Search

The alkali halides are often used as optically-transparent pressure-transmitting media and thermal insulators in laser-heated diamond anvil cell experiments. High P--T equations of state for these materials would allow them to be used simultaneously as sensitive in situ pressure markers, making sample preparation and data analysis simpler. KCl is especially useful for this application because its high melting point, crystallographic simplicity, and low electron density are ideal for use in high-temperature x-ray diffraction experiments. However, the high-temperature static equation of state data for this material is limited in pressure to 8 GPa. As experiments routinely exceed 100 GPa in pressure and thousands of Kelvin, it is necessary for the equation of state of KCl to be determined experimentally at these conditions if it is to be used as an equation of state calibrant in the future. This paper combines new high-pressure, high-temperature data of the B2 phase of KCl (up to 167 GPa and 2400 K) with the previously published room-temperature data of Dewaele et al. [Phys. Rev. B 85, 214105 (2012)] to produce a Mie-Gr\"uneisen-Debye thermal equation of state for this material. The room-temperature equation of state parameters are similar to those reported previously: ${V}_{0}=32.0(3) {\mathrm{cm}}^{3}/\mathrm{mole}, {K}_{0}=24(1)$ GPa, and ${K}_{0}^{\ensuremath{'}}=4.56(5)$. The thermal parameters, ${\ensuremath{\gamma}}_{0}$ and $q$ are 2.9(4) and 1.0(1), respectively. While a $q$ of 1 is expected, the ${\ensuremath{\gamma}}_{0}$ is higher than expected from previous calculations, lower pressure experimental data, and the available shockwave data. Thus, previously-reported equations of state underestimate the pressure of KCl at high temperatures.

ThO_2 is an important material for understanding the heat budget of Earth's mantle, as well as the stability of nuclear fuels at extreme conditions. We measured the in situ high-pressure, high-temperature phase behavior of ThO_2 to ∼60 GPa and ∼2500 K. It undergoes a transition from the cubic fluorite-type structure (thorianite) to the orthorhombic α-PbCl_2 cotunnite-type structure between 20 and 30 GPa at room temperature. Prior to the transition at room temperature, an increase in unit-cell volume is observed, which we interpret as anion sub-lattice disorder or pre-transformation “melting” (Boulfelfel et al. 2006). The thermal equation of state parameters for both thorianite [V_0 = 26.379(7), K_0 = 204(2), αK_T = 0.0035(3)] and the high-pressure cotunnite-type phase [V_0 = 24.75(6), K_0 = 190(3), αK_T = 0.0037(4)] are reported, holding K_0′ fixed at 4. The similarity of these parameters suggests that the two phases behave similarly within the deep Earth. The lattice parameter ratios for the cotunnite-type phase change significantly with pressure, suggesting a different structure is stable at higher pressure.

The high-cosmic abundance of sulfur is not reflected in the terrestrial crust, implying it is either sequestered in the Earth’s interior or was volatilized during accretion. As it has widely been suggested that sulfur could be one of the contributing light elements leading to the density deficit of Earth’s core, a robust thermal equation of state of iron sulfide is useful for understanding the evolution and properties of Earth’s interior. We performed X-ray diffraction measurements on FeS 2 achieving pressures from 15 to 80 GPa and temperatures up to 2400 K using laser-heated diamond-anvil cells. No phase transitions were observed in the pyrite structure over the pressure and temperature ranges investigated. Combining our new P-V-T data with previously published room-temperature compression and thermochemical data, we fit a Debye temperature of 624(14) K and determined a Mie-Gruneisen equation of state for pyrite having bulk modulus K T = 141.2(18) GPa, pressure derivative K ′ T = 5.56(24), Gruneisen parameter γ 0 = 1.41, anharmonic coefficient A 2 = 2.53(27) × 10 −3 J/(K 2 ·mol), and q = 2.06(27). These findings are compared to previously published equation of state parameters for pyrite from static compression, shock compression, and ab initio studies. This revised equation of state for pyrite is consistent with an outer core density deficit satisfied by 11.4(10) wt% sulfur, yet matching the bulk sound speed of PREM requires an outer core composition of 4.8(19) wt% S. This discrepancy suggests that sulfur alone cannot satisfy both seismological constraints simultaneously and cannot be the only light element within Earth’s core, and so the sulfur content needed to satisfy density constraints using our FeS 2 equation of state should be considered an upper bound for sulfur in the Earth’s core.

The Earth's core is comprised mostly of iron and nickel, but it also contains several weight percent of one or more unknown light elements, which may include silicon. Therefore it is important to understand the high pressure, high temperature properties and behavior of alloys in the Fe–FeSi system, such as their phase diagrams. We determined melting temperatures and subsolidus phase relations of Fe–9 wt% Si and stoichiometric FeSi using synchrotron X-ray diffraction at high pressures and temperatures, up to ~200 GPa and ~145 GPa, respectively. Combining this data with that of previous studies, we generated phase diagrams in pressure–temperature, temperature–composition, and pressure–composition space. We find the B2 crystal structure in Fe–9Si where previous studies reported the less ordered bcc structure, and a shallower slope for the hcp+B2 to fcc+B2 boundary than previously reported. In stoichiometric FeSi, we report a wide B2+B20 two-phase field, with complete conversion to the B2 structure at ~42 GPa. The minimum temperature of an Fe–Si outer core is 4380 K, based on the eutectic melting point of Fe–9Si, and silicon is shown to be less efficient at depressing the melting point of iron at core conditions than oxygen or sulfur. At the highest pressures reached, onlymore » the hcp and B2 structures are seen in the Fe–FeSi system. We predict that alloys containing more than ~4–8 wt% silicon will convert to an hcp+B2 mixture and later to the hcp structure with increasing pressure, and that an iron–silicon alloy in the Earth's inner core would most likely be a mixture of hcp and B2 phases.« less

Melting in the Fe–FeO system was investigated at pressures up to 93 GPa using synchrotron X-ray diffraction (XRD) and a laser heated diamond anvil cell (DAC). The criteria for melting were the disappearance of reflections associated with one of the end-member phases upon raising the temperature above the eutectic and the reappearance of those reflections on dropping the temperature below the eutectic. The Fe–FeO system is a simple eutectic at 50 GPa and remains eutectic to at least 93 GPa. The eutectic temperature was bound at several pressure points between 19 and 93 GPa, and in some cases the liquidus temperature was also determined. The eutectic temperature rises rapidly with pressure closely following the melting curve of pure Fe. A detailed phase diagram at 50 GPa is presented; the eutectic temperature is 2500 ± 150 K and the eutectic composition is bound between 7.6 ± 1.0 and 9.5 ± 1.0 wt.% O. The coefficient of thermal expansion of FeO is a strong function of volume and decreases with pressure according to a simple power law.

Partial melting in the Fe-S system was investigated at high pressures because of its importance to understanding the formation, composition, and thermal structure of the Earth's core. Earlier studies at very high pressure (>25 GPa) took place before the discovery of Fe{sub 3}, which compromised the interpretation of those results. Furthermore, they relied on textural criteria for melting that are difficult to apply at high pressure. In this study synchrotron X-ray diffraction was used to monitor coexisting metal and sulfide at high pressures and temperatures, during laser heating in a diamond anvil cell. The criterion for melting was the disappearance of one of the two coexisting phases, and reappearance upon quench. Temperatures of eutectic melting between Fe and Fe{sub 3}S were bracketed in this way up to 60 GPa, and a lower bound was established at 80 GPa. The accuracy of the melting point measured in these studies was improved through modelling of the axial temperature distribution through the thickness of the sample; this indicated an {approx}6% correction to the spectroradiometrically determined temperature. The Fe-Fe{sub 3}S eutectic composition remains close to 15 wt% S up to 60 GPa.

Using a diamond anvil cell and double-sided laser-heating coupled with synchrotron X-ray diffraction, we determined phase relations for three compositions of Fe-rich FeNi alloys in situ at high pressure and high temperature. We studied Fe with 5, 15, and 20 wt.% Ni to 55, 62, and 72 GPa, respectively, at temperatures up to ∼3000 K. Ni stabilizes the face-centered cubic phase to lower temperatures and higher pressure, and this effect increases with increasing pressure. Extrapolation of our experimental results for Fe with 15 wt.% Ni suggests that the stable phase at inner core conditions is hexagonal close packed, although if the temperature at the inner core boundary is higher than ∼6400 K, a two phase outer region may also exist. Comparison to previous laser-heated diamond anvil cell studies demonstrates the importance of kinetics even at high temperatures.

International audience; Earth's core is an iron-rich alloy containing several weight percent of light element(s), possibly including silicon. Therefore, the high pressure-temperature equations of state of iron-silicon alloys can provide understanding of the properties of Earth's core. We performed X-ray diffraction experiments using laser-heated diamond anvil cells to achieve simultaneous high pressures and temperatures, up to similar to 200 GPa for Fe-9 wt % Si alloy and similar to 145 GPa for stoichiometric FeSi. We determined equations of state of the D0(3), hcp + B2, and hcp phases of Fe-9Si, and the B20 and B2 phases of FeSi. We also calculated equations of state of Fe, Fe11Si, Fe5Si, Fe3Si, and FeSi using ab initio methods, finding that iron and silicon atoms have similar volumes at high pressures. By comparing our experimentally determined equations of state to the observed core density deficit, we find that the maximum amount of silicon in the outer core is similar to 11 wt %, while the maximum amount in the inner core is 6-8 wt %, for a purely Fe-Si-Ni core. Bulk sound speeds predicted from our equations of state also match those of the inner and outer core for similar ranges of compositions. We find a compositional contrast between the inner and outer core of 3.5-5.6 wt % silicon, depending on the seismological model used. Theoretical and experimental equations of state agree at high pressures. We find a good match to the observed density, density profile, and sound speed of the Earth's core, suggesting that silicon is a viable candidate for the dominant light element.

High-pressure experiments and theoretical calculations demonstrate that an iron-rich ferromagnesian silicate phase can be synthesized at the pressure–temperature conditions near the core–mantle boundary. The iron-rich phase is up to 20% denser than any known silicate at the core–mantle boundary. The high mean atomic number of the silicate greatly reduces the seismic velocity and provides an explanation to the low-velocity and ultra-low-velocity zones. Formation of this previously undescribed phase from reaction between the silicate mantle and the iron core may be responsible for the unusual geophysical and geochemical signatures observed at the base of the lower mantle.

CuS, or covellite (hexagonal symmetry), was compressed in a diamond anvil cell at room temperature up to a pressure of 45 GPa, and studied using x rays from both a Mo Kα source and a synchrotron. The x‐ray diffraction spectrum of CuS disappears by about 18 GPa. The presence of Cu fluorescence lines in all spectra and the reappearance of diffraction lines upon decompression confirm that CuS undergoes reversible pressure‐induced amorphization at this pressure. A third‐order Birch–Murnaghan equation of state fit to the diffraction data below 11 GPa yields a bulk modulus of 89±10 GPa with a pressure derivative of −2±2 for covellite. Further compression up to 45 GPa shows three to four diffraction lines of very low intensity, implying some high pressure ‘‘ordering’’ of the amorphous phase. The Raman spectra obtained indicate that the changes in structure are probably due to the twisting or the distortion of covalently bonded CuS4–CuS4 units in different directions.

Natural olivine with 12 mol % Fe 2 SiO 4 and synthetic orthopyroxenes with 20% and 40% FeSiO 3 were studied beyond the pressure–temperature conditions of the core–mantle boundary. All samples were found to convert entirely or partially into the CaIrO 3 postperovskite structure, which was recently reported for pure MgSiO 3 . The incorporation of Fe greatly reduces the pressure needed for the transition and establishes the new phase as the major component of the D″ layer. With the liquid core as an unlimited reservoir of iron, core–mantle reactions could further enrich the iron content in this phase and explain the intriguing seismic signatures observed in the D″ layer.

Compressed under ambient temperature, graphite undergoes a transition at approximately 17 gigapascals. The near K-edge spectroscopy of carbon using synchrotron x-ray inelastic scattering reveals that half of the pi-bonds between graphite layers convert to sigma-bonds, whereas the other half remain as pi-bonds in the high-pressure form. The x-ray diffraction pattern of the high-pressure form is consistent with a distorted graphite structure in which bridging carbon atoms between graphite layers pair and form sigma-bonds, whereas the nonbridging carbon atoms remain unpaired with pi-bonds. The high-pressure form is superhard, capable of indenting cubic-diamond single crystals.

Acoustic wave velocities have been measured in RbCl and CsCl to 17 GPa. The congressional wave velocities vary with density as VP (km/s) = 3.4143 (84) + 1.368 (11) Δρ for the high-pressure phase of RbCl and VP (km/s) = 2.8247 (57) + 1.1469 (77) Δρ for CsCl (numbers in parentheses indicate uncertainties on the last digit). The equations of state of RbCl and CsCl have also been determined at room temperature to ∼30 GPa. Eulerian finite strain equation of state parameters are ρ02 = 3.3068 (10) Mg m−3, K0T = 17.9 (10) GPa and K0T′ = 5.23 (29) for RbCl in the B2 structure, and K0T = 17.01 (29) GPa and K0T′ = 5.49 (15) for CsCl. The accumulated set of data for the alkali chlorides under high pressure is used to test several empirical relations between velocity and density. A simple logarithmic relation is shown to allow the most accurate extrapolation from low-pressure velocity data. Gruneisen's parameters are also calculated from the velocity and density data, allowing a complete thermal equation of state to be developed for the alkali chlorides.

MgS (B1 or NaCl structure) was compressed in a diamond anvil cell at room temperature up to a pressure of 54 GPa. Diffraction experiments were performed using X-rays from both a rotating anode generator and synchrotron beam line. The B1 phase of MgS is stable to 54 GPa. A third order Birch-Murnaghan equation of state fitted to the data yields a bulk modulus of 78.9 ± 3.7 GPa with a pressure derivative of 3.71 ± 0.34. Universal equation of state parameters are virtually identical. The thermodynamic parameters obtained are consistent with most of the existing theoretical predictions. For comparison, a bulk modulus of 56.0 ± 1.4 GPa, with a pressure derivative of 5.3 ± 1.0, is calculated for CaS using previously published data. Finally, the pressure-volume data and elastic parameters of several alkali halides and alkaline-earth sulfides and oxides are used to evaluate the applicability of the Born model for these compounds.

