Development of a QuEChERS method for simultaneous analysis of 3-Monochloropropane-1,2-diol monoesters and Glycidyl esters in edible oils and margarine by LC-APCI-MS/MS

Analytica Chimica Acta 1239 (2023) 340712 Available online 7 December 2022 0003-2670/© 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/). Development of a QuEChERS method for simultaneous analysis of 3-Monochloropropane-1,2-diol monoesters and Glycidyl esters in edible oils and margarine by LC-APCI-MS/MS Jorge A. Custodio-Mendoza a , b , ** , Raquel Send´ on a , Ana Rodríguez-Bernaldo de Quir´ os a , Rosa A. Lorenzo a , b , Antonia M. Carro a , b , c , * a Department of Analytical Chemistry, Nutrition and Food Science, University of Santiago de Compostela, 15782, Santiago de Compostela, Spain b Health Research Institute of Santiago de Compostela (IDIS), University of Santiago de Compostela, 15782, Santiago de Compostela, Spain c Instituto de Materiais (iMATUS), University of Santiago de Compostela, 15782, Santiago de Compostela, Spain HIGHLIGHTS GRAPHICAL ABSTRACT •Direct determination using atmosphericpressure chemical ionization (APCI) source. •Comparison of APCI and HESI-II sources in the analysis of 3-MCPDE and GE. •Cost-effective approach with simpler mobile phases and no matrix effect. •Novel QuEChERS approach to extract of 3-MCPDE and GE from margarines. •The presence of glycidyl oleate was confirmed in all analyzed samples. ARTICLE INFO Keywords: APCI Food contaminants Glycidyl HPLC method 3-MCPD QuEChERS ABSTRACT A simple, fast and effective direct method based on HPLC-APCI-QqQ-MS/MS has been developed to simultaneously determine four 3-monochloropropane-1,2-diol monoesters (3-MCPDE) esterified with palmitic, linoleic, stearic, and oleic acid, and two glycidyl esters (GE) with palmitic and oleic acid in margarine and olive oil using a QuEChERS approach. Factors affecting the efficiency of the extraction process were assessed, including type and amount of salt, extraction solvent, test portion amount, and clean-up sorbent. The analytical method was validated according to Food and Drug Administration (FDA) guidelines using matrix-matched calibration with internal standards and showed good results in terms of linearity (r 2 >0.9992), accuracy (80<Recovery<120%), and precision (RSD<15%). The method was successfully applied for the first time to 11 margarine samples for simultaneous analysis of 3-MCPDE and GE. 1. Introduction Vegetable oils are one of the most-used foods in the food industry, so * Corresponding author. Department of Analytical Chemistry, Faculty of Chemistry, University of Santiago de Compostela, 15782, Santiago de Compostela, Spain. ** Corresponding author. Department of Analytical Chemistry, Faculty of Chemistry, University of Santiago de Compostela, 15782, Santiago de Compostela, Spain. E-mail addresses: [email protected] (J.A. Custodio-Mendoza), [email protected] (A.M. Carro). Contents lists available at ScienceDirect Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca https://doi.org/10.1016/j.aca.2022.340712 Received 7 November 2022; Received in revised form 5 December 2022; Accepted 6 December 2022 Analytica Chimica Acta 1239 (2023) 340712 2 it is essential to ensure their quality and safety for human consumption. During the oil refining process, mainly in the deodorization step carried out at low pressures and high temperatures (180–220 ◦C), food processing contaminants, including fatty acid esters of 3-monochloropropane-1,2-diol (3-MCPDE) and glycidyl ester (GE), can be formed [1,2]. Additionally, GE can be formed from monoacylglycerols or diacylglycerols by dehydration at high temperatures or from 3-MCPDE by elimination of HCl in refined edible oils [2]. Lipases can hydrolyze 3-MCPDE in the gastrointestinal system to release free 3-MCPD, which can be absorbed by the body and cause health risks. Toxicological studies have shown the ability of 3-MCPD to induce the formation of tumors in various organs, especially the kidney, and produce infertility [3]. The International Agency for Research on Cancer (IARC) classifies 3MCPD as “Possibly carcinogenic to humans’’ (group 2B) and GE as “Probably carcinogenic to humans” (group 2A) [4]. The concentration of these compounds varies according to the type of oil, but they are especially abundant in palm oils. 3-MCPDE and GE levels are important indexes to evaluate food safety [1,2]. These potentially toxic compounds decrease the nutritional value of foods, harm consumers’ health, and represent a constant concern of organizations such as the USA Food and Drug Administration (FDA) and the European Food Safety Authority (EFSA) [3]. Establishing a maximum level for 3-MCPD esters and GE in foods such as edible oils is appropriate to ensure a high level of human health protection. Institutional concern of these compounds began in the early 2000’s, but solid guidance has not been fully established. The Scientific Panel on Contaminants in the Food Chain (CONTAM Panel) established a tolerable daily intake (TDI) of 2 μ g/kg body weight per day for 3-MCPD and 3-MCPDE. However, even though GE are genotoxic carcinogens, no TDI has been established for these compounds [3,5]. The latest Commission Regulation (EU) No. 2020/1322 sets a maximum level of 20 μ g/kg for 3-MCPD in soy sauce and hydrolyzed vegetable protein and 1250 μ g/kg and 2500 μ g/kg for the sum of 3-MCPD and its esters in vegetable oils and fats, expressed as 3-MCPD. It also sets a maximum level of 1000 μ g/kg for GE as glycidol in vegetable fats and oils [5]. To quantify these contaminants, direct and indirect analytical methods were reported based on liquid and gas chromatographic and mass spectrometry techniques. Indirect methods require the hydrolysis of the 3-MPCD esters and GE to their corresponding free forms, followed by their derivatization and subsequent GC-MS analysis [6–8]. Direct methods are based on determining individual esters by LC-MS, with neither modifications nor derivatization needed [9–16]. However, the nature and characteristics of the target compounds interfere with their identification and determination, as in