Volume 88, Issue 2-3

The high-pressure behavior of rocksalt-structured FeO has been investigated using the LDA + U method, a first-principles computational technique that allows treatment of correlated electrons with strong localized repulsions. Within the local density approximation (LDA) FeO is predicted to be a metal, but with LDA + U, an insulating state is obtained at zero pressure. Electronic and magnetic behavior, the equation of state, and lattice strain are determined for three values of the Coulomb repulsion U. We find two self-consistent solutions, one with rhombohedral and one with monoclinic electronic symmetry. For U = 4.6 eV, the monoclinic solution becomes more stable than the rhombohedral solution at 110 GPa, leading to an insulator-metal transition; with increasing U, metallization occurs at higher pressures. Results from the LDA + U calculation suggest that the high-spin magnetic state should persist to pressures greater than 300 GPa. The method gives improved agreement with experiments for ground state properties as compared to LDA and GGA methods that do not explicitly include a local Coulomb repulsion.

A micro-infrared spectroscopic study of hydrothermal vein quartz known to have anomalous δD signatures has identified two hydrogen reservoirs. In samples that generate an isotopic signature in accordance with that anticipated under the accepted model of quartz crystallization, submicroscopic aggregates of liquid water are the dominant hydrous species. Samples which generate an anomalous δD signature contain, in addition to liquid water, structurally incorporated hydrous species associated with impurity cations. Infrared spectra obtained during in situ stepped heating experiments, coupled with infrared analysis at 25 ℃, demonstrate that hydrogen liberated between 300 and 500 ℃ is chiefly molecular, liquid water. Hydrogen liberated at temperatures greater than 500 ℃ is dominantly that associated with specific structurally incorporated cation defects. Since both defect hydrogen and molecular water are contemporaneously incorporated from the precipitating medium during crystallization, we propose that irregular δD signatures, released following decrepitation at temperatures greater than 500 ℃, are due to isotopically fractionated hydrogen released from interstitial OH defect sites in the quartz structure. δD signatures obtained from stoichiometrically anhydrous minerals are generally interpreted under the assumption that the hydrogen measured comes uniquely from decrepitated fluid inclusions. Instead, we suggest that δD ratios obtained from hydrothermal quartz reflect a sum of the contributions made by individual hydrogen reservoirs, each with a potentially distinctive δD signature. Thus, if the overall δD signature is attributed entirely to fluid inclusion phases, the nature of the precipitating fluid may be misinterpreted. Hydrogen extracted as molecular water at between 300 and 500 ℃ provides a true reflection of the hydrothermal solution associated with crystal precipitation.

Tuzlaite, a hydrated pentaborate from the Tuzla evaporite deposit in Bosnia and Herzegovina, was analyzed for water content and loss upon heating using thermal analysis methods and vibrational spectroscopy. The resulting phases were identified by X-ray diffraction. The heating of tuzlaite results in a gradual loss of water over several dehydration steps. Two coordinated H 2 O molecules escape at 191 ℃. Between 248 and 298 ℃ two hydroxyl groups are eliminated, with an associated structural transformation. A continuous escape of the third water molecule occurs above 300 ℃. A phase relationship model for different hydrated borate minerals implies the possible formation pathway of tuzlaite by sequential polymerization of the borate polyanion.

The response of the laumontite crystal structure to hydration/dehydration was evaluated using Rietveld refinements with XRD data collected under controlled P H₂O conditions at ~28.5 ℃. Refined water contents per unit cell (unit-cell formula: Ca 4 Al 8 Si 16 O 48 ·nH 2 O) ranged between 12.5 H 2 O at 0.11 mbar P H₂O and 17.3 H 2 O at 37.6 mbar. The occupancy of the two awater sites hosting hydrogenbonded water molecules, W5 and W1, ranged from 13% to 100% and from 2% to 86%, respectively. During hydration of W5, between 0.11 and 5 mbar, the unit cell expanded continuously and reversibly from 1327 to 1348 Å 3 . The unit-cell volume remained nearly constant between 5 and 28 mbar. The hydration/dehydration of W1 exhibited hysteresis; hydration occurred at ~29 mbar and dehydration at ~24 mbar. During hydration of W1 at ~29 mbar the unit cell expanded from 1351 to 1384 Å 3 . Further hydration of W1 above 29 mbar resulted in gradual and reversible unit-cell expansion to 1386 Å 3 at 37.6 mbar. Hydration/dehydration of W5 is a continuous reaction typical for zeolites. In contrast, the hydration/dehydration of W1 at room temperature is discontinuous, as manifested by the presence of two laumontite phases during hydration and dehydration. Unit-cell parameters of the two coexisting laumontite phases observed under these conditions are consistent with a vacant W1 site and ~80%-occupied W1 site, respectively. Gradual unit-cell expansion above 29 mbar due to increased P H₂O and increased occupancy of W1 indicate that hydration of the remaining 20% of the W1 site proceeds continuously.

A hydrothermal diamond-anvil cell has been designed for the in situ analysis of the concentration of elements in fluids and melts using synchrotron radiation X-ray fluorescence. This technique permits determination of the solubility of minerals in fluids at high pressures and temperatures (tested to 1.1 GPa and 800 ℃). Further advantages include multi-element analytical capability and applicability for sparingly soluble compounds, for congruent or incongruent dissolution, and for studies of dissolution kinetics. The solubility of AgCl(s) in water was determined at 300 to 450 ℃ and pressures to 760 MPa. The molality of silver in the fluid was obtained from density corrected Kα peak areas by calibration to the known solubility at 300 ℃ and vapor pressure and, standardless, by comparison with Monte- Carlo simulated spectra. The results from both methods were internally consistent and in good agreement with literature data for comparable P-T conditions.

A single crystal high-pressure diffraction study in a diamond-anvil cell shows that monoclinic malayaite (CaSnOSiO 4 ) transforms into a triclinic high-pressure polymorph at P c = 4.95(1) GPa. No discontinuity was observed for the individual crystallographic axes or the volume compressibility. Instead, the A2/a - A1- phase transition reveals itself by significant deviations of the α and γ angles from 90 ℃. The bulk elastic properties of the triclinic phase cannot be distinguished from those of the monoclinic structure within experimental uncertainty (A2/a: V 0 = 389.68(3) Å 3 , K 0 = 121(1) GPa, and K' = 4.2(5); A1̅: V 0 = 390.3(1) Å 3 , K 0 = 118.3(7) GPa, K' = 4). Fitting the compressibility to all data gives values of V 0 = 389.64(3) Å 3 , K 0 = 121.6(7) GPa and K' = 4.6(2). Structure refinements at four different pressures reveal the structural details of the monoclinic A2/a and triclinic A1̅ phases. Below the transition temperature the SiO 4 polyhedra show some non-rigid distortion, whereas the SnO 6 polyhedra remain almost unchanged. At the phase transition, the SiO 4 tetrahedra show further angular twisting while the SnO 6 chains shift parallel to [1 - 01], inducing a reduction in symmetry. Furthermore, at pressures above 5 GPa the Ca atoms are displaced almost parallel to [100], causing a change in coordination from CaO 7 polyhedra to sheets of CaO 8 parallel to (1̅11). At 7.394(4) GPa the cell parameters of the triclinic structure are a = 6.9958(4) Å, b = 8.8080(9) Å, c = 6.4968(4) Å, α = 89.078(7)°, β = 112.745(5)°, γ = 91.230(7)° , V = 369.10(5) Å 3 ; space-group A1̅.

We performed new diffraction experiments to clarify the equation of state (EoS) of stishovite after we suspected systematic errors in previous experimental reports. Using diamond anvil cells, we repeated both single-crystal X-ray diffraction measurements under hydrostatic conditions and powder diffraction measurements using the laser-annealing technique and NaCl pressure medium. The major improvement is the increase in precision of the pressure determination using the quartz and NaCl equations of state. Using both sets of data, the stishovite bulk moduli were refined to K 0 = 309.9(1.1) GPa and K' 0 = 4.59(0.23). We also reinvestigated the mechanism of the phase transformation to the CaCl 2 -structured polymorph of SiO 2 at about 60 GPa. We confirm no volume discontinuity at the transition pressure, but the CaCl 2 form appears slightly more compressible than the rutile-structured form of SiO 2 . This change in compression behavior is used for quantitative analyses of the spontaneous strains of the pressure-induced phase transition.

