BrF5-techniqtue for the Oxygen Isotopic Analysis of Silicates and Water

Publisher Website
Google Scholar
Abstract
The oxygen of quartz, feldspar, biotite and water was extracted as molecular oxygen using BrF5 and by essentially the same procedures as described by Clayton and Mayeda(1963)and by O'Neil and Epstein(1966). A quantitative conversion of the oxygen gas into carbon dioxide for the mass spectrometric analysis was readily attained by spectrographic graphite wound by platinum wire and heated at 700°C in an externally resistance-heated quartz tube. The platinum catalyst is essential to complete the rapid conversion. The reproducibility of the isotopic analysis of the same tank oxygen by this conversion unit was better than 0.1‰ at the 95% confidence interval over a period of one year(Table2.). The isotopic analyses of standard quartzs, MSQ-1 and -2 and standard waters, MSA-1 and -2 reproduced with in 0.2‰(Tables.4 and 8). On our SMOW scale which is based on the δ18O SMOW of MSA-2 and the isotopic fractionation factor, 1.0407(O'Neil and Epstein, 1966), at 25°C between water and CO2, the δ18O SMOW values of MSQ-1 and -2 are +15.25 and +8. 85‰, respectively. The same quartz samples were analyzed to be +15.08 and +8.70‰ by Clayton's laboratory in Chicago(Table 5). MSA-1 and -2 were also analyzed by CO2-H2O isotopic equihbration method, giving the difference between the two waters to be 7.52‰. This is in close agreement with 7.38‰ given by the BrF5method. The isotopic fractionation factor between CO2 and water at the equilibration temperature(25°C)was calculated from our data to be 1.0412. The observed isotopic ratios of biotite showed a variation with grain size of the sample powder and the preheating scheme(Table 7). Satisfactory results were obtained by using sample from 100 to 150 meshes and by 3 hr-preheating at 200°C. The reproducibility and accuracy of the isotopic results of silicates greatly depend on the history of the nickel reaction tubes in which samples are attacked by BrF5. A brand new tube often does not give a correct answer until after quite a few number of samples are processed in it(Table 3). Our experience suggests that repeated attacks by BrF5 of the walls at a higher temperature than usual reaction temperatures, each followed by nitric acid wash, can eliminate quickly the causes of the instable results.

This publication has 0 references indexed in Scilit: