Volume 91, Issue 4

X-ray diffraction (XRD) and Mn K-edge extended X-ray absorption fine structure (EXAFS) spectroscopy were combined to elaborate a structural model for phyllomanganates (layer-type Mn oxides) lacking 3D ordering (turbostratic stacking). These techniques were applied to a sample produced by a common soil and freshwater bacterium (Pseudomonas putida), and to two synthetic analogs, δ-MnO 2 and acid birnessite, obtained by the reduction of potassium permanganate with MnCl 2 and HCl, respectively. To interpret the diffraction and spectroscopic data, we applied an XRD simulation technique utilized previously for well-crystallized birnessite varieties, complementing this approach with single-scattering-path simulations of the Mn K-edge EXAFS spectra. Our structural analyses revealed that all three Mn oxides have an hexagonal layer symmetry with layers comprising edgesharing Mn 4+ O 6 octahedra and cation vacancies, but no layer Mn 3+ O 6 octahedra. The proportion of cation vacancies in the layers ranged from 6 to 17%, these vacancies being charge-compensated in the interlayer by protons, alkali metals, and Mn atoms, in amounts that vary with the phyllomanganate species and synthesis medium. Both vacancies and interlayer Mn were most abundant in the biogenic oxide. The diffracting crystallites contained three to six randomly stacked layers and have coherent scattering domains of 19.42 Å in the c* direction, and of 60.85 Å in the a-b plane. Thus, the Mn oxides investigated here are nanoparticles that bear significant permanent structural charge resulting from cation layer vacancies and variable surface charge from unsaturated O atoms at layer edges.

The FKα emission in samples of triplite-group phosphates (TGP) submitted to electron beams routinely used for electron-probe microanalysis of light elements (5-10 keV, 30 nA, 5 μm in diameter) evolves anisotropically with time. The intensity variations consist of an initial rise of the signal (time constants on the order of few tens of seconds), by up to 20.30% of the zero-time value, followed by a longer-time-constant increase or decrease. The initial signal rise is interpreted as an enhancement of the fluorescence yield of the F ions due to bond breaking. Halogen-formed lattices are known to be sensitive to radiolysis, i.e., destabilization of the atomic arrangement due to perturbations of the electronic structure induced by particle irradiation, leading often to bond breakage and lattice-ion displacements. In strongly ionic structures (alkali halides and alkaline-earth fluorides, notably), bond breakage results from the decay of metastable states in which transient molecular ions (e.g., F 2 - ) are formed. Molecular ions preexisting within ionic lattices under the form of halogen-halogen bonds (e.g., F 0 -F - bonds) are likely involved in the degradation processes under irradiation of materials such as fluoro-phosphates. High-resolution studies of the FKα emission band in TGPs reveal that lattice F ions may have an average fractional charge of .x (0.5 < x < 1, instead of x = 1 for ideally ionic lattices) consistent with the presence of F 0 -F - bonds. Other longer-time-constant processes are attributed to the production of F - interstitials beneath the specimen surface that can accumulate (increasing signal) or desorb (decreasing signal) when they have enough energy to overcome the surface potential barrier. The role of the energy stored within the lattice by charging is critical to the understanding of both the anisotropy of the FKα intensity and the accumulation-desorption processes of the produced F - interstitials. Owing to the differences between specimens in (1) the effective charge of the F ions and (2) the electron-stimulated desorption (ESD) mechanisms leading to the emission of F - and F + ions, it is concluded that standards with structural properties close to those of unknowns should be selected for accurate analysis of F in beam-sensitive minerals or glasses.

First-principles calculations within the local density and pseudopotential approximations were performed to investigate the effects of pressure on the energetics and structural behavior of charged vacancy defects in MgO. The simulations were performed for a supercell containing 216 atoms with their positions being fully optimized. In particular, the formation and migration energies of cation and anion vacancies were shown to substantially increase over the pressure regime of the Earth.s mantle. Our results thus suggest that pressure should suppress intrinsic diffusion mediated by ionic vacancies in MgO over the mantle pressure regime. The calculated three-dimensional data sets for atomic displacements and electron charge density were explored in detail using an interactive visualization system. Although the atomic and electronic structures are highly distorted in the close vicinity of the defects (i.e., in the region covering up to the nearest and next-nearest atoms), the effects are not negligible at farther distances

Despite the abundance of garnet in deforming regions of the Earth, such as subduction zones, its rheological properties are not well defined by laboratory measurements. Here we report measurements of steady-state plastic properties of pyrope in its stability field (temperature up to 1573 K, pressure up to 6.8 GPa, strain rate ~10 -5 s -1 ) using a Deformation-DIA apparatus (D-DIA) coupled with synchrotron radiation. Synthetic pyrope (Py100) and natural pyrope (Py 70 Alm 16 Gr 14 ) are both studied in a dry environment. Transmission electron microscopy (TEM) investigation of the run products indicates that dislocation glide, assisted by climb within grains and dynamic recrystallization for grain-boundary strain accommodation, is the dominant deformation process in pyrope. Both synthetic-and naturalpyropes . stress and strain-rate data, as measured in situ by X-ray diffraction and imaging, are best fitted with the single flow law: where ε̇ is the strain rate, σ = | σ 1 -σ 3 | is the differential stress, R is the gas constant and T the absolute temperature. Synthetic forsterite and synthetic pyrope were stacked along the uniaxial compression direction in the same cell assembly during deformation to compare their strength at mantle condition. Forsterite is observed to be stronger than pyrope, deforming at a rate about 10% slower than the pyrope at 5.2 GPa and 1573 K. San Carlos olivine and natural pyrope were compared in a similar fashion at 6.8 GPa and 1473 K. In this case, San Carlos olivine deformed 2~3 times faster than natural pyrope. The experimental data suggest that pyrope is stronger (by more than a factor of 4) than the dominant mineral (olivine) in the upper mantle when temperatures exceed 1273 K.

This work reports the crystal-chemical characterization of Mn-rich amphiboles from the Grenville Marble of the Arnold mine, Fowler, St. Lawrence Co., New York (U.S.A.), which were previously described by Benimoff et al. (1991) as “manganoan silicic edenite”. According to the new nomenclature scheme for monoclinic amphiboles (Leake et al. 2004) the ideal composition of reference, A Na B (CaMn) C Mg 5 T (Si 7 Al)O 22 (OH) 2 with Ca > 1 apfu, is an end-member of the newly defined Group 5, and is named parvo-mangano-edenite. Re-examination of the original rock specimen showed significant inter- and intra-crystalline compositional variations, which can be expressed by the A Na -1 T Al -1 A ⃞ 1 T Si 1 and B Mn 2 B Ca -2 exchange vectors. The first vector leads to parvo-manganotremolite, ideally A ⃞ B (CaMn) C Mg 5 T Si 8 O 22 (OH) 2 with Ca > 1. The second mechanism was never found to reach Mn dominance; however, crystal-chemical analysis does not provide any evidence of structural limits, and thus the magno-calcic counterparts of the Group 5 amphiboles of this work may occur in similar but Mn-richer genetic environments. The presence of Mn at the B site helps to stabilize the charge arrangement of edenite. The parvo-mangano-edenite crystal with composition closest to the end-member, i.e., A(Na 0.74 K 0.02 )B(Ca 1.27 Mn 0.73 )C(Mg 4.51 Mn 2+ 0.28 Fe 2+ 0.05 Fe 3+ 0.03 Al 0.12 Ti 0.01 )T(Si 7.07 Al 0.93 )O 22 (OH) 2 , has a = 9.8260(5), b = 18.0487(9), c = 5.2840(4) Å, β = 104.55(1)º, V = 907.1 Å 3 (Z = 2); the calculated density is 3.11 g/cm 3 . The parvo-mangano tremolite crystal, with composition A (Na 0.44 K 0.01 ) B (Ca 1.13 Mn 0.83 Na 0.04 ) C (Mg 4.69 Mn 2+ 0.21 Fe 2+ 0.03 Fe 3+ 0.01 Al 0.06 ) T (Si 7.52 Al 0.48 )O 22 (OH) 2 , has a = 9.7807(5), b = 18.0548(9), c = 5.2928(4) Å, β = 104.19(1)º, V = 906.1 Å 3 (Z = 2); the calculated density is 3.08 g/cm 3 . The different compositions are virtually indistinguishable under the optical microscope, but can be identified by a measure of their unit-cell parameters. The single-crystal FTIR spectrum of parvo-mangano-edenite in the OH-stretching region shows two main absorptions at 3711 and 3671 cm -1 , plus shoulders at 3695, 3660, and 3641 cm -1 , respectively. FTIR spectroscopy indicates extensive short-range-order of cations in these amphiboles.

High-pressure single-crystal X-ray diffraction experiments of MgSiO 3 perovskite have been carried out up to 15 GPa in a diamond-anvil cell using synchrotron radiation. Precise crystal structural parameters, including the anisotropic displacement parameters of every atom in MgSiO 3 , are determined under high pressure. In the pressure range up to 15 GPa, the most important responses of the structure are the compressions of SiO 6 and MgO 8 polyhedra and an increase in tilting of SiO 6 octahedra represented by the decrease in angles between octahedra (both Si-O2-Si angle in the a-b plane and Si-O1-Si angle in the b-c plane decrease). The degree of the change in both angles in the a-b and b-c planes is the same. The amplitude of mean square displacement for the Mg atom has the largest value in the structures and its thermal vibration is significantly anisotropic at ambient pressure. Under high pressure, all atoms in the structure have obvious anisotropy of thermal vibration and the largest amplitudes of thermal vibration for Mg, Si, and O2 atoms are directed toward vacant space in the structure. Anisotropy of the structure increases with pressure.

The oxidation behavior of synthetic α-NiS in air has been investigated over the temperature range 670-700 °C. The α-NiS was ground and sieved to give a particle size ranging from 53 to 90 μm. Three oxidation paths were observed: (i) α-NiS + 3/2 O 2 → NiO + SO 2 (ii) 3α-NiS +O 2 → Ni 3 S 2 + SO 2 (iii) Ni 3 S 2 + 7/2 O 2 → 3NiO + 2SO 2 No Ni 3 S 2 (heazlewoodite) was observed over the course of α-NiS oxidation at 670 and 680 °C. The dominant oxidation path at this temperature is path i. At 700 °C, however, all three oxidation paths were observed. As an intermediate oxidation product, Ni 3 S 2 steadily exsolved from α-NiS, reaching a maximum quantity after about 80 min of oxidation, declining afterward, and approaching annihilation at 160 min of oxidation. Experimental results show that the exsolution of Ni 3 S 2 is likely triggered by the loss of one third of S in the α-NiS structure with the release of SO 2 rather than by an intrinsic thermal decomposition of α-NiS to α-Ni 1-x S + Ni 3 S 2 . The eventual annihilation of Ni 3 S 2 was caused by a further oxidation of Ni 3 S 2 to NiO. Oxidation paths 2 and 3 form a typical single chain reaction: The approximate values of k1 are k2 are 3 × 10 -4 s -1 and 5 × 10 -4 s -1 respectively. Oxidation temperature was found to play important roles both in the oxidation kinetics and the oxidation mechanism. By decreasing 10 °C from 680 to 670 °C, the average reaction rate (dy/dt, where y is the reaction extent) over the experiment time scale almost decreased to one third of its original rate (from 3.3 × 10 -5 s -1 to 1.2 × 10 -5 s -1 ). The reaction mechanism in the temperature range 670 to 680 °C is constant with E a = 868.2 kJ/mol.

A series of synthetic sodium-rich saponites and trioctahedral Na, K, and Ba mica solid solutions have been investigated by 29 Si MAS-NMR spectroscopy. The salient result is the major effect of the interlayer charge on 29 Si chemical shift variations detected in NMR components ascribed to tetrahedral Si 3 , Si 2 Al, SiAl 2 , and Al 3 environments. In this analysis, the effect of the octahedral charge is considerably lower. The contribution of the interlayer charge is much more important than the previously reported contribution of the ditrigonal distortion of tetrahedral layers, generated by the misfit between tetrahedral and octahedral layers. This observation rests on the peculiar disposition of tetrahedra in phyllosilicates, with three out of four oxygen atoms exposed to the interlamellar space. A similar effect could be operative in other silicates, like zeolites, in which most of tetrahedra interact with charge compensating cations.

The size distribution of pigeonite and diopside exsolution lamellae on “001,” obtained at temperatures of 1100, 1200, and 1300 °C and annealing times between 2 and 4320 h, was studied by transmission electron microscopy. A total of 5192 pigeonite and 5286 diopside lamellae was studied. At all three temperatures, the size distributions of pigeonite and diopside lamellae are smaller during exsolution compared to the subsequent coarsening process. The final size distributions are time invariant, indicating that a steady-stage distribution is reached. The theory of Ardell (1972a), which assumes volume diffusion as rate-limiting process and takes into account the non-zero volume fraction of the precipitates, describes the experimental size distributions quite well and also leads to the observed exponent of three in the rate law

The high-pressure transformation of MgFe 2 O 4 was studied by Mössbauer and Raman spectroscopy and synchrotron X-ray diffraction using the DAC technique and laser annealing at temperatures of 1500-2000 K. The high-pressure phase of MgFe 2 O 4 was observed from in situ Mössbauer spectra at 17 ± 1 GPa after laser annealing by the appearance of two quadrupole doublets. This indicates a disordered distribution of Mg and Fe in an early stage. The displacive nature of the transformation of the spinel into its high-pressure polymorph was shown at increasing pressure by the redistribution of iron into only one site. After decompression Mössbauer spectroscopy revealed the presence of Fe 2 O 3 in the sample. This was further confirmed by Raman spectroscopy at ambient conditions and by in situ high-pressure XRD, indicating a partial breakdown of the spinel into its constituent oxides MgO and Fe 2 O 3 . The XRD pattern of the high-pressure phase of MgFe 2 O 4 can be indexed in agreement with the CaMn 2 O 4 -type structure, with cell parameters a = 2.775(2), b = 9.283(16), and c = 9.446(5) Å at 23 ± 2 GPa. The multiphase spectra from all three analytical methods suggests that inhomogeneous conditions prevailed in the DAC experiments, resulting in two different reactions at high pressure and temperature, i.e., T < 1800 K: MgFe 2 O 4 → Fe 2 O 3 + MgO and T > 1800 K: MgFe 2 O 4 → hp-MgFe 2 O 4 .