Kinetics and reaction paths of Fe 3+ reduction by H 2 in high-Fe and low-H 2 O silicate melts have been investigated at 800 ℃. Time-series experiments were performed in cold-seal pressure vessels at 50 bars of pure H 2 using rapid-heating and rapid-quench strategies. Within the first minutes of the experiments, a fast partitioning of Na occurred between the gas and the melt due to the reducing conditions. Kinetically decoupled from the Na partitioning, the progression of a front of Fe 3+ reduction within the quenched melt was observed and was identified as a diffusion-limited process. The growth of the reduced layer is accompanied by an increase in concentration of OH-groups suggesting that reduction operates through proton incorporation within the melt. As this growth rate is slightly faster than predicted from the diffusion of molecular H 2 O, a different and mobile waterderived species seems likely. One possible mechanism is the reduction of Fe 3+ by the transport of molecular H 2 . As this process is limited by the flux of H 2 , it will depend on both diffusivity and solubility of H 2 in the melt. Alternatively, migration of protons (H + ) and electronic species within the melt could control the velocity of the reduction front. The increase in concentration of the reactionderived OH groups produces a water over saturation followed by partial dehydration of the melt. This dehydration leads to a change in the redox conditions within the gas that influences the Na partitioning between gas and melt.

New alkali feldspar/felsic magma trace-element partition coefficients (D-values) for Rb, Sr, Ba, Eu, Y, Zr, Nb, Ga, Zn, trivalent REE, Be, Cs, Hf, Pb, Th, and U for 30 samples of peralkalic quartz trachyte and rhyolite are presented. D-values of incompatible elements vary systematically with melt polymerization parameters, increasing with whole-rock silica, but decreasing in rocks with higher agpaitic indices [A.I. = mol (Na + K)/Al]. D-values for Sr and Ba (evaluated to be accurate) vary systematically with crystal chemistry, probably substituting for Na in the Ca-poor alkali feldspar phases. Apparent D-values for Sr and Ba from pre-Quaternary systems are fraught with contamination and analytical errors, respectively, and should be used with caution. D Eu decreases exponentially with A.I., ranging from compatible in weakly peralkalic (A.I. < 1.1) rocks to strongly incompatible in very peralkalic (A.I. > 1.2) rocks. These regular variations strongly suggest that partition coefficients for these elements may be predicted accurately if whole-rock and crystalchemical parameters are known.

Multivariate linear regression (MLR) analysis has been used to develop equations for predicting alkali feldspar/melt trace-element partition coefficients (D-values) for certain large-ion lithophile elements (Rb and Eu), high field-strength elements (Zr and Nb), and rare-earth elements (Y, La, Ce, Pr, Nd, Sm, Eu, Gd, Dy, Er, and Yb) in peralkalic felsic systems. A partitioning equation for Sr based solely on data from Pantelleria is also presented. D Rb , D Zr , and D Nb each can be calculated given whole-rock SiO2 and agpaitic index [A.I. = mol (Na + K) / Al]; alternatively, D Rb can be calculated given feldspar Or content and agpaitic index. D Eu can be calculated given whole-rock SiO 2 . agpaitic index, and whole-rock Na 2 O/K 2 O. Other DREE can be calculated given whole-rock SiO 2 , agpaitic index, and whole-rock CaO. These D-values all increase with increasing whole-rock SiO 2 and decreasing agpaitic index. D Sr for samples from Pantelleria increase with increasing feldspar Ab and can be calculated given feldspar Ab content. The equations formulated will help to constrain better the choice of D-values used in petrogenetic models of peralkalic systems

