SEARCH

SEARCH BY CITATION

Keywords:

  • 2-monochloropropanediol;
  • 3-monochloropropanediol;
  • bound;
  • cyclohexanone;
  • derivatization;
  • foods;
  • free;
  • gas chromatography-mass spectrometry

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Conflict of Interest
  8. References

Assessment of contamination of foods with monochloropropanediols (MCPD) and subsequent mitigation of their formation is an important current issue of a global food security. Methods for the determination of 2- or 3-MCPD in foods at low μg/kg levels require analyte derivatization prior to gas chromatography-mass spectrometry (GC-MS) determination. All existing methods suffer from various drawbacks associated with current derivatization schemes. We have developed a new derivatization scheme, which uses cyclohexanone as a derivatization agent and a sulfonated polymer as a solid-phase acidic catalyst. This derivatization uses a readily available derivatization reagent and does not require any postderivatization workup. The respective 2-MCPD 1,3-dioxane and 3-MCPD 1,3-dioxolane derivatives are stable with storage, produce characteristic molecular ions, and chromatograph well on nonpolar GC columns. This derivatization procedure was applied to the analysis of free 2- and 3-MCPD, bound 2- or 3-MCPD (in the form of fatty acid esters after acidic hydrolysis), and also to simultaneous analysis of free and bound forms. The method was tested on soy sauce, commercial palm oil, palm oil noodles from an instant soup, and olive oil, which was spiked with bound 2- and/or 3-MCPD. The results obtained using derivatization with cyclohexanone agreed with the data obtained using traditional heptafluorobutyryl imidazole derivatization. Additionally, data for soy sauce and palm oil matrices obtained through interlaboratory testing programs had z-scores <1. The method detection limit is 1–3 μg/kg for free 2- and 3-MCPD (sample weight dependent) and 100 μg/kg per fat for bound 2- and 3-MCPD.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Conflict of Interest
  8. References

One of several chemically related contaminants, 3-chloropropane-1,2-diol or 3-monochloropropanediol (3-MCPD) is usually referred to as “chloropropanols.” It was first detected in 1980 by Velisek in acid-hydrolyzed vegetable protein (HVP), which might be present in soy sauce and other similar flavoring materials (Velišek 2009).

A book chapter by Hamlet and Sadd (2009) provides useful background on various aspects of chloropropanols and their esters in food prior to 2009. A recent issue (#3, vol. 113) of European Journal of Lipid Science and Technology provides more recent perspectives on MCPDs and glycidol (e.g., Kuhlmann 2011). Generally considered a food processing-induced toxicant, that is, a toxicant that forms during food processing, 3-MCPD is a nongenotoxic carcinogen and in the EU the tolerable daily intake was set at 2 μg/kg bw (European Commission Scientific Committee on Food 2001). Based on the available occurrence data for 3-MCPD, the maximum levels in foodstuffs in some countries were set from 20 μg/kg to 1 mg/kg (EC 2006; Hamlet and Sadd 2009; Health Canada 2012). These limits were not designed to account for bound 3-MCPD.

It has since been established that 3-MCPD can be found in foods other than those containing soy sauce or HVP, such as cereal or meat products (Hamlet et al. 2002; Breitling-Utzmann et al. 2003; Baer et al. 2010). Recent studies have also identified relatively high levels of bound 3-MCPD (present in the form of 3-MCPD fatty acid esters) in other food commodities such as fats and oils and also in products containing fats and oils such as baby formulas (Hamlet and Sadd 2004; Svejkovska et al. 2004; Karsulinova et al. 2007; Seefelder et al. 2008; Zelinkova et al. 2009). These new findings could result in higher intake of chloropropanols than the original estimates would indicate due to the possibility of cleavage of fatty acids by lipases in vivo (Hamlet and Sadd 2004; Seefelder et al. 2008) Therefore, a need exists for a robust analytical method to conduct surveys of free and bound monochloropropanols.

Most studies have focused on the presence of 3-MCPD and its esters, as it is the predominant isomer, while 2-chloropropane-1,3-diol (2-MCPD) has received limited attention despite indications that it might be present in some foods at levels comparable to 3-MCPD (Robert et al. 2004). A recent publication partially addressed this gap (Hamlet and Asuncion 2011) by providing data on levels of both MCPD isomers in several foodstuffs.

The toxicological significance of 2-MCPD is not known due to a lack of data (Schilter et al. 2011); however, the precautionary principle warrants an investigation of the presence of 2-MCPD in foods.

The lack of data for bound 2-MCPD (Larsen 2009) might be due to the use of procedures that use methanolic sodium methoxide to liberate free MCPD from its esters. Rapid dechlorination during the hydrolysis step (Zelinkova et al. 2009) might interfere with the detection of smaller quantities of these analytes. An alternative explanation for the lack of data would be that bound 2-MCPD has simply been overlooked, as the free compound is usually a relatively minor isomer in some foods. (Meierhans et al. 1998).