Elastic and structural behavior of a natural cubic analcime (space group: Ia3 . d) from Sardinia (Italy) was investigated at high pressure by in situ single-crystal X-ray diffraction. A first-order phase transition is observed in the pressure range between 0.91(5) and 1.08(5) GPa. Unit-cell constants and reflection conditions confirm that the space group of the HP-polymorph is P1̄. No further phasetransition has been observed at least up to 7.14 GPa. Fitting the volume data of the cubic polymorph with a second-order BM-EoS we obtain: V 0 = 2571.2(4) Å 3 , K T0 = 56(3) GPa, and K’ = 4 (fixed). For the triclinic polymorph, a third-order EoS gives: V 0 = 2607(9) Å 3 , K T0 = 19(2) GPa, and K’ = 6.8(7). Axial bulk moduli of the triclinic polymorph, calculated with a linearized BM-EoS, are: K T0 (a) = 29(2) GPa, with K’(a) = 4.9(6) and a 0 = 13.727(10) Å; K T0 (b) = 20(1) GPa, with K’(b) = 5.2(5), and b 0 = 13.751(15) Å; K T0 (c) = 11(1) GPa, with K’(c) = 12.6(6) and c 0 = 13.822(31) Å. The elastic behavior of the HP-polymorph appears to be strongly anisotropic, being K T0 (a):K T0 (b):K T0 (c) = 2.64:1.82:1.00. The relevant structural variations in response to the cubic → triclinic phase transition are due to tetrahedral tilting. The tetrahedral framework distortion gives rise to a change of the eight- and six-ring channels ellipticity and of the extra-framework topological configuration: it appears in fact that for the high-pressure triclinic polymorph the coordination number of some of the Na atoms is seven (2H 2 O + five framework O atoms) instead of six (2H 2 O + four framework O atoms).

A new chemical etching technique has been devised for synthetic Fe-free peridotites composed of forsterite (Fo), enstatite (En), and diopside (Di) Among the etchants were acids, molten carbonates, and borates, but it was found that only molten anhydrous borax (Na 2 B 4 O 7 ) dissolved all phases equally. Molten anhydrous borax was found to be a successful etchant in equally enhancing all the grain (i.e., Fo-Fo, En-En, and Di-Di) and interphase (i.e., Fo-En, Fo-Di, and En-Di) boundaries. From the backscattered electron images of the etched surface, maps of grain- and interphase-boundaries can be obtained semi-automatically for microstructural analysis by using image processing software. An Febearing wehrlite was also etched successfully by molten anhydrous borax, thus showing the usefulness of this technique for enhancing the grain- and interphase-boundaries in many natural peridotites

Poppiite, a new mineral from Gambatesa mine (Val Graveglia, Genova, Northern Italy), is the V 3+ end-member of the pumpellyite group [ [VII] (Ca 7.68 Na 0.27 K 0.03 Rb 0.02 ) Σ8.00 [VI] (V 3+ 1.26 Fe 3+ 1.02 Mg 0.78 Mn 2+ 0.59 Al 0.31 Cu 2+ 0.04 ) Σ4.00 [VI] (V 3+ 6.89 Al 1.07 Ti 0.04 ) Σ8.00 [IV] (Si 11.69 Al 0.31 ) Σ12.00 O 42 (OH) 14 ; C2/m, a = 19.2889(6), b = 6.0444(2), c = 8.8783(3) Å, β = 97.328(2)°, V = 1026.66(6) Å 3 , D meas = 3.36(2) g/cm 3 , and D calc = 3.44 g/cm 3 ]. Poppiite crystals, with size varying from 0.1 to 0.6 mm, are minute, greenish-brown, and prismatic, and are associated with roscoelite, ganophyllite, manganaxinite, goldmanite, and calcite. The strongest lines in the X-ray powder diffraction pattern [dobs (Å), Irel, (hkl)] are: 2.930, 100, (511); 3.817, 70, (202); 2.548, 65, (3.13); 2.551, 62, (420); 1.612, 57, (7.31,424); and 2.367, 51, (222, 403). Poppiite is optically negative, with 2V calc = 44°, n α = 1.768(9), n β 1.804(8), n γ 1.810(9). The pleochroic scheme is α = light yellowish brown, β = deep greenish brown, and γ = brown to reddish brown. The crystal structure was refined using 1918 unique reflections to R = 0.0307. Like the other pumpellyite-group minerals the crystal structure of poppiite consists of chains of edge-sharing octahedra linked by SiO 4 , Si 2 O 7 , and CaO 7 polyhedra

