Simultaneous Determination of Multi-Class Pesticide Residues and PAHs in Plant Material and Soil Samples Using the Optimized QuEChERS Method and Tandem Mass Spectrometry Analysis

New analytical approaches to the simultaneous identification and quantification of 94 pesticides and 13 polycyclic aromatic hydrocarbons (PAHs) in five representative matrices (pepper, apple, lettuce, wheat, and soil) were developed. The analyses were based on gas chromatography coupled with triple quadrupole tandem mass spectrometry (GC-MS/MS). The procedure was optimized by changing the solvent used during the extraction, from acetonitrile to the acetone: n-hexane mixture at a volume ratio of 1:4 (v/v), as well as the use of a reduced amount of water during the extraction of compounds from cereals. An additional modification was the use of florisil instead of GCB in the sample cleanup step. A full method validation study was performed, at two concentration levels (LOQ and 1000 × LOQ), which showed satisfactory results for all analytes from the PAHs group, with recoveries ranging from 70.7–115.1%, and an average RSD of 3.9%. Linearity was tested in the range of 0.001–1.000 mg/kg and showed coefficients of determination (R2) ≥ 0.99 for all PAHs. Satisfactory recovery and precision parameters (LOQ and 100 × LOQ) were achieved for almost all analytes from the pesticide group in the range of 70.1–119.3% with the mean RSD equal to 5.9%. The observed linearity for all analytes in the concentration range of 0.005–1.44 mg/kg was R2 ≥ 0.99, with the exception of famoxadone, chizalofop-p-ethyl, prothioconazole, spirodiclofen, tefluthrin, and zoxamid. The extended uncertainties were estimated, using a top-down approach of 9.9% (average) and 15.3% (average) for PAHs and pesticides samples, respectively (the coverage factor k = 2, the 95% confidence level). Ultimately, the method was successfully applied to determine pesticide residues in commercial samples of fruit, vegetables and grain, and soil samples for PAHs, which were collected from selected places in the Podkarpacie region. A total of 38 real samples were tested, in which 10 pesticides and 13 PAHs were determined. Proposed changes allow us to shorten the sample preparation time (by 20%) and to reduce the consumption of organic solvents (by 17%). The use of florisil for sample cleanup, instead of GCB, improves the recovery of compounds with flat particles.


Introduction
Food quality and safety is becoming an increasingly important issue for consumers worldwide. Equally important topics concern environmental pollution, including soil, water, and air pollution with chemical agents. The presence of pesticides and polycyclic aromatic hydrocarbons (PAHs) in the environment is closely associated with human activity. The first of those compounds are related to the use of plant protection products in the agriculture, while the presence of the second group of substances results from human industrial activities.
Pesticides are compounds widely used to control pests and stabilize plant production [1]. Naturally, these effects provide many benefits, however, the unskillful or excessive use of these chemicals can be harmful not only to consumers' health but also to the environment, and it is associated with the presence of pesticide residues. Pesticides cover an extensive range of chemicals, which are classified on the basis of their active ingredients, chemical structure, mode of action and toxicity. Their action is based on a disruption of: the synthesis of amino acids and proteins in weeds, the nervous system function in insects, cell division, energy production, respiration, growth regulation or development of photosynthesis, the deoxyribonucleic acid (DNA) methylation, and other effects [2,3]. Due to the high persistence of pesticides in the environment, their high biological activity, and a wide spectrum of toxic activity, these substances must be subject to strict legal regulations. Many countries and international organizations have defined Maximum Residue Levels (MRLs) of pesticides for foodstuffs, and in the European Union they are based on Commission Regulation (EC) No. 396/2005, as amended [4].
PAHs are common environmental pollutants that can come from various processes of incomplete combustion and pyrolysis. PAHs and their derivatives belong to a large class of organic compounds whose molecules consist of at least two condensed aromatic rings with a planar spatial structure. Those compounds that contain five or more aromatic rings are called heavy PAHs, while those containing less than five rings are light PAHs [5]. These substances are considered to be highly hazardous as they exhibit genotoxic and carcinogenic properties in the human body [6,7]. Numerous studies showed that these substances are found in food [8][9][10][11], and this may be caused by air and water pollution, the soil up taking, or thermal processing of food and raw products. In order to minimize the risks associated with the consumption of food containing PAHs, monitoring of contamination with these substances is essential; therefore, MRLs were established for these chemicals and included in the Commission Regulation (EC) No 1881/2006 [12], the Commission Regulation (EU) No 835/2011 [13] and the Regulation of the Minister of the Environment, 2016 [14].
