High-pressure experimental and thermodynamic constraints on the solubility of carbonates in subduction zone fluids

Carbon on the Earth's surface can be carried into the mantle by subducting slabs in the form of carbonates and organics. However, not all the subducting carbon could be recycled into the Earth's deep mantle because parts of them will be released from the subducting slabs by decarbonation processes (e.g., metamorphism and dissolution). In comparison with metamorphic decarbonation, the dissolution of carbonates in subduction zone fluids is more essential, but not well constrained. In this study, we systematically constrained the solubility of different carbonates (e.g., calcite/aragonite, dolomite, and magnesite) under high-pressure and high-temperature conditions relevant to subduction zones by combining experimental and thermodynamic simulations. We simultaneously compared the dissolution processes of calcite/aragonite, dolomite, and magnesite in aqueous fluids under the same P–T conditions using the four-hole gasket technique in diamond anvil cell (DAC) experiments. In situ Raman spectra of fluids in the DAC clearly showed that dolomite had a similar solubility to aragonite, while the solubility of magnesite was much lower than that of aragonite and dolomite under the same P–T conditions. Notably, we also found that the dissolution of dolomite was an incongruent process via the reaction CaMg(CO3)2(solid) in H2O = MgCO3(solid) + Ca2+(aqueous) + CO32−(aqueous). Based on the experimental results, we reasonably suppressed some aqueous species in the DEW model and quantitatively calculated the speciation and solubility of calcite/aragonite, dolomite, and magnesite in aqueous fluids at pressures up to 5 GPa and temperatures up to 800 °C. Then, we modeled and compared the decarbonation fluxes between metamorphism and dissolution from typical marine sediment and altered oceanic crust in subduction zones. In the cold or intermediate subducting slabs, decarbonation fluxes by the dissolution of carbonates are about three times larger than that from metamorphism. Considering the total decarbonation fluxes of metamorphism and dissolution, the cold subducting slabs could release a similar magnitude of carbon as the hot ones.