Mineral compositions and reaction textures found in migmatitic gneisses from near Gilsum, New Hampshire, constrain both peak metamorphic pressure (P) and temperature (T) conditions and the PT path. Large K-feldspar porphyroblasts indicate isobaric heating of the samples at P < 4 kb (Spear et al. 1999), and relict cordierite + garnet assemblages record the occurrence of biotite vapor-absent melting; garnet-biotite thermometry yields peak temperatures of 740 ℃ at 3.5 kbar. During melt crystallization, low-Ti biotite + sillimanite replaced cordierite, and production of muscovite indicates P > 4 kb on the cooling path. Four generations of monazite have been identified, three of which have been linked to specific whole-rock reactions. Monazite (4) (the last generation) was produced with xenotime ± apatite during melt crystallization and consumption of garnet and cordierite. Monazite (3) grew in a xenotime-absent mineral assemblage as garnet + muscovite reacted to form sillimanite + biotite. Monazite (2) grew in a xenotime-bearing ± garnet + biotite + chlorite assemblage, as xenotime and chlorite were consumed during garnet production. Monazite (1) has not been linked to a specific reaction; it may be detrital, record an earlier metamorphic event, or represent disequilibrium overgrowths of xenotime. YAG-monazite, YAG-xenotime, and monazite-xenotime thermometry for monazite generations (2)-(4) yield temperatures consistent with major-phase thermometers.

Biotite was oxidized in deep saprolitic weathering profiles developed on granitic rocks in a humid temperate climate in the Youngju-Andong area of South Korea. The mineralogy and chemistry of these oxidized biotites were characterized by chemical analysis, electron microscopy, X-ray diffraction, thermal analysis, and radiogenic Ar analysis. The results showed that a decrease in the b o -dimension, loss of radiogenic Ar, and formation of vermiculite are fairly well correlated with the degree of oxidation of ferrous iron. The chemical composition of oxidized biotite was modified by a non-stoichiometric removal of interlayer and octahedral cations to compensate for the charge imbalance induced by oxidation of Fe. The pervasive loss of cations and radiogenic Ar suggests their diffusion through oxidizing biotite in a non-expanded state. Iron oxidation and cation loss caused a decrease in the b o -dimension with the formation of discontinuities that acted as conduits for the weathering solutions, resulting in partial vermiculitization (<10%). The Fe oxidation was nearly completed in the lower part of the profile, concomitant with mineralogical and chemical modification to oxidized biotite that persists throughout the profile without further notable modification. Cation release from biotite is governed in early stage by the formation of oxidized biotite, and later by its decomposition.

The rate and mechanism of zircon coarsening in quartzite ±H 2 O at 1.0 GPa and 1000 ℃ were characterized by performing time-series experiments in a piston-cylinder apparatus. Anhydrous experiments show no significant growth of zircon or textural modifications after 24 h. In the presence of 1-2 wt% H 2 O, zircon (2 wt%) in coarsely powdered quartz recrystallized in the first 8 hours so that mean crystal size decreased relative to the starting material. After 8 h zircon grew by Ostwald ripening and by coalescence of crystals, and maintained its position on quartz grain boundaries, even while quartz crystals coarsened, by recrystallizing in the fluid. Fitting the experimental data produced the growth-rate equation log (<D> n - <D 0 > n ) = log (<D> 2 - <D 0 > 2 ) = log (t - t 0 ) + log K n = log (t - t 0 ) - 5.9, where <D> is mean diameter in micrometers, the subscript 0 denotes initial value at time zero, t is experiment duration in seconds, and K n is the growth-rate constant with units of μm 2 /s (five experiments, r 2 = 0.988). A value of n = 2 suggests that growth is interface controlled, with the rate limited by dissolution and precipitation on the crystal surface. An equivalent growth-rate equation is <D> - <D 0 > = K 1/n t 1/n = K 1/n n t 1/n = (1.12 × 10 -3 μm/s 1/2 ) t 1/2 , a parabolic rate law equation that can be used to demonstrate that zircon in the presence of H 2 O at 1000 ℃ grows fast enough by Ostwald ripening to produce micrometer-thick overgrowths in <10 y. Rapid coarsening of zircon by Ostwald ripening will occur during thermal metamorphism when fluid is present. Addition of fluid (and possibly melt) greatly increases the rate of coarsening, suggesting that most metamorphic rims form nearly instantaneously during fluid influx (or anatexis), and that U-Pb rim ages date such events.

