MR43B-01
Synthesis and equation of state of perovskites in the Mg3Al2Si3O12- Fe3Al2Si3O12 system
Fe and Al are two of the most abundant minor elements in the lower mantle and may have significant effects on the stability and elastic properties of the region's dominant phase, MgSiO3 perovskite. In order to predict the chemical and physical behavior of lower mantle chemical heterogeneities enriched in Fe and Al, we investigated perovskites synthesized from pyrope-almandine garnet compositions at high pressures. Previous studies of this system above 30 GPa are limited, and no in situ studies were reported before in this pressure range (Ahmed-Zaid and Madon, 1995; Kesson et al., 1995;; Miyajima et al., 1999; Tateno et al., 2005). Five compositions along the Mg3Al2Si3O12 (pyrope)- Fe3Al2Si3O12 (almandine) join were examined: three natural garnets (Alm38, Alm54, and Alm73) and two synthetic garnet glasses (Alm90 and Alm100). X-ray diffraction experiments were performed using the laser-heated diamond anvil cell at the GSECARS (13-ID-D) and HPCAT (16-ID-B) sectors of the Advanced Photon Source. Garnet powders were mixed with 10-15 wt % Au (with Alm54, Alm73, Alm90, and Alm100) or Pt (with Alm38) as pressure calibrant and loaded between foils of NaCl as thermal insulator, pressure medium, and secondary pressure standard. All compositions were initially cold compressed to 74-77 GPa and then laser heated to 1820-2480K. Samples were decompressed gradually to 41-50 GPa with laser annealing at 5-10 GPa intervals. For Alm90 and Alm100, synthesis experiments were also carried out near 41 GPa and then the samples were compressed up to 76 GPa with laser annealing at 10 GPa intervals. Over the pressure range studied, all compositions formed orthorhombic GdFeO3- type perovskites, with trace CaCl2-type SiO2. The unit cell volume of the perovskite structure increases with iron content, with that of the almandine end-member composition 3% larger than that of pyrope-composition perovskite and 4% larger than the corresponding MgSiO3 perovskite. As with MgSiO3 perovskite, the a lattice parameter is the most compressible. The degree of distortion from the cubic structure increases with pressure; however, the addition of a large amount of Fe to the structure appears to retard this effect.
MR43B-02
P-V-T Equation of State of (Al,Fe)-bearing Mantle Perovskite and its Implications for Mantle Models
We have made significant progress on accurate measurements of P-V-T equations-of-state of mantle minerals that are of fundamental importance for developing compositional and mineralogical models of the Earth's mantle. In this study, we report new compression data on (Al,Fe)-bearing mantle perovskite up to simultaneous pressure and temperature of 113 GPa and 2120 K. The mantle perovskite was synthesized in the multi-anvil apparatus at 27 GPa and 2073 K, with chemical compositions expected in a peridotitic mantle. It contains 5.86 wt% FeO and 3.84 wt% Al2O3. The pre-synthesized perovsite mixed with Au powder was compressed in neon pressure medium in a symmetric diamond anvil cell. The sample was heated with a double-sided laser-heating system at the GSECARS 13-ID-D beamline (Advanced Photon Source). We performed 8 heating cycles in the pressure range of 30-113 GPa and temperatures up to 2560 K. In-situ synchrotron X-ray diffraction data were collected within a uniformly heated area, using a MAR-CCD area detector. The diffraction pattern contains peaks of orthorhombic perovskite, internal standard Au, and pressure medium Ne. The triplet (020, 112, and 200 diffraction peaks) of the orthorhombic perovskite is well resolve. The present dataset covers the entire P-T range of the lower mantle and requires no extrapolation to compare the mantle density profile derived from seismic observations. In light of the new P-V-T data on the (Al,Fe)-bearing mantle perovskite combined with our recent density data and spin transition of ferropericlase, we finally discuss the compositional and mineralogical models of the lower mantle.
MR43B-03 INVITED
Unified Analyses for P-V-T Equation of State of MgO: A Solution for Pressure-Scale Problems in High P-T Experiments
It has been long argued that the thermal equations of state (EOS) used for the pressure calibration standard in high-pressure and high-temperature experiments have some distinct uncertainties, which in general increase with increasing pressure. Many studies have tackled this issue from different approaches, but it is yet to be dissolved and often provides critical discrepancies in experimental results for studying the Earth's deep structure. Here we attempt a new and uniform analysis of high P-T EOS of MgO combining only pressure-scale-free reliable experimental data of zero-pressure thermal expansion data, zero-pressure and high-temperature KS data, room-temperature and high-pressure KS data, and shock compression data to 200 GPa and 4000 K. After careful test of several models based on the Mie- Grüneisen-Debye formulation, we successfully determined volume dependence of the Grüneisen parameter quite appropriate to represent the high-pressure and high-temperature behavior of MgO. The new EOS model of MgO can completely reproduce all the analyzed data within considerably small errors. We propose the finally obtained model as a primary pressure-calibration standard applicable to quantitative high- pressure and high-temperature experiments. Research supported in part by JSPS.
