Volume 92, Issue 1

Mid-infrared thermal emission spectra were acquired and are presented for 37 different sulfate minerals representing Strunz classes 6/A-D as well as a few other miscellaneous sulfate-bearing minerals (Strunz class 3/C and 8/J). Sulfate vibrational modes are assigned to each spectrum; also assigned are the modes of component OH, H 2 O, and carbonate where applicable. A discussion also is presented regarding the effect of hydration state on the emissivity spectra; dehydration of the Ca-sulfate mineral series (e.g., gypsum-bassanite-anhydrite), as well as the Mg-sulfate series, causes the high-frequency edge of the sulfate ν 3 band to shift to a larger wavenumber.

Leucite and analcime have open framework aluminosilicate structures, where ion exchange by cation substitution has been previously used to explain the replacement of one phase by another. Using 18 O-enriched NaCl solutions in hydrothermal reactions and run-product analyses using scanning electron microscopy, infrared and Raman spectroscopy, and time-of-flight secondary ion mass spectrometry, we show that the replacement of leucite by analcime is not a solid-state reaction involving cation exchange by volume diffusion. Textural features such as nano-pores and clusters, as well as the detection of high amounts of 18 O in the framework of analcime, suggest that the reaction proceeds by dissolution of leucite and reprecipitation of analcime, where structural O atoms of the leucite framework are exchanged and a new analcime structure forms at a moving interface through the leucite parent crystal. The characteristic high porosity (on a nano-scale) in the analcime product phase results from some of the parent phase being lost to the solution to give a volume deficit reaction. However, external dimensions are maintained during the process to result in the pseudomorphic replacement of an open framework aluminosilicate structure by a coupled dissolution-reprecipitation mechanism.

Spinel single crystals of four compositions along the MgAl 2 O 4 -MgMn 2 O 4 join, with Mn 3+ up to 0.25 apfu, were synthesized by use of a flux-growth method. The crystals were analyzed by electron microprobe, X-ray single-crystal diffraction, and optical absorption spectroscopy. Results revealed that Mg contents vary from 0.90 to 0.99 apfu, Mn 2+ ≤ 0.11 apfu, Mn 3+ varies from zero to 0.25 apfu, and Al-contents from 1.75 to 1.99 apfu. The unit-cell parameter increases linearly from 8.0883(3) to 8.1413(4) Å with increasing Mn 3+ content. The crystals show moderately disordered cation distributions, with i = 0.23(1), and different distribution trends have been observed: the Mg content is constant at the T site and is replaced by Mn 2+ at the M-site; Al decreases while Mn 3+ increases in T- and M-sites. Mn 3+ shows a preference for the T site, and a specific bond distance was refined, IV Mn 3+ -O = 1.88(1) Å. Unpolarized room-temperature single-crystal spectra reveal two relatively broad absorption features at ca. 23 000 and 10 800 cm -1 , which are assigned to spin-allowed d-d transitions in Mn 3+ located at octahedral and tetrahedral sites, respectively. The bond valence approach shows that the bonds are strained in the tetrahedron indicating underbonding in T, whereas the bonds are unstrained in the octahedron. To reduce the M-M repulsion, the steric effect is driven by the movement of the oxygen atoms, which improves the shielding effect around the M-site, thus increasing the distortion of the structure relative to the CCP. As a result the tetrahedron undergoes an isotropic expansion, which constrains the structure to incorporate larger cations such as Mn 3+ rather than smaller cations like Al at the T-site. This behavior, which is in disagreement with predictions based on crystal field energy considerations, illustrates the greater importance of steric factors on the cation distribution in spinels.

Simulations of high-resolution 19 F-decoupled 27 Al and 23 Na magic-angle spinning nuclear magnetic resonance (MAS NMR) spectra of the aluminofluoride minerals, cryolite, cryolithionite, thomsenolite, weberite, chiolite, prosopite, and ralstonite combined with theoretical modeling have given accurate values of chemical shift (δ iso ), and quadrupolar interaction parameters (C q and η), thereby eliminating ambiguities incurred by the complex nuclear interactions. These NMR data have been correlated with local electronic environments in the minerals, which were calculated using Full Potential Linearized Augmented Plane Wave (FP LAPW) modeling based on the structures from X-ray diffraction (XRD) data. This combination of NMR, XRD, and modeling techniques allowed the analysis and optimization of the crystal structures. The electronegativities and distances of neighboring ions, represented here by an environmental parameter χ, are shown to control δ iso of both 23 Na and 27 Al. The calculations using χ, also show that the ions beyond the nearest neighbor play an important role in determining δ iso of 27 Al and 23 Na in these aluminofluoride minerals, and the substitution of OH for F significantly affects the shielding around 27 Al in prosopite and ralstonite. There is a positive correlation between the site distortion at the Na and Al sites and the values of C q in these aluminofluoride minerals.

