- redox melting (1) (remove)
- The carbon speciation in the Earth’s interior as function of pressure, temperature and oxygen fugacity (2011)
- The redox state of the Earth’s interior will influence the speciation of volatile elements both in the mantle and in mantle derived magmas. Carbon is one of the principal elements to be affected in this way because under reducing conditions it forms graphite or diamond, and under oxidizing conditions carbonate (or CO2-bearing) minerals and melts. The cycling and residence time of carbon in the mantle can be strongly effected by the oxygen fugacity because reduced phases such as diamond and graphite are immobile and likely to remain within the mantle and potentially within subducting slabs, while at more oxidizing conditions CO2-rich fluids or melts can migrate and escape from the interior. The carbon cycle in the Earth may therefore depend on the redox state of mantle rocks. Conversely, an influx of CO2-rich fluids or melts may act to oxidize the mantle as an additional aspect of metasomatism. In the first part of this study experiments were performed to measure the oxygen fugacity at which carbon (graphite or diamond) oxidises to carbonate minerals or melts within mantle peridotite assemblages between 2.5 and 11 GPa at 1100-1600 °C. The experiments were performed up to temperatures where carbonate melts evolve towards more silicate-rich compositions. The dilution of the carbonate melt component was found to lower the relative fo2, expanding the melt stability field with respect to reduced carbon. The results allow the fo2 of the diamond formation process to be determined both as a function of pressure, temperature and melt CO2 concentration. These results also have implications for the onset of melting within up welling mantle material. Several studies have indicated that the mantle may become more reduced with depth. This means that the oxidation of elemental carbon (graphite or diamond) may occur in up welling rocks where the oxidized product is a carbonate bearing magma. When the experimental data are compared with current estimates for the fo2 of mantle rocks the implication is that peridotitic mantle will remain in the diamond stability field up to at least 100-150 km depth. Only at depths shallower than 150 km would Fe3+ in mantle silicates react with graphite to produce carbonate rich melts in a redox melting process. Redox melting should limit the depth interval over which carbonate-rich melts can form beneath ridges. Further experiments were performed to determine the fo2 at which diamond oxidises to carbonate in the transition zone and lower mantle. Experiments at 45 GPa were performed using the MADONNA D-DIA (1500 tons) apparatus with sintered diamond anvils installed at the Geodynamics Research Centre, Ehime University in Japan. The measured oxygen fugacity was found to be approximately 3 log units above the iron-wüstite oxygen buffer (deltaIW+3). As the oxygen fugacity of the transition zone and lower mantle is most likely at or below the IW buffer this confines the stability of solid carbonate to the upper mantle or to unusually oxidized regions of the deeper mantle. The oxygen fugacity at which magnesite and diamond coexist showed a slight decrease with pressure, however, implying the possibility that magnesite may become the stable host for carbon at the very base of the lower mantle. The oxygen fugacity at which mantle xenoliths equilibrated can be determined using oxy-thermobarometry equilibria. For garnet-peridotite rocks the only calibrated and tested oxy-barometer employs the equilibrium, 2Fe3Fe23+Si3O12 = 4Fe2SiO4 + 2FeSiO3 + O2 Garnet Olivine Orthopyroxene In the final section of this thesis Fe3+/ΣFe ratios of garnets produced in a peridotite assemblage in equilibrium with carbon and carbonate melts were measured between 3 and 7 GPa. The oxygen fugacity in these experiments was also constrained, which allowed a test of this widely used oxy-barometer to be made at pressures much higher than previously performed. The results indicate that the pressure dependence of this oxy-barometer may be in error and a preliminary recalibration implies that cratonic lithosphere may not be as reduced as previously considered.