Iron-magnesium-silicate perovskite was melted in a laser-heated diamond anvil cell and monitored by thermal analysis. Two signals were identified at each pressure: a lower temperature signal and a higher temperature signal. The lower signal may correspond to the temperature where iron diffuses rapidly. The higher signal would then correspond to melting of magnesium-enriched silicate perovskite. Alternatively, the lower signal may be the solidus where (Fe.14Mg.86)SiO3 undergoes incongruent melting to an iron-magnesium-enriched solid plus a silica-enriched liquid. Then, the higher signal would be the liquidus. The slope for the melting curve (the higher signal) between 30GPa and 94GPa is slightly negative. Fitting to a straight line gives a value for the slope of −2.5 ± 0.6K/GPa. At 30GPa, the lower signal is ≈300K below the melting curve. They converge slightly at higher pressures and differ by ≈140K at 94GPa. Our melting curve is several hundred degrees (K) below previous estimates.

Compressional wave velocities as a function of pressure have been measured in pressure-amorphized anorthite to 14 GPa at room temperature. A weighted linear fit to the data gives a zero-pressure value of Vp = 6.35±0.12 km s−1, with a slope of 0.154±0.016 km s−1 GPa−1. It is shown that these compressional wave velocities compare well with bulk sound speeds inferred from shock-wave experiments on anorthite above the phase transition at 30 GPa, assuming a value of Poisson's ratio in the range of 0.15 to 0.20. All crystalline assemblages which have been suggested as high-pressure phases of anorthite are shown to possess zero-pressure bulk sound speeds in the range 6.9 km s−1 to 8.1 km s−1, well above the value of 5 km s−1 extrapolated from the shock-wave data. It is therefore proposed that the shock Hugoniot of anorthite above 30 GPa represents the properties of the pressure-amorphized phase, rather than those of the equilibrium high-pressure phases.

A previously undocumented effect in the laser-heated diamond anvil cell, namely, the transport of molten species through the sample chamber, over distances large compared to the laser beam diameter, is presented. This effect is exploited to determine the melting behavior of high-pressure silicate assemblages of olivine composition. At pressures where beta-spinel is the phase melted, relative strengths of partitioning can be estimated for the incompatible elements studied. Iron was found to partition into the melt from beta-spinel less strongly than calcium, and slightly more strongly than manganese. At higher pressures, where a silicate perovskite/magnesiowuestite assemblage is melted, it is determined that silicate perovskite is the liquidus phase, with iron-rich magnesiowuestite accumulating at the end of the laser-melted stripe.

The far-infrared reflectivity of Fe0.91O was investigated from atmospheric pressure up to 33 GPa at room temperature using synchrotron Fourier transform infrared reflectivity techniques in conjunction with the diamond-anvil cell from 100-700 cm-1. The frequency of the fundamental transverse optic (TO) mode was found to be nearly independent of pressure up to 4.6 GPa, followed by an increase in the TO frequency with pressure up to the rhombohedral phase transition at {approx}8 GPa. In addition, a second weak mode at 583 cm-1 at atmospheric pressure was well resolved and found to shift to higher frequency and increase in strength with pressure. This localized mode arises from the presence of vacancies in the crystal structure, and the relative strength of this mode suggests pressure-induced charge localization near the vacancy sites. The data was fit with the classical Lorentz model with the addition of a plasmon resonance. This allowed an estimation of the electrical conductivity as well as plasmon-phonon coupling energies. The pressure dependencies of the dielectric properties of westite have been quantified, and their pressure derivatives show a change in sign near the pressure-induced rhombohedral phase transition. Classical theories relating dielectric, vibrational, and elastic properties are evaluated, and in the casemore » of the bulk modulus, the theory fails to reproduce accepted literature values.« less

A laser heating system is described for use with diamond anvil high pressure cells that directly senses and stabilizes visible thermal radiation emitted by hot samples. This technique stabilizes sample temperatures better than other methods and allows superior temperature control. Calibration of the system was checked by measuring the melting temperatures of five metals at ambient pressure. Assuming literature values for spectral emissivity, the calibration was found to be accurate to 3.3% (based upon one standard deviation of the percentage error from published melting temperatures). Performance of the laser heating system was verified by heating iron foil at 13 GPa. With the sample intensity unstabilized, mean temperature was 3003 K with a standard deviation of 144 K, while with it stabilized, mean temperature was 3051 K with a standard deviation of 8 K. For a given wavelength‐dependent emissivity, the difference between the actual temperature and the greybody temperature increases as the temperature in...