the case of GE and 3-MCPDE monoesters of the same fatty acid. An electrospray ionization (ESI) source is often used to simultaneously determine 3-MCPDE and GE with fatty acids in edible oils and oil-based food [11–16,19–25]. However, other sources have been proposed, including heated electrospray ionization (HESI) in the 3-MCPDE analysis [9,10] and atmospheric pressure chemical ionization (APCI) in the GE analysis [16,26,27]. Thermally-focused HESI has been described as a modification of a traditional ESI source that uses electrospray at high temperatures during the nebulization [28]. HESI source applies additional sheath gas to assist the transformation of the ions in the solution into the gas phase [28,29]. Lately, HESI sources, in their different available designs, have gained popularity over conventional ESI sources [30,31]. Recent studies suggest a significant improvement in ionization efficiency when using a HESI source, while other studies report a higher propensity for matrix effect [28–31]. On the other hand, many authors have observed similar performances between APCI and ESI sources regarding different analytical features [28]. Additionally, a significant reduction in matrix effects has been demonstrated when using APCI sources [31]. Concerning the mobile phase, the general compositions do not vary greatly when using ESI, HESI, or APCI sources, except in their acidity, which is necessary to stimulate ionization in the ESI and HESI sources [32]. At the same time, its absence or presence is not necessarily influential in APCI ionization [31,32]. For this work, Thermo Fisher Scientific’s HESI-II and APCI sources were assessed for their performance in determining fatty acid monoesters of 3-MCPD and GE. Based on most abundant fatty acids in oil-based foods (including margarine), monoesters of 3-MCPD and GE with oleic (OL and OLGE), palmitic (PA and PAGE), linoleic (LI), and stearic (ST) acids were selected for this study. The structural similarity of the products causes the MS detector to report similar intensities for both precursor ions and transitions. For this, a chromatographic column capable of separating these analytes and providing complementary selectivity for their determination is needed. The sample preparation step of complex matrices like oils, fats, and oil-based foodstuff is often complicated, requiring large volumes of organic solvents. Previous studies have reported a double SPE procedure using an SPE C18 cartridge first and then an SPE silica cartridge with solvents of different polarities [11–16]. Several approaches were recently proposed and tested to improve sample preparation using other lipid-removal sorbents. Graziani et al. (2017) proposed the extraction of esters using acetonitrile-2-propanol (1:1 v/v), followed by clean-up with a mixture of PSA and C18 powder using vortex agitation. The supernatant was evaporated to dryness under a nitrogen stream, and the dry extract was reconstituted with acetonitrile-2-propanol (1:1 v/v) [9]. In a Abbreviations 3-MCPD 3-monochloropropane-1,2-diol ACN acetonitrile APCI Atmospheric-pressure chemical ionization CONTAM Contaminants in the Food Chain dSPE dispersive solid phase extraction EFSA European Food Safety Authority ESI electrospray ionization EVO extra virgin olive oil FDA USA Food and Drug Administration GE glycidyl ester IARC International Agency for Research on Cancer LC-MS/MS liquid chromatography tandem mass spectrometry Li 1-linoleoyl-3-chloropropanediol LLOQ lower limit of quantification MTBE methyl t-butyl ether OL 1-oleoyl-3-chloropropanediol OL-d5 1-oleoyl-3-chloropropanediol-d5 OLGE glycidyl oleate OLGE-d5 glycidyl oleate-d5 PA 1-palmitoyl-3-chloropropanediol PA-d5 1-Palmitoyl-3-chloropropanediol-d5 PAGE glycidyl palmitate PAGE-d31 glycidyl palmitate-d31 PLLE partitioned liquid-liquid extraction PSA primary secondary amine QCs quality control substances QuEChERS Quick, Easy, Cheap, Effective, Rugged and Safe ST 1-stearoyl-3-chloropropanediol ST-d5 1-stearoyl-3-chloropropanediol-d5 TDI tolerable daily intake ULOQ upper limit of quantification J.A. Custodio-Mendoza et al. Analytica Chimica Acta 1239 (2023) 340712 3 previous study applied to 3-MCPDE diesters, we developed an approach that includes partitioned liquid-liquid extraction (PLLE) with 20% EtOAc/MTBE followed by a solvent exchange after evaporation with nitrogen, using 2% Et 2 O/hexane. Then, two stages of dispersive solid phase extraction (dSPE) were used for clean-up, the first using Si-SAX and PSA and the second using Z-Sep+and PSA. The final extract was evaporated to dryness and rebuilt in 75% MPA/25%MPB (the mobile phase solution) [10]. In response to the need to simplify sample preparation, a direct method for the simultaneous determination of 3-MPCD monoesters and GE in oils was developed and validated. The analysis was carried out by LC-APCI-MS/MS using a 5 μ m Luna OMEGA PS C18 chromatographic column taking only 4 min. This column improves labor efficiency by having comprehensive polar and non-polar retention. For sample preparation, a simplified but effective procedure of QuEChERS is proposed that uses a 15 mg sample, 3 mL of acetonitrile, 3 mL of water, and 150 mg of EMR-lipid sorbent. This novel method was used to assess 11 margarine samples in triplicate for the presence of PA, OL, LI, ST, PAGE, and OLGE. 2. Materials and methods 2.1. Chemicals and materials Acetonitrile (ACN), 2-propanol (IPA), chloroform (CHCl 3 ), methanol (MeOH), acetone (ACO), n-hexane, ethyl acetate (EtOAc) and diethyl ether (Et 2 O) were all LC-MS grade with a purity of ≥99% and purchased from Merck (Darmstadt, Germany). Ammonium sulfate (purity ≥99%), ammonium formate, and ammonium acetate (purity ≥99.99%) from Merck. Ultrapure water was obtained using a Millipore purification system (Millipore, Billerica, MA, USA). SPE sorbents: Bond Elut EMRLipid (EMR-Lipid), C18, primary secondary amine (PSA), and Silica Strong Anion Exchange (Si-SAX) were supplied from Agilent® (Santa Clara, CA, USA), OASIS HLB was purchased from Waters® (Milford, MA, USA) and Supel™ QuE Z-Sep +bulk (Z-Sep+) from SUPELCO (Bellefonte, USA). The standards used were 1-oleoyl-3-chloropropanediol (OL, CAS No. 10311-82-7), 1-palmitoyl-3-chloropropanediol (PA, CAS No. 30557-04-1), 1-stearoyl-3-chloropropanediol (ST, CAS No. 22094-20-8), 1-linoleoyl-3-chloropropanediol (LI, CAS No. 1246833-46-4), glycidyl palmitate (PAGE, CAS No. 7501-44-2), glycidyl oleate (OLGE, CAS No. 5431-33-4), 1-oleoyl-3-chloropropanediol-d5 (OL-d5, CAS No. 1639207-37-6), 1-palmitoyl-3-chloropropanediol-d5 (PA-d5, CAS No. 63326-63-6), 1-stearoyl-3-chloropropanediol-d5 (ST-d5, CAS No. 1795785-84-0), glycidyl palmitate-d31 (PAGE-d31, CAS No. 124681924-8), glycidyl oleate-d5 (OLGE-d5, CAS No. 1426395-63-2), all purchased from Toronto Research Chemical (Toronto, Ontario, Canada). Individual and mixed working standard solutions were prepared in methanol and stored at −18 ◦C. Calibration solutions were prepared at six concentration levels between 100 and 1000 ng/g from the mixed solution by appropriate dilution; the internal standard (IS) containing the deuterated species was added at 250 ng/g. Instruments and devices used were a vortex mixer from Instruments GmbH & Co (Schwabach, Germany), a Centromix II-BL centrifuge from J. P. Selecta (Barcelona, Spain), an AS 82/220.R2 Analytical Balance from RADWAG (Radom, Poland), a EC-1V-130 heater, a sample concentrator from VLM GmbH (Bielefeld, Germany), a HPLC-MS/MS Thermo Scientific™ equipped with a triple quadrupole mass spectrometer TSQ Quantum Access MAX from Thermo Fisher Scientific, (San Jose, CA, USA) provided with an atmospheric pressure chemical ionization APCI source, a heated electrospray ionization source HESI-II, an Accela 1250 pump fitted with a degasser and an autosampler. 2.2. Samples An extra virgin olive oil (EVO) sample was used to develop and validate the proposed method. A palm oil-free margarine sample was used for the matrix extension study. A total of 11 margarine samples composed of mixtures of various vegetable oils, including olive, sunflower, linseed, soybean, corn, rapeseed, and palm oil, in different proportions, were analyzed for 3-MPCD monoesters and GE. All samples were purchased from local supermarkets in Santiago de Compostela, Spain, and stored in their original packet at 4 ◦C until analysis. Once opened, samples were stored in the freezer at −20 ◦C. 2.3. Sample preparation In a conical-bottom glass tube, a 15 mg test portion was accurately weighed, followed by the addition of the internal standard solution, 3 mL of ultrapure water, 3 mL of acetonitrile, and 0.9 g of ammonium sulfate. The mixture was vortex-mixed for 2 min and centrifuged at 2601 rcf for 5 min to separate it into two phases. The upper phase was pipetted into another conical-bottom glass tube containing 150 mg of EMR-Lipid, vortex-mixed for 2 min for clean-up, and centrifuged for 5 min at 2601 rcf. After this, a 2 mL aliquot of the supernatant was evaporated under a nitrogen stream to dryness at 70 ◦C. The extract was reconstituted in 0.25 mL acetonitrile to be analyzed by LC-MS/MS. A scheme of the entire process is presented in Fig. 1. 2.4. Instrumental conditions Analysis was carried out using a Luna OMEGA PS C18 column (5 μ m, 100 Å, 50 ×2.1 mm) equipped with a Security Guard pre-column (4mm ×2 mm i.d.) from Phenomenex (Torrance, USA). The flow rate was set to 400 μ L/min, the column temperature was 30 ◦C, and the sample injection volume was 10 μ L. The mobile phase composition was ACN (A), and water (B), programmed to 85% A at 0 min, increased linearly to 90% An until 5 min, and returned to 85% A at 6 min and kept until 10 min for stabilization. However, the analytes were separated in less than 5 min. Additionally, the performance of the ACQUITY UPLC BEH Shield RP18 Column (1.7 μ m, 130 Å, 100 ×2.1 mm) from Waters (Milford, MA, USA) on the separation of 3-MCPDE and GE was investigated in this paper. Both Luna OMEGA PS C18 and ACQUITY UPLC BEH Shield RP18 columns were tested under the same conditions to compare their separation capabilities for target compounds. Initially, the determination of the compounds by tandem mass spectrometry (QqQ-MS/MS) was carried out with a HESI-II source in positive mode with the operational conditions described in the supplementary material (Table S1). Subsequently, APCI source, also operated in positive mode, were used with the ionization and transfer temperatures set at 350 and 340 ◦C. Nitrogen (>99.98%) was selected as sheath and auxiliary gas set at a pressure of 35 and 10 arbitrary TSQ Quantum units, respectively. Detection was performed in the Multiple Reaction Monitoring mode (MRM) using Argon as collision-induced-dissociation (CID) gas at 18 bar. The monitored range was 200–700 m/z. 2.5. Method validation The FDA guidelines for the Validation of Chemical Methods in Food, Feed, Cosmetics, and Veterinary Products [17] and Bioanalytical Method Validation Guidance for Industry [18] were used for the analytical validation of this method. EVO was used as a blank sample to assess the method’s selectivity, linearity, detection limits, precision, and accuracy. The specificity of the method was assessed using ten different EVO samples and based on the selection of a quantifier transition and two qualifier transitions for each analyte and IS. Sensitivity of the method involve setting a limit of detection (LOD), and the lower and upper limit of quantification (LLOQ, ULOQ). The lowest and highest amount of each analyte that can be quantitatively determined with acceptable precision and accuracy in this method falls between the respective LLOQ and ULOQ. LOD was calculated for analytes concentration at which signalto-noise ratio was 3, and 10 for LLOQ, and the ULOQ was set at 1000 J.A. Custodio-Mendoza et al. Analytica Chimica Acta 1239 (2023) 340712 4 ng/g. Sensitivity of the method was also assessed at the LLOQ in terms of relative standard deviation (%RSD) for each analyte in triplicate. Accuracy of the optimized process was assessed using quality control substances (QCs) at three levels of concentration within the linear range (100, 250 and 500 ng/g). According to FDA guidelines, acceptable criteria for accuracy are recoveries between 80% and 120% of the nominal value. The precision of the method was evaluated by intraday and interday assays in terms of %RSD in quintuplicate. A Matrix extension study was performed using QCs at the same three levels of concentration in a palm oil-free margarine sample to confirm the applicability of the method to this group of samples. 