The authors measured Pb isotope compositions of seven USGS rock referencestandards, i.e. AGV-1, AGV-2, BHVO-1, BHVO-2, BCR-2, BIR-1/1 and W-2, together with NBS 981 using amicromass isoprobe multi-collector inductivel...The authors measured Pb isotope compositions of seven USGS rock referencestandards, i.e. AGV-1, AGV-2, BHVO-1, BHVO-2, BCR-2, BIR-1/1 and W-2, together with NBS 981 using amicromass isoprobe multi-collector inductively-coupled plasma mass spectrometer (MC-ICP-MS) at theUniversity of Queensland. ^(203)Tl-^(205)Tl isotopes were used as an internal standard to correctfor mass-dependant isotopic fractionation. The results for both NBS 981 and USGS rock standardsAGV-1 and BHVO-1 are comparable to or better than double- and triple-spike TIMS (thermal ionizationmass spectrometry) data in precision. The data for BHVO-2 and, to a lesser extent, AGV-2 and BCR-2are reproducibly higher for ^(206)Pb/^(204)Pb, ^(207)Pb/^(204)Pb and ^(208)Pb/^(204)Pb thandouble-spike TIMS data in the literature. The authors also obtained the Pb isotope data for BIR- 1/1and W-2, which may be used as reference values in future studies. It is found that linearcorrection for Pb isotopic fractionation is adequate with the results identical to those correctedfollowing an exponential law or a power law. Precise ^(207)Pb/^(206)Pb, ^(208)Pb/^(206)Pb and^(208)Pb/^(207)Pb ratios can be acquired for sample solutions with Pb>=1 ppb. However, Pb isotoperatios involving ^(204)Pb (i.e., ^(206)Pb/^(204)Pb, ^(207)Pb/^(204)Pb and ^(208)Pb/^(204)Pb) arereliable for solutions with Pb>=40 ppb. The errors for Pb isotope ratio analysis using the MC-ICP-MSare dominated by errors in the analysis of ^(204)Pb, which is commonly ascribed to the difficultyand imprecise correction for a ^(204)Hg isobaric interference. It is found however that the majorerrors on ^(204)Pb come from the tailings of mass ^(203)Tl and mass ^(205)Tl These mass tailingslead to over-subtraction of the baseline for ^(204)Pb, which is measured at +-0.5 amu on both sidesof mass-204 (i.e., at amu 203.5 and 204.5 respectively). Such errors are insignificant for Pb-richsample solutions (i.e., high Pb/Tl ratios), but can be severe for low-Pb sample solutions whenover-'spiked' with Tl. Experiments in this study suggest that a minimum concentration ratio ofPb/Tl>5 in Tl-'spiked' solutions be required to ensure reliable ^(206)Pb/^(204)Pb, ^(207)Pb/^(204)Pband ^(208)Pb/^(204)Pb isotopic ratios. The tailings of ^(203)Tl and ^(205)Tl can also lead toover-subtraction of baselines for ^(202)Hg (at amu 202.5) and ^(206)Pb (at amu 205.5). Therefore,the elegance of using ^(203)Tl and ^(205)Tl isotopes for mass fractionation correction becomes asevere problem in low-Pb rock solution-caution is required. Alternative internal standards for massfractionation correction may be considered. Of course, significant instrumental refinement inabundance sensitivity is in demand.展开更多
文摘The authors measured Pb isotope compositions of seven USGS rock referencestandards, i.e. AGV-1, AGV-2, BHVO-1, BHVO-2, BCR-2, BIR-1/1 and W-2, together with NBS 981 using amicromass isoprobe multi-collector inductively-coupled plasma mass spectrometer (MC-ICP-MS) at theUniversity of Queensland. ^(203)Tl-^(205)Tl isotopes were used as an internal standard to correctfor mass-dependant isotopic fractionation. The results for both NBS 981 and USGS rock standardsAGV-1 and BHVO-1 are comparable to or better than double- and triple-spike TIMS (thermal ionizationmass spectrometry) data in precision. The data for BHVO-2 and, to a lesser extent, AGV-2 and BCR-2are reproducibly higher for ^(206)Pb/^(204)Pb, ^(207)Pb/^(204)Pb and ^(208)Pb/^(204)Pb thandouble-spike TIMS data in the literature. The authors also obtained the Pb isotope data for BIR- 1/1and W-2, which may be used as reference values in future studies. It is found that linearcorrection for Pb isotopic fractionation is adequate with the results identical to those correctedfollowing an exponential law or a power law. Precise ^(207)Pb/^(206)Pb, ^(208)Pb/^(206)Pb and^(208)Pb/^(207)Pb ratios can be acquired for sample solutions with Pb>=1 ppb. However, Pb isotoperatios involving ^(204)Pb (i.e., ^(206)Pb/^(204)Pb, ^(207)Pb/^(204)Pb and ^(208)Pb/^(204)Pb) arereliable for solutions with Pb>=40 ppb. The errors for Pb isotope ratio analysis using the MC-ICP-MSare dominated by errors in the analysis of ^(204)Pb, which is commonly ascribed to the difficultyand imprecise correction for a ^(204)Hg isobaric interference. It is found however that the majorerrors on ^(204)Pb come from the tailings of mass ^(203)Tl and mass ^(205)Tl These mass tailingslead to over-subtraction of the baseline for ^(204)Pb, which is measured at +-0.5 amu on both sidesof mass-204 (i.e., at amu 203.5 and 204.5 respectively). Such errors are insignificant for Pb-richsample solutions (i.e., high Pb/Tl ratios), but can be severe for low-Pb sample solutions whenover-'spiked' with Tl. Experiments in this study suggest that a minimum concentration ratio ofPb/Tl>5 in Tl-'spiked' solutions be required to ensure reliable ^(206)Pb/^(204)Pb, ^(207)Pb/^(204)Pband ^(208)Pb/^(204)Pb isotopic ratios. The tailings of ^(203)Tl and ^(205)Tl can also lead toover-subtraction of baselines for ^(202)Hg (at amu 202.5) and ^(206)Pb (at amu 205.5). Therefore,the elegance of using ^(203)Tl and ^(205)Tl isotopes for mass fractionation correction becomes asevere problem in low-Pb rock solution-caution is required. Alternative internal standards for massfractionation correction may be considered. Of course, significant instrumental refinement inabundance sensitivity is in demand.
