysis of pesticides in oranges and the differences that were observed based on sample preparation approaches.

Example 1: A workflow where preparing peel and pulp separately made a significant difference

During a project where we were evaluating various QuEChERS products for the extraction and cleanup of a list of LC-amenable pesticides (cat.# 31971) in a wide variety of spiked commodities using LC-MS/MS, we observed that when the orange was prepared separately as peel and pulp samples, a significantly larger number of pesticides exhibited higher recoveries in the separated peel and pulp samples compared to the whole orange sample (Figure 1).

For all cases in this first example, the whole orange, peel, and pulp samples were extracted using buffered AOAC method extraction salts (cat.# 25852) and cleanup was performed using a dSPE product containing only primary and secondary amine (PSA) and magnesium sulfate (cat.# 26124).

In the whole orange sample, no additional water was added because oranges have adequate water content, about 86%. However, when the peel and pulp were analyzed separately, the peel did get extra hydration because it contains less than 80% water, which is a threshold used for QuEChERS extractions. Table I summarizes the sample preparation processes and the analytical conditions used for this analysis of pesticides in oranges.

When comparing pesticide recoveries between the whole orange sample and the pulp sample, the pulp results showed a marked increase in pesticides recovered in the target 70-120% range compared to the whole orange sample. It should be noted that the sample preparation procedures for these two samples were identical, so the observed difference is likely due to the parts of the orange that were analyzed. The peel sample also showed an increase in the number of pesticides within the desired recovery range, but it underwent additional hydration, which likely helped with analyte recovery. When evaluating sample preparation approaches, one must consider whether or not whole commodity procedures thoroughly homogenize the sample so that the extraction process is as effective as possible. Could it be in our example that the peel’s relatively low-moisture characteristic affected the performance of the whole orange sample, ultimately degrading overall performance? A clue to the answer may lie in a second experiment we conducted.

Table I: Sample Preparation and Analytical Methods used in LC-MS/MS Analysis of Pesticides in Oranges.

Example 2: A workflow where the differences were not as pronounced.

As a complementary project, we also used a GC-MS/MS to evaluate a list of GC-amenable pesticides (cat.# 32562) in the same set of food matrices spiked with the same concentration of pesticides (10 ng/g) that were studied in the LC-MS/MS work. Once again, a QuEChERS approach was used, but in this case we chose to use optimized procedures for each orange part instead of using the same approach for all. This allowed us to take advantage of tailoring the extraction and dSPE to the characteristics of the individual parts (more detail about this is available in this blog). Table II outlines the sample preparation and analytical conditions used in these different GC workflows.

Table II: Sample Preparation and Analytical Methods used in GC-MS/MS Analysis of Pesticides in Oranges.

In addition to optimizing the extraction conditions for the individual parts, another homogenization step, accomplished using 1⁄16’’ steel ball bearings in a centrifuge tube, was used for the whole orange sample. This step was added in an attempt to mitigate the different natures of the peel and the pulp negatively affecting the overall recoveries. This allowed for better mixing between the pulp and peel parts of the orange. Figure 2 shows the recoveries of the GC-amenable compounds that were included in this analysis of pesticides in oranges. We did not observe as much of a difference between analyzing the orange as a whole versus as individual parts. There was only a 3% increase of pesticide residues from whole orange to orange pulp. Also, the peel performed much better (97% in target zone), which suggest that matrix effects from the orange are less severe in GC-MS/MS analysis.

While this case does not show significantly different overall recoveries when comparing the whole orange to the separate peel and pulp samples, we do notice, in some cases, that there are significantly different pesticide concentrations recovered from the peel as compared to the pulp. When analyzing an orange for levels of incurred pesticides, we observed much higher levels of some pesticides, like fludioxonil, in the peel than we did in the pulp (Figure 3). We also found cypermethrin and diphenylamine in both peel and whole orange, but not in the pulp (Table III). The different parts of a very heterogeneous sample may exhibit different levels of target pesticides, and if it is important to make that distinction, analyzing the sample parts individually, with optimized sample prep processes for each part, is necessary.

