Experimental evidence of the high-PT body-centered cubic Fe stability

Recent (Kraus et al., Science 2021) experiments collected X-ray spectra for iron at extreme pressures and high temperatures. We performed large scale (up to 2000 atoms) ab initio molecular dynamics (AIMD) simulation of iron at presumably experimental conditions and compared calculated spectra for liquid, hexagonal (HCP) and cubic (BCC) with experimental spectra. It appears that the BCC spectrum is a perfect match to the experimental one, interpreted as a liquid spectrum. The sequence of spectrum interpreted as liquid, liquid+HCP, HCP should be interpreted, in fact, as BCC, BCC+HCP, and HCP. In addition, large scale two-phase AIMD determination of melting temperatures of iron performed at 360 GPa and 660 GPa allowed to provide melting temperature of iron up to very high pressures. The simulations performed at the highest level of theory and the most advanced method for the determination of melting temperature for the largest so far computational boxes provide likely the most robust melting curve of iron. This curve is in agreement with the one obtained by the less robust method and validates the method (Belonoshko, FU, Smirnov, PRB, 2021) where the BCC iron is stable prior the melting. Thus, the theory and experiment are in excelent agreement which settles the issue on the BCC stability as the submelting phase in the Earth and exoplanetary iron cores.