Band 97, Heft 10

The Dabie-Sulu Triassic collisional orogen in eastern Asia was created by northward subduction of the Yangtze continental-crust capped plate beneath the Sino-Korean craton. Eclogites, garnet peridotites, and surrounding country rock gneisses and marbles were all subjected to in situ UHP metamorphism, as indicated by the presence of rare but widespread coesite inclusions in eclogitic minerals and in zircon crystals in the country rocks, as well as by virtually identical metamorphic ages of various UHP rock types. Metamorphic P-T estimates, combined with investigations of mineral exsolution textures and high-P polymorphs, indicate that recovered depths of continental subduction may have exceeded 200 km. Parageneses of mineral inclusions in zoned zircon domains combined with U-Pb ages delineate a well-constrained P-T-time path, suggesting exhumation rates of 5-10 km/Myr. A similar P-T-time trajectory has been established for the microdiamond-bearing Kokchetav Massif. Thus far, however, diamond inclusions have not been confirmed from coesite-bearing zircon domains of Dabie-Sulu UHP rocks despite numerous detailed studies. Oxygen isotopes of minerals from many outcrop samples and the Chinese Continental Scientific Drilling (CCSD) project main hole cores indicate that δ 18 O depletion took place in a volume of Proterozoic protoliths exceeding 100 000 km 3 along the northern edge of the Yangtze craton. Evidently, passive-margin sediments and bimodal igneous rocks that had formed during rifting and breakup of the supercontinent Rodinia were subjected to extensive meteoric waterrock interactions attending terminal Neoproterozoic Snowball Earth conditions. Such hydrothermal alteration volatilized and depleted C from the relatively oxidized protoliths, accounting for the rare occurrences of graphite and apparent lack of microdiamond in Dabie-Sulu UHP rocks.

We present an electron backscatter diffraction, cathodoluminescence, and radiogenic U-Pb dating study of large zircon grains (0.8-1.5 mm) that show evidence of intracrystalline deformation, fracturing, grain size reduction and a large spread in U-Pb ages. The samples are from an amphibolite facies deformation zone within granulite facies anorthositic rocks (Bergen Arc, Norway). Large zircon grains show three main lattice distortion types: (I) distortions with rotations around <001> and an orientation change of ~0.3 °/μm subparallel to (100); (II) highly distorted, half circular shaped zones located at grain edges with at least 0.8-1°/μm distortions; and (III) low-angle boundary networks forming deformation zones up to 100 μm wide. Types II and III distortions exhibit significant disturbances of the otherwise homogeneous CL signature. Crystal plastic deformation with the slip system [010](100) resulted in type I distortions. Stress concentrations at grain contacts between rheologically hard grains caused localized crystal plastic deformation with minor amount of microfracturing forming type II distortions. Type III distortions formed by crystal plastic deformation often associated with inclusions using several slip systems. Distortions of types I and II show minor and moderate resetting of the original ca. 900 Ma zircon grains, respectively, due to enhanced pipe diffusion along dislocation walls. In type II distortions, accelerated lattice diffusion through the highly distorted crystal lattice, combined with exceptionally high boundary to volume ratio, caused significant chemical disturbance and age resetting to 410 Ma. Fine-grained aggregates contain grains with low internal deformation and an oscillatory zoned CL signature (Z-grains) or high internal deformation and a disturbed CL signature (D-grains). Z- and Dgrains are interpreted to have formed by heterogeneous nucleation and growth, and fracturing along strain-hardened low-angle boundaries present within types I and II, respectively. Z-grains show a clustered chemical signature with a 437 ± 11 Ma age interpreted to directly date the Caledonian amphibolite facies reworking.

This paper reports in situ observations of the dissolution behavior of the barite (001) surface in pure water at 30-55 °C using hot-stage atomic force microscopy (AFM). The dissolution at 30 and 40 °C occurred in three stages; however, at 55 °C, the dissolution behavior observed at the former temperatures started immediately after injecting water into the AFM fluid cell. The first stage of the dissolution was characterized by the retreat of the initial steps and continued for about 60 min at 30 °C and about 10 min at 40 °C. The second stage of the dissolution was characterized by the splitting of the initial <hk0> one-layer step into two half-layer steps [fast (“f”) and slow (“s”) retreat steps] with different retreat rates and by the formation of etch pits. The large difference in the retreat rate of the “f” and “s” steps led to the formation of a new one-layer step, which showed slightly faster retreat rates than the “s” half-layer step at all temperatures. The splitting of the [010] one-layer step into two half-layer steps was observed only at 55 °C. During the third stage, the development of angular deep etch pits from an initial form with a curved outline differed at each temperature. The deep etch pits grew rapidly at higher temperature, but showed at least two different retreat rates for the (001) plane at each temperature, indicating the development of the pits along different dislocations (screw and edge dislocations). The activation energies (62-74 kJ/mol) for the step and face retreats in this study were significantly higher than those reported in earlier studies. Recalculations performed using only data obtained under similar conditions in previous studies led to activation energies of 66-79 kJ/mol. These results and the earlier report showing that the form of the deep etch pits changed from angular to bow-shaped at about 60 °C may indicate that the activation energy of barite dissolution in water is higher at lower temperatures as compared with higher temperatures, thus changing the rate-limiting step. Whether the vertical and lateral retreat rates of the barite (001) plane differ in dependence of temperature remains unclear; however, the activation energies of the retreat of the (001) face in deep etch pits tended to be slightly higher than that of the lateral retreat rates of steps or other faces in deep etch pits.