3. Results and discussion 3.1. Optimization of the HPLC-APCI-QqQ-MS/MS method No problems were observed using the HESI-II source for PAGE, OLGE, PAGEd31, and OLGEd5 determinations. Different fragments were observed for each GE, including [M+H] + and [M +NH 4 ] + , allowing the selection of a quantifier and two qualifier transitions for these analytes. However, significant signal suppression was observed when infusing PA, OL, PAd5, and Old5, so the low intensity of the observed ions made it impossible to select the transitions necessary for their determination. Said signal suppression has been previously reported when using this source type [9]. It is associated with the fact that thermal focusing causes rapid desolvation, increasing the spray’s concentration while the excess surface charge remains constant. This situation encourages competition for this extra charge, so the matrix effect on the analyte signal is compounded [29,32]. All analytes were successfully determined using the APCI source, archiving ions with enough intensity to set quantitative and qualitative transitions for PA, PAGE, OL, OLGE, Li, ST, PAd5, PAGEd31, Old5, OLGEd5, and STd5 (Table 1). Nevertheless, an interconversion of the 3MCPD monoesters with their GE analog was observed in this work when using the APCI source. Epoxidation of 3-MCPD monoesters has been described before [33], not as a problem but as a resource to determine 3-MCPD monoesters, so it was decided to use the APCI source in this work. However, to simultaneously determine 3-MCPD monoesters and GE using APCI sources, a proper chromatographic separation must be done before epoxidation during MS determination. Luna OMEGA PS C18 and ACQUITY UPLC BEH Shield RP18 Column were tested for 3-MCPD monoesters and GE separation. ACQUITY UPLC BEH C18 column particles incorporate a series of trifunctional ligands that stabilize the stationary phase over a wide pH range [34]. The structure of the 1.7 μ m BEH particles also uses novel final protection processes, improving the shape of the peak for basic analytes, which increases the sensitivity of LC-MS analysis and simplifies sample preparation [34]. On the other hand, the 5 μ m Luna Omega PS C18 column contains different ligands that give it a broad polar and non-polar retention [35]. However, the exact composition of the stationary phase of this column is not available as it is patented information. According to the manufacturer, it has a positive charge, possibly from some amino group incorporated at the end of a short alkyl chain bonded to the silica surface. This positive surface provides a retention factor for acidic compounds through ionic interactions, while for the reversed-phase separation, C18 carries out retention. In this way, the mixed selectivity of this column allows greater separation between compounds with functional diversity. The positive surface of its particles increases the retention of acidic compounds and repel strong alkaline species. In contrast, the C18 ligand increase the resolution of the analytes of interest by stimulating hydrophobic retention. The separation of 3-MCDP monoesters and GE is challenging due to Fig. 1. Scheme of the QuEChERS protocol for simultaneous extraction of four monoesters of 3-monochloropropane-1,2-diol and two glycidyl esters from oilbased foodstuff. Table 1 Retention time, quantifier and qualifier transitions for 3-MCPD monoesters and GE. Analyte RT Quantifier transition (m/z) Qualifier transition (m/ z) IS min Q1 Q3 (CE) Q3 (CE) Q3 (CE) LI 0.50 371.1 95.1 (26 V) 81.1 (23 V) 67.2 (28 V) PAd5 PA 1.21 313.3 57.3 (21 V) 71.3 (18 V) 95.2 (17 V) PAd5 OL 1.19 339.3 95.2 (18 V) 93.2 (28 V) 69.3 (24 V) OLd5 ST 1.18 341.1 95.10 (37 V) 105.0 (40 V) 91.1 (24 V) STd5 PAGE 1.42 313.3 66.4 (24 V) 82.3 (20 V) 98.3 (18 V) PAGEd31 OLGE 1.48 339.2 69.3 (24 V) 95.2 (18 V) 83.2 (20 V) OLGEd5 PAd5 1.39 318.3 57.3 (23 V) 71.3 (19 V) 95.2 (17 V) – OLd5 1.24 344.3 95.2 (20 V) 69.3 (25 V) 81.2 (21 V) – PAGEd31 1.17 344.5 66.4 (24 V) 82.3 (20 V) 98.3 (18 V) – OLGEd5 1.49 344.3 95.2 (19 V) 69.3 (26 V) 83.2 (19 V) – STd5 1.80 346.1 57.2 (26 V) 71.2 (21 V) 95.0 (20 V) – 3-MCPD, 3-monochloropropane-1,2-diol; GE, glycidyl esters; IS, internal standard; CE, collision energy; LI, linoleoyl-3-MCPD; PA, palmitoyl-3-MCPD; OL, oleoyl-3-MCPD; ST, steareoyl-3-MCPD; PAGE, palmitoyl glycidyl ester; OLGE, Oleoyl glycidyl ester. J.A. Custodio-Mendoza et al. Analytica Chimica Acta 1239 (2023) 340712 5 their structural similarities that lead to similar physicochemical properties. Not enough separation of GE and analog 3-MCPD monoesters was achieved when using ACQUITY UPLC BEH C18 with different ratios of aqueous-organic mobile phases (Fig. S1). However, when using OMEGA PS C18 as a stationary phase, different retention times were observed between PA and PAGE, OL and OLGE, and OLd5 and OLGEd5. The best separation was achieved using up to 15% water in the mobile phase. ACN and MeOH were compared, and a better resolution in terms of peak shape was achieved with ACN, so it was chosen as an organic solvent in the mobile phase. Moreover, the addition of formic acid was tested as an attempt to improve the ionization of the analytes. No significant difference was observed between the addition of 0.01% formic acid versus the absence of it. At the same time, a decrease in the ionization of the analytes was observed using 0.02% formic acid, so it was decided to use un-acidified ACN-water as the mobile phase. Although isocratic 15% water-ACN was enough to elute all the analytes, a gradient program described in section 2.4 was used to ensure the complete elution of all other matrix components. 