Volcanology, Geochemistry, Petrology [V]

 CC:711  Monday  1400h

Innovative Applications of Stable Isotopes to Hydrothermal and High-Temperature Processes I

Presiding:  E Sharman, McGill University; J B Chapman, Geological Survey of Canada


Non-traditional Stable Isotope Systematics of Seafloor Hydrothermal Systems

* Rouxel, O J (orouxel@whoi.edu), Woods Hole Oceanographic Institution, Marine Chemistry & Geochemistry Dept MS#25, Woods Hole, MA 02543, United States

Seafloor hydrothermal activity at mid-ocean ridges is one of the fundamental processes controlling the chemistry of the oceans and the altered oceanic crust. Past studies have demonstrated the complexity and diversity of seafloor hydrothermal systems and have highlighted the importance of subsurface environments in controlling the composition of hydrothermal fluids and mineralization types. Traditionally, the behavior of metals in seafloor hydrothermal systems have been investigated by integrating results from laboratory studies, theoretical models, mineralogy and fluid and mineral chemistry. Isotope ratios of various metals and metalloids, such as Fe, Cu, Zn, Se, Cd and Sb have recently provided new approaches for the study of seafloor hydrothermal systems. Despite these initial investigations, the cause of the isotopic variability of these elements remains poorly constrained. We have little understanding of the isotope variations between vent types (black or white smokers) as well as the influence of source rock composition (basalt, felsic or ultrabasic rocks) and alteration types. Here, I will review and present new results of metal isotope systematics of seafloor hydrothermal systems, in particular: (1) determination of empirical isotope fractionation factors for Zn, Fe and Cu-isotopes through isotopic analysis of mono-mineralic sulfide grains lining the internal chimney wall in contact with hydrothermal fluid; (2) comparison of Fe- and Cu-isotope signatures of vent fluids from mid- oceanic and back-arc hydrothermal fields, spanning wide ranges of pH, temperature, metal concentrations and contributions of magmatic fluids enriched in SO2. Ultimately, the use of complementary non-traditional stable isotope systems may help identify and constrain the complex interactions between fluids,minerals, and organisms in seafloor hydrothermal systems.


Subduction-related B and H Isotope Fractionations Across the Mariana arc - Consequences for Recycling

* Shaw, A M (ashaw@whoi.edu), DTM, Carnegie Inst. of Washington, Washington, DC 20015, United States
* Shaw, A M (ashaw@whoi.edu), WHOI, Woods Hole, MA 02536, United States
Hauri, E H (hauri@dtm.ciw.edu), DTM, Carnegie Inst. of Washington, Washington, DC 20015, United States
Stern, R (rjstern@utdallas.edu), University of Texas at Dallas, Richardson, TX 75083, United States
Hawkins, J (jhawkins@ucsd.edu), SIO, UCSD, La Jolla, CA 92039, United States
Gurenko, A (gurenko@whoi.edu), WHOI, Woods Hole, MA 02536, United States

We present new B and H isotope data, along with volatile, trace and major elements for olivine-hosted melt inclusions from a suite of cross chain volcanoes extending across the Mariana arc from Guguan volcano to the Mariana Trough. H and B isotopes, H2O, CO2, S, F and Cl abundances, as well as trace elements, have been determined by SIMS. Our results show that enrichments in fluid-mobile elements generally associated with the subducting slab (e.g., Ba, B) decrease systematically across the arc into the back-arc. However, water contents in cross-chain samples, 230 km above the subducting slab, show similar values to the arc-front samples, implying that water release is a continuous process across the arc and that trace element proxies for slab fluids are decoupled from actual water contents. The isotopic composition of water changes during progressive dehydration, as expected1; δD values are highest at the arc front, ∼ -10‰, and decrease to values as low as -80‰ in the back-arc. Likewise, B isotopes decrease systematically from δ11B values as high as 5.9-8.5‰ at the arc front, consistent with values previously reported for Guguan arc front lavas (δ11B = 5.0-6.2‰2), down to values as low as -14‰ in back-arc melt inclusions, extending beyond the range inferred for MORB (-5 to -9‰3) and to values observed in OIBs (down to -15‰4). Our observations demonstrate that fluid release behind the main volcanic front can be substantial and that the dehydration process has fractionated H and B isotope compositions. Our findings suggest that B and H isotopes of OIBs containing recycled slab components would be low. 1 Shaw et al. (2008) EPSL v.275, 138-145 2 Ishikawa and Tera (1999) Geology v.27, 83-86 3 leRoux et al. (2003) Abstr. Fall AGU #V51A-03 4 Chaussidon and Marty (1995) Science v.269, 383-386


Lithium Isotopes as a Tracer of Fluids in a Subduction Zone: Franciscan Complex, CA

* Penniston-Dorland, S C (sarahpd@geol.umd.edu), Department of Geology, University of Maryland, College Park, MD 20742, United States
Sorensen, S S (sorensen@si.edu), Department of Mineral Sciences, National Museum of Natural History Smithsonian Institution, Washington, DC 20013, United States
Ash, R D (rdash@geol.umd.edu), Department of Geology, University of Maryland, College Park, MD 20742, United States
Khadke, S V (skhadke@umd.edu), Department of Geology, University of Maryland, College Park, MD 20742, United States

Lithium has received a lot of attention as a potential tracer of interactions between fluids and rocks in the high grade metamorphic rocks from subduction zones. This study focuses on the use of Li and its stable isotopes as tracers of reactive fluid flow in subduction zone rocks, focusing on two features found in high grade blocks in mélange which are indicative of fluid-rock reaction: metasomatic actinolite schist rinds on eclogite and garnet amphibolite blocks and blueschist layers within eclogite blocks. The presence of hydrous minerals in the block interiors suggests that they reacted with fluids at or near the peak of metamorphism. Mineral assemblages and textures indicate that the rinds and blueschist layers experienced further hydration reactions during retrogression. Block interiors have δ7Li ranging from -1 to 2‰ while rinds and blueschist layers have δ7Li ranging from 1 to 4‰. Block interiors have Li concentrations ranging from 15 to 40 ppm while rinds and blueschist layers have lower Li concentrations ranging from 7 to 20 ppm. In situ LA-ICP-MS analysis of Li concentrations of mineral phases demonstrates that the Li concentrations of some but not all phases are lower in the blueschist layers and rinds than in block interiors and eclogite layers. These data suggest interaction of block interiors with a Li-rich fluid with relatively low δ7Li, likely of sedimentary origin, during the peak of metamorphism. The rinds and blueschist layers record δ7Li that is higher by 1 to 3‰ than associated block interiors and eclogite layers. These data suggest that during retrogression the block rinds and blueschist layers interacted with fluids from a source with relatively low Li concentrations and relatively high δ7Li, such as altered MORB. A cm-scale traverse across a blueschist-eclogite contact in one of the samples defines an isotopic profile that is consistent with fluid flow along the blueschist layer accompanied by diffusion of Li across the contact.


Sulfur Isotope Fractionation During Magmatic Degassing

* Graham, K A (kathleen.graham@mail.mcgill.ca), Earth and Planetary Sciences, McGill University, 3450 University Street, Montreal, QC H3A2A7, Canada
Wing, B (wing@eps.mcgill.ca), Earth and Planetary Sciences, McGill University, 3450 University Street, Montreal, QC H3A2A7, Canada
Baker, D R (donb@eps.mcgill.ca), Earth and Planetary Sciences, McGill University, 3450 University Street, Montreal, QC H3A2A7, Canada

