Polychlorinated biphenyls (PCBs) and organochlorinepesticides (OCPs) are representative persistent organicpollutants (POPs).1,2 Since some of them are suspected to becarcinogens or endocrine disrupters, the determination of PCBsand OCPs in the environment is important to evaluate their risk.The extraction step is the most critical for the trace analysis oforganic compounds, because incomplete extraction causesinaccurate analytical results, even if the isotope dilution methodis utilized. However, the extraction efficiency of conventionaltechniques, such as Soxhlet extraction, is limited by the boilingpoint of the used solvent. In addition, the conventionalprocedures for the determination of PCBs and OCPs are usuallytedious and time consuming.

Improving the techniques used for the analysis of POPs insolid matrices has been widely investigated to increase therecovery yields of analytes, to minimize waste solvents, and toshorten the analytical procedures and time. Emergingtechniques, microwave-assisted extraction (MAE), pressurizedfluid extraction (PFE) and supercritical fluid extraction (SFE)need a relatively short extraction time and a small amount ofsolvent.36 They sometimes give higher recovery yields of theanalytes compared with conventional Soxhlet extraction orsaponification, because the extraction temperature can beincreased higher than the boiling point of the extraction solventunder atmospheric pressure by performing the extractionprocedures in closed pressure-resistant vessels. Among theemerging techniques, MAE has some advantages. High samplethroughput is realized by a simultaneous sample treatment withmicrowave irradiation in an oven. Possible degradation of somecompounds caused by contact with metals7,8 can be avoidable,because metal vessels are unnecessary. In addition, the MAEapparatus is relatively simple and reliable, because high-

pressure pumps or valves are unnecessary.Although the MAE technique was recently introduced as an

official method for environmental analysis, such as the EPAMethod 35469 and ASTM D6010,10 application of the techniqueto the certification of reference materials is not reported. Theoptimization of MAE procedures and comparison with otherextraction techniques were reported in many articles.1119However, a combination of the MAE technique and the isotope-dilution method has not been employed, except a few cases.16

National Metrology Institute of Japan in National Institute ofAdvanced Industrial Science and Technology (NMIJ/AIST) hasbeen preparing some matrix reference materials, such assediments and biological tissues, for environmental analysis.Primary methods20 have the highest quality of measurement,and are essential to the certification of reference materials.Among the primary methods, only the isotope dilution methodcan be applied for the quantification of analytes in matrixsamples. In addition, we have applied more than two analyticalmethods for their certification to avoid any possible proceduralbias. Since the combination of the high-efficiency extractiontechnique, MAE and a primary measurement method, isotopedilution, is a candidate analytical method for the determinationof PCBs and OCPs in the sediment reference materials, theeffects of the extraction conditions, such as temperature, timeand solvents, on the determination of some chlorinated biphenyl(CB) congeners and OCPs have been studied. The analyticalresults obtained by the optimized MAE technique and thoseobtained by other extraction techniques, such as PFE andSoxhlet extraction, have been compared to evaluate the MAEtechnique as a tool for certification of the matrix referencematerials that we are planning to develop.

793ANALYTICAL SCIENCES MAY 2004, VOL. 202004 The Japan Society for Analytical Chemistry

Evaluation of a Microwave-Assisted Extraction Techniquefor the Determination of Polychlorinated Biphenyls andOrganochlorine Pesticides in Sediments

Masahiko NUMATA, Takashi YARITA, Yoshie AOYAGI, and Akiko TAKATSU

National Metrology Institute of Japan, National Institute of Advanced Industrial Science and Technology, Tsukuba Central 3, Umezono, Tsukuba 3058563, Japan

A microwave-assisted extraction (MAE) technique for the determination of polychlorinated biphenyls (PCBs) andorganochlorine pesticides (OCPs) in marine sediment samples has been investigated. The analytes were extracted underdifferent treatment conditions, such as temperature, time and extraction solvent. They were quantified by an isotope-dilution method, and the observed concentrations and recovery yields obtained under different conditions were compared.The results of a comparison between this MAE and other extraction techniques, such as pressurized fluid extraction,saponification, sonication, and Soxhlet extraction, are also given in this report. The techniques gave comparable resultswith the values obtained by other extraction techniques and the certified values in the samples. However, the observedconcentration values of mono- and dichlorinated biphenyls varied depending on the extraction temperature.

