Volume 95, Issue 10

Chemical zoning according to the substitution Al + (OH,F) = Ti + O is a common phenomenon in natural titanite and can be used to reconstruct its growth conditions. We synthesized titanite, which has a zoning pattern similar to that found in natural titanite that grew at the expense of rutile, visible in element distribution maps as a patchy irregular pattern. Run conditions in the system CaO-Al 2 O 3 - TiO 2 -SiO 2 -NaCl-H 2 O(-CaF 2 ) were 600 °C, 0.4 GPa, 1 to 107 days duration. Natural rutile crystals were placed in a perforated Pt-tube within an outer Au-capsule containing wollastonite, γ-Al 2 O 3 , optionally powdered CaF 2 , and a NaCl brine, to simulate conditions, where rutile is transformed into titanite by a hydrous Ca-Si-Al(-F) fluid, driven by a chemical potential gradient between inner and outer capsule. Spontaneous nucleation of titanite is restricted to a small number of crystals on the rutile surface in the inner capsule. Growth proceeds from sparse isolated titanite crystals after 1 day to assemblages of several 100 μm large crystals in a reaction rim in the long-time runs. Titanite is strongly zoned in Al-Ti and shows up to ~0.5 Al per formula unit (pfu) in experiments containing F; without F, substitution is limited to ~0.25 Al pfu. The range of the Ti-Al exchange is already large in short run times and the same heterogeneity is observed in long-time runs. The Al-Ti distribution (with and without F) in compositional domains (up to several tens of micrometers) is patchy and irregular. No relation of the zoning with time (core-rim) or preferred growth directions was observed. In the case of fluid-mediated growth, mineral growth zoning represents a disequilibrium at supersaturated conditions. The concentration of Ti in the fluid is likely the limiting factor for titanite growth considering the low solubility compared to that of Al, Ca, and Si. Titanium is not transported in significant amounts into the outer capsule, indicating low mobility, in contrast to Al, Ca, and Si, which are transported into the inner capsule. Assuming buffered Al concentration from dissolution of the Al-source, we speculate that the Ti/Al of the fluid at the precipitation site was controlled by the local Ti concentration, which varies over the distance from the rutile surface to the growing titanite crystal and produces the irregular patchy zoning.

Most natural rocks or engineered materials display a multi-scale heterogeneity ranging from the nanometer to the centimeter. Their spatial textural heterogeneity can be approached from chemical element maps acquired using various techniques (SEM, EPMA, SXAM, synchrotron μ-XRF, TEM), depending on the chosen magnification. Chemical map processing that yields quantitative petrographic information is improved here according to newly developed mineral thresholding methods that accommodate mixtures and solid solutions. The complex case of an MX80 compacted bentonite is used as a test case. The 14 major chemical elements of this sample were mapped using an electron probe microanalyzer, and chemical map processing yielded a quantitative map of the 18 mineral species of bentonite with a spatial resolution of a few micrometers. The textural heterogeneity of the solid part of the sample is thus visualized and quantified on an area ranging between 0.1-1 cm 2 . The method also provides a complete modal analysis of the sample. The methodology is expected to have broad applications in Earth and materials sciences.

Visualizing and quantifying the spatial heterogeneity of rock textures (i.e., mineral and porosity distributions) is of great interest in petrology or for understanding petrophysical properties. Spatial heterogeneities are not accurately revealed by usual techniques based on microscopy or bulk physical measurements. Detailed mineral mapping is already available from processing of chemical element maps acquired using an electron probe microanalyzer (Part I). The present paper is devoted to developing a new, coupled method for obtaining porosity maps from the same initial data. According to the difference between measured and theoretical sums of oxide weight percentages, a mean porosity is semi-quantitatively estimated for each pixel of the map (i.e., not fully absolute or accurate). All pores, including nanometer-size ones, are taken into account, whereas a sample area of several square millimeters is analyzed (spatial resolution of a few micrometers). The textural heterogeneities are thus visualized from the complementary maps of solids and voids. By superimposing these two maps, both the mean porosity and a porosity histogram associated with each rock-forming mineral are obtained. Such porosity measurements integrate the pore amounts within mono-crystals larger than the X-ray emission volume or between nanometer-size crystals of a matrix. When porosity changes are associated with a given mineral (various crystal arrangements, dissolution, etc.), pluri-modal distributions appear on porosity histograms. Thresholding each histogram mode then allows these processes to be localized. We used the MX80 bentonite to test this methodology, which represents a useful tool to study the local deformations and alteration of each rock-forming mineral, as well as to model transport properties.

Reaction layer growth between two chemically different solids may be controlled by polycrystalline diffusion kinetics in the growing phase. The kinetics depend on the interplay between volume and grain boundary diffusion. Using spinel formation between single crystals of periclase and sapphire as an example, we quantify the effects of an applied mechanical stress on the bulk-transport properties of the reaction layer. The rate of spinel growth increases fourfold when stress normal to the reaction interface increases from 3 to 30 MPa due to stress-induced changes in grain boundary structure. At low applied stress, low-index (i.e., Σ3) “coincidence site lattice” grain boundaries with slow diffusion coefficients dominate, related to epitactic growth of spinel on sapphire. Increasing stress triggers epitactic growth of spinel on periclase, and causes sapphire-grown spinel grains to rotate out of epitaxy, and grain boundaries with fast diffusion coefficients dominate. This effect outweighs the hitherto emphasized influence of grain size on the bulk transport properties of polycrystals.

Rhomboclase, (H 5 O 2 )Fe(SO 4 ) 2 ·2H 2 O, transforms to a solid crystalline phase, (H 3 O)Fe(SO 4 ) 2 , upon dehydration. The structure of (H 3 O)Fe(SO 4 ) 2 is found to be the same as a recently reported structure determined from single-crystal diffraction by Peterson et al. (2009), who synthesized the same compound using a hydrothermal method. The phase boundary between rhomboclase and (H 3 O)Fe(SO 4 ) 2 as a function of temperature (T) and relative humidity (RH) was determined by environment-controlled in situ X-ray diffraction (XRD) method. The stability of (H 3 O)Fe(SO 4 ) 2 against rhomboclase was further evaluated under a simulated martian condition (constant 50% RH, -20 °C, 6 mbar CO 2 ). Both phases remained after 14 days with no observable transition. This result suggests that hydrate ferric sulfate minerals might not respond to diurnal RH fluctuation under the extremely slowed kinetics expected on the martian surface.

The crystal chemistry of six natural datolites from different localities was investigated by electron microprobe analysis in the wavelength dispersive mode and single-crystal X-ray diffraction. The chemical analyses show no significant site substitution. The single-crystal structure refinements confirm the structural model of datolite previously reported (with a ~ 4.83, b ~ 7.61, c ~ 9.63 Å, and β ~ 90.15°, space group P2 1 /c). Intra-polyhedral bond distances and angles show common features in all samples at room T: (1) the Si-tetrahedron is strongly deformed, with Si-O distances ranging between ~1.57 and ~1.66 Å and O-Si-O angles ranging between ~105.4 and ~115.3°; (2) the B-tetrahedron is almost regular; (3) the Ca-polyhedron is significantly distorted, with bond distances ranging between ~2.28 and ~2.67 Å; (4) only one independent H-site occurs and its refined position suggests a bifurcated hydrogen bonding scheme with O5 as donor and O4 and O2 as acceptors [with O5-H ~ 0.8 Å and (1) O5···O4 ~ 2.99 Å, H···O4 ~ 2.33 Å, and O5-H···O4 ~ 140°, and (2) O5···O2 ~ 2.96 Å, H···O2 ~ 2.36 Å, and O5-H···O2 ~ 131°]. Low-temperature diffraction measurements between 300 and 100 K show that the thermal expansion of datolite is mainly governed by the axial response along [100] and [010], whereas the c-axis length is almost unchanged in this temperature interval. The volume thermal expansion coefficient (α V = V -1 ∂V/∂T) between 100 and 280 K is αV = 1.5(2)·10 -5 K -1 . The higher thermal expansion of the a-axis is due to the layered nature of the structure of datolite: the Ca-O bond distances are the most compressible and expandable, and govern the contraction, upon cooling, along the direction perpendicular to the polyhedral layers. The tetrahedral layer is significantly more rigid and no changes of the tetrahedral tilts are observed from 300 to 100 K.

Density and seismic velocities of the chromitite bodies from the Oman ophiolite were calculated and compared with those of oceanic peridotites. The Oman chromitite is composed of olivine and chromian spinel, and the olivine/spinel volume ratio varies from 20 to 90 vol%. Chromian spinel in the chromitite is enriched in magnesiochromite component, and Fo content of olivine ranges from 94 to 96. Compressibility of a natural chromian spinel [(Mg 0.77 Fe 2+ 0.23 )(Cr 0.46 Al 0.50 Fe 3+ 0.04 )O 4 ] was measured with in situ synchrotron X-ray diffraction experiments, using a diamond anvil cell up to 10 GPa at 300 K. The third-order Birch-Murnaghan equation of state yields the isothermal bulk modulus of KT = 192(7) GPa, its pressure derivative of K' T = 4(1), and the zero-pressure volume of V 0 = 560.6(2) Å 3 . Based on calculations using present and previous studies, the adiabatic bulk modulus of magnesiochromite component was estimated to be 189 GPa. Using petrological and the mineral physics data, density and seismic velocities (V φ , V P , V S ) were calculated. The seismic velocities are higher than those of peridotites in the oceanic upper mantle and decrease with decrease of the spinel/olivine ratio. The contrast between chromitite and oceanic peridotite are 5.0-8.1% for V φ , 2.7-4.9% for VS, and 4.1-6.7% for V P in the Moho transition zone chromitite, and 2.6-6.5% for V φ , 1.4-3.3% for V S , and 2.2-5.2% for VP in the discordant mantle chromitite. This contrast is large enough to be detected by seismological observations and becomes a key to knowing the distributions of chromitite in the oceanic upper mantle.

