Abstract

Phase relations of magnesioferrite (MgFe2O4) have been studied between 8 and 18 GPa and 1000–1600 °C using multi-anvil experiments. At 8–10 GPa and 900–1200 °C, MgFe2O4 breaks down to Fe2O3+MgO. At higher temperatures, a new phase appears along with Fe2O3. Although this new phase is unquenchable, EPMA and TEM data point to a composition with Mg5Fe2O8 or Mg4Fe2O7 stoichiometry. Depending on pressure and temperature, other stoichiometries also appear to be stable together with Fe2O3. In terms of pressure, the stability field of the unquenchable phases + hematite widens with increasing temperature to 3 ± 1 GPa at ~1400 °C, and then narrows to ~1 GPa at 1600 °C. The recoverable assemblage of Mg2Fe2O5+Fe2O3 becomes stable between 11–13 GPa. The Mg2Fe2O5+Fe2O3 assemblage is stable up to at least 18 GPa at 1300 °C without any evidence of a hp-MgFe2O4 phase. In addition, hematite plays an important role in the phase relations of MgFe2O4 by being present over a wide range in pressure and temperature together with a Mg-rich Fe-oxide. Interestingly, hematite incorporates variable amounts of Mg whereby its concentration appears to be a function of temperature. This experimental study has implications for interpreting inclusions in natural diamonds where magnesioferrite occurs by placing a maximum pressure stability on the formation of this phase. Through these inclusions, it also provides constraints on diamond formation and their subsequent evolution prior to eruption. For example, the occasional observation of nano-sized magnesioferrite within (Mg,Fe)O inclusions must have either formed from a high-pressure precursor phase with a different stoichiometry at transition zone or upper lower mantle conditions, or it exsolved directly from the host (Mg,Fe)O under upper mantle conditions (i.e., <9–10 GPa). Since several studies report various non-silicate inclusions with simple oxide compositions, including magnesioferrite, magnetite, or ferropericlase, such inclusions provide evidence for variable redox conditions at the time of entrapment.

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