Abstract Details
(2020) Interlaboratory Calibrations and Methodological Comparisons for the Re-Os Isotope System
Hnatyshin D, Van Acken D, Creaser R, Toma J, Johnson S & Hitzman M
https://doi.org/10.46427/gold2020.1036
The author has not provided any additional details.
06n: Room 2, Wednesday 24th June 08:18 - 08:21
Danny Hnatyshin
View abstracts at 7 conferences in series
David Van Acken View abstracts at 13 conferences in series
Robert Creaser View all 2 abstracts at Goldschmidt2020 View abstracts at 5 conferences in series
Jonathan Toma View all 2 abstracts at Goldschmidt2020 View abstracts at 5 conferences in series
Sean Johnson View abstracts at 5 conferences in series
Murray Hitzman View all 2 abstracts at Goldschmidt2020 View abstracts at 2 conferences in series
David Van Acken View abstracts at 13 conferences in series
Robert Creaser View all 2 abstracts at Goldschmidt2020 View abstracts at 5 conferences in series
Jonathan Toma View all 2 abstracts at Goldschmidt2020 View abstracts at 5 conferences in series
Sean Johnson View abstracts at 5 conferences in series
Murray Hitzman View all 2 abstracts at Goldschmidt2020 View abstracts at 2 conferences in series
Listed below are questions that have been submitted by the community that the author will try and cover in their presentation. To submit a question, ensure you are signed in to the website. Authors or session conveners approve questions before they are displayed here.
Submitted by Yuri Amelin on Wednesday 24th June 05:46
1) What is the cause of the dispersion of your 185Re/187Re ratios measured with MC-ICPMS, and their deviation from the TIMS value? 2) What was the reason to choose MC-ICPMS as an analytical method for Re-Os analysis rather than TIMS? Which method would you choose if the instruments of both kinds were available?
1) The exact cause is unclear, some sort of mass bias effect in the instrument due to complicated nature of the plasma environment in the ICPMS. The problem with Re is that there is no internal way to calibrate since there are only 2 isotopes There are a few techniques (standard bracketing, and correction with tugsten isotopes) that we used to correct our data, however it appears how well they work varies on each individual analysis/day (from how the standards look). However, in the large community there are still ways people are tweaking corrections to get more accurate data. We assume the TIMS values are more accurate since the corrections that are applied are a bit more straightforward and understood. 2) For us, our blank values for TIMS analysis are still a bit too high to warrant using it yet for samples, so we wanted to see if our ICPMS system would be adequate while we improve blanks. Also, since more groups are starting to MC-ICPMS to measure Re, it is rarely addressed whether the ratio's can be considered accurate, since it is often assumed that the corrections (e.g. W, standards) are sufficient, which I think we show will not always be the case. If both instruments are available I think it will be always be safer to use TIMS data. Although you could use a sensitivity analysis on datasets to see if the extra uncertainty in a ICPMS measurement actually matters. I plan to measure on both instruments for the foreseeable future just to get a larger datasets to see long term variations. I hope this was clear, I am willing to disuses anything in more detail if needed!
1) What is the cause of the dispersion of your 185Re/187Re ratios measured with MC-ICPMS, and their deviation from the TIMS value? 2) What was the reason to choose MC-ICPMS as an analytical method for Re-Os analysis rather than TIMS? Which method would you choose if the instruments of both kinds were available?
1) The exact cause is unclear, some sort of mass bias effect in the instrument due to complicated nature of the plasma environment in the ICPMS. The problem with Re is that there is no internal way to calibrate since there are only 2 isotopes There are a few techniques (standard bracketing, and correction with tugsten isotopes) that we used to correct our data, however it appears how well they work varies on each individual analysis/day (from how the standards look). However, in the large community there are still ways people are tweaking corrections to get more accurate data. We assume the TIMS values are more accurate since the corrections that are applied are a bit more straightforward and understood. 2) For us, our blank values for TIMS analysis are still a bit too high to warrant using it yet for samples, so we wanted to see if our ICPMS system would be adequate while we improve blanks. Also, since more groups are starting to MC-ICPMS to measure Re, it is rarely addressed whether the ratio's can be considered accurate, since it is often assumed that the corrections (e.g. W, standards) are sufficient, which I think we show will not always be the case. If both instruments are available I think it will be always be safer to use TIMS data. Although you could use a sensitivity analysis on datasets to see if the extra uncertainty in a ICPMS measurement actually matters. I plan to measure on both instruments for the foreseeable future just to get a larger datasets to see long term variations. I hope this was clear, I am willing to disuses anything in more detail if needed!
Submitted by Bryant Ware on Wednesday 24th June 07:14
When are talking about correcting to a TIMS dataset, is this measuring a subset of your unknowns on a TIMS to use for a correction or using TIMS and then MC-ICPMS data of the spike/standards to correct the unknowns?
If you are going to use ICPMS data for Re I think what you would want to do is run many samples on both instruments to see how well the ICPMS instrument agrees with TIMS (which is assumed to be correct). So in our case, for a limited dataset, it appears our MC-ICPMS data reproduces the TIMS data to within 0.12%. Therefore I would add another 0.12% correction on top of our other MC-ICPMS data correction. Just to clear up what we actually did and plotted. Essentially I did the Re chemistry to completion on several types of samples (shale samples, shale standard, spike solution, Re standard solution) and then split the final solution into 2 cuts, one to be measured on a TIMS (at the University of Alberta) and another on a Neptune MC-ICPMS (University College Dublin). Ideally the 185/187 ratio's we measured should be identical. However, when I plotted them against each other they ratio's do not overlap with each other. What we are trying to show is that the different methods produce different results (even after all corrections), which could affect a dataset in a negative way. In general, if you have TIMS data, you would typically just use that and not worry about the complications associated with MC-ICPMS data. However, if you are using a ICPMS system you would want to measure (long-term) how much the data you measure varies around TIMS data. Let me know if any other clarifications are needed!
When are talking about correcting to a TIMS dataset, is this measuring a subset of your unknowns on a TIMS to use for a correction or using TIMS and then MC-ICPMS data of the spike/standards to correct the unknowns?
If you are going to use ICPMS data for Re I think what you would want to do is run many samples on both instruments to see how well the ICPMS instrument agrees with TIMS (which is assumed to be correct). So in our case, for a limited dataset, it appears our MC-ICPMS data reproduces the TIMS data to within 0.12%. Therefore I would add another 0.12% correction on top of our other MC-ICPMS data correction. Just to clear up what we actually did and plotted. Essentially I did the Re chemistry to completion on several types of samples (shale samples, shale standard, spike solution, Re standard solution) and then split the final solution into 2 cuts, one to be measured on a TIMS (at the University of Alberta) and another on a Neptune MC-ICPMS (University College Dublin). Ideally the 185/187 ratio's we measured should be identical. However, when I plotted them against each other they ratio's do not overlap with each other. What we are trying to show is that the different methods produce different results (even after all corrections), which could affect a dataset in a negative way. In general, if you have TIMS data, you would typically just use that and not worry about the complications associated with MC-ICPMS data. However, if you are using a ICPMS system you would want to measure (long-term) how much the data you measure varies around TIMS data. Let me know if any other clarifications are needed!
Sign in to ask a question.