Supporting dataset on the optimization and validation of a QuEChERS-based method for the determination of 218 pesticide residues in clay loam soil

The dataset presented in this article supports “Optimization and validation of a method for the simultaneous environmental monitoring of 218 pesticide residues in clay loam soil” [1]. A method based on QuEChERS (Quick, Easy, Cheap, Effective, Rugged & Safe) for the extraction of pesticide and some metabolites residues was developed. The quantification of the chemicals was performed by a combination of two complementary LC-MS/MS and GC–MS/MS analyses. Detailed optimization data of the QuEChERS extraction method is provided, including (1) salt combination, (2) acidification of the solvent (3) the amount of the selected acid (Formic Acid, FA) and (4) moisturization of the soil samples prior to extraction. In addition, all the validation data are presented, including the matrix effect, which was evaluated for each analyte using the recommended procedure.


Value of the Data
• The optimization data might be useful to other researchers developing QuEChERS-based extraction methods in soil matrix. • The validated data provided are equally useful for researchers developing methods in other matrices of similar complexity. • The details of the matrix effect of each analyte demonstrates the need of using matrixmatched calibration curves in order to counteract ion suppression or enhancement in chromatography-triple quad mass spectrometry tandems, especially in GC-MS/MS.

Data Description
The data presented here were obtained during the development and validation of a QuEChERS-based extraction method for the detection and quantification in GC-MS/MS and LC-MS/MS of 218 pesticides in soil matrices and supports the main article in Science of the Total Environment entitled "Optimization and validation of a method for the simultaneous environmental monitoring of 218 pesticide residues in clay loam soil" [1] . Table 1 is a list of the analytes presented in alphabetical order together with an identification number from 1 to 218 and the technique in which they are analysed. Thus, compounds are identified numerically with their correspondent label in the following charts. Fig. 1 represents the percentage of compounds against the recovery (%) for AOAC and EN QuEChERS methods, with and without a clean-up step. For recoveries between 70% and 120%, analytes are considered to be successfully extracted under the SANTE 12,682/2019 and the SANCO 825/00 Rev.1 guidance document on residue analytical methods guidelines [2 , 3] , which were followed for the optimization and validation processes. Recoveries in the ranges of 60-70% and 120-130% were considered as well, since further improvement can be achieved during the whole optimization procedures. According the mentioned guides, poor recoveries were considered below 60% and over 130%.
In Fig. 2 we present graphically the results of the comparative study of the recovery percentages obtained for the 218 analytes when they are extracted in the presence of acid (either 1% formic acid or 1% acetic acid) or in the absence of it. Fig. 3 shows the recovery of each compound when 0.5%, 1% and 2.5% of formic acid in the extraction solvent was tested. As stated above, analytes with recoveries between 70% and 120% (relative standard deviation (RSD) < = 20, n = 3) were considered successfully extracted and that area is marked in the graphic. Ranges of 60-70% and 120-130% were also marked. Fig. 4 shows the effect of the percentage of water added to the soil sample in the extraction recovery. Recoveries obtained for dry soil samples (0%) were compared to those obtained for 10%, 20%, 30%, 40% and 50% of moisture. The range from 70 to 120% of recovery is highlighted in the chart along with those of 60-70% and 120-130%. Fig. 5 is the representation of the matrix effect shown by each of the 218 analytes. It shows mean and SD values of ME in percentage for each pesticide and metabolite, which were calculated as follows: ME (%) = ( S m -S b / S s ) x 100, where S m is the signal obtained for each analyte in the soil extract, S b is the response of the non-spiked soil extract and S s is the signal of the standard in the solvent. The effect of the matrix components in the signal was rated as enhancement or suppression whether values of ME were above or below 100%, respectively. No significant matrix effect was considered if values were between 80% and 120. This range is marked in the graphic with a dotted line and the area that it covers had been shaded in grey.
The entire dataset of all these experiments are presented in the files named Fig. 1 -5 raw data included as Supplementary Material of this article.
From individual stock standard solutions (10 0 0 μg mL −1 in ACN) or commercial mixtures (10 μg mL −1 in ACN), an intermediate working mixed solution of 0.833 μg mL −1 was prepared. The P-IS mix working solution was prepared at 1 μg mL −1 in ACN. The working solutions were stored at −20 °C for a maximum period of 5 weeks, and employed to spike soil samples and to prepare calibration curves, either in matrix or solvent.
The QuEChERS salts were acquired in commercial premixes as it also was the Enhanced Matrix Removal-Lipid (EMR-lipid) (Agilent Technologies (Palo Alto, USA). All the solvents employed were of HPLC-MS/MS grade (Honeywell, Charlotte, USA). Ammonium acetate, formic acid and acetic acid were of the maximum purity available and acquired from Fisher Scientific (Loughborough, UK). Ultrapure water was prepared in the laboratory using a Gradient A10 Milli-Q System (Millipore, Bedfore, MA, USA).  2. Acid addition to solvent extraction. The figure shows the number of compounds that, from left to right, had a recovery below 60%, in the range of 60% to 70%, between 70% and 120%, from 120% to 130%, and superior to 130% when the extraction solvent was acetonitrile (orange-coloured bars), acetonitrile-1% acetic acid (brown-coloured bars) and acetonitrile-1% formic acid (dark blue-coloured bars).(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Sample preparation
The extraction was based on the QuEChERS-based using 10 g of dried and sieved soil (with of without of increasing percentages of water), 10 mL of the extractant (ACN, ACN-1% acetic acid, ACN-0.5% formic acid, ACN-1% formic acid, or ACN-2.5% formic acid), and 7.5 g of the QuEChERS extraction salts mixture (either AOAC or EN formulas). The tubes were energetically shaken for 1 min, sonicated in an ultrasonic bath for 15 min, gently shaked for 25 min (rotary shaker), and centrifuged 10 min at 4200 rpm. Finally, supernatant was either filtered (0.2 μm) and directly analysed in GC-MS/MS or dissolved in milliQ water grade (1:1, v/v) and analysed in LC-MS/MS.
When it was necessary to spike the soil samples (either the 218 analytes plus P-IS, or P-IS alone) the samples were left to stand at least 1 h prior to extraction. All optimization experiments were made at a single concentration, 20 ng g −1 (in triplicate). Soil matrix for calibrators and matrix effect samples were extracted without any fortification.

