Exploration of beta-Fe using sintered diamond anvils

Based on the DAC experiments, it has been claimed that a new polymorph of Fe, called beta phase, exists at higher than 35 GPa and 1500 K. We have carried out a series of experiments using a multi anvil press equipped with sintered diamond (SD) to clarify if the beta phase is really existent. The SD anvil assembly was squeezed by the DIA-type press SPEED-1500 installed at SPring-8, Japan. A fine mixture of Fe + MgO (1:1 in wt.) was put into a sleeve of MgO, which was set at the center of a cylindrical Re heater. Diamond powder was packed into the remaining spaces of the heater. In situ X-ray observation was carried out by energy dispersive method using synchrotron radiation. The generated pressure was determined from the measured volume of MgO of the sleeve based on the pressure scale after Jamieson et al. (1982). In the first heating cycle at 662 tonf, Fe initially assumed epsilon at 37.7(4) GPa and 300 K. On heating, gamma appeared at 1400 K and 38.4(2) GPa. The peaks from gamma intensified in accord with weakening of those from epsilon to 1700 K. However, the peaks of epsilon still survived at 1700K presumably due to temperature gradient in the sample. On cooling, the diffraction peaks of gamma began to weaken accompanying with the increase of epsilon peaks at 1300 K and 36.9(1) GPa. In the second heating cycle at 803 tonf, no change in diffraction pattern was observed up to 1300 K. However, the growth of reflections from gamma was recognized at 1600K and 41.1(2) GPa, associated with the reduction of epsilon, and became prompt with increasing temperature. We obtained no evidence for the presence of beta-Fe with the dhcp (Saxena et al. 1995) or the orthorhombic Pbcm structure (Andrault et al. 1997) up to 42.2(3) GPa and ca. 1900 K. All the results obtained in the present study indicate that the epsilon-gamma phase boundary is approximated by an equation of T (K) = 670 + 18P (GPa) over pressures from 14 to 42 GPa.