A gem-grade apatite from Brazil of general composition (Ca,Na) 10 [(P,Si,S)O 4 ] 6 (F,Cl,OH) 2 has been studied using single-crystal X-ray and neutron diffraction together with synchrotron powder X-ray diffraction. Earlier electron microscopy studies had shown the nominally single-phase apatite contains an abundant fluorapatite (F-Ap) host, together with chloro-hydroxylapatites (Cl/OH-Ap) guest phases that encapsulate hydroxylellestadite (OH-El) nanocrystals. While the latter features appear as small (200-400 nm) chemically distinct regions by transmission electron microscopy, and can be identified as separate phases by synchrotron powder X-ray diffraction, these could not be detected by singlecrystal X-ray and neutron analysis. The observations using neutron, X-ray and electron probes are however consistent and complementary. After refinement in the space group P6 3 /m the tunnel anions F− are fixed at z = ¼ along <001>, while the anions Cl− and OH− are disordered, with the suggestion that O-H···O-H··· hydrogen-bonded chains form in localized regions, such that no net poling results. The major cations are located in the 4f A F O 6 metaprism (Ca+Na), 6h A T O 6 X tunnel site (Ca only), and 6h BO 4 tetrahedron (P+Si+S). The structural intricacy of this gem stone provides further evidence that apatite microstructures display a nano-phase separation that is generally unrecognized, with the implication that such complexity may impact upon the functionality of technological analogues.

The thermal expansion of anhydrous Mg 2 SiO 4 wadsleyite and forsterite was comprehensively studied over the temperature ranges 297-1163 and 297-1313 K, respectively, employing X-ray powder diffraction. Experiments were carried out with two separately synthesized samples of wadsleyite (numbered z626 and z627), for which room temperature unit-cell volumes differed by 0.05%, although the determined thermal expansions were identical within error. The high-temperature thermal expansions of wadsleyite and forsterite were parameterized on the basis of the first-order Grüneisen approximation using a Debye function for the internal energy. Values for hypothetical volume at T = 0 K, Debye temperature and Grüneisen parameter are 536.86(14) Å 3 , 980(55) K, 1.28(2) and 537.00(13) Å 3 , 887(50) K, 1.26(1) for z626 and z627, respectively, with the bulk modulus fixed to a literature determination of 161 GPa. For forsterite, the respective values are 288.80(2) Å 3 , 771(9) K, and 1.269(2) with a constrained bulk modulus of 125 GPa. These quantities are in good agreement with literature values obtained independently from sound velocity and heat capacity measurements, giving strong support to the applicability of Grüneisen theory in describing the thermal expansion of wadsleyite and forsterite. In addition, high-temperature structural variations were determined for wadsleyite from Rietveld analysis of the X-ray diffraction data. The pronounced anisotropy in thermal expansion of wadsleyite with a more expandable c-axis, similar to the compressional anisotropy, arises from specific features of the crystal structure consisting of the pseudolayers of MgO 6 octahedra parallel to the a-b plane with cross-linking Si 2 O 7 dimers along the c-axis. Although anisotropic compression and expansion originate from the same structural features, the details of structural changes with pressure differ from those caused by temperature. The longest Mg-O bonds, which are roughly parallel to the c-axis in all three octahedral sites of wadsleyite, dominate the compression, but these bonds do not exhibit the largest expansivities.

Numerical simulation of the evolution of stranded diffusion profiles in partially resorbed garnet crystals from the aureole of the Makhavinekh Lake Pluton, Labrador, yields quantitative determinations of rates of diffusion of Y, rare earth elements (REEs), and Cr at ~700-900 °C, 0.53 GPa. Diffusion coefficients for these trivalent cations are 0.5-1.5 log 10 units smaller than those for major divalent cations measured in the same crystals, but diffusivities for trivalent cations are all equal to one another to within ±0.25 log 10 unit. Integration of these new data with previously published results resolves some prior inconsistencies, and defines the dependence of diffusivities for Y, the REEs, and Cr on temperature, pressure, and oxygen fugacity, while accounting for minor effects of ionic radius and host-crystal composition. Nd, Sm, and Eu-elements that are strongly depleted in rims of relict garnet crystals due to preferential partitioning out of garnet during resorption-evolve small but distinct maxima from initially nearly flat profiles; this “uphill diffusion” results from cross-coupling with Y and the other REEs, which are strongly concentrated in relict garnet rims by resorption. The weak dependence of diffusivity on ionic radius and host-crystal composition, the near-equivalence of diffusivities of Y+REEs with that of Cr, and the strong positive cross-coupling among Y+REEs are all explained by a diffusion mechanism that links the mobility of VIII Y+REEs with that of VI Al; this is likely a consequence of the dominance of the menzerite-(Y) component as a means of incorporating Y+REEs in the garnet structure. However, the variety of possible substitution mechanisms that may enable Y+REE incorporation into garnet, and the degree to which each may be favored in different regimes of temperature, pressure, and oxygen fugacity, imply a potential for great complexity, as the net diffusional flux may result from the superposed effect of multiple diffusion mechanisms, each with different kinetics.

The EPR spectrum of Mn 2+ in microcrystalline calcite geomaterials (e.g., marbles, travertines) possesses exceptional diagnostic characteristics, allowing to relate samples to their origin (natural/synthetic, inorganic, organogenic, …) and to evaluate the details of impurities clustering. This information, beyond their mineralogical and geochemical interest, is of paramount importance for environmental, palaeoclimatic, and cultural heritage studies. Accessing the information hidden in the Mn 2+ EPR spectrum relies on disentangling spurious self-correlation among spin Hamiltonian parameters in the powder spectrum. In the present study, this goal is achieved through a systematic comparison of the temperature dependencies of four different microcrystalline calcite geomaterials. Accordingly, an assessment of the internal correlation structure of the spin Hamiltonian parameters is provided and the most sensitive discriminating parameters, which are able to mark samples, are identified. It has been found that the spin Hamiltonian parameters useful for discrimination purposes are those which are dependent on the ligand field interaction, whereas the Fermi contact interaction, as well as the spin-spin, spin-phonon, and spin-lattice interactions, are not able to “store” information related to formation processes, nor post-depositional events. This characteristic behavior is ascribed to the occurrence of mosaic structure and to the clustering among Mn 2+ and other impurity ions, which are able to induce a strong and variable ligand field interaction. In particular, the proposed method appears fully able to reveal the biogenic origin of microcrystalline calcites and to trace post-depositional events.

