Solid phase extraction

This extraction procedure is more and more applied, being much faster, cheaper, and more versatile than most classical techniques. Moreover, SPE procedures are easily automated using robotics, or can be coupled on-line with chromatographic techniques. The principle of retention (enrichment) is analogous to LC and is suitable for low, intermediate and high polarity solutes. Commonly, SPE media are based on normal phase, reversed phase or ion exchange chromatography. Giving an overview of the different SPE sorbents and their applications is not within the scope of this chapter. Hundreds of SPE cartridges are commercially available and the catalogues of the manufacturers describe their features and applications. The principles and practice of SPE have been described in books by Thurman and Mills [46], Fritz [47] and Simpson [48].

Present designs of SPE devices are the syringe-type cartridge, the Empore disks and the Empore disk cartridges. The syringe cartridge is the most common arrangement and cartridges ranging in mass of sorbent from 50 mg to 10 g are commercially available. Solvent flow through is typically done using a commercially available vacuum manifold. Permeability is no problem because sorbents are mostly based on irregular particles with a particle diameter between 30 and 60 mm. In the Empore disk format, 5-10 mm particles are intertwined with fine threads of Teflon giving a disk with a thickness of ca. 0.5 mm and a diameter in the range 47-70 mm. Manifolds are also commercially available for multiple sample extraction using Empore disks. The SPE disk allows rapid flow rates of sample and of solvents. One litre of water can be passed through an Empore disk in ca. 10min whereas with the syringe cartridge the same volume of water can take more than 1 h. Both approaches need a relatively large volume of desorbing liquid and further concentration is required. In the disk cartridge format, a disk of 10 mm diameter is placed in a cartridge between two polyethylene frits. Only small volumes of solvent are needed for solid phase stripping eliminating the need for an additional evaporation step hereby considerably reducing the risk of losses and contamination. The principle of SPE is illustrated with the determination of triazines in drinking water applying large volume injection capillary GC-MS.

For the enrichment of triazines from water a 10-mm/6mL ODS Empore disk cartridge was chosen. This means that enrichment is based on the reversed phase mechanism. For capillary GC, the operation can be divided in six steps namely wetting and stripping the sorbent, conditioning the sorbent, loading the sample, rinsing the sorbent to elute extraneous material, drying of the sorbent and finally elution of the analytes of interest. The disk was first rinsed with 0.5 mL of methanol (wetting and stripping), followed by 1 mL distilled water (conditioning). Twenty-five milliliter water spiked at the 1 ppb level with eleven triazines was passed through the cartridge at a flow of 50 mL/min (loading). Five milliliter distilled water was then passed through the cartridge to elute polar extraneous solutes (rinsing). After drying under vacuum for 10min (drying), desorption was performed directly in a 2-mL autosampler vial with 0.5 mL ethyl acetate. Forty microliter of the extract was directly injected, this means without further concentration, into the capillary GC system equipped with a programmed temperature vaporizing (PTV) injector operated in the solvent vent mode. The detection limits set by the European Community (0.1 ppb) could easily be reached with scan MS while ion-monitoring allowed determinations down to the ppt or ng/L level.

Several approaches allow the tuning of the selectivity of the SPE process. pH control is important to suppress ionization as illustrated for the analysis of veterinary drug residues in shrimps by Li et al. [49]. For ionic solutes the ion exchange mechanism can be utilized. Chlormequat (chlorocholine chloride) belongs to the family of quaternary ammonium salt pesticides. Chlormequat was extensively used for the control and management of terrestrial and aquatic vegetation since the mid-1950s but its use has been restricted because of its toxicity to man. Nevertheless, in the late 1990s concentrations as high as 7ppm were measured in pears. Because of the zero tolerance for pesticides in baby food, often containing pears, a robust and reliable analytical method down to the ppb level is needed. In the procedure presently used in our laboratory, chopped and homogenized pears are mixed with methanol containing the internal standard d4-chlormequat. The mixture is filtered and an aliquot is extracted by SPE on a strong cationic exchanger (BondElut SCX, Varian, CA, USA). After rinsing with a mixture of MeOH-H2O (1:1), chlormequat and the internal standard are eluted with MeOH-H2O (0.2 M NH4OAC) (4:6). Analysis is performed by LC/MS-MS on a cyanopropyl column with electrospray MS in the positive mode. Monitored ions are m/z 122 and daughters m/z 58:60.

Other selective SPE methods include the use of restricted access media (RAM), immunosorbents and molecularly imprinted polymers (MIP). These sorbents have mainly been applied for biological and environmental samples and food applications to date are rather limited. Pharmaceuticals in milk were determined by Souverain et al. [50] using RAM SPE and SPE with an immunosorbent was applied by Pichon et al. [51] for the determination of phenylurea pesticides in fruit juices.

DSPE in which the solid material added to the liquid sample is more and more applied. The main reason is the high-throughput possibilities DSPE offers. As outlined in Section 3.1.6, DSPE was intensively used in 1999 in our laboratory for the Belgian dioxin crisis [52,53]. Similar procedures are also used in the QuEChERS multi-residue method [1].