MR43B-04
Compression of single-crystal magnesium oxide to 118 GPa and a ruby pressure gauge for helium pressure media
Magnesium oxide (MgO, periclase) is among the most widely studied standard materials for testing experimental and theoretical methods of determining elastic properties. Because of its simple structure and geophysical relevance, knowledge of accurate elastic properties of MgO pertains to problems ranging from experimental pressure scales to interpreting Earth's seismic structure. The pressure-volume equation of state (EoS) of single-crystal MgO has been studied in diamond-anvil cells loaded with helium to 118 GPa and in a non-hydrostatic KCl pressure medium to 87 GPa using monochromatic synchrotron X-ray diffraction at GSECARS (Sector 13, APS). A third-order Birch-Murnaghan fit to the non-hydrostatic P-V data (KCl medium) yields typical results for the initial volume, V0=74.698(7)Å3, bulk modulus, KT0=164(1)GPa, and pressure derivative, K'=4.05(4) using the non-hydrostatic ruby pressure gauge of Mao et al. (1978). However, compression of MgO in helium yields V0=74.697(6)Å3, KT0=159.6(6)GPa, and K'=3.74(3) using the quasi-hydrostatic ruby gauge of Mao et al. (1986). In helium, the fitted equation of state of MgO underdetermines the pressure by 8% at 100 GPa when compared with the primary MgO pressure scale of Zha et al. (2000), with KT0=160.2GPa and K'=4.03. The results suggest that either the compression mechanism of MgO changes above 40 GPa (in helium), or the ruby pressure gauge requires adjustment for the softer helium pressure medium. We provide a revised ruby pressure gauge for helium pressure media against the primary MgO pressure scale, which will be useful for future high-pressure crystallographic studies of minerals compressed with helium in the 25-140 GPa range of the lower mantle.
MR43B-05
Elastic Shear Anisotropy of Ferropericlase in Earth's Lower Mantle: Implications for the D'' Layer
Seismic shear wave velocity anisotropy, observed in the lowermost mantle (D''), is indicative of deformation processes and carries crucial information about mantle dynamics. Unrevaling the origin of these observations can be a key to understanding structure, evolution and dynamics of our planet. At least some of the anisotropy observed in D" is very likely caused by lattice-preferred orientation (LPO) of lower mantle phases, namely ferropericlase or post-perovskite. The strength of seismic anisotropy, resulting from LPO, strongly depends on the single-crystal elastic anisotropy of the constituent phases. To date, the effects of Fe-Mg substitution and Fe spin-state on the elastic shear anisotropy of (Mg,Fe)O remain unknown. We measured the elastic shear anisotropy of single-crystal (Mg0.9Fe0.1)O up to 69 GPa in a Diamond Anvil Cell using Brillouin Scattering. (Mg,Fe)O, even at low iron contents, exhibits significantly stronger elastic shear anisotropy than MgO. Furthermore, the high-spin (HS) to low-spin (LS) transition of ferrous iron in (Mg,Fe)O in the pressure range between 40 and 60 GPa strongly increases the elastic anisotropy. Based on our results, we expect the elastic shear anisotropy to be about 50% larger for a lower-mantle ferropericlase (Mg0.8Fe0.2)O compared to pure MgO. These new findings support the idea that seismic shear anisotropy observed in the lowermost mantle is caused by LPO of ferropericlase, without the need for post-perovskite.
MR43B-06
The Temperature-Pressure-Volume Equation of State of Platinum
High temperature and high pressure equation of state (EOS) of Pt has been developed using measured shock compression data up to 290 GPa (Marsh, 1980; Holmes et al., 1989) and volume thermal expansion data between 100 and nearly 2000 K and 0 GPa (Arblaster, 1997). The contribution of electronic thermal pressure at high temperatures, estimated previously by Tsuchiya and Kawamura (2002), has also been included in the present analysis. The lattice thermal pressures at high temperatures have been estimated based on the Mie-Gruneisen relation with the Debye thermal model and the Vinet isothermal EOS. The four key EOS parameters of Pt, namely the isothermal bulk modulus K0T, its pressure derivative dK0T/dP, the Gruneisen constant γ0, and its volume dependence q, have successfully been optimized in this study. We find there is no need to include a volume dependence of q over a wide pressure range up to more than 300 GPa. The T-P-V data of Pt have also been measured up to 1600 K and 42 GPa, using synchrotron powder X-ray diffraction experiments combined with a Kawai-type multi-anvil high pressure apparatus and sintered diamond anvils. We find the newly developed T-P-V EOS of Pt is fully consistent with not only the shock compression data up to 290 GPa and volume thermal expansion data up to near 2000 K, but also the present measured synchrotron T-P-V data and recently measured T-P-V data of Pt up to 1900 K and 80 GPa (Zha et al., 2008). We note the present EOS has been developed without any pressure scale. Such excellent consistency between the EOS and experimental data over wide temperature and pressure ranges shows that the present EOS can be used as a reliable primary pressure standard for static experiments up to 300 GPa and 3000 K.
