Determination of 1,3-dichloro-2-propanol and 3-chloro-1,2-propandiol in soy sauce by headspace

Analytica Chimica Acta591(2007)

167–172

Determination of1,3-dichloro-2-propanol and3-chloro-1,2-propandiol in soy sauce by headspace derivatization solid-phase microextraction combined with gas chromatography–mass spectrometry

Maw-Rong Lee a,?,Tzu-Chun Chiu a,Jianpeng Dou a,b

a Department of Chemistry,National Chung Hsing University,Taichung40227,Taiwan,ROC

b College of Biological and Agricultural Engineering,Jilin University,Changchun130022,PR China

Received26January2007;received in revised form26March2007;accepted27March2007

Available online31March2007

Abstract

This study proposes a method for identifying1,3-dichloro-2-propanol and3-chloro-1,2-propandiol in aqueous matrices by using headspace

on-?ber derivatization following solid-phase microextraction combined with gas chromatography–mass spectrometry.The optimized SPME exper-

imental procedures for extracting1,3-dichloro-2-propanol and3-chloro-1,2-propandiol in aqueous solutions involved a85?m polyacrylate-coated

?ber at pH6,a sodium chloride concentration of0.36g mL?1,extraction at50?C for15min and desorption of analytes at260?C for3min.

Headspace derivatization was conducted in a laboratory-made design with N-methyl-N-(trimethylsilyl)-tri?uoroacetamide vapor following solid-

phase microextraction by using3?L N-methyl-N-(trimethylsilyl)-tri?uoroacetamide at an oil bath temperature of230?C for40s.This method

had good repeatability(R.S.D.s≤19%,n=8)and good linearity(r2≥0.9972)for ultrapure water and soy sauce samples that were spiked with two analytes.Detection limits were obtained at the ng mL?1.The result demonstrated that headspace on-?ber derivatization following solid-phase

microextraction was a simple,fast and accurate technique for identifying trace1,3-dichloro-2-propanol and3-chloro-1,2-propandiol in soy sauce.

Keywords:1,3-Dichloro-2-propanol;3-Chloro-1,2-propandiol;Solid-phase microextraction;Gas chromatography–mass spectrometry;Soy sauce

1.Introduction

The compounds1,3-dichloro-2-propanol(1,3-DCP)and3-chloro-1,2-propandiol(3-MCPD)both belong to a group of chemicals called chloropropanols and are formed in acid-hydrolyzed vegetable protein(acid-HVP)by the reaction of hydrochloric acid with residual vegetable fat[1].Enzyme-HVP had a much lower level of3-MCPD than acid-HVP[2].Acid-HVP is a common savory ingredient in soy sauce and related products,as it offers higher yield.Therefore,1,3-DCP and3-MCPD were identi?ed in soy sauce[3,4].However,1,3-DCP and3-MCPD are considered to be as potential human carcino-gens and mutagens with other toxic effects[5–9].The European Commissions Scienti?c Committee on Food de?ned2?g kg?1 body weight as a tolerable daily intake(TDI)of3-MCPD[10]. The European Union(EU)established a maximum level of ?Corresponding author.Tel.:+886422851716;fax:+886422862547.

E-mail address:mrlee@https://www.360docs.net/doc/ca1520216.html,.tw(M.-R.Lee).0.02mg kg?1for3-MCPD in soy sauce and acid-HVP[11].The Food and Drug Administration of the United States established regulations for acid-HVP levels of3-MCPD and1,3-DCP to not exceed1mg kg?1and0.05mg kg?1,respectively[12].

Several analytical techniques have been developed to deter-mine chloropropanols in foods including gas chromatography (GC)with electron capture detection(ECD)[13],GC com-bined with mass spectrometry(MS)[4,14–25]and molecular imprinting[26].At present,GC/MS is a major,rapid and sensi-tive approach for analyzing chloropropanols.The conventional pretreatment method for3-MCPD and1,3-DCP in soy sauce is puri?cation with a glass chromatography column loaded with diatomaceous earth or an Extrelut column,followed by con-centration and derivation.The pretreatment procedures are not only tedious and troublesome,but also require large volumes of organic solvents,which in?uence the detection limit of trace analysis[4,14–23].Solid-phase microextraction(SPME)is a rapid and solvent-free technique and does not suffer from the above problems[27,28].SPME is based on the equilibrium of analyte concentration between the sample and the solid-phase

168M.-R.Lee et al./Analytica Chimica Acta591(2007)167–172

?ber coating.SPME has been extensively adopted to extract organic compounds from aqueous matrix[29].