During diagenesis of Jurassic and Cambrian oil source rock shales illite-smectite(-vermiculite) [I-S(-V)] is transformed to illite-tobelite-smectite(-vermiculite) [I-T-S(-V)]. This transformation of S layers to T layers takes place by an increase in tetrahedral charge through Al for Si substitution and subsequent fixation of interlayer NH 4 , accompanied by an increase in Al and a decrease in Fe and Mg in the octahedral sheet. In the present investigation, the distribution of isomorphous cations in octahedral sheets of trans-vacant I-S(-V) and I-T-S(-V) was studied by Mössbauer and Infrared (IR) spectroscopies. Mössbauer spectra have been modeled using numerical values of the Fe 3+ and Fe 2+ quadrupole doublets corresponding to local cation arrangements around Fe 3+ and Fe 2+ in octahedral sheets of micaceous minerals. To interpret IR spectra in the OH-stretching region, frequencies for each pair of cations bonded to OH groups determined for micas and I-S are used. Combination of Mössbauer and IR data by computer simulation provides two-dimensional cation distributions of octahedral cations. The Jurassic and Cambrian I-S(-V) and I-T-S(-V) have clustered octahedral sheets. Ordered clusters of mixed cation composition (Mg, Al, Fe 3+ , and Fe 2+ ) with regular alternation of di- and trivalent cations and Fe 3+ -clusters dispersed over an Al-matrix are found in detrital samples. In diagenetically transformed samples, ordered clusters persist while Fe 3+ -clusters degenerate to either short chains consisting of two Fe-Fe pairs or to isolated Fe-Fe pairs oriented along the b, b 1 , and b 2 directions. The release of Fe and Mg during diagenesis occurs from Fe 3+ clusters and through partial destruction of ordered clusters and of b 1 , b 2 -oriented Mg-Mg pairs. However, as the cation composition and the short-range cation order within the clusters are preserved and the Al for Fe and Mg substitution occurs at cluster edges, the diagenetic transformation of S (and V) to T layers in both the Jurassic and Cambrian I-S(-V) proceeds through a solid-phase transformation and not through dissolution-reprecipitation.

A shock-induced polymorph (TiO 2 II) of anatase and rutile has been identified in breccias from the late Eocene Chesapeake Bay impact structure. The breccia samples are from a recent, partially cored test hole in the central uplift at Cape Charles, Virginia. The drill cores from 744 to 823 m depth consist of suevitic crystalline-clast breccia and brecciated cataclastic gneiss in which the TiO 2 phases anatase and rutile are common accessory minerals. Electron-microprobe imaging and laser Raman spectroscopy of TiO 2 crystals, and powder X-ray diffraction (XRD) of mineral concentrates, confirm that a high-pressure, α-PbO 2 structured polymorph of TiO 2 (TiO 2 II) coexists with anatase and rutile in matrix-hosted crystals and in inclusions within chlorite. Raman spectra of this polymorph include strong bands at wavenumbers (cm.1) 175, 281, 315, 342, 356, 425, 531, 571, and 604; they appear with anatase bands at 397, 515, and 634 cm -1 , and rutile bands at 441 and 608 cm -1 . XRD patterns reveal 12 lines from the polymorph that do not significantly interfere with those of anatase or rutile, and are consistent with the TiO 2 II that was first reported to occur naturally as a shock-induced phase in rutile from the Ries crater in Germany. The recognition here of a second natural shock-induced occurrence of TiO 2 II suggests that its presence in rocks that have not been subjected to ultrahigh-pressure regional metamorphism can be a diagnostic indicator for confirmation of suspected impact structures.

