Goldschmidt Abstracts 2008- R

A772

Goldschmidt Conference Abstracts 2008

Chemodynamic study of chlorinated pesticide in acidic and alkaline soils

The role of geochemists in the era of “Peak Everything”

UZAIRA RAFIQUE*AND SAIMA NASREEN

K.V. RAGNARSDOTTIR

Department of Environmental Sciences, Fatima Jinnah Women University, The Mall Rawalpindi, 46000 Pakistan (*correspondence: [email protected]) ([email protected]) Despite being banned atrazine is the most widely used herbicide and is registered in more than seventy countries worldwide. In Pakistan atrazine has been used extensively for broadleaf weed control in corn and sugarcane production as well as in several minor crops. Parveen and Masud (1988) detected some chlorinated insecticide in cattle drinking water from Karachi (Pakistan), while Jabbar et al. (1993) reported monocrotophos, cyhalothrin and endrin in shallow ground water of Faisalabad (Pakistan), the cotton growing area of the country. Atrazine is retained against leaching losses in soils principally by sorption to organic matter, but the mechanism of sorption is not clearly understood. Quantification of soil is needed for predicting the fate of agricultural chemicals in soils as soil characteristics exert dominating influence on chemicaleosorbent interactions. The present study is an attempt to determine the influence of physico-chemical properties of soil on the adsorption of atrazine in three different soils (Agriculturtal, Garden and Barren) with variable organic matter. Such studies have the potential to contribute significantly to a better understanding of post-application herbicide dynamics in agricultural fields. Composite sample of each soil type was analyzed for bulk density, moisture content, color, pH, electrical conductivity, organic matter and pphosphorus content by standardized methods. Batch experiment for pesticide residue analysis was conducted with variable atrazine initial concentration (10ppb, 30ppb, 50ppb), variable pH (4, 7 and 10) and variable contact time till saturation. The distribution of bound atrazine was determined by Ethyl Aetate extraction method. Atrazine remaining in each soil sample after extraction was determined by UV-Visible spectrophotometer at 223nm. Agricultural soil samples showed optimum adsorption, it may be related to high organic content. Sorption and binding of atrazine to SOM can occur as a result of hydrophobic bonding. pH 4 is found optimum for all soil types and showed maximum atrazine adsorption. The present study also showed that till saturation at 120 hr contact time adsorption increases with increasing contact time. Kinetic studies include calculation of Kd and Koc. The results showed that atrazine adsorption followed psedo first order reaction.

Department of Earth Sciences, University of Bristol, Bristol BS8 1RJ, UK ([email protected]) Peak oil will occur in the next five years upon which extraction will no longer meet oil demand and the amount of petroleum available to society will begin to dwindle. At the same time oil prices will continue to increase, affecting the world GDP. Regional gas production rates have declined in for example the North Sea. Coal production will begin to peak and decline in ~15 years. Heinberg [1] has concluded that because fossil fuels supply ~85% of the world’s total energy, peaks in these fuels will result in the word’s energy supply shrinking within a few years. We are also facing ecological dilemmas related to population pressure, habitat destruction and resource depletion. Phosphate will run out by 2020. Other resources that will decline in the next decades are wild fish, per capita freshwater availability, arable land in agriculture production (due to soil degradation), grain production, and yearly extraction of some metals that are crucial for our present day industry. These include uranium, copper, platinum, silver, gold and zinc. Humanity is thus at the end of a historical era of material abundance and cheap energy. We have reached the inception of over-consumption and resource availability. Europeans are living as if we had three Earths – and Americans as if we had five Earths to support us! A major re-think is imperative. For example, agriculture needs to move from being energy intensive (fertiliser production) and pesticide based (made from oil). Sustainability is a core concept for geochemists to engage with and unite forces with society to develop new technologies, based on more common metals and minerals than our present day “stuff” is made out of. Our involvement can lead to solutions that transform industrial products from being made from scarce minerals to abundant minerals, moving towards nature-inspired design principles that can make industry sustainable [2]. The 21st century will pose many resource and energy difficulties to the world population – and geochemists are ideally suited to play an active role in finding solutions to these challenges. [1] Heinberg R. (2007) Peak Everything. Waking up to a Century of Declines. New Society Publishers. [2] McKenna W. & Braungart M. (2002) From Cradle to Cradle. Remaking the Way We Make Things. North Point Press.

Goldschmidt Conference Abstracts 2008

Iron nanoparticulates in icebergs: A source of bioavailable iron R. RAISWELL1*, L.G. BENNING1, L. DAVIDSON1, M. TRANTER2 AND S. TULACYK3 1

Earth and Biosphere Institute, School of Earth and Environment, Univ. of Leeds, Leeds LS2 9JT, UK (*correspondence: [email protected]) ([email protected]) 2 Bristol Glaciology Centree, School of Geographical Sciences, Univ. of Bristol, Bristol BS8 1SS, UK ([email protected]) 3 Dept of Earth and Planetary Sciences, Univ. of California, Santa Cruz, CA 95064, USA ([email protected]) Ice-hosted sediments from Antarctica contain Fe nanoparticulates that are potentially bioavailable [1] and icebergs locally enhance productivity in the Southern Ocean [2]. Here we show that the rate of supply of potentially bioavailable, nanoparticulate Fe from melting icebergs is comparable to the soluble, bioavailable Fe from aeolian dust [5]. Antarctic glaciers and icebergs contain nano-phase schwertmannite, ferrihydrite and goethite. Concentrations (FeA) of nano-phase Fe estimated by ascorbate extractions removed 75-90% of ferrihydrite and 25-75% of nano-goethite with negligible effects on other Fe minerals [3]. Experimental evidence [4] suggests that this colloidal Fe is bioavailable with intracellular uptake rates of 15-20%. Iceberg calving from Antarctica produces 2.5 Tm3 yr-1 containing 0.5 kg m-3 of sediment [1] with a mean FeA = 0.06% (bioavailable Fe flux of 0.04-0.08 Tg yr-1, Table 1). The flux of aeolian dust to the Southern Ocean is 33 Tg yr-1 and 1-10% of the total Fe believed to be soluble [5] and potentially bioavailable (Fe flux of 0.01 to 0.13 Tg yr-1). We conclude that nanoparticulate Fe in ice-hosted sediment is a regionally significant source of potentially bioavailable Fe to the Southern Ocean. Source

Mass Flux Tg yr-1

Fe Content

Bioavailability

Fe Tg yr-1

FeA 0.115-10% =0.06% 0.15 Total Fe 0.01Dust 33 1-10% =3.5% 0.13 Table 1: Bioavailable Fe Fluxes to the Southern Ocean. Ice

1250

[1] Raiswell et al. (2006) GCA 70, 2765-2780. [2] Smith et al. (2007) Science 317, 478-483. [3] Hyacinthe & Van Cappellen (2004) Mar Chem 91, 227-252. [4] Chen et al. (2003) Mar. Chem. 81, 177-189. [5] Jickells et al. (2005) Science 308, 6773.

A773

Iron-nanoparticulates in ice-hosted sediments: A window into the subglacial environment R. RAISWELL*, L.G. BENNING AND L. DAVIDSON Earth and Biosphere Institute, School of Earth and Environment, University of Leeds, Leeds LS2 9JT, UK (*correspondence: [email protected]) ([email protected]) The nature of the subglacial environment is poorly constrained and in situ sampling is difficult and does not necessarily reflect the range of chemical conditions resulting from the localised effects of regulation, microbial activity, and pressure gradients [1]. However sediments enclosed in icebergs and glaciers retain the mineralogical signatures of the subglacial environment [2]. High resolution transmission electron microscopy, selected area electron diffraction and energy dispersive X-ray analysis have identified Fe nanoparticulates in basal ice from glaciers in Antarctica (Taylor and Canada on granite, dolerite and sandstone bedrock) and Svalbard (Monaco on metasediments). All samples contained nanoparticulate iron phases (schwertmannite, 2-line ferrihydrite and goethite). In surface terrestrial environments schwertmannite, ferrihydrite are unstable at ambient temperatures and near neutral pH and transform to geothite or hematite. Both schwertmannite and ferrihydrite can take months to years to fully transform to geothite, although goethite may initially appear after only a few days [3-4]. However, nanoparticulates enclosed in ice will be preserved for much longer by the low temperatures and limited water contact. Schwertmannite has only been identified in acid mine drainage and its formation requires low pH (< 5) and high sulfate concentrations [4]. Its presence in basal ice from the Taylor and Canada Glaciers clearly indicates the existence of water at the base of these cold, polar glaciers. Furthermore the presence of schwertmannite in ice from a variety of bedorck suggests that site-specific factors do not control its formation. Instead it most likely forms where pyrite is oxidised in microenvironments containing thin films of oxygenated water, creating a low pH with rapid precipitation induced as water is removed by freezing. [1] Tranter et al. (2002) Hydrol. Proc. 16, 959993.[2] Raiswell et al. (2006) GCA 70, 2765-2780. [3] Regenspurg et al. (2004) GCA 68 1185–1197. [4] Bingham et al. (1996) GCA 60, 2111-2121.

A774

Goldschmidt Conference Abstracts 2008

Bimodal volcanism within transtensional extension basins developed along craton-arc suture subsequent to continental collision: An example from southern Africa H.M. RAJESH*, J. GUTZMER AND R.H. BAILIE Department of Geology, University of Johannesburg, Johannesburg 2006, South Africa (*correspondence: [email protected]) Bimodal volcanic rocks occurring along trans-tensional basins that developed subsequent to the collision between a mobile belt and a craton offers unique opportunity to examine the magmatic and tectonic processes associated with the collision/accretion. The Koras Group is such a bimodal volcanic suite on the eastern margin of the poly-deformed and highly metamorphosed Proterozoic Namaqua-Natal Belt (NNB) at its contact with the Archaean Kaapvaal Craton in southern Africa. SHRIMP geochronology results indicate that the Koras Group is comprised of two volcanic successions, a lower, older succession of ~1.17 Ga age and an upper, younger succession of ~1.10 Ga age. The successions are comprised of basalt-basaltic andesites and rhyolitic porphyries sandwiched between immature siliciclastic sediments, with minor intermediate volcanic rocks associated with the upper succession. The basalt-basaltic andesites have slightly higher Nb and low La/Yb ratios, indicating that the subduction-like patterns observed in their multielement diagrams could be the result of a lithospheric mantle source signature created by a subduction process which has been imposed as an inherited signature on the basaltic andesites long after subduction has ceased. The rhyolitic porphyries yield a within-plate geochemical signature. The overall coherence of major and trace element trends of the Koras Group rocks, along with their similar and parellel REE patterns, and reasonably well reproduction of whole-rock compositional trends by MELTS modelling favours a single liquid line of descent that connect rhyolitic rocks to mantle-derived basaltic magmas. Fractional crystallisation was accompanied by varying degrees of assimilation of older crustal material, as evident from Sm-Nd isotopic signatures and depleted mantle (TDM) model ages of 2.16-1.68 Ga. TDM model ages of 1.8–2.5 Ga are typical of the Namaqua Province of NNB, suggesting assimilation of crustal material of this province for the lower volcanic succession. The older TDM model ages for the upper volcanic succession suggest assimilation of older crustal material, possibly from the Kaapvaal Craton.

The effects of chemical weathering on bulk chemistry derived from TIR spectral models E.B. RAMPE1, M.D. KRAFT1, A.D. ROGERS2 AND T.G. SHARP1 1

Arizona State University, School of Earth and Space Exploration, P.O. Box 871404, Tempe, AZ 85287-1404, United States 2 Stony Brook University, Department of Geosciences., 255 Earth and Space Sciences Building (ESS), Stony Brook, NY 11792-2100, United States Chemical trends from rocks and soils on the martian surface can elucidate past and present chemical weathering environments. For example, trends derived from remote sensing data from Mars, including TES, Mini-TES, and GRS, indicate most chemical variability occurs in FeO+MgO concentrations, rather than in CaO+K2O+Na2O and Al2O3 concentrations, as seen in terrestrial chemical weathering trends [1]. The trends in martian data are explained with an acidic weathering model, resulting in the dissolution of Feand Mg-rich igneous phases (olivine and pyroxenes) and the precipitation of amorphous silica [1, 2]. Thermal infrared (TIR) spectral models from spectra of two- and three-component physical mineral mixtures containing one or two igneous minerals (plagioclase, pyroxene, olivine) and small to moderate amounts of one secondary silicate (amorphous silica or montmorillonite clay) were used to calculate model-derived bulk chemistries. For mixtures containing silica, the model-derived chemistries show most chemical variability in FeO+MgO concentrations, as is seen in martian data, even though the actual chemistry of the mixtures does not have variability in FeO+MgO. Previous studies have shown the presence of weathering rinds and silica coatings on terrestrial basalts can cause nonlinear mixing in TIR spectral models [3, 4]. From our experiments, we propose that the presence of amorphous silica in the mineral mixtures causes nonlinear mixing in spectral models, causing variations in the modeled abundances of Feand Mg-rich phases, particularly olivine and pyroxenes. We suggest the precipitation of amorphous silica from chemical weathering of mafic surfaces may enhance the chemical signature of acidic weathering in TIR data. [1] Hurowitz J.A. & McLennan S.M. (2007) EPSL 260, 432443. [2] McLennan S.M. (2003) Geology 31, 315-318. [3] Michalski J.R. et al. (2006) EPSL 248, 822-829. [4] Kraft M.D. et al. (2003) GRL 30, 2288.

Goldschmidt Conference Abstracts 2008

Incomplete mixing in the solar nebula: Implications for early Earth M.C. RANEN* AND S.B. JACOBSEN Dept. of Earth and Planetary Sciences, Harvard Univ., Cambridge, MA 02138 (*correspondence: [email protected]) In order to interpret isotopic differences observed between the Earth and meteorites it is imperative to know the initial ratio of the parent nuclide and bulk daughter reference values in both Earth and meteorite parent bodies. Carbonaceous chondrites have been found to have a negative 142Nd signature (averaging 20 ppm) compared to the bulk silicate Earth [1]. One of the hypotheses to explain this is that it is due to early Sm-Nd fractionation, decay from now extinct 146Sm and subsequent burial of an enriched layer that has remained hidden over the past 4.5 Ga. Here we show that a hidden reservoir is not necessary to explain the 20 ppm variation seen. 146 Sm and 144Sm are both p-process nuclides. The cited initial 146Sm/144Sm ratio of 0.008 is not supported by a wide data review. Instead, we see an early Solar System with an average intial 146Sm/144Sm of 0.009 at 4.567 Ga with a large uncertainity of 0.002. If this variation seen is due to incomplete mixing of fresh 146Sm/144Sm it can lead to differences in 142Nd/144Nd of over 1.4ε units between planetary bodies. It has been shown that there are stable 144Sm (p-process) heterogeneities in Carbonaceous Chondrites [2] which may further complicate this issue. Isotopic differences in bulk meteorites have now been found for a variety of elements. We have measured Ba and Nd isotopic compositions in individual CAIs and with other data have found striking correlations between isotopic anomalies. Isotopes which are made in a neutron burst or supernova have positive correlations with each other and negative correlations with s-process isotopes. Three components mixing is also seen in some correlation diagrams between CAIs, OCs, and CCs having a large amount of refractory inclusions. Here Earth does not lie in the mixing space. All other isotopes with anomalies except for 142Nd cannot be due to planetary differenation processes and are nucleosynthetic in nature. Since 142Nd correlates with these isotopes as expected if it was due to incomplete mixing of material from a supernova the 142 Nd anomalies between the Earth and meteorites can also be thought of as nucleosynthetic which negates the need for a hidden reservoir in the mantle. [1] Boyet M. & Carlson R.W. (1995) Science 309, 576-581. [2] Andreasen R. & Sharma M. (2006) Science 314, 806-809.

A775

Isotopic interpretations for abundant carbonates in the Tertiary red clay of the Chinese Loess Plateau W.B. RAO∗, H.B. TAN AND J.S. CHEN Research Academy, Hohai University, Nanjing, 210098, China (∗correspondence: [email protected]) Why carbonate contents are higher in the Tertiary red clay than in the Quaternary paleosols of the Chinese Loess Plateau is still controversial [1, 2]. In order to explore the cause, we determine Sr isotope ratios and Ca contents of carbonates in the Lingtai profile (Table 1) [3]. Period

Layer (n)

Quaternary

Loess (33) Palesol (32) Red clay (9)

Tertiary

87

Sr/86Sr Mean 0.710990 0.711313 0.711517

Ca content Mean 2.6% 0.8% 3.8%

Table 1: Sr isotope ratios and Ca contents of carbonates in the Lingtai profile (n denotes the number of samples). 87

Sr/86Sr ratios and Ca contents of carbonates are higher in the Tertiary red clay than in the Quaternary loess and paleosols, as shown in Table 1. 87 Sr/86Sr ratios of carbonates are a good proxy of chemical weathering in the Chinese Loess Plateau [3]. High 87Sr/86Sr ratios of carbonates thus show strong chemical weathering in the Tertiary red clay. However, high contents of carbonates in the Tertiary red clay can not indicate weak summer monsoon at that period. The Tertiary red clay is finer than the Quaternary loess and paleosols, thus should have more primary carbonates according to the reference [4]. Most primary carbonates could be in situ evolved into secondary carbonates during the development of the Tertiary red clay. Therefore, high contents of carbonates in the Tertiary red clay were mainly a result of primary carbonate redistribution under strong chemical weathering. [1] Chen et al. (2007) Sci. Sini. 50, 392-403. [2] Guo et al. (2001) PPP. 170, 11-26. [3] Yang et al. (2000) PPP 157, 151159. [4] Wang et al. (2005) Atmos. Environ. 39, 2631-2642.

A776

Goldschmidt Conference Abstracts 2008

Experimental studies on rutile solubility in fluids and melts JENNIFER F. RAPP1, STEPHAN KLEMME2 AND SIMON L. HARLEY1 1

School of GeoSciences, University of Edinburgh, West Mains Road, Edinburgh, EH9 3JW, UK 2 Institut für Mineralogie, Universität Münster, Corrensstr. 24, 48149 Münster, Germany Rutile (TiO2) is an important mineral that preferentially sequesters high field strength elements (HFSE). As such, it has been implicated in the observed depletion of HFSE in arc magmas. It has been suggested that rutile is insoluble in slabderived fluids, and remains residual in the subducted slab. Indeed, experimental data indicates a very low solubility of rutile in pure H2O [2], and this low solubility may result in HFSE-depleted fluids imparting a depleted signature to arc magmas. However, as yet there is scant experimental data on rutile solubility in fluids of more complex compositions [1, 3]. Our systematic experimental study into the effect of specific chemical components on rutile solubility in fluids indicates greatly enhanced TiO2 solubility in halogen-rich brines. In H2O mTi=0.008, whereas in F-bearing brines mTi=0.3. The relative mobility of trace elements (i.e. HFSE) in these fluids can be assessed using our run products. These results suggest that HFSE mobility in metamorphic rocks within subduction zones may be more extensive than previously thought. [1] J. C. Ayers & E. B. Watson (1993) Contrib. Mineral. Petrol. 114, 321-330. [2] Tropper, P. & C. E. Manning (2005) Am. Min. 90(2-3), 502-505. [3] Audetat, A. & Keppler, H. (2005) EPSL 232, 393 - 402.

He, Ne and Ar systematics in single vesicles: Mantle isotopic ratios and origin of the air component in basaltic glasses A. RAQUIN, M. MOREIRA AND F. GUILLON Equipe de Geochimie et Cosmochimie, Institut de Physique du Globe de Paris, 4 place Jussieu, F75005 Paris, France ([email protected]) An outstanding problem in understanding the origin of the gaseous phase, particularly the rare gas compositions, in magmatic rocks, is the ubiquitous atmospheric component in bulk rock samples, and whether this atmospheric component is a late stage contamination of the sample, or a recycled component though sediments or altered oceanic crust [1, 2]. Here we address this problem by analyzing single vesicles from the “popping rock 2πD43” sample from the Mid Atlantic Ridge using a UV laser ablation system. We have determined both elemental and isotopic compositions of He, Ne and Ar in single vesicles as well as Kr and Xe abundances. All vesicles analyzed have an isotopic composition identical to the referred degassed mantle value estimated from this same sample, despite analyzing vesicles from a wide size distribution [3, 4].The 4He/3He and 4He/40Ar ratios are respectively 90,000±3000 and 1.41±0.14. The Ne-Ar systematics indicate that the 40Ar/36Ar value of the degassed mantle is 27,000 ± 4000 as it was suggested by Moreira et al. [5] and that 20 Ne/22Ne reflects more likely the B-Ne (12.5) value than the Solar wind value (13.8). The atmospheric component, which is always detected in bulk samples by crushing, was not detected in the single vesicles. This implies that the recycling of atmospheric noble gases in the mantle cannot explain the “air-like” component of this sample. The addition of the atmospheric component must occur either during the eruption, or after sample recovery. [1] Ballentine & Barfod (2000) EPSL 180, 39-48. [2] Sarda (2004) EPSL 228, 49-63. [3] Sarda & Graham (1990) EPSL 97, 268-289. [4] Burnard et al. (1997) Science 276, 568-571. [5] Moreira et al. (1998) Science 279, 1178-1181.

Goldschmidt Conference Abstracts 2008

A777

Application of statistical methods in geochemical anomalies identification, Baidjan area, North of Iran

Mineralogy and geochemistry of the Qoll-e-Anaroo deposit in the East of Iran

IRAJ RASA1, SOMAYE BOKHARAEI2 AND MOHAMMAD HADI3

IRAJ RASA, MOHAMMAD HADI NEZAMPOUR AND MOHAMMAD HADI POURHAJI

1

Earth Sciences Faculty, Shahid Beheshti University, Tehran, Iran ([email protected]) 2 Sciences Faculti, North Tehran Azad University, Tehran, Iran ([email protected]) 3 Earth Sciences Faculty, Shahid Beheshti University, Tehran ([email protected]) The Baidjan exploration area is located in north of Iran that contains few deposits related to igneous rocks witch important source for Cu, W, Sn and locally Au and U. This area contains several MVT base metal deposits, too. In these deposits types, tectonic structures are important. Also, alteration zone are useful for determination the location of igneous and hydrothermal deposits. Remote sensing technology has been a useful tools for recognition these phenomena. Landsat ETM+ data were used to study alteration zones and tectonic structures. Different approaches were used in this project, such as band rationing, principal component and digital filtering. In this area RGB ETM+541 were make better picture and RGB ETM+321 were make worse picture that these are result of spectrometer of geological phenomena. The structures were identified using low passfilter, edge sharping detector filter, Laplacian detector, Sobel edge detector and specially sun angle filter. WE and N45°E are main orientation of these faults. These faults have homogeneous dispersal in this area. To determination alteration zone, we used the classic ETM+5/ETM+7 ratio for identifying clay mineral zone and ETM+5/ETM+2, ETM+1/ETM+4 and ETM+3/ETM+1 ratio for iron minerals and ferric oxides zone. In result pictures, alteration zone were showed by bright pixel zone. Field control showed these area were associated to clay or iron oxide planes that developed downward of trigenious formation and no have related to hydrothermal alteration. Only one area has true hydrothermal alteration. RGB ETM+541 and RGB ETM+735 pictures were used to determination and separation igneous rock unites. Base on the studies mention above and integration result with field studies, one area is favourable for further exploration. This favourable area in stream sediment exploration was a promising point for REEs, too.

Earth Sciences Faculty, Shahid Beheshti University, Tehran, Iran ([email protected], [email protected], [email protected]) Qoll-e-Anaroo mineral occurrence is located in north east of central Iran. This area has few small ancient mines. Paleozoic carbonate rocks hosted mineralization veins. These veins have N55ºE to N65ºE and N10ºW to N30ºW trends which correlated to shear fractures between the major faults. Geochemical analysis showed Zn and Pb are economic elements in ore samples with content of 40% and 9.8% respectively. Cu, Cd and Ag have high positive correlation with Pb and Zn and negative correlation with Ba, Rb, Y, Cs and W. XRD analysis and thin polished sections showed smithsonite and cerussite are major ore in mineralization veins and galena, sphalerite, anglezite, hemimorphite and Malachite have low abundance in these mineralization veins. Dolomite is a major gangue in mineralization veins and Baroque dolomite recognized in many thin sections. Calcite and Fe oxide and hydroxide and Pyrite are other minerals in these veins. Few Chalcopyrite minerals showed in polished sections. According to above text we achieve to these results: A) Mineralization controlled with tectonic structures, B) Pb and Zn are only economical elements, C) In these veins positive correlation of Cd and Ag and negative correlation of Ba with Zn and Pb can be result of low temperature of mineralization flow, D) Galena and sphalrite minerals showed primary mineralization in sulfide phases. Supergene activity has been replaced carbonate minerals to primary minerals. In addition low abundance of chalcopyrite and malachite minerals can be result of low temperature of mineralization flow and E) In addition Baroque dolomites are evidence for low temperature of mineralization flow, too. So, we suggest that this occurrence could be result of low temperature mineralization fluid that makes a simple sulfide mineralization.

A778

Goldschmidt Conference Abstracts 2008

Mineralization properties of Chah Sorb deposit, Central Iran IRAJ RASA, MOHAMMAD HADI NEZAMPOUR AND ALIREZA POURABDOLLAHI Earth Sciences Faculty, Shahid Beheshti University, Tehran, Iran ([email protected], [email protected], [email protected] The Chah Sorb deposit, located in Yazd Province, Centeral Iran, is hosted mainly by Upper Jurassic carbonate rocks and to a lesser extent by Batonian shale units. The mineral assemblage may be an important indicator for ore genesis. In the Chah Sorb deposit, the mineral paragenesis can be divided into two stages of varying importance which are generally separated by primary mineralization or alteration attected on primary minerals. The primary mineralization stage is characterized by simple and vein or veinlet sulphides consisting of galena, sphalerite and less than pyrite, chalcopyrite. Most galena and Sphalerite forms open space filling, euhedral to subhedral aggregates up to several centimetrs in diameter. The second mineralization stage is represented by altered minerals especially lead and zinc carbonate minerals and iron oxide and hydroxide minerals. Alteration accompanying the deposition of lead in the host carbonate consists of dolomitization with baroque dolomites that showed low temperature condition of mineralization fluid. The structural control of mineralization is major fault N05°W/85°NE trend. Mineralization is characterized by persistent lead and zinc up to 34.42% and 14.8% respectively and a consistent association of anomalous silver and cadmium. Copper have low content, about 450 ppm in average, in these deposits. Simple mineralogy and geochemistry properties, limited alteration, present of baroque dolomite, host rock type, mineralization geometry with geometric relation between mineralization and Jurassic shale unite can be evidence of basin brine fluid affected to sulfide mineralization.

Mass balance of carbon cycling and mineral weathering across a semiarid environmental gradient CRAIG RASMUSSEN Department of Soil, Water and Environmental Science, The University of Arizona, 1177 E. Fourth Street P.O. Box 210038, Tucson, Arizona 85721-0038 ([email protected])

Field Setting and Geochemical Mass Balance A geochemical mass balance was coupled with radiocarbon and terrestrial cosmogenic nuclide analyses to constrain rates of carbon cycling and mineral weathering across the Sonoran Desert Environmental Gradient (SDEG), a semiarid node of the Critical Zone Exploration Network. Regolith profiles were sampled from stable, upland positions in each of the ecosystems spanning the SDEG. Total elemental analysis of soil, saprolite and rock were used to calculate a geochemical mass balance for each soil profile [1]. Soil carbon mean residence time and soil production rates were determined by radiocarbon analysis of bulk soil samples and 10 Be content of saprolite layers, respectively [2, 3].

Figure 1: Mass balance of silica (triangle) and organic carbon (circle) relative to climate parameters across the SDEG.

Discussion of Results Regolith data demonstrated significant variation in the relative importance of organic carbon cycling and mineral weathering with climate. X-ray diffraction indicated weathering reactions dominated by transformation of feldspar to kaolinite. Radiocarbon data indicated rapid turnover of organic carbon, whereas 10Be data indicated moderate rates chemical denudation, suggesting feldspar weathering may dominate atmospheric CO2 sequestration in these systems. [1] Anderson et al. (2002) GSA Bulletin 114, 1143-1158. [2] Trumbore et al. (1996) Science 272, 393-396. [3] Riebe et al. (2003) Geochim. Cosmochim. Acta 67, 4411-4427.

Goldschmidt Conference Abstracts 2008

Magmatic fluid compositions relative to fluorine and chlorine in tungstenmineralizing intrusive systems K.L. RASMUSSEN* AND J.K. MORTENSEN University of British Columbia, Vancouver, V6T 1Z4, Canada (*correspondence: [email protected]) A detailed examination of multiple intrusive phases comprising several Cretaceous-age granitoid intrusions in the Selwyn-Mackenzie Mountains, Northwest Territories (NWT), was conducted in order to compare and contrast magmatic fluid conditions in tungsten-mineralizing and non-tungsternmineralizing intrusive systems. The study area contains numerous W-Cu-(Mo-Au-Sb-Pb-Zn) mineral occurrences, and is best known for two world-class [1] tungsten skarn deposits: Cantung and Mactung. The focus of this study was to characterize magmatic fluid compositions in individual intrusions that could form skarn deposits and to use the data to identify specific intrusive phases that may have been responsible for supplying metals and volatiles to the tungsten mineralizing system. Volatile phases (F, Cl) were measured by electron-probe micro-analysis of magmatic and hydrothermal apatite, biotite, muscovite, and hornblende. Magmatic fluid compositions were approximated with relative ratios using HF, HCl, and H2O activities calculated from mineral compositions following the methods of Zhu & Sverjensky [2]. Preliminary data from at least two tungsten mineralizing intrusions in the study area indicate that these systems are particularly enriched in F and depleted in Cl, and that volatile phases become progressively more enriched in F during cooling as the intrusive system evolves to more differentiated phases. However, samples most proximal to tungsten mineralization have relatively depleted F concentrations, and approximations of the log(aHF/aHCl) and log(aHF/aH2O) of the magmatic fluid for these samples are low in comparison to other, more distal intrusive phases. This indicates that a volatile phase was exsolved dominantly from the intrusive phases most proximal to mineralization, and that this exsolution probably occurred relatively early and at high temperatures (e.g., before apatite saturation: 750-900°C). These data suggest that potential economically significant and/or buried tungsten deposits may be targeted by examining the volatile constituents in magmatic minerals from prospective intrusions. Additional analyses and a more detailed examination of the data are underway for samples from the Cantung mine, as well as two barren intrusions in the study area. A complete discussion of the results of this study will be presented. [1] Lentz (1998) Min. Assoc. of Canada Short Course Series 26, 663. [2] Zhu & Sverjensky (1991, 1992) Geochim. Cosmochim. Acta 56, 3435-3467.

A779

Geochemistry of Newania dolomite carbonatite, Rajasthan, India J.S. RAY Physical Research Laboratory, Navrangpura, Ahmedabad 380009, India (*correspondence: [email protected]) Of the twenty major Indian carbonatite complexes, Newania of Rajasthan is the only known “dolomite carbonatite”. The most intriguing feature of Newania is that unlike most carbonatites it is not associated with alkaline silicate rocks. We have carried out field, geochemical and isotopic investigations of this 2.27 Ga old complex [1] with an aim to understand the origin and evolution of the complex, and the generation of carbonate magmas in general. Field evidences suggest that Newania has seen multiple intrusive events. The intrusion of ankerite carbonatites appears to be coeval with the apatite mineralization event. Major and trace element concentrations indicate extensive fractional crystallization of the parental magmas. The trace element and REE patterns are akin to avarge carbonatites, except that the total contents are lower. δ18OSMOW of carbonate minerals shows a very large variation from 5.8 to 35 ‰, wheras δ13CPDB has a smaller spread from -5.6 to 0.9 ‰. In a δ13C vs δ18O covariation diagram most of the dolomite carbonatites plot in the “primary carbonatite” field [2], whereas most of the calcite/ankerite carbonatites have δ18O > 15‰ indicating their altered nature. Multi-component Rayleigh fractionation model [2] suggests that the dolomite carbonatites have fractionally crystallized from a carbonate magma having initial compositions of δ13CPDB = -4.6 and δ18OSMOW = 8.3 ± 1.5 ‰ at a temperature of ~700oC. This implies that the mantle source compositions of the Newania carbonatites would have been: δ13CPDB = -4.6 ‰ and δ18OSMOW = 6.3 ‰ (2 ‰ lower than that of the magma). The (87Sr/86Sr)i (0.701962 – 0.702316) and εNd(t) (1.7 – 7.4) of these carbonatites (calculated for an age of 2.27 Ga) support the field observations that there exists at least two generations of magmatic activity in the complex. The apatite rich (P2O5 > 10%) carbonatites have a higher (87Sr/86Sr)i and a lower εNd(t). The younger ankerite carbonatites (and apatite rich dolomite carbonatites) appears to have tapped an enriched (in LILE) mantle reservoir whereas the older ones are derived from a more primitive mantle. [1] Schleicher et al. (1997) Chem. Geol. 140, 261-273. [2] Ray & Ramesh (2000) Geochim. Cosmochim. Acta 64, 299-306.

A780

Goldschmidt Conference Abstracts 2008

U-Pb geochronology of authigenic xenotimes from Huronian Supergroup, Canada

Polyphase tectonometamorphic history in the upper plate of TransHudson orogen (southern Baffin Is.)

J.S. RAY1*, R.H. RAINBIRD2 AND W.J. DAVIS2

N.M. RAYNER*, M.R. ST-ONGE, R.G. BERMAN, M. SANBORN-BARRIE AND N. WODICKA

1

Physical Research Laboratory, Navrangpura, Ahmedabad 380009, India (*correspondence: [email protected]) 2 Geological Survey of Canada, Ottawa, Ontario K1A 0E8, Canada U-rich diagenetic xenotime overgrowths on zircons can be dated to yield minimum depositional ages of non-fossiliferous siliciclastic sedimentary rocks [1, 2]. The Huronian Supergroup was deposited during a key interval of Earth history marked by increasing oxygenation of the atmosphere, and deposition of three separate glaciogenic deposits (e.g., Gowganda Formation) [3]. The Huronian is poorly constrained between 2.45 Ga and 2.22 Ga. To better constrain its age, the age of the glaciations and to explore a possible link with oxygenation of the atmosphere, diagenetic xenotime was dated from samples of the McKim, Mississagi, Gowganda, Gordon Lake and Bar River formations. The 207Pb-206Pb age of the majority of xenotime growths, irrespective of their stratigraphic position, can be attributed to large-scale basinal fluid-flow events that significantly postdated deposition. Prominent age peaks at 2.2 Ga and ~1.7 Ga, correlate with intrusion of Nippising diabase sills and orogenic activity to the south, respectively. However, two xenotimes (seven analyses) from the Gordon Lake Formation, stratigraphically above the Gowganda glaciogenic deposits, yield an older age of 2376±13 (2σ) Ma. Although there is a remote possibility that these overgrowths are detrital, their irregular habit and sharp edges support an interpretation in favour of authigenic growth. If this date represents the minimum depositional age for the Gordon Lake Formation, the Gowganda glaciation is older than potentially correlative glaciogenic deposits in southern Africa and therefore cannot be related to a single Paleoproterozoic “Snowball Earth” scenario. [1] McNaughton et al. (1999) Science 285, 78-80. [2] Rasmussen (2005) Earth Sci. Rev. 68, 197-243. [3] Young et al. (2001) Sed. Geol. 141-142, 233-254.

Geological Survey of Canada, 601 Booth St., Ottawa, ON K1A 0E8 (*correspondence: [email protected]) A combined petrographic, thermobarometric and geochronological study of protolith and metamorphic mineral ages documents plutonism, deformation and metamorphism over 100 Ma related to the accretionary and collisional history of an upper plate segment of the Trans-Hudson orogen in southern Baffin Island. The polyphase thermal and deformation history includes penetrative fabric development during southwest-directed D1 deformation, and southwest- to south-directed D2 folding. Non-penetrative strain events include thrust imbrication and localized dextral transcurrent shearing (D3), and open crossfolding (D4). Igneous crystallization ages of variably deformed plutonic rocks provide independent evidence that penetrative strain across the southern Baffin area had ended by about 1.83 Ga, after which thrusting, localized shearing and crossfolding became the dominant deformational style. Ages of metamorphism and deformation were determined on SW Baffin Island by in situ SHRIMP analysis of monazite grains with a variety of textural relationships to fabrics and porphyroblasts. The oldest metamorphic event (M1 = 1873±5 Ma) is not clearly linked to deformation fabrics. Samples containing elongate monazites aligned parallel to the regional S2 fabric and within elongate garnets in the same orientation link metamorphism (M2) with regional D2 deformation. These constrain D2/M2 between 1852 ±9 Ma to ca. 1835 Ma. Peak conditions of 675oC-4.8 kbar to 740oC-5.6 kbar were attained on clockwise P-T-t paths consistent with crustal thickening during M2. Post-D2 metamorphic events at ca. 1.83 Ga and at ca. 1.82 Ga likely coincided with plutonism and localized dextral shearing respectively. Monazite ages of ca. 1.77 Ga are inferred to be related to fluid inflitration. This dataset corresponds exceptionally well to the tectonometamorphic history determined by ID-TIMS further east on southern Baffin Island demonstrating along strike-continuity for at least 600 km. The protracted record of deformation and metamorphism in the Trans-Hudson orogen on Baffin Island is similar in size, duration and character to that of the upper plate of other orogenic belts, in particular that of the Himalaya-KarakoramTibetan orogen of SE Asia.

Goldschmidt Conference Abstracts 2008

Molecular-level 14C analysis: A review and forecast

Dimethyl sulfide production in a saline eutrophic lake, Salton Sea, California

C.M. REDDY Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Instititution, Woods Hole, MA 02543 ([email protected])

Molecular-level 14C Analysis in the Past Decade Over the past decade or so, the advent of novel separation methods [1] and low-level 14C analysis (micrograms of C) by accelerator mass spectrometry (AMS) [2] has provided new insights into numerous environmental studies. The most diverse results have focused on the source apportionment of combustion-derived polycyclic aromatic hydrocarbons (PAHs) in aquatic sediments [3, 4], household dust [5], and aerosols [3, 6]. For example, Manolokis et al. [6] estimated that biomass burning contributed ~10 and 50% of the total PAHs in atmospheric samples collected from southern Europe and central Sweden, respectively. Other studies have employed 14C to differenitate natural verus industrial compounds [7] and the uptake of fossil C into biomass [8-9].

Future for Molecular-level 14C Analysis

Two important methods will change the face of 14C analysis. One will be the ability to make continuous-flow 14C analysis. In one possible approach, a gas chromatograph could be interfaced to a combustion reactor that will convert the organic molecules into CO2 and then deliver the latter directly in an AMS [10]; hence, eliminating the time consuming steps currently necessary to isolate pure compounds for AMS analysis. The other approach will be using low-level 14Clabelled molecules for tracer experiments. Such low amounts are too low to be measured by liquid scintillation, but are easily measured via AMS. With such labels, it will be possible to perform experiments at environmentally relevant concentrations. [1] Eglinton et al. (1996) Anal. Chem. 68, 904-912. [2] Pearson et al. (1998) Radiocarbon 40, 61-75. [3] Reddy et al. (2002) Env. Sci. & Technol. 36, 1774-1782. [4] Mandalakis et al. (2004) Env. Sci. & Technol. 38, 53445349. [5] Reddy et al. (2003) Env.For. 4, 191-197. [6] Mandalakis et al. (2005) Env. Sci. & Technol. 39, 29762982. [7] Teuten et al. (2005) Science, 307, 917-920. [8] Wakeham et al. (2006) GCA 70 1761–1771. [9] Slater et al. (2005) Env. Sci. & Technol. 39, 2552-2558. [10] Schneider et al. (2004) Nucl. Instrum. Meth. B. 223, 149154.

A781

BRANDI KIEL REESE1,2 AND MICHAEL A. ANDERSON1 1

Department of Environmental Sciences, University of California, Riverside, CA 92521 ([email protected]) 2 current address: Department of Oceanography, Texas A&M University, College Station TX 77843 (*correspondence: [email protected]) The Salton Sea (33.3°, -115.8°) is regionally recognized for its high and annually increasing salinity (currently 48), warm surface water temperatures (14oC in winter to 32oC in summer), high algal productivity, and high sulfate concentrations (110 to 135 mmol L-1). It is a large (980 km2), shallow (maximum depth of 15 m), discontinuous polymictic lake located in an arid terminal basin in Southern California. The surrounding areas are dominated by irrigated agriculture, which contributes to high nutrient loading into the Salton Sea. Furthermore, the phytoplankton assemblage in the Salton Sea is typical of marine rather than freshwater environments, due to intentional and unintentional introductions. The concentrations and distribution of volatile organic sulfur compounds (VOSCs), were quantified over a 13-month period (May 2006 through June 2007). The concentrations of dimethyl sulfide (DMS) appear to be the highest reported thus far for any natural body of water, with an average surface (0 to 2 m depth) concentration of 2.5 µmol L-1. DMS concentrations as high as 11 µmol L-1 were measured, and the concentrations of DMS strongly correlated with chlorophyll a (R2 = 0.62, n = 265, p < 0.05). Very high concentrations of dimethylsulfoniopropionate (DMSP), an osmolyte in marine algae, were also measured (average total DMSP of 2.4 µmol L1 ), with concentrations strongly correlated with chlorophyll a (R2 = 0.88, n = 36, p < 0.05). The biomass of the Salton Sea is composed mostly of marine phytoplankton species that are high DMSP producers; based on the correlations of DMS, chlorophyll a, and DMSP, it appears that the DMS in the Salton Sea is directly linked to algal biomass through DMSP. As a result of its very high DMS concentrations, the average estimated volatilization at the Salton Sea (480 µmol m-2 d-1) was greater than the estimates of other lakes and the open ocean. We calculate approximately 9.6 x 105 mol of DMS was volatilized off the surface of the Sea during the course of this study.

A782

Goldschmidt Conference Abstracts 2008

Using conductivity to measure Iron reduction rates from flask to field

Thallium isotope constraints on early Cenozoic climate change

A. REGBERG1, K. SINGHA2, M. TIEN3 AND S. BRANTLEY4

M. REHKÄMPER1, R.G.A. BAKER1,2, S.G. NIELSEN3, T.K. HINKLEY4 AND J.P. TOUTAIN5

1

Penn State University 302 Hosler Bldg ([email protected]) 2 Penn State University 311 Deike Bldg ([email protected]) 3 Penn State University 408 Althouse Bldg ([email protected]) 4 Penn State University 2217 EES Bldg ([email protected]) Researchers at several field sites have noticed anomalous electrical conductivity signatures associated with organic contaminant plumes. It has been hypothesized that signals at these sites are at least in part being generated by bacterial dissimilatory metal reduction. The waters in and around organic contaminant plumes have been shown to quickly become anoxic due to bacterial activity. This creates an ideal situation for the use of metal oxides as terminal electron acceptors in biogeochemical redox processes. Since iron oxides are one of the most abundant and readily utilized metal oxides we have focused our work on dissimilatory iron reduction. We present laboratory data that demonstrates the possibility of quantifiably monitoring iron reduction reactions via changes in fluid conductivity. We conducted three separate lab experiments increasing in complexity from abiotic iron reduction with ascorbic acid to in vitro iron reduction using total membrane fractions (TM) from Shewanella oneidensis MR-1 and ending with in vivo iron reduction using whole cells. We varied the concentration of electron donor in order to vary the reaction rate. Abiotic reduction of goethite with 0.01M ascorbic acid produced a 26% increase in conductivity (53 µS/cm) and 69 µM Fe(II) over the course of a five hour experiment. In vitro experiments with TM and 0.01M sodium formate produced a 70 µS/cm increase in just one hour and much larger concentrations of Fe(II) (647 µM). However, much of this iron adsorbed onto the TM and goethite surfaces and did not contribute to changes in conductivity. Whole cell experiments exhibit similar increases in conductivity from background to 50 µS/cm in the first hour. We use stoichiometrically balanced master equations to model concentrations of conductive and non-conductive species as indicated by conductivity measurements. Furthermore, the rates of conductivity increase in each system are used to obtain model reaction rates. The use of conductivity to quantify reaction rate in flask experiments is then applied to reactions occurring in the Cape Cod aquifer system.

1

Department of Earth Science and Engineering, Imperial College, London SW7 2AZ, UK 2 Department of Mineralogy, The Natural History Museum, Cromwell Road, London, SW7 5BD, UK 3 Department of Earth Science, Parks Road, Oxford OX1 4BH 4 US National Ice Core Laboratory, USGS, Denver CO 80225 5 LMTG, Université de Toulouse, CNRS, IRD, OMP, 14 Av. E. Belin, F-31400 Toulouse, France The thallium (Tl) isotope composition of seawater, as inferred from Fe-Mn crusts, exhibits a sharp 6 ε-unit increase between 60 and 40 Ma and has subsequently remained relatively constant [1]. This record is remarkably similar to the seawater sulphur isotope curve for the same period [2]. Possible causes for the observed 5‰ increase in the S isotope composition of seawater include the termination of North Atlantic volcanism in the early Eocene and a substantial increase in the burial of marine pyrite [2, 3]. Thallium is a highly volatile element, such that up to 45% of dissolved Tl in the oceans may be derived from subaerial volcanism [4]. If volcanic Tl furthermore were to be isotopically lighter than other source fluxes, decling volcanic inputs could account for the observed change in the seawater Tl isotope curve at 60 to 40 Ma and provide new constraints on the effect of volcanism on the sulphur isotope composition of seawaster. To explore this hypothesis, we have analysed the Tl isotope compositions of volcanic aerosols from six volcanoes. Quiescent volcanic emissions are highly enriched in Tl but the Tl isotope compositions are generally similar to the mantle, continental crust, and other marine input fluxes [5, 6]. When incorporated into a model that tracks the evolution of oceanic Tl, changes in volcanic activity therefore cannot explain the observed variations in the Tl isotope composition of seawater. A decrease in the overall input of Tl from 60 to 40 Ma can reproduce the isotopic shift but this also requires a large reduction in the proportion of Tl that is deposited in oxic pelagic sediments relative to mass fraction that is removed by low-temperature alteration of ocean crust, which dominates Tl removal from the modern ocean. [1] Rehkämper et al. (2004) EPSL 219, 77. [2] Paytan et al. (1998) Science 282, 1459. [3] Kurtz et al. (2003) Paleoceanography 18, 1090. [4] Rehkämper & Nielsen (2004) Mar Chem 85, 125. [5] Nielsen et al. (2005) GCA 69, 2007 [6] Nielsen et al. (2006) EPSL 251, 120.