In-situ X-ray absorption spectroscopy at the Fe K-edge was used to characterize the local structural environment of Fe 2+ and Fe 3+ in silicate melts at high temperature (up to 1050 °C) in comparison to their quenched glassy analog at room temperature. Measurements were performed on binary alkali-silicate compositions and on haplogranitic compositions, which were doped with about 5 wt% Fe 2 O 3 . Changes in the structural environment of Fe were evaluated by analyzing both the pre-edge feature and the first maximum of the EXAFS of the spectra. In most cases, the spectra collected at high temperature differed from those of the quenched samples. At reducing conditions, the melts showed slightly higher amounts of low-coordinated Fe 2+ than their glassy counterparts. This finding is consistent with results of earlier studies (e.g., Jackson et al. 1993), but the observed change in speciation is smaller than reported by these authors. At oxidizing conditions, glasses and melts displayed a more heterogeneous behavior. The spectra of alkali-silicate compositions indicate higher amounts of low-coordinated Fe 3+ in the melt, whereas no significant difference between melt and glass was observed for Fe 3+ in haplogranitic compositions, even if the latter are peralkaline. The amount of non-bridging O atoms in the glass/melt system appears to play an important role particularly for Fe 3+ . However, more complex relationships between Fe and other structural components, especially Al, are possible.

To ensure the presence of coesite and its transformed polymorph, quartz, in ultrahigh-pressure (UHP) rocks and to examine the relic of the phase transformation, crystal structures were analyzed by single-crystal X-ray diffraction (XRD) directly using the rock thin section mounted on a slide glass. The rock sample used is a coesite-bearing eclogite from the Sulu UHP terrain, eastern China. The crystal structures were determined successfully by this new method and the presence of coesite and quartz in UHP rocks was identified for the first time by XRD. The R-factor [R(F)] converged to 0.046 for coesite and 0.087 for quartz. The displacement ellipsoids for coesite and quartz are larger than those previously reported for these two phases, which is consistent with expected effects of trapped strain due to the phase transformation from coesite to quartz during exhumation from the Earth.s mantle. This paper is the first report of single-crystal XRD of a rock thin section on a glass slide and establishes the technique, and provides proof-of-concept of the method. Although the mineral species included in a thin section can often be identified by other methods, such as Raman spectroscopy, an advantage of the reported method is that it can be applied to any mineral in a thin section, and not just to the UHP minerals. Moreover, it is applicable to an unknown or new mineral in a thin section, discarding the spots of known minerals and constructing a lattice from the residual spots to find the structure of the unknown phase.

Single-crystal structure determinations of the m = 17 and m = 16 antigorite polysomes confirmed the existence of two basic antigorite structures, i.e., “odd” antigorites, with m = 2n + 1 (m being the number of tetrahedra in a wavelength visible in the [010] projections), and “even” antigorites, with m = 2n. Both structures contain .8-reversals. and “6-reversals”. The “8-reversals” consist of 8-member rings of tetrahedra, where four tetrahedra point in one direction and four in the opposite direction, whereas “6-reversals” involve four tetrahedra and two tetrahedra, respectively. “Odd” antigorites have m tetrahedra and m − 1 octahedra along a wavelength (which coincides with the a translation) and Pm space group. Conversely, b/2 shifts every second 8-reversal cause the structure to be C-centered in “even” antigorites (C2/m space group). In the latter case, m tetrahedra and m − 1 octahedra occur within a wave, but two waves occur within an a periodicity. Transmission electron microscopy (TEM) of antigorite polysomes with m ranging from 15 to 18 is successfully interpreted on the base of previous X-ray data. In particular, [001] selected area electron diffraction (SAED) patterns show primitive or C-centered cells, for m = 2n or 2n + 1, respectively. In addition, [001] high-resolution (HR) images show 8-reversals aligned or offset by b/2 (i.e., stacking based on a primitive or C-centered lattice).