In principle, it is feasible to determine fatty acid esters of MCPDs directly by high-temperature gas chromatography (GC) with mass spectrometry (MS) detection, but such an approach requires thorough removal of matrix components prior to analysis (Zelinkova et al. 2008). Additionally, multiple standards are required, as MCPDs could be present in foods as esters of a combination of different fatty acids.

Recently, work utilizing a direct liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis of fatty acid esters of 3-MCPD was published (Moravcova et al. 2012) and it too required removal of matrix components (triglycerols) due to analyte signal suppression. The LC-MS/MS time of flight (TOF) methods for direct determination of 3-MCPD esters, with (Hori et al. 2012) or without (Haines et al. 2011) sample pretreatment, both use sodiated adducts for quantification. The resulting contamination of the mass spectrometer by nonvolatile salt makes them less robust. So far no direct methods for the determination of 2-MCPD esters have been published.

The main limitation of these direct methods (and other similar “targeted” analyses) is the fact that while it appears that refined edible oils contain only a few major fatty acid MCPD esters (Crews 2012) there is a theoretical possibility that other MCPD esters containing different fatty acid chains could be present. However, these would be not detected due to their removal during matrix clean up or because of their different chemical structure.

Because bound MCPDs exist as derivatives of mono- or diglycerides of various fatty acids, these fatty acids are usually cleaved from a glycerol moiety and the resulting free MCPDs are derivatized for subsequent determination as “MCPD equivalents.” The most widely used cleavage procedures employ sulfuric acid/methanol (Divinova et al. 2004) or sodium methoxide/methanol reagent (Weisshaar 2008), which results in transesterification of fatty acids to their respective methyl esters.

All existing methods for the determination of free 2- or 3-monochloropropanediols (2- or 3-MCPD) in foods at low μg/kg levels require analyte derivatization prior to GC-MS determination. The most common derivatization groups are heptafluorobutyryl (introduced by heptafluorobutyryl imidazole [HFBI] or heptafluorobutyric anhydride), cyclic boronate (introduced by phenylboronic acid [PBA]), and cyclic ketals (introduced by acetone, 3-pentanone or 4-heptanone) (Hamlet and Sadd 2009). These methods suffer from various drawbacks associated with current derivatization schemes. Derivatization with HFBI requires strictly anhydrous conditions (Brereton et al. 2001), an excess of PBA is deleterious to the chromatographic system or has to be removed after the reaction (Weisshaar 2008), and aliphatic ketone derivatives lack characteristic molecular ions of the derivatives (Dayrit and Ninonuevo 2004). This last derivatization is usually accomplished using p-toluenesulfonic acid as a catalyst which must be removed from the reaction medium through additional steps.

Therefore, we set out to test the possibility of reducing of those additional steps by utilizing a solid-phase catalyst (sulfonated polymeric resin [Amberlyst 15] and sulfonated fluoropolymer resin [Nafion] on a silica support), and to improve the quality of the spectral data by derivatization with cyclohexanone, a cyclic ketone which we expected to be more resistant to MS fragmentation than aliphatic ketones.

After our study was completed, a paper was published which described, for the first time, the simultaneous determination of free and bound 3-MCPD in different foodstuffs (Küsters et al. 2010). That procedure used sodium methoxide for fatty acid cleavage from 3-MCPD esters and subsequent derivatization of liberated free 3-MCPD with phenylboronic acid.

The presence of glycidyl esters of fatty acids was confirmed while investigating the occurrence of fatty acid esters of 3-MCPD in foods (Weisshaar 2008; Haines et al. 2011). The interconversion between glycidol and MCPD which takes place during the analysis of samples by the “DGF” (German Society for Fat Science) method (sodium methoxide treatment) might, however, complicate interpretation of results (Kaze et al. 2011).

Such interconversion is avoided in a recently published methods of simultaneous monitoring of esters of glycidol and 2- and 3-MCPD, which uses acidic hydrolysis followed by bromide salts (Ermacora and Hrncirik 2012) and enzymatic hydrolysis procedure coupled to PBA derivatization which was used for determination of ester-bound 2- and 3-MCPD (Chung and Chan 2012).