The study of volatiles is integral to a better understanding of volcanism. Sulfur is one of the volatile constituents in volcanic eruptions, yet the full picture of sulfur behaviour prior to, and during, eruptions remains unknown. Because magma chambers are inaccessible to direct observation, the S isotopic consequences of phenomena such as degassing and diffusion may be used to indirectly constrain processes occurring in magmatic systems. In order to achieve this goal, however, laboratory calibration of S isotope fractionation during magmatic processes is needed. We experimentally studied the effects of rapid sulfur degassing from a melt of the same composition as the 122 BC plinian eruption of Mt. Etna, one of the few recorded basaltic plinian eruptions. All experiments in this study were preformed with this basaltic glass as the starting material, to which was added powdered gypsum (CaSO4 · 2H2O) to create a starting material with approximately 2000 ppm dissolved S. Experiments were performed at an average oxygen fugacity of NNO +1.2. Samples were synthesized by hydrating aliquots of basaltic glass + gypsum + 4-11 wt % H2O at 550 MPa and 1225oC for two hours. These samples were either isobarically quenched to room temperature to provide starting material for degassing experiments at 1 bar, or degassed by lowering the pressure at isothermal conditions. Degassing experiments at 1 bar involved heating the quenched glass to 1200oC, allowing the formation of bubbles and gas loss from the melt. We measured the S contents of the resulting twenty-three experimental run products by electron microprobe analysis in order to quantify sulfur loss during degassing (fraction lost = 1 -S(ppm)final/S(ppm)initial). Estimates of S loss were used in a simple model of open-system Rayleigh isotopic fractionation under equilibrium conditions to predict the S isotopic composition of each degassed experimental glass. In this presentation we will compare these predictions to the measured S isotopic compositions of the experimental run products, and discuss our results in terms of fundamental processes of degassing (e.g., nucleation, diffusion, and growth).


The Fate of Sulfur in Late-Stage Magmatic Processes: Insights From Quadruple Sulfur Isotopes

* Keller, N S (nkeller@whoi.edu), Geology and Geophysics Department, Woods Hole Oceanographic Institution, 360 Woods Hole Rd, Woods Hole, MA 02543, United States
* Keller, N S (nkeller@whoi.edu), Department of Earth, Atmospheric and Planetary Sciences, MIT, 77 Massachusetts Ave, Cambridge, MA 02139, United States
Ono, S (sono@mit.edu), Department of Earth, Atmospheric and Planetary Sciences, MIT, 77 Massachusetts Ave, Cambridge, MA 02139, United States
Shaw, A M (ashaw@whoi.edu), Geology and Geophysics Department, Woods Hole Oceanographic Institution, 360 Woods Hole Rd, Woods Hole, MA 02543, United States

Multiple sulfur isotopes (32S, 33S, 34S and 36S) have recently been shown to be useful tracers of fluid-rock interaction in seafloor hydrothermal systems [1]. Here we present the application of multiple sulfur isotopes to subaerial volcanoes with the aim of unraveling the various processes fractionating sulfur in the upper volcanic system. We take advantage of the fact that the ascent of volcanic gases through a hydrothermal system causes complex isotopic fractionation between the quaduple sulfur isotopes. δ34S is thought to trace the source of sulfur as well as magma degassing; at equilibrium, δ33S follows a mass-dependent fractionation relationship such that two phases in equilibrium with each other have equal Δ33S values (Δ33S ≡ ln(δ33S+1) - 0.515×ln(δ34S+1)). Disequilibrium Δ33S values can indicate isotope mixing and other fluid-rock interactions. The ultimate aim of this study is to assess the use of quadruple sulfur isotopes to obtain quantitative information on the sulfur cycle at convergent plate margins. The sulfur mass balance at convergent margins is poorly constrained, partly because late-stage processes are challenging to quantify and lead to large uncertainties in the global output fluxes. Quadruple sulfur isotopes provide a powerful tool to untangle the convoluted history of volcanic systems. Here we report the first quadruple sulfur isotopic values for H2S, SO2 and native sulfur from arc volcanoes. Fumarolic gases (∼100°C) and sulfur sublimates were collected from Poas and Turrialba, two actively degassing volcanoes in Costa Rica. The gases were bubbled in situ through chemical traps to separate H2S from SO2: H2S was reacted to form ZnS, and SO2 to form BaSO4. Sulfur was chemically extracted from the solid phases and precipitated as Ag2S, which was fluorinated to SF6 and analysed by IRMS. Poas and Turrialba have H2S/SO2 ∼1 and 0.01, respectively. δ34SH2S and δ34SSO2 are similar to gases measured at other arcs [2], - 7.9‰ and 0.6‰ for Poas, and -8.5‰ and +9‰ for Turrialba, likely reflecting a mixture of mantle (δ34S = 0‰) and slab sources which have been degassed to variable degrees. Sulfur sublimate values are similar to those for H2S. Δ33S values are different within each H2S/SO2 pair (Δ33SH2S and Δ33SSO2 are -0.01‰ and -0.02‰ for Poas, 0‰ and -0.07‰ for Turrialba), indicating that at Turrialba, the two gas species are not in isotopic equilibrium. Reaction of the gases with mineral phases, such as sulfur-bearing alteration products in the volcanic edifice (e.g., alunite, anhydrite, sulfides) may explain these differences. [1] Ono et al. (2007), GCA 70 1170-1182, [2] Taylor (1986), RiM 16 185-225