Sizes of cavities and their packing schemes in selected zeolites and clathrasils were studied by means of least squares fitting of circumscribed spheres to them. Resulting packing of spheres of different diameters was analyzed by the coordinates of their centers, their volumes and sphericity, and interpreted by comparison with analogous packings of atoms in inorganic compounds and alloys. The topology is described qualitatively as “negative” structures formed by the cavities. Melanophlogite and dodecasils 3C and 1H are all clathrasils with isolated cavities. They all have pentagonal dodecahedral [5 12 ] cages, associated with other cavity types. The packing of cavities in melanophlogite is analogous to the packing of atoms in the structure of Cr 3 Si, where the Cr atoms form icosahedra around the Si sites. Dodecasil 3C has a cubic arrangement of [5 12 ] cavities, which is described as ABC stacking of kagome nets and its cavity packing is an analog to the packing scheme of atoms in the cubic Laves phase MgCu 2 . Dodecasil 1H has an arrangement of [5 12 ] cavities in an AA stacking of kagome nets and is analogous to the alloy structure type CaZn 5 . Edingtonite and natrolite are built from two types of cavities, tetragonal scalenohedra and [8 6 ] cavities. The [8 6 ] cavities have different shape and packing in the two structures. In paulingite, the largest cavity, a large rhombicuboctahedron [4 12 6 8 8 6 ] with a diameter of 11.02 Å is in a body centered configuration. It is interspaced by six other types of cavities with diameters from 5.10 to 8.20 Å, arranged in a complex pattern. In this structure, there are three non-intersecting channel systems. Their cavity arrangements mimic sodalite and a modified ReO 3 structure. The cubic structure of sodalite has only one type of cavity, a truncated octahedron [4 6 6 8 ] with a diameter of 6.41 Å, arranged in a body-centered packing analogous to α-Fe. Chabasite has a rhombohedral packing of elongate cavities [4 12 6 2 8 6 ], whereas Linde Type A has [4 6 6 8 ] and [4 12 6 8 8 6 ] cavities in a CsCl-type arrangement. The cavity representation used here gives a clear picture of cavities and interconnections even for complicated cases.

The behavior of structural OH in clinopyroxene (augite and omphacite) during successive heating has been investigated by in situ micro-FTIR measurements under temperatures ranging from room temperature to 500 °C at 100 °C increment. The first group of OH band (3620-3640 cm -1 ) exhibits a systematic decrease of peak position upon successive heating, while the other two groups (3520-3535 and 3450-3465 cm -1 ) show only little change. Both augite and omphacite display a decrease of integral absorbance of OH fundamental stretching vibration upon successive heating. The IR spectra of OH band are reversible when the temperature decreases from 500 °C to room temperature, suggesting that changes in IR indicate changes in molecular state of OH. Based on: (1) the decreases of integral absorbance of OH bands with increasing temperature, and (2) comparison with Paterson’s relationship (Paterson 1982), we suggest that OH absorption coefficients are temperature dependent, so it is necessary to apply different absorption coefficients when determining OH content from Beer-Lambert law at different temperatures and sample temperatures should be reported in quantitative IR studies

Temperature dependencies and thermal dynamical behavior of vibrational phonons related to different bonds/motions (e.g., K-O, Si-O, and Al-O) and O-H stretching, bending, and combinations bands, as well as those of NH 4 ions, of muscovite have been investigated in detail at high temperatures using in situ infrared spectroscopy. Phonon modes of muscovite show a complex response to heating and dehydroxylation. The mode Grüneisen parameters of muscovite are largest for low-frequency bands and some bands have negative values. Approaching dehydroxylation, bands associated with K-O stretching, Al-OH bending, and Al-O vibrations exhibit significant variations, indicating modifications of local configurations related to these ions. The O-H stretching feature near 3627 cm -1 shows a change in temperature dependence of the peak position near or during dehydroxylation, and this is attributed to a dramatic loss of OH and changes in local environments. The infrared spectroscopic analyses carried out in situ or on quenched samples do not record the characteristic bands of dehydroxylation-induced H 2 O and these findings suggest that H 2 O is unlikely to be the principal diffusing species. The results suggest that de-ammoniation or loss of NH 4 in muscovite takes place at temperatures near dehydroxylation.

The kinetics of the Fe-oxidation/deprotonation process in a natural Fe-rich phlogopite from Mt. Vulture (Potenza, Italy) was studied under isothermal conditions by in situ high-temperature singlecrystal X-ray diffraction. Isothermal annealing experiments were performed at five temperatures in the range 640-750 °C on five crystals with similar chemical composition and lattice parameters. The Fe-oxidation/deprotonation process at high temperature occurs with a reduction of unit-cell parameters and cell volume. The changes in unit-cell parameters measured at high temperature and during cooling show that the same degree of Fe-oxidation/deprotonation process was achieved at all temperatures. Changes in unit-cell parameters with temperature and time show that the kinetics of Fe-oxidation/ deprotonation in phlogopite follows an exponential law, and the temperature dependence follows the Arrhenius relation. A kinetic analysis was performed and good agreement was obtained with the onedimensional diffusion model. An apparent activation energy of 195(4) kJ/mol was determined.

The crystal chemistry of londonite, a rare Cs-bearing mineral [ideal chemical formula: (Cs,K) Al 4 Be 4 (B,Be) 12 O 28 ; a = 7.3098(2) Å at 300 K, space group P4̅3m] from the granitic pegmatites of the Antsongombato Gem Mine, Betafo district, Madagascar, has been reinvestigated by means of wavelength dispersive X-ray spectroscopy, laser ablation-inductively coupled plasma-mass spectroscopy, inductively coupled plasma-atomic emission spectroscopy, X-ray powder diffraction, and single-crystal neutron diffraction at 300 and 20 K. Single-crystal anisotropic structural refinement at 300 K gave a final agreement index R 1 = 0.0479 for 32 refined parameters and 416 unique reflections with F o > 4σ(F o ). The analysis of the difference-Fourier maps of the nuclear density discounts the presence of hydroxyl groups, as wrongly reported in some of the previous studies. The structural refinements and the chemical analyses suggested that: the tetrahedral 4e site (at x ~ 0.258) is mainly occupied by Be, but a low amount of B (~3%) likely occurs; the tetrahedral 12h site (at x ~ 0.248) is mainly occupied by B, but a significant fraction of Be (~12%) is present; the octahedral 4e site (at x ~ 0.361) is fully occupied by Al; and the 1a site (at x = 0, with CN = 12) is mainly occupied by Cs, with Rb and K. The significantly high amount of B 2 O 3 (~50 wt%) and Cs+Rb(CsO 2 +RbO 2 ≥ 8 wt%) makes the synthetic counterpart of londonite of interest as a potential neutron absorber or a potential host for nuclear waste.

Iron oxide (FeO) is an important component in the mineralogy of Earth’s lower mantle and possibly its core, so its phase diagram is essential to models of the planet’s interior. The melting curve of wüstite, Fe 0.94 O, was determined up to 77 GPa and 3100 K in a laser-heated diamond anvil cell. Melting transition temperatures were identified from discontinuities in the emissivity vs. temperature relationship within the laser-heated spot. The melting curve exhibits no obvious kinks that could be related to a subsolidus transition in wüstite, but there is evidence for a two-phase loop at pressures below 30 GPa. Comparison of these results to previous studies on Fe, Fe-O, and Fe-S confirms that the melting point depression in the Fe-O system remains significantly less, by a factor of 2 or more, than that in the Fe-S system up to pressures exceeding 80 GPa.

An assemblage of primary, extremely As- and Sb-rich, Nb-Ta minerals from the Szklary pegmatite includes columbite-(Fe), columbite-(Mn), tantalite-(Mn), stibiocolumbite, stibiotantalite, an as yet unnamed (As,Sb,U)-rich (Ta,Ti)-oxide, Mn 3 UAs 2 Sb 2 Ta 2 Ti 2 O 20 , and holtite. Anomalous trends of Mn-Fe and Ta-Nb fractionation in the columbite group and crystallization sequences in the primary assemblage can be explained by the contamination of the pegmatite-forming melt by ultramafic and mafic wall-rocks, the competition among these minerals for Ta and Sb and with biotite and tourmaline for Mg, Fe, and Ti, and local variations in melt composition. A hot magmatic fluid, exsolved from the parental melt, reacted with the primary Nb-Ta oxides, inducing two different patterns of alteration. The columbite-group minerals were altered to fersmite, pyrochlore, and bismutopyrochlore locally grading to plumbopyrochlore, whereas stibiocolumbite, stibiotantalite, and the (As,Sb,U)-rich (Ta,Ti)-oxide altered to stibiomicrolite, uranmicrolite grading to betafite, and then to bismutomicrolite or Bi-dominant betafite. In all of the secondary pyrochlore-group minerals, Ta-Nb fractionation is comparable to, or only slightly greater, than that in the primary Nb-Ta oxides, indicating a modest differentiation of the residual melt coexisting with the fluid.

The evolution of the interlayer space of different hydrated, Wyoming-type montmorillonite under the influence of temperature was investigated by means of Monte Carlo molecular modeling simulation methods. Calculations were performed on montmorillonite with monovalent and divalent counterions including Li + , Na + , K + , Rb + , Cs + , Ca 2+ , Ni 2+ , Zn 2+ , and Pb 2+ . For each of these clays we have conducted a series of 16 simulations in which the temperature is increased systematically from 300 to 400 K, then decreased to 260 K, and finally increased to 300 K. SPC/E water model is used to describe the water and hydroxyl behavior. From these simulations we found that the variation in the interlayer spacing of 3-layer hydrated montmorillonite is much greater than that of 2- and 1-layer hydrated montmorillonite. Hysteresis phenomena have been found in the interlayer spacing-temperature curves, especially at high temperature. The influence of temperature on the swelling behaviors of montmorillonite is different in that it depends on the counterions contained in the clay.

Here we present the first study of pyrite textures observed in particulate matter from an active shallow submarine vent site, which is characterized by large spatial and temporal variations in the physico-chemical conditions caused by a combination of geochemical processes and microbial activity. Morphologic characterization of pyrite crystals from suspended particulate matter in the discharged fluid of a shallow submarine vent system near Punta Mita, Nayarit, Mexico, showed diverse crystal morphologies and aggregates; well-defined framboids are only observed within the deposits of finelayered calcite (calcareous tuff) formed around the vents, whereas particulate matter contains diverse pyrite crystal forms as globular, cubic, octahedral, and pyritohedral but no framboids. Available sulfur isotopic data indicate that pyrite is formed as a consequence of microbial sulfate reduction in a hydrothermal reducing environment. The results of our study in a natural system provide evidence of the effect of variations in key parameters, such as redox state, on the pyrite morphologies and framboid size distribution, and support the conclusions of numerous studies that have attempted to explain pyrite crystallization processes that generate different morphologies as a result of large variations in the physico-chemical conditions.