In consequence, the analysis of trace amounts of pesticides and PAHs in food, foodstuffs and environmental samples has become essential. Currently, efforts are focused on developing very flexible analytical methods that would enable the determination of as many substances as possible in one analytical process, which are cheaper (allowing a reduction in the amount of solvents used) and, of course, environmentally friendly (eliminating toxic reagents and replacing them with less harmful ones). An equally important aspect in light of modern methods is also the shorter preparation time of the sample and further analysis using instrumental methods.
The procedure for the determination of pesticides and PAHs consists of many successive stages, including sampling, extraction, purification, and identification and quantification of tested substances. Proper sample preparation has a great influence on the quality and reliability of the obtained results [15].
In recent years, many studies on the selection of appropriate methods for the analysis of pesticides and PAHs, especially in food and soil, have been published ( Table 1).
The most commonly used extraction and purification methods in the analysis of pesticide residues and PAHs are extraction with such reagents as acetonitrile, dichloromethane, ethyl acetate, methanol, as well as combinations, like acetonitrile-water or acetone-n-hexane (1:1), and a wide spectrum of salts and sorbents (Table 1), which effectively remove many co-extracted components from the complex sample matrix, such as dyes, fats, sugars, polyphenols, organic acids, and other. These contaminants could hinder the precise quantitative and qualitative course of the analysis [16,20]. The selection of the sample preparation process is closely related to the type of analyzed product and tested substances, including their physico-chemical properties. The QuEChERS concept has still remained a very popular method in recent years, and it is subject to numerous modifications and optimizations. The most popular instrumental methods for the determination of pesticide residues and PAHs are gas and liquid chromatography, most often coupled with mass spectrometry (Table 1). Table 1. Examples of methods application in pesticides and PAHs analysis from various sample type. 10 mL of ACN:water (1:1), cooled for 20 min, 4 g of MgSO 4, 1 g of NaCl, 1 g of trisodium citrate dihydrate, and 0.5 g of disodium hydrogencitrate sesquihydrate/cooled overnight/60 mg of PSA, 50 mg of C18/filtered through nylon syringe filter, 30  In this work, the effectiveness and efficiency of the modified method based on the QuEChERS procedure [33], followed by gas chromatography coupled with triple quadrupole tandem mass spectrometry (GC-MS/MS) for the simultaneous determination of 94 pesticide residues and 13 substances from the PAH group in samples of plant products and soil were assessed. A detailed validation study was performed in order to assess fitness for purpose of the proposed method. Furthermore, to demonstrate the suitability of the optimized method, real samples were screened for pesticide residues and PAHs.

Results and Discussion
In this study, an alternative procedure of preparing samples for the analysis of pesticide residues and/or PAHs by gas chromatography and/or mass spectrometry in plant material and soil was elaborated. In order to assess the usefulness of the proposed method, validation was carried out with the determination of such parameters as: linearity, recovery, precision, limits of quantification, the working range of the method, and its uncertainty. The validation procedure was performed in accordance with the guidelines of the European Commission contained in the documents: SANTE [34] and the Commission Regulation (EU) [35].
In the present study, 94 pesticides (and their isomers) and 13 PAHs were selected for validation experiments. All included substances were amenable to the GC analysis, and were currently recommended in Polish agriculture or can be potentially used for protection of crops against pests and diseases. The study also covered persistent organochlorine pesticides and PAHs, commonly known as environmental contaminants. All selected analytes represent various structure classes, and they are characterized by highly varied physico-chemical properties.
In order to verify the suitability of the developed method for a specific purpose, it was used to determine the concentration of pesticide residues and PAHs in real samples.

