Project Topics
Get Complete Project Material File(s) Now! »
Regulation of organophosphorus in vegetable oils
Generally, the EU Regulation No 396/2005 sets maximum residue levels (MRLs) of pesticides that are legally tolerated in food or feed. As general default, a MRLs of 10 µg/kg are applied when a pesticide is not specifically mentioned. These limits established for pesticides can be found in the MRL database of the European Commission website.
The french Institute specialized in fats and oils (ITERG) have established a list of the most detected pesticides in vegetable oils according to their recent studies. For this work, this list was used to select the studied OPs. These compounds reported in the Table I.2-1 were present in several vegetable oils. However, in this study we have focused only on three vegetable oils (olive, almond and sunflower) that can be used as raw material to elaborate cosmetic products. Hence in the Table I.2-1 we summarize the update (MRLs) established originally by the EU regulation No. 396/2005 in oil seeds.
Extraction techniques of organophosphorus from vegetable oils
A wide range of OPs are used legally for seeds protection and their residue content in vegetable oils must be accurately monitored for safe consumption. These matrices contain a high level of triglycerides and the possible presence of lipophilic analytes at low concentration [23] which requires complicated sample treatment procedures before chromatographic analysis. Indeed, it is a crucial step in the analytical procedure since even a small residual amount of lipids can damage LC columns or cause signal suppression during MS detection. It is then necessary to simplify the matrix by removing interfering compounds in order to improve detection of pesticide residues and to achieve the lowest limits of detection and quantification [4]. Basically, a sample treatment procedures is required prior to analysis by gas chromatography (GC) or high performance liquid chromatography (HPLC) determination and follows these basic steps:
• The food sample is homogenized or blended to obtain a uniform matrix.
• The pesticide residue will be extracted from the matrix with solvents.
• A cleanup step is used to remove interfering matrix components to decrease the matrix effect during chromatographic analysis.
• The eluent is concentrated and re-constitute in a solvent which is compatible with the GC or HPLC analytical conditions.
The analysis of OPs were carried out by gas chromatography (GC) coupled to different detectors such as flame thermionic detector (FTD)[24], nitrogen/phosphorus detector (NPD) [16,25], flame photometric detector (FPD) [15,26,27] or the more specific mass spectrometry (MS)[28–30] in selected ion monitoring (SIM) mode or in tandem MS/MS [5,23,27,31–33] or by liquid chromatography generally coupled to MS/MS detection [17,28,34–36] with an advantageous features for the analyses of polar pesticides in olive oil. Accordingly prior to the separation and detection, the most widely used techniques to extract OPs in vegetable oils are: liquid-liquid extraction (LLE) [5,15,23,25–30,33–37], solid-phase extraction (SPE) [16,25,38], solid phase microextraction (SPME)[24], matrix solid phase dispersion (MSPD) [26,28,36] or dispersive solid phase extraction (dSPE) [5,23,29,30,33–35] that is usually applied in QuEChERs methods. Other extraction techniques such lower temperature precipitation [17,27], gel permeation chromatography (GPC) [31] or microwave-assisted extraction (MAE)[38] were also applied. The applications of these different methods of separation and detection and of extractions of OPs from vegetable oils are summarized in Table I.4-1.
Liquid-liquid Extraction (LLE)
Liquid-liquid extraction (LLE) is based on the relative solubility of an analyte in two immiscible phases and is defined by the equilibrium distribution/partition coefficient. LLE is traditionally one of the most common methods of extraction, particularly for organic compounds from aqueous matrices. Typically, a separating funnel is used and the two immiscible phases are mixed by shaking and then allowed to separate. To avoid emulsions, in some cases, a salt may be added and centrifugation can be used if necessary [39].
LLE is largely applied to extract OPs from vegetable oils. The most often used solvent in LLE partitioning are acetonitrile or a mixture of acetonitrile and hexane. It has be used without no subsequent clean up steps to extract and analyze some OPs directly from olive oils [15]. In this case, the resulting limit of quantification was relatively high (between 3 and 15 µg/kg), Table I.4-1. However, the extracts obtained after LLE contain a significant amount of residual fat that could interfere with the analysis. Nowadays, most of the methods described for the analyses of OPs in vegetable oils include a subsequent clean-up step. LLE had been used to extract OPs in vegetable oils combined with MSPD [26,28,36], dSPE [5,23,29,30,33–35], GPC [31], that allow the separation of the low molecular mass pesticides from higher molecular mass fat constituents of the oils, such as triglycerides, SPE [25] or with lower temperature precipitation [30,36]. The last extraction technique consists of a precipitation of the fatty component of the oils at lower temperature and generally ACN is used as extraction solvent. However when is applied without any supplementary extraction step, as is shown in Table I.4-1, the recoveries present high RSD (between 15 and 27%) due to the matrix effects [17,27].
