1、Designation: E 2330 04Standard Test Method forDetermination of Trace Elements in Glass Samples UsingInductively Coupled Plasma Mass Spectrometry (ICP-MS)1This standard is issued under the fixed designation E 2330; the number immediately following the designation indicates the year oforiginal adoptio
2、n or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. Asuperscript epsilon (e) indicates an editorial change since the last revision or reapproval.1. Scope1.1 This test method covers a procedure for elementalanalysis of magnesium (Mg
3、), aluminum (Al, titanium (Ti),manganese (Mn), gallium (Ga), rubidium (Rb), strontium (Sr),zirconium (Zr), antimony (Sb), barium (Ba), lanthanum (La),cerium (Ce), samarium (Sm), hafnium (Hf) and lead (Pb) in thebulk of glass samples (irregularly shaped and as small as 200micrograms) for the comparis
4、on of fragments of a knownsource to the recovered fragments from a questioned source.These elements are present in soda lime and borosilicate glassin ngg-1to % levels. Alternative methods for the determinationof elemental analysis of glass are listed in the references. Thisprocedure may also be appl
5、icable to other elements, other typesof glass, or other concentration ranges with appropriate modi-fications of the digestion procedure (if needed for full recoveryof the additional elements) and the mass spectrometer condi-tions. The addition of calcium and potassium, for example,could be added to
6、the list of analytes in a modified analysisscheme.1.2 This standard does not purport to address all of thesafety concerns, if any, associated with its use. It is theresponsibility of the user of this standard to establish appro-priate safety and health practices and determine the applica-bility of r
7、egulatory limitations prior to use.2. Summary of Test Method2.1 The glass fragments are digested using a mixture ofhydrofluoric, nitric and hydrochloric acids. Following aciddigestion, the samples are taken to dryness to eliminate most ofthe silicate matrix and the excess acids. Then an internalstan
8、dard is added as the samples are reconstituted in nitric acid.Dilutions may be utilized to quantitate those elements that arepresent in higher concentrations.2.2 An inductively coupled plasma mass spectrometer isused to measure the concentrations of the identified elements(1.1). The instrument shoul
9、d be adjusted for maximum sensi-tivity, best precision and to minimize oxides and doublycharged ion interferences. The instrument is then calibrated permanufacture recommendations using multi-elemental calibra-tion standards with the same internal standards (Rh and Sc) asthat those added to the samp
10、les.2.3 Reagent blanks are measured along with the samples asdetection limits are usually limited by the background signalsgenerated by the reagent blanks. The limits of detection of themethod are expected to be between 0.5 ngg-1and 25 ngg-1insolution for most elements.3. Significance and Use3.1 Thi
11、s technique is destructive, in that the glass fragmentsmay need to be crushed, and must be digested in acid.3.2 Although the concentration ranges of the calibrationcurves shown in Appendix X1 are applicable to soda lime andborosilicate glass, this method is useful for the accuratemeasurement of trac
12、e elements from a wide variety of glasssamples.3.3 The determination the elemental concentrations in glassallows for additional data that can be compared betweenfragments. A standardized, validated method can aid in theinterchange of data between laboratories.3.4 It should be recognized that the met
13、hod measures thebulk concentration of the target elements. Any extraneousmaterial present on the glass that is not removed beforedigestion may result in a measurably different concentration ofthe elements.3.5 The precision and accuracy of the method should beestablished in each laboratory that emplo