Method Optimization and Validation
Samples of plant material, or other samples containing biological material, are associated with the diversity and complexity of matrices. Due to low concentrations of pesticide residues or PAHs in the samples, the critical stage is the appropriate preparation of sample extracts and reduction in interferences, to improve the analysis parameters [15]. The quantitative and qualitative assessment of the discussed compounds is very difficult and demanding, therefore the analytical methods used for this purpose are constantly developed. Today, specific and selective analytical methods play an important role in ensuring the correct and reliable determination of compounds, especially pesticide residues in food of plant origin, and they include the QuEChERS method combined with GC and HPLC chromatographic techniques with the MS and MS/MS mass spectrometry detector [36].
Validation studies were performed for four representative plant matrices: apple (from the group of plants with high water content, for which the preferred extract purification is conducted with primary and secondary amines (PSA)), wheat (from the group of cereal and seed matrices, purification with PSA and silica gel modified with octadecyl groups (C18EC)), peppers (from the group of pigment matrices, purification with primary and secondary amines (PSA) and florisil for PAHs or GCB for pesticides), and lettuce (from the group of highly pigmented matrices, purification with primary and secondary amines (PSA) and florisil for PAHs or GCB for pesticides), and for the soil (PSA treatment) ( Table 2). Studies were carried out to assess the effectiveness of extraction and cleanup of the 94 target pesticides and 13 PAHs with different physical and chemical properties at two spiking levels (LOQ and 100 × LOQ for pesticides and LOQ and 1000 × LOQ for PAHs) (Details in Tables 3 and 4). Plant and soil samples were prepared according to the optimized QuEChERS procedure. For the discussed samples, a full validation study in terms of evaluation of linearity, recovery, precision, as well as estimation of measurement uncertainty was performed. Details of the validation parameters are presented in Tables 3 and 4. 10 a-salts used for the purification of sample extracts (pesticides); b-salts used for the purification of sample extracts (PAHs)-method modification-GCB replaced with florisil, which prevents the loss of planar compounds such as PAH.              Linearity was evaluated by studying five-level calibration curves plotted for pesticide and PAH standards prepared in blank extracts of the tested matrix, over a concentration range of 0.001-1.000 mg/kg (PAHs) and 0.005-1.44 mg/kg (pesticides). The linearity parameters were excellent with coefficients of determination (R 2 ) ≥ 0.99 for all tested PAHs and almost all pesticides, except for famoxadone, chizalofop-p-ethyl, prothioconazole, spirodiclofen, tefluthrin and zoxamid. Detailed linearity data (linearity range, and R 2 values) can be found in Tables 3 and 4.
Obtained limits of detection (LOD) values ranged from 0.0002 to 0.005 mg/kg and 0.0003 mg/kg for pesticide and PAHs, respectively. Limits of quantification (LOQ) were set at the lowest spiking concentrations, and they were verified by recovery tests (n = 5) and the estimation of the relative standard deviation (RSD) for the obtained results in the tested matrix. Satisfactory recovery and precision parameters were obtained for all validated PAHs and for almost all pesticides (88 of 94 active substances) at this lowest level. These values were within the range of 0.001 mg/kg and 0.005-0.014 mg/kg for PAHs and pesticides, respectively. The highest spiking level for the tested samples was 1.000 mg/kg for PAHs and from 0.5 to 1.44 mg/kg for pesticide residues. All data have been shown in Tables 3 and 4. Nine chromatograms for samples of a different matrix were added in Supplementary Material (Figures S1-S9).
Trueness and precision in terms of mean recovery and RSD were assessed for the proposed method by performing recovery experiments at two concentrations-these levels corresponded to the lowest and highest values of the linear range. The analysis of the spiked samples was performed in five replications at each level, using matched standards prepared in the apple, lettuce, wheat, pepper and soil extracts. The mean recovery of PAHs in samples ranged from 70.7-115.1% with an average RSD of 3.9% (for the soil matrix 71-100%, RSD 0.2-7.5%; cereals 82-109%; RSD 0.2-4.1%; apple 79-115%, RSD 0.4-9.0%; lettuce 93-115%, RSD 1.3-13.7%; and pepper 76-115%, RSD 1.3-7.3%) ( Table 4). These values meet the requirements of European recommendations-the recovery was within the range of 50-120% [35]. On the other hand, for pesticide residues, the average recovery was from 70.  [34], with the recovery ranging from 70-120%, with a relative standard deviation (RSD) of ≤20% (Table 3).