Electron energy loss spectroscopy (EELS) was used with scanning transmission electron microscopy (STEM) to quantify the average Mn valence in natural minerals at the nanometer scale. A method was developed to calibrate the energy-loss scale accurately, providing a comparison between STEMEELS and the X-ray absorption spectroscopy methods that investigate the L-edge chemical shift as Mn valence changes. The chemical-shift measurements were consistent with data reported by previous researchers from both X-ray and electron energy-loss spectroscopy. The L 3 /L 2 white-line intensity ratios also were consistent with previous work. A calibration curve for Mn valence was produced using the L 3 /L 2 white-line intensity ratios from measurements of synthetic standards. The average Mn valence was determined because it is not possible to distinguish Mn 3+ from mixtures of Mn 2+ and Mn 4+ using either method. The white-line intensity method was implemented in automated software that allows for rapid processing of point spectra, and 1-D and 2-D spectrum images. Point analyses of two natural pyrolusite samples indicated a Mn valence of 4.0, and point analyses of romanechite and manganite gave values of 3.8 and 3.4, respectively. An interface between braunite and bementite was used to illustrate 1-D and 2-D spectrum-imaging capabilities. The measured valence of Mn in the braunite and bementite was 2.9 and 2.0, respectively; both consistent with theoretical values. The braunitebementite sample demonstrated the heterogeneity of Mn valence common to natural minerals and the advantages of acquiring quantitative valence information in a known spatial context.

Previous attempts to refine the crystal structure of welshite, known only from Långban, Sweden, have been foiled by its extensive polysynthetic twinning, and without a structure determination, derivation of a reasonable formula has been difficult due its eclectic chemical composition. We report a successful refinement [wR(F 2 ) = 0.0566 for 8048 unique reflections] in the non-centrosymmetric space group P1 of a relatively little twinned crystal (4% chiral) from sample NRM040068: a = 10.394(3) Å, b = 10.777(3) Å, c = 8.896(2) Å, α = 105.953(4)°, β = 96.294(4)°, γ = 124.948(3)°, V = 738.8(3) Å 3 , Z = 1. The refined formula, Ca 3.78 Mg 7.87 Sb 3.00 Mn 1.35 Si 5.73 Be 3.33 Al 1.71 Fe 3+ 0.96 As 0.27 O 40 , is in reasonable agreement with a formula determined by electron microprobe and Mössbauer spectroscopy (Be from the refinement), Ca 3.81 Mg 7.84 Sb 3.03 Mn 1.17 Zn 0.04 Fe 2+ 0.05 Si 5.57 Be 3.24 Al 1.54 Fe 3+ 1.25 As 0.37 O 40 . The P1 symmetry of welshite 040068 is mainly a result of its cation distribution, which is driven by charge ordering on both the octahedral and tetrahedral sites as supported by calculation of electrostatic site potentials. Welshite can accommodate up to 3.46 Be per 40 O atoms by including several T sites with 100% Be occupancies without the formation of unfavorable tetrahedral Be-O-Be linkages that would result in the related centrosymmetric structures of sapphirine, khmaralite, makarochkinite, and høgtuvaite, none of which contain >2.1 Be per 40 O. A generalized formula for welshite is (Ca,Mn) 4 (Mg,Mn,Fe 2+ ,Fe 3+ ) 9 (Sb,Fe 3+ ) 3 O 4 [(Si,As) 6 (Be,Al) 4 (Al,Fe 3+ ) 2 O 36 ]. Compositions of most of the samples, including NRM040068, can be expressed in terms of two idealized Al and Fe end-members Ca 4 Mg 9 Sb 3 O 4 [Si 6 Be 3 Al 3 O 36 ] and Ca 4 Mg 9 Sb 3 O 4 [Si 6 Be 3 AlFe 2 O 36 ], respectively; the latter is the most representative for welshite and could be used in databases. One sample is far more heterogeneous and is irregularly zoned on scales from <1 to 10 μm. Compositional variation in this sample and in a crystal from another lie between the end-members Ca 4 Mg 8.55 Fe 0.45 Sb 3.0 O 4 [Si 5.15 As 0.45 Be 3.45 Al 0.5 Fe 2.45 O 36 ] and Ca 4 Mg 7.05 Fe 3.45 Sb 1.5 O 4 [Si 6.65 As 0.45 Be 3.45 Al 0.5 Fe 0.95 O 36 ], which are related by the coupled substitution MSb 5+ + VI Mg 2+ + IV Fe 3+ = M Fe 3+ + VI Fe 3+ + IV Si 4+ , where M refers specifically to the M7, M3A, and M4A octahedral sites. The lowest measured Sb is 1.454/40 O, which may be close to the limit imposed by the need to maintain local charge balance.

