Minimization of Sample Volume with Air-segmented SampleInjection and the Simultaneous Determination of TraceElements by ICP-MS

2008©TheJapanSocietyforAnalyticalChemistry

Introduction

Recently, in the field of forensic chemistry, more precise and accurate analysis of trace amounts of impurities in forensic samples,i.e.drugs,hairandsoon,isgreatlyrequiredinaddition to the analysis of their distribution pattern and major components.Theconcentrationsandcomponentsofimpurities can identify synthetic processes of the samples and the producingareas.Thepresenceorabsenceofspecificimpurities mayleadtoidentifyingtherootofthesample,comparedwith the components of other samples.  However, these trace impurities in samples can sometimes not be determined with sufficient precision and accuracy.  Therefore, more precise analysesoftraceimpuritiesarerequired.

Gas chromatography (GC) and gas chromatography mass spectrometry (GC-MS) have generally been used to identify organic impurities in forensic samples.1–3_{  In this method, the} impurities are classified based on the peak pattern of chromatograms.  For analyzing inorganic impurities, neutron activation analysis (NAA),4,5_{ atomic absorption spectrometry} (AAS)6_{ and inductively coupled plasma atomic emission} spectrometry (ICP-AES)7_{ have been widely used.  Recently,} synchrotron radiation X-ray fluorescence spectrometry (SR-XRF)8,9_{ was used for small-volume samples, but its operation} needsaspecificlicenseandlargefacilities.Inductivelycoupled plasma mass spectrometry (ICP-MS)10,11_{ is one of the most} powerful tools for the determination of trace and ultra-trace metals.  However, it is sometimes very difficult to measure samples by using conventional/ICP-MS, because the volume of forensic samples is often very small (<1 ml) and the concentration is at the trace level.  Usually, a high efficiency

nebulizer(HEN)isemployedinsteadofaconventionalnebulizer to suppress sample consumption.  However, a HEN is easy to clog,12_{ and the results are sometimes not reproducible. } Solid-phase extraction13_{ is a promising treatment for concentrating} traceinorganiccompounds.Italsotakesmuchvolumewiththe column procedure.  On the other hand, conventional flow injection (FI/ICP-MS)14_{ is applied to the analysis of} small-volume samples in which each injection plug, such as a blank solution, sample solution or rinsing solution is injected sequentially.Thisinjectionmethodisgoodforcontrollingthe exact injection volume and detection time, but plugs cause a carry-over.  On the contrary, in the air-segmented sample injection (ASSI)15–17_{ method, the injected plugs are isolated by} air.Therefore,anycontaminationbetweenasampleandother solutionsissuppressed,thedispersionofplugscanbedecreased and dilution of the sample is also avoided.  ASSI is a very effectivesampleintroductionmethod.WecombinedASSIwith ICP-MS and developed a highly sensitive method with small-volume of samples using a conventional nebulizer.  The proposed method was validated by determining trace elements inariver-watercertifiedreferencematerial(CRM),andapplied tohumanhairsamples.

Experimental

TheICP-MSusedwasanSPQ8000H(SIINanoTechnology Inc., Tokyo, Japan), with a normal nebulizer (TR-30-C2, J. E. Meinhard Associates Inc., CA).  The ASSI system was constructed with an autosampler (Model AT-600, SII NanoTechnology Inc.) and a peristartic pump (Minipuls 2, Gilson Inc., WI).  This system is controlled by MS software (MicroSuction Ver. 1.8 M, SII NanoTechnology Inc.).  The autosamplercouldset50samplevials(volume,2ml),andone

Minimization of Sample Volume with Air-segmented Sample

Injection and the Simultaneous Determination of Trace

Elements by ICP-MS

Osamu N

,*

_**

_{Mitsuko O}

_{,* and Shoji M}

_*

*Graduate School of Natural Science and Technology, Okayama University,

3-1-1 Tsushimanaka, Okayama 700

8530, Japan

**Criminal Investigation Laboratory, Okayama Prefectural Police H.Q., 1-3-2 Tonda, Okayama 700

0816, Japan

The application of inductively coupled plasma mass spectrometry (ICP-MS) to forensic chemistry was studied.  The developedmethod,air-segmentedsampleinjection(ASSI)coupledwithICP-MS,allowedthedeterminationofabout25 elementsatthesub-ppblevelwithonly0.2mlofasamplesolution.Theoptimumsampleflowratewasfoundtobe0.4 mlmin–1_{,alongwithasamplesuctiontimeof30s.Theproposedmethodwasvalidatedbydeterminingtraceelementsin} river-watercertifiedreferencematerial(SLRS-4)issuedbyNationalResearchCouncilCanada.Theanalyticalresultsof theproposedmethodwereingoodagreementwiththecertifiedvalues.Thismethodwassuccessfullyappliedtoahuman hairsample,thevolumeofwhichwas3ml.

(Received November 19, 2007; Accepted February 21, 2008; Published May 10, 2008)



†_{Towhomcorrespondenceshouldbeaddressed.} E-mail:[email protected]

endofaflowline(PTFEtubing,1mmi.d.¥3mmo.d.)was connected to an ICP-MS nebulizer.  The optimized operating parameters ofASSI/ICP-MS are summarized in Table 1.  The microwave oven used was a Multiwave (Anton Paar GmbH, Graz,Austria).

Reagents

Multi-element standard solutions were prepared by diluting mixedstandardsolutionscontaining10mgml–1_{ofmetalionsfor} ICP-MS (XSTC-13: 31 elements and XSTC-1: 16 rare-earth elements,SpexCertiPrepInc.,NJ).Ultrapurewater(18.2MW cm–1_{resistivity)preparedbyanElix3/Milli-Qelementsystem} (Nihon Millipore, Tokyo, Japan) was used throughout. Ultrapure-grade nitric acid (60%) from Kanto Chemicals (Tokyo, Japan) was diluted with ultrapure water to obtain adequateacidconcentrations.

Sample preparation

ACRMofriver-water(SLRS-4)waspurchasedfromNational ResearchCouncilCanada(NRCC).Headhairswerecollected fromaJapanesemale.Afterwashingthemwithwarmwaterand acetone for cleaning the surface, the samples were dried in a desiccator.  Head hairs of 0.110 g (sample A) and 0.015 g (sampleB)wereweighedanddecomposedwitheach3gof0.1M nitricacidusingamicrowaveoven:digestionfor5minat100 W,and15minat800W,andcoolingfor15min(total35min).

Result and Discussion

Comparison of the peak shape obtained with ASSI/ICP-MS and FI/ICP-MS

Although the sample consumption of ASSI/ICP-MS is very little compared with conventional/ICP-MS, the limits of detection (LODs) obtained byASSI/ICP-MS were higher by a factorof2–10fold.Inthecaseofatraceconcentration(about 10 pg g–1_{ level), the reproducibility of ASSI/ICP-MS was} reportedtobeworsethanthatofFI/ICP-MS,whereastheLOD was almost equal.18_{  These disadvantages are caused by using} theareaofthetriangularpeaksignalobtainedinashortsample suctiontimetocalculatetheconcentration.Alow-concentration samplegivesatriangularpeak,anditsareaistoosmalltoobtain reproducible values.  Another reason is an air segment introduced before a sample segment that caused a drift of the baseline.  To improve the reproducibility and accuracy of the measurement, the ASSI conditions were examined.  A longer suctiontimewasexpectedtogiveatrapezoidpeakshapewitha plateau top, but resulted in more sample consumption.  The

effectofthesamplesuctiontimewastestedfrom3to35s.The obtained peak profiles are shown in Fig. 1 along with that obtained by FI/ICP-MS.  A trapezoid-like peak with a comparatively narrow time width was obtained by ASSI/ICP-MS(b),comparedwithatriangularpeakbyFI/ICP-MS(a).In ASSI,diffusionofthesamplezoneissuppressedbyair,anda relatively short suction time is sufficient to obtain a stable trapezoid-likepeak.Therelativestandarddeviation(RSD,%) ofASSI was examined with multi-element solutions of 20, 50 and100pgg–1_{ofeachelementconcentration.Theresultsare} giveninTable2alongwiththatobtainedbyFI/ICP-MS.Ata low concentration of 20 pg g–1_{, ASSI/ICP-MS showed good} reproducibility compared with FI/ICP-MS, though the differenceswerecomparativelysmallerathigherconcentrations (50and100pgg–1_{).Fromtheseresults,theproposedmethod} hassufficientsensitivityandreproducibilitywhentheamountof samplevolumeis1mlorless.

