Peppermint (Mentha piperita L.) a herbaceous plant which is indigenous to Europe but it is widespread cultivated throughout the world. Peppermint is one of the most widely consumed single ingredient herbal teas as well as it is an important row material for the food, cosmetic and farmaceutical industries. Due to a high menthol content, peppermint is often used for flavouring purposes. Apart of menthol, other major constituents include menthone and the phenolic compounds such as rosmarinic acid and several flavonoids. As peppermint is known for the bioactivity and potential health benefits, it is widely used in traditional medicines (McKay and Blumberg 2006). Under cultivation, peppermint is prone to fungal diseases such as mint rust and mint powdery mildew. Unwanted plants (weeds) can interfere with cultivated plants and diminish crop yields. Also, the crop can be damaged by harmful insects pests such as flea beetles, leafhoppers, lygus bugs, mint leaves beetles, tortoise beetles and tortrix moths. Pesticides are applied for preventing crop losses to harmful insects and other pests but each use of pesticides carries some associated risk due to the possibility of leaving residues on harvested crops. Due to the potential consumers’ health concerns, it is necessary to investigate the pesticides in peppermint in order to identify the residues present and quantify their concentrations. Therefore, there is an increasing demand for reliable pesticide analytical methods that are adequately sensitive, selective and accurate to detect and quantify pesticide.

For the purpose of pesticide residue analysis numerous methods have been developed. In the past few years, the miniaturized acetonitrile-base extraction technique know under the acronym name QuEChERS (quick, easy, cheap, effective, rugged and safe) has become one of the main analytical tools in many pesticide residue analysis applications (Lehotay et al. 2010). Although the QuEChERS approach is often considered as the method of choice for sample preparation before analysis by gas and liquid chromatography coupled to mass spectrometry, e.g. GC/MS, GC/MS/MS, LC/MS/MS, GC/TOF MS, LC/TOF MS, there is also an increasing interest in the adaptation of the QuEChERS with gas chromatography combined with element-specific detection such as electron capture (GC/ECD), nitrogen phosphorus (GC/NPD) and pulsed flame photometric (GC/PFPD) as a cost-effective alternative. The QuEChERS approach in conjunction with GC/ECD, GC/NPD or GC/PFPD has already been applied to analysis of pesticide residues in fruits and vegetables (Liu et al. 2011; Srivastava et al. 2011), toasted barley and chickpea flours (González-Curbelo et al. 2012a), American ginseng (Wu et al. 2011), banana leaves (González-Curbelo et al. 2011), traditional Chinese medicine (Xu et al. 2011), choi sum, yardlong beans and aubergines (Chai et al. 2012), cereal samples (González-Curbelo et al. 2012b), cereal-based baby foods (Anagnostopoulos et al. 2010) and low-fatty baby foods (Georgakopoulos et al. 2011). In this work, we report the application of QuEChERS to analysis of pesticide residues in peppermint. The main objective was to study the feasibility of a QuEChERS-based method for the extraction, cleanup and preconcentration of a group of 14 pesticides including eight insecticides, four herbicides and two fungicides from fresh peppermint prior to their determination by GC/ECD/NPD techniques. Detailed validation study was carried out in order to assess fitness for purpose of the proposed method.

Materials and Methods

Certified pesticide analytical standards were obtained from Witko (Łódź, Poland) except for fluazifop-P-butyl which was obtained from Institute of Industrial Organic Chemistry (Warszawa, Poland), and were of the highest available purity. Individual pesticide stock solutions were prepared at approximate concentrations of 1,000 μg mL−1 in residue grade acetone. A single composite pesticides mixture at approximately 10 μg mL−1 was prepared from the stock solutions by dilution with residue grade acetone. Subsequent working standards were prepared by dilutions of the appropriate volumes of the pesticide mixture with acetone. Matrix-matched calibration standards were prepared in peppermint matrix extract in petroleum ether. Pre-weighted mixtures of 4 g anhydrous magnesium sulphate, 1 g sodium chloride, 0.5 g di-sodium hydrogen citrate sesquihydrate and 1 g sodium citrate dehydrate sodium, and pre-weighted mixtures of 150 mg PSA, 45 mg GCB and 900 mg anhydrous magnesium sulfate were purchased from Perlan Technologies (Warszawa, Poland). Acetonitrile (POCH, Gliwice, Poland) and petroleum ether (Chempur, Krupski Młyn, Poland) were distilled from glass before use.

An Agilent Technologies 6890 gas chromatograph was employed for all sample analyses. The system was equipped with electronic pressure control (EPC) and an autosampler. Analytes were separated on a DB-1701 30 m × 0.25 mm × 0.25 μm column (Agilent Technologies, Folsom, USA. The carrier gas was nitrogen (purity 6.0). The column was connected to NPD and ECD using a universal Y-splitter. The ECD and NPD were maintained at 270 and 300°C, respectively. Nitrogen was the make up gas for the ECD (30 mL min−1). For the NPD, hydrogen and air flows were kept at 3 and 30 mL min−1, respectively. The make up gas was nitrogen at 30 mL min−1. The column temperature was held at 100°C for 1 min, increased at 20°C min−1 to 180°C which was held for 4 min, increased at 20°C min−1 to 260°C held for 30 min. Injector temperature was held at 250°C. Sample extract volumes of 2 μL were injected in the splitless mode. HP ChemStation, Rev. A 10.02 chromatography software was used for the instrument control, data acquisition and processing.

For sample extract preparation, the citrate buffered QuEChERS procedure was followed (Lehotay et al. 2010). However certain changes were made to adapt the method to our needs and laboratory resources. The sample size was reduced to 5 g, 10 mL water was added and the final extract was diluted with petroleum ether. To prepare sample extracts for GC/ECD/NPD analysis, 5 g of previously comminuted sample of fresh peppermint was weighted into a 50 mL disposable polypropylene centrifuge tube then 10 mL acetonitrile and 10 mL distilled water were added. The contents were shaken by hand for 1 min then a mixture of 1 g sodium chloride, 0.5 g di-sodium hydrogen citrate sesquihydrate, 1 g sodium citrate dehydrate and 4 g magnesium sulfate was added. Again, the contents were shaken by hand for 1 min and centrifuged at 3000 rpm for 5 min. A 6 mL aliquot of the upper layer was transferred into a 15 mL disposable polypropylene centrifuge tube with pre-weighted 150 mg PSA, 45 mg GCB and 900 mg anhydrous magnesium sulfate. The contents were shaken by hand for 2 min and centrifuged at 3000 rpm for 5 min. For the analysis by GC/ECD/NPD, a 2 mL aliquot of the supernatant was evaporated under a gentle stream of nitrogen and reconstituted in 2 mL of petroleum ether.

Results and Discussion

Determination of residual pesticides in peppermint can be considered as a difficult task since peppermint is known to contain large amounts of chlorophylls, other coloured compounds and peppermint oil. Crude extract of this kind of matrix can give rise to massive interferences in the GC determination and result in ambiguity of identification or complete suppression of chromatographic peaks of certain pesticides. Therefore, adequate cleanup of the extract is needed for reduction of co-extractives in order to enhance sensitivity (with less noise) and facilitate identification and quantification of low concentration of pesticides. For this purpose, a cleanup step was carried out by the combined use of primary secondary amine (PSA), a sorbent which exhibits retaining activity for sugars, fatty acids and other organic acids and graphitized carbon black (GCB), a sorbent which exhibits retaining activity for planar compounds such as natural pigments and sterols. Analytical performance of the selected approach was subjected to validation study in terms of trueness, precision, selectivity, limits of quantification and measurement uncertainty in order to assess fitness for purpose of the method.

Selectivity of the method was evaluated considering the absence of interfering peaks from co-extractives at the retention time of each target pesticide. For this purpose extracts of organically grown peppermint were analyzed (with no detectable pesticides). Linearity of calibration curves was investigated over the concentration range between 0.01 and 1.00 μg mL−1 by GC/ECD/NPD analysis of matrix-matched calibration standards (five levels). Over the studied concentration range, the linearity of GC/ECD/NPD responses was highly satisfactory with coefficients of determination (R2) higher than 0.99 (Table 1).

Table 1 Linearity parameters (range, calibration curve equation and R2) obtained by GC/ECD/NPD analysis of pesticide standards prepared in peppermint matrix extract
Full size table

Accuracy was evaluated by means of spiking recovery determinations at three concentration levels of 0.01, 0.1 and 1.0 mg kg−1 (six replicates per level). Precision was calculated from the recovery experiments and it was expressed in terms of relative standard deviation (RSD; n = 6) at each spiking level. Overall recovery and RSD for each individual pesticide was also determined. The spiking experiments were carried out on organically grown peppermint. Figure 1 shows example chromatograms of peppermint sample spiked with the pesticides at 0.1 mg kg−1. The recovery results are shown in Table 2 indicating that the recoveries were in the acceptance range of 70 %–120 % with RSD ≤ 20 % for all the analyzed pesticides. The limit of quantification (LOQ) was established as the lowest spiking level at which the method fulfilled the aforementioned validation criteria, which was 0.01 mg kg−1 for all the pesticides. The LOQ values were less or equal to the MRLs required by the European legislation which were in the range from 0.01 mg kg−1 (diazinon and fenithrotion) to 2 mg kg−1 (cypermethrin). Uncertainty was estimated following the “top-down” experimental model using overall recovery and precision data. The expanded uncertainty was calculated individually for each pesticide as twice the value of the uncertainty (k = 2, confidence level 95 %), and ranged between 7 % and 30 %, with an average of 14 % (Table 2). These uncertainty values comply with the European Union requirement of a maximum default value of 50 %. As evident from the obtained method performance characteristics, the method is suitable for the intended application.

Fig. 1
figure 1

Example chromatograms of peppermint sample spiked with the target pesticides at 0.1 mg kg−1

Full size image
Table 2 Method performance data—average recovery, RSD and expanded uncertainty
Full size table

Our previous method for analysis of pesticide residues in peppermint entailed extraction of pesticides with dichloromethane/acetone mixture followed by cleanup on column packed with florisil and sodium sulfate and collection of eluates in four fractions which were then analysed by GC/ECD/NPD separately. Using the former method, the total organic solvents consumption was over 350 mL (Sadło et al. 2006). The current method is greatly improved compared with the previous one as it is more straightforward, less time and labour intensive, and more sensitive, selective and accurate, simultaneously. The consumption of organic solvent is approximately 30-fold less, and the usage of environmentally concerned chlorinated solvents is eliminated. The current analytical scope includes 14 pesticides but clearly the method can be expanded to include more pesticides unless limited by GC capillary column resolution or excessive presence of matrix interferents.