Rietveld refinements using synchrotron powder X-ray diffraction data were used to study the crystal structure and dehydration behavior of sepiolite from Durango, Mexico. The room-temperature (RT) sepiolite structure in air compares well with previous models but reveals an additional zeolitic H 2 O site. The RT structure under vacuum retained only ~1/8 of the zeolitic H 2 O and the volume decreased by 1.3%. Real-time, temperature-resolved synchrotron powder X-ray diffraction data and Rietveld refinements were used to investigate the behavior of the sepiolite structure from 300 to 925 K. Rietveld refinements revealed that most of the zeolitic H 2 O is lost by ~390 K, accompanied by a decrease in the a and c unit-cell parameters. Above ~600 K the sepiolite structure folds as one-half of the crystallographically bound H 2 O is lost. Rietveld refinements of the “anhydrous” sepiolite structure reveal that, in general, unit-cell parameters a and b and volume steadily decrease with increasing temperature; there is an obvious change in slope at ~820 K suggesting a phase transformation coinciding with the loss of the remaining bound H 2 O molecule.

Ab-initio thermophysical properties and isotopic separative effects in MgCr 2 O 4 were computed with the CRYSTAL03 code. All calculations were carried out within the LCAO approach (Linear Combination of Atomic Orbitals) by using the Hartree-Fock and the hybrid B3LYP density functional methods. Vibrational frequencies have been computed at the long-wavelength limit corresponding to the center of the Brillouin zone (k → 0). Since thermodynamic functions depend on the phonon dispersion relation, the ab-initio based vibrational analysis of the optical modes has been coupled with the semiempirical Kieffer model for the acoustic modes, which is based on a sine wave dispersion equation. All computed magnitudes (zero point energy, internal energy, Helmholtz free energy, optical vibrational modes, thermal corrections to internal energy, enthalpy and Gibbs free energy, thermal pressure, bulk modulus and its P and T derivatives, entropy, isochoric and isobaric heat capacity, and isotopic separative effects) are consistent with direct observations and/or reasonable estimates, within the accuracy of the method. The computed separative effects are similar to those observed in other Cr 3+ -bearing substances. The crystal field seems to have no influence on the separative effect for an isotopomeric couple with respect to an isolated gaseous couple with Cr 3+ as central cation.

The single-crystal elastic properties of Ca 0.07 Mg 1.93 Si 2 O 6 orthopyroxene (space group Pbca) have been investigated by Brillouin spectroscopy at ambient conditions. The aggregate bulk and shear moduli, K 0,S = 102.5 GPa (1.5) and μ = 74.2 GPa (1.1), respectively, are ~5% and ~3% lower than commonly accepted values for MgSiO 3 end-member (K 0,S = 107.6, μ = 76.8 GPa). These results indicate that the incorporation of small amount of Ca in the orthoenstatite structure does not greatly affect its elastic properties. As a consequence, the increase in bulk modulus reported in natural orthopyroxenes relative to the Mg-end-member is not related to the substitution of Ca in the M2 octahedral sites, but more probably to the substitution of Al in tetrahedral sites.

The thermodynamic properties of crystalline and liquid KAlSi 3 O 8 are used to calculate the fusion curve of sanidine to 6.5 GPa. New values for the enthalpy and entropy of fusion of sanidine at one bar and 1200 °C (ΔH Tf = 63.0 kJ/mol, ΔS Tf = 42.8 J/mol-K) are recommended on the basis of improved heat-capacity equations for KAlSi 3 O 8 crystal, glass, and liquid. On the basis of phase-equilibrium experiments on the congruent melting reaction between 2 and 6.5 GPa, the pressure dependence of the liquid compressibility (K 0 ’ = dK 0 /dP, where K 0 = 1/β 0 ) is constrained to be 12.2 ± 1.0 in a thirdorder Birch-Murnaghan equation of state (EOS). The metastable, one-bar melting temperature (T f ) is additionally constrained to be 1203 ± 26 °C. Determination of the liquid K0’ allows the density and compressibility of KAlSi 3 O 8 liquid to be calculated to 6.5 GPa (2.709 ± 0.014 g/cm 3 at 1600 °C). The uncertainty in K 0 ’ of ±1.0 leads to an error in melt density at 6.5 GPa of ±0.52%. With a K 0 ’ = 12.2, the relatively high compressibility of KAlSi 3 O 8 liquid at 1600 °C (K 0 = 15.8 GPa) drops rapidly with increasing pressure. The dominant mechanism of compression for KAlSi 3 O 8 liquid between 0 and 6.5 GPa most likely involves topological changes and increases in network connectivity with pressure. It is probable that highly compressible liquids, such as hydrous, silica-rich liquids formed by partial melting of a subducted slab, may have K 0 ’ values that exceed 12 (at pressures ≤6.5 GPa).

