In high temperature and high pressure experiments, it is essential to have a reliable pressure calibration standard over a wide temperature and pressure range. The Au pressure scales reported by Jamieson et al.(1982), Anderson et al.(1989), and Shim et al.(2002), and the MgO pressure scales by Jamieson et al.(1982), Matsui et al.(2000), and Speziale et al.(2001) have recently been often used to estimate pressure at high temperatures in both diamond-anvil and multi-anvil experiments. Detailed description and criticism will be made for each of these pressure scales at high temperatures in relation to the physical properties of the deep earth.In addition, the electronic thermal pressure of Au at high temperatures and high pressures are calculated using the uniform electron gas model with the random-phase approximation(Kanhere et al., 1986) for the exchange-correlation potential, in order to estimate the contribution of the electronic thermal pressure to the total pressure at the temperatures and pressures conditions found in multi-anvil, laser heated diamond-anvil experiments, or shock compression measurements of Au.

View full abstract

In situ X-ray diffractions on high-pressure minerals, which are Mg-perovskite, Ca-perovskite, stishovite, CF-type aluminous phase, hexagonal aluminous phase, and scheelite-type zircon (reidite), were carried out using a laser-heated diamond anvil cell (LHDAC) up to a pressure of 80 GPa with X-rays from a synchrotron radiation source at BL10XU of SPring-8 and BL13A of PF. At each pressure increments, the samples were heated using a YAG laser to relax the deviatoric stress in the samples.

View full abstract

[in Japanese]

View full abstract

Water plays an important roles in geodynamic process in the Earth's interior. For example, water affects the physical properties of the Earth's materials, such as elastic velocity, viscosity, plastic deformation, diffusion, and electrical conductivity. The diffusivity of hydrogen in mantle minerals affects a distribution and circulation of water in Earth's interior. The diffusion rate of hydrogen in olivine has been measured at high pressure in the past, but no data exist in the wadsleyite, which is a major constituent mineral in the mantle transition zone. So we tried to determine the diffusion rate of hydrogen in beta-phase of Mg₂SiO₄. We examined the kinetics of hydrogen diffusion in wadsleyite by measuring the OH distribution in polycrystals annealed under hydrothermal conditions. Diffusion experiments were conducted at 17 GPa and 1073-1273 K. A Kawai-type multi-anvil high-pressure apparatus was used, with a 5 mm truncation edge length on the tungsten carbide anvils. The concentration of hydroxyl were determined by Fourier Transform infrared spectrometer(FTIR). In this research, the diffusion experiments were conducted using two methods. One is that wadsleyite was surrounded by NaCl + Mg(OH)₂ powder which was used as the hydrogen source. Another is that dry and hydrous wadsleyite were used as the diffusion couple. The determined value of diffusion coefficients were 1*10^-11 m²/s^-2*10^-10 m²/s, and about 2*10^-13 m²/s ^-3*10^-12 m²/s, respectively.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

To examine the Fe and Al solubilities in MgSiO₃ perovskite, laser-heated diamond anvil cell experiments on the system (Mg,Fe)SiO₃-Al₂O₃ were performed at 30-60 GPa and 1800-2000 K.The synthesized samples were analyzed by synchrotron X-ray diffraction and analytical transmission electron microscopy (ATEM). The results show the following points. The product phases were Mg-perovskite, magneshiowustite and stishovite for 0 and 5 mol% Al₂O₃, while corundum joined the above phase assemblage for 25 and 50 mol% Al₂O₃. The chemical analyses of the perovskite phases showed that the Fe content in perovskite gradually increased with the Al content for low Al contents, but that the increasing rate became higher for higher Al contents. Under an electron microscope metal iron was recognized to coexist with perovskite for Al containing runs. These results indicate that in Mg-perovskite the substitution Mg²⁺ + Si⁴⁺ -> 2Al³⁺ is dominated for low Al contents, while the substitution Mg²⁺ + Si⁴⁺ -> Fe³⁺ + Al³⁺ increases for higher Al contents. This suggests that under the Al-rich circumstances in the lower mantle, Mg-perovskite contains the higher iron content than under the Al-poor circumstances and the part of iron is ferric.

View full abstract

High pressure and high temperature experiments using a multianvil cell and a laser heated diamond anvil cell (DAC) were carried out to study the phase relations and structure variations of CaSiO₃-CaTiO₃ perovskites.Analyses of the quenched product of the multianvil cell experiments showed that perovskite was the only existing phase in the whole area of this system above 12.5 GPa at 1773 K, and perovskite was orthorhombic (Pbmn) for the region from CaTiO₃ to about 40 mol percent CaSiO₃,and for the region from 40 mol percent to 65 mol percent CaSiO₃ it had a double perovskite-like structure whose unit cell is double of CaSiO₃ cubic perovskite. For the region from 65 mol percent to endmember CaSiO₃ the recovered samples were amorphous. There was no two-phase region of perovskites.Meanwhile in-situ synchrotron X-ray diffraction experiments of this system using a laser-heated DAC showed that at 30 GPa and 1800 K perovskite in the intermediate composition has peaks of double cubic perovskite whose indices were all even or all odd, indicating a face-centered cubic lattice, but that the specimes at room temperature have the extra peaks which violate a face-centered cubic lattice. These results combined with the observation by ATEM indicate that Ca(Si,Ti)O₃ perovskite in the intermediate composition is double perovskite with a face-centered cubic lattice under high pressure and high temperature, but it inverts to the lower symmetry phase with a simple cubic lattice by the Si-Ti ordering during quenching. X-ray diffraction peaks at 30 GPa and room temperature showed that perovskites from 50 mol percent to 90 mol percent CaSiO₃ were pseudo-double perovskite.

View full abstract

Subsolidus phase relation in hydrous mid-ocean ridge basalt (MORB) was determined by high pressure and temperature in situ X-ray diffraction study. SPEED1500 was used for this experiment. We used a rhenium tube heater with LaCrO₃ thermal insulator, the magnesia window for the X-ray path, the W3%Re-W25%Re thermocouple for monitoring temperature of the experiments. Pressure was calculated from the lattice constants of gold. We observed that increase of the intensity of the X-ray diffraction peaks of garnet phase clearly at 20.7 GPa and 1200C (Run mb41019), whereas we observed intensity increase of the diffraction lines of Mg-perovskite in three runs made at 21.6 GPa and 1100C (Run mb51017), 23.1GPa and 1200C (Run mb43028), and 25.5 GPa and 1200C (Run mb42027). These observations imply that transformation from the garnet bearing to the Mg-perovskite bearing assemblage occurs at around 21.6 GPa and 1100C. The boundary of the perovskite appearance in the wet MORB composition has a positive slope expressed as P(GPa)=0.0056(+-0.0009)T (C)+15.4(+-1.1) based on the pressure scale by Anderson et al. (1989). The boundary is about 1 GPa lower than that determined under the dry condition by Hirose et al. (1999). There is a crossover between the post garnet phase boundary in MORB and the post spinel phase bouhdary in peridotite at around 1300 C at the around 22 GPa by Anderson gold scale. The result has a significant implication of the subduction of the crustal materials into the lower mantle. In the low temperature conditions of the slabs below 1300C, post garnet transformation occurs at lower temperature compared to the post spinel transformation of the peridotite layer of the slabs. The crustal component in the wet slab can easily descend into the lower mantle.

View full abstract

X-ray absorption fine structure (XAFS) measurements on liquid alkali (Li₂O)-GeO₂ under high pressure and high temperature have been performed using a cubic type high-pressure device and synchrotron radiation. The local structure around Ge ions has been investigated up to 9 GPa at 1273 K. The Ge-O distances obtained with the analysis of the extended X-ray absorption fine structure (EXAFS) and the modification in the X-ray-absorption near-edge structure (XANES) indicate that the present liquid germanate consists of a mixture of fourfold and sixfold Ge around 3 GPa and of the complete sixfold Ge above 4 GPa. The transition pressure is much lower than that observed in germanate glass with room temperature compression.

View full abstract

Segregation of incompatible elements results in a marked difference between the composition of the interiors and the boundaries of grains within the rocks that comprise Earth's mantle. At chemical equilibrium, the molar concentrations of solutes in grain matrices, XGM, and the grain boundaries, XGB, can be related through the grain boundary segregation energy, gamma:
X_GB/(X_GB0-X_GB) = exp (gamma / RT)*X_GM/(1-X_GM)
where X_GB0 is the saturation level of X_GB, R is the gas constant, T is the temperature. This thermodynamic model has not been applied to grain boundaries in Earth materials; grain boundaries have simply been treated as possible sinks for impurities in rocks.
To test whether the thermodynamic model describes segregation to grain boundaries in mantle rocks, we investigated partitioning of Ca between the grain interiors and grain boundaries of olivine in natural and synthetic olivine-rich aggregates. We calculated partitioning curves for Ca (r_i = 0.1 nm) using Eq. (1) for different temperatures; assuming that the absorption energy, gamma, is equal to misfit strain energy, U, and that the segregated Ca is confined to the M sites at the grain boundary plane (X_GB0 = 1). Good overall agreement suggests that the observed Ca partitioning is consistent with the thermodynamic equilibrium model in Eq. (1), with the segregation confined to the grain boundary plane and the partitioning controlled by misfit strain energy. The consistency of the Ca partitioning in both natural and synthetic samples our model predictions suggests that partitioning according to this segregation model accurately describes the partitioning of Ca, and likely other incompatible divalent cations, in Earth's mantle.

