V33B-2210
The use of high resolution analytical methods for the characterization of reaction microstructures
Crystallization of new minerals in metamorphic rocks is strongly controlled by chemical transport rates towards the crystallization sites. Grain and phase boundaries play an essential role in this transport, but information on geometrical and physical properties of boundaries and potential boundary materials are largely absent. In order to gain insights into chemical transport processes during the crystallization of new phases we explored reaction microstructures by using high-resolution analytical techniques. Reactions of type A + plagioclase (Pl) = garnet (Grt) + C form garnet rims between reactants. EBSD microstructure analysis reveals that the reaction rims are composed of grains and subgrains, which form a palisade microstructure perpendicular to the reaction fronts. Based on the known geometry of the grain boundary network, TEM foils can be cut by FIB perpendicular to garnet grain boundaries and TEM investigations reveal distinct nanometer scale compositional patterns across the garnet grain boundaries. In addition, FEG-EPMA and AEM allows to quantitatively measuring micron-scale asymmetric compositional zoning patterns across the reaction rims. These compositional patterns were formed during short-circuit diffusion, which amplifies the necessary mass transfer across the growing reaction rims. Information about the contribution of grain boundary diffusion to bulk material flow in Grt is stored during the formation of these growth zonings and allows for the derivation of Digb/Divol ratios and rim growth rates by diffusion modeling. Reaction progress needs the transfer of material through the interior of reacting Pl towards the reaction fronts. TEM investigations reveal about 100 nm wide pores in Pl and Grt-Pl phase boundaries, which both are filled with a non-crystalline material. This suggests that a coupled diffusion and dissolution/precipitation process, during which internal and surface dissolution/precipitation formed new nano-scale pathways and a non-crystalline transport medium, likely enhanced mass transfer in Pl and along Grt-Pl phase boundaries.
V33B-2211
Growth Controls in Colloform Sulfide Textures
Colloform textures have been described from many of the world's sulfide ore deposits and involve formation of distinct microcrystalline layers. These textures provide information related to sequential stages of ore formation and yet their mechanism and controls of growth are still poorly understood. Samples of colloform sphalerite from the Galmoy Zn-Pb ore body, Ireland and colloform pyrite from Greens Creek, Alaska have been investigated using crystallographic, in-situ S isotope and trace element chemistry analysis. Electron backscatter diffraction (EBSD) results indicate the crystal preferred orientation (CPO) of discrete microcrystalline colloform layers abruptly switch between <100>, <110> and <111> orientations in all of the samples investigated. Examination of the same layers in the Galmoy colloform textures (sphalerite) using in-situ laser S isotope analysis reveals equally striking and abrupt changes in δ34S signature between end-member bacteriogenic (-25‰) and hydrothermal (+10‰) sources. However, there is no correlation between shifting CPO and S isotope signature at Galmoy. The Greens Creek colloform textures (pyrite) preserve a similar, if less dramatic, variation in δ34S signature (-40‰ to -30‰) but again there is no correlation between changing δ34S signature and CPO in discrete layers. Trace element analysis of both suites of colloform samples reveals major changes in trace element sequestration. The variation in trace element abundance however appears to correlate with changes in δ34S signature rather than CPO. At Galmoy Cd and Cl rich layers appear to correspond with a hydrothermal sulfur signature while Sb sequestration at Greens Creek corresponds with a relatively heavy bacteriogenic δ34S (-30‰) signature. While factors such as bacteria induced mineralisation cannot be ruled out, the temperature of the ore forming environment suggests this process is likely to be restricted and therefore not the primary control on CPO changes. In the absence of evidence for alternative mechanisms it is suggested that the most likely factors influencing CPO changes within discrete layers in the colloform textures are temperature and degree of supersaturation.
V33B-2212
Garnet Porphyroblastesis: Growing Inward or Outward?
The microstructure, composition and crystallographic orientation of a garnet porphyroblast in the garnet-zone schist, Imjingang belt, Korea, were investigated in order to delineate chemical and microstructural processes during the crystallization. This garnet hypidioblast is ~1 mm in size, and consists of relatively inclusion- poor core and inclusion-rich rim. The inclusion minerals, mainly composed of quartz together with minor ilmenite and clinozoisite, are distributed in complex patterns. In general, inclusion trails are discontinuous in the core region of garnet, but apparently curved to wrap around the core. The presence of TiO2 needles in the core part suggests that garnet replaced a Ti-bearing precursor such as biotite. Compositional zoning profile of the garnet porphyroblast is characterized by bimodal distribution of the spessartine component: e.g., Mn-poor core and rim bounded by Mn-rich intermediate part. The zoning pattern of grossular varies in an antithetic fashion to that of spessartine. These microstructural and compositional features are different from those of the majority of other garnet porphyroblasts in metapelites, including: (1) relatively inclusion-rich core of syn-kinematic garnet growing mainly at the expense of chlorite; (2) post- kinematic garnet overgrowth replacing the biotite porphyroblast; and (3) monotonous decrease in the spessartine content towards the rim. Electron back-scattered diffraction analyses of garnet reveal multiple, intracrystalline domains, less than 200 μm in size. These domains show small angular differences (1°-2°) in orientation across narrow boundaries, and are common in the Mn-rich intermediate part of garnet. However, they are absent in the Mn-poor core region. The lack of compositional anomalies and nearly identical crystallographic orientations in the intracrystalline domains suggest an absence of multiple nuclei, but the implications for this crystallographic feature are uncertain. All the above-mentioned microstructural, compositional and crystallographic features suggest that the nucleation and initial growth of garnet took place in the intermediate part of garnet porphyroblast which enveloped the rigid, competent precursor in the core, and that subsequent growth of garnet proceeded simultaneously towards the rim and core parts. Alternatively, the nucleation and initial growth of garnet might have occurred at the expense of biotite, under the conditions of small scale of equilibrium for Mn.
V33B-2213
Garnet Polycrystals in a Barrovian Sequence in Dutchess County, New York
An understanding of porphyroblast growth mechanisms is crucial to the study of fluid and element transport during metamorphism. Garnet porphyroblasts are particularly interesting because they preserve a record of metamorphic reaction history and past metamorphic pressure and temperature conditions. Although garnet has long been thought to occur most typically as single crystals, electron backscatter diffraction (EBSD) analysis of garnet has revealed the existence of polycrystalline porphyroblasts that consist of several domains, each with a different lattice orientation separated by high-angle boundaries. New results from an EBSD investigation of metapelitic schists from a complete Barrovian sequence in Dutchess County, New York, document the presence of garnet polycrystals in several different Barrovian zones. The Barrovian sequence allows us to observe garnet growth history during progressive metamorphism because a range of P-T conditions are represented, from 430° C and 3 kbar in the garnet zone to 600° C and 6 kbar in the sillimanite zone. Lutetium-hafnium dating results suggest that garnet grew entirely during Taconic metamorphism at ~430 Ma. Polycrystals were found in the garnet, staurolite and kyanite zones and therefore were either able to grow or survive over a range of P-T conditions. Of 220 garnets analyzed, 10 were polycrystals (4.5%), including both cryptic (no morphological expression) and morphologically distinct polycrystals. Some lattice misorientations (18%) across domain boundaries correspond to coincidence site lattices in cubic minerals, but this incidence is not significantly greater than the 15% predicted for a random distribution in a cubic mineral. Zoning of Fe, Mg, Ca and Mn within polycrystals cuts across domain boundaries and is concentric within the polycrystal as a whole. The formation of polycrystals suggests that closely spaced nuclei may form in a chemically or mechanically heterogeneous matrix, e.g. within a chemically and/or kinetically favorable precursor mineral that has been consumed or a distinct microstructural site.