Derivatization is a common method for improving chromato-graphic separations for polar compounds.The identi?cation of3-MCPD and1,3-DCP generally requires the derivatiza-tion of the hydroxyl group before GC analysis to increase the detection sensitivity.Conventionally,derivatization approaches include butaneboronyl[13],phenylboronyl[14,25],trimethylsi-lyl(TMS)[18]and hepta?uorobutyryl[19–23].A TMS derivatization reaction was carried out on the?ber coating of the SPME in the headspace device that was made in the labora-tory reducing the volume of the organic solvent and the reaction time[30].

Headspace(HS)SPME had been used to extract1,3-DCP and 3-MCPD derivatives[24,25].In this study,HS-SPME followed by on-?ber derivatization was developed for determination of 1,3-DCP and3-MCPD simultaneously.HS-SPME was initially used to extract trace1,3-DCP and3-MCPD from aqueous matrices,which were then derivatized on SPME with N-methyl-N-(trimethylsilyl)-tri?uoroacetamide(MSTFA).The two TMS derivatives were analyzed using GC/MS.The conditions of SPME extraction and derivatization were optimized.The feasi-bility of applying the proposed method was also demonstrated, and the detection limits,linear ranges and precision were elu-cidated.The optimum procedure was utilized to determine the amount of1,3-DCP and3-MCPD in soy sauce.

2.Experimental

2.1.Chemicals and materials

1,3-Dichloro-2-propanol(>98%),3-chloro-1,2-propandiol (>98%)and N-methyl-N-(trimethylsilyl)-tri?uoroacetamide were purchased from Sigma–Aldrich(St.Louis,MO,USA). Disodium hydrogen phosphate(99%)was obtained from Merck (Darmstadt,Germany).Citric acid(99%),sodium chloride (99.8%)and potassium dihydrogen phosphate(99.5%)were obtained from Riedel-deHa¨e n(Seelze,Germany).Methanol, ethyl acetate and hexane(HPLC grade)were purchased from TEDIA(Fair?eld,OH,USA).Hydrochloric acid(37.5%)was obtained from Fisher(Fair Lawn,NJ,USA).One standard mix-ture of the target analytes was prepared with?nal concentrations of136?g mL?11,3-DCP and132?g mL?13-MCPD in ethyl acetate.The other standard mixture was prepared with a?nal concentration of136?g mL?11,3-DCP and1320?g mL?13-MCPD in water.Standard working solutions were prepared by diluting with water and soy sauce,respectively.All stock solu-tions were stored for less than four weeks at4?C in a refrigerator and the fresh working solutions were prepared for each analy-sis.Buffer solutions were prepared at pH4with citric acid and 0.1M HCl,at pH5and6with citric acid and0.1M NaOH,and at pH7and8using Na2HPO4and KH2PO4.All chemicals and reagents used in this study were of analytical grade and were used without further puri?cation.Puri?ed water was obtained using an SG-Ultra clear water puri?cation system(SG Water Company,Barsb¨u ttel,Germany).Four soy sauce samples were purchased from supermarkets in Taichung,Taiwan.2.2.Solid-phase microextraction sampling procedure

The SPME?ber used herein was coated with85?m polyacry-late(PA)(Supelco Company,Bellefonte,PA,USA).The?ber was conditioned under helium at a?ow-rate of1.0mL min?1in the hot injection port of a GC at300?C for2h before use.All analyses were conducted in40mL sample vials and closed with PFTE-coated septa.The SPME?bers were reproducibly placed in the headspace of the vial above the water samples to extract 1,3-DCP and3-MCPD.The working solution of13.6ng mL?1 1,3-DCP and132ng mL?13-MCPD in water was used to opti-mize headspace SPME and derivatization conditions.During extraction,the aqueous matrix was continuously agitated at

a constant velocity of1000rpm with a Te?on-coated stir bar

(0.8cm×2.0cm)on a magnetic stirrer.

