Goldschmidt Abstracts 2007- S

Goldschmidt Conference Abstracts 2007

A864

U4+ and Pu4+ incorporation in zircon and their effect upon helium diffusion in the host lattice I. SAADOUNE AND N.H. DE LEEUW Department of Chemistry, University College London, 20 Gordon Street, London, UK, WC1H 0AJ ([email protected], [email protected])

H.-K. SAARI, S. SCHMIDT AND G. BLANC

Zircon mineral (ZrSiO4) is a potential host matrix for the encapsulation of highly radioactive materials, such as dismantled weapons and nuclear waste from power stations (Ewing, 2001). We report a detailed computer modelling study, based on interatomic potentials, to address two fundamental issues concerning radiation processes in the zircon mineral: first, the physics and chemistry of U4+ and Pu4+ inclusion in the crystalline matrix of zircon and, second, the effect of U4+/Pu4+ dopants on the incorporation and diffusion behaviour of helium in zircon. Our results show that U4+ and Pu4+ can incorporate isomorphously in zirconium sites (Zr4+). The solution energies of U4+ and Pu4+ are 0.18 eV and 0.03 eV respectively, which indicates that Pu4+ is more readily incorporated as a dopant than U4+. However, the positive solution energies show that the inclusion of the two species in the crystalline matrix would occur only under high temperature conditions. Interstitial incorporation of helium in zircon is an energetically unfavourable process, which causes strong deformation in the host lattice due to movement of atoms in the channel walls to accommodate He. Moreover, the presence of Pu4+ and U4+ dopants in the zircon lattice has a direct effect upon the energetics of helium diffusion between interstitial sites. For instance, activation energies calculated for He diffusion are dependent upon the distribution of U4 and Pu4+ in the lattice, and can differ by as much as 0.2 eV from those calculated for the perfect zircon. 3 2.5

E-Es (eV)

2 1.5 1

0.5 0 -0.5

0

0.2

0.4

0.6

0.8

Short-lived naturally occurring radioisotopes (234Th, 7Be, 210Pb) as a tracer for particle transport in the Gironde fluvial-estuarine system (France) UMR CNRS 5805 EPOC, Université Bordeaux1, Avenue des Facultés, 33405 Talence Cedex, France ([email protected], [email protected], [email protected]) Human activities release various chemical species to the environment. Many of these substances bind strongly to sediments, which are then delivered to rivers. Thereafter the fate of pollutants in fluvial environments is linked to particle transport. The short-lived, naturally occurring radioisotopes, 234 Th (T1/2 = 24.1 days), 7Be (T1/2 = 53.3 days) and 210Pb (T1/2 = 22.3 years), can be used as potential tracers to investigate particle transport from land to sea. However, up to now, their use is still rare and for 234Th mainly devoted to marine environments. Uranium (238U) forms 234Th through radioactive decay in the dissolved phase. However, due to the typically low and invariable concentrations of 238U in riverine environments, the formation of 234Th is often insufficient to be used as a tracer. The programme ARTTE “Application Radioisotopes as Tracers of particle transport in environment”, has been carried out to develop the application of 234Th, 7Be and 210Pb in the polluted (e.g. Cd, Hg (e.g. Schäfer et al., 2006)) Gironde fluvial-estuarine system (South-West France). A preliminary investigation in 2005 has revealed adequately elevated and spatiotemporally stable dissolved 238U activities within this system (Saari et al., submitted), allowing the 234Th application. 234 Th, 7Be and 210Pb activities in suspended materials were measured monthly (2006-2007) by J-spectrometry within the Garonne and Lot Rivers, and the Gironde estuary. The first results reveal: (1) 7Be/210Pbxs activity ratios decrease from river to estuary, due to aging of sediments and/or to resuspension of old bottom sediments (2) 234Thxs activities show large variabilities with time, from negligible values up to about 300 mBq g-1, probably due to differences of particle residence time in the river channels.

1

References E-Es(pure)

E-E(U0.0)

x

E-E(U0.5)

E-Es(U0.25)

Figure 1: Diffusion pathways for helium movement along the x-direction of U-doped zircon.

References: Ewing R.C. (2001), The design and evaluation of nuclearwaste forms: Clues from mineralogy. Canadian Mineralogist, 39 697-715.

Schäfer, J., Blanc, G., Audry, S., Cossa, D., Bossy, C., (2006), Appl. Geochim. 21, 515-527. Saari, H.-K., Schmidt, S., Huguet, S., Jour. of Environ. Radioact. Submitted.

Goldschmidt Conference Abstracts 2007

A865

The Formation of the Fe skarn deposit between Camibogazi and Arnastal (Gumushane, NE Turkey): Evidence from mineral chemistry

Modelling pore water B and G11B signatures in the shallow subduction zone forearc: Examples from Barbados, Costa Rica and N. Japan

M. B. SADIKLAR1, F. SIPAHI2, H.-J. BERNHARDT3 1 AND I. UYSAL

DEMIAN SAFFER 1 AND ACHIM KOPF 2

1

Jeoloji Müh., KTU, MF, Tr-61080, Trabzon; ([email protected]) 2 Jeoloji Müh., KTU, GMF, Tr-29000, Gumushane; ([email protected]) 3 Zentrale Elektronen-Mikrosonde, Ruhr University, D-4801, Bochum The Fe skarn deposit in the Camibogazi and Arnastal area is located about 50 km north of Gumushane in NE Turkey. The deposit is hosted by Late Cretaceous Carbonate rocks which comprise dolomitic and sandy limestone, and limestone, adjacent to the Eocene Zigana Granitoid [1]. Petrographically, granitoid consists of syenogranite, monzogranite, quartzmonzonite and granite. Fe skarn deposit has mineral paragenesis of vesuvianite (idocrase), phlogopite, diopside, andradite, actinolite, tremolite, epidote, quartz, magnetite, hematite and less pyrrhotine and pyrite. Vesuvianite, diopside and phlogopite mineralization are only seen in a very narrow field of the Camibogazi Plateau (altitude of c. a. 2450 m). Magnetite, hematite, andradite, actinolite, tremolite, epidote, quartz and less parroting and pyrite are seen in Aransas Plateau. Microprobe analyses indicate that majority of the skarn minerals are calcic and have high Mg/(Mg+Fe). Clinopyroxene consists of diopside (Di96.66-99.04 Hd0.67-2.80 Jo0.29-0.55). Vesuvianite is mostly Mg-rich, and has very low MnO content (< 0.1 %). Phlogopite has high and nearly constant Mg/ (Mg+Fe) ratios, 0.93 to 0.94. The Ba content of phlogopite is low. Magnetite is generally massive, and some of them are martitization. According to vesuvianite-diopside geothermometry, the formation temperatures of skarn minerals in Camibogazi range from 329 to 558 oC. The Fe exoskarn skarn deposit was characterized by early, high-temperature mineral assemblages dominated by anhydrous minerals diopside and garnet, and late lowtemperature assemblages with hydrous minerals vesuvianite, phlogopite and epidote. Types of clinopyroxene and garnet, the abundance of magnetite with minor pyrite, and widespread occurrence of epidote suggest that the Fe skarn deposit between Camibogazi and Arnastal Plateau was formed under relatively oxidized conditions.

Reference [1] Sipahi, F. (1996) MSc. Thesis, KTÜ FBE, Turkey (Unpub.).

1

Pennsylvania State University, 310 Deike Building, University Park, PA 16802, USA – ([email protected]) 2 Research Center Ocean Margins, Univ. Bremen, P.O.Box 330440, 28334 Bremen, Germany ([email protected]) At many subduction zones, pore water geochemistry from boreholes and mud volcanoes indicate a contribution from deep sources. These observations include pore water freshening, but enrichment in volatiles such as B and Li, and decreased G11B, G6Li, and G37Cl. Identifying the location and distribution of source regions for these tracers is one critical step toward characterizing subduction zone fluid transport. Here, we present a simple model to combine (1) heating and compaction that accompany progressive burial of sediment in subduction zones with (2) previously published laboratory experimental data that constrain the distribution coefficient (Kd) for B in marine sediments as a function of temperature, to quantify the expected distribution of [B] G11B within bulk mudstones in subduction zones. In a generic subduction zone with a taper angle of 8º, we evaluate two end-member cases: a “cold” (60 mW m-2) and a “warm” (120 mW m-2) scenario. For the first, simulated values of [B] within the subducted sediment at 30 km from the trench range from 650–1200PM and values of G11B range from 23.4-30.6‰. At 60 km from the trench, [B] ranges from 1010-3340PM and G11B from 18.025‰. For the warm scenario, simulated [B] is 970-2400PM at 30 km and 2250-11480PM at 60 km; G11B ranges from 19.2– 25.2‰ at 30 km and 15.92-19.5‰ at 60 km. These signatures are generally stronger than those observed in pore fluids at shallow depths, as would be expected due to probable readsorption of some B during updip or vertical advection. Ultimately, our results can be used for hydrologic models to quantitatively assess the fluid flow rates and permeabilities required to transport B at rates high enough to produce the observed signature.

A866

Goldschmidt Conference Abstracts 2007

The formation of calcium perovskite from majoritic garnet – Implications for splitting of the 520 km seismic discontinuity

Fulgurites from Garuamukh, Assam, India – FTIR spectroscopic study BHASKAR J. SAIKIA1, G. PARTHASARATHY2, N.C. SARMAH1 AND G.D. BARUAH1 1

A. SAIKIA, D.J. FROST AND D.C. RUBIE Bayerisches Geoinstitut,Universität Bayreuth, D-95440 Germany, ([email protected]) Global seismic observations show complexity in the region of the 520 km seismic discontinuity (Deuss and Woodhouse 2001). In some regions of the mantle this discontinuity seems to be split into two discontinuities at approximately 500 km and 560 km. In addition to the wadsleyite to ringwoodite transition, the exsolution of calcium perovskite (Ca-Pv) from majoritic garnet (gt) may be capable of producing a seismic discontinuity. This, however, would require the reaction to occur over a sufficiently narrow depth interval and to cause a significant change in elastic properties. We have experimentally studied the calcium perovskite forming reaction in high pressure and temperature multi anvil experiments to determine if it is likely to cause an observable discontinuity. Experiments were performed by equilibrating Ca-free majorite gt with Ca-Pv to determine the saturation level as a function of pressure, temperature and garnet majorite content i.e. (Al/Mg+Si). Experimental results show that because the reaction results in a strongly non-linear yield of Ca-Pv with pressure, a significant portion of the reaction is completed over a quite narrow initial depth interval. The sound velocity changes over this interval for a fertile peridotite composition are smaller but of a similar magnitude to those of the wadsleyite-ringwoodite transition and would likely cause a weak discontinuity at approximately 560 km in line with seismic observations. If, however, the mantle contained a significant component of recycled basaltic crust the higher mantle Ca contents would result in a much stronger discontinuity. For this reason the feature observed at 560 km will be sensitive to the proportion of basaltic components in the mantle at transition zone conditions and is therefore an indicator of chemical heterogeneity. References A. Deuss, J. Woodhouse, (2001), Science 294, 354-357.

Department of Physics, Dibrugarh University,Dibrugarh-786 004, India: ([email protected]) 2 National Geophysical Research Institute, Uppal Road, Hyderabad-500 007, India ([email protected]) There are two phenomena that are responsible for making natural glass on Earth: meteorites and lightning. Glass that is made as a result of the collision of a meteorite with the Earth's surface is called meteoritic glass or tektite. Glass (a glassy object, to be exact) that is made as a result of a cloud-toground lightning discharge is called a fulgurite (from the Latin "fulgur" which means lightning). Fulgurites come in a great variety of forms and can be viewed as nature's own works of art. It is worth noting that lechatelierite (natural silica glass) is not present in obsidian, a glass-like material associated with volcanic activity. On the other hand, volcanic activity is known to generate lightning which, if it strikes sandy soil, may produce a fulgurite.The black-brown natural glass found at Garuamukh near Nagaon town (Lat.26º19´N, Long. 94º30´E, Assam, India) is due to the fourth category of formation . The colloidal material comes out from the earth’s surface making a number of branching with fire and foul odour smoke and the upper part of it is gradually solidified to natural glass within a few hours. The solid material have been found to have a temperature of about 80 ºC even after a lapse of two days after occurrence of the event whereas the room temperature during that period was 28 ºC. The increasing of temperature is observed on the spot with the increasing of depth. In order to understand the possible meteoritic or fulgurite origin of the glassy materials, we made an attempt to characterize the sample by using FTIR spectroscopy, and powder x-ray diffraction. The present investigation of the naturally occurring glass sample reveals the characteristic features of amorphous silica with some other minerals as trace. As the structural change taking place in the octahedral symmetry in the formation of the sample so, the possibility of producing this silicate glass is by ultra high- temperature that is higher than the corresponding fusion point of the crystalline state, indicating the nature of a fulgurite.

Goldschmidt Conference Abstracts 2007

A867

Organic compounds in H5 meteorite: Spectroscopic investigation of Dergaon H5 Chondrite

Experimental study on gas emission using fault rock core penetrating Atotsugawa fault, central Japan

BHASKAR J. SAIKIA1, G. PARTHASARATHY2, N.C. SARMAH1 AND G.D. BARUHA1

TOMOHIKO SAITO AND HIDEMI TANAKA

1

Department of Physics, Dibrugarh University, Dibrugarh786 004, India ([email protected]) 2 National Geophysical Research Institute, Hyderabad500007, India ([email protected]) On March 2,2001, at 16:40 local time (GMT+5:30hrs), four fragments of a meteorite fell near Dergaon town (Lat.160 41’ N, long. 930 52’E, Assam, India. Based on the petrography and chemical composition of the stony iron Dergaon meteorite has been classified as H5 chondrite. But from the chemical composition, the previous workers reported that the meteorite is an unusual H chondrite. In this present work we analysed the sample with the help of FTIR and Diffused Reflection Spectroscopy and found the SiO4 tetrahedron at 800-1100cm-1 and aliphatic functional CH2 and CH3 groups in the spectral region around 2800-3000cm-1. In 2800 – 3000 cm-1 region (3.4 Pm region), the hydrated and anhydrous interstellar dust particles attribute CH3 and CH2 stretching vibrations of aliphatic hydrocarbons which is similar to the spectra found from the meteorite in this frequency region. The silicate grain in the interstellar medium and meteorites consist predominantly of olivine [(Mg,Fe) SiO4] or of non-crystalline silicates of intermediate compositions associated with organic matter. The organic matter can extract the poly atomic carbon based molecules by crushing the MgO and (Mg,Fe)SiO4 crystals. Generally four peaks are observed in this region. The symmetric CH2 stretching vibration at 2850 cm-1, symmetric CH3 stretching vibration at 2865 cm-1, asymmetric CH2 stretching vibration at 2922 cm-1 and asymmetric CH3 stretching vibration found at 2958 cm-1. The spectral features in the mid infrared region were found to be identical to that of the interstellar dust gain particles. Most of the strong IR absorption bands are due to olivine. Few weak absorption bands observed in the Dergaon meteorite are due to organic compounds present in the meteorite, the features observed in other H-chondrites by Lawless et al. (1972).

Department of Earth and Planetary Sciense, Univercity of Tokyo, Bunkyoku Hongo 7-3-1,Tokyo, Japan ([email protected], [email protected]) Crustal gas-anomalies with earthquakes have been observed around active faults. However few experimnetal studis have been done for this problem. Therefore fundamental processes are largely unknown. In order to examine the processes, grinding experiments were carried out under wet condition using two types of rocks taken from drilled core penetrating Atotsugawa fault: (1) Weakly Pulverized and Altered Rock (WPAR) and (2) Fault Gouge. Result of the experiments show that positive relationships between surface area (˂ S ) and amount of gas ( n ) (Fig.1). For hydrogen, the relationships can be expressed as: n v 'S  which is in good agreement with those reported by Kameda et al., (2003). For methane, the relationships can be expressed as: n A (1 EXP [k'S ]). The equation corresponds to solution: dn / d's k(A  n). A will be max amount of methane contained in a sample and k shows degree of evolution of gas when samples are crushed. This diferencial equation indictes that methane gas contained in the rock sample were released by fracturing and release ratio( dn /d's ) is proportional to (A - n), which means the amount of methane gas remains in the rocks.

Figure 1. Amount of released gases (methane and hydrogen) during wet grinding experiment.

Reference: Lawless, J. G., Folsome, C. E., and Kvenvolden, K. A. (1972) Organic matter in meteorites. Scientific American 226, 3846

Microstructural observation of thin sections clarified that fluid inclusions are abundant in quartz or calcite in granitic rocks, metasediments, and less in calcarious rock, which released methane gas by grinding. On the contrary, no fluid inclusion is observed in the minerals of andesite that did not released methane gas by grinding. Combining experiment results and microstructural observations, we concled that methane gases were included in fluid inclusion and released by fracturung minerals.

References Kameda J., Saruwatari K., Tanaka H., (2003). Geophys. Res. Lett., 30, doi:10.1029/2003GL018252 Jiang F.L., and Li, G.R., (1981). Geophys. Res. Lett., 8, 473476.

A868

Goldschmidt Conference Abstracts 2007

Adsorption behavior of gold (I and III) complexes on manganese dioxide S. SAKOMOTO1, K. YONEZU1, T. YOKOYAMA2, Y. OKAUE2, A. IMAI1, AND K. WATANABE1 1

Dep. of Earth Resources Engineering, Kyushu University. 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan. [email protected] 2 Dep. of Chemistry, Kyushu University, 4-2-1 Ropponmatsu, Chuo-ku, Fukuoka 810-8560, Japan. It is essential in geochemistry to elucidate the uptake mechanism of trace elements into manganese nodules. It is well known that transition metals such as Co, Cu and Zn are concentrated in manganese nodules. However, only a few investigations have conducted on the uptake of noble metals such as Au, Pd and Pt. It is recognized that adsorption of metal ions on manganese dioxide was investigated to explore their uptake mechanism into marine manganese nodules. In this study, a model compound of manganese nodule was manganese dioxide. As model compounds of gold, [Au(S2O3)2]3- and [AuCl4-n(OH)n]- complexes were selected for convenience because gold(I) and gold(III) ions are present in hydrosphere. The adsorption behavior of [Au(S2O3)2]3- and [AuCl4n(OH)n] complexes on the surface of manganese dioxide was investigated with pH. The adsorption behaviors were quite different. The percentage of [Au(S2O3)2]3- adsorbed increased with decreasing pH (from 8 to 4). Above pH 8, which is the same pH of ocean water, no [Au(S2O3)2]3- was adsorbed on manganese dioxide. Since the isoelectric point of the manganese dioxide was around pH 4-5, the surface negative charges increase with increasing pH above pH 4-5. This opposite correlation between the surface charge and percentage of [Au(S2O3)2]3- adsorbed suggests that the adsorption of [Au(S2O3)2]3- may be controlled by electrostatic force. On the other hand, the percentage of [AuCl4-n(OH)n]adsorbed increased from pH 4 to 6 and showed a maximum at pH 6-8. This result indicates that the adsorption of [AuCl4-n(OH)n]- occurred by overcoming the electrostatic repulsion. In addition, the chemical state of gold adsorbed on manganese dioxide was examined by XPS. The results showed that [Au(S2O3)2]3- and [AuCl4-n(OH)n]- complexes adsorbed on manganese dioxide were spontaneously reduced to elemental gold. [AuCl4-n(OH)n]- may be specifically adsorbed by condensation reaction (the formation of Au-O-Mn bond) between two OH groups in the [AuCl4-n(OH)n]- and two OH groups on manganese dioxide. The adsrobed [AuCl4-n(OH)n]may be distorted due to the regulation by the two OH groups on the manganese dioxide, and then its redox potential may be changed and reduced by water molecules. In contrast, the adsorbed [Au(S2O3)2]3- may disproportionate to form Au(0) and Au(III) species. The resulting Au(III) may be reduced to Au(0) by the above same mechanism. Based on the results of this study, the chemical state of gold in marine manganese nodules should be investigated spectroscopically.

Dissolution and precipitation kinetics of magnesite at conditions relevant for CO2 geologic sequestration GIUSEPPE D. SALDI, ERIC H. OELKERS, OLEG S. POKROVSKY AND JACQUES SCHOTT Géochimie et Biogéochimie Experimentale, LMTG, Université de Toulouse, CNRS, IRD,OMP. 14 ave. Edouard Belin, F-31400 Toulouse, FRANCE. ([email protected]) Understanding carbonate mineral reactivity is of critical importance to assessing and optimizing potential CO2 storage sites (c.f. Oelkers and Schott, 2005). In particular, carbonate dissolution and precipitation rates play a fundamental role in CO2 mineral fixation. To address this need, the dissolution and precipitation rates of magnesite (MgCO3) have been studied as a function of pH (2MgOH2+ concentration, inhibits magnesite dissolution. In addition, a significant increase in dissolution rates is observed with increasing reactive solution ionic strength: rates measured in 1 mol/kg NaCl solutions are on average 5 times faster than those observed in < 0.1 mol/kg NaCl solutions at a pH of ~ 7. Magnesite is observed to precipitate at 150°C only at relatively high degrees of supersaturation. For example measured magnesite precipitation rates are found to be 2x10-12 only at a saturation index š of ~ 30, when the CO2 pressure is at atmospheric or lower values. Moreover, the degree of supersaturation required for magnesite precipitation is observed to decrease with increasing temperature.

References Oelkers E.H. and Schott, J. (2005) Chem. Geol. 217, 183-186. Pokrovsky O.S., Schott J. and Thomas F. (1999) Geochim. Cosmochim. Acta 63, 863-880

Goldschmidt Conference Abstracts 2007

Natural and experimental multiscale F diffusion profiles in biotites from the Serido micaschists, Northeastern Brazil 1

2

R. SALLET , J.D. PRICE , R. MORITZ

3

2

AND D.A. WARK

1

Federal University of Rio Grande do Norte, Brazil, ([email protected]) 2 Earth and Environmental Sciences, Rensselaer Polytechnic Institute, USA, ([email protected], [email protected]) 3 University of Geneva, Department of Mineralogy, Switzerland, ([email protected]) We examined the F distribution in Seridó Fold Belt biotite at the regional, contact, and grain scale through electron probe microanalysis. At the regional scale biotites show largely homogeneous F contents varying from 0.30 to 0.40wt%. The uniform concentration suggests the biotites equilibrated with F-poor fluids (log fH2O/fHF = 4.5 to 5.0) at upper amphibolite metamorphic conditions. At the contact with pegmatoid intrusions, schist biotites are enriched at different levels F= 0.5 to 2.5wt%, similar to those observed for biotites in the pegmatoid biotites. This suggest that schist biotites were locally reequilibrated at 600-700°C by pegmatoid related F-rich fluids (logfH2O/fHF = 3.8 to 4.3). At pegmatoid contacts the F contents of schist biotites decrease to background contents in metric and centimetric diffusion profiles. At the Areias pegmatoid sill the F content of biotites decrease from 0.95wt% to the regional background F=0.30wt% in 2.5m. In the same distance the F contents of amphiboles decrease from 0.47 to 0.32wt% and apatites show a near constant content F=2.8-3.0wt%. At the Pedra Lavrada pegmatoid stock the F content from biotites in a schist enclave decrease from F= 0.43 to 0.37wt% in the narrow range 20 cm. The obtained diffusion profiles can be related to the interplay of three major factors: (i) fluid fH2O/fHF chemical gradient, (ii) thermal gradient and (iii) fluid advection into micaschists. AFSQ-buffered experiments on a mixture of two Seridó biotites (F=0.3 and 2.5wt%) at 750ºC, 0.4 GPa produced strong F zonation. X-ray maps show zontations from F-rich borders to F-poor cores, with F-rich incursions along the cperpendicular direction. A 40 micron profile perpendicular to the c-axis from F=3.0wt% on the border to F=0.5wt% in the core was fitted using Ci= 1 - C0 erf>x (4D)-1/2@ to produce an F diffusivity in biotite D = 2.75 x 10 -15 m2/s. The value is 1 order of magnitude higher than F diffusion in apatite and 4 orders higher than F diffusion on tremolite. The constant F contents of apatites at the Areias sill should be related to its poor cleaved non permeable structure.

A869

The Geochemical Atlas of Europe – Continent-wide distribution patterns of elements R. SALMINEN Geological Survey of Finland, P.O. Box 96, 02151 Espoo Finland, ([email protected]) The European geochemical baseline survey covers 26 countries, and provides invaluable information about the natural and human-induced concentrations of chemical elements in different sample media of the near-surface environment (topsoil, subsoil, humus, stream sediment, stream water and floodplain sediment). This is the first multi-national project of its kind – performed using harmonised sampling, sample preparation and analytical methodologies, and producing high quality compatible data sets across national borders. Over 60 determinands were studied, most for both total and aqua regia extractable concentrations. The results of the project are published in a two-volume set, which is freely available for viewing and downloading from www.gtk.fi/publ/foregsatlas together with the analytical data and photograph archive. Over 400 maps were plotted and interpreted. The geochemical distribution maps show distinct geographical differences in the levels of potentially harmful elements, for example, arising from natural geogenic sources, including lithology and mineralisation, and from natural processes, such as climate, which influence the original concentrations. The data from minerogenic samples reflect mostly geological variation in element distribution throughout Europe. In contrast, stream water data tends to reflect both the results of exogenic processes and the anthropogenic input of elements to the surficial deposits, although some elements give a direct signal related to the chemical composition of the bedrock. In central Europe, nitrate concentrations in stream water form a large anomalous area due to intensive agricultural practices. In northern Europe, stream water is more acidic than in central and southern Europe due to the low base cation capacity of the metamorphosed bedrock and the high concentration of humic and fulvic acids typical of the boreal climate. These low pH waters result in higher solubility of aluminium and some transition metals. Another striking feature is a relatively large anomaly of REEs detected in stream water in the area of southern Fennoscandia extending to northern Germany and Scotland. Reasons for this regional anomaly are unknown. The geochemical maps may be used to delineate areas of ore potential interest across Europe. They may also be used to identify potential geohazard and health risks in particular regions that may warrant more detailed investigation. The observed geochemical variations do, however, illustrate the difficulty in defining a single guideline value for ‘water’, ‘soil’ and ‘sediment’ to be applied throughout Europe.

A870

Goldschmidt Conference Abstracts 2007

Coupled biological, geochemical and physical processes in acidic mine lakes: Model development and application. S.U. SALMON1, C.E. OLDHAM2 AND M. HIPSEY3

U-Pb age and REE data (SHRIMP II) on zircons in mantle xenoliths from alkaline basalts (Vitim area, Transbaikalia): Implication for the upper mantle partial melting A. K. SALTYKOVA 1, L. P. NIKITINA 2 AND D.I. MATUKOV1

1

School of Environmental Systems Engineering, University of Western Australia, ([email protected]) 2 School of Environmental Systems Engineering, Univrsity of Western Australia, ([email protected]) 3 Centre for Water Research, Univrsity of Western Australia, ([email protected]) Existing water quality models for aquatic ecology and hydrodynamics have been extended to include representation of (bio)geochemical processes of relevance for acidic lakes and their sediments. The developed model allows quantitative investigation of biological, geochemical, and physical processes and their intereactions, to assess controls on key water quality parameters, such as acidity. The initial case study for model development was acidic Lake Kepwari, a decommissioned and flooded coal mine pit in Collie, southern Western Australia. Largely without calibration, the numerical model prediction of Lake Kepwari water quality over an annual cycle closely reproduced the patterns of stratification and overturn observed in the lake, as well as evapoconcentration. The model also reproduced major temporal and spatial patterns for non-conservative species, such as NH4, DOC, pH and Al. A sensitivity analysis indicated that even though the lake is now at full capacity, surface and groundwater inflows may still be important factors in the long-term evolution of the lake water quality. The model also allowed us to test the importance of geochemical processes for lake water quality, for example solubility equilibrium control of pH by Al hydroxide phases. The modelling study of Lake Kepwari demonstrated the capability of the model to reproduce the main features of the current water quality in the lake, and highlighted the need for testing against data sets which include e.g. well-constrained water and mass balances over the lake. It should be noted that the generic process descriptions in the developed model can be applied to almost any aquatic system. While the initial field site was a coal mine pit lake, the model is based on sophisticated process description and parameterisation; this allows immediate application to mine lakes of widely varying water quality. We now require a comprehensive validation exercise of model predictions against high quality datasets from acidic water bodies; it is essential to increase our confidence in the simulation results. In particular, there is a need for testing model performance against long-term data sets from acidic systems with well constrained groundwater inflow, sediment fluxes and aquatic food web data, as well as hydrodynamic and geochemical state variables. The result will be an improved tool for management and remediation of acidic water bodies, including design optimisation for future field and laboratory sampling campaigns.

1

Centre of Isotopic Research,VSEGEI, St.Petersburg, Russia; ([email protected]) 2 IPGG RAS, St-Petersburg, Russia; ([email protected]) Isotope dating of the upper mantle ultramafic rocks is difficult because their isotope systems are disturbed and mineral isochrones yield volcanic episode age. We report the U-Pb age and REE pattern of zircons from peridotite nodules in Cenozoic alkaline basalts of the Vitim volcanic area to display complicated evolution of the upper mantle. The zircons 0.06 – 012 mm size were separated from nonmagnetic heavy concentrate. U-Pb and REE analyses have been curried out using SHRIMP-II in the CIR VSEGEI. The protogranular spinel-garnet peridotite formed at P-T conditions of 45-50 kbar and 1400-1500 oC (data on Gar – Opx equilibria, Nikitina, 2000) bears prismatic or rounded zircons, colorless, with abundant fluid-melt and mineral inclusions. CL – images show their oscillatory and sectorial growth zoning and convolut recrystallization domains. The dating results are grouped in four age-spans with concordant 206Pb/238U ages of 1462±19 - 1506±4.0 (PR2), 264.0±7.3- 295.7±0.76 Ma (P1), 164.6±1.6 – 183.4±2.0 (J2) and 135.5±2.7 to 141±3 Ma (K1). Regression through two discordant data points has lower intercept of 138,0±5,7 Ma falling within K1 age group. Its upper intercept of 1891±26 overlaps with other Proterozoic 207Pb/206Pb ages, ranging 1835 -1955 Ma. No volcanic episode age (48 – 28 Ma, Ionov et al. 1992) was noticed. Chondrite-normalized REE distribution patterns in zircons show a LREE and HREE enrichment up to 100 and 1000 normalized units respectively. They are complicated by Ce maximum and Eu minimum. Total REE content and magnitude of the Eu anomaly increase with the age. The ages obtaned are in agreement with main Phanerozoic intraplate magmatic events in the Central Asia region (Yarmoljuk et al., 2000). The studied zircons crystallized during partial melting of the peridotites initiated by gassaturated fluid. Variation of the REE pattern with the zircon age indicates the multiple partial melting.

References Ionov D. A., Jagoutz E., (1992), DAN 301, 1195-1199 (in rus.). Yarmoljuk et al., (2000), Geotectocs, 5, 3-29 1199 (in rus.) Nikitina L. P., (2000), Theophrastum contr., 3, 44-53

Goldschmidt Conference Abstracts 2007

A871

The Dynamics of Core Formation in Terrestrial Planets by Negative Diapirism: Timescales, Heat Distribution and Metal-Silicate Equilibration

Different types of hydrochemical stratification in the acidic mine pit lakes of the Iberian Pyrite Belt

H. SAMUEL AND P. J. TACKLEY

Instituto Geológico y Minero de España (IGME), Ríos Rosas, 23, 28003 Madrid, Spain ([email protected])

Department of Geophysical Fluid Dynamics, Institüt fur Geophysik, ETH Zürich, Switzerland ([email protected]) Geochemical constraints on core formation provided by Hf/W systematics suggests it is a relatively fast process which may have been completed within less than 100 Myrs for the Earth or Mars. In addition, the overabundance of siderophile elements in the Earth’s mantle suggests that metal silicate equilibration occurred during core formation processes. We investigate dynamically the timing and metal-silicate equilibration processes during core formation by negative diapirism. Using numerical modeling, we follow the sinking of iron-rich diapirs through a viscous silicate mantle, in 3D axisymmetric geometry. We carried a parameter study in which shear heating as well as several viscous rheologies were considered and systematically varied. General scaling laws are derived for the diapir sinking velocity as well as for the heat distribution. These scaling laws are subsequently used to investigate the ability of negative diapirism to explain core formation in Terrestrial planets within the timing and metalsilicate equilibration constraints provided by geochemistry and mineral physics.

JAVIER SÁNCHEZ ESPAÑA, ENRIQUE LÓPEZ PAMO, ESTHER SANTOFIMIA AND MARTA DIEZ

Meromixis in pit lakes of the Iberian Pyrite Belt Many acidic pit lakes formed in the abandoned mines of the IPB during the last decades are meromictic, showing a bottom, anoxic water body (called monimolimnion) which is permanently isolated from the rest of the water column, and an upper, oxygenic mixolimnion that periodically circulates.

Results and Discussion On-going research in pit lakes of the IPB has revealed different types of perennial hydrochemical stratification, which ranges from a simple model with two internally homogeneous layers separated by a well defined chemocline, to a more complex, multi-layer model that display numerous steep salinity gradients and different strata with continuously increasing dissolved solids content with depth (Figure 1).

Figure 1: Vertical profiles of electric conductivity (EC) and dissolved oxygen (DO) in the pit lakes of San Telmo (left) and Cueva de la Mora (right). These different models of chemical stratification do not correlate with the age of the pit lakes (15-100 years), their relative depths (13-45%) or the geometry of their pit basins. It is hypothesized that the different vertical trends are the result of the interplay between different types of processes, including (i) physical mixing (e.g., wind-induced advection, heat-driven convection) in the mixolimnion, (ii) geochemical and microbiological reactions (bacterial oxidation of Fe(II), photoreduction of Fe(III), precipitation of Fe(III) phases, metal sorption) in the mixolimnion, (iii) bacterial reduction of Fe(III) and SO42- coupled with fermentation in “reactive” lake bottoms (water/sediment interfaces), (iv) groundwater input at depth, and (v) warm salt-laden water descending into the basin along the pit contours.

A872

Goldschmidt Conference Abstracts 2007

Interaction between barite and celestite cleavage surfaces and carbonate bearing aqueous solutions

Experimentally determined biomediated Sr2+ partitition coefficient for dolomite

N. SÁNCHEZ-PASTOR, C.M. PINA AND L. FERNÁNDEZ-DÍAZ

MONICA SANCHEZ-ROMAN1, CRISOGONO VASCONCELOS1, ANGELA DE LUCA, REBELLO WAGENER2, MICHAEL PLÖTZE1 AND JUDITH MCKENZIE1

Dpto. Cristalografía y Mineralogía. Universidad Complutense de Madrid. E-28040 Madrid, Spain ([email protected]; [email protected]; [email protected]) In nature, crystals of the witherite-strontianite solid solution series often show a significant chemical bimodality, i.e. intermediate compositions in this solid solution are very rare. Such a chemical bimodality can not be easily justified by invoking thermodynamic arguments. However, it has been suggested that witherite-strontianite crystals can result from the replacement of crystals of the barite-celestite solid solution which act as precursors. Consequently, the chemical bimodality in the carbonates might be a compositional feature inherited from the original sulphates. In this case, the replacement implies the coupling of dissolution and crystallisation reactions occurring on the interface between the BaxSr1-xSO4 crystal surfaces and carbonate bearing aqueous solutions. In this work, we present in situ Atomic Force Microscopy (AFM) observations of the interaction between barite and celestite (001) and (210) faces with static Na2CO3 aqueous solutions at room temperature. The concentrations of the solutions used were in the range between 0.001 mol/l and 0.1 mol/l. Typically, such interaction is characterised by the dissolution of the original faces and the rapid nucleation and spreading of islands (a 3 nm in height). The coalescence of the islands leads to the formation of a homogeneous epitaxial layer. Measurements on series of AFM images provide quantitative information about coupled dissolution-growth rates. On the basis of our results, we discuss the viability of the transfer of chemical information at a nanoscale during the replacement of the barite-celestite crystals by whitheritestrontianite.

1

ETH-Zürich, 8092 Zürich, Switzerland ([email protected]) 2 Departamento de Química, PUC Rio de Janeiro, Brazil Aerobic culture experiments with two strains of moderately halophylic bacteria grown in solid medium have been used to determine the Sr2+ partition coefficient for dolomite at different temperatures and chemistries. The Sr2+, Ca2+ and Mg2+ data and results from XRD studies indicate that Sr2+ incorporation into dolomite crystals is not by substitution of Ca2+, but instead, in our experiments, Ca2+ is replaced by Mg2+. Sr2+ enrichment is probably via surface adsorption or occlusion. Thus, the experimental Sr2+ partition coefficient between dolomite and fluid is represented here by the ratio of the Sr2+ concentration in the solid phase (dolomite, “dol”) over that in the fluid (site of dolomite precipitation = bacteriamedium interface, “bmi”), DSrdol = Srdol/Srbmi. Together, our observations suggest that DSrdol is apparently related to microbial activity, temperature and sulfate concentration. Therefore, the highest values calculated in the experiments, in comparison with the theoretical values, have implications for interpretation of depositional environment of ancient dolomite. The results of our cultures experiments show that high very high Sr2+ concentrations in ancient dolomite could reflect microbial mediation.

Goldschmidt Conference Abstracts 2007

Equation of state of fluids from sound velocity measurements in the diamond anvil cell using Brillouin scattering spectroscopy CARMEN SANCHEZ-VALLE1,2 AND JAY D. BASS2 1

Inst. for Mineralogy and Petrology, ETH Zurich, CH-8092 Zurich, Switzerland ([email protected]) 2 Dept. of Geology University of Illinois, 61801 Urbana, USA ([email protected]) Saline-rich aqueous fluids play an important role in metamorphic reactions and chemical transport in a wide range of geological environments. In subduction zones, aqueous fluids expelled from subducting slabs lead to the important geochemical phenomena of mantle wedge metasomatism and arc magmatism and mediate the recycling of elements in the Earth. Reconizing the role of deep fluid in these processes a number of significant but unanswered questions arise as to their chemical composition, the extent of mass transfer, or the mechanism of elemental transport from the slab to the mantle wedge. Answers to these questions depend on quantitative thermodynamic modelling of fluid-mineral interactions, that is greatly limited by the lack of thermodynamic data of complex aqueous fluids at high pressure and temperature conditions [1]. To address this problem, we conducted sound velocity measurements in the diamond anvil cell using Brillouin scattering spectroscopy to determine the equation of state (EoS) of H2O and chlorine-bearing solutions (1m and 3m NaCl) up to 400 ºC and 7 GPa. Sound velocities measured in H2O are in excellent agreement with previous measurements using the Impulsive Stimulated Brillouin Scattering technique [2]. The newly determined equation of state is used to evaluate the pressure and temperature dependences of thermodynamic properties of water and NaCl solutions, including thermal expansion coefficients, isothermal and adiabatic compressibilities and heat capacities. The results are combined with previous experimental and theoretical EoS of H2O-NaCl mixtures to provide an internally consistent dataset for the thermodynamical properties of the most relevant aqueous systems involved in subduction process.

References [1] Manning, C.E. (2004) Earth Planet. Sci. Lett. 223, 1-16. [2] Abramson, E.H. and Brown, J.M. (2004) Geochim. Cosmochim. Acta 68, 1827-1835.

A873

Effect of Al3+ on the elastic properties of ferropericlase at high pressure CARMEN SANCHEZ-VALLE1, KOSTANTIN D. LITASOV2, EIJI OHTANI2 AND JAY D. BASS3 1

Inst. for Mineralogy and Petrology, ETH Zurich, CH-8092 Zurich, Switzerland ([email protected]). 2 Inst. of Mineralogy, Petrology and Economic Geology, Tohoku University, Sendai, 980-8578 Japan ([email protected]; [email protected]) 3 Dept. of Geology University of Illinois, 61801 Urbana, USA ([email protected]) Main features of the Earth’s mantle structure revealed by seismic observations, including the radial profile, lateral heterogeneity, and anisotropy are largely determined by the elasticity of the materials present in this region [1]. Minealogical models indicate that the lower mantle consists of approximately 20% ferropericlase (Mg,Fe)O and more than 70% of silicate perovskite [(Mg,Fe)SiO3], containing minor amounts of aluminium. Although the effect of Al3+ incorporation on the elastic properties of MgSiO3 perovskite has been constrained from several studies [2,3], nothing is known about how this element can affect the sound velocities and elasticity of (Mg,Fe)O. In this contribution we present high-pressure measurements on the sound velocities and single-crystal elastic properties of aluminium-bearing ferropericlase (containing 2.4 atom% Al3+ and 8.7 atom% Fe2+) to 25 GPa at room temperature. Measurements were performed by Brillouin scattering in samples compressed in the diamond-anvil cell using alcohol-water mixtures as pressure transmitting media. The pressure-dependence of the aggregate compressional (VP) and shear (VS) wave velocities, as well as, the adiabatic bulk (KS), shear (μ) moduli and elastic anisotropy were obtained. In view of these results and previous elastic data for (Mg,Fe)O [4], the effect of Al3+ incorporation on the elastic properties ferropericlase and the implications for the chemical composition of the Earth’s lower mantle will be discussed.

References: [1] Trampert, J. (2004) Science 306, 853-856. [2] Jackson, J.M. (2004) Geophys. Res. Lett. 31(10), L10614. [3] Daniel, I., et al., (2004) Geophys. Res. Lett. 31(15), L15608. [4] Jacobsen, S.D. (2002) J. Geophys. Res. 107, (B2) 2037.

A874

Goldschmidt Conference Abstracts 2007

Comparative study of organic Cd and Zn complexation in lake waters SYLVIA SANDER1, LÉTICIA GINON1, BARRY ANDERSON1 1 AND KEITH A. HUNTER

Interactions of uranium with Fe(III)(hydr)oxides under sulfate reducing conditions R.K. SANI1, B.M. PEYTON2, T. GINN3 AND A. DOHNALKOVA4

1

University of Otago, NZ; ([email protected]) 1

The complexing capacity and associated stability constants of Zn and Cd in three alpine lakes in New Zealand has been determined using Competitive Ligand Equilibration-Cathodic Stripping Voltammetry (CLE-CSV) with ammonium pyrrolidine dithiocarbamate (APDC) [1] as competing ligand for Zn and Anodic Stripping Voltammetry (ASV) was used to measure the complexing parameters for Cd [2]. We compared depth profiles, seasonal variability, pH dependence and selectivity of the natural organic ligands for both elements and took a view to similarities and differences with Cu speciation in the same lakes Probably the most noticeable and interesting result of our study on the speciation of Cd an Zn in three freshwater lakes in southern New Zealand is the apparent difference of the Cdbinding ligand composition with depth, but also differences in summer and winter. While our data gives evidence for a rather unspecific binding affinity towards Cd in winter and in deep water, the specificity seems to be increased at the surface in summer. Both the higher stability constant and selectivity against competing elements could indicate towards a compound of recent biological origin. Ligands with a lower stability found in winter at the surface and all year around in the deep water samples are probably non-specific degradation products. The chemistry of Zn is much more uniform throughout the depth profiles and seasons, with Zn speciation being dominated by strong organic complexes. On average 95%r5% of Zn is present as organic complex, compared to only 70% r 22% of Cd being organically complexes. Zn binding ligands also seem to be very selective towards Zn, and thus, probably produced in response to the presence of this element on the water. The pH dependence of the complexation of both metal ions is indicating that the composition of ligands is not uniform and functional groups may consist potentially of carboxylic groups (pKa 3-5) and groups being incompletely deprotonated at pHs as high as 8.2 (e.g. phenolic and amino groups). Comparing the speciation of Cd and Zn with that of Cu in the same lakes, it strikes that the concentrations of all ligands for all three elements is very similar with an over all average of 6.8 nM r 3.1 nM (SD) although the total metal concentration vary in the range of two orders of magnitude, with Cd being the least prevalent element. However, the different complex stability constants, inorganic speciation and ligand selectivities for Cd, Zn and Cu in combination with their concentration is leading to the fact, that probably the highest risk of toxicity for organisms is coming from cadmium.

References [1] M. Ellwood, K. Hunter, J. Kim, Mar. Fresh. Res., 2001, 52, 217. [2] K. W. Bruland, Limnol. Oceanogr., 1992, 37, 1008.

South Dakota School of Mines and Technology, 501 East St. Joseph Street, Rapid City, SD 57701-3995 USA ([email protected]) 2 Montana State University, Bozeman, MT, USA ([email protected]) 3 University of California Davis, CA, USA ([email protected]) 4 PNNL, Richland, WA, USA ([email protected]) Uranium (U) contamination is a global problem in water supply and related ecosystem health, and it is the primary radioactive metal contaminating subsurface environments at many sites and is of particular concern because of its carcinogenicity. The high solubility of its hexavalent form can result in U transport to sensitive receptors such as drinking water sources. One potential method of treating U contamination is by using natural dissimilatory metal reducing bacteria including sulfate reducing bacteria (SRB) to reduce soluble U(VI) to insoluble U(IV) (as uraninite, UO2). This reduction can lead to in situ U immobilization; however, it is vital to understand the reactivity and stability of microbiallymediated uraninite. Our recent published HR-TEM, SAED, and EDS analysis data showed that SRB transformed U(VI) into colloidal uraninite with particle diameters of 3–5 nm. Further, we recently showed that under lactate-limited sulfatereducing conditions, this biogenic uraninite can be oxidized by common Fe(III)-(hydr)oxides such as hematite, goethite, and ferrihydrite. In addition to the oxidation of biogenic uranium under biotic conditions, in this paper we review the oxidation of aqueous phase biogenic uraninite by Fe(III)-(hydr)oxides under abiotic conditions. These aqueous uraninite particles were separated from SRB using 0.2 Pm membrane filter, and treated with hematite and ferrihydrite under strict anaerobic conditions. Our results suggest that Fe(III)-(hydr)oxides oxidized uraninite; however, the reactivity of biogenic uraninite was greater with ferrihydrite as compared to the more crystalline structure of hematite. Thermodynamic calculations also indicate congruent favorability of these reoxidation reactions. The results have strong implications for field application of in situ biological reduction of U(VI) since oxidation of bioreduced uraninite by Fe(III)-(hydr)oxides in situ may limit U immobilization even under reducing conditions.

Goldschmidt Conference Abstracts 2007

Reactivity of heavy rare gases with silicates at deep Earth conditions

Geographical distribution of helium isotope ratios and seismic tomography in Japan

C. SANLOUP 1,2, B. SCHMIDT 3, A. DEWAELE 4, G. GUDFINNSSON 5 AND M. MEZOUAR 6 1

Université Pierre et Marie Curie, Paris-6, France ([email protected]) 2 School of Geosciences/CSEC, University of Edinburgh, UK. 3 Universitäet Göettingen, Germany ([email protected]) 4 CEA-DPTA, Bruyères-le-Châtel, France ([email protected]) 5 Bayerisches Geoinstitut, Universitäet Bayreuth, Germany ([email protected]) 6 European Synchrotron Radiation Facility, Grenoble, France ([email protected]) The incompatible [1-3] and inert [4-5] character of atmospheric rare gases varies with pressure and temperature. This variation is poorly understood, yet it challenges our current appreciation of the formation of Earth’s atmosphere and mantle dynamics. The levels of the noble gas Xe in the atmospheres of Earth, and Mars are much lower than those of other noble gases, a puzzle known as the “missing Xe problem” [6]. One solution would be that Xe is hidden in other phases in a deep Earth reservoir. From a geochemical point of view, not only Xe, but all rare gases could be partly hiding in the deep Earth [7]. X-ray diffraction were conducted in situ on the SiO2-Xe and San Carlos Olivine-Xe systems, and completed by chemical analysis and Raman spectroscopy on quenched samples. We have shown that at high T and P, the normally unreactive Xe can bond covalently with oxygen in quartz5, displacing Si atoms in the crystal lattice. A similar behavior of Xe is observed in the olivine lattice. Xe could therefore be trapped in deep Earth silicates. The results provide an answer to the paradox of "missing Xe" in the atmospheres of the Earth and Mars. Preliminary results will be shown on silicate-Ar systems, and discussed in relationship with the observed solubility drop of Ar observed at high pressure in silicate melts1-3.

References [1] Chamorro Perez E., Gillet P., and Jambon A. (1996), Earth Planet. Sci. Lett. 145, 97-107. [2] Chamorro Perez E. Gillet P., Jambon A. et al. (1998), Nature 393, 352-355. [3] Bouhifd M. A. and Jephcoat A. (2006), Nature 439, 961964. [4] Pettersson M., Lundell J., and Räsänen M. (1999), Eur. J. Inorg. Chem. 505, 729-737 and ref. therein. [5] Sanloup C., Schmidt B., Chamorro Perez E. et al. (2005), Science 310, 1174-1177. [6] Anders E. and Owen T. (1977), Science 198, 453-465. [7] Ballentine C. J., Marty B., Lollar and Cassidy M. (2005), Nature 433, 33-37.

A875

Y. SANO1 AND J. NAKAJIMA2 1

Center for Advanced Marine Research, Ocean Research Institute, the University of Tokyo, Tokyo 164-8639, Japan ([email protected]) 2 Research Center for Prediction of Earthquakes and Volcanic Eruptions, Graduate School of Science, Tohoku University, Sendai 980-8578, Japan The 3He/4He ratios (Ra) of natural gas, volcanic fluid, and groundwater are compiled in the Japanese Islands and their geographical distributions are discussed in the tectonic frame work of subduction zone together with precisely-determined seismic velocity structures. In Northeastern (NE) Japan where typical island arc signatures are developed, there is a clear contrast of 3He/4He ratios perpendicular to the trench axis, low-Ra in the frontal arc and high-Ra in the volcanic arc. This may reflect the presence or absence of magma with high-Ra in the shallow crust [1]. As a carrier of primordial helium, source melt may be generated in low-V zone of the wedge mantle by dehydration of Pacific slab at about 150 km deep and may flow upward sub-parallel to the slab, which is well constrained by S-wave velocity perturbation [2]. In the Chugoku and Shikoku districts of Southwestern (SW) Japan, there is a geographical contrast of Ra similar to NE Japan except for the region at about 100 km from the volcanic front where medium-Ra was found. High-Ra observed in volcanic arc of the Chugoku district may be attributable to the mantle helium derived from the magma source generated below the Philippine Sea slab [3]. Medium-Ra in the Shikoku district is explained by dehydration of the young slab with a moderate aging effect [4]. These features are again consistent with the results of seismic tomography [5]. In the Kinki district of SW Japan, anomalously high-Ra was observed in the frontal arc region that was called by “Kinki Spot”[6]. Since the high-Ra is located at much wider region from the volcanic front when compared with NE Japan, the melt generated below the Philippine Sea slab may penetrate into the fissure of the slab tear and may arrive at the shallow crust by upwelling flow.

References [1] Sano and Wakita (1985) JGR 90, 8729-8741. [2] Nakajima et al. (2001) JGR 106, 21843-21857. [3] Sano et al. (2006) PAGEOPH 163, 745-757. [4] Dogan et al. (2006) Chem. Geol. 233, 235-148. [5] Nakajima and Hasegawa (2007) EPSL 254, 90-154. [6] Wakita et al. (1987) JGR 92, 12539-12546.

Goldschmidt Conference Abstracts 2007

A876

D/H ratios in n-alkanes as a proxy for paleoclimatic changes in a Brazilian lacustrine rift sequence

Emplacement of the Monchique alkaline massif (SW Portugal): Microstructures and magnetic fabric constraints

E.V. SANTOS NETO, A.A. FERREIRA AND A.L.D. SPIGOLON

H. SANT’OVAIA1, C.R. GOMES2 AND L.G. PEREIRA2

Petrobras R&D Center, 950, Horácio Macedo Ave, 21941915, Rio de Janeiro, RJ., Brazil ([email protected]) Meteoric water is considered the main hydrogen source for primary producers (Schiegl and Vogel; 1970), and fossil hydrocarbons can preserve most of their source D/H signature throughout geologic time (Santos Neto and Hayes, 1999; Li et al., 2001, Dawson et al., 2004), and even during catagenesis (Schimmelmann et al., 2006). Examination of n-alkanes from organic extracts of representative core samples of an early Brazilian Cretaceous paleolake sediments showed that GDSMOW of n-C20 to n-C32 varied mostly between § -60‰ to -100‰ (Fig. 1). Spikes up to -125‰ probably are related to coelution of compounds. The cyclic fluctuations of GD in n-alkanes are significant, correlative to inorganic indicators of environment changes, and largely independent of secondary exchanges. This suggests that D/H ratios are reflecting periods of more and less water evaporation, during the lake evolution, and can be an useful proxy for paleoclimatic studies. Figure 1: Graph showing the įD (‰) variation for the nparaffins ranging from 20 to 32 carbon atoms -40 -50 -60

įD (‰)

-70 -80 -90 -100 -110 -120 -130 20

21

22

23

24

25

26

27

28

29

30

31

32

n-paraffin carbon atom number

References Dawson, D., Grice, K., Wang, S.X., Alexander, R., Radke, J. (2004) Org. Geochem., 35, 1387-1399. Li, M., Huang, Y., Obermajer, M., Jiang, C., Snowdon, L.R., Fowler, M.G. (2001) Org. Geochem., 32, 1387-1399. Schiegl, W.E., Vogel, J.C. (1970) Earth Planet. Sci. Letters, 7, 307-313. Santos Neto, E.V., Hayes, J.M. (1999) AAPG Bull., 3, 496518. Schimmelmann, A., Sessions, A.L. Mastalerz (2006) Annu.Rev.Earth Planet. Sci, 34, 5-1-533.

1

GIMEF, Dep. de Geologia FCUP, Centro de Geologia Univ. Porto, Portugal ([email protected]) 2 CGUC, Dep. de Ciências da Terra, Univ. Coimbra, Portugal The Monchique massif aged of Upper Cretaceous (40Ar/39Ar 72.7r2.7 Ma, Valadares et al., 2005) covers an area of 80 km2 and has an east-west elongate shape. This massif is composed by two principal units: the central one corresponding to a coarse grained nepheline syenite and a marginal unit represented by a heterogeneous syenite. The central unit contains large crystals of alkali feldspars, nepheline (25% to 40%), pyroxene, biotite and sphene. The marginal unit has a variable granulometry and is poorer on nepheline (10 to 20%) (Gonzalez-Clavijo & Valadares 2003). Gomes & Pereira (2004) considered this massif as an intrusion in an antiform structure and controlled by a NE-SW shear zone. All the thin sections display microstructures typical of magmatic or submagmatic state and no microstructures indicate of strain developed under near-solidus conditions were observed. Preliminary studies of Anisotropy of Magnetic Susceptibility (AMS) were carried out measuring 102 oriented core samples from 12 sites, using an Agico Kappabridge (KLY-4S), at the Geology Dep., Porto University. The high average bulk susceptibility (23.83E-3 SI) together with previous paleomagnetic works (Gomes & Pereira, 2004) indicates that magnetite controls the magnetic behaviour. Magnetic anisotropy magnitude (Kmax/Kmin) is quite low, 1.065 on average. AMS fabric patterns show subvertical magnetic foliations associated with subhorizontal magnetic lineations. They have both a tendency to display NE-SW trends on the east and west sectors of the massif, passing to an E-W trend on the centre of the massif, drawing a sigmoid shape. Considering the AMS fabric as magma flow indicators, we propose that Monchique massif is a subvertical intrusion and that the steeply dipping magnetic foliations related to gentle dipping lineations represent an E-W trending of the magmatic flux controlled by a NE-SW left strike slip fault. References Gomes C.S.R. & Pereira L.C.G. (2004). Paleomagnetismo do Maciço de Monchique (Sul de Portugal): implicações tectónicas. Cad. Lab. Xeol. de Laxe, 29: 291-297. Gonzalez-Clavijo E.J. & Valadares, V. (2003). O maciço alcalino de Monchique (SW Portugal): estrutura e modelo de instalação na crosta superior. Comun. Inst. Geol. e Mineiro 90, 43-64. Valadares V., Azevedo M.R. & Terrinha, P. (2005). Geoquímica e Geocronologia do Complexo Alcalino de Monchique (SW Português). Actas da XIV Semana de Geoquímica: 85-88.

Goldschmidt Conference Abstracts 2007

Sr isotopic composition in Variscan granitoids, Silurian metasediments and waters from the Boticas area (Northern Portugal) 1

M. SARAIVA , M. E. P.GOMES

1

2

AND M. R. AZEVEDO

1

Department of Geology, University of Trás-os-Montes e Alto Douro, Apartado 1013, 5001-801 Vila Real, Portugal ([email protected], [email protected]) 2 Dep. of Geosciences, University of Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal ([email protected]) The processes of water-rock interaction controlling the Sr isotopic composition of superficial and groundwaters from the Boticas sanitary landfill and from the Carvalhelhos mineral waters are investigated in this work. The Boticas region is composed of several intrusions of Variscan granitoids emplaced into Silurian metasediments. The granitoids, consisting of quartz, K-feldspar, plagioclase (albite oligoclase) and micas show Sr content, and present-day 87 Sr/86Sr isotopic ratios varying between 37-105 ppm and 0.749238-0.886953, respectively. In the Silurian metasediments, quartz and mica are modally important mineral phases and plagioclase is absent. Their Sr contents range from 41-60 ppm and their 87Sr/86Sr ratios from 0.760103-0.782684. The surface and groundwaters from the Boticas sanitary landfill have bicarbonate compositions with relatively high calcium and magnesium contents (mixed-type), whilst the Carvalhelhos mineral waters show sodium bicarbonate compositions. The range of Sr and 87Sr/86Sr values in the Boticas waters (Sr = 0.0042-0.056 ppm; 87Sr/86Sr = 0.718106-0.732276) is much wider than those of the Carvalhelhos mineral waters (Sr = 0.072-0.076 ppm; 87Sr/86Sr = 0.726255-0.726322). The observed differences appear to be controlled by two major factors: (a) lithological variations of the bedrock (granites/ metasediments in the Boticas landfill and essentially granitic in Carvalhelhos) and (b) residence time. All the analysed waters show 87Sr/86Sr isotopic ratios higher than those of the rainwater (Sr poor and nonradiogenic; 87Sr/86Sr = 0.710040 - 0.715645). It is therefore likely that the Sr isotopic signatures of the analysed waters reflect variable degrees of interaction between rainwater and the radiogenic bedrock. The superficial waters with less residence time tend to show the lowest 87Sr/86Sr isotopic ratios, whilst the groundwaters are generally more radiogenic. In both the Boticas and the Carvalhelhos groundwaters, hydrolysis reactions involving the plagioclase from the granites appear to be the major mechanism of water-rock interaction. The absence of plagioclase in the Silurian metasediments suggests that the Variscan granitoids have a significant influence in the Sr-isotopic composition of the waters of the region, although the potential participation of biotite hydrolysis reactions cannot be ruled out in the case of the Boticas groundwaters.

A877

Adsorbed soil gas and microbial studies for Hydrocarbon prospecting: Jamnagar sub-basin, Gujarat, India T. SATISH KUMAR, M.A. RASHEED, G. KALPANA, D.J. PATIL AND A.M. DAYAL Stable Isotope and Surface Geochemical Prospecting Lab, National Geophysical Research Institute, Hyderabad 500 007, India ([email protected]) This study was designed to combine and compare Adsorbed gas (acid released gas) and Microbial methods with sub-surface geology. These techniques provides to understand the hydrocarbon generation, migration and accumulation processes within a basin prior to drilling. Sub -surface soil sample were collected on a detailed grid over the Jamnagar sub-basin, India. The area is largely covered and prominently exposed by the Deccan Trap (basaltic rocks), whose thickness varies from few hundreds to thousands of meters. Traps are underlain by thick Mesozoic sediments (100- 4000 m), which can form potential source rock for hydrocarbon. Deccan Trap volcanicity during Late Cretaceous may have generated the requisite thermal conditions and acted as a catalyst in Mesozoic hydrocarbon-generation process. Sedimentation in marine intertonguing environments is considered to have been favorable phenomena for hydrocarbon generation and entrapment (Kumar et al., 2004). Adsorbed soil gas of light hydrocarbons (C1 to C5) shows variotions in concentrations of Jamnagr sub-basin. The C1, 6C2+ and C3 concentration range from 3-518 ppb, 0-977 ppb and 0 – 331 ppb, respectively. Corelation factor is more than 0.9 between C1 - C2, C1 - C3, C2 - C3 and C1 - 6C2+, which indicates that light hydrocarbon components are generated from a thermogenic source. The compostional ratio of C3/C1 G103 shows that majority of samples fall in oil zone (Jones&Drodz, 1983). The propane Oxidizing Bacteria counts varies from 60 – 68600 CFU/gm. Adsorbed soil gas and microbes intensities indicate that this is warm area for hydrocarbon. The adsorbed soil gas data and microbial population are concordant and coincide with major structural features in basin. This study clearly demonstrates that adsorbed soil gas and microbial methods with geology can be applied sucessfully to demarcate the hydrocarbon reservoir zone for future exploration.

References Kumar B, Patil D.J, Kalpana G. and Vishnu Vardhan C (2004) Search and Discovery Article #10138 Jones V.T. and Drozd R.J. (1983) Geophysics. 4, 210-228.

Goldschmidt Conference Abstracts 2007

A878

A K-Ar age reset of frictionally melted gabbro and detect for degassed noble gas 1

2

KEIKO SATO , TAKEHIRO HIROSE , HIDENORI KUMAGAI1, HAJIMU TAMURA 3, KAZUO MIZOGUCHI4 AND TOSHIHIKO SHIMAMOTO 2

Sulfur/Selenium ratios in Bushveld Complex, South Africa D. SAVARD, S.-J. BARNES, L.P. BÉDARD AND R. COX 1,2,3,4

University of Quebec at Chicoutimi, Qc., Canada, ([email protected]; [email protected]; [email protected]; [email protected])

1

Radiometric age is defined as a cooling age or a closure age of a particular mineral or of minerals. With determination, quantitative analysis of thermal history of a rock or a geologic event (fault movement) could be performed with comparison of various minerals of radiometric age. Contrastingly, other method, e.g. ESR, FT and U-He age, the closure temperature of K-Ar system is rather high, a kind of melted-rocks under more high temperature, such as pseudotachylytes, are widely applied to evalution of fault activity with K-Ar dating method. To obtain reliable ages of fault activities, it is a key issue whether the ages were initialized by re-equilibrating their Ar isotopes with atmospheric Ar besides to be rejuvenesced, and other noble gas isotopes are re-equilibrating from original rock ratio to atmospheric ratio. In the case that such a rejuvenescence can explicate one of a geochemical phenomena of field observation, analyses of apparent K-Ar ages could be applied for fault rocks for thermal history. Here we report a preliminary experiment that effectuates high temperature frictional melting experiments with using a high-velocity friction apparatus in Kyoto Univ. in order to test whether age is reset by frictional heating of fault or not. A pair of gabbro chunks cut into hollowed cylinder was slid at a slip velocity of 1.3 m/s and a normal stress of 1.4 MPa. Under this condition, local frictional melting started into 5 sec. and temperature around an artificial fault reached around 1100 C in ca. 20 sec., then fault plane is well above the closure temperature of K-Ar system. Due to rapid equilibrium of volatiles during frictional melting of rocks, we expect that the K-Ar age should be reseted. This age-reset is only observed in glass completely melted. In natural field observation, such completely melted glassy materials, are faced on fault planes. Such glassy materials age are regarded to be reset during fast fault movement by re-equilibration of Ar isotopes to atmospheric one at the event. An apparent incomplete initialize was observed in the part including some un-melted gabbro fragments. In the part, mechanically fractured pieces are mixed in the glass matrices in frictional zones. That’s why the Ar isotope ratios might not be re-equilibrated in the experiments. However, U-He age is also initialized as for reequilibration of He isotopes, which is more sensitive and degassed earlier than Ar isotopes such as low speed frictional and low temperature (un-melted), just crushed rock.

Introduction Over the past decade S/Se ratios have been used in modeling the processes leading to magmatic ore deposits. Sulfur and selenium have many chemical similarities but the use of S/Se ratio is based on the idea that Se is less mobile. Thus, magmatic sulfides ore deposits have S/Se ratio close to mantle values (a2500-4000). If sulfides are altered S may be preferentially removed resulting in rocks with low S/Se. It has been suggested that the Pt-rich Merensky Reef in the Bushveld Complex formed when Pt-bearing disseminated sulfides in the underlying cumulate rocks were dissolved by Cl-rich fluids. The S and Pt were deposited at the level of the reef when this fluid dissolved into the intercumulate silicate liquid. An alternative model for the formation of the reef is that sulfides segregated from the mafic magma a little above the reef then collected on the cumulate pile. We have investigated these models with the idea that if the Pt and S were collected from bellow the reef then the S/Se ratios of the cumulate pile should be less than normal magmatic values and S/Se ratios in the reef should be higher than magmatic values. Alternatively if the sulfides in the reef segregated from a mafic magma they will have S/Se ratios close to mantle.

Results and conclusion The S/Se ratios of the reef rocks is approximately 2000 and the S/Se ratios of the rocks below the reef are in most case >2000, within or close to S/Se mantle range (Fig. 1). Therefore S/Se ratios are consistent with the magmatic model for the formation of the reef. Figure 1: Statigraphic distribution of S/Se ratios in Bushveld (Union section) 0DQWOHUDQJH

 

+WNP

IFREE, JAMSTEC, Yokosuka, Kanagawa 237-0061, Japan; ([email protected], [email protected]) 2 Graduate School of Sci., Kyoto University, Kyoto, 606-8502, Japan, ([email protected], shimamoto@ kueps.kyoto-u.ac.jp) 3 KOCHI, JAMSTEC, B200 Mononobe, Nangoku, Kochi 7838502, Japan; ([email protected]) 4 Earthquake Research Depart., NIED, 3-1, Tennodai, Tsukuba, Ibaraki, 305-0006, Japan, ([email protected])

0DJPD 5HHI



66H PDQWOH 66HPDQWOH 66H!PDQWOH 0HUHQVN\5HHI

UHHIOHYHO

   





66HUDWLR





Goldschmidt Conference Abstracts 2007

Sorption removal influence on the transformation of dissolved uranium (VI) runoff in the river water – seawater mixing zone ALLA V. SAVENKO

A879

Li-Sr-Lu-Hf isotope and trace element systematics of eclogites from Bulgaria IVAN P. SAVOV1, MICHAEL BIZIMIS2, RALF HALAMA3, STEVEN SHIREY1, ERIK HAURI1 AND IVAN HAYDOUTOV4 1

Department of Geology, Moscow M.V. Lomonosov State University, Moscow, Russia, ([email protected]) Special attention is paid to radio ecologic research, when studying chemical transformation of continental runoff in river mouth areas. Sorption-desorption processes are especially important for trace elements characterized with low coefficients of biologic accumulation. Experimental simulation of interaction between waters of various salinity which contain uranium (VI) and terrigenic clay minerals was provided aimed at revealing the role of sorption-desorption processes in transformation of runoff of dissolved uranium in the river water – seawater mixing zone. The resultant action of sorption-desorption processes was estimated based on experimental parameters of isotherms of uranium sorption at natural clays of different composition. Calculations showed that sorption removal of essential amounts of dissolved uranium occurs when river suspended solids interact with salinous waters. These amounts approach 0.07–0.08, 1.5–2.9, and 0.13–0.16 nmole/g at sea-side boundary of mixing zone for kaolinite, montmorillonite, and polymineral Gzhel clay varieties, respectively. The relative abundances of clay minerals in the sample of polymineral Gzhel clay correspond to average abundances in suspended solids for rivers of the world, with only exception for montmorillonite. When accounting of correction coefficient for difference in total amounts of clay minerals in the Gzhel clay (53%) and in the global solid runoff (67%), the value of specific sorption of uranium at the Gzhel clay rises from 0.13–0.16 nmole/g up to 0.16–0.20 nmole/g or 39–48 ng/g at transition from river water to seawater. The total amount of suspended solids removed to the Ocean was estimated varying from 15.7 to 18.5 billion ton per year [Walling, Webb, 1987; Lisitsiyn, 1978]. Consequently, the amount of sorption removal of uranium values 610–890 ton per year at the river–sea barrier, when applying the Gzhel clay as the analog of suspended solids in continental runoff. Values of uranium sorption at montmorillonite clay are an order of magnitude higher than values for the Gzhel clay (1.5– 2.9 nmole/g). As the result the presence of montmorillonite in suspended solids causes essential increase of uranium removal from solutions. River suspended solids contain ~3% of montmorillonite, thus additional removal of uranium was estimated at 170–380 ton per year. Hence the total amount of uranium removed by sorption at terrigenic solids is equal to 780–1270 ton per year. The amount of dissolved uranium entering seawater with river runoff was estimated at 20 000 ton per year. Considering this estimation the sorption removal of uranium is insignificant (4–6%). This work was supported by Russian Fond of Fundamental Investigations (project 05–05–66810).

DTM-Carnegie Institution, Washington DC, USA. ([email protected]) 2 NHFML-Florida State University, Tallahassee, FL, USA. 3 Univeristy of Maryland, College Park, MD, USA. 4 Geological Institute, B.A.S., Sofia, Bulgaria. Eclogites from the Sredna Gora Terrane and the Rhodope (UHP) Massif lie in southern Bulgaria and northern Greece and are part of the Alp-Himalayan orogenic belt. Two groups of eclogites were studied. Group 1 eclogites (Verila Mts., Ograzden Mts., E. Rhodope Mts.) have 1-10 x chondrite (CI) HREE concentrations and [La/Sm]N = 0.2-1 and variable 87 Sr/86Sr (0.7035-0.7098). This group also shows MORB-like trace element systematics and Nb/Zr (0.07-0.26), Nd/Pb (~1), Hf/Yb (0.22-0.48) and Nb/U (~29) ratios. Group 2 eclogites (Central Rhodope Mts.) have 30 x (CI) chondrite HREE abundances, 87Sr/86Sr ratios between 0.7057 and 0.7077; enriched LREE ([La/Sm]N = 1.1-2.5), elevated LILE contents and Nb/Zr (0.25-0.94), Nd/Pb (~9.5) and Nb/U (~ 32) ratios, but relatively low Hf/Yb (0.17-0.20). One interpretation of the data is that group 1 eclogites represent metamorphosed altered oceanic crust and group 2 eclogities record interaction with crustal melts/fluids possibly in an arc crust environment. in situ analyses by SIMS (Cameca 6f) of garnets and omphacites show depleted LREE and little core to rim trace element variations. The Li abundances of all garnets (1-2.5 ppm) and omphacites (45-56 ppm) are in the typical range for eclogites [1]. Mineral separates and bulk rocks from group 1 have extremely light G7Li (+1.2 and –12.9 ‰) compared to group 2 (G7Li = -0.26 to + 5.7 ‰). Our data confirm earlier studies reporting very light G7Li in eclogites from the Alps[1]. The Lu/Hf isotope systematics of omphacite-garnet pairs suggest that eclogites from Cental Rhodope Massif are of Late Proterozoic (Pan African) age [Beden eclogite = 567 Ma], almost identical to ophiolite successions throughout the basement of the Balkan Terrane to the north. Eclogites from Ograzden Mts. and Verila Mts. record Carboniferous metamorphic overprint [Gega eclogite = 293 Ma; Krainitsi eclogite = 320 Ma]. The Variscan orogeny in Bulgaria was established based on stratigraphy and abundant 300-315 Ma granitic intrusions [2]. The reported Lu/Hf age of eclogites from the Ograzden and Verila Mts. (in addition to zircon metamorphic rim geochronology [2]), implies the existence of previously not known high-grade metamorphic event similar to the one recorded in western and central Europe. These results on well-preserved Bulgarian eclogites provide a sensitive record of the pre-Mesozoic subduction events in the southern margin of the European continent. References [1] Zack, T. et al. (2003), Earth Planet Sci. Lett., 208, 279290. [2] Carrigan, C. et al (2005), Lithos, 82, 125-147.

Goldschmidt Conference Abstracts 2007

A880

Effects of cations and pressure on the structure of aqueous solutions as evidenced by IR OH bands

Geochemestry of Kairouan plain, Tunisia RIM SAYADI1, RACHIDA BOUHLILA2 AND MUSTAPHA BESBES3

T. SAWAI AND S. NAKASHIMA 1

Department of Earth and Space Science, Graduate School of Science, Osaka University, 1-1 Machikaneyama-cho, Toyonaka-shi, Osaka 560-0043, Japan. ([email protected]) In order to study effects of cations and pressure/temperatures on the structure of aqueous solutions, infrared (IR) spectra were measured for salt solutions under different pressures/temperatures. IR spectra of synthetic salt solutions (LiCl, NaCl, KCl, RbCl, CsCl, MgCl2 , CaCl2, SrCl2 , BaCl2 ) were measured by attenuated total reflection (ATR)-IR spectroscopy at room temeprature under atmospheric pressure. The peak positions of OH stretching bands shifted to higher wavenumbers with increasing salt concentrations. Figure 1: The peak positions of OH stretching bands. 



ZDYHQXPEHU>FP @

  

/L&O 1D&O .&O 5E&O &V&O 3XUH :DW HU

   



   FRQFHQWUDW LRQ>PRO O@





Ecole Nationale d’Ingénieurs de Tunis, [email protected] 2 Ecole nationale d’Ingénéieurs de Tunis, [email protected] 3 Ecole nationale d’Ingénéieurs de Tunis, [email protected] The aquifer of Kairouan, located in the central part of Tunisia, is the main water resource for drinking supply, agriculture and industrial use in the region. It is composed of a shallow and a deep units and its main feeding comes from natural recharge in wadiis. Because of the demographic, agricultural and industrial development in the region, the aquifer knows an increasing exploitation with more than 5000 wells and 400 drillings. The actual exploitation volume exceeds the aquifer capacity and thus can threaten the water quality there. The main risks are dealing with deep saline water upcoming and Sebkha water intrusion. Another risk can come from irrigation water recycling which can only be very saline because of the climate aridity in the region. In this work, we present the main results of a geochemical study of Kairouan aquifer. Major elements were used to identify the water types and a statistical study was performed to determine the salinity origin as well as the inter connection between the surface water and the two aquifer layers. The richness of the mineralization by different major elements has been explained by the probability of rocks alteration, meteoric contribution and salt migration in the atmosphere from Sebkhas that constitute the Kairouan groundwater outlet.

Reference Sayadi R. (2006). Géochimie de la plaine de Kairouan. MSc in Ecole Nationale d’ingénieurs de Tunis-Tunisie, 200p.

By taking difference spectra from the pure water, the higher wavenumber component (3430 cm-1) increases, while the lower wavenumber component (3170 cm-1) decreases for most of the salt solutions. The addition of these salts to pure water resulted in the increase of water molecular clusters with longer H-bond distances. In order to measure IR transmission spectra of fluids at different pressurre/temperature conditions, a high pressure/temperature IR transmission cell has been developped to be used under an IR microscope. First, IR spectra of pure water was measured at different pressures from 0.1 to 15.6 MPa. The OH stretching peak position did not change significantly upto 15.6 MPa. The above salt solutions are now being examined at different pressure/temperature conditions.

Goldschmidt Conference Abstracts 2007

Palaeoweathering chemistry of some flood basalt hosted fossil soils and their palaeoclimatic significance M. R. G. SAYYED Department of Geology, Poona College, Camp, Pune 411 001, India ([email protected]) Ancient soils have been largely used to understand the palaeoclimates because the chemical weathering was in response to the contemporaneous reactions between the meteoric water equilibrated with the atmospheric gases and the protoliths. The present paper attempts to compare chemical weathering patterns of the fossil interbasaltic soils (from the published geochemical data) from some flood basalt provinces. These include from Late Triassic Ischigualasto basin (NW Argentina), end-Cretaceous Deccan Traps (India), British Tertiary Basaltic Province and Tertiary lavas of Northern Ireland. The published geochemical data of the basalts and modern soils (Mumbai) developed upon them from Deccan Traps have been used for comparison. In general the CIA values do not show many variations from the Deccan boles but if studied from two spatially separated areas they have appreciable differences; the Ischiguasto palaeosols, however, show a wide range of CIA values. The higher CIA values in the Skye and Mull boles from British Tertiary basaltic province, than the Deccan boles, indicate their quite intensive weathering conditions while Tertiary interbasaltic formation of Northern Ireland conspicuously shows very high CIA values, indicating highly intense and/or prolong chemical weathering. Hydrolysis show good correlation with MAP and CIA but the Tertiary interbasaltic formation shows exceptionally higher hydrolysis. When plotted on the triangular diagrams the Deccan boles and the Tertiary interbasaltic formation show distinctive fields. The Skye and Mull boles, Ischigualasto palaeosols and Mumbai modern soils plot close together. Thus from the fossil interbasaltic soil geochemistry it is evident that the palaeoweathering characteristics are different in different flood basalt provinces and the different chemical weathering characteristics should be correlated with the time gap (quiescence periods) between the successive lava flow eruptions and the palaeolatitudinal positions influencing the palaeoclimates.

A881

Age evolution of Gold concentrations in metamorphic rocks A.M. SAZONOV1 AND E A. ZVIAGUINA2 1

Siberian Federal University, Krasnoyarsk, Russia ([email protected]) 2 Siberian Federal University, Krasnoyarsk, Russia ([email protected]) The results of examination of gold-bearing rocks in Achaean, Proterozoic and Oligocene metamorphic complexes are presented below. The territories under study are located within the blocks of the basement and the folded frame of old platforms (the Siberian platform, the East–European platform) and the Oligocene folded structure of the Eastern Pamir (Central Asia). The distribution of gold in metamorphic rocks depends on the next regularities. The oldest super-crustal metamorphic rocks of the granulite facies are characterized with higher concentrations of gold in comparison with clark (12.18 – 55.3 ppb). Charnokites formation, granitization, migmatization and diaphthorezis in the granulite series lead to gold carrying out – 20-55 ton / km3. These series could be one of the sources of metal for Proterozoic and Phanerozoic mineral deposits. Gold concentration in the Proterozoic and Phanerozoic metamorphic complexes is 2-5 times more than concentration in the surrounding aleuropelite sedimentary rocks. Gold is introduced with deep fluids into the metamorphism area. Rocks of the amphibolite facies of zoning metamorphic belts are characterized with the lowest concentration of gold (2.92-4.5 ppb). Double exceeding of gold grades is typical for the greenschist facies (4.25-6.19 ppb) and epidote-amphibolite facies (4.4-6.7 ppb) in comparison with the amphibolite facies. Accumulation of gold takes place under marginal P-T conditions of metamorphism, in the rocks near the boundaries between facies or subfacies (8.21-24.00 ppb). The distribution of gold in the zoning metamorphic Oligocene rocks of the S-E Pamir is similar to Proterozoic rocks, but gold grades in these units are more than 20% higher in comparison with Proterozoic series.

Goldschmidt Conference Abstracts 2007

A882

Patterns and processes of long-term drainage exports from a pine chronosequence on restored lignitemining dumps in Lusatia, Germany WOLFGANG SCHAAF Chair of Soil Protection and Recultivation, Brandenburg University of Technology, P. O. Box 101344, D-03013 Cottbus, Germany; ([email protected]) Opencast lignite mining results in severe multiple disturbances of ecosystem functions. In the Lusatian lignite mining district (E-Germany), an area of 80.000 ha is characterized by overburden dumps, tailings and mining lakes due to intensive lignite mining. Pyrite oxidation and AMD result in initially phytotoxic site conditions. On 60% of this area pine forests have been restored after intensive amelioration measures. Characteristic features of the soils developing at these sites are sandy texture, poor water and nutrient storage capacity, high acidity and salinity, secondary mineral formation, and a high content of geogenic organic matter in various forms and spatial distribution (Schaaf and Hüttl 2006). We studied the relevant soil structures and processes along a chronosequence of mine sites as well as the effects of its specific properties on element budgets and drainage exports at various scales using intensive field measurements, small-scale monolith sampling, and soil column experiments. Pyrite oxidation itself and labile secondary minerals as well as the content and distribution of geogenic organic matter and its related physico-chemical properties affects water and element transport, element transformation and release processes, nutrient cycling, and soil development. The disturbance effect on element exports is declining over time, but can be clearly identified over a period of at least 60 years with considerably increased leaching rates from the restored ecosystems (Schaaf 2001).

References Schaaf, W., (2001). What can element budgets of false-time series tell us about ecosystem development on post-lignite mining sites? Ecological Engineering 17, 241-252. Schaaf, W. and Hüttl, R. F., (2006). Direct and indirect effects of soil pollution by lignite mining. Water, Air and Soil Pollution - Focus 6, 253-264.

Numerical study on shock effects in impact-rocks due to porosity S. SCHADE AND K.WÜNNEMANN Museum für Naturkunde, Invalidenstr. 43a, 10115 Berlin, [email protected] ([email protected])

Motivation and method The heterogeneous distribution of crystalline shock polymorphs is commonly observed in impact rocks, and results from impedance contrasts between different minerals (e.g. Heider and Kenkmann, 2003) and pore spaces (e.g. Kieffer, 1976, Steward, 2007). We present numerical simulations of pore collapse due to shock wave loading to analyse the dependency of pre-shock internal rock structure on the resulting variety of shock features. We use the 2D hydrocode iSALE (e.g. Wünnemann et al., 2006) and the analytic EOS (ANEOS) for dunite and quartzite. Pore spaces are represented by simple geometries, aligned by varying their sizes, distances and numbers.

Fig.1 Peak pressure distribution after collapse of a rhombic (left) and a cubic (right) pore.

Results Significantly enhanced high temperatures and pressures occur when the collapse of a pore is completed. The maximum and distribution of peak pressures dependents on the geometry (Fig. 1), number and alignment of pores. Results will be furthermore compared analytically with impacted terrestrial and meteoritic rocks.

References Heider and Kenkmann (2003). MAPS 38, 1451-1460 Kieffer S.W. et al. (1976), Contrib. Miner. Petrol. 59, 41-93 Steward S.T. et al. (2007), 38th LPSC, #2413 Wünnemann K. et al. (2006), Icarus 180, 514-527

Goldschmidt Conference Abstracts 2007

Novel cosmogenic iosotope tools and progress towards historic time-scales JOERG M. SCHAEFER1, GUNTHER KORSCHINEK 2 , THOMAS FAESTERMANN 2, KLAUS KNIE3, ROBERT FINCKEL4, FERIDE SERIFIDDIN4, GREG HERZOG5, GISELA WINCKLER1, CHRISTIAN SCHLUECHTER6, GEORGE DENTON7 AND SAMUEL NIEDERMANN8 1

Lamont-Doherty Earth Observatory, Palisades, NY-10964, USA 2 Technische Universität München, 85748 Garching, Germany 3 Universität Wien, 1010 Wien, Austria 4 UC Berkeley/LLNL, USA 5 Rutgers University, Piscataway, USA 6 Universität Bern, Switzerland 7 University of Maine, Orono, USA 8 GFZ Potsdam, Germany Terrestrial cosmogenic isotope techniques have been revolutionizing earth sciences over the last decade and the methodological progress is accelerating. Related to the US/EU framework of the CRONUS initiative, substantial advances have been made over the last two years in developing new cosmogenic isotope systems and in sharpening exisiting cosmogenic tools. We present analysis and production rate calibrations of several cosmogenic isotope systems focusing on terrestrial 53 Mn, a long-lived radio isotope previously limited to extraterrestrial applications. This novel tool will allow new insights into landscape dynamics in arid regions on the million year time scale. On the other end, we present recent progress in the 10Be technique allowing precise and accurate dating of Holocene climate changes including processes less than two hundred years in age.

A883

Weathering rates determined on eroding moraine crests using cosmogenic 10Be and base cation depletion M. SCHALLER, J.D. BLUM, AND T.A. EHLERS Department of Geological Sciences, University of Michigan, 1100 N, University Ave., Ann Arbor, MI 48109-1005, USA ([email protected]: [email protected]; [email protected]) Quantification of weathering rates is important for the understanding of landscape evolution and soil formation. Chemical weathering rates are influenced by many factors such as climate, vegetation, physical erosion, parent material composition, and soil age. Weathering rates of soils are typically calculated from non-eroding surfaces of known deposition age (e.g. Bain et al., 1993; Taylor and Blum, 1995) or surfaces assumed to erode at steady-state (e.g. Green et al., 2006; Riebe et al., 2001). Unfortunately, the assumption of non-eroding or steady-state landscapes is not always justified (e.g. moraine crests). In this study we integrate cosmogenic radionuclides (10Be) and base cation depletion in two ~2 m depth profiles from moraines of different age to quantify soil mixing, denudation, and chemical weathering. Depth profiles were sampled from terminal moraines in the Wind River Mountains, Wyoming. The two profiles were collected on the crests of the Pinedale (~21 ka) and Bull Lake (~140 ka) moraines. Calculations of weathering rates are based on base cations (Ca, K, Mg, Na) and reported as g(oxide) cm-2 yr-1. Weathering rates assuming no physical erosion of the moraine surfaces are 1.0 x 10-5 and 3.1 x 10-5 g(oxide) cm-2 yr-1 for the Pinedale and Bull Lake moraines, respectively. Measurement of in situ-produced cosmogenic 10 Be in each depth profile suggest denudation rates of 0.01 mm yr-1, and soil mixing depths of 40 to 60 cm. Results to date suggest the assumption of steady-state denudation is not applicable to the investigated moraines due to their young age. Calculation of weathering rates based on transient denudation of the moraines needs to be investigated further. Prelimenary Results suggest that weathering rates accounting for physical erosion are higher than rates based on the assumption of no erosion. The increase in weathering rates might be larger for older surfaces as more weathered material has been eroded. Taking into account physical erosion, the decrease of weathering rates over time might be less than previously determined by assuming no erosion (e.g. Taylor and Blum, 1995).

References Bain D.C., Mellor, A., Robertson-Rintoul M.S.E., and Buckland S.T. (1993), Geoderma 57(3) 275-293 Green E.G., Dietrich W.E., and Banfield J.F. (2006), Earth and Planet. Sci Letters 242 155-169 Riebe C.S., Kirchner J.W., Granger D.E. and Finkel R.C. (2001), Geology 29 5111-514 Taylor A. and Blum J.D. (1995), Geology 23(11) 979-982

A884

Goldschmidt Conference Abstracts 2007

Precise ages for the Triassic/Jurassic boundary and Hettangian recovery from northern Peru

Estimating magnesium and silicon isotope fractionation with firstprinciples lattice dynamics

U. SCHALTEGGER1, B. SCHOENE1, A. BARTOLINI2, J. GUEX3 AND M. OVTCHAROVA1

E A SCHAUBLE1, E TONUI1, R B GEORG2, E D YOUNG1 2 AND A N HALLIDAY

1

Université de Genève, Section des Sciences de la Terre, Geneva, Switzerland ([email protected], [email protected], [email protected]) 2 Université "Pierre et Marie Curie" - Paris VI, CNRS-UMR 5143, Paris, France ([email protected]) 3 Université de Lausanne, Département de Géologie et Paléontologie, Lausanne, Switzerland ([email protected]) The end-Triassic extinction is considered to be one of the most important biotic crises our planet has experienced, involving the disappearance of about 80% of the known species. Flood basalt volcanism of the Central Atlantic Magmatic Province (CAMP) has repeatedly been invoked to have caused this catastrophic event. New precise zircon U-Pb ages are proposed for the Triassic-Jurassic and the HettangianSinemurian boundaries. The ages were obtained with IDTIMS techniques from single chemical-abraded zircons from volcanic ash layers within the Pucara Group, Aramachay Formation, in the Utcubamba valley, northern Peru. Ammonite biostratigraphy is combined with U-Pb dating to indicate 206Pb/238U ages of c. 201.5 Ma for the TriassicJurassic and of c. 199.5 Ma for the Hettangian-Sinemurian boundaries. The former is established on a tuff located 1 m above the last local occurrence of the topmost Triassic genus Choristoceras and the latter was obtained from a tuff collected within the Badouxia canadensis beds. Our new ages document total duration of the Hettangian of c. 2 Ma, which has fundamental implications for the interpretation and significance of the ammonite recovery after the end-Triassic extinction. The new zircon 206Pb/238U age for the T/J boundary is 2 Myr older than the currently accepted 206Pb/238U date for this boundary, and is in agreement with the published database of 40 Ar/39Ar ages, taking into account the systematic 0.6-1.0% age difference between the two isotopc systems. The most precisely dated CAMP rock – the North Mountain Basalt in Eastern United States – is slightly younger than our date for the T/J boundary, with a 206Pb/238U date of 201.27 ± 0.06 Ma (Schoene et al., 2006). Resolving the question of synchronicity between CAMP volcanism and the T/J extinction, and thereby supporting a causal relationship between these two events, requires further high-precision UPb dating of potentially older basaltic units within the CAMP.

References Schoene B., Crowley J.L., Condon D.C., Schmitz M.D. & Bowring S.A. (2006) Geochim. Cosmochim. Acta 70, 426445

1

Dept. of Earth and Space Sciences, UCLA, CA 90095, USA ([email protected]) 2 Earth Sciences, Oxford University, OX1 3PR, UK Silicon and magnesium are major elements that undergo differentiation in a variety of igneous, metamorphic, and lowtemperature environments. In this study first-principles methods are used to calculate equilibrium isotopic fractionations involving these elements, in order to assess potential applications of isotope measurements to understanding geochemical processes. Calculations focus on silicate and oxide minerals, and aqueous Mg2+ (by analogy to crystalline Mg(H2O)6•SO3). Isotopic fractionations are calculated using density functional perturbation theory (DFPT), with norm-conserving pseudopotentials and a gradient-corrected functional (PBE). The energies driving fractionation are determined from changes in the density of states of vibrational (phonon) frequencies when isotopes are substituted into each crystal structure. Comparison with measured infrared and Raman spectra suggests that the DFPT-PBE method underestimates frequencies by 3-5%, and a fitted scale factor is used to adjust model frequencies. 1-5 phonon wave vectors appear to give a sufficiently large sample of the phonon density of states for each crystal. The models predict measurable equilibrium isotopic fractionation of both magnesium and silicon in common minerals at metamorphic and igneous temperatures. At 1000ºC, 26Mg/24Mg will be higher in spinel than silicates (0.6‰ for spinel-forsterite). This agrees with preliminary measurements of peridotite mineral separates (Tonui et al., this meeting). Silicates will have higher 26Mg/24Mg than carbonates (0.5‰ for diopside-dolomite at 600ºC). Aqueous Mg2+ will also have higher 26Mg/24Mg than coexisting carbonates (by ~2-3‰ at 25ºC), in qualitative agreement with speleothem data [1]. There may be a small (0.1‰ at 1000ºC) fractionation between forsterite and diopside, with diopside tending to be heavy. 30 Si/28Si will also vary, tending to be highest in quartz. This agrees with previous Si-isotope fractionation models [2]. Quartz-zircon fractionation is expected to be ~0.5‰ at 800ºC, compared with 0.3‰ observed in a granodiorite. Forsteritediopside fractionation is small (~0.1‰ at 1000ºC, diopside being light), consistent with xenolith data. Our results indicate that processes of igneous and metamorphic differentiation and metasomatism are likely to cause detectable Si- and Mgisotope signatures.

References [1] Galy et al. (2002) EPSL 201, 105-115. [2] Méheut et al. (2007) GCA in press.

Goldschmidt Conference Abstracts 2007

Molecular and geochemical investigation of sediments covered with white mats at the Logatchev hydrothermal vent field R. SCHAUER1, H. ROY1, H.-H. GENNERICH2, A. MEYERDIERKS1 AND R. AMANN 1 1

Max Planck Institute for Marine Microbiology, Germany ([email protected]) 2 University of Bremen, Germany ([email protected]) Microbial mats have been described for different hydrothermal vent sites, e.g. along the East Pacific Rise (Taylor et al., 1999), in the Western Pacific (Takai et al., 2003), and at the Loihi Seamount (Moyer et al., 1995). In our study, we investigated the microbial community of two different sediments covered with white mats at the Logatchev hydrothermal vent area on the Mid-Atlantic Ridge (14°45`N). The white areas were occasionally interspersed with mussels at diffuse venting sites. Temperature measurements showed 2-3 °C in the surface layer of both sediments increasing to up to 99°C in 28 cm depth. The corresponding temperature profiles revealed a linear temperature gradient indicating a high convective heat flux and no fluid flow. In contrast, the temperature of the sediments outside of the white-covered areas increased from 2-3°C to a maximum of only 8°C in 28 cm depth. To investigate the microbial diversity within the sediment surface layer (0-1 cm) of the white-covered areas 16S rRNA clone libraries were constructed. The analysis of about 100 clones per site resulted in 10-13 detected phyla. Using operational taxonomic units (OTUs), members of the Gamma-, Delta-, and Epsilonproteobacteria as well as of the phylum Bacteroidetes were identified as mat specific. CARD-FISH experiments confirmed that members of these four phylogenetic groups constitute the majority of microorganism in the surface sediment layer of the “white mat” community. Typical filamentous sulphur-oxidizing bacteria of the genera Beggiatoa and Thiothrix were not detected. In conclusion, the microbial community of the white mats was found to be highly diverse. However, mat specific groups were identified, which phylogeneticaly affiliated mainly with cultured sulphur-oxidizing bacteria. The high temperature in the deeper sediment layers correlates with the appearance of the white mats on the sediments. This suggests that a high subsurface temperature could cause processes supporting the formation of white mats on hydrothermal sediments.

References Moyer, C.L., Dobbs, F.C., Karl, D.M., (1995). Appl. Environ. Microbiol. 61, 1555-1562 Takai et al., (2003). FEMS Microbiol. Lett. 218, 167-174 Taylor, C.D., Wirsen, C.O., Gaill, F., (1999). Appl. Environ. Microbiol. 65, 2253-2255

A885

U-Th magma residence times of the Plinian Mercato eruption at Mt. Somma-Vesuvius (Southern Italy) B. SCHEIBNER1, G. WÖRNER1, A. HEUMANN1 AND L. CIVETTA² 1

Geowissenschaftliches Zentrum Göttingen (GZG) – Universität Göttingen, Goldschmidtstr. 1, D-37077 Göttingen, Germany ([email protected]) ²Istituto Nazionale di Geofisica e Vulcanologia, Osservatorio Vesuviano and Dept. Scienze Fisiche, Università di napoli Federico II, Via Diocleziano 328, I-80124 Naples, Italy Plinian phonolite eruptions are a permanent dangerous threat for the population close-by to explosive volcanoes and have an effect on global climate changes. Hence, phonolite residence times in the crust are critical for assessing the potential hazard and frequencies of such explosive phonolite volcanic systems. So far, previous U-Th-studies on phonolites suggest time scales for the whole differentation process forming a basanite to a phonolite of hundred of years up to hundred of thousend years. However, it is still unclear how long an differentiated phonolite and potentially explosive magma can reside in the crust without eruption or solidifying. Our combined petrological, geochemical and isotope study on late-stage phenocrystic unzoned phonolite Ca-garnets from a phonolite sample of the Pomici di Mercato eruption from Mt. Somma-Vesuvius (Southern Italy) permit for the first time the precise determination of the residence time of a differentiated phonolite magma in the shallow crust: These Ca-garnets, glass and whole-rock from the 8.1 ka-old phonolite Pomici di Mercato eruption define an U-Th-isochrone age of 14.4 ± 1.1 ka (2ı). It follows that explosive phonolite magma resided for more than 5 ka before eruption beneath Mt. Somma-Vesuvius in the shallow crust. In addition, given the recorded frequencies of eruptions from this volcano over the time, the Pomici di Mercato eruption was preceded by an exceptionnel long repose time. This could be explained by the closedsystem behaviour of the Mercato reservoir, that was not refilled since Ca-garnet crystallisation and was not triggered by an input of new volatile-rich, less-differentiated magma.

A886

Goldschmidt Conference Abstracts 2007

Noble gases in fluid inclusions in speleothems Y. SCHEIDEGGER1,2, R.KIPFER1, R.WIELER2, S.BADERTSCHER3 AND M. LEUENBERGER4

Selenite reduction by nano-crystalline green rust, magnetite, siderite and mackinawite: EXAFS identification of Se species

1

Water Ressources and Drinking Water, Eawag, Switzerland ([email protected]) 2 Isotope Geochemistry and Mineral Resources, ETH, Zürich, Switzerland 3 Geological Institute, University of Bern, Switzerland, 4 Climate and Environmental Physics, University of Bern, Switzerland Speleothems contain various paleoenvironmental proxy records on long time scales with high temporal resolution. Air and water inclusions formed during speleothem growth both represent potentially valuable archives of past climate conditions. First semi-quantitative analyses of air inclusions indicate that the major components (N2, O2, Ar) are present in near-atmospheric abundances. In contrast, trace gases (CO2, N2O, CH4) are enriched in the inclusions and hence the cave air. Therefore, similar to air trapped in ice cores, air inclusions in speleothems have the potential to reconstruct past climate conditions in regions where no ice is available. The concentrations of dissolved noble gases in water inclusions directly reflect the atmospheric pressure as well as the temperature and salinity of the water at the time of formation of the inclusions. To infer past climate conditions, the parameters T, p, S and “excess air” have to be determined from four measured noble gas concentrations (Ne, Ar, Kr, Xe), analogous to the noble-gas thermometer employed in lakes and groundwater. The separation of noble gases from air and water inclusions is crucial to determine paleotemperatures from noble gases is water inclusions. Microscopical investigations show that air and water inclusions differ in size and position within the speleothem. Hence, a sufficient separation should be achievable by extracting noble gases in two subsequent steps (crushing and heating). First results show that crushing mainly opens inter-granular air inclusions whereas heating cracks intra-granular water and air inclusions. However, the noble gas signature after heating is dominated by noble gases from water inclusions. The salinity of the drip water before incorporation into the speleothem - determined by microthermometry - ranges from 7 – 15 ‰ and therefore can be assumed to be equal to that of freshwater. Since tritiogenic and radiogenic 3He in the water inclusions are probably negligible, 3He is purely atmospheric, in contrast to 4He. This would allow 3He to be used as an additional constraint to reconstruct the paleoenvironmental conditions. With the potential of a quantitative separation of noble gases from air and water inclusions before analysis and additional measurements of 3He, the determination of paleotemperatures from noble gases in fluid inclusions seems to be possible. In summary, dissolved noble gases from water inclusions, as well as the trace gas abundance from air inclusions are emerging proxies that will complete the available methods to reconstruct past climate conditions from speleothemes.

A. C. SCHEINOST1AND L. CHARLET2 1

Institute of Radiochemistry, FZD, Dresden, Germany, and The Rossendorf Beamline, ESRF, Grenoble, France ([email protected]) 2 Geochemistry Group, LGIT, Grenoble, France Selenium oxoanions are abiotically reduced on Fe(II/III) hydroxide (green rust) and Fe(0) metal surfaces to Se oxidation states of 0, -I or –II with relatively slow reaction kinetics [1, 2]. The unequivocal phase characterization by EXAFS spectroscopy, which is a key for the determination/prediction of Se solubility, was hindered by incomplete reduction, insufficient spectral data ranges and the similarity in short-range structures of the potential mineral phases. In a recent study [3], however, we were able to clearly identify the reduction product of selenite co-sorbed with Fe2+ to montmorillonite as nano-particulate, trigonal Se(0) by collecting low-noise EXAFS spectra at 15 K (Rossendorf Beamline at ESRF, Grenoble, France) and by employing advanced data analysis methods [4, 5]. Here, we present data on the reduction of selenite by green rust, magnetite, siderite and mackinawite. In the presence of green rust, selenite was completely reduced to trigonal Se(0) within one day, which was further converted to ferroselite (orthorombic FeSe2) within one month. The associated Fe(II) oxidation and magnetite formation quantified by Moessbauer spectroscopy allowed us to establish the following reaction scheme: 32 [FeII12FeIII5(OH)34 Cl-5 nH2O](s) + 38 SeIVO32- œ 19 [FeIISe-I2](s) + 175 [FeIIFeIII2O4](s) + 160 Cl- + 502 H2O + 84 H+ Similar to green rust, mackinawite and magnetite reduced all selenite within one day, whereas reduction by siderite was incomplete. Depending on type of mineral and Se-to-mineral ratio, trigonal Se(0), ferroselite or achavalite (hexagonal FeSe) formed. The observed differences in reaction end products and reduction kinetics may significantly affect Se bioavailability and Se migration.

References [1] Myneni, S. C. B., Tokunaga, T. K., and Brown, G. E. (1997) Science 278, 1106-1109. [2] Scheidegger, A. M., Grolimund, D., Cui, D., et al. (2003) J Phys. IV 104, 417-420. [3] Charlet, L., Scheinost, A. C., Tournassat, C., et al. (2007) Geochim. Cosmochim. Acta, in review. [4] Scheinost, A. C., Rossberg, A., Vantelon, D., et al. (2006) Geochim. Cosmochim. Acta 70, 3299-3312. [5] Rossberg, A., Reich, T., and Bernhard, G. (2003) Anal. Bioanal. Chem. 376, 631-638.

Goldschmidt Conference Abstracts 2007

Formation of Liesegang rings in borosilicate glass during experimental alteration D. SCHEITER, A. JANSSEN AND T. GEISLER Institut für Mineralogie, University of Münster, 48149 Germany ([email protected]) Borosilicate glasses are currently the standard material for the immobilization of high-level nuclear waste. Numerous studies have been carried out to study the aqueous durability of such glasses, which have indicated that under ambient conditions the corrosion of the glass is initially controlled by ion exchange and later by hydrolysis of the glass network, forming a so-called “leached layer”, before the glass dissolves. For the present study we have synthesized a borosilicate glass with a complex composition, including network modifiers such as Na, Ce, Al, Mn, Li, Ca, Mg, and Ti (WAK glass), from which we cut cuboids with edge lengths of about 2.6 mm. These cuboids were treated in a 1 M HCl solution at temperatures between 100 and 220 °C for 4 days. For one experiment (at 150°C) the solution was enriched with 47.5 at.% 18O. After the experiments, the formerly translucent, yellowish glass cuboids showed a milky-white color, but their external shape has not changed, as in pseudomorphism. Interestingly, the measured weight and elemental loss values increase regularly from 100 to 150°C, and then decrease towards a local minimum between 170 to 190°C before they increase again. Backscattered electron images on cut glass samples revealed reaction rims (“leached layers”) with complex oscillatory banding structures and bifurcations (Liesegang rings), which form a layer between the unaltered glass and an unpatterned outer rim. Whereas the outer rim consists of almost pure SiO2, the patterned layer retained some Ce, Ti, and Al, located now within distinct bands. The occurrence of Liesegang-like banding structures suggests, at a first glance, a diffusion-controlled alteration process. However, infrared spectroscopic measurements on the alteration rim formed in the 18O-enriched solution show a large mass shift of the Si-O vibrational modes (even close to the interface between the unaltered glass and the alteration layer) when compared with measurements on the alteration layer formed in “normal” 16O-enriched solution. This indicates a high concentration of 18O throughout the alteration rim. Furthermore, the chemical interface between the unaltered glass and the alteration rim is sharp on a micrometer scale with no apparent chemical diffusion profiles. Such features are not compatible with a process that is solely controlled by diffusion of cations and hydrogen species. Here we suggest that during the first step the borosilicate glass dissolved congruently followed immediately by the precipitation of amorphous SiO2 at an inward moving interface. Once the SiO2 layer became too thick, the communication between fluid and the reaction interface ceased and the SiO2 layer increasingly incorporated Ce, Ti, and Al. In a second step the Liesegang bands were formed by a diffusion-controlled process.

A887

Metamorphic reworking of the Congo Craton in Uganda VOLKER SCHENK, PETER APPEL, NIELS JÖNS, DENNY LOOSE, ANDREAS SCHUMANN AND HEIKE WEGNER Institut für Geowissenschaften, Universität Kiel, Germany ([email protected]; [email protected]; [email protected]; [email protected]) The Basement Complex of Uganda (BC) forms the NE corner of the Archean Congo craton bordering the Pan-African Central African Fold Belt (CAFB) in the N and the East African Orogen (EAO) in the E. We studied the tectonic history of the Archean basement and of its Neoproterozoic cover series by means of metamorphic P-T paths and metamorphic ages (SHRIMP dating of zircon and U-Th-total Pb chemical dating of monazite). A clockwise P-T evolution during a first metamorphism (M1) along the northern craton border is constrained by prograde sillimanite and later formed kyanite included in garnet, peak-metamorphic Grt-Sil-Bt assemblages and the late-stage formation of cordierite rims around garnet. Peak conditions (9-10 kbar; ca. 940 °C) were followed by decompression to about 8 kbar and subsequent near-isobaric cooling. A second metamorphism (ca. 750°C at 7 kbar) is evident from new garnet overgrowing Bt-Sil/Ky-Crd symplectites formed during retrogression after M1. Dating revealed ca. 2.6 Ga for M1 and ca. 650 Ma for the reworking, indicating that the granulites formed the deep cratonic crust for about 2 Ga prior to their Pan-African uplift. At the eastern craton border, aluminous Opx (up to 11.5 wt% Al2O3 when coexisting with Grt) and assemblages with Spl-Spr-Qtz-Grt-Osumilite (pseudomorph), Opx-Sil-Grt or inverted pigeonite point to extreme metamorphic temperatures (>1000°C) at 8-10 kbar in a very large area (including the Labwor Hills). In contrast, late-stage Grt-CpxQtz coronas in charnockitic and metabasic rocks were formed at 6-7.6 kbar only after near-isobaric cooling to temperatures of 650-680°C. The reaction history in metapelites points to an anti-clockwise P-T evolution. UHT metamorphism and magmatism of high-temperature A type granitoids are spatially and timely associated and lasted for about 70 Ma between 700 and 630 Ma. The intrusion of the magmas is interpreted as the cause of the UHT metamorphism. Neoproterozoic cover series (NW Uganda) have not experienced granulite facies metamorphism unlike the basement rocks on which they are resting. They are Grt-KyMs-Bt-Pl-Qtz bearing micaschists metamorphosed at 8 kbar and 650-680°C during a clockwise P-T path. Monazites show a one-stage growth history and give ages of 621±26 and 633±27 Ma for the metamorphic event that was caused by thickening of the cratonic crust.

A888

Goldschmidt Conference Abstracts 2007

The Lu-Hf systematics of meteorites: Consistent or not?

Lipid biomarkers as indicators for environmental stress in cyanobacterial mats of Abu Dhabi, United Arab Emirates

E.E. SCHERER1, C. MÜNKER2, AND K. MEZGER1 1

Universität Münster, Germany ([email protected]) Universität Bonn, Germany

2

The discrepancy between the O176Lu value determined from terrestrial minerals (1.867 ± 0.011 ×10-11yr-1; [1-3]) and those derived from meteorite WR isochrons (1.93-1.98 ×10-11yr-1; [2,4-8]) remains unsolved. Meteorites that violate isochron assumptions are not ideally suited for O176Lu measurements [9]. Nevertheless, some internal (mineral) isochrons may yield valuable insights about the early solar system. Amelin [9] compared the Lu-Hf and U-Pb systematics of phosphates in the primitive achondrite Acapulco and the ordinary chondrite Richardton (H5), resulting in a O176Lu that is consistent with terrestrial mineral data, but yielding imprecise initial 176Hf/177Hf values (i). Constraining the Richardton intercept further using the WR of [8] yields i = 0.279849 ±32, which is higher than expected for basaltic eucrite- and chondrite WR isochrons (i | 0.27964 ±4, wtd. mean of [5-7] & this work]), but would make sense if a positive correlation between 176Hf/177Hf and 176Lu/177Hf already existed in the earliest solar system materials. Photoexcitation by gamma irradiation could have temporarily increased the effective decay rate of 176Lu [10], creating anomalously steep Lu-Hf isochron slopes and high apparent O176Lu values for meteorites. Albarède et al. [10] argued that internal gamma sources such as 26Alo26Mg and 60 Feo60Coo60Ni were not abundant enough to do this. For an external gamma source, such as the young Sun or a gamma ray burst from a supernova, significant photoexcitation would have been restricted to objects less than a few cm in radius, i.e., before the material coalesced into planetary bodies [10]. Igneous rocks crystallized in planetary interiors would not be expected to show anomalously high O176Lu. However, internal isochrons of the eucrite Juvinas (present study, [6 & 7], i = 0.27969 ±4, O176Luapparent = 1.93 ±0.02 ×10-11yr-1), and the angrite SAH99555 ([11], i = 0.279682 ±30, O176Luapparent = 1.99 ±0.02 ×10-11yr-1) show exactly that, suggesting that perhaps an internal source of gamma radiation did indeed exist [11] or that there is a different reason for the apparent 176Hf excesses.

References [1] Scherer, E.E. et al. (2001) Science, 293, 683-687. [2] Scherer, E.E. et al. (2003) Meteor. Plan. Sci. 38 (7) suppl., A136. [3] Söderlund, U. et al. (2004) EPSL. 219, 311-314. [4] Patchett, P.J. and Tatsumoto M. (1980) Nature 288, 571-574. [5] Tatsumoto, M. et al. (1981) Mem. Natl. Inst. Polar. Res. Spec. Issue 20, 237-249. [6] Blichert-Toft, J. et al. (2002) EPSL. 204, 167-181. [7] Bizzarro, M. et al. (2003) Nature 421, 931-933. [8] Patchett, P.J. et al. (2004) EPSL. 222, 29-41. [9] Amelin, Y. (2005) Science 310, 839-841. [10] Albarède, F. et al. (2006) GCA 70, 1261–1270. [11] Thrane, K. et al., (2006) Meteor. Plan. Sci. 41 (8), suppl. A173.

A.-K. SCHERF1, R.M.M. ABED2 AND J. RULLKÖTTER1 1

Institute of Chemistry and Biology of the Marine Environment (ICBM), Carl von Ossietzky University of Oldenburg, Oldenburg, Germany ([email protected]) 2 Max Planck Institute for Marine Microbiology, Bremen, Germany Intertidal cyanobacterial mats in the coastal flats of the Arabian Gulf in Abu Dhabi (United Arab Emirates) grow under locally different conditions. Salinity ranges from 60 to 200 due to the high evaporation rates of the tidal waters caused by shadowless sunshine exposure for nearly 12 hours a day. We compared the phospholipid fatty acid profiles of six mats from lower, middle and upper tidal zones. The analysed mats show differences in their texture and horizontal layering. As the colours of the layers are attributed to the activity of physiologically different bacterial populations we sliced the mats accordingly and compared the results of single-layer analyses with the data for the whole mats. Phospholipids are main constituents of cell membranes and are easy to extract and identify. Differences in concentrations and distributions will reflect changes in microbial populations and reactions on environmental stress, e.g. desiccation. Former studies had shown that lipids may be involved in the protection against environmental stress as they maintain the membrane fluidity. In our mats, increased ratios of saturated to unsaturated fatty acids, cyclopropyl to monounsaturated fatty acids and trans- to cismonounsaturated fatty acids are interpreted as response to environmental stress in the region. The trans- to cismonounsaturated fatty acid ratios reached up to 1.6, e.g. for elaidic acid in our mats.

References Abed, R.M.M., Kohls, K., de Beer, D., (2007). Environmental Microbiology. doi:10.1111/j.1462-2920.2007.01254.x. Kieft, T.L., Ringelberg, D.B., White, D.C., (1994). Applied and Environmental Microbiology 60, 3292-3299.

Goldschmidt Conference Abstracts 2007

A889

Werner Schreyer and the MASHsystem: Results from investigations on natural rocks and experiments

What governs the transition from miaskitic to agpaitic assemblages in peralkaline rocks?

HANS-PETER SCHERTL1, THOMAS FOCKENBERG1 AND MICHAEL BURCHARD2

JULIAN SCHILLING, MICHAEL A.W. MARKS AND GREGOR MARKL

1

Ruhr-Universität Bochum, Institut für Geologie, Mineralogie und Geophysik, D-44780 Bochum; Germany. ([email protected]; [email protected]) 2 Universität Heidelberg, Mineralogisches Institut, Im Neuenheimer Feld 236, 69120 Heidelberg, Germany; ([email protected]) Nearly iron-free metamorphic rocks of sedimentary and metasomatic origin are found in several parts of the world (e.g. Sar-e-Sang/Afghanistan, Dora-Maira-Massif/Western Alps, Eastern Alps/Austria). All these rocks have experienced metamorphic conditions at moderate temperatures and relatively high to ultrahigh pressures. Chemically, these rocks can be well described with the MgO-Al2O3-SiO2-H2O (MASH) experimental model system, and thus provide an excellent independent check on it. In so-called leucophyllites from the Eastern Alps the assemblage chlorite + quartz is found which is stable to only 600°C (Chopin and Schreyer, 1983). Towards higher pressures these phases react to the “whiteschist” assemblage talc + kyanite (Sar-e-Sang) with a large P-T field ranging from 5 to 40 kbar (Schreyer, 1988). At temperatures of 750 - 800°C this is replaced by the ultra-highpressure assemblage pyrope + coesite (Schreyer and Yoder, 1964) found in the Dora-Maira Massif (Chopin, 1984; Schertl et al., 1991). Particularly here, several in part new highpressure phases near Mg-end-member compositions were found such as pyrope, Mg-staurolite, Mg-chloritoid, ellenbergerite, and Mg-dumortierite. At all times, Werner Schreyer recognized the significance and interaction of nature and experiment. From experimental studies in the MASH-system he and his working group mapped out petrogenetic grids which allow HP-metamorphic sequences in natural rocks to be seen in a geodynamic context, while at the same time these natural occurrences also served as a template for delineating stability/compatibility fields of the observed nearly iron-free phases and assemblages in the laboratory (e.g. Fockenberg, 1995).

References Chopin C. (1984), Contrib. Mineral. Petrol. 86, 107-118. Chopin C. and Schreyer W. (1983), Am. J. Sci. 283-A, 72-96. Fockenberg T. (1995), Boch. geol. geotechn. Arb. 44, 39-44. Schertl H.-P., Schreyer W. and Chopin C. (1991), Contrib. Mineral. Petrol. 108, 1-21. Schreyer W. (1988), Min. Mag. 52, 1-26. Schreyer W. and Yoder H.S. Jr. (1964), Carn. Inst. Wash. Year Book 66, 376-380.

University of Tuebingen, Wilhelmstr. 56, 72074 Tuebingen, ([email protected]) Peralkaline igneous rocks (molar (Na+K)/Al>1) are characterized by exceptionally high contents of generally incompatible elements like Na, Li, Be, Rb, REE and HFSE (Ti, Zr, Hf, Nb, Ta). However, based on their mineral assemblages, they can be subdivided into two groups: In the case of peralkaline miaskitic rocks, HFSE are incorporated in minerals like zircon, titanite and ilmenite, whereas in agpaitic varieties, these elements occur in complex Na-(Zr,Ti)-silicates such as eudialyte, rinkite and låvenite (Sørensen, 1997). Commonly, peralkaline igneous complexes worldwide show either a miaskitic or an agpaitic character, or at least distinct melt batches within the same complex tend to develop to either miaskitic or agpaitic assemblages. Still, it is not clear, which processes cause either miaskitic or agpaitic assemblages to crystallize. In the special case of the Eocene Tamazeght complex (Moroccan High Atlas mountains), a countinous transition from miaskitic nepheline syenites to agpaitic ones, including late-stage agpaitic pegmatites and veins, is observed. The earlier miaskitic rocks are characterized by nepheline, K-feldspar, aegirine-rich clinopyroxene, sodalite, titanite and rarely developed zircon. In contrast, the agpaitic varieties bear eudialyte instead of titanite and zircon, without showing any replacement textures between these minerals. However, the major mineral assemblages in the agpaites show no obvious difference compared to their miaskitic counterparts. In this study, we compare electron microprobe data for nepheline, sodalite and clinopyroxen as well as mineral stable isotope data (O, H) of both miaskitic and agpaitic samples. Additionally, eudialyte from agpaitic samples was analyzed by electron microprobe. Preliminary fluid inclusion investigations indicate two types of inclusions in both nepheline and eudialyte, including a methane- and hydrogene-bearing population, which is known to be typical of peralkaline rocks (Potter & Konnerup-Madsen, 2003). Based on this, we will try to shed light on the relevant changes of parameters being responsible for the transition from miaskites to agpaites.

References Potter, J. and Konnerup-Madsen, J. (2003): in: Petford, N. and Mc Caffrey, K.J.W. (eds) Hydrocarbons in Crystalline Rocks. Geological Society, London, Special Publications, 214, 141-173 Sørensen, H. (1997): Mineralogical Magazine, (1997) 61, 485498

A890

Goldschmidt Conference Abstracts 2007

In situ cosmogenic 36Cl production rate calibration on basaltic flows of Mount Etna (Sicily, 38°N)

Europa’s subsurface ocean N. SCHILLING1, F.M. NEUBAUER2 AND J. SAUR3 1

Institut für Geophysik und Meteorologie, Universität zu Köln, Cologne, Germany ([email protected]) 2 Institut für Geophysik und Meteorologie, Universität zu Köln, Cologne, Germany ([email protected]) 3 Institut für Geophysik und Meteorologie, Universität zu Köln, Cologne, Germany ([email protected]) Recent geological and geophysical observations from the Galileo spacecraft indicate that Europa, one of the icy moons of Jupiter, harbors a salty ocean of liquid water beneath its icy surface. Magnetic field perturbations observed by the Galileo spacecraft at Europa, are consistent with induced magnetic fields from the interior of the moon. These magnetic fields are very likely caused by electromagnetic induction in a subsurface water ocean with high electrolytical conductivity. An additional process, which also generates magnetic field perturbations at Europa and therefore competes with induction, is the interaction of Europa’s oxygen atmosphere with the magnetospheric plasma in which the moon is embedded. We have developed a time dependent 3D model to study simultaneously the electrical conductivity distribution inside Europa and the time-varying interaction of Jupiter’s magnetosphere with Europa’s atmosphere. By comparing our simulation results to the Galileo spacecraft measurements by Europa, we place the so far strongest constraints on the conductivity and the thickness of Europa’s subsurface ocean. We find an internal ocean close to the surface with a conductance of 50,000 S or larger to be most suitable to explain the magnetic flyby data. These results suggest that a global subsurface ocean at Europa with terrestrial “sea water” would have to be thicker than 10 km.

I. SCHIMMELPFENNIG1, L. BENEDETTI1, R. PIK2, P. BURNARD2, P.H. BLARD3 AND D. BOURLÈS1 1

CEREGE, UMR 6635 CNRS, Aix en Provence, France ([email protected]) 2 CRPG, UPR 2300 CNRS, Vandoeuvre-lès-Nancy, France 3 CALTECH, Pasadena, CA 91125, USA One of the CRONUS-EU goals is to provide high quality calibration sites from independently dated surfaces. Several previous studies have been conducted on 36Cl production rate calibration (e.g. Stone et al. 1996, Phillips et al. 2001), which, however, used different protocols and yielded 36Cl production rates with up to 40% discrepancies. The objectives of this study are 1- to understand the source of these discrepancies and 2- to calibrate 36Cl production rates from its most abundant target elements 40Ca, 39K and 35Cl. As a first step we focused on testing the chemical protocol by performing a sequential 36Cl extraction experiment on whole rock grains and Ca rich plagioclase from the same sample. The sample was collected at Mt. Etna on a pahoehoe flow which has a K-Ar fossil exposure time of 10±3 kyr. Cosmogenic 3He of this sample was also precisely measured within cogenetic olivine phenocrysts (Blard et al. 2005) and yields an exposure time of 10.4±1.5 kyr. Both, total Cl and 36Cl concentrations from the first dissolution steps are high, 5800 ppm (whole rock) and 450 ppm (plagioclase) Cl, and 107 - 106 atoms 36Cl/g of rock dissolved. After about 20% dissolution of the plagioclase sample, Cl is almost completely removed (1-3ppm) and 36Cl concentrations reach a plateau value of 2*105 atoms/g of rock. Using the Stone et al. (1996) and Evans et al. (1997) 36Cl production rates for the target elements Ca and K, respectively, this plateau concentration yields an exposure age which is in agreement, within uncertainty, both with K-Ar dating and cosmogenic 3He ages. On the contrary, in the whole rock sample total Cl concentrations remain high (>330ppm) resulting in a considerable 36Cl production from capture of low-energy neutrons by 35Cl, an additional and still not well-constrained 36Cl production mechanism. The resulting exposure ages from the whole rock are 20-30% higher than the independent 3He ages. To better constrain the different sources of 36Cl in basaltic rocks and their production rates we will perform a similar experiment on separated sanidine (K-rich feldspar) from an independently dated basaltic flow of another volcano.

References Stone J.O., et al. (1996), Geochim. Cosmochim. Acta 60 679692; Phillips F.M., et al. (2001), Chem. Geol. 175 689-701; Blard P.H., et al. (2005), EPSL 236 613-631; Evans J.M. et al. (1997), Nucl. Instr. and Meth. in Phys. Res. B 123 334-340

Goldschmidt Conference Abstracts 2007

Simultaneous determination of Se and Te in different geological matrices with DRC-ICP-MS THOMAS SCHIRMER, ANDREA KOSCHINSKY AND MICHAEL BAU Jacobs University Bremen gGmbH, Campus Ring 8, D-28759, Bremen, Germany ([email protected], [email protected], [email protected]) An investigation of ferro-manganese crusts conducted by Hein, Koschinsky et al. (2003) had shown an extreme enrichment of tellurium, but no comparable concentrations of the geochemically very similar element selenium. This Te enrichment was related to a surface oxidation process of Te(IV) to Te(VI) on the Fe-Mn oxide surfaces, which apparently does not take place for Se(IV). Therefore the element pair Se and Te seems to have an interesting potential as paleo-proxy to investigate in situ redox conditions. In a next step Se and Te are investigated in more detail in different geological matrices to gain knowledge of the usability of this element pair to describe geopaleontogical redox processes. Very little Te data of different geological materials have been available so far. A first challenge to be addressed is the precise determination of these two elements with quadrupole DRCICP-MS. In this context an analytical approach was developed, using different methods for sample digestion, enrichment and analysis. Two well known problems connected with the determination of Se and Te with standard quadrupole ICP mass spectrometry are low sensitivities (max. ~3 kcps and ~15 kcps for 10 ppb Se and Te respectively) and severe 40 80 + interferences, especially on Se (e.g. Ar2 on Se). This fact is accompanied by the overall low concentrations in the different matrices, especially in water. Another problem with respect to the acid pressure digestion is the instability of Se in chloride matrix and at temperatures above 80 °C. All these analytical problems where addressed by optimizing the ICP-MS conditions, the digestion procedure and by the use of thiol cotton separation for samples with very low Te and Se concentrations, first described by Mu-Qing and Gui-Qin (1983). Data for Te and Se in various sample types including marine ferromanganese crusts, weathered basalt and dolomites will be presented.

References Hein, J. R., A. Koschinsky, and Halliday, A. (2003). Geochimica et Cosmochimica Acta 67 (6): 1117–1127. Mu-Qing, Y. and L. Gui-Qin (1983). Talanta 30 (4): 265-270.

A891

Influence of Hydrogen on electrical conductivity in enstatite E. SCHLECHTER1, R. STALDER1 AND H. BEHRENS2 1

Dept. of Geoscience, University of Göttingen, ([email protected]), ([email protected]) 2 Institute of Mineralogy, University of Hannover, ([email protected]) Hydrogen has a significant influence on several physical properties of upper mantle rocks. Models for electrical conductivity, for instance, are mainly based on conductivity data for olivine. Therefore, systematic laboratory data on electrical conductivity of other H-bearing nominally anhydrous minerals (e.g. enstatite, MgSiO3) are needed. An important mechanism for electrical conduction in enstatite requires the presence of protons, because they are the major mobile charge carriers. Key factors for this conduction mechanism are concentration and mobility of protons. In this study, we measured electrical conductivity and H-diffusivity in pure, Al-, and Fe-doped synthetic enstatite. As incorporation of additional cations, such as Al and Fe, create new point-defects in the crystal lattice, transport mechanisms might depend on the defect chemistry. Diffusion experiments were conducted at 1 bar and 800°C to 1000°C in air. Electrical conductivity was measured at same conditions with an impedance analyzer (0.1 - 10000 Hz), and H-concentration was monitored by FTIR-spectroscopy. Initial results show that H-diffusivity from dehydration experiments in Fe-bearing enstatite is about one order of magnitude faster than in pure enstatite, whereas it is slower in Al-bearing enstatite. A similar behaviour was observed for electrical conductivity. Application of the Nernst-Einstein equation to the diffusivity results for pure enstatite are in good agreement with our preliminary conductivity data; this supports the hypothesis that H is the major mobile charge carrier.

Goldschmidt Conference Abstracts 2007

A892

Cation sorption at the smectite edges: From transition metals to Y and Lu M.L. SCHLEGEL CEA, DANS/DPC/SCP/ Laboratory for the Reactivity of Surfaces and Interfaces, F-91191 Gif-sur-Yvette, France ([email protected])

Introduction Clay minerals can adsorb trace elements in soils and weathering formations, a sorption property which is also a key-point for the use of clay materials in nuclear waste repositories. Retention can occur either by adsorption on the clay basal plane or at the layer edges, depending on physicochemical conditions. Building on previous identification of sorption sites of transition metals [1-3], this study shows how the geometry of Y and Lu surface complexes on layer edges of clay minerals can be identified.

Materials and methods Suspensions of purified clay minerals (hectorite or Almontmorillonite, 2 and 1 g/L, respectively) were reacted in 0.5 M NaCl at pH 6 (Al-montmorillonite) and 7 (hectorite) with sorbates at concentrations of 50 μM (Y) and 100 μM (Lu). Self-supporting films of reacted clay minerals were obtained by slowly filtering suspension aliquots on cellulose nitrate filters. Yttrium K-edge and Lu L3-edge polarized EXAFS (PEXAFS) spectra of the self-supporting films were collected in fluorescence mode on the FAME beamline (ESRF, France). PEXAFS data were reduced, and analyzed using standard procedures.

Results and interpretation Small but significant angular dependences were observed for all P-EXAFS data, meaning that the Y and Lu binding environments are anisotropic. Coordination spheres of 7 O at 2.36 Å and 8 O at 2.27 Å, were observed for Y and Lu, respectively, comparable with d(Ln-O) distances measured by EXAFS spectroscopy [4]. Lutetium sorbed on Almontmorillonite was surrounded by an Al-shell at 3.35 Å, consistent with Lu sharing edges with Al octahedra and partially incorportated in a gibbsite-like interlayer. Both Y and Lu sorbed on hectorite were surrounded by cationic shells. Modelling of these cationic contributions yielded one out-ofplane Si/Mg shell at 3.16 Å (Y) or 3.04 Å (Lu), and two inplane (Mg/Si) shells at 3.50, and 3.97 Å for Y, or 3.33 and 3.91 Å for Lu. These distances and angular orientations collectively suggest that Y and Lu form inner-sphere surface complexes at the edges of hectorite platelets, but slightly above/below the hectorite octahedral plane, and tend to share edges with silicate groups of the clay tetrahedral sheet.

References [1] M.L. Schlegel, et al. Am. J. Sci. 301, 798-830 (2001). [2] M.L. Schlegel, A. Manceau. GCA 70, 901-917 (2006). [3] M.L. Schlegel, A. Manceau. Environ. Sci. Technol., 41, 1942-1948 (2007). [4] K. Dardenne, et al. Radiochim. Acta 89, 469-479 (2001).

Interaction between colloidal and organically complexed Iron (III) species in seawater by ultrafiltration C. SCHLOSSER AND P. CROOT IFM-Geomar, Kiel, Germany ([email protected], [email protected])

Method and Theory Ultrafiltration experiments (Viavflow 50 - 10kDa filter) were performed using 55Fe in combination with artifical and natural organic ligands (in excess) in seawater. The activity of the 55Fe in the permeate and retentate was measured using standard liquid scintilation methods. The permeation (Pc) and retention coefficient (Rc) were calculated as follows:

PC

§ log c p · ¸¸ 1  ¨¨ © log CF ¹

§ log cR · ¸¸ RC 1  ¨¨ © log CF ¹

Where CF = concentration factor (CF). Some results are shown in the Table below: Table 1: PC and RC values of the Fe (III) species Ligand

Pc

Rc

None - UV irradiated Seawater Natural - Antarctic Seawater Natural - Coastal Seawater 2-(2-Thiazolylazo)-p-cresol (TAC) Ethylenediaminetetraacetic acid (EDTA) Desferrioxamine_B (DFO_B) 2-Keto-D-Gluconic Acide Protoporphyrine IX Phytic Acide

0.49 0.96 0.93 0.99 1.00 0.97 0.88 0.89 0.97

0.44 0.58 0.63 0.75 0.85 0.97 0.59 0.49 0.63

Discussion These 2 parameters allow us to differentiate between very strong, strong and weak ligands based on their ultrafiltration properties. Pc=Rc=1 are significant for very strong ligands (e.g. DFO_B), where all added Fe is organically complexed. Strong ligands (e.g. TAC, EDTA) have additionally a colloidal fraction, but no interaction between both species (Pc=1, Rc<1). Pc=Rc<1 are observable for weaker ligands (e.g. Protoporphyrin IX) with an identifiable interaction between both species.

Conclusion Using this approach helps us better understand the relationship between colloidal and dissolved Fe concentrations in natural seawater and how to quantify the interconnections between natural organic complexes and colloidal iron. The results strongly suggest that information on colloidal Fe concentrations is dependent on CF and this needs to be taken into account when reporting ultrafiltration data. This approach is particularly useful for assessing the bioavailability of iron in natural seawater as most colloidal iron is believed to be weakly bioavailable.

Goldschmidt Conference Abstracts 2007

Hydrothermal AFM and mixed-flow reactor studies of ZnS dissolution

Architecture of phosphatic and calcitic brachiopod shell materials – A comparison.

C. SCHLUETER1, G. JORDAN1, O. POKROVSKY2 AND W.W. SCHMAHL1

W. W. SCHMAHL1, C. MERKEL1, E. GRIESSHABER 1, K. KELM2 AND C.LUETER3

1

Sektion Kristallographie, Departement für Geo- und Umweltwissenschaften, Theresienstr.41, 80333 München, Germany, ([email protected], [email protected], [email protected]) 2 LMTG-OMP CNRS, Univ. P. Sabatier, Toulouse, France, ([email protected]) Sulfide minerals are the major contributors to acid mine drainage (AMD) and heavy metal release. While the weathering of iron sulfides has been studied intensively only little attention has been paid to the zinc sulfides. To understand the mechanisms and kinetics of sphalerite reactions in aqueous solutions, a combination of high resolution microscopy methods and chemical analytical methods proved to be perfect as it correlates nano-scale observations with macro-scale data. Here, in situ hydrothermal AFM (HAFM) measurements and mixed-flow reactor experiments were conducted at pH 1-2 and a temperature of 125 °C. While HAFM measurements were performed on the (110) surfaces, sphalerite powder was used for mixed-flow reactor experiments. Mixed-flow reactor experiments showed an initial maximum of zinc and sulfur release. This initial maximum in dissolution was followed by an exponentially decreasing part, which converted into a steady state after ca. 10,000-15,000 min. Within the accuracy of the data, the mixed-flow experiments revealed no incongruent or not stoichiometric dissolution. in situ HAFM measurements showed the formation of a thin altered layer on the surface during the initial state of dissolution followed by a development of facets. From the mixed-flow reactor experiments it can be concluded that the leached layer observed in HAFM experiments can only be caused by a minimal shift in the chemical composition of the surface. The HAFM experiments further showed the decomposition of the cleavage morphology into facets with three prominent orientations. These observations indicate that the observed initial maximum in zinc and sulfur concentration was caused by a change of the surface morphology from one morphology with high dissolution rates to another morphology with lower dissolution rates. As proved by SEM investigations, small particles on the crystals can be excluded as a reason for the observed rate maximum. Until now tipsurface convolution disabled crystallographic indexing of the facets by AFM. Indexing the facets via LEED remained difficult because the facets were not persistent while LEED sample conditioning. Further experiments are in progress. In summary, the two different experimental approaches yielded complementary information which in combination allowed one to obtain a significantly more comprehensive understanding of the mechanisms and kinetics of sphalerite dissolution.

A893

1

Department for Earth and Environment, LMU Munich, Munich, Germany, ([email protected]) 2 CAESAR Research Institute, Bonn, Germany, ([email protected]) 3 Museum für Naturkunde der Humboldt-Universität, Berlin, Germany, ([email protected]) Brachiopods have developed shell materials employing the two principal mineral groups of hard biologic tissue: calcium carbonate and calcium phosphate, each with distinct hierarchical architecture. The “phosphatic” shells consist of a predominantly chitin fibrous matrix, which is reinforced by isometric hydroxyapatite or similarly sized amorphous Caphosphate particles attached to the fibres [1]. The fibre composit is not unlike vertebrate bone. A laminated structure is created by changing volume ratios of chitin and reinforcing particles. In contrast, the composit structure of calcitic brachiopod shells employs inorganic single-crystal fibres reinforced by thin intercrystalline and extremely thin intracrystalline organic membranes [2, 3]. The pronounced crystallographic texture of the fibres [2, 4] appears to originate from growth selection. Over lengths well exceeding hundred micrometers the fibres change crystallographic orientation continuously in the order of one degree. While the laminated organic fibre/inorganic nanoparticle strategy provides a degree of shell flexibility the inorganic fibre/polymer membrane composit provides a high stiffness and micro- and nanohardness.

References [1] C. Merkel, E. Griesshaber, K. Kelm, R. Neuser, G. Jordan, A. Logan, W. Schmahl (2007), J.G.R.- Biogeosci. in press [2] W.W. Schmahl, E. Griesshaber, R. Neuser, A. Lenze, R. Job, U. Brand (2004), Eur. J. Mineral. 16 693-697 [3] E. Griesshaber, W. W. Schmahl, R. Neuser, T. Pettke, M. Blüm, J. Mutterlose, U. Brand (2007), Am. Min., 9, in press

Goldschmidt Conference Abstracts 2007

A894

Lu-Hf geochronology of eclogites from the Dabie-Sulu terrain: Constraints on the timing of eclogitefacies metamorphism 1

1

2

A. SCHMIDT , S. WEYER , Y. XIAO , J. HOEFS G.P. BREY1

2

AND

1

Institut für Geowissenschaften, JWG Universität Frankfurt, ([email protected]) 2 2 Geowissenschaftliches Zentrum Göttingen The Qinling-Tongbai-Dabie-Sulu belt was formed by the collision of the Yangtze and the Sino-Korean cratons, and is one of the largest ultrahigh-pressure metamorphic belts worldwide. Micro-diamond and coesite was found in eclogites and country rocks of this area, indicating a deep subduction of continental material at mantle depth. The eastern UHP part of the belt is subdivided in the Dabie and the Sulu terrain. The latter was offset 500km to the North by the Tan-Lu fault. Various geochronometers (U-Pb, Sm-Nd, Rb-Sr, Ar-Ar) have been applied to date the metamorphic events in this collision zone. However, UHP and eclogite facies metamorphism are controversely constrained to a Triassic age between 210 Ma and 245 Ma. Here we present a new approach using the Lu-Hf isotope system to date garnet+clinopyroxene mineral assemblages in three eclogites from the Dabie terrain (sample localities are Bixiling, Shima and Lidu) and three eclogites from the Chinese Continental Scientific Drilling Program (CCSD) in the Sulu terrain. Garnets of our samples are very homogeneous and virtually non-zoned. The Lu-Hf isochrons yield tightly grouped ages between 219.6 Ma and 223.4 Ma for both terranes (Dabie Shan and Sulu), yielding a mean age of 222.0 Ma for all six samples. This age is in good agreement with earlier estimates on the timing of UHP metamorphism. However, in contrast to previous studies our results indicate a tight age range for eclogites of the entire Dabie complex. Based on U-Pb zircon ages, Liu et al. (2006) and Hacker et al. (2006) recently proposed three metamorphic events within age ranges of 244236 Ma, 230-220 Ma and 220-205 Ma, respectively. Liu et al. (2006) interpreted their ages as distinct eclogite facies events. In contrast, our Lu-Hf results show no evidence of multiple or long lasting eclogite facies conditions. Alternatively, Hacker et al. (2006) proposed a “precursor” UHP, a “main” UHP and an amphibolites facies metamorphic event for the Dabie-Sulu area. Our Lu-Hf ages fall into their proposed UHP event (220230 Ma), however, define a much smaller range. This implies a very rapid UHP event on a large regional scale. Alternatively, not UHP, but the transformation of pre-curser rocks into eclogite or closure of the Lu-Hf system in garnet in the Dabie and Sulu terrain occurred within a very limited timescale.

References Hacker, B.R. et al.; Tectonics 25 (2006); TC5006, doi:5010.1029/2005TC001937 Liu, D. et al.; EPSL 250 (2006); pp650-666

Using 222Rn as environmental tracer for assessing groundwater/surface water interaction AXEL SCHMIDT AND MICHAEL SCHUBERT Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany ([email protected]; [email protected]) The exchange of water between lakes or rivers and hydraulically connected aquifers provides a major pathway for chemical transfer between the respective water bodies. For instance, the migration of dissolved carbon, oxygen, and/or nutrients coupled to such exchange processes has to be considered a main driver for biogeochemical processes on both sides of the surface water/groundwater interface. Furthermore, dissolved contaminants, such as dissolved NAPL, pharmaceuticals, or heavy metals, are not only influential on the aquatic life or the biological properties of the affected water body, but also on its overall water quality. Generally speaking, if the status of a surface water or groundwater resource is to be assessed or its fate to be predicted it must not be looked at as a separate aquatic system but interactions at the surface water/groundwater interface have to be taken into account. The groundwater flux into a meromictic lignite mining lake (Lusatia Mining District, Germany) was quantitatively assessed by means of a geochemical tracer technique using the naturally occurring radio isotope 222Rn. The noble gas radon makes an ideal environmental tracer because of its chemically inert behaviour and its ubiquitous presence in groundwater, where it appears in concentrations well above the concentrations found in surface waters (Nazaroff and Nero, 1988). In a long-term project radon concentrations in the water of the studied mining lake and in the groundwater sampled from surrounding monitoring wells were determined monthly over a two-year period. Evaluation and interpretation of the data sets allowed for assessing the dynamics of the local groundwater/surface water exchange processes. It could be shown that there is a high variability in the groundwater/surface water interaction rate, depending on changes of the (seasonal) precipitation rate, even within very short time scales. The radon technique used has a high potential to improve the quality of the investigation of limnic environments. As a additional advantage the possibility of on-site determination of radon concentration in water samples (Schubert et al., 2006) allows for straight forward decisions concerning the further strategy of ongoing sampling campaigns.

References Nazaroff W.W., Nero A.V. jr., (1988), Radon and its Decay Products in Indoor Air. John Wiley & Sons, New York/NY/USA. Schubert M., Bürkin W., Peña P., Lopez A., Balcázar M., (2006), Rad. Meas. 41. 492-497.