The working range of the method was established by determining the limit of quantification and the highest level of spiking, whose acceptance criteria for accuracy, precision and linearity were met (Tables 3 and 4).
The expanded measurement uncertainty was estimated by identifying all possible sources of uncertainty throughout the analytical process and calculating the uncertainty associated with each of them, following the "top-down" experimental approach [37]. The results showed that repeatability and recovery are the most important sources of uncertainty. The remaining components of the uncertainty budget, such as uncertainty related to weighing, dilution of standards, and purity of standards, are below 2% in the optimized method. The expanded uncertainty was calculated by multiplying the complex standard uncertainty by the coverage factor k. A probability of 95% for the coverage factor k = 2 was assumed. Uncertainty ranged from 4.2 to 18.8%, with an average uncertainty of 9.9% and from 4.2 to 31.4% with an average uncertainty of 15.3% for PAHs and pesticides, respectively (Tables 3 and 4). According to the SANTE document [34] the uncertainty criterion for pesticide analysis is 50%.
The proposed modification of the PN-EN 15662: 2018 [33] standard method, known as the QuEChERS procedure, consists of changing the solvent used during the extraction from acetonitrile to the acetone: n-hexane mixture at a volume ratio of 1:4 (v/v). This change eliminates the last stage of sample preparation consisting of evaporation of acetonitrile immediately before the pesticides/PAHs analysis by gas chromatography and/or mass spectrometry, which in turn allows us to shorten the sample preparation time (by 20%) and to reduce the consumption of organic solvents (from 12 mL up to 10 mL per sample, i.e., by 17%). Moreover, it should also be noted that this method enables the simultaneous preparation of sample extracts both for the analysis of PAHs and pesticide residues (the standard method applies only to pesticides).
The use of a reduced amount of water during the extraction of compounds from cereals ensures a better separation of the organic and inorganic phases. On the other hand, in the case of soil samples, a hydration step was necessary (due to the very low water content in the samples). The addition of 10 mL of water did not reduce the extraction efficiency or affect the accuracy of the results. For both the soil and wheat matrices, cold water was added to the samples to compensate for the effect of the exothermic reaction (after adding magnesium sulfate at the extraction step) and loss of thermolabile compounds [38].
On the other hand, the use of florisil at the stage of sample cleanup (recommended for pigment and high-pigment matrices in the QuEChERS method), instead of the commonly used GCB, improves recovery (with an increase from 10% to 100% in the case of lettuce and peppers) of compounds with flat particles that are adsorbed on the surface of carbon (GCB). The effectiveness of replacing GCB with florisil, leading to better validation results, has already been confirmed in previous studies [39].
The use of an extraction mixture consisting of acetone and n-hexane (1:4 v/v) enables elimination of a more toxic organic solvent, i.e., acetonitrile (NDS of 600 mg/m 3 , 70 mg/m 3 , and 72 mg/m 3 for acetone, acetonitrile, and hexane, respectively).