The influence of СО 2 and Н 2 О on the morphology of diamond dissolution in carbonate melts was studied experimentally at pressure 5.7-7.0 GPa and temperature 1400-1750 °С, using a BARS multi-anvil apparatus. It has been established that diamond dissolution in fluid-free carbonate melts starts with the development of positive trigons on the {111} diamond faces, followed by truncation of crystal edges by trigon-trioctahedral surfaces, and finally by the transformation of diamond into spherical dodecahedroid-like morphology. Diamond dissolution in СО 2 -bearing carbonate melts also begins with the formation of positive trigons on the {111} faces and development trigon-trioctahedron surfaces on the edges. Dissolution form changes from trigon-trioctahedron to dodecahedroid with increasing loss of initial weight. Addition of more than 8 wt% of Н 2 О into the carbonate medium changes the orientation of the trigons and the secondary morphology of diamond. At the beginning of the process, negative trigons and ditrigonal (shield-shaped) dissolution layers developed on the {111} faces. Dissolution form of diamond in water-bearing melts is tetrahexahedroid, which is most similar to rounded natural diamonds. The results obtained allow us to regard the morphology of trigons and dissolution forms as an indicator of the composition of the diamond dissolution medium. The experiments suggest that the morphology of diamond during dissolution is controlled by the presence of water in a system. Our data show that the СО 2 /(СО 2 +Н 2 О) ratio by weight value was <0.81 during natural diamond dissolution.

This study presents far infrared spectra in the range 650-70 cm -1 of 18 common and rare carbonate minerals. Mineral samples of known provenance are selected and physically characterized to determine the purity of the crystalline phase and their composition. The fine ground mineral powders are embedded in polyethylene pellets, and their transmittance spectra are collected with a Fourier spectrometer. The far infrared spectra of different carbonate minerals from the same structural group have well-defined similarities. Observed shifts generally manifest the mass effect of the constituent metal cations. Remarkable spectral differences occur for different carbonates in the far IR region and may serve as fingerprints for mineral identification and are more useful identifiers of carbonate species than those in any other infrared range. For some of the minerals studied here, like kutnohorite, artinite, gaylussite, and trona, no far infrared spectra to that extend (up to 70 cm -1 ) have been found in literature.

Diamond-trap experiments in combination with laser ablation ICP-MS analyses of frozen samples have been used to directly determine the composition of fluids and melts in equilibrium with crust and mantle mineral assemblages. Here we: (1) describe an improved freezing cell that utilizes electronic cooling elements to facilitate routine laser ablation measurement on diamond traps and (2) demonstrate that the major element stoichiometry of subsolidus aqueous fluids in equilibrium with crystalline mineral assemblages can be measured using the diamond-trap method with comparable precision and accuracy to single-crystal weight-loss experiments

The new mineral françoisite-(Ce), (Ce,Nd,Ca)[(UO 2 ) 3 O(OH)(PO 4 ) 2 ]·6H 2 O is the Ce-analog of françoisite-(Nd). It has been discovered simultaneously at the La Creusaz uranium deposit near Les Marécottes in Valais, Switzerland, and at the Number 2 uranium Workings, Radium Ridge near Mt. Painter, Arkaroola area, Northern Flinders Ranges in South Australia. Françoisite-(Ce) is a uranylbearing supergene mineral that results from the alteration under oxidative conditions of REE- and U 4+ - bearing hypogene minerals: allanite-(Ce), monazite-(Ce), ±uraninite at Les Marécottes; monazite-(Ce), ishikawaite-samarskite, and an unknown primary U-mineral at Radium Ridge. The REE composition of françoisite-(Ce) results from a short aqueous transport of REE leached out of primary minerals [most likely monazite-(Ce) at Radium Ridge and allanite-(Ce) at La Creusaz], with fractionation among REE resulting mainly from aqueous transport, with only limited Ce loss due to oxidation to Ce 4+ during transport.