Biotite dissolution in the laboratory and in nature was examined and compared to elucidate certain aspects of the weathering processes. Batch dissolution experiments of fresh biotite [(K 0.91 Na 0.01 )(Mg 0.40 Fe 2.07 Mn 0.05 Al 0.14 Ti 0.19 )(Si 2.82 Al 1.18 )O 10 (OH) 2 ] in granite were carried out at 150 ℃ for 1 to 56 days. Examination by scanning electron microscope (SEM) equipped with an energy dispersive X-ray spectrometer (EDS) revealed that, in the early stage, dissolution proceeds from the edges of crystals inward and secondary minerals such as Fe oxide are precipitated mostly at the edges, with only a few secondary minerals found on the basal surfaces. A dissolution experiment using a mixture of biotite and muscovite, done at 150 ℃ for 7 days, indicated that hematite crystals formed mostly at the edges of biotite but not on muscovite. This observation suggests that released Fe is precipitated before it diffuses into the bulk solution. Because a dissolution rate at the edge is larger by two orders of magnitude than at the basal surface (Turpault and Trotignon 1994), precipitation before diffusion of dissolved elements to bulk solution well explains the preferential secondary mineralization at the edges in the laboratory experiments. SEM-EDS of fresh to slightly weathered biotites revealed that early stage weathering proceeds in the same way as in the laboratory with the edges being preferentially weathered and secondary minerals being precipitated mostly at the edges. Because of the similarity between the occurrence of secondary minerals in the laboratory experiments and in nature, the laboratory results elucidate the early stage weathering conditions, namely that (1) supersaturation with respect to secondary minerals in a solution occurs around biotite, i.e., the solution is poorly connected to a main flow pathway of water, and (2) once supersaturation is achieved, secondary minerals are precipitated mainly at the edge before some released elements diffuse into the solution. The immobility of water immediately adjacent to primary minerals partly explains the large difference in dissolution rate between the laboratory and natural samples. A phlogopite dissolution experiment was done at 150 ℃ for 5.5 days to examine the effect of biotite composition on vermiculitization. High-resolution transmission electron microscopy revealed that vermiculite layers are formed by layer-by-layer transformation within phlogopite after 5.5 day dissolution but not within biotite after 56 day dissolution or within slightly weathered biotite. A comparison of our data with results of other studies indicates that the formation of vermiculite layers between biotite layers occurs even in the early stage when biotite is not Fe-rich [more than 0.8 Mg per O 10 (OH) 2 ]. When biotite is Fe-rich [less than 0.4 Mg per O 10 (OH) 2 ]-like the present one- released Fe is precipitated as, for example, Fe hydroxide, and vermiculite is rarely formed because of limited availability of Mg. Thus, a higher Mg content in biotite facilitates the formation of vermiculite, at least in the early stage. Because vermiculite dissolves at a much slower rate than biotite, Mg-rich biotite dissolves at a slower rate than Fe-rich biotite. In the late stage, biotite dissolution still continues from the edge and within biotite, which results in a fine comb-like texture for weathered biotite; vermiculite occurs as a domain within chemically and structurally altered biotite as well, as at the edge even during weathering of Mg-poor biotite.

The partitioning of Fe and Mg between garnet coexisting with olivine, wadsleyite, ringwoodite and magnesiowüstite solid solutions has been measured between 9 and 19 GPa and 1400-1700 ℃. In addition to studying exchange reactions involving the pyrope-almandine garnet solid solution, majoritic garnets forming in a natural peridotite bulk composition have been also examined. Metallic Fe was added to all bulk compositions to buffer the Fe 3+ concentration at its lowest possible level. Partitioning data between pyrope-almandine garnet and ringwoodite were combined with data for garnet-magnesiowüstite and ringwoodite-magnesiowüstite partitioning to refine thermodynamic mixing properties for these solid solutions at 18 GPa and 1400 ℃. A multiple non-linear regression employing cation exchange data between these 3 solid solutions yielded the following well-constrained values for the differences in symmetric solid-solution interaction parameters: W mw FeMg - W ring FeMg = 8.8(2) kJ/mol, W mw FeMg - W gt FeMg = 12.8(5) kJ/mol,W ring FeMg - W gt FeMg = 4.0(4) kJ/mol. These differences were then solved simultaneously to give, on a single site basis: W mw FeMg = 13.2(3) kJ/mol, W ring FeMg = 4.4(2) kJ/mol, W gt FeMg = 0.3(3) kJ/mol. In a peridotite composition, the occurrence of majorite and grossular garnet components produced a measurable influence on the Fe-Mg partitioning, however, the combined effect was relatively small because each individual component affected the partitioning in an opposite direction. These data, combined with previously determined phase relations in the Mg 2 SiO 4 -Fe 2 SiO 4 system, were used to calculate the influence of garnet Fe-Mg partitioning on the pressure intervals of divariant (Mg,Fe) 2 SiO 4 phase transformations in the transition zone. Results show that the presence of garnet reduces the pressure interval over which the (Mg 0.9 ,Fe 0.1 ) 2 SiO 4 olivine to wadsleyite transformation occurs in a peridotite composition to 0.35 GPa (10 km) in comparison to 0.5 GPa in a garnet-free system. This width, however, is still greater than the estimated width of 4 km for the 410 km discontinuity observed in some regions of the Earth as inferred from high frequency reflected and converted seismic waves. The existence of garnet in a peridotite composition in the mantle cannot, therefore, be the only explanation as to why the 410 km discontinuity is, in some regions of the Earth, apparently sharper than experimental estimates for the (Mg 0.9 ,Fe 0.1 ) 2 SiO 4 olivine to wadsleyite transformation. The wadsleyite to ringwoodite transformation is reduced to 0.8 GPa (25 km) compared to 1 GPa in the garnet-free system.

