Toward a universal carbonate clumped isotope calibration: Diverse synthesis and preparatory methods suggest a single temperature relationship

Abstract

Carbonate clumped isotope (Δ47) thermometry has been applied to a wide range of problems in earth, ocean and biological sciences over the last decade, but is still plagued by discrepancies among empirical calibrations that show a range of Δ47-temperature sensitivities. The most commonly suggested causes of these discrepancies are the method of mineral precipitation and analytical differences, including the temperature of phosphoric acid used to digest carbonates. However, these mechanisms have yet to be tested in a consistent analytical setting, which makes it difficult to isolate the cause(s) of discrepancies and to evaluate which synthetic calibration is most appropriate for natural samples. Here, we systematically explore the impact of synthetic carbonate precipitation by replicating precipitation experiments of previous workers under a constant analytical setting. We (1) precipitate 56 synthetic carbonates at temperatures of 4–85 °C using different procedures to degas CO2, with and without the use of the enzyme carbonic anhydrase (CA) to promote rapid dissolved inorganic carbon (DIC) equilibration; (2) digest samples in phosphoric acid at both 90 °C and 25 °C; and (3) hold constant all analytical methods including acid preparation, CO2 purification, and mass spectrometry; and (4) reduce our data with 17O corrections that are appropriate for our samples. We find that the CO2 degassing method does not influence Δ47 values of these synthetic carbonates, and therefore probably only influences natural samples with very rapid degassing rates, like speleothems that precipitate out of drip solution with high pCO2. CA in solution does not influence Δ47 values in this work, suggesting that disequilibrium in the DIC pool is negligible. We also find the Δ47 values of samples reacted in 25 and 90 °C acid are within error of each other (once corrected with a constant acid fractionation factor). Taken together, our results show that the Δ47-temperature relationship does not measurably change with either the precipitation methods used in this study or acid digestion temperature. This leaves phosphoric acid preparation, CO2 gas purification, and/or data reduction methods as the possible sources of the discrepancy among published calibrations. In particular, the use of appropriate 17O corrections has the potential to reduce disagreement among calibrations. Our study nearly doubles the available synthetic carbonate calibration data for Δ47 thermometry (adding 56 samples to the 74 previously published samples). This large population size creates a robust calibration that enables us to examine the potential for calibration slope aliasing due to small sample size. The similarity of Δ47 values among carbonates precipitated under such diverse conditions suggests that many natural samples grown at 4–85 °C in moderate pH conditions (6–10) may also be described by our Δ47-temperature relationship.