PP52B-03
Enhancing the Accuracy of Carbonate δ18O and δ13C Measurements by SIMS

Friday, 18 December 2015: 10:50
2012 (Moscone West)
Ian J Orland1, Reinhard Kozdon2, Ben Linzmeier1, Jody Wycech3, Maciej Sliwinski1, Kouki Kitajima4, Noriko Kita5 and John W Valley1, (1)University of Wisconsin Madison, WiscSIMS, Dept. of Geoscience, Madison, WI, United States, (2)Lamont Doherty Earth Observatory, Palisades, NY, United States, (3)University of Wisconsin Madison, Madison, WI, United States, (4)University of Wisconsin, Madison, WI, United States, (5)UW-Madison, Madison, WI, United States
Abstract:
The precision and accuracy of carbonate δ18O & δ13C analysis by multicollector SIMS is well established if standards match samples in structure and major/minor element chemistry. However, low-T- and bio-carbonates used to construct paleoclimate archives can include complex internal structures and some samples analyzed at WiscSIMS (and other SIMS labs) have a consistent, sample-dependent offset between average SIMS δ18O measurements and bulk δ18O analyses by phosphoric-acid digestion. The offset is typically <1‰, but recent work has discovered samples where the offset is greater — up to 1.8‰ (average SIMS δ18O values < corresponding conventional measurements). Notably, δ13C offsets have not been observed even in samples with a δ18O offset.

We conducted tests to characterize the δ18O offset in different low-T carbonate materials. Multiple potential causes were examined: perhaps the measured offset is real and conventional analyses include material that SIMS excludes (and vice versa); analytical errors and inter-lab (mis)calibration; depth-profiling effects; porosity; and the effects of variable minor element composition. One explanation implicates water and/or organic matter within carbonate that is ionized during SIMS analysis, but sometimes removed for bulk analysis. Two diagnostic tools help monitor such contaminants during SIMS analysis: 1) simultaneous measurement of [16O1H], and 2) secondary ion yield. Offsets of 0.3 to 1.8‰ in δ18O correlate to [16O1H] for 7 studies of Nautilus, foraminifera, pteropods and speleothems. Offsets were not observed in all foraminifera. For Nautilus, foraminifera, otoliths, and speleothems we also tested pre-treatment techniques (e.g. vacuum roasting, hydrogen peroxide), for which there is no agreed procedure in conventional bulk analyses. For SIMS analyses, pre-treatments had varied influence on the δ18O value, [16O1H], the concentration of “organic markers” like 12C14N and 31P, and mineralogy (of aragonite samples).

<< Previous Abstract | Next Abstract >>