The authors of the works related to the development of methods for the determination of PAHs and pesticide residues, both separately and simultaneously, were guided by the basic issues, such as: selection and amount of the sample, type of extraction solvent, pH effect, type and amount of salt and sorbents used in the purification phase and the type of analyte separation technique ( Table 1). The advantage of the most popular extraction solvent, acetonitrile, is its compatibility to chromatographic applications, although it interferes with specific GC detectors, e.g., for nitrogen, and is definitely less volatile than other organic solvents, which may extend the evaporation and concentration stage [40,41]. Moreover, due to the relatively low solubility of lipids in acetonitrile, the co-extraction of lipids with this solvent is quite low, however, there may be limitations concerning the availability of pesticides from lipids and loss of non-polar pesticides [40]. Other non-halogenated solvents used, such as acetone, ethyl acetate or hexane, are much less polar in relation to acetonitrile; therefore, polar pesticides of medium and high polarity have a much better solubility in acetonitrile and thus higher recoveries when it is used for extraction. In addition, ethyl acetate is also characterized by the ability to extract lipids and waxes, and gives lower recoveries when compared to acid/alkaline pesticides and generally lower purification efficiency in DSPE [42]. Extraction with hexane is more characteristic for hydrophobic components in aqueous matrices, and for very low extraction of polar matrix components (proteins, amino acids, carbohydrates, etc.) [43]. Table 1 shows examples of sorbents and salts used in the purification stage. The most frequently used compounds were PSA (primary secondary amine), octadecyl sorbent C18, SAX (quaternary amine), Z-Sep (sorbent based on modified silica gel with oxide zirconium) and GCB (graphitized carbon black) [16,17,20,25,30]. These are sorbents that are responsible for the removal of polar organic acids, some sugars, lipids and sterols, carotenoids and chlorophyll [44]. Other commercially available sorbents are Z-Sep Plus (is a sorbent based on modified silica gel with zirconium oxide dual bonded on the same silica particles) [45,46], EMR-Lipid ("enhanced matrix removal" to selectively remove lipids from extracts of fatty foods) [47,48], ChloroFiltr (polymetric sorbent to removal of chlorophyll) [49] and ENVI-Carb (is used to eliminate polar compounds, pigments and polyphenols) [50]. In case of flat molecules, the use of GCB introduces strong limitations in the recovery of analytes, which was also confirmed in our research. For this reason we decided to replace this sorbent with florisil. This change resulted in a significant improvement in recoveries (from 10% to 100%). Generally, florisil is magnesium silicate SiO 2 + MgO, a polar compound classified as amphoteric. It is intended to be used for the isolation of hydrophilic polar substances from non-aqueous, non-polar mixtures, for analyses of samples with a high content of lipids, waxes or oils, for adsorption of pesticides from environmental samples, or for separation of aromatic compounds from aliphatic-aromatic mixtures, as well as for similar applications. [51]. The salts used for the extraction and purification, or their combinations (MgSO 4 , NaCl, Na 2 SO 4 ) enable a good separation of the aqueous and organic phases, as well as better recoveries, especially of polar compounds, which may otherwise be lost in the aqueous phase [39,52]. On the other hand, the compounds used in the described procedures, trisodium citrate dihydrate and disodium hydrogen citrate sesquihydrate, affect the recovery of acid-labile analytes [44].
The last steps in the procedure for the determination of PAHs and pesticides involves instrumental analysis, most often conducted using the HPLC-MS/MS, GC-MS/M, or LC-MS/MS systems (Table 1), and which enable the analysis of the components within the desired detection limits. The limits of quantification determined in this validation experiment for the tested matrices are much lower than the maximum permissible levels (MRLs) specified in the Regulations for individual analytes, therefore the method can be used for routine tests.
In conclusion, the possibility of introducing new methods or modifying the existing ones, on a basis of the use of various solvents, salts, buffers and sorbents, enables the implementation of a wide spectrum of analyzed analytes and matrices in the analysis of pesticides and PAHs.

Real Samples
In order to demonstrate the suitability of the optimized method for routine quantifications, real samples were screened for pesticide and PAHs residues. The study included 28 samples of plant material: apple (n = 8), wheat (n = 7), lettuce (n = 6), and pepper (n = 7) ( Table 5), which were purchased at the local commercial shops, as well as 10 samples of soil. Soil samples came from the Podkarpacie region (Poland) and were collected from a city center, a housing estate, recreation areas, banks of a river (San), farmland, railway tracks, sewage treatment plants, power plants, and steelworks, as detailed in Table 6.
