“Carbon in the core” revisited

In order to investigate the existence of carbon in the core, we performed high-pressure melting experiments using a Kawai-type multi-anvil apparatus by two methods: (1) quench experiments up to 14 GPa and 2200 °C and (2) in situ X-ray diffraction experiments up to 29 GPa. From the quench experiments, carbon solubility in molten iron and liquidus phase relations in the Fe–C system was investigated. The carbon solubility in molten iron is 8.5 wt% at 5 GPa and 2000 °C, where graphite is the liquidus phase. At 10–14 GPa, the carbon solubility in molten iron coexisting with diamond is about 7 wt% at 2000 °C and is pressure insensitive. At 5 GPa, Fe3C (6.7 wt% C) is the liquidus phase below 1400 °C. Above 10 GPa, Fe7C3 (8.4 wt% C) appears as the liquidus phase below 1700 °C and Fe3C melts incongruently to liquid iron and Fe7C3 at temperatures below 1500 °C. From in situ X-ray diffraction observations, the incongruent melting of Fe3C was found to occur at least up to 29 GPa, and liquidus temperatures of the Fe3C composition were found to be 1825 and 1925 °C at 21.0 and 29.2 GPa, respectively. Based on our results, it is concluded that large amount of carbon could be incorporated into the Earth's core if the Earth forming material was rich in carbon. Our result also shows that Fe7C3 would be the first crystallizing phase from the liquid of the outer core, implying that Fe7C3 could be a potential constituent of the solid inner core.