After extraction equilibrium had been reached,the?ber was transferred into the headspace derivatization system as described by Lee et al.[30,31].The derivatization system comprises a modi?ed10mL test tube with an inner tube(5cm×0.3cm) and a heating mantle.A3?L of MSTFA was placed in a test tube that was partially submerged in6mL of silicone oil.The lower portion of the system was immersed in an oil bath that was maintained at230?C for40s using a heating mantle.The SPME ?ber following extraction was pierced through the rubber sep-tum into the inner tube of the system,whereas the1,3-DCP and 3-MCPD adsorbed on the?ber were immediately derivatized with the MSTFA vapor from the bottom of the tube.Heating the inner tube in oil bath generated the MSTFA vapor.The deriva-tization procedures must be conducted in a hood with safety equipment,adding as little derivatization reagent as possible to prevent explosion of the glass apparatus at an excessive vapor-ization pressure of MSTFA.It is con?rmed to be safe for the derivatization device in our laboratory when the volume of the derivatization reagent is less than10?L.Following the deriva-tization,the SPME?ber was immediately removed and inserted into the GC injector.The desorption temperature was260?C and the desorption time was3min in all of the runs.No carry-over was observed after this desorption time.Triplicate analyses were performed for each data point in all experiments.

The water samples and soy sauce samples(5mL)were adjusted to pH6with buffer solution(pH6)and1.8g NaCl, and used in SPME extraction.

2.3.Gas chromatography–mass spectrometry

Samples were analyzed using a Varian Saturn2000ion-trap mass spectrometer(Varian Company,Palo Alto,CA,USA)with a Varian Star3400CX gas chromatograph.The column was a30m×0.25mm i.d.fused-silica capillary column DB-5MS and a stationary phase with a thickness of0.25?m(Agilent Company,Palo Alto,CA,USA).The?ow of helium carrier gas was maintained at a rate of1mL min?1using an electronic pressure control.Throughout the whole analysis,the injector was operated in the splitless mode with an injector tempera-ture of260?C.The transfer line was maintained at250?C.The oven temperature was initially set to40?C(held for3min); increased linearly at25?C min?1to80?C,increased to110?C

M.-R.Lee et al./Analytica Chimica Acta 591(2007)167–172169

at a rate of 5?C min ?1,and ?nally increased to 270?C at a rate of 25?C min ?1.Electron ionization (EI)and chemical ionization (CI)with methanol as the reagent gas were used as ionization modes.The ion trap was maintained at 200?C for EI and 100?C for CI,respectively.The mass spectra were obtained in a mass-to-charge ratio (m /z )scan range from 40to 350u.Extract ions mode was used to reduce background interference.The total analysis time of a single run was 17min.3.Results and discussion

3.1.Gas chromatography–mass spectrometry analysis A highly sensitive GC/MS technique was developed to deter-mine trace 1,3-DCP and 3-MCPD in aqueous matrices.Two ionization modes of MS,EI with electron energy of 70eV and CI with electron energy of 230eV ,were applied to analyze 1,3-DCP and 3-MCPD TMS derivatives.The characteristic ions of 1,3-DCP TMS derivative were m /z 185([M ?CH 3]+),m /z 165([M ?Cl]+)and m /z 151([M ?CClH 2]+)and those of 3-MCPD TMS derivative were m /z 239([M ?CH 3]+)and m /z 219([M ?Cl]+)in the normal EI mass spectrum.When methanol was used as the reagent gas in positive chemical ionization (PCI)mode,the characteristic ions were m /z 201([M +H]+),m /z 185([M ?CH 3]+)and m /z 165([M ?Cl]+)for the 1,3-DCP TMS derivative,and m /z 255([M +H]+),m /z 239([M ?CH 3]+)and m /z 219([M ?Cl]+)for the 3-MCPD TMS derivative.

A standard mixed solution of 1,3-DCP and 3-MCPD (200?L)in ethyl acetate and 200?L of the silylating agent,MSTFA,were added to a 0.7mL brown glass vial.The vial was capped with a PTFE-faced butyl septum.After 50min at 80?C in the water bath the vial was taken out and cooled to room tem-perature.A 200?L of hexane was also added to the vial which was vortex mixed for 5min.A 1?L of supernatant was injected into GC to compare the ionization ef?ciencies of various MS ionization modes.A 1?L of mixed solution of 1,3-DCP and 3-MCPD TMS derivatives was directly injected into GC,and the peak area of two derivatives in various ionization modes were compared.The results in Fig.1indicate that the full scan mode of EI had greater ionization ef?ciency than that of PCI.There-fore,EI-MS was adopted to analyze 1,3-DCP and 3-MCPD TMS

derivatives.