A mineral assemblage involving grandidierite, ominelite, boralsilite, werdingite, dumortierite (locally Sb,Ti-rich), tourmaline, and corundum, along with the matrix minerals K-feldspar, quartz, and plagioclase, was found in a veinlet cutting leucocratic granulite at Horní Bory, Bory Granulite Massif, Moldanubian Zone of the Bohemian Massif. Zoned crystals of primary grandidierite to ominelite enclosed in quartz are locally overgrown by prismatic crystals of boralsilite and Fe-rich werdingite. Boralsilite also occurs as separate cross-shaped plumose aggregates with Fe-rich werdingite in quartz. Grandidierite is commonly rimmed by a narrow zone of secondary tourmaline or is partially replaced by the assemblage tourmaline + corundum ± hercynite. Grandidierite (X Fe = 0.34-0.71) exhibits dominant FeMg -1 substitution and elevated contents of Li (120-1890 ppm). Boralsilite formula ranges from Al 15.97 B 6.20 Si 1.80 O 37 to Al 15.65 B 5.29 Si 2.71 O 37 and the formula of werdingite ranges from (Fe,Mg) 1.44 Al 14.61 B 4.00 Si 3.80 O 37 to (Fe,Mg) 1.22 Al 14.86 B 4.25 Si 3.55 O 37 . Dumortierite and Sb,Ti-rich dumortierite occur as zoned crystals with zones poor in minor elements (≤0.12 apfu Fe+Mg) and zones enriched in Sb (≤0.46 apfu) and Ti (≤0.25 apfu). Secondary tourmaline (X Fe = 0.44-0.75) of the schorlmagnesiofotite- foitite-olenite solid solution occurs as a replacement product of grandidierite, rarely boralsilite. Other accessory minerals in the veinlet include monazite-(Ce), ilmenite, rutile, ferberite, srilankite, löllingite, arsenopyrite, and apatite. Formation of the borosilicate-bearing veinlet post-dates the development of foliation in the host granulite and is related to the decompressional process. The assemblage most probably originated from a H 2 O-poor system at T ~ 750 °C and P ~ 6-8 kbar. Textural relations as well as geological position of the borosilicate veinlet suggest that it represents the earliest intrusion related to pegmatites in the Bory Granulite Massif. Younger granitic pegmatites in the area are characterized by high contents of B, Al, P, Fe, and minor concentrations of W, Ti, Zr, Sc, and Sb. All pegmatite types probably formed within a short time period of ~5 Ma.

Thorneite, Pb 6 (Te 2 6+ O 10 )(CO 3 )Cl 2 (H 2 O), is a new tellurate from Otto Mountain near Baker, California, named in honor of Brent Thorne. The new mineral occurs on fracture surfaces and in small vugs in brecciated quartz veins. Thorneite is directly associated with acanthite, cerussite, gold, hessite, iodargyrite, khinite, wulfenite, and three other new tellurates: housleyite, markcooperite, and ottoite. Various other secondary minerals occur in the veins, including three other new secondary tellurium minerals: paratimroseite, telluroperite, and timroseite. Thorneite is monoclinic, space group C2/c, a = 21.305(1), b = 11.059(1), c = 7.564(1) Å, β = 101.112(4)°, V = 1748.8(4) Å 3 , and Z = 4. Crystals are prismatic to bladed with elongation and striations parallel to c and typically occur in parallel and random aggregates. It is yellow and transparent, with pale yellow streak and adamantine luster. Mohs hardness is estimated at 2. The mineral is brittle, with an irregular to splintery fracture and good {100} cleavage. The calculated density is 6.828 g/cm 3 . Thorneite is biaxial (+), with large 2V, but indices of refraction are too high to be measured. The optic orientation is Y = b, Z ^ a = 29° in obtuse β. No pleochroism was observed. Electron microprobe analysis provided PbO 73.90, ZnO 0.03, TeO 3 20.35, Cl 2.29, H 2 O 1.28 (structure), CO 2 2.29 (structure), O≡Cl -0.52, total 99.62 wt%; the empirical formula (based on O+Cl = 16) is (Pb 5.94 Zn 0.01 )(Te 6+ 2.08 O 10 )(C 1.00 O 3 )[Cl 1.16 O 0.34 (OH) 0.50 ](H 2 O). The strongest powder X-ray diffraction lines are [d obs in Å (hkl) I]: 10.43 (200) 35, 3.733 (5̅11, 2̅02, 002) 27, 3.595 (4̅21) 33, 3.351 (112) 66, 3.224 (511, 131) 100, 3.093 (2̅22, 3̅31) 30, 2.900 (6̅21) 44, 2.133 (821, 622, 223, 731, 242) 38. The crystal structure (R 1 = 0.028) contains edge-sharing octahedral tellurate dimers, [Te 2 6+ O 10 ] 8- that bond to Pb atoms, which in turn are linked via bonds to Cl atoms, CO 3 triangles, and H 2 O molecules.

Markcooperite, Pb 2 (UO 2 )Te 6+ O 6 , is a new tellurate from Otto Mountain near Baker, California, named in honor of Mark A. Cooper of the University of Manitoba for his contributions to mineralogy. The new mineral occurs on fracture surfaces and in small vugs in brecciated quartz veins. Markcooperite is directly associated with bromian chlorargyrite, iodargyrite, khinite-4O, wulfenite, and four other new tellurates: housleyite, thorneite, ottoite, and timroseite. Various other secondary minerals occur in the veins, including two other new secondary tellurium minerals: paratimroseite and telluroperite. Markcooperite is monoclinic, space group P2 1 /c, a = 5.722(2), b = 7.7478(2), c = 7.889(2) Å, β = 90.833(5)°, V = 349.7(2) Å 3 , and Z = 2. It occurs as pseudotetragonal prisms to 0.2 mm with the forms {100} and {011} and as botryoidal intergrowths to 0.3 mm in diameter; no twinning was observed. Markcooperite is orange and transparent, with a light orange streak and adamantine luster, and is non-fluorescent. Mohs hardness is estimated at 3. The mineral is brittle, with an irregular fracture and perfect {100} cleavage. The calculated density is 8.496 g/cm 3 based on the empirical formula. Markcooperite is biaxial (+), with indices of refraction α = 2.11, β = 2.12, γ = 2.29 calculated using the Gladstone-Dale relationship, measured α-β birefringence of 0.01 and measured 2V of 30(5)°. The optical orientation is X = c, Y = b, Z = a. The mineral is slightly pleochroic in shades of orange, with absorption: X > Y = Z. No dispersion was observed. Electron microprobe analysis provided PbO 50.07, TeO 3 22.64, UO 3 25.01, Cl 0.03, O≡Cl -0.01, total 97.74 wt%; the empirical formula (based on O+Cl = 8) is Pb 2.05 U 0.80 Te 6+ 1.18 O 7.99 Cl 0.01 . The strongest powder X-ray diffraction lines are [d obs in Å (hkl) I]: 3.235 (120, 102, 1̅02) 100, 2.873 (200) 40, 2.985 (1̅21, 112, 121) 37, 2.774 (022) 30, 3.501 (021, 012) 29, 2.220 (221, 2̅21, 212) 23, 1.990 (222, 2̅22) 21, and 1.715 (320) 22. The crystal structure (R 1 = 0.052) is based on sheets of corner-sharing uranyl square bipyramids and tellurate octahedra, with Pb atoms between the sheets. Markcooperite is the first compound to show Te 6+ substitution for U 6+ within the same crystallographic site. Markcooperite is structurally related to synthetic Pb(UO 2 )O 2 .