3.2. Development and optimization of QuEChERS procedure All experiments in this study were performed in triplicate. Different organic solvents were tested for oil dilution and the extraction of the analytes, including MeOH, CHCl 3 , ACN, IPA, ACO, nhexane, Et 2 O, and EtOAc. Only CHCl 3 , ACN, and IPA proved to sufficiently extract the analytes, even though other matrix components were coextracted. CHCl 3 was discarded due to its toxicity, and IPA, ACN, and a 1:1 (v/v) mixture were tested in terms of recovery. Results (Fig. 2A) show that better recoveries were achieved using ACN. This solvent is also used in the mobile phase, so it was decided to use ACN as the extraction solvent. Aqueous to ACN ratios in the liquid-liquid partition were assessed at three levels, 1:3, 1; 1, and 3:1 (v/v) in terms of recovery, achieving better results with 1:1 (v/v) for most analytes (Fig. 2B). The test portion amount was tested at two levels (15 and 150 mg), achieving similar results in terms of recovery for all the analytes, so the smallest test portion should be used. A salting-out effect is an essential tool in QuEChERS since it increases the ion strength of the aqueous phase, enhancing the mass transfer phenomenon into the organic phase and the aqueous-organic phase separation. Ammonium salts have proved to be good salting-out agents and are source-friendly [36], so three different ammonium salts were assessed at different concentrations. Fig. 3 shows that the higher chromatographic peak areas were achieved when using ammonium sulfate compared to ammonium, formate, and acetate. Moreover, the effect of the ammonium sulfate was tested at 5, 10, 15 and 25% w/w in triplicate and in terms of recovery. It was observed that the efficiency of the method varies linearly with the addition of said salt for the analytes. Since there is no significant difference between 15% and 25%, it was decided to use 15% w/w ammonium sulfate as the salting-out agent. A clean-up step using silica-based sorbents is included in QuEChERS to remove other matrix eluents. In this work, six sorbents were tested, including Bond Elut EMR-Lipid®, Oasis ® HLB, Z-sep+, primary secondary amine, C-18, and Strong Anion Xchange silica sorbents. Most of these sorbents were purchased in bulk except for EMR-Lipid and Oasis HLB sorbents, which were purchased as SPE cartridges and cut in the lab to get the sorbent. Fig. 4 shows that the best results in terms of recovery were achieved using EMR-Lipid at 150 mg. The specific absorbability of this sorbent of C5 and long-chain hydrocarbons, as well as glyceride and waxes, promotes stronger selective adsorption for lipids compared with other conventional sorbents reported in previously published analysis [37–39]. 3.3. Analytical features Validation of the proposed method was assessed following FDA guidelines for analytical method validation [17,18]. The selectivity and specificity of the process (Table 1) are based on the selection of one quantifier and two qualifier transitions from a parent ion to a product ion at a specific collision energy. Moreover, as discussed before, 3-MCPD monoesters and GE of the same fatty acid shared the same parent ion, so the chromatographic separation of the analyte is a crucial factor in this work. Successfully, different retention times are given for these analytes using Luna OMEGA PS C18 (Table 1). No matrix interference was observed in the retention times of the analytes or the IS (Fig. 5). Chromatographic signals of 3-MCPD seem to be 10 times smaller than those of the GE analog, perhaps because of the transformation of 3-MCPD into GE during ionization, nonetheless, sensibility was sufficient for simultaneous determination in a wide range of concentrations. Table 2 presents analytical features, including limits, linearity, accuracy, and precision. LOD and LLOQ were 3 and 10 ng/g for GE and 30 and 100 ng/g for 3-MCPD monoesters. All the analytes achieved excellent determination coefficients (r 2 ≥0.9992) within the linear range (LLOQ to 1000 ng/g). Accuracy assessed using QCs in triplicate resulted in recoveries spanning 84.8%–118.7%. Excellent precision is confirmed by intraday and interday assays, both in quintuplicate with a coefficient of variation (%RSD) lower than 12.4%. Even when the APCI source is less susceptible to matrix effect when compared to other sources [31], a matrix-matched calibration with internal standard in a margarine sample was used to avoid possible matrix interferences due the matrix complexity. The matrix extension study supports the applicability of the developed method in the analysis of 3MCP monoesters and GE in margarine samples (Table S2). Previously published articles (Table S3) on the simultaneous determination of GE and 3-MCPD monoesters reported quantification limits Fig. 2. Selection of extraction solvent (A), aqueous-organic phases ratio (B) and test portion (C). IPA, isopropanol; ACN, acetonitrile. In triplicate using an EVO sample spiked at 0.5 μ g/g. J.A. Custodio-Mendoza et al. Analytica Chimica Acta 1239 (2023) 340712 6 similar to or higher than those presented here [11,12,16,20–27]. Hidalgo-Ruiz et al. (2021) achieved lower LOQ for most of the 3-MCPD monoesters by using a much larger test portion size (more than sixty times) than that of this work, but reporting similar LLOQs for LI and GE [21]. Other works focused on the determination of 3-MCPD monoesters [21–24], often alongside 3-MCPD diesters, and also reported equal or higher limits than those of this work, except for Chai et al. (2016) for the determination of 3-MCPD monoand di-esters in beef flavoring samples [20]. Chai et al. also accomplished lower LOQs by increasing the test portion amount to 1g [20]. However, despite the potential improvement in method sensitivity with larger test portion sizes, its miniaturization is preferred as it follows the trends of green analytical chemistry while allowing volume reduction of solvents, adsorbents, and waste. Moreover, most methods developed for GE determination reported limits much higher than those reported here [16,26]. Blumhorst et al. (2013) presented lower limits in their collaborative study on edible oils [27]; however, they are instrumental limits calculated in the absence of a sample. So, they are related to the instrument response and not to the application of the entire analytical method. The work presented in this paper is more accurate and precise than those reported before in terms of % recovery and variation coefficients ≤12.4% [10,11,15,19–22,24,25]. 