Recently the need to reduce solvent usage has led to microextraction techniques [40], such as dispersive liquid−liquid microextraction (DLLME). This technique was emerged in 2006 and was described by Rezaee et al.[41]. It had shown high recovery and enrichment factors in comparison with classic LLE. In the DLLME technique, a mixture of an organic solvent as the extractant and a disperser solvent is rapidly injected into an aqueous sample so that the turbulence produced causes the formation of fine droplets, which are dispersed through the aqueous sample. The emulsified droplets have a great interstitial area and, consequently, the equilibrium is reached rapidly and the extraction is almost instantaneous [42]. Recently this technique was used to extract 3 OPs from several vegetable oils (olive, flaxseed, walnut and coconut) by using as extractant magnetic water prior to the analysis by GC-MS/MS. This procedure led to recoveries included between 78 and 138% and to limits of quantification between 0.7 and 1.27µg/kg [32]. Even if the recoveries were higher than 100%, which is probably due to the matrix effects, this method of extraction coupled with a more specific detector (MS-MS) allowed an important decreased of the limits of quantification as compared to classical LLE [15].
Solid phase extraction (SPE)
As it requires a lower volume of solvent than LLE and it imply simple manipulations which are less time consuming and that could be automatized, solid phase extraction (SPE) was developed in 1970 as an alternative approach to LLE for separation, purification, pre-concentration and solvent exchange of analytes. SPE can be used directly as an extraction technique for liquid matrices, or as a cleanup steps for solvent extracts. A SPE method consists in four successive steps, as illustrated in Figure I.3-1. First, the solid sorbent should be conditioned using an appropriate solvent. Typically, for reversed phase sorbent, methanol is frequently used, followed by water or an aqueous buffer whose pH and ionic strength are similar to that of the sample. The second step is the percolation of the sample through the sorbent (in this step the analytes are retained on the sorbent). The third step consist in the washing of the sorbent with an appropriate solvent, to eliminate matrix components which have been retained without displacing the analytes. The final step is the elution of the analytes of interest by an appropriate solvent that allows to recover the analyte of interest without removing the retained matrix component [39]. In the SPE different sorbents can be used (e.g., florisil, alumina, aminopropyl, graphitized carbon black or silica gel). SPE has been used without any additional step to extract 18 OPs from olive using classical sorbents such as silica gel and C18 silica [16], the obtained recoveries were over 100% with RSD until 16%. However when this technique was combined with an additional extraction step such LLE [25] or with matrix accelerated extraction MAE [38] used also to extract OP from olive oil, the obtained recoveries and the RSD were lower with also lower LOQs by using the same detector (NPD). Hence, the combination of several extraction steps allow better recovery yields and cleaner extracts thanks to the reduction of the matrix effects. Different SPE methodologies such as, SPME [24], MSPD [26,28,36] or dSPE [5,23,29,30,33–35]) were used to extract several OPs from vegetable oils and the performances of these techniques are described below.
MIP-SPE of OPPs
As shown by the conditions of synthesis reported in Table II.5-1 , MIPs for SPE were mainly prepared by bulk polymerization. The resulting monolith was ground to obtain 25-50 µm particles that were packed between two frits in disposable cartridges and applied as conventional SPE sorbent (C18 silica, polymers…) to the extraction of OPPs from real samples.
Except in one case for which three OPPs were studied as template before eventually choosing omethoate as template [81], the reported works described the use of a unique OPP to prepare a MIP for this molecule and then for its selective extraction from real samples.
In more than 75% of the reported studies, MAA was used as monomer without any preliminary studies related to the selection of this monomer. The computational screening of monomers was only reported by Bakas et al. [65,71,81] that allows them to select a unique monomer that presents the highest interaction energy with the template. In one of these studies, several MIPs were synthesized using the several selected monomers (IA, MAA, TFMAA) and SPE was carried out to definitively select IA, since leading to a MIP that provides the highest retention and the best selectivity for omethoate [81].
