Fig. 1. Previous results from high pressure melting studies in the Fe–FeS system. Gray curve: Usselman (1975); dotted curve: Williams and Jeanloz (1990); dashed curve: Boehler (1996); solid curve: (Fei et al., 1997) and (Fei et al., 2000) and Li et al. (2001).

Fig. 2. Modelled temperature distribution in an opaque diamond anvil cell sample that is laser heated on both sides. In this model identical Gaussian temperature distributions are imposed on two surfaces of the sample, with a peak temperature of 2000 K (1700 K above ambient). The sample thickness (vertical) d = 7 μm, and the beam diameter ; only the central 30 μm of the solution is shown. The dotted lines indicate the width of the synchrotron X-ray beam (5 μm). For these parameters, within the volume probed by the X-ray beam the temperature increase (T − T_ambient) at the sample midplane is 5% lower than that measured at the surface. The temperature scale in the image has been adjusted for high contrast in the region of interest; lower axial gradients exist in the cooler portions of the laser heated spot.

Fig. 3. X-ray diffraction spectra of a Fe + Fe₃S mixture during and after laser heating in a diamond anvil cell. NaCl was used as a pressure medium/insulator. The two-dimensional detector image has been caked into a rectilinear projection; the horizontal axis is the scattering angle (2θ), and the vertical axis is the azimuthal angle around the 2θ = 0 direction. Vertical bars are used to indicate positions of Fe₃S, hcp-Fe, and B2-NaCl diffraction peaks. The data collection sequence proceeds from top to bottom. Prominent Fe₃S peaks in the 1710 K diffraction spectrum disappeared in the partially molten sample at 2100 K, and then they reappeared upon quench to 300 K.

Fig. 4. Melting data in the Fe–S system at high pressure. Previous results using both the multi-anvil press ([Fei et al., 1997], [Fei et al., 2000] and [Li et al., 2001]) and the laser heated diamond anvil cell (Boehler, 1996) are compared to our results. Two experimental melting curves of pure Fe ([Boehler, 1993] and [Shen et al., 1998]) are also shown for reference.

Fig. 5. Compositions in melting experiments in the Fe–FeS system. Filled circles: eutectic temperatures determined from multi-anvil experiments ([Fei et al., 1997], [Fei et al., 2000] and [Li et al., 2001]); gray line: pressure range and composition of starting material used by Boehler (1996); large open squares: starting compositions used in this study; small open squares: phases observed identified by X-ray diffraction of quenched partially melted samples in this study. All of the partial melts in this study were solid Fe + sulfide melt; the presence of Fe₃S upon quench indicates a Fe–Fe₃S eutectic extending to 60 GPa.

High pressure melting data for the Fe–Fe₃S system