V33B-2214
Characteristics of Polycrystalline Garnets in Micaschists From the Southern Menderes Massif (Turkey) and the Solitude Range (BC, Canada)
Electron backscatter diffraction (EBSD) analysis of garnets in metamorphic rocks has revealed the presence of grain boundaries within what appear, based on morphology, to be single crystals. There have only been a few previous studies that have described these types of polycrystals in nature. In this study we analyzed garnets from two suites of metamorphic rocks: kyanite-staurolite schist from the Solitude Range, SW Rocky Mountains (BC, Canada), and mica schist from the southern Menderes Massif (western Turkey). Garnets from both sites are growth zoned and formed during a single metamorphic event, although the Solitude Range garnets record in their zoning and inclusion textures a change from chloritoid-present to staurolite- present (chloritoid-out) reaction history. The garnet-bearing rocks from these sites formed at P-T conditions of 430-550 C, 7-8 kbar (Menderes) and 550-600 C, 6-7 kbar (BC). Less than 10% of the garnets analyzed are polycrystals, but all polycrystals detected have similar characteristics: high-angle misorientation boundaries that crosscut inclusions and inclusion trails. Most polycrystals have 2-3 domains (crystals), but one complex polycrystal was comprised of 16 distinct lattice domains. In most cases, misorientation boundaries crosscut growth zoning, but one Menderes polycrystal exhibited distinct zoning in each domain. Most polycrystals likely formed early in the garnet growth history as closely-spaced nuclei coalesced, but clustering (coalescence) continued throughout the history of garnet crystallization in these rocks.
V33B-2215
Non-uniqueness of Porphyroblast CSDs
Crystal size distributions (CSDs) of porphyroblasts cannot uniquely identify crystallization mechanisms: distinct mechanisms can yield indistinguishable CSDs, and a single mechanism can generate CSDs with markedly different characteristics. Porphyroblast CSDs record of rates of nucleation and growth, integrated over the crystallization interval. They therefore have the potential to elucidate processes operating during metamorphic crystallization. Attempts have been made to extract crystallization mechanisms from porphyroblast CSDs by comparing measured size distributions to the predictions of forward models for specific reaction mechanisms (e.g., Eberl et al., 2002, Am Min 87:1235). Our numerical simulations of diffusion-controlled nucleation and growth demonstrate, however, that common porphyroblast CSDs cannot be uniquely linked to any particular mechanism. This work employed a 3D finite-difference model that tracks the evolution in time and space of the supersaturation of a single rate-limiting component (e.g., Al), during dissolution of a reactant assemblage (e.g., chl+qtz) and diffusion-controlled nucleation and growth of a product porphyroblast (e.g., grt). Nucleation rates are functions of local supersaturation and temperature; growth rates depend upon local diffusional fluxes driven by concentration gradients. By varying initial conditions (distribution of reactants, temperature and pressure at onset of reaction, equilibrium concentration of the diffusing element in the intergranular medium at the onset of reaction, volume fraction of interconnected porosity) and kinetic parameters (pre-exponential constants and thermal- acceleration factors for nucleation and diffusion), this model produces, for a single crystallization mechanism, a wide range of CSDs that span the range of distributions measured in natural occurrences. But if these CSDs were to be interpreted in the framework of prior studies, for many the analysis would spuriously imply the operation of processes other than diffusion-controlled nucleation and growth (e.g., size-dependent growth, Ostwald ripening). We conclude that crystallization mechanisms cannot be uniquely determined from porphyroblast CSDs alone. Such determinations require approaches that assess and integrate a variety of textural and microchemical features.