In most of the cases, the presence of specific cavities was proven by binding experiments in a pure solvent spiked with increasing amounts of the target molecule. These experiments allow the affinity of the binding sites of the MIP to being compared with those of the NIP and then to evaluate the presence of specific cavities in the MIP. This approach was also used by Zhu et al. to evaluate the best monomer and solvent among three to produce cavities of high affinity for monocrotophos [77]. If binding experiments can also be used to determine the affinity of the MIP towards other OPPs [67,77], the ability of a MIP to trap several OPPs has been mainly done by measuring extraction recoveries on MIP and on NIP after the application of a SPE procedure previously optimized by studying the target compound alone. In a SPE procedure, different parameters can be studied such as (i) the nature of the percolated solution that must favor the retention, (ii) the composition of the washing solution that constitutes a key parameter for differentiating to differentiate the MIP from and the NIP and (iii) the nature and the volume of the elution solution to recover the target analyte. This is particularly well illustrated with the results reported by Bakas et al. [65] and related to the selective extraction of fenthion and four other OPPs from olive oil using a MIP produced with fenthion as template and AAM as monomer in dimethylformamide. After studying the retention of fenthion on MIP/NIP in different solvents, heptane was selected for its ability to favor the retention of this compound on MIP and different solvents were further tested as washing solvent to select the one that allows the retentions of between MIP and NIP to being differentiated. As shown by results reported in Figure II.5-1, the use of dichloromethane allows fenthion to being partially removed from the NIP (40%) during the washing step (Figure II.5-1a) while maintaining the retention on the MIP (Figure II.5-1b). To improve the selectivity of the procedure, an increasing amount of acetonitrile was added in dichloromethane. This study showed that the use of 5% acetonitrile in dichloromethane (v/v) allows 98% fenthion to being removed from the NIP (Figure II.5-1c) while maintaining its retention on the MIP (Figure II.5-1d).
SPE procedure applied in pure media
Different studies were carried out on the synthesized polymers to optimize the SPE procedure, as the selection of the percolation solvent or the washing solution. Before the percolation, the cartridges were conditioned with 4 mL of the used percolation solvent. Then 1 mL of percolated solvent (toluene, DCM, hexane or mix of hexane, DCM and ACN (70/29/1, v/v/v)) spiked with 1 mg/L of PE was passed through MIP/NIP 1 cartridges. To study the washing solvents, a spiked solution of hexane using six OPs at 1 mg/L was used as percolation solution on the 6 synthesized MIPs/NIPs. Three washing steps were included in an SPE procedure: 1 mL of hexane and DCM (80/20, v/v), 1 mL of hexane, DCM and ACN (80/18/2, v/v/v) and 1 mL of hexane, DCM and ACN (80/15/5, v/v/v). The second procedure applied to the six synthesized MIPs consisted in a single washing step with 1 mL of a mixture of hexane and DCM (95/5, v/v). After the washing step, the cartridge was dried by 5 mL of air. Finally, the OPs were eluted with 1 mL of ACN. Each fraction resulting from each step was evaporated to dryness by a nitrogen stream and was resuspended in 0.5 mL of ACN before injection in the LC-DAD system.
Preliminary extraction procedure for the vegetable oils
Before the SPE procedure using MIP or NIP sorbents, an LLE was performed on oil samples. This LLE procedure was described by the ITERG (French Institute specialized in fats and oils) and used before an SPE step using a C18 sorbent. LLE was carried out using 3 x 1 mL of a mixture of ACN and DCM (90/10, v/v) for 200 mg of oil. The obtained oil extract was evaporated to dryness under nitrogen stream and was spiked with 2.5 mg/kg of ten OPs in 1 mL of hexane and passed through the MIP 2 and NIP 2 cartridges. After a conditioning step with 4 mL of hexane, the oil extract was percolated and 1 mL of a mixture of hexane and DCM (95/5, v/v) was used for the washing step. Finally, the OPs were eluted in 1 mL of ACN. The elution fraction was directly injected in the LC-MS/MS and LC-UV systems. For the clean-up on C18, 12 mL of MeOH and 12 mL of ACN were passed through the cartridge for conditioning, then 3 mL of oil extract resulting from the LLE step were percolated, and 1.5 mL of MeOH was used for the elution step. The elution fraction was recollected and evaporated under nitrogen stream. Finally, the dry extract was suspended in ACN, before its analysis by LC-MS.