View full abstract

The core of the terrestrial planets consists of Fe and the lighter alloying elements. Oxygen and sulfur are the candidates for the major alloying elements. Dreibus and wanke (1985) showed the composition of the Martian core is sulfur-rich. Ringwood (1977) suggested that Mars is more oxidized than the Earth. Therefore, it is important to study the melting relation of Fe-FeS and Fe-FeO-FeS systems at high pressure and temperature. We investigated the Fe-FeS and Fe-FeO-FeS systems at 23GPa, 850-1300C using a uniaxial 3000ton press installed in Tohoku University. The solidus temperature of the Fe-FeS system at 23GPa is about 1075C, whereas that of the Fe-FeO-FeS system is about 900C at the same pressure, suggesting the Martian core is partially molten.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

Stable phase and equation of state of iron-silicon alloys, Fe-8.7 wt.% Si and Fe-17.8 wt.% Si, have been investigated using diamond-anvil cell techniques up to 196 GPa and 124 GPa, respectively. The unit cell volumes at room temperature were measured by X-ray diffraction method using synchrotron radiation and an imaging plate (IP). Fe-8.7 wt.% Si transformed from bcc phase into the bcc and hcp phases at 16 GPa and room temperature, and the phase transformation to the hcp phase was completed at 36 GPa. The high-pressure phase of Fe-8.7 wt.% Si with hcp structure was stable up to 196 GPa. No phase transition in bcc-Fe-17.8 wt.% Si was observed up to 124 GPa. The pressure-volume data were fitted to a third-order Birch-Murnaghan equation of state (BM EOS) with zero-pressure parameters: V₀ = 22.2(8) A³, K₀ = 198(9) GPa, and K'₀ = 4.7(3) for hcp-Fe-8.7 wt.% Si and V₀ = 179.41(45) A³, K₀ = 207(15) GPa and K'₀ = 5.1(6) for Fe-17.8 wt.% Si. The density and bulk sound velocity of hcp-Fe-8.7 wt.% Si indicate that the inner core would contain a few weight % of Si.

View full abstract

Density of silicate melt at high pressure is one of the most important properties to understand magma migration in the planetary interior. Because of experimental difficulties, the density of magma at high pressure is poorly known. Katayama (1996) recently developed a new in situ density measurement method based on the density dependency of X-ray absorption in the sample. In this study we made, as a first step, preliminary measurements of the density of basaltic glass by the absorption method.Experimental TechniquesMORB composition (SiO₂-Al₂O₃-FeO-MgO-CaO-Na₂O) glass was prepeared by cuenching by falling into water the melt heated at 1603K. the glass was grided into and used as starting material. X-ray absorption method was fallowing.
I/I0 = exp(-mrt)
I0 is intensity of incident beam, I is the intensity of the transmitted beam. m is the mass absorption coefficient, r is the density of the sample, and t is the X-ray path lengh. If t and m were known, We got the density of the sample, by mesuring I and I0. In order to know t, we used a single diamond cyrinder as sample cupsele, which is less absorbed and less deformed. t is as following t = 2*(r2-x2)1/2 r is hemisphere of cyrinder inner, and x is distance from center of the capsule.
For generation of high pressure and high temprature, We used cubic-type multi-anvil apparatus (SMAP180) installed on the beamline (BL22XU) at Spring-8, Japan. Tungsten carbide anvils with a square center flat of 6*6 mm2 were used. Boron-epoxy pressure medium was used. Monocro X-ray was used, whose energy was 25 keV and whose size is 0.1*0.1 mm2 by 2 slits, Experiment were up to 5 GP at 300 K and heated up to 773K . And down to 1atm at 773KResults and DiscussionUp to 5GPa at 300 K, the density was increasingAt 500 c ,density of basalt glass was densified by 20 %At down path the density was decreasing. we caculated bulk modulus of basaltic glass K=46(4) (GPa), when K'=4.