Laser Combustion Analyses Of Pyrite Grains From Arghash Gold Prospect, Iran

* Alirezaei, S (s-alirezaei@sbu.ac.ir), Saeed Alirezaei, Faculty of Earth Sciences, University of Shahid Beheshti, Tehran, Iran, Tehran, 15875-4731, Iran (Islamic Republic of)
Ashrafpour, E (e-ashrafpour@sbu.ac.ir), Esmaeel Ashrafpour, Parskan East Company, Tehran, 15875-4731, Iran (Islamic Republic of)
Ansdell, K M (k.ansdell@usask.ac.ca), Kevin M Ansdell, Department of Geological Sciences, University of Saskatchewan, Saskatoon, S7N 5E2, Canada

The Arghash gold prospect is located in the Sabzevar zone in northeastern Iran. The basement of the Sabzevar zone consists of Precambrian metamorphosed rocks covered by Paleozoic epicontinental sediments and Upper Cretaceous ophiolitic mélange. Tertiary magmatic rocks, with chemical compositions characteristic of continental arc magmas, overlie or crosscut the older rocks. The Arghash gold prospect includes five gold- bearing vein systems, hosted in Eocene-Oligocene, intermediate-felsic volcanic and intrusive rocks of volcanic rocks. Pyrite is the main sulfide mineral in the hypogene ore. Four generations of pyrite (Py-I-Py-IV) were identified through detailed microscopic observations and electron microprobe analyses. Py-I is euhedral to anhedral fine- to coarse-grained pyrite, which is disseminated in the vein quartz and locally is associated with minor chalcopyrite, marcasite, arsenopyrite and, native gold grains (<30 μm). This pyrite generation is characterized by low contents of As, Sb, and Au. Py-II is a framboidal pyrite which fills microfractures in quartz and calcite and is characterized by concentric arsenic-rich and arsenic-poor bands, the former also rich in gold (up to 963 ppm). Arsenian pyrite overgrowths (Py-III) form <10 μm thick rims on Py-I and contain 1.2-7.9 wt. percent As, up to 6000 ppm Ni, and up to 1980 ppm Au. Py-IV is a coarse-grained, anhedral, barren pyrite, forming veinlets <5mm thick in quartz. The trace elements in this pyrite were below detection limits. Sulfur isotope analyses were first carried out on bulk pyrite concentrates from auriferous vein materials and adjacent altered wall rocks. The δ34S values for ten pyrite concentrates vary from -4.3 to +21.8 per mil. The wide variations of δ34S values could be related to different generations of pyrite and the occurrence of more than one generation in the samples, and possibly different origins for sulfur in the hydrothermal system. To characterize the isotopic compositions of various generations of pyrites, in-situ laser combustion analyses were performed on two polished slabs at the Scottish Universities Environmental Research Center (SUERC) using a SPECTRON LASERS 902Q CW Nd-YAG laser (1W power). The δ34S values for Py-I to Py-III vary between -5.8 and +0.1 per mil; these are comparable to depleted to less enriched δ34S values in the bulk analyses. A magmatic source, or derivation of sulfur from older igneous rocks, is proposed for the origin of sulfur in the Py-I to Py-III generations. Py-IV is highly enriched in 34S (δ34S= +8.9 to +23.7). This accounts for the high enriched δ34S values in the bulk analyses. Such high values may reflect contributions of isotopically heavy sulfur from a source enriched in 34S, such as evaporites. Gypsum, anhydrite, gypsiferous marl, and halite-bearing beds occur locally in the Eocene and older sedimentary units.