Birnessite is an important scavenger of trace metals in soils and aqueous environments. The basic birnessite-type structure consists of sheets of Mn octahedra separated by ~7 or ~10 Å (“buserite”) interlayer regions filled with cations and water. Synthetic birnessite-like structures were produced through cation exchange reactions with synthetic Na-birnessite. The unheated, synthetic Mg 2+ , Ca 2+ , and Ni 2+ layer structures have an ~10 Å interlayer spacing, whereas the other cation-exchanged synthetic birnessites and the related mineral chalcophanite have an interlayer spacing of ~7 Å. The Li + , Na + , K + , Cs + , and Pb 2+ synthetic birnessites each contain two to three structurally different water sites, as evidenced by multiple H 2 O bending and stretching modes in the infrared spectra. The complexity of the water bands in these spectra is likely related to disordering of cations on the interlayer sites. H-birnessite contains structural water and either hydroxyl, hydronium (H 3 O + ), or both. The small difference in the width of the water stretching modes between room temperature and -180 °C indicates that the water molecules in birnessite-like structures are predominantly structurally, rather than dynamically, disordered. Most of the synthetic birnessites, including Na- and K-birnessite, undergo significant water loss at temperatures below 100 °C. There is a linear relationship between the temperature at which most of the water is lost from a given cation-exchanged birnessite and the heat of hydration of the interlayer cation. This finding implies that the interlayer water is strongly bound to the interlayer cations, and plays an important role in the thermal stability of birnessite-like structures.

Hydrothermal AFM has been used to study the thermal dehydration reaction on gypsum (010) surfaces in solutions at different saturation states, and in the absence of a bulk liquid phase. Experiments were carried out at temperatures ranging from 25 to 130 °C. Whereas supersaturated solutions (β = 1.8.5) caused gypsum growth in the entire temperature range, solutions close to equilibrium (β = 1.02) caused various responses of the gypsum surface. The most prominent was a sharp transition from fast growth to very fast dissolution at ~120 °C suggesting a sudden nucleation of a phase more stable than gypsum. No structural relation could be found between the parental gypsum (010) surface and the crystallizing phase. In the absence of a bulk liquid phase, dehydration takes place via the nucleation and spreading of etch-pit like pattern. Laterally, the thermal etch pits spread in an unrestricted way. In the vertical direction, pit growth was limited to a few micrometers. Dehydration by monolayer pits and nucleation of the dehydration process at monolayer steps on the (010) surface were never observed. Thus, unlike growth or dissolution, surface energy related to kink sites or individual point defects seems to be insufficient to trigger dehydration. The temperature-dependent lateral pit growth yields an activation energy of 119 ± 11 kJ/mol. The product phase disintegrates at the parental surface into nano-size particles without any morphologically noticeable transition zone.

The thermally induced structural modifications of the natural zeolite stellerite [Ca 8 Al 16 Si 56 O 144 ·58H 2 O, a = 13.5947(4), b = 18.1823(6), and c = 17.8335(6) Å, V = 4408.1(3) Å 3 , space group Fmmm, framework type STI] were studied in a temperature-resolved X-ray powder diffraction experiment, using synchrotron radiation, in the temperature range RT-976 K. In the initial stage of heating (below 430 K) Stellerite Phase A (space group Fmmm) is stable, and the cell volume decreases only 0.6% to this temperature. Between 430 and 490 K most of the water is released, the symmetry lowers, and a phase transition to the collapsed so-called Phase B (space group Amma) is observed. In this phase rotation of the 4254 Secondary Building Units causes cell volume contraction and deformation of the channel system. This new phase, at 530 K, shows the statistical breaking of T-O-T bridges in the four-rings and the migration of tetrahedral atoms to new .face-sharing. tetrahedra, which partially occlude both the channels parallel to [100] and to [001]. This framework deformation is interpreted as due to the strain induced by calcium atoms on the framework to achieve better coordination after the release of water. The new structure is stable up to 750 K and the total volume decrease is about 8%. The dehydration process causes a similar framework deformation and the transition to a collapsed phase characterized by the statistical breaking of T-O-T bridges in all zeolites with STI-type frameworks. However, comparing the results obtained with different thermal kinetics, it is possible to assume that the experimental conditions play a primary role in the mechanisms of dehydration and of framework bridge breaking.

The molar volume (Vm) and chemical composition (x) of saline aqueous inclusions and gas inclusions in minerals can be calculated satisfactorily from microthermometric and other analytical data. For complex gas-bearing aqueous inclusions, however, calculation of Vm-x properties requires additional input of the volume-fractions of the inclusion phases (φ). Traditional estimation of φ in non-fluorescing inclusions involves measuring area-fractions of the phases projected in the microscope and then making rough corrections for the third dimension. The uncertainties in the results are unknown and therefore the accuracies of the calculated Vm-x properties are also unknown. To alleviate this problem we present a new, routine method to estimate φ using the petrographic microscope in conjunction with a spindle-stage. Inclusions in normal thick-sections are rotated stepwise and their projected areas and area-fractions are plotted against rotation angle. The resulting data arrays are systematically related to inclusion orientation, to inclusion shape, and to φ. The dependency on orientation is minimized when area fractions are measured at the position where the inclusions project their largest total areas. The shape dependency is accounted for using a new objective classification of inclusion projections, based on parameters from digital image processing. The method has been verified with synthetic fluid inclusions of known φ. For individual liquid + vapor inclusions with regular (not .negative-crystal.) shapes, the new procedure yields φ with a relative accuracy of ±4%. This degree of accuracy permits Vm . x properties of gas-bearing, aqueous fluid inclusions to be calculated with sufficient certainty for many geochemical applications. Even better accuracy (e.g., down to ±0.6%) can be obtained by combining results from several inclusions in the same homogeneously trapped petrographic assemblage.