Effect of the sample flow rate on the sensitivity

Toimprovethesensitivityandreproducibilityinthismethod, thesamplesuctiontimeisanimportantfactor,butthelongeris thesamplesuctiontime,thelargeristhesampleconsumption. Theaimistodecreasethesamplevolumedowntobelow1ml. Sincerepeatedmeasurementsarerequired,atleast3times,one measurementmustbedonewithbelow0.3ml.Thesampleflow rate and the sample suction time were examined.  When the nebulizergasflowrateisfixed,theioncountsofICP-MSisnot proportionaltothesampleflowrate.19_{Here,thesamplesuction} time was fixed at 10 s, and the sample flow rate was changed from 0.23 to 0.89 ml min–1_{. A multi-element solution of each} 50pgg–1_{wasanalyzed,andtheintegrationareafor5softhe} central part of the trapezoid-like peak signal was used to calculatetheconcentration.TheresultsforMnandUareshown inFig.2.FromFig.2,whenthesampleflowratewaslessthan 0.35 ml min–1_{, the integration ion counts gradually decreased,} butathigherthan0.35mlmin–1_{theintegrationioncountswere}

Fig. 1 PeakprofilesobtainedbyFI/ICP-MS(a)andASSI/ICP-MS (b) and a flow image in a tubing.  Sample suction time, 35 s; flow rate,1.2mlmin–1_{;sample,U,50pgg}–1_.

Table 1 Fundamental operating parameters for ASSI/ICP-MS ICP-MS: SII NanoTechnology Inc. Model SPQ8000H

Incident power/kW 1.1

Coolant gas Ar 15 l min–1

Carrier gas Ar 0.49 l min–1

Auxiliary gas Ar 1.0 l min–1

Sampling depth/mm 10

ASSI: SII NanoTechnology Inc. Model AT600 autosampler Sample flow rate/ml min–1 _1.2

Sample suction time/s 3 Rinse with 0.1 M HNO3/s 5

Peak integration time/s 10

Dwell time/ms 100

almost constant, and sufficient sensitivity was obtained.  The sampleflowratewasselectedtobe0.4mlmin–1_.

Effect of the sample suction time on the repeatability

Undertheconditionsthatthesampleflowratebefixedto0.4 mlmin–1_{andthesampleinjectiontimebe35s,theintegration} timezoneofthecentralpartofthetrapezoid-likesignalchanged from 5 to 28 s, and the integration ion counts were measured. Thesamplesweremeasured5times;theRSDoftheintegration ioncountsobtainedateachintegrationtimeisplottedinFig.3. Consequently,whentakingtheintegrationtimefor20sormore, the RSD values were reasonable and sufficient. To obtain the integrationtimefor20s,thesamplesuctiontimewasrequired tobeabout30s;thesamplesuctiontimewasselectedtobe30s. Effect of a suction pump on the signal noise

In FI/ICP-MS, to keep the sample and the carrier flow rates constant, a double-plunger pump is usually used to propel the solutions.  In thisASSI/ICP-MS, a peristaltic pump (PP) was usedbecauseofacombinationtotheautosampler.However,PP generated a pulsation that caused noises on the signals.  The performances of three kinds PP were compared with a multi-elementsolutionof100pgg–1_{.TheresultsareshowninFig.4} along with conventional/ICP-MS signals, which did not use a pump.  At this low concentration level, the

conventional/ICP-MSmeasurementalsocausednoisesonthesignal.WithpumpA, whichwasofthenon-variable-pressuretype,itwasimpossible tosuppressthepulsation.Onthecontrary,pumpBorC,which wasofthevariable-pressuretype,couldgivemorestablesignals byoptimizingthepressureappliedonthetube.Afteradetailed comparison,pumpBwasselectedforfollowingexperiments.