Timroseite, Pb 2 Cu 5 2+ (Te 6+ O 6 ) 2 (OH) 2 , and paratimroseite, Pb 2 Cu 4 2+ (Te 6+ O 6 ) 2 (H 2 O) 2 , are two new tellurates from Otto Mountain near Baker, California. Timroseite is named in honor of Timothy (Tim) P. Rose and paratimroseite is named for its relationship to timroseite. Both new minerals occur on fracture surfaces and in small vugs in brecciated quartz veins. Timroseite is directly associated with acanthite, cerussite, bromine-rich chlorargyrite, chrysocolla, gold, housleyite, iodargyrite, khinite-4O, markcooperite, ottoite, paratimroseite, thorneite, vauquelinite, and wulfenite. Paratimroseite is directly associated with calcite, cerussite, housleyite, khinite-4O, markcooperite, and timroseite. Timroseite is orthorhombic, space group P2 1 nm, a = 5.2000(2), b = 9.6225(4), c = 11.5340(5) Å, V = 577.13(4) Å 3 , and Z = 2. Paratimroseite is orthorhombic, space group P212121, a = 5.1943(4), b = 9.6198(10), c = 11.6746(11) Å, V = 583.35(9) Å 3 , and Z = 2. Timroseite commonly occurs as olive to lime green, irregular, rounded masses and rarely in crystals as dark olive green, equant rhombs, and diamond-shaped plates in subparallel sheaf-like aggregates. It has a very pale yellowish green streak, dull to adamantine luster, a hardness of about 2½ (Mohs), brittle tenacity, irregular fracture, no cleavage, and a calculated density of 6.981 g/cm 3 . Paratimroseite occurs as vibrant “neon” green blades typically intergrown in irregular clusters and as lime green botryoids. It has a very pale green streak, dull to adamantine luster, a hardness of about 3 (Mohs), brittle tenacity, irregular fracture, good {001} cleavage, and a calculated density of 6.556 g/cm 3 . Timroseite is biaxial (+) with a large 2V, indices of refraction > 2, orientation X = b, Y = a, Z = c and pleochroism: X = greenish yellow, Y = yellowish green, Z = dark green (Z > Y > X). Paratimroseite is biaxial (-) with a large 2V, indices of refraction > 2, orientation X = c, Y = b, Z = a and pleochroism: X = light green, Y = green, Z = green (Y = Z >> X). Electron microprobe analysis of timroseite provided PbO 35.85, CuO 29.57, TeO 3 27.75, Cl 0.04, H 2 O 1.38 (structure), O≡Cl -0.01, total 94.58 wt%; the empirical formula (based on O+Cl = 14) is Pb 2.07 Cu 2+ 4.80 Te 6+ 2.04 O 12 (OH) 1.98 Cl 0.02 . Electron microprobe analysis of paratimroseite provided PbO 36.11, CuO 26.27, TeO 3 29.80, Cl 0.04, H 2 O 3.01 (structure), O≡Cl -0.01, total 95.22 wt%; the empirical formula (based on O+Cl = 14) is Pb 1.94 Cu 2+ 3.96 Te 6+ 2.03 O 12 (H 2 O) 1.99 Cl 0.01 . The strongest powder X-ray diffraction lines for timroseite are [d obs in Å (hkl) I]: 3.693 (022) 43, 3.578 (112) 44, 3.008 (023) 84, 2.950 (113) 88, 2.732 (130) 100, 1.785 (multiple) 33, 1.475 (332) 36; and for paratimroseite 4.771 (101) 76, 4.463 (021) 32, 3.544 (120) 44, 3.029 (023,122) 100, 2.973 (113) 48, 2.665 (131) 41, 2.469 (114) 40, 2.246 (221) 34. The crystal structures of timroseite (R1 = 0.029) and paratimroseite (R 1 = 0.039) are very closely related. The structures are based upon edge- and corner-sharing sheets of Te and Cu polyhedra parallel to (001) and the sheets in both structures are identical in topology and virtually identical in geometry. In timroseite, the sheets are joined to one another along c by sharing the apical O atoms of Cu octahedra, as well as by sharing edges and corners with an additional CuO 5 square pyramid located between the sheets. The sheets in paratimroseite are joined only via Pb-O and H bonds.