The enthalpies of formation of seven uranyl oxide hydrate phases and one uranate have been determined using high-temperature oxide melt solution calorimetry: [(UO 2 ) 4 O(OH) 6 ](H 2 O) 5 , metaschoepite; β-UO 2 (OH) 2 ; CaUO 4 ; Ca(UO 2 ) 6 O 4 (OH) 6 (H 2 O) 8 , becquerelite; Ca(UO 2 ) 4 O 3 (OH) 4 (H 2 O) 2 ; Na(UO 2 )O(OH), clarkeite; Na 2 (UO 2 ) 6 O 4 (OH) 6 (H 2 O) 7 , the sodium analogue of compreignacite, and Pb 3 (UO 2 ) 8 O 8 (OH) 6 (H 2 O) 2 , curite. The enthalpy of formation from the binary oxides, ΔH f-ox , at 298 K was calculated for each compound from the respective drop solution enthalpy, ΔH ds . The standard enthalpies of formation from the elements, ΔH 0 f , at 298 K are -1791.0 ± 3.2, -1536.2 ± 2.8, -2002.0 ± 3.2, -11389.2 ± 13.5, -6653.1 ± 13.8, -1724.7 ± 5.1, -10936.4 ± 14.5, and -13163.2 ± 34.4 kJ/mol, respectively. These values are useful in exploring the stability of uranyl oxide hydrates in auxiliary chemical systems, such as those expected in U-contaminated environments.

Electron-beam irradiation causes permanent damage to hydrous, silica-rich glasses. The extent of electron-beam damage is quantified using data generated by SIMS analysis of points subjected to previous electron microprobe analysis (EPMA). Even optimum EPMA conditions cause damage to the glass, manifest as a marked depletion in alkali ions at the surface of an irradiated sample. Deeper in the sample, an enrichment in alkali ions to above-baseline levels is followed by a decay back to baseline. The depth of the final decay correlates with species diffusivity and increases in the order K-Li-Na. H-bearing species are also affected by electron beam irradiation, but in the opposite sense to the alkalis, i.e., they are enriched at the surface. Migration of alkaline earth cations is not observed because of their low diffusivities. Ion depletion or enrichment results from simple migration of ions toward or away from electrons implanted by the beam. Migration depth depends on species diffusivity and heating caused by the electron beam, and therefore increases with increasing electron beam current. Because of the reverse behavior of H, the mobile hydrous species in the presence of an electric field is probably OH - . The extent of electron beam damage to glasses may increase with total water content. Critically, SIMS measurements of H, Li, Na, D/H, and 6 Li/ 7 Li after electron-probe analysis are compromised by the damage. Despite the damage caused by the electron beam, use of appropriate electron-beam conditions (e.g., 2 nA, 15 kV) gives volatiles by difference accurate to ~0.6 wt%.

Ag-rich tourmaline crystals were synthesized at 750 °C, 200 MPa H 2 O, and f O₂ = log (NNO) - 0.5, starting with an oxide mix of dravite composition to which various reagents, including AgF and AgCl, were added. Tourmaline containing up to 7.65 wt% Ag2O was synthesized, and this is the first time a tourmaline is described that contains significant amounts of Ag at the ninefold-coordinated X site. Crystal structure refinement and chemical analysis (EMPA) give the optimized formula X (Na 0.58 Ag 0.18 □ 0.24 ) Y (Al 1.54 Mg 1.46 ) Z (Al 5.34 Mg 0.66 ) T (Si 5.90 Al 0.10 )O 18 (BO 3 ) 3 V (OH) 3 W (O 0.53 F 0.47 ), with a = 15.8995(4) and c = 7.1577(4) Å, and R = 0.036 for a crystal (~20 × 100 μm) with approximately 2.2 wt% Ag 2 O. Refining Na ↔ Ag at the X site clearly indicates that Ag occupies this site. The X-O 2 distance of ~2.52 Å is slightly longer than tourmaline with ~(Na 0.6 □ 0.4 ), reflecting the slightly larger ionic radius of Ag compared to Na. Releasing the occupancy at the Y site gives ~Al 0.98 (~12.7 e.), which can be explained by occupation of Mg and Al. On a bond-angle distortion vs. <Y-O> distance diagram, the Ag-rich olenite- dravite lies approximately on the V site = 3 (OH) line in the figure, defining the relation between bond-angle distortion (σ oct 2 ) of the ZO 6 octahedron and the <Y-O> distance. No H could be found at the O1 site by refinement, in agreement with the Mg-Al disorder between the Y site and the Z site. Synthetic tourmaline contains no Ag when only AgCl is added; the compatibility of Ag in tourmaline, therefore, is largely a function of the F/Cl ratio of the fluid medium. A positive association of Ag at the X site with Al at the Y site and with F suggests that tourmaline might be useful for exploration in Cornwall-type polymetallic ore deposits associated with F-rich peraluminous granites or at other Ag-, F-, and B-enriched deposits such as Broken Hill, Australia. Preliminary electron microprobe analyses of tourmaline from Cornwall and Broken Hill, however, failed to detect Ag at the 3σ detection level of 0.08 wt% Ag 2 O.