MR43B-07
Mie-Grüneisen-Einstein Equation of State as a Practical High Temperature Pressure Scale
Over the years, various models have been used for smoothing experimental PVT data and parametrization of thermoelastic functions (see, e.g., Anderson et al., 1989, JAP, 65, 1534-1543; Saxena, Zhang, 1990, PCM, 17, 45-51; Jackson, Rigden, 1996, PEPI, 96, 85-112; Pavese, 2002, PCM, 29, 43-51). The Mie-Grüneisen- Debye formula for the thermal pressure can be obtained by differentiation of the Helmoholtz free energy, and is therefore a strictly thermodynamic formula. In the high-temperature limit it is easy to calculate the thermal part of the free energy by the Einstein model, keeping in mind that the Einstein temperature, TE, is related to the Debye temperature, TD, as TE = 0.775TD. Obtained by direct differentiation of the free energy, the bulk moduli, entropy, heat capacity and other thermodynamic functions, are by construction internally consistent. An important point is the choice of the analytical form of the volumetric dependence of the Einstein (or Debye) temperature, and correct account of intrinsic anharmonicity, which makes the Einstein temperature depend not only on volume, but also on temperature. Here we propose a simple analytical form for the pressure as a function of temperature and volume. This equation can be used for constructing a practical pressure scale, as has been verified on numerous examples (Dorogokupets, Oganov, 2007; Dorogokupets, Dewaele, 2007; Fei et al., 2007; Hirose et al., 2008; Matsui, 2008; Ueda et al., 2008; Sun et al., 2008; Ono et al., 2008; Wu et al., 2008; etc.).
MR43B-08
Water Incorporation in Wadsleyite and its Influence on the Olivine-Wadsleyite Phase Boundary
Wadsleyite (wad) (Mg2SiO4), the high P polymorph of olivine (ol), belongs to the group of NAMs, Nominally Anhydrous Minerals, which incorporate water as hydroxyl in their structure via point defects. The transition from ol to wad is thought to cause the 410-Km discontinuity in the Earth's mantle. So far controversial results exist on (i) the hydrogen incorporation mechanism in wad (Smyth, 1987 and 1994; Jacobsen et al., 2005) and (ii) the influence of water on the depth of the ol-wad phase boundary (Chen et al., 2002; Frost & Dolejs 2007). To know how und how much water is incorporated in wad (Mg2SiO4), we performed experiments under hydrous conditions using a rotating multi-anvil press at 1200 ° C and 13.3-13.5 GPa. The recovered crystals were investigated by electron microprobe (EMPA), FTIR-spectroscopy in the OH stretching region at ambient condition and in-situ in a diamond anvil cell (DAC), Raman-spectroscopy (Raman), Single Crystal X-Ray diffraction (SC-XRD), Secondary Ion Mass Spectrometry (SIMS) and Transmission Electron Microscopy (TEM). To obtain information on the influence of the structurally bounded water on the stability of wad multi-anvil experiments were performed in the system Mg2SiO4-Fe2SiO4 under hydrous and anhydrous conditions. The water was quantified by Raman as 9230(±1150) ppm yielding to the molar absorption coefficient, ε, for water in wad of 34000 (±5000) (l mol-1H2Ocm-2). It represents the first direct calibration for water in wad. Its value is comparable to that of forsterite (Koch-Müller et al., 2006; Thomas et al., 2008) and implies that previous calibrations, e.g. Paterson (1984), underestimate the water concentration of NAMs. Results from EMPA, SC-XRD and TEM reveal that wad incorporates water mainly via vacancies of the M3 site. Based on the HP-IR- spectra in combination with polarized IR spectra we propose that hydrogen is bounded to O1 and the OH dipole is oriented along the O1-O4 (3.105 Å) unshared edge of a vacant M3 octahedron. In the system (Mg2SiO4), water incorporation causes a shift of the ol-wad phase boundary of 0.7 GPa at 1200 ° C to lower pressures. Experiments in the system Mg2SiO4-Fe2SiO4 indicate that water not only shifts the phase boundary to lower pressures but also affects the shape of the two-phase region. Details of the phase relations will be discussed in the presentation. References Chen J, Inoue T, Yurimoto H, Weidner DJ (2002) Geophys Res Lett 29: 1875, DOI: 10.1029/2001GL014429. Frost DJ & Dolejs D (2007) Earth and Planet Sci Lett 256: 182-195. Jacobsen SD, Demouchy S, Frost DJ, Boffa-Ballaran T, Kung J (2005) Am Min 90: 61-70. Koch-Müller M, Matsyuk SS, Rhede D, Wirth R, Khisina N (2006) Phys Chem Minerals 33: 276-287 DOI 10.1007/s00269-006-0079-9. Paterson MS, (1984) Bull Minéral 105. Smyth JR (1987): Am Min 72: 1051-1055. Smyth JR (1994): Am Min 79: 1021-1024. Thomas SM, Thomas R, Davidson P, Reichart P, Koch-Müller M, Dollinger G. (2008) Am Min (in press).