Volume 86, Issue 3

Coexisting minerals of the hollandite-cryptomelane series, piemontite-group, and abswurmbachite-braunite series occur as a small nodule in a quartzose schist (metachert) from the epidote-amphibolite facies area of the Sanbagawa metamorphic belt, central Shikoku, Japan. Stron­tium, Ba. K. and Pb contents of the hollandite-cryptomelane minerals are 0.00-0.42, 0.07-0.99, 0.00-0.75 and 0.00-0.05 apfu (atoms per formula unit: O = 16), respectively. The piemontite-group minerals can be divided into common piemontite (Ca > Sr in the ten-coordinated A2-site) and strontiopiemontite (Sr > Ca). Strontium. Ba and Pb contents of the strontiopiemontite reach 0.71, 0.26 and 0.20 apfu (O = 12.5), respectively, suggesting the existence of two possible new end- members for epidote-group minerals: “Ba-piemontite [BaCa(Mn 3+ , Fe 3+ , Al) 3 Si 3 O 12 (OH)]” and "Pb- piemontite [PbCa(Mn 3+ , Fe 3+ , Al) 3 Si 3 O 12 (OH)]”. The abswunnbachite-braunite series minerals (Cu = 0.30-0.53 apfu: O = 12) contain up to 0.34 apfu Mg, indicating the exchange Mg(Mn 2+ , Cu) -1 . Distribution coefficients for Sr between the hollandite-cryptomelane series and the piemontite-group minerals are up to 20. implying that the hollandite-type structure with its large open tunnel is more suitable for incorporation of Sr than the A2-site of the piemontite-group minerals. A synthetic alumi­nosilicate analogue of the hollandite-cryptomelane compound is stable under P > 8-10 GPa. and suggests the hollandite-type aluminosilicate may be a reservoir of Sr and other large-ion-lithophile elements in the upper mantle.

An important prerequisite for interpreting microstructures in plutonic and metamorphic rocks is understanding crystallization processes. This study uses crystal size distribution (CSD) and grain shape parameters to investigate mineral-reaction-controlled crystal growth and melt solidification in migmatites of the Bayerische Wald (Variscan Orogeny, Germany). Partially molten rocks were studied because they allow the simultaneous investigation of crystallization of a melt and the growth of solid products of a metamorphic reaction. CSD and shape factors were obtained for three different minerals: (1) cordierite in a diatexite, where the melt remained in the system; (2) K-feldspar in a segregated K-feldspar + quartz leucosome; and (3) plagioclase from a segregated plagioclase + quartz leucosome. Cordierite crystals exhibit a nearly linear CSD pattern and subeuhedral crystal shapes. K-feldspar grains display an approximately linear CSD except for a marked increase at large crystal size classes. K-feldspar grains are characterized by strongly lobate grain boundaries. The CSD of plagioclase crystals is nearly linear, with a slight decrease at small crystal size classes. Plagioclase shape factors also indicate lobate grain boundaries, however, with lower values than K-feldspar. Cordierite CSD and shape data are consistent with interface-controlled crystallization. The CSD indicates that nucleation and growth is the rate limiting step during the cordierite-producing reaction, in contrast to porphyroblast growth in solid rocks where mass transfer is typically rate limiting. CSD combined with information on the duration of crystallization yields growth rates of 10 -17 to 10 -18 m/s. The plagioclase data are consistent with magmatic crystallization and are well-described by the communicating neighbor theory. The K-feldspar data also indicate magmatic crystallization. The “overproduction” of large grains of K-feldspar reflects the interplay between nucleation and growth rates at the initial stage of crystallization. Because of different environments between solid and anatectic rocks, the CSD may help to decipher metamorphic from anatectic migmatites.