3.4. Analysis of margarine samples This new QuEChERS protocol was used to simultaneously determine and quantify two GE and four 3-MCPD monoesters with fatty acids in 11 margarine samples by HPLC-APCI-QqQ-MS/MS. The results are presented in Table 3 and Fig. S2. All samples were positive for at least one GE or 3-MCPDE. PA and PAGE were below the LLOQ in all samples. This could be due to the current trend to promote healthy nutrition, which would imply avoiding or reducing the use of palm oil in processed foods such as in the blend of oils used in margarine production. LI and OL were only quantifiable in two samples at levels between 0.24 and 0.28 μ g/g, and 0.14–0.16 μ g/g, respectively. ST was present in all samples and quantified in nine of them at concentrations among 0.10–0.54 μ g/g. ST was also the analyte with the highest amounts found in the analyzed samples. Regarding GE, OLGE was quantifiable in all samples at levels from 0.03 to 0.22 μ g/g. These values are in accordance with those previously reported by Hidalgo-Ruiz et al. [21]. Since the analysis of 3-MCPDE and GE in margarines has been mostly performed using indirect methods [6,7], an estimation of the equimolar contribution of these analytes to the presence of free 3-MCPD and GE were evaluated using equations (1) and (2). The results are present in Table 3. Equivalent 3MCPD =∑ i n=1 [3MCPDEi] Molecular weight 3MCPDEi ∗110.539 g/mol⋯ (1) Fig. 3. Study of type of salt (A) and amount of ammonium sulfate (B). AF, ammonium formate; AA, ammonium acetate; AS, ammonium sulfate. In triplicate using an EVO sample spiked at 0.5 μ g/g. Fig. 4. Study of type of clean-up sorbent (A) and amount of EMR-lipid (B). In triplicate using an EVO sample spiked at 0.5 μ g/g. J.A. Custodio-Mendoza et al. Analytica Chimica Acta 1239 (2023) 340712 7 Equivalent Glycidol =∑ i n=1 [3MCPDEi] Molecular weight 3MCPDEi ∗74.08 g/mol⋯(2) The results obtained in this work for the individual and simultaneous quantification of PA, OL, LI, ST, PAGE and OLGE are in accordance with those reported by Hidalgo-Ruiz et al. [21] for the analysis of 3-MCPDE and GE in fatty matrices (including margarines). Equivalent 3-MCPD and glycidol ranged from 0.07 to 0.16 μ g/g and 0.01–0.05 μ g/g. These Fig. 5. Chromatogram of four monoesters of 3-MCPd and two GE in a margarine sample enriched at 0.5 μ g/g. Table 2 Analytical method validation parameters of QuEChERS-HPLC-APCI-QqQ-MS/MS method for analysis of four monoesters of 3-MCPD and two GE in oil-based foodstuff. Analyte LLOQ r 2a Accuracy Precision (n =3)% recovery Intraday (n =5) % RSD Interday (n =5)% RSD ng g −1 L-Qc M-QC H-QC L-Qc M-QC H-QC L-Qc M-QC H-QC LI 100 0.9996 111 95 98 7 5 5 9 4 6 PA 100 0.9999 110 112 104 1 4 7 11 8 8 OL 100 0.9993 91 115 103 10 10 3 5 12 8 ST 100 0.9992 107 109 119 2 7 5 6 4 4 PAGE 10 0.9995 109 100 93 12 10 7 9 4 12 OLGE 10 0.9998 85 87 104 11 2 3 12 12 5 3-MCPD, 3-monochloropropane-1,2-diol; GE, glycidyl esters; CV, coefficient of variation; L-QC; low quality control at 100 ng/g; M-QC; middle quality control at 250 ng/g; high quality control at 500 ng/g. a Linear range from LLOQ to 1000 ng/g. J.A. Custodio-Mendoza et al. Analytica Chimica Acta 1239 (2023) 340712 8 estimations agree with the 2016 EFSA report [2] and those reported by previous works [21]. Nevertheless, since most of these articles are based on the indirect determination, the levels of 3-MCPD reported include 3-MCPD released from both monoand diesters, so levels of total 3-MCPD similar and higher than those reported here have been found [6,7]. This could be related to the contribution of 3-MCPD diesters to total 3-MCPD. Based on the last annual report on food consumption in Spain [40], the per capita domestic consumption of margarine was 0.55 kg per person. Since the composition of this margarine is not specified in this report, the risk of exposure was estimated based on the highest results of the 3-MCPD equivalent obtained in this study. The consumption of this margarine would not imply a concern in intake, which would be well below the established TDI of 2 μ g/kg body weight per day. However, this margarine sample would be within the maximum limits set by the European Regulation Commission [5]. Since margarine is a product used in preparing other foods or culinary activities, it could represent a risk of increased consumption of 3-MCPD and GE. And so, it will be crucial to study the increase in the occurrence of these contaminants in margarines when cooked. 4. Conclusions A novel method for simultaneously determining 3-MCPD monoesters and glycidyl esters in margarine has been successfully developed and validated. The APCI source allows this analysis to be performed using a much simpler mobile phase without acidification. The miniaturization of the test portion size in the proposed QuEChERS method follows the trends of green analytical chemistry. The use of decreased volumes of extraction solvents and clean-up sorbents compared to other sample preparation methods currently used significantly impacts the profitability of this method. This method showed excellent performance in terms of limits of determination, linearity, accuracy, precision, and suitability for analyzing the 3-MCPDE and GE levels in margarine samples, according to FDA validation guidelines. Determining food contaminants such as glycidyl esters and 3-MCPD with fatty acids in margarine is crucial in ensuring food quality. Beyond their domestic consumption, margarines are widely used in the food industry as raw material in the production of other products such as cakes and pastries due to their lower price than fats of animal origin (butter) or oils of vegetable origin such as extra virgin olive oil. So, it will be imperative to monitor the potentially increased exposure to these contaminants after cooking. Here, the