The results of the analyses concerning pesticide residues were interpreted by comparing the MRL values in force in Poland [4] and by verifying the correct use of chemical preparations on the basis of the current "Register of plant protection products authorized for marketing and use" [14] and "Labels-instructions for the use of plant protection products" approved for marketing and use with the permit of the Ministry of Agriculture [53]. 17.5% (five samples) were free of pesticide residues, while the majority of samples, in which residues were found, contained more than one pesticide (two active substances detected in seven samples (25%), and three active substances found in two samples (7%). Most frequently, the determined pesticide residues belonged to fungicides, with boscalid being the most frequently detected compound. Four apple samples exceeded the established MRLs for boscalid. However, there was no violation of the law concerning the use of plant protection products not recommended for a given crop in any of the samples. The test results are shown in Table 5, where the MRL values for individual substances are also given.    Table 6, which concerns the analysis of soil samples for PAHs, shows that the least contaminated material were samples two, three and seven, which came from a housing estate, an arable field and the San River bank, respectively. In samples two and three, all determined compounds were < LOQ, and in sample seven, only two compounds above the LOQ were detected, i.e., phenanthrene and pyrene, at a level of 0.001 mg/kg. On the other hand, the most polluted samples were from the steelworks (No. five), the sewage treatment plant (No. six) and the city center (No. eight). The highest level of pyrene pollution, 0.092 mg/kg, was recorded in the smelter, and of indeo (one, two, three) pyrene in the city center, also at a level of 0.092 mg/kg. Furthermore, the results presented in the table show that the most frequently detected compounds from the PAH group were phenanthrene and pyrene, and acenaphthylene was not detected above the LOQ in any collected material. Examples of chromatograms of real samples are presented in the Supplementary Material ( Figures S3, S6 and S9).
The obtained results of actual soil samples were compared to the permissible PAH content in the soil according to the Minister of the Environment [54,55], depending on a relevant soil type applicable to a given soil. The data are presented in Table 6.
From a review of scientific publications it can be concluded that the presence of PAHs or pesticides is common, and in many cases the concentration levels of these substances exceed the permissible MRLs.
Kubecki, M., (2010) [56], Majkowska, E. et al., (2016) [57], and Wang, D., et al. (2018) [28] showed a relationship between the location of the sampling site and the occurrence of cases of exceeded levels for these substances. It is worth remembering that the problem of hydrocarbon pollution also affects the food we eat. The surprising results on PAHs concentrations in food were published by EFSA under the Commission Order 2005/108/EC. The obtained data showed that 47.3% of 1000 tested samples were above the detection limit for the presence of benzo(a)pyrene [58,59].
In case of pesticide residues, there are also reports on the monitoring of these substances. In our previous work [60] we showed the presence of pesticides in agricultural products from south-eastern Poland. The residues of these xenobiotics were detected in 84 (25.6%) of the analyzed samples, and in seven samples (2.1%) they exceeded the MRL level. Most often, pesticide residues were detected in samples of fruit, herbs, and vegetables.
The problem of pesticide residues in food and soil is common, and this is also confirmed by other publications: 35 active substances were found in vegetables and fruit (lettuces, oranges, peppers, tomatoes, and carrots) [23], 93 pesticides in flour wheat, lettuce and apples, 13 pesticides in soil and water [61], 105 pesticides in cereals [62], and 35 pesticides of various classes in tropical fruit [30]. Additionally, the European Food Safety Authority publishes numerous scientific papers or reports [63,64] on the presence of pesticide residues in food, which further highlights this problem. On a basis of the above information, it can definitely be concluded that it is necessary to continuously monitor pesticide residues and PAHs, as well as to implement appropriate measures to prevent overexposure to these substances.

Chemicals and Standards
Acetone and n-hexane of high purity were purchased from Honeywell Specialty Chemicals Seelze GmbH (Germany). Kits of salt and sorbents (SampliQ) for extraction and purification (Table 4) were obtained from Agilent Technologies (Santa Clara, CA, USA). Florisil was purchased from Sigma-Aldrich Chemie GmbH (Saint Louis, MO, USA). Water for sample preparations was purified through SolPure XIO P (ELKAR, Kęty, Poland).
The individual pesticide analytical standard (94 active substances) were obtained from Dr. Ehrenstorfer (Augsburg, Germany), Supelco ® (Bellefonte, PA, USA) and the Institute of Organic Industry (IPO) (Warsaw, Poland). A certified mixture of 13 PAHs standard solutions in acetone (EPA 525 PAHs Mix B) was purchased from Supelco ® (Bellefonte, PA, USA). Triphenyl phosphate (TPP) was used as the internal standard (IS), and it was obtained from Dr. Ehrenstorfer (Augsburg, Germany).