View full abstract

With increasing pressure, CaSiO₃ wollastonite transforms to walstromite which dissociates into Ca₂SiO₄ larnite + CaSi₂O₅ titanite, and finally transforms to CaSiO₃ perovskite which is accepted as one of major constituents of the lower mantle. Among the phases, CaSi₂O₅ titanite and CaSiO₃ perovskite convert to triclinic phase and amorphous phase, respectively, on release of pressure. CaSi₂O₅ component is dissolved into CaTiSiO₅ titanite with increasing pressure, and a complete solid solution is formed in the system at about 9 GPa. The titanite solid solutions of all the compositions except for CaSi₂O₅ endmember are quenched at ambient conditions. Thermodynamic properties and stability relations of these Ca-silicates have not yet been fully clarified, and particularly enthalpy measurements of Ti-bearing silicates have been scarcely reported. In this study, we have successfully applied drop-solution calorimetric method combined with gas-bubbling technique to measure enthalpies of titanite solid solutions in the system CaTiSiO₅-CaSi₂O₅ that are dissolved very slowly in lead borate solvent. Using the measured enthalpies, we have estimated enthalpy of CaSi₂O₅ titanite, and have calculated high pressure phase relations in the system CaO-SiO₂-TiO₂. xCaTiSiO₅-(1-x)CaSi₂O₅ titanite solid solutions (x=0.1, 0.2, 0.5, 0.75, 1.0) were synthesized using a multianvil apparatus, confirming single phase titanite solid solutions by powder X-ray diffraction. Pellets of the titanite solid solutions were dropped into 2PbO.B₂O₃ solvent in a Calvet microcalorimeter kept at 705C. In the drop-solution run, Ar gas was passed though the solvent to produce bubbles that enhanced the sample dissolution. The measured enthalpies of the titanite solid solutions decrease linearly with increasing x, which indicates that the solid solution is ideal. The extrapolated enthalpy of CaSi₂O₅ titanite was used to obtain enthalpies of transitions for CaSiO₃ walstromite + SiO₂ coesite -> CaSi₂O₅ titanite (44.9+-2.5 kJ/mol) and 1/3Ca₂SiO₄ larnite + 1/3CaSi₂O₅ titanite -> CaSiO₃ perovskite (66.0+-4.5 kJ/mol). These enthalpy data were used to calculate stability field of titanite solid solution in the CaTiSiO₅-CaSi₂O₅ system, and the calculated results are generally consistent with experimental data by Knoche et al. (1998). The present calculation combined with some quench experiments also indicates that transition boundary for larnite + titanite (R) Ca-perovskite has a slope of 1.8MPa/K at 14.6GPa and 1600C. Our calculated phase relations suggest that coexisting larnite and titanite in Kankan diamonds (Joswig et al., 1999) occurred at depth of about 300-420 km in the mantle, and that Si-rich titanite found in Kokchetav ultrahigh pressure metamorphic belt (Ogasawara et al., 2002) occurred at depth deeper than about 200 km.

View full abstract

[in Japanese]

View full abstract

In high-P/T in situ X-ray observation using Kawai-type apparatus, generated pressure is usually determined from the unit cell volume calculated from d-spacings of diffracted planes of pressure standard materials, such as cubic crystals of Au, Pt, NaCl and MgO.
However, possible deviatoric stress exerting on the standard causes serious error to the calculated pressure. In order to determine the appropriate pressure, it is indispensable to evaluate stress state of the standard material. In hydrostatic condition, strains of d-spacings are uniform over diffraction lines. Therefore, it can be interpreted that difference among strains of d-spacings is caused by deviatoric stress. If elastic constants of standard material are known, we can calculate stress from strain using Hooke's law. In the conference, we would like to discuss a new simple (and fully numerical) method established by the present study to estimate magnitude of deviatoric stress in a Kawai-type apparatus.

View full abstract

Introduction Plagioclase is one of the major minerals in the earth crust. Information of its melt and glass provide a cru for understanding properties of magma. In this study, the structures of plagioclase glasses densified by static compression were investigated by Raman spectroscopy and X-ray diffraction measurement.Experimental Plagioclase glasses with composition of Ab100, Ab50An50, and An100 were densified under the conditions of 4.0GPa-500C, 7.7GPa-500C and 7.7GPa-20C by a cubic anvil press. The densities of these glasses were measured by heavy liquid technique. Raman spectra and X-ray diffraction intensities are also measured.Results and Discussions By compression at 500C, densities increased with increasing pressure. The density increase increases with An content. The pressure variation of Raman spectra and FSDP (First Sharp Diffraction Peak) position of X-ray diffraction profiles may indicate that the density increase of plagioclase glass is due to the decrease of T-O-T angle, increase of 3-membered rings of TO₄ tetrahedra, and increase of 6-fold Al. Large density increase of An100 glass may suggest that increase of 6-fold Al largely contribute density increase of plagioclase glass.

View full abstract

The structural changes and densification mechanisms of silica glass with compression up to 15GPa at 300 degrees centigrade has been investigated by infra-red absorption and X-ray diffraction measurement. The density increases with apllied pressure. But the rate of increase changed between 7.5GPa and 10GPa. It may suggest that are two the densification mechanisms change at this pressure range. Infra-red spectra of densified silica glass showed a systematic shift of the high frequency bands (nu=~1100cm^-1) and a remarkable decrease in the intensity of the low frequency bands (nu=~470cm^-1). They can be attributed to a decrease in Si-O-Si angle and possibly to an existence of high coordinated silicon. Structures of these densified silica glass have been analyzed based on the radial distribution function curves (RDF(r)) derived from the X-ray scattering intensities. It showed almost no pressure variations in atomic distances and angles, coordination number. However, the intensity of FSDP (First Sharp Diffraction Peak), whichi is attributed to the intermediate range structure of the glass (ring structure of the glass), decreases to 10GPa. It may indicate increase in a number of non-bridging oxygen. And linear change in the FSDP position with increasing pressure provides evidence for increase of small ring structure (three or four memberd rings of SiO₄ tetrahedra).