The room temperature static compression of the high pressure B2 phase of KCl has been measured in a diamond anvil cell to 56 GPa. Birch-Murnaghan finite strain analysis extrapolated to zero pressure gives V 02 V 01 = 0.8483 ± 0.0057 and K 02 = 28.7 ± 0.6 GPa, with K 02 ' constrained to a value of four. Universal equation of state parameters are virtually identical. Although these results generally concur with previous work at lower pressures, significant differences exist in the zero pressure equation of state parameters extrapolated from the two sets of data. Recent ab initio structure calculations predict compressions in good agreement with our experimental results. Born models utilizing exponential repulsive potentials account well for the data, even over the large (50%) range of compression observed.

The static compression of the high pressure (B1) phase of zinc sulfide has been measured at room temperature using X-ray diffraction through a diamond anvil cell between 11 and 45 GPa. The volume, the bulk modulus, and its derivative of the B1 phase, extrapolated to zero pressure, are V 02 V 01 = 0.851 (±0.021), K 02 = 85.0 (±3.8) GPa , and K'02  4 based on a third order Birch-Murnaghan finite strain analysis. Alternatively, a universal equation of state fit yields V 02 V 01 = 0.895 , K02 = 47.5 (±8.2) GPa, and K02 = 6.2 (±1.1). The large discrepancy between these two sets of zero pressure parameters highlights the difficulty in extrapolating properties from a high pressure equation of state. Thin samples of the high pressure phase of zinc sulfide were discovered to transmit light in the visible region. Optical absorption spectra were recorded from 15.9 to 36.2 GPa. With the increase of a few GPa in pressure past the phase transition at 15 GPa, the direct band gap narrows to approximately 2.1 eV. Further compression induces a slight, but poorly resolved, increase of direct gap energy with compression ( ≈ 10 meV GPa ).

[1] We present a multitechnique approach to experimentally determine the elastic anisotropy of polycrystalline hcp Fe at high pressure. Directional phonon measurements from inelastic X-ray scattering on a sample with lattice preferred orientation at 52 GPa in a diamond anvil cell were coupled with X-ray diffraction data to determine the elastic tensor. Comparison of the results from this new method with the elasticity determined by lattice strain analysis of radial X-ray diffraction measurements showed significant differences, highlighting the importance of strength anisotropy in hcp Fe. At 52 GPa, we found that a method which combines results from inelastic scattering and pressure-volume measurements gives a shape in the velocity anisotropy close to sigmoidal (with a faster c and slower a axis) a smaller magnitude in the anisotropy and compared to velocities based on the lattice strain method which gives a bell shape velocity distribution with the fast direction between the c and a axes. We used additional results from nuclear resonant inelastic X-ray scattering to constrain errors and provide additional validation of the accuracy of our results.

Estimation of the thermal elastic effect is necessary for the calibration of the pressure and temperature conditions during laser-heated diamond anvil cell experiments, since above 800K, the standard technique of using ruby florescence to measure pressure fails. Continuum calculations based upon the thermoelastic equations for an elastic medium were used to estimate the thermal pressure resulting from a radially symmetric temperature gradient in an elastic sphere with zero displacement on its surface. This calculation corresponds to the thermal pressure generated in a laser-heated diamond anvil cell sample that is compressed without a pressure medium. This solution must fall between circumstances where the sample is held at constant pressure and where the sample is held at constant volume. It is shown here that the thermal pressure in an elastic medium with a Gaussian temperature gradient is approximately 40–60% of the thermodynamic value of the thermal pressure in a material raised to some constant temperature with the volume constrained to be constant. Even though the thermal pressure correction can be significant in terms of the total pressure that the sample experiences, these calculations indicate that the correction can be estimated to approximately 10%.

Copper selenide (CuSe) was compressed in a diamond anvil cell at room temperature up to a pressure of 52 GPa and studied using energy dispersive x-ray diffraction and Raman spectroscopy. CuSe is nearly isostructural with copper sulfide (CuS), and a previous study indicates that copper sulfide undergoes reversible pressure-induced amorphization at 18 GPa. The intensity of the x-ray diffraction peaks for CuSe decrease slowly, however, they never completely disappear up to a pressure of 52 GPa. The third-order Birch–Murnaghan equation of state fit to the data yields K0=96.9±5.3 GPa and K0′=4.1±0.5. Vinet’s universal equation of state yields essentially identical parameters. Raman spectroscopy demonstrates that upon compression, the S–S bond in CuS compresses differently than the Se–Se bond in CuSe, possibly accounting for the different high pressure behavior of these two very similar compounds.