Goldschmidt Conference Abstracts 2008

Nanoscale mineralogy of Ag in sulfides from Cu deposits in northern Chile: Implications for ore genesis, exploration, and metallurgical recovery M. REICH1*, S.L. CHRYSSOULIS2 AND C. PALACIOS1, 1

Dept. of Geology, University of Chile, Santiago, Chile (*correspondence: [email protected]) 2 Advanced Mineral Technology Laboratory (AMTEL), London, Ontario, Canada Little is known about the mineralogical form of occurrence of “invisible” precious metals in stratabound (“manto-type”) Cu deposits in northern Chile, where Ag can be enriched up to thousands of ppm. A few studies report bulk analyses of sulfide concentrates revealing that Ag is hosted by primary and secondary Cu and Fe sulfides such as chalcopyrite, bornite, pyrite, chalcocite, digenite, and covellite. However, no comprehensive set of data is available regarding the relative distribution, speciation, solubility, and limits of incorporation of Ag (and other trace metals) in the sulfides. This information is of upmost importance to determine: (a) the saturation state of the parent ore-forming hydrothermal solutions with respect to Ag (b) the ability of different sulfides to extract economically important amounts of precious metals from solution during Cu ore deposition. Preliminary bulk (INAA) data from the Mantos Blancos deposit show that maximum amounts of Ag are present in mineral separates of chalcocite and covellite (1057 and 812 ppm, respectively), followed by bornite (510 ppm) and chalcopyrite (336 ppm). Bulk analyses allow constraining the ranges of concentration in different sulfides, but they do not provide information about the mineralogical form of Ag (solid-solution versus nanoparticulate), the distribution of Ag in the sulfide grains, and the spatial relation between Ag and other trace metals. Therefore, we are analizing the trace metal content of sulfides in situ, using secondary-ionization mass spectrometry (SIMS) along with high-resolution transmission electron microscopy (HRTEM). The speciation of Ag in different sulfides is determined (Ag+, Ag0, Ag2S, (Ag, Au)) as well as its close association with other trace metals such as Au, Sb, Se, and Te. The precise determination of the mineralogical form of precious metals in their sulfide host(s) is a key factor to improve our understanding of the distribution of precious metals in the ore body and its relation with ore-forming processes. In addition, by knowing the mineralogical form and distribution of the incorporated metals, new exploration guidelines and more cost-efficient metallurgical extraction strategies can be developed.

A783

Whole rock and single grain provenance analysis of Early Paleozoic sandstones from northern Peru and Southern Bolivia C.R. REIMANN1*, H. BAHLBURG1, J. BERNDT2 AND E. KOOIJMAN2 1

Geological-Palaentological Institute, University of Muenster, Germany (*correspondence: [email protected]) 2 Institut fuer Mineralogie, University of Muenster, Germany The Ordovician sedimentary basins of the northern central Andes (southern Peru, northern Bolivia) developed in a backarc position, with the arc on the Arequipa Massif in the west and the Amazon craton in the east. This plate-tectonic setting appears to have changed into a passiv margin in the Early Devonian. As there are only very few outcrops of intrusive and extrusive Early Paleozoic rocks, detailed information considering the plate-tectonic evolution and the duration of Early Paleozoic magmatism can mainly be found in the coeval and younger siliciclastic strata. In a provenance study we address these problems by geochemical analysis of Ordovician (Sandia/Amutara Fm.) and Devonian (Cabanillas Gr.) sandstones, exposed across southern Peru and northern Bolivia. A particular emphasis is placed on single grain analysis of tourmalines and rutiles, and U-Pb LA-ICP-MS isotope geochemistry of detrital zircons. Whole-rock REE-patterns in sandstones of both intervals reflect an upper continental crust composition. The major part of the detritus is well recycled. Over 70% of tourmalines and rutils show a chemical composition typical of a metasedimentary origin. Thus, the majority of the detritus has undergone at least one resedimentation event before final deposition. We conclude that the Ordovician and Devonian basins were only to a minor degree influenced by immature detritus, indicative of an active margin plate-tectonic setting. The detrital zircons from the Ordovician sandstones are well rounded to subhedral. C. 50% have a magmatic origin. The U-Pb LA-ICP-MS zircon data show an age distribution with the most distinctive peak between 0.5-0.7Ga (Brasiliano event), and smaller peaks at 1-1.3Ga, 1.4-1.5Ga and 1.5-1.8Ga. The detrital zircons from the Devonian sandstones are mostly subhedral to euhedral and >70% are of magmatic origin. The most prominent peak lies between 0.4-0.5Ga (Famatinian event). These zircons were derived from a western source on the Arequipa Massif, the only place where respective intrusives occur. The oldest (476.93±33Ma) and youngest zircons (406.5±11.4Ma) from this Famatinian peak are narrowing down the duration of Early Paleozoic arc magmatism to the Early Ordovician to very Early Devonian.

A784

Goldschmidt Conference Abstracts 2008

Secondary weathering phases and apatite (U-Th)/He ages

Oxidative weathering and the Fe-S balance of the late Archean ocean

PETER W. REINERS1, STUART N. THOMSON2, BRETT J. TIPPLE2, S. LYNN PEYTON1, JEFFREY M. RAHL3 AND ANDREAS MULCH4

C. REINHARD1, R. RAISWELL2, A. ANBAR3 AND T. LYONS1 1

Geosciences, Univ. of Arizona, Tucson, AZ, USA 2 Geol. & Geophys., Yale Univ., New Haven, CT, USA 3 Geology, Washington & Lee Univ., Lexington, VA, USA 4 Geosciences, Univ. of Hannover, Hannover, Germany

Dept. of Earth Sciences, University of California, Riverside, CA, USA ([email protected]) 2 School of Earth and Environment, University of Leeds, Leeds, LS2 9JT, UK ([email protected]) 3 Dept. of Geological Sciences, Arizona State University, Tempe, AZ, USA ([email protected])

Apatite (U-Th)/He dating provides unique constraints on low-T thermal histories of rocks, but in some settings produces anomalous ages. Possible (nonanalytical) origins of this include complications arising from alpha ejection and implantation or reheating. Here we present observations from several settings suggesting a role for secondary grainboundary phases with high parent nuclides. Detrital apatite from Paleogene clastic sediments that have not been heated above ~30-40 °C from the Sierra Nevada, the Bighorn Basin, and the Pyrenees shows anomalous ages that are often younger than depositional age and inversely correlated with U and Th across large ranges of U-Th concentrations. In several cases, single-grain analyses from individual conglomerate clasts show correlations that are distinct from those of other clasts, suggesting parent nuclide mobility at the cm-scale. Many apatites in these samples are similar to those in soils, showing several features distinct from those in most unweathered crystalline rock. These include secondary clayoxyhydroxide phases on grain exteriors and filling cleavage fractures and c-axis-parallel dissolution pits. Induced fissiontrack distributions indicate that the secondary weathering phases often have high U relative to apatite. Given relatively low U-Th concentrations (~10s of ppm) in most apatite and often high concentrations in oxyhydroxides, late-stage addition of secondary phases may have a significant influence on the parent nuclide budget and therefore age of a grain. If the secondary phases have higher U-Th than the grain on which they precipitate, then regardless of when they form relative to cooling-age closure, measured ages will be “too-young” and U-Th concentrations relatively high. If, however, the secondary phases are not analyzed with the grain (e.g., due to removal during mineral separation), apparent ages will be “too-old” due to He implantation. The U-Th concentration of secondary phases is likely to vary by setting, but may be higher in old (e.g., Precambrian) rocks with radiation-damaged U-Th-rich phases such as zircons. In such cases grain boundary secondary phases may have the potential to implant significant amounts of He into apatite.

Archean ocean chemistry is thought to have been characterized by extremely low levels of both oxygen and sulfate, and as a result Archean seas are presumed to have been rich in dissolved Fe2+. The most striking consequences of this are the presence of abundant marine siderite and the episodic deposition of banded iron formation (BIF), which attests to temporally extended sourcing and transport of dissolved Fe2+ across wide oceanic distances. With the notable exception of a transient pulse of BIF deposition following Neoproterozoic low-latitude glaciation, BIFs are absent from the rock record after ~1.8 Ga. This observation has been interpreted to reflect an increase in the oxidation state of the atmosphere and surface ocean, which resulted in the removal of dissolved Fe2+ via iron (oxyhydr)oxide precipitation and/or FeS2 formation. The latter may have occurred in marine environments supporting microbial sulfate reduction or inorganically during the venting of hydrothermal fluids charged with oceanic sulfur. We present a case study focusing on iron mineral speciation in the late Archean Mt. McRae Shale and the underlying Mt. Sylvia Formation (Hamersley Basin, Western Australia). Data for the Mt. Sylvia Formation and lower Mt. McRae Shale show sulfur limitation of pyrite formation and are consistent with reducing, iron-rich aqueous chemistry. However, data for the upper Mt. McRae Shale indicate ironlimited pyrite formation during deposition, and primary authigenic enrichments in molybdenum suggest free sulfide concentrations in excess of 10 µM. Both observations imply oxidative weathering and transport of SO42- and MoO42-, although a photolytic source for SO42- cannot be ruled out and the presence of small but significant non-mass-dependent sulfur isotope anomalies indicates very low oxygen levels. More broadly, the data speak to the possibility of reactive iron limitation and vacillations between the dominance of iron and sulfur in Archean oceanic environments as a function of secular or sporadic changes in the input and output fluxes of both elements to/from the ocean.

1

Goldschmidt Conference Abstracts 2008

A785

Radionuclides, organic complexing agents and solid phases: Exploring their interactions

Lithological and flow control of metals in a mountainous river basin, NE Portugal

ESTELA REINOSO-MASET*, PAUL J. WORSFOLD AND MIRANDA J. KEITH-ROACH

A.R. REIS, A. PARKER2 AND A. ALENCOÃO1

BEACh Group, SEOES, University of Plymouth, Plymouth, PL4 8AA, UK (*correspondence: [email protected]) Characterising the geochemistry and speciation of major contaminant radionuclides is crucial in order to understand their behaviour and migration in complex environmental systems. Organic complexing agents used in nuclear decontamination have been found to enhance migration of radionuclides at contaminated sites [1]; however, the mechanisms of the interactions in complex environments are poorly understood [2]. In this work, radionuclide speciation was explored alongside sorption behaviour, to identify ternary interactions between radionuclides, organic complexing agents and sand as a representative environmental solid phase. Four key radionuclides encompassing different oxidation states (Cs and Sr as important fission products; Th and U(VI) as representative actinides) and three anthropogenic organic complexing agents with different denticities (EDTA, NTA and picolinic acid) were selected for this project. Electrospray ionisation–mass spectrometry (ESI–MS) was used to characterise the speciation of radionuclide complexes over the pH range 4 to 10. Interactions with sand were examined at circumneutral pH using inductively coupled plasma–mass spectrometry (ICP–MS) and dissolved organic carbon analysis. The results showed that Sr, Th and U interacted with all three ligands and complexation affected their sorption to sand, even in the presence of a seven-fold excess of natural DOC. With EDTA and NTA the solution phase species were identified as: [SrH2EDTA] and [SrNTA]–; [ThEDTA] and [ThNTA2]2–; and [(UO2)2EDTA] and [UO2NTA]–. These neutral and anionic species interacted with the sand to a greater extent than the uncomplexed radionuclides, thus complexation lowers radionuclide solubility. With picolinate, the cationic [Cs2Pic]+ and [Sr2Pic3]+ species were observed by ESI–MS. The increased solubility of Cs was consistent with this species; however, picolinate did not impact on the sorption of Sr. These results will be explored in the context of sorption kinetics, emphasizing the mechanisms by which the organic co-contaminants affect radionuclide mobility in the environment. [1] Means, J.L. et al. (1978) Science 200, 1477-1481. [2] Keith-Roach, M.J. (2008, in press) Sci. Tot. Environ.

1

Department of Geology, University of Trás-os-Montes e Alto Douro, Vila Real, Portugal, and Centre for Geophysics, University of Coimbra, Coimbra, Portugal ([email protected], [email protected]) 2 Department of Soil Science, University of Reading, Reading, UK ([email protected]) The distribution of Cd, Cr, Ni, Cu, Pb and Zn was investigated in fluvial waters and bottom sediments (<63µm fraction) in two different zones of the Corgo mountainous rural hydrographic basin (NE Portugal). These two zones correspond to relatively large valleys, originating from accumulation of Quaternary alluvial deposits on top of Cambrian rocks, in the Campeã Valley (Zone I), and on top of Hercynian granites, in the Vila Pouca de Aguiar Valley (Zone II). Both are agricultural valleys with low population density (Zone I-33 inhabitants/km2; Zone II-146 inhabitants/km2). Farmers mainly use manures, and in smaller quantities, phosphate fertilizers in agriculture. Two sampling campaigns were carried out: a) at the end of the wet period (WP) (May), in high-flow conditions; b) at the end of the dry period (DP) (October), in low-flow conditions. In sediment samples the potentially-available fraction and the residual fraction were analysed by a 3-step chemical sequential extraction procedure and aqua regia digestion, respectively. The results show that in the sediment fractions Pb, Cu and Zn are similar in both zones; Cr and Ni are higher in the schist zone; and Cd is higher in the granite zone. In fluvial water, Cu and Zn are slightly higher in the DP in schists and slightly higher in the WP in granites; Pb occurs only in the DP and almost only in schists; Cd occurs almost only in the WP, with higher contents in schists; Cr and Ni are high in the DP in schists, and in the WP are slightly higher in granites. The main observations resulting from the combination of data on dissolved and particulate metal contents point towards a general diluting effect for metals, except for Cd, in high flow conditions in the schist zone. A contribution of groundwater flow to summer river discharge, associated with a probable slower water flow in soils and sediments from this zone, resulting in an increase in pore water mineralization, should be considered. The higher levels of metals in water from the WP in the granite zone suggest an increase in weathering and transport in winter, with a contribution from finer sediments in suspension to the transport of associated metals. For Cd this behaviour is more noticeable.

A786

Goldschmidt Conference Abstracts 2008

Re/Os fractionation during generation and evolution of hydrocarbons

The partitioning of molybdenum between aqueous liquid and vapour at high temperatures and pressures

L. REISBERG1, R. MICHELS2 AND Y. HAUTEVELLE2

K.U. REMPEL*, A.E. WILLIAMS-JONES AND A.A. MIGDISOV

1

CRPG CNRS UPR2300, BP20, 54501 Vandoeuvre-lesNancy, France, [email protected] 2 UMR CNRS 7566-G2R, Faculté des Sciences, Campus V. Grignard, BP236, 54501 Vandoeuvre-les-Nancy, France, [email protected] Recent results [1] suggest that the Re-Os isotopic system can be used to date the emplacement of petroleum deposits. However, to fully exploit this potentially exciting new application of the Re-Os geochronometer, it is essential to identify the physical processes that fractionate the parent(Re)/daughter(Os) ratio in hydrocarbon systems. We are currently examining two processes that could potentially cause Re/Os fractionation: 1) fractional precipitation of the asphaltene component of petroleum; and 2) black shale maturation and hydrocarbon expulsion. To investigate the first process, we have separated the various asphaltene fractions according to their density and their molecular size. We find a rapid, linear decrease in Re concentration with decreasing asphaltene density. Natural precipitation of asphaltenes could thus provide a means of fractionating the Re/Os ratio in petroleum, if Os concentrations do not follow exactly the same distribution pattern observed for Re. To investigate the second process, we have performed artificial maturation experiments on the well-known Woodford shale source rock of the western United States. We find that Re and Os are both retained nearly completely by the experimental residue. This explains the earlier observation [2] that hydrocarbon maturation does not affect the depositional age determined by Re-Os analysis of black shale source rocks. On the other hand, the Re and Os concentrations of the extracted petroleum fluids are more than 150 times lower than those of the black shale residue. Thus mass balance considerations suggest that the Re/Os ratio of the extracted fluid could differ greatly from that of the black shale parent rock, despite the fact that the Re/Os ratio of the residue is essentially unchanged. [1] Selby & Creaser (2005) Science 308, 1293-1295. [2] Creaser et al. (2002) GCA 66, 3441-3452.

McGill Univeristy, 3450 University St., Montreal QC, Canada (*correspondence: [email protected]) In the study of porphyry Mo ore-forming systems, there is increasing debate over whether the fluid responsible for the dissolution and transport of Mo is a liquid, a vapour, or both. In order to help resolve this issue, it is necessary to have information on how Mo will tend to distribute between these two phases. With this in mind, we have conducted an experimental study designed to determine the partition coefficients of Mo between aqueous liquid and vapour at 300 to 370°C and saturated vapour pressure. The apparent distribution coefficient, log Kd, increases with increasing temperature and pressure, but decreases with increasing Mo concentration of the liquid phase. For experiments conducted with a starting solution of 100 ppm Mo, vapour-phase Mo concentration exceeds that of the liquid at temperatures above about 360°C. The concentration of Mo in aqueous vapour at any given temperature is independent of the Mo concentration of the liquid, indicating that the dominant gaseous Mo species is MoO3•nH2O. Thermodynamic modeling indicates that bimolybdate (HMoO4-) is the dominant species in the liquid at the conditions considered, and thus the solubility of Mo in this phase was dependent on pH in our experiments. In order to determine the true partitioning of Mo between vapour and liquid, we calculated Henry’s constants from the fugacity of MoO3•nH2O and the molality of pH-independent molybdic acid (H2MoO4). These are 2.10, 21.05 and 116.09 bar kg mol-1 at 300, 325 and 350°C, respectively. Our results indicate that in the two-phase, aqueous liquid and vapour systems common to porphyry Mo ore-forming environments, both phases are likely to be important in the transport and deposition of Mo.

Goldschmidt Conference Abstracts 2008

The effects of paleoclimatic change on sediment fluxes in the vicinity of the Gao-Ping submarine canyon, Taiwan REBECCA RENDLE-BUEHRING1*, STEPHAN STEINKE1, JAMES T. LIU2 AND HUI-LING LIN2 1

MARUM-Research Center Ocean Margins, University Bremen, Postfach 330440, 28334 Bremen, Germany (*correspondence: [email protected]) 2 Institute of Marine Geology and Chemistry, National Sun Yat-sen Iniversity, Kaohsiung 804-24, Taiwan, ROC This multi-proxy study aims to look at the effects of late Holocene paleoclimatic variations in monsoonal intensity (humid vs. dry periods), on sediment mass transfer between the Gao-ping shelf and the open ocean regime of the South China Sea (SCS) in order to identify significant climatecoupled terrigenous sediment export and depositional events. For this, two pelagic, undisturbed, long cores (ORI 799G24 and ORI 732-8G) and short box-cores (ORI 811-K8, 811K31) are analysed. Results show that the sediments consist of homogenous olive grey silty clay with little bioturbation and low carbonate concentrations (<5%) indicating a predominantly terrestrial origin of the sediment constituents. Initial chemical analyses, using XRF-spectrometry, of terrigenous elements from core ORI 732-8G display significant variations over the sedimentary sequence studied. The Rb/Sr ratios, which are used as an indicator of chemical weathering, reveal distinct trends which may be associated to short-term changes in chemical weathering conditions due to periods of strengthened or weaker summer monsoonal intensity. AMS-14C dating from planktonic foraminifera constrain the ages of the sediments. Stable oxygen isotope, mineralogical, micropaleontological, Mg/Ca and alkenone analyses form future work in progress.

A787

Thermochronology of the Bushveld Complex: Rapid cooling confirmed P.R. RENNE1,2*, R. MUNDIL1, W.S. CASSATA2, J.M. FEINBERG3 AND R.K.W. MERKLE4 1

Berkeley Geochronology Center (*correspondence: [email protected]) 2 Dept. Earth & Planetary Science, U.C. Berkeley 3 Dept. Geology & Geophysics, Univ. Minnesota 4 Dept. Geology, Univ. Pretoria The thermal history of the Bushveld Igneous Complex (BIC) of South Africa, the world’s largest known mafic intrusion, has received little attention. Thermal modelling suggests emplacement of the mafic portion of the BIC in a strikingly brief interval of <75 ka [1], consistent with the BIC being the intrusive component of a continental flood basalt province. Replicate 40Ar/39Ar plateau ages have been obtained for single crystals or crystal fragments of hornblende (hb), biotite (bi) and plagioclase (pl) from 8 rock units of the Rustenburg Layered Suite (RLS) and associated veins, and a granite from the Lebowa Granite Suite (LGS). Hb and bi plateau ages from the RLS are indistinguishable, and U/Pbconsistent constants [2] yield ages indistinguishable from a Pb/Pb age of 2059 Ma determined for titanite from a xenolith [3], implying a cooling rate >1000°C/Ma over the range ~650300 °C. Multiple crystal fragments of one RLS pl sample yields variably discordant age spectra with plateaux ranging from ~2060 Ma to <2000 Ma, consistent with partial degassing by the nearby ~1.3 Ga Pilanesberg Complex. Closure temperatures ca. 305-340 °C [4] were determined using the approach pioneered by Berger and York [5], and we infer that the range in plateau ages reflects variable subsampling of larger scale 40Ar/K gradients imposed by the partial degasssing event. Limited 40 Ar/39Ar data from RLS veins suggests possibly slightly lower cooling rates. A U/Pb zircon Concordia age of 2057 ±3 Ma from an LGS granite establishes the LGS as an integral component of the BIC which must be accounted for in mass balance calculations of “missing” (i.e., inferred to have erupted) magma [e.g. 1]. In comparison with the U/Pb zircon age, a weighted mean (2055 ±1 Ma) of 8 40Ar/39Ar hb plateau ages suggests much slower cooling (~100 °C/Ma) for the LGS than the RLS, probably indicative of elevated ambient temperatures over the time span of BIC intrusion. [1] Cawthorne & Walraven (1998) J. Petrol. 39, 1669-1687. [2] Mundil et al. (2006) Eos Abstr. AGU 57, V21A-0543 [3] Buick et al. (2001) J. Geol. Soc. Lond. 158, 3-6. [4] Cassata et al. (2008) GCA, this volume [5] Berger & York (1981) GCA 45, 795-811.