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Conflict of Interest
  8. References

Reagents

The reagents used were as follows: 3-MCPD 98% was from Alfa Aesar; 2-MCPD 98%, 2-MCPD-d5 98%, isotopic purity >99 atom % D, 3-MCPD-d5 dipalmitate 98%, isotopic purity 98 atom % D were supplied by Toronto Research Chemicals Inc. (Toronto, Ontario, Canada) and 3-MCPD-d5 98%, isotopic purity 99 atom % D was from CDN Isotopes (Pointe-Claire, Quebec, Canada). Amberlyst 15 and Nafion SAC-13, 10-20% on a silica support and cyclohexanone 99.8% were from Aldrich-Sigma (St. Louis, MO). Sulpfric acid 98%+ (trace metal use, A-510-500) was from Fisher Scientific while anhydrous sodium sulfate was from EMD (Gibbstown, NJ). Sodium sulfate was muffled at 650°C for 12 h before use. All other reagents were of analytical grade. Water was obtained from a Barnstead NANOpure Diamond purification ystem.

Standards

All stock and spiking deuterated standard solutions of 2- and 3-MCPD were prepared in ethyl acetate and stored at −18°C. Calibration solutions were prepared in 4 mL vials with septa and contained 2, 5, 25, 100, 500, and 1000 ng of 2- and 3-MCPD and 100 ng 2-MCPD-d5 and 3-MCPD-d5 in each vial. The volume of ethyl acetate used for dispensing standards was in the range of 20–120 μL, 100 μL of cyclohexanone was added to each vial, and the total volume of the organic phase was adjusted to 1.0 mL using isooctane. Sodium sulfate (0.2 g) and Nafion/silica (10 mg) was added, vials were capped and heated in a heater block at 45°C for 1 h. After cooling to room temperature, the organic phase was pipetted in to a 2 mL vial for GC/MS analysis.

Food samples

Extra virgin olive oil and instant oriental noodles soup containing palm oil were collected from retail outlets in Ottawa, Ontario, Canada, while refined palm oil was supplied by Dyets Inc. (Bethlehem, PA). Reference materials: Soy sauce material # 2626 was obtained from Food Analysis Performance Assessment Scheme (FAPAS), The Food and Environment Research Agency Sand Hutton, York, U.K. Samples of naturally contaminated palm oil, olive oil spiked with bound 3-MCPD, and a sodium chloride standard of free 3-MCPD were a gift of Dr. L. Karasek, Institute for Reference Materials and Measurements (IRMM), Geel, Belgium.

Typical sample preparation

Our method consists of four distinct steps: (1) separation of free and bound MCPD by partitioning between water and the organic phase, (2) extraction of free MCPD from the aqueous phase, (3) hydrolysis of oil and conversion of bound MCPD into free species followed by a subsequent extraction of free MCPD, and (4) derivatization with cyclohexanone which is done independently for both steps 2 and 3.

Separation of free and bound MCPD

Three grams of a homogenized food sample, 19 mL of water (solid samples were further homogenized in situ with a small Polytron probe), and 15 mL of 10% ethyl ether/hexane mixture (v/v) were added to a 50 mL fluorinated ethylene propylene (FEP) centrifuge tube (Nalgene, Oak Ridge type), and spiked with 2-MCPD-d5 and 3-MCPD-d5 (for the quantification of free MCPDs) at the level of 100 μg/kg. The mixture was shaken for 30 min on a horizontal shaker and centrifuged at 15,000g at room temperature for 10 min. The organic phase was transferred to another 50 mL centrifuge tube, extraction was repeated one more time and organic extracts were combined. (The tube containing the sample and the aqueous phase was saved for the determination of free MCPD.)

A quantity of 10 mL of water was added to the combined organic phases, the mixture was shaken for 30 min on a horizontal shaker and centrifuged at 15,000g at room temperature for 10 min. The aqueous layer was removed with a pipette, and 1 g of sodium sulfate was added. After 30 min, the dried organic phase was transferred to a 100 mL round-bottom (r.b.) flask, the centrifuge tube was rinsed twice with diethyl ether and the organic phase was evaporated using a rotary evaporator. The oily residue was transferred to a 10 mL vial using a small amount of ether, the bulk of the ether was evaporated in the fume hood and the remaining solvent was removed in a vacuum desiccator. The oily residue was used for the determination of bound MCPD.

In the first step, water, not sodium chloride solution, was used as an extraction medium to prevent the formation of possible artifacts of chloropropanols. These artifacts might occur through a reaction of chloride ion with other compounds present in foods, such as glycidol esters (Weisshaar and Perz 2010), thus forming MCPD esters. Also, back extraction of 10% ether/hexane with water was found to be necessary to remove any free MCPD (about 5% of the amount present in the aqueous phase) which might partition into the organic layer. In this extraction step, due to the combination of solvent and high centrifugation speeds, the formation of emulsions was not observed.