(Received November 28, 2003; Accepted March 17, 2004)

To whom correspondence should be addressed.E-mail: mas-numata@aist.go.jp

Experimental

Samples and reagentsTwo marine sediment samples were used in this study.

Sediment D was collected from a bay of Kyusyu island as acandidate matrix reference material for environmentalanalysis.21 A certified reference material for the analysis oforganic pollutants in marine sediment, SRM1944,22 wasobtained from the National Institute of Standards andTechnology (USA).

Surrogate solution and syringe spike solution were preparedfrom solutions of each isotope labeled compounds (CB15, 2870, 101, 170, 180, 194, 209-13C12: 50 g/ml, WellingtonLaboratories, Canada; -HCH-13C6, 4,4-DDE-13C12, 4,4-DDD-d8: 100 g/ml, Cambridge Isotope Laboratories, USA; 4,4-DDT-13C12: 100 g/ml, Dr. Ehrenstorfer, Germany; chlorinatedbiphenyl (CB) congener numbers are represented as IUPACnumber).23 The calibration solutions for gas chromatograph/high resolution mass spectrometry (GC/HRMS) were preparedfrom native compounds (CB15, 28, 70, 101, 180, 194, 209;purity > 98%, Cambridge Isotope Laboratories, USA; -HCH: >99%, Wako Pure Chemical Industries, Osaka, Japan; 4,4-DDE:neat, Supelco, USA; 4,4-DDD, 4,4-DDT: > 99%, Dr.Ehrenstorfer, Germany) and the PCB-13C12 solutions describedabove. Some PCB and pesticide analysis-grade reagents(acetone, toluene, dichloromethane, hexane, and anhydroussodium sulfate: Kanto Chemical, Tokyo, Japan) were used forthe extraction procedures.

Microwave-assisted extraction The sediment sample (2.5 g of sediment D or 0.8 g of

SRM1944) was accurately weighed into a Teflon PFAextraction cell (approximately 100 ml volume, GreenChemPlus, CEM, USA). After the addition of the surrogate solution(0.4 ml of 2,2,4-trimethylpentane solution), the sample wasextracted with 20 ml of hexane/acetone (1:1, v/v), acetone,dichloromethane, toluene or hexane. Because less-polarsolvents do not absorb microwave energy efficiently, 5 ml ofwater was added in the case of toluene or hexane extraction.The cells were covered with pressure-resistant holders, andwere heated with sporadic irradiation of microwave (150 200W per cell) at 100 145C for 5 30 min (temperature ramp: 15min) using a microwave extraction system MARSX (CEM,USA). Upon termination of the microwave irradiation, the cellswere air-cooled. After cooling, the organic solvent layer wascentrifuged at 3000 rpm for 3 min to remove the solid residues.The obtained supernatant was cleaned up by a method ofSchantz et al.24 with some modifications.21,25 The supernatantwas treated with activated copper powder and anhydroussodium sulfate to remove elemental sulfur and water. Afterfiltration with a PTFE membrane filter (pore size, 0.2 m), thesolution was passed through a solid phase extraction (SPE)silica cartridge (500 mg silica in a 3 ml cartridge, InternationalSorbent Technology, England). Then, PCBs and OCPs wererecovered with 15 ml of a dichloromethane/hexane mixture(1:9, v/v). The eluent was concentrated and then loaded onto anormal phase liquid chromatography (nHPLC) system (Table1). For GC/HRMS analysis, the collected two fractions (3 17min fraction, PCBs, 4,4-DDE and 4,4-DDT; 20 25 minfraction, more polar OCPs [4,4-DDD and -HCH]) were mixed.

Pressurized fluid extractionThe sediment sample and 12 g of anhydrous sodium sulfate

were weighed in an extraction cell (stainless-steel; volume, 11

ml). After mixing the contents by shaking, the surrogatesolution was added into the cell. The content was extractedwith hexane/acetone (1:1, v/v) for 30 min using a PFE system,ASE 200 (Dionex, USA) at 150C, 15 MPa. The extraction wasrepeated once under the same condition. The obtained extractwas cleaned up by an activated copper-powder treatment, SPE,and nHPLC, as described above.25

Soxhlet extractionA mixture of the sediment sample and 10 g of anhydrous

sodium sulfate was placed into a filter paper thimble, and thesurrogate solution was added. The sample was loaded on anautomated Soxhlet extraction system, B-811 (BCHI,Switzerland), and then extracted with 250 mL of n-hexane/acetone (1:1, v/v) for 36 h. The extract was cleaned upby SPE and nHPLC, as described above.

