Band 91, Heft 8-9

The release of Zr, U, and Pb from nearly metamict zircon and its recrystallized analog and of Zr from fully crystalline and slightly radiation-damaged zircon in H 2 O + HCl ± SiO 2 fluids was investigated in situ at temperatures between 200 and 500 °C using a hydrothermal diamond-anvil cell and time-resolved synchrotron-radiation XRF analyses. Dissolution of nearly metamict zircon proceeded much faster than that of zircon with little or no radiation damage and resulted in a 1.5 to 2 log units higher Zr molality in 6 to 7 m HCl fluids. Extensive recrystallization of the almost fully amorphous material started at 260 to 300 °C in H 2 O + HCl, and at about 360 °C if quartz was added, and was coupled with a decrease of the Zr concentration in the fluid by more than an order of magnitude. Recrystallization in 7 m HCl had little effect on the aqueous U and Pb concentrations, whereas addition of quartz caused a more sluggish decrease of the Zr concentration in the fluid upon recrystallization and lowered the release of U. The data presented here support the interpretation that enhanced Zr mobility in low-grade metamorphic rocks may be related to dissolution of metamict zircon by aqueous fluids and illustrate the significance of the silica activity on the kinetics of dissolution and recrystallization during zircon-fluid interaction.

Magnetotactic bacteria comprise several aquatic species that orient and migrate along geomagnetic field lines. This behavior is based on the presence of intracellular ferrimagnetic grains of the minerals magnetite (Fe 3 O 4 ) or greigite (Fe 3 S 4 ). Whereas the structural and magnetic properties of magnetite magnetosomes have been studied extensively, the properties of greigite magnetosomes are less well known. Here we present a study of the magnetic microstructures, chemical compositions, and threedimensional morphologies and positions of Fe-sulfide crystals in air-dried cells of magnetotactic bacteria. Data were obtained using several transmission electron microscopy techniques that include electron holography, energy-filtered imaging, electron tomography, selected-area electron diffraction, and high-resolution imaging. The studied rod-shaped cells typically contain multiple chains of greigite magnetosomes that have random shapes and orientations. Many of the greigite crystals appear to be only weakly magnetic, because the direction of their magnetic induction is almost parallel to the electron beam. Nevertheless, the magnetosomes collectively comprise a permanent magnetic dipole moment that is sufficient for magnetotaxis. One of the cells, which is imaged at the point of dividing, contains multiple chains of both equidimensional Fe-sulfide and elongated Fe-oxide crystals. The equidimensional and elongated crystals have magnetic properties that are consistent with those of greigite and magnetite, respectively. These results can be useful for obtaining a better understanding of the function of magnetotaxis in sulfide-producing cells, and they have implications for the interpretation of the paleomagnetic signals of greigite-bearing sedimentary rocks.

We applied different independent techniques (electron microprobe analysis, structure refinement, and X-ray absorption spectroscopy) to unravel the possible mechanisms of Sc incorporation in the pyrope-grossular join. Samples were synthesized at elevated pressure and temperature by adding 5 wt% of Sc 2 O 3 to selected nominal compositions (pyrope, pyrope 60 grossular 40 , pyrope 20 grossular 80 , and grossular). In this way, the site of incorporation was not pre-determined, and only depends on the availability of a mechanism for local charge-balance. The EXAFS spectra of the two end-members could be analyzed by a multi-shell fit procedure, whereas the chemical heterogeneity of the Sc-doped solid-solution terms prevented this approach. However, the available information allows detection of different mechanisms of incorporation, which are active as a function of the bulk composition. In pyrope, Sc mainly enters the dodecahedral X site, and the local charge balance is achieved by incorporation of Mg at the adjacent tetrahedral Z site. Local charge-balance requirements suggest that a Z site occupied by Mg bridges two X sites occupied by Sc. When the entrance of Ca provides relaxation of the averaged structure, Sc may enter all the three available cation sites via the coupled heterovalent exchange X Sc 1 Z Sc 1 X Mg -1 Z Si -1 and the homovalent exchange Y Sc 1 Y Al -1 . In the samples of this work, there is an apparent limit in the Sc incorporation at the Y site, which is in contrast to the favored mechanism of incorporation in Sc-doped andradites. This limit can be explained in terms of relative dimensions of the structural sites when Al is the dominant Y cation. These results must be taken into account when evaluating trace-element behavior in garnets for geochemical purposes. In particular, they explain why D Sc can be treated together with D REE in models based on the elastic strain theory in garnets close to the pyrope composition, but deviate from the parabolic fit in grossular-rich garnets.

Investigation of scandium incorporation in garnets along the synthetic Ca 3 Fe 2 3+ Si 3 O 12 -Ca 3 Sc 2 Si 3 O 12 (adr.CaSc) join, based on the same multi-technique approach used in the companion paper (Oberti et al. 2006a), shows that (1) Sc is incorporated exclusively at the Y octahedron; (2) the local coordination of Sc is slightly different in Sc-poor than in Sc-rich compositions (Sc-O = 2.06 Å in CaSc 10 vs. 2.10 Å in CaSc 30-90 ); (3) the local coordination of Ca is also slightly different in Sc-poor than in Sc-rich compositions [Ca1,2-O are 2.34(2) and 2.48(2) Å in CaSc 10 and 2.36(2) and 2.50(2) Å in CaSc 90 , with Δ fixed at 0.14 Å in all the samples]; (4) the linear increase of the unit-cell edge along the join derives from multiple changes in the geometry of the different polyhedra and from the rotation of the tetrahedron around the 4̄ axis (α rotation), and cannot be modeled from extrapolation of the behavior observed along the Ca 3 Al 2 Si 3 O 12 -Ca 3 Fe 2 3+ Si 3 O 12 (grs.adr) join. CaSc-rich garnets, where a large X dodecahedron coexists with a large Y octahedron and a Z tetrahedron occupied by Si, similar to pyrope-grossular garnets, have the highest α values observed to date in calcium silicate garnets. Slightly lower α values are observed in pyrope and almandine, but correspond to a different structural arrangement, where a small X dodecahedron coexists with a small Y octahedron. These results further confirm the efficiency of a combined short- and long-range approach for understanding the properties of garnet solid solutions.

A re-examination of the chemistry and structure of nearly all the known occurrences of arrojadite and related minerals (dickinsonite and sigismundite) allowed understanding of the main substitution vectors and cation ordering schemes ruling the crystal-chemistry of these very complex phosphates. Electron microprobe analyses were done with a careful choice of the standards and of experimental conditions, and were coupled with LA-ICP-MS in situ analysis for Li, Be, and B. Structure refinement was done in a space group (Cc) with a lower symmetry than those used in previous studies (C2/c and its equivalents), which allowed a better understanding of the structure details and of cation ordering. The combined approach yielded a new formula for the arrojadite group, namely A 2 B 2 Ca 1 Na 2+x M 13 Al(PO 4 ) 11 (PO 3 OH 1-x )W 2 , where A are either large divalent cations (Ba, Sr, Pb) plus vacancy, or monovalent (K, Na) cations; and B are either small divalent cations (Fe, Mn, Mg) plus vacancy, or monovalent (Na) cations. The number of hydroxyl groups in the arrojadite formula is generally 3 apfu, and can be lowered to 2 apfu in particular when the sum of non-(P,Al) cations is higher than 20 apfu. We present in this paper the complete characterization of three samples (two of which are new members) that are crucial to fix the cornerstones of arrojadite crystal-chemistry. The sample from Rapid Creek (Yukon Territory) is the holotype for arrojadite-(KNa), and has unit formula K 0.83 Na 5.01 (Ca 0.91 Sr-) Σ=0.92 (Fe 2+ 9.34 Mg 2.69 Mn 2+ 1.03 Zn 0.01 Li 0.01 ) Σ=13.08 (Al 1.04 Ti 0.02 ) Σ=1.06 (OH 1.97 F 0.03 ) Σ=2.00 [(P 11.99 Si 0.01 T 1 )O 47 (OH) 1.00 ] [ideally, A1 K A2 Na B1 Na B2 Na Na1,2 Na 2 Na3 ⃞ Ca Ca M Fe 13 Al (PO 4 ) 11 P1x (PO 3 OH) W(OH,F) 2 ] and unit-cell dimensions: a = 16.5220(11), b = 10.0529(7), c = 24.6477 (16) Å, β = 106.509(2)°, V = 3932.2(7) Å 3 (Z = 4). The sample from Horrsjöberg (Värmland, Sweden) is the holotype material for arrojadite-(SrFe), and has unit formula Sr 0.93 Na 3.20 (Ca 0.59 Ba 0.20 Pb 0.03 K 0.03 ) Σ=0.85 (Fe 2+ 6.64 Mg 3.61 Mn 2+ 3.33 Zn 0.07 Li 0.01 ) Σ=13.66 (Sc 0.04 Al 1.00 ) Σ=1.04 (OH 1.10 F 0.90 ) Σ=2.00 [(P 11.95 Si 0.02 ) Σ= 11.97 O 47 (OH) 1.00 ] [ideally, A1 Sr A2 ⃞ B1 Fe 2+ B2 ⃞ Na1,2 Na 2 Na3 ⃞ Ca Ca M Fe 2+ 13 Al (PO 4 ) 11 P1x (PO 3 OH) W (OH,F) 2 ], and unit-cell dimensions a = 16.3992(7), b = 9.9400(4), c = 24.4434(11) Å, β = 105.489(1)°, V = 3839.76(46) Å 3 . The sample from Branchville (Connecticut) is the holotype material for dickinsonite-(KMnNa), and has unit formula K 0.50 Na 5.78 (Ca 0.51 Sr 0.05 Ba 0.01 Pb 0.01 ) Σ=0.58 (Mn 2+ 9.70 Fe 2+ 3.72 Li 0.31 Mg 0.06 Zn 0.01 ) Σ=13.80 (Al 0.91 Fe 3+ 0.09 Ti 0.01 ) Σ=1.00 (OH 1.97 F 0.03 ) Σ=2.00 [(P 12.02 Si 0.01 ) Σ=12.03 O 47 (OH) 0.21 ] [ideally, A1 K A2 Na B1 Mn B2 ⃞ Na1,2 Na 2 Na3 Na Ca Ca M Mn 13 Al (PO 4 ) 11 P1x (PO 4 ) W (OH, F) 2 ] and unit-cell dimensions a = 16.6900 (9), b = 10.1013 (5), c = 24.8752 (13) Å, β = 105.616(2)°, V = 4038.9(7) Å 3 .

A systematic chemical and structural investigation of arrojadite-group minerals has provided new data allowing us to design a consistent nomenclature scheme, approved by the IMA CNMMN (vote 05-D). The cornerstones of this scheme are: (1) the fundamental structural formula for the arrojadite group is A 2 B 2 Ca Na 2+x M 13 Al (PO 4 ) 11 (PO 3 OH 1-x ) W 2 , where A are either large divalent cations (Ba, Sr, Pb) plus vacancy, or monovalent (K, Na) cations, B are either small divalent cations (Fe, Mn, Mg) plus vacancy, or monovalent (Na) cations. (2) The dominant cation at the M sites defines the rootname: Fe 2+ , arrojadite; Mn 2+ , dickinsonite; Mg (if any), name to be proposed. (3) Two suffixes are added to the root-name according to the dominant cation of the dominant valence state at the A1 and B1 sites (the mono- or divalent nature of the cation used implicitly specifies the dominant occupancy by Na or the dominant vacancy, respectively, at the A2 and B2 sites, according to two heterovalent substitutions). (4) A third suffix is added in case the sum of non-(P,Al) cations exceeds 20.5 apfu [which implies that the Na3 site is more than half occupied and the total (OH,F) content is less than 2.5 apfu]. (5) Prefixes may be added to the root-name in the case of dominance of F over OH at the W site or of Fe 3+ over Al at the Al site. The compositional range explored is quite large, with either K, Sr, Ba, or Pb as the dominant cation at A1; Na, Fe, or Mn dominant at B1; Na or vacancies dominant at A2 and B2; Na or vacancy at the Na3 site; and F or OH at the W site. Lithium can amount up to 1 Li pfu and is partitioned into the M sites, preferentially M1. As a consequence of this new nomenclature scheme, the mineral name “sigismundite” is abolished and the corresponding composition must be referred to as arrojadite- (BaFe). In addition, arrojadites.dickinsonites from classic localities are identified as arrojadite-(KFe) (Nickel Plate Mine), dickinsonite-(KMnNa) (Branchville), fluorarrojadite-(BaFe) (Sidi-bou-Kricha), and we define the new members arrojadite-(KNa) (Rapid Creek), arrojadite-(PbFe) (Sapucaia), and arrojadite-(SrFe) (Horrsjöberg).

Synthetic samples with different chemical compositions along the hedenbergite-aegirine (CaFe 2+ Si 2 O 6 -NaFe 3+ Si 2 O 6 ) solid-solution series have been investigated by single-crystal X-ray diffraction and 57Fe Mössbauer spectroscopy. All compounds show C2/c symmetry, both at 298 K and at low temperature (100 K). The structural changes within the hedenbergite-aegirine series are dominated by the M1 site while the M2 site plays a minor role. Replacement of Fe 2+ by Fe 3+ increases the polyhedral distortion of the M1 site and causes an increased repulsion between neighboring M1 sites. The changes in M1-site geometry also induce distinct alterations within the kinking state of the tetrahedral chains, but the changes in tetrahedral bond lengths and angles are small. In addition to the single-crystal X-ray diffraction experiments, a large number of synthetic samples were investigated by 57 Fe Mössbauer spectroscopy at 298 K and, for three selected compositions, between 80 and 700 K. Here, substantial line broadening of the Fe 2+ resonance absorption was observed as an aegirine component is substituted. Two different groups of local distortion environments were observed for Fe 2+ within the solid-solution series, which change relative proportions and numeric value of the quadrupole splitting as a function of chemistry and temperature. This line broadening cannot be ascribed to discrete next-nearest-neighbor (NNN) configurations of Ca and Na as has been done in the literature. Above ~250 K, additional resonance absorption appears in the spectra of samples with aegirine components between 20 and 75 mol%. This absorption gains intensity with increasing temperature, while the 57 Fe hyperfine parameters approach values intermediate between Fe 2+ and Fe 3+ . This effect is ascribed to fast electron delocalization between Fe 2+ and Fe 3+ at elevated temperature.

Stacking structures in pyrophyllite, Al 2 Si 4 O 10 (OH) 2 , were investigated mainly by using high-resolution transmission electron microscopy (HRTEM). The specimens examined were large lath-shaped crystals (Berosovska, Urals, Russia) and massive aggregates of fine platy crystals (Nohwa, southwest Korea). Both specimens showed powder X-ray diffraction (XRD) patterns similar to those reported previously as the 2M polytype. The common stacking sequence in the two specimens is not monoclinic with two-layer periodicity as previously reported, but a uniform orientation of the 2:1 layers and near complete disorder of two alternative directions of interlayer displacement, i.e., lateral displacement between the two tetrahedral sheets across an interlayer region. The directions of interlayer displacement are about ± 2π/3 from that of the intralayer shift (lateral displacement between the two tetrahedral sheets within a 2:1 layer). Simulation of powder XRD patterns by this stacking model closely approximates the experimental pattern. Elongation of the lath-shaped Berosovska crystals corresponds to the direction of the intralayer shift, as seen in illite-1M. 2:1 layers with different orientations, and interlayer displacement almost parallel to the intralayer shift, were occasionally observed as stacking faults. Such disorder occurs more frequently in the massive Nohwa specimen than in the Berosovska specimen. Sub-micrometer domains of the 2M stacking sequence with regular alternation of the two directions of interlayer displacement were found in the Nohwa specimen.

Textures of hydrothermal quartz revealed by cathodoluminescence using a scanning electron microscope (SEM-CL) reflect the physical and chemical environment of quartz formation. Variations in intensity of SEM-CL can be used to distinguish among quartz from superimposed mineralization events in a single vein. In this study, we present a technique to quantify the cathodoluminescent intensity of quartz within individual and among multiple samples to relate luminescence intensity to specific mineralizing events. This technique has been applied to plutonic quartz and three generations of hydrothermal veins at the porphyry copper deposit in Butte, Montana. Analyzed veins include early quartz-molybdenite veins with potassic alteration, pyrite-quartz veins with sericitic alteration, and Main Stage veins with intense sericitic alteration. CL intensity of quartz is diagnostic of each mineralizing event and can be used to fingerprint quartz and its fluid inclusions, isotopes, trace elements, etc., from specific mineralizing episodes. Furthermore, CL intensity increases proportional to temperature of quartz formation, such that plutonic quartz from the Butte quartz monzonite (BQM) that crystallized at temperatures near 750 °C luminesces with the highest intensity, whereas quartz that precipitated at ~250 °C in Main Stage veins luminesces with the least intensity. Trace-element analyses via electron microprobe and laser ablation-ICP-MS indicate that plutonic quartz and each generation of hydrothermal quartz from Butte is dominated by characteristic trace amounts of Al, P, Ti, and Fe. Thus, in addition to CL intensity, each generation of quartz can be distinguished based on its unique trace-element content. Aluminum is generally the most abundant element in all generations of quartz, typically between 50 and 200 ppm, but low-temperature, Main Stage quartz containing 400 to 3600 ppm Al is enriched by an order of magnitude relative to all other quartz generations. Phosphorous is present in abundances between 25 and 75 ppm, and P concentrations in quartz show little variation among quartz generations. Iron is the least abundant of these elements in most quartz types and is slightly enriched in CL-dark quartz in pyrite-quartz veins with sericitic alteration. Titanium is directly correlated with both temperature of quartz precipitation, and intensity of quartz luminescence, such that BQM quartz contains hundreds of ppm Ti, whereas Main Stage quartz contains less than 10 ppm Ti. Our results suggest that Ti concentration in quartz is controlled by temperature of quartz precipitation and that increased Ti concentrations in quartz may be responsible for increased CL intensities.

The aim of the present study was to investigate the detailed evolution of the SAz-1 smectite in 1 M KOH at 80 °C at a solid/liquid ratio of 1/80. AFM observations indicated no change in crystal size or shape. XRD measurements at 40% relative humidity revealed changes in expandability of the smectite. The (001) reflection profile of smectite was modeled using a trial-and-error approach. The results indicate that with increasing run time, the number of non-expandable layers with zero or one water layer increases and that the coherent scattering domain size of the smectite decreases. IR spectroscopy of the reacted smectite suggests that there is no change from the initial clay products. The dehydroxylation temperature showed a slight decrease from 619 to 605 °C. STA measurements demonstrated that the cis-vacant character of the octahedral sheet remained nearly unchanged throughout the experiment. Determination of the average layer charge showed a continuous increase from 0.32 to 0.42 eq/(Si/Al) 4 O 10 , whereas the layer charge distribution indicated the appearance of high charged smectite layers with a charge of ~ 0.6 eq/(Si/Al) 4 O 10 and the disappearance of the low charged layers. XPS and SEM measurements indicate an increase of Al in the smectite samples. Isotope data support the theory of an internal diffusion mechanism by gradual changes in δ 18 O values. From these data it appears that KOH solutions provoke a mineralogical change in the 2:1 layer of the smectite minerals that increases the layer charge by increasing the Al content. This mineralogical change does not involve dissolution/crystallization processes and then must show solid-state transformation of the clays at 80 °C.

Light-induced structural changes in single crystals belonging to the β-As 4 S 4 -As 8 S 9 series and in a crystal of synthetic β-As 4 S 3 were monitored step by step by determining the unit-cell dimensions. A marked increase of unit-cell volume as a function of exposure time was observed for all the crystals belonging to the β-As 4 S 4 -As 8 S 9 series except for stoichiometric alacranite (As 8 S 9 ). No significant change upon long exposures to light was observed for the synthetic β-As 4 S 3 crystal. Crystal structure refinements were carried out for crystals with different composition at selected steps of the light-induced process. The structural results clearly showed that the percentage of the As 4 S 5 molecule in the structure increases when a crystal is exposed to light. Therefore, the increment of the unit-cell volume induced by light exposure appears to be related to a random replacement of As 4 S 5 for As 4 S 4 in the structure according to the reaction 5As 4 S 4 + 3O 2 → 4As 4 S 5 + 2As 2 O 3 . The results obtained in the present study combined with a critical review of data previously published indicate that the As 4 S 4 molecule is able to incorporate sulfur to convert to As 4 S 5 upon exposure to light, whereas either As 4 S- or As 4 S 5 molecules do not go undergo any modification. It appears that the extent of sulfur incorporation is strictly controlled by the type of molecular packing as well as by the kind of molecule. A final, complete conversion to pararealgar was observed only for pure β-As 4 S 4 , whereas nonstoichiometric As 8 S 9-x crystals initially containing variable amounts of β-As 4 S 4 microdomains convert only partially to pararealgar upon light exposure.

Most carbon on Earth is bound within minerals, and increasing the uptake of atmospheric carbon dioxide into minerals may reduce the greenhouse gas content of the atmosphere. We document carbon disposal through the mineralization of mine tailings at Clinton Creek, Yukon Territory, and Cassiar, British Columbia. We confirm crystallographic binding of carbon in these tailings and quantify carbon dioxide uptake using quantitative phase analysis with the Rietveld method for X-ray powder diffraction data. Planar disorder in the structures of the kaolinite-serpentine group minerals makes Rietveld refinements of X-ray powder diffraction data for serpentinites problematic. Using structureless pattern fitting and with the addition of a known quantity of a well-crystallized material, the problem of structural disorder is overcome by considering the serpentine minerals as amorphous phases. We test the accuracy and precision of this refinement method using synthetic serpentine-rich mine tailings of known composition. Estimates of the abundance of hydrated magnesium carbonates in these tailings have a precision of approximately 5% relative for mineral species present in amounts greater than 10 wt%. Precise estimates of carbonate mineral content and crystallographically bound atmospheric CO 2 are made for samples of serpentine-rich tailings from Clinton Creek and Cassiar. Results for mine tailings are also compared to mineralogically similar samples from a carbonate playa at Atlin, British Columbia. The potential for decomposition of metastable hydrated magnesium carbonate phases to geologically stable magnesite may represent long-term stability of the products of mineral sequestration in mine tailings.

To obtain higher pressures in a Kawai-type apparatus, we explored and tested high-pressure cell assemblies suitable for experiments using sintered diamond (SD) anvils. As a result, we succeeded in generating pressures exceeding 55 GPa at temperatures of ~1000 °C in a Kawai-type apparatus (SPEED Mk-II) at SPring-8. Using the optimized cell assembly, we examined the stability field of MnGeO 3 perovskite, an analog of MgSiO 3 perovskite, which was recently found to transform to a new high-pressure form under the P-T conditions near the core-mantle boundary. From our in situ X-ray observations, MnGeO 3 perovskite was found to be stable at pressures up to 56.57 GPa at temperatures of 800.1050 °C.

A thermodynamic model to predict phase equilibrium of methane and carbon dioxide hydrate in aqueous electrolyte solutions is presented. Using the Pitzer model to account for the variation of water activity due to electrolytes and dissolved gas in aqueous solutions, we extended the model based on ab initio molecular potential developed recently by us for the CH 4 -H 2 O and CO 2 -H 2 O binary systems to the CH 4 (or CO 2 )-H 2 O-salts system. Comparison of the model with extensive experimental data indicates that this model can accurately predict the phase equilibrium of CH 4 hydrate and CO 2 hydrate in various electrolyte solutions (such as aqueous NaCl, KCl, CaCl 2 , NaCl + KCl, NaCl + CaCl 2 solutions, and seawater) from zero to high ionic strength (about 6 m) and from low to high pressures.

Single-crystal polarized Raman spectra (40 to 4000 cm.1 at 4 ≤ T ≤ 860 K) were measured for hemimorphite, Zn 4 Si 2 O 7 (OH) 2 ·H 2 O, to determine the behavior of H 2 O molecules and OH groups in the channels. All observed stretching modes of the O-H vibrations could be assigned. Low-temperature Raman spectra show the presence of two different spatially divided regions in the crystal where the H 2 O molecules and hydroxyl groups have slightly different positions in the channel. Two structural phase transitions were observed at 20.30 and 80.100 K, respectively, which reflect temperature replacement and reorientation of the H 2 O molecules in the cavity. At T > 100 K, H 2 O molecules are rotationally disordered around the c axis. An analysis of combined modes consisting of external and internal H 2 O vibrations permits the frequency of translational T(H 2 O) modes at 30, 53, 71, 90, 118, and 148 cm -1 to be determined. The T(H 2 O) modes also can be observed directly at low frequencies in the Raman spectra. The dehydration process of hemimorphite was investigated by Raman measurements at elevated temperatures of the crystal. One of two hydroxyl groups in the dehydrated crystal is hydrogen-bonded to neighbor oxygen and disordered statistically.

Stacking structure in disordered talc, Mg 3 Si 4 O 10 (OH) 2 , has been investigated by using highresolution transmission electron microscopy (HRTEM) and by comparison between experimental and simulated powder X-ray diffraction (XRD) patterns. The talc specimen investigated was massive aggregates of fine platy crystals from the Shirakashi mine, Nagasaki-Prefecture, Japan, formed from regional metamorphism. HRTEM observations revealed that the orientation of the 2:1 layer, which is described by the direction of the lateral shift of a/3 from the lower tetrahedral sheet to the upper tetrahedral sheet within the 2:1 layer (intralayer shift), is almost completely disordered. In contrast, lateral displacement between the adjacent tetrahedral sheets across the interlayer region (interlayer displacement) is relatively ordered. The interlayer displacement parallel to the intralayer shift in lower or upper 2:1 layers tends to be avoided. A Gandolfi camera was used to record XRD patterns from a small fragment of the aggregates, to avoid preferred orientation and artifacts in stacking sequence by grinding. XRD patterns were simulated using the DIFFaX program and compared with experimental patterns. The experimental pattern from hk = 02, 11, and 1 .1 reciprocal lattice rows is explained by a mixture of the stacking sequence with almost random directions of intralayer shift and a small amount of 1A dominant stacking sequence. The peak profiles for the reflections indexed with 20l, 13l, or 1 .3l are reproduced by considering their variance, and a slightly different interlayer displacement from that reported in talc-1A.

Electrical conductivities of polycrystalline garnets ranging in chemical composition from almandine (Fe 3 Al 2 Si 3 O 12 ) to pyrope (Mg 3 Al 2 Si 3 O 12 ) were measured at 10 GPa and 19 GPa at temperatures ranging from 300 to 1700 °C using complex impedance spectroscopy in a multianvil device. Mössbauer spectroscopy of each sample was carried out both before and after the electrical measurements to characterize the oxidation state of Fe in the almandine bearing garnets. Similar to the behavior of other ferromagnesian silicates, the substitution of Fe for Mg along this compositional join dramatically increases electrical conductivity, but this compositional effect is reduced with increasing temperature. Conductivities increase with increasing total Fe content, as the average Fe 2+ -Fe 3+ distance decreases. At 10 GPa, activation energies for conductivity vary smoothly with composition and increase rapidly toward the pyrope end-member composition, where it reaches a value of 2.5 eV. The results are consistent with an electrical conductivity mechanism involving small polaron mobility in the Fe-bearing garnets at 10 GPa. At 19 GPa, however, there is virtually no change in the activation energy as a function of Fe-Mg substitution for the pyrope-rich garnets. These higher pressure measurements reflect a mechanism involving oxygen related point defects, as conductivities increase with pressure at constant T for each garnet, and the effect of pressure is greater for the more Mg-rich garnets. The data also allow for a more quantitative evaluation of the effect of chemical composition, specifically Fe-Mg substitution, on the electrical conductivity profile of the mantle, using a recently developed laboratory- derived model. We apply the model using these data to a portion of the transition zone between 520 and 660 km, in which we vary the garnet composition from Py 100 to Py 85 Alm 15 . Although only a minor effect on bulk mantle conductivity results, we conclude that the overall garnet composition may, however, be important in characterizing the magnitude of any EC discontinuity with respect to the above-lying mantle.

The crystal structures of the 1M and 2M polytypes of tyrolite have been solved from single-crystal X-ray diffraction data. The structure of tyrolite-1M [monoclinic, P2/c, a = 27.562(3), b = 5.5682(7), c = 10.4662(15) Å, β = 98.074(11)°, V = 1590.3(3) Å 3 ] has been refined to R 1 = 0.086 on the basis of 2522 unique observed reflections collected using synchrotron radiation at the Swiss-Norwegian beamline BM01 of the European Synchrotron Research Facility (SNBL at the ESRF). The structure of tyrolite-2M [monoclinic, C2/c, a = 54.520(6), b = 5.5638(6), c = 10.4647(10) Å, β = 96.432(9)°, V = 3154.4(6) Å 3 ] has been refined to R 1 = 0.144 on the basis of 2666 unique observed reflections obtained from a non-merohedrally twinned crystal using in-house X-ray radiation and a STOE IPDS II imageplate diffractometer. The structures are based upon complex nanolayers consisting of Cu, As, and Ca coordination polyhedra. The core of the nanolayer is a copper arsenate substructure consisting of A and B sublayers. The B sublayer consists of chains of edge-sharing Cu octahedra running along the b axis. The A sublayer contains trimeric units of Cu octahedra sharing corners with AsO 4 tetrahedra. Two adjacent A sublayers are linked by the octahedral chains of the B sublayer resulting in formation of the 18 Å thick ABA slab. The ABA slab is sandwiched between sublayers of Ca 2+ cations and H 2 O molecules. Adjacent nanolayers are connected by hydrogen bonds to the interlayer species (carbonate anions and H 2 O molecules). The structures of tyrolite-1M and tyrolite-2M differ by the stacking sequence of the nanolayers only. The adjacent nanolayers in tyrolite-2M are shifted by b/2 = 2.8 Å in comparison to the relative position of the nanolayers in tyrolite-1M. The structural formula of tyrolite can be written as [Ca 2 Cu 9 (AsO 4 ) 4 (OH) 8 (CO 3 )(H 2 O) 11 ](H 2 O) x where x = 0-1.

We have investigated trace-element partitioning between clinopyroxene and silicic melts ranging from 62 to 69 wt% SiO 2 in piston-cylinder experiments at 1.1 GPa and 1145.1155 °C. Run products consist of diopsidic clinopyroxene (with 3.5 wt% Al 2 O 3 and ~1 wt% Na 2 O) coexisting with silicic, aluminous, and alkaline melts. Clinopyroxene-melt partition coefficients for rare earth elements (REE) are significantly higher (e.g., D Sm = 0.76 to 3.1) than previous studies of clinopyroxene-basaltic melt partitioning. The maximum D REE is also shifted toward middle REE (e.g., Tb). Although the partition coefficients of Sr and Ti (D Sr = 0.23.0.39, D Ti = 0.28.1.0) are also elevated relative to clinopyroxenebasaltic melt, partition coefficients for Zr and Li (0.14.0.23 and 0.16.0.25, respectively) are within the range of literature values. The high partition coefficients for REE, Sr, and Ti in this study do not reflect crystal-compositional effects. For instance, no relationship is observed between IV Al and these partition coefficients. REE and Ti partition coefficients correlate poorly with jadeite component of the clinopyroxene; however, the elevated Na 2 O contents of both the melt and clinopyroxene make it difficult to discriminate fully the effect of jadeite component from the effect of changes in melt structure. The non-bridging oxygen to tetrahedral cation ratio (NBO/T; Mysen et al. 1985) varies from 0.08 to 0.17 and roughly correlates with cpx/melt D REE , consistent with a recent study (Gaetani 2004) that concluded that melt structure significantly affects partitioning when NBO/T < 0.49. Notably, we find a strong correlation between the molar Ca 2+ /(M + + M 2+ ) of the melt and the optimum D (D 0 ) for the REE, indicating that melt structure, and not jadeite content, is the dominant control on clinopyroxene-melt partitioning in these experiments. The partition coefficients for the REE, Ti, Zr, and Sr are in good agreement with observed clinopyroxene- silicic glass partitioning within mantle xenoliths (Chazot et al. 1996; Vannucci et al. 1998), indicating these results may apply to lithospheric melting. Furthermore, the origin of hump-shaped REE patterns observed in residual clinopyroxene from abyssal peridotites could be explained if shallow depleted melts of peridotite are silica-rich. A two-stage fractional melting model assuming clinopyroxene- silicic melt partitioning at shallow mantle depths from 10 to 30 km beneath a mid-ocean ridge can reproduce the observed hump-shaped REE patterns. If so, hump-shaped REE patterns may provide evidence for the existence and extraction of silica-rich melt created in the shallow mantle.

We report techniques for in-situ abundance measurements of the light-lithophile elements (LLE; Li, Be, and B) in silicate glasses by laser-ablation inductively coupled mass spectrometry (LA-ICP-MS), and compare the analytical performance of a sector field and quadrupole mass analyzer for these measurements. LA-ICP-MS is shown to be an effective means of determining LLE abundances at spatial scales between 25 and 100 µm. Detection limits depend on instrumental sensitivity and ablation rate, but can be in the low- to sub-ng/g range. Measured ion yields for 7 Li, 9 Be, and 11 B ion, normalized to 43 Ca as an internal standard, remain largely constant during ablation, although 11 B shows a relative increase once the ablation crater aspect ratio exceeds ~1. Surficial contamination, particularly of B, can be removed rapidly via a short pre-ablation (~20 pulses) immediately prior to analysis. The sector field ICP-MS provided considerable improvements in analytical performance over the quadrupole mass analyzer, although longer magnet settling times result in greater duty cycle losses. Calculated detection limits for a given set of ablation conditions are 20.90 times lower, and useful yields (the ratio of atoms ablated to counts detected) are 20.75 times greater for the sector field instrument. Analysis of reference glasses shows that LA-ICP-MS provides accurate measurements of Li, Be, and B contents in silicate glasses over a range of compositions (komatiite to rhyolite). LA-ICP-MS also offers similar accuracy and precision and marked improvements in sample throughput compared to secondary ion mass spectrometry (SIMS) analysis of LLE abundances, although SIMS has higher useful yield and thus provides better spatial resolution.

Stornesite-(Y), end-member formula Y⃞ 2 Na 6 (Ca 5 Na 3 )Mg 43 (PO 4 ) 36 , is a new Y-dominant analog of the meteoritic mineral chladniite. A representative electron microprobe analysis is SiO 2 = 0.02, P 2 O 5 = 48.11, SO 3 = 0.05, MgO = 23.16, MnO = 0.24, FeO = 15.55, Na2O = 5.04, CaO = 5.66, SrO = 0.02, Y 2 O 3 = 1.43, Yb 2 O 3 = 0.24, UO 2 = 0.01, Sum = 99.53 wt%, which gives Y 0.68 Yb 0.06 Na 8.69 Ca 5.40 Sr 0.01 Mg 30.71 Fe 11.56 Mn 0.18 Si 0.02 S 0.04 P 36.22 O 144 . Overall, Y + REE range from 0.542 to 0.985 atoms per formula, and atomic Mg/(Mg + Fe) ratio from 0.684 to 0.749. Single-crystal X-ray diffraction gives trigonal symmetry, R3̅, a = 14.9628(27) Å, c = 42.756(11) Å, V = 8290(4) Å 3 , calculated density = 3.196 g/cm3, Z = 3. The mineral is isostructural with synthetic chladniite, but the (0, 0, 0) site is dominantly occupied by Y instead of Ca. Bond lengths are considerably shorter than for Ca sites; Y and Yb are fully ordered at this site, which is our rationale for recognizing stornesite-(Y) as a distinct species. The strongest lines in the powder pattern [d in Å, (I), (hkl)] are 3.67 (40) (0 3 6, 3 0 6), 3.52 (40) (0 0 12, 3 1 2, 1 3 2̅), 2.94 (60) (0 1 14, 3 2 2̅, 2 3 2), 2.73 (100) (2 0 14, 0 3 12, 3 0 12), 1.84 (40) (1 5 14, 5 11̅4̅ , 0 6 12, 6 0 12). The mineral is optically uniaxial +, n ω = 1.6215(10) and n ε = 1.6250(10) at 589 nm. Its color is pale yellow in standard thin sections. Stornesite-(Y) is found as inclusions in fluorapatite nodules in two paragneiss specimens from Johnston Fjord, Stornes Peninsula (whence the name) and in a third from Brattnevet, Larsemann Hills. Associated minerals are wagnerite, xenotime-(Y), monazite-(Ce), P-bearing K-feldspar, biotite, sillimanite, quartz, and pyrite; it is commonly altered to rusty material and secondary phosphates. Grains are anhedral, subhedral, or locally euhedral with hexagonal or rhombic outlines; maximum dimensions are 1 × 0.25 mm. It is inferred to have formed at 800.860 °C, 6.7 kbar by reaction of biotite with an anatectic melt locally enriched in P by interaction with fluorapatite.

X-band electron paramagnetic resonance (EPR) spectra of gamma-irradiated crystals of carbonatebearing fluorapatite from the Levant mine, Cornwall, England, revealed the presence of two previously characterized centers (i.e., an O − defect and an O − ⃞ F defect, where ⃞ F represents a vacancy at the F site), a CO 2 − radical, and a new oxygen-associated hole-like center in the anion column. The O − ⃞ F center in carbonate-bearing fluorapatite is stable at room temperature, whereas in carbonate-free fluorapatite the stability of this radical is shifted to lower temperatures (<225 K). The CO 2 − radical herewith first reported in carbonate-bearing fluorapatite is characterized by an axial symmetry at room temperature but a weakly orthorhombic symmetry at 77 K, similar to its counterpart in carbonate-bearing hydroxylapatite. This CO 2 − radical most likely formed from Type A carbonate ions by the loss of an O atom and trapping of an electron during gamma irradiation. The single-crystal EPR spectra of the new holelike center are characterized by the absence of any hyperfine interactions and a strongly orthorhombic symmetry. The spin Hamiltonian parameters of this new center suggest a structural model involving the trapping of a hole by a substitutional oxygen ion sandwiched between two fluorine ion vacancies in the anion column and strongly disturbed by vacancies at the neighboring Ca 2 and O 3 sites, suggesting a complex substitution of the type: □ F O 2- □ F + □ Ca2 + CO 3 2- → F-F-F-+ Ca 2+ + PO 4 3- .

EXAFS spectroscopy has been used to determine the medium-range structural environment (within a radius of ≈7 Å) of trace levels of Zn 2+ ions within the octahedral sheets of trioctahedral 2:1 phyllosilicates. EXAFS signatures of trace quantities (800.4000 ppm) of Zn in natural and synthetic trioctahedral 2:1 phyllosilicates (talcs and biotite) were analyzed using ab initio FEFF calculations (FEFF 8.10 code) in which various second-neighbor atomic configurations around a Zn-containing octahedral site were examined. Comparison of the results of these model calculations with observed Zn K-edge EXAFS spectra provides constraints on the distribution of Zn within the octahedral sheet. Zn was found to be randomly distributed within the octahedral sheets of the synthetic talc containing 4000 ppm Zn. In the natural biotite containing 800 ppm Zn, the distribution of Zn could not be determined because of significant Fe content, which precluded unambiguous FEFF analysis. Finally, FEFF analysis of single-scattering and multiple-scattering contributions to the EXAFS data of the Zn-dilute talc sample, followed by comparison with EXAFS data from a natural dioctahedral illite sample containing 140 ppm Zn, allowed unambiguous identification of spectral features characteristic of Zn 2+ ions incorporated within the octahedral sheets of trioctahedral phyllosilicates. These results indicate that EXAFS spectroscopy can be used to distinguish between incorporation of trace levels of Zn 2+ ions within the octahedral sheet of di- or trioctahedral phyllosilicates. They also provide a strong basis for determining the speciation of Zn at trace concentration levels in natural phyllosilicates from Zn-contaminated soils and sediments using Zn K-edge EXAFS spectroscopy. Because the crystal chemistry of other potentially toxic first-row divalent transition elements (i.e., Co 2+ , Ni 2+ , and Cu 2+ ) is similar to that of Zn 2+ , this FEFF-based EXAFS approach could be used to (1) complement existing FTIR and polarized-EXAFS approaches in assessing the distribution of these elements within the octahedral sheets of di- and trioctahedral plyllosilicates and (2) better determine the speciation of these elements in contaminated soils and sediments.

Cation ordering in two important iron nickel sulfide minerals, pentlandite and violarite, was studied by neutron powder diffraction using samples prepared with isotopically enriched 60 Ni. Pentlandite of composition Fe 4.8 Ni 4.2 S-, annealed at 150° C for 1 month has a disordered Ni/Fe distribution with Ni occupying 57(2)% of the octahedral site and 46(1)% of the tetrahedral site with a unit cell repeat of 10.1075(1) Å. After annealing for a further 2 months at 150 °C the Ni/Fe distribution is still disordered with 53(2)% Ni in the octahedral site and 48(2)% Ni in the tetrahedral site and the cell parameter increased by 0.21% possibly due to slight readjustment of the Fe/Ni ratio. Synthetic violarite, FeNi 2 S 4 , prepared at 300 °C, exhibits ordering of Ni onto the tetrahedral site, with Fe and the remaining Ni sharing the octahedral site. In situ neutron diffraction heating experiments in which pentlandite and violarite were heated up to 300 °C for 12 hours did not alter the Ni/Fe distribution between tetrahedral and octahedral sites. Violarite did not exhibit any evidence of magnetic ordering upon cooling to .173 °C. The interatomic distances in the two structures indicate that in pentlandite Fe 2+ and Ni 2+ are high spin and in violarite Fe 2+ is low spin with Ni 3+ with an inverted spinel structural type.

Five expanded vermiculite samples known to have originated from the former vermiculite mine near Libby, Montana were analyzed using powder XRD to quantify their amphibole content. An expanded vermiculite with no detectable amphibole based on our XRD methods discussed herein was spiked with 100, 500, 1000, 2500, 5000, 7500, and 10 000 ppm Libby amphiboles to determine the detection limit of XRD and to develop a calibration method. The 110 amphibole peaks were not detectable in the 100 and 500 ppm samples; thus, defining the detection limit. The 110 peak areas for the remaining samples were measured and used to form a calibration curve relating peak area to ppm amphibole. Of the five Libby samples, each had detectable levels of amphibole and their amphibole content ranged from 1171 to 9218 ppm. The XRD method provides a bulk measurement of the total amphibole content. To determine the amount of asbestiform amphiboles, morphological considerations obtained by microscopy techniques are required. From the results of a previous polarized light microscopy study, the five expanded vermiculites in this study would contain from 390 to 3073 ppm asbestiform amphiboles and not be subject to regulation on the basis of the 1% rule used by regulatory agencies in the U.S.A.