Telluroperite, Pb 3 Te 4+ O 4 Cl 2 , is a new tellurite from Otto Mountain near Baker, California. The new mineral occurs on fracture surfaces and in small vugs in brecciated quartz veins in direct association with acanthite, bromine-rich chlorargyrite, caledonite, cerussite, galena, goethite, and linarite. Various other secondary minerals occur in the veins, including six new tellurates, housleyite, markcooperite, paratimroseite, ottoite, thorneite, and timroseite. Telluroperite is orthorhombic, space group Bmmb, a = 5.5649(6), b = 5.5565(6), c = 12.4750(14) Å, V = 386.37(7) Å 3 , and Z = 2. The new mineral occurs as rounded square tablets and flakes up to 0.25 mm on edge and 0.02 mm thick. The form {001} is prominent and is probably bounded by {100}, {010}, and {110}. It is bluish-green and transparent, with a pale bluish-green streak and adamantine luster. The mineral is non-fluorescent. Mohs hardness is estimated to be between 2 and 3. The mineral is brittle, with a curved fracture and perfect {001} cleavage. The calculated density based on the empirical formula is 7.323 g/cm 3 . Telluroperite is biaxial (-), with very small 2V (~10°). The average index of refraction is 2.219 calculated by the Gladstone-Dale relationship. The optical orientation is X = c and the mineral exhibits moderate bluish-green pleochrosim; absorption: X < Y = Z. Electron microprobe analysis provided PbO 72.70, TeO 2 19.26, Cl 9.44, O≡Cl -2.31, total 99.27 wt%. The empirical formula (based on O+Cl = 6) is Pb 2.79 Te 4+ 1.03 O 3.72 Cl 2.28 . The six strongest powder X-ray diffraction lines are [d obs in Å (hkl) I]: 3.750 (111) 58, 2.857 (113) 100, 2.781 (020, 200) 43, 2.075 (024, 204) 31, 1.966 (220) 30, and 1.620 (117, 313, 133) 52. The crystal structure (R 1 = 0.056) is based on the Sillén X 1 structure-type and consists of a three-dimensional structural topology with lead-oxide halide polyhedra linked to tellurium/lead oxide groups. The mineral is named for the relationship to perite and the dominance of Te (with Pb) in the Bi site of perite.

Raman spectroscopy is used to identify the sulfur speciation (sulfur valence state) in “technical” iron-free soda-lime and potassium-silicate glasses and in “natural” glass compositions such as basalt, andesite, and rhyodacite. The presence of sulfate (S 6+ ) is marked in Raman spectra of oxidized synthetic and natural glasses by bands at ~990 and ~1000 cm -1 , respectively. The presence of sulfide (S 2- ) in the reduced technical glasses is marked in Raman spectra by a band at 2574 cm -1 indicating that S 2- is present as HS - in these Fe-free glasses. Such a band is absent in the Raman spectra of reduced basaltic, andesitic, and rhyodacitic glasses. However, an additional band at ~400 cm -1 appears in the Raman spectra of the reduced S-bearing “natural” glasses when compared to that of S-free reduced “natural” reference glasses indicating that S 2- is most likely complexed with Fe in these glasses. Thus, the dissolution mechanism of S 2- appears to be different in Fe-free and Fe-bearing glasses and S 2- is dissolved as either HS - -species or Fe-S complexes, respectively. The data shown here demonstrate the potential of Raman spectroscopy in identifying the sulfur valence state in silicate glasses. In addition, S 2- is dissolved as completely different complexes when comparing “technical” iron-free and “natural” iron-bearing, hydrous silicate glass compositions.

Calcium aluminate glasses are important materials where AlO - 4/2 is the only network former. Aluminum in crystals or glasses between CaO and Al 2 O 3 can have different environments as a function of the CaO/Al 2 O 3 ratio. Using X-ray absorption at the Al K- and L-edges, Raman and 27 Al NMR spectroscopies, we have determined the structural surroundings of Al in glasses, crystals, and melts in this binary system. Aluminum is in octahedral coordination at high-Al 2 O 3 content (>80 mol%) and essentially in fourfold coordination with 4 bridging O atoms (BOs) at Al 2 O 3 contents between 30 and 75 mol%. At around 25 mol% Al 2 O 3 , Al is in tetrahedral coordination with two BOs. The presence of higher-coordinated species at high-Al 2 O 3 contents and their absence at low Al 2 O 3 imply different viscous flow mechanisms for high- and low-concentration Al 2 O 3 networks.

The results of in situ synchrotron X-ray powder diffraction experiments conducted during the synthesis of iron-deficient Na + /H 3 O + and K + /H 3 O + jarosites at temperatures in the range 80 to 120 °C are presented. They demonstrate that samples can be prepared in which the iron-site vacancies are ordered. The ordering is accompanied by a lowering of symmetry, from rhombohedral, space group R3̅m, to monoclinic, C2/m. The implications for magnetic properties are discussed.