Red vesuvianite crystals occur with hausmanite and calcite in a hand specimen from Långban, Sweden. Backscattered electron imaging revealed that a very few of the vesuvianite crystals exhibit cores of higher average atomic number (vesuvianite I) than the bulk (vesuvianite II). Average electron microprobe compositions of vesuvianite I show 3.32 wt% MnO (1.43 Mn 2+ apfu), 3.13 wt% Bi 2 O 3 (0.41 Bi 3+ apfu), 0.82 wt% Cl (0.71 Cl apfu), and 0.56 wt% As 2 O 5 (0.15 As 5+ apfu). The average vesuvianite II composition shows considerably more Mn (4.14 wt% MnO, or 1.76 Mn 2+ apfu) but less Bi (0.62 wt% Bi 2 O 3 , or 0.08 Bi 3+ apfu) and Cl (0.13 wt% Cl, or 0.11 Cl apfu) and about the same amount of As (0.62 wt% As 2 O 5 , or 0.16 As 5+ apfu). The crystal structures of vesuvianite I (a = 15.595, c = 11.779 Å) and II (a = 15.585, c = 11.778 Å) were refined in space group P4/nnc to R 1 values of 0.0445 and 0.0167, respectively. The results show Bi at the new Bi3 position and Cl at the new Cl site; Bi3 can only be occupied when there is a vacancy at the nearby (~0.4 Å) X3 position, and Cl when there is a vacancy at the nearby (~0.4-0.5 Å) O10 position. Bond distances and site occupancies suggest that the Cl atom at the Cl site is bonded to the Bi atom at the Bi3 position when it is occupied. Arsenic occupies the T1 site and is coordinated by oxygen atoms at two O7B and two O11 positions. Manganese replaces Ca at the X1 and X2 position and dominates the Y1 position in vesuvianite I, whereas in vesuvianite II it replaces Ca at the X1 site and fills the Y1A position. In both structures Mn also occurs subordinate to Al + Mg at the Y3 position. Bond-valence analysis indicates that the Mn at the Y1 and Y1A positions is divalent rather than trivalent as required for manganvesuvianite.

A multi-analytical approach using electron microprobe analysis, X-ray structural refinement, and polarized single-crystal Fourier transform infrared spectroscopy was used to characterize short-range and long-range structures of a Pb 2+ -bearing, Mn 3+ -rich pargasite. Site populations, derived from the results of site-scattering refinement and stereochemical analysis, demonstrate that Pb 2+ is strongly ordered at the A-site in the monoclinic C2/m amphibole structure. This finding is in agreement with the observed ordering of Pb 2+ in the rare Pb 2+ -bearing P2/a amphibole joesmithite, but is in contrast to a Pb 2+ preference for the M4 site suggested by some studies on element partitioning between C2/m amphiboles and melts. Mn 3+ is strongly ordered at the M2 site in structure of the present amphibole. Contrasting results obtained for mean M2-O bond lengths and reported local Mn 3+ -O bond lengths as well as between bond-length distortion of the mean M2 polyhedron and the local Mn 3+ -centered M 2 O 6 octahedron in pargasite indicates that the structural relaxation of this polyhedron is complete or nearly so.

The texture and mineralogy of solid phases resulting from biogeochemical metal reduction of U(VI)-FeOOH slurries was investigated over a period of four years. Solid-phase reaction products were analyzed with EXAFS, TEM, and XRD following fermentative reduction of uranium-loaded ferric hydroxide precursors with 0.01 and 0.05 cation mole fraction (CMF) U by cultures of Thermoanaerobacter sp. strain TOR-39. Only minor changes could be distinguished between 3 and 51 months for most slurries. Magnetite, goethite, uraninite, and minor akaganéite were present after 3 months at both U-CMFs. Akaganéite was not detected by XRD after 3 months, but was still observed by TEM after 50 months. Increasing uranium in the starting slurries led to a greater proportion of oxidized iron in the solid-phase products. Euhedral goethite and subhedral to euhedral magnetite were observed at all times. Uraninite was observed in clusters of <10 nm particles without any particular relationship to the iron minerals. HRTEM imaging indicated that even the smallest uraninite particles were well crystallized, with textures that remained consistent throughout the duration of experiments. X-ray absorption spectra after 3 months indicated 100% and 96.4% U(IV) in 0.01 and 0.05 CMF U slurries, respectively. EXAFS spectra were consistent with uraninite at both uranium levels, plus additional non-uraninite U(IV) for 0.05 CMF U. One 0.05 CMF U culture slurry was found to have a lower pH and a more oxidized final iron mineral assemblage; in this case uraninite was not observed by XRD, but large (101 nm average diameter) rounded uraninite grains were observed by TEM. These grains were observed in chains or aggregates often connected by necks, in textures suggestive of biological influence. HRTEM demonstrated each grain was composed of poorly oriented, primary, 2-5 nm uraninite crystallites. Uraninite crystal growth occurred by nanoparticle aggregation, but ripening was not observed even though incubation temperatures were held at 65 °C for 20 days. Thus, previous studies of biogenic nanoparticulate uraninite short-term reactivity are likely to be representative of systems aged over a period of years.