QuEChERS salts selection
For these experiments we employed modifications of the QuEChERS (quick, easy, cheap, effective, rugged and safe) method, initially designed for the extraction of pesticides in fruits and vegetables [4] . The two main official variants of the original QuEChERS, the AOAC [5] and the EN variants [6] , were compared. During this step, ACN was used as the extraction solvent. QuECh-ERS extraction salts used for each method consisted on 1.5 g of NaOAc and 6 g MgSO 4 for AOAC version and 4 g MgSO 4 , 1 g NaCl, 1 g Sodium Citrate dihydrate and 0.5 g Sodium hydrogencitrate  Table 1 . Bold-dotted lines shows the recovery limits recommended by the SANTE guide as acceptable (70% and 120%). Since the same guide also permits an expanded 60-130% when the results are highly reproducible these limits are also marked with dotted lines. For clarity the graph has been divided in 8 panels.    Fig. 4 we show the effect of soil moistening on the recovery percentages of the 218 analytes. Various percentages of soil moisture were tested (0%, 10%, 20%, 30%, 40%, and 50% water). The compounds are numbered according to the relation in Table 1 . Bold-dotted lines shows the recovery limits recommended by the SANTE guide as acceptable (70% and 120%). Since the same guide also permits an expanded 60-130% when the results are highly reproducible these limits are also marked with dotted lines. For clarity the graph has been divided in 8 panels. sesquihydrate for EN variant. Both methods were tested followed or not by an additional cleanup step using the Enhanced Matrix Removal sorbent (EMR, Agilent Technologies) [7] . Five mL of the supernatant were treated with 1 g of EMR, which had been previously activated with 5 mL of water. Then it was centrifugated and 3.5 g MgSO 4 were added to 8 mL of supernatant to remove the remaining water. Finally, all extracts produced with AOAC, EN QuEChERS versions with and without clean-up were analysed by LC-MS/MS and GC-MS/MS [1] .