Fig.1.Mass chromatograms of 1?L 136?g mL ?11,3-DCP and 132?g mL ?13-MCPD TMS derivatives in full scan mode of

EI-MS.

Fig.2.Extraction ef?ciencies of 26.4?g mL ?11,3-DCP and 27.2?g mL ?13-MCPD in water with various ?ber coatings direct immersion for 15min.Extraction at 30?C,desorption at 260?C for 3min.

3.2.Optimization of solid-phase microextraction conditions The adsorption characteristics of ?ber coatings,the extrac-tion method,the extraction time,the extraction temperature,the headspace of the sample vial,the desorption temperature,the desorption time,the sample matrix and various other fac-tors in?uence the extraction ef?ciency of SPME [27].Various coatings have different absorption properties toward various analytes.The choice of coating is essential to the SPME method.In this study,the performance characteristics of 85?m of polyacrylate (PA),75?m of carboxen/polydimethylsiloxane (CAR/PDMS),100?m of polydimethylsiloxane (PDMS)and 65?m of polydimethylsiloxane/divinylbenzene (PDMS/DVB)were compared using direct immersion (DI)for 15min in 26.4?g mL ?1of 1,3-DCP and 27.2?g mL ?1of 3-MCPD mixed solution.A moderately polar PA ?ber with more of these two compounds was used to extract 1,3-DCP and 3-MCPD in aqueous matrices (Fig.2).In DI-SPME,the ?ber was intro-duced in the aqueous phase.The relative signals area counts of 13.6ng mL ?11,3-DCP and 132ng mL ?13-MCPD working solution were compared using SPME with DI and HS extraction for 15min and then on-?ber derivatization with 3?L MSTFA at 230?C for 40s.Fig.3indicates that sensitivity of HS-SPME exceeds that of DI-SPME for 1,3-DCP extraction and is slightly less than that of DI-SPME for 3-MCPD extraction.

HS-SPME

Fig.3.Peak areas of 13.6ng mL ?11,3-DCP and 132ng mL ?13-MCPD in water produced by DI-SPME and HS-SPME.Extraction at 30?C for 15min,desorption at 260?C for 3min,MSTFA 3?L,derivatization at 230?C for 40s.

170M.-R.Lee et al./Analytica Chimica Acta591(2007)

167–172

Fig.4.Effect of solution volume on peak areas of13.6ng mL?11,3-DCP and 132ng mL?13-MCPD in water produced by SPME–GC/MS.Extraction at 50?C for15min,desorption at260?C for3min,MSTFA3?L,derivatization at230?C for40s.

reduces interference from the matrix in real samples.Hence, HS-SPME was used to extract low volatile and strongly polar 1,3-DCP and3-MCPD in this study.

SPME is based on partition equilibrium between the concen-tration of the analytes in a sample and that in the solid-phase ?ber coating[32].The extraction temperature governs the mass transfer rate of1,3-DCP and3-MCPD from water into?ber. Extraction temperatures from30to70?C were maintained for 15min.The extracted amount reached a maximum at50?C,and decreased gradually as the temperature increased further,prob-ably because the partition coef?cients of1,3-DCP and3-MCPD on?ber coating decreased.Extraction was performed from5 to25min to determine the effect of extraction time at50?C. Fresh samples were used for each extraction.The results indi-cated that the peak areas of1,3-DCP and3-MCPD derivatives increased with increase of time and reached partition equilib-rium in15min.The headspace of the vial affected extraction ef?ciencies of the sample by SPME.The absolute amounts of 1,3-DCP and3-MCPD in the gas phase were relevant to the headspace volume of the sample vial,5,10,15,20and25mL of13.6ng mL?11,3-DCP and132ng mL?13-MCPD working solution were separately added to a40mL of vial.Fig.4reveals that the peak areas of1,3-DCP and3-MCPD decreased as the volume of the solution increased.The concentration of the ana-lytes in the gas phase was too low to be adsorbed by the SPME ?ber coating when the volume of solution was less than5mL. Therefore,5mL of1,3-DCP and3-MCPD water solutions in a 40mL vial were extracted for15min using headspace SPME at 50?C.