Free MCPD

Four grams of NaCl was added to a 50 mL FEP centrifuge tube containing a sample (the aqueous phase from the previous step involving extraction of bound MCPD), the mixture was shaken for 30 min on a horizontal shaker and centrifuged at 15,000g at room temperature for 10 min. Eight grams of Extrelut NT was placed in a new FEP centrifuge tube and 7.0 g of the supernatant (containing ~0.8 g of an original food sample) from the centrifugation was added. The tube was capped, shaken by hand for 1 min and left to stand for 30 min. The extraction column (30 × 2.5 cm inside diameter [ID], with a Grade C frit) was prepared as follows: 1 cm layer of sodium sulfate, followed by 28 g of a mixture of sodium/magnesium sulfate (1:6, w/w), sample adsorbed on Extrelut NT and 1 cm layer of sodium sulfate. The column was tapped to pack the layers. The column was eluted with the stopcock open with 70 mL of 10% ethyl ether in hexane followed by 15 mL of 95% ethyl ether in hexane and that eluate was discarded. The column was further eluted dropwise with 150 mL of 95% ethyl ether in hexane and the eluate collected in a 250 mL r.b. flask. The solvent was evaporated to a volume of 2 mL (180 mmHg, 35°C) and transferred (with two 1 mL ether rinses of the r.b. flask) to a 10-mL vial. The ether was removed under a stream of N2 at 35°C to near dryness, 0.1 mL of cyclohexanone followed by 0.9 mL of isooctane was added to the vial and sample was derivatized as for standards.

Step 2 used diatomaceous material, Extrelut NT for column liquid/liquid extraction (Brereton et al. 2001). The solvent was 95% diethyl ether in hexane which extracts MCPDs as efficiently as ether, but with less water residue (Xu et al. 2006). This step was optimized by adding a layer of a mixture of sodium and magnesium sulfates as drying agents directly into the column. Pure magnesium sulfate could not be used due to clogging of the column.

Bound MCPD

Hundred milligrams of the oily residue from step 1 was placed in a 10-mL screw top vial, 1 mL of tetrahydrofuran was added and 2-MCPD-d5 and 3-MCPD-d5 was spiked at the level of 1000 μg/kg of oil. A volume of 1.8 mL of sulfuric acid solution in methanol (3.1 g of concentrated acid per 100 mL of methanol) was added and the vial was heated at 40°C in a heater block for 16 h. After cooling to room temperature, 0.5 mL of a saturated sodium hydrogen carbonate was added and the vial was vortexed for 2 min. The 10-mL vial was placed in a custom cylindrical glass sleeve (~15 × 2.5 cm ID) with a ground glass joint and organic solvents were removed by a rotary evaporator (25 mmHg, 55°C bath).

To the vial, 2 mL of hexane was added, vortexed for 2 min, phases were allowed to separate and the hexane layer was removed and discarded using a pipette. The extraction with hexane was repeated one more time. To the vial, 0.8 g Extrelut NT was added, the vial was capped, shaken by hand for 1 min, and left to stand for 30 min.

The extraction column (6 mL blank polypropylene reservoir [Varian], with a polypropylene frit) was prepared as follows: a layer of 2 g of a mixture of sodium/magnesium sulfate (1:6, w/w), followed by the sample, which had been adsorbed on Extrelut NT and 0.2 g layer of sodium sulfate. The column was tapped to pack the layers. The column was eluted in a SPE manifold (Supelco) with the valve fully open with 6 mL of 10% ethyl ether in hexane mixture (v/v), followed by a 100% diethyl ether. The first 5 mL of eluate was discarded, the elution speed was changed to 1 drop per second and elution was continued until 12 mL had been collected.

The ether was removed under a stream of N2 at 35°C to near dryness, 0.1 mL of cyclohexanone followed by 0.9 mL of isooctane was added to the vial and the sample was derivatized as for standards.

In step 3, an acidic hydrolysis was employed (Divinova et al. 2004) as we had confirmed earlier reports (Zelinkova et al. 2009) of significant losses of 3-MCPD during the sodium methoxide/methanol mediated transesterification reaction (Weisshaar 2008). In blank samples processed by us (data not shown), the average loss of 3-MCPD was 76%. Under such conditions, 2-MCPD was more stable with a loss of 42%. A similar order of stabilities of 2- and 3-MCPD under alkaline conditions was reported earlier (Doležal and Velišek 1992, 1995).

GC-MS operating conditions

Carrier gas: helium, constant flow 1.2 mL/min, oven temperature profile: initial, 60°C (1 min), rate (1), 15°C/min to 120°C, rate (2), 6°C/min to 150 °C, hold 1 min, rate (3), 50°C/min to 300°C, hold 6 min; inj. volume: 1 μL; inj. temperature 280°C, injection mode: splitless, split time: 1 min; ionization mode: 70 eV EI+; source temperature 230°C; scan mode: selected ion monitoring (8–10 min); ions (m/z): for 2- and 3-MCPD 147, 149, 161, 190, and for 2- and 3-MCPD-d5 152, 195. For quantitation the following signals were used (m/z): 2- and 3-MCPD 147, 2- and 3-MCPD-d5 152. Dwell time was 50 msec for each ion.