Solvent acidification
Following the optimization of the salts, we tested the influence of the acidification of the acetonitrile in the extraction efficiency. First, it was necessary to decide the acidificant, and formic acid and acetic acid, both at 1% in ACN were assayed and compared with the non-acidified ACN   Table 1 . Dotted lines represent the tolerance range in which it is considered that no significant matrix effect exists. For clarity the graph has been divided in 4 panels.

Water addition to the soil sample
The effect of the moisture of the sample was studied by adding different volumes of water to 10 g of soil sample prior to the extraction in order to achieve 10%, 20%, 30%, 40% and 50% (v/p). For this purpose, 1, 2, 3, 4, and 5 mL of ultrapure water were added to each sample 1 h before the extraction, once each sample was left to stand for another 1 h after the fortification with the pesticide and/or P-IS mixes.

Matrix effect
For matrix effect experiments, 5 level calibration curves (0, 6.25, 12.5, 25 and 50 ng g −1 ) were prepared in solvent and matrix in triplicate. Soil matrix was extracted using the optimized, recommended procedure (AOAC salt combination, ACN-2.5% FA and air-dried soil samples). According to the technique they were going to be analysed by, curves in solvent were prepared in either ACN-2.5% FA or ACN-2.5% FA-H 2 O, 1:1 (v/v) and either in matrix or matrix-H 2 O, 1:1 (v/v) for GC-MS/MS and LC-MS/MS, respectively.

Instrumental analyses
For the determination and quantification of the total amount of compounds, samples were analysed by liquid chromatography and gas chromatography, both coupled to triple quadrupole mass spectrometry. LC-MS/MS analyses were performed using a 1290 Infinity II LC System and a Triple Quad 6460 mass spectrometer (Agilent Technologies, Palo Alto, CA, USA). The Agilent Technologies, Poroshell 120 EC -C18 column (2.1 × 100 mm, 2.7 μm) equipped with a guard precolumn and pre-filter (2.1 × 5 mm, 1.8 μm) was used for the chromatographic separations. GC-MS/MS analysis were achieved with a GC System 7890B equipped with a 7693 Autosampler and Triple Quad 7010 mass spectrometer (Agilent Technologies). The chromatographic separation in GC was performed using two fused silica ultra-inert capillary columns Agilent HP-5MS (15 m x 0.25 mm i.d., 0.25 μm film thickness), connected by a purged union to allow the backflushing. A detailed description of the operation conditions, spectrometric parameters and the optimization procedure of both methods can be found in the main article [1] .

Method validation parameters
The validation of the developed method was performed following the recommendations of the European Union SANTE 12,682/2019 and the SANCO 825/00 Rev.1 guidance document on residue analytical methods (EC, 2010; EC, 2019b), which were followed in the absence of specific guidelines for the analysis of pesticide residues of pesticides in soil.
The linearity in the response was studied by injecting standards into the soil extract or in the soil extract diluted with water (1:1, v/v) in GC-MS/MS and LC-MS/MS, respectively, at nine concentration levels: 0.3, 0.5, 1.0, 2.5, 5, 10, 20, 50, and 100 ng mL-1. Accuracy (% recoveries) and precision (% relative standard deviation) were estimated by recovery experiments in spiked soil samples (in quintuplicate) at 7 concentration levels: 0.5, 1.0, 2.5, 5, 10, 20 and 50 ng g −1 . Values were considered acceptable when recoveries were between 70% and 120% and RSDs ≤20%. The limit of quantification (LOQ) was set as the lowest concentration level that has acceptable accuracy and precision and the limit of detection (LOD) was selected as the lowest point of the calibration curve that meets all the agreements, had a signal-to-noise ratio (S/N) > 3 and an accuracy between 80% and 120%.
The confirmation of the identity of the analytes in the samples was performed with the acquisition of two MS/MS transitions, the quantification (Q) transition and the confirmation (q) transition, with a maximum ion ratio tolerance of ±30% and agreement of the retention time with a maximum deviation of ±0.1 min between the analyte in the sample and the reference standard. It should be noticed that for analytes with chiral isomers as cypermethrin, the sum of their isomers is provided as so is the residue definition. Nevertheless, when a single enantiomer is included in the residue definition, such as lambda-cyhalothrin, they were determined and quantified separately.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships which have, or could be perceived to have, influenced the work reported in this article.