Goldschmidt Conference Abstracts 2007

In situ studies of aqueous fluid properties at high pressures and temperatures using hydrothermal diamond-anvil cells CHRISTIAN SCHMIDT GeoForschungsZentrum Potsdam, Telegrafenberg, 14473 Potsdam, Germany ([email protected]) Aqueous fluids and their interaction with mineral assemblages play a key role for the mobility of elements and, thus, material flux in the lithosphere. The invention of the hydrothermal diamond-anvil cell (HDAC) [1] has extended the PT range, in which aqueous fluid properties can be studied in situ, up to the conditions of the lower crust and dehydrating slabs. Modifications of the HDAC design [2,3] permit application of synchrotron-radiation X-ray techniques to investigate the behavior of dissolved heavy elements with Xray emission energies as low as ~4 keV and absorption edge energies as low as ~5 keV. Time-resolved SR-XRF analyses have been used to study the solubility of minerals containing elements with Z•22 in aqueous fluids, the kinetics of fluidmineral interaction, and trace element partitioning between aqueous fluids and silicate melts [e.g., 4, 5]. A minimum detection limit of about 2 ppm for Ti (KD energy=4.5 keV) was achieved in recent in situ SR-XRF experiments on the solubility and dissolution kinetics of rutile in H2O+NaAlSi3O8 and H2O+Na2Si3O7 fluids to 800 °C and 1.3 GPa. Complexation and speciation of heavy elements in aqueous melts and fluids have been investigated in situ by X-ray absorption techniques, e.g., SR-XAFS analyses of the La L3edge in a study of the hydration structure of aqueous La3+ (0.007 m) to 300 °C and 160 MPa [6] or SR-XANES measurements on Fe2+ in water-saturated haplogranitic melt up to 700 °C and 500 MPa [7]. Raman spectroscopy and Bassett-type HDACs have been used in a number of studies to obtain information on the behavior of light elements at high PT, e.g., on the speciation of silica to 900 °C and 1.4 GPa [8] or the complexation of boron to 600 °C and 2 GPa [9]. However, utilization of Raman spectroscopy as a quanititative tool is still in an early stage. Much remains to be learned about fluids in deeper portions of the lithosphere, and further method developments (e.g., to determine electrical conductance and density of such fluids) are needed. References [1] Bassett, W.A. et al. (1993), Rev. Sci. Instrum. 64 23402345; [2] Bassett, W.A. et al. (2000), Z. Kristallogr. 215 711717; [3] Schmidt, C. and Rickers K. (2003) Amer. Mineral. 88 288-292; [4] Schmidt, C. et al. (2007), Lithos 95 87-102; [5] Schmidt, C. et al. (2006) Amer. Mineral. 91 1211-1215; [6] Anderson A.J. et al. (2002) Amer. Mineral. 87 262-268; [7] Wilke, M. et al. (2006) Chem. Geol. 229 144-161; [8] Zotov, N. and Keppler, H. (2002) Chem. Geol. 184 71-82; [9] Schmidt, C. et al. (2005) Geochim. Cosmochim. Acta 69 275281

A895

Trace elements in foraminiferal calcite: New understanding based on in situ analysis D.N. SCHMIDT1 AND S. A. KASEMANN2 1

Department of Earth Science, University of Bristol, Bristol Uk; ([email protected]) 2 Grant Institute of Earth Science, University of Edinburgh, Edinburgh UK; ([email protected]) Trace element concentrations in foraminiferal carbonates are a widely used as proxy to assess past climates and ocean circulation. Planktic foraminifers strongly discriminate against trace element incorporation, resulting in Me/Ca ratios which are significantly lower than ambient sea water concentrations. Physiological (e.g. photosynthesis, respiration, and calcification) and ecological processes (e.g. depth migration) can modify the trace element uptake into the foraminiferal calcite and thus exert an important influence on Me/Ca ratios. in situ analysis of spatially resolved elemental ratios increase our understanding of biological processes influences in incorporation of these proxies in foraminifers. The applicability of the proxy and the precision of palaeoclimate assessment will depend on our understanding of the incorporation of these trace elements into foraminiferal carbonate. To this end, we have performed trace element analysis using Secondary Ionization Mass Spectrometry and Electron Microprobe on planktic foraminifers from sediment samples and sediment traps. The sediment trap samples from the Canary Islands and the western equatorial Atlantic allow for direct comparison between ambient conditions and Me/Ca ratios. We analysed symbiont bearing (G. ruber, G. sacculifer) and asymbiotic (G. truncatulinoides, G. tumida) species which are living in different depth habitats. Mg/Ca, Ba/Ca, Li/Ca and Sr/Ca ratios were analysed in profiles and spot measurements in and across several chambers and specimens. The profiles at a resolution 2.0 Pm clearly display the chemical variability within the multi-layered foraminiferal test. Additionally, we document Me/Ca variability during the development of the organism and, using specimens from different seasons with optimal and suboptimal growth rates, assessed the influences of foraminiferal growth rates on element incorporation. Foraminiferal test displays strong heterogeneity in Mg/Ca, B/Ca, Ba/Ca, and Li/Ca but not in Sr/Ca ratios. B/Ca within one specimen range from 20 to 120 mmol/mol, Ba/Ca from 1.2 to 4.5 mmol/mol, Mg/Ca ratios from 0.6 to 1.2 mmol/mol. The proloculus, the first chamber, displays significantly higher Mg/Ca and B/Ca ratios but not Sr/Ca and Ba/Ca which can not be explained by environmental conditions alone.

Goldschmidt Conference Abstracts 2007

A896

Influence of impactors on the chemical composition of the Earth G. SCHMIDT Institute of Nuclear Chemistry, University of Mainz, Germany ([email protected])

First direct sampling of superhot supercritical vapor phase fluids at the Mid-Atlantic Ridge K. SCHMIDT1, A. KOSCHINSKY1, D. GARBE-SCHÖNBERG2 3 AND R. SEIFERT 1

The relative abundances of highly siderophile elements (HSE: Os, Ir, Ru, Pt, Rh, Pd) in the Earth’s primitive upper mantle (PUM) and the continental upper crust (UCC) is a key issue for understanding their origin and the influence of impactors on the chemical composition. The past twelve years we have measured the HSE in many mantle suites of the Earth by neutron activation. Estimates of Rh/Ir, Ru/Ir, Pd/Ir, and Pt/Os derived from most mantle suites indicates modestly suprachondritic compositions for average PUM. The enigmatical mantle Ru/Ir ratio of 2.01 ± 0.12 (Schmidt 2004) in comparison of 1.51 ± 0.04 of the known different chondrite groups has been observed by other groups as well (Pattou et al. 1996; Becker et al. 2006). From highly melt depleted peridotites (Schmidt et al. 1998) and metasomatized mantle xenoliths (Schmidt et al. 2003) we conclude that the effects of melt depletion, refertilization and metasomatism can be discounted. However, the Os, Ir, Ru, Pt, and Pd pattern on PUM perfectly match IVA iron meteorite data (Charlotte) recently derived by Walker et al. (2005). The question raises if HSE in PUM are added to the accreting Earth by a late bombardment of (a) iron meteorites, (b) some unsampled inner solar system materials (more highly fractionated than enstatite chondrites), or (c) excess Rh, Ru, Pt, and Pd in the depleted mantle may have predated the late veneer, or (d) added by crustal recycling (Schmidt et al. 2000), or (e) HSE are mixed from the differentiated outer core into the mantle after core formation (Snow and Schmidt 1998)? The HSE and Ni pattern of the upper continental crust (UCC) is strongly fractionated. Highly fractionated Ru/Ir is unparalleled in terrestrial magmatic systems. In fact, the HSE and Ni systematics closely resembles IIIAB iron meteorites (many impact craters on Earth are produced by this type of iron meteorite projectiles, e.g. Schmidt et al. 1997), pallasites, and the evolved suite of Martian meteorites (Shergotty, Nakhla, Zagami), possibly representing the elemental pattern of the Martian crust (Jones et al. 2003). Probably the UCC preserves an imprint of such meteorites. Calculations show that a projectile with a radius of 54 ± 4 km as an upper limit would yield the abundances of HSE and Ni in the UCC.

References Becker H. et al. (2006) Geochim. Cosmochim. Acta 70, 4528. Pattou L., Lorand J. P., Gros M. (1996) Nature 379, 712-715. Schmidt G. (2004) Meteoritics & Planet. Sci., 39, 1995-2007. Schmidt G. et al. (1997) Geochim.Cosmochim.Acta 61, 2977. Schmidt G. et al. (1998) Eos Trans. AGU. 79(45). Schmidt G. et al. (2000) Chemical Geology 163, 167-188. Schmidt G. et al. (2003) Chemical Geology 196, 77-105. Snow J. E. and Schmidt G. (1998) Nature 391, 166-169. Walker R. J. et al. (2005) LPSC XXXVI, 1313. Jones J. H. et al. (2003) Chemical Geology 196, 21-41.

Jacobs University Bremen, Earth and Space Sciences, 28759 Bremen, Germany; ([email protected]) 2 Universität Kiel, Insitute of Geosciences, 24098 Kiel, Germany; ([email protected]) 3 Universität Hamburg, Insitute of Biogeochemistry, 20146 Hamburg; ([email protected]) Within the German Special Priority Program 1144 "From mantle to ocean: energy-, material-, and life-cycles at spreading axes", unique hydrothermal fluids were sampled at two newly discovered hydrothermal vent fields located at the spreading axis of the Mid-Atlantic Ridge at 4°49'S. Turtle Pits and Sisters Peak, both located at 3000 m water depth, emanate vapor phase fluids with stable sampling temperatures of about 390°C-407°C. In addition, temperature measurements recorded short-term spikes of up to 464°C, by far the highest temperature ever measured in submarine hydrothermal fluids. The fluids, representing a supercritical vapor phase sampled near the critical point of seawater, are characterized by a Cl depletion of 50% compared to seawater. The estimated temperature and depth of the initial phase separation with ~430°C, and 360 bar agree with the depth of the reaction zone about 500 m below the seafloor (constrained by Si concentrations). The overall fluid geochemistry is similar to those observed in other phase-separating systems at mid-ocean ridges, with an extremely high mobility of Fe and enhanced H2 concentrations. Most chloride-normalized element concentrations are greater than seawater, indicating advanced water-rock interaction. Despite a similar chlorinity and major element composition, different fluid samples emanating from the same vent orifice show a very heterogeneous endmember composition with respect to metals such as Cu, Zn, Pb and rare earth elements, indicating the mixing of the superhot vapor phase with another liquid or solid phase. Small differences in the fluid geochemistry between Turtle Pits and Sisters Peak can be related to a somewhat deeper zone of phase separation beneath the Sisters Peak field. Total and Cl-normalized major and trace element concentrations will be evaluated with respect to possible mixing proccesses of different types of fluids in the subsurface. The respective influences of vapor-brine fractionation, (low-Cl) water-rock equilibria, and a different elemental speciation in supercritical vapors on the fluid geochemistry of the Turtle Pits and Sisters Peak fluids will be discussed.

Goldschmidt Conference Abstracts 2007

Fluids, melts, and supercriticality in the MSH system and element transport in subduction zones M.W. SCHMIDT AND E. MELEKHOVA Dep.Earth Sciences, ETH Zürich, Switzerland ([email protected], [email protected]) The dichotomy of aqueous fluids and hydrous melts which includes fluid saturated melting is known to be limited in pressure, as the critical temperature of the miscibility gap between the two liquid phases strongly decreases with pressure. The critical P-T curve of the immiscibility gap intersects with the fluid saturated solidus at a critical pressure between 1 (SiO2-H2O) and 12 GPa (MgO-SiO2-H2O), nevertheless, the critical pressures for natural rock systems are largely unknown and under discussion. This discussion stems from the imperfection of all available experimental methods, which will be critically reviewed. Our new experimental results on the MSH peridotite model system suggest that the entire system goes supercritical around 12 GPa, where phase A and chondrodite are stable at the solidus in the Mg-rich portion of MSH. Supercriticality moves stepwise from the SiO2-rich side to the MgO-rich side, and hitherto insufficiently defined equilibria involving enstatite are needed to define the MSH phase diagram completely. While supercriticality eliminates fluid saturated melting, it is irrelevant to fluid-absent melting, and thus to devolatilization of deeply subducted oceanic mantle. Even above 12 GPa, phase E has a realistic chance to undergo fluid absent melting in thermally relaxing subducted lithosphere. Secondly, while the question of supercriticality of the fluid-melt pair is petrologically challenging, it is, also in the presence of a fluid, of moderate practical relevance to the real world. In subduction scenarios, we do expect to have one liquid (either fluid, or melt or supercritical liquid), compositionally buffered by and coexisting with the residual minerals. The main questions relate then to solubilities of key elements in and to the viscosity and wetting behavior of this liquid.

A897

Is “ballen quartz” diagnostic for shock metamorphism? M. SCHMIEDER AND E. BUCHNER Institut für Geologie und Paläontologie, Universität Stuttgart, Herdweg 51, D-70174 Stuttgart, Germany Ballen quartz has been observed in impactites from a number of terrestrial impact structures, predominantly in impact melt rocks, suevites, and target rock clasts affected by strong post-shock heating. In thin section, ballen are generally roundish to oval-shaped and typically range in size from several microns to ~200 μm. Ballen rims may vary in brightness and colour, and appear as curved open fractures, fractures filled by submicroscopic phases, or very faint lines. Accordingly, ballen quartz, with its typical crackled “fishscale” pattern, has turned out to be a distinctive textural feature suggestive for shock metamorphism [1-5]. In addition to the reports available in literature, we detected ballen quartz in impact melt rocks from Manicouagan, Rochechouart, Kara, Terny, and strongly heated crystalline basement clasts from the Ries crater and Ilyinets. As ballen quartz is known from very small but also the largest terrestrial impact structures (e.g., Tenoumer or Chicxulub), one must assume that ballen quartz should have been primarily present at all terrestrial impact sites. Three types of ballen quartz (homogenous ballen, optically individual ballen, and microcrystalline ballen) are distinguished by their optical properties corresponding to different shock stages [2]. Various models of ballen quartz formation have been proposed to date, such as by prograde silica transformations in the solid state [3;4] or by the melting of silica, lechatelierite formation, and retrograde recrystallisation [1;5]. However, it is widely accepted that quartz ballen represent pseudomorphs after cristobalite [3-5]. The formation mechanism of ballen quartz, nevertheless, still appears to be somewhat enigmatic. Domains of typical “granular cristobalite” in siliceous volcanic rocks [6], fulgurites [7], and heated industrial quartz ceramics [8] - i.e., non-impact materials of emphasised thermal history - display essentially the same textural pattern, which further suggests that ballen quartz originates from granular cristobalite and that ballen texture develops during cristobalitization of silica (and the high-low inversion of cristobalite [5]) at high temperatures. Therefore, ballen quartz might not be restricted to impactites, and we put into question that the sole presence of ballen quartz can be considered as reliable evidence for shock metamorphism. References [1] French, B.M., (1998), LPI Contrib. 954. [2] Bischoff, A. and Stöffler, D., (1984), J. Geophys. Res. 89, B645-B656. [3] von Engelhardt, W., (1972), Contrib. Min. Petrol. 36, 265292. [4] Stöffler, D. (1993), Meteoritics 28, 444. [5] Carstens, H., (1975), Contrib. Min. Petrol. 50, 145-155. [6] Swanson, S.E, Naney, M.T., Westrich, H.R. and Eichelberger, J.C., (1989), Bull. Volcanol. 51, 161-176. [7] Rogers, A.F., (1946), J. Geol. 54, 117-122. [8] Akh’yan, A.M., (1986), Refr. Ind. Ceram. 27, 515-520.

A898

Goldschmidt Conference Abstracts 2007

Natural mass-dependent Cd isotopic variations determined by TIMS

Zircon U-Th ages from Laacher See indicate coeval crystallization of coerupted carbonatite and silicate magmas

A.-D. SCHMITT1,2, S. J. G. GALER1 AND W. ABOUCHAMI1 1

Max-Planck-Institut für Chemie, Postfach 3060, 55020 Mainz, Germany ([email protected]; [email protected]) 2 Université de Franche-Comté, Département des Géosciences, 16 route de Gray, 25030 Besançon Cédex, France ([email protected]) We present a high-precision double spike (DS) technique for Cd isotopes in which the Cd is measured by thermal ionisation mass spectrometry (TIMS). Since we observe anomalous odd-even isotope mass bias during measurement, we utilize even isotopes of Cd only. The "natural" fractionation is expressed as deviations in 112Cd/110Cd (in parts per 104) from our JMC Cd shelf standard. The commonly-used "JMC Münster Cd" standard lies at -1.11 H112/110Cd on this scale. The external reproducibility (2SD) for double-spiked JMC Cd shelf (100 ng) is ±0.14 H112/110Cd (i.e. ±7 ppm/amu) which is a factor of 4 to 10 times better than that reported in previous MC-ICP-MS studies (e.g. Wombacher et al., 2003; Lacan et al., 2006; Ripperger and Rehkämper, 2007). We have analyzed H112/110Cd in over sixty samples from different terrestrial reservoirs and environments (MORB, OIB, loess, Fe-Mn nodules, continental and hydrothermal sulphides). Our study confirms the limited variations found by Wombacher et al. (2003) in terrestrial materials—the total range is ~5 H112/110Cd, but most samples lie between -1.0 and +1.0. Three mid-T hydrothermal sulphides differ by ~4 H112/110Cd, suggesting significant Cd isotopic fractionation occurs in this setting, though other sulphides are much more tightly clustered. Analyses of 31 hydrogenous Fe-Mn deposits (and phosphorites) worldwide range from -0.6 to +2.0, and display isotopically heavier Cd (by 1 to 2 H112/110Cd) in the top 2000 m than in the deeper ocean. These differences might reflect shallow inorganic scavenging of Cd by Fe-Mn oxides and its remineralization at depth. However, allowing for interlab bias, three shallow seawater samples of Ripperger and Rehkämper (2007) have H112/110Cd of +0.5 to +2.2. These data are quite consistent with our shallow Fe-Mn nodule data, implying that only minor Cd isotope fractionation occurs on Fe-Mn oxide precipitation from seawater. The principal cause of the heavy Cd in the surface ocean then appears to be partitioning of light Cd into phytoplankton, entirely consistent with the sense and magnitude of the fractionation factor (7 ± 3 H112/110Cd) determined by Lacan et al. (2006). Overall, these considerations suggest that H112/110Cd is potentially useful as a paleoproductivity proxy, in much the same way as Cd/Ca. References Lacan F., François R., Ji Y. and Sherrell R. M. (2006), Geochim. Cosmochim. Acta 70, 5104-5118. Ripperger S. and Rehkämper M. (2007), Geochim. Cosmochim. Acta 71, 631-642. Wombacher F., Rehkämper M., Mezger K. and Münker C. (2003), Geochim. Cosmochim. Acta 67, 4639-4654.

A. K. SCHMITT1 AND G. WÖRNER 2 1

Department of Earth and Space Sciences, University of California Los Angeles, USA ([email protected]) 2 Geowissenschaftliches Zentrum Göttingen, Abteilung Geochemie, Universität Göttingen, Germany ([email protected]) In order to constrain the origins of spatially associated carbonatite and silicate magmas and to quantify the time scales during which they interacted in a shallow magma system, we are employing U-Th dating of individual zircon crystals by secondary ionization mass spectrometry (SIMS). Laacher See (Germany) erupted a suite of crystal-rich xenolithic clasts that range from carbonate-free intermediate (dominantly syenitic) compositions to carbonatites (sövites). Because of the youthful age of the Laacher See eruption (~12,900 a BP), SIMS U-series geochronology can be applied which provides unrivaled temporal resolution in the order of ~103 – 104 years for individual grains or grain domains that are analyzed in situ (thin-section) at ~30 Pm lateral resolution and crater depths <5 Pm. Our initial results from three syenitic sövite clasts yield overlapping U-Th zircon isochron ages that average 15.8±2.7 ka (1V; MSWD = 0.45; 30 spot analyses). This age overlaps within uncertainty with the U-Th isochron age for zircons from early erupted Laacher See differentiated phonolite and late-crystallized zircons in vesicles of carbonate-free syenite ejecta clasts (17.1±1.3 ka; Schmitt, 2006). These results indicate coeval crystallization of carbonatite and differentiated silicate magmas that were stored at the top of a shallow magma body. Thus far, the zircon record lacks unequivocal evidence for protracted (several 10’s to 100’s of ka) pre-eruptive crystallization within a long-lived shallow Laacher See magma chamber and rather suggests crystallization and differentiation at most within few ka prior to eruption. The geochronological results are also in line with trace element evidence that indicate a close chemical affinity between carbonatites and Laacher See phonolites (Liebsch, 1996). Coeval zircon crystals in both, Laacher See carbonatites and phonolites/syenites have indistinguishable initial (230Th)/(232Th) ratios (0.885±0.013 and 0.894±0.010, respectively). This further supports the hypothesis of consanguineous carbonatite and silicate melts in the Laacher See magma chamber that originated by liquid immiscibility.

References Liebsch H., (1996) PhD thesis, University of Göttingen, 1-111. Schmitt A.K., (2006) Geology 34, 597-600.

Goldschmidt Conference Abstracts 2007

A potential site for long-term in situ cosmogenic 3He and 21Ne production rate calibration on Fuerteventura, Canary Islands

High resolution S-XRF study of stardust impact tracks S. SCHMITZ1, F.E. BRENKER1, L. VINCZE², B. VEKEMANS², M. BURGHAMMER3 AND C. RIEKEL3 1

JWG University, Dept. of Mineralogy, Frankfurt M., Germany ([email protected]) 2 University of Ghent, Dept. of Analytical Chemistry, Ghent, Belgium 3 ESRF, Grenoble, France

Introduction Stardust was the first sample return mission which captured extraterrestrial solid particles on a flight-by through the coma of a comet (Wild 2) which enables the investigation of cometary dust in its direct contextual setting [1] with the best available analytical system available in any laboratory on Earth. The nature of cometary material is of great importance for the understanding of the history of the early solar nebula and the protoplanetary disk.

Data processing Several keystones, each including an impact track with a terminal particle (TP) at its end, were studied [2]. Track No. C2044, 37 was investigated with high resolution synchrotron XRF and XRD (S-XRF/XRD) techniques at beamline ID13 at the ESRF (Grenoble, France) with 13 keV to determine the elemental abundances from Ca to Se. The focus spot size was 200 nm enabling an investigation with very high spatial resolution of grains having sub micron size.

Results The evaluation of the diffraction pattern obtained from the TP yield a mixture of olivine and pentlandite and/or phyrrotite [3]. S-XRF data reveal several chemical hot spots along the track with most of these fragments having sizes of only several 100 nm. Some have Fe concentrations of more than 20 wt. % indicating Fe–Sulphide as possible phases along the track. These fragments are also high in Mn and Cr in comparison to the other hotspots. Fe-Ni sulphide or - metal is also assumed for a fragment containing up to 2.5 wt. % Ni. For the TP, S-XRF reveals a complex chemical structure showing band like Fe- and Cr-features inside the particle. Feand C1 normalized abundances for the TP show relative enrichments for Cr and Mn, depletions for Ni and CI abundance for Ca, Ti, Cu and Zn. References [1] Brownlee et al., (2003), JGR 108 (1), 1-15. [2] Flynn et al., (2006), Science 314, 1731-1735. [3] Zolensky et al., (2006), Science 314, 1735-1739.

A899

B.S.H. SCHNEIDER 1; J.R. WIJBRANS 1, F.M. STUART2 AND J.P.T. FOEKEN 2. 1

Vrije Universiteit, Amsterdam, The Netherlands ([email protected]) 2 SUERC University of Glasgow, United Kingdom ([email protected]) The use of terrestrial cosmogenic nuclides (TCN) as a tool for quantitative geomorphology has rapidly increased over the last decade or so. Production rates are poorly understood and are perhaps the biggest constraint on their use as absolute chronometers. The CRONUS-EU network aims to improve TCN production rates. In this particular study we are focusing on potential calibration sites on Fuerteventura in the Canary Islands (29°N) because sets of scaling factors used for determining production rates so far diverge most prominently in at latitudes below 40° (Dunai, 2000). Furthermore most reported production rates are of late Pleistocene or even Holocene age making them susceptible to the influence of short term variations in the Earth’s magnetic field. The production rate of cosmogenic 3He can be determined from the analysis of olivine and pyroxene phenocrysts from wellpreserved flow tops of independently-dated basaltic lavas. We have identified several suitable flows from the most recent Series IV volcanic units from the north side of Fuerteventura (Coello et al., 1992). 40Ar/39Ar incremental heating experiments of 9 flows yield eruption ages of 50 - 400 ka with total uncertainties of 5 % and less. Analysis of 3He in pyroxene and olivine phenocrysts from sea-level samples have potential as calibration for short- to mid-term production rates at low latitude, as well as allow a comparison of elemental control on cosmogenic 3He production. Cosmogenic 21Ne accumulation in 400 ka flows should be high enough to measure with reasonable accuracy. We aim to present new cosmogenic 21Ne production rate estimates from the oldest flows.

References Coello, J., Cantagrel, J.-M., Hernan, F., Fuster, J.-M., Ibarrola, E., Ancochea, E., Casquet, C., Jamond, C., Diaz de Teran, J.-R., Cendrero, A. (1992), Volcanol. Geotherm. Res. 53, 251-274. Dunai, T.J. (2000), EPSL 176, 157-169.

A900

Goldschmidt Conference Abstracts 2007

Multiple hydrothermal fluid flow in the NW Rhenohercynian

The Cr isotope label of the solid Earth and hydrothermal pathways

J. SCHNEIDER1, M. CHATZILIADOU2, S. SINDERN2, CH. HILGERS2 AND U. KRAMM2

R. SCHOENBERG, S. ZINK, M. STAUBWASSER AND F. VON BLANCKENBURG1

1

Department of Geology and Centre for Archaeological Sciences, K.U.Leuven, Celestijnenlaan 200E, B-3001 Leuven, Belgium ([email protected]) 2 Institut für Mineralogie und Lagerstättenlehre, RWTH Aachen, Wüllnerstr. 2, D-52062 Aachen, Germany ([email protected]) The NW Rhenohercynian belt of the Central European Variscan orogen, studied in the northern Eifel area, was affected by several pre-, syn- and postorogenic hydrothermal events. Investigation of core material from the RWTH-1 drill hole in Aachen revealed mineralogical, geochemical and structural evidence of intense Variscan hydrothermal activity. Recently published temperatures of up to 370 °C (Lögering et al. 2006) but low illite ‘crystallinites’ can only be explained by a short duration (< 10000 a) of this hydrothermal event, which was initiated by dewatering of sedimentary rocks during Variscan compressional tectonics. This synorogenic fluid flow did not result in significant large-scale element transfer, except for nitrogen which was incorporated into clay minerals as ammonium. Important postorogenic, sandstone-hosted Pb-Zn deposits in the northern Eifel have been dated at 170 r 4 Ma (Schneider et al. 1999). This Jurassic event caused large-scale element mobilisation in the Palaeozoic sedimentary rocks, as indicated by the homogeneous Pb isotope signatures of the sulfide ores. However, new geochronological data give evidence for at least one Cretaceous hydrothermal remobilisation event in the NW Rhenohercynian. Direct RbSr-dating of non-economic vein sphalerites from the Sandgewand fault (Hastenrath near Stolberg) yields an age of 134 r 2 Ma, while the Pb isotope signatures of these ores are identical to the Jurassic ones. Thermal spring waters in Aachen represent the youngest stage of hydrothermal events in the NW Rhenohercynian. They are characterized by high concentrations of particlebound Zn. Isotope ratios of Pb and Sr determined directly on water and sinter samples indicate several metal sources, such as postorogenic base metal ores but also evolved DevonianCarboniferous sediments and a yet unidentified source characterized by unradiogenic Pb isotope signatures.

References Lögering, M.J., Kolb, J., Meyer, F.M. and Schwarzbauer, J.C. (2006):. Abstract TSK 11, 22-24, Göttingen, Germany Schneider, J., Haack, U., Hein, U.F. and Germann, A. (1999): Proc. 5th Bienn. SGA Meeting, London, 22-25 August, 1999: 1287-1290.

Institute for Mineralogy, Leibniz University of Hannover, Callinstrasse 3, D-30167 Hannover, Germany The Cr isotope compositions of Earth’s major silicate rock reservoirs and Cr-bearing minerals of metamorphic and hightemperature hydrothermal origin were obtained by highprecision double-spike MC-ICP-MS Cr measurements. The principle igneous reservoirs of the Earth share a common stable Cr isotope composition. Within the 95% confidence interval, Student’s t-tests do not allow to distinguish the Cr isotope compositions of seven mantle lherzolites (G53/52CrSRM979 = –0.104±0.110‰; 2 SD), ten ultramafic cumulate rocks (–0.119±0.113‰) and six continental and oceanic basalts that represent partial mantle melts (–0.151±0.050‰). Thus, unlike suggested for Fe isotopes [1], partial mantle melting does not appear to fractionate Cr isotopes. The average G53/52CrSRM979 value of – 0.082±0.058‰ for 12 different chromitite seams from the Bushveld and Great Dyke layered igneous intrusions is identical to that of the silicate igneous Earth reservoirs. This again supports the notion that Cr isotopes do not fractionate during partial mantle melting, since chromites crystallize from Cr-rich partial mantle melts that enter an overlying magma chamber containing an evolved Cr-depleted liquid. The Cr isotope compositions of Cr(III)-bearing uvarovite and fuchsite minerals from amphibolite-facies metamorphic terraines are also equal to those of igneous rock reservoirs. Since partial mantle melting and metamorphism do not involve redoxchanges for Cr and differences in the Cr-ligands of solids and melt, high-temperature isotope fractionation appears to be too small to be detectable. However, crocoite (PbCr(VI)O4), a mineral that forms in oxidizing zones of high-temperature hydrothermal systems, shows G53/52CrSRM979 values from +0.640 to +1.037‰. Precipitation experiments of crocoite from Cr(VI) solutions reveal (equilibrium) isotope fractionation with '53/52CrcrocoiteCr(VI)aq § 0.1 to 0.2‰. This is too small to explain the heavy Cr isotope compositions of crocoite. One possible explanation is that oxidation of dissolved hydrothermal Cr(III) to Cr(VI) is accompanied by isotope fractionation. Alternatively, Cr behaves like its chemical twin Se that only fractionates during reduction [2]. Partial Cr(VI) reduction [3] may explain the heavy compositions of the crocoites, which in turn allowed quantification of redox-changes along the hydrothermal pathway.

References [1] Weyer, S. et al. (2005), Earth and Planet. Sci. Letters 240, 251ff. [2] Johnson, T.M. and Bullen, T.D. (2004), Reviews in Mineralogy&Geochemistry 55, 289ff. [3] Ellis, A.S. et al. (2002), Science 295, 2060ff.

Goldschmidt Conference Abstracts 2007

Experimental study on crustal wehrlites of the Oman ophiolite

Recent developments in U-Pb thermochronology

STEPHAN SCHOENBORN1, JUERGEN KOEPKE1, SANDRIN T. FEIG1, FRANCOISE BOUDIER2 AND ERIC HELLEBRANDT 3

B. SCHOENE1,2 AND S.A. BOWRING.1

1

Institut für Mineralogie, Leibniz Universität Hannover, Germany ([email protected]) 2 ISTEEM, University Montpellier, France ([email protected]) 3 SOEST, University of Hawaii, USA ([email protected]) In the Wadi Haymiliyah of the Oman ophiolite (Halyn block), discordant wehrlite bodies ranging in size from decameters to hundreds of meters intrude the layered gabbro series at different crustal levels. Common feature of all wehrlite bodies is the dominance of olivine and clinopyroxene with high Mg# (MgO/(MgO+FeO)*100, molar) varying between 85 and 95. SIMS analyses of clinopyroxenes from different wehrlite bodies reveal REE and trace element patterns suggesting that the clinopyroxenes were crystallized from tholeiitic, MORB-type melts which were highly depleted in incompatible trace elements. Some wehrlites from the upper section bear mm-sized poikilitic pargasite as magmatic phase, implying that the associated melts showed a high water activity, at least in a late stage. In order to evaluate the genesis of the Oman crustal wehrlites, we combined experimental work with investigations on natural rocks. The experimental study under controlled fO2 and aH2O at pressure of 200 MPa is aimed to constrain the physical conditions of wehrlitic magmas within the lower oceanic crust. Especially, we want to clarify whether the crustal wehrlites from the Oman ophiolite are derived from wet primitive tholeiitic magmas, as suggested by a model of Feig et al. (2006). Starting material are mixtures of natural olivines and clinopyroxene separated from wehrlites from the lower Wadi Haymiliyah section, and a synthesized glass representing the "lost" equilibrium melt. Experiments were performed in AuPd capsules (pre-saturated with iron) at temperatures between 1020 and 1160°C with fO2 varying between QFM and QFM+2 (QFM=quartz fayalite magnetite buffer). First experiments at nominal dry conditions above 1140°C (fO2: ~ QFM) reveal the stability of plagioclase (newly formed crystals), while the doped clinopyroxene became unstable (decomposition by reaction with the melt). Thus, we were not able to reproduce the characteristic wehrlite paragenesis (olivine plus clinopyroxene without plagioclase) under dry conditions, implying that wehrlitic crystal mushes can not be generated in dry, primitive tholeiitic systems at crustal pressure by simple accumulation of fractionating crystals. The corresponding experiments at wet conditions are in progress.

References Feig S.T., Koepke J. and Snow, J., (2006), Contrib. Mineral. Petrol. 152. 611-638.

A901

1

Department of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA, USA ([email protected]) 2 Université de Genève, Section des Sciences de la Terre, Geneva, Switzerland ([email protected]) Investigations can now be designed to exploit and explore volume diffusion in accessory minerals with the goal of generating accurate temperature-time paths for rocks. Despite the abundance of high-U minerals with high to moderate nominal closure temperatures for the diffusion of Pb (e.g. titanite, apatite, and rutile; Tc ~ 400-650 ˚C), U-Pb thermochronology has received little attention compared to lower temperature systems such as 40Ar/39Ar and U/Th-He. However, understanding thermal histories of rocks in the 400650 °C temperature range is crucial for deconvolving the thermal histories of middle and lower crustal rocks. We have investigated the potential and limitations of U-Pb thermochronology through single- to sub-grain ID-TIMS analysis of apatite and titanite in conjunction with BSE and CL imaging and numerical modeling. The precision of U-Pb thermochronology is usually limited by relatively low ratios of radiogenic lead to initial (a.k.a. common) lead. In systems that have undergone complex rock-fluid interactions, the common practice of using the isotopic composition of Pb in “coexisitng” low-U minerals such as feldspar for this correction may be inaccurate. This is especially important in slowlycooled rocks where the common Pb composition due to fluid interaction may change in time and/or space. In minerals where volume diffusion controls the distribution of daughter product in a mineral, one expects intra-grain diffusion gradients such that the smallest grains and the rims of large grains record the youngest dates. We have explored relationships between grain size and date (here called a-t curves) in apatite and titanite in combination with numerical diffusion modeling. Given some geologic constraints, utilizing a-t curves in two minerals from the same rock will allow, in some cases, the generation of unique and non-linear temperature-time paths regardless of the effects of post-cooling reheating from magmatic intrusions. Numerical modeling can also be used to evaluate the effect of differences in U zonation, cooling rate, and grain shape on closure temperature and a-t curves. In many cases these minerals do not form a-t curves. This is likely because U-Pb thermochronometers are often involved in metamorphic reactions at a range of temperatures, and therefore characterizing the internal zonation, trace element patterns, and petrographic setting of these minerals may be crucial for generating accurate thermal histories of rocks. This work in conjunction with studies exploiting the U/Th-He method will become more useful in characterizing the behavior of U and its daughter products in high-U thermochronometers and determing thermal histories of rocks from lower-crustal to near surface conditions.

Goldschmidt Conference Abstracts 2007

A902

REE systematics of fluorites in (per-)alkaline systems

Structural Chemistry of Cationdoped Bacteriogenic UO2

JOHANNES SCHOENENBERGER, JASMIN KOEHLER AND GREGOR MARKL

ELEANOR SCHOFIELD1, RIZLAN BERNIER-LATMANI2, HARISH VEERAMANI2, JOSH SHARP2, DAVID L. CLARK3, STEVEN D. CONRADON3 AND JOHN R. BARGAR1

Universität Tübingen, Inst. f. Geowissenschaften, D-72074 Tübingen, Germany, ([email protected], [email protected], [email protected]) The mid-Proterozoic Gardar Province (South Greenland) is a failed rift province hosting 12 major (per-)alkaline intrusions [1]. High fluorine contents occur in a great variety of magmatic rocks all over the Gardar Province. Rare earth element (REE) patterns of fluorite and associated minerals like cryolite (Na3AlF6), calcite and siderite provide information about the magma source, crystallisation history, fluid-melt interaction and late-magmatic processes. We focused on fluorites from three intrusions representing the whole compositional range of the Gardar Province: the Ilímaussaq intrusion which is mainly composed of silicaundersaturated agpaitic nepheline syenites; the Motzfeldt intrusion with its mostly miaskitic nepheline syenites; and the Ivigtut intrusion famous for its now mined-out cryolite deposit hosted by an A-type granite [1]. All chondrite-normalized REE patterns of fluorite show a negative Eu anomaly which is attributed to feldspar fractionation. The REE patterns of the Motzfeldt and Ilímaussaq intrusions in the eastern part of the Gardar province show a positive Y anomaly if Y is plotted between Dy and Ho according to [2]. In contrast, the patterns of the Ivigtut intrusion (in the western part of the province) exhibit a negative Y anomaly. This regional difference might be ascribed to heterogeneities in the magma source as XRF data of basaltic to rhyolitic dykes in the Ilímaussaq and Motzfeldt region are characterized by higher Y contents than dykes from the Ivigtut region. Primary magmatic fluorite of the Motzfeldt and Ilímaussaq intrusions show almost identical patterns with an enrichment in light REE suggesting an early crystallisation [3]. Later crystallised fluorites of veins are less concentrated in REE and exhibit flatter patterns or enrichment of the middle/heavy REE typical of the late-magmatic to hydrothermal crystallisation stage. The REE patterns of minerals from Ivigtut are slightly enriched in heavy REE and indicate a strong interaction with a fluid phase as evidenced by the very pronounced tetrad effect. Our preliminary results show that REE in fluorite reflect the source but can also act as tracers for the transition from early magmatic to hydrothermal conditions during the crystallisation of (per-)alkaline intrusions. The tetrad effect reflects strong water-rock interaction and the fluid’s influence on the fluorite genesis.

References [1] Upton et al. (2003), Lithos 68 43-65. [2] Bau and Dulski (1995), Contr. Min. Petr. 119 213-223. [3] Möller et al. (1976), Mineral. Dep. 11 111-116.

1

Stanford Syncrhotron Radiation Laboratory, 2575 Sand Hill Road, Menlo Park, CA 94025 ([email protected], [email protected]) 2 Ecole Polytechnique Federale de Lausanne, Ch 1 644 Station 6, Ch-1015, Lausanne, Switzerland ([email protected], [email protected], [email protected]) 3 Los Alamos National Laboratory, MS-G756, ADSMS, Los Alamos, NM 87545 The chemical stability of bacteriogenic uraninite, “UO2”, is one of the seminal issues governing its success as an in situ immobilization strategy in remediated subsurface locations. Little detail is known about the structure and reactivity of this material, but based on comparison to its closest abiotic analog, UO2+x (0 < x < 0.25), we expect that it is complex and disordered, likely to exhibit non-stoichiometry, and capable of structurally incorporating common ground water cations and U(VI). These subtle changes in mineralogy are expected to substantially impact its stability in ground water. In this study, the product of microbial U(VI) reduction under varying conditions of pH, carbonate and divalent cation concentration (Mg2+, Mn2+, and Ca2+) was investigated. To facilitate x-ray scattering measurements, a gentle aqueous cleaning method was developed to separate the biooxide and organic components without altering the biooxides. The local and long-range atomic structures and nano-scale structures of the wet oxides have been measured using EXAFS, WAXS, SAXS and TEM. The material exhibits a range of particle sizes with a mean around 4 nm. The first oxygen shell is distorted, consistent with a nonstoichiometric composition. Mn2+ and Mg2+ were found to attenuate the particle size of bacteriogenic UO2+x and to be structurally incorporated. These findings suggest that ground water composition can have a pronounced impact on the structure and properties of bacteriogenic uraninte.

Goldschmidt Conference Abstracts 2007

Do particulate 231Pa/230Th ratios depend on water depth?

Trace element behavior in sulfidic porewaters of the Oder estuary, SW Baltic Sea

JAN C. SCHOLTEN1 AND J. FIETZKE2

F. SCHOLZ1 AND T. NEUMANN1

1

Marine Environment Laboratories, International Atomic Energy Agency, 4 Quai Antione 1er, 98000 Monaco ([email protected]) 2 IFM-GEOMAR, Wischhofstr. 1-3, 24148 Kiel, Germany ([email protected]) The natural radionuclides 230Th (T1/2 = 75.6 ky) and 231Pa (T1/2 = 32.1 ky) are widely used as tracers for paleoceanographic process studies. Both isotopes are produced in seawater by radioactive decay of uranium isotopes (234U, 235 U) and due to their particle reactivity they adsorb on sinking particles and are removed to the sediments. Since 231Pa is less particle reactive than 230Th i.e. 231Pa has a longer residence time (W) in the water column compared to 230Th (W = 230Th: 2040y; W = 231Pa: 80-200y) 231Pa is preferentially removed in areas of higher particle flux like e. g. ocean margins (boundary scavenging); and the variable 231Pa/230Th ratios stored in the sediment record are believed to reflect changes in the paleofluxes (or paleoproductivity) during the geological past. Further processes which may influence 231Pa/230Th ratios are the composition of sinking particles, i.e. the higher affinity of 231 Pa to adsorb on biogene opal compared to 230Th, and the dissolved 231Pa/230Th in water masses. In a study of 231Pa/230Th ratios in surface sediments from the southeast Atlantic off Namibia we find 231Pa/230Th ratios higher than the production ratio (P) of both isotopes (PPa/Th= 0.093) at the ocean margin and lower ones in the open ocean, a distribution typical of boundary scavenging. For this data set we observe a strong negative correlation (r2 = 0.82) between 231 Pa/230Th ratios in surface sediments and water depths. Also for previously published data from the Pacific and Arabian Sea we find negative correlations between 231Pa/230Th ratios in surface sediments and water depths, however, these correlations are less strong. We will discuss to what extent changes in the fractionation between 231Pa and 230Th with depth, high opal content, and/or boundary scavenging may cause 231Pa/230Th ratios to depend on water depth off Namibia.

A903

1

Institut für Mineralogie und Geochemie, Universität Karlsruhe, 76131 Karlsruhe, Germany ([email protected])

The diagenesis of trace elements in anoxic sediments of the Achterwasser lagoon, which is part of the Oder estuary in the Baltic Sea, is examined and evaluated in the context of pyrite formation. Certain trace elements show the same distribution pattern within the pore water regime suggesting a similar diagenetic behavior. The release of Zn, Cu and Cd into the pore water occurs prior to the reduction of Fe and Mn, indicating that they are released from organic matter in the thin oxic surface layer of the sediments. While a considerable part of these elements is recycled into the water column another part diffuses downcore and is removed as soon as H2S appears in pore water. Since degrees of trace element pyritization (DTMP) for Zn and Cd range well below 1%, it is inferred that they become trapped as monosulfide. Copper appears as both monosulfide and impurity in pyrite (DTMP: 12%). The accumulation of V, Cr, Co and As in the pore water starts concomitantly to that of Mn and Fe suggesting that these elements are liberated through reductive dissolution of Mnand Fe-oxyhydroxides. Increasing concentrations of dissolved V, Cr, Co and As with depth demonstrate that the fixation of these elements is not quantitatively controlled by the process of pyrite formation. While pyritization of Co and As is intense with median DTMP values of 35% and 17%, respectively, Cr (DTMP: 4%) is less and V (DTMP: <1%) almost not incorporated into pyrite. The coinstantaneous increase of DOP and DTMP with depth suggests that, in analogy to Fe pyritization, addition of trace metal sulfide complexes on preexisting pyrite grains may be responsible for increasing DTMP values for Co, Cu and Zn. Arsenic does not form sulfide complexes but rather As(III) oxyanions under reducing conditions, suggesting that again adsorption of complexes is responsible for increasing pyritization with depth.

A904

Goldschmidt Conference Abstracts 2007

Different sources of hydrocarbon pollution in surface sediments of the Campeche Sound, Gulf of Mexico, revealed by biomarker analysis BARBARA SCHOLZ-BÖTTCHER1, FELIPE VAZQUEZGUTIERREZ2 AND JÜRGEN RULLKÖTTER1 1

Institute of Chemistry and Biology of the Marine Environment (ICBM), Carl von Ossietzky University of Oldenburg, P.O. Box 2503, D-26111 Oldenburg, Germany ([email protected]) 2 Laboratorio de Fisicoquímica Marina, Instituto de Ciencias del Mar y Limnologia, Universidad Nacional Autonoma de Mexico, Apartado Postal 70-305, Mexico 04510. D. F., Mexico ([email protected]) The Campeche Sound in the Gulf of Mexico is the largest oil field in Mexico. Oil production, refinery activities, heavy ship traffic as well as natural petroleum seeps are potential sources of pollution of water and sediment in the area (GarcíaCuéllar et al., 2004). Embedded in a grid system of a long term environmental monitoring program (e.g., Vázquez and Virender, 2004) surface sediments from the shelf area of the Campeche Sound and also from the adjacent abyssal plain further west were studied. The nonaromatic hydrocarbon biomarkers in these surface sediments, several crude oils and drill cuttings samples were investigated by GC-MS. The overlapping fossil hydrocarbon envelopes indicate multiple sources like discharged drilling fluids, the heavy ship traffic or land run-off. Samples taken close to known asphalt seeps exhibit biomarker patterns virtually identical to those of the reference crude oils, and thus have apparently escaped from petroleum reservoirs with fractured cap rocks. These sediments and the asphalts both lack of angiosperm biomarker oleanane (cf. Moldowan et al., 1994) as expected for crude oils from a Jurassic source rock. Other surface sediments, superimposed on the fossil hydrocarbons, contain biomarkers neither typical of mature fossil fuels nor of typical immature organic matter in marine surface sediments. Their origin from drill cuttings recovered from Tertiary to Cretaceous deposits and disposed of into the sea would be a reasonable explanation. The regional distribution of indicative biomarkers will be shown and discussed.

References García-Cuéllar, J.A., Arreguín-Sánchez, F., Vázquez, S.H., Lluch-Cota, D.B., (2004). Intersciencia 29, 311-319. Moldowan, J.M., Dahl, J., Huizinga, B. J., Fago, F. J., Hickey, L. J., Peakman, T. M., Taylor, D. W. (1994). Science 265, 768-771. Vázquez, G.F., Virender, K.S., (2004). Marine Pollution Bulletin 48, 87-90.

The timescale of the Earth’s accretion and volatile loss: New constraints from Pd-Ag systematics M. SCHÖNBÄCHLER1,2, R. W. CARLSON2, M. F. HORAN2 2 AND E. H. HAURI 1

Impact and Astromaterials Research Centre (IARC), Imperial College, London SW7 2AZ, UK 2 Dept. of Terrestrial Magnetism, Carnegie Institution of Washington, Washington, DC 20015, USA The extinct radionuclide 107Pd decays to 107Ag with a halflife of 6.5 Myr and is a useful chronometer to study early solar system processes [1]. It provides constraints for the accretion and subsequent chemical differentiation of planetary bodies in a similar way as the 182Hf-182W system. In contrast to the refractory elements Hf and W, there is a distinct difference in volatility between the more refractory element Pd and moderately volatile Ag. Thus the Pd-Ag system can be used to study accretion and core formation but also volatile depletion. In particular, the Pd-Ag decay system has been successfully applied to study the formation and differentiation of iron meteorites [1]. Internal isochrons determined for these meteorites generally yield initial 107Pd/108Pd ratios in the range of 1.5 – 2.5 x 10-5 [1]. Combined with recent results for carbonaceous chondrites, this suggests that the initial 107 Pd/108Pd ratio of the solar system may be in the range of 6 x 10-5 [2]. Here we report new high precision Ag isotope data for a variety of terrestrial basalts from different geological settings. The data indicate an H107Ag of 2.1 ± 0.6 for the Ag isotopic composition of bulk silicate Earth relative to the NIST SRM 978a Ag standard. In contrast, the CV3 chondrite Allende has a well-constrained H107Ag of -0.4 ± 0.5. Assuming that the Earth accreted from material with similar volatile depletion as the CV3 chondrites and that Ag isotopes are not significantly affected by stable isotope fractionation, it is possible to model the Earth accretion and core formation. A simple two-stage model suggests that the last complete metal-silicate equilibration took place no longer than 15 Myr after the start of the solar system. This is at odds with the results of the wellinvestigated Hf-W system, where the same model yields a time of ~30 Myr. Different scenarios can be considered to reconcile the apparent discrepancy: (1) the Earth accreted from material having greater volatile depletion than CV3 chondrites (Pd/Ag > 20). Some ordinary chondrites feature such high Pd/Ag ratios. (2) Metal-silicate equilibration was incomplete and affected the Pd-Ag and Hf-W system differently. (3) Significant volatile depletion took place prior to the giant impact, for example during the accretion of the Earth.