[1] Iron (Fe) and coexisting Fe3S were studied simultaneously using synchrotron X-ray diffraction and a laser-heated diamond anvil cell (DAC). The thermal equation of state (EOS) of Fe3S was investigated up to pressures of 80 GPa and temperatures of 2500 K. Fitting a third-order Birch-Murnaghan EOS to the room temperature data yielded bulk modulus K0 = 156(7) GPa (values in parentheses are standard deviation) and pressure derivative K′0 = 3.8(3) calibrated against NaCl in the B2 structure. The room temperature data were also calibrated against the EOS of hcp-Fe for comparison and aid in the determination of the thermal pressure contribution of Fe3S. This fit yielded bulk modulus K0 = 113(9) GPa and pressure derivative K′0 = 5.2(6). The thermal pressure contribution of Fe3S was assumed to be of the form ΔPthermal = αKTΔT, where αKT is constant. The best fit to the data yielded αKT = 0.011(2) GPa K−1. Iron and Fe3S coexisted in the high-pressure, high-temperature experiments, and a density relationship between Fe and Fe3S was found to be linear and independent of temperature. Extrapolation of the data to the core-mantle boundary (CMB), using an assumed temperature of 3500 K at the CMB, a 2% volume change associated with melting, and applying a small adjustment to account for the nickel content of the core indicates that 14.7(11) wt % sulfur is adequate to resolve the density deficit of the outer core.

[1] We studied the nuclear resonant x-ray scattering of iron hydride (FeHx) up to 52 GPa. Coupled with hydrostatic x-ray diffraction data, the partial phonon density of states measured by nuclear resonant inelastic x-ray scattering provides information on sound velocities and the Fe contribution to thermodynamic parameters. In particular it constrains the aggregate shear velocity and shear modulus for comparison to seismic observations. We found that VS (km/sec) = 0.023*P (GPa) + 3.2. A loss of magnetism was observed with synchrotron Mossbauer spectroscopy at 22 GPa, lower than theoretically predicted but consistent with the observed anomalous velocity behavior. Results confirm that FeHx could be a major light element bearing phase for explaining the core density deficit relative to pure Fe. Formation of FeHx by reaction with water would be expected to leave a signature in the mantle.

Magnesiowüstite [(Mg,Fe)O] is the second most abundant mineral of Earth's lower mantle. Understanding its stability under lower mantle conditions is crucial for interpreting the physical and chemical properties of the whole Earth. Previous studies in an externally heated diamond anvil cell suggested that magnesiowüstites decompose into two components, Fe-rich and Mg-rich magnesiowüstites at 86 GPa and 1,000 K. Here we report an in situ study of two magnesiowüstites [(Mg 0.39 ,Fe 0.61 )O and (Mg 0.25 ,Fe 0.75 )O] at pressures and temperatures that overlap with mantle conditions, using a laser-heated diamond anvil cell combined with synchrotron x-ray diffraction. Our results show that addition of Mg in wüstite (FeO) can stabilize the rock-salt structure to much higher pressures and temperatures. In contrast to the previous studies, our results indicate that Mg-rich magnesiowüstite is stable in the rock-salt structure in the lower mantle. The physical and chemical properties of magnesiowüstite should change gradually and continuously in the lower mantle, suggesting that it does not make a significant contribution to seismic-wave heterogeneity of the lower mantle. Stable Mg-rich magnesiowüstite in lowermost mantle can destabilize FeO in the core–mantle boundary region and remove FeO from the outer core.

lower the density of iron, but significantly changes its compressibility neither in the bcc phase, nor at high pressures in the hcp phase. Upon comparison with the Preliminary Reference Earth Model, the calculated equations of state (EOS) of hcp-Fe85Si15, using the Mie-Gruneisen EOS, indicate that an outer core containing about 8-10 wt.% Si and inner core containing about 4 wt.% Si in iron would satisfy the seismological constraints. Addition of silicon into iron increases the bulk sound velocity of iron, consistent with silicon being a light element in the Earth's core. INDEX TERMS: 1015 Geochemistry: Composition of the core; 3919 Mineral Physics: Equations of state; 3924 Mineral Physics: High-pressure behavior; 3954 Mineral Physics: X ray, neutron, and electron spectroscopy and diffraction; KEYWORDS: high pressure, light elements, iron-silicon alloy, Earth's core, X-ray diffraction, equation of state

Because of its close packing and high symmetry, the NiAs (B8) type is considered to be an important structure in high pressure mineral physics. To provide systematic insight on the behavior of materials assuming this structure under extreme compression, a compression study of NiAs-type NiS is presented. A fit of the data of the Birch-Murnaghan equation of state yields a bulk modulus and its pressure derivative of KoT = 156 (10) GPa and K'oT = 4.4 (12), respectively. Above 45GPa, X-ray diffraction lines of this phase were unreadable, but laser-heating of the sample induced a phase change at these pressures. Because the new phase could not be recovered at ambient conditions, and only two diffraction lines were observed at high pressures, its structure remains unknown.

