Polytype distribution of circumstellar silicon carbide: microstructural characterization by transmission electron microscopy

Silicon carbide (SiC) is a particularly interesting species of presolar grain because it is known to form on the order of a hundred different polytypes in the laboratory, and the formation of a particular polytype is sensitive to growth conditions. Astronomical evidence for the formation of SiC in expanding circumstellar atmospheres of asymptotic giant branch (AGB) carbon stars is provided by infrared (IR) studies. However, identification of the crystallographic structure of SiC from IR spectra is controversial. Since >95% of the presolar SiC isolated from meteorites formed around carbon stars, a determination of the structure of presolar SiC is, to first order, a direct determination of the structure of circumstellar SiC. We therefore determined the polytype distribution of presolar SiC from the Murchison CM2 carbonaceous meteorite using analytical and high-resolution transmission electron microscopy (TEM). High-resolution lattice images and electron diffraction of 508 individual SiC grains demonstrate that only two polytypes are present, the cubic 3C (β-SiC) polytype (79.4% of population by number) and the hexagonal 2H (α-SiC) polytype (2.7%). Intergrowths of these two polytypes are relatively abundant (17.1%). No other polytypes were found. A small population of one-dimensionally disordered SiC grains (0.9%), whose high density of stacking faults precluded classification as any polytype, was also observed. The presolar origin of 2H α-SiC is unambiguously established by tens-of-nanometers-resolution secondary ion mass spectroscopy (NanoSIMS). Isotopic maps of a TEM-characterized 2H α-SiC grain exhibit non-solar isotopic compositions of ¹²C/¹³C = 64 ± 4 and ¹⁴N/¹⁵N = 575 ± 24. These measurements are consistent with mainstream presolar SiC thought to originate in the expanding atmospheres of AGB carbon stars. Equilibrium condensation calculations together with inferred mineral condensation sequences predict relatively low SiC condensation temperatures in carbon stars. The laboratory observed condensation temperatures of 2H and 3C SiC are generally the lowest of all SiC polytypes and fall within the predictions of the equilibrium calculations. These points account for the occurrence of only 2H and 3C polytypes of SiC in circumstellar outflows. The 2H and 3C SiC polytypes presumably condense at different radii (i.e., temperatures) in the expanding stellar atmospheres of AGB carbon stars.     

The breakdown of potassium feldspar at high water pressures

The equilibrium position of the reaction between sanidine and water to form “sanidine hydrate” has been determined by reversal experiments on well characterised synthetic starting materials in a piston cylinder apparatus. The reaction was found to lie between four reversed brackets of 2.35 and 2.50 GPa at 450 °C, 2.40 and 2.59 GPa at 550 °C, 2.67 and 2.74 GPa at 650 °C, and 2.70 and 2.72 GPa at 680 °C. Infrared spectroscopy showed that the dominant water species in sanidine hydrate was structural H₂O. The minimum quantity of this structural H₂O, measured by thermogravimetric analysis, varied between 4.42 and 5.85 wt% over the pressure range of 2.7 to 3.2 GPa and the temperature range of 450 to 680 °C. Systematic variation in water content with pressure and temperature was not clearly established. The maximum value was below 6.07 wt%, the equivalent of 1 molecule of H₂O per formula unit. The water could be removed entirely by heating at atmospheric pressure to produce a metastable, anhydrous, hexagonal KAlSi₃O₈ phase (“hexasanidine”) implying that the structural H₂O content of sanidine hydrate can vary. The unit cell parameters for sanidine hydrate, measured by powder X-ray diffraction, were a = 0.53366 (±0.00022) nm and c = 0.77141 (±0.00052) nm, and those for hexasanidine were a = 0.52893 (±0.00016) nm and c = 0.78185 (±0.00036) nm. The behaviour and properties of sanidine hydrate appear to be analogous to those of the hydrate phase cymrite in the equivalent barium system. The occurrence of sanidine hydrate in the Earth would be limited to high pressure but very low temperature conditions and hence it could be a potential reservoir for water in cold subduction zones. However, sanidine hydrate would probably be constrained to granitic rock compositions at these pressures and temperatures. Received: 6 May 1997 / Accepted: 2 October 1997     