References [1] Chen J. H. and Wasserburg G. J. (1996), in Earth Processes: Reading the Isotopic Code, Geophysics Monograph ’95, pp. 1-20. [2] Schönbächler et al. (2006), Meteorit. Planet. Sci. Suppl. 41, A5330.

Goldschmidt Conference Abstracts 2007

A905

Geomorphology and chemical weathering on basaltic islands

Development and application of novel organic proxies

HERDIS H. SCHOPKA AND LOUIS DERRY

STEFAN SCHOUTEN, JOHAN W.H. WEIJERS, FRANCIEN PETERSE, MARCEL T.J. VAN DER MEER AND JAAP S. SINNINGHE DAMSTÉ

Department of Earth and Atmospheric Sciences, Cornell University, Ithaca, NY 14853, ([email protected]) We are investigating the relationship between geomorphology and chemical weathering on actively eroding basaltic oceanic islands in the tropics, using the Hawaiian islands as a natural experiment. Our field sites include the islands of Kaua’i and Hawai’i, which represent the extremes in ages and soil development on the islands. Hawai’i is largely dominated by constructive volcanic shield topography. On Kaua’i early shield topography has been heavily dissected by fluvial erosion, creating a landscape of high relief characterized by serrated ridges and deep fluvial valleys. Less than ~10% of Hawaii’s surface has undergone fluvial erosion while only ~20% of Kauai’s original shield topography remains. This system, where relief increases substantially with age as the volcanic edifice is eroded by fluvial erosion, is very different from the conventional school of thought which postulates that after relief is created by an initial disturbance subsequent erosion will smooth the relief and lower it. We wish to investigate how the competing effects of increasing age and weathering are offset by increasing incision and development of relief. Initial observations indicate a link between Si concentration in river water and hillslope, where watersheds with steeper slopes exhibit higher concentrations of Si than do streams draining more gently sloping topography under similar flow conditions. In the Waimea Canyon on Kaua’i we have observed a nearly 4-fold increase in Si concentrations in the trunk of the river as it plunges over the edge of the Koke’e Plateau (the only remnant of Kauai’s original volcanic shield, ~1000 m a.s.l.) into the canyon (~300 m a.s.l.) and mixes with tributaries originating from the western wall of the canyon. The low Si concentrations in river water coming off the plateau reflect interaction with highly weathered and Sidepleted soils. The Si concentrations in the canyon tributaries tend to be 5-6 times higher than in the river water before it flows off the plateau, indicating that the tributaries tap sources of groundwater that has interacted with fresh rock at depth in the volcanic pile. Streams draining young bedrock on Hawai’i, where the river profile still follows to a large degree the original shield topography, have nearly as low Si concentrations as rivers draining old, stable shield surfaces, indicating that rivers in an early stage of incision do not tap groundwater sources. Groundwater that discharges directly to the ocean is an important geochemical pathway on Hawai’i and our data indicate its Si-concentrations are similar to the highest Si concentrations observed in the Waimea Canyon tributaries.

Royal Netherlands Institute for Sea Research, Department of Marine Biogeochemistry & Toxicology, PO Box 59, 1790 AB Den Burg, The Netherlands Sediments contain a myriad of organic compounds which potentially contain information on past microbial communities and the environments in which they lived. Research over the last few decades has improved our knowledge on the structure and origin of these compounds and their use as qualitative and quantitative proxies for past climatic conditions is rapidly increasing. Recently we developed several novel proxies based on the distribution and isotopic composition of terrestrial and aquatic biomarkers. Analysis of different soils from all over the globe showed a significant correlation between the distribution of branched tetraether lipids, derived from bacteria, and mean annual air temperature. We used this correlation to reconstruct continental temperature changes in several climatically important periods such as glacialinterglacial changes in the Quaternary and Cenozoic periods of global warming and cooling. Further validation of this proxy is currently underway by e.g. analysing soils surrounding hot springs. These soils are geothermally heated thereby providing a natural temperature gradient in soils of similar composition. Another proxy which we are currently developing is the relative deuterium content of algal biomarkers as a proxy for sea water salinity. First results show a significant correlation between the GD of long chain alkenones with salinity and GD of growth water. This has been used to reconstruct past salinity changes in the Holocene Black Sea and the Aegean Sea during time of sapropel S5 deposition. Further culture studies are now underway to examine other factors influencing the GD of alkenones and to test other biomarkers as potential paleosalinity proxies.

Goldschmidt Conference Abstracts 2007

A906

Peering at the subsurface biosphere through a diamond window

Using 222Rn for assessing nutrient transfer into the sea via SGD

MATTHEW O. SCHRENK

M. SCHUBERT1, A. SCHMIDT1, J. SCHOLTEN2, M.M. RUTGERS VAN DER LOEFF3 AND M. SCHLUETER3

Carnegie Institution of Washington-Geophysical Laboratory, 5251 Broad Branch Rd. NW, Washington, DC 20015, USA ([email protected]) Investigations of the subsurface biosphere have pushed the depth limits of microbial ecosystems to greater than 800 meters below the seafloor in marine sediments and 3-4 km into the continental lithosphere. In subsurface environments, the mode-of-growth for microorganisms is attached to minerals in structures known as biofilms. In many rockhosted, subsurface environments, organisms are confronted with multiple stressors including not only high pressures, but elevated temperatures and low energy fluxes. Subsurface microorganisms live at a precarious boundary between geologically-supported growth and cell death and remineralization. A significant limitation in the study of deep ecosystems has been an inability to distinguish and quantify microbial activities under conditions found in their native habitats. I will describe a research plan aimed at improving our ability to observe and characterize biogeochemical processes under conditions relevant to the deep subsurface environment. This plan relies upon the establishment of a one-of-a-kind research facility optimized for conducting high pressure experimental microbiology, borrowing from tools developed for materials science and hydrothermal geochemistry applications. Preliminary results show that the physiological status of microorganisms indigenous to the deep subsurface can be distinguished in vivo using molecular probes and geochemical measurements. A future goal is to make the system modular, allowing for its use in both laboratory and field-based studies. An analytical suite necessary to characterize biogeochemical transformations at microbe-mineral interfaces can be employed to follow the products of high pressure experiments, but will also be amenable to a range of studies in near-surface endolithic environments. The data obtained from these experiments will be important aid in deciphering both the extent and the biogeochemical consequences of a deep subsurface biosphere.

1

Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany ([email protected], [email protected]) 2 Marine Environment Laboratories, International Atomic Energy Agency, Monaco ([email protected]) 3 Alfred-Wegener Institut, Bremerhaven, Germany ([email protected], [email protected]) Submarine Groundwater Discharge (SGD) has to be considered an important pathway of contaminant/nutrient transfer from aquifers into the coastal sea. One of the main reasons for temporal changes in the SGD rate is the altering hydrological gradient between groundwater and sea determined by the tidal cycle. While SGD is known to sum up to about 6 % of the global river discharge into the oceans, the actual SGD rate in a particular coastal area is usually hard to be quantified due to the generally complex spatial SGD distribution. However, naturally occurring stable and radioactive isotopes have shown to be useful environmental tracers in SGD studies (e.g. Burnett and Dulaiova, 2003). At Cabbé, Bay of Roquebrune (Mediteranian Sea, France), the temporal SGD pattern was investigated over a 24 h period using naturally occurring 222Rn as an environmental tracer. At a location at the shore, known to represent a SGD spring, 222 Rn measurements of the sea water were carried out as time series using a portable radon monitor (Dulaiova et al., 2005). Tidal range, water temperature, and salinity were monitored using a CTD. Discreet samples for nitrate analyses were taken for assessing the nutrient transfer into the sea via SGD. In addition, off-shore surface water samples were taken along two near-shore transects. Temporal variations in 222Rn show a clear dependence on the tidal cycle. The results are backed up by the temperature and salinity data. High tide periods are characterized by lower 222 Rn concentrations, higher water temperatures, and higher salinities consistent with a decreased SGD rate. Variations of nitrate concentrations basically follow those of 222Rn, suggesting a strong dependence of the nutrient input into the sea on the SGD rate. In conjunction with the transect data and based on the nitrate and radon endmembers representative for groundwater (29.7 μM and 28.3 kBq/m3, respectively) and off-shore water (both below detection limit) a nitrate flux was estimated. The results suggest that SGD may be an important nitrate source for the coastal area investigated.

References Burnett W.C. and Dulaiova H. (2003). J.Env.Rad. 69. 21-35. Dulaiova H., Peterson R.G. and Burnett W.C. (2005). J. Rad. Nuc. Chem. 263. 361-365.

Goldschmidt Conference Abstracts 2007

Biogeochemical investigation of asphalt seepage at the Chapapote Knoll in the southern Gulf of Mexico FLORENCE SCHUBOTZ1, THORSTEN WILHELM2, HANS-JÜRGEN HOHNBERG1, SABINE KASTEN2, MATTHIAS ZABEL1, GERHARD BOHRMANN1 1 AND KAI-UWE HINRICHS 1

DFG-Research Center Ocean Margins & Dept. of Geosciences, University of Bremen, 28359 Bremen, Germany ([email protected], [email protected], [email protected], [email protected], [email protected]) 2 Alfred Wegener Institute for Polar and Marine Research, 27570 Bremerhaven, Germany ([email protected], [email protected]) Seepage of asphalt-laden heavy oils in 3000 meter water depth is fueling a unique cold seep habitat at the Campeche Knolls in the Southern Gulf of Mexico (MacDonald et al., 2004). We studied the Chapapote Knoll, one of the various salt domes in this area, which is covered with asphalt beds that extend over more than 1km2. During the expedition of RV Meteor M72/2 in 2006, we recovered a diverse set of asphalt samples and benthic organisms inhabiting this extreme environment to further explore mechanistic details of asphalt leakage and accumulation and its role for fueling benthic ecosystems. The asphalt samples are highly diverse in terms of their content of gaseous hydrocarbons. Maximum concentrations of gaseous hydrocarbons were associated with methane hydrate. Samples with very low concentrations of methane and higher gaseous hydrocarbons have a brittle and porous structure, which we interpret as signs of postdepositional alteration. The relative distribution of individual hydrocarbons in the C1-C5 range is highly diverse in the asphalts. We interpret this diversity to be reflective of the postdepositional history, which could be due to outgassing, leaching or biodegradation. Oily sediments recovered by gravity coring show evidence of a sulfate/methane transition zone at several meters subseafloor depth due to sulfate-dependent turnover of hydrocarbons. Stable carbon isotope analysis of methane in these cores indicate biological activity close to sulfate methane interfaces at 6m depth. Ongoing work is aimed at the investigation of the biodegradation state of the asphalt samples at the seafloor. Incubation experiments with 13C-labeled aromatic and aliphatic compounds are being carried out to assess the importance of biologically induced postdepositional alteration processes.

References MacDonald I.R, Borhmann G., Escobar E., Abegg F., Blanchon P., Blinova V., Brückmann W., Drews M., Eisenhauer A., Han X., Heeschen K., Meier F., Mortera C., Naehr T., Orcutt B., Bernahrd B., Brooks J., de Faragó M., (2004), Science 304, 999-1002.

A907

Iron isotope fractionation during the volcanic evolution of Hekla, Iceland JAN A. SCHUESSLER1, RONNY SCHOENBERG1AND OLGEIR SIGMARSSON2 1

Institut für Mineralogie, Leibniz Universität Hannover, Callinstr. 3, D-30167 Hannover, Germany ([email protected]) 2 Laboratoire Magmas et Volcans, CNRS - Université Blaise Pascal - OPGC, 5 rue Kessler, Clermont-Ferrand, 63038 France The aim of this study is to systematically investigate potential iron isotope fractionation in the Earth’s crust by magmatic processes. High precision iron isotope analysis by high resolution MC-ICP-MS were performed on a suite of rock samples representative for the volcanic evolution of the Hekla volcano (Iceland). The whole series of Hekla’s rocks results from several processes [1]. (i) Basaltic magmas rise and induce partial melting of meta-basalts in the lower part of the Icelandic crust. The resulting dacitic magma evolves to rhyolitic composition through crystal fractionation. (ii) The basaltic magma itself differentiates by crystal fractionation forming a basaltic andesite magma. (iii) Andesites are produced by mixing of basaltic with dacitic melts. These processes were traced by iron isotopes: (i) During the differentiation from the dacites to the rhyolites the G56/54Fe value increases successively. This increase (e.g., 0.15‰ for the 4000 B.P. eruption) can be described by a Rayleigh fractionation model using a constant bulk fractionation factor between the silicate liquid (L) and all mineral phases (M) of '56/54FeM-L = -0.1‰. (ii) No isotopic fractionation was found between the basalts and the basaltic andesites which have an average G56/54FeIRMM-014 value of 0.068 ± 0.057 (2SD), identical to mean basaltic values reported by other studies [2-5]. This observation is consistent with the limited change in iron concentration in the remaining silicate liquid during crystal fractionation and small mineralmelt Fe isotope fractionation factors expected at high temperatures. (iii) The iron isotope composition of the andesites is matching the basaltic andesites and the less evolved dacites, compatible with a mixing process. The observations suggest that the iron isotope composition of the crust can be slightly modified by magmatic processes and that magma differentiation processes are reflected in the iron isotope composition of some evolved silicic rocks.

References [1] Sigmarsson et al. (1992) Contrib. Min. Pet. 112(1), 20-34. [2] Beard et al. (2003) Chem. Geol. 195(1-4) 87-117. [3] Poitrasson et al. (2004) EPSL 223(3-4), 253-266. [4] Weyer et al. (2005) EPSL 240(2) 251-264. [5] Schoenberg and von Blanckenburg (2006) EPSL 252(3-4) 342-359.

Goldschmidt Conference Abstracts 2007

A908

Neutron capture-induced 150Sm anomalies in IAB Iron meteorites and Winonaites T SCHULZ.1,2,3, D. UPADHYAY 2, K. MEZGER 2, C. MÜNKER 3,2 AND H. PALME 1 1

An HRTEM and XRD investigation of 2:1 clay mineral diagenesis in the Jeanne d’Arc Basin, offshore eastern Canada D. SCHUMANN1, R. HESSE1, S.K. SEARS2, AND H. VALI1,2

Institut für Geologie und Mineralogie, Universität zu Köln, Zülpicher Str. 49b, 50674 Köln, Germany 2 Zentrallabor für Geochronologie, Institut für Mineralogie, Universität Münster, Corrensstr. 24, 48149 Münster, Germany 3 Mineralogisch-Petrologisches Institut, Universität Bonn, Poppelsdorfer Schloss, 53115 Bonn, Germany

1

Isotope anomalies produced by neutron capture in nuclides with large neutron capture cross sections can be used to constrain the irradiation history of extra-terrestrial matter [e.g. 1]. This study investigates the exposure history of IAB irons and Winonaites, which are presumed to come from the same body [2], using the increase in 150Sm as a result of neutron capture by 149Sm. For this purpose, Samarium was separated from 200-500 mg samples using standard dissolution and ion exchange techniques. The isotope ratios were measured on a Triton TIMS in static mode. Analytical uncertainties on the reported ratios are <12 ppm (2 sd; n=8) for the standard and <35 ppm for the samples (2 se). Figure 1 shows the neutron capture-induced isotopic shifts in 150Sm/152Sm measured on three IAB silicates (1.1-1.8 İunits) and two Winonaites (1.0-5.5 İ-units). The 15 corresponding neutron fluences for the IABs (1.5-2.5 x 10 2 15 2 n/cm ) and the Winonaites (1.4-7.4 x 10 n/cm ) are similar to those of chondrites [1]. The similar fluences calculated for the IABs are difficult to reconcile with their different cosmic ray exposure ages [e.g. 3], unless one invokes a common regolith history on the parent body followed by high shielding depths within their meteoroids (so that they were only insignificantly affected by exposure to cosmic rays after ejection from the parent body).

We have undertaken an HRTEM and XRD study of 2:1 clay mineral assemblages from argillaceous rocks of the Adolphus D-50 (AD-D50) and the Northern Ben Nevis P-93 (NBN-P93) wells, Jeanne d’Arc Basin, offshore eastern Canada. X-ray diffraction patterns of the <0.1μm size fractions after air drying and ethylene glycol solvation show that the 2:1 clay mineral assemblages span the major steps in the evolution of smectite to illite, consistent with the results of Abid et al. (2004). Clay mineral assemblages at shallow (1565 m) to intermediate well depths (3135 m) consist of smectite, randomly interstratified I-S (R0), weakly ordered I/S (WR1), and short-range ordered I-S (R1) phases. HRTEM images of the same samples after n-alkylammonium exchange show the presence of multiple phases of low- and high-charge expandable 2:1 clay minerals, an R1-ordered phase and illite. The low- and high-charge expandable 2:1 clay minerals decrease with burial depth, whereas the R1 ordered I-S and illite become the predominant 2:1 clay minerals. Microanalysis of individual 2:1 clay mineral particles as well as TEM images of freeze-etch replicas will be presented and will contribute to our overall understanding of the structural changes that occur in I/S during burial diagenesis, especially in the structure of weakly-ordered I/S. This clay mineral study is part of a broader research project attempting to link fluid flow events within the Jeanne d’Arc Basin to a number of anomalous patterns of illitization in strongly faulted areas of the basin. There is evidence that points to a fluid-flow driven increase in the percentage of illite in I/S (involving K-bearing fluids) rather than a simple temperature-dependence of the illitization of I/S clays.

0.27606

HaH 193

predicted neutron capture line

150Sm/152Sm

0.27602

Dept. Earth & Planetary Sciences, McGill Univ., 3450 University St., Montréal, QC H34 2A7 ([email protected]) ([email protected]) 2 Fac. EM Research,, McGill Univ., 3640 University St., Montréal, QC H3A 2B2 ([email protected]) ([email protected])

0.27598

Landes [~240]

0.27594

0.27590

0.51645

CC [~5] CD [~520] NWA 1457 Winonaites [~20-80] IAB silicates STD 0.51650 149

0.51655

0.51660

152

Sm / Sm Fig. 1: HaH = Hammadah al Hamra; CC = Caddo County; CD = Canyon Diablo; NWA = North West Africa; STD = terrestrial standard; Cosmic ray exposure age [Myr].

References [1] Hidaka et al., EPSL 180, 29-37 (2000) [2] Benedix et al., Meteoritics 35, 1127-1141 (2000) [3] Niemeyer, GCA 43, 1829-1840 (1979)

References Abid, I., Hesse, R. & Harper, J.D., (2004), Can. J. Earth Sci. 41, 401-429.

Goldschmidt Conference Abstracts 2007

High-precision Pb isotope measurements discriminate different subduction components along the Solomon island arc 1,2

2

S. SCHUTH , C. MÜNKER , E.E. SCHERER S. KÖNIG1,2

1

AND

A909

Ultrastructure of bone: Hollow apatite crystals, solution chemistry and organic inhibitors HENRY P. SCHWARCZ School of Geography and Earth Sciences, McMaster University, Hamilton, ON L8S 4K1, Canada

1

([email protected]) ([email protected])

2

Along the Solomon island arc, SW Pacific, the Indian plate is subducting beneath the Pacific plate. Geophysical evidence indicates the presence of a fossil slab of Pacific oceanic crust that was subducting until ca. 6 Ma [e.g., 1]. To assess the influence of subducted oceanic crust and Ontong Java Plateau material along the arc, we determined Pb isotope compositions of representative arc magmas covering the complete southern island arc chain (ca. 1000 km). Most of their Hf-Nd isotope compositions overlap with those of the Australian-Indian mantle domain, indicating that the active trench does not mark the boundary between the AustralianIndian and Pacific mantle domains [2]. In contrast to Hf-Nd, Pb isotope compositions of the lavas are dominated by subduction components. To achieve sufficient analytical resolution, Pb isotope compositions were obtained using MCICP-MS and normalization to Tl that was added prior to measurement. High-precision Pb isotope data for silicate reference materials obtained by Tl normalization using MC-ICP-MS are in good agreement with previously published data, supporting the reliability of the analytical method. An external reproducibility of better than ±130 ppm (2ı) can be achieved. The measured 206Pb/204Pb, 207Pb/204Pb, and 208Pb/204Pb in the Solomon arc magmas range from 18.351 to 18.853, 15.479 to 15.564, and 38.105 to 38.462, respectively, indicating the absence of significant amounts of subducted pelagic sediments. The high-precision Pb data clearly reveal a bimodal distribution, reflecting the presence of two distinct types of subduction component, originating from either the currently subducting Australian plate or the fossil Pacific plate. Notably, domains with Australian-Indian type Pb isotope signatures are confined to the central New Georgia group, where the active Woodlark spreading center is being subducted. Due to a lower geothermal gradient, all other segments of the subarc mantle have apparently not yet been fluxed by subduction components from the Australian-Indian plate that has been subducting since ca. 6 Ma. Lead isotope compositions of some samples show evidence that material from the Ontong Java Plateau was subducted together with the Pacific plate, consistent with geophysical evidence [e.g., 1]. References [1] Mann, P. & Taira, A. (2004) Tectonophysics 389: 137-190. [2] Schuth, S. et al. (2004) Contrib Mineral Petrol 148: 288304.

The utrastructural nature of bone is a matter of long debate. While all studies confirm that an apatitic mineral component makes up at least 60 wt % of bone, the morphology and even the composition of this phase is uncertain, due to the small size of the crystals (10’s of nm in maximum dimension). TEM analysis of ion-milled and focused ion-beam milled sections of bone reveals that most of the apatite in bone is in the form of apparently hollow fibers 5 to 10 nm in diameter and several hundred nm long, which are oriented parallel to collagen fibers. Lesser amounts of apatite occur in the gap zones spaced at 67 nm along collagen fibrils. The core of the apatite fibers is assumed to be filled with a protein, probably collagen, making up c. 5% of the total collagen in bone. The fibers are constructed of single crystals of apatite, with their caxes oriented parallel to the fibers. Boyde (1974) found similar structures in dentine. Although bathed in an extracellular fluid (ECF) which is supersaturated with respected in apatite, the crystals have remarkably uniform dimensions, implying that, after initial formation, further deposition of apatite is strongly inhibited. Osteopontin (OPN) and other molecules in the ECF (e.g., Pampena et al. 2004) are known to inhibit apatite growth. Both Ca and PO4 ions are essential to the vital activity of all metazoans. The ECF of invertebrates is also supersaturated in hydroxyapatite (HA), with saturation index values up to 1012. The relative scarcity of apatite as a biomineral suggests that strong inhibition of HA formation must occur in the ECF of invertebrates as well.

References Boyde, A. (1974) Transmission electron microscopy of ion beam thinned dentine. Cell Tissue Research, 152: 543-555 Pampena, D et al. (2004) Inhibition of hydroxyapatite formation by osteopontin phosphopeptides. Biochem. J. (2004) 378, 1083–1087

A910

Goldschmidt Conference Abstracts 2007

Integrated air quality assessment – A synthesis of elemental and organic air pollution indicators L. SCHWARK AND E. LEHNDORFF Department for Geology and Mineralogy, University of Cologne, Germany; ([email protected]) In a multidisciplinary approach atmospheric quality in the Greater Cologne Area (GCA) was studied using pine needles as bioreceptors. This contribution describes accumulation behaviour, transport fractionation, source allocation, and the spatial distribution of major and trace elements including platinum group elements. Special emphasis is placed on synthesizing the inorganic and organic pollutant load on pine needles for improved differentiation of traffic, industrial and power plant emissions. The interpretation is corroborated byG13C and G15N isotopes and environmental magnetic data. The GCA comprises various element emission sources, mainly lignite fueled power plants, urban regions of high traffic density and reduced air mixing, domestic heating and large industrial complexes along the Rhine Valley. We will first demonstrate that element accumulation on pine needles occurs in a systematic and predictable manner by analyses of 3 to 50 month old needles taken separately in summer and winter at six key locations. This allows for interpretation of spatial distribution maps constructed from 71 locations covering 3000 km² in the GCA. Absolute concentrations of elements and element enrichment factors were used for spatial analyses. Normalization to average dust composition was achieved using the sum of REE, as Ti was regionally enriched in the volcanic province of the “Siebengebirge”. Source characterization based on element concentrations revealed that Mo was best suited to identify petrochemical emissions. Cu enrichment results from petrochemial emission and traffic sourcs. PGE showed a predominantly traffic origin for Pt and Pd, the latter also affected by fertilizer application in agricultural areas. Ru and Re were less traffic dependent, the former being associated with industrial the latter with agricultural emission. Traffic pollution was clearly depicted by enhanced Sb and Ba concentrations, whereby Ba showed a higher degree of dispersal. Fe and V were shown also to be related to traffic emissions in urban areas but as well originate from power plant emissions. Verificaton of emission sources was achieved by polycyclic aromatic hydrocarbon (PAH) pollution indicators. Lignite-fueled power plants emit preferentially phenanthrene, cyclopentenophenanthrene, and dibenzothiophene; urban traffic is characterized by higher load of alkylated phenanthrenes and different isomer patterns of e.g. methylphenanthrenes. Correlation of element and PAH pollution indicators was excellent although certain pollutant sources were depicted by one group only. For example Pb/Zn mining activity in the Rhenish Massif was only detected by elevated Cd concentration. Inorganic and organic multiproxy air quality analyses allows for very reliable and independent source allocation even in areas with multiple emission sources.

Constraints on the revision of the K decay constants WINFRIED H. SCHWARZ AND MARIO TRIELOFF Universität Heidelberg, Im Neuenheimer Feld 236, D-69120 Heidelberg ([email protected]) As the accuracy of current K-Ar age calculations is seriously limited by the uncertainties of the 40K decay constants [1,2], a reevaluation is required. 40K has a dual decay to 40Ar and 40Ca with a branching ratio of 10.48/89.52 [1,3] and a total decay constant of 5.543·10-10 a-1 (calculated with data from [3] and a 40K/K ratio of 0.01167% [4]). Thus for conventional K-Ar and Ar-Ar dating (where the age calculation depend on mineral standards dated by conv. K-Ar technique) all three values are needed in the age equation. To revise the decay parameters for 40K, the systematic offset of Ar-Ar and U/Pb mineral ages of rapidly cooled rocks were determined [e.g. 5]. Using the H chondrite parent body cooling history [6,7] it is possible to calculate the age offset of U/Pb and Ar-Ar ages for c. 4.6 Ga old rocks – it is c. 30 Ma [8], significantly smaller than 1% which is noted in [5] or [9]. A biotite from the Great Dyke Intrusion in Zimbabwe/Africa with an age of c. 2.5 Ga, lead to an age discrepancy of c. 20 Ma. The age data for the c. 2.0 Ga old Vredefort impact structure from [10] recalculated for a new NL25 standard age [11] lead to a difference of c. 17 Ma. These new age data lead to a decay constant of c. 5.520·10-10 a-1, only a little smaller than that defined by [1] this result will not change including the available literature data [e.g. 5,9]. It is slightly different from the determination of the total decay constant of 5.554·10-10 a-1 via LSC by [12] and [13], but different 40K/K and branching ratios were used for these calculations. Using the geochronological data and the age equation for K-Ar dating it is not possible (though not important for calculating ages) to decide whether the branching or the 40K/K ratio is wrong by about 1%. Assuming a 40K/K ratio of 0.01167% would lead to the decay constant to 40 Ar of c. 0.575·10-10 a-1 and thus a branching ratio of c. 10.42/89.58. Nevertheless an independent redetermination of the 40K/K and the branching ratio is desirable.

References [1] Steiger R.H. and Jäger E. (1977) Earth Planet. Sci. Lett. 36, 359-362. [2] Begemann F. et al. (2001) Geochim. Cosmochim. Acta 65 (1), 111-121. [3] Beckinsale R.D. and Gale N.H. (1969) Earth Planet. Sci. Lett. 6, 289-294. [4] Garner E.L. et al. (1975) J. Res. Natl. Bur. Stand. A, 79A, No. 6, 713-725. [5] Kwon J. et al. (2002) Math. Geol. 34, 457-474. [6] Trieloff M. et al. (2003) Nature 422, 502-506. [7] Schwarz W.H. and Trieloff M. (2006) Met. Planet. Sci. 41, Nr 8, Suppl. A161 (5132). [8] Trieloff M. et al. (2001) Earth Planet. Sci. Lett. 190, 267-269. [9] Krumrei T.V. et al. (2006) Chem. Geol. 227, 258-273. [10] Trieloff M. et al. (1994) S. Afr. J. Geol. 97, 365–384. [11] Schwarz W.H. and Trieloff M. (2007) Chem. Geol., in press. [12] Malonda A.G. and Carles A.G. (2002) App. Rad. Isot. 56, 153-156. [13] Kossert K. and Günther E. (2004) App.Rad.Isot. 60, 459-464.

Goldschmidt Conference Abstracts 2007

The AsO6 polyhedron in arsenates, statistics and the novel compound Tl1+Tl3+As4O12

Detector strategies to measure Osmium isotope ratios in small samples by NTIMS

K. SCHWENDTNER1, U. KOLITSCH2 AND E. TILLMANNS1

JOHANNES SCHWIETERS1, DIETMAR TUTTAS1, CLAUDIA BOUMAN1, JOHN C. LASSITER2 AND TODD B. HOUSH2

1

Institut für Mineralogie und Kristallographie, Althanstraße 14, 1090 Wien, Austria ([email protected]) 2 Naturhistorisches Museum Wien, Burgring 7, 1010 Wien, Austria Inorganic arsenates comprise a large number of compounds, demonstrated by 1145 entries in the ICSD (version 2006/2), however, only 23 compounds (R < 8 %) contain the rare AsO6 octahedral unit. Hydrothermal synthesis (Teflon-lined stainless steel autoclave, 7 d, 493 K) using the starting materials Tl2CO3 and arsenic acid led to the growth of hexagonal platelets of the novel compound Tl1+Tl3+As4O12 (single-crystal X-ray diffraction, P-31m, Z = 1; a = 4.848(1), c = 11.091(2) Å, V = 225.75(8) Å3). The new compound is closely related to the following three structure types: M2+As2O6 (M2+ = Ca, Mn, Co, Ni, Cd, Hg, Pb), Hg2As2O6 and LiAsO3. All of these, as well as the novel compound, are built of layers of edge-sharing AsO6 octahedra, which are connected by the usually octahedrally coordinated metal cations. In the case of Tl1+Tl3+As4O12 Tl3+ is octahedrally coordinated whereas the coordination polyhedron of the larger Tl1+ can be described as a trigonal antiprism.

Tl3

Tl1 A

A

T l A

Considering the scarce data available for AsO6 polyhedra a statistical analysis of bond lengths in AsO6 groups was conducted. A total of 40 AsO6 polyhedra in 38 different compounds were analysed. The bond lengths in the AsO6 octahedra range between 1.736 and 1.918 Å, with a mean value of 1.827(29) Å. The individual bond lengths are very strongly influenced by the next nearest neighbours. As-OH bond lengths are considerably shortened to a mean value of 1.767(14) Å, which gives the perfect bond valence contribution of 1.00 for this As-O bond. Exactly the opposite effect is found in protonated AsO4 tetrahedra, where the AsOH bond lengths are elongated to a mean value of 1.719(27) Å. As-O bond lengths to other AsO4 and AsO6 polyhedra are very close to the mean value with 1.834(33) and 1.824(20) Å, respectively. The longest mean distances are to edges of other metal octahedra with 1.846(23) Å. No clear correlation between the bond-length distortion and mean bond lengths of individual polyhedra could be established. Financial support by a DOC-FFORTE Fellowship of the Austrian Academy of Sciences (ÖAW) is gratefully acknowledged.

A911

1

Thermo Fisher Scientific, Hannah-Kunath-Str. 11, 28199 Bremen, Germany; ([email protected]) 2 Department of Geological Sciences, The University of Texas at Austin, USA Negative Thermal Ionization Mass Spectrometry (NTIMS) is the method of choice for high precision Osmium isotope ratio measurements of small sample sizes. Because of its high selectivity and high ionization efficiency optimum sample utilization as well as low backgrounds are achieved. One of the limiting factors when analyzing small samples with Faraday detectors, is the noise level of the current amplifiers. The noise level is reduced as the resistor value and the gain of the current amplifier is increased. Today’s instruments use 1011Ohm amplifiers as a standard. Larger resistor values maybe used to reduce the noise level, however there is a practical limit due to the limited insulation resistance of the PCB as well as the insulation resistance of the Faraday cup and the feedthroughs itself. Single collector ion counting or even multicollector ion counting detectors overcome the noise problem and promises higher precision touching the limit of counting statistics. Today’s multicollector instruments are equipped with a single ion counting channel as a standard and very small samples can be measured in single collector by peak jumping procedures. The laminated magnet of the TRITON instrument from Thermo Fisher Scientific allows fast peak jumping sequences with short settling times between peak jumps. Nevertheless, this still is a sequential method, whereas the Multi-IonCounting (MIC) aproach allows a simultaneaous measurement of all isotopes of interest and thus maximises sample utilization and also eliminates the effect of signal fluctuations on the isotope ratio measurement. In this experimental study, the performance of all three different detector strategies are compared on the basis of measured data of Osmium samples with different ion beam intensities ranging from >1 Mcps to a few Kcps. The different detector strategies are: (i) Faraday detectors combined with 1012 Ohm amplifiers, (ii) fast peak jumping single collector ion counting using an SEM, and finally (iii) multi-ioncounting using multiple channeltron detectors. The best operation range and the pros and cons of each detector strategy for different signal intensities are discussed.

Goldschmidt Conference Abstracts 2007

A912

Experimental study of the Na-incordierite thermometer at different fluid compositions (NaOH-H2O; NaCl-H2O) M. SCOLA, P.W. MIRWALD AND P. TROPPER Institute of Mineralogy and Petrography, Univ. Innsbruck, Austria ([email protected]) Orthorhombic cordierite, (Mg,Fe)2Al4Si5O10*n(Na, H2O, CO2 etc.), a characteristic mineral phase of medium to high grade metapelites, has thermometric properties. A previous study in the model system Mg-cordierite – 1n Na(OH) (Mirwald, 1986) showed that the Na-content is inversely related to temperature, is largely pressure independent, and stabilises cordierite by about 75MPa relatively to the upper stability boundary according to the reaction (1) Crd + H2O = Tc + Q + Ky (Mirwald, 1984). In this study we investigated the Na-incorporation in Mgcordierite using different NaCl solutions (XH2O= H2O/H2O+NaCl; XH2O: 1, 0.8, 0.6, 0.4 and 0.2) which implies, in addition, different H2O activities covering the P-T range 0.2 – 1.1GPa and 650 to 750°C. Thus, the present experiments were conducted in a slightly acidic fluid milieu. The experiments were carried out in a piston cylinder and a hydrothermal apparatus. Microprobe and X-ray diffraction were used for sample characterisation. The prelimenary data confirm the inverse temperature relation without a singnificant pressure dependence. Previous Na-T relation: Na(apfu) = 0.406+/-0.003 –0.000471*T(°C); this study: Na(apfu) = 0.398+/-0.015 – 0.000466*T(°C). Again, the Na incorporation stabilises cordierite by about 75MPa, varying H2O activities have no significant influence on the position of reaction (1). Further experiments performed on quartzphyllite samples and application of the Na-in-cordierite thermometer to highgrade rocks are presented by Wyhlidal et al. (2007) at the same meeting.

References Mirwald, P.W. (1984): Fortschr. Mineral., 62, Beih. 1: 156157. Mirwald, P.W. (1986), Fortsch. Mineral. 64, 119 Wyhlidal,.S., Thöny,W.F., and Tropper, P. (2007), this volume.

Water-rock interactions at the new Gotthard rail base tunnel, Switzerland ULRIKE SEELIG1 INGRID STOBER2 AND KURT BUCHER1 1

University of Freiburg, Germany; ([email protected]) 2 Geological Survey of BW; ([email protected]) In Switzerland the new 57 km long rail base tunnel traversing the Central Alps is currently under construction. From the 11.35 km long Amsteg-Sedrun section in the northern part of the tunnel 110 water samples from fractures and open fissures have been collected and analyzed. Geologically this section lies completely in the steeply dipping crystalline basement of the Aar massif. Depending on the lithology the tunnel water samples show different compositions and the total of dissolved solids is varying from 171 to 4014 mg/l. The most prominent cation is sodium. Magnesium is in most samples close to or below the detection limit of 0.02 mg/l. The majority of the waters has also a very low calcium content. Only a few samples have elevated Ca-concentration, which can be related to the dissolution of fracture anhydrite. The major anions represent a mixture of bicarbonate, chloride and sulphate with varying concentrations. Chloride together with an equivalent amount of sodium is contributed to the waters from leaching of crushed fluid inclusions and grain boundaries. Sulphate originates mainly from the oxidation of sulphide minerals and to a minor extend from the dissolution of anhydrite locally present in fissures. One striking feature of the tunnel waters is their surprisingly high fluoride content of up to 29 mg/l. In lowTDS waters the fluoride concentration can even exceed the chloride concentration. Fluoride is contributed by fluorite dissolution, which can be found both as secondary fracture mineral and in the granite matrix. The low calcium content of the fluorite-buffered waters leads to the observed extremely high fluoride concentration. The other remarkable feature of the tunnel waters is their very high pH throughout all lithologies with values up to 10.4. It results from dissolution of calcite under closed system conditions, which leads to a strong decrease of the partial pressure of CO2. Calcite was found frequently in the granitoid host rocks. Calcite dissolution is also the source of bicarbonate in the waters. The dissolved calcium is then removed via a combination of plagioclase alteration and formation of zeolites. The hydrolysis of albite (anorthite content of the plagioclase is below 5 mole%) releases sodium as major cation into the water. Zeolites, especially laumontite, are present as thick mats on fracture surfaces in the tunnel. Zeolite formation is then responsible for the removal of calcium from the water.

Goldschmidt Conference Abstracts 2007

A913

Rare earth elements as natural tracers in the Thau basin karst system (Southern France)

Diverse active microbial communities in a tidal flat sediment as deciphered by a multidisciplinary approach

J.L. SEIDEL1, T. IBRAHIM 2 AND F. ZWAHLEN3

M. SEIDEL1, A. GITTEL1, K. BISCHOF2, J. KÖSTER1, H. SASS3 AND J. RULLKÖTTER1

1

UMR HydroSciences, CNRS-IRD-UM1-UM2, Université Montpellier 2, France; ([email protected]) 2 Catchment Science Center, University of Sheffield, Sheffield, United Kingdom; ([email protected]) 3 CHYN, Université de Neuchâtel, Neuchâtel, Suisse ; ([email protected]) Rare Earth Elements (REE) have been more frequently used as natural tracers in hydrogeology (Smedley, 1991; Johannesson et al., 1996), with the analytical progresses for their determination at very low concentrations(ng/L). These elements cannot be considered as perfect conservative tracers, but REE patterns of groundwaters are mostly inherited from aquifer rocks through which they flow. As a consequence, REEs can be useful to study flow pathways and water mixing processes at different scales. Because of the few REE studies realized in karst environment, the method was applied to the karst system of the Thau basin (southern France), which is characterized by a high complexity of water mixing processes (Aquilina et al., 2002). Three endmembers waterbodies( hydrothermal, karstic and sea or brackish waters) are present. The Vise submarine spring constitutes one of the main permanent outlet of the aquifers. Rain, surface and ground waters were sampled from 2005 to 2007 over the Thau basin karst system. Major elements were determined by ionic chromatography. Trace elements and REE, after preconcentration, were analyzed by ICP-MS. Characteristic REE profiles enable to differenciate several types of water and to follow the hydrodynamics of the submarine spring. The study allowed to individuate the different concerned aquifers and to illustrate their possible temporal connections in function of the hydrologic cycle. Further studies are under progress to investigate REE complexation with organic matter.

References Aquilina L., Ladouche B., Doerfliger N., Seidel J.L., Bakalowicz M., Dupuy C., Le Strat P. (2002), Chem.Geol. 192 :1-21 Johannesson K. H., Stetzenbach K. J., Hodge V. F. & Lyons W. B. (1996), Earth and Planet. Sci. Lett. 139 : 305-319. Smedley P. L. (1991), Geochim. Cosmochim. Acta 55 : 27672779

1

Institute of Chemistry and Biology of the Marine Environment (ICBM), Carl von Ossietzky University of Oldenburg, P.O. Box 2503, D-26111 Oldenburg, Germany ([email protected]) 2 Dept. of Molecular Ecology, Max Planck Institute for Marine Microbiology, Celsiusstraße 1, D-28359, Bremen 3 School of Earth, Ocean and Planetary Sciences, Cardiff University, Cardiff CF1O 3 YE, Wales, UK Up to six meter long sediment cores were recovered from a backbarrier tidal flat area of the island of Spiekeroog in the Northwest German Wadden Sea and analyzed by a combination of complementary geochemical, molecular biological and microbiological methods to investigate the activity and composition of microbial communities. One of our analytical approaches is the analysis of intact polar lipids (IPLs) using HPLC-ESI-MS. These diagnostic membrane lipids are rapidly degraded after cell lysis and thus are considered suitable biomarkers to trace viable microorganisms. The quantities of the detected IPLs decreased only slightly with depth and correlate well with the total cell counts obtained by DAPI staining. Furthermore, phospholipids with alkyl diether and mixed alkyl-acyl side-chains become dominant with increasing depth. In accordance with the molecular biological results this may indicate a substantial proportion of sulphate-reducing bacteria (SRB) in the microbial community even in deeper sediment layers because these lipids were detected in mesophilic SRBs (Rütters et al., 2001). Whereas in the uppermost layers phospholipid-type diethers of archaeal origin were absent, archaeol-containing phospholipids were detected throughout the deeper part of the sediment column. In layers with low contents of methane high numbers of ANME-2 and ANME-1 archaea were found using molecular biological methods. The major IPLs detected in a deep sulphate-methane transition zone were phospholipids with archaeol and hydroxyarchaeol cores which support the identification of ANME-2 consortia. The isotopic composition of these diethers is currently examined to show whether they are constituents of cell membranes from archaea mediating AOM.

References Rütters, H., Sass, H., Cypionka, H., Rullkötter, J. (2001), Archives of Microbiology, 176, 435-442.

Goldschmidt Conference Abstracts 2007

A914

Concentrations and signatures of stable isotopes of methane and hydrogen in hydrothermal fluids of the Mid-Atlantic Ridge 1

1

1

R. SEIFERT , S. WEBER , M. WARMUTH , A. KOSCHINSKY2 AND W. MICHAELIS1 1

Institut für Biogeochemie und Meereschemie, Universität Hamburg, Germany ([email protected]) 2 School of Engineering and Science, Jacobs University Bremen, Germany

Introduction We have measured concentrations of dissolved methane and hydrogen in a set of fluid samples recovered by ROV from hydrothermal systems located on the Mid-Atlantic Ridge (MAR) at 15°N and between 4° and 10°S under the auspices of the German SPP 1144 project. Fluids investigated originate from different settings comprising peridotitic to basaltic host rocks, hot and diffuse vents, and water depths between 1500m and 3000m.

Results and discussion

Evaluation of geothermometers for a zircon-rutile-corundum intergrowth W. SEIFERT AND D. RHEDE GeoForschungsZentrum Potsdam (GFZ), Potsdam, Germany ([email protected] and [email protected]) A corundum megacryst (7x5 mm) from residues of an alkali basalt from SE Saxony (Germany) includes zircon and rutile. It provides a suitable specimen to test recently published geothermometers “Ti-in-zircon” and “Zr-in-rutile” (Watson et al., 2006), and may be revealing about the origin of this mineral paragenesis, which is uncommon for basalts. Rim-core-rim traverses of the trace elements Ti and Zr in zircon and rutile, respectively, were performed using an electron microprobe JXA-8500F under different analytical conditions (see figures below, detection limits in brackets). Calculated temperatures are about 1100 °C for zircon (distance of analysis points to rutile >60 μm), but about 1015 °C for the core of rutile and 940 °C for the rim of rutile. The shape of the traverses are discussed in consideration of secondary fluorescence, temperature, and chemistry of the crystallizing medium.

Fluid endmember concentrations calculated based on the Mg content were up to 3.5 and 19 mmol/L for CH4 and H2, respectively, for fluids from the Logatchev hydrothermal field. Similar exceptional high concentrations with 1.4 mmol/L of CH4 and 11.6 mmol/L of H2 were also found in fluids of the recently discovered smoking crater Drachenschlund located at 08°18’S and characterised by ultramafic host rocks alike the Logatchev hydrothermal field. Much lower gas concentrations were found in the fluids of basalt hosted systems. Among these, a black smoker of the Turtle Pits field (4°49’S) emanating vigorously boiling fluids with temperatures of up to 407°C at 3000m depth showed the highest fluid endmember concentrations of 0.3 mmol/L H2 and 0.018 mmol/L CH4. Black smoker fluids with no indication of phase separation revealed significantly lower gas concentrations. Signatures of stable C isotopes for CH4 covered a range from į13C -9.2 to -13.7‰ for all fluids sampled from vents at about 3000m water depth with no systematic difference between the various settings. Only CH4 in fluids of the Lilliput hydrothermal field (9°33’S) emanating at 1500m water depth with temperatures < 20°C revealed į13C values of about 32‰. This might be attributed to distinct p/T conditions during abiotic CH4 generation or to a significant microbial contribution. į2H values of H2 were used to estimate temperatures of emanating fluids.

Conclusion Hydrogen dominates over methane among reactive gases within black smoker fluids. Host rock composition and phase separation processes strongly affect the gas concentrations in hydrothermal fluids but not the stable isotope distributions within the gases.

Reference Watson E. B., Wark D.A. and Thomas J.B., (2006). Contrib. Mineral. Petrol. 151. 413-433

Goldschmidt Conference Abstracts 2007

Lithium isotopes of the early solar system and terrestrial planets

Re-Os isotope systematics of kimberlites from SW Greenland: Implications for an isolated lithospheric mantle during 500m.y.

