Volume 97, Issue 1

The presence of boron in the structure of diamond is rare in nature, and even when present, reported values are ≤0.5 ppm. This study used various spectroscopic methods and time-of-fight (ToF-) SIMS to characterize and analyze for boron in natural type IIb blue diamonds, including the well-known Hope and the Blue Heart diamonds, and on one high-pressure, high-temperature annealed natural stone. Infrared spectroscopy measurements reveal uncompensated boron values as large as 1.72 ± 0.15 ppm, which is significantly higher than the previously reported maximum of 0.5 ppm. ToF-SIMS analyses gave spot total boron concentrations as high as 8.4 ± 1.1 ppm for the Hope diamond to less than 0.08 ppm in other blue diamonds. By comparison, a type Ia diamond did not show detectable boron. ToF-SIMS analyses revealed strong zoning of boron in some diamonds, which was confirmed by mapping the uncompensated boron using synchrotron infrared spectroscopy. This greater range of boron concentrations compared to previous studies might be explained by the larger number of natural diamonds analyzed here, 78, compared to <10 samples reported in the literature. The samples in this study are all gem-quality diamonds, including some Intense to Fancy-Deep blue diamonds; color intensity, however, only loosely correlates with the boron content. Boron is also likely responsible for the phosphorescence emissions of type IIb diamonds, in the red at 660 nm and in the blue-green at 500 nm. Our results are consistent with previous work suggesting that the emissions are caused by donor-acceptor pair recombination processes involving boron and other defects. The exact nature of the phosphorescence processes is still not fully understood, but likely involves complex steps of charge carrier trapping and detrapping.

Mejillonesite, ideally NaMg 2 (PO 3 OH)(PO 4 )(OH)· H 5 O 2 , is a new mineral approved by the CNMNC (IMA 2010-068). It occurs as isolated crystal aggregates in thin zones in fine-grained opal-zeolite aggregate on the north slope of Cerro Mejillones, Antofagasta, Chile. Closely associated minerals are bobierrite, opal, clinoptilolite-Na, clinoptilolite-K, and gypsum. Mejillonesite forms orthorhombic, prismatic, and elongated thick tabular crystals up to 6 mm long, usually intergrown in radiating aggregates. The dominant form is pinacoid {100}. Prisms {hk0}, {h0l}, and {0kl} are also observed. The crystals are colorless, their streak is white, and the luster is vitreous. The mineral is transparent. It is non-fluorescent under ultraviolet light. Mohs’ hardness is 4, tenacity is brittle. Cleavage is perfect on {100}, good on {010} and {001}, and fracture is stepped. The measured density is 2.36(1) g/cm 3 ; the calculated density is 2.367 g/cm 3 . Mejillonesite is biaxial (-), α = 1.507(2), β = 1.531(2), γ = 1.531(2), 2V (meas) = 15(10)°, 2V (calc) = 0° (589 nm). Orientation is X = a, Z = elongation direction. The mineral is non-pleochroic. Dispersion is r > v, medium. The IR spectrum contains characteristic bands of the Zundel cation (H 5 O 2 + , or H + ·2H 2 O) and the groups P-OH and OH−. The chemical composition is (by EDS, H 2 O by the Alimarin method, wt%): Na 2 O 9.19, MgO 26.82, P 2 O 5 46.87, H 2 O 19, total 101.88. The empirical formula, based on 11 oxygen atoms, is Na 0.93 Mg 2.08 (PO 3 OH) 1.00 (PO 4 ) 1.06 (OH) 0.86 · 0.95H 5 O 2 . The strongest eight X-ray powder-diffraction lines [d in Å(I)(hkl)] are: 8.095(100)(200), 6.846(9) (210), 6.470(8)(111), 3.317(5)(302), 2.959(5)(132), 2.706(12)(113), 2.157(19)(333), and 2.153(9) (622). The crystal structure was solved on a single crystal (R = 0.055) and gave the following data: orthorhombic, Pbca, a = 16.295(1), b = 13.009(2), c = 8.434(1) Å, V = 1787.9(4) Å 3 , Z = 8. The crystal structure of mejillonesite is based on a sheet (parallel to the b-c plane) formed by two types of MgO 6 octahedra, isolated tetrahedra PO 4 and PO 3 OH whose apical vertices have different orientation with respect to the sheet. The sheets are connected by interlayer, 5-coordinated sodium ions, proton hydration complexes, and hydroxyl groups. The structure of mejillonesite is related to that of angarfite, NaFe 5 3+ (PO 4 ) 4 (OH) 4 ·4H 2 O and bakhchisaraitsevite, Na 2 Mg 5` (PO 4 ) 4 ·7H 2 O.