14、ys the method. Con-fidence intervals or a similar statistical quality statement shouldbe quoted along with any reported concentration value.4. Apparatus4.1 ICP-MSAn ICP-MS instrument is employed. Sincethere are many manufacturers, the specification of the instru-ment should be noted within the analy
15、sis results.4.2 The instrument should be tuned prior to the analysisusing the manufacturers recommendations covering the massrange of the identified elements. The instrument should beadjusted for maximum sensitivity, best precision and to mini-mize oxides and doubly charged ion interferences. The in
16、stru-ment is then calibrated per manufacture recommendations1This test method is under the jurisdiction of ASTM Committee E30 on ForensicSciences and is the direct responsibility of Subcommittee E30.01 on Criminalistics.Current edition approved Aug. 1, 2004. Published August 2004.1Copyright ASTM Int
17、ernational, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.using multi-elemental calibration standards with the sameinternal standard as that added to the samples.4.3 ReplicatesThe minimum number of measurementreplicates per sample should be three with considerat
18、ion foradditional measurements, if practicable4.4 Standard Reference GlassesA minimum of two dif-ferent reference glasses of known elemental compositionshould be used.5. Calibration5.1 Two calibration curves as well as two check standardsare used (Appendix X1). The calibration standards must beNIST
19、traceable.5.2 For the trace element standards calibration curve, Rh (50ngg-1) is used as an internal standard and the elements aregrouped according to the expected concentrations. The firstgroup consists of Mg, Ti, Mn, Sr, Zr, Sb, Ba and Pb with aconcentration range of 0 to 150 ngg-1. The second gro
20、upconsists of Ga, Rb, La, Ce, Sm and Hf with a concentrationrange of 0 to 50 ngg-1ppb.5.3 The check standard for the calibration is 50 ngg-1for thefirst group and 5 ngg-1for the second group.5.4 The high standards calibration curve has Sc (60 ngg-1)as an internal standard and is composed of three co
21、ncentrationslevels of Mg, Al, and Ba from 0 to 150 ngg-1.5.5 The check standard for the high standard calibrationcurve is 60 ngg-1in all elements including Sc.5.6 The system calibration must be checked daily (or on thedays the instrument is in use for analysis) and prior to theperformance of an anal
22、ysis, as well as during the analysis afterevery ten samples. This is accomplished by the analysis of thecheck standards as a continuing calibration verification (CCV).5.7 The system is recalibrated any time that the control fallsoutside the acceptable parameters established by the laboratoryor analy
23、st for this procedure (10 % tolerance is recommended).5.8 Method Detection Limit (MDL) and Limit of Quantita-tion (LOQ)The limits of detection of the method (MDL)aredetermined for each element by measuring the three procedureblanks on two non-consecutive days. Multiply by three thestandard deviation
24、 (three instrumental replicates) of the mea-sured intensities calculated by the calibration curve for thatelement in the day that it is measured. To calculate the limits ofquantitation for the method (LOQ), multiply by ten thestandard deviation (three instrumental replicates) of the mea-sured intens
25、ities calculated by the calibration curve for thatelement in the day that it is measured. The measured intensitiesmust be converted to concentration units using the appropriatecalibration curve and the standard deviations calculated fromthe concentrations. To calculate these limits of detection andq
