The morphological stability of hydrous liquid droplets at grain boundaries of eclogite minerals in the basaltic part of a subducting slab were investigated based on experimental constraints for the dihedral angles. We measured the dihedral angles in the eclogite–H2O system at a temperature of 1000 °C and at pressures ranging from 4 to 16 GPa. The dihedral angle of hydrous liquid versus garnet–garnet (θGG–L) increased with increasing pressure from 46° at 4 GPa to 66° at 12 GPa, although it showed a weak decreasing trend at pressures higher than 12 GPa. The dihedral angle of hydrous liquid versus clinopyroxene–clinopyroxene (θCC–L) was almost constant with increasing pressure (61 and 59° at 4 and 10 GPa, respectively). The dihedral angle of hydrous liquid versus garnet–clinopyroxene (θGC–L) was 73–76° at 4–10 GPa, and it was always higher than θGG–L and θCC–L. By applying the morphological stability criteria for liquid in a system with two solid phases (garnet and clinopyroxene) and bond percolation theory for a three–dimensional lattice of tetracoordination, we found that the hydrous liquid was isolated at the grain edges and corners of eclogite minerals in the cold slab under a wide range of pressure conditions of the upper mantle from 4–14 GPa when the grain size of garnet was equal to that of clinopyroxene. Thus, basaltic crust containing hydrous liquid droplets may carry water to the lowermost upper mantle and the mantle transition zone when the slab is cold.
Crystal structures of the cubic and tetragonal spinel phases of Zn2GeO4 and the modified spinel phase of Zn2SiO4 were refined by Rietveld analysis of synchrotron powder X–ray diffraction data. The Zn2GeO4 cubic spinel phase was found to have an inverse spinel configuration. The Zn2GeO4 tetragonal spinel phase is isostructural to Zn2TiO4, where half of Zn occupies the tetrahedral site, and the remaining Zn and Ge are ordered in two octahedral sites. In the modified spinel phase of Zn2SiO4, Zn occupies the octahedral sites. Independent individual cation–oxygen distances in the Zn2GeO4 tetragonal spinel and Zn2SiO4 modified spinel phases were calculated using the bond valence repartition method, and are in reasonable agreement with the refined structures.
In addition to the conventional 14C and Th/U dating methods, thermoluminescence (TL) dating has been applied to calcite, but has been less popular partly because the luminescence responses for different types of radiation are unclear. To report more reliable TL ages for calcite, the fundamental characteristics of its response to radiation exposure were investigated and related to chemical composition. Relative TL factors for calcite after beta and gamma irradiation normalized with quartz, hereafter termed the beta and gamma factors, were measured as 0.19–0.34 and 0.16–0.33, respectively. These lower values than for quartz may be caused by differences in common substitution elements in calcite (20Ca, 25Mn, and 26Fe) versus quartz (3Li, 11Na, 13Al, and 14Si), and the interaction between mediums with different atomic numbers and radiation energies. The beta factor is higher than the gamma factor for some samples. These samples show relatively higher concentrations in lighter elements (up to Ba); thus, the concentration of minor elements may cause differing behavior between beta and gamma rays. The gamma factor may depend on Mn concentration; however, the elements most affecting the beta factor remain unknown. The accumulated dose from alpha rays is affected by sample thickness because of the spatial energy density around the center of the alpha track and luminescence detection range. Thus, for accurate alpha efficiency measurements, evaluation of the effective alpha ray range and luminescence detection thickness is important. The alpha efficiency against the gamma factor, known as the k–value, increases with Mn concentration. Previous studies have suggested that the alpha efficiency is lower than beta and gamma efficiency because the ionization density produced by alpha particles is so great that the thermoluminescence traps in the tracks’ central core become saturated. This leads to a much greater proportion of the ionized electrons being wasted compared with beta and gamma radiation. Thus, we concluded that luminescence traps increase with increasing Mn concentrations.