In this work, we present an improved method for the semi-quantitative determination of sulfur species in S-bearing minerals by electron microprobe analysis. For calibration, we analyzed several sulfate and sulfide standard minerals such as baryte, celestine, chalcopyrite, and pyrite, and correlated the results with theoretical calculations retrieved from density functional theory (DFT). We applied this method to natural sodalite-group minerals from various localities. In addition, we applied the more common Raman spectroscopy to some samples and show that this method cannot be applied to sodalite-group minerals to determine their sulfur speciation. We show that even though sodalite-group minerals have a complex crystal structure and are sensitive to the electron beam, electron microprobe analysis is a reliable tool for the analysis of their sulfur speciation. The natural sodalite-group minerals show systematic variations in sulfur speciation. These variations can be correlated with the independently determined oxidation state of the parental magmas thus making S-bearing sodalitegroup minerals good redox proxies, although we show that the electron microprobe analysis of the sulfur speciation is matrix-dependent, and the sulfur speciation itself depends on crystal chemistry and structure, and not only on f O₂ .

The elastic compressional (P) and shear (S) wave velocities in NaCl were measured up to 12 GPa at 300 K, and up to 8 GPa at 473 and 673 K, by combining ultrasonic interferometry, in situ synchrotron X-ray diffraction, and X-ray radiographic techniques in a large-volume Kawai-type multi-anvil apparatus. The simultaneously measured sound velocity and density data at 300 K and high pressures up to 12 GPa were corrected to transform the adiabatic values to isothermal values and then used to estimate the 300 K equation of state (EOS) by a least-squares fit to the fourth-order Birch-Murnaghan finite strain equation, without pressure data. For a fixed isothermal bulk modulus K T0 of 23.7 GPa at 0 GPa and 300 K, we obtained the first and the second pressure derivatives of K T0 , K′ T0 = 5.14 ± 0.05 and K″ T0 = -0.392 ± 0.021 GPa −1 , respectively. A high-temperature and high-pressure EOS of NaCl was then developed using the Mie-Grüneisen relation and the Debye thermal model. To accomplish this, the simultaneously measured sound velocities and densities up to 8 GPa at both 473 and 673 K, as well as previously reported volume thermal expansion data of NaCl at 0 GPa were included in the fit. This resulted in a q parameter of 0.96, while holding the Grüneisen parameter and the Debye temperature, both at 0 GPa and 300 K, fixed at 1.56 and 279 K, respectively. Our EOS model accurately modeled not only the present measured K T data at pressures up to 12 GPa and temperatures between 300 and 673 K, but also the previously reported volume thermal expansion and the temperature dependence of K T , both at 0 GPa. The new temperature-pressure-volume EOS for NaCl, presented here, provides a pressure-independent primary pressure standard at high temperatures and high pressures.

Dolomite occurs in a wide range of rock compositions, from peridotites to mafic eclogites and metasediments, up to mantle depths of more than 200 km. At low-temperatures dolomite is ordered (R3̄), but transforms with increasing temperature into a disordered higher symmetry structure (R3̄c). To understand the thermodynamics of dolomite, we have investigated temperature, pressure, kinetics, and compositional dependence of the disordering process in Fe-bearing dolomites. To avoid quench effects, in situ X-ray powder diffraction experiments were performed at 300-1350 K and 2.6-4.2 GPa. The long-range order parameter s, quantifying the degree of ordering, has been determined using structural parameters from Rietveld refinement and the normalized peak area variation of superstructure Bragg peaks characterizing structural ordering/disordering. Time-series experiments show that disordering occurs in 20-30 min at 858 K and in a few minutes at temperatures ≥999 K. The order parameter decreases with increasing temperature and X Fe . Complete disorder is attained in dolomite at ~1240 K, 100-220 K lower than previously thought, and in an ankeritic-dolomite s.s. with an X Fe of 0.43 at temperatures as low as ~900 K. The temperature-composition dependence of the disorder process was fitted with a phenomenological approach intermediate between the Landau theory and the Bragg-Williams model and predicts complete disorder in pure ankerite to occur already at ~470 K. The relatively low-temperature experiments of this study also constrain the breakdown of dolomite to aragonite+Fe-bearing magnesite at 4.2 GPa to temperature lower than ~800 K favoring an almost straight Clapeyron-slope for this disputed reaction.

Devitrification of silicic volcanic rocks is a relatively common process, resulting in the production of microcrystalline silica and feldspar components. Here we investigate how the products of pervasive devitrification may be characterized using the combined techniques of X-ray powder diffraction, electron microprobe analysis, and X-ray fluorescence analysis to provide a new calibrated approach to calculating the crystallinity and mineral modes in both glassy vitrophyre and devitrified volcanics. Using the integrated areas of the X-ray diffraction peaks associated with both the crystalline and amorphous components, the relative proportions of groundmass crystallites and amorphous material from both glassy and devitrified material can be calculated. A detailed calibration indicates a linear relationship among the ratio of the integrated counts and bulk crystallinity. Mineral proportions are also calculated from X-ray fluorescence measurements of whole-rock and groundmass separates and are well correlated to crystallinities calculated from both X-ray diffraction and electron microprobe image analysis for vitrophyre samples. Devitrification products in a pervasively devitrified sample are tridymite, quartz, sanidine, and a Ca-rich aluminosilicate component. Mineral analysis and X-ray mapping by electron microprobe analysis indicates that the Ca-rich aluminosilicate component appears to be the dominant metastable or amorphous phase in the devitrified sample with proportions calculated from X-ray mapping (~32%) in reasonable agreement with the calculated proportion of amorphous material determined by means of X-ray diffraction (~38%). These results demonstrate the robustness of this combined X-ray diffraction and electron microprobe imagery technique for quantifying and characterizing crystallization in complex samples.

