JpGU-AGU Joint Meeting 2020

Presentation information

[E] Oral

S (Solid Earth Sciences ) » S-GC Geochemistry

[S-GC48] Volatiles in the Earth - from Surface to Deep Mantle

convener:Takeshi Hanyu(Japan Agency for Marine-Earth Science and Technology, Research Institute for Marine Geodynamics), Gray E Bebout(Lehigh University), Hirochika Sumino(Department of Basic Science, Graduate School of Arts and Sciences, The University of Tokyo), Yuji Sano(Division of Ocean and Earth Systems, Atmosphere and Ocean Research Institute, University of Tokyo)

[SGC48-10] . The conditions of sub-lithospheric diamond formation constrained from ferric iron-rich exsolution in ferropericlase inclusions in diamond .

*Caterina Melai1, Daniel Frost 1, Tiziana Boffa Ballaran1, Catherine McCammon1, Katharina Marquardt 2 (1.Bayerisches Geoinstitut, Universitaet Bayreuth, 2.Imperial College, London)

Keywords:Deep carbon cycle, Ferropericlase , diamond formation

The rocks-salt structured Mg-Fe oxide, ferropericlase, is one of the most common minerals found as an inclusion in sub-lithospheric diamonds, where it appears with a wide range of Fe/(Fe+Mg) ratios. It is important to be able to understand the formation conditions of such inclusions as they provide some of the only direct evidence for processes that are part of the deep mantle carbon cycle. In a number of TEM investigations exsolution of Fe3+ –rich phases, normally described as being part of the magnetite-magnesioferrite solid solution, have been identified within such ferropericlase inclusions. This is somewhat surprising because it implies that ferropericlase becomes saturated in Fe2O3. At low pressures this would indeed result in the exsolution of magnetite-magnesioferrite solid solution but this only occurs at oxygen fugacities that are higher than those at which diamond or graphite would be stable. At pressures higher than 8 GPa, however, experiments have shown that other mixed valence iron oxide phases with the stoichiometries Fe4O5 and Fe5O6 are stable, which may well cause a decrease in the maximum Fe2O3 content of coexisting ferropericlase. In order to understand the exsolution of Fe2O3-bearing oxides from ferropericlase inclusions in diamonds and to determine the conditions or the changes in conditions required for this to occur, it is essential to be able to describe phase relations in the FeO-MgO-Fe2O3 system at high pressures and temperatures. This information will not only allow the oxygen fugacity of diamond formation to be assessed but can potentially provide information on the post entrapment conditions experienced by diamonds.

In this project we examine the conditions at which mixed valence iron oxides become stable and determine the maximum Fe3+/Fetot ratio of coexisting (Fe,Mg)1-xO ferropericlase using both experimental measurements and thermodynamic modelling. Multianvil experiments have been performed between 6 – 25 GPa and 1200-1800°C using a starting composition of (Mg86Fe14)O and (Mg50Fe50)O plus varying amounts of Fe2O3 (20%, 10%, 5%). Pt powder was added to the experiments to act as a redox sensor and minor amounts of Ni, Cr, Mn and Na were also added. Samples were then analysed using scanning electron microscopy, the electron microprobe, Mössbauer spectroscopy and X-ray diffraction.

In the recovered experiments, ferropericlase was found to coexist with magnetite-magnesioferrite solid solution up to 12 GPa and Mg2Fe2O5-Fe4O5 solid solution at higher pressures. In all experiments the oxygen fugacities could be determined from three different equilibria simultaneously, and in the calculation of the oxygen fugacity a ferropericlase model in the FeO-Fe2/3O-MgO system was employed and exchange of Mg and Fe2+in magnetite was accounted for.
In this model only at pressures above 16 GPa the calculated oxygen fugacity of (Fe,Mg)4O5 exsolution is compatible with the diamond stability field. As it is likely that the exsolution observed in natural ferropericlase inclusions occurred after entrapment in the diamond, it must have occurred either as pressures increased after entrapment or temperatures decreased. The first scenario might imply that the diamonds were then further subducted after formation, while the second scenario might arise if the diamonds were then later stored in the subcontinental lithosphere. From precise determinations of the Fe3+/Fetot ratios of ferropericlase inclusions coexisting with exsolved mix valence iron oxides and through characterisation of the exsolved oxides themselves, it should be possible to differentiate between these scenarios.