The distribution of F between amblygonite (Amb, LiAlPO 4 F)-montebrasite (Mbr, LiAlPO 4 OH) solid solutions and metaluminous haplogranitic melt has been calibrated at 585 °C and 200 MPa H 2 O. The partition coefficient for F between the crystalline phase and melt, D F Mbr/melt , is linear between 0 to ~10 wt% F in amblygonite, which contains 13 wt% F at the end-member: C F Mbr = 3.65C F melt + 0.07, r 2 = 0.995, n = 6. Values of D F Amb/melt decrease sharply above 10 wt% F in amblygonite as the amblygonite reaches saturation in F at 200 MPa H 2 O. In natural occurrences, however, the vast majority of primary amblygonite- montebrasite solid solutions contain ~4-7 wt% F, well within the linear range of the calibrated exchange reaction, and the montebrasite-bearing assemblages are among the last to crystallize. If the F contents of the montebrasite are magmatic, then these most-fractionated residual melts of the LCT (Li-Cs-Ta, and mostly peraluminous S-types) rare-element class generally contained up to ~1.0-1.8 wt% F near the end of their crystallization. The modest F contents of pegmatites are consistent with the common association of Li aluminosilicates and with the general paucity of topaz in these occurrences. In topaz-bearing granites of Western Europe, however, high-F amblygonite (~10- 11 wt% F) reflects >3 wt% F in melt during crystallization of these magmatic phases.

Water loss was measured from H 2 O-undersaturated granitic melts in Au 75 Pd 25 capsules during 6- 43 h piston-cylinder experiments using NaCl-pyrex glass-crushable alumina assemblies at 1050 to 1200 °C and 1.0 GPa. Experiments performed when capsules were surrounded only by alumina demonstrated severe water loss, in some cases more than 70% of the initial water added. It has long been known that surrounding capsules with pyrophyllite powder reduces water loss, but the efficacy of this technique has not been quantified previously. Our results confirm that by surrounding Au 75 Pd 25 capsules with pyrophyllite in the assembly, the loss of water is significantly reduced at 1200 °C, 1.0 GPa. When ~5.6 wt% H 2 O is added to the sample, the loss of water decreases from ~60 to less than 20% relative by pyrophyllite addition, a value slightly higher than the uncertainty of SIMS analysis (10% relative). When about ~2 wt% H 2 O is added to a sample, the use of pyrophyllite in the assembly causes the loss of water to drop from more than 70 to 0% relative (no H 2 O loss within analytical uncertainty).

Evidence for liquid immiscibility on the Moon has been documented in melt inclusions in minerals that crystallized from mare basalts, within mesostasis in mare basalts and in at least one sample of a lunar plutonic rock. This study focuses on trace-element partitioning between immiscible melt pairs occurring as inclusions in plagioclase. These inclusions are blade-like in shape and range in length from 2 to 150 μm. They consist of two coexisting glasses separated by a sharp meniscus. The colorless, low-index glass (felsite) is spherical and immersed in a dark-brown, high-index glass (high-Fe basalt). The Fe-rich basaltic component of the inclusion is quartz + hypersthene normative with low Mg/(Mg + Fe) and variable SiO 2 content (between 34 and 45 wt%). The felsic component has a SiO 2 content between 72 and 82 wt% and variable K 2 O (4.4 to 8.5 wt%) In most cases, the partitioning behavior of the trace elements agrees with behavior either predicted or measured in experimental and natural systems. The high charge density elements preferentially partition into the basaltic component. High P 2 O 5 and a wider solvus increase the D basalt/felsite for these elements. In contrast to many experimental studies and in agreement with studies of natural silicate liquid immiscibility, the D basalt/felsite for Ba indicates a preference for the felsic component. This difference in Ba behavior between experimental and natural samples has been attributed to differences in melt polymerization and compensation for charge unbalance within the polymerized melt structure. The apparent differences in D basalt/felsite for Sr and divalent Eu between the Apollo 11 high-Ti basalts and Apollo 12 low-Ti basalts may be a result of either subtle differences in bulk composition or the extent of plagioclase plating on inclusion walls. Based on our partitioning data from melt inclusions and the chemical characteristics of lunar felsites, the latter cannot be a product of simple fractional crystallization. Silicate liquid immiscibility can account for fractionation of Ba/ La, K/U, and Ba/U in the lunar felsites and the limited variation in Zr/La. However, it cannot account for the REE pattern of the lunar felsites or the fractionation of U/La. These characteristics must be attributed to whitlockite crystallization prior to the onset of liquid immiscibility.

