Volume 86, Issue 4

We explore the implications of recent mineral physics measurements of diffusion coefficients and melting temperatures of lower mantle materials on the rheological and geothermal structure of Earth’s lower mantle. We show that MgSiO 3 perovskite is significantly stronger than MgO periclase and therefore the rheology of the lower mantle depends strongly on the geometry of a weaker phase, periclase. We calculate viscosities of the lower mantle for two cases: (1) where periclase occurs as isolated grains and (2) where periclase occurs as continuous films, using mineral physics data and models of two-phase rheology. We find that the effective viscosity for the former is about ~10-1000 times higher than the latter. We therefore suggest that the rheology of the lower mantle is structureand hence strain-dependent, leading to weakening at large strains due to the formation of continuous films of periclase. Overall depth variation of viscosity depends not only on the pressure dependence of creep but also on the geothermal gradient. Both MgSiO 3 perovskite and periclase have relatively small activation energies (E* = gRT m with g = 10-14, where R is the gas constant and T m is melting temperature), and therefore the depth variation of viscosity is rather small, even for a nearly adiabatic temperature gradient. However, the geothermal gradients consistent with the geodynamical inference of nearly depth-independent viscosity are sensitive to the pressure dependence of viscosity which is only poorly understood. A superadiabatic gradient of up to ~0.6 K/km is also consistent with mineral physics and geodynamical observations.

Dissolution rates of a well-characterized sample of powdered talc were measured in solvents that mimic fluids found in the human lung. These experiments found that talc dissolution rates were the same for pH controlled aqueous solvents, phosphate buffered saline solution, and modified Gamble’s solutions. Variation of solvent chemistry, including the addition of organic chelators and proteins at intercellular fluid concentrations, does not markedly affect the measured dissolution rate of talc at 37 °C. The data further indicate that the dissolution mechanism for talc in aqueous solutions is independent of pH over the range of 2 to 8. The dissolution rate at 37 °C, determined by measuring the silicon release rate per unit surface area of talc in a mixed-flow reactor system, is 1.4 (±1.0) × 10 -11 mol Si/m 2 ·s. Application of a geometric shrinking particle model using this dissolution rate results in an estimated lifetime (upper limit) of approximately 8 years for a 1 μm talc particle under pulmonary conditions. Talc dissolves considerably faster than quartz, but slower than chrysotile and olivine in the body. These data can be used to place constraints on the role of particle dissolution in the disease models associated with airborne respirable mineral particles.

Microbial biomineralization in a weathered volcanic ash deposit from the 1914 to 1915 A.D. eruption of Sakurajima volcano was investigated by transmission electron microscopy (TEM) and energy dispersive X-ray analysis (EDX). The solution chemistry of pore water was also analyzed to elucidate saturation conditions. In addition, experimental incubations of bacteria collected from the volcanic ash were performed to confirm bacterial mineralization. TEM revealed that the weathered volcanic ash contains significant amounts of spherical to rod-shaped bacteria ranging from 1.3 × 108 to 2.6 × 108 cell/g, most of which have cell wall surfaces that are completely covered or decorated by either massive aggregates of allophane-like granular materials or irregular aggregates of smectitelike fibers and/or flakes. EDX confirmed that the granular minerals have chemical compositions similar to proto-imogolite allophane, whereas the smectite-like fibers and/or flakes show a wide range of chemical compositions corresponding to the compositional field between allophane and nontronite. The volcanic ash contains about 22 wt% of pore water, which is slightly acidic, relatively low redox potential, and enriched in Si, Na, Cl-, and SO 4 2- ions. The saturation indices (SI) calculated by the PHREEQC geochemical code indicate that the pore water is almost saturated with respect to amorphous Al(OH) 3 , ferrihydrite, amorphous silica, and cristobalite, and significantly oversaturated with respect to silicate minerals in the order: halloysite < kaolinite < montmorillonite < allophane < nontronite. The allophane-like granular minerals seems to be preferentially precipitated by bacterial interaction with Al and Si ions in the pore water as a metastable phase. The poorly ordered smectite-like fibers and/or flakes may be transformed from the allophane-like materials as a intermediate phase between allophane and nontronite by the driving force originated from the greatest SI value of nontronite. The experimental incubation confirmed that amorphous silica containing a small amount of Fe is formed on the bacterial cell surfaces in liquid media with both Fe and Si ions. Likewise, beidellite-like smectite associated with the bacterial surfaces is produced in liquid media containing both Al and Si ions. However, no minerals are produced in the same media containing no metal ions or no bacteria. These results imply that bacteria play an important role in the accumulation of metal ions and in the formation of silicate minerals during weathering of volcanic ash.

The dissolution behavior of two smectite minerals, hectorite (trioctahedral) and nontronite (dioctahedral), was observed in situ, in acid solutions, using atomic force microscopy. As expected, the crystallites dissolved inward from the edges, and the basal surfaces appeared to be unreactive during the timescale of the experiments. The hectorite (010) faces appeared to dissolve about 6× more slowly than the lath ends, usually broken edges. The edges visibly dissolved on all sides, and appeared to roughen somewhat. On the other hand, the (010), (110), and (11̄0) faces on nontronite crystals were exceptionally stable, so that any dissolution fronts originating at broken edges or defects would quickly become pinned along these faces, after which no more dissolution was observable. These observations can be explained by using periodic bond chain theory to predict the topology of the surface functional groups on the edge faces of these minerals. If a certain amount of predicted surface relaxation is allowed on the (110) and (11̄0) faces of nontronite, an important difference between the exceptionally stable faces and the others becomes apparent. That is, the oxygen sites connecting the octahedral and tetrahedral sheets are all fully bonded on the nontronite (010), (110), and (11̄0) edge faces, whereas all hectorite edge faces and nontronite broken edges would have coordinatively unsaturated connecting O atoms. This explanation for the differential reactivity of these crystal faces implies that the rate limiting step of the dissolution process is the breaking of bonds to connecting O atoms.

Batch cation-exchange experiments were performed by placing 4 g of clinoptilolite-rich rock into 100 mL of 1 M concentrations of NaCl or CsCl solution for differing times and temperatures. The solutions were analyzed for the outgoing cations Na, K, Ca, and Mg by ICP-AES. The goal of these experiments was to determine the behavior (i.e., effect of temperature, release rate, and time) of Na, K, Ca, and Mg exchanging out of the solid and into the liquid in either NaCl or CsCl solutions. For the Na-exchanged samples, the concentration of the outgoing divalent cations increased with time and increased temperature, with the majority of the exchange occurring in the first 10 h. Potassium behaved in a different manner, with approximately 20% exchange occurring within 0.5 h. With increased exchange time, K concentrations actually decreased slightly, indicating some reexchange back into the solid from the liquid. After 24 h, samples in the four storage conditions averaged approximately 20% K, 80% Ca, and 50% Mg exchange out of the sample and into the liquid. For the Cs-exchanged samples, the concentration of the outgoing divalent cations increased with time and increased temperature, with the majority of the exchange occurring in the first 10 h. The outgoing monovalent cations behaved in a different, and somewhat unpredictable, manner. The majority of the exchange for both Na and K occurred within the first 0.5 h. Increased time had little effect on exchange; in fact, both exhibited some re-exchange back into the liquid. Also, more exchange occurred at lower than higher temperatures. With increased exchange time, K concentrations decreased slightly, indicating some re-exchange back into the solid from the liquid. After 24 h, samples in the four storage conditions averaged approximately 85% Na, 50% K, 80% Ca, and 40% Mg exchange out of the sample and into the liquid.

Four grams of a clinoptilolite-rich rock, crushed to -180 mesh, were exchanged in deionized water (DI), and in 0.1, 0.01, and 0.001 M solutions of SrCl 2 ·6H 2 O at 5, 21, 50, and 90 °C for 0.5, 5, 24, and 240 h yielding 64 sample solutions, which were vacuum-filtered to remove solids. The solutions were analyzed by ICP-AES for the outgoing cations (i.e., Na + , K + , Ca 2+ , and Mg 2+ ). The higher molarity solutions were expected to contain greater concentrations of the outgoing cations than the DI or lower molarity concentration solutions. This trend was exhibited by the monovalent and divalent cations in samples of all concentrations. Outgoing monovalent cations exchanged to greater extents with increasing time and temperature. Outgoing divalent cations of 0.01 and 0.001 M concentrations exchanged at low temperatures contained greater concentrations of outgoing divalent cations than solutions exchanged at high temperatures. The concentrations of the monovalent cations were slightly greater in the 0.001 M sample than in DI, whereas divalent cations had a similar concentration between the 0.001 M and DI treatments. Overall exchange of cations out of the clinoptilolite was favored at higher temperatures.

Natural clinoptilolite (Cpt: Na 0.085 K 0.037 Ca 0.010 Mg 0.020 Al 0.182 Si 0.818 O 2 · 0.528 H 2 O) from Castle Creek, Idaho, and its cation-exchanged variants (Na-Cpt, NaK-Cpt, K-Cpt, and Ca-Cpt) were studied by high-temperature calorimetry. The hydration enthalpy for all the clinoptilolites is about -30 kJ/mol H 2 O (liquid water reference state) at 25 °C. The energetic stabilization effect of hydration on each clinoptilolite can be largely correlated to its hydration capacity. The higher the average ionic potential of the extra-framework cations, the larger the hydration capacity of the clinoptilolite. This trend may be attributed to the small size as well as the efficient water-cation packing of high field strength cations in the zeolite structure. The hydration properties of these clinoptilolites are compared with those previously reported in the literature. The dehydration conditions as well as the measurement direction (dehydration of the initially hydrated sample or rehydration of the dehydrated zeolites) are important factors to control to obtain consistent thermodynamic properties for hydration. The standard enthalpy for formation of the clinoptilolites from the constituent elements at 25 °C based on two framework O atoms was obtained from the calorimetric data: -1117.57 ± 0.95 kJ/mol Cpt, -1130.05 ± 1.00 kJ/mol Na-Cpt, -1109.49 ± 1.04 kJ/mol NaK-Cpt, -1094.21 ± 1.12 kJ/mol KCpt, and -1153.78 ± 1.07 kJ/mol Ca-Cpt. Their molar entropy was determined by a summation method based on the thermodynamic properties of the component oxides. Thus the standard free energy based on two framework O atoms was derived: -1034.01 ± 1.05 kJ/mol Cpt, -1044.19 ± 1.10 kJ/mol Na-Cpt, -1027.26 ± 1.13 kJ/mol NaK-Cpt, -1014.89 ± 1.21 kJ/mol K-Cpt, and -1064.95 ± 1.16 kJ/mol Ca-Cpt.

Calorimetric measurements were made on natural samples of heulandite having the composition Ca 0.86 Na 0.37 K 0.06 Al 2.14 Si 6.86 O 18 ·6.1H 2 O and stilbite Ca 1.01 Na 0.12 Al 2.12 Si 6.88 O 18 · 7.27H 2 O both from the same locality (Nidym River, E. Siberia, Russia). Enthalpies of formation and hydration were studied by calorimetry in lead borate solvent at 975 K. The enthalpies of formation from oxides and elements at 298 K of heulandite are: -238.7 ± 4.9 kJ/mol, and -10656.3 ± 8.6 kJ/mol, and of stilbite are -232.0 ± 8.3 kJ/mol and -11017.9 ± 10.9 kJ/mol respectively. The integral hydration enthalpies including the enthalpies of phase transitions during heulandite and stilbite dehydration are -209.4 ± 5.9 kJ/mol and -229.2 ± 7.4 kJ/mol at 298 K, respectively. The molar Gibbs free energies of formation for idealized calcium-sodium and pure calcium heulandites and stilbites were calculated by combining these new calorimetric data with thermodynamic quantities from the literature. Equilibrium temperatures for the reaction: Ca-stilbite = Ca-heulandite + H 2 O, calculated on the basis of these thermodynamic data, agree with experimental phase equilibria.

This paper describes a preliminary study that attempts to determine the oxidation state of Fe (Fe 3+ /ΣFe) with the electron microprobe (EMP) by measuring the self-absorption induced shift of the FeLα peak emitted from minerals and glasses. In transition metals of the first row, the L-spectra exhibit common distortions, namely peak position shifts, peak shape alterations, and changes in the Lβ/Lα ratios, caused by the large difference in the self-absorption coefficients (μ/ρ) on either sides of the L 3 absorption edges that are in close proximity to the Lα peak maxima. Measurements performed on α-Fe 2 O 3 and FexO oxides have shown that self-absorption effects are stronger for the later oxide, leading to enhanced Fe 2+ Lα peak shift toward longer wavelengths as the beam energy increases. First measurements performed on silicates have confirmed that enhanced self-absorption of FeLα occurs on Fe 2+ sites. The measurements consisted of plotting the FeLα peak position at a fixed beam energy (15 keV) against the total Fe concentration for two series of Fe 2+ - and Fe 3+ - bearing silicates. In a first step, these data have shown that both Fe 2+ Lα and Fe 3+ Lα peaks shift continuously toward longer wavelengths as the Fe concentration increases, with enhanced shifts for Fe 2+ Lα. For silicates containing only Fe 2+ or Fe 3+ , no effects of the site geometry were detected on the variations of the FeLα peak position. A second set of plots has shown the variations of the peak position relative to the previous Fe 2+ -Fe 3+ curves of step 1, as a function of the nominal Fe 3+ /ΣFe, for a series of reference minerals (hydrated and non-hydrated) and basaltic glasses. Data from chain and sheet silicates (e.g., pyroxenes, amphiboles, micas) exhibited strong deviations compared to other phases (e.g., garnets, Al-rich spinels, glasses), due to reduced self-absorption of FeLα. Intervalence-charge transfer (IVCT) mechanisms between Fe 2+ and Fe 3+ sites may be the origin of these deviations. These crystal-structure effects limit the accuracy of the method for mixed Fe 2+ -Fe 3+ valence silicates. Precisions achieved for further Fe 3+ /ΣFe measurements strongly depend on the total Fe concentration. For basaltic glasses containing an average of 8 wt% Fe and 10% Fe 3+ /ΣFe, the precision is about ±2% (absolute). For low Fe concentrations (below 3.5 wt%), the uncertainty in the peak position measured by the EMP spectrometers leads to error bars that are similar to with the separation of the curves fitted to the Fe 2+ and Fe 3+ plots, which is propagated as prohibitive lack of precision for Fe 3+ /ΣFe (>70% relative). A major limitation of microbeam methods in general deals with beam damage. This aspect has been carefully studied for basaltic glasses, and optimal beam conditions have been established (in general, electron doses higher than those corresponding to 130 nA and 30 μm beam diameter should be avoided to prevent large beam induced oxidation phenomena). Additional work, in progress, concerns: (1) other beam-sensitive phases such as hydrated glasses; and (2) minerals in which FeLα is affected by large matrix effect corrections (e.g., Cr- and Ti-rich oxides where FeLα is strongly absorbed), for which the self-absorptioninduced shift of FeLα is different from that of common silicates and glasses.

Pink nanofibers were extracted from rose quartz from 29 different pegmatitic and massive vein localities throughout the world. Their width varied from 0.1 to 0.5 μm. On the basis of optical absorption spectra of the fibers and the initial rose quartz, we conclude that these nanofibrous inclusions are the cause of coloration of massive rose quartz worldwide. These fibers do not occur in the rare, euhedral variety of pink quartz. Redox and heating experiments showed that the pink color of the fibers is due to Fe-Ti intervalence charge transfer that produces an optical absorption band at 500 nm. Based on the XRD patterns and characteristics of pleochroism, the best match for these inclusions is dumortierite. However, FTIR and Raman spectra consistently did not exactly match the standard dumortierite patterns, suggesting that this fibrous nano-phase may not be dumortierite itself, but rather a closely related material.

Cathodoluminescence (CL) spectrometry represents a promising technique for the analysis of trace-element concentrations and distributions in minerals. However, a higher precision and a standardization of the recording conditions are required to use CL spectral data quantitatively. A significant step towards a more quantitative treatment of CL spectra is presented in this study. A procedure to correct the spectra for the various efficiencies as a function of the wavelength of the CL detector is proposed using low-pressure mercury-vapor and quartz-iodine lamps. CL spectra presented in this study are thus corrected for the system response. Apatite CL spectra, which are commonly composed of two broad bands centered at 3.5 and 2.2 eV, are deconvoluted to isolate component bands and determine their areas. The crystallographic control by prismatic or basal sections of apatite on spectral intensities is significant and only prismatic sections should be used. Signal decrease associated with electron bombardment (electron beam aging) is exponential and appears drastic in the first hundred seconds but continues even after 15 minutes of beam bombardment. All observed CL bands could be correlated with a specific activator [rare earth elements (REE) or manganese]. The 3.5 eV band is composed of three bands at 3.59 eV, 3.29 eV, and 2.87 eV. Ion microprobe results and comparison between CL and photoluminescence data support Ce 3+ activation for the origin of these bands. The relationship between CL band intensity and REE concentration measured by ion microprobe analysis demonstrates that CL also can provide semi-quantitative data for Gd 3+ , Ce 3+ , Dy 3+ , and Sm 3+ when recording conditions are strictly controlled.

The hydroxide contents of spessartine-almandine garnets from the Rutherford no. 2 (Virginia), Himalaya (California), and George Ashley Block (California) pegmatites were determined by infrared spectroscopy. The hydroxide content of garnet increases from the wall zone to the core zone of the Rutherford no. 2 and Himalaya pegmatites, consistent with increasing H 2 O activity during pegmatite crystallization. However, the absolute OH contents differ by about two orders of magnitude for these two suites of garnets, possibly due to the elevated Ca content of the Rutherford no. 2 and the differences in the depth of emplacement. The garnets from the George Ashley Block show significant excursions from this correlation at the positions within the pegmatite where Kleck and Foord (1999) identified disruptions in major- and minor-element trends that they associated with re-injections of magma and subsequent flushing of the dike system. Ease of measurement as well as a relative amplified sensitivity compared to the Mn and Fe trends, make hydrogen an excellent tracer for the evolution of granitic pegmatites.

Ab initio NMR gauge-including atomic orbital (GIAO) calculations were used to constrain assignments of resonances in 27 Al NMR spectra of F-bearing alkali aluminosilicate glasses. The effect of bond angles within the range 126-150° on the chemical shift was investigated using cluster models of next-nearest atoms that are charge balanced by hydrogen atoms. GIAO calculations used geometries obtained through optimization at fixed Al-O-Si bond angles. The calculated peak positions for all of the 4-fold coordinated Al species yielded calculated 27 Al NMR peak positions in good agreement with the experimental data, suggesting that any or all of the species AlF 4 - , AlF 3 O(SiH 3 ) - , AlF 2 O 2 (SiH 3 ) 2 - , and AlFO 3 (SiH 3 ) 3 - may be present. Three of the investigated 5-fold coordinated species AlF 5 2- , AlF 3 O 2 (SiH 3 ) 2 2- , and AlF 2 O 3 (SiH 3 ) 3 2- fit the experimental requirements well, whereas the remaining 5-fold coordinated species that were tested [AlF 4 O(SiH 3 ) 2- , and AlFO 4 (SiH 3 ) 4 2- ] did not.

Several micro-techniques (confocal laser-Raman microprobe, optical absorption micro-spectroscopy, high-resolution transmission electron microscopy, electron microprobe analysis) were employed in the detailed characterization of radiohaloes in biotites from two Variscan rocks from Germany. The studied biotites are intermediate members of the phlogopite-annite series with Mg/ Fe 2+ ratios in the range 1.6-1.0. Radiohaloes in biotite resulted from the impact of 4 He cores (α- particles) emitted from actinide-bearing inclusions. Monte Carlo simulations yielded α ( 238 U, 235 U, and 232 Th series) penetration ranges in biotite between 12.5 and 37.3 μm, which are in reasonable agreement with the observed radii of radiohaloes in natural biotites. The coloration pattern of a radiohalo closely correlates with the calculated distribution pattern of point defects generated in displacive events. Calculated point defect densities in the range from < 10 -5 to at most 10 -2 dpa (displacements per lattice atom) suggest that there are only scattered point defects in a mainly preserved biotite lattice. This is consistent with HRTEM studies that did not reveal any indication for initial volume amorphization in the haloes. However, general Raman band broadening and intensity loss suggest that the short-range order in radiohaloes is significantly disturbed. The darkened color of radiohaloes, when compared with the un-irradiated host biotite, is caused by increased light absorption over the complete visible range due to increased point defect density. No additional color centers were found, and the absorbances of the VI Fe 2+ , Fe 2+ -Fe 3+ , and Fe 2+ -Ti 4+ centers seem hardly to be changed. Both Raman and optical absorption spectra obtained from radiohaloes retain a clear orientational dependence. The results suggest that the formation of point defects rather than ionization is the main process causing the coloration of radiohaloes in natural biotites. The haloes represent an early stage of structural radiation damage, characterized by significantly disturbed short-range order but still widely preserved long-range order of the structure.

Geochemical study of the Alta stock and adjacent contact aureole rocks provides information concerning the source, composition, and physical-chemical conditions of infiltrating fluids. Special emphasis was given to boron (B) as a tracer of fluid-rock interactions due to the occurrence of borate minerals (ludwigite, kotoite, and szaibelyite) in skarn deposits around the stock. In addition, thin section alpha-track mapping implies significant B enrichments in fluid-altered minerals within the stock, stockwork veins and related selvages, igneous sills near the stock, contact skarns, and in marbles up to 500 m from the stock. Forsterite, clinohumite, lizardite, and malachite contain between 50 and 1200 ppm B. Diopside, calcite, clintonite, phlogopite, brucite, hedenbergite, tremolite, and other minerals host B to a lesser extent. Aureole B enrichments correlate well with major and other trace-element enrichments, and support existing models of element transport in magmatic fluids with lateral down-temperature flow. Large variations in mineral B concentrations reflect changes in B concentrations of these fluids through time. Mass-balance calculations indicate that magmatic fluids emanating from the Alta pluton could supply most B in the Alta aureole. It is estimated that the emplaced magma had an initial B concentration between 7-10 ppm; indicated exhalative losses of B from the pluton are on the order of 50%. We estimate that the exsolved fluids had a time-integrated B concentration of 160 ± 40 ppm, although much higher concentrations may have attended local borate mineralization.

Uranium-rich zircons from a Paleoproterozoic, high-grade deformation zone in the Fennoscandian Shield, central Sweden, show an almost complete resetting of the U-Pb system in early Phanerozoic time. A mylonitic gneiss in the deformation zone contains two types of highly discordant (>70%), U-rich zircons: large, brown, cloudy prisms, and small milky-white irregularly shaped grains. The gneiss also contains mostly clear prismatic zircon of lower U content with mildly discordant to concordant U-Pb ages. Laser Raman spectroscopy reveals that the dark cathodoluminescent areas in brown zircons have a highly metamict crystal structure, whereas the structures of both the dark cathodoluminescent milky-white grains and the bright cathodoluminescent clear prisms have higher degrees of crystallinity. Age dates obtained by U-Pb SIMS analysis of 40 zircons of the three types described above range continuously from concordant at 1871 ± 11 Ma to 98% discordant at 384 ± 15 Ma. The strongly discordant zircons clearly have suffered severe disturbance at about the time of the Caledonian orogeny. However, Caledonian metamorphic temperatures and pressures in this region did not exceed 150-200 °C and 1-3 kbar, too low to strongly disturb the U-Pb systematics in nonmetamict zircon by thermal means alone. Independent evidence indicates that saline fluids were circulating in the Paleoproterozoic basement rocks at this time, possibly driven by hydrological gradients generated in front of the encroaching Caledonian orogenic wedge. These low-temperature saline fluids are inferred to be responsible for causing both strong Pb loss in the mostly metamict brown zircons via a diffusive process, and the formation of small milky-white zircon via a lowtemperature recrystallization or dissolution/re-precipitation process.

There are both natural and experimental observations of the coexistence of three pyroxenes- orthopyroxene (Opx) + pigeonite (Pig) + augite (Aug)-with plagioclase (Pl). Commonly, the assemblage occurs as an intermediate product in the following fractionation trend of a mafic magma: Opx + Aug + Pl → Opx + Aug + Pig + Pl → Aug + Pig + Pl. To clarify the phase-equilibria constraints on the existence of this mineral assemblage, we have graphically analyzed the change in topology of an isobaric-isoplethic section, Ol-Aug-Pl-Qtz [with fe = 25-50%, where fe = Fe/(Fe + Mg), and An = 50%], arising from an increase in the fe-value of silicate liquid. The analysis shows that the stability field of the mineral assemblage Opx + Aug + Pig + Pl is restricted, and can only crystallize in the interval between two invariant points-T 1 4 (Ol + Opx + Pig + Aug + Pl + L) and T 2 4 (Qtz + Opx + Pig + Aug + Pl + L)-that emerge successively during expansion of a liquidus volume of pigeonite within the isobaric-isoplethic section Ol-Aug-Pl-Qtz. At fe-values lower than T 1 4 , the 3- pyroxene assemblage cannot exist due to the absence of a contact surface between the primary volumes of plagioclase and pigeonite. At fe-values greater than T 2 4 , the assemblage is unstable due to separation of the primary volumes of orthopyroxene and augite.

Potassic white micas were synthesized in the K 2 O-MgO-Al 2 O 3 -SiO 2 -H 2 O system along the pseudobinary join muscovite-aluminoceladonite (mu-Alcel). Composition of run products as measured by electron microprobe analysis are in the range mu 89 -Alcel 11 to mu 01 -Alcel 99 . Cell parameters were determined on powder samples by full-profile Rietveld refinement, using both a single-polytype and a multi-polytype model. The results of both analysis models are in full agreement, and show that the phengite cell parameters have a distinct dependence on the celadonite content: the c parameter shows a monotonic decrease over the full compositional range, whereas the a and b parameters both increase in the Alcel 0 -Alcel 60 range but decrease in the Alcel 60 -Alcel 100 range. The monoclinic b angle decreases slightly with increasing celadonite content. The overall behavior of the cell parameters indicates a decrease of the ditrigonal distortion of the tetrahedral 6-rings, and an increased trioctahedral character of the structure at high celadonite compositions. The molar volume along the solid solution join shows a maximum at about Alcel 30 . Molar volume vs. composition can be fitted by a symmetric function for the excess volume yielding a molar volume for end member aluminoceladonite of 13.957 ± 0.006 J/bar, for muscovite 14.076 ± 0.004 J/bar, and a symmetric positive deviation from ideal volumes of mixing with W = 0.198 ± 0.025 J/bar, and r 2 = 0.941. The use of an asymmetric excess volume function does not significantly improve the fit quality (r 2 = 0.945).

The phase transitions in deuterated lawsonite were investigated with high-resolution, time-offlight neutron diffraction between 2 and 500 K. From the analysis of spontaneous strain data, the thermodynamics of the phase transition at 273 K are not changed by the deuteration process. Shifts in atomic positions with temperature indicate continuous changes for a framework oxygen and for one of the deuterium atoms, whereas for the other deuterium atom, a more discontinuous behavior was observed in the average structure. Comparison of O···D and O···O bond lengths with IR data from a non-deuterated lawsonite permits a detailed analysis of assignments of O-H stretching modes

Stokes’ viscometry combined with in situ X-ray radiographic observation, using the 6-8 type multi-anvil press and synchrotron radiation, has been applied to the viscosity measurement of the Fe- FeS melt up to pressures of 7 GPa. The viscosity is found to be about 2 × 10 -2 Pa-s at 5 to 7 GPa and temperatures about 1350 K, in marked contrast to previous viscosity measurements, which showed high viscosity, 0.5 to 14 Pa-s, at 2 to 5 GPa (LeBlanc and Secco 1996). Our viscosity data, however, is consistent with all other evidence, which include 1 atm viscosity data, X-ray structure analysis, and ab initio simulations. Recent viscosity measurements (Dobson et al. 2000) also showed the viscosity of Fe-FeS melt to be about 10 -2 Pa-s at 2.5 GPa. Thus, we are confident that the viscosity of the Fe- FeS melt is close to a typical value (10 -2 Pa-s) of viscosity for liquid metal even at high pressures.