The assemblage strontium anorthite, quartz, and kyanite was reacted with H 2 O+CaCl 2 solutions at 500 °C and pressures between 460 and ~1300 MPa using a hydrothermal diamond-anvil cell. Information on the kinetics was obtained in situ based on time-resolved synchrotron-radiation X-ray fluorescence analyses of the Sr concentration in the fluid. The reaction products (anorthite or zoisite) were studied using transmission electron microscopy to obtain information on the reaction mechanism and mineral-fluid partitioning of strontium. The time required for equilibration was primarily controlled by the reaction mechanism, but not discernibly affected by pressure or chloride concentration. Nucleation and growth of zoisite at the expense of strontium anorthite was much faster than the Sr-Ca exchange reaction of strontium anorthite to anorthite, and resulted in chemically homogeneous crystals. The anorthite had developed a high nanoporosity during the reaction, which is indicative of coupled dissolution-precipitation. A zoisite-fluid exchange coefficient was obtained for the Sr-Ca fractionation at 500 °C and ~1300 MPa. At low bulk Sr/Ca, this value is in very good agreement with literature data, which are based on zoisite syntheses from oxide and hydroxide mixtures in chloridic fluids at 600 °C, 2 GPa and analyses after quench. This suggests that the Ca-Sr ratios in fluid and zoisite were not affected by back reactions during quenching. The constrained anorthite-fluid Sr partition coefficient for 500 °C, 460 MPa is, likewise, consistent with literature data, but determination of mineral-fluid partition and exchange coefficients can be hampered by quench phases in nanopores if coupled dissolution-precipitation acted as reaction mechanism.

Constraining mass transfer of the rare earth elements (REE) and high field strength elements (HFSE) from subducted oceanic crust and metasediments to the mantle wedge is fundamental toward interpreting processes that affect trace element mobility in subduction zone environments. Experimental studies of the partitioning of trace elements involving aqueous fluids at P-T conditions appropriate for slab-mantle wedge conditions are complicated by the difficulties in retrieving the fluid. Here we present the results from an application of an in situ technique that permits quantitative determination of element concentrations in aqueous fluid at geologically relevant supercritical conditions. We focus on pressures and temperatures appropriate for devolatilization-induced element transfer in subduction zone environments, and conditions obtained during regional metamorphism. In this study, we used a hydrothermal diamond-anvil cell (HDAC) and in situ synchrotron X-ray fluorescence (SXRF) to quantify the concentration of Y, an important trace element often used as a proxy for the heavy REE in geologic systems, in a xenotime-saturated 2 M HCl-aqueous fluid at 1.19 to 2.6 GPa and 300-500 °C. At these pressures and temperatures the solubility of yttrium ranges from 2400 to 2850 ppm. We find that the concentration of Y decreases with increasing fluid density. These new data, combined with published data generated from experiments done at lower pressure, in fluids of nearly identical composition and also NaCl-H 2 O fluids, constrain the effects of pressure and temperature on the ability of aqueous fluid containing Cl to scavenge and transport Y and, by analogy, the HREE. Although the physical properties of Y are similar to the high field strength elements, Y exhibits geochemical behavior that is analogous to the heavy rare earth elements (HREE).

Elastic and anelastic anomalies associated with spin-state transitions of Co 3+ in NdCoO 3 and GdCoO 3 perovskites with the Pnma structure have been investigated using resonant ultrasound spectroscopy (RUS) at high frequencies (0.1-1.5 MHz) and dynamic mechanical analysis (DMA) at low frequencies (0.1-50 Hz). Analysis of spontaneous strains related to octahedral tilting transitions and calculated using lattice parameter data from the literature show that the sequence of changing spin states with increasing temperature, low spin → low spin + high-spin → intermediate spin, is accompanied by significant variations in shear strain due to changes in ionic radius of the Co 3+ ions. This implies significant spin state/strain coupling, which, in turn, leads to renormalization of the shear modulus. In NdCoO 3 the shear strains are small, so the coupling is weak and the shear modulus increases with falling temperature in a manner that scales semi-quantitatively with an empirical spin order parameter. In GdCoO 3 the shear strains are much greater and softening of the shear modulus by up to ~35% in the low-spin + high-spin field arises through the influence of both the spin state/strain coupling and the order parameter susceptibility. Enhanced acoustic dissipation is also observed in the low-spin + high-spin field and is tentatively attributed to mobility of transformation twin walls, which are likely to have structures modified by changes in local Co 3+ spin-state populations. Co 3+ is isoelectronic with Fe 2+ and, although the spin-state transitions observed in cobaltate and silicate perovskites are not the same, the large shear strains associated with octahedral tilting in (Mg,Fe)SiO 3 at high pressures suggest that the effects of a high-spin → intermediate-spin transition of Fe 2+ would be closely analogous to those shown by GdCoO 3 . A spin-state transition of Fe 3+ in silicate perovskite would have a similar influence, due to the change in radius and its influence on octahedral tilting via the Goldschmidt tolerance factor.