Comparison of the observed far-infrared (FIR) spectrum of Cl - -containing hydrotalcite, [Mg 3 Al(OH) 8 ]Cl·3H 2 O, to a power spectrum calculated using molecular dynamics (MD) computer simulation, provides a greatly increased understanding of the structure and vibrational dynamics in the interlayers of layered double hydroxides. Good agreement between the observed FIR band positions and the simulated power spectrum illustrates the capability of this combination of experimental and computational techniques to effectively probe the structure and dynamics of water in nano-pores and other confined spaces. The simulation model assumes an ordered Mg 3 Al arrangement in the octahedral sheet and no constraints on the movement of any atoms or on the geometry and symmetry of the simulation supercell. Calculated anisotropic components of the individual atomic power spectra in combination with computed animations of the vibrational modes from normal mode analysis allow for reliable interpretations of the observed spectral bands. For the vibrations related to octahedral cation motions, bands near 145, 180, and 250 cm -1 are due dominantly to Mg vibration in the c direction (perpendicular to the hydroxide layers), Al vibration in the c direction, and Mg and Al vibrations in the a-b plane (parallel to the hydroxide layers), respectively. The low frequency vibrational motions of the interlayer are controlled by a network of hydrogen bonds formed between interlayer water molecules, Cl - ions, and the OH groups of the main hydroxide layers. The bands near 40-70 cm -1 are related to the translational motions of interlayer Cl - and H 2 O in the a-b plane, and the bands near 120 cm -1 and 210 cm -1 are largely due to translational motions of the interlayer species in the c direction. The three librational modes of interlayer water molecules near 390, 450, and 540 cm -1 correspond to twisting, rocking, and wagging hindered rotations, respectively. The spectral components of the interlayer Cl - motions are remarkably similar to those of bulk aqueous chloride solutions, reflecting the structural and dynamic similarity of the nearest-neighbor Cl - environments in the interlayer and in solution.

Quantitative determination of the abundance of Si(nAl) tetrahedral structural units (where n = 0, 1, 2, 3, or 4) through analysis of 29 Si magic angle spinning nuclear magnetic resonance (MAS NMR) spectra was used to assess the state of Si-Al disorder in the zeolites analcime [(NaAl) x Si 48-x O 96 ·16H 2 O], chabazite [(Ca 0.5 ,Na,K) x Al x Si 12-x O 24 ·12H 2 O], and wairakite [CaAl 2 Si 4 O 12 ·2H 2 O]. Short-range Si-Al ordering in chabazite and analcime is a regular function of Al mol fraction and is fully consistent with Al avoidance, as has generally been reported for zeolites not subjected to heat treatment. The results of this study and previously reported 29 Si MAS NMR spectra suggest that natural analcime samples are more Si-Al ordered than either their synthetic counterparts or chabazite. Cluster variation method (CVM) calculations were used to calculate the configurational entropy (S CON ) due to Si-Al disorder in chabazite and analcime. The calculations predict that long-range Si-Al ordering develops when Al occupies 5 out of 12 tetrahedral sites in chabazite and synthetic analcime and 17 out of every 48 tetrahedral sites in natural analcime. The difference between the calculated entropies and ideal entropies of mixing was used to derive activity-composition relationships for Si-Al substitution in these frameworks. Comparison between calculated values of S CON and the results of calorimetric and phase equilibrium studies on analcime indicate that the CVM accurately assesses SCON. The 29 Si MAS NMR spectrum obtained for natural wairakite indicates that this mineral is largely Si- Al ordered, but comparison with a previously published spectrum indicates that natural and synthetic wairakites can exhibit significant variation in Si-Al disorder.