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Conflict of Interest
  8. References

Our method was tested on replicates of (a) soy sauce, (b) commercial palm oil, (c) palm oil noodles from an instant soup, and (d) olive oil which was spiked with bound 2- and/or 3-MCPD. Samples a, b, and c included FAPAS and/or IRMM proficiency material. Results were compared with analyses of samples that were extracted in the same way, but derivatized with heptafluorobutyrylimidazole (HFBI).

The full method, steps 1 through 3 can be used for the simultaneous determination of free 2- and 3-MCPD and 2- and 3-MCPD esters. For the determination of only free MCPDs step 2 followed by derivatization can be used. Similarly for the determination of only bound MCPDs step 3 followed by derivatization can be used.

Use of these single steps is appropriate when only one form of MCPD is present in a given matrix. For example, soya sauce contains only free MCPDs while vegetable oil samples are likely to contain only bound MCPDs. Many other matrices can contain both forms of MCPDs (Hamlet and Sadd 2009).

To avoid the laborious manipulations necessary when using p-toluenesulfonic acid as a catalyst in the formation of cyclic cyclohexanone ketals of 2- and 3-MCPD, two solid-phase catalysts were tested (sulfonated polymeric resin [Amberlyst 15] and sulfonated fluoropolymer resin [Nafion] on a silica support). Under very mild condition, 45°C for 1 h, the formation of expected derivatives (cyclic cyclohexanone ketals of 2- and 3-MCPD) was observed for both catalysts. Increasing the temperature to 55°C did not increase the yield of the derivatization for both catalysts, Amberlyst or Nafion. Nafion afforded a marginally cleaner chromatogram and as it was easier to manipulate than Amberlyst, it was retained as the catalyst of choice for further study. The formation of derivatives proceeded when no drying agents were present in the reaction mixture. However, to make the derivatization more robust, a small quantity of sodium sulfate was added during derivatization. We tested this approach by spiking a calibration solution with 10 μL of water prior to heating in a heating block. The presence of extra water did not reduce the efficiency of analyte derivatization.

The mass spectra representing the 1,3-dioxane derivates of 3-MCPD-d5 and 3-MCPD and 1,3-dioxolanes of 2-MCPD-d5 and 2-MCPD are shown in the Figure 1. As expected, the spectra contain molecular ions (m/z = 195 for deuterated and m/z = 190 for native compounds), however, the intensity of these ions is only about 10% of the base peak. The spectra of all compounds are very similar as fragmentation takes place predominantly in the cyclohexanone ring. The base peak at m/z = 152 for deuterated and m/z = 147 for native compounds, arises through loss of a C3H7 radical and the formation of a more stable allylic 1,3-dioxane or 1,3-dioxolane cation. Spectra of native compounds also have useful diagnostic ions at m/z = 149 and 161.

image

Figure 1. Mass spectra of cyclic cyclohexanone ketals of (A) 3-MCPD-d5, (B) 3-MCPD, (C) 2-MCPD-d5, and (D) 2-MCPD (1 ppm of each analyte derivatized as for standards). MCPD, monochloropropanediols

Download figure to PowerPoint

No interferences from phthalate compounds (intense m/z 149) were observed as they do not elute in the time window used for collecting the analytical data. Similarly, interferences which could originate from siloxanes in the channel m/z 147 were not observed.

The derivatives are stable on storage as a decrease in analyte response of only about 10% was observed after storing of derivatized sample extracts for 2 months at −15°C. The calibration curves (concentration of 2- and 3-MCPD 2, 5, 25, 100, 500, 1000 μg/L; concentration of 2- and 3-MCPD-d5 100 μg/L) were linear with r2 higher than 0.995. The injection of 2 μg/L mixed standard of 2- and 3-MCPD gave signal to noise (S/N), peak-to-peak, between 46 and 12 for quantifier (147) and qualifier ions (149, 190). A typical chromatogram of a 5 μg/L standard is shown in Figure 2.

image

Figure 2. GC-MS chromatograms of a mixed standard of 3-MCPD and 2-MCPD 5 μg/L each (2- and 3-MCPD-d5 100 μg/L). GC-MS, gas chromatography-mass spectrometry; MCPD, monochloropropanediols.

Download figure to PowerPoint

Method spikes were done using cold-pressed virgin olive oil as this type of oil was expected to have very low levels of chloropropanols (Zelinkova et al. 2006). The oil was spiked with free native 2- and 3-MCPD at levels of 500 μg/kg The average recoveries for 2- and 3-MCPD were 95% and 100% with respective relative standard deviations of 0.4% and 0.8%.