All analytical standard compounds were of >98% purity, and they were prepared in concentrations of about 1000 µg/mL in acetone, then the intermediate stock standard mixture at approximately 10 µg/mL (of each compound) was prepared from the stock solutions by dilution in acetone. Subsequent working standards mixtures of 0.001-1.00 µg/mL (for PAHs) and 0.005-1.44 µg/mL (for pesticide) were prepared in an acetone: n-hexane (1:4 v/v). The IS was prepared by dissolving the TPP in acetone: n-hexane (1:4 v/v) to obtain a 1000 µg/mL solution. Calibration of matrix-matched standards was performed by mixing working standards solutions with blank sample extracts containing one g of sample per one mL of solvent. The pesticide and PAHs standards at appropriate concentrations were used to calibrate the GC-MS/MS system and spike samples in validation experiments. The mixture of analytical standards, working solutions and IS were stored in glass bottles closed with parafilm at 4 • C in dark pending the analysis.

Sample Preparation Procedure
The method for preparing samples for the analysis of pesticide residues and/or PAHs by gas chromatography with mass spectrometry in plant material and soil was a modification of the standardized method provided in PN-EN 15662: 2018-06. A Vegetable Food Multimeter was used to determine pesticide residues using GC and LC based analysis after acetonitrile extraction/partitioning and dispersive solid phase extraction (DSPE), as described in QuEChERS modular method [33]. The proposed solution is to change the solvent used during the extraction from acetonitrile to the acetone: n-hexane mixture at a volume ratio of 1:4 (v/v). An additional change in the procedure was the use of a reduced amount of water during the extraction of compounds from cereals, and the use of florisil instead of GCB at the sample cleanup step, as it prevents the loss of planar compounds such as PAH. The quoted methodological standard does not describe the procedure for the extraction and cleanup of soil samples for the determination of PAHs and pesticides. In this work, as a novelty, a solution dedicated to this type of matrix was proposed.

The Sample Preparation Scheme Is as Follows
Samples of plant material or soil (of five g or ten g depending on the matrix; Table 2) with water added, if necessary (Table 2), were extracted with 10 mL mixture of acetone: n-hexane (1:4 v/v) for approx. 1 min (vortex shaking; Vortex BenchMixerTM, Benchmark, Edison, NY, USA). Next, buffer salts were added containing: four g of anhydrous magnesium sulfate (MgSO 4 ), one g of sodium chloride (NaCl), one g of anhydrous trisodium citrate, and 0.5 g of disodium sesquihydrate citrate, and then samples were shaken again vigorously for one min. In the next step, the samples were centrifuged at the centrifuge speed > 4000 rpm (5804R, Eppendorf, Hamburg, Germany) for five min. The next step was the cleanup of the samples using the dispersive solid phase extraction (DSPE). For this purpose five mL of sample extract (upper organic layer) were taken and extracted with the mixture of salts I, II, III or IV (depending on the matrix; Table 2). The samples were shaken vigorously for approx. 30 s (for mixtures of salts I and II) or approx. two min (for mixtures of salts III and IV), and then the samples were centrifuged at the centrifuge speed > 4000 rpm for five min.
The analysis of each sample was performed in five replications, with IS at a concentration of 1000 µg/mL in the amount of 50 µL added to each of them immediately before the chromatographic analysis. The sample extracts obtained this way were ready for analysis by GC-ECD/NPD, GC-MS or GC-MS/MS.

Chromatographic Analysis
A 7890A gas chromatograph (Agilent Technologies, USA) gas chromatography coupled with triple quadrupole tandem mass spectrometry, model 7000 (GC-MS/MS), was used to analyze sample extracts. Chromatographic separations were conducted using the HP-5 MS Ultra Inert column (30 m × 0.25 mm I.D.× 0.25-µm; Agilent Technologies, USA). Analyses were conducted in the selected ion monitoring mode (SIM) based on the use of one of the quantitative ions for determination of PAHs and the multiple reaction monitoring (MRM) mode with three mass transitions for each pesticide. Analyzed compounds were identified according to their qualitative ions and retention times. The following analysis parameters were used: samples injected in a splitless mode, injected volume-2 µL, carrier gas-helium (5.0 purity, flow 2.1 mL/min), the MS ionization was carried out in the electron ionization mode at 70 eV. For pesticide residues, the temperature was 280 • C for the transfer line, 300 • C for the ion source, 150 • C for the quadrupoles, and 70-280 • C for the oven. For PAHs analysis, the temperature was 320 • C for the transfer line, 320 • C for the ion source, 150 • C for the quadrupoles, and 80-320 • C for the oven. Software Mass Hunter, version B.07.06, was used for data acquisition, control and data processing of the analysis results. MRM transition, SIM and retention times of tested substances can be found in the Supplementary Material (Tables S1 and S2) attached to this article. The total run time was 42 and 26 min for pesticide and PAHs, respectively.