In recent years, regulatory agencies emphasize more and more the need for development and use of analytical methods able to determine in food products as many residues as possible out of the whole pattern of insecticides, fungicides and other compounds applied in agricultural practice. At present, single residue methods, i.e., the determination of one pesticide, e.g., chlormequat, or selective residue methods (MRMs), i.e., the determination of a relatively small number of chemically related compounds, e.g., N-methylcarbamate insecticides, are intensively applied for pesticide residue determinations in large number of samples whose pesticide treatment history is known. The use of single residue methods and sMRMs will definitely continue but the development and utilization of MRMs, i.e., the determination of as many pesticides as possible with preferably a single sample preparation method and one GC and one LC method both hyphenated to MS, recognizes a momentum.

In the past, a variety of capillary GC-MS-based MRMs have been developed. For example, working group 4 of the Technical Committee (TC 275) of the European Committee for Standardization (CEN) provides information on five MRMs for non-fatty foods (EN 1528:1996) [54]. All methods require extraction with organic solvents such as acetone [55-57], acetonitrile [58] and ethyl acetate [59] and with the exception of Ref. [59] they all require partitioning in a solvent mixture and further clean-up by column chromatography or GPC is advised. The multi-residue capillary GC-MS method that was applied in our laboratory until the application of stir bar sorptive extraction (SBSE - see Section 3.2.5) was a method used by the laboratories of the Dutch Inspectorate for Health Protection [59]. This method is similar to the Luke method [55] but the extraction procedures have been miniaturized to reduce solvent consumption. Other MRMs for the GC determination of residues in fruit and vegetables include the procedures described by Stan et al. [60,61] for 385 pesticides using extraction of the samples with acetone followed by salting-out and a multi-bed clean-up and the procedure described by Fillion et al. [62] for 251 pesticide and degradation products. The sample preparation of the later method comprises extraction with acetonitrile, a salting-out step, clean-up by SPE on octadecyl and on carbon/aminopropyl silica and a concentration step.

An MRM for animal tissues was recently described by Pang et al. [63] using gel permeation chromatographic clean-up followed by GC-MS and LC-tandem MS. The procedure on clean-up will be described in Section 4.4. Today, the QuEChERS method is by far the best developed MRM for a wide variety of pesticides in several common non-fatty and, since recently, fatty food matrices. The QuEChERS process, as initially developed for fruit and vegetables, is outlined in the flow diagram in Figure 9 [1]. DSPE is a key aspect of the method and the type of adsorbents, as well as the pH and polarity of the solvent, can be adjusted for differing matrices and difficult analytes.

Homogenization in Dry Ice 10 g Sample

10 mL Acetonitrile

Add IS solution

Dispersive SPE

PSAm + MgSO4

(Add 0.1% HAc and Analyte Protectants)

CGC-MS or LC-MS Figure 9 Flow diagram of the QuEChERS procedure.

In step 1, a 10-g sample obtained by homogenization in a powerful chopping device using dry ice is vigorously shaken with 10 mL acetonitrile for 1 min. Salts are added in step 2 to induce the phase separation. MgSO4 reduces the water phase while NaCl controls the ionic strength of the extracting solvent. After shaking for 1 min, an internal standard solution is added (step 3). Step 4 is DSPE and typically 50 mg adsorbent is added to a 1-mL aliquot of the acetonitrile phase. The sample is centrifuged, and acetic acid and GC protectors (sorbitol, gulonolactone and ethyl glycerol) are added before injection. For LC analysis, GC protectors are not needed.

In recent years, several modifications of the QuEChERS method have been made for specific applications and matrices. Different extraction solvents can be applied and DSPE, commonly performed with PSAm (primary secondary amine) which mainly removes sugars and fatty acids, can be complemented with graphitized carbon black (remove pigments and sterols) and/or octadecylsilica (remove lipids). An important note is that "QuEChERS-kits" have been assembled commercially by Sigma Aldrich/Supelco (Bellefonte, Pennsylvania), Restek (Bellefonte, Pennsylvania) and United Chemical Technologies (Bristol, Pennsylvania).

The QuEChERS technique has proven successful for the determination of pesticides in a variety of fruit and vegetables, e.g., [64-66] and also recently in fatty matrices such as olive oil [67], and milk and eggs [68]. The technique was also used to analyse acrylamide in various food matrices [69] and ivermectin in salmon and antioxidants in pet food [70].

In the European Community (EU), QuEChERS round robin tests have been conducted with over 100 laboratories participating. These tests showed that the method can be transferred among laboratories with success [71]. In the US, the QuEChERS method has been adopted First Action as AOAC International Official Method 2007.01.

The QuEChERS method is an excellent method to determine pesticides at the maximum residue limits (MRLs) set by regulatory authorities for human consumption. However, for very low MRLs such as set by the European baby food directive (10ng/g), the method is inadequate. In a recent article, Lehotay et al. [72] described the analysis of a mixture of vegetables and fruit spiked with 36 organophosphorus and organochlorine pesticides at the 10-ng/g level. With the original QuEChERS method only 12 of the 36 pesticides could be determined at the 5-ng/g level. A new technique called Solvent in silicone tube extraction (SISTEx) was introduced. This is a combination of sorptive extraction and solvent collection and similar in its operation as membrane-assisted extraction. The number of elucidated pesticides increased to 24 with a 44-fold improvement in detection threshold.

0 0

Post a comment