The optimized measurement conditions are summarized in Table 3. The sample consumption under these conditions was abletodecreaseto0.2mlbyonemeasurement.

LOD and linearity of the calibration graph

Undertheoptimizedconditions,acalibrationgraphwasmade for about 30 elements.  In this ASSI/ICP-MS, because 15 elements could be measured simultaneously, the measurement was done twice. The RSDs at 3 different concentration levels wereexamined,andthereproducibilityofthemeasurementwas evaluated(Table4).WhentheupperlimitofRSDissettobe 5%, this method is applicable to quantitative analysis at a concentrationlevelof100pgg–1_{fortransitionelementsand50} pgg–1_{forrare-earthelements,respectively.}

The linearity of the calibration graph was shown with the correlationcoefficient(r2_{)giveninTable5.LODwasdefined} as the concentration corresponding to three-times standard Fig. 2 Effect of the sample flow rate on the peak area.  Element

concentration,50pgg–1_{;samplesuctiontime,10s.}

Table 2 RSDa _{(%) at different concentrations obtained by}

FI/ICP-MS and ASSI/ICP-MS

Al 27 28.0 4.2 4.2 1.1 3.4 0.9 V 51 2.1 1.7 1.8 1.6 1.4 1.3 Cr 53 10.0 3.0 5.2 2.9 1.8 2.5 Mn 55 16.8 1.8 3.1 3.2 1.5 0.9 Ni 60 12.6 7.0 2.8 4.6 4.2 2.5 Cu 63 1.3 1.4 1.2 1.1 1.0 0.7 Zn 66 5.9 3.7 5.2 0.9 3.8 0.8 As 75 4.0 6.9 7.5 7.8 2.5 3.5 Ag 107 28.3 4.2 10.0 3.2 10.3 3.2 Cd 111 28.0 8.6 4.2 6.1 4.5 5.4 Hg 202 4.7 6.0 4.4 3.5 3.8 3.3 Tl 205 1.5 6.3 2.7 3.8 1.7 4.6 Pb 208 8.3 3.8 3.7 2.1 3.6 3.2 Bi 209 5.7 3.7 2.2 2.3 2.6 3.8 U 238 7.0 5.4 0.8 2.8 2.6 2.0 Element m/z 20 pg g –1 FI ASSI 50 pg g–1 FI ASSI 100 pg g–1 FI ASSI

a. Relative standard deviation (n = 5).

Fig. 3 Effectofthepeakintegrationtimeattheplateaupartofthe signalonRSD.EachRSD(%)istheaverageoffivemeasurements. Element concentration, 50 pg g–1_{; flow rate, 0.4 ml min}–1_{; sample}

suctiontime,35s.

Fig. 4 Peakprofileswithdifferentperistalticpumps.Flowrate,1.1 ml min–1_{ (conventional/ICP-MS), 0.4 ml min}–1_{ (pump A, B, C);}

sample,Mn,100pgg–1_.

Table 3 Optimized conditions for ASSI/ICP-MS Sample flow rate/ml min–1 _0.4

Sample suction time/s 30

Rinse with 0.1 M HNO3/s 5

Peak integration time/s 20

Dwell time/ms 100

deviation(s)oftheblankintensityfortracemetalscontainedin 0.1MHNO3.TheresultsaresummarizedinTable5,alongwith thoseobtainedbyconventional/ICP-MS.Someelements,such as Al and Zn, were affected by contamination at the concentration range tested (0 – 500 pg g–1_{). The LODs of the} proposed ASSI/ICP-MS with a small volume of the sample solutionweregreatlyimprovedtopracticallyequalthelevelof thoseobtainedbyconventional/ICP-MS.