The method detection limit was calculated using the above-mentioned FAPAS soy sauce samples and fortified olive oil samples, not the water standard. Using signal to noise (S/N) criterion 3:1 (height/average noise), the limit of detection (LOD) was calculated to be 1 μg/kg for free 2- and 3-MCPD. For bound 2- and 3-MCPD the LOD was 100 μg/kg (in the oil matrix).

The accuracy of the method for free 3-MCPD was tested on a FAPAS soy sauce material # 2626 with an assigned concentration of 3-MCPD at 22.2 μg/kg and a satisfactory range of 12.4–32.0 μg/kg (no suitable reference material for determination of bound MCPDs was available to us). Analysis of replicates (n = 4) of FAPAS gave an average 3-MCPD concentration of 18.9 μg/kg with a standard deviation of 0.1 μg/kg. The method was applied to the analysis of palm oil noodles containing 14.9% fat (from a retail oriental soup package) and commercially available palm oil.

A typical chromatogram of cyclohexanone derivatives of 2- and 3-MCPD (from a sample of a oil from dry oriental noodles, which contained bound chloropropanols) is shown in Figure 3. Figure 4 depicts a chromatogram of cyclohexanone derivatives of free 3-MCPD in a FAPAS soy sauce (2-MCPD was not detected).

image

Figure 3. GC-MS chromatograms of a palm oil (from oriental style noodles sample) containing: 3-MCPD, 1420 μg/kg and 2-MCPD 1280 μg/kg (2- and 3-MCPD-d5 1000 μg/kg). GC-MS, gas chromatography-mass spectrometry; MCPD, monochloropropanediols.

Download figure to PowerPoint

image

Figure 4. GC-MS chromatograms of a soy sauce sample containing: 3-MCPD, 18.9 μg/kg, 2-MCPD <1 μg/kg (2- and 3-MCPD-d5 100 μg/kg). GC-MS, gas chromatography-mass spectrometry; MCPD, monochloropropanediols.

Download figure to PowerPoint

The data obtained by derivatization with cyclohexanone were compared to our earlier (unpublished) data obtained by derivatization with HFBI, Table 1. As seen from the table, derivatization with cyclohexanone produced very similar results when compared to HFBI and with low overall relative standard deviations.

Table 1. Comparison of concentrations of 2- and 3-MCPD in three matrices obtained by derivatization with cyclohexanone and HFBI
DerivatizationMatrix3-MCPD (μg/kg)Standard deviation2-MCPD (μg/kg)Standard deviationNumber of determinations
  1. MCPD, monochloropropanediols; HFBI, heptafluorobutyryl imidazole; FAPAS, Food Analysis Performance Assessment Scheme.

  2. a

    FAPAS satisfactory range: 12.4–32.0 μg/kg 3-MCPD. This matrix was analyzed only for free MCPDs.

  3. b

    This matrix was analyzed only for bound MCPDs.

  4. c

    This matrix was analyzed for both free and bound MCPDs.

HFBISoy saucea20.10.1ND 4
Cyclohexanone18.90.1ND 4
HFBIPalm oilb1803885293
Cyclohexanone178538100873
HFBIPalm oil noodlesc171623914161154
Cyclohexanone1419321282526

Additionally, we have analyzed three matrices from the proficiency test conducted by IRMM in 2009 (naturally contaminated palm oil, spiked olive oil and 3-MCPD standard in 20% NaCl). We obtained a very good agreement between derivatization with cyclohexanone and HFBI, Table 2. It is worth noting that 2-MCPD was present in the naturally contaminated palm oil used for this proficiency study at the level of about 4 ppm.

Table 2. Comparison of concentrations of 3-MCPD in three matricesa obtained by derivatization with cyclohexanone and HFBI
Sample3-MCPDUnitsz-scoreRelative bias (%)Derivatization method
  1. MCPD, monochloropropanediols; HFBI, heptafluorobutyryl imidazole; IRMM, Institute for Reference Materials and Measurements.

  2. a

    Material obtained from IRMM proficiency test. Samples were analyzed only for free (standard solution) or bound MCPDs (palm oil and spiked olive oil) (http://irmm.jrc.ec.europa.eu/html/interlaboratory_comparisons/3_MCPD/EUR_24356_EN_3-MPCD_esters_in_edible_oil.pdf).