Saponification The analytical procedures were followed as a Japanese official

method for PCB determination, Endocrine DisruptingChemicals Interim Investigation Manual26 (Chapter I:Determination of polychlorinated biphenyls). After theaddition of the surrogate solution, the sediment sample wastreated with 50 ml of an ethanolic potassium hydroxide (1 M)solution and 10 ml of water at room temperature by shaking for1 h. The CB congeners were extracted with hexane. Then, thehexane layer was shaken with sulfuric acid (98%) to removemost of the pigments. The hexane layer was passed through acolumn containing 5 g of silica gel (moisture content, 5%;Kanto Chemical) to remove any polar constituents. The columnwas washed with 40 ml of hexane to recover CB congeners.

Sonication The analytical procedures were followed as a Japanese official

method for OCP determination, Endocrine Disrupting

794 ANALYTICAL SCIENCES MAY 2004, VOL. 20

Table 1 Operational parameters of normal phase HPLC

Instrument Gulliver system, JascoUV detector: Jasco UV-975

Analytical column YMC-Pack NH2 10 mm i.d. 150 mmGuard column YMC-Guard Pack NH2 10 mm i.d. 30 mmMobile phase A: hexane; B: dichloromethaneGradient 100% A (7.5 min) 91% A, 9% B (30.0 min)Flow rate 4.0 ml min1Injection volume 950 lDetection 245 nm

Fig. 1 Flow diagram of sample preparation for the determination ofPCB congeners and OCPs in sediment samples. Sox, Soxhletextraction; Sap, saponification/hexane extraction.

Chemicals Interim Investigation Manual26 (Chapter II:Determination of organochlorine pesticides, polybrominatedbiphenyls and benzo[a]pyrene). After the addition of thesurrogate solution and 5 ml of water, the sediment sample wasextracted with acetone by mechanical shaking and ultrasonicirradiation (10 min each). Then, the extraction procedures wererepeated twice. The obtained extract was treated with a sodiumchloride aqueous solution (5%, w/w) and hexane. The hexanelayer containing OCPs was passed through a silica-gel column(described previously). The column was eluted with 40 ml ofhexane, followed by 40 ml of hexane/acetone (19:1, v/v) torecover the OCPs.

Determination of PCBs and OCPs by GC-MSAnalyses of the CB congeners and OCPs in the extracts

obtained from different extraction and cleanup techniques (Fig.1) were performed using a GC/HRMS system (AutoSpec,Micromass, UK). After the syringe spike was added to thesample solution, the solution volume was reduced to 0.2 ml bymeans of a rotary evaporator and nitrogen gas stream. Aportion of the solution (1.0 l) was analyzed with the system(Table 2). The CB congener peaks on the chromatogram wereassigned by following some reports.27,28 Representativechromatograms of the analytes and their internal standards areshown in Fig.2.

Analytical results are represented as dry-mass base in thisreport. The moisture contents of the samples were determinedgravimetrically. The samples were dried at 105C for 6 h, andthe moisture contents were calculated from weighing before andafter drying.

Results and Discussion

Effect of the extraction solvent The extraction solvent, temperature, time, moisture content of

sample, microwave power, cycles of extraction would affect theextraction efficiency in the MAE technique.4 In this study,effects of relatively important factors (solvent choice, extractiontemperature and time) were investigated. The effects of the

extraction solvents on the PCB and OCP analysis in sediment Dare given in Tables 3 and 4. Since we applied the isotope-dilution method, differences between the calculatedconcentration values were not significant among the testedsolvents, except for extraction with hexane, in contrast with theresults obtained from the absolute calibration method inprevious reports.3,5,12,14,16 An increase in the extractionefficiency by the addition of water to less-polar solvents wasreported.14 However hexane-extraction gave a significantly lowrecovery of the analytes and surrogates.