View full abstract

[in Japanese]

View full abstract

Perovskite-type oxides such as Ln³⁺ doped SrCeO₃ and BaCeO₃ where Ln³⁺ represents trivalent lanthanide elements, have been found to exhibit appreciable proton conduction in hydrogen-containing atmosphere at high temperature. Among these, the chemical stability of SrZrO₃-based oxide is much better, and Yb-doped SrZrO₃ oxide shows the highest proton conductivity. The proton conductivity depends on the amount of dopant Yb ions and shows the maximum value when substituted for Yb ions by 10mol%. The oxide ion vacancies and/or electron holes are produced by substitution of tetravalent zirconium ions by trivalent ytterbium ions. In our previous studies, the mass loss and gain of oxygen in the Yb-doped SrZrO₃ was observed in heating and cooling process, respectively.In this study, temperature dependence of XAFS spectra of Perovskite-type proton conductor SrZr_0.9Yb_0.1O_3-δ at Sr and Zr K-edges and Yb L3, L2 and L1-edges have been investigated. The chemical shift to lower energy and change in XANES spectra are observed with increasing temperature. No change in XANES spectra and in threshold energies of the Yb L3-, L2- and L1-edge was observed. This indicates that the effective charges (oxidation states) of Sr and Zr ions change with the oxygen loss while the effective charge of Yb ion were maintained. The unit cell volumes increase with increasing temperature with a slight bent between 700 and 950 K. This bent of the unit cell volumes vs. temperature is considered to be due to loss of oxygen, which is consistent with the result of a weight loss in TGA observed in the same temperature range. The mean-square relative displacements (MSRD) of the backscattering atoms with respect to the absorbing atom were also determined by EXAFS. Its temperature dependence increases with a slight bent between 700 and 950 K, which is considered to be due to a local structure change accompanied by the loss of oxygen rather than that by the phase transition at 948 K. It is considered that the changes of local structures and effective charges of Sr and Zr ions must play an important role in the protonic conductivity.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

La₂/3TiO₃-based compounds with an A-site-deficient perovskite-type ABO3 structure exhibit high ionic conductivities and dielectric constants. The orthorhombic (Cmmm) - tetragonal (P4/mmm) phase transition occurs reversibly and continuously at 623(2) K in the La_0.63Ti_0.92Nb_0.08O₃. Cation vacancies in the compounds are ordered in a lattice site at lower temperature. The temperature dependence of EXAFS Debye-Waller factors in La_0.63Ti_0.92Nb_0.08O₃ perovskite with cation deficiency was investigated with the cumulant expansion method. The measurements of the Ti and Nb K-edge and La L3-edge EXAFS spectra were carried out in the transmission mode at temperature up to 800 K. The obtained local bond distances of 1.95(1) A for Ti-O, 1.98(1) A for Nb-O and 2.65(1) A for La-O are significantly smaller than the expected values on the Shannon's radii. This characteristic is attributable to the presence of the cation deficiency. The effective pair potentials, V(u)=au2/2+bu3/3!, were evaluated and the potential coefficients a and b for the Ti-O bond in La_0.63Ti_0.92Nb_0.08O₃ are 6.6(1) eV/A2 and -42(3) eV/A3, respectively. The Ti-O bond in the A-site-deficient La_0.63Ti_0.92Nb_0.08O₃ has relatively larger anharmonic and soft potential coefficient than that in ordinary CaTiO₃ perovskites.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

Mellite (Al₂[C₆(COO)₆]16H₂O) was synthesized by a reaction of an aqueous solution of mellitic acid with an aqueous solution of aluminum nitrate hydrate at room temperature. The crystal structures of synthetic mellite investigated at room and low temperatures (298 and 128 K) are orthorhombic, space group Ibca, Z = 8, unit cell parameters over the temperature 298-128 K ranging from a = 15.5412(4) - 15.5983(8) A, b = 15.5543(6) - 15.6333(6) A, c = 23.145(1) - 23.186(1) A, V = 5600.7(3) - 5643.9(4) A³. Increases of a and b cell parameters with lowering temperature lead to the expansion of unit cell volume, whereas c cell parameter is slightly decreased by the deformation of Al1-O1-H angles. Comparison of the bond lengths between 298 K and 128 K shows that all Al-O and C-O bonds measured at 128 K are longer than those measured at 298 K. However, hydrogen bond distances remain constant between 298 K and 128 K. The C-C bonds in benzene ring are also almost elongated under the low temperature although the changes of C-C bond distances are smaller than that of Al-O and C-O bonds. When the condition again reverts to room temperature, the bond distances and unit cell volume at the low temperature immediately shrinks to those at the starting temperature. Since the contraction of thermal vibrations with lowering temperature causes the reductions of the degree of polarization in electron clouds of Al³⁺, C⁴⁺ cations and O^2- anions, their reductions are responsible for decreasing the covalent character for the Al-O and C-O bonds in this mellite. On the other hand, bond strength for hydrogen bonds is little influenced by the lowering temperature as compared with that for ionic bonds in mellite. Therefore, hydrogen bonds become so strengthened relative to ionic bonds that the bond distances within Al(H₂O)₆ octahedra and [C₆(COO)₆] anions are isotropically expanded in the low temperature.