The crystal structure of cafetite, ideally Ca[Ti 2 O 5 ](H 2 O), (monoclinic, P2 1 /n, a = 4.9436(15), b = 12.109(4), c = 15.911(5) Å, b = 98.937(5)°, V = 940.9(5) Å 3 , Z = 8) has been solved by direct methods and refined to R 1 = 0.057 using X-ray diffraction data collected from a crystal pseudo-merohedrally twinned on (001). There are four symmetrically independent Ti cations; each is octahedrally coordinated by six O atoms. The coordination polyhedra around the Ti cations are strongly distorted with individual Ti-O bond lengths ranging from 1.743 to 2.223 Å (the average <Ti-O> bond length is 1.98 Å). Two symmetrically independent Ca cations are coordinated by six and eight anions for Ca1 and Ca2, respectively. The structure is based on [Ti 2 O 5 ] sheets of TiO 6 octahedra parallel to (001). The Ca atoms and H 2 O groups are located between the sheets and link them into a three-dimensional structure. The structural formula of cafetite confirmed by electron microprobe analysis is Ca[Ti 2 O 5 ](H 2 O), in contrast to the formula (Ca,Mg)(Fe,Al) 2 Ti 4 O 12 .4H 2 O suggested by Kukharenko et al. (1959). The wrong chemical formula suggested for cafetite by Kukharenko et al. (1959) is probably due to admixtures of magnetite or titanomagnetite in their samples. Cafetite is chemically related to kassite, CaTi 2 O 4 (OH) 2 , but differs from it in structure and structural formula.

The crystal structure of calcioaravaipaite, PbCa 2 Al(F,OH) 9 , was initially solved by direct methods in the monoclinic space group A2/m (R = 12.4%). Further study demonstrated the OD nature of the structure, and showed that the crystal was twinned. The structure was solved in the triclinic space group C1̅ , a = 7.722(3), b = 7.516(3), c = 12.206(4) Å, α = 98.86(1), β = 96.91(1), γ = 90.00(1)°, V = 694.8(3) Å 3 , Z = 4, yielding R = 5.1% for 1420 reflections with F o > 4σ(F o ). Calcioaravaipaite belongs to a family of order-disorder (OD) structures formed by equivalent layers of symmetry C2/m. Two maximum-degree-of-order (MDO) polytypes are possible. MDO1 results from a regular alternation of stacking operators 2 1/2 and 2 -1/2 and yields a monoclinic structure with C2/c, a = 7.72, b = 7.52, c = 24.12 Å, β = 96.99°. MDO2 results from the sequence 2 1/2 / 2 1/2 / 2 1/2 /… and yields a triclinic structure with a = 7.72, b = 7.52, c = 12.21 Å, α = 98.86, β = 96.91, γ = 90.00°. The structure of calcioaravaipaite is comprised of two kinds of alternating polyhedral slabs parallel to (001). Slab 1 consists of a fluorite-like double layer of edge-sharing (CaF 8 ) distorted cubes and slab 2 is a composite of face- and edge-sharing (PbF 12 ) polyhedra and outlying (AlF 6 ) octahedra, the latter sharing faces and edges with the (PbF 12 ) polyhedra, but no elements with one another. Aravaipaite and calcioaravaipaite share a common fluorite-type layer; however, in aravaipaite the presence of Pb 2+ rather than Ca 2+ in this layer results in slabs of strikingly different polyhedral configuration.