Germanium Isotopic Fractionation in Iron Meteorites : Comparison with Experimental Data

* Luais, B (luais@crpg.cnrs-nancy.fr), Centre de Recherches Petrographiques et Geochimiques, 15, rue Notre-Dame des Pauvres, BP 20, Vandoeuvre-les-Nancy, 54501, France
Toplis, M (toplis@dtp.obs-mip.fr), Laboratoire Dynamique Terrestre et Planétaire, 14 avenue Edouard Belin, Toulouse, 31400, France
Tissandier, L (tix@crpg.cnrs-nancy.fr), Centre de Recherches Petrographiques et Geochimiques, 15, rue Notre-Dame des Pauvres, BP 20, Vandoeuvre-les-Nancy, 54501, France
Roskosz, M (mathieu.roskosz@univ-lille1.fr), Laboratoire de Structure et Propriétés de l'Etat Solide, Université de Lille1, Batiment C6, Villeneuve d'Ascq, 59655, France

Magmatic and non-magmatic iron meteorites are thought to be formed respectively by processes of metal- silicate segregation, and complex impacts on undifferentiated parent bodies. These processes are inferred from variations of siderophile element concentrations, such as Ge, Ni, Ir. Germanium is moderately siderophile, with metal-silicate partition coefficients which depend on oxygen fugacity. Germanium is also moderately volatile, and fractionation would be expected during high temperature processes. In order to investigate the extent of elemental and isotopic fractionation of germanium during metal-silicate equilibria and impact processes, we use a double approach including (1) Ge isotopic measurements of iron meteorites from non-magmatic and magmatic groups [1], and (2) experimental investigations of the isotopic fractionation associated with germanium transfer from an oxidized silicate liquid to a metallic phase under various fO2 conditions. Experiments were performed in a 1 atm vertical drop quench furnace, with starting materials corresponding to a glass of 1 bar An-Di euctectic composition doped with ∼ 4,000 ppm reference Ge standard, and pure Ni capsules as the metal phase. The assembly was heated at 1355°C for t =2 to 60 hrs over a range of fO2 from 4 log units below, to 2.5 log units above, the IW buffer. Metal and silicate phases were then mechanically separated. For isotopic measurements, the metal phase of these experiments and the selected iron meteorites were dissolved in high-purity dilute nitric acid. Chemical purification of Ge, and isotopic measurements using the Isoprobe MC-ICPMS follow Luais (2007). Germanium isotopic measurements of Fe-meteorites show that δ74Ge of magmatic irons are constant (δ74Ge=+1.77±0.22‰, 2σ), but heavier than non-magmatic irons (IAB : +1.15±0.2‰; IIE : -0.27 to +1.40±0.2‰). Time series experiments at the IW buffer show that there is a clear continuous increase in δ 74Ge in the metal as a function of time t, values being isotopically lighter than the reference Ge standard after 2 hours (δ74Ge ∼ - 0.5 ‰), reaching a δ 74Ge of ∼ +3‰ after 60 hours. For a series of experiments running for 24 hours, an increase in δ74Ge (from ∼ -0.3 to +10‰) is observed with increasing fO2. The slight enrichment of light isotopes in the metallic phase at very low fO2 and short t could reflect the consequence of Ge diffusion from the silicate to the metal. This small extent of fractionation can be compared with the nearly constant δ74Ge values in magmatic Fe-meteorites, indicating that diffusion-induced metal-silicate segregation would produce very small if detectable isotopic fractionation. Therefore, Ge isotopic composition of Fe-meteorites would be close to that of their parent bodies. On the other hand, strong enrichment in heavy isotopes at high fO2 and/or long t is indicative of Ge loss by evaporation, and would explain the large isotopic fractionation toward heavy δ 74Ge values for the IIE non-magmatic irons, interpreted as impact-induced evaporation of Ge. [1] Luais B (2007) EPSL 262 p. 21