Such a low recovery was improved by using more polarsolvents. Although extraction with acetone gave the highest(not significant) concentration values of some compounds, the

795ANALYTICAL SCIENCES MAY 2004, VOL. 20

Table 2 Operational parameters of GC/HRMS

GC conditions:Mobile phase Helium, 276 kPa (constant pressure)Injection Splitless, 200CColumn HT-8 (SGE), 0.22 mm i.d. 50 mm,

film thickness: 0.25 m60C (2 min) 15C min1 170C 3.5C min1 300C (6 min)

MS conditions:Ionization mode Electron ionization (35 40 eV)

250C

Mass resolution > 10000Detection mode Selected ion monitoringMonitoring ion 222.0003 (CB15), 234.0406 (CB15-13C12)(target compound) 257.9587 (CB28), 269.9986(CB28-13C12)

289.9224 (CB70), 301.9626 (CB70-13C12)323.8834 (CB101), 335.9237(CB101-13C12)393.8025 (CB180), 405.8428 (CB170,180-13C12)427.7646 (CB194), 439.8038 (CB194-13C12)497.6826 (CB209), 509.7229 (CB209-13C12)218.9116( -HCH), 224.9317 ( -HCH-13C6)235.0081(4,4-DDD), 243.0583 (4,4-DDD-d8)235.0081(4,4-DDT), 247.0484 (4,4-DDT-13C12)317.9351(4,4-DDE), 329.9735(4,4-DDE-13C12)

Column temperature

Ionization temperature

Fig. 2 Example of a GC/HRMS chromatogram of PCB congeners and OCPs in the sample solutionobtained by the MAE technique. Extraction temperature, 115C; extraction time, 10 min; solvent,hexane/acetone (1:1); GC/HRMS parameters are represented in Table 2.

variations of the values were larger in most cases. Judging fromthe color of the obtained extract, this phenomenon may havebeen caused by large amounts of the concomitants extractedwith acetone. Hexane/acetone (1:1, v/v) was used for thefollowing experiments, because the recovery yields of the mostanalytes were the highest among the tested solvents. Also, thesolvent was used for a temperature-dependent experiment byreason of safety, because the vapor pressure of dichloromethaneand acetone is higher.

Effect of the extraction temperature and time The effects of the process temperature and time on the

analytical results and the recovery yields of the surrogates in thesediment D are shown in Figs. 3 and 4. Because the behaviorsof CB70, CB180, CB194 and 4,4-DDT were almost the sameas the temperature and time dependences of CB101 and CB28,they were omitted from the graphs. The stabilities of the testedanalytes were checked in advance. The hexane/acetone solutionof the analytes was irradiated with microwaves, and heated at

the highest tested temperature (145C) for 10 min. Because therecovery yields of all analytes were higher than 95%,degradation of the analytes would be negligible.

The observed values of the concentration of most CBcongeners reached plateaus at under 115C. The observedvalues of the concentration and the recovery yields of most CBcongeners and 4,4-DDE reached plateaus within 10 min at115C. Instead of elongation of the MAE process, the effect ofrepeating the extraction was also investigated. The extractionresidue was extracted again, and the obtained supernatants werecombined. Although it gave a slightly higher recovery yields ofthe analytes, the concentration values were not improved.

The reason for the relatively low reproducibility of 4,4-DDDquantification is unclear. The use of a deuterium-substitutedcompound as a surrogate of 4,4,-DDD could cause an error,because the deference in the physical properties betweendeuterium and hydrogen are larger than in the case of 13C and12C. Because 4,4-DDD is one of the most polar analytes, theabsorption on sediment particles, the inner surface of glasswareor the stationary phase of chromatograph may enhance such aneffect.

As shown in Fig. 3, the obtained concentration values ofCB15, 4,4-DDD and -HCH in sediment D depended on theextraction temperature. Such the temperature dependence wasalso observed in the case of other mono- and dichlorinatedbiphenyl congeners. However, there is no significant differencebetween the recovery yields of these isotope labeled compoundsunder high and low temperature extraction, On the other hand,when the extraction residue at 100C was extracted at 145C,the amounts of native CB15 and native polar OCPs in theextracted solution were almost equivalent to the differencebetween the observed values at both temperatures, and theconcentrations of their surrogates in the extract were almostnegligible (data not shown). These results would mean thatthese compounds (mono- and di-CBs, and relatively polarOCPs) bind the sediment particle firmly. Also, the exchangebetween the isotope labeled compound in solution and nativecompounds adsorbed on the particles may be very slow at lowtemperature. This phenomenon may be caused by the trappingof aged pollutants into the three-dimensional structure of clayminerals or humic substances.29,30 In particular, thedetermination of accurate mono- and di-CB concentrationswould be problematic, because the concentration values did notreach a plateau, even at the maximum temperature in the tested

796 ANALYTICAL SCIENCES MAY 2004, VOL. 20

Table 3 Concentrations of CB congeners and OCPs in sediment D, and the recovery yields of surrogates obtained by MAE with polar solvents

a. Extraction: 115C, 10 min; values as average of 3 measurements (3 extractions); error, SD. b. Recovery yields of surrogates through extraction and cleanup.