Palm oil7.38mg/kg−0.8 HFBI
8.33mg/kg−0.2 Cyclohexanone
Spiked olive oil3.92mg/kg−0.7 HFBI
4.56mg/kg0.0 Cyclohexanone
Standard solution432μg/L 3.6HFBI
427μg/L 2.4Cyclohexanone

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Conflict of Interest
  8. References

The new derivatization protocol using cyclohexanone/Nafion is simple, rapid, robust, and produces derivatives with good spectral data. It is applicable to determination of both free and bound 2- and 3-MCPD with corresponding LODs of 1–3 μg/kg (dependent on sample amount) and 100 μg/kg.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Conflict of Interest
  8. References
  • Baer, I., B. de la Calle, and P. Taylor. 2010. 3-MCPD in food other than soy sauce or hydrolyzed vegetable protein (HVP). Anal. Bioanal. Chem. 396:443456.
  • Breitling-Utzmann, C. M., H. Kobler, D. Herbolzheimer, and A. Maier. 2003. 3-MCPD – occurence in bread crust and various food groups as well as formation in toast. Dtsch. Lebensm.-Rundsch. 99:280285.
  • Brereton, P., J. Kelly, C. Crews, S. Honour, and R. Wood. 2001. Determination of 3-chloro-1,2-propanediol in foods and food ingredients by gas chromatography with mass spectrometric detection: collaborative study. J. AOAC Int. 84:455465.
  • Chung, S. W. C., and B. T. P. Chan. 2012. Simultaneous determination of 2- and 3-monochloropropan-1,3-diol esters in foods by enzymatic hydrolysis and GC-MS detection. Chromatographia 75:10491056.
  • Crews, C. 2012. MCPD and glycidyl esters in food products. Report/White Paper 2012. Commissioned by the Risk Assessment of Chemicals in Food Task Force and the Process-related Compounds & Natural Toxins Task Force, ILSI Europe Report Series 2012:124.
  • Dayrit, F. M., and M. R. Ninonuevo. 2004. Development of an analytical method for 3-monochloropropane-1,2-diol in soy sauce using 4-heptanone as derivatizating agent. Food Addit. Contam. 21:204209.
  • Divinova, V., B. Svejkovska, M. Dolezal, and J. Velisek. 2004. Determination of free and bound 3-chloropropane-1,2-diol by gas chromatography with mass spectrometric detection using deuterated 3-chloropropane-1,2-diol as internal standard. Czech J. Food Sci. 22:182189.
  • Doležal, M., and J. Velišek. 1992. Kinetics of 3-chloro-1,2-propanediol degradation in model systems in Proceedings of Chemical Reactions in Foods II Prague, Czech Republic, 24–26 September, pp. 297302.
  • Doležal, M., and J. Velišek. 1995. Kinetics of 2-chloro-1,3-propanediol degradation in model systems and in protein hydrolysates. Potravinarske Vedy 13:8591.
  • EC. 2006. Pp. 524 in Commission regulation No. 1881/2006 setting maximum levels for certain contaminants in foodstuffs. Official Journal of the European Union, L 364/5. Available at: http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2006:364:0005:0024:EN:PDF.
  • Ermacora, A., and K. Hrncirik. 2012. Novel method for simultaneous monitoring of 2-MCPD, 3-MCPD and glycidyl esters in oils and fats. J. Am. Oil Chem. Soc. 30:18.
  • European Commission Scientific Committee on Food. 2001. Opinion of the scientific committee on food on 3-monochloro-propane-1,2-diol (3-MCPD) updating the SCF opinion of 1994. Adopted on 30 May 2001. European Commission Scientific Committee on Food, Brussels, Belgium.
  • Haines, T. D., K. J. Adlaf, R. M. Pierceall, I. Lee, P. Venkitasubramanian, and M. W. Collison. 2011. Direct determination of MCPD fatty acid esters and glycidyl fatty acid esters in vegetable oils by LC–TOFMS. J. Am. Oil Chem. Soc. 88:114.
  • Hamlet, C. G., and L. Asuncion. 2011. Pp. 345–355 in Single-laboratory validation of a method to quantify bound 2-chloropropane-1,3-diol and 3-chloropropane-1,2-diol in foodstuffs using acid catalysed transesterification, HFBI derivatisation and GC/MS detection. Eur. J. Lipid Sci. 113:345355.
  • Hamlet, C. G., and P. A. Sadd. 2004. Chloropropanols and their esters in cereal products. Czech J. Food Sci. 22:259262.
  • Hamlet, C. G., and P. A. Sadd. 2009. Chloropropanols and chloroesters. Pp. 175214 in R. H. Stadler and D. R. Lineback, eds. Process-induced food toxicants. John Wiley and Sons, Inc., Hoboken, NJ.