Plane-wave pseudopotential density functional methods using the Perdew-Burke-Ernzerhof exchange- correlation functional were used to investigate theoretically the structures of five NaOH hydrate phases through optimization of lattice parameters and atomic coordinates. Although all the calculations were carried out with P1 symmetry, we find in four of the five cases that the experimentally determined space group is maintained to high accuracy. Particular focus is placed on the coordination environments of Na + and OH - . The Na-O distances are, in general, overestimated; however, the sodium ion coordination polyhedra are well reproduced by the theoretical calculations, including the fivefold coordinated sodium atom in the α-NaOH·4H 2 O structure. The theoretical calculations correctly predict that α-NaOH·4H 2 O is lower in energy than the metastable β-NaOH·4H 2 O phase; thus, the a phase is stable even in the absence of proton disorder. The octahedral coordination environment around OH- is calculated accurately, including the distances of the weak OH - -OH 2 hydrogen bonds in which the hydroxide ion acts as the proton donor. This work provides further evidence of the reliability of the Perdew-Burke-Ernzerhof exchange-correlation functional in hydrogen bonded systems, providing a direct, unambiguous test of the elusive hydroxide-water interaction.

Single-crystal X-ray diffraction data, collected from a sartorite crystal from Lengenbach (Binntal, Valais, Switzerland), yielded a ninefold superstructure: space group P2 1 /c; a = 37.71(2), b = 7.898(3), c = 20.106(8) Å, β = 101.993(7)°; R 1 = 6.08% for 6293 reflections with I > 2σ I . The sample is an N 1,2 = 3,3 sartorite homologue and the refined formula Pb 8 Tl 1.5 As 17.5 S 35 compares very well with the empirical formula Pb 8.2 Tl 1.4 As 17.5 Sb 0.5 S 35 obtained from electron microprobe analyses. It has a high Tl content, up to 6.5 wt%. In a coupled substitution approximately 1.5 Tl + replace about 0.5 As 3+ and 1 PbS. The refined structure has 35 S atoms pfu instead of the expected 36 S (= 9 × 4 S, from PbAs 2 S 4 ). The incorporation of substantial amounts of Tl + into PbAs 2 S 4 is essential for the type and periodicity of superstructures in sartorite. The refined superstructure can be interpreted as a socalled “lock-in” structure with a composition that yields a commensurate lattice for a mineral that usually has an incommensurate lattice. No commensurate periodicity could be found for a second crystal with about 0.5 Tl apfu.

We report data on the composition and crystal structure of the most K-rich (3.61 wt% K 2 O) natural clinopyroxene yet discovered. The studied crystal was found as a tiny inclusion in garnet from a garnet-clinopyroxene rock of the Kumdy-Kol microdiamond mine, Kokchetav complex, Northern Kazakhstan. Microprobe analysis yields the formula (Ca 0.61 Fe 0.13 Mg 0.04 Mn 0.01 K 0.17 Na 0.05 ) (Al 0.61 Mg 0.39 )(Si 1.61 Al 0.39 )O 6.00 . Lattice parameters are: a = 9.773(1), b = 8.926(1), c = 5.269(1) Å, β = 105.75(1)°. The structure was refined up to R all = 2.42% using 982 independent reflections. Substitution of K for Ca causes significant modification of the average structure. No evidence for an additional M2’ position was found. Crystal-chemical characteristics are compared with published data on both natural and synthetic K-bearing clinopyroxenes.

Many basaltic meteorites are being discovered in old and new meteorite suites including those from cold- deserts (e.g., Antarctica) and hot-desert environments. It is important to establish the specific planetary body source. Proven techniques for establishing planetary parentage include stableisotopic signatures (especially oxygen), certain elemental ratios in bulk samples, and certain elemental ratios in specific minerals. Some of these techniques are expensive, require considerable sample preparation, and are adversely affected by weathering processes on the parent body or on Earth. We have been seeking key major and minor elemental ratios (in pyroxene, olivine, and feldspar) that can be measured by the electron microprobe on standard thin sections. These ratios may be preserved in unweathered portions of mineral grains and thus “see through” weathering processes. In addition, if the sample is too small to provide a representative bulk composition, it may still have key information recorded in individual minerals. We have found that some of the most useful chemical parameters are Fe/Mn (atomic) in olivine or pyroxene and the percent anorthite (%An) in plagioclase solid solutions. A plot of Fe/Mn in pyroxene and/or olivine verses %An defines compositional fields that are significantly different for Earth, Mars, Moon, 4 Vesta, and the angrite parent body. This method may be especially powerful when combined with oxygen isotope data.