DichloromethaneAcetoneHexane/acetone(1:1)Recoveryg, %Concentrationa/ng g1 Recoveryg, %Concentrationa/ng g1 Recoveryg, %Concentrationa/ng g1

CB15 2.27 0.08 81 12 2.29 0.08 77 45 2.28 0.13 83 20CB28 33.9 0.2 73 11 34.1 1.1 66 16 33.9 0.4 61 5 CB70 60.5 1.1 74 7 61.3 1.0 69 13 59.5 0.1 65 12CB101 30.6 1.0 78 9 30.9 0.8 74 15 31.0 0.7 68 13CB180 8.58 0.58 85 4 8.59 0.26 70 8 8.04 0.26 67 13CB194 1.92 0.13 89 3 1.77 0.07 65 13 1.76 0.05 61 3 CB209 1.51 0.35 93 6 1.12 0.05 59 11 1.16 0.06 58 5 -HCH 4.70 0.39 67 12 5.16 0.21 48 5 5.45 0.02 46 12

4,4-DDE 5.71 0.12 82 9 6.03 0.08 71 8 5.80 0.17 70 164,4-DDD 13.2 0.9 63 10 13.6 0.6 58 15 13.0 0.6 48 104,4-DDT 5.59 0.92 107 42 6.44 1.51 91 34 5.71 0.67 68 10

Table 4 Concentrations of CB congeners and OCPs in sediment D, and the recovery yields of surrogates obtained by MAE with different solvents

a. Extraction: 115C, 10 min; values as average of 3 measurements (3 extractions); error, SD. b. Recovery yields of surrogates through extraction and cleanup.c. The peak of 4,4-DDT-13C12 on the chromatogram was overlapped on other peaks.

Toluene Hexane

Concentrationa/ng g1

Recoveryb,%

Concentrationa/ng g1

Recoveryb,%

CB15 2.27 0.04 86 12 1.87 0.07 50 1 CB28 33.0 1.0 70 4 28.2 1.3 45 3 CB70 58.7 0.7 71 5 50.7 2.0 47 5 CB101 29.8 1.1 75 5 26.2 0.9 50 5 CB180 7.46 0.32 68 4 7.12 0.48 49 9 CB194 1.53 0.07 67 1 1.54 0.18 44 4 CB209 1.08 0.17 64 0 1.36 0.51 41 3 -HCH 5.23 0.04 65 5 4.78 0.67 40 8

4,4-DDE 4.35 0.03 136 12 4.31 0.51 54 9 4,4-DDD 12.5 0.1 20 1 12.0 2.3 43 144,4-DDT n.d.c 6.89 4.97 52 27

conditions (145C). Such a temperature dependence has notbeen reported, because the extraction temperature was up to130C and only tri- to deca-CB congeners were the target ofanalysis in most of previous studies. Although the contributionof the mono- and di-CBs to the total toxicity of PCBs is low,such a possible error should be considered in the case of a PCBpollution source apportioning based on the congener profiles.

Method comparisonPrincipally, the isotope-dilution method has the highest

accuracy among quantification methods that can be applied tothe analysis of matrix samples. However, the equilibriumbetween native analytes adsorbed on a solid matrix and isotope-labeled surrogates is not realized in some case, as in the case ofCB15. Because the achievement of isotope equilibrium cannotbe evaluated practically, the recovery yield should be as high aspossible for the accurate determination of analytes. In general,higher temperature extraction, such as MAE and PFE, gives amore complete recovery and isotope equilibrium compared withlow-temperature processes, such as sonication and Soxhlet.Extraction by semi-continuous processes, such as Soxhlet andPFE, would give a higher recovery of the analytes, but an error

may be caused by loss of surrogates or insufficient time for theexchange of surrogates and native analytes adsorbed into thematrices, compared with batch processes, such as sonication andMAE. In the case of MAE, a decrease in recovery of theanalytes may be caused by re-absorption on extraction residuesin the period of cooling before solid-liquid separation. In thisstudy, the analytical results obtained using MAE werecompared with those obtained by other extraction techniques toevaluate the feasibility of the MAE technique. As shown inTables 5 and 6, the analytical results obtained using the MAEtechnique agreed with the certified values concerning theconcentration of the analytes, and the results obtained by otherextraction techniques, except for CB15 and -HCH. Therecovery yields of the surrogates obtained by the MAEtechnique (typically 70 to 100%) were almost the same, orslightly higher, compared with other extraction techniques. Therelatively high analytical results of CB15 and -HCH obtainedby the MAE and PFE techniques would reflect an improvementof the recovery from the matrices.