Measurements of Fe 3+ /ΣFe in geological materials have been intractable because of lack of access to appropriate facilities, the time-consuming nature of most analyses, and the lack of precision and reproducibility in most techniques. Accurate use of bulk Mössbauer spectroscopy is limited by largely unconstrained recoilless fraction (f), which is used to convert spectral peak area ratios into valid estimates of species concentrations and is unique to different mineral groups and compositions. Use of petrographic-scale synchrotron micro-XANES has been handicapped by the lack of a consistent model to relate spectral features to Fe 3+ /ΣFe. This paper addresses these two deficiencies, focusing specifically on a set of garnet group minerals. Variable-temperature Mössbauer spectra of the Fe 2+ -bearing almandine and Fe 3+ -bearing andradite end-members are used to characterize f in garnets, allowing Fe 3+ /ΣFe to be measured accurately. Mössbauer spectra of 19 garnets with varying composition were acquired and fit, producing a set of garnet-specific standards for Fe 3+ analyses. High-resolution XANES data were then acquired from these and 15 additional previously studied samples to create a calibration suite representing a broad range of Fe 3+ and garnet composition. Several previously proposed techniques for using simple linear regression methods to predict Fe 3+ /ΣFe were evaluated, along with the multivariate analysis technique of partial least-squares regression (PLS). Results show that PLS analysis of the entire XANES spectral region yields the most accurate predictions of Fe 3+ in garnets with both robustness and generalizability. Together, these two techniques present reliable choices for bulk and microanalysis of garnet group minerals.

A synthetic single crystal of pure orthoenstatite (MgSiO 3 , space group Pbca) has been investigated at high pressure for structural determinations by in situ single-crystal X-ray diffraction using a diamond-anvil cell. Ten complete intensity data collections were performed up to 9.36 GPa. This study significantly improved the accuracy of structural parameters in comparison to a previous high-pressure structural study, allowing a more detailed examination of structural behavior of orthoenstatite at high pressures and a comparison to other more recent structural studies performed on orthopyroxenes with different compositions. The structural evolution determined in this work confirms the high-pressure evolution found previously for other orthopyroxenes and removes some ambiguities originating from the less accurate published data on the MgSiO 3 structure at high pressure. The structural compression is mostly governed by significant volume decrease of the Mg1 and Mg2 octahedra, affecting in turn the kink of the tetrahedral chains, especially the TB chain of larger SiO 4 tetrahedra. The Mg2 polyhedron undergoes the largest volume variation, 8.7%, due especially to the strong contraction of the longest bond distance (Mg2-O3B), whereas Mg1 polyhedral volume decreases by about 7.4%. The compressional behavior of the tetrahedral sites is quite different from previously published data. The TA and TB tetrahedral volumes decrease by about 2.8 and 1.8%, respectively, and no discontinuities can be observed in the pressure range investigated. Using the data on the pure orthoenstatite as reference, we can confirm the basic influences of element substitutions on the evolution of the crystal structure with pressure.

Experimental high-pressure investigations on benitoite in the diamond-anvil cell reveal a secondorder phase transition at a critical transition pressure P c = 4.24(3) GPa, as determined from synchrotron powder diffraction, single-crystal X-ray diffraction, and Raman spectroscopy. Diffraction experiments indicate a non-isomorphous transition from P6̄c2 to P31c space-group symmetry with a′ = a√3 and c′ = c relative to the P6̄2c subcell below P c . The high-pressure polymorph is characterized by a larger compressibility compared to the compressional behavior of benitoite below P c . Fitting second-order Birch-Murnaghan equations of state to the experimental data sets, the parameters obtained are V 0 = 372.34(4) Å 3 , K 0 = 117.9(7) GPa, with a0 = 6.6387(3) Å, K a = 108.1(7) GPa, and c 0 = 9.7554(4) Å, Kc = 143.3(1.1) GPa for the low-pressure form (P < P c ), and V 0 = 376.1(4) Å 3 , K 0 = 88.9(1.6) GPa, with a 0 = 11.516(4) Å, K a = 95.4(1.8) GPa, and c 0 = 9.826(4) Å, K c = 77.2(1.6) GPa for the high-pressure form (P > Pc). One of the most significant structural changes is related to the coordination of Ba atoms, changing from an irregular [6+6] coordination to a more regular ninefold. Simultaneously, the Si 3 O 9 rings are distorted due to no longer being constrained by mirror-plane symmetry, and the Si atoms occupy three independent sites. The higher compressibility along the c-axis direction is explained by the relative displacement of the Ba position to the Si 3 O 9 rings, which is coupled to the lateral displacement of the non-bridging O2-type atoms of the ring unit. A symmetry mode analysis revealed that the transition is induced by the onset of a primary order parameter transforming according to the K 6 irreducible representation of P6̄c2.

The structural behavior of Cr 2 O 3 was investigated up to ~70 GPa using single-crystal X-ray diffraction under a quasi-hydrostatic pressure (neon pressure medium) at room temperature. The crystal structure remains rhombohedral with the space group R3̄c (No. 167) and upon compression the oxygen atoms approach an ideal hexagonal close-packing arrangement. An isothermal bulk modulus of Cr 2 O 3 and its pressure derivative were found to be 245(4) GPa and 3.6(2), respectively, based on a third-order Birch-Murnaghan equation of state and V 0 = 288.73 Å 3 . An analysis of the crystal strains suggest that the non-hydrostatic stresses can be considered as negligible even at the highest pressure reached.