View full abstract

[in Japanese]

View full abstract

Mn³⁺ distribution in octahedral M1, M2 and M3 sites of piemontite and its effect on the structural changes were investigated in terms of X-ray Rietveld analysis of Ca₂Al_3-pMn_pSi₃O₁₂(OH)-piemontite synthesized by hydrothermal experiments at 200-350 MPa and 500-550 C. Piemontites were synthesized almost as single-phase from p=0.5, 0.75, 1.0 and 1.1 starting materials, whereas those crystallized from p=1.5 and 1.75 starting materials were associated with minor amounts of bixbyite and parawollastonite. The maximum Mn³⁺-content was 1.3 apfu. The site preference order of Mn³⁺ is M3>M1>>M2. Mn³⁺ enters M1 site even if M3 site is not filled by Mn³⁺ and Mn³⁺ occupancy of M1 site attains up to 0.30. K_D (=(Mn/Al) in M1/(Mn/Al) in M3) ranges between 0.024 and 0.120. If both Mn³⁺ and Fe³⁺ distribute in the octahedral sites, K_D of Mn³⁺ would become smaller. The a-, b- and c-dimensions vary with p as follows: a (A)=0.037p²-0.062p+8.879; b (A)=0.005p²+0.085p+5.585, c (A)=0.021p³-0.008p²-0.022p+10.155. The nonlinear variations of a- and c-dimensions with increasing p are caused by those of the M1-O distance (=0.115p²-0.081p+1.930 A) and O5-Si3-O6 angle (=11.322p²-10.165p+100.200 deg), respectively. The variations of the M3-O1 and M3-O2 distances reflect the distortion of M3 site, and that of the M1-O1 the M1 site deformation.

View full abstract

The crystal structures of two rare earths calcium hydrous carbonate minerals, kimuraite-(Y) and lokkaite-(Y), are determined with specimens from Hizen-cho, Saga Prefecture, Japan. The data were collected with MoKa radiation on a CCD detector. The structures were solved and refined with Shelx-97. The refinements converged to R1 = 5.79 and 8.47 % for kimuraite-(Y) and lokkaite-(Y), respectively. Kimuraite, CaY₂(CO₃)₄.6H₂O, is orthorhombic, with space group Iba2, a = 24.015(2)A, b = 6.0425(6)A, c = 9.2596(9), Z=4. It consists of the tengerite-type corrugated sheet of nine-coordinated Y-polyhedra and CO₃ triangles, and the layer of calcium ions and water molecules. The sheet and the layer stack alternately in the crystal structure of kimuraite-(Y). Lokkaite-(Y), CaY₄(CO₃)₇·9H₂O, is orthorhombic, with space group Abm2, a = 6.1044(8)A, b = 39.176(5)A, c = 9.2193(12), Z=4. In lokkaite, a pair of the corrugated sheets is located between the Ca-H₂O-layers. The symmetry of the corrugated sheet in lokkaite-(Y) is lower than that of tengerite-(Y), while the sheet in kimuraite-(Y) keeps the symmetry. These results confirm the previous prediction on the structural relationships among tengerite-(Y), kimuraite-(Y) and lokkaite-(Y).

View full abstract

Crystal structure of brodtkorbite (Cu₂HgSe₂) was solved by Monte Carlo and R-factor method on the basis of the XRD data reported by Paar et al.(2002) using Structure Model-Assembly Program.

View full abstract

Sapphire of intense orange-pink hue came out in the jewellery market.It was doubtful of so-called "surface diffusion" at first due to curious color distribution. However,it made clear that color of sapphire was induced by heating with diffusion of Be from external source. The results of investigation of natural and treated sapphires by new technique inductive coupled plasma mass spectrometer.

View full abstract

The specimen used in this study was a single crystal of hydrous ringwoodite synthesized by Ohtani and Mizobata (1998) using a multi-anvil apparatus at conditions of 1680C and 22 GPa. Electron microprobe analysis showed a chemical composition of 42.83wt% SiO₂, 56.42 wt% MgO, yielding a total wt% of 99.25 (H₂O excluded) with Mg/Si being 1.97. The H₂O content measured by SIMS was 0.2(0.004) wt %. The unit cell content is calculated to be Mg_1.97SiH_0.03O₄. Sets of X-ray diffraction intensities up to 7.9 GPa were measured with a single crystal of 35x35x24 micron using synchrotron radiation at the beam line BL-10A, Photon Factory, High Energy Accelerator Reasearch Organization, Tukuba, Japan. The wave lengths were calibrated by the unit cell constants of a ruby standard crystal. The modified Merrill-Bassett type diamond anvil pressure cell was used. The 4:1 fluid mixture of methanol and ethanol was used for pressure medium and SUS301 plate was used for gasket. The pressure was calibrated using the ruby fluorescence method. The crystallographic data and the final atomic parameters at ambient condition and at 7.9 GPa are given in Table 1. The atomic parameters were refined by SHELEX-93. Interatomic distances and angles are given in Table 2. The compression of T-O distance was 0.25 % while the compression of M-O was 1.8 %. The compression of the unit cell axial length was 1.3 %. The compressibility of the unit cell is close to the compressibility of the MO₂ octahedron. This fact indicate that the compression of the crystal structure of hydrous ringwoodite is governed by the compression of MO₆ octahedron