We report data that allow us to evaluate the method of trapping immiscible, saline aqueous fluids (i.e., vapor and brine in the NaCl-KCl-HCl-FeCl 2 -AuHCl 2 -H 2 O system) as synthetic fluid inclusions in pre-fractured quartz cores in order to quantify the concentrations of Au, Fe, K, and Na, among coexisting three-phase, immiscible fluids (i.e., haplogranite melt, brine, and vapor) at magmatic conditions. Coexisting vapor and brine were trapped experimentally at 800 °C and 100-110 MPa as synthetic fluid inclusions in both quartz microfractures and quenched silicate melt (i.e., glass), and also sampled indirectly using the recovered quenched aqueous fluid. Quartz-hosted and glass-hosted brine inclusions were analyzed by laser-ablation inductively-coupled-plasma mass spectrometry (LAICPMS) and instrumental neutron activation analysis (INAA), respectively. Quenched aqueous fluid from each experiment containing a quartz core was recovered and analyzed by atomic absorption spectrophotometry (AAS). The composition of aqueous fluids trapped as quartz-hosted inclusions, glass-hosted inclusions, and those recovered after quench yield consistent and precise data, at the 2σ uncertainty level, for the elements of interest. The overlapping Au, Fe, K, and Na concentrations in aqueous fluids trapped and analyzed via three entirely different instrumental techniques (i.e., LAICPMS, INAA, and AAS) suggest strongly that quartz microfractures heal on a slow enough time scale to permit entrapment of fully equilibrated aqueous fluids at our experimental PTX conditions. The data evince clearly that the chemical composition of fluids in quartz microfractures at the time of self-healing represents equilibrium conditions; hence, synthetic fluid inclusions in experiments with low thermal gradients across the charge provide a reasonable estimate of fluid composition at least at the experimental conditions examined in this study.

The stability of Mg-Fe-Al biotite has been investigated with reversed phase-equilibrium experiments on four equilibria. Experimental brackets in pure H 2 O and H 2 O-CO 2 mixtures for the equilibrium: phlogopite + 3 quartz = enstatite + sanidine + H 2 O (1) are in good agreement with previous experiments in mixed-volatile fluids (Bohlen et al. 1983) and H 2 O-KCl solutions (Aranovich and Newton 1998), while indicating a reduced stability field for phlogopite compared to previous data in pure H 2 O (Wood 1976; Peterson and Newton 1989). Aluminum solubility in biotite has been determined in the Fe-, Mg-, and Fe-Mg systems from reversed phase-equilibrium data for the equilibria: 3 eastonite + 6 quartz = 2 phlogopite + 3 sillimanite + sanidine + H 2 O (2) 3 siderophyllite + 6 quartz = 2 annite + 3 sillimanite + sanidine + H 2 O (3) over the P-T range ~600.750 °C and 1.1.3.4 kbar. Over the investigated temperatures, the brackets define nominal Al saturation levels of 1.60 ± 0.04 in Mg-biotite, 2.08 ± 0.05 in Fe-biotite, and 1.81 ± 0.03 in biotite with Fe/(Fe + Mg) = 0.43.0.44. The slight decrease in Al with increasing T and decreasing P suggested by the data is less than experimental uncertainties. Compared to biotite on the Phl.Ann join, Al-saturated biotites have a markedly larger stability field, particularly in the Fe-system. This effect has been quantified in the Fe-system with one reversal between 691.709 °C at 2.4 kbar for the equilibrium: biotite + sillimanite + quartz = almandine + sanidine + H2O (4) The combined experimental results place tight constraints on the thermodynamic properties of phlogopite, annite, eastonite, and siderophyllite. The resulting nonzero (ΔH298 = .9.4 kJ/mol, with ΔS = ΔV = 0) energetics for the internal equilibrium: Eastonite + 2/3 Annite = 2/3 Phlogopite + Siderophyllite (5) reflect strong Fe-Al affinity in biotite, which has a marked effect on thermobarometers involving biotite.