The heat capacity of three synthetic polycrystalline almandine garnets (ideal formula Fe 3 Al 2 Si 3 O 12 ) and one natural almandine-rich single crystal was measured. The samples were characterized by optical microscopy, electron microprobe analysis, X-ray powder diffraction, and Mössbauer spectroscopy. Measurements were performed in the temperature range 3 to 300 K using relaxation calorimetry and between 282 and 764 K using DSC methods. All garnets show a prominent λ-type heat-capacity anomaly at low temperatures resulting from a paramagnetic-antiferromagnetic phase transition. For two Fe 3+ -free or nearly Fe3 + -free synthetic almandines, the phase transition is sharp and occurs at 9.2 K. Almandine samples that have ~3% Fe 3+ show a λ-type peak that is less sharp and that occurs at 8.0 ± 0.2 K. The low-T CP data were adjusted slightly using the DSC results to improve the experimental accuracy. Integration of the low-T CP data yields calorimetric standard entropy, S°, values between 336.7 ± 0.8 and 337.8 ± 0.8 J/(mol·K). The smaller value is recommended as the best S° for end-member stoichiometric almandine, because it derives from the “best” Fe 3+ -free synthetic sample. The lattice (vibrational) heat capacity of almandine was calculated using the single-parameter phonon dispersion model of Komada and Westrum (1997), which allows the non-lattice heat capacity (C ex ) behavior to be modeled. An analysis shows the presence of an electronic heat-capacity contribution (C el , Schottky anomaly) superimposed on a larger magnetic heat-capacity effect (C mag ) around 17 K. The calculated lattice entropy at 298.15 K is S vib = 303.3 J/(mol·K) and it contributes about 90% to the total standard entropy at 298 K. The non-lattice entropy is S ex = 33.4 J/(mol·K) and consists of S mag = 32.1 J/(mol·K) and S el = 1.3 J/(mol·K) contributions. The C P behavior for almandine above 298 K is given by the polynomial [in J/(mol·K)]: C P = 649.06(±4) - 3837.57(±122)⋅T -0.5 - 1.44682(±0.06)·10 7 ·T -2 + 1.94834(±0.09)·10 9 ·T -3 which is calculated using the measured DSC data together with one published heat-content datum determined by transposed-drop calorimetry along with a new determination in this work that gives H 1181K - H 302K = 415.0 ± 3.2 kJ/mol. Using our S° value and the CP polynomial for almandine, we derived the enthalpy of formation, ΔH° f , from an analysis of experimental phase equilibrium results on the reactions almandine + 3rutile = 3ilmenite + sillimanite + 2quartz and 2ilmenite = 2Fe + 2rutile + O 2 . A ΔH° f = -5269.63 kJ/mol was obtained.

Witzkeite, ideally Na 4 K 4 Ca(NO 3 ) 2 (SO 4 ) 4 ·2H 2 O, is a new mineral found in the oxidation zone of the guano mining field at Punta de Lobos, Tarapacá region, Chile. It occurs as colorless, tabular crystals up to 140 μm in length, associated with dittmanite and nitratine. Witzkeite is colorless and transparent, with a white streak and a vitreous luster. It is brittle, with Mohs hardness 2 and distinct cleavage on {001}. Measured density is 2.40(2) g/cm 3 , calculated density is 2.403 g/cm 3 . Witzkeite is biaxial (-) with refractive indexes α = 1.470(5), β = 1.495(5), γ = 1.510(5), measured 2V = 50-70°. The empirical composition is (electron microprobe, mean of five analyses, H 2 O, CO 2 , and N 2 O 5 by gas chromatography; wt%): Na 2 O 12.83, K 2 O 22.64, CaO 7.57, FeO 0.44, SO 3 39.96, N 2 O 5 12.7, H 2 O 4.5, total 100.64; CO 2 was not detected. The chemical formula, calculated based on 24 O, is: Na 3.40 K 3.95 Ca 1.11 Fe 0.05 (NO 3 ) 1.93 (SO 4 ) 4.10 (H 4.10 O 1.81 ). Witzkeite is monoclinic, space group C2/c, with unit-cell parameters: a = 24.902(2), b = 5.3323(4), c = 17.246(1) Å, β = 94.281(7)°, V = 2283.6(3) Å 3 (Z = 4). The crystal structure was solved using single-crystal X-ray diffraction data and refined to R 1 (F) = 0.043. Witzkeite belongs to a new structure type and is noteworthy for the very rare simultaneous presence of sulfate and nitrate groups. The eight strongest X-ray powder-diffraction lines [d in Å (I in %) (h k l)] are: 12.38 (100) (2 0 0), 4.13 (19) (6 0 0), 3.10 (24) (8 0 0), 2.99 (7) (8̄ 0 2), 2.85 (6) (8 0 2), 2.69 (9) (7̅ 1 3), 2.48 (12) (10 0 0), and 2.07 (54) (12 0 0). The IR spectrum of witzkeite was collected in the range 390-4000 cm -1 . The spectrum shows the typical bands of SO 4 2- ions (1192, 1154, 1116, 1101, 1084, 993, 634, and 617 cm -1 ) and of NO 3 - ions (1385, 1354, 830, 716, and 2775 cm -1 ). Moreover, a complex pattern of bands related to H 2 O is visible (bands at 3565, 3419, 3260, 2405, 2110, 1638, and 499 cm -1 ). The IR spectrum is discussed in detail.