simultaneous determination of four 3-MCPDE and two GE in margarine is performed in less than 40 min from sampling to data analysis. All samples were positive for OLGE. To the best of our knowledge, this is the first simultaneous study of 3-MCPD monoesters and GE esters in margarine samples. Funding The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. CRediT authorship contribution statement Jorge A. Custodio-Mendoza: Investigation, Validation, Formal analysis, Visualization, Writing – original draft, Writing – review & editing. Raquel Send´ on: Methodology, Writing – review & editing, Supervision. Ana Rodríguez-Bernaldo de Quir´ os: Methodology, Writing – review & editing, Supervision. Rosa A. Lorenzo: Conceptualization, Methodology, Writing – review & editing, Supervision. Antonia M. Carro: Conceptualization, Methodology, Formal analysis, Data curation, Writing – original draft, Writing – review & editing, Supervision. Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: A. M. Carro reports financial support was provided by Spanish Ministry of Science, Innovation, and Universities. Data availability Data will be made available on request. Acknowledgments This work was supported by the Spanish Ministry of Science, Innovation, and Universities (Project RTI2018-096450-B-I00) and FEDER funds. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.aca.2022.340712. References [1] A. Eisenreich, B.H. Monien, M.E. G¨ otz, T. Buhrke, A. Oberemm, K. Schultrich, K. Abraham, A. Braeuning, B. Sch¨ afer, 3-MCPD as contaminant in processed foods: state of knowledge and remaining challenges, Food Chem. (2022) 134332. [2] W.W. Cheng, G.Q. Liu, L.Q. Wang, Z.S. Liu, Glycidyl fatty acid esters in refined edible oils: a review on formation, occurrence, analysis, and elimination methods. Comp. Reviews in Food Sc. and, Food Saf. 16 (2017) 263–281. Table 3 Content of 3-MCPD monoesters and GE in margarine samples. Sample LI PA OL ST PAGE OLGE 3-MCPD* Glycidol* ng/g ±SD ng/g ±SD ng/g ±SD ng/g ±SD ng/g ±SD ng/g ±SD ng/g ng/g M1 ND – ND – ND – 203.8 17.2 ND – 43.4 20.4 59.7 9.5 M2 ND – ND – ND – 284.3 20.2 ND – 27.8 5.9 83.4 6.1 M3 ND – ND – ND – 538.2 26.9 ND – 100.4 22.1 157.8 22.0 M4 ND – ND – ND – 542.7 26.0 ND – 54.9 3.2 159.1 12.0 M5 ND – ND – 157.4 16.3 309.1 52.1 ND – 70.3 57.7 137.0 15.4 M6 ND – ND – ND – 438.8 9.0 ND – 82.3 56.0 128.7 18.0 M7 ND – ND – 143.6 19.9 96.8 19.8 ND – 216.5 13.9 70.7 47.4 M8 ND – ND – ND – 346.0 5.5 ND – 43.4 59.7 101.4 9.5 M9 242.9 10.4 ND – ND – ND – ND – 78.9 14.1 72.0 17.3 M10 ND – ND – ND – ND – ND – 157.9 47.8 – 34.5 M11 276.5 16.9 ND – 38.6 7.5 149.7 6.0 ND – 191.0 31.4 125.8 41.8 3-MCPD, 3-monochloropropane-1,2-diol; GE, glycidyl esters; LI, linoleoyl-3MCPD; PA, palmitoyl-3MCPD; OL, oleoyl-3-MCPD; ST, steareoyl-3-MCPD; PAGE, glycidyl palmitate; OLGE, oleoyl palmitate. *Estimated from their esterified forms. ND, not determined. J.A. Custodio-Mendoza et al. Analytica Chimica Acta 1239 (2023) 340712 9 [3] EFSA Panel on Contaminants in the Food Chain (CONTAM), Risks for human health related to the presence of 3-and 2-monochloropropanediol (MCPD), and their fatty acid esters, and glycidyl fatty acid esters in food, EFSA J. 14 (2016), e04426. [4] International Agency for Research on Cancer, Some chemicals present in industrial and consumer products, Monogr. Eval. Carcinog. Risks Hum. 101 (2012) 349–374. [5] E.U. 2020/1322, COMMISSION REGULATION (EU) 2020/1322 of 23 September 2020 amending Regulation (EC) No 1881/2006 as regards maximum levels of 3monochloropropanediol (3-MCPD), 3-MCPD fatty acid esters and glycidyl fatty acid esters in certain foods, Off. J. Eur. Union L310 (2020) 2–5. [6] R. Jędrkiewicz, A. Głowacz, J. Gromadzka, J. Namie´ snik, Determination of 3-MCPD and 2-MCPD esters in edible oils, fish oils and lipid fractions of margarines available on Polish market, Food Control 59 (2016) 487–492. [7] K.M. Goh, Y.H. Wong, M.Y. Ang, S.C.M. Yeo, F. Abas, O.M. Lai, C.P. Tan, Comparison assessment between SIM and MRM mode in the analysis of 3-MCPD ester, 2-MCPD ester and glycidyl ester, Food Res. Int. 121 (2019) 553–560. [8] X. Zheng, W. Fu, K. Zheng, B. Gao, L. Lin, W. Liu, Z. Lin, Q. Fang, A novel method for the simultaneous determination of esterified 2-/3-MCPD and glycidol in foods by GC-MS/MS, Food Control 123 (2021) 107766. [9] G. Graziani, A. Gaspari, A. Chianese, L. Conte, A. Ritieni, Direct determination of 3chloropropanol esters in edible vegetable oils using high resolution mass spectrometry (HRMS-Orbitrap), Food Addit. & Contam.: Part A 34 (2017) 1893–1903. [10] J.A. Custodio-Mendoza, R.A. Lorenzo, I.M. Valente, P.J. Almeida, M.A. Lage, J. A. Rodrigues, A.M. Carro, Development of a partitioned liquid-liquid extractiondispersive solid phase extraction procedure followed by liquid chromatographytandem mass spectrometry for analysis of 3-monochloropropane-1, 2-diol diesters in edible oils, J. Chromatogr. A 1548 (2018) 19–26. [11] S. MacMahon, E. Mazzola, T.H. Begley, G.W. Diachenko, Analysis of processing contaminants in edible oils. Part 1. Liquid Chromatography−Tandem mass spectrometry method for the direct detection of 3-monochloropropanediol monoesters and glycidyl esters, J. Agric. Food Chem. 61 (2013) 4737–4747. [12] K. Yamazaki, M. Ogiso, S. Isagawa, T. Urushiyama, T. Ukena, N. Kibune, A new direct analytical method using LC-MS/MS for fatty acid esters of 3-chloro-1,2propanediol (3-MCPD esters) in edible oils, Food Addit. Contam. 30 (2013) 52–68. [13] J.K. Beekman, S. MacMahon, The impact of infant formula production on the concentrations of 3-MCPD and glycidyl esters, Food Addit. Contam. Part A 37 (2020) 48–60. [14] J.K. Beekman, S. Popol, M. Granvogl, S. MacMahon, Occurrence of 3-monochloropropane-1,2-diol (3-MCPD) esters and glycidyl esters in infant formulas from Germany, Food Addit. Contam. Part A 38 (2021) 1656–1671. [15] N. Ahmad, M.A. Shabbir, M.S. Butt, M. Shahid, T.H. Robert, Quantification of 3monochloropropane-diol fatty acid esters by liquid chromatography-mass spectrometry method with their subsequent reduction in processed foods, Pakistan J. Agric. Sci. 57 (2020) 763–771. [16] M. Aniołowska, A. Kita, Monitoring of glycidyl fatty acid esters in refined vegetable oils from retail outlets by LC–MS, J. Sci. Food Agric. 