A laser-heating system was constructed that can stably heat through a diamond-anvil cell, probe regions as small as 6.7 μm in diameter, and accurately measure visible spectral radiation at frequencies up to 30 Hz. Spectra can be analyzed to measure temperature and to identify phenomena of interest such as melting. Reflecting optics were used where dispersion is critical. Temperature is stabilized by feedback from the thermal emissions of laser-heated samples and by attenuating the laser beam using a liquid-crystal variable waveplate and a fixed polarizer. The response time of the stabilizer is estimated to be 1.25 ms. The laser-heating system is suitable for experiments at the pressures and temperatures of the Earth's lower mantle and core. With this apparatus, a magnesium-iron-silicate perovskite, (Mg .88 Fe .12 ) SiO 3 , was melted in a diamond-anvil cell at pressures between 25 GPa and 85 GPa. Natural bronzite and synthetic enstatite were the starting materials. Polycrystalline samples with no pressure medium and single crystals in a NaCl pressure medium were melted. Melting was determined in situ by thermal analysis and corroborated by the appearance of glass in temperature-quenched samples. Corrections for radial temperature gradients in the samples were obviated by aperturing the collecting optics of the spectrometer. Corrections for axial temperature gradients were estimated from a simple model. The slope of the melting curve between 35 GPa and 85 GPa is 5.2 ± 0.8 K/GPa and sub-adiabatic. The melting temperature extrapolated to the core-mantle boundary is 4502 ± 176 K. Both are significantly less than recent estimates.

Lattice strains have been measured as a function of the angle, ψ, from the diamond cell stress axis in a sample of gold and rhenium at pressures of 15–37 GPa. Experiments were conducted using X-ray transparent gaskets made from beryllium. The differential stresses supported by gold and rhenium have been characterized to 37 GPa. It is also shown that proper choice of the diffraction geometry allows recovery of a quasi-hydrostatic compression curve under these highly non-hydrostatic conditions. X-ray elastic moduli have also been determined, and while good agreement with previous data is achieved for gold, there is a large discrepancy between the present results and extrapolated ultrasonic data for rhenium.

The compressibilities of ScS and a series of Sc(1)(-)(x)()S metal defect structures (x > 0) were studied using a diamond anvil cell and synchrotron radiation at pressures up to 73 GPa. The results show that the bulk modulus increases with x, in contrast to the usual view that a less dense material is more easily compressed. The results are in qualitative agreement with existing band structure calculations designed to probe the mechanism of the ready metal atom loss from ScS at high temperatures. They are also in agreement with new calculations, both of the tight-binding and the first-principles FLAPW type. Numerically the latter show an increase in the bulk modulus of 30% between ScS and Sc(0.875)S and are in exact numerical agreement with the experimental results.

The compressibility of α-Mns (alabandite) was determined by x-ray analysis using a Mao-Bell type diamond anvil cell. The zero pressure bulk modulus (K0) is 74±2 GPa with the pressure derivative of the bulk modulus (K′o) fixed at four. Allowing (K′o) to vary yielded a statistically better fit with K0 = 88±6 GPa and k′0 = 2.2±0.6. Our data combined with the data of McCammon (1991) gave Ko = 73±1 GPa with k′o fixed at four. A fit with k′o allowed to vary yielded ko = 75±2 GPa and k′o = 3.7±0.4. Alabandite transformed from the B1 structure (NaCl-type) to an unknown high-pressure phase at 26 GPa. The high-pressure phase has lower than hexagonal symmetry and it is stable to at least 46±4 GPa.

A new technique actively controls thermal radiation and monitors sample properties during laser-heating in a diamond anvil cell. The technique can be described as a qualitative application of thermal analysis. Discontinuities in temperature, laser power, visible thermal radiation, or in their derivatives as functions of time can be associated with the enthalpy of phase transitions (such as melting) or with changes in material properties (such as emissivity).

The compressional wave velocities of polycrystalline NaCl and KCl have been measured to over 17 gigapascals, with the use of Brillouin scattering and the diamond anvil cell. This pressure corresponds to 40% compression for NaCl and 60% compression for KCl (including the volume change across the B1-B2 transition). The data obey Birch's Law, which predicts that the velocity of each material is linear with density, except across the B1-B2 phase transition in KCl. This deviation from Birch's Law can be rationalized in terms of an interatomic potential model wherein the vibrational frequencies of the nearest neighbor bonds decrease when going to the eight-coordinated B2 structure from the six-coordinated B1 structure.

The infrared reflectance of iron was studied using high-pressure synchrotron radiation methods up to 50 GPa at room temperature in a diamond anvil cell of 1000-8000 cm(-1) (1.25-10 microm). The magnitude of the reflectivity shows a weak pressure dependence up to the transition from the body centered cubic (alpha) to hexagonal close packed (epsilon) phase transition, where a discontinuous change in both the slope and magnitude of the reflectivity was observed. Reflectance spectra were corrected for diamond absorption and treated with a Kramers-Kronig analysis to extract the optical constants; the emissivity of iron was derived from Kirchoff's law. The pressure and wavelength dependence of the emissivity is characterized by an empirical function for 1.5-1.9 microm; this wavelength range is useful for spectroradiometric temperature measurements from 1000 K up to approximately 2500 K. Alpha-Fe is a nonideal emitter; however, epsilon-Fe behaves as an almost perfect greybody in the infrared up to the highest pressures of the measurements. Temperature measurements based on the spectroradiometry of iron samples should take into account the wavelength dependent emissivity below the alpha-epsilon phase transition at approximately 13 GPa.

