Project Details
Projekt Print View

Experimental study of coupled redox equilibria in a peridotite–metasediment–eclogite mélange and implications for diamond formation

Subject Area Mineralogy, Petrology and Geochemistry
Term from 2020 to 2022
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 445062436
 
Final Report Year 2022

Final Report Abstract

In this project, we experimentally simulated the conditions where different carbonated lithologies are reduced solely by the exchange of oxygen with adjacent peridotite in a subduction zone environment to produce diamond. These experiments were carried out with simplified chemical compositions, including being Fe-free in order to remove ambiguities in the interpretation of the results. Pure forsterite (Mg2SiO4) was used as a peridotite proxy. Carbonated eclogite was represented by SiO2-saturated mixtures of MgCO3 + SiO2 ±Al2O3 or alternatively by MgO + CaCO3 + SiO2 + Al2O3, to simulate interactions involving Ca-poor or Ca-rich pyroxene, respectively, along with garnet. Mixtures of CaCO3 + SiO2 + Al2O3 were used as a proxy for metasediment, which yielded calcic garnet, kyanite and stishovite at the conditions of the experiments. A graphite disc was inserted between the two lithologies to try and limit direct physical contact between the layers and to serve as a carbon source for the carbonation reactions. The experiments were performed at 10 GPa in order to assure that diamond could form, which could be readily distinguished from the graphite present in the starting configuration. Results of these experiments are described and discussed. The combination of a metasediment layer and a graphite disc produced no reaction between the two. The addition of a forsterite layer (peridotite proxy) on the other side of the graphite led to a series of metasomatic reactions, related to the redox contrast between the two layers. Initially, magnesite and low-Ca pyroxene were produced within the forsterite layer near the contact with the graphite and the capsule walls. Simultaneously, carbonate decomposed in the metasediment/eclogite layer adjacent to the graphite disc. Eventually, a reaction zone with olivine, low-Ca pyroxene, magnesite, garnet, and high-Ca pyroxene developed in the peridotite layer near its contact with the capsule. Diamond crystallized in the metasediment layer near the capsule walls and formed a monomineralic layer at the interface with the graphite disc, as a result of reduction of carbonate. On the peridotite side, dissolution and carbonation of the graphite had occurred and it is an important observation that no diamond was identified at the contact between the graphite and metasomatized peridotite. Melt migration from the metasediment/eclogite layer produced a series of well-defined mineral zones in the peridotite, in part related to Ca-metasomatism, which caused local wehritization. A model has been developed to describe the series of redox interactions. The most important result of our experiments is that we have identified simultaneous carbonation of peridotite and decarbonation of juxtaposed metasediment or eclogite. These involve redox reactions in Fe-free bulk compositions and thus redox interaction between these lithologies may occur in the absence of a metal phase and even in a Fe-free system. Therefore, our study suggests that there are additional possibilities to form diamonds in the mantle that are unrelated to variations in the Fe3+/Fe2+ ratio in minerals or melts and these should be taken into consideration when modelling redox processes and carbon cycling in the deep Earth. Assemblages of diamond-bearing eclogites and eclogite-like rocks are stable in a subsolidus subduction mélange. The diamond content of the metasedimentary and eclogitic domains depends on their original carbonate concentration, and hence, the occurrence of diamond-enriched metasediments and eclogites is not unexpected. On the other hand, the amount of diamond produced in peridotite according to the redox freezing mechanism is controlled by the reducing capacity of the peridotite, which is rather limited since it is related to the amount of a metallic phase or the partial oxidation of Fe2+ in silicate phases. This means that in most cases, the diamond content of peridotite should be low.

Publications

 
 

Additional Information

Textvergrößerung und Kontrastanpassung