Abstract

This study presents an internally consistent model for the behavior of Fe3+/∑Fe during partial melting of spinel lherzolite. The Fe3+/∑Fe ratio for olivine is calculated on the basis of point defect thermodynamics, and the oxidation states of iron in the other solid phases are calculated using Fe3+/Fe2+ distribution between olivine and orthopyroxene, clinopyroxene, or spinel. Conservation of mass is used to relate the Fe3+/Fe2+ ratio of partial melt to the concentrations of Fe3+ and Fe2+ in the initial and residual solids as a function of pressure, temperature, and oxygen fugacity. Results from isobaric batch melting calculations demonstrate that the Fe3+/∑Fe ratio of the partial melt decreases with increasing melt fraction. Conversely, the Fe3+/∑Fe ratio of the partial melt increases with increasing melt fraction during decompression batch melting. The relative oxygen fugacity of the upper mantle depends on both the oxidation state of iron and mantle potential temperature. Results from incremental decompression melting calculations in which 1% melt is produced for each 100MPa of decompression and then removed from the residual solid indicate that relative oxygen fugacity calculated from the oxidation state of iron in basaltic glass does not represent a unique value for the oceanic upper mantle but, rather, reflects conditions in the lower portion of the melting regime. A 100°C change in mantle potential temperature produces a change in relative oxygen fugacity of ∼0.8logunits, similar to the global range inferred from mid-ocean ridge basalt glasses. It is necessary, therefore, to compare relative oxygen fugacity calculated from basaltic glass with proxies for potential temperature before drawing conclusions on heterogeneity of the oxidation state of iron in the oceanic upper mantle. Results from model calculations also suggest that the sub-arc mantle is intrinsically more oxidizing than the oceanic mantle because it is cooler. The global correlation between Fe3+/∑Fe and H2O in basaltic glasses may, in part, reflect the solidus-lowering influence of the latter. Secular variations in relative oxygen fugacity of the upper mantle due to cooling of the Earth over geologic time do not appear to be large enough to have influenced the nature of volcanic outgassing significantly enough to have contributed to the Great Oxidation Event.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call