[1] The phase relations of an Fe10wt%Ni alloy were investigated in a diamond anvil cell up to 86 GPa and 2382 K. Adding nickel into iron stabilizes the fcc phase to higher pressures and lower temperatures compared to pure iron, and a region of two-phase coexistence between fcc and hcp phases is observed. Iron with up to 10 wt% nickel is likely to be in the hcp structure under inner core conditions. The axial ratio (c/a) of hcp-Fe10wt%Ni has a weak pressure dependence, but it increases substantially with increasing temperature. The extrapolated c/a ratio at ∼5700 K and ∼86 GPa is approximately 1.64, lower than a theoretically predicted value of nearly 1.7 for hcp-Fe at 5700 K and inner-core pressure. A lower c/a ratio should have an effect on the longitudinal anisotropy of the hcp phase, and hence, may influence the interpretation of the seismic wave anisotropy of the inner core.

CW laser heating through a diamond anvil cell by means of a focused Nd:YAG laser has resulted in the achievement of temperatures in the 1500-5000 K range for pressures of 10-100 GPa. Temperatures are determined radiometrically, with an accuracy of about 200 K, and temperature variation across the laser-heated spot is derived by means of spatial filtering with a slit that can be scanned. Melting temperatures are thus determined either on the basis of observed temperature at the liquid-solid interface or on that of the peak temperature at which glass is first produced with increasing laser power.

An experimental determination of the melting of Mg0.9Fe0.1SiO3 under lower mantle conditions is presented. Samples were compressed and laser heated (Nd-YAG) in a Mao-Bell-type diamond cell. The temperature fields generated in the sample are measured using a simple tomographic technique and spectroradiometer. Mg0.9Fe0.1SiO3 in the perovskite structure melts at 3000 ± 300 K independently of pressure over the range of 22–65 GPa. The melting curve of Mg0.9Fe0.1SiO3 represents an experimental upper bound upon the geotherm from 700 to 1500 km. This melting curve corresponds to a homologous temperature of 0.70 or 0.85 for one- or two-layer mantle convection, respectively. This is consistent with either one- or two-layer mantle convection and a lower mantle of constant viscosity (based upon homologous temperature arguments).

Results from room-temperature static-compression experiments are presented for the $B2$ phase of NaCl between 25 and 70 GPa (250---700 kbar) pressure. The volume and bulk modulus of the $B2$ phase, extrapolated to zero pressure, are $\frac{{V}_{02}}{{V}_{01}}=0.929(\ifmmode\pm\else\textpm\fi{}0.071)$ and ${K}_{02}=36.2(\ifmmode\pm\else\textpm\fi{}4.2)$ GPa based on a finite-strain analysis of the new data (${V}_{01}$) is the zero-pressure volume of the $B1$ phase). Our results are in good agreement with ab initio calculations of the NaCl ($B2$) equation of state. Also, we find that the bulk modulus probably decreases across the $B1\ensuremath{-}B2$ transition, in accord with previous predictions. We illustrate the use of high-pressure compression studies on different polymorphs for empirically determining interatomic potentials of ionic crystals.

The compression of Au has been measured at room temperature to 70 GPa (700 kbar) using x‐ray diffraction through a diamond‐anvil cell and the ruby‐fluorescence pressure scale. Based on these data, the isothermal bulk modulus and its pressure derivative at zero pressure are K0T =167 (±11) GPa, and K′0T=5.5 (±0.8). These results are in excellent agreement with ultrasonic measurements of the elastic constants as well as an equation of state based on shock‐wave data. Hence, this study represents an independent experimental confirmation of both the ruby fluorescence pressure scale, and the predicted equation of state of the proposed Au pressure calibration standard. We derive a thermal equation of state for gold by inverting all equation‐of‐state data simultaneously. From this, we extend the gold pressure‐calibration standard to cover the range 0–200 GPa in pressure and 300–3000 K in temperature.

We have investigated the phase relations in the iron-rich portion of the iron-silicon (Fe-Si) alloys at high pressures and temperatures. Our study indicates that Si alloyed with Fe can stabilize the body-centered cubic (bcc) phase up to at least 84 gigapascals (compared to ∼10 gigapascals for pure Fe) and 2400 kelvin. Earth's inner core may be composed of hexagonal close-packed (hcp) Fe with up to 4 weight percent Si, but it is also conceivable that the inner core could be a mixture of a Si-rich bcc phase and a Si-poor hcp phase.