Validation with a certified reference material (CRM) of a river-water sample

Theproposedmethodwasappliedtoriver-waterCRM(SLRS-4)forvalidation.Theabsolutecalibrationgraphmethodandthe internalstandard(IS)method,inwhicheach1ngg–1_ofSc(for V,CoandNi)andRh(forAsandCd)and5ngg–1_{ofW(forPb} and U) were added as IS elements, were performed.  The analyticalresultsaregiveninTable6.Differencesbetweenthe certifiedandobtainedvaluesbythismethodforAsandVwere due to the influence of spectrum interference with contained polyatomic species of Cl.  In the case of Co and Ni,

compensationbytheinternalstandardelementswaseffective. Application to human hair

Thismethodwasappliedtohumanhairsamples.Headhairs of0.110g(sampleA)and0.015g(sampleB)wereweighedand decomposed with each 3 g of 0.1 M nitric acid using a microwave oven, as shown in the sample preparation section. SampleAwasdilutedto20gwith0.1Mnitricacid,20_whereas sampleBwasanalyzedwithoutdilution.SampleAwastested here as a typical case, in which large amounts of the sample wereavailable,andsampleBwastestedasarepresentativecase in which only small amounts of the sample were available in forensic chemistry.  Trace elements of these samples were measured by the absolute calibration graph method.  The analyticalresultsaregiveninTable7.Traceelementsinsample A(20mlofthefinalsolutionwasavailable)couldbemeasured byconventional/ICP-MS,andtheanalyticalresultswereingood agreement with that obtained by the proposed ASSI/ICP-MS. Ontheotherhand,sincethefinalsolutionwaslimitedtoonly 3 ml in the case of sample B, the concentration could not be Table 4 RSD (%) at different concentrations obtained by ASSI/ICP-MS and conventional/ICP-MS

Al 27 14.1 11.3 3.5 2.1 3.0 4.0 4.0 0.3 1.3 V 51 2.9 3.0 1.7 1.6 2.3 2.0 2.8 1.5 2.3 Cr 53 7.0 3.2 4.9 7.2 4.5 7.7 6.8 1.3 5.9 Mn 55 3.9 12.6 2.8 5.6 1.8 4.5 3.7 1.2 0.7 Ni 60 3.0 30.0 3.7 12.1 6.4 7.8 2.8 1.1 3.0 Cu 63 2.4 7.5 1.0 1.5 1.7 3.2 0.9 1.2 0.9 Zn 66 11.8 32.0 1.6 51.2 3.1 15.8 9.2 2.3 1.4 As 75 18.2 7.6 9.5 13.6 4.0 3.1 9.8 2.0 4.3 Ag 107 8.3 3.8 8.0 5.8 2.3 1.6 2.3 1.8 2.4 Cd 111 21.0 9.3 2.4 10.7 5.1 10.8 6.3 3.3 1.4 Hg 202 13.4 14.5 8.9 11.9 5.2 13.6 7.0 5.3 3.4 Tl 205 13.1 1.8 4.9 6.4 3.0 2.6 3.3 0.5 0.7 Pb 208 2.5 28.7 6.2 4.8 2.3 3.2 2.2 0.6 0.4 Bi 209 6.5 3.6 4.7 1.2 1.3 1.8 2.1 0.8 1.2 U 238 6.3 1.2 1.9 4.9 2.3 1.7 1.8 1.0 2.2 Element m/z 50 pg g–1 ASSI 3 sa _{30 s}b 100 pg g–1 ASSI 3 sa _{30 s}b 500 pg g–1 ASSI 3 sa _{30 s}b

Conventionalc _Conventionalc _Conventionalc

5 pg g–1 ASSI 3 sa _{30 s}b 50 pg g–1 ASSI 3 sa _{30 s}b 100 pg g–1 ASSI 3 sa _{30 s}b