Goldschmidt Conference Abstracts 2008

A788

Gold deposition onto arsenian pyrite 1

1

2

Vegetation and climate variability in the Galapagos Islands over the last 2000 years: A high-resolution pollen record

1

D. RENOCK , A. DEDITIUS , M. REICH , S.E. KESLER , R.C. EWING1 AND U. BECKER1 1

Dept. of Geological Sciences, University of Michigan, Ann Arbor, MI., USA ([email protected]) 2 Dept. of Geological Sciences, University of Chile, Santiago, Chile ([email protected])

ALEJANDRA RESTREPO1, MARK BUSH2, JESSICA CONROY3 AND JONATHAN OVERPECK 1

Arsenian pyrite is widely known for its dual role as an important host for “invisible” gold in ore deposits, and a major source of arsenic and heavy metal environmental pollution. Despite recent studies to characterize the structural position of nanoscopic gold [1] within arsenian pyrite, little is known about the atomic scale mechanisms that are responsible for uptake of Au and/or the associated release of As from the mineral. Solution deposition of Au onto fine-grained arsenian pyrite surfaces has been investigated using in situ atomic force microscopy and X-ray photoelectron spectroscopy. The spatial distribution of Au correlates with As-enriched domains on the surface, similar to the positive correlations between Au and As content in pyrite grains [2]. Ab initio quantum mechanical methods were employed to describe possible reaction paths for the adsorption and reduction of Au3+ on arsenian pyrite surfaces, and for Au+1 substitution for (Fe-S)+1 vacancies within pyrite and arsenian pyrite structures. Au incorporation energies were shown to be >1 eV more favorable for Au substitution into arsenian pyrite versus pure pyrite, depending on the As-Au atomic distance. These results can be used to improve the understanding of how surface reactions contribute to the formation of precious metal ore deposits and control iron sulfide surface reactivity. [1] Reich et al. (2005) Geochimica et Cosmochimica Acta 69, 2781-2796. [2] Fleet, M. E. & Mumin, A. H., (1997) American Mineralogist 82, 182-193.

([email protected]) ([email protected]) 3 ([email protected]) 2

A fossil pollen record from El Junco Crater Lake on the island of San Cristobal within the Galapagos Islands, provides the first continuous high-resolution terrestrial paleoclimate record from within the El Nino-3 region. The record provides a c. 3-8 year resolution record of vegetation change around the crater. The limited flora of the Galapagos allows many pollen and spore types to be identified to genus or species level, e.g. Scalesia, Darwiniothamnus, Alternanthera, Acalypha, Bursera graveolens, and Cyathea wetherbyana. Some taxa such as Poaceae and Psidium show responses highly characteristic of human landuse, and may be taken as indicators of human activity. Also evident are pollen that have undoubtedly blown in from the Andes, e.g. Alnus, Podocarpus, and Hedyosmum; an influx that provides an index of trade-wind activity. Major shifts in the fossil pollen record correspond to the onset (c. 1340 AD) of the Little Ice Age and also a dry period at the peak of that event (c. 1610-1690 AD). A turning point toward warmer and wetter climates is evident at c. 1800 in both the fossil pollen and diatom data sets recovered from these sediments. Strong changes associated with land clearance are evident from 1930 onwards. An intensification of warm, wet conditions is suggested by the expansion of Bursera graveolens after 1970. Wavelet analyses and time series analyses of pollen representation relative to sea-surface temperature are used to develop a chronology of ENSO variability since mediaeval times.

Goldschmidt Conference Abstracts 2008

Linking erosion rates and climatic variations in the Gulf of Lions, France: A geochemical approach S. REVILLON1,2, S. BERNE1, G. BAYON1, B. DENNIELOU1 2 AND C. HEMOND 1

IFREMER Géosciences Marine, 29280, Plouzané, France ([email protected]) 2 IUEM Domaines océaniques, 29280, Plouzané, France As a part of an integrated source-to-sink approach, we use geochemical tools to characterize the source of detrital sediments on continental margins and assess how variations in the sediment provenance during the past have been connected to global climate changes. We chose the deltaic margin of the Gulf of Lions because high sedimentation rates in this area allow accessing a high-resolution record of climate variability in Western Europe during the Late Quaternary period. During European project PROMESS 1, Hole PRGL1-4 was drilled on the continental slope of the Gulf of Lions in the Mediterranean Sea. ~ 300m of sediment cores were recovered covering the last five climatic cycles. Such a deep penetration allowed reconstructing climatic variations during the last 500 ka. Here, we report geochemical results for sediment samples from Hole PRGL1-4 focussing on the last climatic cycle. Trace elements, Sr and Nd isotopic compositions were analysed on both silt and clay fractions. Mainly Sr and Nd isotopic compositions vary through time indicating a change in the composition of sediments deposited in the Gulf of Lions and a change in the sediment sources. This variability can be related to climatic forcing and, in particular correlate well with the imprint of so-called Heinrich events. During these cold episodes, 87Sr/86Sr ratios increase and εNd decrease toward more radiogenic values. These variations can be interpreted in term of changes in the contribution of detrital material from distinct sources. Our results suggest either, an increase of the Saharan particles contribution or a decrease in the sediment flux from the Rhône River during the same period or a combination of the two processes. These also suggest an increase in the aridity over the African continent in combination with a decrease in the erosion rates in the Rhone River catchments area.

A789

Rheology of serpentines, seismicity and mass transfer in subduction zone B. REYNARD1, N. HILAIRET1, I. DANIEL AND Y. WANG2 1

Laboratoire des Sciences de la Terre, CNRS UMR 5570, Ecole normale superieure de Lyon, Universite Claude Bernard Lyon 1, 46 allee d'Italie 69364, Lyon cedex 07, 69364, France 2 Center for Advanced Radiation Sources, The University of Chicago, 5640 S. Ellis Ave., Chicago, IL 60637, United States Serpentinites have a lower density and lower viscosity than "dry" ultramafic rocks and it was proposed, based on numerical simulations, that they play a major role in mantleslab decoupling, and in downward (sink) or upward (exhumation) motion of eclogites and ultra-high pressure (UHP) rocks in subduction zones. Rheological data on antigorite, the stable variety of serpentine in subduction zones, are obtained over a P-T range of 1-4 GPa and 200-500°C that cover most of its stability field. The experiments were carried out in a D-DIA apparatus installed at GSECARS on the 13BM-D line of APS. The determined stress-strain curves were fitted to a power-law equation including both temperature and pressure dependence. The results confirm that serpentinites acts as a weak layer that allows significant mass transfer along the "serpentinized channel" and dynamic processes such as mantle slab decoupling, and mantle wedge convection. Regardless of the temperature, serpentinized mantle at the slab surface has a low viscosity that allows localizing the deformation and impeding stress build-up. It will limit the downdip propagation of large earthquakes, and allow viscous relaxation as an origin of post-seismic deformations and slow earthquakes. The low viscosity of serpentinized faults in the oceanic lithosphere makes them possible sites for subduction initiation. Models of growth and transport of a serpentinized channel using available kinetic and present rheological data explain high exhumation rates of eclogites and limited thickness of the channel at great depths (≥ 50 km), and slower exhumation at in a thick hydrated mantle corner at shallower depths.

Goldschmidt Conference Abstracts 2008

A790

Stable Calcium isotope ratios 44/42 (δ Ca) in bones and teeth for the detection of dairying by ancient humans 1,2

L.M. REYNARD , R.E.M. HEDGES G.M. HENDERSON2

1

AND

1

Research Laboratory for Archaeology and the History of Art, University of Oxford, Dyson Perrins Building, South Parks Road, Oxford, United Kingdom, OX1 3QY ([email protected], [email protected]) 2 Department of Earth Sciences, University of Oxford, Parks Road, Oxford, OX1 3PR ([email protected]) The transition to agriculture and pastoralism from hunting and gathering is considered one of the pivotal points in human development, leading to profound changes in society, economy, nutrition, and ultimately our modern way of life. Geochemical techniques, particularly stable isotope analysis, can be used to provide information not available from traditional archaeological techniques, and offer the possibility of greater insight into the timing and breadth of the adoption of new foods and their relationship to the “Neolithic revolution”. Dairy products offer advantages of more efficient land use, improved nutrition, and more reliable and constant access to protein, so that understanding the adoption of dairy and its timing is key to developing a fuller understanding of the Neolithic. We have measured stable calcium isotope ratios (δ44/42Ca) of bones and teeth for the direct detection of dairy consumption by prehistoric humans. Dairy products have lower δ44/42Ca than other dietary calcium inputs [1], and this results in lower δ44/42Ca of the dairy consumer. We have measured the δ44/42Ca of human and animal bones from a range of archaeological sites by MC-ICP-MS with standard bracketing. Results from pre- and post-agricultural times at the key Near Eastern site of Abu Hureyra, Syria (11,100 –7,300 BP) show a δ44/42Ca signal attributable to dairy consumption by ancient humans, with a changing pattern through time. Work on intra- and inter-tooth δ44/42Ca variability is in progress as this material is expected to form a robust archive of in vivo isotope ratios. [1] Chu N.-C., Henderson G.M., Belshaw N.S., & Hedges R.E.M. (2006) Appl. Geoch. 21, 1656-1667.

Optimization of a 46Ca-43Ca doublespike to study radiogenic and stable isotope variations B.C. REYNOLDS* AND B. BOURDON ETH Zurich, Institute of Isotope Geology and Mineral Resources, Switzerland (*correspondence: [email protected]) While Ca isotopes are relatively heavy compared with traditional stable isotopes such as H, C and O, studies have clearly demonstrated the existence of mass-dependent fractionation between stable calcium isotopes (e.g. 40, 42 or 44). The usual notation used for reporting Ca isotope composition is the δ44Ca notation which is defined as variations of the 44Ca/40Ca ratio of a sample compared to a standard in part per thousand (‰). However, the β-decay of 40 K produces 40Ca (half life = 1.28 Ga and a branching ratio of 89% [1]) that can lead to easily detectable 40Ca isotope anomaly in biotites and granitic rocks [e.g. 2]. In order to separate radiogenic anomalies from stable isotope variations, the stable isotopes variations must be determined without using 40Ca, and isotope variations need to be expressed relative to 42Ca. Here we present the optimization of a 46Ca-43Ca doublespike which can be used to determine both the stable and radiogenic isotope variations. This spike is distinct from earlier work [3, 4]. We strive for an improved analytical precision available through the carefully selected double-spike and utilization of the TRITON thermal ionization mass spectrometer (TIMS). It was found that previous spike-sample mixtures, whilst minimizing error propagation of the spike ratios, had small intersection angles of the natural and measured fractionation lines. This angle was optimized in the choice of available spike compositions to provide more orthogonal geometries at low spike-sample ratios (θ > 60º). This optimization is required to separate the instrumental fractionation from estimates of the unspiked natural fractionation factor, and thus minimise error propagations. The application of the double spike technique is also applicable to high-resolution MC-ICPMS for the determination of stable Ca isotope variations. Initial results documenting its applicablity and comparing TIMS and MCICPMS results will be shown. [1] Steiger & Jaeger (1977) EPSL 36, 359-362. [2] Marshall & DePaolo (1989) GCA 46, 2537-2545. [3] Fletcher et al. (1997) Int. J. Mass Spectrom. Ion Proc. 163, 1-17. [4] Gopalan et al. (2007) Int. J. Mass Spectrom. Ion Proc. 248, 9-16.

Goldschmidt Conference Abstracts 2008

Azores and MAR basalts similarity, a multidimensional analysis L.P.RIBEIRO, F. PIMENTEL, M.P.ABREU AND THE EMEPC TEAM EMEPC, R. Costa Pinto 165, 2770-047 Paço D’Arcos Portugal ([email protected]) Basalts from the MAR near the Azores attest a 1000 km long asymmetric geochemical anomaly that begins near the Kurchatov FZ, maxes out near the Azores and diminish gradually southward. Basalts possess enriched composition in trace element and Sr isotopes relative to N-MORB, and is of general agreement, that the regional geochemical signature results from the interaction between the Azores Magmatic Anomaly (AMA) and MAR [4, 5, 6]. Principal Component Analysis (PCA) has been carried successfully by [3], on isotopes from MAR basalts (22º-35ºN), identifying the presence of E-MORB near 35ºN and N-MORB elsewhere. Also, [2] PCA coupled trace elements ratios from OIB and MORB with mantle end-member composition. Following [2], PCA (using software from [1]) was applied to eight incompatible trace element ratios (X/Th) to basalt from Azores region in a multidimensional diagram configured for OIB end-members (HIMU, EMI, EMII and Hawaii) and NMORB from North Atlantic. The testing samples with average basaltic composition (MgO> 7%) are from: the Azores Islands, W of the MAR (ODP Leg82) and MAR at 38.7º and 39.7ºN [7]. The eight variables are reduced to two main principal components (F1 and F2) describing 91% of the variance of the data. F1 bears a maximum positive variance for N-MORB and a minimum negative variance for HIMU. F2 accounts for element ratios variance, ranking variables according with elements incompatibility. Results evidence that the Azores islands plot in the range of HIMU, EMI and EMII end-members, strengthening the existence of source heterogeneities. Also inter-island compositional variability agrees with the existence of heterogeneities beneath the islands, although geographically spaced islands show compositional similarity, e.g. São Miguel-Faial-Flores. Finally, except for ODP 33ºN N-MORB’s, MAR and ODP EMORB’s PCA evidences a long living (>30 Ma) regional geochemical anomaly that could have resulted from similar interaction as the nowadays between AMA and MAR. [1] PCA http,//biomonitor.ist.utl.pt/ajsousa/Andad.html Andad program [2] Allègre et al. (1995) EPSL 129, 1-12. [3] Debaille et al. (2006) EPSL 241, 844-862. [4] Dosso et al. (1999) EPSL 170, 269-286. [5] Gente et al. (2003) G3 4, 1-23. [6] Schilling (1975) EPSL 25, 103-115. [7] Georoc+PetDB.

A791

Numerical modeling of two-phase flow in geodynamics: State of the art, benchmark and perspectives GUILLAUME RICHARD* AND HARRO SCHMELING Earth Sciences Institute, J.W. Goethe Univ., 60438, Frankfurt/Main, Germany (*correspondence: [email protected]) The dynamics of two-phase (or two-component) media is a well-studied field with numerous natural applications to solid earth physics. Some important geophysical problems falling within the scope of the two-phase (or multi-phase) approach are for example: Magma segregation and dynamics, Earth’s Core formation, tectonic plate generation, water transport at subduction zone, etc. For some years, in addition to the commonly used set of equations derived by McKenzie [1], a new theoretical formulation [2] is available to describe two-phase flow mixtures. In parallel, improvements of computers efficiency have made numerical solving of such a complex system more and more tractable. The combination of these two facts result in the existence of a large number of numerical codes not only using different numerical techniques to solve for the equations governing two-phase flow dynamics but also solving for different sets of governing equations. For the sake of clarity, a benchmark of available codes has been launched. We have designed a set of 1-D and 2-D experiments that allowed to compare the efficiency (computer time) and accuracy of numerical schemes and the possibilities offered by different theoretical approaches. These experiments have been chosen to be general enough to permit the comparison of codes initially build to deal with very a large range of settings. Among all two-phase geodynamical systems, the percolation of geophysical fluids at subduction zone is particularly complicated to model. This complexity comes from the richness and variety of processes taking place in subducting factories (melting, dehydration, diapirism, dykes, phase transition, dip angle, trench migration, etc.) and incidentally suggests paths to follow. A review of principal numerical approaches will be presented with special emphasis on the field constrains required by geodynamicists in their quest towards the big picture of subduction zones. [1] McKenzie D. (1984) J. Petrol. 25, 713-765. [2] Bercovici et al. (2001) J. Geophys. Res. 106(B5), 8887-8906.

A792

Goldschmidt Conference Abstracts 2008

Reconciling 14C timescales for marine isotope stage 3

Structural geology and geometry of subducting and subducted slabs

D.A. RICHARDS1, J.W. BECK2, D.L. HOFFMANN1, P.L. SMART1, J.S. SINGARAYER1, A.J. RIDGWELL1 AND P.J. VALDES1

S.W. RICHARDS

1

University of Bristol, School of Geographical Sciences, Bristol, BS8 1SS, UK ([email protected]) 2 University of Arizona, Tucson, AZ 87521, USA Radiocarbon is used to constrain the chronology of a vast array of paleoclimate records, geomagnetic/solar modulation of cosmogenic nuclide production, and dynamics of the past carbon cycle. Carbon-14 timescales have an effective range of > 50 ka, but for the period > 26 ka there is currently a range of alternative records. Here, we reconcile some of the differences between existing archives using a suite of MC-ICPMS U-Th and AMS 14C ages for a new speleothem record from the Bahamas. This represents an attempt to duplicate another stalagmite record (GB89-24-1) from the same cave [1], which attracted much attention because it revealed surprisingly elevated and variable ∆14C (> 1000‰) from 45 to 40 ka. These features are not recorded in recent records from the marine realm (e.g. Cariaco Basin [2]) Adoption of an improved low-blank 14C protocol revealed procedural blank artefacts that significantly influenced the older section of GB-89-24-1. The new record, GB-89-25-3, exhibits elevated ∆14C, up to 600‰ with abrupt 200‰ shifts, for most of the period 45 to 28 ka, but the large peak at ~44 ka is not present. The record is broadly similar to that of Cariaco Basin [2], however, there are some significant differences in timing and amplitude, particularly between 28 and 32 ka. While we recognise that both records can be influenced by varying reservoir effects, much of the difference can be reconciled by adjusting the Cariaco record to the ice core chronology, GICC05 [3]. The ∆14C pattern revealed by the Bahamas and Cariaco records corresponds well with the geomagnetic intensity record GLOPIS-75 [4]. While the latter dominates the temporal pattern of 14C production, ∆14C is also influenced by carbon cycle dynamics. Hence, we explore the nature of ocean-circulation and air-sea exchange during the last glacial period by comparing the ∆14C estimated from the geological record with that from a suite of transient runs using GENIE, a fast, intermediate-complexity Earth-system model. [1] Beck et al. (2001) Science 292, 2453-2458. [2] Hughen et al. (2006) Quat. Sci Rev. 25, 3216-3227. [3] Andersen et al, Clim. Past 3, 1235-1260. [4] Laj et al. (2004) in Timescales of the Paleomagnetic Field, AGU. p255–265.

Research School of Earth Sciences, Australian National University, ACT, Australia ([email protected]) Characteristics of subducting plates such as slab age, structure and angle of subduction are key to understanding the geological evolution of the overriding plate. Recent advances in computer technologies have allowed us to interpret and visualize the geometry of subducting slabs with new precision. Results show that slabs are not simple uniform sheets of crust and lithosphere that penetrate the Earth’s asthenosphere; rather, they exhibit extremely complex geometries including faults, tears, folds and boudinage. The shape and structure of any given slab or slab segment can be used to help interpret the evolution of subduction zones and associated orogens over time. The 3D geometry and 4D evolution of: (1) the subducted Nazca slab below South America, (2) the subducted NW IndoAustralian slab below SE Asia and (3) remnant subducted slab(s) below India and the Pamirs are presented here to bring to attention the complexity in slab geometry and how this geometry can be used to explain aspects of the geological evolution of each region including changes in the geochemical evolution of arc volcanics. Specific features that will be presented include a recumbent, isoclinal fold in the Nazca slab some 400 to 600 k below the south central Andes with its lower, overturned limb resting near the 660 km discontinuity. This folding can be explained by changes in the relative motion between the Nazca and South American plates at around 120 Ma ago. The formation of vertical and horizontal tears, folds and boudinage of the Indo-Australian plate will also be presented.

Goldschmidt Conference Abstracts 2008

A793

Detecting biosignatures associated with minerals by Geomatrix-Assisted Laser Desorption/Ionization Fourier Transform Mass Spectrometry (GALDI-FTMS)

Geochemistry and geochronology of the Grenvillian granitic pegmatites of the Fort-Coulonge area, Québec: Complexity of the origin of the magma

C. DOC RICHARDSON1, J. MICHELLE KOTLER1, NANCY W. HINMAN1, TIMOTHY R. MCJUNKIN2 3 AND JILL R. SCOTT

CAROLINE RICHER* AND DAVID R. LENTZ

1

Geosciences Department, University of Montana, Missoula, MT 59812 2 Industrial Technology, Idaho National Laboratory, Idaho Falls, ID 83415 3 Chemical Sciences, Idaho National Laboratory, Idaho Falls, ID 83415 The ability to detect carbon signatures that can be linked to complex, possibly biogenic, organic molecules is imperative in research into the origin and distribution of life in our solar system particularly when used in conjunction with inorganic, mineralogical, and isotopic signatures that provide strong evidence for geochemical influences of living organisms on their environment. Ideally, the method used to detect these signatures must (i) accurately and automatically translate the organic and other information into usable forms, (ii) precisely distinguish such information from alternative compositions, (iii) operate with high spatial resolution coupled with precise location abilities, and (iv) require little to no sample preparation because of the potential for contamination. Geomatrix-assisted laser desorption/ionization (GALDI) in conjunction with a Fourier transform mass spectrometer (FTMS) has been used to determine the presence of bio/organic molecules (BOM) associated with different minerals and mineraloids including oxide, sulfate, carbonate, chloride, and silicate minerals. BOM is defined as an organic structure that can be produced by living organisms or derived from another organic compound made by living organisms (i.e., degradation product). GALDI requires no sample preparation because the mineral matrix assists desorption. Ultimately, however, the detectability of BOM is controlled by the desorption efficiency, ionization efficiency, and the specific experimental conditions. Results from experiments with combinations of known BOM and mineral standards indicated that the detectability of BOM increased with decreasing concentration, contrary to most analytical procedures. Results suggest that BOM when combined with certain minerals is more easily detected than when combined with other minerals. Such conclusions can guide selection of appropriate samples for sample return missions.

Department of Geology, University of New Brunswick, PO Box 4400, Fredericton, New Brunswick, CANADA, E3B 5A3 (*correspondence: [email protected]) The Fort-Coulonge area is located in the Central Metasedimentary Belt (CMB) in the western part of the Grenville Province about 100km north-west of Ottawa, Canada. Two types of late orogenic granitic pegmatites are found in this area, intruding ductily deformed and high grade metamorphic rocks formed during the Proterozoic Grenvillian orogeny: (1) pink magnetite-bearing, locally zoned and foliated, aplitic to pegmatite (2 to 5 cm); (2) white unzoned, aplitic to pegmatite (2 to 5 cm) characterized by graphic and granophyric texture, white feldspars (K-feldspar and plagioclase), 1-5% biotite, with accessory molybdenite and tourmaline. The multiphase pegmatitic intrusions were generated at depth due to anatexis associated with late highgrade metamorphism during the Grenvillian Orogeny. The pegmatites of the western part of the Grenville previously dated at 937 to 980 Ma using whole-rock isochrons and at 1010 to 1060 Ma using U-Pb zircons which are 30 to 80 Ma younger than the peak metamorphism at this level of the Ottawan phase of the Grenville Orogeny. Both pegmatites show some characteristics of NYF-type pegmatite and A-type granite such as low CaO, MgO, Sc, Cr and high Zr. All pegmatite samples have high silica and alkalis contents, with SiO2 ranging from 63-80 wt.% and total K2O+Na2O varying from 5.1-10.2 wt.%. They also have Al2O3/(CaO-Na2O-K2O)>1 , which indicates that they are peraluminous. The white pegmatite has lower Sr values (155-254 ppm) compared to the pink pegmatite (199-659 ppm). Thorium is frequently more abundant than uranium and Th/U ratio varies from 0.77 to 25.2, but is usually closer to 2 to 4. Generally indicators of fractionation, Rb/Sr and TiO2/Zr ratio do not correlate with Th and (or) U values. These pegmatites fall in the anomalous granite field using RbBa-Sr systematics. High Ba content in both pegmatites may be related to a small degree (less than 15%) of melting of a hybridized lower crustal protolith. The magma shows important crustal contamination characterized by high Rb and Th contents and by relatively enriched Ce and Sm relative to their adjacent elements. Importance of feldspar fractionation in the evolution of the magma is shown plotting Y/NB versus Ba/La with a trend parallel to the alkali feldspar fractionation.

Goldschmidt Conference Abstracts 2008

A794

Water and magma: New insights into an old problem PASCAL RICHET Physique des Minéraux et des Magmas, IPGP, 4 place Jussieu, 75005 Paris, France Although not for the right reasons, the fundamental importance of water in volcanic processes was already recognized in Antiquity. At the end of the 18th century, the role played by water solubility was correctly perceived and the problem of the source of magma raised. Even though much progress has of course been made since then, it can be said that much of the work currently done is still addressing questions posed two centuries ago. For example, water solubility has long been studied thanks to the possibility of quenching hydrous melts as glasses. By contrast, the influence of water on melt properties used to be much less well known because of the great difficulties of making measurements at the high pressures at which water solubility is significant. The slowness of water exsolution close to the glass transition makes it in fact possible to perform at room pressure accurate physical measurements on water-bearing supercooled liquids, a major advantage being that only hydration has to be performed at high pressure. This procedure has allowed a wealth of information to be gathered during the past decade and, thus, to complement the limited high-pressure information available. Much effort has been devoted to viscosity, a property for which the depressing effects of water clearly increase with the degree of polymerization of the anhydrous liquid. Besides, the fragile character of melts increase with water content and can be accounted for with the configurational entropy theory of relaxation processes. Contrasting with this strong influence, the heat of mixing between water and aluminosilicate melts is almost ideal. As for volume properties, water exerts a moderate influence on thermal expansion whereas its effects on compressibility are strong and markedly different for felsic and mafic magmas. The effects of water on physical properties are thus complex. They may give rise to unexpected contrasts between silica-rich and silica-poor melts, for instance in terms of rheology or rate of ascent.

A tale of two liquids: Kinetic isotope fractionation in water and in silicate melts F.M. RICHTER Department of the Geophysical Sciences, The University of Chicago, Chicago, IL 60637, USA (*correspondence: [email protected])

Isotope Fractionation by Chemical Diffusion Recent experimental results [1-3] regarding kinetic isotope fractionations by chemical diffusion in water and in molten silicates will be reviewed. One might have thought that the isotopic fractionations would be larger at the low temperatures of liquid water than in the high temperatures of molten silicates. But this is not the case. As a general rule, isotopic fractionations by chemical diffusion in molten silicates are much larger than in water.

Isotope Fractionation by Evaporation Evaporation of water or a silicate melt can result in isotopic fractionation. Recent results [4] of the isotopic fractionation of silicon and magnesium of evaporation residues relevant to the interpretation of the thermal history Ca- Al-rich inclusions in primitive meteorites will be reviewed.

Isotope Fractionation by Thermal (Soret) Diffusion I will present new laboratory data documenting that a temperature difference of only 100°C across molten basalt fractionates the isotopes of all the major elements (except of course aluminum) by many per mil.

Molecular Calculations I will illustrate the potential of molecular calculations [5] to explain what we observe during the diffusion of ionic species in water. [1] Richter et al. (2003) Geochim. Cosmochim. Acta 67, 39053923. [2] Richter et al. (2006) Geochim. Cosmochim. Acta 70, 277-289. [3] Richter et al. (2008) Geochim. Cosmochim. Acta 72, 206-220. [4] Richter et al. (2007) Geochim. Cosmochim. Acta 71, 5544-5564. [5] Bourg & Sposito (2007) Geochim. Cosmochim. Acta 71, 5583-5589.

Goldschmidt Conference Abstracts 2008

Perturbing Phytoplankton: A polar view on anthropogenic change

New constraints on the pyrolitic model under lower mantle conditions A. RICOLLEAU1*, Y. FEI1, E. COTTRELL2, H. WATSON1, L. ZHANG1, G. FIQUET3, A. AUZENDE3, M. ROSKOSZ4, G. MORARD5 6 AND V. PRAKAPENKA

R.E.M. RICKABY*, K. HENDRY, J. YOUNG AND P. HALLORAN Department of Earth Sciences, University of Oxford, Parks Road, Oxford, OX1 3PR (*correspondence: [email protected]) The input of carbon dioxide to the atmosphere from fossil fuel burning since the Industrial revolution represents a perturbation to the ocean-atmosphere system of geological proportions. In order to predict the future response of phytoplankton to abrupt change, it is possible to impose a perturbation to a laboratory culture or mesocosm experiment or to look to the geological record for signatures of biological response to analogous past periods of major carbon release. Alternatively, we have investigated rapidly accumulating sediments which hold the potential to resolve the phytoplankton response, already underway, to the anthropogenic perturbation of the last two centuries. We shall present high resolution geochemical and microfossil records from a North Atlantic open ocean and a coastal site on the Antarctic peninsula, which indicate that the phytoplankton communities are already starting to respond to the current environmental change. The Antarctic peninsula has experienced rapid warming with associated retreat of 87% of marine and tidewater glacier fronts. Our evidence from diatom opal Zn/Si and diatom assemblages indicate that the recent accelerated input of freshwater from retreating glaciers has resulted in greater diatom productivity, likely as a result of enhanced stratification of the water column. Increasing diatom production and export in these coastal waters, which are thought to contribute up to 50% of total Southern Ocean productivity, could represent a signficant additional sink of carbon and a negative feedback to carbon release. Between 1780 and the modern day, in the high latitude North Atlantic, whilst the coccolithophore assemblage remained invariant, the average mass of calcite produced by the dominant species of coccolithophore (Coccolithus pelagicus) increased by up to 40%. The net impact of such an increase in calcification on pCO2, will depend on the extent of response in other oceans, and on ballasting. Nonetheless, a global increase in calcification by coccolithophores could represent a positive feedback to the anthropogenic emissions. In light of this field evidence, we will explore how the contrasting carbon metabolism of diatoms and coccolithophores, reminiscent of their evolutionary histories, may determine their response and feedback to elevated carbon in the ocean and atmosphere.

A795

1

Geophysical Laboratory, Carnegie Institution of Washington, USA (*correspondence: [email protected]) 2 Smithsonian Institution, National Museum of Natural History, Department of Mineral Sciences, USA 3 Institut de Physique du Globe de Paris, IMPMC, France 4 Laboratoire de Structure et Proprietes de l'Etat Solide, Lille, France 5 European Synchrotron Radiation Facility, Grenoble, France 6 Center for Advanced Radiation Sources, University of Chicago, USA Composition of the Earth’s lower mantle remains an essential question in Earth sciences. The composition of the lower mantle is generally modelled by comparing mineral physics data and seismic observations, using measured properties of the lower mantle end-member phases such as perovskite and ferropericlase. In this study, in situ X-ray diffraction patterns were collected from 30 GPa to 110 GPa and from 1400 K to 2500 K using KLB-1 peridotite as starting material. The experiments were performed in laser-heated diamond-anvil cell at the Advanced Photon Source and at the European Synchrotron Radiation Facility. The sample was loaded in neon pressure medium. We report molar volume changes as a function of pressure and temperature for the three lower mantle phases, i.e., magnesium-iron silicate perovskite, ferropericlase and calcium silicate perovskite. We notice the high spin to low spin transition of iron in the ferropericlase at room temperature. The equations of state (EOS) of the two perovskites and ferropericlase are determined using Au as internal pressure standard. We used the Fe/Mg partitioning values between Mgperovskite and ferropericlase from the literature. Using the volume data and chemical composition information, we calculate the densities of peridotite along geotherm of the Earth. Our calculated density profile allows us to make direct comparison with the seismic observations. The partitioning coefficient does not affect the trend of the density profile in any significant way. The result shows that the pyrolite density profile cannot match the seismic density profile of the entire lower mantle.

A796

Goldschmidt Conference Abstracts 2008

Nanocrystalline anatase-electrolyte solution interface: A surface structural interpretation with the MUSIC and CD Models MOIRA K. RIDLEY1 AND MICHAEL L. MACHESKY2 1

Texas Tech University, Lubbock, TX ([email protected]) Illinois State Water Survey, Champaign, IL ([email protected])

2

Mineral surface reactivity and pH-dependent charging in aqueous media is related to oxide surface structure and variations in the coordination of surface functional groups at mineral surfaces. Relatively little is known about how interface reactions affect the chemical and physical properties of nanoparticles. To develop a fundamental understanding of the surface reactivity of nanosized particles controlled studies with monodispersed crystalline anatase (TiO2) particles were performed. Bulk surface titrations were completed with commercially available anatase particles ranging in size from 200 to 3 nm diameter. The titrations were performed in NaCl media at ionic strengths from 0.005 to 0.3 molality. The focus of this contribution will be our efforts to rationalize the bulk surface titration data utilizing surface complexation modeling (SCM) that incorporates all available molecular information (e.g., Ti-O bond lengths), with crystallographic information acquired from abberation corrected electron microscopy (ACEM). ACEM imaging revealed that the [100] face predominates on the 20 nm particles, and edges/facets include [010], [001], and [0-11]. A MUSIC model based description of surface protonation permitted rationalization of the experimentally observed decrease in pHznpc values with increasing particle size. The surface charging curves were fit with a charge distribution (CD) model and a three-plane description of the Stern layer. Moreover, the titration data suggested an innersphere species for the so-called “inert” Na electrolyte cations. From the SCM, a bidentate innersphere Na-surface species predominated, with the spatial charge distributed between the anatase surface (0plane) and the 1-plane. Conversely, the best-fit for the Cl- ions was a monodentate outersphere complex. As conventional, the outersphere complex was treated as a point charge at the Stern layer (2-plane). The capacitance value of the Stern layer increased with increasing particle size, equating to a narrower compact layer; thus, reproducing the experimental data which showed an enhancement in positive surface charge for the larger nanoparticles (>20nm).

Biogeochemistry of advective tidal flat porewaters THOMAS RIEDEL1, MELANIE BECK1, KARSTEN LETTMANN2 AND HANS J. BRUMSACK1 1

Microbiogeochemistry, ICBM, Carl-von-Ossietzky University, Oldenburg, Germany ([email protected]) 2 Physical Oceanography, ICBM, Carl-von-Ossietzky University, Oldenburg, Germany ([email protected]) Microbial life is abundant in marine sediments. Its activity has an impact on pore water chemistry and – on geological timescales - on sediment diagenesis. In open ocean sediments the limited supply of organic carbon and/or nutrients gives rise to low microbial activities at depth. Our studies on NW German tidal flat sediments have shown that advective systems are, in contrast, far more active than the diffusion controlled marine sediments. For our studies we permanently installed 5 m long pore water samplers along a transect from the top of the tidal flat towards the low water line to monitor changes in pore water chemistry in space and time. During high tide the sediment at the top of the tidal flat is recharged with seawater and therefore with organic matter and electron acceptors such as oxygen, iron, manganese or sulfate. During the following low tide a deep pore water flow is established at the tidal flat margin. Flowing towards the low water line, the pore waters are strongly altered. Within about 30 to 40 m of flow path the pore water profiles show changing redox states from oxygenated to sulfate reducing. Close to the low water line, where pore water seeps are occurring, even methanogenesis can be detected. These seeps contribute to the trace metal and nutrient budget of the open water column. The quantification of this contribution is a major goal of this study and will help to understand trace metal and nutrient cycling in coastal systems affected by tides. A modeling approach coupled with geophysical data shows that the residence time of pore waters in the tidal flat ranges between several days to a few weeks. This makes tidal flats an easily accessible small scale study site for the deep biosphere, which in this case is encountered at shallow depths, in contrast to sub-seafloor settings.

Goldschmidt Conference Abstracts 2008

Processing of the smallest Wild 2 particles in hot silica aerogel FRANS J.M. RIETMEIJER Department of Earth and Planetary Sciences, MSC 03 2040, 1University of New Mexico, Albuquerque, NM 871310001, USA The comet 81P/Wild 2 agglomerates impacted the underdense silica aerogel tiles of the Stardust collector at 6.1 km/s and, when decelerating to rest velocity, each one carved its own track that is diagnostic of the weakly constructed nature of the particle mixture. In these unique bulbous tracks most nanometer-scale particles that numerically dominated many of the agglomerates were forced into the wall of the track that was expanding into hot aerogel. The entrained particles were flash-heated under reducing conditions to at least ~2000oC.

Observations Mostly 10-20 micron sized clumps of melted aerogel extracted along the tracks have a vesicular glass matrix that almost uniformally contains up to several weight percent MgO, Al2O3, K2O, CaO, and MnO, in various combinations, and ditto trace amounts of P, Cr and Ti. Chemically pure silica glass is rare. The glassy matrix also contains Mg-rich olivine and pyroxene single-crystals up to ~500nm that show no signs of melting or interactions with the silica-rich glass. Rare, ~100nm ‘FeS’ (pyrrhotite?) fragments are present. The characteristic feature is the ubiquitous ‘shotgun’ distribution of numerous mostly Fe-S(±Ni) compound, and metallic iron and Fe,Ni-compoud inclusions. They range from a few nm to ~150 nm in size. The smallest inclusions are chemically homogenous but larger ones have a low-S (~10-15 el%) Fe-S core and high-S (~ 30-35 el%) Fe-S rim. Nickel is often absent or <2 el%. High Ni, up to 10 el%, contents are present.

Discussion and Conclusions The observations support pervasive melting of comet nanoparticles with a chondritic bulk compostion. The closest analog to these particles are Mg-Fe-Si principal components, Mg,Fe-silicates and pyrrhotite grains, all up to <500nm, in chondritic aggregate IDPs. Chemially, the IDPs have a “Mg,Fe silicate (+ other elements)” and a “sulfide” reservoir. When melted during impact the fomer reservoir in comet particles was fully assimilated in the pure silica melt wherein “sulfide” melts were immiscible. Upon quenching they yielded numerous inclusions. Under the appropriate timetemperature conditions the large Fe-S melt droplets separated into a core and mantle with Fe-S compositions defined by metastable eutectic compositions in the Fe-S phase diagram, including ‘mackinawite’. Much is information is still present in the controlled chaos that is preserved in the glass matrix.

A797

Mantle flux at lower crust / upper mantle interface: A carbon and hydrogen isotopes study of granulites T. RIGAUDIER1*, V. GARDIEN1, C. LECUYER1, L. SIMON2 1 AND F. MARTINEAU 1

UMR 5125 PEPS, Univ. Lyon1, 69622, Villeurbanne, France (*correspondence: [email protected]) 2 UMR 5023 Laboratoire d’Ecologie des Hydrosytèmes Fluviaux, Univ. Lyon1, 69622, Villeurbanne, France Stable isotopes analysis of C and H of biphasic primary and secondary FI in granulites facies rocks formed in collision and subduction zones dated from Archaean (3,3Ga) to Miocene (16Ma) were performed. The amount of CO2 extracted from primary FI (≈635 ppm) confirms that the formation of granulites is associated with a flux of CO2 at the base of the crust. The δ13C values around -7‰ PDB for the CO2 together with the δD values of H2O (average value -78‰) present in the primary FI indicate a mantle origin for the fluid involved in the formation of granulites during peak metamorphism. The scattered δ13C values (between -2.5 and 8.9‰) of primary FI from all granulites correspond to the mixing of a "juvenile" carbon source from mantle-derived CO2 (δ13C=-5±2‰) and a superficial carbon source. However, the situation for hydrogen seems to be more complex. The D/H ratios of the granulites formed in subduction zones are within the range which is considered to be that of mantle derived magmas. For the granulites formed in collision zones, the δD values (-99‰) of recent granulites (-16 Ma) suggest a contamination of juvenile water by meteoric waters of high elevation whereas δD values (-60‰) for Archean (-3.3 to -1.7 Ga) granulites suggest a contamination of juvenile water by crustal (metamorphic) waters during granulitic metamorphism.

A798

Goldschmidt Conference Abstracts 2008

Transformations across the bedrock/saprolite – regolith boundary S.G. RIGGINS1,2*, S.P. ANDERSON1,2 AND A.E. BLUM3 1

Institute of Arctic and Alpine Research (INSTAAR), University of Colorado, Boulder, CO 80309, USA (*correspondence: [email protected]) 2 Department of Geography, University of Colorado, Boulder, CO 80309, USA ([email protected]) 3 United States Geological Survey (USGS), Boulder, CO 80303, USA ([email protected]) This study examines the transformations that occur as material transitions from relatively unweathered bedrock/saprolite to weathered regolith at two hillslopes, one on Osborn Mountain in the Wind River Range, WY, USA and another on Bodmin Moor, Cornwall, UK. Both sites experienced a similar periglacial climate during the majority of their formation and we expect that frost processes were responsible for weathering-front advance during that time. Currently these hillslopes experience different climates. Osborn Mountain is relatively cold and dry with periglacial processes still operating and Bodmin Moor is relatively warm and wet. At both sites, mineralogy and chemistry of bedrock/saprolite is distinctly different than the regolith. Most primary minerals decrease in abundance above the bedrock/saprolite boundary while secondary minerals have the opposite trend. We use a mass balance approach to quantify solid phase chemical mass gains and losses by calculating tau (τ) values with Zr as an immobile element. τ values for most major elements show loss above the bedrock/saprolite boundary. There is an exception at Osborn Mountain where Mg, Fe, and Ca have positive τ values, indicating the site has received eolian inputs. Feldspar weathering is different at each site. At Bodmin Moor, plagioclase has an average absolute loss of 45% while Osborn Mountain has an average absolute loss of 13%. We interpret this to indicate that plagioclase is weathering at a faster rate at Bodmin Moor than at Osborn Mountain. The observed weathering differences at Bodmin Moor compared to Osborn Mountain are likely due to the differences in recent climate conditions. The much warmer, wetter conditions at Bodmin Moor currently favor higher chemical weathering rates and a weathering front advance dominated by chemical processes while than the colder, drier conditions at Osborn Mountain favor a weathering front advance dominated by physical processes.

Time constants of chemical transfers in a forested ecosystem inferred from short-lived nuclides 226 228 ( - Ra,210Pb and 137Cs) S. RIHS*, B. THIEN, J. PRUNIER, M-C. PIERRET, D. LEMARCHAND, P. STILLE AND F. CHABAUX Centre de Géochimie de la Surface, UMR 7517, Université Louis Pasteur, 1 rue Blessig, Strasbourg, France (*correspondence: [email protected]) The time constants of radionuclide transfers between several compartments of a forested ecosystem was inferred using natural short-lived nuclides of the U and Th decay series as well as anthropogenic 137Cs fallout. This study was performed on podzolic and brown acidic soils and the surrounding vegetation of the experimental Strengbach watershed located in the Vosges Mountains (France) (http://ohge.u-strasbg.fr/). Analyses of 226Ra, 228Ra, 210Pb, 137 Cs, 230Th and 232Th were performed either by gammaspectrometry or by mass spectrometry (TIMS). Two soil profiles developed from the same bedrock but under two types of vegetation (beech and spruce) were sampled down to 120 cm depth. Plant fractions including roots, barks, leafs and needles were likewise analysed. All the radionuclides display a strongly constrasted behaviour between the "beech-soil" and the "spruce-soil", as well as a fractionation between the vegetation and the underlying soils. In beech-soil, both (226Ra/230Th) and (228Ra/232Th) activity ratios show a non-uniform distribution, with disequilibrium values higher than 1 above 20cm depth, lower than 1 in the 40-60cm range and close to equilibrium below 60cm. In constrast, (226Ra/230Th) and (228Ra/232Th) activity ratios in spruce-soil are close to radioactive equilibrium along the whole profile. Such contrasted behaviour is also shown by vertical migration of atmospheric fallout of 137Cs and 210Pb. The calculated velocity of 137Cs migration in the spruce-soil (0.2 cm/y) fall into the previously reported range. In the beech-soil, however, almost no migration of deposited 137Cs or 210Pb occurred. We take advantage of the different half-lives of the two Ra isotopes to build a coupled transport reactive model of the Ra in the soil-plant system able to quantify timescales of the radionuclide transfers between the distinct ecosystem compartments. [1] Almgren & Isaksson (2006) J. Envir. Radio. 91, 90.

Goldschmidt Conference Abstracts 2008

First exposure ages from Mt Chimborazo, Ecuador V. RINTERKNECHT1*, P. HALL2, C. SCHNABEL3 AND K. WILCKEN4 1

School of Geography and Geosciences, University of St Andrews, St Andrews, KY16 9AL, UK (*correspondence: [email protected]) 2 Instituto Geofisico, Escuela Politécnica Nacional, Quito, Ecuador 3 Natural Environment Research Council-Cosmogenic Isotope Analysis Facility, East Kilbride, G75 0QF, UK 4 Scottish Universities Environmental Research Centre, East Kilbride, G75 0QF, UK Moraine sequences present around Mt Chimborazo, Ecuador suggest wider glacier extents at time of colder and possibly wetter conditions but the timing of these glacial stages is poorly known. We have sampled boulders for surface exposure dating with cosmogenic nuclides from moraines belonging to the late glacial group and the full glacial group. Here we report 36Cl exposure ages on boulder samples prepared at the NERC-CIAF, UK and measured by AMS at the SUERC, UK. We used the program CHLOE3 [1] and production rates estimated by Phillips et al. [2] scaled for latitude and altitude to calculate the exposure ages. No corrections for ash cover or erosion have been applied. The uncertainties correspond to the analytical uncertainties and do not include the uncertainties associated with the estimated production rates used to calculate the exposure ages. Three boulders from the late glacial moraine have exposure ages ranging from 8.5 ± 0.5 36Cl kyr to 9.9 ± 0.5 36Cl kyr. Two boulders from the full glacial stage moraine have similar exposure ages as boulders from the late glacial moraine. Additional analyses on samples from the late glacial moraine will be carried out. [1] Phillips F. M. & Plummer M. A. (1996) Radiocarbon 38, 98-99. [2] Philips et al. (2001) Chem. Geol. 175, 689-701.

A799

Origin of riverine dissolved inorganic carbon: Comparison between West Indies, Reunion and Iceland K. RIVÉ1,2, P. AGINIER1 AND J. GAILLARDET2 1

Lab. de Géochimie des Isotopes Stables, IPGP ([email protected]) 2 Lab. de Géochimie et Cosmochimie, IPGP Volcanic rocks are responsible of about 30% of the global consumption rates of CO2 by chemical weathering of silicate rocks (Dessert et al. 2003). The origin of CO2 implied in the weathering reactions is however still poorly known, particularly in volcanic areas. We propose here a study of three volcanic areas: West Indies, Reunion and Iceland. The studied areas have high weathering rates, but differ in the geodynamic context (subduction, hot spot, and ridge), climate (tropical versus subpolar climate) and vegetative cover. To estimate chemical weathering, we used dissolved inorganic carbon (DIC) and its δ13C, and major ions in the rivers, soil solutions and hydrothermal springs. Our data show two main origins for the DIC: biogenic for 58-72% and magmatic for 42-28%. In West Indies (n=110), the proportion of magmatic CO2 is lower than in Reunion (n=39) and in Iceland (n=40). The contribution of each reservoir varies in the rivers according to the geodynamic context, the ongoing or recent volcanic activity, the proximity from the volcanic edifices and the vegetative cover. These observations illustrate that a significant amount of magmatic CO2 is exported to the ocean via rivers, which could lead to reconsider the balance of magmatic CO2 emissions by the volcanoes at global scale, and its role in weathering reactions. [1] Dessert et al. (2003) Chemical Geology 202, 3-4, 257-273.

A800

Goldschmidt Conference Abstracts 2008

Strontium and Cesium desorption from reacted Hanford sediments

Chemical and textural records during snowball garnet growth

NELSON RIVERA1, PEGGY O’DAY1, SUNKYUNG CHOI1, AARON THOMPSON2 AND JON CHOROVER2

M. ROBYR

1

School of Natural Sciences, University of California, Merced ([email protected], [email protected]) 2 Department of Soil, Water and Environmental Science, University of Arizona Flow-through column experiments were conducted with Hanford sediment to ascertain mechanisms that control the fate and mobility of contaminant Sr and Cs in the subsurface at Hanford (WA). Sediment was reacted with a high ionic strength, caustic synthetic tank waste leachate (STWL) solution containing 10-3 m Sr and Cs. Reactant STWL was flushed through the column until breakthrough, and then flow was stopped, allowing the solution to react for 1 day to 3 months. After reaction, columns were flushed with a simulated background pore water (BPW) solution and effluent Sr, Cs, and dissolved elements were measured for up to 200 pore volumes. Strontium exhibits greater retention over Cs during contaminant uptake, with Cs showing breakthrough two times faster than Sr. During flush out with BPW, initial Cs release is rapid and reaches a steady state after ~5 pore volumes. Depending on stopped-flow reaction time, 5-15% of total sorbed Cs is eluted after 200 pore volumes. Strontium desorption is retarded until ~25 pore volumes for all reaction times. However, Sr exhibits more rapid desorption than Cs after the initial retardation and up to 55% total Sr (sorbed and native in sediment minerals) is eluted from the column after 200 pore volumes. A three-step sequential extraction of reacted column sediments showed 65% of Sr removed by an extraction targeting exchangeable ions (1.0 M MgCl2) and 75% of Cs extracted by stronger extractants (acidic ammonium oxalate and aqua regia). Analyses of EXAFS spectra indicate that Sr in reacted sediments is present as native Sr in feldspars, as sorbed Sr, and as Sr associated with a neoformed feldspathoid or zeolite phase. Current and prior work indicates that the fate of Sr and Cs are controlled by a combination of sorption/desorption and dissolution/precipitation processes operating at different rates, with different reactivities for each contaminant, during both uptake and release. Longer reaction and aging times create recalcitrant neoformed phases that sequester Sr and Cs, which are more resistant to desorption. In a field setting such as Hanford, the rate of fluid migration and neutralization during initial reaction of tank wastes with sediment may strongly influence the subsequent behavior of contaminant migration.

Geol Science Dept., The University of Texas at Austin, Austin, TX 78712-0254, USA ([email protected]) The growth history of two populations of snowball garnets from the Lukmanier Pass area (central Swiss Alps) was examined through a detailed analysis of chemical zoning, crystallographic orientation and 3D geometry. The first population was collected in the hinge of a chevron-type fold and shows an apparent rotation of 360°. Microstructural and chemical data reveal a modification of the stress field regime during garnet growth occuring after 270° of rotation and for XMn = 0.009. Crenulated inclusion trails indicate that the last 90° of the spiral curvature was formed under a non-rotational regime associated with flexural folding. EBSD maps reveal a crystallographic central domain exhibiting 270° of rotation and distinct smaller crystallographic domains at the end of the spirals. A second population collected on the limb of the folds exhibits rotations that do not exceed 270°. Here, the garnets do not record any evidence for a modification of the stress field regime during garnet growth, and a single crystallographic orientation is observed for the whole spiral. X-ray compositional maps show strong correlations in the zoning of Mn, Ca, Fe and Mg. These correlations indicate that these cations achieved local equilibrium during garnet growth and justify the use of Mn concentration as a time marker. Compositional information points out that a particular event in the garnet growth history affected all the garnets simultaneously. This event occurs at XMn = 0.009 and is characterized by an unexpected increase in Ca concentration. In contrast, rim chemical data indicate that the growth of the garnets from the limb ceased at XMn = 0.009. Near the rim of these latter garnets, concentration profiles show an abrupt decrease of Ca concentration. The contrasting behavior of Ca at XMn = 0.009 between samples from both sampling localities suggests dissolution of a Ca rich phase in the limb of the fold, and its precipitation in the hinge of the fold. As plagioclase is the main phase that incorporates Ca in these rocks, flexural folding is interpreted as being responsible for pressure dissolution of plagioclase along the limb of the folds and reprecipitaion of Ca through garnet crystallization in the hinge of the fold.

Goldschmidt Conference Abstracts 2008

A801

Climate during the last glacial cycle in the Pyrenees inferred from 10Be dating and paleoglacier modeling

The realtime kinetics and mechanisms of the amorphous to crystalline CaCO3 transformation

Á. RODÉS1*, M.A. PLUMMER2, R. PALLÀS1, R. BRAUCHER3 AND D.L. BOURLÈS3

J.D. RODRIGUEZ-BLANCO*, S. SHAW AND L.G. BENNING

1

RISKNAT Group, Fac. de Geologia, Univ. de Barcelona, 08028 Barcelona, Spain (*correspondence: [email protected]) 2 Idaho National Laboratory, Idaho Falls, ID 83415-2107, USA 3 CEREGE UMR 6635, Plateau de l’Arbois 13545, Aix en Provence, France Extracting quantitative, temporally referenced, paleoclimatic information from glacial geomorphic features is often limited by 1) the poorly dated glacial records commonly available and 2) a lack of constraints on the relative roles of precipitation and temperature in producing a given glacier extent. 10Be surface exposure dating of glacial features addresses the first problem in the Noguera Ribagorzana valley (South-Central Pyrenees) [1] and allows us to correlate the glacial record with other paleoclimate proxies around the Pyrenees. To address the second problem, we explored use of the Plummer & Phillips 2D numerical model [2] as a means of constraining paleoclimatic conditions by the inverse approach. Model results matching Last Glacial Maximum (LGM) and Greenland Stadial 1 (GS-1) moraines, combined with information from other proxies appear to provide further constraints on likely paleoclimatic conditions (fig. 1).

School of Earth & Environment., Univ. Leeds., Leeds, LS2 9JT UK (*correspondence: [email protected]) Amorphous CaCO3 (ACC) is an unstable precursor to crystalline CaCO3 phases –vaterite (V), calcite (Cc) and aragonite (A) [1]. ACC precipitates instantaneously in highly supersaturated CaCO3 aqueous solutions and then rapidly transforms (e.g., < 3 min at 25°C [1]) to V, Cc or A. The fast kinetics of the reaction represents a significant challenge when quantifiing the nucleation, growth and transformation of ACC. We have used in situ, time resolved synchrotron-based X-ray diffraction combined with high-resolution imaging to characterise the homogeneous nucleation and growth of ACC and its transformation to V and Cc under a variety of environmentally relevant conditions. Experiments were conducted at 1 to 25°C by fast and controlled mixing of equimolar Ca2+:Mg2+ (1:0 to 1:9) and CO32- aqueous solutions. The reactions were characterised using Energy Dispersive XRay Diffraction (ED-XRD) and Small and Wide Angle Scattering (SAXS/WAXS) with patterns collected every 15 sec.

Figure 1: Photomicrograph of V and Cc crystals in ACC matrix (~ 50 nm ø spheres in the background). Insert: Degree of reaction (α) determined from EDXRD for the crystalissation of Cc from ACC at ~ 15°C. Figure 1: Paleoclimatic conditions of the Pyrenees during the last glacial cycle deduced from inverse modeling (grey areas). [1] Pallàs et al. (2006) Quaternary Science Reviews 25, 29372963. [2] Plummer & Phillips (2003) Quaternary Science Reviews 22, 1389-1406.

The results show that although the ACC to Cc transformation is fast (< 400 sec at ~ 15 °C, inset above), a highly quantitative evaluation of kinetic rates, mechanisms and activation energies at sub-minute time scales is achievable. [1] Ogino et al. (1987) Geochim. Cosmochim. Acta 51, 27572767.

A802

Goldschmidt Conference Abstracts 2008

Surface properties of nanosize oxides by high temperature electrophoresis V. RODRIGUEZ-SANTIAGO1, M.V. FEDKIN1, J. ROSENQVIST2, M.L. MACHESKY3, D.J. WESOLOWSKI2 1 AND S.N. LVOV * 1

The Earth and Mineral Sciences Energy Institute, Department of Energy and Mineral Engineering & Department of Materails Science and Engeneering, The Pennsylvania State University, University Park, PA 16802 (*correspondence: [email protected]) 2 Chemical Sciences Division, Oak Ridge National Laboratory, P. O. Box 2008, Oak Ridge TN 37831-6110 ([email protected]) 3 Center for Watershed Science, Illinois State Water Survey, 2204 Griffith Dr., Champaign, IL 61820 ([email protected]) Surface properties – zeta potential and isoelectric point (IEP) – of SnO2 and SiO2 were studied over the temperature range from 25 to 260oC. Two illumination approaches (i.e., bright- and dark-field) were used to obtain the hightemperature electrophoretic mobility; both using a specially designed hydrothermal electrophoresis cell [1, 2]. Bright-field illumination worked better for large particles, while dark-field illumination provided increased particle size detection down to the nanometer range, as well as more accurate measurements close to the IEP. Nanometer detection was possible by observing the light reflected from the particles’ surface instead of tracking the particles themselves. The electrophoretic mobility of SnO2 was measured over the pH range 2 to 12, and the pH value corresponding to zero zeta potential (IEP) was determined at each temperature. In the case of SiO2, the surface was always negatively charged at all temperatures thus no IEPs were observed. The obtained IEP values for SnO2 were fitted to the 1pK surface complexation model, as well as compared to the values of the pH points of zero charge (PZC) predicted by the Multisite Complexation (MUSIC) model for this material. The temperature dependence of the IEP values of SnO2 closely follows the trend of temperature change of the water dissociation constant. A similar surface behavior has been observed for other oxides [1,3]; furthermore, this trend may provide an empirical tool for predicting surface properties of oxides at elevated temperatures. [1] Zhou et al. (2003) Rev. Sci. Instr., 74, 2501-2506. [2] Fedkin et al. (2003) Langmuir, 19, 3797-3804. [3] Machesky et al. (2001) J. Colloid Int. Sci., 200, 298-309.

Integrated modelling of biological and weathering processes at the continental scale CAROLINE ROELANDT, YVES GODDÉRIS, MARIE-PAULE BONNET AND FRANCIS SONDAG LMTG, CNRS-IRD-Université Toulouse III, Toulouse, France ([email protected]) Continental weathering is a key process of the surficial biogeochemical cycles at the geological timescale, and is generally modelled as a function of global climate through simple parametric laws. However, recent evidences for rapid changes in continental weathering at the decennial to centenial scale in response to the ongoing global climatic change ask for the development of mechanistic models coupling the fast responding continental biospheric processes to the weathering reactions. Here we present the coupling of the LPJ global biospheric model to a reactive transport model of the weathering reactions in the field (WITCH). The coupled LPJ-WITCH model runs with a spatial resolution of 0.5°lat x 0.5°long, and is forced by database output supplying the mean annual rainwater flux and chemistry, the air temperature, the bedrock lithology, and the soil characteristics. The model is run over the Orinoco watershed, and its main output are the main base cation composition of the runoff and drainage waters (thus the CO2 consumption), which compares reasonably well with the available measurements in the main stream. Sensitivity studies to global climate and land use changes are performed, illustrating the efficiency of such large scale mechanistic models. Also a sensitivity test to increased erosion rate is performed when tropical forest is replaced by grasslands, by reducing the thickness of the soil layer.

Goldschmidt Conference Abstracts 2008

Mantle lithosphere in basalts from the Kenya Rift

A803

Iron reducing bacterial enrichments from arctic marine sediments at Tempelfjorden, Svalbard, Norway

NICK ROGERS

YUL ROH1, SUNG-KEUN RHEE2, BYUNGNO PARK1, KYU-CHEOL YOO3, JAE IL LEE3, MATTHIAS FORWICK4 3 AND HO IL YOON

Department of Earth and Environmental Sciences, CEPSAR, The Open University, Milton Keynes, MK76AA, UK ([email protected]) 1

The Kenya branch of the African Rift cuts across the east African plateau and is characterised by basaltic volcanism with many similarities to OIB. Lavas range in composition from transitional tholeiite to alkali basalt, basanite and nephelinite and the mean rate of magma production is ~0.03 km3 yr-1, similar to that of small ocean islands. Age progression of the onset of basaltic activity can be related to the counter-clockwise rotation of the Africa plate over a stationary hot-spot located beneath the Tanzania craton. Finally the East African plateau itself is dynamically supported from sub-lithospheric depths and mantle tomographic studies reveal the possible presence of a mantle plume beneath the Tanzanian. Critical compositional features of Kenya basalts contrast with those expected of plume-related magmas and reflect a clear lithospheric control. Nd and Sr isotope ratios change with the age of the underlying lithosphere and critical trace element ratios are distinct from those characteristic of OIB. Zr/Hf is higher than OIB with similar Zr and REE contents and Th/U ratios are distinctly lower. By contrast U-series isotopes in Kenya basalts are characterised up to 40% 230Th excess, consistent with REE fractionation that indicates residual garnet during melt generation. The amount of 230Th excess is too great to be generated by simple batch or fractional melting and imply a dynamic melting regime beneath the rift axis in which porosity is low and mantle upwelling rates are or the order of 1 cm yr-1. The data imply thermal erosion and reactivation of the mantle lithosphere by the underlying mantle plume, such that lithospheric material can undergo dynamic melting.

Chonnam National University, South Korea Chungbuk National University, South Korea 3 Korea Polar Research Institute, South Korea 4 University of Tromso, Norway 2

Svalvard fjords, affected by turbid overflow emanating from outlet glaciers, are modern analogues for Quaternary deglacial settings. We sampled the marine sediments at 8 stations along the axis of Tempelfjorden, Svalbard to better understand the response of the discharge of glacial meltwater including suspended sediments to the biogeochemistry and microbial community. Marine sediments were enriched for iron-reducing bacteria with acetate or lactate as the electron donors. The enrichments could reduce both Fe(III)-citrate and iron oxyhydroxide with either acetate or lactate as the electron donor. The enrichments were dominated by Clostridium sp., Tissierella sp., and Alkaliphilssp. And the enrichments also contained strictly fermentative microorganisms and novel microrganisms. Iron-reducing bacteria enriched from the marine sediments showed active Fe(III) reduction at 8°C and 25°C. They reduced an iron oxyhydroxide, akaganeite, as an electron acceptor using lactate as an electron donor and formed nm-sized magnetites. Dynamic bacterial community succession was observed at 8 stations and archaeal community was quite distinct in the sediments affected by deglaciation compared to control sediments. This study indicated that microorganisms in marine sediments contribute significantly to global cycles of organic and inorganic matters because of their abundance.

A804

Goldschmidt Conference Abstracts 2008

Precambrian metaigneous xenoliths of Vestfjella: Implications for lithospheric architecture in western Dronning Maud Land, Antarctica I. ROMU*, M. KURHILA AND A. LUTTINEN1 Department of Geology, P.O.Box 64, FI-00014 University of Helsinki, FINLAND (*correspondence: [email protected]) ([email protected], [email protected]) Lamproite- and basalt-hosted xenoliths from Kjakebeinet (73º47’ S, 14º53’ W) and Muren (73º43’ S, 15º02’ W), south Vestfjella, represent unique samples of the unexposed continental crust at the rifted margin of western Dronning Maud Land. The lamproite-hosted xenolith suite indicates 1) heterogeneity of the crust, 2) mainly igneous protoliths, and 3) granulite facies metamorphic conditions. Two leucocratic tonalite gneiss samples yielded U-Pb SHRIMP zircon ages of ~1.3–1.0 Ga (J. Jacobs, personal communication, 2006). The mafic granulites exhibit compositional affinities to Proterozoic lower crustal xenoliths from Lesotho, South Africa [1], and to mantle-derived lower crustal material in general [2]. Interpretation of mineral–whole-rock Sm-Nd isotope data on the mafic granulites is complicated by lamproite overprint; the data are accordant with equilibration of the Sm-Nd -system during Grenvillian and Jurassic magmatic events. In contrast, whole-rock Sm-Nd isotopic data on basalt-hosted leucocratic metatonalite xenoliths show highly negative εNd values (ca. –50 @ 180 Ma) indicative of Archean origin [3]. Overall, the present results imply proximity of the Archean-Proterozoic lithospheric boundary in south Vestfjella. New U-Pb SIMS zircon data on the basalt-hosted and lamproite-hosted felsic metaigneous xenoliths will be available in May 2008 and are expected to facilitate precise dating of tectonic and magmatic events related to the evolution of the poorly exposed crust of East Antarctica. [1] Rogers & Hawkesworth (1982) Nature 299, 409-413. [2] Condie (1998) Lithos 46, 95-101. [3] Luttinen & Furnes (2002) Journal of Petrology 41, 1271-1305.

Multi-component isotopic mixing in the Ethiopian Rift: Modeling plume contributions to recent magmatism T. ROONEY1,3*, B. HANAN2, T. FURMAN3 AND D. GRAHAM4 1

Dept. of Geological Sciences, Michigan State University, East Lansing, MI 48824, USA (*correspondence: [email protected]) 2 Dept. of Geological Sciences, San Diego State University, San Diego, CA 92182-1020, USA 3 Dept. of Geosciences, The Pennsylvania State University, State College, PA 16801, USA 4 College of Oceanic & Atmospheric Sciences, Oregon State University, Corvallis, OR 97331, USA Primitive basalts erupted in the Wonjii Fault Belt (WFB), Debre Zeyit (DZ) and Butajira (BJ) areas of the northern and central main Ethiopian Rift define pseudo-binary mixing trends in Nd-Sr-Pb-Hf multi-isotope diagrams. The arrays for the sub-populations WFB, BJ and DZ extend away from a common Afar plume (AP) end-member towards distinct depleted mantle (DM)–African continental lithosphere (ACL) binary-like components. Preliminary Hf isotopes confirm that the Afar Plume component has a C-like and not HIMU radiogenic isotope signature. We choose to model the data trends by 3-component mixes between AP, ACL, and DM end-members. Principal component analysis using the Sr, Nd, and Pb data and the 3 end-member components shows that 98.72% of the total variation is accounted for within the plane of the first two eigenvectors, providing support that the model is justified. The Nd-Sr-Pb inverse isotopic ternary mixing model shows that contribution to magmatism from the Afar plume decreases while the influence of DM correspondingly increases as a function of the radial distance from Erta ‘Ale volcano in Afar, the presumed plume location today. The contribution of the ACL component is nearly constant, but is unusually strong near the Boru-Toru structural high. 3He/4He values correlate positively with the weight fraction of the Afar plume component and, hence, inversely with distance from Erta ‘Ale. We further tested the model by using the weight fractions of the 3 end-member components and published averages for the trace element composition of OIB, DM, and continental lithosphere to show that we can replicate the spider plots for the actual basalt data.

Goldschmidt Conference Abstracts 2008

Predicting arsenic behavior in highiron subsurface environments

Sulfur and Oxygen isotopes: Evidence of H2S spring sources, Southern Mexico

ROBERT A. ROOT1, PEGGY O’DAY1, JANET HERING2, KATE CAMPBELL3 AND DIMITRIOS VLASSOPOULOS4

L. ROSALES-LAGARDE1,2*, A. CAMPBELL1, P.J. BOSTON1,2 AND K.W. STAFFORD1,2

1

School of Natural Sciences, University of California, Merced ([email protected], [email protected]) 2 EAWAG, Dübendorf, Switzerland ([email protected]) 3 USGS, Menlo Park, CA , USA ([email protected]) 4 SSPA, Portland, OR, USA ([email protected]) Arsenic sorption depends on biotically and abiotically driven pH and redox processes, and can be controlled in lowsulfur subsurface environments by the activity of Fe. Reaction-path modeling of redoximorphic sediments suggests that phosphate can play an important role by competing with arsenic for sorption sites on Fe surfaces and by potential precipitation of Fe phosphate phases. Modeling of an Ascontaminated system at oxic conditions with very low sulfate [log(SO42-) = -7], high Fe [(log(Fe2+) = -3], and low phosphate [log(HPO42-) = -5.5] predicts that ferrihydrite is the metastable Fe phase. At the modeled pH range (4-10), HxPO4x-3 will compete with HxAsO4x-3 for sorption sites on ferrihydrite. As conditions become suboxic (pe < -1) at pH > ~7, a green rust phase, modeled here as (GRICO3) [Fe(II)4Fe(III)2(OH)12]2+[CO3·3H2O]2- is predicted and overlaps with the stability fields of dissolved As(V) and As(III). If present, GRICO3 provides a substrate for sorption in suboxic sediments, and may also intercalate HxPO4x-3 and HxAsO4x-3. At pH < ~7, GRICO3 is not stable and Fe is present as Fe2+(aq). At modeled conditions of high Fe and high phosphate [log(HPO42-) = -3.5], phosphate will compete more strongly with arsenate for sorption sites on ferrihydrite at pe > 1. As conditions become suboxic (pe < -1) and pH = ~7, the GR phase may not form, as vivianite [Fe3(PO4)2. 8H2O] is the predicted stable Fe phase and the GR stability field is shifted to higher pH (> ~7.5). Model predictions were compared to sediment data from field investigations at North Haiwee Reservoir (Olancha, CA, USA). An Fe(II,III) phase similar to GRICO3 was identified by X-ray absorption spectroscopy. The GR-like phase formed at depths above and overlapping a release to pore waters of As, P, and Fe, determined by in situ gel probe sampling. Field investigations indicate that phosphate can be released to porewaters at concentrations greater than contaminant arsenic (10-25µM versus 2-5µM) and may inhibit re-adsorption of As. As conditions become more reducing (pe < -2), it is predicted that the number of sorption sites will decrease by reductive dissolution of the host Fe phase (GR) and phosphate will out compete As(III), releasing As to porewater.

A805

1

Earth and Environmental Science Department, New Mexico Tech 801 Leroy Place, Socorro, NM, USA 87801 (*correspondence: [email protected]) 2 National Cave and Karst Institute Carslbad, NM, USA Anomalously high H2S concentrations up to 240 ppm exhale from springs on the surface and in caves in TabascoChiapas states, Mexico. The proposed sources are basinal fluids [1] and gas from an active volcano [2]. Further regional sampling of spring water seeks to elucidate the extent of the H2S generation and its relation to the subsurface evaporites [4]. Although δD-δ18O analysis suggests that meteoric water is the main water-source of the H2S-springs, the water composition ranges from 1400 to 6600 mg/l TDS (brackish water); [Cl-] = 250 to 2000 ppm; and [SO42-] = 250 to 4000 ppm. This composition may be reached by interaction of groundwater with Late Jurassic salt and Lower Cretaceous anhydrite through lateral faults and anticline axis fractures. H2S degassing from the spring water suggests the presence of an open system. The similarities of the δ34S and δ18O between the spring soluble sulfate, the brine-related oil sulfate, and the subsurface anhydrite propose a common origin. These isotopes do not show any evidence of a volcanic H2S input to the springs. A relatively constant fractionation from 28 to 45‰ δ34S-VCDT between the aqueous sulfides and sulfates of the H2S springs points to an equilibrium sulfate reduction fractionation process. These values are inside the sulfatesulfide fractionation range reported for microbial [4] sulfate reduction suggesting this mechanism can be producing the H2S. The driving force for reduction, (temperature, pH, or the presence of organic matter) is not yet defined. The near neutral pH of the water (6.4 to 7.5) and a temperature at most three degrees Celsius above atmospheric temperature, points to the latter. The lack of change in the sulfate suggests there is an infinite reservoir of this compound. The produced H2S reoxidizes to H2SO4, reacting with the limestone host rock in caves to produce gypsum/selenite with the same isotopic signature as the precursor H2S. Further data is being collected to clarify the H2S source. [1] Hose et al. (2000) Chemical Geology 169, 399-423. [2] Spilde et al. (2004) GSA Abs. 36, 5, 106-11. [3] MenesesRocha (2001) AAPG Mem. 75, 183-216. [4] Johnston (2007) GCA 71, 3929–3947.

A806

Goldschmidt Conference Abstracts 2008

Melting of residual eclogites with variable proportions of quartz/coesite

Bond strength and the compression of framework minerals

A. ROSENTHAL1*, G.M. YAXLEY1, D.H. GREEN1, J. HERMANN1 AND C.S. SPANDLER1,2

N.L. ROSS, R.J. ANGEL, J. ZHAO AND D. WANG

1

Research School of Earth Sciences, The Australian National University, Canberra, 0200 ACT, Australia (*correspondence: [email protected]) 2 Institute of Geological Sciences, University of Bern, 3012 Bern, Switzerland Eclogite and pyroxenite layers and veins within the peridotitic mantle may be important in genesis of many magmas. Understanding high pressure melting of residual eclogites with varying amounts of quartz/coesite (qz/co) will improve understanding of the melting behaviour of heterogeneous mantle assemblages. We are investigating the phase and melting behaviour of a residual eclogite composition, which crystallizes garnet (grt) + clinopyroxene (cpx) at 3.0 & 4.0 GPa with no qz/co (REC). The melting behaviour of REC reveals the anhydrous solidus of a coesite-free eclogite. The subsolidus phases of REC are grt and cpx at 3.0 & 4.0 GPa. The solidus is at 1325±25°C at 3.0 GPa, and at 1475°C±25°C at 4.0 GPa. Melt compositions and proportions with cpx > grt are controlled by grt-cpx cotectics. At 3.0 GPa, cpx/grt ratios decrease with increasing % melting. At 4.0 GPa, cpx/grt ratios are lower at a given % melting than at 3.0 GPa. We are also studying eutectic-like melting in coesitebearing eclogites with varying proportions of qz/co. REC10 and REC20 are identical to REC, but have 10% and 20% higher SiO2 contents, respectively. Both REC10 & 20 crystallise sub-solidus as grt + cpx + qz/co at 3.0 GPa. The solidi of both are similar at 3.0 GPa, at 1275±25°C. Low-% melting (<20%) is eutectic-like until qz/co melts out. Like REC, grt-cpx cotectics control high degree melting for all compositions, with cpx abundance always exceeding grt. However, for REC, the proportion of cpx always exceeds those in REC10 & 20, but the % melt is always lower than for REC10 & 20 at given temperature. Melts vary with increasing % melting from andesitic to basaltic for REC & REC10, but from dacitic to basalticandesitic for REC20. At <25% melting, melt in REC has lower Mg# at given % melting than melt in REC10 & 20. The contrary is given at higher % melting. Garnet and cpx in REC have higher Mg# at a given % melting than grt and cpx in REC10 & 20. Towards higher % melting (>25% melting), variations in cpx Mg# diminish. These variations of residual grt and cpx Mg# and melt Mg# are principally because melts from REC (qz/co-free) are more mafic at a given % melting. These outcomes reveal the strong dependence of Mg# of residual mineral phases & melt on the presence or absence of qz/co in melting of eclogites.

Crystallography Laboratory, Department of Geosciences, Virginia Tech, Blacksburg, VA 24061, USA It is commonly assumed that the relationship between bond strength and bond length for a particular pair of atoms is a simple and single-valued for a given coordination environment; longer bonds are weaker. This is the basis of the concept of bond valence. In strongly-bonded oxide minerals, the range of bond lengths found for a given polyhedron is so small that it was long thought that the polyhedral bulk moduli were essentially independent of structure type. This view is incompatible with the discovery that the response of the perovskite structure to high pressures is controlled by the equipartition of bond-valence strain between the A and B cation sites within the structure [1]. The same appears to be true, within experimental uncertainties, for all framework structures with rigid-unit modes. In perovskites, this explicitly implies that the octahedral compressibility depends not only upon the octahedral cation, but also upon the compressibility of the extra-framework site. Thus the octahedral compressibility of a B cation site must change as the A-site cation is changed, whether or not the B-O bond lengths change as a result of the substitution on the A site. The strength of bonds is thus dependent upon the crystal environment and not solely upon the bond length. The observation of a plateau effect in the variation of octahedral compressibilities in perovskite solid solutions suggests that the bond-valence matching principle is followed not just globally, but on a local scale as well. Such observations should allow the change with pressure of the excess thermodynamic properties of solid solutions to be directly related to the microscopic (atomic scale) evolution of the structure. [1] Zhao J., Ross N.L., & Angel R.J. (2004) Acta Cryst. B60, 263-271.

Goldschmidt Conference Abstracts 2008

C60 ToF-SIMS: A tool for highresolution mapping of elements and organic compounds

Characterization of simple fluids under confinement and at free interfaces using neutron scattering techniques

D. ROST*, T. HENKEL, A. KING AND I. LYON The University of Manchester, SEAES, Williamson Bldg. Oxford Rd., Manchester, M13 9PL, United Kingdom (*correspondence: [email protected])

Experimental A recent development in time-of-flight secondary ion mass spectrometry (ToF-SIMS) is the deployment of heavy cluster primary ion guns. Hereby desorption of secondary particles is less destructive and larger (organic) molecules can be detected intact, complementing the elemental information. Furthermore, the reduced penetration depth of the cluster ions causes less sputter damage like layer mixing and hence allows for shallower depth profiling. We recently equipped one of our ToF-SIMS instruments [1, 2] with a 40 kV C60 primary ion gun from Ionoptika Ltd [3] for application in cosmochemistry. Measurements can be routinely performed with a lateral resolution of ~2 µm. Submicrometer resolution is possible although at the expense of intensity. Ultimately, this method enables the measurement of the 3D distribution of most major, minor and trace elements plus organic compounds on the (sub)micrometer scale. Our instrument is also equipped with a digital micro camera and a secondary electron detector to ease correlation with results from other techniques, e.g. electron microbeam methods. Additional non-resonant laser post ionization is almost operational. Its application will distinctively reduce detection limits and hence improve our capabilities to measure trace elements and isotopic compositions [2].

Application Our first project is the examination of the aerogel capture medium used in the Stardust cometary sample return mission [4]. Surfaces of flight aerogel exposed to the cometary particle flux as well as unexposed will be analyzed. The aim of this study [5] is to investigate: (1) If direct deposition of organic compounds from the coma of comet Wild 2 can be identified on exposed aerogel surface. (2) If micrometer-sized impact structures are present, potentially accompanied by local enrichments of cometary matter including organic compounds. First results will be presented. [1] Braun et al. (1998) Rapid Commun. Mass Spectrom. 12, 1246–1252. [2] Henkel et al. (2007) Rev. Sci. Instrum. 78, 055107. [3] Hill et al. (2006) Appl. Surf. Sci. 252, 7304–7307. [4] Brownlee et al. (2006) Science 314, 1711–1716. [5] Rost et al. (2008) LPS XXXIX #2110.

A807

GERNOT ROTHER1, DAVID R. COLE1, ARIEL A. CHIALVO1 AND KENNETH C. LITTRELL2 1

Chemical Sciences Div., ORNL, Oak Ridge, TN 37831 ([email protected], [email protected], [email protected]) 2 Neutron Sciences Div., ORNL, Oak Ridge, TN 37831 ([email protected]) The properties of simple and complex fluids confined in nano-porous earth and engineered materials can deviate significantly from bulk behaviour under the same thermodynamic conditions. Neutron scattering techniques can resolve the structure and mobility of imbibed fluids and of molecules in pores. We used SANS to study the sorption properties of supercritical fluids inside mesoporous silica aerogel. The Adsorbed Phase Model [1] allows, for the first time, a means to quantify the physical properties of the adsorbed phase formed by fluids inside porous media in terms of the mean density and volume of the sorption phase. Knowledge of these quantities permits the model-free calculation of the absolute adsorption, a property relevant to application of the equation of adsorption and molecular modeling. The APM model has been applied to the SANS data obtained for sorption of carbon dioxide and sulphur hexafluoride in silica aerogel at supercritical conditions. Our results show clear evidence for fluid depletion for conditions above the critical density. The density and volume of the sorption phase change in a sensible, complex fashion as a function of pressure and temperature. As an extension to the SANS effort we recently explored the use of neutron reflectivity to assess the density profile of fluids near solid interfaces. Our initial study focused on the behavior of sulphur hexafluoride near a Si/SiO2 interface to confirm the existence of a fluid depletion zone near the surface at fluid densities near and above the critical density. Ultimately, the combination of SANS and neutron reflectivity data and molecular dynamics modeling will provide a more fundamentally-rigorous understanding of the interplay between surface and confinement effects. [1] G. Rother et al. (2007) J. Phys. Chem. C 111, 15736.

A808

Goldschmidt Conference Abstracts 2008

K-Si-metasomatism of 3.4-3.3 Ga volcanoclastic sediments: Implications for Archean seawater evolution VIRGILE ROUCHON AND BEATE ORBERGER Université Paris Sud, CNRS UMR-IDES 8148, Département des Sciences de la Terre, Bât. 504/509, 91405 Orsay Cedex, France ([email protected]) The effects of K-Si-metasomatism during the formation of Early Archean replacement cherts were quantified for two examples, the Msauli chert (MC, Barberton greenstone belt, South Africa) and the Kittys Gap chert (KGC, Pilbara craton, Western Australia). The precursor composition of the KGC is similar to dacites of the Duffer formation (Pilbara craton), while the MC have an Al-depleted komatiites precursor, similar to those from the Weltevreden Formation (Barberton greenstone belt). Based on mass balance calculations, the KGC volcanoclastic deposits had an initial porosity of up to 85 vol.% and those of the MC about 65 vol.%. Secondary porosities (27 vol.%: MC, 8 vol.%: KGC) result from Kmetasomatism proportional to the dissolution of Fe, Ca, Mgrich glass and precursor minerals. Komatiites have a higher chemical exchange potential than dacites, each gram releasing 1.2 mmol Fe2+, 2.8 mmol Mg2+, 1.4 mmol Ca2+ and 1.1 mmol Na+ together with 4.4 mmol O2- to seawater. During Kmetasomatism, one gram of komatiite takes up 0.67 mmol of K+ and 2.7 mmol of H+. KGC take up most silica (82 mmol/g of precursor). It is suggested that this silica enrichment operated in the water column and at the sediment-water interface by sorption on the surface of detrital particles and particulate organic matter, as a result of seawater silicasaturation. The formation of K-rich phyllosilicates by interaction with seawater during the early diagenetic alteration of the volcanoclastic particles. was favored by acidic conditions (pH 5.5-6.5) and hot temperatures (> 70°C). The widespread occurrence of K-Si-metasomatism in volcanic and sedimentary rocks can be regarded as a general alteration process of the Early Archean seafloor, as major controlling process for the seawater composition.

Contamination and heterogeneity in the mantle beneath the alkaline Monteregian Province (Québec): Evidence from geochemical and Nd-Sr isotope data E. ROULLEAU* AND R. STEVENSON GEOTOP research center, CP 8888 Succ. Centre Ville Montreal (Qc), H3C 3P8, Canada (*correspondence: [email protected]) ([email protected]) The Monteregian Igneous Province is a product of continental magmatism and constitutes nine plutons with associated dykes that were emplaced in southern of Québec ca 124 Ma ago (+/-1.5 Ma). Previous geochemical and isotopic studies have suggested that the magmas that formed these intrusions were derived from shallow or deep mantle sources with variable crustal contamination. Geochemical and isotopic (Nd-Sr isotopes) analyses of mafic units from the Monteregian intrusions reveal compositions similar to those of ocean island basalts (OIB) associated with mantle plume dynamics. This includes enriched abundances of light Rare Earth Elements (REE) and High Field Strength Elements (HFS). The isotopic compositions of the intrusions are characterized by lower Nd and higher Sr isotope ratios than those of the normal depleted mantle. Part, but not all, of this variation likely reflects crustal contamination of the magmas during their ascent through the continental crust. This contamination can be modelled using assimilation-fractional crystallization (AFC) calculations based on contamination by incorporation of Grenville, St. Lawrence Lowlands or Appalachian crusts. This modelling yields good results for samples with lower incompatible element concentrations (e.g Nb-poor), but poor results for samples with higher incompatible element (Nb-rich) samples. This suggests that incompatible element enriched samples preserve an original mantle signature that is equivalent to the HIMU plume component observed in present-day ocean island basalts whereas contaminated samples preserve evidence of contamination of the plume during its ascent through the continental crust.

Goldschmidt Conference Abstracts 2008

Sources and biogeochemical cycling of Iron isotopes in marine environments

Measurement of Germanium isotope composition in marine samples by hydride generation coupled to MC-ICP-MS

OLIVIER ROUXEL Marine Chemistry & Geochemistry Dept., Woods Hole Oceanographic Institution, Woods Hole, MA02543, USA ([email protected]) Considered esoteric only a few years ago, research into the stable isotope geochemistry of iron is moving into the biogeoscience mainstream. Although initial attention focused on the potential use of Fe isotopes as biosignatures, they are now emerging as powerful oceanographic and paleoceanographic proxies. In particular, Fe isotopes are providing useful information about redox conditions in modern and ancient marine environments, mainly because biotic and abiotic redox processes are among the principal factors that fractionate Fe isotopes. While Fe isotope studies in seawater still present a challenging analytical problem, Fe isotopes provide a new approach to important questions concerning Fe sources, Fe speciation and its bioavailability in the ocean. Here, I will review and present new results of Fe isotope systematics in modern marine environments, in particular: (1) riverine input to the ocean, which has been generally considered to have low δ56Fe values (down to -1‰). Our recent study of two estuaries along US East coast suggests that flocculation processes produce minimal Fe-isotope fractionation. However, we also found that dissolved Fe flux to the ocean may be characterized by more positive δ56Fe values (up to 0.3‰) relative to the crust than previously reported; (2) redox cycling of Fe in substerranean estuaries and shelf sediments, which can result in very light Fe-isotope composition in surface pore waters (down to -5‰); (3) atmospheric Fe sources to the open ocean, with crustal δ56Fe values. Our recent time-series study of marine particulates from the Saragasso sea (sediment traps) yielding δ56Fe=0.07±0.03‰ (1σ, n=36) confirm such assumptions; (4) hydrothermal Fe sources from mid-oceanic ridges, back-arc systems and volcanic seamounts, which are shifted to low δ56Fe values from 0.0 down to -1.8‰ compared to igneous rocks. Additional Fe-isotope fractionation may be also produced in hydrothermal plumes through the water column during partial sulfide precipitation and Fe oxidation and stabilization of Fe in seawater by organic-ligands. Further advances will focus on the direct determination of Fe-isotope compositions in seawater in order to provide new constraints on the oceanic sources and cycling of Fe, an element that is essential for ocean biogeochemistry.

A809

OLIVIER J. ROUXEL1, RAPHAELLE ESCOUBE1,2 AND OLIVIER F.X. DONARD2 1

Marine Chemistry & Geochemistry, Woods Hole Oceanographic Institution, Woods Hole, MA02543, USA ([email protected]) 2 IPREM, U. Pau et Pays de l’Adour, CNRS UMR 5254, Hélioparc 64053 Pau, France Germanium (Ge) is a trace metalloid in seawater that is of particular interest in marine biogeochemistry because of its strong chemical similarity to silicon. In addition, Ge stable isotope ratios in biogenic silica may have strong potential as paleoceanography proxies. We recently reported Ge-isotope composition of igneous rocks and deep-sea clays and defined a bulk Earth 74/70Ge value of 0.5‰ (relative to NIST3120a). Based on 74Ge values of modern marine sponges, it has been also suggested that Ge-isotope composition of seawater is enriched in heavy isotopes relative to the crust ( 74Ge around 2.2‰). Here, we report the development of a new technique to measure Ge isotope composition in seawater. Ge is coprecipitated with magnesium at pH~9-10 and Mg precipitates are filtered and dissolved in 0.25N HNO3. The samples are then analyzed using a continuous flow hydride generation system coupled to a MC-ICP-MS (Thermo-Finnigan Neptune) operating at WHOI. The instrumental mass bias was corrected by the “standard-sample bracketing” method. We initially applied this technique to measure Ge-isotope composition in seafloor hydrothermal fluids. Ge isotopic composition of low temperature (~50°C) hydrothermal fluids from Loihi Seamount in the Pacific Ocean show systematically positive δ74Ge values ~1.5‰. In contrast, δ74Ge values of associated hydrothermal deposits, composed essentially of Fe-oxyhydroxides, range from -0.7 to 0.6‰. The enrichment of light Ge-isotopes in Fe-oxyhydroxide precipitates at Loihi relative to the fluid (up to 2.5‰) is consistent with previous experimental study of Ge-isotope fractionation during Ge-sorption onto Fe-oxyhydroxides (goethite). These results suggest that seafloor hydrothermal fluids may represent a source of isotopically heavy Ge in the ocean and that this isotope signature may be affected by Ge precipitation upon mixing with seawater. Additional studies will focus on Ge-isotope systematics in rivers and deep seawater in order to establish a reliable isotope mass balance of Ge in the ocean.

A810

Goldschmidt Conference Abstracts 2008

A melt inclusion investigation of crustal contamination: A case study at Paricutin Volcano

The ridge filter: How melt supply and magma chambers modulate mantle compositions in MORB

M.C. ROWE1*, D.W. PEATE1, I. UKSTINS-PEATE1 AND T. HOUSH2

K.H. RUBIN* AND J.M. SINTON

1

University of Iowa, Department of Geoscience, Iowa City, IA 52242, USA (*correspondence: [email protected]) 2 University of Texas at Austin, Dept. of Geological Sciences, Austin, TX 78712 Geochemical variability observed in lavas at Paricutin Volcano (Mexico) is often cited as a classic example of assimilation and fractional crystallization processes. Increases in whole rock SiO2 (wt%), δ18O, and 87Sr/86Sr, characteristic of crustal assimilation, are observed over the 9 year lifespan (1943-52) of Paricutin Volcano [1]. In this study, the geochemical variations of olivine and orthopyroxene-hosted melt inclusions are utilized to characterize discrete batches of magma present and to examine the timing of crustal contamination during the evolution of the volcanic system. Two compositional groups of melt inclusions are identified in each of the samples included in this study; a high-SiO2 (58-65 wt%) and low-SiO2 (50-60 wt%) group. Melt inclusions in the high-SiO2 group are compositionally similar to scoria groundmass glass, but retain higher SO3 concentrations (0.017-0.118 wt%) than reported in scoria glass [2]. K2O/TiO2 ratios overlap between high- and low-SiO2 groups within a given sample, possibly suggesting an evolution through fractional crystallization. Melt inclusions, similar to whole rock compositions, record a rapid change in composition during 1947, characterized in part by a dramatic increase in K2O/TiO2, rather than a gradual evolution over the volcano’s lifespan. Importantly, there is no compositional overlap between melt inclusions in early and late erupted lavas. Melt inclusion geochemistry suggests that 1) any contamination occurred early, prior to significant crystallization of either olivine or orthopyroxene and that 2) contamination was either a rapid process occuring over a limited timespan, or the change in composition records a transition to erupting a more evolved and contaminated batch of magma, discrete from that erupted from 1943-46. [1] McBirney et al. (1987) Contrib Mineral Petrol 95, 4-20. [2] Luhr (2001) Contrib Mineral Petrol 142, 261-283.

Geology & Geophysics, SOEST, Univ. of Hawaii, Honolulu, HI 96822 USA (*correspondence: [email protected]) Mid Ocean Ridges (MOR) have long been appreciated by geochemists for their ability to depict underlying mantle compositions in the lavas they erupt. In many ways, compositions of MORB provide the least complicated, most widespread and most continuous picture of upper mantle compositions of any rock type on Earth’s surface. Yet not all ridges sample the mantle in the same way. The effects of magma production, accumulation, mixing and differentiation are important considerations for how ridge volcanism represents the underlying mantle and how one uses MORB to extract the highest definition picture of spatial compositional variations in the mantle. This presentation will describe ways that MOR volcanism modifies spatial and temporal patterns of mantle compositional variance at normal and hot-spot influenced ridges. Rubin and Sinton [1] demonstrated that the depth, size, mixing efficacy and thermal state of mid-ocean ridge magma chambers is strongly dependent upon melt supply, resulting in non-intuitive yet systematic spatial variations in MORB compositions around the world. Broadly speaking the effect on mantle signatures, such as radiogenic isotope compositions and highly incompatible trace element ratios, is diminished compositional variance as melt supply increases. Mean extent of magma differentiation also increases with melt supply, yielding predictable, coupled major element and mantle tracer patterns in MORB. However, high eruption rates at high magma supply also better preserve temporally variable parental melt compositions in erupted lavas (e.g., at fast spreading ridges [2]). Low melt supply preserves a fuller range of mantle chemical variance, although low eruption frequency reduces the ability of erupted lavas to reflect temporality in parental melt variations. Interestingly, the relative range of values of a mantle tracer such as 87Sr/86Sr in MORB does not vary much through the global ocean ridge system away from hot spots, when only lavas with the most magnesian melts are considered. This suggests that all ridges process and mix parental magmas to similar degree prior to the stage where melts pool and differentiate in shallow, melt-rich magma lenses. [1] Rubin & Sinton (2007) EPSL 260, 257-276. [2] Bergmanis et al. (2007) G-cubed 8, Q12O06.

Goldschmidt Conference Abstracts 2008

Finding peaks in geochemical data sets JOHN F. RUDGE Lamont-Doherty Earth Observatory, Columbia University, Palisades, New York 10964, USA ([email protected]) Histograms of many geochemical data sets contain peaks, and these peaks can have physical significance. A notable example is the peaks in the distribution of crustal ages obtained from a worldwide compilation of U-Pb zircon data [1]. The three main peaks at “magic ages” of 1.2 Ga, 1.9 Ga, and 2.7 Ga have been associated with major pulses in continental growth. More recently, using a compliation of helium isotopic data [2], it has been suggested that the histograms of 4He/3He ratios of ocean island basalts are also peaked, and that these peaks can be related to the crustal peaks [3]. Such a relationship would provide an important link between mantle depletion and crustal formation. These observations raise some important general statistical questions that are relevant for any compliation of geochemical data: How can we tell if a peak in a histogram is really there? How can we separate real peaks from peaks that arise as artifacts of the sampling process? Statistical methods for addressing these questions will be described, with a particular focus on the problem of bandwidth selection in kernel density estimation, and a method for identifying significant features known as SiZer [4]. These techniques have been used to reassess the helium-continental crust correlation, which it seems may not be supported by the current data. [1] Condie K.C. (1998) Earth Planet. Sci. Lett. 163, 97-108. [2] Abedini A.A. et al. (2006) USGS-NoGaDat http,//pubs.usgs.gov/ds/2006/202 [3] Parman S.W. (2007) Nature 446, 900-903. [4] Chaudhuri P. & Marron J.S. (1999) J. Amer. Stat. Assoc. 94, 807–823.

A811

Precise U-Th-Pb geochronology of carbonatites and mantle perturbations ALEXEI S. RUKHLOV Alberta Geological Survey, Alberta Energy Resources Conservation Board, 4999 – 98 Avenue, Edmonton, Alberta T6B 2X3, Canada ([email protected]) Found on every continent, carbonatites range in age from the Archean through to the Recent, and hense can provide important insights into the terrestrial evolution (Bell et al. 1982; Bell and Blenkinsop, 1987; Kwon et al. 1989; Bizzaro et al. 2002). Carbonatite magmatism repeatedly occurred in relatively restricted areas of several cratons, and the distinct age groupings were linked to different geological events (Kukharenko, 1967; Vartiainen & Woolley, 1974; Doig, 1970; Larsen et al. 1983; Nielsen, 1985; Dahlgren, 1994; Belyaev et al. 1976; Bell, 2001). Of the known 500 or so carbonatites, however, few have been precisely dated. Here we present new precise U-Pb, Th-Pb and Pb-Pb ages from several carbonatite complexes from Canada, Russia, and Finland which can be used to establish important reference points in defining mantle perturbations. Events at 2617 Ma and 586 Ma are now firmly established for the Baltic Shield, while several distinct events of 2680 Ma, and between 1897 1083 Ma are documented for the Canadian Shield. Other, less well defined, events in Canada include magmatism at 568 Ma from Manitou Islands and an event at 357 Ma from the Canadian Cordillera. East European Craton events also include carbonatitic magmatism at 2074 Ma, 1963 Ma, 1792 Ma and between 386 and 377 Ma. Most of these events can be correlated with extensive mafic magmatism generally considered to be the result of plume activity and associated continental fragmentation. Selected ages from carbonatite complexes are presented in the table below. Complex Siilinjarvi

Age ±2σ Ma 2617.4 ± 9.6

Method U-Pb zircon

Cargill Goldray Tiksheozero Chernigovka Firesand River Valentine Manitou Islands Turiy Peninsula

1896.8 ± 1.4 1886.0 ± 0.9 1959 ± 16 2074.4 ± 6.3 1142 ± 23 1114.7 ± 1.1 568 ± 24 377.0 ± 3.7

Kandaguba

385.6 ± 3.1

Seblyavr

382.3 ± 5.3

U-Pb baddeleyite U-Pb zircon U-Pb zircon U-Pb zircon U-Pb kimzeyite U-Pb baddeleyite Pb-Pb isochron zircon 206Pb*/238U baddeleyite Th-Pb isochron calcite-apatite Th-Pb isochron carbonates

Goldschmidt Conference Abstracts 2008

A812

Reactive transport of salicylate in a goethite-coated sand column B. RUSCH, K. HANNA* AND B. HUMBERT LCPME/CNRS-Nancy University, 54600, Villers-lès-Nancy, France (*correspondence: [email protected]) ([email protected])

Methodology The salicylate sorption at oxide/water interface and correlation with the hydrodynamic transport parameters is the main objective of this study. For this purpose, the surface complexation reactions were examined under static and dynamic conditions by conducting batch and column tests, vibrational spectroscopy and modeling approach.

Results and Discussion Prior to salicylate sorption experiments, the surface properties of the synthesized Goethite-coated quartz were determined. Sorption isotherm, sorption edges, solid-tosolution ratio effect on sorption were studied and described using macroscopic models and the MINTEQ database incorporated in PHREEQC-2 [1]. 7,0

1,0

6,5 6,0 Fe

0,6

O

5,5

≡FeOH + HL- + H+ ↔ FeHL + H2O

O HO

≡FeOH + HL- ↔ FeOL3- + 2H+

5,0

pH

Concentration C/C0

0,8

O

0,4

4,5

Fe O

4,0

HO

0,2

3,5 0,0

3,0 0

10

20 30 Volume V/Vp

40

50

Figure 1: Breakthrough curve of salicylate (♦) and modeling results (—); pHoutflow (--). The hydrodynamic parameters of the column were determined by a non-reactive tracer injection and by using the Convection-Dispersion Equation (θ = 0.58 cm3/cm3; D = 0.0048 cm²/min; λ = 125 µm). Column breakthrough curves displayed two steps at high flow rate (0.3 cm/min). On the basis of spectroscopic observations (Raman and Infrared), the mononuclear salicylate surface complexes (monodentate and/or bidentate) are identical at any height in the column. However, a binuclear monodentate complex is expected in batch. Assuming one type of site and two surface complexation reactions per site, the transport of salicylic acid in a goethite-coated sand column can be described by surface complexation modeling. [1] Parkhurst D.L. & Appelo C.A.J. (2000) PHREEQC (Version 2)

Assessing parent magma compositions and H2O contents in the central Oregon Cascades D. RUSCITTO1*, P. WALLACE1, A. KENT2 AND I. BINDEMAN1 1

Dept. of Geological Sciences, Univ. of Oregon, Eugene, OR 97403, USA (*correspondence: [email protected]) 2 Dept. of Geosciences, Oregon State Univ., Corvallis, OR 97331, USA We analyzed olivine–hosted melt inclusions in mafic tephra deposits from six volcanic centers located in the central Oregon High Cascades by FTIR, LA–ICPMS, and EMP to relate pre–eruptive volatile contents (H2O, CO2, S, Cl) to major and trace element compositions and mineralogy. Host olivine crystals (Fo77–86) have δ18O = 5.35 ±0.2‰; values typical for mantle–derived magmas. Melt inclusion compositions have 49–54 wt% SiO2 and 4.3–7.2 wt% MgO. Maximum H2O and CO2 contents range from 1.6–3.8 wt% and 860–1500 ppm, respectively and suggest maximum pressures of inclusion entrapment between 2–4 kbar. Magmatic H2O concentrations are lower in tholeiitic compositions than in calc–alkaline samples. High incompatible element abundances relative to N–MORB indicate that these magmas formed by melting of an enriched mantle source that was overprinted with even higher abundances of incompatible and fluid mobile elements derived from the underlying subducted slab. Geochemical trends, petrographic observations, and correlations between maximum H2O contents and major elements suggest that these melts have experienced multiphase crystal fractionation (olivine and plagioclase dominated) during transit from the mantle to the surface. We investigated a variety of models, including equilibrium olivine and plagioclase addition, forward crystallization modeling from whole rock primitive compositions from the Cascades, and least–squares mixing calculations to assess total extents of crystallization. Minimum (olivine–only) and maximum (~50:50 ol+plag) estimates of crystallization range from 11–17% and 31–40%, respectively. After correction for the effects of crystallization, we estimate that the original magmas contained 1.2–2.8 wt% H2O. Batch melting calculations for parental compositions suggest relatively large degrees of melting (15–23%) of an enriched mantle source. Comparison with wet mantle melting models indicate maximum temperatures in the mantle wedge that are near or above the dry peridotite solidus [1]; these temperature and melting estimates suggest that flux–melting may be important beneath the central Oregon High Cascades. [1] Portnyagin et al. (2007) EPSL 225, 53–69.

Goldschmidt Conference Abstracts 2008

Distribution of Pt, Os, Ir during liquid metal segregation under extremely reducing conditions

Provenance of early Paleozoic siliciclastic sandstones from NW Argentina – Results from an ongoing study on detrital zircons

TRACY RUSHMER GEMOC, Department of Earth and Planetary Sciences, Macquarie University, NSW 2092, Sydney, Australia ([email protected]) Experimental deformation studies have been performed on a natural, partially molten H6 ordinarly chondrite (Kernouvé) to enhance our understanding of early differentiation processes. One of the goals of the experiments is to determine the partitioning of siderophile elements during deformation and dynamic liquid metal segregation. We report results from a set of experiments in which deformation was performed under extremely reducing conditions. The study was conducted at temperatures between 925°C and 950°C, at 1.3 GPa confining pressure with a strain rate of 10-4/s. Major element analyses show that in zones of deformation the compositions of both silicate and metal phases are considerably modified. Metallic compositions include (Mg,Fe)S, which is found with Si-bearing FeNi metal (+/- P), and Fe-Ni-S quench metal. Texturally, Fe-Ni-S appears to have been liquid and lines grain boundaries and cracks. Fe-NiSi compositions are found in the shear zones produced during the deformation experiment. Si-bearing FeNi metal and (Fe,Mg,Ca,Mn)S are found with silicate glass, forsterite (Fo92-96) and enstatite (En98). We also report highly siderophile element (HSE) concentrations measured in the Fe-Ni, FeS and Fe-Ni-Si metallic phases by LA-ICPMS. Earlier work has shown that the D’s of many of the HSE are dependent on the sulfur content of the liquid metal with HSE concentrated in the Fe-Ni solid rather than FeNiS liquid. This also holds true under these reducing conditions however, Pt (not analyzed in earlier runs due to the use of Pt jackets) behaves noticeably differently to either Os or Ir. While Os and Ir are predictable in their distribution, Pt is highly concentrated in the modified Fe-Ni-Si +/- P bearing metal. D values for Pt, Os and Ir between FeNi (solid) and FeS (liquid) outside of the shear zones are approximately 20 for Os, to 70 for Ir and slightly lower for Pt and Au at 4 and 2.5 respectively, as expected. In the shear zones where the metal is reduced, D values for Os and Ir between FeNiSi +/- P and FeS liquid are as with FeNi and FeS, ranging from 11 and 55 for Os and Ir. Au also remains similar at 3.3. Pt, however, jumps consistently to much higher values. These data suggest that Si and possibly P play and important role in concentrating Pt. Further experiments in idealized systems are needed.

A813

T. RÜSING1, C. AUGUSTSSON1, E. KOOIJMAN2, J. BERNDT2 AND U. ZIMMERMANN3 1

Geologisch-Paläontologisches Institut, Westfälische Wilhelms-Universität, Corrensstrasse 24, 48149 Münster, Germany ([email protected]) 2 Institut für Mineralogie, Westfälische Wilhelms-Universität, Corrensstrasse 24, 48149 Münster, Germany 3 Department of Geology, University of Johannesburg, South Africa Paleogeoraphic reconstructions can be greatly aided by provenance studies on sedimentary rocks. We will present initial results of an ongoing varietal study on detrital zircons from siliciclastic sandstones of the Cambrian Mesón Group, Ordovician Purna Turbidite Complex and Las Vicuñas Formation and the Silurian Salar del Rincón Formation in NW Argentina (ca. 23-24°S, 65-67°W). The sediments were deposited along the western Gondwana margin, with both tectonically active and passive margin phases. Cathodoluminescence images of the zircon grains give insight into grain morphology, zoning and growth phases and we use them for a first approach to classify the zircons. This analysis also gives insight into the magmatic or metamorphic history of the grains. The majority of the zircons in the studied units are 50-150 µm long, slightly abraded to rounded and have oscillatory zoning, typical for a magmatic origin. Zircons with more than one growth phase are rare. The degree of rounding point to low and high degrees of abration for different grains, which possibly can be indicative of detrital input both from local and regional source areas. First LA-ICPMS in situ U-Pb data for 50-100 zircons in each of 5 analysed samples indicate three distinguished U/Pb age-peaks at ca. < 550 Ma, ca. 600850 Ma and ca. 1.2-1.6 Ga. These peaks can be observed throughout most sandstones of the different ages and from varying locations, that have been analysed so far. The results imply a long and continuous influx of clastic material from crustal source areas in Gondwanan realms.

A814

Goldschmidt Conference Abstracts 2008

Melting of an enriched lithology on the SE Indian Ridge and possibly other ridges from Pb-Th-U isotopes

Sorption characteristics of dissolved organic Phosphorus onto Iron (oxy)hydroxides in seawater

C.J. RUSSO1, K.H. RUBIN1 AND D.W. GRAHAM2

K.C. RUTTENBERG* AND D.J. SULAK

1

SOEST, Univ. of Hawaii, Honolulu, HI 96822, USA (*correspondence: [email protected]) 2 COAS, Oregon State University, Corvallis, OR 97331, USA Our recent studies of compositional variations along the eastern Southeast Indian Ridge (SEIR; i.e. east of Kerguelen and Amsterdam-St. Paul hotspots) reveal along-axis gradients in (230Th/232Th) and Th/U (κ), with high (230Th/232Th) and low κ values occurring along the western SEIR, and low (230Th/232Th) and high κ in the east near the AustralianAntarctic Discordance. Similar along-axis gradients are observed in other mantle tracers such as time integrated Th/U (κPb; calculated from the radiogenic component of measured 208 Pb/204Pb and 206Pb/204Pb [1]), as well as 3He/4He [2]. But κPb variations are much smaller than κ variations and they converge to κ≈κPb≈4.0 in the east; in the west κ<<κPb. The high κ and κPb endmember sits on the geochron, indicative of a long time integrated history of coherent Th-U-Pb abundances as a closed system at roughly the Bulk Earth value. These along-axis chemical gradients indicate that melts of the enriched, high Th/U, mantle dominate eastern SEIR MORB and, moving west, are variably mixed to a lesser degree with melts of a relatively low Th/U, depleted MORB mantle-like (DMM) component. The highest κ and κPb, lowest 3 He/4He lavas are also lower in (230Th/238U). Melt modeling using the constant upwelling rate applicable to this portion of the SEIR suggests that the high Th/U mantle melts much more extensively and/or rapidly than the DMM component. It therefore has the potential to dominate other radiogenic isotope and incompatible trace element compositions of MORB in the region, even if it comprises a volumetrically smaller fraction of the mantle. Similar κ and κPb values are highly unusual in MORB, but do rarely occur at some hotspots, such as Reunion [3] and Samoa [4], indicating that mantle with this composition is not unique to the eastern SEIR. It may even exist beneath other ridges, allowing one to speculate that such mantle may play a role in melt supply fluctuations along other ridge systems that are not easily rationalized by upwelling rate variations. [1] Mahoney et al. (2002) J. Petrology 43, 1155-1176. [2] Graham et al. (2001) Nature 409, 701-703. [3] Vlastelic et al. (2006) EPSL 248, 379-393. [4] Sims & Hart (2006) EPSL 245, 743-761.

Department of Oceanography, University of Hawaii, Honolulu, HI 96822, USA (*correspondence: [email protected]) ([email protected]) Sorption of phosphorus (P) onto particulate surfaces significantly influences dissolved P concentrations in aquatic environments. We present results of a study contrasting the sorption behavior of several dissolved organic phosphorus (DOP) compounds and phosphate onto three commonly occurring iron (oxyhydroxides): ferrihydrite, goethite, and hematite. The DOP compounds were chosen to represent a range of molecular weights and structures, and include: Adenosine tri-phosphate (ATP), Adenosine Monophosphate (AMP), Glucose-6-phosphate (G6P), and Aminoe Ethyl Phosphonic Acid (AEP). All DOP compounds and phosphate displayed a decreasing trend of sorption density as a function of the degree of crystallinity of the iron (oxy)hydroxide substrate, with ferrihydrite adsorbing the most, hematite the least, and goethite adsorbing a quantity intermediate between these two end-members. The affinity of phosphomonoesters ATP and AMP exceeded that of G6P, and AEP was intermediate. Without exception, DOP compounds showed less affinity for sorption than did free orthophosphate. This suggests that in aquatic systems enriched in reactive iron minerals, as suspended particulates in the water column or in benthic sediments, DOP bioavailability may become enhanced relative to orthophosphate. Since uptake of P from DOP requires enzymatic cleavage of orthophosphate, a system enriched in DOP relative to orthophosphate may impact community structure of the ecosystem.

Goldschmidt Conference Abstracts 2008

Noble gases in pumices: Magmatic or atmosphere-derived?

A new approach for estimating more accurate groundwater gas concentrations

L. RUZIÉ AND M. MOREIRA Institut de Physique du Globe de Paris, Laboratoire de Géochimie et Cosmochimie, 4 pl. Jussieu, 75005 Paris, France ([email protected]) Noble gases are chemically inert and therefore can only fractionate during physical processes such as diffusion. For that reason, they provide strong constraints on degassing processes. Pinti et al. [1] have studied noble gases trapped in pumices from Japanese and Lipari volcanoes. They suggested that the noble gas signature was linked to a mechanism of air incorporation during the eruption. Thus, noble gases could represent a potential tool for understanding the dynamics of the plinian eruptions. We have decided to focus on wellstudied eruptions to improve our knowledge on the origin of rare gases trapped in pumice. Therefore, we measured by mass spectrometry noble gas abundances and isotopic compositions in pumices from Martinique island (Lesser Antilles), and from several volcanoes around the world. Gases were extracted by step crushing under vacuum and then purified and analyzed with our mass spectrometers. All pumices show enrichments (up to 40 times) in neon over argon compared to the atmospheric ratio. Krypton and xenon abundances vary with the different arcs, but are similar for a same eruption. These features are explained by a model of fractional degassing with several steps from the mantle wedge towards the surface. When the magma fragmentation occurs in the magmatic chamber or in the conduit, pumices are formed and trap the gases in their vesicles. Then, pumices spend few seconds in the plinian column before being deposed. We therefore have a different interpretation than the one of Pinti et al. on the origin of noble gases in pumices. Rather than a contamination during eruption in the atmosphere with preferential incorporation of neon, we suggest than pumices can trap gases during magma fragmentation. [1] Pinti & Wada et al. (1999) Journal of Volcanology & Geothermal Research 88, 279-289.

A815

C. RYAN1, K.L. MCLEISH2, J.W. ROY3 AND A. CHU2 1

Dept Geoscience, University of Calgary, Calgary AB ([email protected]) 2 Dept Civil Engineering, University of Calgary, Calgary AB ([email protected]) 3 Water Science and Technology, Environment Canada, Burlinton, ON ([email protected]) We combine a passive gas diffusion sampler with fieldmeasured total dissolved gas pressures (TDGPs) to provide improved estimates of groundwater gas concentrations. The passive gas diffusion sampler is deployed downhole forperiod of up to 3 months. Upon recovery, the gas sample is isolated in a gas-tight syringe for transport to the laboratory where it can be directly injected into a gas chromatograph. This sampling and analytical approach was used for a wide variety of geochemical environments, and appeared to be less prone to systematic bias than a traditional sampling approach. Gas concentrations estimated from gas chromatographic analysis were multiplied by field-measured TDGPs. Groundwater in Alberta was consistently ‘gas-charged’ (i.e. TDGP greater than atmospheric). Total dissolved gas pressures were typically higher when biogeochemical gases (e.g. CO2 and CH4) were observed. When the groundwater in monitoring wells was isolated from the atmosphere using a packer, substantially high TDGPs (more than 2000 mm Hg) were measured. This suggests groundwater gas concentrations are being routinely underestimated.

A816

Goldschmidt Conference Abstracts 2008

Li and B isotopes as tracers for crust/mantle vs. slab influences in western Mexico

Chemical weathering of carbonates and silicates in the Han River basin, South Korea

JEFFREY G. RYAN1, IVAN P. SAVOV2,3 AND STEVEN SHIREY3

J.-S. RYU1, K.-S. LEE1, H.-W. CHANG2 AND S.H. SHIN1

1

Dept Geology, Univ. South Florida, Tampa, FL 33620 ([email protected]) 2 School of Earth and Environment, Leeds University, Leeds LS2 9JT, UK 3 DTM-Carnegie Institution of Washington, 5251 Broad Branch Rd. NW, Washington DC 20015 While both B and Li are enriched in sediments and altered ocean crust reaching trenches, their signatures in continental arcs often reflect diverse source and process effects. In the western Trans-Mexican volcanic belt (TMVB), a young, hot downgoing plate minimizes inputs of low-T° slab components, permitting the study of mantle and crustal contributions to the Li and B signatures of erupted lavas. B contents and B/Be ratios are low (<10 ppm; B/Be < 10) in mafic TMVB lavas, while evolved lavas reach values ≥ 50 ppm, indicating B-rich crustal inputs [1]. δ11B in TMVB lavas are surprisingly high, with values up to +10.6 ‰ in some Colima Graben minettes. This heavy δ11B signature, observed primarily in rift-related alkaline lavas, may reflect relict slab B remobilized during extension, as the low P/T° conditions under which high δ11B boron can be transferred from slab to mantle are unlikely to occur during TMVB subduction. Evolved lavas have markedly lower δ11B ( -1.6 to -9.9‰), indicating that high B crustal rocks can have very low δ11B signatures, so crustal assimilation can result in lowered δ11B. All TMVB lavas thus far examined seem to reflect mixing between high δ11B alkaline lava (old slab?) and low δ11B crustal end-members. In contrast, Li and δ7Li are insensitive to either present or older slab influences in the western TMVB. Li contents are similar in mafic lavas from across the arc, and while Li/Yb ratios are elevated in western TMVB centers, this appears to reflect Yb depletions due to residual amphibole ± garnet during melting. δ7Li in the western TMVB, at +3 to +5‰, is similar to arcs globally [2], and likely indicates that here, as elsewhere, the upper mantle Li reservoir dominates, and/or the signature for slab-derived Li is not markedly different [3,4]. Evolved TMVB lavas move to somewhat lower 7Li, reaching values similar to those measured in entrained crustal xenoliths, and approaching average crustal δ7Li [5]. [1] Hochstaedter et al. (1996) GCA. [2] Tomascak et al. (2002) EPSL. [3] Tomascak et al. (2000) Geology. [4] Benton et al. (2004) G3. [5] Teng et al. (2005) Chem Geol.

1

Division of Earth and Environmental Science, Korea Basic Science Institute ([email protected], [email protected], [email protected]) 2 SEES, Seoul National University ([email protected]) A detailed investigation of the fluvial geochemistry of the Han River system allows us to estimate the rates of chemical weathering and the consumption of CO2. The Han River drains approximately 26,000 km2 and is the largest river system in South Korea in terms of both water discharge and total river length. It consists of two major tributaries: the North Han River (NHR) and the South Han River (SHR). Distinct differences in basin lithology (silicate vs. carbonate) between the NHR and SHR provide a good natural laboratory in which to examine weathering processes and the influence of basin geology on water quality. The concentrations of major elements and the Sr isotopic compositions were obtained from 58 samples collected in both summer and winter along the Han River system in both 2000 and 2006. The concentrations of dissolved loads differed considerably between the NHR and SHR; compared with the SHR, the NHR had much lower total dissolved solids (TDS), Sr, and major ion concentrations but a higher Si concentration and 87Sr/86Sr ratio. A forward model showed that the dissolved loads in the NHR came primarily from silicate weathering (55 ± 11%), with a relatively small portion from carbonates (30 ± 14%), whereas the main contribution to the dissolved loads in the SHR was carbonate weathering (82 ± 3%), with only 11 ± 4% from silicates. These results are consistent with the different lithologies of the two drainage basins: silicate rocks in the NHR versus carbonate rocks in the SHR. Sulfuric acid derived from sulfide dissolution in coal-containing sedimentary strata has played an important role in carbonate weathering in the SHR basin, unlike in the NHR basin. The silicate weathering rate (SWR) was similar between the NHR and SHR basins, but the rate of CO2 consumption in the SHR basin was lower than in the NHR basin due to an important role of sulfuric acid derived from pyrite oxidation. This work was supported by a grant (code 3-2-3) from the Sustainable Resources Research Center of 21st Century Frontier Research.