HANS-MICHAEL SEITZ 1, GERHARD P. BREY1, JUTTA ZIPFEL 2, ULRICH OTT3 AND STEFAN WEYER1 1

Institut für Geowissenschaften, J.W. Goethe Universität Frankfurt, Germany, ([email protected]) 2 Sektion Meteoritenforschung Forschungsinstitut und Naturmuseum Senckenberg, Frankfurt, Germany 3 Max-Planck-Institut für Chemie, Mainz, Germany Despite large mineralogical, textural and chemical differences between carbonaceous chondrites, ordinary chondrites and achondrites, the range of G7Li is restricted from +0.5 to +5.4‰. Hydrous alteration and metamorphism are extremely variable in chondrites but Li isotopes show no obvious correlation with petrological types. Neither carbonaceous or ordinary chondrites, nor achondrites exhibit a correlation between their Li abundances and G7Li, suggesting that Li isotopes did not fractionate during early stages of our solar system nor during magmatic differentiation on smaller asteroidal bodies. Based mainly on the composition of basaltic and ultramafic rock types from Earth, Seitz et al. [1] and Magna et al. [2] suggested a G7Li-value for BSE (bulk silicate Earth) of ~ +4‰. Elliott et al [3] derived a weighted average of +3.4‰ for unmetasomatised mantle peridotites. Seitz et al. [4] and [3] concluded that a G7Li of ~ +4‰ should also be the value for the inner planets, because rocks from the Moon and Mars give on average similar values. McDonough et al. [5], on the other hand, suggested a G7Li-value of 0‰ for the solar system, an average of a restricted number of measurements on chondrites. More recent work by McDonough et al. [6] revealed a slightly heavier average composition (G7Li = +1.3‰). However, such low values are not supported by our measurements which comprise a much wider range of chondrite types. We find an average of +3.3‰ for carbonaceous chondrites. Ordinary and enstatite chondrites give a lighter average of +2.3‰. The previously suggested ‘BSE’ value of ~ +4‰ by [1,3,4] was based mainly on basaltic rock types. The average ҏG7Li-value of unmetasomatised peridotites [3,7] and of olivines from relatively pristine spinel peridotites [1] is +3.3‰, indistinguishable from carbonaceous chondrites. We therefore suggest, in agreement with Elliott et al. [3], that the G7Li-value for BSE and also for the inner planets is +3.3‰. In turn, the different isotope signatures of carbonaceous chondrites on one side and of ordinary chondrites on the other are interpreted to reflect distinct reservoirs in the early solar nebula.

References [1] Seitz et al. (2004) Chem. Geol. 212: 163-177. [2] Magna et al. (2006) EPSL 243, 336-353. [3] Elliott et al. (2006) Nature, 443, 565-568. [4] Seitz et al. (2006) EPSL 243, 6-18. [5] McDonough et al. (2003) LPSC XXXIV #1931. [6] McDonough et al. (2006) LPSC, XXXVII, #2416. [7] Jeffcoate et al. (2007) GCA, 71, 202-218.

A915

R. SENDA1, K. SUZUKI1, H. KAWABATA1, AND I. KANEOKA 2 1

Institute for Reseach on Earth Evolution (IFREE), Japan Agency for Marine-Earth Science and Technology (JAMSTEC), Yokosuka, Japan ([email protected]) 2 ERI, Univ. Tokyo, Bunkyo-ku, Tokyo, Japan Kimberlite magma is well known to entrain many xenoliths and diamonds, which suggest that the magma originated from a source at the depth deeper than 200km. However, the nature of this source material is still highly debated. Here, we report the Re-Os data for kimberlite dykes from the Sarfartoq area, SW Greenland. In this area, the alkaline ultramafic dykes intrude into the boundary between the Archean craton and the Palaeo-proterozoic orogen. The kimberlites were erupted around 600Ma with a large carbonatite complex, which is located at the center of a widely distributed kimberlite swarm (Larsen et al., 1983; Larsen and Rex, 1992). The Re-Os whole rock reference age of 1100 Ma is significantly older than the eruption ages of the Sarfartoq kimberlites, determined by K-Ar and U-Pb methods (Larsen et al., 1983; Bizzaro et al., 2002). It implies that the kimberlite source mantle was separated from convecting mantle at the time of the Re-Os reference isochron age and remained isolated until eruption. The Re-Os TMA ages of the kimberlites, however, are similar to the eruption ages. Thus, in this case it appears that Re-Os TMA ages record the eruption. The initial 187 Os/188Os ratio of the kimberlites, 0.113, is in the range of the continental lithospheric mantle. The ȖOs values of the kimberlites at 1100 Ma are also similar to those of the lithospheric mantle peridotite xenoliths from Somerset Island (Irvine et al., 2003), rather than the primitive upper mantle estimated by Meisel et al. (1996), supporting the isolated source hypothesis. References Larsen L. M., Rex D. C. & Secher K., (1983), Lithos 16, 215221. Larsen L. M., & Rex D. C., (1992), Lithos. 28, 367-402. Bizzaro M., Simonetti A., Stevenson R. K., & David J., (2002), Geology 30, 771-774. Irvine G. J., Peason D. G., Kjarsgaard B. A., Calson R.W., Kopylova M.G., & Dreibus G., (2003), Lithos 71, 461488. Meisel T., Walker R.J., Morgan J.W., (1996) Nature, 383, 517-520.

Goldschmidt Conference Abstracts 2007

A916

Numerical simulation on the thermal evolution and differentiation of iron meteorites’ parent body HIROKI SENSHU1

2

AND TAKAFUMI MATSUI

1

Research Center for the Evolving Earth and Planets, Tokyo Institute of Technology, Tokyo, Japan ([email protected]) 2 Department of Complexity Science and Engineering, Graduate School Frontier Science, University of Tokyo, Tokyo, Japan ([email protected]) According to Hf-W chronometry, the parent bodies of iron meteorites had experienced silicate-metal separation within the first a few myr since CAI formation (e.g., Markowski et al., 2006). In this case short-lived radioisotopes such as 26Al and 60Fe may have been survived and played a key role in thermal history of parent body (Urey, 1955). However the thermal history and the mechanism of silicate-metal separation have been unclear so far. Thus, in this study, we developed a numerical model on the thermal evolution and differentiation of planetesimal-sized (from a few km to about 100km-sized) celestial body to clarify the timing and mechanism of the formation of metallic core, which is one of the most plausible candidate of the origin of iron meteorites. Our model takes into account the internal heating due to short-lived radioisotopes, radiative cooling from the surface, the porosity effect on the thermal conductivity, sintering of silicate media (pores shrink due to the effect of pressure and temperature), and silicate-metal separation due to permeable flow of molten metal through silicate media. Since the melting temperature highly depends on the sulfur content, we also consider the change in sulfur content along the phase diagram in Fe-FeS system (Kullerud, 1967). According to our numerical results, there are two conditions for the planetesimal-sized body to form a metallic core. At first, the parent body should form within the first 2 myr from CAI formation, otherwise it would have insufficient heat source. Next, the radius of the body should be larger than about 7 km, otherwise the body would be cooled rapidly. Once these conditions are satisfied, molten metal, into which sulfur preferentially dissolves, migrates toward the center of the body to form metallic core. Afterward, as the core cools down precipitation of pure iron takes place in accordance with the phase diagram in Fe-FeS system. Finally a pure iron ‘inner core’ forms at the center of the metallic core, which would be the origin of iron meteorites. Resulting cooling rate of metallic core at 800 K is slower than several 100 K/myr depending on the planetary size, which is consistent with the cooling rate of iron meteorites obtained from the analytical study, from 1 to 500 K/myr (e.g., Wood, 1964).

References Kullerud, G. (1967), “Research in Geochemistry”, 286-321. Markowski, A., et al. (2006), EPSL 250, 104-115. Urey, H. C. (1955), Proc. Natl. Acad. Sci. 41, 127-144. Wood, J. A. (1964), Icarus 3, 429-459.

Syn-plutonic dykes and magma mingling: An example from the Alvand plutonic complex, SanandajSirjan metamorphic belt, Iran ALI A. SEPAHI Department of geology , Bu Ali Sina University, Hamadan, Iran, ([email protected]) The Alvand plutonic complex is composed of various mafic to felsic plutonic rocks (gabbro-diorite-tonalite-granite) formed by repeated injections of magmas of crustal (anatectic) and mantle origin. Two extreme end-members of magmas in the complex include: 1) M-type magmas generating gabbrodiorite-tonalite suite and mafic-intermediate dykes and 2) crustal anatectic magmas generating S-type granites and related rocks. There are many NE-SW trending microdioriticmicrotonalitic syn-plutonic dykes in the complex which in the some places they are disaggregated to mafic microgranular enclaves (MME). Their appearance, mineralogical and geochemical affinities are similar to other dioritic rocks of the complex but they are finer grained, richer in quartz-ocelli and mafic clots and they are more silicic than their coarse grained equivalents. Regarding to field, petrographic, mineralogical and geochemical features a suite of pure anatectic to hybrid and to mantle type plutonic rocks can be distinguished in the Alvand plutonic complex. The dioritic-tonalitic rocks are hornblende-rich (30-50 %) but anatectic monzogranitesgranodiorites are hornblende-free. The evidence of hybridization due to magma mingling in the region is better understood by characteristics of mafic end member than felsic one. This is due to volumetric abundance of two original magmas so that felsic magma has been more abundant than mafic one at the time of hybridization, and therefore, the mafic magma has been surrounded by the felsic magma, so that, solidified parts of the felsic magma have been disintegrated into mafic magma. Since that the felsic magma has been near the end of its crystallization (solidification) history its mixing with mafic magma has been limited. The geochemical properties of both of the felsic and mafic rocks of the region have been slightly affected by hybridization process.

Goldschmidt Conference Abstracts 2007

Geochemistry and petrogenesis of the Tamuteh leucogranites in SW Saqqez, northwestern Iran A. A. SEPAHI1, S. F. ATHARI2 AND M. MOAZZEN3 1

Department of Geology, Bu Ali Sina University, Hamadan, Iran ([email protected]) 2 Department of Mining, University of Kurdistan, Sanandaj & Young Researchers Club of Kurdistan, Iran ([email protected]) 3 Department of Geology, University of Tabriz, Iran ([email protected]) The Tamuteh area is located in the southwest of Saqqez in the north of the Kurdistan Province, western Iran. The Tamuteh leucogranites are situated in the Sanandaj-Sirjan Metamorphic Belt, Zagros Orogen and are composed of various granitic rocks including syenogranite, monzogranite, granodiorite, tonalite and microtonalite. They are cut by lamprophyric dykes and silicic veins. Geochemically, these leucogranites are subalkaline (calc-alkaline), metaluminous and weakly peraluminous and their characteristics resmble the I-type granites. According to the geochemical classification scheme of Frost’s et al. (2001), Tamuteh leucogranites are magnesian to weakly ferroan and calc-alkalic to calcic. They contain low concentartions of Rb, Y, Zr, Th, U, Ce, FeOtot+MgO and high concentrations of SiO2, Ba and high Sr/Y and Ba/Rb ratios. They show general trends of decreasing contents of Al2O3, Fe2O3, MgO, P2O5 and TiO2 with increasing SiO2 but CaO, Na2O and K2O amounts show no regular patterns. According to tectonic discrimination diagrams of Pearce et al. (1984), Tamuteh leucogranites were generated in a volcanic arc setting. Behaviour of trace elements in the Tamuteh leucogranites is similar to those of well-known volcanic arc granites from Chile and Jamaica. Normalized trace element patterns show enrichment in LILEs (Rb, Ba, K, Th and Ce) relative to HFSEs (Nb, Zr and Y) and similar to calc-alkaline subduction related rocks from orogenic belts, including that they have had much interaction with crustal materials. The Tamuteh leucogranites also have high Sr and low Y and Rb contents, therefore can be related to mantle-derived magmas. The calc-alkaline, I-type leucogranite or syenogranitic to tonalitic composition and the presence of lamprophyric dykes indicate the formation of two magmas experiencing subsequent mixing/mingling precesses. Therefore, these leucogranites may have resulted from contamination of mantle-derived magmas by continental crust during an ancient subduction event. References Frost, B. R., Barnes, G. G., Collins, W. J., Arculus, R. J., Ellis, D. J. and Frost, C. D. (2001), Journal of Petrology, 42, 2033-2048. Pearce, J. A., Harris, N. B. and Tindle, A. G. (1984), Journal of Petrology, 25, 956-983.

A917

Large 14C age offsets between fine aragonite fraction and coexisting planktonic foraminifera in shallow Caribbean sediments S. SEPULCRE, K. TACHIKAWA, L. VIDAL AND E. BARD CEREGE, UPCAM3, CNRS, Collège de France, Europôle de l’Arbois, 13545 Aix-en-Provence, France ([email protected] ) 14

C is a widespread dating tool in paleoceanography. Age models are usually derived from coarse fraction planktonic foraminifera (>150 μm), but many proxies used in paleoclimatic research such as fine aragonite or organic components belong to the fine fraction (<63μm). Several studies have shown time-lags between records from various granulometric fractions (Paull et al., 1991; Thomson et al., 1995; Mollenhauer et al., 2005) due to different processes such as changes in sediment sources or abundances, sedimentation rates, bioturbation, reworking... We studied the temporal phasing between the coarse and the fine fractions from sediments retrieved in the Northern Caribbean Sea. Detailed stratigraphies over the past 40 kyr of G18O and 14C of aragonite fine fraction and planktonic foraminifer Globigerinoides ruber, and their respective abundances were performed on core MD032628 (Walton Basin, 17°21N, 77°42W, 846m water depth). G18O records are nearly in phase for both fractions, with a slight lead of the fine fraction during the last deglaciation. 14C ages are identical within errors over the past last 5 kyr. The age difference increases through time with a fine fraction being younger than G. ruber. The discrepancies range between 1.33 kyr at the end of the last deglaciation up to 5 kyr during the LGM. Bioturbation and variations in accumulation rates are likely causes for the observed discrepancies (Bard et al., 1987; Wheatcroft, 1992; Bard, 2001). Indeed, G. ruber abundance is highest during the LGM and decreases over the last deglaciation. Inversely, the fine fraction is minimum during the LGM and reaches its maximum during the Holocene. Bioturbation has probably mixed a “lighter G18O-younger 14C” fine fraction of the last deglaciation downward with “heavier G18O-older 14C” G. ruber. The relative influence of various mechanisms (size-dependant bioturbation, sedimentation rates, abundances) has been tested by using a numerical model (Bard et al., 1987). References Bard E., (2001), Paleoceanography 16, 235-239. Bard E. et al., (1987), Climate Dynamics 1, 101-112. Mollenhauer G. et al., (2005), Paleoceanography 20, PA1016. Paull C. et al., (1991), Quat. Res. 35, 274-290. Thomson J. et al., (1995), Radiocarbon 37, 91-101. Wheatcroft R., (1992), Limn.Oceano. 37, 90-104.

A918

Goldschmidt Conference Abstracts 2007

The geochemistry of Ni isotopes in mafic-ultramafic layered complexes D.S. SERGEEV1, A.S. SERGEEV2, E.M. PRASOLOV2,3, I.N. KAPITONOV3 AND S.A. SERGEEV2,3

Development of a mosaic-like micro-pattern during Mg-calcite crystal growth INGO SETHMANN

1

University of Neuchatel, Institute of Geology, Switzerland; ([email protected]) 2 Saint- Petersburg State University, 7/9 Universitetskaya em., 199034 St.-Petersburg, Russia 3 Centre of Isotopic Research, VSEGEI, 74 Sredny pr., St.-Petersburg, Russia; ([email protected]) Introduction For better understanding the natural processes involved in the formation of Ni ore deposits in igneous layered complexes, we analyzed Ni isotopes in a section through the 2.45 Ga Burakovsky layered mafic-ultramafic massif (NW Baltic Shield, Russia). In order to constrain the mechanisms of Ni isotope fractionation, we also analysed Ni isotopes in samples that have undergone Ni-enriching by the Mond process, where synthesis gas (95-97% CO and H) is used for highly effective Ni extraction involving metal-carbonyl formation. Analyses were made on whole-rock (WR) samples and on separated pyroxenes and Ni-rich sulphide minerals with a Neptune MC-HR-ICP-MS. The measuring precision achieved for GNi was at r0.1‰ with a maximum observed fractionation value for G62Ni/60Ni of 5‰. Results 1) WR samples from the section through the Burakovsky Massif (BM) (from -700 to -50m) show a clear fractionation trend of enrichment of light Ni isotopes from the base towards the roof of the massif (change of G62Ni/60Ni of +0.3‰ at the base, towards -1.5‰ at the top). 2) Ni-rich sulphide secondary minerals, which are forming the upper layer of the intrusion, have the lightest measured Ni isotope composition in the natural samples, in particular those which are associated with carbon minerals (G62Ni/60Ni=-2‰). 3) The Ni isotope ratios in pyroxenes changes independently of their position in the section. This may indicate the importance of a secondary, low temperature process. 4) Materials from the technological process of metallic nickel production show the most intensive 62Ni/60Ni and 61 Ni/60Ni fractionation (G62Ni/60Ni from +0.5 to –4.5‰). Lighter isotopes accumulate in the chamber’s top, where the gas phases condense. 5) Gas analyses from the BM WR samples show up to 41.5 vol% of (N2+CO) component. Interpretation The observed trends of Ni isotopic fractionation in the BM rock, similar to those observed in the technological Mond process, as well as the presence of trapped (N2+CO) gasses, suggest a model of natural kinetic Ni fractionation involving reducing gasses and natural metal-carbonyl formation. Such gasses may either have a juvenile origin or may be formed from heating (intrusion contact, metamorphism) and water vaporization of carbon-bearing geological units.

Institut für Mineralogie, Universität Münster, Corrensstr. 24, D-48149 Münster, Germany. ([email protected]) In natural calcium carbonate precipitation systems, Mg2+ ions play a major role due to their inhibiting effect on calcite growth and the formation of CaCO3-MgCO3 solid solutions, resulting in the modification of precipitation kinetics and rates, thermodynamic stability of the precipitate, as well as in changes of crystal morphology. Marine sedimentary and biomineralized Mg-calcite crystals have been reported to contain up to about 20 mole% MgCO3, although such highMg-calcite crystals are considered to be metastable under seawater-like conditions. These natural Mg-calcite crystals usually show micro-morphologies different from pure calcite crystals. To explore deeper into the conditions of high-Mg-calcite formation, Mg-calcite growth under different solution conditions has been directly observed using atomic force microscopy. Within a certain range of solution conditions, dependent on the Mg/Ca ratio in solution, on the saturation state, as well as on the presence of other ions in solution, an intriguing observation of crystal restructuring has been made: After fast crystal growth by step advancement as well as by two-dimensional nucleation and spreading of monolayerislands on calcite cleavage faces, straight ridges of subnanometre height appeared at the crystal surface. These ridges do not grow on the crystal surface, but they pop up by bulging of the newly precipitated crystal layers. Ridge formation starts at one point and usually proceeds in one direction until they connect to another ridge or stop at a major crystal defect. Ridges occur in three crystallographically defined directions, which probably correspond to the equivalent planes of the prismatic {11 2 0} form. Through progressive development of new ridges, the growing crystal becomes subdivided into mosaic blocks that can be reduced down to a few hundreds of nanometres in diameter. When advancing steps stop at ridges, they pile up to form multilayer steps that separate the mosaic blocks from each other, which may finally lead to crystal growth as aligned fibres. EDX analyses yielded a bulk content of about 8-10 mole% MgCO3 in the mosaic calcite, which should increase its solubility in water. However, dissolution of a mosaic takes place preferentially along the ridges, which may suggest a shift of the bulk strain and possibly exsolution-like dislocation of magnesium ions towards preferential crystallographic planes, where the accumulated strain gives rise to the ridges. The described mechanism of crystal restructuring may have implications for the formation kinetics of CaCO3-MgCO3 solid solutions, and it may play a role in the formation of crystal ultrastructures, such as in biominerals.

Goldschmidt Conference Abstracts 2007

A919

Metal sulfide complexing – What we know and what we should know

A geochemical scenario for evolution of the Nain-Baft back arc basin

T.M. SEWARD

H. SHAFAII MOGHADAM1, M. RAHGOSHAY 2 , H. WHITECHURCH 1 AND R. MONTIGNY1

Institut fur Mineralogie und Petrographie, ETH Zurich, CH-8092 Zurich, Switzerland Reduced sulfur occurs ubiquitously in aqueous fluids within and on the Earth’s crust over a wide range of conditions extending from ambient to high temperatures and pressures. In all these environments, metal ions may be complexed to varying degrees by simple inorganic ligands such as HS-, S2-, and SxS2- (polysulfide) as well as by thio-organic electron donors. Metal sulfide complexes and clusters play an important role in element sequestration and transport and biogeochemistry in aquatic (e.g. wetland) systems as well as in the ocean water column and in particular, in anoxic marine environments (e.g. the Black Sea dead zone). In seafloor and subaerial hydrothermal systems, metal-sulfide complexing (i.e. transport) and precipitation are in important not only in ore formation but also in defining many aspects of biodiversity in seafloor and surface discharge environments. At more extreme conditions within the Earth’s crust, metal sulfide complexes are considered to partition into volatile rich fractions during phase separation of saline, magmatic (hydrothermal) fluids. In volcanic gas environments, volatile metal sulfide complexes account significantly for the transition metal transport budget to the atmosphere. Despite the overall importance of metal sulfide complexing in understanding many geo(bio)chemical phenomena, the stability and stoichiometry of many aqueous metal sulfide complexes are not well known, particularly at ambient temperature where the experimental determination of reliable equilibrium constants is fraught with problems. A few metal-sulfide complex are quite well defined over a range to temperatures extending to supercritical conditions such as with the formation of silver and gold hydrosulfide/sulfide complexes. Both Ag and Au are soft Lewis acis and thus form very stable molecular moieties with reduced sulfur. Reliable equilibrium thermodynamic data also exist for sulfide complexes of Cu, Zn, Cd and Hg and some other elements (e.g. As, Mo) but there is a dearth of information available for many other metals with respect to the stability of the complexes as well as their related clusters and their structures. Our current state of knowledge will be discussed together with an assessment of the crucial gaps in our knowledge in metalsulfide interactions under a variety of conditions.

1

Ecole et Observatoire des Sciences de la Terre, Institute de Physique du Globe, Strasbourg, France, ([email protected], [email protected], [email protected]) 2 University of Shahid Beheshti, Faculty of Earth Sciences, Tehran, Iran, E-mail: [email protected] The Nain-Baft ophiolitic complex is categorized as an oceanic back arc basin, opened during the middle Cretaceous in the active continental margin of the Central Iranian continental block (the Sanandaj-Sirjan zone), due to the oblique subduction of the Tethys Ocean beneath the Iranian block. This extensional basin is behind the Mesozoic magmatic arc of the Sanandaj-Sirjan zone but crosscutting by the Eocene magmatic rocks of the Urumieh-Dokhtar magmatic belt. The Upper Cretaceous is believed to be the time of crystallization or metamorphism of the rock units in this oceanic basin based upon the K-Ar age determination. Clinopyroxene grains in the basaltic rocks of the Nain-Baft ophiolitic complex are characterized by low content of TiO2 content, similar to those of basalts, erupted in arc related environments. The basaltic-gabbroic rocks and the more acidic terms of this belt are of both tholeiitic (IAT) and calc-alkaline affinity. Depletion in HFSE and enrichment in LILE are the more distinctive features. Fractional crystallization and various degrees of depletion/enrichment in the mantle source associated with different degrees of partial melting are the main mechanisms, responsible for the geochemical evolution of the rock series in these ophiolitic complexes.

A920

Goldschmidt Conference Abstracts 2007

An experimental approach to hightemperature iron isotope fractionation

Coeval Pan-African granitization and migmatization of the north Sudan Basement

A. SHAHAR1, C. E. MANNING1 AND E. D. YOUNG1,2

C.K. SHANG1, M. SATIR1, H. TAUBALD1 AND G MORTEANI 2

1

Department of Earth and Space Sciences, University of California Los Angeles ([email protected], [email protected]) 2 Institute of Geophysics and Planetary Physics, University of California Los Angeles ([email protected]) There is considerable variation in 57Fe/54Fe among igneous minerals but the cause is uncertain. A central problem in interpreting natural iron isotope data is that equilibrium fractionation factors among coexisting minerals are poorly known. We performed experiments to establish the iron isotope equilibrium partitioining between fayalite and magnetite at a range of temperatures. These minerals are surrogates for mantle spinel and olvine. Experiments were performed in a piston cyldiner apparatus at 1 GPa and at 600, 700, and 800qC. Starting synthetic magnetite was spiked with 54 Fe to track the approach to equilibrium (e.g., Matsuhisa et al., 1978). The experiments were conducted with added quartz to ensure that oxygen fugacity was fixed at the quartz-fayalitemagnetite equilibrium. The results show that there is a resolvable fractionation between the fayalite and magnetite and that it decreases with increasing temperature: at 600qC 'mgt-fa= 0.45‰ at700qC '= 0.31‰, and at 800qC '= 0.28‰ +/- 0.017‰. The figure below shows our data compared with the predicted fractionation factors (Polyakov and Mineev, 2000). At higher temperatures the theory and experiments are in fair agreement, but as temperature decreases there is a more substantial difference. This study has experimentally determined for the first time inter-mineral 57Fe/54Fe equilibrium fractionation at high temperature. The results show that the magnitudes of these fractionations are large and require consideration when interpreting iron isotope ratios observed in natural igneous samples.

1

University of Tuebingen, Institute of Geosciences, Wilhelmstrasse 56, 72074 Tuebingen, Germany [email protected] ; [email protected] 2

Technical University of Munich Gmain Nr.1, 84424 Isen/Germany [email protected] We have carried out geochronological and isotopic investigations (zircon U-Pb and Pb evaporation and Nd isotope analyses of North Sudan Basement granitoids and migmatites to characterize sources and to find comprehensive answers to the apparently misleading field chronology, where granitoids (quartz-monzonite, granodiorite, tonalite and granites) intrude (quartz-monzonite, granodiorite, tonalite and granites) migmatite-gneisses; thus suggesting

post migmatization emplacement. Whole rock and biotite separates were also analysed for Rb and Sr isotope composition, to determine migmatizationreheating and/or cooling ages. Zircon U-Pb isotope dilution analyses for representative granitoid that intrude or form lenses and pods interbeded with migmatites, yield ages of 602 ± 3.1 Ma and 602 ± 28 Ma, while zircon 207Pb/206Pb evaporations data give similar mean ages of 602 ± 3.5 Ma and 599 ± 3.8 Ma. U-Pb analyses for migmatite zircon fractions from representative samples yield ages of 599 ± 12 Ma to 606 ± 7 Ma and zircon 207Pb/206Pb evaporation analyses give similar mean ages of 603.8 ± 2.1 Ma and 603.5 ± 3.7 Ma. We interpret these identical

zircon age data as crystallization ages for the North Sudan basement granitoids and migmatites Both rock types have crustal signatures (İNd = -3.5 to -4.8 for migmatites and İNd = -2.4 to -8.9 for granitoids). These basement rock types are different from the older (717.6 ± 0.61 Ma and 707.3 ± 0.97 Ma) mantle sourced (İNd = +5.6 to +6.9) granitoid Ring complexes. Rb/Sr whole rock isochron for migmatites give an age of 583 ± 21 Ma, being identical within error limits to basement rock zircon ages. Biotite Rb/Sr cooling ages for both migmatites and granitoids are younger and very similar, (566 ± 11 Ma to 570 ± 17 Ma) with a mean age of 567.5 ± 2.8 Ma and are interpreted as

the metamorphic-cooling ages. While isotope data indicate crustal sources for North Sudan basement migmatite and associated granitoid melt, their similar Pan-African zircon age data and identical biotite

cooling ages suggest concomitant crystallization, with inference for coeval indicate coeval crystallization, which References Matsuhisa Y., Goldsmith, J. R., and Clayton, R. N. (1978), Geochimica et Cosmochimica Acta 42 173-182. Polyakov V.B. and Mineev, S.D. (2000), Geochimica et Cosmochimica Acta 64 849-865.

together with identical biotite cooling ages, suggest concomitant granitization and migmatization of the North Sudan basement.

Goldschmidt Conference Abstracts 2007

Amino acids in shungite matter of Precambrian sedimentary rocks of Karelia S.N. SHANINA AND YE.A. GOLUBEV Institute of Geology, Syktyvkar, Russia ([email protected]) Amino acid composition of Karelian shungites (Shunga, Zazhogino and Nigozero deposits) of the Onega area from Palaeoproterozoic with various carbon contents (from 3 up to 98 %) was investigated. It was established, that the general contents of amino-acids in calculation on percentage of organic matter increase with concentration decrease of organic carbon. The lowest concentrations of amino-acids were established in rocks with the migratory organic matter (0.0005 mg\gɝɋorg). In shungite rocks containing sedimentary organic matter, amino acid contents were increased by 1–2 order (0.002–0.02 mg\gɋorg). In the area of Shunga deposit the group composition of amino acids for sedimentary and mixed organic matter look as follows: aliphatic § hydroxil. In migration organic substance the relation varies: aliphatic > hydroxil. In sedimentary shungite of Zazhogino deposit the group composition of amino-acids is submitted substantially by aliphatic differences (60 %), and in migratory-clastic shungite substance of Nigozero deposit the concentration of amino acids containing acid groups is increased: aliphatic > acid > hydroxil. Among individual amino acids increased contents of serine, glicine and alanine are characteristic. In a sample with the least content of organic carbon (lidite, Shunga deposit) was established high tyrosine quantity and presence of lisine (basic amino acid) which is absent in high-carbon shungite. Additional researches of sedimentary shungite remains (Zazhogino deposit) after bitumoid extraction have shown that more than 60% of amino-acids were removed from a sample as a result of extraction by chloroform and methanol-benzene. Content of amino acids up to extraction is 0.075 mg\g, after extraction – 0.029 mg\g. This fact allows us to believe that the significant part of amino acids in shungite structure includes in bitumoid organic matter. Our results well coordinate with previous data about distribution of bitumoids in shungite rocks (Solov’eva et al., 2000). The more content of carbon in shungite, the less amount of bitumoids it contains, and accordingly amino acids. Amino acids in shungite matter of sedimentary origin are the best kept; with increase of carbon contents the quantity of amino acids sharply decrease, but does not disappear totally. That is connected with originality shungite structure formed nanoporous fullerene-like carbon particles. References Solov’eva A.B., Rozhkova N.N., Glagolev N.N. and Zaichenko N.L. (2000) Geochemistry International 38 685–688.

A921

Fluid composition in veins of the Outokumpu drilling site, Finland L. SHARIF, J. KOLB AND F.M. MEYER Institute for Mineralogy and Economic Geology, RWTH Aachen University, Wuellnerstr. 2, 52062 Aachen, Germany. ([email protected]) The Outokumpu drilling site is located in the vicinity of the Outokumpu ore deposit in SW Finland. The Proterozoic Outokumpu formation, consisting of mica schist with black schist interlayers, serpentinite, skarn rocks, and pegmatitic gneisses, is thrust over an Archaean basement. The borehole (2.5 km deep) transects the Outokumpu formation and was intended to encounter the Archaean basement. Core samples with quartz and carbonate veins were taken in order to study fluid inclusions. The wallrocks were metamorphosed in the amphibolite facies. Three different vein types can be discerned: (i) quartz-filled with a biotite-rich alteration halo, (ii) quartz-filled without alteration zone, and (iii) carbonate-filled with an Mg-Hbl alteration zone. The quartz veins show variable sizes in length and thickness ranging from a few mm to several cm, which range from lenses and thin to relatively large veins. The type (iii) veins have a smaller thickness of a few mm. The investigated fluid inclusions in the quartz veins are primary and pseudosecondary showing one to three phases containing vapor and/or liquid. These inclusions occur together on intragranular trails, in clusters, or as single inclusions in quartz. Locally, pure gas inclusions form intragranular trails. Transgranular trails which indicate secondary inclusions are rare. They also consist of mixed aqueous and carbonic inclusions. The lowest measured homogenization temperature (Th) is at about 250°C but most of the measured Th are in the range of 300-400°C. The eutectic temperatures (Te) have values of about -22°C and final melting temperatures (Tm) of about – 12°C which indicate a salinity of 15–19 wt.% NaCl eq. in the aqueous solution. The Te for gas inclusions have values of about -56°C and a Th of 26°C which points to a CO2-rich composition. The minimum formation temperatures around 300-400°C and the alteration mineralogy (Hbl, Bt) suggest a formation of the hydrothermal quartz veins at conditions of the lower amphibolite facies. This saline and carbonic fluid composition is typical for metamorphic rocks. The investigation of the fluid inclusions provides a documentary of the fluid development and the possible hydrothermal mobilisation/remobilisation of the Outokumpu ores during later metamorphism.

A922

Goldschmidt Conference Abstracts 2007

How well do we know the initial Nd isotopic state of the Earth?

Remediation of heavy metals with species and green vegetables

M. SHARMA AND R. ANDREASEN

S. SHARMA1,K. S. PATEL1, V. K. JENA1 AND S. C. PATEL2

Department of Earth Sciences, Dartmouth College, Hanover, NH 03755 USA. ([email protected], [email protected]) As is evident from the extensive studies of CAIs and presolar grains in meteorites, the material in the Solar System is a mixture of debris from widely different nucleosynthetic sources. The differences in isotopic composition of bulk meteorite samples may therefore reflect variations in the mixing ratio of various components in different “planetary” materials. It would then follow that models of planetary evolution that are based on small shifts in isotopic ratios due to radioactive decay (Boyet and Carlson, 2005) are strongly dependent on the initial isotopic state and a “bulk solar” value that cannot be properly assumed. A key issue to understand the evolution of early Earth has been whether the available samples of meteorites accurately reflect the Nd isotopic composition of bulk Earth (Andreasen and Sharma, 2006; Ranen and Jacobsen, 2006). We have addressed this question using high precision Sr, Ba, Nd, and Sm isotopes in large samples of chondrites and a eucrite (Andreasen and Sharma, 2006; 2007). High-precision barium isotopic compositions of large samples of St. Severin (LL6) and Juvinas (eucrite) are identical to the terrestrial values. In contrast, Murchison and Allende (carbonaceous chondrites) reveal excesses in 135Ba and 137Ba of around +39 and +22 parts per million, respectively; no anomalies are resolvable in 130,132,138Ba. High precision Sr isotopic compositions of all meteorites are identical within error. The data are consistent with carbonaceous chondrites having an excess in r-process 135,137Ba with respect to Earth, Eucrite Parent Body, and ordinary chondrites. The carbonaceous chondrites, however, display no variation in the r- and s-process Sm and Nd isotopes suggesting that the rprocess sources of Ba and the lanthanides were decoupled. The homogeneity of Ba and Sm isotopes in the Earth, Eucrite Parent Body and ordinary chondrites indicates that the Solar Nebula that fed planetesimals between ~1 to ~2.4 AU was well mixed with respect to these elements. It was heterogeneous beyond ~2.7 AU where carbonaceous chondrite parent bodies formed. These observations also indicate that the best estimate of the Nd isotopic composition of the Earth is obtained from ordinary chondrites and not from carbonaceous chondrites, as is normally assumed. Since the terrestrial upper mantle shows a 142Nd anomaly of +18±8 parts per million with respect to the ordinary chondrites, this is further evidence that the upper mantle retains a memory of early Earth differentiation and sequestration of a reservoir with an average Sm/Nd ratio lower than that of chondrites.

References Boyet M. and Carlson R.W. (2005) Science, 309, 576-581. Andreasen R. and Sharma M. (2006) Science, 314, 806-809. Ranen M.C. and Jacobsen S.B. (2006) Science, 314, 809-812. Andreasen R. and Sharma M. (2007) Astrophys. J., In Press.

1

School of Studies in Chemistry, Pt. Ravishankar Shukla University, Raipur-492010, CG, India, ([email protected]) 2 Department of Earth Science, Indian Institute of Technology, Powai, Mumbai-400 076, India The common spices and green vegetables (i.e. Amaranth, Anise, Coriander, Methi and spinach) grown in the contaminated soil of the pyrite belt area, Mandala, central India has been studied for remediation of heavy metals (HMs) i.e. As, Cr, Mn, Fe, Ni, Cu, Zn, Hg, Pb. These plant have remarkable accumulation capacity for these elements. The total mean soil contents of metals i.e. As, Cr, Mn, Fe, Ni, Cu, Zn, Hg and Pb were found to be 53.0, 1380, 1832, 8900, 667, 743, 153, 4.8 and 387 mg kg-1, respectively. The total mean contents of HMs i.e. As, Cr, Mn, Fe, Ni, Cu, Zn, Hg and Pb in these plants were 0.48-2.81, 107–5869, 76–1700, 716-53381, 56–8837, 19.9-165, 59–196, 0.9–11.2, 4.6–54 mg kg-1 of DW respectively. The species: Anise, Coriander and Spinach were found to be hyperaccumulator for Cr (> 0.1); Cr (>0.5), Fe (> 5%), Ni (> 0.8%) and Cr (>0.4%), Ni (>0.2%). A sequential extraction procedure was used to fractionate heavy metals in highly contaminated soils into the conceptual metal pools: exchangeable; carbonate bound; bound organically; bound in Fe-Mn oxides; and residual. The leachable and total contents of metals in soil, total metal contents in six plant species and their correlations are discussed. Among five plant leaf tested, Anise, Coriander and Spinach showed the extremely high accumulation tendency towards the heavy metals (i.e. Cr, Fe, Ni), may be due to micellization with essential oils, and sequestrilization with dibasic acids i.e. oxalic, succinic, etc. The species: Anise, Coriander and Spinach accumulated Cr; Cr, Fe and Ni; Cr and Ni at extremely high levels and considered as hyperaccumulator. These spices provide new plant resource for exploring the mechanism of metal hyper accumulation to use in the phytoremediation of the heavy metal contaminated soils.

Goldschmidt Conference Abstracts 2007

Subduction of hydrated lithosphere: 300 ppm H2O in subducting olivine would eliminate the metastable olivine wedge 1

1

T. G. SHARP , T. DIEDRICH , F. C. MARTON W. DUFRANE1

2

AND

1

School of Earth and Space Exploration, Arizona State University, Tempe, AZ 85287-1404; ([email protected]) 2 Department of Science and Technology, Bergen Community College, Paramus, NJ, 07652-1595. In subducting slabs, where temperatures may be low enough to inhibit equilibrium transformation of olivine, the persistence of metastable olivine into the earth’s transition zone would reduce subduction rates and possibly result in deep focus earthquakes through transformational faulting. Several studies have used experimental transformation kinetic data to model the depth range of metastable olivine persistence under subduction zone conditions (Kirby et al. 1996; Mosenfelder et al. 2001, Marton et al. 1999, 2005). H2O in olivine has been shown to enhance growth rates in the olivinewadsleyite and olivine-ringwoodite phase transformations (Kubo et al. 1998, 2004) Recent modeling of the depth range of metastable olivine persistence has included a power law dependence of water concentration (Hosoya, 2005). Here we present new experimental results on the transformation of hydrous San Carlos olivine (300 wt-ppm H2O) to ringwoodite at 18 GPa. Our experiments show that 300 ppm H2O greatly enhances ringwoodite growth rates, resulting in an activation enthalpy of 186 kJ/mole. Even at temperatures as low as 700 °C, we obtain measurable growth rates, implying that rapid ringwoodite growth at temperatures corresponding to the interiors of cold subducting slabs. Combining our kinetic data with thermal modeling of subduction zones shows that even for very old and fast subduction zones, 300 ppm H2O would eliminate the metastable olivine wedge. Alternatively, if deepfocus earthquakes are triggered by olivine-transformational faulting in cold subducting slabs, then the olivine in such slabs cannot be significantly hydrated.

References Hosoya, T. et al., (2005) Geophysical Research Letters 32. Kirby S.H. et al., (1996) Reviews Of Geophysics 34, 261. Kubo T. et al., (2004) American Mineralogist 89, 285 Kubo T. et al., (1998) Science 281, 85 Marton, F.C. et al., (1999) Geophysical Research Letters 26, 119. Marton, F.C. et al., (2005) Phys. Earth Planet. Int. 149, 53. Mosenfelder, J.L., et al., (2001) Physics Earth Planetary Interiors 127, 165.

A923

Chlorine isotope distribution on Earth Z.D. SHARP AND J.D. BARNES Department of Earth & Planetary Sciences, University of New Mexico, Albuquerque, New Mexico 87131 ([email protected]) With our ability to analyze 37Cl/35Cl ratios (G37Cl values) of small amounts of solid materials, it is now possible to use chlorine isotope geochemistry of low Cl-concentration solids as a tracer of fluid movement in the subduction process. Cl is a hydrophilic ion that should faithfully track aqueous fluids throughout the crust and mantle. The G37Cl values of primative carbonaceous chondrites, pristine MORB glasses and samples of sub-continental mantle origin are all close to 0‰, equal to average crustal values (Sharp et al., Nature ’07). Notable variations do exist, however. Primative sodalite inclusions from Allende (CV chondrite) average -1.3(±0.6)‰, and MORB glass from 12°N, EPR have G37Cl values of -1.0‰. The sodalite may represent a separate nebular Cl reservoir or have low G37Cl due to an equilibrium isotopic fractionation. The 12°N, EPR data may represent mantle heterogeneity, as is seen in other geochemical systems. The overall homogeneous G37Cl value of 0‰ for Earth means that even small deviations from this value require a distinct source or fractionation process to explain the non-zero values. Chlorine isotope fractionation between phases is generally very small, and cannot change the G37Cl of a system by more than ~0.5‰. Two mechanisms which cause appreciable fractionation have been identified. The first is vaporization of NaCl, where liquid-vapor fractionation at 800°C is ~0.7‰. While not large, fractionation in a Rayleigh distillation process in the solar nebula could easily exceed 2‰, even at high T, if 90% of a sample is vaporized. Far more remarkable is the fractionation between hydrochloric acid and vapor, where fractionations in excess of 8‰ are observed in simple boiling experiments. This fractionation process explains extremely variable G37Cl values (covering a range of 16‰) found in high temperature, acidic volcanic fumaroles in Central America (Barnes et al., this meeting). Excepting these unusal fractionation processes, chlorine isotope ratios should maintain fidelity throughout subduction and return in arcs. Non-fractionated samples from Central America and the Izu Bonin arcs have values clustering around -2‰, presumably recording the value of their source. References Barnes, J.D., Sharp, Z.D. and Fischer, T.P. (2007) Chlorine stable isotopes as a geochemical tracer along the Central American and the Izu-Bonin-Mariana arcs. This volume. Sharp, Z.D. et al. (2007) Chlorine isotope homogeneity of the mantle, crust and carbonaceous chondrites. Nature (in press)

Goldschmidt Conference Abstracts 2007

A924

How well do trace element proxies predict slab fluid behavior?

An experimental study of the origin of reaction textures in mantle xenoliths

A.M. SHAW1, E.H. HAURI2, R.J. STERN3 AND J. HAWKINS4 1

Woods Hole Oceanographic Institution, Woods Hole MA 02543 ([email protected]) 2 DTM, Carnegie Institution of Washington, Washington DC 20015 ([email protected] 3 University of Texas at Dallas, Richardson TX, 75083 USA 4 Scripps Inst. of Oceanography, La Jolla CA, 92093, USA We present volatile, major and trace element data for olivine-hosted melt inclusions from a suite of cross chain seamount samples extending across the Mariana arc from Guguan volcano to the Mariana Trough. H2O, CO2, S, F and Cl abundances, as well as trace elements, have been determined by SIMS [1]. Our results show that there is a strong decoupling of water and slab fluid tracers such as Ba/La and B contents, beyond the main arc system. As might be anticipated, trace element enrichments generally associated with the subducting slab show a systematic decrease across the arc into the back-arc. However, water contents in cross-chain samples, 230 km above the subducting slab, show similar values to the arc-front samples, implying that water release is a continuous process across the arc. This finding is consistent with experimental results [2] and melt inclusion studies across the Central American arc [3]. In contrast to H2O contents, we observe significantly higher CO2 contents in the cross chain samples (up to 820 ppm) as compared to the arc samples (max CO2= 550ppm), suggesting either enhanced decarbonation at depth, or that the cross chain samples have experienced less degassing. Our observations show that fluid release can be substantial behind the main volcanic front and that a reevaluation of the common use of slab-fluid tracers is clearly needed.

References [1] Hauri, E. et al. Chem. Geol. 183, 99-114 (2002). [2] Schmidt, M.W. and Poli, S. Earth Planet. Sci. Lett. 163, 361–379 (1998). [3] Walker et al. (2003) Contrib. Min. Petrol. 146, 62-77.

C.S.J. SHAW 1 AND D.B. DINGWELL 2 1

Department of Geology, University of New Brunswick, Canada. ([email protected]) 2 Earth and Environmental Sciences, University of Munich, Germany Sieve-textured clinopyroxene and spinel are common in mantle lherzolite and harzburgite xenoliths. Various interpretations have been put forward to expalin the origin of this texture, including partial melting, mantle metasomatism and host magma – xenolith reaction during transport. Shaw et al (2006) suggested that sieve texture was a result of a two stage process involving breakdown of orthopyroxene with generation of a secondary Si-Al-alkali-rich melt, followed by migration of this melt along grain boundaries where it dissolved clinopyroxene and spinel in an incongruent reaction. In this study, we have performed a set of time series experiments to test the above hypothesis. The experiments were performed under both reducing and oxidized conditions at 1200 and 1156 oC at one atmosphere, using a synthetic leucitite melt and discs of natural lherzolite. Experiment durations ranged from 1 to 15 days. Sieve texture was developed on clinopyroxene and spinel in all experiments, but only on grains in direct contact with melt derived from symplectic reaction zones around orthopyroxene. Reaction zone thickness does not show a simple relation to experiment duration, however reaction zones developed at 1156oC are narrower than those formed at higher temperature. Our results show that sieve texture development on clinopyroxene and spinel in orthopyroxene-bearing mantle xenoliths is the result of a multistage reaction process. In the first step, orthopyroxene undergoes incongruent dissolution to produce a silica and alkali-rich melt together with olivine and clinopyroxene. As this melt migrates along grain boundaries it causes incongruent dissolution of clinopyroxene and spinel. The incongruent dissolution reactions involve complete dissolution of the reacting mineral followed by nucleation and growth of the secondary phases once the reacting melt is saturated with them. These secondary phases have a smaller volume than the primary grain,s leaving behind a reaction zone with interstial melt, i.e. a sieve textured zone. The reaction of orthopyroxene, clinopyroxene and spinel with infiltrated host magma results in a range of melt compositions that are indistinguishable from those interpreted to be due to very small degrees of partial melting.

References Shaw, C.S.J, Heidelbach, F., Dingwell, D.B. (2006), Contrib. Mineral. Petrol. 151, 681-697.