V33B-2216
Charge Contrast Imaging of Zircon in Amphibolite Facies Rock of the Central Appalachian Piedmont, SE Pennsylvania
Charge contrast imaging using an environmental scanning electron microscope (ESEM) provides high resolution images of internal zoning in zircon crystals in uncoated thin sections. To locate all zircon (and monazite) crystals within a thin section we utilize an automated routine which combines back-scattered imaging and an integrated energy dispersive spectroscopy (EDS) system. The brightest objects in a given field are located and a two-second EDS spectrum is obtained, along with information on size, shape, and location of the bright features. Objects are thus readily sorted by size and individual grains can be selected for further analysis. Differential charging of compositional domains within zircon crystals occurs under low vacuum conditions (70 – 240 Pa, H20) and is imaged using secondary electrons. Here we examine the internal and external morphology of the zircon crystals in upper amphibolite facies rocks from SE Pennsylvania in the central Appalachian Piedmont, to gain insight into the nature of the protolith and to provide context for planned ion-probe analysis. The Chester Park gneiss has been mapped and described as consisting of meta-volcanic, meta-plutonic and/or meta-sedimentary rock. Zircon in the Chester Park gneiss contains rounded, detrital cores, some of which exhibit concentric oscillatory zoning, and euhedral to subhedral metamorphic overgrowths. These observations are consistent with a supra-crustal origin for this unit. Similar zircon is present in the Fairmount member of the Wissahickon Formation, which is present along strike from the Chester Park gneiss. In contrast, zircons present in the Waterworks gneiss, which underlies the Philadelphia Museum of Art, are euhedral and exhibit oscillatory zoning indicating magmatic origin for this rock.
V33B-2217
The Scrutinization Of Shock Textures By a FIB-TEM Technique
Some meteorites record high-grade (shock) metamorphism. The shock metamorphism was induced by the collisions of planetesimal in the solar nebula. A part of the meteorites was deformed and heated by friction- induced melting along fractures or localized concentration of stress. Melting along fractures is called gshock-melt veinsh. Silicate minerals in the shock-melt veins were transformed to their high-pressure polymorphs. These high-pressure polymorphs are extraterrestrial analogues of the phases in Earthfs lower mantle. It is important to clarify their mechanisms of formation either by solid-state phase transformation and/or if also by localized fractional crystallization in mineral melt pockets to understand the evolution of Earthfs lower mantle. In addition, the magnitude of the equilibrium pressure and the time scale of the dynamic events could be meaningfully estimated. However, successful investigations require precise surgical cutting and removal of the fine-scale assemblages in nm-thin slices to warrant successful TEM-investigation. We applied a combination technique of a focused ion beam (FIB) system and TEM to investigate the mechanisms. FIB-TEM technique allowed us to precisely study sub-micron textures. This resulted in novel findings and revision of previous models.
V33B-2218
Crystal Packing in a Mush – the Role of Grain Shape
Particle packing controls the physical properties of accumulations of grains. Grain size and shape in a crystal mush affects the amount of pore space and its interconnectivity, controlling permeability and mechanical strength during compaction. The spatial distribution and aspect ratio of grains also has an influence on mechanical strength during expansion, with application to the survival of entrained fragments of crystal aggregate during decompression-driven vesiculation in erupting magma. At present very little is known about the controls of grain shape, and in particular aspect ratio, on the physical properties of a crystal mush layer. We use packings of rice and beads as analogue models to investigate how the packing is affected by particle shape and the accumulation rate of particles at the top of the mush. Alignment factors, derived from eigenvectors, are used as a measure of the preferred grain orientations. Results show that grain anisotropy is an important factor when considering the structure and behaviour of a crystal mush. In addition, it is believed that high alignment factors seen in outcrop, which are sometimes attributed to dissolution and reprecipitation of material under compressive strain, may in fact be largely due to the original crystal shape. This means that estimates of the volume of fluid/melt liberated during compaction may have to be reconsidered.