A new mineral krasheninnikovite, ideally KNa 2 CaMg(SO 4 ) 3 F, is found in the sublimates of an active fumarole at the Second scoria cone of the Northern Breakthrough of the Great Tolbachik Fissure Eruption, Tolbachik volcano, Kamchatka, Russia. It is associated with tenorite, thenardite, hematite, euchlorine, blödite, vergasovaite, and fluorophlogopite. Krasheninnikovite forms long-prismatic to acicular crystals up to 3 mm long and up to 20 μm thick. The crystals are combined in sheaf-like, radiating or open-work matted aggregates forming nests up to several cm 3 or crusts. Krasheninnikovite is transparent, colorless in individuals and white in aggregates. The luster is vitreous. The mineral is brittle; the thinnest needles are flexible and elastic. The Mohs hardness is 2½-3. Cleavage was not observed. D meas is 2.68(1), D calc is 2.67 g/cm 3 . Krasheninnikovite is optically uniaxial (-), ω = 1.500(2), ε = 1.492(2). The IR spectrum is unique. The chemical composition (wt%, electron microprobe data) is: Na 2 O 15.48, K 2 O 6.92, CaO 11.51, MgO 9.25, MnO 0.15, FeO 0.04, Al 2 O 3 0.23, SO 3 53.51, F 3.22, Cl 0.16, -O=(F,Cl) 2 -1.39, total 99.08. The empirical formula, calculated on the basis of 13 (O+F+Cl) apfu, is: K 0.67 Na 2.27 Ca 0.93 Mn 0.01 Mg 1.04 Al 0.02 (SO 4 ) 3.04 F 0.76 Cl 0.02 O 0.06 . Krasheninnikovite is hexagonal, space group P6 3 /mcm, a = 16.6682(2), c = 6.9007(1) Å, V = 1660.36(4) Å 3 , Z = 6. The strongest reflections of the X-ray powder pattern [d, Å I(hkl)] are: 4.286 22(121); 3.613 24(040); 3.571 17(221); 3.467 42(131); 3.454 43(002); 3.153 100(140), 3.116 22(022), 2.660 39(222), 2.085 17(440). The crystal structure was solved on a single crystal and refined on a powder sample by the Rietveld method, R wp = 0.0485. The krasheninnikovite structure is unique. It is based upon a heteropolyhedral pseudoframework consisting of CaO 6 octahedra, MgO 5 F octahedra, and SO 4 tetrahedra; K and Na cations are located in cavities. Krasheninnikovite is named in honor of the Russian geographer, ethnographer, and naturalist S.P. Krasheninnikov (1711-1755), one of the first scientists who researched Kamchatka. The type specimen is deposited in the Fersman Mineralogical Museum of the Russian Academy of Sciences, Moscow.

The crystal structure of pseudojohannite from White Canyon, Utah, was solved by charge-flipping from single-crystal X-ray diffraction data and refined to an R obs = 0.0347, based on 2664 observed reflections. Pseudojohannite from White Canyon is triclinic, P1̄, with a = 8.6744(4), b = 8.8692(4), c = 10.0090(5) Å, α = 72.105(4)°, β = 70.544(4)°, γ = 76.035(4)°, and V = 682.61(5) Å 3 , with Z = 1 and chemical formula Cu 3 (OH) 2 [(UO 2 ) 4 O 4 (SO 4 ) 2 ](H 2 O) 12 . The crystal structure of pseudojohannite is built up from sheets of zippeite topology that do not contain any OH groups; these sheets are identical to those found in zippeites containing Mg 2+ , Co 2+ , and Zn 2+ . The two Cu 2+ sites in pseudojohannite are [5]- and [6]-coordinated by H 2 O molecules and OH groups. The crystal structure of the pseudojohannite holotype specimen from Jáchymov was refined using Rietveld refinement of high-resolution powder diffraction data. Results indicate that the crystal structures of pseudojohannite from White Canyon and Jáchymov are identical.

The surface oxidation of pyrite can create a (hydr)oxide layer. This configuration constitutes a natural two-semiconductor (tandem) photoelectrochemical cell. Here, we show that an illuminated hematite-pyrite cell produces photocurrent, H 2 , and O 2 by water splitting. Photocurrent is also observed with illumination of hematite alone. The observed current densities are in the same order of magnitude as estimates of banded iron formation deposition rates, and are 400 to 1000 times higher than needed to oxidize, over geologic time, all of the surface water thought to have existed on Mars. Mineral-based water splitting constitutes a potential source of O 2 prior to the evolution of oxygenic photosynthesis on Earth. Semiconducting minerals deserve study as photochemical sources of oxidizing power in low-oxygen environments.

In situ synchrotron X-ray powder diffraction measurements using a Paris-Edinburgh pressure cell were performed to investigate the nature of the high-pressure breakdown reaction of magnetite (Fe 3 O 4 ). Refinement of diffraction patterns reveals that magnetite breaks down via a disproportionation reaction to Fe 4 O 5 and hematite (Fe 2 O 3 ) rather than undergoing an isochemical phase transition. This result, combined with literature data indicates (1) that this reaction occurs at ~9.5-11 GPa and 973-1673 K, and (2) these two phases should recombine at yet higher pressures to produce an h-Fe 3 O 4 phase.

Bartelkeite from Tsumeb, Namibia, was originally described by Keller et al. (1981) with the chemical formula PbFeGe 3 O 8 . By means of electron microprobe analysis, single-crystal X-ray diffraction, and Raman spectroscopy, we examined this mineral from the type locality. Our results show that bartelkeite is monoclinic with space group P2 1 /m, unit-cell parameters a = 5.8279(2), b = 13.6150(4), c = 6.3097(2) Å, β = 127.314(2)°, and a revised ideal chemical formula PbFeGe VI Ge 2 IV O 7 (OH) 2 ·H 2 O (Z = 2). Most remarkably, bartelkeite is isostructural with the high-pressure P2 1 /m phase of lawsonite, CaAl 2 Si 2 O 7 (OH)·H 2 O, which is only stable above 8.6 GPa and a potential host for H 2 O in subducting slabs. Its structure consists of single chains of edge-sharing FeO 6 and Ge 1 O 6 octahedra parallel to the c-axis, cross-linked by Ge 22 O 7 tetrahedral dimers. The average <Ge-O> bond lengths for the GeO 6 and GeO 4 polyhedra are 1.889 and 1.744 Å, respectively. The Pb atoms and H 2 O groups occupy large cavities within the framework. The hydrogen bonding scheme in bartelkeite is similar to that in lawsonite. Bartelkeite represents the first known mineral containing both 4- and 6-coordinated Ge atoms and may serve as an excellent analog for further exploration of the temperature-pressure-composition space of lawsonite.