Experiments in the system K 2 O-(NH 4 ) 2 O-Al 2 O 3 -SiO 2 -H 2 O-HCl have been conducted between temperatures of 400 and 600 °C at pressures of 200, 400, 500, and 1500 MPa. The run products consisted of solid solutions of the K-feldspar.-buddingtonite, muscovite- tobelite series, and quartz. The run products were characterized by electron microprobe, powder X-ray diffraction with Rietveld analysis, and infrared spectroscopy. It can be shown that complete solid solution exists for both series for temperatures at least above 400 °C. The grain size of the synthesized phases was usually small and rarely exceeded 10 μm. In addition the micas were quite thin and the thickness was lower than 100 nm. Due to the small grain size and the (NH 4 ) 2 O-content, reliable electron-microprobe results were difficult to obtain. Rietveld analyses did show that the synthesized micas consisted of a complex mixture of 1M, 2M1, 2M2, 3T, and 2Or polytypes. The lattice parameters of the feldspar and mica solid solutions are linear combinations of the end-members and show practically no excess volume. Reasonable compositions of the solid solutions can be derived from the lattice parameters. The mica solid solutions are not binary mixtures but at elevated pressure increasing amounts of pyrophyllite component were observed. By using the NH 4 -deformation mode at around 1430 cm -1 and the OHstretching vibrations ν OH at 3600-3700 cm -1 the NH 4 -contents of the micas could be determined also by IR spectroscopy.

Melanophlogite, a clathrasil, possesses a framework of corner-linked silica tetrahedra forming framework cavities that can enclose small guest molecules. Synchrotron X-ray diffraction experiments of the guest-free melanophlogite have been conducted at pressures up to 12 GPa and temperatures up to 1473 K. Upon compression at room temperature, melanophlogite gradually lost its crystallinity and became completely X-ray amorphous at ~8 GPa. The amorphization process was similar to those of denser silica polymorphs, but it reached completion at a much lower pressure (e.g., quartz becomes X-ray amorphous at ~30 GPa). The decreased amorphization pressure of melanophlogite may be attributed to its lower framework density and the ease of bending of its Si-O-Si linkages, thereby accelerating the collapse of the structure under high pressure. Determination of cell volumes of melanophlogite prior to its amorphization yielded a room-temperature bulk modulus of 26.3 ± 1.7 GPa, which is consistent with the relatively large compressibilities reported for the structurally similar zeolites. When heated at ~8 GPa, the amorphous phase started to crystallize at 873 K into coesite, the stable silica phase at these pressure and temperature conditions. Thus the occurrence of pressure-induced amorphization in melanophlogite appears to result from the kinetic hindrance to its transformation to the thermodynamically stable coesite.

We present here a first principle computational study of the effect of cation disorder on the elasticity of spinel. Our calculated elastic moduli and structural parameters are comparable with reported experimental and theoretical results. We find that bulk modulus for MgAl 2 O 4 spinel does not soften at pressure as high as 27 GPa, while shear modulus c S [ ½ (c 11 - c 12 )] decreases with pressure. Disorder increases both the bulk modulus and shear modulus in aluminate spinel, but decreases both for the silicate spinel. The elastic properties of ringwoodite are significantly affected by Si-Mg disorder; a 10% disorder decreases the seismic velocities by 3.5%. Thus, a small amount of disorder will significantly affect seismic observations.

A smectite and a rhyolitic obsidian specimen were subjected to hydrothermal alteration for 4 months at 150 °C in two solutions having different AlCl 3 concentrations. The solids were analyzed by means of X-ray diffraction, thermogravimetry, electron microscopy (with microanalysis), and Fourier transform infrared spectroscopy. The solutions were chemically analyzed and the ion activities were plotted on activity diagrams. They are broadly coherent with the reaction products. The smectite specimen produced mixed-layer kaolinite-smectite (K-S), with a higher kaolinite proportion for the solution with higher Al content. The glass specimen produced glass + boehmite + kaolinite in the low-Al solution, and kaolinite in the high-Al solution. Thus, K-S probably forms only from smectite precursors. The formation of K-S follows the same mechanism of stripping of one of the tetrahedral sheets from smectite described for natural specimens by Dudek et al. (2006). Comparison of the synthetic and natural K-S indicates great similarity, although the arrangement of the kaolinite layers in the synthetic specimens is more ordered and some of their kaolinite infrared bands are more intense. The results of Dudek et al. (2006) and this study show a wide range of chemical and structural “maturity” of kaolinite in K-S, as kaolinization proceeds, that varies depending on the reaction conditions.