V33B-2219
SEM and TEM Investigation of Crystal Nucleation in Silicate Melts During Isothermal Decompression
Isothermal constant-rate decompression experiments were performed on H2O-saturated, nearly aphyric (<2 v.% crystals) natural rhyodacite to determine rates of crystal nucleation and growth relevant to natural magma ascent during volcanic eruptions. Experiments quenched at different pressures along a given decompression trajectory allow investigation of factors that influence progressive changes in crystal texture, such as time-varying crystal growth rate, crystal synneusis and ripening, and crystallographically controlled growth rate anisotropy. Textural analyses at the millimeter to nanometer scale using SEM and TEM are examining the incipient stages of crystal growth and addressing aspects fundamental to the nucleation process, such as whether crystal nucleation occurs homogeneously or heterogeneously. Samples are prepared for TEM using conventional thinning techniques as well as focused ion beam (FIB) milling and liftout. FIB samples are smaller (typically 6x8 μ m in area), but more uniform in thickness than conventionally milled samples. SEM backscattered electron imaging of an experiment quenched at 87 MPa following decompression from 130 MPa at a rate of 10 MPah-1 suggests conditions of extremely rapid crystal growth (10-6 mm s-1) and comparatively sluggish nucleation (<1 mm-3 s-1). In contrast, preliminary TEM results indicate that nucleation rates of silicate crystals are ~5 orders of magnitude greater than as interpreted from SEM imaging. No silicates <100 nm wide have been observed, suggesting that nucleation occurs in discrete events rather than continuously. Although crystals appear uniformly distributed and spatially unrelated at the SEM scale, crystals are strongly clustered in TEM. Juxtaposition of Fe-Ti oxides with silicates is provisionally interpreted to reflect heterogeneous nucleation followed by epitaxial growth.
V33B-2220
Kinetic Element Fractionation in Orthopyroxene Reaction Rims
The contact of Mg-rich rocks to Si-rich rocks or melts is a major geochemical contrast that leads to the formation and advancement of reaction fronts due to reactions like olivine + quartz = orthopyroxene. Detailed experimental investigation focused on this reaction showed that microtextures in Opx reaction rims give evidence of either dilative or compressive local growth regimes on the two sides of the rims caused by the total volume change of the reaction and the ratio of diffusive mass flow of the SiO2 and MgO components. If the reaction starts from olivine solid solutions in addition distinct microchemical patterns evolve. Thanks to advancements in microanalytical techniques like field emission electron microprobe, focused ion beam preparation, or dual beam techniques (ion beam/electron beam) we are now able to study the combination of microtextural and microchemical patterns in great detail. In rim growth experiments using natural Fo-rich Ol and Qtz as reactants the two most important minor and trace elements are Fe and Ni. According to equilibrium fractionation both should be enriched in Ol vs. Opx. Nevertheless, experimental evidence shows that these two elements do not behave in identical way. Ni is retained in Ol at the reaction front. Fe, in contrast, is expelled, and after transport along grain boundaries of the Opx rim is enriched at the Opx-Qtz interface, where the Fe-concentration in Opx is not buffered. Thus, there are two ways to maintain local equilibrium at the Ol/Opx interface – retain or expel. Which of the two is more efficient in energy minimization depends on the transport properties of the grain boundaries in the reaction rim. Thus, we can speak of kinetic fractionation in reaction rims. This kind of kinetic fractionation is followed as a general principle in diffusion-controlled reaction rim growth. For additional investigation, forsterite doped with various elements (like Cr, Ca, Zn) was synthesized and used in Opx rim growth experiments, where each dopand develops a distinct zonation pattern. Investigation of microtextures and microchemical patterns may have implications on global geological processes. Recently the hypothesis of large scale SiO2 metasomatism (Ol + SiO2 component = Opx) in the upper mantle due to eclogite-derived SiO2-rich melts became prominent and is supported by a growing amount of evidence. Little, however, is known about the actual mechanisms at work in diffusion-reaction metasomatism that are at the base of kinetic major and minor element fractionation.
V33B-2221
Quantitative fabric analysis of experimentally deformed volcanic rocks
The quantitative analysis of patterns as a geometric arrangement of material domains with specific geometric or crystallographic properties such as shape, size or crystallographic orientation has been shown to be a valuable tool with a wide field of applications in geo- and material sciences. We have developed a collection of automated methods that can easily be applied for different problems. For example, in experimental volcanology an adequate description of the ductile-brittle transition of highly crystalline lavas is crucial to a better understanding of explosive eruptions at dome-building volcanoes. During lava emplacement, ductile shearing and brittle fracturing of these highly crystalline melts may produce complex crystal and fracture fabrics. Such fabrics can be reproduced in experiments (Lavallée et al., 2008) and have been quantitatively analyzed with respect to anisotropy of fabric complexity and fabric inhomogeneity (Gerik & Kruhl, 2008). Such analyses are important since they provide valuable information on the material behavior at different stages of deformation. Experimentally deformed volcanic rock samples showed an anisotropy increase in the groundmass pattern whereas the anisotropy decreased in the crystal pattern. This can be explained by fracturing of platy plagioclase crystals, which leads to more equant fragment shapes and, consequently, to a decrease in crystal pattern anisotropy. The increase in groundmass anisotropy results from an alignment of the original crystals perpendicular to the direction of compression. Such a quantitative approach also allows for a comparison of natural rocks and experimentally deformed samples and, therefore, potentially enables a deeper-rooted investigation of natural processes, based on experiments.
