Project Details
Projekt Print View

Redox melting processes in the mantle under reducing conditions

Subject Area Mineralogy, Petrology and Geochemistry
Term from 2016 to 2020
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 320586258
 
Final Report Year 2019

Final Report Abstract

In this project, we experimentally investigated the process of redox melting under mantle conditions in the presence of reduced, methane-bearing fluids and its potential link with diamond formation. Experiments were carried out in the pressure range of 3-7 GPa, corresponding to depths in the mantle down to ~225 km. An experimental technique was established for producing and maintaining reduced C–O–H fluids at high-pressures and high-temperatures, including the ability to confirm the low oxygen fugacity (ƒO2) ex situ. Experiments buffered by Fe-FeO (IW) or Mo-MoO2 can be maintained for at least 24 hours at temperatures <1300°C at 5.0 GPa using gold outer capsules and an internal olivine capsule for the sample. The olivine container plays two important roles: 1) it prevents Fe-loss from the sample by shielding it from direct contact with the metal capsule, and 2) it serves as a fluid trap, allowing characterisation of the fluid phase by Raman and IR spectroscopy. The ∆logƒO2 of the sample lies between IW+0.4 and IW+1.5, due to the presence of carbon in the sample fluid. An extensive network of fluid inclusions was trapped in the olivine container and analysis by Raman spectroscopy reveals that they are composed essentially of methane and minor amounts of ethane (C2H6), with rarely detectable H2. Additional H2 appears to have been incorporated into the olivine structure. Surprisingly, no H2O was detected although some must have been present during the experiment since FTIR measurements indicate that the olivine contains OH on all four possible defect sites. At 5 GPa and IW-buffered conditions (~IW+0.8), the harzburgite solidus was determined to lie at ~1230 °C and in experiments with Mo-MoO2 buffering, melt was detected at 1300°C at 7 GPa. As melt fractions are < 1 %, this represents the incipient melting regime where melts should be enriched in incompatible elements. This is manifested in high P2O5 contents of 0.9-1.9 wt %. The melts are hypersthene normative at 5 GPa and nepheline normative at 7 GPa. Assuming that low microprobe totals reflect the presence of volatile components, the melts appear to contain ~10 wt % H2O (no C was detected in careful spectrometer scans). If incipient melts remain trapped in the peridotite matrix, they could serve as an additional sink for H2O in the upper mantle that can be readily mobilised through deformation. Major element ratio systematics in terms of molar (Mg+Fe)/Si and Al/Ca exhibit similarities and differences with published experimental results for the regime of major melting (>10% melt fraction) in anhydrous and hydrous peridotite compositions over a range in pressures. The regime of major melting was not attainable with our experimental configuration. A particular highlight of this project was the crystallisation of diamond in a number of experiments, even at 5 GPa. Diamond formation was spontaneous (i.e. without the presence of seed crystals) and crystallised from a CH4-rich fluid. Our method for diamond synthesis is the subject of European and international patent applications with the European Patent Office. The incorporation of H2 into olivine and orthopyroxene along with H incorporation as OH provide local sinks for H2, which promotes diamond formation by driving the reaction CH4 = C + 2H2 to the right. This avoids the need for a significant oxygen source to produce diamond via CH4 oxidation.

Publications

 
 

Additional Information

Textvergrößerung und Kontrastanpassung