The desorption temperature and desorption time in?uence the amount of analytes that is desorbed from the?ber coating, which are usually optimized to develop a new SPME method [33,34].Desorption temperatures of240–280?C were applied. The results indicate that the peak areas of1,3-DCP and3-MCPD increased with the desorption temperature and then decreased slightly as the temperature rose further,reaching a maximum desorption amount at260?C.Desorption times of1–5min were examined by leaving the?ber in the injector for an increasing period and maintaining the temperature of the injector at260?

C.Fig.5.Effect of pH on peak areas of13.6ng mL?11,3-DCP and132ng mL?1 3-MCPD in water produced by SPME–GC/MS.Extraction at50?C for15min, desorption at260?C for3min,MSTFA3?L,derivatization at230?C for40s. The amounts of the1,3-DCP and3-MCPD desorbed increased with desorption time and reached a maximum after3min.There-fore,a desorption temperature of260?C and a3min desorption time were used in the experiment.

The changes in the sample matrix affected the signal inten-sity of the analyte that was obtained by SPME.The pH was varied from4to8,and was used to investigate how pH affects the extraction of aqueous1,3-DCP and3-MCPD(Fig.5).The result indicates that the extraction increased with pH to a maxi-mum at pH6,and then,decreased as the basicity of the solution increased further.The two compounds1,3-DCP and3-MCPD were easily extracted using SPME when they were in the molec-ular state at pH6.Sodium chloride of0.5,1.0,1.5,1.8and2.0g were added separately to the water samples to yield?nal con-centrations of NaCl of0,0.10,0.20,0.30and0.36g mL?1to study further the effect of sample matrix on the signal intensity of analyte.The results indicate that the peak areas of1,3-DCP and3-MCPD TMS derivatives increased with the amount of NaCl to a maximum at0.36g mL?1.Therefore,extraction was performed with spiking with0.36g mL?1NaCl and adjusting the pH of the solution to

Fig.6.Effect of derivatization time on peak areas of13.6ng mL?11,3-DCP and132ng mL?13-MCPD in water produced by SPME–GC/MS.Extraction at 50?C for15min,pH6,sodium chloride concentration of0.36g mL?1,desorp-tion at260?C for3min,MSTFA3?L,derivatization at230?C for40s.

M.-R.Lee et al./Analytica Chimica Acta591(2007)167–172171 Table1

Linear ranges,slopes,intercepts,standard errors,limits of detection and precisions at low and high concentrations produced by SPME–GC/MS for1,3-DCP and 3-MCPD in water and soy sauce

Compound Linear range

(ng mL?1)Slope(mean±S.D.)

(n=7)

Intercept(mean±S.D.)

(n=7)

r2(n=7)Standard error a

(n=7)

R.S.D.a

(%,n=8)

R.S.D.b

(%,n=8)

LOD

(ng mL?1)

LOD

(ng g?1)

1,3-DCP c 1.36–13608702±15334284±738380.998415747251250.27

3-MCPD c13.2–13200815±1026139±472310.99721431293167 4.22

1,3-DCP d 1.36–136011253±2124408±100060.997920347031790.411e

3-MCPD d13.2–13200866±14113958±669230.997815213661910 4.62 3.87f

a The concentration of1,3-DCP and3-MCPD was13.6and132ng mL?1,respectively.

b The concentration of1,3-DCP and3-MCPD was680and6600ng mL?1,respectively.

c The matrix was ultrapure water.

d Th

e matrix was soy sauce.

e Data from reference[23].

f Data from reference[25].

3.3.On-?ber derivatization of solid-phase microextraction

The major parameters that determine the ef?ciency in deriva-tization,derivatization time,oil bath temperature and amount of reagent,are investigated after obtaining the optimum condi-tions of SPME.In an oil bath study at temperature from200 to250?C,the maximum derivatization was obtained at230?C. When the oil bath temperature exceeded230?C,the desorption of the analytes from the?ber reduced the overall extraction ef?-ciency.The effects of1–5?L of MSTFA on derivatization were studied at230?C for10s.The peak areas of two TMS deriva-tives increased with the volume of MSTFA to a maximum at 3?L.The full derivatization time of1,3-DCP and3-MCPD was 40s.When the derivatization time exceeded40s,the desorp-tion of the analytes from?ber reduced the extraction ef?ciency (Fig.6).Therefore,HS-SPME was performed following the on-?ber derivatization of1,3-DCP and3-MCPD by adding3?L MSTFA at230?C for40s.

