Solid phases of FeSi to 45 GPa and 2500 K

FeSi remains solid up to 2300+-200 K between 22 and 45 GPa in a laser-heated diamond anvil cell, showing that addition of silicon does not cause a large amount of melting point depression (the melting temperature of Fe ranges from 2300 +- 200 K to 2700 +- 200 K between 22 and 45 GPa). While the phase that is stable at core conditions remains an open question, our data constrain the epsilon-B2 transition to 29 +- 1 GPa at all temperatures from 1200 K to 2300 K. Room temperature diffraction data are fit to a Birch-Murnaghan equation of state with V0 = 21.28(4) Å^3, K0 = 235(13) GPa and K0' = 3.2(6) (1-sigma uncertainties), tied to the equations of state of Ar and Ne used as pressure standards and media [Errandonea, et al., PRB, 73, 092106 (2006); Lin, et al., JGR, 108, 2045 (2010)]. The relatively low value of K0' matches that of epsilon-FeSi from Knittle and Williams, GRL, 15, 445 (1995), and suggests that an iron-silicon alloy may not be compatible with the density profile of the Earth's outer core.