We report silician magnetite from banded iron formation (BIF) in the Dales Gorge Member of the Brockman Iron Formation, Hamersley Group, Western Australia. Magnetite mesobands typically consisting of individual ~100 μm microlaminae are revealed to be composed of silician magnetite overgrowths on magnetite. Silician magnetite overgrowths contain from 1 to 3 wt% SiO 2 , whereas (low-Si) magnetite domains contain less than 1 wt% SiO 2 . Silicon solid solution is present in the magnetite crystal lattice as determined by in situ micro-X-ray diffraction and high-resolution transmission electron microscopy. Three textures are distinguished in magnetite mesobands: (1) magnetite sub-microlaminae with silician magnetite overgrowths, (2) recrystallized magnetite fragments with silician magnetite overgrowths, and (3) a complex intergrowth of magnetite and silician magnetite. All three textures are found in magnetite mesobands from the BIF4-5 and BIF12-16 macrobands of the Dales Gorge type-section drill core DDH-47A from Wittenoom, Western Australia. Magnetite domains contain numerous submicrometer-to-micrometer inclusions of quartz, carbonate, stilpnomelane, and apatite, whereas silician magnetite overgrowths are devoid of mineral inclusions. The presence of mineral inclusions in magnetite indicates the BIF oxide precipitate was not chemically pure iron oxyhydroxide/oxide. Magnetite domains display textures formed during soft sediment deformation that are the earliest and best preserved relict sedimentary structures in this BIF. Silician magnetite is the dominant iron oxide in the Dales Gorge BIF and is present in many other sub-greenschist facies BIFs worldwide. We suggest the former presence of organic matter creates reducing conditions necessary to stabilize silician magnetite. Thus, silician magnetite is a potential biosignature in BIFs.

Despite being studied for more than one century, no consensus exists regarding the ultimate mechanism(s) of the thermal decomposition of dolomite [(CaMg(CO 3 ) 2 ]. To shed light on such a reaction, dolomite single crystals were calcined in air between 500 and 1000 °C, and in situ, in a TEM (high vacuum), following irradiation with the electron beam. In situ TEM shows that the decomposition involves the initial formation of a face centered cubic mixed oxide (Ca 0.5 Mg 0.5 O) with reactant/product orientation relationships [001] dolomite //<111> oxide , <4̄41> dolomite //<100> oxide , {112̄0} dolomite //{110} oxide , {112̄8} dolomite //{110} oxide and {101̄4} dolomite ^{100} oxide ~12°. This phase undergoes de-mixing into oriented crystals of Mg-poor CaO and Ca-poor MgO solid solutions upon long-term e-beam exposure. Ex situ TEM, XRD, 2D-XRD, and FESEM analyses show the formation of porous pseudomorphs made up of oxide nanocrystals with similar parent/product orientation relationships, but with limited Ca/Mg substitution (up to ~9-11%) due to de-mixing (spinodal decomposition) of the metastable (Ca,Mg)O precursor. High ion diffusivity at T > 500 °C (ex situ experiments) favors the formation of pure CaO and MgO crystals during coarsening via oriented aggregation and sintering. These results show that the thermal decomposition of dolomite is topotactic and independent of pCO 2 . Formation of Mg-calcite nanocrystals (up to ~8 mol% Mg) during the so-called “half decomposition” is observed at 650-750 °C. This transient phase formed topotactically following the reaction of CaO nanocystals (solid solution with ~9 mol% Mg) with CO 2 present in the air and/or released upon further dolomite decomposition. With increasing T, Mg-calcite transformed into calcite, which underwent decomposition following the known topotactic relationship: {101̄4} calcite //{110}CaO and <4̄41> calcite //<110>CaO. These observations solve the long standing controversy on the mechanism of the “two-stage” decomposition of dolomite, which assumed the direct formation of calcite during the so-called “half decomposition.”

The transition from diamond to graphite is a key equilibrium for interpreting ultrahigh-pressure metamorphic rocks. Despite widespread interest, there remain significant systematic differences between the best available experimental determinations of P and T (Kennedy and Kennedy 1976) and numerous thermodynamic calculations of the transition. At temperatures below 1400 K, calculated equilibrium pressures are lower than extrapolations of the experiments by as much as 5 kbar. At 3000 K, calculated pressures vary from more than 8 kbar above to almost 20 kbar below the position of the extrapolated transition. A revised curve based on a critical review of the experimental and thermodynamic data is consistent with expanded experimental brackets and the preferred calorimetric data. It is steeper than the transition proposed by Kennedy and Kennedy (1976) and previous calculations and passes through 16.2 kbar, 298 K; 33.9 kbar, 1000 K; 63.5 kbar, 2000 K; and 98.4 kbar, 3000 K. The revised curve implies that the minimum pressure for formation of diamond-bearing crustal rocks is 3-4 kbar less than implied by extrapolation of the experiments. Because the revised transition is steeper than most previous calculations, the triple point among graphite, diamond, and liquid carbon may be as much as 40 kbar higher than previously estimated.

Structural characteristics of Fe 3+ oxide/silica co-precipitates were investigated. The association between these materials is relevant to practically all natural aqueous systems due to the prevalence of iron and silicon in the Earth’s crust. Crystallographic information is very difficult to obtain from these precipitates due to the nanocrystalline nature of ferrihydrite and the amorphous structure of precipitated silica. Several previously undetermined key insights were gained into the structure of iron oxide/silica co-precipitates through this examination. The distribution of iron and silicon throughout co-precipitate particles is illustrated along with the influence of their association. Evidence to the governing factor behind differences in apparent crystallinity is also presented. This information culminates in the formulation of a precipitation pathway, displaying the formation of the co-precipitates.

Compositional dependence of alkali diffusivities in silicate melts is modeled by explicitly considering mixed alkali and “pseudo-alkali” effects. The well-known mixed alkali effect describes that the presence of light alkalis (Li, Na) retards diffusion of heavy alkalis (K, Rb, Cs) and vice versa, which can be attributed to stronger interaction between dissimilar alkalis and reduction of favorable sites for the alkali in question. Due to the same reasons, Ca, which has similar ionic radius as Li and Na but carries more charge and greater field strength, also impedes the migration of light alkalis. In this regard Ca behaves as a “pseudo-alkali,” and I refer to the influence caused by Ca on Li and Na diffusion as the “pseudo-alkali effect,” which belongs to the category of mixed cation effects in glass literature. Pseudo-alkali effect is expected to exist also between other size-matching divalent cations and alkali cations, such as Ba blocking K diffusion. Mixed alkali and pseudo-alkali effects are manifested mainly by an increase in the activation energy for diffusion. Sodium diffusivities for melt compositions ranging between albite, orthoclase, and anorthite reveal that the amplitude of the pseudo-alkali effect is much larger than that of the mixed alkali effect. Furthermore, the activation energy increases more rapidly as more Ca substitutes for Na, but the mixed alkali effect approaches saturation as K substitutes for sodium. A general model based on mixed alkali effect and pseudo-alkali effect, together with a special treatment relating Cs diffusion to melt polymerization, successfully characterizes existing experimental data on alkali diffusion in silicate melts. Enthalpies of mixing between end-member melts (e.g., Ab-Or melts and Ab-An melts) inferred from the model agree well with previous experimental investigations.

Congruent evaporation of a crystalline material in vacuum is an extreme reaction in that backward reactions and transport processes in the reactant can be neglected. The evaporation is strongly governed by surface processes and intrinsic nature of the substance. A thorough knowledge of the atomistic evaporation mechanism is fundamental for better understanding reaction kinetics between gas and condensed materials in general. We have conducted a series of evaporation experiments of forsterite in vacuum for crystallographically oriented surfaces at 1500 to 1810 °C. The (100), (010), and (001) surfaces developed their own morphology characterized by evaporation pits and grooves originated from dislocation outcrops. Nominal overall evaporation rate (average retreat rate of a surface) shows significant anisotropy with the maximum difference by a factor of five below 1740 °C. The overall evaporation rates for individual surfaces are fitted with respective Arrhenius relationships, giving the highest activation energy for (100), intermediate for (001), and the lowest for (010). The anisotropy decreases to within 50% at ~1800 °C, which is caused by enhancement of evaporation from (010) owing to preferential evaporation around dislocation outcrops. “Intrinsic evaporation rates” estimated by subtracting contributions of initial roughness and the preferential evaporation around dislocations from the nominal overall evaporation rates show substantial anisotropy even at ~1800 °C. The “intrinsic evaporation rate” for (010) is adequately fitted by an Arrhenius relationship over the examined temperature range giving a single activation energy of 655 kJ/mol. The prevalence of steps with submicrometer to nanometer-scale height shows that forsterite evaporates mostly by layer-by-layer mechanism. The only exception is the (001) surface above ~1650 °C, on which such steps are absent except for surface-parallel minor facets which are rapidly diminishing with time. The (001) surface is inferred to evaporate by direct detachment mechanism at high temperatures. The change of evaporation mechanisms for (001) at around 1650 °C corresponds to a rough-smooth transition kinetically induced by an atomistic evaporation process.

X-ray absorption spectroscopy was used to investigate the oxidation state of uranium in various Uand Th-bearing Al-rich CaSiO 3 perovskite samples synthesized at high-pressure and high-temperature using a multi-anvil press apparatus. X-ray absorption near edge spectroscopy (XANES) spectra collected at the U L III - and Th L III -edges using both micro- and macro-focused beams show U 4+ in the Al-rich CaSiO 3 perovskite. The structure of the U- and Th-bearing Al-rich CaSiO 3 perovskite samples have been cross-checked by XANES spectra collected at the Ca K-, Al K-, and Si K-edges. Al K and Si K spectra suggest that Al incorporates exclusively on the Si site of the CaSiO 3 perovskite. Ca K spectra of the (U,Th)-bearing Al-rich CaSiO 3 perovskite samples were succesfully compared to FEFF8.2 ab initio models of a tetragonal CaSiO 3 perovskite with space group P4/mmm. Our results confirm previous assumptions of the coupled substitution of CaSi 2 by UAl 2 in CaSiO 3 perovskite and that U and Th can be incorporated separately or together in CaSiO 3 perovskite by means of this mechanism. The possible occurrence of the U- and Th-bearing Al-rich CaSiO 3 perovskite are discussed as a potential candidate to locally host a large amount of actinides in the Earth’s deep mantle. The study of a phase that can act as a storage mineral for heat-producing actinide elements such as uranium and thorium is fundamental to the understanding of the geodynamics and thermal behavior of Earth.

Thermal analysis experiments in the environment of an extremely low water vapor concentration provide insight into the first steps of the rehydration mechanism in smectite when completely dehydrated and the interlayer region is collapsed. The relative structural and compositional controls on dehydration and rehydration reactions are compared from a well-characterized suite of samples that vary with respect to chemical composition, octahedral and tetrahedral substitution, octahedral cation site vacancy, and degree of dehydroxylation. Techniques including multi-cycle heating-cooling thermogravimetric analysis and nitrogen gas adsorption on various smectite samples preheated at different temperatures followed by rehydration at ambient conditions were used to characterize the interaction of water molecules with completely dehydrated montmorillonite, beidellite, and nontronite smectite types. Beidellite with high-Al 3+ tetrahedral substitution results in electrostatically undersaturated basal oxygen atoms that exert strong repulsion between the tetrahedral sheets of adjacent 2:1 layers. The interlayer region of dehydrated or dehydroxylated beidellite is capable of being spontaneously rehydrated even in low water vapor environments. In completely dehydrated montmorillonite and nontronite, the external surface area of the crystallites is a primary control on water adsorption at low humidity when the molecules form a shell around the exchangeable cations present on external surfaces. The potential of montmorillonite and nontronite to reopen a collapsed interlayer is significantly lower than beidellite because of their crystal-chemical features that result in 2:1 layer and interlayer cation attraction. With increasing water vapor partial pressure, the hydration potential of interlayer cations provokes a reopening of the interlayer. In a dehydroxylated nontronite, the undersaturated residual oxygen atom strongly bonds the interlayer cation within the ditrigonal ring of the tetrahedral sheet, resulting in a permanent interlayer collapse. The specific surface area calculated from a conventional N 2 gas adsorption measurement using the BET model represents the external surface area of a dehydrated smectite crystallite and can be converted into the mean crystallite thickness. The mean crystallite thickness of a completely dehydrated smectite increases with an increase in preheating temperature.

We report results from high-pressure single-crystal X-ray diffraction and Mössbauer absorption experiments on magnetite. Based on high-quality diffraction data, we have obtained accurate information on the crystal structure of magnetite below 25 GPa, which enables an unambiguous interpretation of the Mössbauer data using constrained area ratios and a full transmission integral fit that avoids area distortion due to thickness effects. Based on our analysis, all aspects of the electronic and magnetic properties of magnetite reported previously below 25 GPa at ambient temperature can be explained solely by the enhanced delocalization of 3d electrons of iron atoms. For instance, we present evidence that the compression-induced metallization changes the sign of the charge carrier spin polarization at 15 GPa.

Infrared (IR) absorption spectra of antigorite were measured up to 27 GPa and 320 °C using synchrotron IR radiation to elucidate OH group behavior under high-pressure (HP) and high-temperature (HT) conditions. The absorption bands attributable to the OH stretching modes of outer OH groups (OH outer ) and inner OH groups (OH inner ) show positive pressure dependencies. The shift rate of the OHinner band is almost constant at all pressure ranges. In contrast, that of the OH outer band increases slightly at about 6 GPa. This discontinuous change of the shift rate is consistent with the anomalous behavior of the OH outer upon compression, which was predicted in the previous first-principle calculation study. Specifically, the pressure dependence of the OH outer band shows that the hydrogen ion of an OH outer interacts not only with the nearest basal oxygen ion of the SiO 4 tetrahedron but also with the second nearest two basal oxygen ions upon compression. The latter interaction becomes dominant over the former interaction at about 6 GPa. Pressure-induced amorphization was indicated from IR spectra measured at 300 °C and 25.6 GPa. This P-T condition is out of the thermodynamic stability field of antigorite. A broad absorption band, which is close to the broad band attributable to natural hydrous silicate glass, appeared after amorphization, which suggests that the pressure-induced amorphization of antigorite does not induce dehydration. Hydrogen atoms are retained in amorphized antigorite as OH groups.

The single-crystal Raman spectra of the natural mineral paulmooreite Pb 2 As 2 O 5 from the Långban, Filipstad district, Värmland province, Sweden, are presented for the first time. It is a monoclinic mineral containing an isolated [As 2 O 5 ] 4- dimer. Unpolarized single-crystal spectra of the natural and synthetic samples compare favorably with each other and are characterized by strong bands around 186 and 140 cm -1 and three medium bands at 800-700 cm -1 . Band assignments were made based on band symmetry and spectral comparison between experimental band positions and those resulting from Hartree-Fock calculation of an isolated [As 2 O 5 ] 4- anion complex. Spectral comparison was also made with lead arsenites such as synthetic PbAs 2 O 4 and Pb 2 (AsO 2 ) 3 Cl and natural finnemanite to determine the contribution of the terminal and bridging O in paulmooreite. Bands at 760-733 cm -1 were assigned to terminal As-O vibrations, whereas stretches of the bridging O occur at 562 and 503 cm -1 . The single-crystal spectra showed good mode separation, allowing bands to be assigned a symmetry species of A g or B g .

The behavior of the zeolite laumontite during dissolution in acidic aqueous solutions has been studied by fluid-cell atomic force microscopy, with a focus on the role of the framework Al atoms in controlling the dissolution mechanisms. Isothermal single etch pits dissolution rates have been measured in situ at three different temperatures (11, 23, and 40 °C) on the (110) surface, and at one temperature (22 °C) on (2̄01). The experimentally derived apparent activation energy of the development of the etch pits (E = 11.9 ± 4 kJ/mol) on the (110) surface and their morphology can be interpreted considering the structural features of laumontite, especially the crystallochemical distribution of the Al atoms in the framework. Preliminary comparison of the dissolution features of laumontite with those measured on zeolites with different Si/Al ratios and Al distribution further confirm the controlling role of the structural Al in the dissolution process.

A high-pressure phase of CaRhO 3 stable between perovskite and post-perovskite in P-T space was synthesized at 17 GPa and 1650 °C using a multi-anvil apparatus. The crystal structure of CaRhO 3 was solved by the structure prediction evolutionary algorithm and was refined by Rietveld analysis of the synchrotron powder X-ray diffraction pattern, along with transmission electron microscopy observations. The structure is monoclinic with lattice parameters of a = 12.5114(1) Å, b = 3.1241(1) Å, c = 8.8579(1) Å, β = 103.951(1)°, V = 336.01(1) Å 3 with space group P2 1 /m. The structure contains edge-sharing RhO 6 octahedral chains along the b-axis. The six RhO 6 octahedral chains make a unit, which stacks up alternatively with the CaO 8 polyhedral layer along the [101] direction to form the structure of CaRhO 3 intermediate phase.

A consistent methodology for obtaining the enthalpy of formation of Fe 2+ -containing binary and multicomponent oxides using high-temperature oxide melt solution calorimetry has been developed. The enthalpies of wüstite (FeO) and magnetite (Fe 3 O 4 ) oxidation to hematite (Fe 2 O 3 ) were measured using oxidative drop solution calorimetry in which the final product is dissolved ferric oxide. Two methods were applied: drop solution calorimetry at 1073 K in lead borate solvent and at 973 K in sodium molybdate, each under both oxygen flowing over and bubbling through the solvent, giving consistent results in agreement with literature values. The enthalpies of formation of all three iron oxides from the elements were obtained using a thermodynamic cycle involving the directly measured oxidative dissolution enthalpy of iron metal in sodium molybdate at 973 K and gave excellent consistency with literature data. The methodology was then applied to the magnetite-maghemite system. The enthalpy of mixing of the Fe 3 O 4 -Fe 8/3 O 4 spinel solid solution is exothermic and, represented by a subregular (Margules) formalism, ΔH mix = x(1 - x)[-63.36 ± 8.60(1 - x) + 17.65 ± 6.40x] kJ/mol, where x is the mole fraction of magnetite. The entropies of mixing of the solid solution were calculated for different assumptions about the distribution of cations, charges, and vacancies in these defect spinels. The different models lead to only small differences in the entropy of mixing. Calculated free energies of mixing show no evidence for a solvus in the magnetite-maghemite system.

Static compression data of (Mg 0.83 ,Fe 0.17 )O and (Mg 0.75 ,Fe 0.25 )O ferropericlases have been measured up to 58 GPa along 300, 700, and 1100 K isotherms, using synchrotron powder X-ray diffraction experiments combined with a Kawai-type, multi-anvil, high-pressure apparatus and sintered diamond anvils. High-temperature and high-pressure equations of state for these two ferropericlases, which have high-spin Fe 2+ ions, were developed using measured compression data below 47 GPa, based on the Mie-Grüneisen relation and the Debye thermal model, combined with the 300 K Birch-Murnaghan equation. When the isothermal bulk modulus (K 0T ) and the Debye temperature (Θ 0 ) are fixed at 160 GPa and 500 K, respectively, the optimized equation-of-state parameters for these two phases are as follows: the pressure derivative of the bulk modulus (K' 0T ), the Grüneisen constant (γ 0 ), and the q parameter are 4.08 ± 0.02, 1.53 ± 0.04, and 0.7 ± 0.2, respectively, for (Mg 0.83 ,Fe 0.17 )O; and 4.22 ± 0.03, 1.64 ± 0.04, and 0.7 ± 0.2, respectively, for (Mg 0.75 ,Fe 0.25 )O. We found that calculated pressures with these equation-of-state parameters accurately reproduce the measured pressures of each ferropericlase below ~50 GPa for the isotherms of 300, 700, and 1100 K. Furthermore, the compression curve indicates that for each ferropericlase at each isothermal compression of 300, 700, and 1100 K, an abrupt volume reduction occurs at ~50 GPa. This volume reduction becomes more pronounced with increasing pressure, as a result of the progressive transition from high-spin to low-spin of the Fe 2+ ions in each ferropericlase.

Microprobe analyses of antigorite show that (Al+Cr) and inferred Fe 3+ correlate inversely with Si apfu in a Tschermaks substitution. This observation suggests that the uptake of Fe 3+ is not simply related to f O₂ . For Si = 1.95 apfu estimated Fe 3+ = 0.032 apfu (or 0.95 wt% Fe 2 O 3 ). Such estimates of Fe 3+ require high analytical accuracy and precision, and assume a fixed polysomatic formula (e.g., m = 17) and freedom from interlayer sheet-silicate impurities. In many cases the estimates appear to be high. An alternative measure of Fe 3+ is provided by the partitioning of total Fe and Mg between antigorite and olivine in well-equilibrated natural antigorite-olivine-magnetite parageneses. Extrapolation of Nernst and Roozeboom partition plots to Fe-free olivine permits an estimate of the Fe 3+ content of the average antigorite in this paragenesis, namely 0.42 or 0.64 wt% Fe 2 O 3 . The partition estimates are in good agreement with the results of Mössbauer spectroscopy performed here on 14 antigorites from metaperidotites, together with four from the literature. These spectra reveal a range of 0.16 to 1.94 in wt% Fe 2 O 3 in metaperidotite antigorite, with an average of 0.83. In two olivine-bearing rocks, antigorite has Fe 3+ /ΣFe ratios of 0.13 and 0.15, which corresponds to wt% Fe 2 O 3 = 0.47 and 0.54, respectively. Larger amounts of Fe 2 O 3 occur in some, but not all, vein antigorites. The prograde formation of antigorite in serpentinite from lizardite is accompanied by loss of some cronstedtite component and the precipitation of additional magnetite. The Roozeboom Mg/Fe partition plot is concave down rather than up; in other words the partition coefficient K D is a function of the X Mg of olivine. This behavior has been found in other olivinemineral pairs. It can be interpreted to reflect strongly non-ideal solution behavior of MgFe-olivine at low temperatures, viz. W G ≈ 8.5 kJ assuming a symmetrical solution. MgFe-brucite appears to be similarly non-ideal.

A new mineral species, markascherite (IMA2010-051), ideally Cu 3 (MoO 4 )(OH) 4 , has been found at Copper Creek, Pinal County, Arizona, U.S.A. The mineral is of secondary origin and is associated with brochantite, antlerite, lindgrenite, wulfenite, natrojarosite, and chalcanthite. Markascherite crystals are bladed (elongated along the b axis), up to 0.50 × 0.10 × 0.05 mm. The dominant forms are {001}, {100}, and {010}. Twinning is found with the twofold twin axis along [101̄]. The mineral is green, transparent with green streak and vitreous luster. It is brittle and has a Mohs hardness of 3.5~4; cleavage is perfect on {100} and no parting was observed. The calculated density is 4.216 g/cm 3 . Optically, markascherite is biaxial (-), with n α >1.8, n β > 1.8, and n γ >1.8. The dispersion is strong (r > v). It is insoluble in water, acetone, or hydrochloric acid. An electron microprobe analysis yielded an empirical formula Cu 2.89 (Mo 1.04 O 4 )(OH) 4 . Markascherite, polymorphic with szenicsite, is monoclinic, with space group P2 1 /m and unit-cell parameters a = 9.9904(6), b = 5.9934(4), c = 5.5255(4) Å, β = 97.428(4)°, and V = 328.04(4) Å 3 . Its structure is composed of three nonequivalent, markedly distorted Cu 2+ (O,OH) 6 octahedra and one MoO 4 tetrahedron. The Cu1 and Cu2 octahedra share edges to form brucite-type layers parallel to (100), whereas the Cu3 octahedra share edges with one another to form rutile-type chains parallel to the b axis. These layers and chains alternate along [100] and are interlinked together by both MoO 4 tetrahedra and hydrogen bonds. Topologically, the structure of markascherite exhibits a remarkable resemblance to that of deloryite, Cu 4 (UO 2 )(MoO 4 ) 2 (OH) 6 , given the coupled substitution of [2Cu 2+ + 2(OH − )]2+ for [(U 6+ + □) + 2O 2- ] 2+ . The Raman spectra of markascherite are compared with those of two other copper molybdate minerals szenicsite and lindgrenite.

Natural and synthetic hydrous amorphous silicas were investigated with single-pulse 29 Si magic angle spinning (MAS) NMR and with vibrational spectroscopic methods. Samples included a volcanically derived silica coating on young basalt from Kilauea, Hawaii, as well as hyalite (opal-AN), silica sinters, and synthetic silica gels and silicic acid. Pulse delays of up to an hour were employed for silica samples with slow spin lattice relaxation rates, and nearly fully relaxed spectra (90-100%) were demonstrably achieved for all samples. 29 Si NMR spectra consisted of two broad, overlapping peaks at -111 and -102 ppm and a smaller peak at -92 ppm, corresponding to Q 4 , Q 3 , and Q 2 sites, respectively. The Hawaiian silica coating and silicic acid samples displayed high Q 3 and Q 2 contents; in particular, the structural Si-OH content of the coating was unusually high for a natural silica (5.4 ± 0.4 wt% H 2 O). Saturation-recovery spectra of the Hawaiian silica with increasing delay times were consistent with “stretched exponential” relaxation behavior and three-dimensional distribution of paramagnetic centers. Attenuated total reflectance infrared (ATR-IR) and Raman spectra of the silica powders indicated fully amorphous structures, and displayed hydrous (SiO 3 OH) and anhydrous silicate vibrational bands in positions consistent with previous work. Raman spectra of some samples indicated modest grain to grain heterogeneity. Inferred Si-OH contents from ATR-IR band ratios were strongly correlated with hydroxyl contents calculated from NMR spectra. The high Si-OH content of the Hawaiian silica coating suggests it is diagenetically immature and has not been exposed to elevated temperatures.

Chromschieffelinite, Pb 10 Te 6 O 20 (OH) 14 (CrO 4 )(H 2 O) 5 , is a new tellurate from Otto Mountain near Baker, California, named as the chromate analog of schieffelinite, Pb 10 Te 6 O 20 (OH) 14 (SO 4 )(H 2 O) 5 . The new mineral occurs in a single 1 mm vug in a quartz vein. Associated mineral species include: chalcopyrite, chrysocolla, galena, goethite, hematite, khinite, pyrite, and wulfenite. Chromschieffelinite is orthorhombic, space group C222 1 , a = 9.6646(3), b = 19.4962(8), c = 10.5101(7) Å, V = 1980.33(17) Å 3 , and Z = 2. Crystals are blocky to tabular on {010} with striations parallel to [001]. The forms observed are {010}, {210}, {120}, {150}, {180}, {212}, and {101}, and crystals reach 0.2 mm in maximum dimension. The color and streak are pale yellow and the luster is adamantine. The Mohs hardness is estimated at 2. The new mineral is brittle with irregular fracture and one perfect cleavage on {010}. The calculated density based on the ideal formula is 5.892 g/cm 3 . Chromschieffelinite is biaxial (-) with indices of refraction α = 1.930(5), β = 1.960(5), and γ = 1.975(5), measured in white light. The measured 2V is 68(2)°, the dispersion is strong, r < v, and the optical orientation is X = b, Y = c, Z = a. No pleochroism was observed. Electron microprobe analysis provided: PbO 59.42, TeO 3 29.08, CrO 3 1.86, H 2 O 6.63 (structure), total 96.99 wt%; the empirical formula (based on 6 Te) is Pb 9.65 Te 6 O 19.96 (OH) 14.04 (CrO 4 ) 0.67 (H 2 O) 6.32 . The strongest powder X-ray diffraction lines are [d obs in Å (hkl) I]: 9.814 (020) 100, 3.575 (042,202) 41, 3.347 (222) 44, 3.262 (241,060,113) 53, 3.052 (311) 45, 2.9455 (152,133) 55, 2.0396 (115,353) 33, and 1.6500 (multiple) 33. The crystal structures of schieffelinite (R 1 = 0.0282) and chromschieffelinite (R 1 = 0.0277) contain isolated Te 6+ O 6 octahedra and Te 2 6+ O 11 corner-sharing dimers, which are linked into a three-dimensional framework via bonds to Pb 2+ atoms. The framework has large channels along c, which contain disordered SO 4 or CrO 4 groups and H 2 O. The lone-electron pair of each Pb 2+ is stereochemically active, resulting in one-sided Pb-O coordination arrangements. The short Pb-O bonds of the Pb 2+ coordinations are all to Te 6+ O 6 octahedra, resulting in strongly bonded layers parallel to {010}, which accounts for the perfect {010} cleavage.

The growth rate and internal organization of bimineralic diopside (CaMgSi 2 O 6 )-merwinite (Ca 3 MgSi 2 O 8 ) reaction rims produced by a solid-state reaction between monticellite (CaMgSiO 4 ) and wollastonite (CaSiO 3 ) single crystals was determined at 900 °C, 1.2 GPa, and run durations from 5 to 65 h using conventional piston-cylinder equipment. Overall reaction rim thickness ranges from 3.8 to 20.9 μm and increases with the square root of time, indicating that rim growth is diffusion controlled. Symmetrical makeup of the internal microstructure implies that rims grow from the original interface toward both reactants at identical rates, indicating that diffusion of MgO across the rim controls overall growth, with an effective bulk diffusion coefficient of D Di Mw bu+lk,MgO = 10 −16.3 ± 0.2 m 2 /s. At the initial stages, a “lamellar type” microstructure of alternating palisade-shaped diopside and merwinite grains elongated normal to the reaction front is generated, which gradually transforms into a “segregated multilayer type” microstructure with almost perfectly monomineralic Mw|Di|Mw layers oriented parallel to the reaction fronts at long run durations. This is due to changes in relative component mobilities. Whereas the “lamellar” microstructure develops when MgO is substantially more mobile than the other components, formation of the “segregated multilayer” microstructure requires additional mobility of at least one of the other components, CaO or SiO 2 . We assume that a significant change in relative component mobilities is caused by continuous entrance of minute amounts of water from the piston-cylinder solid pressure medium through the capsule walls, as revealed by the presence of OH-defects in a reactant after the runs, and supported by water-containing powder experiments that only produce monomineralic Mw|Di|Mw layers. Traces of water have a major influence on relative component mobilities, internal rim organization, and the microstructural development of reaction zones.

A thermodynamic model for the growth of a reaction rim with lamellar internal microstructure is derived. Chemical mass transfer across the rim and the chemical segregation within the reaction fronts, at which the lamellar microstructure is produced, are considered as the only dissipative processes involved in rim growth. Depending on their relative rates, either one of these processes may be rate limiting. Rim growth is parabolic when mass transfer across the growing rim is rate limiting, and it is linear when the material redistribution within either one or both of the reaction fronts is rate limiting. The transition between these two extreme scenarios is continuous. The controlling factors are the characteristic length scale of the lamellar microstructure and the relative rates of chemical mass transfer across the rim and within the reaction fronts. Reaction rims with both lamellar and layered internal microstructures have been produced experimentally in the ternary system CaO-MgO-SiO 2 . Based on the thermodynamic extremum principle, parameter domains can be discerned, where reaction rims preferably develop lamellar microstructures or, alternatively, a sequence of monophase layers. For a given set of kinetic parameters, formation of the layered microstructural type is more likely during the initial growth stages, and the lamellar type is preferred at later growth stages.

The infrared spectra of natural samples of muscovite, biotite, and phlogopite are characterized to pressures of ~30 GPa, as is the Raman spectrum of muscovite to ~8 GPa. Both far-infrared and midinfrared data are collected for muscovite, and mid-infrared data for biotite and phlogopite. The response of the hydroxyl vibrations to compression differs markedly between the dioctahedral and trioctahedral micas: the hydrogen bonding in dioctahedral environments increases with pressure, as manifested by shifts to lower frequency of the hydroxyl-stretching vibrations, whereas cation-hydrogen repulsion likely produces shifts to higher frequency of the hydroxyl vibrations within trioctahedral environments. An abrupt decrease in frequency and increase in band width of the hydroxyl-stretching vibration in muscovite is observed at pressures above ~18-20 GPa, implying that the previously documented pressure-induced disordering is associated with the local environment and shifts in location of the hydroxyl unit in this material. The far-infrared vibrations of muscovite indicate that its compressional mechanism changes above 5-8 GPa, as the K-O stretching vibration with a zero-pressure frequency near 112 cm -1 shifts in its pressure dependence from 6.9 cm -1 /GPa below this pressure range to 0.78 cm -1 /GPa above it. Thus, it appears that the magnitude of interlayer compression is decreased above this pressure, and hence that the compression of muscovite may become less strongly anisotropic. The mid-infrared bands that are primarily produced by vibrations of the tetrahedral layer broaden under pressure in both muscovite and biotite: within biotite, a spectral region that may be associated with higher coordination of tetrahedral cations increases in amplitude above about 25 GPa. The corresponding bands in phlogopite undergo less broadening, and their behavior is fully reversible on decompression.

Hydromaghemite and ferrimagnetic ferrihydrite are two terms that designate the same phase, which exhibits strong ferrimagnetism and >3% water loss between 110 and 350 °C. Its X-ray diffraction patterns, both in the real and reciprocal space, are consistent with the structural model of ferrihydrite of Michel et al. (2010), which includes tetrahedrally coordinated iron. This phase can be readily produced via transformation of 2-line ferrihydrite in strictly aerobic conditions without the intermediate formation of magnetite, and where additives act, with different efficiency, as sorbents retarding the fast transformation, via aggregation, into hematite.

The XRD pattern of hydromaghemite, previously interpreted as a mixture of hydroxylated maghemite, ferrihydrite, and hematite (Barrón et al. 2003; Liu 2008), has been reinterpreted as being from a new form of ferrihydrite called “ferrimagnetic ferrihydrite” (ferrifh, Michel et al. 2010; Barrón et al. 2012). Although ferrihydrite sensu stricto and ferrifh have distinct XRD patterns, their profiles have been fit with the same akdalaite-type model using atomic pair distribution (PDF) analysis (Michel et al. 2007, 2010). This ambivalence shows that the crystal structure problem is ill-conditioned, as often the case for the study of nanostructured materials by PDF (Billinge and Levin 2007).