Upper Cretaceous basic pyroclastic rocks, which overlie the Upper Pillow Lavas of the Troodos ophiolitic complex, Cyprus have been altered to bentonites. The resulting smectite is Fe-rich montmorillonite and Fe-rich beidellite, with moderate Mg contents. The smectite is trans-vacant and contains abundant exchangeable K. The presence of K is linked with hydrothermal alteration, which affected the higher members of the Troodos ophiolitic suite. The smectite displays significant compositional heterogeneity, which involves substitution of Fe for Al and to a lesser degree substitution of Mg for Al, and that reflects the influence of microenvironmental conditions on smectite formation. The layer charge of the smectite is controlled mainly by the tetrahedral charge, whereas the influence of octahedral charge is of lesser importance, because of Fe for Al substitutions, which does not create a charge deficit. Although the parent pyroclastic rocks were basic, the bentonites contain abundant Sipolymorphs and Si-rich zeolites, from dissolution of abundant radiolarian frustules, which increased the Si-activity of the pore waters, and also produced the partial replacement of smectite by palygorskite at a later stage. Dissolution of frustules was facilitated by the high heat flow from the ocean floor and by the circulation of hydrothermal fluids. The crystal chemistry of smectite and the bulk mineralogy of the bentonites influence the physical properties and industrial applications of the Cyprus bentonites, as well as their response to acid treatment.

The barium dioctahedral layer silicate, ganterite, was identified from the Lincoln Hill dumortierite deposit near Oreana, Nevada, based on electron microprobe, electron-backscatter diffraction (EBSD), and Raman spectrum microanalyses. This phase occurs with dumortierite, barite, and muscovite in a vein specimen formed by hydrothermal alteration. Back-scattered electron images of the muscovite from this locality show extensive zonation of the BaO content with regions of high Ba concentrations up to 15 µm in dimension. Electron microprobe analyses of these regions reveal a composition (Ba 0.53 K 0.37 Na 0.05 ) Σ=0.95 (Al 2.00 Ti 0.01 ) Σ = 2.01 [Al 1.51 Si 2.49 O 10 ](OH) 2 or, ideally, (Ba 0.5 K 0.5 )Al 2 (Al 1.5 Si 2.5 )O 10 (OH) 2 . This composition corresponds to the recently described mica, ganterite. Complete solid solutions between muscovite and ganterite were observed that range from 0.60% up to 18.12 wt% BaO. The electronbackscatter diffraction and Raman spectra of this phase are essentially indistinguishable from those of muscovite confirming that ganterite has a mica structure.

The transformation of pentlandite, (Ni,Fe) 9 S 8 , to violarite, (Ni,Fe) 3 S 4 , has been investigated under mild hydrothermal conditions, at constant values of pH (range 3 to 5) controlled by the acetic acid/sodium acetate buffer. At 80 °C, 20(4) wt% of the pentlandite transforms to violarite in 33 days; with the addition of small amounts of Fe 3+ (CH 3 COO) 2 (OH) and H 2 S the reaction reaches 40(4) wt% completion in this time. At 120 °C and a pressure of 3.5 bars the reaction is complete in 3 days at pH 3.9. Electron backscatter diffraction and backscattered electron imaging reveal that the reaction textures are typical of a coupled dissolution-reprecipitation reaction, rather than a solid state electrolytic process as has been previously reported. The gap between the dissolution front and the precipitation front of violarite is less than 400 nm. The violarite produced by these hydrothermal transformations is texturally similar to supergene violarite, being fine grained, porous and finely cracked.