View full abstract

Many of Mn₂O₃ compounds have a corundum structure with space group of R3c. Compounds with r(M)/r(O)>0.87 are characterized by A-type rare earth structure. Those with 0.60< r(M)/r(O)<0.87 are C-type rare earth structure. Bixbyite (Mn,Fe) ₂O₃ has a cubic structure of C-type rare earth type. Pure M₂O₃ has an orthorhombic structure of Pbnm which is a little deformed structure of cubic structure. Jahn-Teler effect du to Mn³⁺ induces the distortion from the cubic site symmetry of ideal octahedron. Prewitt et at., (1969) proposed Mn₂O₃ changed from C-type to corundum structure under compression. Shono et al., (1997) reported a high-pressure phase at about 20GPa. In the present experiment structure analysis of bixbyite has been conducted by single crystal diffraction study through using diamond anvil cell with laboratory X-ray source and SR at BL02B2 Spring-8 and PF BL-10A. Powder X-ray diffraction study was executed at PF BL-18C with increasing pressure up to 40 GPa in order to elucidate the compressibility and pressure-induced phase transition. Diffraction peaks of high-pressure phase were observed above 20GPa. No abrupt change was found in the lattice constants and the lattice constants continuously change. The high-pressure phase was not quenchable, inferring a reversible phase transformation. The isothermal Birch-Murnagham equation of state was applied to P-V-T data. The observed data of Ko=176.5(4.8)GPa and Ko=7.56(0.67) were indicates a little smaller than those of Al₂O₃, V₂O₃, Cr₂O₃ and Fe₂O₃. These data imply Mn₂O₃ is less rigid in comparison of corundum structure. Bixbyite has two different metal sites of Mn1 and Mn2, which have the site symmetry of (8a) and .2. (24d), respectively. The former is an almost ideal octahedron of Mn1O6 but the latter has a largely distorted octahedron. All six Mn1-O bonds are equivalent. But three different Mn2-O bonds,, Mn2-O(1), Mn2-O(2) and Mn2-O(3) show different compression with pressure. Mean Mn2-O distance more decrease than Mn1-O. Both octahedra do not show a remarkable Jahn-Teller effect due to Mn³⁺. A high-pressure phase has monoclinic symmetry rather than rhombohedral such as corundum, ilmenite LiNbO3 type structure. Charge dispropor- tionation such as 2Mn³⁺=Mn²⁺+ Mn⁴⁺ is now tested by measuring the magnetic moment under high pressure.

View full abstract

Phase A (Mg₇Si₂O₈(OH)₆) is a dense hydrous magnesium silicate phase in which two fully occupied hydrogen sites have been found. Previously (Kagi et al., 2000) carried out neutron diffraction measurements to 3.2 GPa in fluorinert (3M Company) pressure-transmitting medium using the Paris-Edinburgh opposed-anvil cell. Reliable refinements were not possible using data collected beyond this pressure because of the relative complexity of the diffraction pattern and the substantial peak broadening due to deviatoric stress introduced above the freezing point of Fluorinert, . Recently however an encapsulated TiZr gasket has been developed by Marshal and Francis (2002) and this allows the use of deuterated methanol-ethanol mixtures as a pressure transmitting medium. We applied this new gasketting technique to the deuterated Phase A sample and obtained neutron diffraction patterns at pressures up to ca. 8 GPa on the POLARIS beam line at the UK pulsed neutron facility, ISIS at the Rutherford Appleton Laboratory. The quasi-hydrostatic condition minimized peak broadening and allowed us to take advantage of the higher resolution at low d-spacing at POLARIS to refine high quality structural models for Phase A with the Rietveld technique, without resorting to constraints involved in the previous study. Furthermore, the bulk modulus of Phase A was determined to be 106.2 (1.6) GPa by the single crystal X-ray diffraction method at high pressure. X ray diffraction intensities were collected by a four-circle diffractometer with monochromatized MoKa X-ray (50kV, 40mA) or synchrotron light source at Photon Factory, KEK. This bulk modulus gave consistent pressure values from the unit cell volumes of Phase A and coexisting brucite. The Raman spectra of Phase A measured at high pressures using a diamond anvil cell and a relationship between OH stretch frequency and atomic positions of hydrogen atoms will be presented.