V33B-2222
High and Ultra High Resolution Neutron Computed Tomography as a new Method for Analyzing Rock Textures
X-ray computed tomography is a well-established method for fabric analysis in many fields of science. Neutron Computed Tomography has very recently achieved high resolution comparable to X-ray CT, and has already proven to be a very powerful tool in geologic texture determination. While X-rays show attenuation monotonously rising with the atomic number of the elements, the attenuation of thermal and cold neutrons is determined by crystalline structures and by the internal structure of the nuclei themselves. The resulting behaviour is in no way monotonous or regular across the periodic system, often showing high contrast variation even between isotopes of neighbouring or the same elements. With good penetration of metals, and very high sensitivity for hydrogen, neutron imaging is often able to deliver complemetary information to X-ray examinations which can be ideally combined with each other. Beam formation and high resolution detection are considerably more difficult than for X-rays, and can currently only be performed at very few places in the world. This talk will give an overview about the method of neutron imaging, describe its peculiarities and chances, and present tomographic measurements performed at the FRM II reactor of Technische Universitaet Munich. Details will be provided about the current limit in resolution, the envisaged ways to improve the resolution, and first measurements on the way to ultra high resolution will be presented. The presentations by Hess et al. : Analyzing 3d-structures of syntectonic magmatic rocks) and Scheu et al.: High Resolution Neutron Computer Tomography of Vesicular Pyroclasts: Interplay of Vesicles and Crystalline Phases - both in this session - will provide more details about current examinations and will also give an outlook on future measurements and applications in geosciences.
V33B-2223
Determination of Bubble Sizes in Volcanic Melts by Confocal Microscopy and Subsequent 3D Image Analysis
Bubbles in silicic melts are studied for many characteristics including size, shape and volume to understand vesiculation processes behind volcanic eruptions. This work presents a new and innovative method of reconstructing bubbles in volcanic melts using confocal microscopy. A sample of volcanic glass produced in a laboratory experiment was studied using the Laser Scanning Confocal Microscope (LSCM). LSCM have unique optical sectioning capability that can be used to conduct non-invasive studies of the surface and subsurface of a sample. A stack of images representing virtual sections so obtained is studied using image analysis techniques to reconstruct the volumes of the bubbles present inside the melt. The previously known technique employed in obtaining images from the confocal microscope resulted in low contrast images making it particularly difficult or impossible to segment various physical features. For this study the authors suggested the application of a new image filtering technique. The image stack is filtered using a Gaussian blurring technique to provide good contrast around the vesicles edges and then a 3D active contour segmentation technique. In turn the contours were processed using a level-set algorithm to achieve the final segmentation and registration of the vesicular volumes. The filtration is done manually and the segmentation is semi-automatic. There is a possibility of modifying the technique to completely automate the process to make it applicable to a large number of samples.
V33B-2224
Comparison of Trace-Element Concentrations in Rhyolitic Glass From Laser Ablation and Solution Data: Effects of Microlites and Trace Phases.
