Internally consistent thermodynamic database for iron to the Earth's core conditions

Tetsuya Komabayashi,Yingwei Fei
DOI: https://doi.org/10.1029/2009JB006442
2010-01-01
Abstract:An internally consistent thermodynamic database for pure iron has been established to pressures (P) up to 360 GPa and temperatures (T) up to 7000 K from existing static experimental data and thermochemical measurements. The database includes body-centered cubic (BCC) phases (alpha or delta phase), the face-centered cubic (FCC) phase (g phase), the hexagonal close-packed (HCP) phase (e phase), and the liquid phase. We describe fundamental thermodynamic relations as the Gibbs free energy divided into thermochemical and thermophysical terms. The thermochemical data were evaluated from existing metallurgy databases together with experimentally determined phase relations. The thermophysical term is obtained from the pressure-volume-temperature equations of state (EoS) for the phases. We constructed an EoS of the FCC phase from our recent internally-heated diamond anvil cell (DAC) experimental data and assessed the EoS of the liquid phase from existing laser-heated DAC experiments together with density data at P = 1 bar, 0.2 GPa, and along the Hugoniot. The HCP-FCC-liquid triple point is located at P = 90 GPa and T = 2800 K. The calculated melting temperature of HCP iron at the inner core boundary (P = 330 GPa) is 4900 K and the density change at melting is -1.2%. The core density deficits at the inner core boundary are 8.1 wt.% and 5.3 wt.% for the liquid outer core and solid inner core, respectively. The calculated melting temperature is much lower than that from dynamic shock wave experiments, suggesting that the HCP structure may not be stable in the inner core. We included a hypothetical high-pressure BCC phase which could be stabilized above 220 GPa by a solid-solid transition of high-P BCC-HCP phases. This hypothetical BCC phase should have a large entropy to give a high melting temperature in order to reconcile the existing discrepancies between the static and shock wave experimental studies.
What problem does this paper attempt to address?