Conventionalc _Conventionalc _Conventionalc

Y 89 15.5 8.4 12.9 4.1 3.0 4.3 2.9 1.3 1.9 La 139 9.4 7.3 4.6 7.0 3.5 3.5 3.2 0.6 2.6 Ce 140 23.2 9.9 12.5 8.6 1.0 3.7 4.6 1.0 4.6 Pr 141 24.6 5.6 6.9 2.7 1.4 2.3 5.8 1.7 1.6 Nd 146 13.2 20.2 12.9 9.0 5.1 5.6 9.3 2.4 3.4 Sm 147 31.2 11.4 18.2 6.8 8.2 6.3 6.5 4.0 3.3 Eu 153 26.6 1.8 12.0 6.1 3.6 2.0 3.1 2.0 2.2 Gd 157 33.7 9.9 3.6 5.7 2.3 2.2 6.5 1.6 2.1 Tb 159 15.1 6.8 0.9 3.1 2.3 1.4 2.3 1.7 2.0 Dy 163 29.0 13.0 13.2 6.0 3.2 4.7 2.7 1.9 0.3 Ho 165 9.6 5.4 4.1 3.2 1.0 4.0 3.4 3.5 1.8 Er 166 21.5 7.2 11.6 7.3 4.4 2.1 4.7 3.4 2.3 Tm 169 13.0 5.3 4.9 4.2 4.0 3.3 3.4 2.1 2.3 Yb 172 34.1 19.2 13.5 18.7 4.7 2.5 7.0 4.4 2.1 Lu 175 7.0 6.7 6.5 4.4 1.3 1.8 3.3 2.2 1.3 Element m/z

measured by conventional/ICP-MS.  However, the proposed method, ASSI/ICP-MS, could measure many trace elements, eveninsuchasmallvolumeofsamples.

References

 1. L.Stroemberg,H.Bergkvist,andE.A.M.K.Edirisinghe, J. Chromatogr.,1983,258,65.

 2. K. Tanaka, T. Ohmori, T. Inoue, and S. Seta, J. Forensic Sci.,1994,39,500.

 3. T. Inoue, K. Tanaka, T. Ohmori,Y. Togawa, and S. Seta, Forensic Sci. Int.,1994,69,97.

 4. T.Kishi,T.Inoue,S.Suzuki,T.Yasuda,T.Oikawa,andT. Niwaguchi,Eisei Kagaku,1983,29,400.

 5. S.Suzuki,S.Hirai,andT.Mitsugashira,Bunseki Kagaku, 1994,43,845.

 6. Y.SuzukiandY.Marumo,Bunseki Kagaku,1993,42,417.  7. Y. Suzuki, M. Kasamatsu, S. Suzuki, K. Ohashi, and A.

Kawase,Bunseki Kagaku,2001,50,335.

 8. S.Muratsu,T.Ninomiya,Y.Kagoshima,andJ.Matsui,J. Forensic Sci.,2002,47,944.

 9. S. Suzuki, Y. Suzuki, H. Ohta, M. Kasamatsu, and T. Nakanishi,Anal. Sci.,2005,21,775.

10. S.Suzuki,H.Tsuchihashi,K.Nakajima,A.Matsushita,and T.Nagao,J. Chromatogr.,1988,437,322.

11. Y.Marumo,T.Inoue,andS.Seta,Forensic Sci. Int.,1994, 69,89.

12. H.LiuandA.Monastery,Anal. Chem.,1994,66,3233. 13. Y. Takaku, Y. Kudo, J. Kimura, T. Hayashi, I. Ota, H.

Hasegawa,andS.Ueda,Bunseki Kagaku,2002,51,539. 14. K.-H. Lee, M. Oshima, and S. Motomizu, J. Flow Inject.

Anal.,2002,19,39. 15. D.Bachman,Analyst,1993,118,815. 16. J.M.CraigandD.Bachman,Analyst,1994,119,1677. 17. K.-H.Lee,M.Oshima,andS.Motomizu,Bunseki Kagaku, 2000,49,529. 18. K.-H.Lee,M.Oshima,T.Takayanagi,andS.Motomizu,J. Flow Inject. Anal.,1999,16,255.