Goldschmidt Conference Abstracts 2007

Geochemical and microbiological controls on the corrosion and transport of depleted uranium in soil S. SHAW1, A.C. BAXTER2, S.A. JACKMAN2 AND I.P. THOMPSON3 1

School of Earth and Environment, University of Leeds, Leeds, UK. 2 Oxford University, Earth Science, Oxford, UK. 3 NERC Centre of Ecology and Hydrology, Oxford, UK. Assessment of the environmental impact of depleted uranium (DU) from munitions in oxic soil is hampered by limitations on information regarding uranium mobilisation and transport. Soil geochemistry, mineralogy and microbiology have been shown to have a significant effect on the behaviour of uranium within many natural environments. The aim of this study is to determine the relative influence of geochemical and microbiological factors within the context of the overall biogeochemical processes, including the speciation and mineralogy of the uranium present during the DU metal breakdown. In this study 2 soil columns were collected from field sites (Kirkcudbright and Eskmeals) in the UK where DU metal samples had been buried for |8 years. The Kirkcudbright soil is organic and clay rich, whereas the Eskmeals soils is a quartz-rich dune sand. Samples of soil up to 50cm from the DU metal were analysed by Sequential Extractions (SE), Scanning Electron Microscopy (SEM), mini-focus X-ray Adsorption Spectroscopy/ X-ray Fluorescence (XAS/XRF), X-ray Diffraction and pore water analysis (ICP-MS). The microbial population in the soil was also characterised using culturing and Biolog techniques. The SE and XAS data indicate that wihtin the soils the uranium is only present as U(VI). In the Eskmeals site the contaminated soil is dominated by schoepite (UO2)4O(OH)6.6H2O) coating the quartz grains (fig. 1). In the Kirkcudbright soils mini-focus XAS and Figure1. Schoepite coating on soil XRF analysis indicate particle. the U is more dispersed and associated with ferric oxyhydroxide minerals. The contaminated zone around the corroding metal is relatively small in the Eskemeals soil, with total [U] <10ppm within 10cm of the DU metal. The bacterial community counts from the soil are not affected by the U concentration, but there is a significant decrease in the metabolic diversity within the contaminated zone. The pore water in the contaminated zone also contains high concentrations of oxalic acid indicating a significant biological response.

A925

The distribution pattern study of rare earth elements in Choghart iron ore mine, Bafq area, Iran M.R. SHAYESTEHFAR1 A. ZARRABI1

2

AND A. YAZDI

1

Mining Engineering Department, Shahid Bahonar University of Kerman, 76175-133 Kerman, Iran. ([email protected], [email protected]) 2 Iranian Central Iron Ore Company, Mining Manager of Sechahun mine, Iran. ([email protected] ) From the metallogenic points of view, Iran is divide into several provinces and belts out of which, metallogenic province of Saghand-Golpayegan-Bafq has a special importance. The presence of rare earth elements along with the elements such as Fe, P, U and Ti is known only in this province. Metallogenic province of Saghand-GolpayeganBafq is made of the different mines, out of which, the Choghart iron ore mine is one of them. This mine is one of the most important iron ore mines of Iran which could be important due to presence of the REE. The Choghart iron ore mine is situated in central Iran plateou and its geographical coordinate is as: 55º28/02// longitude and 31º42/00// latitude. In order to study the behaviour of REE in Choghart mine, samples from magnetite and apatite have been analysed by NAA. Aquired results show that, as a whole, magnetites from this mine are not much enriched from the REE. All magnetites show less LREE enrichment and only the aquired sample from the borehole with degree of high purity of iron shows more enrichment in LREE. (La/Lu)cn ratio shows that the fractionation between LREE and HREE occur nearly in all magnetites but magnetite with degree of high purity of iron shows more extent of fractionation between LREE and HREE in Choghart mine. Also these conclusions show that Choghart mine apatites are enriched from REE in such away that the percent of REE in these apatites is about 1.54%. The great slope of spidergrams and amount of (La/Lu)cn ratio show intense fractionation between LREE and HREE, depletion of HREE and enrichment of LREE in these apatites. Therefore apatites from this mine have very high values in order to obtain LREE or Ce group. We come to know that in mineral processing stage of iron ore of Choghart mine, if the apatite concentration with degree of high purity produces, then the extraction of these elements will be considerable. Keywords: Rare earth elements, magnetite, apatite, Choghart mine, metallogenic province.

References Shayestehfar M.R, Zarrabi A, Sharafi A, Yazdi A., 2006. Petrology, petrography and mineralographical studies of Choghart iron ore mine, Bafq area, Iran, Geochemi. Cosmochim. Acta 70, P. A578.

Goldschmidt Conference Abstracts 2007

A926

Cretaceous-Cenozoic exhumation of Dabashan from apatite fission track thermochronology and its implication for growth of the northeastern Tibetan plateau margin 1

C.B. SHEN

2

AND L.F. MEI

Pedogenic origin dolomite developed within calcium concretion of Tertiary red clay at Loess Plateau, China XF SHENG, J CHEN AND JF JI Departmen of Earth Sciences, Nanjing University, Nanjing 210093, China ([email protected])

1

Faculty of Earth Resources, China University of Geosciences, Wuhan, China ([email protected]) 2 Key Laboratory of Theory and Technology of Petroleum Exploration and Development in Hubei Province, China ([email protected]) As a part of Qinling collisional orogenic belt, Dabashan located at the northeast margin of Sichuan craton basin and closely adjacent to the eastern Tibetan plateau margin, is an ideal region for better understanding the tectonic evolution of Qinling orogeny roundly and stepwise migration of the northeastern Tibetan plateau margin. Apatite fission track dating and time-temperature thermal history modeling were carried to analyze on 10 samples in the Dabashan. The new data yield ages ranging from 73±8 to 33±4 and mean track lengths between 11.4±1.6 to 12.7±1.6. Thermal history models based on the AFT data taken together with published K-Ar data (Zheng et al., 2006) indicate that the cooling and exhumation process of Dabashan can be divided into three stages since Cretaceous. The three stages are identified as (1) a phase of rapid uplifting cooling at 120-110 Ma, (2) following by a period of relative thermal stability during which rocks remained at temperatures with in AFT partial annealing zone (ᨺ60-110ºC), (3) ending with a new stage of accelerated uplifting during ᨺ10Ma to resent. The first rapid uplifting cooling phase has been suggested to reflect the large-scale transtensional deformation of Qinling orogeny belt, which is attributed to the Cretaceous eastward tectonic escape and Pacific backarc extension (Hu et al., 2006). Following the transition phase the subsequent slow cooling phase pattern implies a net reduction in horizontal compressional stress corresponding to increased extension rates along the continental margin due to the decrease in plate convergene. The last accerelated uplifting event is response to eastward growth of Tibetan plateau uplift and lateral growth, which continues today. Numerous studies of Tibetan Plateau suggest that the onset of deformation in eastern and northern margin of Tibetan Plateau occurred in the latest Miocene (Tapponnier et al., 2001; Kirby et al., 2002; Zheng et al., 2006; Enkelmann et al., 2006).

References Enkelmann E. et al. (2006). GSA. 118, 651-671. Hu S.B. et al. (2006). Tectonophysics. 420, 409-429. Kirby E. et al. (2002). Tectonics. 21, 1-20. Tapponnier P. et al. (2001). Science. 294, 1671-1677. Zheng D.W. et al. (2006). ESPL. 248, 198-208.

Previous research has shown that the land-origin dolomite was almost deposited at the high salinity environments. Only a few researches reported the un-salina origin dolomite was existed in the paleosol of Hawaii for a particular geologic reason and not any data regarding the authigenic dolomite in Chinese Loess has been recorded. The carbonate concretions are widely existed in Tertiary red clay layers developed at the Xifeng area of Loess Plateau, central China. With application of the XRD and the highresolution SEM and TEM, lots of dolomite in rhombohedron in company with palygorskite has been detected, distributing along the dissolved interspaces within the calcium concretion. The morphological characters of those minerals indicate they are authigenic origin and come into being during the pedogenic period of the red clay. The former appearance needs enough Mg2+in environment. In this case, Mg2+is thought from the condensation of the outcome of the thorough weathering of the carbonate and silicate after the action of the heavy precipitation during the pedogenic period. The latter is widely considered only formed under the extremely dry climate. Electron Microprobe analysis farther testifies that the dolomite is with low Mg concentration and the lower ratio of Mg/Ca. Moreover, the cations such as Mg2+, Sr2+, Na+ etc. in dolomites at the different carbonate concretions are in a wide range, indicating the dolomites don’t come into being in the unitary soil solution, but at the dividable micro-environments with different salinity. It can be concluded that, when the red clay come into being, the climatic condition was characterized as the alternation of heavy precipitation-high temperature and dryhigh temperature. The result has also provided an example to explain the cause of formation of dolomite in terrigenous fresh-water environment. This work was funded by the NSF of China through Grant 40503010.

Goldschmidt Conference Abstracts 2007

Surface complexation of Uranium(VI) on Fe/Mn (hydr)oxides DAVID M. SHERMAN1, CAROLINE L. PEACOCK1,2 AND CHRIS G. HUBBARD1

A927

The Os isotope heterogeneities in the mantle sequence of the Bangong Lake ophiolite, northwestern Tibet R.D. SHI, X.C. ZHI, B.H. DING AND L. CHEN

1

Department of Earth Sciences, University of Bristol, Bristol BS8 1RJ, UK ([email protected]) 2 Present Address: School of Ocean and Earth Science, University of Southampton, Southampton SO14 3ZH, UK The fate and mobility of U in oxic soils, groundwater and the marine environment is controlled by sorption of UO2++ onto nanocrystalline iron and manganese (hydr)oxides. Thermodynamic models for such reactions are needed before we can understand the aqueous geochemistry of U. However, we cannot develop useful thermodynamic models until we have a molecular-level understanding of U sorption and complexation. Previous EXAFS studies of UO2++ sorption on FeOOH have been interpreted as indicating an inner-sphere >Fe(OH)2UO2 surface complex formed by edge sharing with FeO6 polyhedra (2E complex). However, this complex is at odds with the known surface site densities of FeOOH phases. On goethite, for example, 2E complexes can only occur on the {210} and {010} planes (space group setting Pnma) which comprise only a small fraction of the surface area. Here, we show that previous EXAFS spectra have been misinterpreted owing to the neglect or incomplete inclusion of multiple scattering. The dominant UO2++ surface complex on FeOOH is (>FeOH)2UO2(H2O,OH)3 and (>FeOH)2UO2CO3 (2C complex) resulting from bidentate corner-sharing with two adjacent FeO6 surface polyhedra. Ab initio calculations of UO and U-Fe distances in the (2C) complexes are consistent with the EXAFS results. On goethite, 2C complexes can form on the {101} planes which comprise nearly all of the reactive surface area; the 2C complex explains the high sorption capacity (>2 wt%) for U(VI) on goethite and, apparently, ferrihydrite. We developed a surface complexation model to fit a series of sorption experiments of U on goethite and ferrihydrite. Using a 1pK formalism and Basic Stern model for electrostatics, we found we could fit our data using (>FeOH)2UO2(OH) and (>FeOH)2UO2CO3 (2C) complexes, consistent with EXAFS results. An analogous model for U sorption onto hexagonal birnessite was developed using the (>MnO)3UO2OH and (>MnO)3UO2CO3 complexes assuming that UO2++ complexes over vacancies in the phyllomanganate layer. Implications of our models for the fate of U(VI) in soils and aquatic environments will be discussed.

CAS Key Lab. of C-M Materials and Environments, School of Earth and Space Sciences, USTC, Hefei 230026, China Two type harzburgites occur in mantle sequence of the Bangong Lake ophiolite, NW Tibet. Type I is Cpx-bearing harzburgite with Cr# = 20~25 of spinel reflecting 10~15% degrees of partial melting; Type II is Cpx-free harzburgite with Cr# = 69~73 of spinel representing 35~40% degrees of partial melting formed in the suprasubduction zone (SSZ) [1]. In Type I, the 187Os/188Os ratios (0.12446 to 0.12853) are chondritic and the 187Re/188Os ratios (0.44 to 1.77) are suprachondritic reflecting that Re gain recently by basaltic melt percolating during the formation of the Bangong Lake crust 167 Ma ago [2], while in Type II, part of 187Os/188Os ratios (0.12166 to 0.12896) are subchondritic, this can be explained by Re depletion during ancient partial melting. The old Os isotope model age (>950 Ma) of one Cpx-free harzburgite in a young ophiolitic mantle show that ancient Os isotopic heterogeneities can survive in the Earth upper mantle. Osmium concentrations tend to decrease from Type I (Cpx-bearing harzburgites, 5.25-5.45ppb) to Type II (Cpx-free harzburgites, 4.4-4.45ppb, one sample is 1.32ppb) indicating that Os does not behave compatibly during melt percolation as it is observed during partial melting, but becomes dissolved and mobilized by the percolating melt [3]. But the heavy serpentinization likely caused the Os loss by the sulphide decomposition in a SSZ environment with high fO2. This study shows that Type I and II harzburgites of the Bangong Lake ophiolitic mantle have complex and different evolution. However, the Os isotopic features are consistent with a model where the two type harzburgites belong to the same melting regime producing the Bangong Lake oceanic crust. This scenario is also observed early in Troodos ophiolite [3] .

Acknowledgment This work was supported by the NSFC (Grant Nos. 40672051, 40572036).

References [1] Shi R.D. et al. (2005). Acta Petrol Mineral, 24(5): 397-408 (in Chinese with English abstract). [2] Shi R.D. (2007). Chin Sci Bull, 52(2): 1272-1278. [3] Buchl A. et al. (2004). Geochim Cosmochim Acta, 68(16): 3397-3408.

Goldschmidt Conference Abstracts 2007

A928

Genesis of tholeiitic and calcalcaline series of Zao volcano, NE Japan arc, Japan

Anthropogenic contamination of bivalves revealed by Cd isotopes ALYSSA E. SHIEL1, KRISTIN J. ORIANS1, DANIEL COSSA 1 AND DOMINIQUE WEIS

T. SHIBATA1, Y. TATSUMI2 AND A. SAKAYORI3 1

Beppu Geothermal Research Laboratory, IGS, GRL, Kyoto University, Japan ([email protected]) 2 Institute for Frontier Research on Earth Evolution, JAMSTEC, Japan ([email protected]) 3 The Graduate School of Education, Kanazawa University, Japan ([email protected]) Genesis of co-existing tholeiitic series (TH) and calcalkaline series (CA) in island arc and their relationship has been a focus of attention because it is critical to understanding magma genesis in island arc. The Zao is Quaternary volcano situated on the Tohoku Backbone Ranges, Northeastern Japan arc, and both of TH and CA co-exist in continuous volcanic activity (Sakayori, 1991). We investigated the Sr, Nd, Pb isotope and trace element compositions to discuses genesis of TH and CA from Zao volcano. The trace element compositions of the studied samples show the typical characteristics of island arc magma in the diagram of MORB normalized pattern, such as enrichment of LILEs and negative Nb spike. Positive Pb and Sr spikes are also apparent. The LILEs enrichment and negative spikes of Nb are relatively larger in CA than TH. Although abundance ratios of Cs/Nb of TH are restricted to 0.09 - 0.20, those of CA can be divided into two ranges, such as 0.20 – 0.25 and 0.42 – 0.57. The isotopic compositions of TH are more enriched compared to CA. In all the diagrams presenting the relationship of the isotope compositions, TH and CA make different liner trends, which indicate the mixing relation. Furthermore, Pb isotopic compositions suggest that two depleted endmember is necessary to explain the trend of TH and CA. The depleted endmember of TH is relatively enriched than that of CA. On the other hand, tow enriched endmembers are required from the relationships between parent/daughter and the isotope ratios, because TH and CA show different linier trends in those relation and the liner trends are diverse in the direction of isotopically enriched side. The both of the enriched direction of TH and CA is differ from the mixing trend of mantle wedge and recycled materials form subducting slab observed from Northeastern Japan (Shibata and Nakamura, 1997). From the observations in the above, it can be concluded that four components are necessary to explain the chemical characteristics of TH and CA from Zao volcano and the different two enriched components are derived from different crustal materials, although the sources of those are not obvious yet.

References Sakayori A., (1991), Bull. Volcanol. Soc. Japan, 36, 79-92. Shibata T. and Nakamura E., (1997), J. Geophys. Res., 102, 8051-8064.

1

PCIGR, Earth and Ocean Sciences, Univ. of British Columbia, 6339 Stores Rd, Vancouver, BC V6T 1Z4, Canada; ([email protected]) 2 IFREMER, Centre de Nantes, BP 21105, F 44311 Nantes Cedex 03, France We are studying Cd isotopic compositions in bivalves to assess their natural and anthropogenic variability and sources. Oysters (Crassostrea gigas) from BC were collected from Desolation Sound (BC coast) and Barkley Sound (west coast of Vancouver Island). For comparison, we analyzed bivalves collected off the coasts of France, oysters (C. gigas) from the Gironde estuary and northern Brittany and mussels (Mytilus galloprovincialis) from the Gulf of Lion. Cd isotopes were measured by dynamic multi-collection using a Nu Plasma MC-ICPMS and sample-standard bracketing technique together with external normalization to correct for instrumental mass bias. Results are reported in del notation (G), normalized to a mass difference of one atomic mass unit. Our reference Cd standard (High Purity Standards, Inc., lot 291012) has the same Cd isotopic signature as JMC Cd [1,2]. Reproducibility is estimated from repeat analysis of a secondary Cd standard (High Purity Standards, Inc., lot 502624), 0.37 ± 0.03 permil/amu (2SD; n=31). BC oysters have Cd isotopic signatures consistent with those reported for seawater from the N Pacific [3,4] suggesting that the high Cd concentrations (4.8-15.8 ppm tissue dw) found in bivalves along this coast result from natural coastal upwelling. Oysters from the BC coast have a slightly lighter signature than those from the west coast of Vancouver Island, potentially resulting from a larger anthropogenic contribution consistent with this geographical location. Mediterranean mussels have Cd isotopic signatures within the range of NW Mediterranean seawater values [3]. Oysters from northern Brittany and the Gironde estuary have signatures significantly lighter than the literature value for the Atlantic [4]. The Gironde oysters have the lightest signatures consistent with pollution resulting from industrial evaporationcondensation processes, as documented by the shift to G= –0.16 permil/amu for two dust samples from a Pb-Zn refinery plant in northern France [2]. These results demonstrate the ability of Cd isotopes to trace anthropogenic pollution.

References [1] Wombacher F. et al., (2003), GCA 67, 4639-4654. [2] Cloquet C. et al., (2006), Environ. Sci. Technol. 40, 25252530. [3] Lacan F. et al., (2006), GCA 70, 5104-5118. [4] Ripperger S. and Rehkämper M., (2007), GCA 71, 631642.

Goldschmidt Conference Abstracts 2007

A929

Geochemical variation and residence time of groundwater in Mt. Fuji area, central Japan

CO2-water-basalt interactions: Experimental and mineralogical study

NAOTATSU SHIKAZONO

S.N. SHILOBREEVA1 AND I. MARTINEZ2

3-14-1,Hiyoshi, Kohoku-ku, Yokohama 223-8522 Japan ([email protected]) Many groundwater samples were collected from Mt.Fuji area, central Japan which is totally composed of basaltic materials. The samples were analyzed for Na+, K+, Ca2+, Mg2+, Si, Al3+, Fe2+, Fe3+, TC(total dissolved carbon), Cl-, NO3-, SO42- and TP(total phosphorus). Analytical data plotted against altitude indicate that alkali and alkali earth element (Na+, K+, Ca2+, Mg2+, Ca2+) and Si concentrations increase with decreasing altitude, indicating that the dissolution of silicates in basaltic materials control the trends. Ca and Mg concentrations positively correlate with each other and the correlation coefficient is 1.64. This is consistent with CaO/MgO molal ratio of basalt which is 1.47. Therefore, it is inferred that Ca and Mg in groundwater were derived mainly from the congruent dissolution of basalt. Ca/Si concentration ratio determined by the dissolution reactions of basalt accompanied by the precipitation of allophane is 0.29 which is lower than 0.48 estimated from the analytical data on groundwater. This lower value could be due to the precipitation of silica mineral (SiO2). The agreement between theoretical and analytical results indicate that Ca, Mg and Si concentrations of groundwater are governed by dissolution and precipitation reactions. In order to interpret groundwater chemistry and estimate residence time of groundwater the simplified coupled dissolution kinetics-fluid flow model was used. Assuming reasonable values of parameters (reactive surface area, mass of groundwater, temperature etc) and using rate constant experimentally determined, residence time of groundwater in southeastern part of Mt.Fuji area(Kakitagawa site) was estimated to be several years to 30years). This estimated residence time is consistent with isotope data (herium isotope and tritium concentration).

1

Vernadsky Institute of Geochemistry and Analytical Chemistry, Russian Academy of Sciences, Moscow, Russia ([email protected]) 2 IPGP, Laboratoire de Géochimie des Isotopes Stables, 4 place Jussieu, Paris, France ([email protected])

Introduction Development of a better understanding of fluid-rock interaction is important in many areas of applied science and basic research. Investigation of reaction of basalt with H2O and CO2 at different P-T-fO2 conditions is important both for planetary processes including processes in subduction zones of the Earth and mineral storage of CO2 including the problem of reducing the anthropogenic CO2 emission into the atmosphere. We have carried out experimental, mineralogical and carbon stable isotope study of this reaction.

Experimental Method Carbonation experiments were run by exposing natural basalt powders (100-200μm grain size) to CO2 and H2O under various conditions in a hydraulic press (total pressure 1kbar and 400-500°C). Oxalic acid (H2C2O4) was used as a source of CO2 and H2O. Additional amounts of H2O were added in the platinum capsules. The thus produced CO2 and H2O were then reacted with the basalt powder at high pressure and temperature. Gas phase and solid products of experiments were analyzed by manometry, step heating and massspectrometry techniques. Mineralogical study of the products of experiments by Scanning Electron Microscopy allowed the identification and observation of carbonates and water-bearing minerals.

Discussion and Conclusions Our experimental results document the effect of the H2OCO2 medium on basalts under these P,T conditions and that calcite forms during the carbonation of plagioclase. We observed that carbonation reaction in basic rocks is different from that in ultrabasic rocks. Pressure, temperature and molar fractions of H2O and CO2 are important parameters whose influences were quantified. Measured į13 C of gas and solid phase allowed to discuss kinetic and thermodynamic factors of carbon isotope fractionation in such reactions and are discussed in terms of mass balance.

A930

Goldschmidt Conference Abstracts 2007

A study on the formation mechanism of Temagami Iron-Formations, Canada

High-grade Ag-Cu-Sn-In mineralization in the NishizawaAshio area, Tochigi Prefecture, central Japan

Y. SHIMADA1, A. YASUMATSU1, Y. MOTOMURA1, R. OKAZAKI1, T. NAKAMURA1, H. OHMOTO2 , Y. OKAUE1 1 AND T. YOKOYAMA

MASAAKI SHIMIZU1, SATOSHI MATSUBARA2, MARINA SHIMIZU1, YOSHINORI KYOUNO3, AKIRA HARADA4 AND NIGEL J. COOK5

1

Fac. Sci., Kyushu University; Ropponmatsu, Chuo-ku, Fukuoka, 810-8560, Japan; ([email protected]) 2 Penn State University In order to elucidate systematically the formation mechanism of BIF (Banded Iron-Formations), a BIF sample collected at Temagami, Canada was characterized by optical microscope, EPMA, Xray microscope, SIMS, and chemical analysis. This BIF can be divided into three layers visually; black layer, white layer, red layer. Main minerals in the three layers were magnetite, dolomite and quartz with fine hematite particles respectively. To estimate the formation temperature of each three mineral and the formation environment (especially oxidationreduction conditions), the oxygen isotope ratios (18O/16O) and contents of rare earth elements were measured by SIMS and chemical analysis. The isotope equilibrium temperature of dolomite and magnetite located in the white layer were 250-350ºC. In the rare earth elements pattertns for all the layers, a positive Eu anomaly appeared suggesting contribution of the anoxic submarine hydrothermal fluid for the formation of the BIF. On the other hand, a negative Ce anomaly, which indicates a character of oxic sea water, was observed for a part of samples of the red layer. These dolomite and magnetite may be formed under reduction conditions at high temperature, that is, from mixed water of the large amount of anoxic hydrothermal water of 250-350ºC and the small amount of oxic sea water. While magnetite and dolomite located in the black layer shows lower formation temperature of 100 ºC than that of magnetite and dolomite in the white layer. The latter magnetite and dolomite may be formed from hydrothermal water with high mixing percentage of seawater. As mentioned above, each mineral in the layers might be precipitate from the fluids formed by mixing of large amount of anoxic submarine hydrothermal fluid with high temperature and small amount of oxic seawater with low temperature. The mixing degree may be different among minerals in the BIF.

1

Department of Earth Sciences, University of Toyama, Japan ([email protected]) 2 Department of Geology and Paleontology, National Museum of Nature and Science 3 Omochanomachi Branch, Ashikaga Bank, Japan 4 D. O. C. Consultant, Japan 5 Natural History Museum, University of Oslo, Norway

Mineral Identification and Observation Polymetallic (Au-Ag-Cu-Pb-Zn-Fe-As-Sb-Bi-Sn-In-W) vein-type mineralization in the Nishizawa-Ashio area occurs in strongly altered late Neogene felsic volcanic rocks. Recently Ishihara (2006) and Ishihara et al. (2006) reported that indium had been exploited in ore concentrates (e.g., 1200 tons In at Ashio). At Nishizawa, roquesite and “sakuraiite” (CuZn2InS4) were identified as discrete In minerals, associated with Pb,Se-bearing matildite, electrum (Au:Agล 5:5–7:3), chalcopyrite, sphalerite (1-2% FeS), cassiterite and quartz. Canfieldite-argyrodite, Se-bearing acanthite, proustitepyrargyrite, Ag-bearing tetrahedrite-freibergite (Cu:Ag=5.4:4.6-4.2:5.7), smithite and ferbeite (Fe:Mnล 9:1) are also observed. Roquesite frequently occurs as radiating prismatic crystals (up to 0.5 mm), and sometimes as intergrowths with “sakuraiite”.

Conclusions In-bearing sphalerite as solid solution between sphalerite and roquesite has been previously reported (e.g., Burke and Kieft, 1980). The roquesite-“sakuraiite” intergrowths at Nishizawa reveal no solid solution between roquesite and sakuraiite, and instead imply that In-bearing sphalerite represents solid solution between sphalerite and incompletely defined“sakuraiite” (with petrukite as an intermediate phase?).

References Burke, E. A. J. and Kieft, C. (1980), Canad. Mineral., 18, 361-363. Ishihara, S. (2006), Shigen-Chishitsu, 56(1), 95-96. Ishihara, S., Hoshino, K., Murakami, H. and Endo, Y. (2006), Resource Geology, 56(3), 347-364.

Goldschmidt Conference Abstracts 2007

A931

Geochemical study on Bousei, Hotta and Smetanin Seamounts near the Japan Trench in Northwestern Pacific Ocean

The influence of Philippine Sea Plate on the composition of mantle beneath Kyusyu, SW Japan arc: Along-arc variation of B data

G. SHIMODA1, O. ISHIZUKA1, K. YAMASHITA2, 3, M. YOSHITAKE3, M. OGASAWARA1 AND M. YUASA1

M. SHIMONO1 M. MIYOSHI1 T. FUKUOKA2 T. SANO3 AND T. HASENAKA1

1

Geological Survey of Japan, AIST, Tsukuba, Japan ([email protected], [email protected], [email protected], [email protected]) 2 Institute for Study of the Earth's Interior, Okayama University, Misasa, Japan ([email protected]) 3 Department of Earth and Planetary Sciences Kobe University, Kobe, Japan ([email protected]) The Northwestern Pacific plate is characterized by existence of a large number of seamounts (Koppers et al., 1998). It is widely accepted that these seamounts were produced by extensive magmatic activity during the Cretaceous in the South Pacific that is considered to be one of the largest volcanic events in Earth’s history. As this magmatism would represent large-scale mantle upwelling, the magmatism could provide a key constraint to understand chemical evolution of the Earth. Due to this importance, many researches have been conducted on the south Pacific islands, and thus present-day magmatism in the South Pacific is well constrained. In contrast, little constraint has been given to the past volcanic activity in this region. It has been reported that there are two peaks in the volcanic activity in the South Pacific during the Cretaceous (after Winterer, 1976). As the volcanic peaks should be essential to understand the large mantle up-welling, an investigation on the volcanism during the Cretaceous should be important. The Cretaceous age seamounts near the Japan Trench (e.g., Takagi et al., 1989) are thus suitable examples to study the magmatism in the South Pacific. For the evaluation of the origin of these seamounts, we collected samples from these seamounts during Geological Survey of Japan DainiHakureimaru cruise. Drilling and dredging were conducted at 12 sites and fresh samples, alkali basalts and andesites, were recovered from three Seamounts; Bousei, Hotta and Smetanin Seamounts. The phenocrysts are mainly plagioclase and pyroxene with/without hornblende. The geochemical character of the rocks from the Bosei and Hotta are HIMU-type. In contrast, those from the Smetanin show less HIMU signature. In this presentation, we will discuss the origin of Seamounts along Japan Trench and constrain the volcanic activity in the South Pacific.

References Winterer, E. L (1976) in The Geographics of the Pacific Ocean basin and its Margin. Geophys. Monogr. 19, 269278. Koppers A. A. P., Staudige H., Wijbrans J. R., and Pringle M. S. (1998) Earth Planet. Sci. Lett. 163. 53–68 Takigami, Y., Kaneoka, I., Ishii, T. and Nakamura, Y. (1989) Palaeogeography, Palaeoclimatology, Palaeoecology, 71. 71-81.

1

Graduate School of Science and Technology, Kumamoto University, Japan ([email protected], [email protected], [email protected]) 2 Department of Environmental Systems, Rissho University, Japan (tfukuoka@ris. ac.jp) 3 National Science Museum, Tokyo, Japan ([email protected]) Volcanism in Kyusyu island is associated with subduction of Philippine Sea Plate (PSP) under Eurasian plate. KyusyuPalau Ridge, which subducts nearly at right angles to central part of Kyusyu, marks the boundary between young (15-26 Ma) PSP in the north and old (37-115 Ma) PSP in the south. Thus volcanic rocks from Kyusyu are good samples to test the compositional difference resulting from the subduction of ocenic plates with different ages. We studied the difference in subduction component using boron element, As boron is enriched in altered oceanic crust and ocean sediment, it is one of the best indicators to show the contribution of fluid phase coming from the subducting plate to the mantle beneath volcanic arc. The release of the fluid phase from the oceanic plate is believed to occur in two manners. One is by aqueous fluid at high temperature, and the other is by the dehydration process, i.e. the breakdown of hydrous phases at high pressures. In this case, fluid is continously released from the plate with increasing pressure, because various phases were involved in producing boron-rich fluid at different pressures. The ratio of B/Nb was used to see the contribution of the subducting plate. It eliminates the influence of different degrees of partial melting in the mantle and fractional crystallization and crustal contaminationin the crust. The across-arc variation as observed by B/Nb ratios in volcanic rock is almost absent in the northern Kyusyu (NK), whereas it shows a gradual decrease in B/Nb from the volcanic front to the back-arc side in southern Kyusyu (SK). In addition, the B/Nb ratio of rocks from the volcanoes at the volcanic front is relatively high in the SK (~7.5), whereas it is relatively low in the NK (~3.5). The lack of lateral variation in subduction component in volcanic rocks from the NK indicates a possibility that the relrease of fluid phase is completed before the oceanic plate reaches the volcanic front. The relatively low number in B/Nb ratio also supports this hypothesis. The young age of oceanic plate suggests a high temperature gradient in the depth profile of the subducting plate. In contrast, the volcanic rocks from the SK show a normal across-arc variation of subduction component typically related with the subduction of relatively old and cold oceanic plate.

Goldschmidt Conference Abstracts 2007

A932

Oxygen and sulfur isotope characterisitics of the Salmagora Complex, Kola Peninsula D.B. SHIN

1

AND M.J. LEE

2

1

Kongju Nat’l Univ, 314-701, Korea, ([email protected]) KIGAM, Daejeon, 305-350, Korea, ([email protected])

2

Introduction Oxygen and sulfur isotope study on the rock of dunitewehrlite-melilitolite (type 0) from the early stage of magma differentiation and on the rock of pyroxeneite-ijolite (type 1 and type 2) from the later stage, which are classified based on the mineralogical and geochemical characterisitcs, is performed for the contribution to the petrogenetic model of the Salmagora ultramafic-alkaline-carbonatite complex in the Kola Peninsula.

Experimental Method 25 bulk rock samples were analyzed for oxygen isotope using silicate oxygen preparation line and Finnigan MAT 252 mass spectrometer in Indiana University and splits of the samples used for oxygen isotope were also analyzed for sulfur isotope composition by CF-mass spectrometer.

Inorganic arsenic speciation in contaminated soils, in Korea MIYOUNG SHIN, HYEON YOON, MISUN PARK AND CHEOLHO YOON Korea Basic Science Institute, Seoul center, ([email protected])

Materials and method The selected site for this study is an area of arsenic contaminated region (abondoned mine) and near land. Total arsenic concentration in soil samples and reference materials were as follows. 0.1g of sample was accurately weighed into acid washed digestion vessels containing 9ml of HCl and 3ml of HNO3, and the resultant solution was heated at 140ºC for 1hr in a digestion block. The solutions were allowed to cool and then diluted to 100ml with deionized water. The extraction of the arsenic species for speciation analysis were as follows. 0.1g of soil sample and 15ml of extractant (1M of phosphoric acid + 0.1 M of ascorbic acid) were reacted in a microwave system (Milestone 1200 Mega, power 60W, 15min). This solution is then cooled at room temperature and centrifuged at 2500 rpm for 30min. Total arsenic and arsenic speciation of the supertanant solution was anayzed by ICP-AES and SPE-HG-ICP-AES respectively.

Result and discussion

Fig. 1. Variations of oxygen isotope compositions of the Salmagora Complex.

Results and Discussions In Figure 1, we can observe that oxygen isotope compositon systematically increases from type 0, type 1 and to type 2. Especially each rock type of pyroxenite, melteigite, ijolite and urtite in type 2 has higher value than that in type 1 by +0.6 to +2.4ಽ. Sulfur content was below analytical limit in type 0. The isotope composition varies from -9.3 to -4.8‰ for only melteigite in type 1 and from -4.2 to +1.0‰ in type 2. The increases in oxygen isotope compositon from type 0 to type 1 can be explained by common magmatic differnetiation process [1] with the influence of crustal contamination, while type 2 seems to indicate another highly evolved magma. Similarly, the low sulfur isotope values in type 1 may also be the effect of crustal contamination and the higher values of type 2 indicates the effect of more evolved magma.

Reference [1] Epstein, S. and Taylor, Jr. H.P. (1967). Researches in Geochemistry. John Wiley, 633p.

Total arsenic concentration in samples were measured between 109.97 and 11758.31ppm. In all samples, the majority of arsenic was present as AsV, while AsIII only account for < 3% of the total arsenic. The extraction efficiency, relative to the total arsenic concentration, varied from 25 – 108%. Most of the samples - except to one samples – showed recovery efficiency of above 86%. Only one samples exist both AsIII and AsV and recovery efficiency is not good (respectively 3.18% (AsIII) and 22.18% (AsV)). This sample is tailing which is mostly composed of calcite mineral. Concentration of the Ca is about 24%. AsIII of the tailing sample is presumed to be adsorbed on the calcite surfaces.

Reference S. Garcia-Manyes et al., (2002), Talanta, 58, 97-109. C. Hutton et al., (2005), Minelogical Magazine, 69(5), 577589 Myron Georgiadis et al., (2006), Environmental pollution, 141, 22-29

Goldschmidt Conference Abstracts 2007

A933

Minor and trace element geochemistry of a branching coral Acropora sp. skeleton

Mantle and crustal processes in the Hadean and Archean: Evidence for the onset of subduction at 3.8 Ga

K. SHIRAI1, N. TAKAHATA2, H. AMAKAWA2, T. WATANABE3, T. KAWASHIMA3, K. SOWA3, Y. SANO2

STEVEN B. SHIREY1, BALZ S. KAMBER2, MARTIN J. WHITEHOUSE3, PAUL A. MUELLER4 AND ASISH R. BASU5

1

HADEEP, Ocean Research Institute, the University of Tokyo ([email protected]) 2 Center for Advanced Marine Research, Ocean Research Institute, the University of Tokyo 3 Dept. Natural History Sciences, Graduate School of Science, Hokkaido University Incorporations of chemical elements and isotopes, into coral skeletons are influenced by ambient water conditions which may provide important information on past climate in the tropics. The geochemistry of Acropora is significantly important for the past-temperature reconstruction because Acropora is one of the main genera constituting the coral reefs. Acropora may also provide unique opportunity to evaluate the effects of physiological processes on the elemental incorporation. Branching coral Acropora consists of fast-growing axial corallite and slowly growing radial corallite at the visible scale. On the other hand, at the micro-scale, there are several types of skeletal elements precipitated under different biological mechanisms. However, geochemistry of branching corals has not been well understood. To investigate the mechanisms of elemental incorporation into Acropora skeletons, chemical and isotopic compositions in the skeleton were analyzed at various spatial resolutions. The chemical profiles of both axial and radial corallite along with growth axes were measured by conventional ICPMS and Stable Isotope Mass Spectrometry. The tip and basal parts of Acropora skeletons were also analyzed at micro-scale. The Mg/Ca, Sr/Ca, Ba/Ca, and U/Ca ratios were measured in ~8Pm diameter spots by using NanoSIMS, and Mg, Sr, Ca, and S distributions were analyzed by Electron Probe Micro Analyzer (EPMA), with a spatial resolution of ~2Pm. Based on the elemental distribution obtained by EPMA, we found that the Acropora’s skeleton is composed of more than three types of the skeletal elements, “Framework”, “Infilling” and “High-Mg Low-S” skeletons. Observation of skeletal structure revealed that the skeletal porosity decreased with distance from the tip, because “Infilling” skeletons possibly filled the space between “Framework” skeletons. Micro-scale elemental analyses (EPMA and NanoSIMS) revealed that “Infilling” skeletons have lower Mg/Ca and higher Sr/Ca and U/Ca than “Framework” skeletons. Since the “Infilling” skeletons were probably formed under the slower calcification rate than “Framework” skeletons, the elemental fractionation pattern between two skeletal elements is consistent with the model of elemental incorporations dependent on calcification rate. The chemical profiles of axial corallite along with the growth were significantly affected by the proportions of “Infilling” skeletons.

1

Department of Terrestrial Magnetism, Carnegie Institution, Washington DC 20015; [email protected] 2 Department of Earth Sciences, Laurentian University, Sudbury ON P3E 2C6, Canada; [email protected] 3 Laboratory for Isotope Geology, Swedish Museum of Natural History, Stockholm S-105 05 Sweden; [email protected] 4 Department of Geological Sciences, University of Florida, Gainesville FL 32611; [email protected] 5 Earth and Environmental Sciences, University of Rochester, Rochester NY 14627; [email protected] Considerable geochemical evidence supports initiation of plate tectonics on Earth shortly after the end of the Hadean. The present upper mantle retains old heterogeneities, some of which likely derive from subduction in the early Eoarchean. Nb/Th and Th/U ratios of mafic-ultramafic rocks from the depleted upper mantle begin to change from 7 to 18.2 and 4.7 to 2.9 (respectively) at 3.6 Ga. This signals the appearance of subduction-altered slabs in general mantle circulation from subduction initiated at 3.8 Ga. Juvenile crustal rocks begin to show derivation from progressively depleted mantle with typical igneous ˢ Nd:ˢ Hf = 1:2 after 3.6 Ga. Cratons with stable mantle keels that have subduction imprints begin to appear at 3.5 Ga. These changes all suggest that extraction of continental crust by plate tectonic processes was progressively depleting the mantle from 3.6 Ga onwards. Neoarchean subduction appears largely analogous to present subduction except for the production of large cratons with thick mantle keels. The earliest Eoarchean juvenile rocks and Hadean zircons have compositions that reflect the integrated effects of separation of an early enriched reservoir and fractionation of perovskite from the Mars-sized, impact-derived magma ocean, rather than separation of voluminous continental crust or oceanic plate tectonics. Hadean zircons most likely were derived from a continent-absent, initially mafic to ultramafic protocrust that was multiply remelted between 4.4 and 4.0 Ga under wet conditions to produce more evolved (felsic) rocks. If the protocrust was produced by global mantle overturn at ca 4.4 Ga, then the transition to plate tectonics resulted from radioactive decay-driven mantle heating. Otherwise, such protocrust would have been the typical product of mantle convection and the transition to plate tectonics resulted from cooling and stabilization of lithospheric plates.

A934

Goldschmidt Conference Abstracts 2007

The influence of neutron irradiation and thermal annealing on helium diffusivity in apatite

Geochemistry of Late Cretaceous tholeiitic volcanism and oceanic island arc affinities of the Chagai arc

DAVID L. SHUSTER1 AND KENNETH A. FARLEY2

REHANUL HAQ SIDDIQUI1, MUHAMMAD ASIF KHAN2 3 AND MUHAMMAD QASIM JAN

1

Berkeley Geochronology Center, 2455 Ridge Rd. Berkeley, CA 94707 USA, ([email protected]) 2 Division of Geological and Planetary Sciences, Caltech, Pasadena, CA 91125 USA, ([email protected]) Accurate extrapolation of helium diffusivity over temperature and time is essential for quantifying long term erosion rates and the topographic evolution of Cenozoic orogens using (U-Th)/He thermochronometry. Naturally occurring radiation damage was recently shown to influence helium diffusion in apatite, such that the kinetics is controlled by the parent nuclide conentrations as an evolving function of time [1]. This implies that apatite helium ages are sensitive to lower temperatures and shallower depths than indicated by the diffusion kinetics presently observed in a given mineral. Here, we present results of controlled experiments to quantify the effects of (a) synthetic irradiation and (b) thermal annealing on helium diffusion kinetics in both synthetic and natural apatites. Exposure to a 1 MeV equivalent neutron fluence (ĭeq, 1MeV, Si) of 2×1018 n/cm2 (90 hours in the cadmium lined in-core irradiation tube, CLICIT, facility of the Oregon State University TRIGA reactor) caused the helium closure temperatures (Tc; 10 ºC/My) to increase by up to +27 ºC. The ¨Tc negatively correlates with the initial Tc, where apatites with higher initial Tc were less perturbed by the neutron irradiation than samples with lower initial Tc. Conversely, simply heating natural apatites to 550ºC for 1 hour caused Tc in all cases to decrease to 47±7 ºC regardless of the initial Tc (the maximum observed ¨Tc= -44 ºC). The resulting Tc agrees well with diffusion parameters constrained [1] for radiation damage-free apatite, Tc = 52 ºC. These results clearly demonstrate that exposure to radiation causes the retentivity of helium in apatite to increase, whereas exposure to temperatures at which thermal annealing occurs causes the retentivity to decrease. The experiments suggest that after 1 hr at 550ºC, effectively all natural radiation damage was annealed in each sample to yield a common Tc. This closure temperature (~47ºC) would correspond to the diffusion kinetics in the damage-free apatite structure. From radiation damage theory we estimate the damage caused by the 90 hr neutron irradiation to be roughly equal to the alpha recoil damage corresponding to [4He] ~10-8 mol/g, or roughly the present 4He concentration of Durango apatite. This implies that Durango apatite should presently have Tc ~27ºC above the completely annealed state, or ~74ºC, which is in excellent agreement with observations. This study highlights a need to quantify the kinetic effects of damage accumulation and annealing on helium diffusion in minerals used for (U-Th)/He thermochronometry.

Reference [1] Shuster, D.L., R.M. Flowers, and K.A. Farley, (2006). Earth and Planetary Science Letters,. 249,. 148-161.

1

Geoscience Laboratory, Geological Survey of Pakistan, Shahzad Town, Islamabad Pakistan, ([email protected]) 2 National Centre of Excellence in Geology University of Peshawar, Pakistan, 3 Quaid-e-Azam University Islamabad, Pakistan The major part of the Chagai arc occur in the westernnorth part of Pakistan and a small part of it also extends towards north in Afghanistan and west in Iran. The Late Cretaceous volcanic rocks which are designated as Sinjrani Volcanic Group, is the most wide spread and the oldest unite of the Chagai arc. This volcanic Group is mainly composed of basaltic to andesitic lava flow and volcaniclastics including agglomerate. The petrological studies of various lava flow revealed that these are mainly basaltic-andesites (53.27-55.93 wt.% SiO2) with minor basalts (49.57-52.14 wt.% SiO2) and andesites (59.12-.59.88 wt.% SiO2). Petrochemical studies based on major and trace elements suggest that these are medium to low K tholeiites. The trace element show variable enrichment in LILE and depletion in HFSE relative to N-MORB. Their primordial mantle-normalized trace element patterns show marked negative Nb anomalies with positive spikes generally on K, Ba and Sr which strongly confirm their island arc signatures. The chondrite normalized REE patterns shows minor but variable enrichment of LREE and positive Eu anomalies. The Zr/Y versus Zr, and Cr versus Y studies, lowers Mg # and lower abundances of Ni and Co suggest that the parent magma of these rock suites was generated by about 15-30 % melting of depleted sub-arc mantle source, and fractionated in an upper level magma chamber en-route to eruption. These volcanics exhibit lower 87Sr/86Sr ratios (0.7038-0.7049), which are consistent with a depleted mantle source and closely correlate with oceanic island arcs rather than continental margin type arcs. On the basis of these studies it is concluded that the Chagai arc was initially developed as an oceanic island arc which was formed due to the intra-oceanic convergence in the Ceno-Tethys during the Late Cretaceous rather than constructed on the southern continental margin of Afghan block, as previously claimed by several workers.

Goldschmidt Conference Abstracts 2007

The age of the Danube fault, 40 years after W. Schreyer W. SIEBEL1, H.P. HANN1, CH. BERTOLD1, C.K. SHANG1, J. ROHRMÜLLER2 AND K.WEMMER3

A935

Peat deposits from Central Europe to the East European Plains investigated by uranium-series dating M. SIERRALTA AND M. FRECHEN

1

Universität Tübingen ([email protected]) 2 LfU, Martredwitz ([email protected]) 3 Universität Göttingen ([email protected]) As a result of his PhD study, Werner Schreyer published several papers about petrographic work in the Bavarian Forest where he investigated Variscan basement rocks close to the Danube valley. He also addressed the age of the Danube fault based on his observation from Natternberg, geological map sheet Deggendorf, and finally (in 1967) concluded that the Danube fault was formed during late-Variscan times and not, as suggested earlier (1961), during the Tertiary period. The Danube fault stretches for about 200 km and is one of the most impressive fault lines in central Europe. As revealed by our recent field mapping, ancient motion along this fault has produced intensive cataclastic deformation along the Donaustauf segment of the fracture zone. The strain localised in this zone resulted in pervasive brittle deformation of the primary rock type, K-feldspar dominated granite. The cataclastic material was ultimately subjected to argillic alteration and K-feldspar was almost completely transferred into illite and other phyllosilicates. The crystallization age of the granite (known as “Kristallgranit”) derived from an unaltered sample is 325 Ma (Pb- evaporation method) whereas the age of argillic alteration is constrained by K-Ar dating of illite fine-fractions (<2 Pm) at 266 and 255 Ma. The new ages bracket the time of deformation and imply near surface exhumation of the fault rocks already during the Permian period, confirming Werner Schreyer’s conclusion from 1967. Post-Cretaceous movement along the Danube fault, as indicated by offset of Mesozoic and Tertiary strata, did not re-open the K-Ar illite system. The illite ages also suggest that hydrothermal fluorite mineralization, genetically connected with the Danube fault, was an early Permian process.

