Experimental study of carbonate formation in oceanic peridotite

Interactions of CO2-rich aqueous fluids with mantle peridotite have major implications for geochemical budgets and microbial life in the shallow oceanic lithosphere through the formation of carbonate minerals and reduced carbon species. However, the underlying mechanisms controlling the transformation of CO2 to carbonates in ultramafic-hosted hydrothermal systems remain incompletely understood. A long-term laboratory experiment was conducted at 300 °C and 35 MPa to investigate serpentinization and carbonate formation pathways during hydrothermal alteration of peridotite. Powdered harzburgite was initially reacted with a Ca-rich aqueous fluid for 14,592 h (608 days) and changes in fluid composition were monitored with time. Once the system reached a steady state, a CO2(aq)-rich fluid was injected and allowed to react with the system for 5907 h (246 days). Fluid speciation and mineral analyses suggest that serpentinization of harzburgite in the CO2-poor system led to the precipitation of serpentine, brucite, magnetite, and minor calcite, in addition to other minor phases including chlorite and sulfur-poor Ni sulfides. The addition of the CO2(aq)-rich fluid caused dolomite, Ca-rich dolomite, and high-Mg calcite to form at the expense of olivine, calcite, and brucite, while serpentine remained unreactive. Replacement textures and mineral assemblages mimic those documented in carbonate-altered seafloor serpentinites, particularly those from the Mid-Atlantic Ridge and the Iberia Margin. In contrast to thermodynamic predictions, magnesite did not form in the experiment because the dissolution of clinopyroxene, in combination with the lack of serpentine reactivity, maintained low Mg/Ca ratios in solution. Clinopyroxene dissolution and unreactive serpentine may similarly maintain low Mg/Ca ratios in submarine serpentinization systems and limit magnesite formation in subseafloor environments. Results of this study suggest that the formation of Ca-Mg carbonates by mineral carbonation is favorable in subseafloor serpentinization systems and likely represents a significant, but poorly quantified, carbon sink in hydrothermally altered oceanic lithosphere at slow-spreading mid-ocean ridges.