96 (2016) 4056–4061. [17] Food and Drug Administration, U. S. Guidelines for the validation of chemical methods in food, feed, cosmetics, and veterinary products. Food and Drug Administration: silver Spring, MD, USA 1–39 (1) (2019). Available online, https://www.fda.gov/media/81810/download. September 26 th 2021 st . [18] Food and Drug Administration, U. S. Bioanalytical method validation guidance for industry, U S Dep. Health Hum. Serv. 1–41 (2018) 1. Available online, htt ps://www.fda.gov/files/drugs/published/Bioanalytical-Method-Validation-Guid ance-for-Industry.pdf. September 26th 2021 st . [19] J.K. Beekman, K. Grassi, S. MacMahon, Updated occurrence of 3-monochloropropane-1, 2-diol esters (3-MCPD) and glycidyl esters in infant formulas purchased in the United States between 2017 and 2019, Food Addit. Contam. 37 (2020) 374–390. [20] Q. Chai, X. Zhang, E. Karangwa, Q. Dai, S. Xia, J. Yu, Y. Gao, Direct determination of 3-chloro-1, 2-propanediol esters in beef flavoring products by ultra-performance liquid chromatography tandem quadrupole mass spectrometry, RSC Adv. 6 (2016) 113576–113582. [21] J.L. Hidalgo-Ruiz, R. Romero-Gonz´ alez, J.L.M. Vidal, A.G. Frenich, Determination of 3-monochloropropanediol esters and glycidyl esters in fatty matrices by ultrahigh performance liquid chromatography-tandem mass spectrometry, J. Chromatogr. A 1639 (2021) 461940. [22] K. Hori, N. Koriyama, H. Omori, M. Kuriyama, T. Arishima, K. Tsumura, Simultaneous determination of 3-MCPD fatty acid esters and glycidol fatty acid esters in edible oils using liquid chromatography time-of-flight mass spectrometry, LWT–Food Sci. Technol. 48 (2012) 204–208. [23] J.K. Leigh, S. MacMahon, Extraction and liquid chromatography–tandem mass spectrometry detection of 3-monochloropropanediol esters and glycidyl esters in infant formula, J. Agric. Food Chem. 64 (2016) 9442–9451. [24] H. Li, D. Chen, H. Miao, Y. Zhao, J. Shen, Y. Wu, Direct determination of fatty acid esters of 3-chloro-1, 2-propanediol in edible vegetable oils by isotope dilution-ultra high performance liquid chromatography-triple quadrupole mass spectrometry, J. Chromatogr., A 1410 (2015) 99–109. DOI. [25] H. Zhou, Q. Jin, X. Wang, X. Xu, Direct measurement of 3-chloropropane-1, 2-diol fatty acid esters in oils and fats by HPLC method, Food Control 36 (2014) 111–118. [26] A. Becalski, S.Y. Feng, B.Y. Lau, T. Zhao, Glycidyl fatty acid esters in food by LCMS/MS: method development, Anal. Bioanal. Chem. 403 (2012) 2933–2942. [27] M.R. Blumhorst, M.W. Collison, R. Cantrill, H. Shiro, Y. Masukawa, S. Kawai, K. Yasunaga, Collaborative study for the analysis of glycidyl fatty acid esters in edible oils using LC-MS, J. Am. Oil 90 (2013) 493–500. [28] A. Laaniste, I. Leito, A. Kruve, ESI outcompetes other ion sources in LC/MS trace analysis, Anal. Bioanal. Chem. 411 (2019) 3533–3542. [29] H. Stahnke, S. Kittlaus, G. Kempe, C. Hemmerling, L. Alder, The influence of electrospray ion source design on matrix effects, J. Mass Spectrom. 47 (2012) 875–884. [30] I. Kourtchev, P. Szeto, I. O’Connor, O.A.M. Popoola, W. Maenhaut, J. Wenger, M. Kalberer, Comparison of heated electrospray ionization and nanoelectrospray ionization sources coupled to ultra-high-resolution mass spectrometry for analysis of highly complex atmospheric aerosol samples, Anal. Chem. 92 (2020) 8396–8403. [31] L.M. Morrison, J.B. Renaud, L. Sabourin, M.W. Sumarah, K.K. Yeung, D.R. Lapen, High-throughput quantitation of neonicotinoids in lyophilized surface water by LCAPCI-MS/MS, J. AOAC Int. 101 (2018) 1940–1947. [32] A.K. Rahn, V.A. Yaylayan, Cyclic acyloxonium ions as diagnostic aids in the characterization of chloropropanol esters under electron impact (EI), electrospray ionization (ESI), and atmospheric pressure chemical ionization (APCI) conditions, J. Agric. Food Chem. 61 (2013) 5946–5952. [33] N. Kaze, H. Sato, H. Yamamoto, Y. Watanabe, Bidirectional conversion between 3monochloro-1, 2-propanediol and glycidol in course of the procedure of DGF standard methods, J. Am. Oil Chem. Soc. 88 (2011) 1143–1151. [34] L.S. New, E.C. Chan, Evaluation of BEH C18, BEH HILIC, and HSS T3 (C18) column chemistries for the UPLC-MS-MS analysis of glutathione, glutathione disulfide, and ophthalmic acid in mouse liver and human plasma, J. Chromatogr. Sci. 46 (2008) 209–214. [35] Z. Kadlecov´ a, P. Kozlík, E. Tesaˇ rov´ a, M. Gilar, K. Kalíkov´ a, Characterization and comparison of mixed-mode and reversed-phase columns; interaction abilities and applicability for peptide separation, J. Chromatogr. A 1648 (2021), 462182. [36] A. Nemeˇ skalov´ a, M. Bursov´ a, D. Sýkora, M. Kuchaˇ r, R. ˇ Cabala, T. Hloˇ zek, Salting out assisted liquid-liquid extraction for liquid chromatography tandem-mass spectrometry determination of amphetamine-like stimulants in meconium, J. Pharmaceut. Biomed. Anal. 172 (2019) 42–49. [37] L. Han, J. Matarrita, Y. Sapozhnikova, S.J. Lehotay, Evaluation of a recent product to remove lipids and other matrix co-extractives in the analysis of pesticide residues and environmental contaminants in foods, J. Chromatogr. A 1449 (2016) 17–29. [38] F. Zhu, X. Wu, F. Li, W. Wang, W. Ji, Z. Huo, Y. Xu, Simultaneous determination of 12 antibacterial drugs in cream disinfection products with EMR-Lipid cleanup using ultra-high-performance liquid chromatography tandem mass spectrometry, Anal. Methods 11 (2019) 4084–4092. [39] P. Luo, X. Liu, F. Kong, L. Tang, Q. Wang, W. Li, W. Xu, S. Wen, L. Chen, Y. Li, Multi-residue determination of 325 pesticides in chicken eggs with EMR-Lipid clean-up by UHPLC–MS/MS and GC–MS/MS, Chromatographia 83 (2020) 593–599. [40] 2021 MAPA, Ministerio de Agricultura y pesca, alimentaci´ on y medio ambiente, Informe del consumo alimentario en Espa˜ na 2020 (2021). v2-nov2021-baja-res_ tcm30-562704.pdf">, https://www.mapa.gob.es/es/alimentacion/temas/con sumo-tendencias/informe-anual-consumo-2020-v2-nov2021-baja-res_tcm30-562 704.pdf. October 24 th 2022 st . J.A. Custodio-Mendoza et al.