Critical evaluation of the thermodynamic data for titanite, in conjunction with a heat capacity equation that takes into account the P2 1 /a ↔A2/a transition and a more complete P-V-T data set for titanite, indicates that: (1) enthalpy of formation values greater than -2600 kJ/mol are not supported by calorimetry; (2) calculated internally consistent feasible solutions converge to enthalpy of formation from the elements values between -2608 and -2600 kJ/mol and 113 to 121 J/mol⋅K for the entropy; (3) the practice of adjusting the enthalpy of formation to fit phase equilibrium experiments may be erroneous, in contrast, it is the currently accepted entropy of 129.20 ± 0.84 J/mol⋅K that may need revision to a smaller value. Consequently, we optimize standard-state properties for end-member titanite (P2 1 /a structure) consistent with the thermodynamic data in the program QUILF. In addition, we use a modified version of the program to calibrate equilibria among titanite (CaTiOSiO 4 ), Fe-Mg-Ti ilmenite and spinel, Ca-Mg-Fe pyroxenes and olivine, and quartz. Calculations at 1 and 3 kbar, and 650, 850, and 1100 °C, in the system CaO-MgO-FeO-Fe 2 O 3 -TiO 2 -SiO 2 , suggest that the reactions: augite + ilmenite = titanite + spinel phase + quartz and augite + ilmenite + quartz = titanite + orthopyroxene impose well defined f O₂ , a SiO₂ , and X Fe Opx restrictions to the assemblages (1) titanite + spinel phase + quartz, (2) titanite + orthopyroxene, (3) augite + ilmenite, and consequently titanite stability. The absence of quartz (a SiO₂ < 1 with respect to Qz) favors the sub-assemblage Fe-Mg-Ti spinel + titanite over augite + ilmenite, and the latter over titanite + orthopyroxene. Nonetheless, in the absence of quartz, no titanite-bearing assemblage is stable relative to olivine + orthopyroxene + augite + ilmenite + Fe-Mg-Ti spinel. From the phase relations, we can also infer that in quartz-saturated rocks: (1) titanite can coexist with orthopyroxene only at low X Fe Opx values, regardless of f O₂ conditions, whereas the assemblage titanite + orthopyroxene + spinel phase requires high f O₂ , relative Mg-enrichment (from high to moderate X Fe Opx ), and temperatures ≤650 °C. (2) The association of titanite with a spinel phase is generally indicative of relatively oxidizing conditions. (3) The assemblage titanite + olivine requires fO2 condition close to FMQ and relative Fe-enrichment (from moderate to high X Fe Opx ), and temperatures ≤650 °C.

