Fractionation of Carbon Isotopes Between C-O-H Fluids and Melts in High Temperature Systems - Experimental Developments and Outlook

Whether in gases or fluids, as solid or liquid carbonates, dissolved in magma or precipitated in its elemental form, carbon is present in every domain on Earth. The pathways of carbon across atmospheric-, surface-, subduction- and deeper reservoirs of our planet are complex, but can be illuminated by tracing stable carbon isotope ratios. Carbonates take a key role in connecting the surface to the deep carbon cycle. At moderate temperatures, carbon compounds dissolve in fluids, above 1000 °C, carbonates dissolve in or form melts and mobilize carbon inside the Earth. Towards the crust, carbon compounds tend to be oxidized (e.g. CO₂, CO₃^2-) while in the deeper mantle (> 6-8 GPa), reduced states are dominant and cause carbonate reduction to CH₄, FeC or C dissolved in metal, graphite or diamond.^[1]

Recent experimental studies show large carbon isotope fractions at temperatures relevant for the mantle and early Earth environments (i.e. magma ocean surfaces). High temperature equilibrium fractionations have been constrained for CH₄-CO₂-CO^[2], carbonate - graphite^[3], and FeC - graphite^[4] systems, most pairs amounting to a few ‰ at 1000 ^oC. The recognition of kinetic carbon isotope fractionation during elemental carbon precipitation from C-O-H fluids revealed an unexpected high-temperature fractionation mechanism of ~5 ‰ for lower crust and mantle temperatures^[5]. In this light, carbon isotope fractionation may yield surprises in other experimentally underexplored processes. 

We present internally consistent experimental data on high temperature carbon isotope fractionation between carbonate or silicate melts, carbonate, C-O-H-fluids, carbide and graphite. Our results suggest that at high temperatures (>1000 °C) the bonding environment of CO₃-groups (i.e. either in depolymerized silicate- or carbonate melt, in which carbon is anionic CO₃^2-, or as calcite) causes no resolvable differences leading to a universal ∆¹³C (CO₂ -CO₃^2-) fractionation function. Similarly, we suggest that granitic melts with all carbon as molecular CO₂will show no isotope fractionation with an oxidized high temperature fluid. We further discuss challenges of experimental setups under reducing fO₂conditions and the intent of equilibrating silicate melt with reduced C-O-H-fluids, which is experimentally unconstrained and required to understand on one hand the magmatic outgassing of the Earth and how to reconstruct the source isotope composition, on the other hand in a magma-ocean setting, where reduced species are key for the evolution of primitive carbon reservoirs and their isotopic ratios (i.e. mantle carbon).

 

^[1] Rohrbach, A., Schmidt, M. (2011). Nature 472, 209–212.^[2] Kueter, N., Schmidt, M. W., Lilley M. D., Bernasconi, S.M. (2019b). EPSL 506, p.64-75. ^[3] Kueter, N., Lilley, M.D., Schmidt, M.W., Bernasconi, S.M. (2019a). GCA253, 290–306. ^[4] Satish-Kumar, M., So, H., Yoshino, T., Kato, M., Hiroi, Y. (2011). EPLS 310, 340–348. ^[5]Kueter, N., Schmidt, M. W., Lilley M. D., Bernasconi, S.M. (2020). EPSL 529,115848