Water and the density of silicate glasses

A review of published and newly measured densities for 40 hydrous silicate glasses indicates that the room-temperature partial molar volume of water is 12.0 ± 0.5 cm³/mol. This value holds for simple or mineral compositions as well as for complex natural glasses, from rhyolite to tephrite compositions, prepared up to 10–20 kbar pressures and containing up to 7 wt% H₂O. This volume does not vary either with the molar volume of the water-free silicate phase, with its degree of polymerization or with water speciation. Over a wide range of compositions, this constant value implies that the volume change for the reaction between hydroxyl ions and molecular water is zero and that, at least in glasses, speciation does not depend on pressure. Consistent with data from Ochs and Lange (1997, 1999), systematics in volume expansion for SiO₂–M₂O systems (M=H, Li, Na, K) suggests that the partial molar thermal expansion coefficient of H₂O is about 4 × 10⁻⁵ K⁻¹ in silicate glasses. Received: 30 June 1999 / Accepted: 5 November 1999     

A TEM investigation of natural metamict zircons: structure and recovery of amorphous domains

The nature of the amorphous regions and their recovery processes in two natural metamict zircon samples from Sri Lanka have been studied by high resolution and analytical transmission electron microscopy. Samples untreated and annealed at different temperatures were investigated. Nanoprobe analyses on untreated samples and samples annealed at 1000 K show that within experimental uncertainties, no chemical segregation occurred. In samples annealed at higher temperatures (≥1100 K) recovery occurs in a two-stage process and leads to different microstructures, which depended on the initial amount of metamictization. In highly amorphized samples, recrystallization starts at 1200 K. Randomly oriented ZrO₂ grains embedded in a silica-rich matrix are detected. At higher temperature (16 h at 1600 K), the assemblage transforms into a polygonal texture of small zircon grains. Some untransformed zirconia grains and pockets of silica-rich glass are still present, however. In partially metamict samples, recovery starts at 1100 K. The small surviving oriented zircon domains grow at the expense of the surrounding amorphous material. At 1200 K, new zirconia grains nucleate with random orientations. After 1 h annealing at 1400 K, the zircon structure is restored and the microstructure coarse-grained. The proportion of crystalline zirconia and silica-rich glass has dramatically decreased. Received: 15 November 1999 / Accepted: 1 March 2000     

Pb diffusion in rutile

Diffusion of Pb was measured in natural and synthetic rutile under dry, 1 atmosphere conditions, using mixtures of Pb titanate or Pb sulfide and TiO₂ as the sources of diffusant. Pb depth profiles were then measured with Rutherford Backscattering Spectrometry (RBS). Over the temperature range 700–1100 °C, the following Arrhenius relation was obtained for the synthetic rutile: D=3.9 × 10⁻¹⁰exp(−250 ± 12 kJ mol⁻¹/RT) m²s⁻¹. Results for diffusion in natural and synthetic rutile were quite similar, despite significant differences in trace element compositions. Mean closure temperatures calculated from the diffusion parameters are around 600 °C for rutile grains of ∼100 μm size. This is about 100 °C higher than rutile closure temperature determinations from past field-based studies, suggesting that rutile is more resistant to Pb loss through volume diffusion than previously thought. Received: 28 June 1999 / Accepted: 29 December 1999     

Phase relations in the system fayalite–magnetite at high pressures and temperatures

The phase relations in the Fe₂SiO₄–Fe₃O₄ binary system have been determined between 900 and 1200 °C and from 2.0 to 9.0 GPa. At low to moderate pressures magnetite can accommodate significant Si, reaching X_Fe2SiO₄=0.1 and 0.2 at 3.0 and 5.0 GPa respectively, with temperature having only a secondary influence. At pressures below 3.5 GPa at 900 °C and 2.6 GPa at 1100 °C magnetite-rich spinel coexists with pure fayalite. This assemblage becomes unstable at higher pressures with respect to three intermediate phases that are spinelloid polytypes isostructural to spinelloids II, III and V in the Ni-aluminosilicate system. The phase relations between the spinelloid phases are complex. At pressures above ≈8.0 GPa at 1100 °C, the spinelloid phases give way to a complete spinel solid solution between Fe₃O₄ and Fe₂SiO₄. The presence of small amounts of Fe³⁺ stabilises the spinel structure to lower pressures compared to the Fe₂SiO₄ end member. This means that the fayalite–γ-spinel transition is generally unsuitable as a pressure calibration point for experimental apparatuses. The molar volumes of the spinel solid solutions vary nearly linearly with composition, having a small negative deviation from ideal behaviour described by W_v=−0.15(6) cm³. Extrapolation yields V°₍₂₉₈₎ = 41.981(14) cm³ for the Fe₂SiO₄-spinel end member. The cell parameters and molar volumes of the three spinelloid polytypes vary systematically with composition. Cation disorder is an important factor in stabilising the spinelloid polytypes. Each polytype exhibits a particular solid solution range that is directly influenced by the interplay between its structure and the cation distributions that are energetically favourable. In the FeO–FeO_1.5–SiO₂ ternary system Fe₇SiO₁₀ (“iscorite”) coexists with the spinelloid phases at intermediate pressures on the SiO₂-poor, or Fe³⁺-poor side of the Fe₂SiO₄–Fe₃O₄ join. On the SiO₂ and Fe³⁺-rich side of the join, orthopyroxene or high-P clinopyroxene coexists with the spinelloids and spinel solid solutions. The assemblage pyroxene+spinel+SiO₂ is stable over a wide range of bulk composition. The stability of spinelloid III is of particular petrologic interest since this phase has the same structure as (Mg,Fe)₂SiO₄–wadsleyite, indicating that Fe³⁺ can be easily incorporated in this important phase in the Earth's transition zone, in addition to silicate spinel. This has important implications for the redox state of the Earth's transition zone and for the depth at which the olivine to spinel transition occurs in the mantle, potentially leading to a shift in the “410 km” seismic discontinuity to shallower depths depending on the prevailing redox state. In addition, a coupled tetrahedral substitution of Fe³⁺+OH for Si+O could provide a further mechanism for the incorporation of H₂O in wadsleyite. Received: 10 January 2000 / Accepted: 12 May 2000     

Carbonation and decarbonation of eclogites: the role of garnet

Carbonates are potentially significant hosts for primordial and subducted carbon in the Earth's mantle. In addition, the coexistence of carbonate with silicates and reduced carbon (diamond or graphite), allows constraints to be placed on the oxidation state of the mantle. Carbonate-silicate-vapor reactions control how carbonate + silicate assemblages may form from carbon-bearing vapor + silicate assemblages with increasing pressure. In olivine-bearing rocks such as peridotite, considered the dominant rock type in the upper mantle, the lowest-pressure carbonate-forming reactions involve olivine (±clinopyroxene) reacting with CO₂ (e.g., Wyllie et al. 1983). In eclogitic rocks, the essential mineral assemblage is omphacitic clinopyroxene + garnet, without olivine. Therefore, alternative carbonate-forming reactions must be sought. The carbonation of clinopyroxene via the reaction dolomite + 2 coesite = diopside + 2 CO₂ was studied experimentally by Luth (1995). The alternative possibility that garnet reacts with CO₂ is explored here by determining the location of the reaction 3 magnesite + kyanite + 2 coesite = pyrope + 3 CO₂ between 5 and 11 GPa in multi-anvil apparatus. At the temperatures ≥1200 °C, carbonation of eclogitic rocks with increasing pressure will proceed initially by reaction with clinopyroxene, because the pyrope-carbonation reaction lies at higher pressures for a given temperature than does the diopside-carbonation reaction. Diluting the pyrope component of garnet and the diopside component of clinopyroxene to levels appropriate for mantle eclogites does not change this conclusion. At lower temperatures, appropriate for “cold” slabs, it is possible that the converse situation will hold, with initial carbonation proceeding via reaction with garnet, but this possibility awaits experimental confirmation. Decarbonation of an eclogite under “normal mantle” geothermal conditions by a decrease in pressure, as in an ascending limb of a mantle convection cell, would be governed by the formation of clinopyroxene + CO₂. At higher pressure than this reaction, any CO₂ produced by the breakdown of magnesite reacting with kyanite and coesite would react with clinopyroxene to produce dolomite + coesite. Release of CO₂ from eclogite into mantle peridotite would form carbonate at sub-solidus conditions and produce a dolomitic carbonate melt if temperatures are above the peridotite-CO₂ solidus. Received: 4 May 1998 / Accepted: 23 December 1998     

Kinetics of dehydration of Ca-montmorillonite

Isothermal thermogravimetric experiments have been carried out to determine the reaction kinetics of the dehydration processes in fuller's earth, a natural Ca-montmorillonite. Dehydration in swelling clays is a complex reaction, and analysis of the thermogravimetric data using empirical rate equations and time-transformation analysis reveals that the nature of the rate controlling mechanism is dependent upon both the temperature regime of the sample as well as the extent of reaction. For fuller's earth, we find that the dehydration kinetics are dominated by a nucleation and growth mechanism at low temperatures and fractions transformed (stage I), but above 90 °C the last stages of the reaction are diffusion controlled (stage II). The activation energy for dehydration during stage I is around 35 kJ · mol⁻¹, whereas the removal of water during stage II requires an activation energy of around 50 kJ · mol⁻¹. These two stages of dehydration are associated with primary collapse of the interlayer (stage I) and movement of water that is hydrated to cations within the interlayer (stage II). Received: 28 August 1998 / Revised, accepted: 27 January 1999     

Al zoning in orthopyroxene in a sapphirine quartzite: evidence for >1120 °C UHT metamorphism in the Napier Complex, Antarctica, and implications for the entropy of sapphirine

A unique sapphirine + orthopyroxene + quartz granulite from Mt. Riiser-Larsen in the Tula Mountains of Enderby Land, East Antarctica, preserves two generations of coarse and texturally equilibrated orthopyroxene and sapphirine coexisting with quartz. Initial subhedral orthopyroxene porphyroblasts retain core compositions enriched in Al₂O₃ (12.2 ± 0.5 wt%) compared with their rims and finer orthopyroxene (9.6 ± 0.5 wt% Al₂O₃) that forms granoblastic textures with sapphirine. Sapphirine and quartz also form symplectites on and along cleavage planes within orthopyroxene. These compositional and textural features are consistent with the reaction [2MgAl₂SiO₆=Mg₂Al₄ SiO₁₀ + SiO₂] leading to the formation of sapphirine + quartz at the expense of aluminous orthopyroxene. Calculations in the MAS and FMAS systems and theoretical considerations involving the phases enstatite, sapphirine, sillimanite, quartz and cordierite indicate that the reaction above progresses from left to right with decreasing temperature in the orthopyroxene + sapphirine + quartz field, at pressures of ca. 8–10 kbar. The temperature difference required to account for the ca. 2.5–3 wt% decrease in Al₂O₃ in orthopyroxene is at least 60–80 °C, and implies peak temperatures for the initial assemblage of at least 1120 °C if the second granoblastic assemblage equilibrated at 1040 °C, the P–T conditions required by the sapphirine + quartz association and other P–T-sensitive assemblage indicators in the Napier Complex. It is not possible to distinguish whether the two assemblages are simply related by cooling and re-equilibration or reflect a polyphase evolution involving the superposition of a second UHT event on an earlier, even higher temperature, UHT metamorphism. Preliminary thermodynamic modelling of the reaction above incorporating the observed range in orthopyroxene Al₂O₃ zoning indicates that present estimates for the entropy of high-temperature sapphirine are potentially too high by 15–18% compared with sapphirine entropy estimates that are consistent with MAS system experiments. The Mt. Riiser-Larsen sapphirine–quartz rocks preserve the first definitive record of regional metamorphic temperatures in excess of 1120 °C in the Napier Complex, or indeed any UHT granulite terrain worldwide. Similarly high peak temperatures may be retrieved from detailed studies of sapphirine–quartz granulites from other regions, further expanding the thermal realm of crustal metamorphism, but progress will critically depend on the experimental acquisition of new entropy data for sapphirine. Received: 3 September 1998 / Accepted: 8 November 1999     

The correlation of reaction and isotope fronts and the mechanism of metamorphic fluid flow

 Infiltration of a metabasite sill from Islay, Scotland by an H₂O-CO₂ fluid caused (1) modification of δ¹⁸O and (2) carbonation at the sill margins. Maps of δ¹⁸O and reaction progress were constructed from a 20 × 47.7 metre sample grid across the sill. The grid consisted of 300 samples, spaced at m, dm and cm intervals, many of which were analysed for both δ¹⁸O and reaction progress. The δ¹⁸O was determined by laser fluorination of whole rock silicate powders and reaction progress was determined by rapid field-based measurement of % calcite (“fizz-o-meter”, Skelton et al. 1995). Reaction and isotope fronts outlined tube-like features that emanate from the sill margin and discrete nodes that, although detached from the sill margin in two dimensions, are thought to represent sections through similar tubes in three dimensions. We envisage that these protrusions are the fossil record of metamorphic “fluid pathways” whereby fluid permeated the sill. Isotope and reaction fronts are found to correlate spatially as predicted by a modified form of the chromatographic equation which describes this envisaged geometry, that is where isotopic and reactive transport in the fluid phase are facilitated by advection along specific fluid pathways and transverse diffusion in the surrounding rock. These fluid pathways consist of bundles of anastomosing grain boundary channels or micro-cracks, which are thought to propagate through transient cyclic infiltration, reaction, porosity enhancement and fracturing. This mechanism is self-perpetuating and accentuates random perturbations at the sill margin to form the observed tubes. We argue that this is the earliest stage of the infiltration process which has affected metabasites of the SW Scottish Highlands and that subsequent shear deformation of the reacted rims of these pathways, has caused their re-orientation and juxtaposition to form the reacted sill margins described by Skelton et al. (1995). Received: 17 February 1998 / Accepted: 6 December 1999     

The normal grain growth behaviour of nominally pure calcitic aggregates

The normal grain growth behaviour of four different, but all nominally pure, calcite powders (99%+ analytic grade calcite, 99.7% chalk, 99.97% crushed Iceland Spar, 99.95%+ chelometric grade calcite) has been investigated as a function of temperature (550, 600, 650, 700 °C) and confining pressure (100, 190 MPa) under both “dry” and hydrostatic (P _fluid = P _total) conditions. The initial particle size of both the analytic grade and chelometric grade calcite was about 5 μm, and that of the chalk was about 3 μm, while the experiments on the Iceland Spar were conducted on powders of three different initial particle sizes (3.4, 7.5, 38.5 μm). On each material, at each pressure/temperature condition 6 to 15 experiments, equally spaced in log time from 15 minutes to 50 days, were conducted. Under dry conditions all four materials recrystallized to aggregates which contained less than 2% porosity and which had a grain size of between 4 and 20 μm (depending on the initial particle size). Subsequently the aggregates coarsened by normal grain growth, with the kinetics of the growth process being controlled by the rate at which the grain boundaries could drag the residual pores with them as they migrated. Under nominally identical conditions both the mechanism and rates of pore drag differed greatly for the different materials, implying that this process is highly sensitive to trace solute impurity concentrations. This sensitivity renders the task of providing a systematic account of dry calcite grain growth kinetics highly problematic. Under hydrostatic conditions all the powders followed the same normal grain growth kinetics in which the growth process was rate-controlled by diffusion through the pore fluid on the grain boundaries. An activation enthalpy of 162.6 kJ mol⁻¹ and an activation volume of 34.35 cm³ mol⁻¹ was obtained for this process. Received: 23 May 1996 / Accepted: 8 July 1997     

Garnet-bearing granite from the Třebíč pluton, Bohemian Massif (Czech Republic)

Summary Garnet occurs as a significant mineral constituent of felsic garnet-biotite granite in the southern edge of the Třebíč pluton. Two textural groups of garnet were recognized on the basis of their shape and relationship to biotite. Group I garnets are 1.5–2.5 mm, euhedral grains which have no reaction relationship with biotite. They are zoned having high X_Mn at the rims and are considered as magmatic. Group II garnets form grain aggregates up to 2.5 cm in size, with anhedral shape of individual grains. The individual garnet II grains are usually rimmed by biotite and have no compositional zoning. The core of group I garnets and group II garnets contains 67–80 mol% of almandine, 5–19 mol% of pyrope, 7–17 mol% of spessartine and 2–4 mol% of grossular. Biotite occurs in two generations; both are magnesian siderophyllites with Fe/(Fe + Mg) = 0.50–0.69. The matrix biotite in granites (biotite I) has high Ti content (0.09–0.31 apfu) and Fe/(Fe + Mg) ratio between 0.50 and 0.59. Biotite II forms reaction rims around garnet, is poor in Ti (0.00–0.06 apfu) and has a Fe/(Fe + Mg) ratio between 0.61 and 0.69. The textural relationship between biotite and garnet indicates that garnet reacted with granitic melt to form Ti-poor biotite and a new granitic melt, depleted in Ti and Mg and enriched in Fe and Al. In contrast to the host durbachites (hornblende-biotite melagranites), which originated by mixing of crustal melts and upper mantle melts, the origin of garnet-bearing granites is related to partial melting of the aluminium-rich metamorphic series of the Moldanubian Zone.     

Petrologic and geochemical investigation of carbonates in peridotite xenoliths from northeastern Tanzania

Primary carbonates in peridotite xenoliths from the East African Rift in northeastern Tanzania occur as intergranular patches with accessory minerals (olivine and spinel), as patches with accessory magmatic minerals (nepheline), and as round monomineralic inclusions in primary olivine grains. All are characterized by calcitic compositions (Ca/Ca + Mg + Fe from 0.83 to 0.99), extremely low SiO₂ + Al₂O₃ + Na₂O + K₂O, low trace element abundance [total rare-earth element (REE) abundance <25 ppm], uniform extinction, and lack of reaction textures with the host xenolith. Calculated Fe–Mg exchange coefficients between carbonate and primary olivine indicate disequilibrium in most samples. Combined with the lack of significant reaction textures, this suggests that the carbonates were introduced shortly before or during eruption of the host magma. A global compilation of electron microprobe analyses of mantle-derived carbonates (in xenoliths, xenocrysts, and megacrysts) reveals compositional clusters near end member calcite, end member magnesite, and stoichiometric dolomite. Eutectic liquid compositions are less common, suggesting that many carbonate inclusions reported worldwide may be crystalline precipitates. Likewise, the calcites in this study are not interpreted to represent quenched carbonatitic melts, but are interpreted instead to be crystalline cumulates from such melts. These inferences are consistent with recent experiments, which show that carbonatitic melts cannot become more calcitic than CaCO₃∼80 wt%. Low trace element abundance may be a diagnostic feature of cumulate carbonate, and in combination with petrography and major element composition, serve to distinguish it from quenched carbonated liquid. Received: 30 July 1999 / Accepted: 5 February 2000     

Growth, annealing and recrystallization of zircon and preservation of monazite in high-grade metamorphism: conventional and in-situ U-Pb isotope, cathodoluminescence and microchemical evidence

Zircon and monazite from granulite- to amphibolite-facies rocks of the Vosges mountains (central Variscan Belt, eastern France) were dated by ion-microprobe and conventional U-Pb techniques. Different granulites of igneous (so-called leptynites) and sedimentary origin (kinzigites) and their leucosomes were dated at 334.9 ± 3.6, 335.4 ± 3.6 and 336.7 ± 3.5 Ma (conventional age 335.4 ± 0.6 Ma). Subsequent growth stages of zircon were distinguished by secondary electron (SEM) and cathodoluminescence (CL) imaging: (1) subsolidus growth producing round anhedral morphologies and sector zoning; (2) appearance of an intergranular fluid or melt phase at incipient dehydration melting that first resulted in resorption of pre-existing zircons, followed by growth of acicular zircons or overgrowths on round zircons consisting of planar growth zoning; (3) advanced melting producing euhedral prismatic zircons with oscillatory zoning overgrowing the sector zones. Two further lithologies, the Kaysersberg granite and the Trois-Epis units, were both formerly considered as migmatites. The intrusion of the Kaysersberg granite was dated at 325.8 ± 4.8 Ma. The Trois-Epis unit was found to be the product of volume recrystallization of a former granulite, which occurred under amphibolite-facies conditions 327.9 ± 4.4 Ma ago. The amphibolite-facies overprint of the Trois-Epis zircons led to the complete rejuvenation of most of the zircon domains by annealing and replacement/recrystallization processes. Annealing is assumed to occur in strained lattice domains, which are possibly disturbed by high trace element contents and/or large differences in decay damage between adjacent growth zones. Investigation of cathodoluminescence structures reveals that the replacement occurs along curved chemical reaction fronts that proceed from the surface towards the interior of the zircon. The monazite U-Pb system still records the age of high-grade metamorphism at around 335 Ma. The chemical reagent responsible for the rejuvenation of zircon obviously left the monazite unaffected. Received: 19 February 1998 / Accepted: 19 October 1998     

Stability of dense hydrous magnesium silicate phases in the systems Mg2SiO4-H2O and MgSiO3-H2O at pressures up to 27 GPa

We conducted high-pressure phase equilibrium experiments in the systems MgSiO₃ with 15 wt% H₂O and Mg₂SiO₄ with 5 wt% and 11 wt% H₂O at 20 ∼ 27 GPa. Based on the phase relations in these systems, together with the previous works on the related systems, we have clarified the stability relations of dense hydrous magnesium silicates in the system MgO-SiO₂-H₂O in the pressure range from 10 to 27 GPa. The results show that the stability field of phase G, which is identical to phase D and phase F, expands with increasing water contents. Water stored in serpentine in the descending cold slabs is transported into depths greater than 200 km, where serpentine decomposes to a mixture of phase A, enstatite, and fluid. Reaction sequences of the hydrous phases which appear at higher pressures vary with water content. In the slabs with a water content less than about 2 wt%, phase A carries water to a depth of 450 km. Hydrous wadsleyite, hydrous ringwoodite, and ilmenite are the main water reservoirs in the transition zone from 450 to 660 km. Superhydrous phase B is the water reservoir in the uppermost part of the lower mantle from 670 to 800 km, whereas phase G appears in the lower mantle only at depths greater than 800 km. In cold slabs with local water enrichment greater than 2 wt%, the following hydrous phases appear with increasing depths; phase A to 450 km, phase A and phase G from 450 km to 550 km, brucite, superhydrous phase B, and phase G from 550 km to 800 km, and phase G at depths greater than 800 km. Received: 4 August 1999 / Accepted: 1 March 2000     

The petrogenesis of Merensky Reef potholes at the Western Platinum Mine, Bushveld Complex: Sr-isotopic evidence for synmagmatic deformation

Potholes represent areas where the normally planar PGE-rich Merensky Reef of the upper Critical Zone of the Bushveld Complex transgresses its footwall, such geometric relationships being unusual in layered intrusions. The recognition of vertical dykes of Merensky pyroxenite in the footwall suggests downward collapse of crystal mush into pull-apart sites resulting from tensional deformation due to the loading effects of major new magma additions. In contrast, crosscutting anorthosite veins display physical and isotopic evidence of upward emplacement. The Merensky Reef and its footwall have distinct initial Sr-isotope ratios (R ₀ > 0.7066 and <0.7066, respectively), which may be used to constrain these processes related to pothole formation. Merensky Reef in potholes (R ₀ = 0.7069−0.7078) shows no isotopic evidence of assimilation of, or reaction with, footwall material. Discrete, discordant replacement bodies of anorthosite extend from the footwall lithologies to cross-cut the Merensky Reef and its hanging wall. The initial Sr-isotope ratio in these replaced rocks is totally reset to footwall values (R ₀ = 0.7066), and immediately adjacent stratiform lithologies are slightly modified towards footwall values. In contrast, Neptunian pyroxenitic (Merensky) dykes cross-cutting the footwall lithologies, with a large surface area to volume ratio, and low Sr content, do not display footwall-like Sr-isotope initial-ratios (R ₀ = 0.7077), and thus show no evidence for assimilation of or reaction with footwall material. Furthermore, pegmatoidal replacement pyroxenite (“replacement pegmatoid”), at the base of the Merensky Reef within potholes, has a high initial-ratio (R ₀ > 0.7071), and so models of pervasive metasomatism by footwall material are not applicable. This isotopic evidence indicates that there was no active interaction of footwall material with the overlying magma during, or after, the formation of Merensky Reef potholes, a basic tenet of existing pothole formation hypotheses involving footwall mass-transfer. In contrast, the isotopic data are entirely consistent with an extensional model for pothole formation, with the more radiogenic Merensky magma migrating laterally to fill extensional zones in the footwall layers. Received: 11 October 1997 / Accepted: 21 December 1998     

The formation of columnar fiber texture in wollastonite rims by induced stress and implications for diffusion-controlled corona growth

 The evolution of columnar fiber texture was studied in wollastonite reaction rims synthesized by the reaction calcite + quartz=wollastonite + CO₂. Experiments were performed at 850 to 950 °C at 100 MPa in dry CO₂ and were evaluated by scanning and transmission electron microscopy. Rim growth rates are interpreted as controlled by the diffusion of the SiO₂ component through the rims from the quartz–wollastonite to the wollastonite–calcite interface. The temperature dependence of rim growth rates yields an apparent activation energy of 314 ± 53 kJ mol⁻¹. The columnar fibrous wollastonite crystallizes at the quartz–wollastonite interface and comprises the largest parts of the rims. Ultimately, at the growth front strain contrast centers are present in the quartz. The strained volume extends about 200 nm into the quartz grains. We suggest that this might signify deformation of the quartz lattice due to wollastonite crystallization. Wollastonite fiber thickness was measured from TEM images along traverses that represent intermediate positions of the growth front during the experiments. The average thickness is in the 100–200 nm range. Fiber thickness increases with increasing growth temperature. At a given temperature, the thickness of the fibers at the growth front slightly decreases with time, i.e., the number of fiber tips per unit area in the growth front increases. The decrease of the fiber thickness is well fitted by a parabolic rate law. The generation of the columnar fiber texture is interpreted as an effect of induced stresses at the growth front, resulting from the volume increase due to the local reaction. This volume increase forces SiO₂ to diffuse along the growth front to the grain boundaries between the wollastonite fibers. These serve as fast diffusion pathways through the rims. The fiber thickness monitors the diffusion distances in the growth front and thus the height of the induced stress gradients. Since interface reactions are usually associated with volume changes, growth rates of reaction rims and zones in coronas are not only controlled by the diffusive mobility of the components but also by the volume restraints on the interface reactions. Received: 19 July 2002 / Accepted: 14 February 2003     

寿山叶蜡石矿床中叶蜡石的多型及其转变

叶蜡石具有1Tc和2M_1两种多型。寿山叶蜡石矿床中的叶蜡石不但具有普遍报道的2M_1型及1Tc与2M_1的混合型,而且具有同类矿床中少见的、但在本区却大量出现的1Tc型。本文对寿山叶蜡石多型变体的X射线衍射特征、镜下特征及化学成分特征作了写实性描述,指出本区叶蜡石存在着2M_1型向1Tc型转变的客观事实,并可能存在二者的中间类型;探讨了叶蜡石多型转变的条件,提出“表生溶液”是实现该转变的主要因素。     

Fe–Mg partitioning between ringwoodite and magnesiowüstite and the effect of pressure, temperature and oxygen fugacity

 The partitioning of Mg and Fe between magnesiowüstite and ringwoodite solid solutions has been measured between 15 and 23 GPa and 1200–1600 ^∘C using both Fe and Re capsule materials to vary the oxidation conditions. The partitioning results show a clear dependence on the capsule material used due to the variation in Fe³⁺ concentrations as a consequence of the different oxidation environments. Using results from experiments performed in Fe capsules, where metallic Fe was also added to the starting materials, the difference in the interaction parameters for the two solid solutions (W _FeMg ^mw−W _FeMg ^ring) is calculated to be 8.5±1 kJ mol⁻¹. Similar experiments performed in Re metal capsules result in a value for W _FeMg ^mw−W _FeMg ^ring that is apparently 4 kJ higher, if all Fe is assumed to be FeO. Electron energy-loss near-edge structure (ELNES) spectroscopic analyses, however, show Fe³⁺ concentrations to be approximately three times higher in magnesiowüstite produced in Re capsules than in Fe capsules and that Fe³⁺ partitions preferentially into magnesiowüstite, with K _{D Fe3+} ^ring/mw estimated between 0.1 and 0.6. Using an existing activity composition model for magnesiowüstite, a least–squares fit to the partitioning data collected in Fe capsules results in a value for the ringwoodite interaction parameter (W _FeMg ^ring) of 3.5±1 kJ mol⁻¹. The equivalent regular interaction parameter for magnesiowüstite (W _FeMg ^mw) is 12.1±1.8 kJ mol. These determinations take into account the Fe³⁺ concentrations that occur in both phases in the presence of metallic Fe. The free energy change in J mol⁻¹ for the Fe exchange reaction can be described, over the range of experimental conditions, by 912 + 4.15 (T−298)+18.9P with T in K, P in kbar. The estimated volume change for this reaction is smaller than that predicted using current compilations of equation of state data and is much closer to the volume change at ambient conditions. These results are therefore a useful test of high pressure and temperature equation of state data. Using thermodynamic data consistent with this study the reaction of ringwoodite to form magnesiowüstite and stishovite is calculated from the data collected using Fe capsules. Comparison of these results with previous studies shows that the presence of Fe³⁺ in phases produced in multianvil experiments using Re capsules can have a marked effect on apparent phase relations and determined thermodynamic properties. Received: 13 September 2000 / Accepted: 25 March 2001     

Characterization of NH4-phlogopite (NH4) (Mg3) [AlSi3O10] (OH)2 and ND4-phlogopite (ND4) (Mg3) [AlSi3O10] (OD)2 using IR spectroscopy and Rietveld refinement of XRD spectra

 A synthesis technique is described which results in >99% pure NH₄-phlogopite (NH₄) (Mg₃) [AlSi₃O₁₀] (OH)₂ and its deuterium analogue ND₄-phlogopite (ND₄) (Mg₃) [AlSi₃O₁₀] (OD)₂. Both phases are characterised using both IR spectroscopy at 298 and 77 K as well as Rietveld refinement of their X-ray powder diffraction pattern. Both NH₄ ⁺ and ND₄ ⁺ are found to occupy the interlayer site in the phlogopite structure. Absorption bands in the IR caused by either NH₄ ⁺ or ND₄ ⁺ can be explained to a good approximation using T_d symmetry as a basis. Rietveld refinement indicates that either phlogopite synthesis contains several mica polytypes. The principle polytype is the one-layer monoclinic polytype (1M), which possesses the space group symmetry C2/m. The next most common polytype is the two-layer polytype (2M ₁ ) with space group symmetry C2/c. Minor amounts of the trigonal polytype 3T with the space group symmetry P3₁12 were found only in the synthesis run for the ND₄-phlogopite. Electron microprobe analyses indicate that NH₄-phlogopite deviates from the ideal phlogopite composition with respect to variable Si/Al and Mg/Al on both the tetrahedral and octahedral sites, respectively, due to the Tschermaks substitution ^VIMg²⁺+^IVSi⁴⁺↔^VIAl³⁺+^IVAl³⁺ and with respect to vacancies on the interlayer site due to the exchange vector ^XII(NH₄)⁺+^IVAl³⁺↔^XII□+^IVSi⁴⁺. Received: 30 August 1999 / Accepted: 2 October 2000