Enstatite (MgSiO3) sample was prepared by conventional solid state reaction at 1500 °C, followed by cooling to ambient temperature with different cooling rates (0.04–500 °C/s). Quantitative analysis of enstatite phases in the samples by Rietveld refinement and 29Si MAS NMR revealed that a large amount of metastable protoenstatite (30–40 mol%) along with clinoenstatite was preserved in all samples. The presence of protoenstatite was also confirmed by micro–Raman spectroscopy. The crystal structures of protoenstatite and clinoenstatite at room temperature were refined using the Rietveld method, and were essentially the same as those reported by previous studies. As the solid state reaction method is commonly used to prepare synthetic clinoenstatite, the present result implies that the ‘clinoenstatite’ starting materials used in previous phase equilibrium studies of MgSiO3 pyroxene might contain overlooked protoenstatite, which could affect the interpretation for phase transition and thus those results should be reexamined.
Samples of the diatom Nitzschia cf. frustulum, collected from Lake Yogo, Siga Prefecture, Japan, were cultured in the laboratory. Organic components of the diatom cell were removed by washing with acetone and sodium hypochlorite. The remaining frustules were studied by scanning electron microscopy coupled with energy dispersive X–ray spectroscopy (SEM–EDX), Fourier–transform infrared (FTIR) spectroscopy, and synchrotron X–ray diffraction. The results showed that the spindle–shaped diatom frustule was composed of hydrous amorphous silica. Pressure–induced phase transformation of the diatom frustule was investigated by in situ Raman spectroscopic analysis. With exposure to 0.3 GPa at 100 °C, the Raman band corresponding to quartz occurred at ν = 465 cm−1. In addition, a characteristic Raman band for moganite was observed at 501 cm−1. From the integral ratio of Raman bands, the moganite content in the probed area was estimated to be approximately 50 wt%. With increased pressure and temperature, the initial morphology of diatom frustules was totally changed to a characteristic spherical particle with a diameter of about 2 µm. Increasing pressure to 5.7 GPa at 100 °C resulted in the appearance of a Raman band assignable to coesite at 538 cm−1. That is, with compression and heating, hydrous amorphous silica can be readily crystallized into quartz, moganite, and coesite. First–principles calculations revealed that a disiloxane molecule with a trans configuration is twisted 60° with a close approach of a water molecule, which leads to a trans to cis configuration change. It is therefore reasonable to assume that during crystallization of hydrous amorphous silica, an Si–O–Si bridging unit with the cis configuration would survive as a structural defect, and could subsequently crystallize into moganite by maintaining that geometry. This hypothesis is adaptable to the phase transformation from hydrous amorphous silica to coesite as well, because coesite has four–membered rings and is easily formed from hydrous amorphous silica under high pressure and temperature conditions.
The Tokoro Belt is a subduction complex located in eastern Hokkaido, Japan. The Nikoro Group, a constituent of the Tokoro Belt, is composed mainly of Late Jurassic to Early Cretaceous igneous rocks intercalated with bedded chert and limestone. These rocks have been regarded as fragments of seamounts. Here, we report new whole–rock geochemistry, clinopyroxene major and trace element compositions of the basalts and gabbros, and re–evaluate the origin and geodynamic setting of the Nikoro Group. The gabbros showed ophitic texture and contain fresh, large oikocrystic clinopyroxenes. Fe–Mg partitioning between the clinopyroxenes and whole–rock (i.e., melt) can be regarded as in equilibrium. The trace element composition of the clinopyroxenes within the gabbros and the whole–rock geochemistry of the gabbroic rocks are almost identical to the basalts. This evidence suggests that the whole–rock compositions were not extensively modified and the gabbroic rocks represent the melt composition. Whole–rock trace elements indicated Enriched Mid–Ocean Ridge Basalt (E–MORB)–type patterns as well as ‘garnet signatures’ [e.g., (Sm/Yb)N > 1]. After correction of the gabbros and basalts for the fractionation effect to Mg# = 0.72, the whole–rock chemistry suggests a significantly shallow lithosphere–asthenosphere boundary depth of ~ 0–km. Whole–rock (Sm/Yb)N ratios also confirm the same results. These geochemical results constrain the geodynamic setting of the Nikoro Group; the greenstone most likely originated from a plume–influenced ridge in the Pacific Ocean basin during the Middle Jurassic to Early Cretaceous.