In the system CaO-MgO-FeO-Fe 2 O 3 -TiO2--SiO 2 , equilibria among titanite, pyroxene(s), olivine, spinel phase, ilmenite, and quartz constrain f O₂ , T, a SiO₂ , a TiO₂ , a Fe , and the compositional variables, μTiFe 2+ Fe 3+ -2 , μCa(Mg,Fe) -1 , and consequently μMg,Fe -1 among the silicate and oxide phases. These equilibria can provide the framework for a better understanding of relevant titanite-bearing natural assemblages. In addition, such equilibria can constrain the variables that operate during crystallization of igneous and metamorphic rocks. Titanite-bearing assemblages with pyroxene(s), Fe-Mg-Ti ilmenite and spinel, olivine, and quartz are rare in metamorphic rocks but are common in igneous rocks where they are more commonly reported in plutonic than in volcanic rocks. Probably because, at high temperatures, titanite cannot coexist with the relative common in volcanic rocks sub-assemblage augite + orthopyroxene + olivine + spinel phase + ilmenite. Moreover, decreasing pressure and temperature conditions appear to favor titanite. Thus, it is not surprising that titanite is commonly observed in slowly cooled rocks, albeit, most typically in association with amphibole. We argue that the titanite + amphibole association is favored by high a H₂O and, because a H₂O typically increases during crystallization of a pluton, titanite + amphibole (and consequently titanite) is more common in plutonic than in volcanic rocks. We have used a modified version of the program QUILF that includes our thermodynamic data for titanite and appropriate titanite-bearing assemblages to estimate the equilibrium crystallization conditions of samples from: (1) a tonalite from the Myoken-Zan granitic complex in Japan; (2) the Sandwich Horizon in the Skaergaard intrusion in Greenland; (3) rhyodacites, dacites, and andesites from the Sajama, Porquesa, and Parinacota volcanoes in the Andes; (4) the Fish Canyon Tuff in Colorado; and (5) a quartz-rich, calc-silicate gneiss adjacent to the Horse Creek Anorthosite Complex in Wyoming. The calculations suggest that: (1) these titanite-bearing assemblages crystallized at values of f O₂ that ranged from slightly below FMQ to more than 3 logf O₂ units above FMQ at 500- 800 °C; and (2) the stability of titanite is controlled by T, f O₂ , a SiO₂ , and the compositions of the coexisting oxides and silicates.

To investigate the substitution of Ti 4+ for Si 4+ on tetrahedral sites in clinopyroxene, a series of Ti- bearing clinopyroxene compositions was synthesized in the system CaO-MgO-SiO 2 -TiO 2 (CMST) at pressures from 1 atm to 2 GPa and temperatures from 800 to 1200 °C. Along the join CaMgSi 2 O 6 (diopside)-CaMgTiSiO 6 , clinopyroxene crystallizes, depending on coexisting phases, as binary solid solutions (diopside-CaMgTiSiO 6 ) or ternary solid solutions (diopside-CaMgTiSiO 6 - enstatite). The solubility of Ti increases with increasing temperature, whereas it decreases with increasing pressure. The binary clinopyroxene has an invariant composition (CaMgSi 2 O 6 ) 93 (CaMgTiSiO 6 ) 7 (mol%) corresponding to a Ti content of 2.6 wt% at 1000 °C and 1 atm. Magnesium substitutes in the M2 site in ternary clinopyroxene and its solubility mainly in­creases with increasing temperature. The substitution of Si 4+ by the larger Ti 4+ cation on tetrahedral sites causes a linear increase hi a, b, c, β, and V by 0.0020 Å, 0.0012 Å, 0.0016 Å, 0.01°, and 0.28 Å 3 per mol% CaMgTiSiO 6 , respectively.

The kinetics of Fe 2+ -Mg exchange between the M1 and M2 sites (Fe M2 2+ + Mg M1 = Fe 2+ M1 + Mg M2 ) were followed by ordering experiments in an augitic clinopyroxene (Ca 0.841 Na 0.019 Mg 0.888 Fe 2+ 0.151 Ti 0.011 Al 0.028 Cr 0.003 Fe 3+ 0.054 Mn 0.006 )(Si 1.914 Al 0.086 ) from a dike on Alicudi (Aeolian Islands, Italy). The same single crystal was annealed in silica tubes in an iron-wüstite buffered Ar atmosphere, at 700, 750, 800, 850, and 950 °C, followed by quenching after each thermal treatment, and was used for the collection of X-ray diffraction data and for microprobe analysis. Kinetic constants (K) for each isotherm were calculated from the distribution coefficient K D [= (Fe 2+ /Mg) M1 /(Fe 2+ /Mg) M2 ] measured during a series of experiments carried out by increasing annealing times until exchange equilibrium was achieved. Calculations of K were performed using the Ginzburg-Landau equation, the kinetic model of Mueller, and the mathematical formulation proposed by Sha and Chappell (1996) to check the responses of various kinetic models for a Ca-rich pyroxene on the same data set. The activation energy of the Fe 2+ -Mg exchange spans a narrow range, yielding values of 200 (±28) kJ/mol, independent of the method used to calculate K.

The crystal structure of α-eucryptite, LiAlSiO 4 , is reinvestigated by combining 29 Si NMR spectroscopy and single-crystal X-ray diffraction. The silicon coordination in α-eucryptite of three different samples (two of natural, one of synthetic origin) is shown to be identical on a local scale by means of 29 Si MAS NMR spectroscopy. This method also suggests a well ordered arrangement of Si and Al within the crystal structure of α-eucryptite and the presence of two symmetrically non-equivalent sites for silicon. The refinement of the structure of a natural crystal using single-crystal X-ray diffraction confirms these observations and demonstrates the presence of long-range Al-Si order and the acentric space group R3.

Transmission electron microscopy (TEM) and energy dispersive X-ray (EDX) analysis were used to characterize naturally weathered layers on the surface of volcanic glass, bytownite, hypersthene, and secondary precipitates in volcanic ash erupted from the Sakurajima volcano, Japan. TEM analysis of the volcanic ash revealed sharply, bounded, leached layers on the surface of the volcanic glass and bytownite that were thin structureless coatings mostly ≤0.1 μm thick. EDX analysis showed that the leached layer on the volcanic glass surface is preferentially depleted in Si and enriched in Al relative to its parent matrix, whereas the leached layer on the bytownite surface is extremely depleted in Al and enriched in Si relative to the original bytownite matrix. These chemical characteristics of the weathered layers indicate that incipient dissolution of volcanic glass and bytownite proceeded by preferential leaching of Si and Al, respectively. On the surface of hypersthene, a noncrystalline weathered layer generally <0.01 μm in thickness, which has nearly the same composition as that of the parent matrix, was observed. This weathered layer was produced by precipitation of noncrystalline hydrous ferric oxide with partly developed to poorly crystallized Fe- Si-Mg rich phyllosilicate. The volcanic ash sample contains small amounts of noncrystalline secondary precipitates exhibiting three distinct morphologies: (1) aggregates of very fine fibers, (2) aggregates of fine fibers with crinkled fringes, and (3) spherical forms composed of roughly curled fringes. These noncrystalline precipitates are enriched in Al and Si and contain variable amounts of Fe, depending on their morphology. The Fe content of these materials decreases drastically in the sequence morphology 1 → morphology 2 → morphology 3, which is consistent with the transformation from Al-, Si-, and Fe-rich fine fibers to spherical halloysite by elimination of Fe from the fibers. These alterations of the volcanic ash took place in the crater of the Sakurajima volcano by interaction with near-neutral to weakly acidic solutions under relatively low-temperature conditions.

The thermal transformation of natural, single-crystal diaspore (D) to corundum (C) and water has been analyzed in detail by means of transmission electron microscopy (TEM). Pieces of singlecrystal diaspore were partially dehydrated at temperatures ranging from 400 to 600 °C and were prepared in cross-section for TEM studies. The transformation of diaspore was studied at the faces (100) D , (010) D , and (001) D , with special attention to the microstructural details at the reaction interface. Irrespective of the transformation direction, the dehydration product is highly porous. Nanometer- sized pores form two-dimensional arrays arranged parallel to the basal plane of corundum forming lamellae that are separated by thin regions of dense corundum. The periodicity of the lamellar structure is 3.7 nm, which causes distinct satellite reflections in diffraction patterns in the case of dehydration at moderate temperatures (<450 °C). At higher temperatures, the degree of regularity, as well as the satellite reflection intensities, decrease. During transformation, the diaspore crystal cleaves parallel to (010) D , and micro-cracks are developed parallel to (100) D and to (001) D in the corundum due to the misfit of the lattices. The entire crack system as well as the lamellar pore system play an important role in the dehydration process because they serve as effective transport paths for the water vapor. The rapid dehydration at the (010) D faces is attributed to fast-propagating cracks along [010] D . The slow transformation processes along [100] D and [001] D are discussed in terms of difficulties in crack propagation and in generating suitable crack systems, respectively.

Six kaolinite [Al 2 Si 2 O 5 (OH) 4 ] samples were synthesized under different conditions of temperature, pressure, and pH from two different starting materials. Chemical composition and properties of the samples were characterized using classical methods (electron microprobe, atomic absorption spectrometry, X-ray diffraction, differential and thermal analyses, and Fourier transform infrared spectrometry). All synthetic kaolinite samples contained various amounts of a boehmite impurity. The defect density was different for each kaolinite, ranging from high to low. The enthalpy of formation of these kaolinites at 25 °C was investigated by drop solution calorimetry into molten lead borate at 700 °C. All data were corrected for impurities. Whatever the synthesis conditions and the kaolinite properties, the enthalpy of kaolinite dissolution into molten lead borate at 700 °C and the standard enthalpy of kaolinite formation from the oxides and from the elements at 25 °C are constant: 372.3 ±1.0 kJ/mol , -46.6 ±2.6 kJ/mol, and -4115.3 ±4.1 kJ/mol respectively. Using entropy data from the literature, the standard Gibbs free energy of kaolinite formation from the elements at 25 °C is -3793.9 ±4.1 kJ/mol. This value is in excellent agreement with most of the literature data obtained for natural kaolinites. Furthermore, the standard Gibbs free energy of kaolinite formation at 25 °C and 1 atm is very close to that obtained using the same method for the San Juanito dickite, which is commonly used as a standard mineral, the value for kaolinite being slightly more negative than the value for dickite. This trend is also true for all the temperature and pressure range of kaolin minerals occurrences. Thus, dickite is a metastable phase relative to kaolinite, and kaolinite seems to be thermodynamically more stable than dickite, as already proposed by DeLigny and Navrotsky (1999) and Anovitz et al. (1991). The natural occurrence of dickite must result from specific reaction paths and be controlled by kinetic factors.

Plane wave pseudopotential methods are used to investigate the structures and total energies of AlOOH and FeOOH in the five canonical oxyhydroxide structures: diaspore (goethite), boehmite (lepidocrocite), akaganeite, guyanaite, and grimaldiite. The local density approximation was used in conjunction with ultrasoft pseudopotentials in full optimizations of both AlOOH and FeOOH in each of these structures. Structures are in reasonably good agreement with experiment, with lattice parameters and bond lengths within 3% of experimental values. Neither AlOOH nor FeOOH have been identified in the grimaldiite or guyanaite structures, however we find that total energies for AlOOH and FeOOH in these structures are comparable to or lower than the total energies of the commonly observed polymorphs (with the exception of FeOOH in the grimaldiite structure, which is anomalously high energy). Estimated zero-point energy corrections do not alter this result. For diaspore and boehmite, we also provide calculations using the generalized gradient approximation and norm-conserving pseudopotentials to assess the extent to which the results depend on the particular level of theory used. We find that diaspore is predicted to have a lower energy, consistent with experimental observation, using all but one methodological combination where the generalized gradient approximation is combined with ultrasoft pseudopotentials. Thus, although one may reasonably conclude that the differences in total energies of the various (Al,Fe)oxyhydroxide polymorphs are small, current electronic structure methods do not appear to be fully capable of accurately resolving these small differences. These findings provide further confirmation that the structures of oxyhydroxide polymorphs and surface precipitates are more likely to be a function of kinetics than of intrinsic lattice stability.

Collection of S 2p XPS spectra of millerite (NiS), using a conventional (AlKα) and a synchrotron photon source, demonstrates the presence of one surface species on millerite (NiS) and spectral deconvolution indicates a second surface contribution. The observed surface core-level shift (binding energy = 161.1 eV) is attributed to a surface monomeric species (S 2- ) whereas the second contribution probably is a surface dimeric species (S 2 2- , binding energy = 162.3 eV). Surface dimers, if present, indicate surface reconstruction of millerite surfaces upon cleavage. Monoclinic (Fe 7 S 8 ) and hexagonal (Fe 10 S 11 ) pyrrhotite are non-stoichiometric due to vacancies on metal sites. The conventional S 2p XPS spectra of these phases, which sample primarily bulk sulfur states, reveal contributions from fivefold and sixfold coordinated S atoms, the proportion of which is consistent with structural refinement studies. The more intense signal is derived from S in fivefold coordination (80-85%) and the remainder represents the sixfold contribution. Comparison of a highly surface sensitive S 2p XPS spectrum of pyrrhotite (photon energy tuned to 210 eV) with a conventional S 2p XPS signal (AlKα source) indicates the presence of a monomeric surface species (S 2- ). Spectral deconvolution of the surface sensitive spectrum indicates another contribution near 162 eV, the origin of which is uncertain. It may represent S atoms in sixfold coordination, surface dimers (S 2 2- ) or both.

Single-crystal electron diffraction patterns of nanocrystals in two samples of synthetic six-line ferrihydrite (6LFh) were obtained using electron nanodiffraction to produce diffraction patterns from areas between ~0.7 and 5 nm across. One of the samples was synthesized at 75 °C by a conventional method; the other was synthesized by a recently developed room-temperature technique. Structures of individual nanocrystals were investigated by comparing experimental and simulated electrondiffraction patterns. The most common structure in each sample is based on double-hexagonal (ABAC) stacking of close-packed oxygen layers, and is similar to the “defect-free ferrihydrite” structure proposed by Drits and co-workers. Nanocrystals with maghemite- or magnetite-like structures and structures similar to those in two-line ferrihydrite (2LFh) occur less commonly in both samples. The sample synthesized at room temperature has a small amount of hematite, which was not found in nanodiffraction patterns from the 75 °C sample. The most common structure in 6LFh is distinct from those previously determined by nanodiffraction of 2LFh, indicating that 6LFh is not simply a more crystalline form of 2LFh.

Atomic structures of planar defects found in oxybiotite from the Ruiz Peak ash flow, Jemez Mountains, New Mexico, were investigated using high-resolution transmission electron microscopy (HRTEM) and electron nano-beam chemical analyses. Two kinds of planar defects were observed. One defect contains more Si and Al than the host biotite. HRTEM images recorded along two directions indicated that this defect consists of an unbranched tetrahedral double sheet, as reported, e.g., in hexagonal BaAl 2 Si 2 O 8 . The other defect is rich in Fe and Mg. Complete determination of the atomic structure was not successful, but this defect probably contains several octahedral sheets parallel to (001) of biotite. Investigation of the relationship between these defects and the polytypic features of the host biotite indicated that the tetrahedral double sheets shorten and the Fe/ Mg-rich defects lengthen the period of the polytype unit including the defect. Increase of trivalent cations by oxidation of Fe 2+ in the trioctahedral sheet destabilizes the structure of biotite, which is probably the cause of the formation of these defects. Such a defective structure may be a precursor of decomposition products of biotite by oxidation at high temperature, and our results emphasize the importance of re-examining the microstructures of the specimens previously investigated in experiments on oxidation and decomposition of biotite.

Oxygen-17 magic angle spinning, non-spinning, 1 H-decoupled, 1 H- 17 O cross-polarization, and triple-quantum magic-angle spinning nuclear magnetic resonance (NMR) spectroscopy experiments were performed at two different magnetic fields on crystalline potassium hydrogen disilicate (KHSi 2 O 5 ) and hydrous and anhydrous potassium tetrasilicate (K 2 Si 4 O 9 ), sanidine (KAlSi 3 O 8 ), and haplogranite (KAlSi 3 O 8 -NaAlSi 3 O 8 -SiO 2 ) glasses. Crystalline KHSi 2 O 5 has two bridging oxygen sites and one non-bridging oxygen site, and the NMR parameters for these two types of sites have been determined. The non-bridging oxygen (silanol) site is well-resolved and observed in a crystalline compound with an isotropic chemical shift (δ iso ) of 60 ppm, a quadrupolar coupling constant (C Q ) of 3.5 MHz, and an asymmetry parameter (η) of 0.35. Knowledge of the non-bridging-oxygen site parameters allows for comparisons to glassy or other materials that might have silanol-type structures. A similar peak can be found in the hydrous potassium silicate spectrum but not in the hydrous potassium aluminosilicate glass spectrum.

A single Landau free energy expansion is used to describe phase transitions in perovskites, from a cubic parent structure to tetragonal and orthorhombic structures with space groups related to the M 3 and R 25 points of the Pm3̅m reciprocal lattice. This expansion permits relationships between symmetry-adapted forms of the spontaneous strain and individual order parameter components to be predicted. Data from the literature for (Ca,Sr)TiO 3 perovskites are analyzed in the light of these predictions. Shear strains for I4/mcm, Pnma, and Cmcm structures tend to conform to the predicted pattern. The Pm3̅m↔I4/mcm transition has nearly tricritical character as a function of temperature in CaTiO 3 and more nearly second-order character as a function of composition at the Sr-rich end of the solid solution. Coupling with the volume strain appears to be both temperature and composition dependent, which may be a general feature of phase transitions in perovskites. Renormalization of fourth-order terms by changing the volume coupling coefficients could be responsible for the unusual order parameter evolution shown by CaTiO 3 and for changes in thermodynamic character of the phase transitions as a function of composition. The pattern of strain variations also correlates closely with patterns of variations in heat capacity from the literature, suggesting revisions to the subsolidus phase diagram for the (Ca,Sr)TiO 3 solid solution above room temperature.

Crystals of liddicoatite-elbaite tourmaline from a pegmatite in Jochy, Madagascar are composed of o{021̄1}, r{101̄1}, c{0001}, a{112̄̄0}, and m{101̄0} sectors, which correspond to the prominent crystal faces, respectively. Therefore, the sectors were produced during growth, not by strain after growth. The o, m, and r sectors of one specimen are biaxial between crossed polars [2V(-) = 30°, 20°, and 15°, respectively] and triclinic, as indicated by X-ray diffraction. The a sector is optically biaxial and orthorhombic. The c sector is optically uniaxial and essentially trigonal as indicated by single-crystal X-ray diffraction. The o, r, and c sectors are of liddicoatite component, whereas the a sector of the one specimen corresponds to fluor-elbaite. Another crystal specimen comprises a and m sectors, which are polysynthetically twinned, resulting in striations parallel to the c axis on the prism faces, and of liddicoatite. All five sectors have vacancies in the X-site (Ca, Na, ⃞ ).

We show experimental evidence that the zoned Mg-Fe-Ca carbonates, magnetite, and Fe sulfides in Martian meteorite ALH84001 may have formed by simple, inorganic processes. Chemically zoned carbonate globules and Fe sulfides were rapidly precipitated under low-temperature (150 °C), hydrothermal, and non-equilibrium conditions from multiple fluxes of Ca-Mg-Fe-CO 2 -S-H 2 O solutions that have different compositions. Chemically pure, single-domain, defect-free magnetite crystals were formed by subsequent decomposition of previously precipitated Fe-rich carbonates by brief heating to 470 °C. The sequence of hydrothermal precipitation of carbonates from flowing CO 2 -rich waters followed by a transient thermal event provides an inorganic explanation for the formation of the carbonate globules, magnetite, and Fe sulfides in ALH84001. In separate experiments, kinetically controlled 13 C enrichment was observed in synthetic carbonates that is similar in magnitude to the 13 C enrichment in ALH84001 carbonates.