26、uantitation, the average from the results for the two days istaken.NOTE 1Table X2.1 illustrates the values for MDL, LOQ and expectedvalues for a set of 50 soda-lime glass samples all expressed as ngg-1of theelemental concentration in the final solution6. Procedure6.1 Prior to crushing the glass samp
27、le for the digestion,wash samples separately in MeOH for 10 min and rinse 3 timeswith high purity water (18 MV-cm). Wash in 1.6 molL-1HNO3for 30 min, rinse 3 times with high purity water andair-dry overnight.6.2 The samples are crushed between clean polystyreneweighing boats using a pestle, taking c
28、are not to puncture theboats.6.3 Approximately 2 to 3 mg of each sample should beaccurately weighed (with a precision of 6 1g or better) andquantitatively transferred into a labeled polypropylene test tubewith a cap. Each sample should be weighed in triplicate forthree determination for each glass e
29、xhibit.with a precision of 61 g or better.6.4 All volumes are delivered using pipetters. The followingmixture is added to each sample, and standards: 150 L ofconcentrated HNO3(16 molL-1), 300 L of concentrated HF(29 molL-1), and 150 L of concentrated HCl (12 molL-1). Thisacid mixture is also used to
30、 prepare the reagent blanks.NOTE 2All reagents are at least trace metal quality for ICP-MS.NOTE 3Normally, the mixture turns pale yellow, and if not, the acidreagents may have lost their strength and should be replaced before addingthem to the samples.6.5 The tubes are capped, vortex mixed, and plac
31、ed in anultrasound bath to assist in the digestion. The tubes are thenuncapped and placed in a dry bath block or an oven, at 80C (orgreater but below the temperature of the softening of thedigestion tubes), and taken to dryness (about 24 h).6.6 The samples are reconstituted by adding 800- L ofHNO3(4
32、.0 molL-1), 20 L of an internal standard Rh stocksolution (10 gg-1in HNO30.8 molL-1) and 680 L of highpurity water and the tubes are capped.6.7 The tubes are vortex mixed and left to stand overnight.6.8 A 2.500 mL volume of high purity water is added, thetubes are capped and vortex mixed. A 50 L ali
33、quot is removedand the remaining digest solution (undiluted) is analyzed usingthe trace element standardscalibration curve (or curves).NOTE 4The 50 L aliquot of the above solution is transferred to apolypropylene test tube. A 30 L volume of a scandium (Sc) internalstandard (10 gg-1 in 0.8 molL-1 HNO
34、3) and 4.920 ml of 0.8 molL-1HNO3 are added. The solution is vigorously mixed before analysis. Thissecond dilution is analyzed for magnesium, aluminum, and barium withthe high standards calibration curve.7. Precision and Bias7.1 An interlaboratory study was conducted in 2001. Eachof four laboratorie
35、s tested four standard reference glasses using5 replicate sample measurements of NIST 612, NIST 614,NIST 621 and NIST 1831.7.2 The bias and precision results for each of the glasses aretabulated in Appendix X3. The terms repeatability and repro-ducibility are used as specified in Practice E 177. The
36、 95 %limits were calculated by multiplying the respective standarddeviations by a factor of 2.8.8. Keywords8.1 forensic science; glass comparisons; glass measurement;trace elemental analysisE2330042APPENDIXES(Nonmandatory Information)X1. ISOTOPESX1.1 Trace ElementsX1.1.1 For trace calibration standa
37、rds, Rh (50 ngg-1) is usedas an internal standard. The isotopes are specified in TablesX1.1 and X1.2.X1.2 High StandardsX1.2.1 For calibration standards, Sc (60 ngg-1) is used as aninternal standard. The isotopes are specified in Table X1.3.TABLE X1.1 Group I:25Mg,26Mg,47Ti,55Mn,86Sr,88Sr,90Zr,91Zr,