View full abstract

Introduction It has been well known, natural garnets exhibit zoning of many elements and especially the distribution of trace heavy elements in its crystal structure attracts interest since such information may include the history of the mineral genesis. SR-XRF imaging was carried out to reveal the distribution of trace heavy elements in garnet from the Sanpou mine, Okayama. From the results, the garnet contained Sn, W and rare earth elements (REE) as trace heavy elements and the zoning of these elements were parallel to garnet crystal shape and their distribution varied from core to rim. The purpose of the present study is what produced such characteristic zoning in garnet structure. So first, we carried out single-crystal X-ray structure analysis to reveal the size of the crystallographic sites of constituent atoms, and then, estimated the sites of trace elements based on their ionic radii. In order to obtain direct information of the site of the trace elements, we have utilized local structure information obtained by XAFS analysis. Result & Discussion Single-crystal X-ray analysis revealed that garnet from the Sanpou mine is grandite garnet (Ca₃(Fe_0.71Al_0.29)₂(SiO₄)₃) and possess a cubic unit cell with space group Ia-3d and lattice constant of a=12.0065(6)&Aring;. Sn K-edge XAFS analysis showed that tetravalent Sn was coordinated by the 6 oxygen and the Sn-O distance was 2.00&Aring;.This results agreed well with the coordination and bond length of the octahedral site determined by the structure analysis, so the substitution of Sn for the octahedral site was apparent.For other elements except for Sn, the crystal-chemical structural environment was examined based on the ionic radii. Element which has ionic radii close to the size of the crystallographic site is able to substitute for the site. Therefore, we compared the ionic radii of the trace heavy elements with the size of each site. First, the sizes of the 3 atomic sites in the garnet were determined using metal-oxygen distance obtained from the structural analysis fixing the ionic radius of oxygen as 1.40&Aring;. The metal-oxygen distance for the dodecahedral, octahedral, and tetrahedral sites were calculated as 1.00&Aring;, 0.60 and 0.25&Aring;, respectively. Then they were compared with the ionic radii after Shannon and Prewitt(1969,1970). As for REE zoning, the zoning position shifted outside with increasing the atomic number. This tendency corresponded to that of the lanthanide contraction, so it seemed that ionic radii played important role in the zoning during the crystal growth.

View full abstract

A single crystal of gibbsite (Al(OH)₃)was studied by in-situ monochromatic Laue method in order to characterize new unknown phase of alumina (Al₂O₃), X-phase, on dehydration process from gibbsite into corundum (a-Al₂O₃). Experiments were performed at BL4B-1, KEK, using X-ray diffraction equipment for micro-size crystal with Imaging Plate and a heating stage. A specimen of gibbsite crystal was put on the heating stage, heated onto 600 C, and exposed into X- rays monochromatized in 1.0A by Si (111) double crystal monochromator and collimated into 100 micrometer. The measurement time and interval of heating temperature were synchronized. Diffraction from X phase of alumina was observed on a Laue photograph around 200 C area, and there reflections were broadened, that means the size of crystallite of X- phase is very small as well as gamma-alumina.

View full abstract

Effects of ionic size in octahedron and two kinds of tetrahedron sites, M-, T1-and T2-sites, on the thermal expansion behavior of cordierite were systematically investigated by molecular dynamics simulations. The thermal expansion along the a axis can be controlled by the ionic size in M-site, and that along the c axis can be controlled by that in M- and T2-sites. However, the ionic sizes in M- and T2-sites have the opposite effect each other.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

Humite group minerals from Sri-Lanka were examined with a high resolutiontransmission electron microscope(HRTEM) and by a single crystal oscillationcamera with imaging plate and rotating anode X-ray generator. It was observedthat the number of chondrodite lamellae were always multiple of three unit cellsin the specimen in which chondrodite made frequent intergrowth with humite.Several week spots which correspond to the periodicity of 28.2A parallel to c*were found in the specimen of humite, indicating the existence of the structurewhich is built of alternating modules of humite and chondrodite. Clinohumitespecimens sometimes reveal complex lamellar faults.

View full abstract

Hureaulite is one of the most persistent low temperature transition metal phosphate hydrates derived usually from hydrothermal attack of the primary phosphate triphylite-lithiophilite series. A single crystal of Mn-hureaulite[Mn₅(HOPO₃)(PO₄)₂(H₂O)₄] has been readily synthesized hydro-thermally in reaction mixture of 4H₃PO₄:3(CH₃COO)₂Mn4H₂O:100H₂O at 180^oC for 3 days. Similar hydro-thermal technique was applied in order to produce a single crystal Fe-hureaulite [Fe₅(HOPO₃)(PO₄)₂(H₂O)₄], however obtained compounds were Fe₃(PO₄)₂H₂O and NH₄Fe₄(PO₄)₃ phases, only. This result agrees with the previous reports that the substisution of Mn by Fe in the hureaulite structure is limited to a maximum of about 40%. Structural features for a variety of transition series phosphates prepared in this study are also discussed.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract