Goldschmidt Abstracts 2008- O

Goldschmidt Conference Abstracts 2008

A693

Origin and evolution of mantle heterogeneities

Productivity and sediment supply from the E-O boundary in Tanzania

R.J. O’CONNELL1*, S.B. JACOBSEN2 AND J.B. KELLOGG3

A. O’HALLORAN, C.J. NICHOLAS AND R. GOODHUE

1

Dept EPS, Harvard Univ, Cambridge MA 02138, USA (*correspondence: [email protected]) 2 Dept EPS, Harvard Univ, Cambridge MA 02138, USA ([email protected]) 3 J.B. Kellogg, Cate School, Carpinteria CA 93014, USA ([email protected]) Mantle heterogeneities arise from differentiation at midocean ridges, subduction zones, and melting associated with hotspots. The residues from of these processes, as well as the recycled products, are circulated in the mantle where they evolve isotopically, mechanically and spatially. We model the evolution of heterogeneities in a statistical model in which the isotopic composition, length scale and depth of heterogeneities evolve. The change in length scale of heterogeneities depends on their effective viscosity (dehydrated melting residues are strong, recycled crust and unerupted plume material is weak) as well as their location in the mantle; deformation rates are high in the upper mantle, and lower in the higher viscosity deeper mantle. Melting residues and recycled material mixes to varying depths depending on the details of mantle flow, and heterogeneities may follow a range of paths. Melting at ridges samples the upper mantle on a time scale of ~300 My; processing the volume of the mantle takes ~10 times longer. This leads to a preferential processing of the upper mantle. Exchange with the deeper mantle is from deep subduction, and from plume conduits that rise from depth, as well as convective stirring. The rates of deformation in the mantle and of vertical transport of heterogeneities are estimated from geographically realistic mantle circulation models with plates. A statistical assemblage of mantle heterogeneities is constructed based on these processes. The assemblage is sampled to produce MOR basalts and continental crust, which reflect averages of the component heterogeneities. The model successfully reproduces observations of Pb, Nd and Sr isotopes [1]. A feature of the model is a ubiquitous component similar in composition to FOZO or C that is made up of heterogeneities with small length scales (<15 km). This mixes with residues of melting (which are larger) to produce observed data arrays. The model allows the effects of heterogeneity properties, deformation rates, sampling rates and material transport rates to be assessed, and it captures the essentially statistical nature of mantle evolution and sampling. [1] Kellogg et al. (2007) EPSL 262, 328-342.

Department of Geology, Trinity College, University of Dublin, Dublin 2, Ireland The Eocene-Oligocene transition was one of intense global climate change, including global cooling, ice-growth, attendant sea-level fall, and accelerated extinction occuring across the boundary. We use nitrogen isotopes and mineralogical data gathered from Tanzanian passive margin, organic-rich, hemipelagic marine clays, cored by the Tanzania Drilling Project (Sites 12 and 17). Recent data [1] has recorded extinctions in the foraminifera and nannoplankton, as well as the two-step onset of the delta18O maximum, recorded in deep-sea cores [2, 3] due to early Oligocene glaciation. We show that terrigenous input to the coastal zone increased in the early Oligocene (Fig. 1). This increase may be due to eustatic sea-level fall and/or to enhanced terrestrial run-off.

Figure 1: Quartz increases, indicating increased terrigenous input. Our initial data suggest an increase in productivity during the early Oligocene. Further nitrogen isotopic data may ascertain whether this signals a boom in ocean ecology or if, in fact, it marks a recovery from the late-Eocene extinctions. [1] Pearson et al. (2008) Geology 36, 179-182. [2] Coxall et al. (2005) Nature 433, 53-57. [3] Zachos et al. (1996) Paleoceanography 11, 251-266.

A694

Goldschmidt Conference Abstracts 2008

The influence of Aluminum on the partitioning of water between clinopyroxene and basaltic melt

Interactions of U(VI) with secondary mineralization products from the bioreduction of Fe(III) oxides

J.A. O’LEARY1, G.A. GAETANI2 AND E.H. HAURI1

E.J. O’LOUGHLIN1*, S.D. KELLY1, M.I. BOYANOV2 1 AND K.M. KEMNER

1

Department of Terrestrial Magnetism, Carnegie Institution of Washington, Washington DC, 20015 USA ([email protected], [email protected]) 2 Department of Geology & Geophysics, Woods Hole Oceanographic Institution, Woods Hole MA 02543 USA ([email protected]) The distribution of water between basaltic melt and nominally anhydrous minerals has implications for the transport and storage of water during peridotite melting. Although there is evidence that mineral-melt partitioning is effected by the aluminum content of the crystals [1, 2], experimental data are only available for a limited range of bulk compositions. To isolate the effect of clinopyroxene aluminum content on partitioning behavior we conducted high temperature hydrous melting experiments in aluminum-free and aluminum-bearing systems. Our results indicate that the capacity of clinopyroxene to incorporate hydroxyl through a substitution involving Mg vacancies is limited to ~300 ppm. For the same concentration of water dissolved in the melt, the water content of aluminum-bearing clinopyroxene is ~500 ppm. This indicates that approximately 40% of the hydroxyl is charge balanced by aluminum. Experiments were carried out in a piston-cylinder device at 1.5 GPa, 1275 °C, with oxygen fugacity controlled at Ni-NiO using a solid buffer. Starting compositions consisted of either a natural basalt or a synthetic, aluminum-free basalt. Experiments were carried out with up to ~6 wt% H2O dissolved in the melt. The volatile content of each phase present in the experiments was determined in situ by SIMS. Our experimental results suggest that the mechanism for incorporation of OH into pyroxene is strongly effected by the presence of aluminum. Al-bearing experiments have partition coefficients Dmelt/cpx = .019-.036, in agreement with previous reports [1, 2]. Partition coefficients for Al-free composition experiments are much lower, Dmelt/cpx = .006-.012. The water content of pyroxene in all Al-free composition experiments is the same within error, 315 ± 12 ppm H2O, and is similar to water saturated Al-free pyroxene [3]. [1] Aubaud et al. (2004) GRL, V. 31, L20611, doi, 10.1029/2004GL021341. [2] Hauri t al. (2006) EPSL 248, 715-734. [3] Rauch & Kepplar (2002) CMP 143, 525-536.

1

Biosciences Division, Argonne National Laboratory, Argonne, IL, 60439, USA (*correspondence: [email protected]) ([email protected], [email protected]) 2 Faculty of Physics, University of Sofia, Sofia, 1000, Bulgaria ([email protected])

Introduction Biogenic Fe(II) phases (magnetite, green rust, siderite, vivianite, etc.) provide a reservoir of reducing capacity in the subsurface that may contribute to the reduction of contaminants such as U(VI). In this study we examined the potential for reduction of U(VI) in the presence of biogenic green rust, (BioGR) biogenic magnetite (BioMAG), and biogenic siderite (BioSID) resulting from the bioreduction of Fe(III) oxides (lepidocrocite and ferrihydrite) by Shewanella putrefaciens CN32.

Experimental Methodology Suspensions of biogenic Fe(II) phases were pasteurized (70°C for 1 h) to eliminate the potential for microbial reduction of U(VI) and washed repeatedly to remove any soluble reductants. The resulting suspensions had initial total Fe(II) concentrations of 60 mM, 42 mM and 66 mM for BioGR, BioMAG, and BioSID, respectively. The suspensions were then spiked with a stock solution of uranyl chloride to achieve an initial U(VI) concentration of 500 µM.

Discussion of Results Within 48 h, the total solution-phase U(VI) concentrations decreased from 500 M to 1.5 µM in the U-bioGR system, to 392 µM in the U-bioMAG system, and to 472 µM in the UbioSID system, as determined by ICP-OES. Analysis of the samples by U LIII X-ray absorption fine structure spectroscopy (XAFS) indicated that despite a stoichiometric excess of Fe(II), no more than 6% of U(VI) was reduced in the U– BioSID system, and no more than 22% of U(VI) was reduced in the U–BioMAG system. For comparison, in the U–BioGR system, no less than 80% of U(VI) was reduced to U(IV). Thus, differences in the reactivity of the different biogenic Fe(II) phases with respect to the reduction of U(VI) were substantial, with BioGR >> BioMAG > BioSID.

Goldschmidt Conference Abstracts 2008

A695

Eoarchean mafic crust in the Nuvvuagittuq greenstone belt

Insights on Hadean Geodynamics from diamond stability constraints

JONATHAN O’NEIL1*, DON FRANCIS1, RICHARD W. CARLSON2 AND ROSS K. STEVENSON3

CRAIG O’NEILL

1

Earth & Planetary Sciences Department, McGill University, Montreal, Quebec, Canada, H3A 2A7 (*correspondence: [email protected]) 2 Department of Terrestrial Magnetism, Carnegie Institution of Washington, 5241 Broad Branch Road, N.W., Washington, DC 20015 USA 3 GEOTOP-UQAM-McGill, PO Box 8888, St. Centre-ville, Montreal, Quebec, Canada, H3C 3P8 Our understanding of the early differentiation of the Earth’s crust and mantle is limited by the dearth of preserved rocks with ages older than 3.8 Ga. Recent geochronology work on a felsic unit within gabbroic rocks of the Nuvvuagittuq greenstone belt gives a minimum age of ~3.8 Ga for the supracrustal assemblage. The Nuvvuagittuq greenstone belt is a volcano-sedimentary sequence composed mainly of a peculiar cummingtonite-amphibolite (“fauxamphibolite”) intruded by numerous ultramafic and gabbroic sills. The faux-amphibolite consists of mafic gneisses (5 to 15 wt% MgO) composed of cummingtonite + plagioclase + biotite + quartz ± garnet, with a compositional layering defined by variations in the proportions of biotite and cummingtonite. The faux-amphibolites have a similar major element composition to the gabbro sills that intrude them, but are relatively depleted in Ca and exhibit LREE-enriched profiles in contrast to the flat to slightly LREE-depleted profiles of the gabbro and ultramafic sills. The dominance of cummingtonite in the faux-amphibolites appears to reflect their anomalously low Ca content, whose origin remains unclear. All samples of the faux-amphibolite yield negative εNd(3.8 Ga) values ranging from -5.3 to -0.3, whereas samples of the gabbro and ultramafic sills have mostly positive εNd(3.8 Ga) values ranging from -2 to +4, indicating derivation from a mantle already depleted at 3.8 Ga. Sm-Nd isochrons for the sills are consistent with an age of 3.8 Ga and, like most terrestrial rocks, their 142/144Nd ratios are indistinguishable from the terrestrial standard. Preliminary analyses of the fauxamphibolite and the dated felsic unit, however, suggest that they may have slight 142Nd deficits relative to the terrestrial standard. The fact the faux-amphibolites are intruded by the gabbro sills suggests that they may represent an older gabbroic crust, possibly derived from an enriched reservoir complementary to the depleted reservoirs created during the early differentiation of the Earth.

GEMOC ARC National Key Centre, Department of Earth and Planetary Science, Macquarie University, Sydney, NSW, Australia ([email protected])

Hadean Diamonds The recent discovery of diamond inclusions in zircons of Hadean age (4.4-4.0Ga) poses a paradox. The zircons are preserved in younger (Archaean) assemblages in the Jack Hills, Western Australia. Their oxygen isotopes suggest they formed in shallow, probably hydrated, felsic magma chambers, at temperatures of ~680oC [1, 2]. These shallow, hot conditions are not consistent with the stability of diamond. Recent examples of diamond inclusions in zircon come from ultra-high pressure massifs in Kazakhstan, where diamond replaced the original graphite [3]. A number of scenarios for exist for explaining their coexistence under Hadean conditions. We examine a number of tectonic scenarios, the most plausible involve some form of subduction, sagduction (vertical tectonism), or a whole-scale mantle overturn event.

Figure 1: Finite element model of subduction under early conditions, showing the stability of diamonds in the downthrust lithosphere (circle) despite high surface heat flux.

Geodynamic Simulations We use a Lagrangian intergration point finite element method to recreate candidate scenarios for graphite->diamond conversion for thermal conditions appropriate for the Hadean. The PT path of near-surface zircons into the diamond stability field, and their exhumation to the surface, provides an important constraint on Hadean lithospheric dynamics. These scenarios have contemporary analogues in exhumed continental subduction zones such as western Norway. We discuss the key features and similarities of each geodynamic model, and summarize physically plausible models for the global Hadean tectonic regime. [1] Menneken M. et al. (2007) Nature 448, 917-920. [2] Mojzsis et al. (2001) Nature 409, 178-181. [3] De Corte (2000) Island Arc 9, 428-438.

A696

Goldschmidt Conference Abstracts 2008

Collisional erosion and the nonchondritic composition of the Earth

Aragonite as a precursor phase of cyanobacterial calcite precipitation and the influence of EPS on the nucleation process: A STXM study

HUGH O’NEILL1 AND HERBERT PALME2 1

Research School of Earth Sciences, Australian National University, Canberra 0200, Australia (*correspondence: [email protected]) 2 Institut für Geologie und Mineralogie, Universität zu Köln, Zülpicherstrasse 49b, 50674 Köln, Germany The “chondritic assumption” that underlies estimates of the bulk composition of the Earth rests on the view that terrestrial planets are derived from the solar composition having undergone only those cosmochemical fractionation processes that are observed among the “chondritic” or undifferentiated meteorites. These fractionations include various depletions in volatile elements, variable Mg-Sirefractory ratios and a cosmochemical metal/silicate fractionation. Planetary embryos, accreted in local feeding zones within 106 years, may preserve some of these chemical variations. However, modern models of planetary formation postulate that terrestrial planets then grow from ~102 embryos sourced across wide heliocentric distances, involving energetic collisions, in which material is lost as well as gained. This stage should average out some of the “chondritic” fractionations, but potentially introduces two non-chondritic chemical fractionation processes: post-nebular volatilisation and preferential collisional erosion. That post-nebular volatilisation was widespread is demonstrated by the nonchondritic Mn/Na ratio in all the small, differentiated, rocky bodies for which we have samples, including the Moon and Mars. The Bulk Silicate Earth (BSE) has chondritic Mn/Na, but shows several other compositional features in its pattern of depletion of volatile elements suggestive of post-nebular volatilization. The extent of collisional erosion during the formation of the Earth may be estimated from the whole Earth Fe/Mg ratio of 2.2±0.1, which is significantly greater than the solar ratio of 1.9±0.1, implying net loss of ~10% silicate relative to metal during the Earth’s accretion. If this collisional erosion preferentially removed differentiated crust, the assumption of chondritic refractory element ratios in the BSE would not be valid, with the BSE depleted in elements according to their geochemical incompatibility. In the extreme case, the BSE might only have half the chondritic abundances of the highly incompatible heat-producing elements Th, U and K. Such an Earth model resolves several geochemical paradoxes: the depleted mantle occupies the whole mantle, is completely outgassed in 40Ar, and produces the observed 4He flux through the ocean basins.

MARTIN OBST1,2*, ADAM P. HITCHCOCK2, JAMES J. DYNES2,3, JOHN R. LAWRENCE3, GEORGE D. W. SWERHONE3 AND KARIM BENZERARA4 1

Canadian Light Source, 101 Perimeter Road, Saskatoon SK, S7H 0X4, Canada (*correspondence: [email protected]) 2 BIMR, McMaster University, Hamilton, Canada 3 Environment Canada, Saskatoon SK, Canada 4 IMPMC, CNRS, Paris, France The global carbon cycle is influenced by many environmental processes such as the precipitation of carbonbearing minerals like carbonates. Although the importance of unicellular planktonic cyanobacteria in the process of lacustrine and marine CaCO3 precipitation has been investigated for decades, the mechanism(s) of mineral nucleation still have not been fully clarified. Synchrotronbased Scanning Transmission X-ray Microscopy (STXM) so far is the only technique which allows for the investigation of this important biogeochemical question with good chemical sensitivity at a spatial resolution of <30 nm. Cyanobacteria were cultured under varying nutrient concentrations and treated with different amounts (3-6 mM) of Ca2+. Whole cells and focused ion beam milled ultrathin sections were spectromicroscopically analyzed by STXM.

Discussion of Results

Ca2+-adsorption within different types of extracellular polymers (EPS) was quantified using the NEXAFS fingerprints of the C-1s and the Ca-2p absorption edges. Furthermore, aragonite was identified to be an unstable precursor phase of the thermodynamically more stable calcite which forms the final precipitate. The potential for aragonite nucleation was shown to be dependent on the nutrient concentrations and decreased with increasing culture age. This presentation will discuss how the combination of soft X-ray spectromicroscopy and geomicrobiological experiments helped to understand the biogeochemical process of bacterially mediated mineral nucleation. We thank K. Kaznatcheev, C. Karunakaran, D. Bertwistle (CLS-SM), D. Kilcoyne, T. Tyliszczak (ALS); P. Gasser (EMEZ, ETHZ), L. Holzer (3D-Mat, EMPA). Study supported by the Swiss National Science Foundation (PBEZ2-115172), NSERC, Canada Research Chair. CLS supported by NSERC, CIHR, NRC and the University of Saskatchewan. ALS supported by DoE-BES.

Goldschmidt Conference Abstracts 2008

Formation and oxidation of FeS(aq) molecular clusters: Decoupling iron sulfide mineral dissolution and oxidation reactions H.D. ODURO1,2 AND G.K. DRUSCHEL1 1

Dept. of Geology, University of Vermont, 180 Colchester Ave. Burlington, VT 05405 ([email protected]) 2 Dept. of Geology, University of Maryland, College Park, MD 20742 Iron sulfide molecular clusters, FeS(aq), are a group of polynuclear iron-sulfide complexes formed under reducing conditions at different modern anoxic or suboxic interfaces in various environmental systems, and are necessary precursors for the formation of iron sulfide minerals [1-5]. Batch and agarose gradient experiments at circumneutral pH show nonstiochiometric, non-oxidative dissolution of mackinawite, pyrrhotite, pyrite, and marcasite minerals with formation of FeS(aq), illustrating a significant role for FeS(aq) in the dissolution of Fe-S minerals. Batch and agarose gradient experiments where non-oxidative dissolution products were subjected to oxic conditions indicate no significant formation of intermediate sulphur or sulfoxyanion species in contrast to studies of direct iron sulfide mineral surface oxidation. The dynamic conditions between the formation of the clusters as well as other species in suboxic conditions may thus be a critical component to understanding the cycling of iron and sulfur in a host environments. This suggests significant differences in the reaction pathways for the overall oxidation of iron and sulfur in Fe-S minerals depending on the specific redox environment they occupy. [1] Theberge & Luther (1997). [2] Rickard & Luther (1997). [3] Wolthers et al. (2003). [4] Butler et al. (2004). [5] Butler et al. (2005).

A697

Mineralogical and geochemical study of PM10 in Strasbourg A. OEHLER1, R. GIERÉ1, P. STILLE2, B. GROBÉTY3 AND V. DIETZE4 1

Universität Freiburg, Germany ([email protected]) Université Louis Pasteur, 7517 Strasbourg, France 3 Universitas Fribourgensis, Pérolles, 1700 Fribourg, Suisse 4 Deutscher Wetterdienst, Freiburg, Germany 2

The study focused on the mineralogy and geochemistry of airborne particulates (PM10) collected in June 2007 near the centre of Strasbourg at 30 m above ground. An active sampler (Digitel) was used to collect PM10 on a teflon filter for subsequent analysis of the bulk chemical composition by ICPMS. In the entire suite of samples one sample stands out (collected on June 13/14): its bulk chemical composition is characterized by a distinctly larger mass concentration of metals, such as, Pb, Fe and Zn (e.g., ~3x higher than on June 07/08). A second set of PM10 samples was collected with an active sampler on carbon filters, which were characterized by SEM using automatic single-particle analysis (Genesis, EDAX). EDS spectra were used for mineralogical classification of the particles. This method revealed a distinct increase in anthropogenic particles on June 13/14 relative to June 7/8. The number of transition metal-bearing particles (e.g., Mn- and Ti-oxide; Fe- and Zn-sulphate) increased by up to seven times. A passive sampler (Sigma-2) was used to collect particulate matter, ranging in size from 2.5 to 10 µm, on a transparent collection plate suitable for automated singleparticle optical microscopy. The mass concentrations, calculated from the optical data, revealed a roughly 5-fold increase in anthropogenic particles on June 13/14 compared to the other samples. This considerable difference between the two sampling periods was caused by different meteorological conditions, especially wind direction (N-NE on June 7/8 vs. SSW on June 13/14), as confirmed by backward trajectories (calculated by NOAA, HYSPLIT MODEL).

A698

Goldschmidt Conference Abstracts 2008

Bona fide biosignatures: Insights from combined NanoSIMS-SIMS

The surface chemistry of multi-oxide silicates

D.Z. OEHLER1*, F. ROBERT2, M. CHAUSSIDON3 1 AND E.K. GIBSON

E.H. OELKERS*, S.V. GOLUBEV, C. CHAÏRAT, O. S. POKROVSKY AND J. SCHOTT

1

Astromaterials Research and Exploration Science, NASA Johnson Space Center, Houston, TX 77058 (*correspondence: [email protected]) 2 Laboratoire d’Etude de la Matière Extraterrestre, Muséum National d’Histoire Naturelle, Paris, 75005, France 3 CRPG-CNRS, BP 20 54501 Vandoeuvre-lès-Nancy, France Bona fide biosignatures are those for which both biogenicity and syngeneity can be established. NanoSIMS, particularly in combination with SIMS, is providing new insights into both these criteria. NanoSIMS yields in situ, 50 nm-resolution, element abundance and ratio maps of trace quantities of carbonaceous and mineral materials. In 2006, NanoSIMS was used to characterize C, N, S, Si and O occurrence in undisputed Late Precambrian (~ 0.8 Ga) microfossils excellently preserved in chert [1]. Results provided a baseline for interpreting less well preserved organic remains in older and/or extraterrestrial samples. More recently, NanoSIMS was used to assess spheroidal and spindle-shaped structures in Early Archean (~ 3 Ga) cherts [2]. Element distributions are similar to those in the ~ 0.8 Ga microfossils but distinct from those of carbonaceous material within a hydrothermal vein. Also, spatial relationships among C, Si and O in the Archean and 0.8 Ga structures may reflect silica nucleation on organic surfaces [1], supporting prior conclusions that the structures are the same age as the enclosing cherts. Initial δ13C analysis of the individual Archean structures, using SIMS, supports a biogenic interpretation. If further study corroborates this assessment, then the age, size, and complexity of the spindles would suggest that life on Earth was moderately advanced by 3 Ga. Pairing NanoSIMS with other techniques such as SIMS will provide further insights into biogenicity and syngeneity that can be applied to any minute C-bearing residues that might be found in ancient terrestrial or extraterrestrial samples. [1] Oehler et al. (2006) Astrobiology 6(6), 838-850. [2] Oehler et al. (2008) 39th LPSC, Abstract #1303.

LMTG, UPS/OMP/CNRS UMR 5563, 14 av. Edouard Belin, 31400 Toulouse, France (*correspondence: [email protected]) The surface chemistry of natural wollastonite, diopside, enstatite, forsterite, and albite in aqueous solutions was characterized using both surface titrations in limited residence time batch reactors and electrokinetic techniques [1, 2]. The compositions of all investigated surfaces vary dramatically with pH. Ca and Mg are preferentially released from wollastonite, diopside, enstatite, and forsterite at pH less than ~8.5-10 but preferentially retained at higher pH. As such, the surfaces of these minerals are Si-rich/divalent metal poor except in strongly basic solutions. The preferential removal of divalent cations from these surfaces is coupled to proton consumption. The number of protons consumed for each divalent cation removed varies from 1.5 to 4 depending on the mineral. This suggests the creation of additional ‘internal’ adsorption sites through the preferential removal of divalent metal cations. Similarly, Na and Al are preferentially removed from the albite surface at all pH; mass balance calculations suggest that three protons are consumed by the preferential removal of each Al atom from this surface. Electrokinetic measurements yield isoelectric points of 2.6, 4.4, 3.0, 4.5, and <1 for wollastonite, diopside, enstatite, forsterite, and albite, respectively, consistent with the predominance of SiO2 in the surface layer of all of these silicates at acidic pH. Taken together these observations suggest fundamental differences between the surface chemistry of simple versus multi-oxide minerals including 1) a dependency of the number and identity of multi-oxide silicate surface sites on the aqueous solution composition, and 2) multi-oxide dissolution rates and surface charge exhibit different pH variations [3, 4]. [1] Pokrovsky & Schott (2000) GCA 64, 3299-3312. [2] Chaïrat et al. (2007) GCA 71 5888-5900. [3] Oelkers (2001) GCA 65, 3703-3719, [4] Oelkers & Schott (2001) GCA, 65, 1219-1231.

Goldschmidt Conference Abstracts 2008

A699

Near-equilbrium albite dissolution kinetics

Origins of hematite, magnetite and siderite in banded iron formations

ALEXIS D. OFFNER1, ROLF S. ARVIDSON1 AND ANDREAS LUTTGE1,2

H. OHMOTO1*, M. HOASHI2, T. OTAKE1, D. BEVACQUA1 1 AND Y. WATANABE

1

Department of Earth Science, Rice University, Houston TX 77005 ([email protected]) 2 Department of Chemistry, Rice University, Houston TX 77005 Extensions of BCF theory indicate that if undersaturation does not exceed a critical value (i.e., ∆G ≥ ∆Gcrit), hollow cores of screw dislocations with large Burgers vectors will not open to form macroscopic etch pits [1]. This analysis has been expanded to a general framework of crystal dissolution that explains the oft-observed nonlinear decrease in crystal dissolution rate as equilibrium is approached [2]. These relationships, when applied to albite dissolution (Fig. 1), also imply the existence of two distinct near-equilibrium reaction mechanisms with large differences in specific rate: a fast mechanism in which etch pits serve both as areas of accelerated bulk removal and as step-generators (stepwaves) as well, and a slow mechanism in which step-generation is inactive [3, 4]. However, both ∆G and ∆Gcrit are functions of temperature. We have thus used novel experimental and analytical techniques to explore these potential mechanisms, and here present preliminary results under near equilibrium conditions < 100°C.

Figure 1: Albite cleavage surface after dissolution, pH 9. [1] Cabrera et al. (1954) Phys. Rev. 96, 1153. [2] Lasaga & Luttge (2001) Science 291, 2400–2404. [3] Beig & Luttge (2006) GCA 70, 1402–1420 [4] Luttge (2006) J. Electron. Spectrosc. Relat. Phenom. 150, 248–259.

1

Astrobiology Research Center and Dept. of Geosciences, Penn State University, University Park, PA 16802 USA (*correspondence: [email protected]) 2 Center for Promotion of Science, Kagoshima University, Kagoshima, 890-0065 Japan Geological, geochemical and mineralogical investigations by previous investigators and our group suggest that Lake Superior-type BIFs have accumulated in large, nearly completely-closed basins, much like the modern Black and Red Seas. The basins possibly developed by rifting of a thick continental crust and hosted Fe2+-rich brine pools. TEM and other data suggest the primary Fe3+-bearing compounds were colloidal ferric hydroxide particles (T < ~60°C), which were subsequently transformed to hematite during diagenesis. Nucleation of silica and ferric hydroxides occurred episodically, but simultaneously throughout the basin, probably at the interface between the Fe2+-rich brine and overlying O2-rich seawater, to form basin-wide continuous microbands. Algoma-type BIFs, occurring in submarine volcanic terranes, possess many characteristics of black-smoker-type VMS deposits, including foot-wall alteration. The Marble Bar Jasper/Chert unit (MBC) in the Pilbara Craton, Western Australia is a 3.46 Ga low-grade Algoma-type BIF. Hematite in the MBC mostly occurs as a nano-sized (~100 - 600 nm) particle with varying morphology and habit. TEM and SEM observations suggest the hematite particles formed directly as single crystals of hematite at T >60°C, during the rapid mixing of high-temperature (>150°C) Fe2+-rich submarine hydrothermal fluids with O2- and CO2-rich seawater near the vents at depths >400 m. Hematite in the MBC is frequently enclosed in magnetite (Fe3O4) and siderite (FeCO3) crystals, indicating that the hemitite formed earlier than these minerals. The magnetite most likely formed during the early diagenesis of hematite- and silica-rich sediments by reaction between hematite and Fe2+-rich hydrothermal solutions: Fe2O3 + Fe2+ + H2O = Fe3O4 + 2H+. The siderite also formed during diagenesis of the chemical sediments by reaction between hydrothermal Fe2+ and CO2 (and HCO3-)-rich pore fluids; the CO2 and HCO3- were derived from a CO2-rich Archean atmosphere and the decay of organic matter in the chemical sediments. These and various other mineralogical and geochemical data of the MBC suggest that the oceans and atmosphere were already oxygenated 3.46 Ga ago.

A700

Goldschmidt Conference Abstracts 2008

Modeling the chemical evolution of the atmosphere and oceans

Bimodal magmatism in the Andriamena belt, Madagascar. Implications for continental crust evolution in Neoproterozoic times

HIROSHI OHMOTO1* AND ANTONIO C. LASAGA2 1

Astrobiology Research Center and Dept. of Geosciences, Penn State University, University Park, PA 16802, USA (*correspondence: [email protected]) 2 Geokinetics, State College, PA 16801, USA We have simulated the long-term (4 billion years) changes in the concentrations of atmospheric gases (O2, CO2, H2, CH4, H2S, SO2), the concentrations of aqueous species in the oceans (O2, CO2, HCO3-, SO42-, H2S, Fe2+, and other redox sensitive elements), and the amounts and isotopic compositions of organic C, carbonate C, sulfide S and sulfate S in sedimentary rocks under a variety of scenarios for the growth of continental crust, volcanic flux, and H2 escape to the space. Our model is not a box model, but is a dynamic model [1] where equilibrium (i.e. input = output) is not assumed for each subsystem. The results of these simulations put severe constraints on possible scenarios for the evoluton of biosphere and environment on early Earth. For example, if the atmosphere was H2-rich and the volcanic flux of reduced volcanic gas was >10X the present value, within ~500 Ma of the emergence of photoautotrophs (oxygenic or non-oxygenic) the atmospheric (and volcanic) CO2 would have been completely converted to organic matter (and carbonates) in sediments, resulting in the Snowball Earth. If the flux of reducing volcanic gas was less than ~7X of the present value and if the land area was greater than ~10% of the present value, the atmospheric pO2 would have risen to ~1 PAL in less than 30 Ma after the emergence of oxygenic photoautotrophs. If the volcanic flux gradually decreased and the land area increased to the present values, the pO2 would have remained at ~1 PAL. The pCO2 would have gradually decreased from 1000 PAL to the present level over a 4 Ga period, while the concentrations of sulfur, Fe, and other redox species in the oceans have remained essentially the same as those in the present oceans. Such a scenario is consistent with the sedimentary records of concentrations and isotopic compositions of C, S, and Fe. [1] Lasaga & Ohmoto (2004).

M. OHNENSTETTER1*, A. RATEFIARIMINO2 AND D. RAKOTOMANANA2 1

CRPG/CNRS, BP 20, 54501, Vandoeuvre les Nancy, France (*correspondence: [email protected]) 2 PGRM, Ex Laboratoire des Mines, Route d'Andraisoro, Antananarivo 101, Madagascar The Andriemena greenstone belt belongs to the Tsaratanana sheet which was trust onto the central northern Antananarivo block. The Andriemena belt is associated with N-S trending upper crustal migmatites, comprising amphibolite to granulite-facies orthogneiss to paragneiss. The synformal structure of the greenstone belts is the result of the ca. 600-550 Ma collision between India and Congo blocks. Before the final collision, the 820-740 episode is related to the development of an arc magmatic sequence with gabbro and granitoid intrusives. Ultramafic rocks within the Andriamena belt, dated at 787±6 Ma, occur within small scale intrusives bodies, the shape of which is controlled by subvertical large scale shear zones. Dunite, harzburgite, orthopyroxenite and pyroxenite are found in narrow N-S conduits or within larger magma chambers, however less than one km wide. Within the orthogneiss commonly surrounding the ultramafic massifs, a charnokite magma suite has been found. The suite comprises a high proportion of intermediate terms which show high TiO2, K2O and P2O5, LILE elements and incompatible elements as Zr (up to 1230 ppm). Rocks are also characterized by more or less prominent negative anomalies of Rb, K, Sr, P and Ti, indicative of strong fractionation processes. The presence of a charnokite type suite devoid of negative anomalies of Nb and Ta may be indicative of a rift environment during Neoproterozoic times. Strong interaction between crustal melt components and mantle magma is otherwise attested by their respective trace element evolution and isotopic data.

Goldschmidt Conference Abstracts 2008

A701

Mass-dependent isotopic fractionation of Sr in geochemical samples by MC-ICP-MS

Interactions of Actinides with Microorganisms and Organic Ligands

TAKESHI OHNO, TAKAHIRO WAKABAYASHI AND TAKAFUMI HIRATA

T. OHNUKI1, T. OZAKI1, F. SAKAMOTO1, N. KOZAI1, T. NANKAWA1, Y. SUZUKI1 AND A. J. FRANCIS2

Department of Earth and Planetary Sciences, Tokyo Institute of Technology, Japan ([email protected]) The study of naturally occurring isotopic fractionation of Sr has a potentially significant influence in geochemical research fields combined with traditional studies using radiogenic isotopes as a tracer, because information on isotopic fractionation may provide another constraint on a formation process of a sample. In this study, we present a method to determine 88Sr/86Sr and 87Sr/86Sr simultaneously. The former variation reflects the mass-dependent isotopic fractionation through the physicochemical processes, and the latter originates from decay of the parent nuclide 87Rb as well as the mass-dependent isotopic fractionation. In order to determine the mass-dependent isotopic fractionation, the mass-discrimination effect on 88 Sr/86Sr was externally corrected by an exponential law using Zr. For the radiogenic growth of 87Sr/86Sr, the mass-dependent isotopic fractionation effect on 87Sr/86Sr was corrected by a conventional correction technique using the 88Sr/86Sr ratio. The reproducibility of the 88Sr/86Sr and 87Sr/86Sr measurements for a high-purity Sr chemical reagent was 0.06‰ (2SD, n = 20) and 0.07‰ (2SD, n = 20), respectively. Isobaric interferences and matrix elements were chemically eliminated by an extraction chromatography using Sr spec. resin. We examined Sr isotopic fractionation during the separation procedure. The cumulative Sr isotopic value of the eluent indicates that no detectable isotopic fractionation of Sr was found through a few percent loss of Sr during the chromatographic separation. Strontium isotopic ratios (88Sr/86Sr and 87Sr/86Sr) were measured on six geochemical reference materials (igneous rock: JB-1a and JA-2; carbonate mineral: JLs-1, JDo-1, JCp-1 and JCt-1) and one seawater sample. The resulting 87Sr/86Sr ratios obtained here were consistent with previously published data within the analytical uncertainties. The resulting 88Sr/86Sr ratios for igneous rock samples did not vary significantly within the samples, whereas the carbonate samples showed enrichments of the lighter Sr isotopes over the seawater sample. The 88Sr/86Sr ratio of geochemical samples could reflect the physico-chemical processes for the sample formation. Also, a combined discussion of 88Sr/86Sr and 87 Sr/86Sr of samples will render multi-dimensional information on geochemical processes.

1

Japan Atomic Energy Agency, Tokai, Ibaraki, 319-1195 Japan 2 Brookhaven National Laboratory, Upton, NW, 11973, USA Trivalent actinides, such as Am(III) and Cm(III), are highly toxic because they emit high-energy alpha-particles and have long half-lives. Plans to dispose of HLW and TRU wastes in geological disposal repositories have raised a number of concerns about polluting the environment through dissolution and subsequent mobilization of actinides; thus, long-term assessments of safety are required. TRU wastes contain cellulosic materials, scintillation fluids, waste oils, decontamination reagents, and chemical reagents. Among them, organic acids, such as citric acid, EDTA, and NTA form stable complexes with multivalent radionuclides, enhancing their mobility by increasing their solubility. However, little is known of the interaction of actinides with microorganisms and organic ligands. In this paper, we summarize our findings on the interactions of actinides with microorganism and organic ligands. Malic acid formed complexes with Eu(III), analogues of Am(III) and Cm(III), but degradation of malic acid by Pseudomonas fluorescens was observed when the ratio of malic acid to Eu(III) was higher than 100. One of the metabolites associated with Eu(III), that were produced by the degradation of malic acid, was determined as pyruvic acid by ESI-MS. The effect of bacterial organic exudates on the sorption of Eu(III) and Cm(III) by Chlorella vulgaris was studied. The pH dependence of log Kd of Eu(III) and Cm(III) for cellulose, major component of C. vulgaris cell, differed from that for C. vulgaris. However, the log Kd of Eu(III) and Cm(III) for cellulose in the solution containing exudates from C. vulgaris cells in a 0.5% NaCl solution showed a similar pH dependence to that by C. vulgaris. These findings strongly suggest that interaction of microorganisms with the actinide organic complexes affect the migration of actinides.

Goldschmidt Conference Abstracts 2008

A702

Chemical reactions and element partitioning at the core-mantle boundary

Contrast occurrence of banded iron formations in western part of Isua Spracrustal Belt, West Greenland

E. OHTANI

Y. OHTOMO AND T. KAKEGAWA

Faculty of Science, Tohoku University, Sendai 980-8578, Japan ([email protected]) Recent advances in high-pressure spectroscopic techniques have led to the discovery of the spin transitions and a post-perovskite phase [1, 2]. Spin transition occurs in magnesiowustite, and perovskite in lower mantle conditions. The transitions might cause changes in partitioning behaviour between these minerals [1] although the pressure interval in spin transitions can be large at high temperature [3]. The postperovskite phase with CaIrO3 structure is stable at pressures above 110 GPa and at high temperature. The effects of Al and Fe on phase transition and element partitioning between postperovskite and magnesiowustite are also under debate [4, 5]. In this talk, I will review these recent results on the phase relations at the base of the lower mantle including our new results on Mg-Fe partitioning between post-perovskite and magnesiowusitute, which shows a large compositional dependency in the partition coefficient. The reactions between metallic iron and silicates are also essential process at the core-mantle boundary. I also summarize the recent results on the reactions between metallic iron and silicate/oxide under the lower mantle/CMB conditions [6, 7]. I also show our recent results on dissolution of O, Si, and H in metallic iron as candidates for the light elements in the core, and dissolution of K as a possible heat source in the core [8, 9]. Significant amount of O, Si, and H can be dissolved in molten iron at the CMB conditions, whereas dissolution of K might be limited and K may not be important as a heat source in the core. [1] Badro et al. (2004) Science 305, 383-386. [2] Murakami et al. (2004) Science 304, 855- 858. [3] Lin et al. (2007) Science 317, 1740-1742. [4] Mao et al. (2004) PNAS 101, 15867– 15869. [5] Murakami et al. (2005) GRL, 32, L03304, doi,10.1029/2004GL021956. [6] Sakai et al. (2006) GRL, 33, doi, 10.1029/2006GL026868. [7] Asahara et al. (2007) EPSL, 257, 435-449. [8] Hirao et al. (2006) GRL, 33, L08303, doi, 10.1029/2005GL025324. [9] Sano et al. (2008) GRL, Vol. 35, L03303, doi,10.1029/ 2007GL031718.

Institute of Mineralogy, Petrology, and Economic Geology, Graduate School of Science, Tohoku University, Sendai, Japan ([email protected]) Banded iron formations (BIF) are widespread in Isua Spracrustal Belt, West Greenland (ca. 3.8Ga). Detailed geological surveys of BIFs in the western side of Isua were performed to find as to if (1) indicative materials for their depositional environments and (2) remnants of life were preserved. In the northern part of low-strain area, BIFs were interbedded with graphitic schist. Laminations of graphite, quartz and magnetite/cummingtonite/anthophyllite layer are observed in microscopic scale in graphitic schist. Mg-Fe chlorite, ilmenite, zircon, monazite and apatite are distinguished as minor components. REE patterns of such graphitic schist are similar to other Archean shales. These geochemical characteristics and existence of detrital zircon suggest that graphitic schist was clear clastic marine sediments containing remnants of 3.8 Ga biota, probably analogous to reported metasediments [1]. Note that magnetite in BIFs also contains microscopic graphite inclusions. In the east side of high-strain area, a few m thick BIFs occur adjacent to ultramafic rocks [2]. These BIFs consist of quartz and magnetite/amphibole layer, without having graphite. Garnet-chlorite-biotite schist appears next to BIFs in particular at the boundary with Ameralik dike. Geochemical and mineralogical studies indicate that such Al-rich sequences are metasomatic origin, suggesting the absence of clastic components in the eastern BIFs of the examined area. These magnetite-BIF in high-strain area could be relatively deep-sea chemical sediments contrasting to the clastics-rich northern BIFs. [1] Rosing, M. T. (1999) Science 283, 674-676. [2] Furnes et al. (2007) Science 315, 1704-1706.

Goldschmidt Conference Abstracts 2008

Kinetics and textural development of quartz veins

Rare Earth Elements (REE) geochemistry of Iron mineralization of Sarikaya (Yozgat, Turkey)

ATSUSHI OKAMOTO1 AND NORIYOSHI TSUCHIYA2

N. OKSUZ1 AND S. KOC2

1

Graduate School of Environmental Studies, Tohoku Univ. Japan (okamoto @mail.kankyo.tohoku.ac.jp) 2 Graduate School of Environmental Studies, Tohoku Univ. Japan ([email protected]) Quartz veins generally contain various textures. To understand the relationship between developments of vein textures, crystallization mechanism and physico-chemical environments in fluid-filled cracks, we conducted hydrothermal flow-through experiments at 150-430 °C and 30 MPa. The reaction tube is diameter of 10 mm and length of 31 cm, that was set to produce a horizontal flow path. First, we determined precipitation rate constant from dissolution of 31 g quartz sand (1 - 2 mm) at flow rate of 1.01 g/min. Si concentration of solutions in outlet increases 0.6 to 290 ppm in temperatures of 150-390 °C. From the Arrhenius plot, the activation energy is estimated to be 51.1 kJ/mole, which is consistent with previous estimates for temperature <300 °C [1]. Our result also suggests that the reaction was surface-controlled in advective flow. Second, we employed precipitation experiments of quartz with granite plates (4.5 x 4.5 x 30 mm) at 430 °C for 50 hours, under which the degree of supersaturation, CSi/CSi,eq = 3-4. We found two types of quartz growth occurred simultaneously. On granite surfaces, quartz grew epitaxially on pre-existing quartz grains. In other parts, sedimentation of euhedral quartz crystals (size of 0.01-0.5 mm) occurred. These textures in different sites are similar to elongate-blocky and blocky textures in natural quartz veins, respectively. Our results suggest that nucleation could occur even under relatively low supersaturation, and that the vein textures would be controlled by the ratio of quartz surface area / mass of water (crack aperture). [1] Rimstdt & Barnes (1980) Geochim Cosmochim Acta 44, 1683-1699.

A703

1

Bozok University, Department of Geology, 66100, Yozgat, Turkey ([email protected]) 2 Ankara University, Department of Geology, 06100, Ankara, Turkey ([email protected]) Iron mineralization in Sarıkaya (Yozgat) was observed in three different locations named Büyükören, Karabacak and Uzunkuyu-Atkayası, situated 25 kilometers to the west of Sarıkaya district. This process of mineralization is syngenetically related to the rocks with calc-alkaline composition which metamorphosed into amphibolite as a result of the metamorphism of basalts. In the paragenesis of mineralization the main mineral components are magnetite and hematite, and to a small extent pyrite and chalcopyrite. Among the mentioned regions, magnetite mineralization is commonly observed particularly in Büyükören and Karabacak, whereas hematite mineralization is seen in the region of Uzunkuyu-Atkayası [1]. The REE data from ore samples in the study area were used to organize chondrite-normalized spider diagrams separately for each region. Accordingly, negative Eu anomaly was detected in Büyükören region, whereas negative Ce as well as both negative and positive Eu anomalies were detected in the regions of Uzunkuyu-Atkayası and Karabacak. The fact that LREE values were higher than those of HREE in the spider diagram suggests that mineralization was influenced by a felsic rock [2]. This further indicates the presence of a hydrothermal mineralization [3-9]. In the ore samples from the research region, (Eu/Sm)CN ratios vary between 0.1-2.65, whereas (Sm/Yb)CN ratios between 0.15-10.72. (Eu/Sm)CN<1 and (Sm/Yb)CN>1 ratios point the fact that Fe-containing sediments formed the mineralization [10]. These values showed that the all the ore formation in the region was not formed from sedimentary Fe layers. The (La/Lu)N ratio in the ore samples of the region was observed as varying within the range of from 0.52 to 59.43, which points to the mid-acidic and high-temperature conditions of the magnetite mineralization in the region as well as its magmatic origins. [1] Oksuz (2007) Ankara Unv. Doctoral thesis. [2] Khan & Naqvi (1996) Mineral Deposita. [3] Ozmen & ve Koc. (2001) [4] Singh & Rajamanı (2001) Geochimica et Cosmochimica Acta.Economic Geol. [5] Jiang et al. (2002) Chemical Geology. [6] Peng & Palmer (2002) Economic Geol. [7] Sagiroglu & Sasmaz (2004) Journal of Asian Earth Sciences. [8] Fitzgerald & Gillis (2006) Marine Geology. [9] Ghaderi et al. (2006) Research School of Earth Sciences. [10] Fernandez & Moro (1998) Mineralium Deposita.

A704

Goldschmidt Conference Abstracts 2008

Distributions of Th isotopes in the North Pacific Ocean A. OKUBO1, H. OBATA2, T. GAMO2 AND J. ZHENG1 1

National Institute of Radiological Sciences, Ibaraki, Japan ([email protected], [email protected]) 2 Ocean Research Institute of University of Tokyo, Tokyo, Japan ([email protected], [email protected]) We investigated the vertical distribution of thorium isotopes in mid-latitudes of the Pacific Ocean especially 230Th as a test case of scavenging of metals, and discuss the control factor of the distribution of 230Th. Sea water samples were collected on board the R/V Hakuho-Maru during KH-00-3 (The Bootes expedition) on Jun-July, 2000 and KH-03-1 (The Hydra expedition) on JunAugust, 2003. For 230Th measurements, large-volume water samples of 250 litters for each depth were obtained. After separation with anion exchange, 230Th was determined by alpha spectrometer. For 232Th measurements, clean sampling techniques were applied. 232Th was determined by SF-ICP-MS (Finnigan Element 2). 230 Th section shows lateral uniformity in the upper ca. 3000 m, but below 3000 m, 230Th distributions show geographical variations. Regarding the stratified hydrography in the deep layer, the depletion of 230Th could be attributed to local bottom scavenging at seafloor and diffusion of low 230Th sea water. 232 Th section along 20˚N did not indicate definite lateral transport effect. Assuming no lateral transport of 232Th in the Central Pacific, atmospheric 232Th input is calculated to be 1.2 µg/m2/year from 232Th concentration in water column and 230 Th residence time. This calculated input corresponds with the estimation based on dust flux and averaged 232Th concentration in the dust (2−5 µg/m2/year [1]). [1] Roy-Barman et al. (1996).

Using trace elements to constrain relationships between minerals and glasses in mixed-magma volcanic rocks P.H. OLIN* AND J.A. WOLFF SEES, Washington State Univ., Pullman, WA 99164-2812 USA (*correspondence: [email protected]) Assessing geochemical relationships between minerals and glasses in mixed-magma volcanic rocks can be challenging because of the scrambling of the different magmas and crystal cargos. These and subsequent processes preclude assuming simple genetic relationships between any crystal and the adjacent glass. An example is the 309±6 ka, ~13 km3 DRE Fasnia Member of the Diego Hernández Formation, Tenerife. The Fasnia represents the eruption of a mixture of two distinct end member phonolite magmas and a small amount of mafic magma that resulted in a significant volume of hybrid phonolite magma. Microprobe and LA-ICP-MS analyses of crystals and glass in Fasnia pumices have revealed three distinct pyroxene groups (salites with ~1wt% Na2O, Na-salites with >1.5wt% Na2O, and titanaugites), titanites with a range in compositions (32.3 to 36.7 wt% TiO2), and three dominant phonolitic glass compositional fields that were not resolvable by bulk pumice analyses; low-Zr phonolite (500 to 850 ppm Zr), high-Zr phonolite (>1800 ppm Zr), and hybrid phonolite (1100 to 1800 ppm Zr). Clearly the titanaugites crystallized in the mafic magma, however confidently matching salites, Nasalites, and titanites with their respective parental magmas required a more rigorous approach. We applied the lattice strain trace element partitioning model using REE in the salites, Na-salites, titanites, and ~500 glass analyses. First, apparent REE mineral-melt partition coefficients for pyroxenes and titanites were calculated using all the glasses. Then the apparent partition coefficients for each mineral-glass pairing were evaluated using a linear form of the lattice strain model equation. Mineral glass pairs with the best linear correlation coefficient were identified as equilibrium pairs. Using this approach we are able show that salites crystallized in low-Zr phonolite magma and Na-salites crystallized in hybrid phonolite. Similarly, the titanites crystallized in low-Zr and hybrid phonolite. This suggests that high-Zr phonolite was crystal-free and may represent a filter-pressed or super-heated liquid.

Goldschmidt Conference Abstracts 2008

Developing models to predict silicate mineral dissolution AMANDA ALBRIGHT OLSEN, SUSAN L. BRANTLEY AND JAMES D. KUBICKI Center for Environmental Kinetics Analysis, Penn State ([email protected], [email protected], [email protected]) The goal of this project is to develop a model that allows prediction of the rate of silicate mineral dissolution over a wide range of pH and temperature. Such a model would be useful for rate prediction when data is unavailable as well as to elucidate inconsistencies among published data for mineral dissolution kinetics. We describe an empirical model and an approach to seek theoretical constraints on this model. Previous experiments have identified key factors which affect silicate dissolution rates and have been included in an empirical model. Casey and Westrich [1] showed that the rate of water exchange around an octahedral cation is correlated to the mineral dissolution rates in orthosilicates; a similar but weaker correlation has been observed in pyroxenes [2]. This does not necessarily suggest that water is exchanging at this cation in the the dissolution mechanism since this correlation: the correlation could be an indirect measure of metal-O bond strength within a mineral structure. It has also been suggested that minerals with a lower connectedness (the number of bridging oxygens per tetrahedron) dissolve more quickly than minerals with a higher degree of polymerization. We compiled data from approximately 70 laboratory studies of silicate dissolution (see ChemXSeer (http://chemxseer.ist.psu.edu/), an online database focused on environmental kinetics data). Minerals considered include orthosilicate, pyroxene, amphibole, sorosilicate, and phyllosilicate groups. We observe a strong correlation between log r and pH, ksolv, connectedness, and temperature. We are furthermore using ab initio quantum mechanical modeling of hydrolysis of cations in silicate structures as a function of cation identity to establish a theoretically based underpinning to the empirical model. We are using this approach to calculate rate constants as a function of pH using the POLYRATE program [3]. [1] Casey & Westrich (1992) Nature 355, 157-159. [2] Banfield et al. (1995) GCA 59, 19-31. [3] Nangia & Garrison (2008) J. Phys. Chem. A.

A705

Carbonate precipitation potential by cyanobacterial activity in high Arctic cryptoendolithic habitats C.R. OMELON1*, W.H. POLLARD2, F.G. FERRIS3 AND P.C. BENNETT1 1

Department of Geological Sciences, The University of Texas at Austin, Austin, Texas, 78712, USA (*correspondence: [email protected]) 2 Department of Geography, McGill University, Montréal, Québec, H3A-2K6, Canada 3 Deparment of Geology, University of Toronto, Toronto, Ontario, M5S-3B1, Canada Cryptoendolithic habitats are host to a wide variety of microorganisms including cyanobacteria, algae, fungi and heterotrophic bacteria and are found in both hot and cold deserts. Low pH conditions are normally associated with these habitats due to the generation of acidity by fungi, but recent studies [1, 2] reveal elevated pH conditions associated with cyanobacterial-dominated microbial communities. These high pH conditions are result from the activation of a carbon concentrating mechanism in these microorganisms to generate OH- ions that are expelled from the cell into the surrounding microenvironment. In addition to enhancing dissolution of the sandstone framework including etching of quartz grains and dissolution of silica-rich intergranular cements [1] elevated concentrations of Ca2+ ions in proximity to these communities [2] suggest that they can induce carbonate precipitation during photosynthesis [1]. Although described previously from aquatic environments [3], little is known about mechanisms for carbonate biomineralization in terrestrial habitats. The current study investigates the potential for carbonate precipitation by cyanobacteria isolated from high Arctic cryptoendolithic habitats around Eureka, Ellesmere Island, Nunavut (80˚00’N, 85˚55’W). Through examination of specific surface functional groups, potentiometric titration experiments and TEM observations, we propose that the same mechanism for carbonate biomineralization can occur here, whereby binding of metal cations onto cell surfaces – combined with excess production of OH- during cyanobacterial photosynthesis – generate conditions suitable for mineral precipitation in this high pH setting to create unique terrestrial biosignatures in silicate rocks. [1] Büdel et al. (2004) Geobiology 2, 261-268. [2] Omelon et al. (2007) Microb Ecol 54, 740-752. [3] Thompson et al. (1998) Geology 18, 995-998.

A706

Goldschmidt Conference Abstracts 2008

Accessory (REE+Y)XO4 phases: substitutions of As, S, Nb, Zr, Th, Ca and Sr in endogenous systems M. ONDREJKA1* AND P. UHER2 1

Dept. of Mineralogy and Petrology, Comenius Univ., Mlynská dolina G, 842 15 Bratislava, Slovak Republic (*correspondence: [email protected]) 2 Dept. of Mineral Deposits, Comenius Univ., Mlynská dolina G, 842 15 Bratislava, Slovak Republic ([email protected] ) The solid solution systems of accessory (REE+Y)XO4 phases with the monazite- or xenotime-type structure show a wide range of substitutions with apparent compositional variations of elements, especially As, S, Nb, Zr, Th, Ca, Sr in endogenous, magmatic to post-magmatic/ hydrothermal and metamorphic environments. Sulphur, arsenic and niobium are commonly incorporated into REE, Y, Th-bearing phosphate (monazite, cheralite and xenotime) and silicate minerals (thorite, coffinite, huttonite) as sulphate (SO4)2−, arsenate (AsO4)3− and/or niobate (NbO4)3− anionic groups. The As, S and Nb content of these accessory minerals from common magmatic and metamorphic rocks is usually negligible. However, our investigation of various granites, rhyolites and metamorphic rocks revealed a systematic presence of As (up to almost complete monazitegasparite, xenotime-chernovite and partially thorite-chernovite solid solutions), locally with increased S content (up to 8.4 wt. % SO3) as well as accompanying Ca or Sr (up to 7.4 wt. % SrO) and Nb constituent (up to 18 wt. % Nb2O5). The presence of Zr (up to 13 wt. % ZrO2) was documented in some intermediate solid solutions of the metastable system thoritezircon-xenotime-(Y). Small concentrations of S, As, Nb, Th, Ca as well as Sr in the (REE+Y)XO4 phases of granitic and rhyolitic rocks seem to be a primary magmatic feature, whereas large contents of these elements indicate usually a subsolidus nature, probably due to post-magmatic, hydrothermal or metamorphic overprint of the primary rocks. These phases represent typical rock-fluid interaction products. The extensive incorporation of Zr into thorite originated probably as a product of partial latemagmatic to early subsolidus alteration and/or recrystallization of zircon, together with xenotime-(Y).

Stage II of multiple-sulfur isotope records in South Africa SHUHEI ONO1*, BRADLEY GUY2, DOUGLAS RUMBLE3 2 AND NICOLAS BEUKES 1

Massachusetts Institute of Technology, Cambridge, MA 02139 (*correspondence: [email protected]) 2 University of Johannesburg, Auckland Park, South Africa 3 Carnegie Institution of Washington, Washington, DC 20015 Multiple-sulfur isotope records of the period between 2.09 and 2.45 Ga, is called Stage II, which is characterized by small non-zero ∆33S values (up to +0.5 ‰) and represents the transition from Archean mass-independent fractionation (MIF) to post-Archean mass-dependent fractionation [1]. The small non-zero ∆33S values of Stage II may reflect either input from oxidative weathering of Archean MIF continental sulfides, or atmospheric sulfur chemistry under an atmosphere with intermediate oxygen level [2]. We measured multiple-sulfur isotope ratios (δ33S, δ34S, and δ36S) of pyrite sulfur from the 2.41 Ga Koegas Formation and the 2.32 Ga Timeball Hill Formaiton in South Africa, in order to document multiple-sulfur isotope records of the Stage II by using improved high-precision analytical protocol [2]. The Koegas Formation yield MIF sulfides with ∆33S values up to +1.9 ‰, representing MIF sulfur from the youngest dated rock. The ∆33S values for pyrite sulfur in a series of drill cores of the Timeball Hill Formation range from +0.02 to +0.20 ‰, which is much narrower than previously reported values from Geophysical Laboratory [3]. The new range of 33S (determined by new high-precision protrol) is within that of Phanerozoic sulfides formed by mass-dependent (biological and hydrothermal) processes (e.g. [2, 4]). Our ∆36S data from the Timeball Hill Formation show ∆36S/∆33S ratios of -7.5, also supportting the mass-dependent origin of the small nonzero ∆33S values [2]. Therefore, there is no or very little signature of S-MIF in the 2.32 Ga Timebal Hill formation. We conclude that the transition period between MIF and MDF is shorter than 90 million years. The level of seawater sulfate rose sufficiently high within 90 million years to dilute inputs of oxidative weathering of Archean MIF suflide. This may imply that Proterozoic style sulfur cycle was established shortly after the deposition of the 2.41 Ga Koegas Formation. [1] Farquhar & Wing (2003) Earth Planet. Sci. Lett. 213, 1-13. [2] Ono et al. (2006) Geochim. Cosmochim. Acta 70, 22382252. [3] Bekker et al. (2004) Nature 427, 117-120. [4] Johnston et al. (2005) Science 310, 1477-1479.

Goldschmidt Conference Abstracts 2008

A707

Silicon cycle in weathering sequences of volcanic soils in Cameroon using δ30Si and Ge/Si

Sorption characteristics of oxyanions during and after allophane precipitation

S. OPFERGELT1,2*, B. DELVAUX1, D. CARDINAL2, C. DELVIGNE2 AND L. ANDRÉ2

E.M. OPISO*, T. SATO AND T. YONEDA

1

Université catholique de Louvain, Unité Sciences du sol, Croix du Sud 2/10, B 1348 Louvain-la-Neuve (*correspondence: [email protected]) 2 Musée Royal de l'Afrique Centrale, Chaussée de Louvain 13, B 3080 Tervuren In soils, silicon released by mineral weathering can be incorporated into secondary clays, adsorbed onto iron or aluminum oxides, or taken up by plants. These processes may significantly impact the Si isotopic signature and Ge/Si ratio of Si dissolved phase exported to water streams. Here we test these proxies, and report on detailed δ30Si of sand (>50µm), silt (2-50µm), clay (<2µm), and amorphous Si (ASi) fractions and Ge/Si ratios of clay fractions from basaltic ash soils differing in weathering stage. The δ30Si values were measured by MC-ICP-MS Nu Plasma in medium resolution, operating in dry plasma with Mg doping: δ30Si vs NBS28 ± 0.12‰ (± 2 SD), while Ge/Si ratios were determined after Ge measurements by HR-ICP-MS and Si by ICP-AES. Compared to fresh basaltic ash (δ30Si = -0.38‰; Ge/Si = 2.21µmol/mol), clay fractions displayed a gradient from -1.19‰ to -2.37‰ (δ30Si) and from 4.10 to 5.25µmol/mol (Ge/Si) from young (Y) to old (O) volcanic soil (vs), confirming the trend of lighter isotopic compositions and Ge enrichment with increasing weathering. Sand and silt fractions displayed Si isotopic signatures close to fresh ash (-0.33‰) or heavier (+0.18‰) due to dust quartz contribution evaluated by isotopic balance. Relatively to Ovs, clay fractions in Yvs were isotopically heavy and Ge-depleted in surface horizons, denoting a heavier Si source for clay neoformation therein. ASi fractions were generally isotopically close the fresh ash. However, relatively to Yvs, ASi fractions in Ovs were isotopically heavy in surface horizons, within the range of plant phytoliths. Combined with phytolith content in surface, this supports a biogenic Si input in surface horizons. In addition to Si provided by volcanic glass weathering, clays may have been partially impacted by this input. Isotopic balance calculations indicate that (i) bulk soil signatures were determined by Si proportions in mineral fractions rather than dissolved Si from irrigation water or Si adsorbed onto iron oxides, (ii) Si adsorbed onto Fe-oxides was mainly impacted by biogenic Si and irrigation water. Combining δ30Si and Ge/Si ratio is promising to identify major processes impacting the Si cycle in soil.

Laboratory of Environmental Geology, Graduate School of Engineering, Hokkaido University 060-8628 Japan (*correspondence: [email protected]) Allophane naturally exists as aluminosilicate hydrate commonly present in volcanic ash soil, pumice deposits and stream sediments. It has great retention capacity for anions over a wide range of pH (4-8) because of its high surface area and variable charge characteristics. However, the retention by adsorption due to electrostatic interaction with the variable charge would be unstable in terms of long term stability. Therefore, it is crucial to understand the mechanism of oxyanion incorporation during allophane formation because it has soil/geochemical implication for predicting their mobility in earth’s surface environment. The reactivity of Cr(VI) and As(V) on synthetic allophane like particles were investigated during and after allophane precipitation in batch experiments. Solutions containing AlCl3 and Na4SiO4 (with Si/Al ratio of 1.0 and 0.67) and oxyanions (Cr(VI) and As(V)) (10 ppm) were mixed and adjusted to near neutral pH with 1M NaOH. The mineral suspensions were shaken for 1 hr and incubated at 80 C for 5 d. The precipitates were collected by centrifugation at 15000 rpm for 30 mins and the supernatant were analysed for residual oxyanion concentration using ICP-AES. Chemical extraction of freeze dried samples was carried out using TAO and Na2HPO4 solutions and were analysed for Si, Al, Cr and As concentration. ATR-FTIR analyses were conducted to investigate the sorption of Cr and As onto allophane. The adsorption experiments indicated that more Cr and As uptake on allophane with initial Si/Al ratio of 0.67 and after precipitation respectively. The final Si/Al ratio of the samples were around 0.52 and 0.79 for Cr and around 0.5 for As. The ATR-FTIR spectroscopy of paste and freeze dried samples showed that Cr and As were surface sorbed after allophane precipitation while coprecipitated Cr and As are presumed to be in tetrahedral coordination by substituing for Si in the tetrahedral layer of the allophane structure. The structurally fixed Cr and As are more resistant to release than surface sorbed Cr and As.

A708

Goldschmidt Conference Abstracts 2008

Selenium speciation in rhizosphere soils and Aster eatonii roots: Implications for plant uptake

The role of Fe-oxides and quartz in the sequestration of carbonaceous matter in 2.72 Ga Banded Iron Formations, Dhawar Craton, India

LIBBIE L. ORAM1*, DANIEL G. STRAWN2 AND GREGORY MÖLLER3

B. ORBERGER1, R. WIRTH2, C. WAGNER3, A. NORET1, M. MASSAULT1, J.P. GALLIEN4, M. JAYANANDA5, V. ROUCHON1, E. QUIRICO6 AND G. MONTAGNAC6

1

University of Idaho, Environmental Science Department, PO Box 442339, Moscow, ID, USA 83844-2339 (*correspondence: [email protected]) 2 University of Idaho, Department of Plant, Soils, and Entomological Science, PO Box 442339, Moscow, ID, USA, 83844-2339 ([email protected]) 3 University of Idaho, Department of Food Science and Toxicology, PO Box 441052, Moscow, ID, USA, 838441052 ([email protected]) As a result of phosphate mining activities, soils and waters in the Western Phosphate Resource Area, USA contain elevated concentrations of selenium (Se). However, the bioavailability, water solubility, and toxicity of Se are dependent on speciation, which is not well understood in complex environments, such as plant rhizospheres. Research that addresses Se cycling in rhizosphere soils is required to more accurately predict Se bioavailability and environmental impact. This research investigates rhizosphere influence on Se speciation using both molecular and macroscopic experiments on natural environmental samples. X-ray spectroscopy, laboratory extractions, and greenhouse studies are used to determine variation in Se speciation in rhizosphere soil and bulk soil. Multiple Se oxidation state maps of Aster eatonii roots with attached rhizosphere soil show reduced Se species in the rhizosphere soils with oxidized Se species inside roots. Results may indicate that rhizosphere processes are oxidizing and mobilizing Se from reduced insoluble species in soils to species such as Se(VI) that are actively accumulated in plant roots. In greenhouse studies, Se(VI) was the predominant species detected in rhizosphere soil, plant roots, and control soils spiked with Se(IV) sorbed onto goethite, suggesting that Se(IV) is readily oxidized and mobilized in both rhizosphere soil and in bulk soil. Currently, laboratory extractions are being conducted to provide information on whether rhizosphere processes facilitate Se mobilization. By integrating rhizosphere Se speciation information with known ecological and biogeochemical processes, we are developing a better understanding of reaction processes allowing plant uptake of Se. This detailed information will facilitate a better understanding of Se biogeochemical cycling in the rhizosphere environment.

1

Université Paris Sud, UMR IDES 8148, Bât 509, 91405 Orsay, France ([email protected]) 2 GFZ, Telegrafenberg, Potsdam, Germany 3 Lab. PPMM, UPMC, UMR 7160 4 Place Jussieu, 75252 Paris Cedex, France 4 LPS, UMR, Gif-sur Yvette, France 5 Bangalore University, India 6 UJF, Grenoble, France 7 Lab Sc. Terre, ENS, Lyon, France The Archean BIF, composed of dark grey Fe-oxide and white microquartz bands, (Dharwar Craton, Lower Bababudan Group) is a typical Archaean hydrothermal seawater precipitate, as shown by its REE element characteritics. The massive Fe-oxide bands consist of euhedral crystals of former magnetite in part replaced by hematite (± goethite). A porous network of euhedral former crystal shapes walled by hematite and filled in with hematite spherules, is observed in the silica close to the Fe-oxide bands. The quartz bands, generally thicker than the Fe-oxide bands, contain a few discontinuous Fe-oxide alignments of micrometric magnetite-hematite spherules, which fill in negative crystal shapes of quartz. In the quartz bands, carbonaceous matter (CM) occurs as nanometric intergrowth in the magnetite-hematite spherules (CM1), and as nanometric nodules at the interface between the quartz and the spherules (CM2). Furthermore, CM embeds the hematite spherules in the porous network (CM3). Low N contents (0.09 ppm) and high C/N ratios are typical for CM1, while CM2 shows high N contens (0.9 ppm) and low C/N ratios. This corresponds to a higher maturity degree for CM1 than for CM2. Raman spectroscopy confirms the high maturity (as indicated by the vitrinite reflectance of 2-3%) performed of CM1 from the quartz matrix, while CM3 has a lower maturity degree, as CM2. δ13C vary between –17 and –26 ‰, being significantly higher in Fe-oxide rich layers than in the silica ones. Biologically derived CM (~δ13C=-24‰) is protected in silica layers. Lower δ13C -19 ‰ and low N-contents are the result of CM degradation through a catalytic effect of the Feoxides. The acidification during CM decay may be the mechanism for the transformation of magnetite to hematite under reducing conditions.

Goldschmidt Conference Abstracts 2008

Magmatic evolution of the Quaternary volcanics from Hudson and Lautaro volcanoes, Austral Andean Cordillera Y. ORIHASHI1, S. NAKAI1, H. SHINJOE2, J. A. NARANJO3, A. MOTOKI4 AND CHRISTMASSY GROUP 1

Earthquake Research Institute, the University of Tokyo ([email protected], [email protected]) 2 Tokyo Keizai University ([email protected]) 3 Servicio Nacional de Geologica y Mineria, Chile ([email protected]) 4 Dept. Mineralogia e Petrologia Ignea, Universidade de Esado do Rio de Janeiro ([email protected]) Near the Chile Triple Junction (CTJ), an about 350 km long volcanic gap in recent subduction-related volcanic activity separates two distinct volcanic zones in the Andean Cordillera; Southern Volcanic Zone (SVZ) and Austral Volcanic Zone (AVZ). Geochemical characteristics of the Quaternary volcanics drastically change across this gap; andesite with adakitic affinity in AVZ and mainly arctholeiites in SVZ [1]. This distinction has been attributed to incorporation of contributions of slab melting in south of the CTJ, caused by subduction of the Chile ridge [2]. Therefore, the detail information about geochemical variations and spatiotemporal distribution of volcanic activity provides an important key to figuring out the magmatic evolution in the mantle wedge influenced by the active ridge subduction. Here, we newly report twenty-nine whole rock compositions of major and trace elements, including boron, and Sr, Nd and Pb isotope compositions for the Quaternary volcanics from Hudson volcano in SVZ and Lautaro volcano in AVZ near the CTJ. With precise K-Ar age data in [3], we will address their magmatic evolution. Based on the above data of the Hudson and Lautaro volcanoes, we suggest that the magma source in Hudson volcano might be generated by addition of slabderived fluid toward mantle wedge having E-MORB affinity rather than E-MORB-like asthenospheric injection through slab window, although the addition rates were much smaller than those of the other SVZs due to a hotter slab subduction. As for the Lautaro volcano, our data also suggest that the magmatic sources were generated by slab melting triggered by the Chile ridge subduction as pointed out by previous literatures (e.g. [2]). [1] Stern (2004) Rev. Geol. Chile, 31, 161-206. [2] Ramos & Kay (1992) Tectonophys. 205, 261-282 [3] Orihashi et al. (2004) Rev. Geol. Chile 31, 207-224.

A709

Seasonal climate change as revealed by ion microprobe analysis of δ18O in Soreq Cave (Israel) speleothems I.J. ORLAND1, M. BAR-MATTHEWS2, N.T. KITA1, A. AYALON2, A. MATTHEWS3 AND J.W. VALLEY1 1

Univ. of Wisconsin, Madison, WI, 53706, USA (*correspondence: [email protected]) 2 Geological Survey of Israel, Jerusalem, 95501, Israel 3 Hebrew University of Jerusalem, 91904, Israel Soreq Cave, Israel, contains a record of continuous speleothem (cave formation) growth from 185ka to present [1]. The speleothems preserve geochemical signals of climate as they grow; fluctuations in their oxygen isotope (δ18O) composition reflect changes in cave-air temperature, local rainfall characteristics, and possibly CO2 concentrations [2, 3]. The growth rate of Soreq Cave speleothems, however, limits the temporal resolution of conventional sampling methods (0.5mm drill-samples) to a decadal, or longer, timescale. The analytical capabilities of the WiscSIMS CAMECA 1280 ion microprobe, with a spatial resolution of 10µm while measuring δ18O in carbonates at 1σ = ±0.15‰, allow us to analyze sub-annual growth bands in a speleothem. Sample “2-6” is a drip-formed stalagmite (5.5cm radius) composed of low-magnesium calcite. Eight U-series ages indicate growth from ~2200 to 900 years BP. Fluorescent confocal imaging of “2-6” reveals distinct concentric banding that occurs in light/dark couplets. Within single couplets we observe consistently smooth increases of δ18O by as much as 2‰ (∆18O) between the light and dark bands. Modern records of precipitation and cave drip water indicate that these bands may correspond to the annual cycle of wet and dry seasons in the region [4]. Thus, the variability of δ18O across each growth band is caused by mixing of seasonal rainfall in the vadose seepage zone above Soreq Cave. Analysis of δ18O across annual bands reveals: 1) changes in “seasonality” across the sample; 2) a rhythmic signal of climate variation; and 3) maximum values of ∆18O decrease from 2‰ to ~0.5‰ between 2100 and 1500 years BP, corresponding to an estimated decrease in annual rainfall from 1000 to 600mm coincident with the fall of the Roman Empire in the Levant region. [1] Bar-Matthews et al. (2003) GCA 67, 3181-3199. [2] Hendy (1971) GCA 35, 801-824. [3] Mickler et al. (2006) GSA Bull. 118, 65-81. [4] Ayalon et al. (1998) J. Hydro 207, 18-31.

A710 10

Goldschmidt Conference Abstracts 2008

Be dating of erratics at Hell Creek (Coast Mountains, BC, Canada)

Modeling oxygen isotopes in the Mediterranean

M. ORTUÑO1*, S. EVANS2, J.J. CLAGUE3, R., PALLÀS1, A. RODÉS, A. CUMAN4, R. BRAUCHER5 5 AND D. BOURLÈS

A.H. OSBORNE1*, P.J.VALDES2 AND D. VANCE1

1

RISKNAT group, Fac. Dept. Geodinàmica i Geofísica, Univ. de Barcelona, Spain (*correspondence: [email protected]) 2 Dept. of Earth Sciencies, Univ. of Waterloo, ON, Canada 3 Dept. of Earth Sciencies, Simon Fraser Univ., BC, Canada 4 Dipt.Geoscienze, Univ.degli Studi di Padova, Italy 5 CEREGE UMR 6635, Plateau de l’Arbois 13545, Aix en Provence, France Researchers have suggested that the Cordilleran ice sheet decayed by retreat at its margins and by thinning and downwasting in its interior. According to this model, high areas in the interior were deglaciated much earlier than adjacent valleys and plateaux. We tested this idea by sampling and dating erratics at high elevation (~2,200 m asl) in the southern Coast Mountains at Hell Creek. 10Be exposure ages for four erratics are summarized in Fig. 1. Samples 2 and 4 yield an exposure age of 18.9 ± 1.5 10Be ka. These ages are in agreement with 14C ages of proglacial sediments formed at the southern periphery of the Cordilleran ice sheet at the time it began to recede (end of Vashon advance). Samples 1 and 3 are, respectively, older and younger. The older age may result from 10Be inheritance from previous exposure during the Port Moody interstade. Field observations suggest that the age of sample 3 may reflect postdepositional movement of the erratic or degradation of the rock surface by localized weathering. Our results highlight the usefulness of the 10Be method for refining the chronology of deglaciation in western Canada, but also point out the importance of careful sample selection and of dating many samples.

1

Bristol Isotope Group, Department of Earth Sciences, Univ. Bristol, Bristol, UK (*correspondence: [email protected]) ([email protected]) 2 BRIDGE, School of Geographical Sciences, Univ. Bristol, Bristol, UK ([email protected]) Planktonic foraminifera from deep sea cores consistently show excursions towards more negative δ18O values during Plio-Pleistocene Mediterranean anoxic events. Anoxia has thus been linked to enhanced freshwater supply to the surface ocean, often attributed to an increased intensity of the African monsoon or a northward shift of the Intertropical Convergence Zone. Such mechanisms are supported by the precessional pacing of these events. However, anoxia also appears to be associated with greater precipitation over Europe and over the eastern Mediterranean, implying teleconnection between monsoonal climate drivers and those at higher latitudes. Each of the proposed freshwater inputs should produce a different pattern of δ18O in the surface ocean but the dispersal of δ18O in the Mediterranean is poorly understood and at present cannot be used to distinguish between the large-scale regional climate signals. Here we present the preliminary attempts to reproduce modern Mediterranean circulation using NEMO (the Nucleus for European Modeling of the Ocean). We outline the parameterisation of δ18O in the model, and how we will compare model results to observational data for the present day [1, 2]. We also test the response of the model to changes in freshwater input. Further experiments will represent the various scenarios put forward for development of anoxia in the Mediterranean in order to better understand what the sedimentary records are telling us about mid and low latitude climate conditions in the Quaternary. [1] Pierre C. (1999) Marine Geology 153 41-55. [2] Schmidt, Bigg & Rohling E.J. (1999) Global Seawater Oxygen-18 Database http://data.giss.nasa.gov/o18data/

Figure 1: 10Be ages of the four erratics sampled at Hell Creek. Canadian chronoestratigraphic stages in calibrated years are shown for comparison.

Goldschmidt Conference Abstracts 2008

A711

Extra-thick plates: Basis for a single model of mantle magmagenesis, all the way from MORB to kimberlite

Oxygen mass transfer from trapped gas phase and its biogeochemical consumption

MILES F. OSMASTON

S.E. OSWALD1* AND G.U. BALCKE1,2

The White Cottage, Ripley, Woking, Surrey GU23 6JT, UK ([email protected]) Tectonic evidence [1, 2], briefly outlined, that cratons have tectospheric keels that approach 660km, demands an appraisal of how this radical change reflects upon the genesis and significance of mantle-derived magmas, noting that this perspective extends (less deeply) all the way to MORs [2]. We show that without calling on extraneous heat sources magmagenesis by thick-plate splitting yields many rewarding insights. The basic model [3] envisaged a split-induced diapir in a deep, narrow, mantle crack, and needed thick plates to be fully applicable. We present four simple variants of this model, adapted to each of MORB, OIB, CFB and kimberlite. Source compositions are still important but processing is central and (variably) thick plates provide the column-space to do it in, with a varying result. Among the notable features are:(a) Melting in the diapir decreases again as wall cooling asserts control. Enlarged by cumulate intergrowths, the solids form a ‘log-jam’ in the crack (familiar to engineers), and melt is forced through it (primary segregation). So this depth varies with current parameters (wall temperature, splitting rate), the jam providing xenoliths when ruptured. The force to do so and ability to extract melt increases with jam depth (kimberlite). (b) Reduced pressure at the foot of the diapir causes incipient melting of mantle accessories, trace element contents being drawn, and gases diffuse, along melt pathways, resulting in light-isotope enhancement (OIB). MOR continuity promotes self-cancelling of this effect, so a common source is possible. (c) Heated by an eruption, the big volume increase at the gt-sp peridotite phase change in the walls may close the crack, prising it apart elsewhere. In our MOR variant this is the pushapart force. In the IOB case, it may prolong volcanic chains. (d) In the MOR variant, crystal accretion to narrow-crack walls uniquely explains the straightness, segmentation and seismic anisotropy [4], strongly supporting our basic model. [1] Osmaston (2006) Proc. ICAM IV, OCS Study MMS 2006003, p.105-124. Also at, http,//www.mms.gov/alaska/icam. [2] Osmaston (2007) XXIV IUGG, IASPEI JSS 011 Abstr. #2105. http,//www.iugg2007perugia.it/webbook/ [3] Osmaston (2005). GCA 69, (10S) A439. [4] Osmaston (1995) XXI IUGG, Boulder, Colorado, Abstracts p. A472. (N.B.‘c-axes’ should read ‘a-axes’)

1

Department of Hydrogeology, Helmholtz Centre for Environmental Research – UFZ, 04318 Leipzig, Germany (*correspondence: [email protected]) 2 metanomics GmbH, 10589 Berlin, Germany ([email protected]) Contaminated aquifers are often depleted by the biogeochemical consumption of oxygen, and the attenuation of contaminants could be enhanced by additional supply of oxygen. Gas sparging is a remediation technique that supplies addional oxygen to ground water via injection of air or oxygen gas. For the performance of this method the mass transfer of gases from the gas phase trapped after injection is a key process, similar to the processes occurring for trapped air at the capillary fringe. Our study investigated gas sparging in this respect experimentally as well as numerically. We extended an existing numerical model called KBD [1], which describes the mass transfer processes and gas phase development after injection in a kinetic framework. A series of laboratory columns experiments with sandy aquifer material included use of SF6 as a partitioning tracer, measurement of breakthrough curves of dissolved gases and dissolved tracers for a number of oxygen gas pulses. Further experiments established a chemical oxygen consuming reaction and proved that the degradation reaction, which depends on the transfer of oxygen into the aqueous phase, itself influences the fate of the gas phase noticeably. Oxygen transfer into the aqueous phase is slowing down with number of gas pulses and strongly depends on the reverse transfer and accumulation of nitrogen. The composition of the entrapped gas phase, the volumetric changes of the entrapped gas phase and the transport of oxygen through the columns could be reproduced. An analysis extending to a field situation included a range of biological degradation rates and studied injection of pure oxygen versus air. The results give evidence that partitioning tracers and the naturally occurring nitrogen can contribute additional information on the gas dissolution process and thus the transfer of oxygen and the stimulation of biodegradation. [1] Holocher et al. (2003) Environmental Science & Technology 37(7), 1337-1343.

A712

Goldschmidt Conference Abstracts 2008

Anomalous S isotope fractionations during reactions with an organic surface: I. Theoretical investigations

A multi-proxy approach to look at the paleoclimatic variability in the upwelling zone off NW Africa

TSUBASA OTAKE1*, ANTONIO C. LASAGA2, YUMIKO WATANABE1 AND HIROSHI OHMOTO1

CAROLA OTT1*, STEPHAN STEINKE1, REBECCA RENDLE-BUEHRING, HELGE MEGGERS2, BABETTE BOECKEL2, K.-H. BAUMANN2 AND BARBARA DONNER1

1

Penn State Astrobiology Research Center and Department of Geosciences, The Pennsylvania State University, University Park, PA 16802, USA (*correspondence: [email protected]) ([email protected], [email protected]) 2 Geokinetics, State Collge, PA 16803, USA Ab initio calculations were performed to investigate: (1) the validity of mass-dependent approximations [1] during equilibrium isotope fractionations and (2) a new surface reaction model to produce anomalous isotope fractionations. We use the term “anomalous sulfur isotope fractionation” when the ∆33S and 33 θ’ (≈δ33S/ δ34S) values of a sample fall outside of 0±0.2‰ and 0.51±0.01‰, respectively, and/or when the ∆36S and 36θ’ (≈δ36S/ δ34S) values of a sample fall outside of 0±0.4‰ and 1.9±0.1‰, respectively. We have calculated the vibrational frequencies, reduced partition function ratios, and fractionation factors for all four stable isotopes of S in simple gaseous and aqueous compounds at T = 0 − 650°C. Mass-dependent coefficients, (33α − 1)/(34α − 1) and (36α − 1)/(34α − 1) values, converge to 0.515 and 1.89, respectively, at T > 500°C, but deviate from these values with decreasing temperature and depending on the sulfur species pairs. For example, at T = 0°C, (33α − 1)/(34α − 1) values range from 0.505 to 0.517 and (36α − 1)/(34α − 1) values range from 1.88 to 1.96. However, these deviations do not fall in the range of “anomalous isotope fractionation”. On the other hand, we have recognized that the combination of small chemisorption energies (<30 kJ/mole) with possible discontinuities in the number of bound energy levels for different sulfur isotopes may lead to anomalous isotope effects in heterogeneous reactions between a surface and sulfur-bearing species. The magnitude of anomalous fractionation effects during a heterogeneous reaction increases with increasing temperature. We have performed ab initio calculations for SO2 adsorption on a kerogen surface. The results indicate the possibility of creating large anomalous sulfur isotope fractionations (e.g., δ33S/δ34S ≈ 1.08, δ36S/δ34S ≈ 0.84, ∆33S = 7.0 − 13.6‰, ∆36S = -13.0 − -25.2‰) by heterogeneous reactions between organic matter and sulfurbearing solutions under hydrothermal conditions. [1] Bigeleisen and Mayer (1947) J. Chem. Phys. 15, 261-267.

1

MARUM-Research Center Ocean Margins, University Bremen, Postfach 330440, 28334 Bremen, Germany (*correspondence: [email protected]) 2 FB5 Department of Geoscience, University Bremen, Klagenfurther Straße, 28334 Bremen, Germany This study applies a multi-proxy approach, using alkenone- and Mg/Ca-based (from G. bulloides) sea surface temperature (SST), alkane and TOC estimates to reconstruct the paleoceanographic conditions of the upwelling area off NW-Africa for the past ~55 kyrs. Investigations were carried out on two sediment cores (GeoB8507-3 and GeoB9601-3) south of ~20°N in the NE Atlantic. In this region, upwelling is seasonal, occurring during winter and early spring [1]. The alkenone-derived SST record of GeoB9601-3 clearly shows a cooling in surface water during the time intervals of Heinrich Events 1-5. This correlates with the GeoB8507-3 SST record which indicates an increase in temperature during the last glacial compared to modern conditions. However, the SST record based on Mg/Ca ratios shows a cooling. These observed opposing trends in the SST records could be related to the overall varying ecological preferences of the different plankton groups. However, our findings could also reflect seasonal differences in productivity in response to changes in the location and intensity of the upwelling during the last glacial. [1] Wooster, Bakun & McLain (1976) Journal of Marine Research 34, 131-141.

Goldschmidt Conference Abstracts 2008

Tracing magmatic sources of ash beds in the Late Permian to Middle Triassic Nanpanjiang Basin (South China): Insights from Hf isotopes on zircons from volcanic ash beds 1

2

2

M. OVTCHAROVA , H. BUCHER , T. GALFETTI , U. SCHALTEGGER1, A. BRAYARD2, N. GOUDEMAND2 3 AND A. STRACKE

A713

Isotopic evidence for a “beheaded” mantle plume in the Western Mediterranean: A new model for Italian volcanism J.P. OWEN* Department of Geoscience, University of Calgary, 2500 University Drive NW Calgary, Alberta T2N 1N4 Canada (*correspondence: [email protected])

1

Earth Sciences, University of Geneva, Switzerland ([email protected]) 2 Institute and Museum of Paleontology, University of Zürich, Switzerland ([email protected]) 3 Earth Sciences, ETH Zürich, Switzerland The ammonoid-rich rocks from the Early Triassic Luolou Fm. and Anisian Baifeng Fm. (Nanpanjiang Basin, South China) are well calibrated by U-Pb ages on volcanic zircons [1, 2]. Permian and Early Triassic carbonate-dominated rocks, with reduced sedimentation rate in the Early Triassic, are overlain by a thick, prograding turbiditic complex of Anisian age. This transition indicates a major change of subsidence rate, concomitant with a burst in volcanic activity between latest Early Triassic and early Anisian. Few other volcanic ash layers occur throughout late Permian and Triassic. The Hf isotopic composition of dated zircons from volcanic ash beds indicate that the source of magmatic liquids is changing together with the sedimentary regime: (1) volcanic zircons from Late Permian up to Late Smithian ash beds have εHf(t) of +1 to -4 (Hf TDM = 1.4 Ga), typical for a mixed magma source; (2) zircons from an early Spathian volcanic ash layer of the Luolou Fm. (250.55 ± 0.50 Ma) up to the early Anisian transition beds (246.83 ± 0.31 Ma) have uniformly low εHf(t) between -4 and -7 (Hf TDM = 1.7 Ga), suggesting a crustal source of magmas; (3) zircons from a late middle Anisian ash bed of the Baifeng Fm. (244.60 ± 0.36 Ma) have εHf(t) values of +2 to +7 (Hf TDM = 0.9 Ga), indicating a significant participation of juvenile material in the magma generation. This drastic change follows the transition of a mixed carbonate-silicilastic outer platform to a terrigeneous trough with high subsidence rate. The study shows that Hf isotopes in volcanic zircons from stratigraphically well-defined ash beds not only monitor shortlived changes in the tectonic regime of a sedimentary basin, they are also a good tracer for melt transfer from the mantle into the upper crust during orogenic processes. [1] Galfetti et al. (2007) EPSL 258, 593-604. [2] Ovtcharova et al. (2006) EPSL 243, 463-475.

One of the current controversies regarding Cenozoic Italian volcanism is the origin of a geographically widespread, depleted isotopic end-member sampled by a diverse range of mantle-derived magmas. Previous work has shown that this end-member is characterized by low 87Sr/86Sr (< 0.7040), high 143 Nd/144Nd (> 0.5130), and 206Pb/204Pb ratios that are similar to those of volcanic rocks associated with deep-seated mantle upwellings. As a result, several authors have proposed the presence of an asymmetric mantle plume located beneath the Western Mediterranean [1]. However, opponents cite tomographic studies that show the presence of a broad, highvelocity anomaly at about 400-600 km depth, interpreted as subducted oceanic lithosphere [2]. The new model presented here reconciles the plume-like isotopic signatures with the apparent roadblock caused by the accumulation of subducted slabs in the transition zone. In this model, a late Cretaceous deep-seated mantle plume head contaminated the upper mantle in the Western Mediterranean and imparted a distinct, depleted isotopic signature. The plume was then “beheaded” (cut off from its source) during the midTertiary by a build-up of lithospheric material subducted prior to and during the Alpine-Betic and Apennine-Maghrebides orogenies. Hundreds of km of oceanic and thinned continental lithosphere were consumed during closure of the Tethyan Ocean and subsequent opening of the Western Mediterranean. These processes resulted in the isolation of a widespread, homogeneous, fossil plume head trapped in the asthenosphere above a thick layer of subducted slabs. This model may explain the absence of strong thermal anomalies and associated flood basalts in the Western Mediterranean. Support for the model comes from rheological and numerical studies indicating the limited ability of a plume tail to penetrate physical boundary zones (such as subducted slabs) in the mantle [3]. [1] Bell et al. (2004) EOS 85-50, 541-547. [2] Lucente et al. (1999) J. of Geophys. Res. 104(B9), 20307-20327. [3] Davies (1995) EPSL 133, 507-516.

Goldschmidt Conference Abstracts 2008

A714

Where precipitation matters: Be and 26Al-derived hillslope denudation rates in the Atacama Desert, Chile

Toward understanding early Earth evolution: Prescription for approach from terrestrial noble gases and light elements records in lunar soils

J.J. OWEN*1, K. NISHIIZUMI2, R. AMUNDSON1, W. DIETRICH3 AND K. YOO4

M. OZIMA1*, Q.-Z. YIN2, F. PODOSEK3 AND Y. MIURA4

10

1

Dept. of ESPM, Univ. of California, Berkeley, CA 94720, USA ([email protected]) (*correspondence: [email protected]) 2 Space Sciences Lab, Univ. of California, Berkeley, CA 94720, USA ([email protected]) 3 Dept. of EPS, Univ. of California, Berkeley, CA 94720, USA ([email protected]) 4 Dept. of Plant and Soil Sciences, Dept. of Geological Sciences, Univ. of Delaware, Newark, DE 19716, USA ([email protected]) A fundamental goal of geomorphology is to quantitatively understand how landscapes respond to climate. We have quantified denudation rates on gentle, granitic, and semiarid to hyperarid hillslopes in the Atacama Desert, northern Chile, using 10Be and 26Al. Unlike more humid parts of the world, where mean annual precipitation (MAP) seems to exert little control on denudation rate [1], denudation rate increases as a power law function with increasing precipitation, going from 1 m/My at MAP<1 mm/y to 19 m/My at MAP~100 mm/y. This corresponds with a shift in soil formation and transport processes from abiotic (salt shrink-swell) in the hyperarid region to biotic (bioturbation) in the semiarid region. On hillslopes approaching dynamic steady state, we can calculate rates of physical and chemical erosion using a soil mass balance equation, soil chemistry, and the denudation rates above. This has been done in only a few other locations, including Australia [2, 3] and California [3]. Soil flux (Q) on some hillslopes is proportional to both soil thickness (H) and slope (S): Q=-KhHS. Kh is thought to account for differences in climate and intensity of soil transport processes but has remained essentially a black box. Our data lets us begin opening the box. We calculated Kh for each hillslope and compared it to other Kh values available. We observe an increase in Kh of nearly one order of magnitude with increasing MAP in the Atacama (going from abiotic to biotic hillslopes). However, once bioturbation has become the dominant transport mechanism, K values plateau at 0.0050±0.0005 m/y, regardless of MAP. [1] Riebe et al. (2004) EPSL 224, 547-562. [2] Heimsath et al. (2005) Geology 33, 949-952. [3] Yoo et al. (2007) JGR 112, F02013.

1

Graduate School of Earth and Planet. Science, University of Tokyo, Tokyo 113-0033, Japan ([email protected]) 2 Department of Geology, University of California Davis, Davis, CA 95616, USA ([email protected]) 3 Dept. of Earth and Planet. Sciences, Washington University, St. Louis, MO 63130, USA ([email protected]) 4 Earthquake Research Institute, University of Tokyo, 1130032, Tokyo, Japan ([email protected]) Because of the almost total lack of geological record on the Earth’s surface before 4 Ga, the history of the Earth during this period is still enigmatic. Here we describe practical approach to tackle these formidable problems. We propose that examinations of lunar soils for light elements such as He, N, O, Ne, and Ar would shed a new light on this dark period in the Earth history, and resolve some of the most fundamental questions in earth science such as the onset time of the geomagnetic field, the appearance of oxygen atmosphere, and the secular variation of an Earth-Moon dynamical system. Due to a strong dynamic coupling between the Earth and the Moon, theoretical studies have concluded that the spinlocking between the Earth and Moon took place in a few tens of million years (Ma) after the formation of the Earth-Moon system about 4.5 billion years ago, and hence-forth the Earth has been facing only to the nearside of the Moon (e.g. [1]). Also, theories have suggested that due to tidal energy dissipation, the Moon has been receding from the Earth. Recent theoretical studies concluded that the distance between the Earth and the Moon was about a half of the present distance circa 4 Ga ago (e.g. [2, 3]). These suggest the possibility that there has been interaction between the Earth’s atmosphere and the nearside lunar surface in ancient time. Such interaction would further be enhanced [4], if the geomagnetic field were less developed or even absent in ancient times [5]. Therefore, we suggest that close examinations of lunar soils for lighter elements would resolve long standing fundamental problems such as the onset time of the mature geomagnetic field as well as the dynamic evolution of the Earth-Moon system. [1] Murray & Dermott (1999). [2] Abe & Ooe (2001). [3] Bills & Ray (1999). [4] Ozima et al. (2005). [5] Tarduno et al. (2007).