Phase stabilities and Fe/Sr/La partitioning between magnesite (MgCO3) and mantle silicates at lower mantle conditions.

Carbonates appear to be one group of the main carbon-bearing minerals in the Earth's interior. Inclusions of carbonates in diamonds of lower mantle origin support the assumption that they are present even in the Earth's lower mantle. Although the carbonates' phase diagrams have been intensively studied, their stability in presence of mantle silicates at deep mantle conditions (>25 GPa) remains unclear. Furthermore, the carbonate inclusions show a high REE enrichment. This raises questions on the distribution of trace elements between carbonates and silicates and on the possible role of carbonates as trace element carrier in the Earth's mantle.Numerous studies show that magnesite is likely to be the major solid carbonate carried by subduction into the Earth's lower mantle. We investigated the stability of MgCO3 in presence of mantle silicates and the Fe, Sr and La partitioning in high-pressure and high-temperature experiments. One set of experiments was conducted with multi-anvil presses at BGI, Bayreuth, at conditions ranging from 24 GPa to 30 GPa and 2000 K. The investigated reaction is between natural magnesite and (Mg,Fe)SiO3-glasses doped with either Sr or La. Preliminary data from the multi-anvil press at 24 GPa and 2000K show the onset of carbonate melting which is consistent with the previous study of the melting curve in the enstatite-magnesite system [1]. Decomposition of MgCO3 is not observed, in contrast to experiments using magnesite and SiO2 as starting materials [2], suggesting that MgCO3 is stable at these conditions in the presence of silicates phases. The silicate glass react to bridgmanite (Mg,Fe)SiO3 as well as stishovite SiO2 and magnesiowüstite (Mg,Fe)O. The Fe-Mg partitioning coefficient between bridgmanite and magnesite calculated in this study is ~2 and in agreement with previous experiments at similar conditions [3].Laser-heated diamond anvil cell (LH-DAC) experiments were performed at University of Potsdam [4] at conditions 30 to 40 GPa and 1800 to 2300 K. The run products were characterized in-situ at high-pressure by XRD and XRF mapping at the P02.2 beamline at PETRA III. Our data show a transformation of the starting silicate glass into bridgmanite. We also observed stishovite and magnesiowüstite in the center of the hotspot where the temperature had reached >2000 K. In this case, the presence of magnesiowüstite might be the result of MgCO3 decomposition at higher temperature. Additional TEM analyses on the post-mortem sample will allow us to further characterize the different phases present in the laser-heated hotspot.[1] Thompson et al. (2014) Chemistry and mineralogy of the earth's mantle. Experimental determination of melting in the systems enstatite-magnesite and magnesite-calcite from 15 to 80 GPa. American Mineralogist 99(8-9), 1544-1554.[2] Drewitt et al. (2019) The fate of carbonate in oceanic crust subducted into Earth's lower mantle. EPSL 511, 213-222[3] Martinez, et al. (1998). Experimental investigation of silicate-carbonate system at high pressure and high temperature. Journal of Geophysical Research: Solid Earth, 103(B3), 5143-5163.[4] Spiekermann et al. (2020). A portable on-axis laser heating system for near-90° X-ray spectroscopy: Application to ferropericlase and iron silicide. Journal of Synchrotron Radiation. (accepted)