The spatial resolution of LA-ICPMS allows characterization of fine scale heterogeneities in volcanic glasses and phenocrysts, providing potentially valuable insight into magmatic processes. This study compares thin section LA-ICPMS data to whole rock and microsample glass solution data from vitrophyric portions of high temperature rhyolitic lavas and rheoignimbrites in the Central Snake River Plain volcanic province near the borders of Idaho, Nevada, and Oregon, USA. 1.0 mg and 200 mg aliquots of glass fragments (< 1 mm diameter) were hand picked from fresh rhyolitic vitrophyre to exclude phenocryst fragments. Glasses were digested using a conventional, open-vial, mixed- acid dissolution method in order to minimize uptake of insoluble phases such as zircon, and to isolate the glass signature. LA-ICPMS data was collected using a Thermo-Finnigan Element 2 ICPMS coupled to a New Wave UP213 laser ablation system. Troughs of 12-20 microns wide by 500-800 microns long were ablated to a depth of less than 10 microns on polished thin sections. This resulted in a sample mass of ~10-3 to ~10-5 milligrams. The Element 2 was set to scan through the mass range of the elements of interest 25 times per run, resulting in raw counts per second (CPS) data which was then averaged over the course of the run and internally normalized to Si concentration data from electron microprobe analysis. This procedure allows for both the manual exclusion of scans that clearly hit microlites or inclusions and minimalizes the effects of variations in sample delivery due to fluctuations in ablation rate and carrier gas flow. Any ablation run requiring the exclusion of more than 5 scans, and/or those with high standard deviation in CPS, were discarded. Results show significant concentration discrepancies between the different types of sample for elements which are compatible in observed phenocrysts in the rocks. Ba, Sr, and Zr concentrations in the glass aliquots are generally higher and more variable than the laser data, indicating the presence of zircon and feldspar microcrysts in the glass separates. Microlites are pervasive in CSRP glasses and influence the trace element concentrations determined by LA. Microlites form prior to and during eruption, emplacement, and cooling of volcanic deposits, and presumably were not present prior to eruption; LA analytical protocols must be designed accordingly if the goal is to determine trace element concentrations in 'magmatic' liquid. Potential implications for a variety of geochemical modeling include estimation of crystal-liquid partition coefficients from mineral-glass pairs, and minor or trace element- based geothermometry.
V33B-2225
Trace Elements in Molybdenite as Indicators of Tectono-Metallogenic Settings
Molybdenite serves as a robust Re-Os geochronometer for directly dating ore formation in a wide range of ore systems. Previous work has shown a strong relationship between Re concentrations, tectonic setting, and ore-forming processes (Stein et al. 2001; Stein, 2006; Zimmerman et al. 2008). Preliminary rare earth element analyses of molybdenites indicate a correlation between REE patterns and tectonic setting. As molybdenite is widespread and occurs in a variety of ore deposit types its trace element chemistry may prove a promising exploration tool. An inexpensive sampling method was used to determine if there is a direct link between molybdenite minor/trace element chemistry and commodity enrichment, ore-forming process, and tectonic setting. Through the analysis of over 110 molybdenite (and one rheniite) samples from different ore systems around the world, trends in minor and trace element enrichments were found. Polytype was determined for most samples using XRD, and for each sample a suite of 64 elements was determined using solution ICP-MS. In addition to geographically localized enrichments of specific trace elements, the REE profiles of the molybdenites have a signature which appears to correspond to the associated magmatic setting. Furthermore, molybdenites are generally enriched in trace elements specific to the commodity being mined at any given deposit. We are working toward using these ICP-MS trace element data to predict commodity associations so that trace element geochemistry in molybdenite might prove an inexpensive tool for mineral exploration. Stein, H.J., Markey, R.J., Morgan, J.W., Hannah, J.L., and Scherstén, A. (2001) The remarkable Re-Os chronometer in molybdenite: how and why it works: Terra Nova, v. 13, no. 6, p. 479-486. Stein, H.J. (2006) Low-rhenium molybdenite by metamorphism in northern Sweden: recognition, genesis, and global implications: Lithos, v. 87, p. 300-327. Zimmerman, A., Stein, H.J., Hannah, J.L., Kozelj, D., Bogdanov, K., and Berza, T. (2008) Tectonic configuration of the Apusini-Banat-Timok-Srednogorie belt, Balkans-South Carpathians, constrained by high precision Re-Os molybdenite ages: Mineralium Deposita, v. 43, p. 1-21.