797ANALYTICAL SCIENCES MAY 2004, VOL. 20

Fig. 3 CB congener (A) and OCP (B) concentrations as a function of the MAE process temperature.Extraction time, 10 min; solvent, hexane/acetone (1:1); values as an average of 3 measurements (3extractions): , CB15; , CB28; , CB101; , CB209; , -HCH; , 4,4-DDE; , 4,4-DDD; error bars,SD (standard deviation).

Fig. 4 CB congener (A) and OCP (B) concentrations as a function of the MAE process time. Extractiontemperature: 115C, solvent, hexane/acetone (1:1); values as an average of 3 measurements (3 extractions);

, CB15; , CB28; , CB101; , CB209; , -HCH; , 4,4-DDE; , 4,4-DDD; error bars, SD(standard deviation).

Conclusion

As a result of evaluations, the MAE technique was found to besuitable as an accurate alternative to conventional extractiontechniques. Although more investigation would be necessary todetermine mono- and di-CB congeners accurately, the highlyefficient extraction technique, MAE, would be the preferredtechnique for the certification of analytes, such as PCBs andOCPs. We are planning to develop some certified referencematrix materials for environmental analysis, such as PCBs andOCPs in marine sediments. The accurate, high sample-throughput, and low solvent consumption technique, MAE, willbe used as one of the extraction techniques in certifying thematrix reference materials by combining the isotope dilution

method.

References

1. V. Lang, J. Chromatogr., 1992, 595, 1.2. S. Tanabe, H. Iwata, and R. Tatsukawa, Sci. Total Environ.,

1994, 154, 163.3. C. S. Eskilsson and E. Bjrklund, J. Chromatogr. A, 2000,

902, 227.4. J. R. Dean and G. Xiong, Trends Anal. Chem., 2000, 19, 553.5. V. Camel, Analyst, 2001, 126, 1182.6. M. M. Schantz, S. Bwadt, B. A. Benner Jr., S. A. Wise,

and S. B. Hawthorne, J. Chromatogr. A, 1998, 816, 213.7. V. van de Valk and Q. T. Dao, Chemosphere, 1988, 17, 1735.8. Q. Wu, A. Majid, and W. D. Marshall, Green Chemistry,

2000, 2, 127.9. EPA Method 3546, Microwave Extraction, Test methods

for evaluating solid waste, EPA, Washington D.C., 1999.10. ASTM (American Society of Testing and Materials)

D6010-96, Standard Practice for Closed Vessel MicrowaveSolvent Extraction of Organic Compounds from SolidMatrices, ASTM International, PA, USA, 2001.

11. V. Lopez-Avila, R. Young, and W. F. Beckert, Anal.Chem., 1994, 66, 1097.

12. V. Lopez-Avila, R. Young, J. Benedicto, P. Ho, R. Kim,and W. F. Beckert, Anal. Chem., 1995, 67, 2096.

13. R.-A. Dring and St. Gth, Fresenius J. Anal. Chem., 2000,368, 684.

14. A. Pastor, E. Vzquez, R. Ciscar, and M. de la Guardia,Anal. Chim. Acta, 1997, 344, 241.

15. N. Carro, Y. Saavedra, I. Garca, and M. Llompart, J.Microcolumn Sep., 1999, 11, 544.

16. O. Zuloaga, N. Etxebarria, L. A. Fernndez, and J. M.Madariaga, Talanta, 1999, 50, 345.

17. A. M. Cicero, E. Pietrantonio, G. Romanelli, and A. DiMuccio, Bull. Environ. Contam. Toxicol., 2000, 65, 307.

18. S. Jayaraman, R. J. Pruell, and R. McKinney,Chemosphere, 2001, 44, 181.

19. D. Martens, M. Gfrerer, T. Wenzl, A. Zhang, B. M.Gawlik, K.-W. Schramm, E. Lankmayr, and A. Kettrup,Anal. Bioanal. Chem., 2002, 372, 562.