3.4.Linear range,limits of detection and precision

The limits of detection(LODs),quanti?cation(LOQs)and con?rmation(LOCs),quanti?cation range,the method precision and the recovery of the proposed method with spiked ultrapure water and soy sauce were evaluated.Table1presents the per-formance of this method under the optimum conditions.Seven points on calibration curves were plotted at1.36,6.8,13.6, 68,136,680and1360ng mL?1for1,3-DCP and at13.2,66, 132,660,1320,6600and13200ng mL?1for3-MCPD in water and soy sauce,respectively.The response was highly linear in water and soy sauce with a coef?cient of determination over 0.9972.The linear range experiments provided the necessary information to estimate LOD,based on the lowest detectable peak with a signal-to-noise(S/N)of three.The LOD of1,3-DCP, 0.35ng g?1,was less than1ng g?1,as stated by Abu-El-Haj et al.[23],and the LOD of3-MCPD,3.91ng g?1,was similar to that3.87ng g?1,presented by Huang et al.[25]for a density of soy sauce of1.18g mL?1.LOQ was based on an S/N of10.The LOCs of1,3-DCP in both matrices were2.7and4.1ng mL?1 and the LOQs were0.90and1.37ng mL?1.The LOCs of3-MCPD in both matrices were42.2and46.2ng mL?1and the LOQs were14.05and15.38ng mL?1.3.5.Application

The proposed method for determining amounts of1,3-DCP and3-MCPD in real samples was tested with soy sauce samples. Experiments were performed under the optimum conditions. Four soy sauce samples were analyzed using the developed method.No target analyte was found in the samples.However, this result was expected because the amount of chloropropanols in soy sauce is regulated.The four soy sauce samples were spiked with13.6ng mL?1of1,3-DCP and132ng mL?1of3-MCPD to assess the matrix effects.The relative recoveries of1,3-DCP and 3-MCPD were in the ranges84–99%and83–94%,respectively. Therefore,this method can be applied to soy sauce samples at the lowest permitted levels of1,3-DCP and3-MCPD in soy sauce, which are0.05and0.02mg kg?1,respectively[11,12].

4.Conclusion

This study proposed a method based on headspace SPME combined with GC/MS for simultaneously determining trace amount of1,3-DCP and3-MCPD in aqueous samples.Chro-matographic shape and sensitivity were improved using on-?ber derivatization following SPME combined with GC/MS.Detec-tion limits were0.41and4.62ng mL?1in soy sauce for1,3-DCP and3-MCPD,respectively.The proposed method exhibited a relative standard deviation of under19%.The method was improved by using an internal standard to improve precision and chemometrics to yield the optimal parameters.In conclu-sion,this solvent-free extraction and minimized derivatization reagent approach is a simple,fast and accurate procedure for determining trace chloropropanols in aqueous matrices. Acknowledgements

The authors would like to thank the National Science Council of the Republic of China for?nancially supporting this research under contract no.NSC92-2113-M-005-024.

References

[1]C.Crews,S.Hasnip,S.Chapman,P.Hough,N.Potter,J.Todd,P.Brereton,

W.Matthews,Food Addit.Contam.20(2003)916.

172M.-R.Lee et al./Analytica Chimica Acta591(2007)167–172

[2]http://ec.europa.eu/food/food/chemicalsafety/contaminants/scoop3-2-

9?nal report chloropropanols en.pdf.

[3]R.Macarthur,C.Crews,A.Davies,P.Brereton,P.Hough,D.Harvey,Food

Addit.Contam.17(2000)903.

[4]P.J.Nyman,G.W.Diachenko,G.A.Perfetti,Food Addit.Contam.20(2003)

909.

[5]B.S.Lynch,D.W.Bryant,G.J.Hook,E.R.Nestmann,I.C.Munro,Int.J.

Toxicol.17(1998)47.