Validation Process
In order to determine the usefulness of the developed method, a validation was carried out with the parameters of linearity, recovery, precision, and limits of detection and quantification assessed, and measurement of uncertainty. The validation procedure was performed in accordance with the European Commission guidelines specified in Guidance SANTE/12682/2019-Guidance document on analytical quality control and method validation procedures for pesticides residues analysis in food and feed [34] and Commission Regulation (EU) (2011) No 836/2011 of 19 August 2011 amending Regulation (EC) No 333/2007 laying down the methods of sampling and analysis for the official control of the levels of lead, cadmium, mercury, inorganic tin, 3-MCPD and benzo(a)pyrene in food stuffs [35].
Validation studies were performed for four representative plant matrices: apple, wheat, pepper, lettuce, and for soil ( Table 2).
Linearity was studied by GC-MS/MS analysis at five different concentration levels over the ranges of 0.005-1.440 µg/mL and 0.001-1.000 µg/mL for pesticide residues and PAHs, respectively. Calibration standards were prepared in sample extracts, and they were applied in three repetitions per level over three days.
The limit of quantification (LOQ) was set at the lowest spiking concentration that has been validated with satisfactory recovery and precision parameters. Limits of detection (LOD) were calculated using a signal-to-noise ratio criteria, equal to three for all tested substances [34].
In the recovery experiments, samples were spiked with the appropriate volumes of working standard solutions of pesticides and PAHs, at two levels: LOQ and 1000 × LOQ. The samples were prepared using the method described above, and next they were analyzed for each spiking level in five replicates (n = 5). Precision was calculated from the recovery experiments, and it was expressed in terms of relative standard deviation (RSD%) at each spiking level. Recovery per level and overall recovery were determined for every tested substance. For quantitative methods, the established mean recovery for spiked samples should be between 70-120%, with a relative standard deviation (RSD) below or equal to 20% for pesticide residues and 50-120% for PAHs [35].
Uncertainty of measurement (U) was estimated on the basis of the results obtained in the validation process. The major uncertainty sources were the repeatability of recoveries from spiked samples and uncertainty of the average recovery calculated from the rectangular distribution. The relative expanded uncertainty was calculated by using the coverage factor k = 2 at the confidence level of 95% [65].

Conclusions
In this article, the sample preparation procedure (based on the QuEChERS method [33] has been optimized to separate pesticide residues and PAHs from five representative matrices and analysis by GC-MS/MS system. The novelty in this research was to change the solvent used during the extraction from acetonitrile to the acetone: n-hexane mixture at a volume ratio of 1:4 (v/v). An additional change was a reduction in the amount of water used during the extraction of compounds from cereals, and the use of florisil instead of GCB at the sample cleanup step. The method was validated for a total of 94 substances from the pesticide and 13 compounds from the PAH group. Overall linearity, precision, and accuracy parameters were highly satisfactory, with the exception of six pesticides. The extended uncertainty of the method was also acceptable. The proposed analytical procedure is efficient, accurate and repeatable, and therefore, is suitable for simultaneous determination of multiclass pesticide residues and PAHs in plant material and soil samples. It is also safer, allowing for a reduction in the consumption of organic solvents (by 17%) and time of sample preparation (by 20%). The use of florisil for sample cleanup, improves recovery of compounds with flat particles. The developed method was successfully used to test a total of 38 real soil and plant material samples. The results of the analysis of samples revealed cases of the MRLs violation for pesticides (the case of boscalid), however, no substances not approved for use in the tested crops were found. As for PAHs, no MRL exceedance was detected. Nevertheless, the obtained results indicate the need for the further monitoring of these compounds.