The crystal structure of tooeleite, Fe 6 (AsO 3 ) 4 (SO 4 )(OH) 4 ·4H 2 O, has been solved from high-resolution synchrotron XRD powder data recorded on a sample from Tooele county, Utah. The structure is monoclinic, space group C2/m, a = 8.9575(1), b = 6.4238(1), c = 9.7912(1) Å, β = 96.032(1)°, V = 560.27(3) Å 3 , d calc = 3.16 g/cm 3 . The structure was solved by direct methods and atomic positions, site occupancies, and isotropic displacement parameters were refined by the Rietveld method. The AsO 3 pyramids bond to FeO 6 octahedra by both edge- and corner-linkage, forming layers that intercalate SO 4 groups. Assignment of structural H 2 O and OH groups were done from bond-valence analysis. Tooeleite is the only arsenite-sulfate mineral known and has been recently identified as the main constituent of stromatolite-like deposits in the Carnoulès acid mine, Gard, France.

Zincalstibite, a new mineral occurring within the cavities of marble of the Apuan Alps, Tuscany, Italy, has chemical formula Zn 2 AlSb(OH) 12 , space group P3̄, a = 5.321(1), c = 9.786(2) Å. It is associated with sub-millimeter tufts of white crystals of mimetite and sub-millimetric stalactite aggregates of opal and an amorphous copper-silicate phase (possibly crisocolla). The crystals are trigonal prismatic, with forms {110}, {001}, elongated [001], generally less than 10 × 10 × 40 ÷ 50 μm, with few larger crystals. They are colorless, transparent, with vitreous luster, white streak and {001} cleavage. The stronger reflections are [(hkl), d (Å), I rel ]: (002), 4.904, 100; (100), 4.620, 35; (101), 4.179, 57; (103, 110), 2.669, 31; (112, 112̄), 2.343, 88; (114, 114̄), 1.805, 57. Zincalstibite is structurally related to cualstibite, as evidenced by the structural determinations and refinements we present for both the minerals. They are built up by layers of isolated Sb(OH) 6 octahedra alternating along c with trioctahedral layers, which contain Zn and Al cations and Cu and Al cations in zincalstibite and cualstibite, respectively. In cualstibite the ordering of Al and Cu within these trioctahedral layers results in a supercell, with a cual = 9.150(2) Å ≈ √3 a zinc . The name of zincalstibite is related to its chemical composition and points to its relationships with cualstibite. Both the mineral and its name were approved by the IMA Commission for New Minerals and Mineral Names (IMA 1998-033).

Keilite, (Fe,Mn,Mg,Ca,Cr)S, from the Zakłodzie enstatite achondrite is described. Forming xenomorphic grains up to 0.5 mm in diameter, the keilite is associated with troilite (or pyrrhotite), Fe-Ni metal, an (Fe,Zn,Mn)S phase, enstatite (with relict forsterite in cores), plagioclase and accessory schreibersite, oldhamite, graphite, sinoite, and an SiO 2 polymorph. It is brittle and possesses a good cleavage similar to that of galena, parallel to (001), (010), and (100). X-ray diffraction structural data reveal the following: cubic space group Fm3m, α = β = γ = 90°, a = 5.1717 (18) Å, unit-cell volume V = 138.32 (8) Å 3 ; D = 3.958 g/cm 3 ; Z = 4. The chemical formula based on 63 electron microprobe point analyses is: (Fe 0.437 , Mn 0.356 , Mg 0.160 , Ca 0.017 , Cr 0.019 , Zn 0.001 )S 1.008 . Compared with previously described keilites from enstatite chondrites, the Zakłodzie keilite is richer in (Mn,Ca,Cr)S and poorer in the Fe- and Mg-end-members and, consequently, it is nearer to alabandite and oldhamite. This is the first detailed description of keilite from a meteorite that is not an enstatite chondrite.