  • Hamlet, C. G., P. A. Sadd, C. Crews, J. Velisek, and D. E. Baxter. 2002. Occurrence of 3-chloropropane-1,2-diol (3-MCPD) and related compounds in foods: a review. Food Addit. Contam. 19:619631.
  • Health Canada. 2012. Canadian standards (maximum levels) for various chemical contaminants in foods. Available at: http://www.hc-sc.gc.ca/fn-an/securit/chem-chim/contaminants-guidelines-directives-eng.php.
  • Hori, K., N. Koriyama, H. Omori, M. Kuriyama, T. Arishima, and K. Tsumura. 2012. 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. 48:204208.
  • Karsulinova, L., B. Folprechtova, M. Dolezal, J. Dostalova, and J. Velisek. 2007. Analysis of the lipid fraction of coffee creamers, cream aerosols, and bouillon cubes for their health risk associated constituents. Czech J. Food Sci. 25:257264.
  • Kaze, N., H. Sato, H. Yamamoto, and Y. Watanabe. 2011. Bidirectional conversion between 3-monochloro-1,2-propanediol and glycidol in course of the procedure of DGF standard methods. J. Am. Oil Chem. Soc. 88:11431151.
  • Kuhlmann, J. 2011. Determination of bound 2,3-epoxy-1-propanol (glycidol) and bound monochloropropanediol (MCPD) in refined oils. Eur. J. Lipid Sci. Technol. 113:335344.
  • Küsters, M., U. Bimber, A. Ossenbrüggen, S. Reeser, R. Gallitzendörfer, and M. Gerhartz. 2010. Rapid and simple micromethod for the simultaneous determination of 3-MCPD and 3-MCPD esters in different foodstuffs. J. Agric. Food Chem. 58:65706577.
  • Larsen, J. C. 2009. 3-MCPD esters in food products. ILSI Europe, Brussels, Belgium.
  • Meierhans, D. C., S. Bruehlmann, J. Meili, and C. Taeschler. 1998. Sensitive method for the determination of 3-chloropropane-1,2-diol and 2-chloropropane-1,3-diol by capillary gas chromatography with mass spectrometric detection. J. Chromatogr. A 802:325333.
  • Moravcova, E., L. Vaclavik, O. Lacina, V. Hrbek, K. Riddellova, and J. Hajšlova. 2012. Novel approaches to analysis of 3-chloropropane1,2-diol esters in vegetable oils. Anal. Bioanal. Chem. 402:28712883.
  • Robert, M.-C., J.-M. Oberson, and R. H. Stadler. 2004. Model studies on the formation of monochloropropanediols in the presence of lipase. J. Agric. Food Chem. 52:51025108.
  • Schilter, B., G. Scholz, and W. Seefelder. 2011. Fatty acid esters of chloropropanols and related compounds in food: toxicological aspects. Eur. J. Lipid Sci. Technol. 113:309311.
  • Seefelder, W., N. Varga, A. Studer, G. Williamson, F. P. Scanlan, and R. H. Stadler. 2008. Esters of 3-chloro-1,2-propanediol (3-MCPD) in vegetable oils: significance in the formation of 3-MCPD. Food Addit. Contam. 25:391400.
  • Svejkovska, B., O. Novotny, V. Divinova, Z. Reblova, M. Dolezal, and J. Velisek. 2004. Esters of 3-chloropropane-1,2-diol in foodstuffs. Czech J. Food Sci. 22:190196.
  • Velišek, J. 2009. Chloropropanols. Pp. 539562 in R. H. Stadler and D. R. Lineback, eds. Process-induced food toxicants. John Wiley and Sons, Inc., Hoboken, NJ.
  • Weisshaar, R. 2008. Determination of total 3-chloropropane-1,2-diol (3-MCPD) in edible oils by cleavage of MCPD esters with sodium methoxide. Eur. J. Lipid Sci. Technol. 110:183186.
  • Weisshaar, R., and R. Perz. 2010. Pp. 158–165 in Fatty acid esters of glycidol in refined fats and oils. Eur. J. Lipid Sci. Technol. 112:158165.
  • Xu, X., Y. Ren, P. Wu, J. Han, and X. Shen. 2006. The simultaneous separation and determination of chloropropanols in soy sauce and other flavoring with gas chromatography-mass spectrometry in negative chemical and electron impact ionization modes. Food Addit. Contam. 23:110119.
  • Zelinkova, Z., B. Svejkovska, J. Velišek, and M. Doležal. 2006. Fatty acid esters of 3-chloropropane-1,2-diol in edible oils. Food Addit. Contam. 23:12901298.
  • Zelinkova, Z., O. Novotny, J. Schurek, J. Velišek, J. Hajšlova, and M. Doležal. 2008. Occurence of 3-MCPD fatty acid esters in human breast milk. Food Addit. Contam. 25:669676.
  • Zelinkova, Z., M. Dolezal, and J. Velisek. 2009. Occurrence of 3-chloropropane-1,2-diol fatty acid esters in infant and baby foods. Eur. Food Res. Technol. 228:571578.