38、121Sb,137Ba,206Pb,207Pb,208PbCalibrationStandardCalibration StandardFinal Concentration,ngg-1S00.0000S11.000S210.00S350.00S475.00S5150.0TABLE X1.2 Group II:71Ga,85Rb,139La,140Ce,147Sm,178HfCalibrationStandardCalibration StandardFinal Concentration,ngg-1S00.0000S10.100S20.500S31.000S45.00S550.0TABLE
39、X1.325Mg,26Mg,27Al,137BaCalibrationStandardCalibrationStandardFinalConcentration,ngg-1HS00.000HS150.00HS275.00HS3150.0E2330043X2. LIMITSX3. BIAS AND PRECISIONTABLE X2.1 Expected Limits of Detection (MDL and LOQ)for theElements Described in this MethodElementLimit ofDetectionMDL, nggLimit ofQuantitat
40、ion,LOQ, nngRange ofConcentrations,A(nng in solution for2 to 4 mg samples)Mg 7.5 28 10 3 103-33 104Al 5.0 16 dl - 40Ti 0.18 0.60 40 - 240Mn 0.17 0.57 dl - 350Ga 0.07 0.24 dl - 2Rb 0.04 0.14 dl - 30Sr 0.06 0.20 10 - 75Zr 0.91 3.02 dl - 150Sb 0.099 0.33 dl (for non boro-silicate glass)Ba 0.04 0.13 0.9
41、 - 50La 0.023 0.08 dl - 11Ce 0.055 0.18 dl - 75Sm 0.01 0.035 dl - 0.5Hf 0.15 0.51 dl - 4Pb 0.05 0.19 dl - 40AFrom actual measurement of a set of 50 soda-lime glass samples from vehiclewindows, total ranges could include higher concentrations of some elements.TABLE X3.1 Bias and Precision Found in SR
42、M NIST 612 Obtained from the Interlaboratory StudyElementReported Value,gg-1Average,gg-1Bias, %Repeatability-within,sr(%)Reproducibility-between,sR(%)RepeatabilityLimit-within,r (%)ReproducibilityLimit-between,R (%)Ti 50.1A49.93 -0.34 5.5 18 15 51Mn 39.6A39.58 -0.05 3.7 3.2 10 9.1Ga 36.24B36.69 1.2
43、4.0 7.3 11 21Rb 31.4C32.43 3.3 2.2 12 6.1 33Sr 78.4C76.55 -2.4 2.2 12 6.3 33Zr 35.99B40.91 14 4.6 13 13 35Sb 38.44B36.46 -5.2 3.7 8.9 10 25Ba 41A40.46 -1.3 2.7 5.4 7.6 15Ce 39A39.32 0.82 2.8 6.5 7.9 18Sm 39A39.36 0.92 2.6 5.4 7.3 15Hf 34.77B38.64 11 3.9 9.2 11 26Pb 38.57B37.72 -2.2 5.6 5.9 16 17ARep
44、orted in Ref (17).BCertified by NIST.CValues given by NIST for information only.E2330044REFERENCES(1) Duckworth, D. C., Bayne, C. K., Morton, S. J., and Almirall, J. R.,“Analysis of Variance in Forensic Glass Analysis by ICP-MS: VarianceWithin the Method,” J. of Analytical and Atomic Spectrometry, 1
45、5(7),2000, pp. 821-828.(2) Montero, S., Hobbs, A., French, T., and Almirall, J. R., “ElementalProfiling of Glass Fragments by ICP-MS as Evidence of Association:Analysis of a Case,”Journal of Forensic Sciences2003, 48(5)11011107.(3) Duckworth, D. C., Morton, S. J., Bayne, C. K., Montero, S., Koons, R
46、.D., and Almirall, J. R., “Forensic Glass Analysis by ICP-MS: AMulti-Element Assessment of Discriminating Power Via Analysis ofVariance (ANOVA) and Pair-Wise Comparisons,” Journal of Analyti-cal and Atomic Spectrometry 2002. 17 (7), 662668.(4) Almirall, J. R., “Glass as Evidence of Association,” Cha
47、pter in MuteWitness; When Trace Evidence Makes the Case, Edited by MaxHouck, Academic Press, 2001.TABLE X3.2 Bias and Precision Found in SRM NIST 614 Obtained from the Interlaboratory StudyElementReported Value,gg-1Average,gg-1Bias, %Repeatability-within,sr(%)Reproducibility-between,sR(%)Repeatabili
48、tyLimit-within,r (%)ReproducibilityLimit-between,R (%)Ti 3.1A9.41 204 18 56 50 157Mn 4.41B20 - 55 -Ga 1.3A1.12 -14 3.7 4.3 10 12Rb 0.855C0.90 5.3 2.4 9.8 6.7 27Sr 45.8C45.33 -1.0 2.0 13 5.5 35Zr 7.40B90 - 251 -Sb 1.06A5.78B445 131 - 367 -Ba 3.72 9.1 16 25 45Ce 0.75 3.0 13 8.4 38Sm 0.83 4.4 5.6 12 16
49、Hf 0.80 5.2 8.6 15 24Pb 2.32C2.78 20 17 17 48 47AValues given by NIST for information only.BResults from a single lab.CCertified by NIST.TABLE X3.3 Bias and Precision Found in SRM NIST 621 Obtained from the Interlaboratory StudyElementReported Value,gg-1Average,gg-1Bias, %Repeatability-within,sr(%)Reproducibility-between,sR(%)RepeatabilityLimit-within,r (%)ReproducibilityLimit-between,R (%)Ti 84A90.97 8.3 3.6 18 10 52Mn 22.09 14 19 39 53Ga 2.03 3.6 5.0 10 13Rb 38.47 2.1 12 5.9 33Sr 106 2.5 11 7.0 29Zr 52A67.45 30 3.0 12 8.3 34Sb 1.90B13 - 36 -Ba 1070A1005 -