The 17 O NMR spectra of glasses quenched from melts at high pressure are often difficult to interpret due to overlapping peaks and lack of crystalline model compounds. High-pressure aluminosilicate glasses often contain significant amounts of [5] Al and [6] Al, thus these high-pressure glasses must contain oxygen bonded to high-coordinated aluminum. The 17 O NMR parameters for the minerals jadeite, pyrope, grossular, and mullite are presented to assist interpretation of glass spectra and to help test quantum chemical calculations. The 17 O NMR parameters for jadeite and grossular support previous peak assignments of oxygen bonded to Si and high-coordinated Al in high-pressure glasses as well as quantum chemical calculations. The oxygen tricluster in mullite is very similar to the previously observed tricluster in grossite (CaAl 4 O 7 ) and suspected triclusters in glasses. We also present 27 Al NMR spectra for pyrope, grossular, and mullite.

Carbonado diamonds from the Central African Republic were investigated using spectroscopic observations and C-isotopic analysis. Based on photoluminescence (PL) spectra, carbonado samples were classified into two groups: Group-A, which exhibits an intense PL band at 504 nm; and Group- B, which exhibits PL bands at 504, 575, and 638 nm at room temperature. PL spectra measured at 120 K gave well-resolved side-band structures of 504 nm bands. Consequently, the 504 nm band of Group-A can be assigned to the 3H center attributable to self interstitials in diamond, whereas the 504 nm band from Group-B can be assigned to the H3 center attributable to a vacancy (□) trapped at nearest-neighbor substitutional nitrogen (N) pairs. The PL band at 575 nm, which is attributable to neutral N-□ pairs, is known to increase its width with increasing residual stress in diamond aggregates. The average FWHM of the 575 nm band and the standard deviations were 3.80 and 0.54 nm for Group-A carbonado, and 2.80 and 0.38 nm for Group-B carbonado. These values suggest that Group-A carbonado samples have higher residual stress than the Group-B samples. The presence of an H3 center and the lower residual stress in Group-B carbonado both suggest that they originated from higher temperatures compared to Group-A. Radiation halos were observed in cathodoluminescence (CL) images of both Group-A and Group-B samples. The CL halos are traces of radiation damage from radioactive nuclides. The texture of the haloes suggests that the radiation damage was a secondary event after formation of the carbonado diamonds. The average C-isotopic composition of Group-A is -23.6 ± 0.52. in δ 13 C PDB and that of Group-B is -26.3 ± 0.65‰. Group-A carbonado was enriched systematically in 13 C compared with Group-B carbonado. Isotopic fractionation might occur as a result of the different thermal history of carbonado.

Li 0.52 Mg 0.96 Sc 0.52 Si 2 O 6 orthopyroxene was synthesized by the flux method and its structure studied with single-crystal X-ray diffraction. The crystal is orthorhombic with space group Pbca and unit-cell parameters a = 18.259(5), b = 8.883(2), c = 5.271(1) Å, and V = 854.9(3) Å 3 . The structure refinement shows that the M1 and M2 sites are occupied by (0.48 Mg + 0.52 Sc) and (0.48 Mg + 0.52 Li), respectively. While the O3-O3-O3 kinking angle (165.40°) of the silicate tetrahedral A chain appears to be normal when compared with reported data, the kinking angle (151.92°) of the B chain is the largest of all orthopyroxenes examined at ambient conditions. This is the first orthopyroxene structure that contains more than 50% trivalent and monovalent cations in the M1 and M2 sites, respectively, and displays a kinking angle of the tetrahedral B chain that is greater than 150°. Our study demonstrates the stability of the new pyroxene structure type predicted by Pannhorst (1979) at room temperature.

To study the ambient analog of the deep mantle MgSiO 3 perovskite to post-perovskite phase transition, high-temperature drop calorimetry experiments of perovskite and post-perovskite phases of CaIrO 3 system as well as high-pressure phase equilibrium experiments in the CaIrO 3 system were made. The enthalpies for dissociation of CaIrO 3 (298 K) to CaO + Ir + O2 (1573 K) were 486.7 ± 9.2 kJ/mol for post-perovskite and 454.5 ± 12.5 kJ/mol for perovskite. From the difference between them, the phase transition enthalpy from perovskite to post-perovskite at 298 K is -32.2 ± 15.5 kJ/mol. This gives 2.7 ± 15.6 kJ/mol as formation enthalpy of CaIrO 3 perovskite from CaO + IrO 2 at 298 K. Using the phase transition enthalpy and volume change of .0.48 ± 0.02 cm 3 /mol determined in this study, the phase equilibrium boundary is calculated to be P (GPa) = 0.040 T (K) - 67.1. The strongly positive slope agrees with that obtained in high-pressure experiments. This is consistent with a large positive Clapeyron slope of post-perovskite phase transition in MgSiO 3 recently reported from experimental and theoretical studies.