Optimized extraction procedure on MIP for the vegetable oils
Optimization of the SPE procedure was necessary to reach the MRLs established by the regulation (EC) No 396/2005 of the European Parliament and of the Council of 23 February 2005. LLE was carried out as in the previous section. The obtained oil extract was evaporated to dryness under nitrogen stream, diluted with 10 mL of hexane and was spiked with a low concentration of 100 µg/kg of three OPs (MTH, MAL, DIZ). After conditioning the MIP/NIP with 4 mL of hexane, 1 mL of the oil extract was percolated through MIP/NIP cartridgs and different volumes of washing solution hexane and DCM (95/5, v/v) were tested: 0.4, 0.65, 0.8 and 1 mL. Finally, the OPs were eluted with 1 mL of ACN. The elution fraction was evaporated to dryness under nitrogen stream and suspended in 100 µL of ACN before injection in the LC-MS/MS system.
Table of contents :
Remerciements
Liste des publications et communications
Résumé
Abstract
Liste des abréviations
Introduction générale
PART I: BIBLIOGRAPHIC STUDY
Chapter I: Presence of organophosphorus pesticides in vegetable oils
I.1. Pesticides, generalities
I.2. Vegetable oils
I.2.1. Regulation of organophosphorus in vegetable oils
I.2.2. Organophosphorus pesticides
I.2.3. Physical and chemical properties
I.3. Extraction techniques of organophosphorus from vegetable oils
I.3.1. Liquid-liquid Extraction (LLE)
I.3.2. Solid phase extraction (SPE)
I.3.3. Dispersive Solid Phase Extraction (dSPE)
I.3.4. Matrix Solid Phase Dispersion (MSPD)
I.3.5. Solid Phase Microextraction (SPME)
I.4. Conclusions
References
Chapter II: Molecularly imprinted polymers for the determination of organophosphorus pesticides in complex samples
II. Review
II.1. Abstract
II.2. Introduction
II.3. Synthesis of MIPs
II.4. MIP characterization
II.5. MIP for selective extraction
II.5.1. MIP-SPE of OPPs
II.5.2. Other extraction methods
II.6. MIP used as sensors
II.7. Miscellaneous applications
II.8. Conclusions
References
PART II: EXPERIMENTAL
Chapter III: Synthesis and application of molecularly imprinted polymers for the selective extraction of organophosphorus pesticides from vegetable oil
III. Article 1
III.1. Abstract
III.2. Introduction
III.3. Materials and methods
III.3.1. Chemicals
III.3.2. Apparatus and analytical conditions
III.3.3. Synthesis of the MIPs
III.3.4. SPE procedure applied in pure media
III.3.5. Extraction of OPs from the vegetable oils
III.4. Results and discussions
III.4.1. Development of the LC-UV and LC-MS analyses
III.4.2. Screening of the synthesis conditions
III.4.3. Potential of the MIPs towards OPs
III.4.4. Study of the capacity of the MIP in pure media
III.4.5. Extraction of OPs from different oils
III.4.6. Optimization of the SPE procedure using almond oil
III.5. Conclusions
References
Chapter IV: Synthesis and characterization of molecularly imprinted silica for the selective extraction of organophosphorus pesticides from almond oil
IV. Article 2
IV.1. Abstract
IV.2. Introduction
IV.3. Materials and methods
IV.3.1. Chemicals
IV.3.2. Apparatus and analytical conditions
IV.3.3. Synthesis of molecularly imprinted silica sorbents
IV.3.4. Characterization of four MISs in pure medium
IV.3.5. MIS applied to almond oil extract
IV.4. Results and discussions
IV.4.1. Choice of conditions of synthesis of MISs
IV.4.2. Comparison of the synthesized MISs
IV.4.3. Repeatability of the extraction procedure
IV.4.4. Optimization of the extraction of OPs from almond oil
IV.4.5. Influence of a LLE step prior MISPE
IV.4.6. Sensitivity of the applied analytical method
IV.5. Conclusions
References
Chapter V: Reduction of matrix effects using molecularly imprinted silica applied to the extraction of organophosphorus pesticides from vegetable oils
V. Article 3
V.1. Abstract
V.2. Introduction
V.3. Materials and methods
V.3.1. Chemicals
V.3.2. Apparatus and analytical conditions
V.3.3. Synthesis of molecularly imprinted silica (MIS) sorbents
V.3.4. Solid phase extraction procedure in pure media
V.3.5. Extraction procedure of OPs from vegetable oils
V.4. Results and discussions
V.4.1. Repeatability of the MIS synthesis
V.4.2. Study of the capacity of the MIS in pure medium
V.4.3. Extraction of DMT and FSX from various vegetable oils
V.4.4. Evaluation of the matrix effects
V.4.5. Sensitivity of the applied method on the three vegetable oils
V.5. Conclusions
References
