Calibration of the dolomite clumped isotope thermometer from 25 to 350 °C, and implications for a universal calibration for all (Ca, Mg, Fe)CO3 carbonates

Magali Bonifacie, Damien Calmels, John M. Eiler, Juske Horita, Carine Chaduteau, Crisogono Vasconcelos, Pierre Agrinier, Amandine Katz, Benjamin H. Passey, John M. Ferry, Jean Jacques Bourrand

Research output: Contribution to journalArticlepeer-review

105 Scopus citations


Carbonate clumped isotope thermometry is based on the temperature-dependent formation of 13C18O16O22− ion groups within the lattice of solid carbonate minerals. At low temperatures the bonds between rare, heavy 13C and 18O isotopes are thermodynamically favored, and thus at equilibrium they are present in higher than random abundances. Here we calibrate the use of this temperature proxy in a previously uncalibrated carbonate phase — dolomite [CaMg(CO3)2] — over a temperature range that extends to conditions typical of shallow crustal environments, by determining the Δ47 values of CO2 extracted from synthetic or natural (proto)dolomites grown at known temperatures from 25 to 350 °C and analyzed in two different laboratories using different procedures for sample analysis, purification and post-measurement data treatment. We found that the Δ47-1/T2 dependence for (proto)dolomite is linear between 25 and 350 °C, independent of their Mg/Ca compositions or cation order (or the laboratory in which they were analyzed), and offset from, but parallel to, the theoretical predictions of the Δ63 dependence to temperature of the abundance of the 13C18O16O2 isotopologue inside the dolomite and calcite mineral lattices as expected from ab-initio calculations (Schauble et al., 2006). This suggests that neither the equilibrium constant for 13C–18O clumping in (proto)dolomite lattice, nor the experimental fractionation associated with acid digestion to produce CO2, depend on their formation mechanism, degree of cation order and/or stoichiometry (i.e., Mg/Ca ratio) and/or δ18O and δ13C compositions (at least over the range we explored). Thus, we suggest the following single Δ47-1/T2 linear regression for describing all dolomite minerals: Δ47CDES90=0.0428(±0.0033)∗106/T2+0.1174(±0.0248)(r2=0.997),with T in kelvin and Δ47 in the Carbon Dioxide Equilibrium Scale (CDES) of Dennis et al. (2011) and referring to CO2 extracted by phosphoric acid digestion at 90 °C. The listed uncertainties on slope and intercept are 95% confidence intervals. The temperature sensitivity (slope) of this relation is lower than those based on low temperature acid digestion of carbonates, but comparable to most of those based on high temperature acid digestion (with however significantly better constraints on the slope and intercept values, notably due to the large range in temperature investigated here and the large number of Δ47 measurements performed, n = 67). We also use this dataset to empirically determine that the acid fractionation factor associated with phosphoric acid digestion of dolomite at 90 °C (Δ*dolomite90) is about +0.176‰. This is comparable to the Δ*calcite90 experimentally obtained for calcite (Guo et al., 2009), suggesting that the acid fractionation Δ* for acid digestion of dolomite and calcite are the same within error of measurement, with apparently no influence of the cation identity. This hypothesis is also supported by the fact that our dolomite calibration data are indistinguishable from published calibration data for calcite, aragonite and siderite generated in similar analytical conditions (i.e., carbonate digested at T ⩾ 70 °C and directly referenced into CDES), demonstrating excellent consistency among the four (Ca, Mg, Fe)CO3 mineral phases analyzed in seven different laboratories (this represents a total of 103 mean Δ47 values resulting from more than 331 Δ47 measurements). These data are used to calculate a robust composite Δ47-T universal relation for those carbonate minerals of geological interest, for temperatures between −1 and 300 °C (that is statistically indistinguishable from the one described by dolomite only): Δ47CDES90=0.0422(±0.0019)∗106/T2+0.1262(±0.0207)(r2=0.985)Thus, in order to standardize the temperature estimates out of different laboratories running high temperature digestion of (Ca, Mg, Fe)CO3 carbonate minerals, we recommend the use of this single Δ47-T calibration to convert Δ47CDES data into accurate and precise temperature estimates. More widely, this study extends the use of the Δ47 thermometry to studies of diagenesis and low-grade metamorphism of carbonates with unprecedented precision on temperature estimates based on Δ47 measurements — environments where many other thermometers are generally empirical or semi-quantitative.

Original languageEnglish
Pages (from-to)255-279
Number of pages25
JournalGeochimica et Cosmochimica Acta
StatePublished - Mar 1 2017


  • Calcium carbonate
  • Carbonate geochemistry
  • Clumped isotopes
  • Dolomite
  • Magnesium carbonate
  • Thermometry

Fingerprint Dive into the research topics of 'Calibration of the dolomite clumped isotope thermometer from 25 to 350 °C, and implications for a universal calibration for all (Ca, Mg, Fe)CO<sub>3</sub> carbonates'. Together they form a unique fingerprint.

Cite this