19. J.W.OlesikandL.C.Bates,Spectrochim. Acta,1995,50B, 285.

20. E. Fujimori, K. Inagaki, and H. Haraguchi, Bunseki Kagaku,1999,48,57.

Table 5 LODs for trace metals by ASSI/ICP-MS and conventional/ICP-MS Al 0.2004 0.1994 0.2072 V 0.9998 26.3 0.9993 4.3 0.9993 13.0 Cr 0.9791 133.2 0.9932 38.6 0.9860 15.7 Mn 0.9931 9.1 0.9973 2.5 0.9991 1.8 Ni 0.7547 0.8567 0.9982 11.2 Cu 0.5492 0.5397 0.5218 Zn 0.0230 0.0001 0.4819 As 0.9997 22.8 0.9999 6.3 0.9999 3.9 Ag 0.9996 6.8 0.9996 1.7 0.9998 0.5 Cd 0.9998 8.0 1.0000 2.9 0.9998 2.3 Hg 0.9862 46.1 0.9909 22.9 0.9913 11.5 Tl 1.0000 1.4 1.0000 0.2 0.9999 0.2 Pb 0.8468 0.9270 0.9989 3.3 Bi 0.9999 2.5 0.9999 0.8 1.0000 0.4 U 1.0000 1.7 1.0000 0.6 0.9999 0.3 Y 0.9999 1.90 1.0000 0.11 1.0000 0.17 La 1.0000 0.88 1.0000 0.17 0.9999 0.14 Ce 0.9998 0.88 1.0000 0.13 0.9997 0.28 Pr 0.9993 0.24 1.0000 0.30 0.9998 0.19 Nd 0.9992 2.69 0.9998 0.67 1.0000 0.48 Sm 0.9994 5.22 1.0000 1.52 0.9988 0.78 Eu 0.9999 0.48 1.0000 0.17 0.9998 0.21 Gd 0.9993 2.05 1.0000 0.92 0.9996 0.18 Tb 1.0000 0.62 1.0000 0.16 0.9998 0.04 Dy 0.9988 0.84 1.0000 0.83 0.9997 0.87 Ho 1.0000 0.41 0.9998 0.15 0.9999 0.05 Er 0.9979 0.98 0.9999 0.24 0.9991 0.42 Tm 1.0000 0.51 1.0000 0.19 0.9998 0.14 Yb 0.9995 1.59 1.0000 0.48 0.9995 0.78 Lu 0.9999 0.35 0.9999 0.20 1.0000 0.08 Element ASSI/ICP-MS Suction time, 3 s r2a LOD/ pg g–1 r2a LOD/ pg g–1 Suction time, 30 s Conventional/ ICP-MS LOD/ pg g–1 r2a

a. Correlarion coefficient of calibration graph.

Table 7 Analytical results for human hair by ASSI/ICP-MS

V 53 ± 2 64 ± 3 68 ± 2 Cr 242 ± 5 134 ± 6 209 ± 2 Mn 117 ± 2 78 ± 1 106 ± 7 Ni 120 ± 5 70 ± 3 201 ± 7 As 46 ± 3 30 ± 1 54 ± 2 Sr 707 ± 48 541 ± 3 806 ± 15 Ag 14 ± 0 6 ± 3 19 ± 1 Cd 13 ± 1 6 ± 1 15 ± 2 Ba 132 ± 8 105 ± 2 144 ± 5 Pb 597 ± 31 446 ± 2 595 ± 6 Element Found Reference valuec

Sample Aa _{Sample B}b

Unit, ng g–1_{; n = 3.}

a. A 0.11-g portion of head hairs was decomposed and diluted to 20 g with 0.1 M nitric acid.

b. A 0.015-g portion of head hairs was decomposed to 3 g with 0.1 M nitric acid.

c. Analytical results for sample A obtained by conventional/ICP-MS.

Table 6 Analytical results for the river-water certified reference material (SLRS-4)a V 575 ± 5 114 ± 1 970 ± 30 507 ± 16 320 ± 30 Co 65 ± 1 34 ± 0 44 ± 1 33 ± 1 33 ± 6 Ni 1338 ± 14 725 ± 8 913 ± 22 692 ± 17 670 ± 80 As 803 ± 13 751 ± 13 867 ± 8 897 ± 9 680 ± 60 Cd 16 ± 1 15 ± 1 12 ± 2 13 ± 2 12 ± 2 Pb 86 ± 2 82 ± 2 77 ± 2 83 ± 2 86 ± 7 U 52 ± 1 49 ± 1 53 ± 2 57 ± 3 50 ± 3 Element Certified value Conventional/ICP-MS abs.cal.b _int.std.c ASSI/ICP-MS abs.cal.b _int.std.c Unit, pg g–1_{; n = 3.}

a. National Reseach Council Canada, Saint Lawrence River Standard # 4. b. Absolute calibration graph method.