Justin Chaillot, Samir Kassi, Thibault Clauzel, Marie Pesnin, Mathieu Casado, Amaëlle Landais, Mathieu Daëron 

Joint measurements of the 18O/16O and 17O/16O ratios of carbonate minerals and waters are increasingly used to investigate various geochemical, physical and biological processes. Diverse analytical methods, each of them technically challenging in one way or another, have been developed or refined in recent years to measure oxygen-17 anomalies (Δ17O) with instrumental precisions of 10 ppm or better. A critical underpinning of all these methods is how the international carbonate reference materials currently anchoring the VPDB 18O/16O scale are linked to the primary VSMOW-SLAP scale in (18O/16O, 17O/16O) space. For now, however, substantial systematic discrepancies persist between different groups and methods, even after all measurements are nominally standardized to VSMOW-SLAP.

Here we take advantage of VCOF-CRDS, a novel spectroscopic method combining the ease and simplicity of near-infra-red absorption measurements in pure CO2 with metrological performance competitive with state-of-the-art IRMS techniques, to precisely characterize, based on previously reported equilibrium fractionation factors between water and CO2, the relative triple oxygen isotope compositions of international water standards (VSMOW2, SLAP2, GRESP) and CO2 produced by phosphoric acid reaction of carbonate standards (NBS18, NBS19, IAEA603, IAEA610, IAEA611, IAEA612). The robustness of our results derives from the demonstrated linearity of our measurements (RMSE ≈ 1 ppm), but also from the fact that, when equilibrated with or converted to CO2, all of these reference materials yield analytes with closely comparable oxygen-18 compositions. In light of these observations, we revisit potential causes of the large inter-laboratory discrepancies reported so far. Collectively reconciling the different types of measurements constraining the relative 17O/16O ratios of the two standards most often used to normalize carbonate analyses (NBS18, IAEA603) is a matter of high priority.