Content

DGGV-E-Publikationen

Title: Si isotope thermometry in silicified carbonate

Authors:
Michael Tatzel1,2, Marcus Oelze2, Moritz Liesegang3, Maria Stuff4, Michael Wiedenbeck2

Institutions:
1Universität Göttingen; 2Deutsches Geoforschungszentrum GFZ, Potsdam; 3Freie Universität Berlin; 4Bundesanstalt für Materialforschung und -prüfung, Berlin

Event: GeoKarlsruhe 2021

Date: 2021

DOI: 10.48380/dggv-tbwb-kx09

Summary:
Cherts, including silicified carbonates, are one of the most detailed and alteration resistant archives of near-surface environments. Yet, the information disclosed in form of stable isotope ratios of Si and O cannot be confidently translated into conditions prevailing at the Earth surface in deep time. Thermometry based on δ18O is compromised by the lack of knowledge about the fluid’s δ18O value and attempts to determine Si sources or temperatures from δ30Si remain unsatisfying.

We investigated carbonate silicification in a Lower Cambrian silicified zebra dolomite that we analyzed for δ30Si by laser ablation MC-ICP-MS and δ18O using SIMS. Successively replaced carbonate layers show systematically decreasing δ18O values from 14.4 to 13.4 ‰ and systematically decreasing δ30Si values from 0.9 ‰ to ca. -2.0 ‰. We show that quantitative Si precipitation in a closed system best explains these data, requiring positive ε30Si values, which has long been proposed for thermodynamic equilibrium using ab-initio models. We exploit the modal abundances of the successively formed silica phases to quantify the fractional Si depletion from the fluid and to infer the Ɛ30Si values. Using a temperature calibration based on an ab-initio model (Dupuis et al., 2015), we determine the temperatures of carbonate replacement to be approx. 60°C and calculate the fluid δ18O to have been approx. -11 ‰, which is consistent with a meteoric water source. This approach opens a new avenue for determining initial fluid δ18O values in deep time and could thus solve long-standing disputes about hot vs. temperate Precambrian oceans.



Back to list