[6]F.Bastos,J.Bessa,C.Pacheco,P.De Marco,P.M.L.Castro,M.Silva,R.

Ferreira Jorge,Biodegradation13(2002)211.

[7]E.K.Weisburger,B.M.Uliand,J.Nam,J.J.Gart,J.H.Weisburger,J.Natl.

Cancer Inst.67(1981)75.

[8]T.Shiozaki,Y.Mizobata,H.Sugimoto,T.Yoshioka,T.Yoshiharu,T.

Sugimoto,Hum.Exp.Toxicol.13(1994)267.

[9]L.Silhankova,F.Smid,M.Cerna,J.Davidek,J.Velisek,Mutat.Res.103

(1982)77.

[10]Scienti?c Committee on Food,Opinion of the Committee on food on3-

monochloro-propane-1,2-diol(3-MCPD).Updating the SCF opinion of 1994;adopted on30May2001.Available from http://europa.eu.int./comm/ food/fs/sc/scf/out91en.pdf.

[11]Commission Regulation(EC)No.466/2001,Setting maximum levels for

certain contaminants in foodstuff,https://www.360docs.net/doc/ca1520216.html,mun.(2001).

[12]Position Paper on Chloropropanols for the thirty-third Session of the Codex

Committee on Food Additives and Contaminants,Joint FAO/WHO Food Standards Programme,2001.

[13]R.L.Pesselman,M.J.Feit,J.Chromatogr.439(1988)448.

[14]L.E.Rodman,R.Dudley Ross,J.Chromatogr.369(1986)97.

[15]C.G.Hamlet,Food Addit.Contam.15(1998)451.

[16]P.Brereton,J.Kelly,C.Crews,S.Honour,R.Wood,A.Davies,J.AOAC

Int.84(2001)455.[17]D.C.Meierhans,S.Bruehlmann,J.Meili,C.Taeschler,J.Chromatogr.A

802(1998)325.

[18]I.Kissa,J.Chromatogr.605(1992)134.

[19]C.G.Hamlet,S.M.Jayaratne,W.Matthews,Food Addit.Contam.19(2002)

15.

[20]P.J.Nyman,G.W.Diachenko,G.A.Perfetti,Food Addit.Contam.20(2003)

903.

[21]K.O.Wong,Y.H.Cheong,H.L.Seah,Food Control17(2006)408.

[22]W.-c.Chung,K.-y.Hui,S.-c Cheng,J.Chromatogr.A952(2002)

185.

[23]S.Abu-El-Haj,M.J.Bogusz,Z.Ibrahim,H.Hassan,M.Al Tufail,Food

Control18(2007)81.

[24]S.Hasnip,C.Crews,N.Potter,P.Brereton,H.Diserens,J.-M.Oberson,J.

AOAC Int.88(2005)1404.

[25]M.Huang,G.Jiang,B.He,J.Liu,Q.Zhou,W.Fu,Y.Wu,Anal.Sci.21

(2005)1343.

[26]M.K.P.Leung,B.K.W.Chiu,https://www.360docs.net/doc/ca1520216.html,m,Anal.Chim.Acta491(2003)

15.

[27]https://www.360docs.net/doc/ca1520216.html,galante,M.A.Felter,J.Agric.Food Chem.52(2004)3744.

[28]Z.Zhang,M.J.Yang,J.Pawliszyn,Anal.Chem.66(1994)844A.

[29]K.J.Hageman,L.Mazeas,C.B.Grabanski,https://www.360docs.net/doc/ca1520216.html,ler,S.B.Hawthorne,

Anal.Chem.68(1996)3892.

[30]M.R.Lee,Y.S.Song,B.H.Hwang,C.C.Chou,J.Chromatogr.A896(2000)

265.

[31]M.R.Lee,R.J.Lee,Y.W.Lin,C.M.Chen,B.H.Hwang,Anal.Chem.70

(1998)1963.

[32]C.L.Arthur,J.Pawliszyn,Anal.Chem.62(1990)2145.

[33]L.Moens,T.De Smaele,R.Dams,P.Van Den Broeck,P.Sandra,Anal.

Chem.69(1997)1604.

[34]https://www.360docs.net/doc/ca1520216.html,l′a n,J.Pawliszyn,J.Chromatogr.A873(2000)63.