20. T. J. Quinn, Metrologia, 1997, 34, 61.21. M. Numata, T. Yarita, Y. Aoyagi, and A. Takatsu, Anal.

Chem., 2003, 75, 1450.22. Certificate of Analysis for SRM1944: New York/New

Jersey Waterway Sediment, National Institute of Standardsand Technology, Gaithersburg, MD, 1999.

23. R. Guitart, P. Puig, and J. Gmez-Cataln, Chemosphere,1993, 27, 1451.

24. M. M. Schantz, J. J. Nichols, and S. A. Wise, Anal. Chem.,1997, 69, 4210.

25. T. Yarita, A. Nakama, M. Numata, Y. Aoyagi, M.Yamazaki, and A. Takatsu, Bunseki Kagaku, 2003, 52,1011.

26. Endocrine Disrupting Chemicals Interim InvestigationManual (water, sediments, aquatic organisms) (inJapanese), Environment Agency of Japan, 1998.

27. B. Larsen, M. Cont, L. Montanarella, and N. Platzner, J.Chromatogr. A, 1995, 708, 115.

28. G. M. Frame, Fresenius J. Anal. Chem., 1997, 357, 701.29. J. J. Pingnatello and B. Xing, Environ. Sci. Technol., 1996,

30, 1.30. G. L. Northcott and K. C. Jones, Environ. Pollut., 2000,

108, 19.

798 ANALYTICAL SCIENCES MAY 2004, VOL. 20

Table 5 Concentrations of CB congeners and OCPs in the NIST SRM1944 obtained by different extraction techniques

Official methodbConcentrationd

/ng g1

Certified valuescConcentratione

/ng g1

MAEaConcentrationd

/ng g1

CB15 28.2 0.8 15.5 0.3CB28 84.1 3.4 75.6 3.8 80.8 2.7CB70 71.6 1.8 70.1 2.5CB101 61.6 2.2 61.6 1.1 73.4 2.5fCB180 40.2 3.4 38.8 0.6 44.3 1.2CB194 10.4 0.9 10.5 0.6 11.2 0.33CB209 7.45 0.18 7.36 0.28 6.81 0.33-HCH 0.21 0.21 0.14 0.02

4,4-DDE 78.9 2.2 80.3 2.3 86.0 12.04,4-DDD 123 6 111 7 108 16

a. Solvent: hexane/acetone (1:1), 130C, 10 min.b. CB congeners and OCPs were determined by saponification and ultrasonic extraction, respectively.c. Certified values calculated from results obtained by Soxhlet extraction and PFE (Ref. 18).d. Values as average of 3 measurements (3 extractions); error, SD. e. Error: expanded uncertainties.f. CB101 + CB90.

Table 6 Concentrations of CB congeners and OCPs in sediment D obtained by different extraction techniques

a. Solvent: hexane/acetone (1:1), 130C, 10 min.b. CB congeners and OCPs were determined by saponification and ultrasonic extraction, respectively.c. Solvent: hexane/acetone (1:1), 36 h extraction.d. Solvent: hexane/acetone (1:1), 150C, 15 MPa, 30 min 2 cycles. e. Values as average of 3 measurements (3 extractions); error, SD.

MAEaConcent-

ratione/ng g1

Official methodbConcent-

ratione/ng g1

SoxhletcConcent-

ratione/ng g1

PFEdConcent-

ratione/ng g1

CB15 2.29 0.05 1.80 0.10 2.28 0.06 2.41 0.06CB28 34.2 0.4 31.0 1.5 34.6 0.7 35.3 0.4CB70 61.2 0.9 59.2 1.3 61.3 0.8 62.8 0.3CB101 30.4 0.5 30.7 1.1 32.6 1.0 32.8 0.1CB180 8.27 0.57 7.84 0.61 10.0 1.5 9.57 0.14CB194 1.82 0.19 1.75 0.11 1.86 0.25 1.81 0.09CB209 1.18 0.07 1.39 0.26 1.14 0.07 1.30 0.24-HCH 5.02 0.34 3.73 0.53 5.04 0.36 4.80 0.33

4,4-DDE 5.85 0.01 5.63 0.05 5.83 0.22 5.81 0.184,4-DDD 12.4 0.6 12.5 0.8 12.5 2.1 13.6 0.74,4-DDT 5.53 1.01 6.73 1.68 4.90 0.04 6.02 0.37