	Peaks	tR (min)	Precursor Ion	Product Ion	Collision Energy (CE)
1.	Fludioxonil	18.16	248	127	30

Column	Rxi-5ms, 30 m, 0.25 mm ID, 0.25 µm (cat.# 13423)
Standard/Sample
	Whole orange QuEChERS extract (318 ng/g)
	Orange peel QuEChERS extract (594 ng/g)
	Orange pulp QuEChERS extract (6.8 ng/g)
Diluent:	Acetonitrile
Injection
Inj. Vol.:	1 µL splitless (hold 0.5 min)
Liner:	Topaz 4.0 mm ID single taper inlet liner w/wool (cat.# 23447)
Inj. Temp.:	250 °C
Oven
Oven Temp.:	90 °C (hold 1 min) to 330 °C at 8.5 °C/min (hold 5 min)
Carrier Gas	He, constant flow
Flow Rate:	1.4 mL/min

Detector	TSQ 8000
Transfer Line Temp.:	290 °C
Analyzer Type:	Quadrupole
Source Temp.:	280 °C
Tune Type:	PFTBA
Ionization Mode:	EI
Instrument	Thermo Scientific TSQ 8000 Triple Quadrupole GC-MS
Sample Preparation	Sample Preparation (adapted from EN and AOAC QuEChERS methods): Orange (Whole): 10 g of homogenized orange was fortified with an internal standard and mixed with 10 mL acetonitrile and two steel ball bearings (1/16'') and shaken for 10 minutes. Q-sep QuEChERS extraction salts for EN 15662 (cat.# 25849) were added, and the sample was shaken for 1 minute. The sample was centrifuged for 5 minutes, and the supernatant was removed and further cleaned up with prefilled Q-sep QuEChERS dSPE tubes containing 150 mg MgSO4, 50 mg PSA, and 50 mg C18-EC (cat.# 26125). Orange Peel: 10 g of homogenized orange peel was fortified with an internal standard, hydrated with 4 mL water, and briefly shaken by hand. Acetonitrile (10 mL) was added, and the sample was shaken for an additional 1 minute. Q-sep QuEChERS extraction salts for EN 15662 (cat.# 25849) were added, and the sample was shaken for 1 minute. The sample was centrifuged for 5 minutes, and the supernatant was removed and further cleaned up with prefilled Q-sep QuEChERS dSPE tubes containing 150 mg MgSO4, 25 mg PSA, and 25 mg C18-EC (cat.# 26216). Orange Pulp: 15 g of homogenized orange pulp was fortified with an internal standard. Acetonitrile with 1% acetic acid (15 mL) was added, and the sample was shaken for 1 minute. Q-sep QuEChERS extraction salts for AOAC 2007.01 (cat.# 25851) were added, and the sample was shaken for an additional 1 minute. The sample was centrifuged for 5 minutes, and the supernatant was removed and further cleaned up with prefilled Q-sep QuEChERS dSPE tubes containing 150 mg MgSO4 and 50 mg PSA (cat.# 26124). The dSPE supernatant of all samples was analyzed directly (no further dilution) within 24 hours of the initial extraction.

Download PDF

Table III: Incurred pesticides (ng/g) in different parts of oranges. (ND = not detected)

Conclusion

This article is intended to introduce an idea for your lab to consider: if a given sample has different parts that exhibit very different characteristics, and if you are experiencing poor recoveries when analyzing the commodity as a whole, it may be valuable to evaluate the parts separately. We have noticed in our own studies that complex samples can give poor recoveries. To improve recoveries, testing distinct parts of the sample (e.g., peel and pulp) using optimized sample preparation procedures for each part may help. In summary, when adapting or developing methods for the analysis of pesticides in oranges—or other heterogeneous sample commodities—it might be beneficial to explore whether or not an individual component or whole-sample approach yields the best results.