References Schreyer, W., (1967), Geol. Bavarica 58: 77-85.

Leibniz Institute for Applied Geosciences, GGA-Institut, Stilleweg 2, 30655 Hannover, Germany, ([email protected]) Interstadial and interglacial peat deposits are widespread in a transection between Central Europe, Lithuania, Russia, and Siberia, and enable us to reconstruct the vegetation history of the Northern Hemisphere. Multidisciplinary studies including lithostratigraphy, palaeontology, and palynology were performed by partners all over the transect. The reliable chronological frame for the reconstruction of climate and vegetation history was investigated by uranium series dating. The suitability of the peat layers for dating strongly depends on two essential prerequisites: (1) during the initial formation process any thorium was absent and (2) a geochemically closed system behaviour excluding uranium and thorium migration after deposition. However, peat may contain varying amount of admixed thorium by dust and clay minerals, while water passing through the peat layer can cause migration of uranium. Therefore, dating has to be carefully checked for the fulfilments of the prerequisites. Uranium series dating was performed on peat and organogenic deposits applying the thermal ionisation mass spectrometry (TIMS). Coeval peat samples were burned to ash and prepared by the leachate/leachate technique, spiked, and chemically separated for measurements of the isotopic composition. Once isotope activities are determined, age calculation depends on the evaluation techniques used to obtain reliable ages. The evaluation procedure includes (1) estimation of the thorium index by the isochron method to (2) correct the activity ratios for admixed detrital thorium, (3) calculation of corrected single 230Th/U ages, (4) checking corrected ages with the Chisquare test, and (5) calculation of weighted mean of isochron derived detritally corrected age for the deposit. The study of a variety of sections of burial peat on the transect gave a widespread overview on the suitability of peat deposits. Several case studies were investigated by uranium series dating. At the Netiesos section located in Lithuania, the age determination for the peat failed owing to the impossibility of determining an isochron to correct the single ages for the admixed detrital thorium. Furthermore, the investigation of the peat section of Gröbern, Germany, failed due to wide spreading activity ratios clearly demonstrating open system behaviour. Comparisons with radiometrically investigated uranium series ages retrieve the possible reasons. An isochron derived detritally corrected age of 219 ±8 ka was successfully determined for the peat layer from Krivosheino in Siberia, which is in excellent agreement with independent age control.

Goldschmidt Conference Abstracts 2007

A936

Pb-226Ra disequilibrium in basalts from Surstsey Island (Iceland) and implications for magma transport time

210

Climate changes and volcanic signals during the Bronze Age: A stalagmite record Z. SIKLOSY1, A. DEMENY1, T.W. VENNEMANN2, E. HEGNER3, J. KRAMERS4 AND SZ. LEEL-OSSY5

O. SIGMARSSON 1

Laboratoire Magmas et Volcans, OPGC-Université Blaise Pascal-CNRS, 5 rue Kessler, 63038 Clermont-Ferrand (France) and Institute of Earth Sciences, University of Iceland, Reykjavik (Iceland). ([email protected]) Primitive basalts with radioactive disequilibrium between isotopes of the 238U decay chain may provide constraints on the timescales of mantle melt migration. Few results still exist on the 210Pb-226Ra disequilibria. Due to the short half-live of 210 Pb (22.3 years), only basalts significantly younger than 100 years old can be studied for 210Pb-226Ra disequilibria generated in the magma pluming system or the mantle. Most lavas measured so far show either 210Pb-226Ra equilibria or 210Pbdeficit which have been attributed to the degassing of 222Rn in shallow magma chambers. Excess 210Pb has also been observed in a few cases and explained by accumulation of 222 Rn that decays to 210Pb. Icelandic tholeiites from the last century are in radioactive equilibrium with (210Pb/226Ra) equal to unity. These basalts are fed from shallow magma chambers having residence time exceeding 100 years. In contrast, primitive alkaline basalts (MgO =7-12 %) from Surtsey island had (210Pb/226Ra) ranging from 0.45 ±0.04 to 0.82 ±0.06 at the time of eruption. These large 210Pb deficits are unlikely to result from shallow magma degassing since no magma chamber existed beneath this volcanic island which was born during the 1963-67 eruption. The 210Pb-226Ra disequilibria increase from the beginning towards the end of the eruption when the most primitive basalts were produced, and decrease systematically with increasing Th content. These same basalts show a negative correlation between Pb and Cu abundances which are inconsistent with exsolution of sulfur rich liquid or crystallisation of sulphides as a fractionation mechanism of 210 Pb and 226Ra. The large deficit of 210Pb in Surtsey lavas were thus most likely generated during mantle partial melting. In such a case, the time of melt transport from the source region to surface is constrained to be significantly shorter than 100 years.

Institute for Geochemical Research, Hungarian Academy of Sciences, Budapest, Budaorsi út 45, H-1112, Hungary, ([email protected]) 2 Institute of Mineralogy and Geochemistry, University of Lausanne, Switzerland, ([email protected]) 3 Institute for Mineralogy, Petrology and Geochemistry, University of München, ([email protected]) 4 Institute of Geological Sciences, University of Bern, Switzerland, ([email protected]) 5 Eotvos University, Budapest, Hungary, ([email protected]) In this study we present high-resolution and multi-proxy records of a Holocene stalagmite showing volcanic signals detected the first time in a stalagmite by REE analyses. Stable oxygen and carbon isotope data profile along the speleothem deposited during the last 5000 years in the Mecsek Mts. (SHungary) suggest relatively stable conditions in most of the studied period. However, a significant d18O decrease (>2 %o) between approx. 3800 and 3500 years BP occurs in the record. Fluid inclusion water shows also significant D-depletion, supporting cooling. Combined isotope and trace element measurements indicated coupled temperature and precipitation quantity changes occuring in the above period. Rare earth elements (REEs) were also measured by LA-ICP-MS technique, and unlike the longer trends shown by C and O isotopes, the REE and Y distributions indicate sudden changes at the beginning of the cooling period. The La/Y ratio of this segment shows similarities with the volcanic rocks of of the Thera (Santorini) eruption that occurred at about 3650 years BP. As an independent indicator, 87Sr/86Sr ratios show slight decrease at the REE peak, supporting the inferred volcanic signal. The climatic conditions ameliorated rapidly (within ~100 years) to close to present day conditions as reflected by the C, O and H isotope compositions. However, some of the trace elements show marked changes following the recovery. Elements indicating detrital material within the carbonate matrix (e.g. Si, Al, Th) show marked elevations, along with 87Sr/86Sr increase, suggesting increased amount of silicious material transported by dripwaters. However, the Mg content is also higher at this section part, thus, the increase of detrital material amount may be related to lower carbonate precipitation rate, rather than to stronger weathering of silicate rocks. This study was financially supported by the Hungarian Scientific Research Fund (OTKA T 049713) and the Hungarian Scholarship Board (travel grants to Z.S.).

Goldschmidt Conference Abstracts 2007

A geochemical gradient along the North Mid-Atlantic Ridge revisited: New Hf and Pb isotope data SERGEÏ SILANTYEV1, LAURE DOSSO2 AND BARRY HANAN3 1

Vernadsky Institute, Russian Academy of Sciences - 117975 Moscow, Russia ([email protected]) 2 CNRS UMR 6538, IFREMER, BP70, 29280 Plouzané, France ([email protected]) 3 Department of Geological Sciences, San Diego State University, 5500 Campanile Drive, San Diego, CA 92182, U.S.A. ([email protected]) Basalts dredged along the Mid-Atlantic Ridge north of the 15°20’N fracture zone and the so-called 14°N anomaly (Bougault et al., 1988), from 15° 44’ to 17° 28’ show a well documented gradient from enriched to depleted chemical characteristics (Dosso et al. 1991) where Sr isotopic compositions vary from 0.70288 to 0.70217, respectively. We will present new Hf and Pb isotope data for this region of the MAR. Preliminary results show that the Hf and the Pb isotope ratios range between the ambient local MORB mantle source that defines the 14°N bathymetric and C-like (Hanan and Graham, 1996) geochemical anomaly and an extremely depleted mantle source unlike any other so far recognized along the North Atlantic ridge. The results will be discussed in terms of the regional tectonic framework.

References Bougault H., Dimitriev L., Schilling J.-G., Sobolev A., Joron, J.-L., Needham H.D., (1988), Earth Planet. Sci. Lett. 88, 27-36. Dosso L., Hanan B., Bougault H., Schilling J.-G., Joron J.-L., (1991), Earth Planet. Sci. Lett. 106, 29-43. Hanan, B.B. and Graham, D.W. (1996), Science 272, 991-995.

A937

Detection of biomarkers in oils using ToF-SIMS SANDRA SILJESTRÖM1, 2, TOMAS HODE3, JUKKA LAUSMAA2, JAN TOPORSKI4, VOLKER THIEL5 AND PETER SJÖVALL2 1

Department of Geology and Geochemistry, Stockholm University, Sweden ([email protected]) 2 Department of Chemistry and Materials Technology, SP Technical Reseach Institute of Sweden, Sweden 3 Department of Geology, Portland State University, USA 4 WITec, Germany 5 Department of Geobiology, University of Goettingen, Germany To identify biological traces in geological samples molecular biomarkers are widely used. Traditionally GC-MS (Gas chromatography-mass spectrometry) has been used for the analysis. In recent years when concentrations of target compounds were low surface sensitive ToF-SIMS (Time of Flight – Secondary Ion Mass Spectrometry) analysis has been applied. We present the results of a study using GC-MS and ToF-SIMS for the characterization of biomarkers in oils and oil extracts, aiming to prepare for the analysis of smallest quantities in oil samples recovered from fluid inclusions. The biomarkers steranes and hopanes were characterized using polyatomic primary ion source (Bi3+) ToF-SIMS. These compounds were present in hexane extracts from natural oil samples by GC-MS analysis. The same extracts were analyzed by ToF-SIMS, allowing identification of individual biomarkers in this complex sample extract. This information was then used to interpret ToF-SIMS spectra derived from the crude oil by direct analysis. Although these spectra show the presence of a vast variety of components and compound fragments, the target biomarker components could be identified. This initial study was done in preparation for the investigation of oil-bearing fluid inclusions in rocks from different environments and of different ages. Fluid inclusions may provide a wealth of yet undiscovered information due to their isolation from the environment. The capability to characterize smallest biomarker quantities from oil-bearing fluid inclusions contributes to the understanding of the environment in which the oil was formed and could provide insight into the biodiversity of early Earth.

A938

Goldschmidt Conference Abstracts 2007

Arsenic contamination of drinking water in some localities of Vila Real – Northern Portugal

Global Ca-isotope signatures in postSnowball Earth cap-carbonates J.C. SILVA1, T.F. NÄGLER1, I.M. VILLA1, K. KYSER2, A.N. SIAL3, G.M NARBONE2, N.P JAMES2, DA SILVA4 AND M.A. FILHO

C. SILVA, H. SOUSA, O. LOPES AND P. FAVAS University of Trás-os-Montes e Alto Douro, 5001-801 Vila Real, Portugal, (pjcf @utad.pt) The occurrence of problems resulting from arsenic contamination of drinking water in Vila Real (Northern Portugal) was recognized by the Supplier Company of domestic water himself. One of the most significant water quality problems that will be present for a long time to come is the Rebordolongo site, which has arsenic contamination far in excess of any other area. This preliminary study discusses the geological occurrence, the origin, the environmental impact and the health risks of arsenic in drinking water of the public water supply of Rebordolongo. The approach taken by this study was to compare the water quality were from boreholes and wells with the water quality in the homes of the consumers. Hydrochemical analyses have confirmed the presence of dissolved arsenic in drinking waters, with concentrations exceeding the current WHO potable water guideline. The total arsenic content of all these waters varies between <10 Pg/L, and 28 Pg/L. The field work have confirmed placed near the one of the wells an abandoned wolfram mine, the waste from which is typically rich in arsenopirite. The mineralogical composition of quartz-veins and waste samples from the mining site were undertaken to assess their potential as a subregional source of arsenic contamination. Tailings from the site are predominantly composed of quartz detritus with abundant arsenopyrite. The sulphides are mainly arsenopyrite and pyrite, with some galena and chalcopyrite. The relatively high arsenic concentrations in the waters from the mine site may reflect the oxidation/weathering of arsenopyrite and other sulphide-bearing mine tailings. This allows for the solubilization of metals. Arsenic concentrations in domestic water were higher than those in the water from boreholes, probably suggesting direct mixing of effluent from mining area. Meanwhile, the domestic waters do not contain more than 14 Pg As/L. However, arsenic is highly concentrated in the iron oxyhydroxides sampled in the bottom of the water reservoir. Upon contact with air, these waters locally form oxyhydroxides deposits, sorbing effectively arsenic. These deposits contain 0.83% As. Therefore, one can conclude that the environmental and health impact of these waters in Rebordolongo is not negligible. A complementary phase that is in progress intends to characterise arsenic and heavy metals speciation more clearly and determine the link with iron oxyhydroxydes.

1

Universitat Bern ([email protected], [email protected], [email protected]) 2 Queens University ([email protected], [email protected], [email protected]) 3 Univeridade Federal de Pernambuco ([email protected]) 4 CPRM ([email protected]) G44/40Ca(seawater) values of Neoproterozoic cap-carbonates in NW Canada and NE Brazil overlying glacigenic units are reported. While cap-carbonates (Rapitian and Jaccoca Fms) overlying ca 0.74 Ga Rapitian diamictites display G44/40Ca(seawater) values increasing from –1.9‰ to –0.7‰, capcarbonates (Raventhroat-Hayhook and Acaua Fms) overlying ca 0.635 Ga Marinoan diamictites display G44/40Ca(seawater) values decreasing from -0.9‰ to –1.8‰, followed by a progressive up-section increase to values near 0‰. The G44/40Ca(seawater) values displayed by the post-Marinoan capcarbonates are undistinguishable from those of their correlative Kailber and Maiberg Fms (NW Namibia, Kasemann et al., 2005). We suggest that changes in the G44/40Ca(seawater) values of the studied cap-carbonates reflect changes in the Ca-isotope composition of their coeval seawater. The similar G44/40Ca(seawater) values displayed by coeval carbonates from a deep to shallow marine profile (lower Hayhook Fm) suggest a homogeneous Ca-isotope composition of oceans. Differences in the degree of diagenetic overprint of these samples along with the sub-greenschits facies of metamorphism of the Brazilian cap carbonates suggest preservation of the sedimentary Ca-isotope compositions even after diagenesis and burial. The global nature of the G44/40Ca(seawater) values, along with varying Caisotope compositions with stratigraphy displayed by the postMarinoan and Post-Rapitian cap-carbonate indicate that Caisotope stratigraphy can be used to correlate cap-carbonate successions worldwide.

Reference Kasemann, S., et al. (2005), Earth and Plan. Sci. Letters, 231, 73-86.

Goldschmidt Conference Abstracts 2007

Macroscopic and infrared spectroscopic investigation of the synergistic effect between small organic ligands and desferrioxamineB on the dissolution of goethite ANNA A. SIMANOVA, PER PERSSON AND JOHN S. LORING Department of Chemistry, Umeå University, Umeå, Sweden ([email protected], [email protected], [email protected]) In several recent papers, Kraemer et al. [e.g. 1-2] have suggested that the dissolution of goethite by desferrioxamineB (des-B) is enhanced in the presence of small organic ligands such as oxalate that actively shuttle iron from the goethite surface to bulk solution. The goal of the present study was to further an understanding of this phenomenon in both the oxalate- and malonate-goethite-des-B systems using macroscopic and spectroscopic methods. All experiments were performed at 25 qC in the absence of visible and ultraviolet light. The amount of oxalate and malonate adsorbed and the concentration of Fe(III) dissolved were studied after a reaction time of one week at total ligand concentrations between 0 and 5 μmol/m2 and covering the pH range 3 to 9.5. Several different types of in situ infrared spectroscopic experiments were performed at constant pH values of both 4 and 6, total oxalate and malonate concentrations of 1 μmol/m2, and des-B concentrations of between 0.1 μmol/m2 and 1 μmol/m2. First, spectra were collected to monitor the adsorption and surface speciation of oxalate and malonate on the goethite surface in the absence of des-B as a function of time, and the desorption of these ligands was followed spectroscopically to clarify the relative labilities of the different types surface complexes. Second, spectra were collected as a function of time after the simultaneous addition of 1 μmol/m2 concentrations of both oxalate or malonate and des-B to identify the complexes present at the goethite surface. Third, oxalate and malonate were reacted with goethite in the absence of des-B, and then spectra were collected as a function of time after repeated 0.1 These μmol/m2 additions of des-B every 24 hours. experiments have revealed the type of surface complex responsible for the synergistic effect between these small organic ligands and des-B on the dissolution of goethite. The results are also discussed with respect to their broader implications toward mechanisms for ligand-promoted mineral dissolution.

References [1] Reichard P.U., Kraemer S.M., Frazier S.W. and Kretzschmar R.K.,(2005) Plant and Soil 276, 115-132. [2] Cheah S.-F., Kraemer S.M, Cervini-Silva J., Sposito G., (2003) Chemical Geology 198, 63-75.

A939

Variable Calcium isotopic fractionation factor in natural carbonated water NEIL G. SIME AND ALBERT GALY Department of Earth Sciences, Cambridge University, UK ([email protected]; [email protected]) Carbonates precipitated inorganically under laboratory conditions are fractionated relative to the coexisting solution by up to about -1‰ in their 44Ca/42Ca ratio [1,2]. Such Ca isotopic fractionation (44/42Dcarb) is dependent upon the rate of carbonate precipitation with a temperature dependency as a knock-on effect [2] and has been atributed to kinetic effects. The mechanism by which Ca gets enriched in the light isotope in carbonate is rather controversial [1,2] but the experimental approach can be reconciled with the data obtained for biomineralisation. Evidence for a kinetic effect behind the preferential incorporation of light isotopes in inorganic carbonate is more elusive since studies at catchment scale [35] are complicated by the role of the vegetation and the incorporation or release of Ca by the biomass. Here we report į44/42Ca for paired dripping water and speleothems previously studied for Mg isotopic fractionation [6]. The 44/42Dcarb ranges from 0.9998 to 0.9988 while the į44/42Ca of the dripping water are either indistinguishible from the values of the country rock or enriched toward heavier values by up to 0.22‰. The largest 44/42Dcarb is found in a slow dripping speleothem while fast dripping stalagtatites precipitate carbonate weakly factionated. Qualitatively, our data support Lemarchand et al. [2] model but with a much greater 44/42Dcarb value for very low rate of carbonate precipitation. While this data is not direct evidence for an equilibrium fractionation of Ca isotopes, it is worth noticing that the incorporation of O and Mg isotopes in all of these speleothems occurs at equilibrium [6] and clearly suggests that the mechanisms responsible for Mg and O isotopic fractionations are distinct from those responsible for Ca isotopic fractionation.

References [1] Gussone N., Eisenhauer A., Heuser A., Dietzel M., Bock B., Böhm F., Spero H. J., Lea D. W., Bijma J., and Nägler T. F. (2003), Geochim. Cosmochim. Acta 67 1375-1382. [2] Lemarchand D., Wasserburg G. and Papanastassiou D. (2004), Geochim. Cosmochim. Acta 68 4665–4678 [3] Schmitt A.D., Chabaux F. and Stille P. (2003), Earth and Planet. Sci. Lett. 213 503–518. [4] Wiegand B. A., Chadwick O. A., Vitousek P. M., and Wooden J. L. (2005), Geophys. Res. Lett. 32 L11404. [5] Tipper E. T., Galy A., and Bickle M. J. (2006), Earth Planet. Sci. Lett. 247 267-279. [6] Galy A., Bar-Matthews M., Halicz L. and O’Nions R.K. (2002), Earth Planet. Sci. Lett. 201 105-115.

A940

Goldschmidt Conference Abstracts 2007

Lithospheric geodynamics with thermo-chemical density anomalies and mineral phase transitions switched on

Elemental and Sr isotope investigations of human tooth enamel by laser ablation-(MC)-ICP-MS: Successes and pitfalls

N.S.C. SIMON, Y.Y. PODLADCHIKOV AND L.H. RÜPKE

ANTONIO SIMONETTI1, MICHELE R. BUZON2 AND ROBERT A. CREASER1

Physics of Geological Processes, University of Oslo, P.O. Box 1048 Blindern, N-0316 Oslo, Norway, ([email protected]) Density is an important physical property of rocks, which e.g. determines the isostatic response of the lithosphere to deformation. The density of a piece of mantle depends on pressure, temperature and composition. These three parameters inevitably change during geological processes such as rifting, but can also vary in apparently passive settings, e.g. at the lithosphere-asthenosphere boundary. The density distribution of the shallow mantle is non-linear and discontinuous due to complex mineralogy and, most importantly, phase transitions. The two shallow phase transitions that have the strongest effect on density are the plagioclase-in transition above ~50 km and the garnet-in reaction below ~40 – 100 km. The depth of these phase transitions strongly depends on the bulk composition of the mantle. The depth of the plagioclase-in boundary is mostly pressure dependent and predominantly controlled by bulk Na2O/Al2O3, whereas the garnet-in transition has a steep and curved Clapeyron slope with a strong depth-dependence on bulk Cr2O3/Al2O3. In an extending lithosphere, the density changes in the lithospheric column due to phase transitions can cause significant deviations from purely thermal subsidence. The amount of subsidence depends on the composition of the mantle, in addition to classical parameters such as the thickness of the crust, the initial geotherm and the amount of stretching. The phase-transition effect is most pronounced for thin crust, strong mantle thinning/upwelling and relatively fertile mantle compositions rich in aluminum and sodium, and can match the effect of thermal expansion. This could explain the pronounced syn-rift uplift and accelerated post-rift subsidence observed in some basins. The incompatible nature and strong mobility of sodium may lead to large changes in bulk Na2O, and therefore density, during minor melting (Na extraction) or metasomatism (Na addition). Densification of portions of the mantle, e.g., due to melt infiltration and shifts in the location of phase boundaries, might also contribute to the formation of gravitational instabilities and small scale convection, e.g. at the lithosphereasthenosphere boundary. The combination of 2-D finite element deformation modeling and thermodynamic phase diagram calculations allows us to evaluate and quantify the feedback between geochemical variations and lithospheric deformation.

1

Dept. Earth & Atmospheric Sciences, University of Alberta, Edmonton, AB, T6B 1S6, Canada ([email protected]) 2 Dept. Sociology & Anthropology, Purdue University, West Lafayette IN 47907-2059, USA ([email protected]) Sr isotope analysis of human tissue has proven to be an important tool for examining ancient human migration. Previous studies have typically conducted Sr isotope measurements by TIMS; however, Sr isotopic measurement by solution mode MC-ICP-MS greatly increases sample volume throughput with little (if any) detriment to the quality of individual analyses (e.g. Buzon et al., in press). Recent studies have also reported reliable Sr isotope measurements and elemental concentrations within geologic and biologic samples at high spatial resolution using various LA-(MC)ICP-MS instrument configurations. The relatively rapid nature of an individual laser ablation Sr isotope analysis (i.e. minutes) could possibly revolutionize (if successful) the manner in which migration studies of ancient civilizations are conducted in the future. We investigated the elemental and Sr isotope ratios of tooth enamel from remains at the New Kingdom period Egyptian colonial site of Tombos (Sudan) by both laser ablation and solution mode-(MC)-ICP-MS analysis. Elemental abundances for both solution and laser ablation modes of analysis were obtained using a ELAN6000 quadrupole ICPMS. Sr isotope measurements obtained in both solution and laser ablation modes were acquired with a NuPlasma MCICP-MS instrument. Laser ablation analyses were conducted using a UP213 system (New Wave Research). Elemental abundances determined by both solution mode and laser ablation-ICP-MS analysis yield comparable values; however, 87Sr/86Sr values obtained by laser ablation are higher compared to their solution mode counterparts. This discrepancy is related to the production of a molecular interference- Ca+P+O (overlaps 87Sr); hence the higher 87 Sr/86Sr values recorded during ablation analyses. Laser ablation studies of enamel may provide relatively precise 87 Sr/86Sr values rather quickly but cannot be used for accurately deciphering historical population migrations (Simonetti et al. in press).

References Buzon M.R., Simonetti A. and Creaser R.A. (in press), J. Archaeol. Sci. Simonetti A., Buzon M.R. and Creaser R.A. (in press), Archaeometry.

Goldschmidt Conference Abstracts 2007

Characterizing degassing and magma recharge from measurement of short-lived U-series isotopes in volcanic gases and lavas KENNETH WW SIMS1 AND PIERRE-JEAN GAUTHIER2

Anthropogenic signatures in sediments of the fast growing urban area of Natal (NE-Brazil) – A study of heavy metals and organic components S. SINDERN1, H.C. OSKIERSKI1, J.C. SCHWARZBAUER2, R.F.S. LIMA3 AND R.A. PETTA3

1

Department of Geology and Geophysics – Woods Hole Oceanographic Institution Woods Hole, MA 02543, USA ([email protected]) 2 Laboratoire Magmas et Volcans – Observatoire de Physique du Globe de Clermont-Ferrand (CNRS, UBP) – 5 rue Kessler – 63000 Clermont-Ferrand, France ([email protected]) Magma recharge and degassing are directly linked to volcanic activity. Determining the timescales of these shallow level processes is fundamental to our understanding of physical eruption dynamics and hazard assessment. At many quiescent but potentially dangerous volcanoes, persistent degassing is the most common manifestation of volcanic activity. While it is generally agreed that these gas fluxes are mostly sustained by the exsolution of volatiles that were initially dissolved in the magma at depth, it has been unclear whether these gases originated from small ascending magma batches (ultimately recycled at depth once degassed) or large magma reservoirs beneath active volcanoes. Setting constraints on magma residence times in shallow degassing reservoirs and/or feeding systems is critical to distinguish between these two end-member models. Measurement of radioactive disequilibria between short-lived 238U-series isotopes in volcanic gases and rocks can provide previously unobtainable constraints on the timescales of shallow level magma dynamics (e.g. magma recharge and degassing). We present new data on radioactive disequilibria (222Rn, 210 Pb, 210Bi, and 210Po) in gases and tephras collected from inside the Santiago Crater of volcano Masaya (Nicaragua). To our knowledge these are the first measurements of 222Rn from magmatic gases. Masaya’s (210Po/210Pb) and (222Rn/210Pb) are best explained by 222Rn enrichment and subsequent decay in gas bubbles of a magma body having a residence time between 3 and 150 years. Our measurements of radioactive disequilibria in Masaya's gases also display significant variability through time (periodically since 2000) that are correlated with observed variations in activity and gas fluxes at the surface. Taken together, these observations suggest that degassing occurs in a large magma reservoir beneath the active Santiago crater and that input of deep pristine magma into this reservoir controls the eruptive activity at the surface.

A941

1

Institut für Mineralogie und Lagerstättenlehre, RWTH Aachen University, Wüllnerstr. 2, D-52062 Aachen, Germany ([email protected]) 2 Lehrstuhl für Geologie, Geochemie und Lagerstätten des Erdöls und der Kohle, RWTH Aachen University, Lochnerstr. 4-20, D-52062 Aachen, Germany ([email protected]) 3 Depto. de Geologia, Campus Universitário, Natal/RN CEP: 59072-970, Brazil ([email protected]) In this study the effect of anthropogenic discharges on the composition of sediments in the Potengi – Jundiai river system near the fast growing city of Natal, NE-Brazil, is investigated. In general, the sediments of the Rio Potengi – Jundiai river system in the studied area are not severely polluted. Rather they represent an incipient stage of anthropogenic accumulation. A previous study (Sindern et al. 2006) has shown that heavy metals such as Zn, Pb, Cu, Cd and in part also Sn, Hg and Ag have enhanced values relative to reference elements such as Al and Fe. Sources are domestic and animal waste, combustion products and hydrocarbons. These heavy metals are mainly bound to organic matter. The elements Cr, Ni and V are characteristic of weathering heavy minerals in crystalline rocks exposed in the catchment area of the river system and are not significantly added from anthropogenic sources. Additionally, a comprehensive spectrum of organic compounds was identified in sediment as well as water samples of the Potengi-Jundiai river system. Individual organic components were characterized dominantly to be of biogenic origin. However, huge amounts of indicative substances, e.g. fatty acids, suggest a significant emission of natural organic matter as the result of anthropogenic waste discharge. This might reflect incomplete of insufficiently working waste water treatment. Further on, well known xenobiotics, such as plastiziers, pharmaceuticals or pesticides, were detected only to a minor extend, although this type of anthropogenic contamination has frequently been found in riverine systems. The anthropogenic impact on the organic emission within the riverine system differs significantly from the organic signature of rivers from other regions, e.g. Europe or North America.

References Sindern, S., Lima, R.F.S., Schwarzbauer, J., Petta, R.A. (2006): Environ. Geol., Online published, DOI 10.1007/s00254-006-0510-z

Goldschmidt Conference Abstracts 2007

A942

Geochemistry and PGE potential of Bangur Gabbro from the Baula-Nuasahi Mafic Ultramafic Complex, Orissa (India) RAVI SHEKHAR SINGH AND S. N. CHARAN National Geophysical Research institute,Hyd.,India ([email protected], [email protected] )

Geological set-up The Baula-Nausahi mafic ultramafic complex comprises of (1) Gabbro-Anorthosite unit, (2) Peridotite unit (with three chromite bands Ganga, Lakshmi and Sankar), (3) Pyroxenite unit and (4) Bangur Gabbro (~3.1 Ga) (Auge et al. 2003).

Geochemistry The Zr/TiO2*0.0001 vs Nb/Y diagram shows these gabbros in the field of basalt. SiO2, Na2O and Al2O3 shows negative trend while MnO and Fe2O3 show positive trend with Mgo which indicates the differentiation trend of magma and is reflected in mineralogy. Trace element pattern of Bangur gabbro shows the positive Rb and Y anomaly and negative Nb anomaly. The positive anomaly of Rb and Y can be attributed to the development of secondary amphiboles due to alteration. The REE pattern suggests that these gabbros were derived from slightly enriched source. The metal ratio diagram shows the PGE potential of Bangur Gabbro as almost all the samples plots in the field of Layered complex. So the samples plotted above the extrusive rocks field also show enough potential as the chrome-spinel in Bangur gabbro may have affected the ratio.

Discussion and Conclusion The hydrothermal processes have lead to the formation of secondary minerals as well as remobilization of Base metal sulphide. The metal ratio diagram shows the potential for PGE and supports the further investigation to delineate the mineralized zone within Bangur gabbro.

References: Auge T, Cocherie A, Genna A, Armstrong R, Guerrot C, Mukherjee MM, Patra RN; (2003), Precambrian Res. 121: 85–101 Sarah-Jane Barnes; (1990), Journal of Geochemical exploration, 37, 91-99

The Alteration Mineralogy and Mass Change of Volcanics from Zigana (Gümüúhane, NE-Turkey) F. SIPAHI, M. B. SADIKLAR AND B. PRACEJUS Jeoloji Müh., KTU, MF, Tr-61080, Trabzon; ([email protected], [email protected]) The Late Cretaceous volcanic rocks around Zigana Mountain (Gümüúhane) at the eastern part of Black Sea Region (NE-Turkey) belong to the east Black Sea metallogenic province, and have intensive hydrothermal alteration although weathering alteration is limited. The basement of the study area is formed by the Late Cretaceous basalt, andesite and their pyroclastics. These rocks are overlied by the Late Cretaceous aged dacitic rocks, namely Dacite-I and Dacite-II described by [1] and [2]. These volcanic rocks are bimodal in character and have tholeitic to calc-alkaline affinity, and have developed in volcanic arc environment. The Late Cretaceous rocks are the host of VMS and vein type deposits in the study area and region. Volcanic rocks in the study area have been altered to the sericite/illite–chlorite facieses, and contain sericite/illite, chlorite, quartz, carbonate minerals (ankerite and calcite), iron-oxide, and rare kaolinite, smectite and epidote as the alteration products. Sericitization/illitization is the most common type of the hydrothermal alteration associated with the Zigana Volcanics, and chloritization is the second. Pyritisation is seen all volcanics, and the most commonly in dacites. In some fields, limonitisation is occasionally present. Epidotization is rare, and especially seen at basalt and andesite. Isocon method was applied to estimate the mass gains and losses of the Zigana Volcanics as a result of hydrothermal alteration. According to this, basalt and andesite, Dacite-I, and Dacite-II have 2-61 % mass gain, 71 % mass gain and 42 % mass loss, and 44 % mass gain and 32 % mass loss, respectively. Namely, both mass gain and mass loss have occurred in volcanics during the hydrothermal alteration of the parent materials. Illitization-chloritization-kaolinitization increase generally from least altered rock to highly altered rock, whereas carbonatization decreases. The relation between metals such as Cu, Pb, Zn and sericitization/illitization, chloritization and silisification shows that fluids which cause sericitization and chloritization did not increase the amount of these metals and, in fact, it can be said that this fluids are poor in these metals. This also shows that they developed under different hydrothermal conditions. This study was supported by DAAD scholarship.

References [1] Sipahi, F. and Sadiklar, M. B. (2004). Eur. J. Min. 16/1, 134. [2] Sipahi, F. (2005). PhD. Thesis, KTÜ FBE, Turkey (Unpub.).

Goldschmidt Conference Abstracts 2007

Results of FTIR studying microdiamonds from gneisses and calc-silikate rocks from mine Kumdi-Kol, Northern Kazakhstan E. SITNIKOVA AND V.S. SHATSKY Institute Geology and Mineralogy, Siberian Branch of the RAS, 3prosp. Koptyuga, Novosibirsk, 630090, Russia ([email protected], [email protected]) Statistically representative selection of microdiamonds from gneiss and calc-silicate rocks of the Kumdy-Kol deposit was first time studied by means of FTIR spectroscopy. Grayish yellow diamonds of the calc-silicate rocks are about 10-90 mkm in size and have cuboidal morphology. The morphology of microdiamonds from gneiss is more versatile. The crystals of octahedral, cubooctahedral and cuboidal morphologies were found there. They also have yellow color with grayish tint. The size of the gneiss microdiamonds varies between 15 and 150 mkm. The size of studied diamonds were about 60-150 mkm. The selected diamonds have been studied with Bruker VERTEX-70 FTIR spectrometer equipped by Hyperion IRmicroscope. IR spectra demonstrate that all of the studied diamonds contain nitrogen defects, which are manifested through the absorption at 1135 cm-1 (C-defects) and 1282 cm-1 (A-defects). Concentrations of nitrogen in the diamonds from different rocks are similar and vary from 700 to 2500 ppm (±20%). The nitrogen aggregation in the studied crystals varies from 30 up to 50% (±6-18%). The absorption at 3107 cm-1 indicates the presence of hydrogen defects. The intensity of absorption at 3107 cm-1 in gneiss diamonds is 5-20, while in calc-silicate rock diamonds - 2-10. Our data indicate that the studied microdiamonds belong to the Ib-IaA type. IR spectra of diamonds from gneiss and calc-silicate rocks differ from each other in additional absorption lines. The spectra of calc-silicate rock diamonds testify to the presence of carbonate (1430 cm-1) and silicate (1090 cm-1) inclusions, water (banding vibrations at 1630 (1650)±5ɫɦ-1 and stretching at 3420 cm-1). The spectra of the gneiss diamonds do not demonstrate the presence of carbonates, silicates and/or water. The differences of inclusion assemblages in microdiamonds from gneiss and calc-silicate rocks are the evidence of the extremely low mobility of the fluids/melts in the course of UHP metamorphism.

A943

“Blood Coltan”: Fingerprinting of columbite-tantalite ores M.A. SITNIKOVA1, F. MELCHER1, T. OBERTHÜR1, F. HENJES-KUNST1, A. GERDES2, H.BRÄTZ3 AND D.W.DAVIS4 1

BGR, Stilleweg 2, D-30655 Hannover, Germany ([email protected], [email protected], [email protected], [email protected]) 2 Institute of Geosciences, Petrology & Geochemistry Altenhöferallee, D-60438 Frankfurt am Main, Germany ([email protected]) 3 Mineralogical Institue, Würzburg, Germany, Am Hubland, D-97074 Würzburg, Germany ([email protected]) 4 Department of Geology, University of Toronto, 22 Russell Street, Toronto ON M5S 3B1, Canada ([email protected]) The term “Blood Coltan” was coined for columbitetantalite ores in the Congolese civil war, as their sale supported the conflict, especially in the eastern provinces of the DRC. Following the United Nations initiative to fingerprint the origin of conflict materials, the German Ministry for Economic Cooperation and Development decided to fund this pilot study on fingerprinting coltan ores. So far about 160 samples have been obtained from the world’s major coltan producing areas. Special attention is directed toward the Ta-Nb-Sn provinces in Africa: DR Congo, Rwanda, Mozambique and Namibia. Using state of the art analytical tools, we are investigating a wide range of mineralogical and chemical parameters obtained from columbite-tantalite ores, in an attempt to distinguish between different ore provinces, down to the individual deposit scale. Methods employed include XRF (bulk major and trace elements), XRD (structure), ore microscopy, fully automated electron microscopy (Mineral Liberation Analysis), electron microprobe analysis (major and minor), laser ablation-ICP-MS (trace elements, isotopes), and TIMS (U-Pb dating). Elevated concentrations of U, and low amounts of common Pb in columbite-tantalite, facilitate the application of the U-Pb system to date columbite-tantalite. Consistent results of TIMS and LA-ICP-MS analyses in three different laboratories prove that columbite-tantalite yields concordant and reliable ages. In the samples analyzed so far from Africa, four age populations are evident: Archaean (>2.6 Ga), Palaeoproterozoic (1.9-2.1 Ga), early Neoproterozoic (“Kibaran”; 0.98-0.93 Ga), and early Palaeozoic (ca. 0.5 Ga). The bulk of the current Central African columbite-tantalite production is from Kibaran pegmatites associated with late “G4” tin granites. Trace element concentration patterns, mineral assemblages, and zoning characteristics in these pegmatites are clearly different from rare element pegmatitehosted Ta mineralization in Alto Ligonha, Mozambique.

Goldschmidt Conference Abstracts 2007

A944

Study of trace elements reactivity in polluted soils: Measure of Cd, Zn, Cu and Pb lability by using DGT and isotopic dilution methods 1

2

1

1

Y. SIVRY , L. DENAIX , M. FABREGUE , DR J. RIOTTE , C. ZOUITEN1, V. SAPPIN-DIDIER2 AND B. DUPRE1 1

LMTG-UMR 5563 UR, 154 CNRS, Univ. P. Sabatier, IRD, Observatoire Midi-Pyrénées-14, av. Edouard Belin-31400 Toulouse, France; ([email protected]) 2 INRA-UMR Transfert sol plante et cycle des éléments minéraux dans les écosystèmes cultivés-71, Av. Edouard Bourlaux BP 81-33883 Villenave d’Ornon, France Industrial production and more precisely the metallurgy led to the dissemination of chemical contaminants in the environment. According to the metallurgical processes, this contamination can be more or less strong and induce, locally, grounds, water and plants pollutions. In this study we try to quantify the labile pool of various trace elements in cultivated gardens, located around an industrial area. The concentrations of heavy metals (As, Cu, Zn, Cd and Pb) measured in various cultivated vegetables are significatively higher than the european reference values (EC466/2001). The accumulation in plants differs in leaf, roots and fruits vegetables and is clearly correlated with the pH of the soils. In order to describe the heavy metal soil/plant exchanges, we tried to determine the concentrations of metals in the labile pool of the soils and to relate it to phytoavailability. An isotopic dilution method (ID) using spiking with stable isotopes [1] and [2] has been performed in these polluted soils to determine the labile pool whereas a Diffusive Gradient in Thin film (DGT) method realised in the same soils permitted to mimic the heavy metal uptake by the roots of the plants [3]. The comparison between these two complementary methods allowed to describe the soil-solution transfer and demonstrated the different behaviours of the measured metals: relationships between labile pool measured by DGT and phytoaccumulation are observed for some of then, such as cadmium or zinc, but not for lead. The kinetic aspects of the soil/solution exchanges were also described by performing ID and DGT, from 1h to 2 weeks. Modelization of these kinetiks gave us the parameters describing the exchange properties and the desorption of the metal from soil to solution. These parameters variated with metal concentration, pH or organic carbon content in the soils.

References [1] Hammer D., Keller C., McLaughlin M. J., Hamon R. E. (2006), Env. Pol. 143 407-415 [2] Sivry Y., Riotte J. and Dupré B. (2006), JGE 88, Issues 1-3, 144-147 [3] Lehto N J, Davison W, Zhang H, Tych W.(2006) J. Environ. Qual. 35 1903-1913

Organic microanalysis by time-offlight secondary ion mass spectrometry PETER SJÖVALL SP Technical Research Institute of Sweden, Chemistry and Materials Technology, P.O. Box 857, SE-501 15 Borås, Sweden, ([email protected]) Time-of-flight secondary ion mass spectrometry (TOFSIMS) is based on the analysis of positive or negative secondary ions emitted from a solid surface during bombardment by energetic ions. The technique has during the past decades become a well established surface analysis technique in materials science, in particular in the field of semiconductors and polymeric materials [1]. This is due to the attractive combination of high analytical sensitivity (in terms of absolute detection limits) and excellent imaging capability, with lateral resolution in the ~100 nm range. The application of the technique to more complex materials and molecular analysis was, however, for a long time hampered by the fact that the ionization process on which the technique is based lead to extensive fragmentation of and low yields (sensitivities) for organic analyte molecules of masses higher than a few 100 u. This was in turn due to the properties of the primary ions used for accomplishing the sputtering and ionization of the sample under study. During the last few years, there has been a strong development around TOF-SIMS for the analysis of organic materials and molecules. In particular, the development of new cluster primary ion sources (e.g., Au3+, Bi3-7+, C60+) have lead to dramatic improvements (typically by a factor 100-1000) in the detection sensitivity for organic molecules in the range from a few 100 u to 2000 u [2]. Due to the developments in instrumentation, in combination with a rapidly growing data base of reference spectra and the use of multivariate statistical tools for spectrum interpretation, TOF-SIMS is now emerging as a highly useful analytical tool for organic microanalysis, for example, in the field of biomedicine. The presentation will give an introduction to TOF-SIMS, and illustrate the capabilities of the technique for analysis of organic materials. Opportunities for applications in the field of organic geochemistry will also be discussed.

References [1] Benninghoven, A. (1994), Angew. Chem. Int. Ed. 33 10231043 [2] Kollmer, F. (2004), Appl. Surf. Sci. 231-232 153-158

Goldschmidt Conference Abstracts 2007

Fluid inclusions, REE and sulfur isotope geochemistry of the Lavrion carbonate hosted ore deposit, SE Attica, Greece 1

2

N. SKARPELIS , V. LÜDERS

AND D.A. BANKS

3

1

Department of Geology & Geoenvironment, University of Athens, 157 84 Zografou. Greece ([email protected]) 2 GeoForschungsZentrum Potsdam, Telegrafenberg, D-14473 Potsdam, Germany ([email protected]) 3 School of Earth Sciences, University of Leeds, Woodhouse Lane, Leeds LS2 9JT, UK ([email protected]) The Pb-Ag-Zn sulfide mineralization of the ca. 3000 year old mining district of Lavrion, comprises mainly mantos and skarn-type massive sulfides. The area is a part of the AtticoCycladic Belt (ACB). Late Miocene granitoid dykes and a granodiorite stock intruded the footwall of the detachment fault that separates the Basal Unit from the overlying Cycladic Blueschist Unit. Late Miocene porphyritic S-type granitoid rocks occur as sills along or within the hangingwall close to the detachment fault. The manto-type ores are spatially associated with the detachment fault, shear bands within marbles, and the shear contacts between marbles and the intercalated metaclastics of the Basal Unit, indicating that the most important structural control of the Lavrion mineralization is related to the large-scale back-arc Miocene extension in the Aegean. The mineralizing event postdated the mylonitic deformation stage of the marbles, as proved by the alignment of ore bodies with the mylonitic foliation planes or by their crosscut relationship. Wall-rock alteration is characterized by carbonatization and minor silicification. Ore deposition involved mainly marble dissolution and replacement, and open space filling. Fluid inclusion studies provide evidence for deposition of sulfides of the manto-type mineralization at temperatures of about 280oC from fluids having salinities between 14 and 17wt% NaCl eq. Quartz and fluorite deposition occurred at lower temperatures (250 to 125 oC) from fluids having highly variable salinities (1-19 wt% NaCl eq.). The į34SCDT values of galena, sphalerite and pyrite range between -4.8 to +3.99 per mil. Most paired samples of sphalerite-galena, sphaleritepyrite and galena-pyrite show either reversed fractionations or unreasonably high temperature values, indicating either sulfur isotope disequilibrium or noncontemporaneous precipitation of sulfides. Chondrite-normalized REE patterns in fluorites show slightly increasing LREE and decreasing HREE contents. All fluorite samples possess weak negative Ce and pronounced positive Eu anomalies. A meteoric water influx late into the granodiorite stock is documented by the fluid inclusions in quartz of widespread extensional quartz veins. There is no clear evidence whether a direct genetic link exists between the manto-type massive sulfides and the Late Miocene igneous activity in the footwall of the detachment or along the detachment fault. Ag-rich sulfide mineralization forming tension gashes within hydrothermally altered hornfelses postdates contact metamorphic phenomena.

A945

Adsorption energy trends on UO2 and ThO2 surfaces F. N. SKOMURSKI, R. C. EWING, AND U. BECKER Department of Geological Sciences, University of Michigan, Ann Arbor, Michigan, U.S.A. ([email protected], [email protected], [email protected]) Determining the interaction of adsorbates such as water and oxygen with surfaces of uranium and thorium dioxide is important for the long-term storage of these isostructural nuclear fuel materials. The semi-conducting versus insulating nature of UO2 and ThO2, respectively, makes comparison of surface-adsorbate interactions with these materials possible as a function of electronic structure in addition to surface structure. Previously, the quantum mechanical code CASTEP was used to calculate surface energies for UO2 and ThO2, and the (111) surface was found to be the most stable relative to the (110) and (100) surfaces in both cases (Skomurski et al., 2006). In this study, CASTEP was used to investigate the interaction of water and oxygen with UO2 and ThO2 slabs of finite thickness representing each crystallographic orientation. The effect of model set-up on adsorption energy trends is discussed for single versus double-sided models in a periodic setting. On a defect-free (111) surface, the adsorption of molecular water is found to be more favorable than dissociated water for both UO2 and ThO2. On the more reactive (110) surface, however, the opposite trend is favored. The adsorption of molecular and atomic oxygen is investigated as a function of spin configuration to determine the very first steps of oxidation on UO2 surfaces. On both the (111) and (110) surfaces, the adsorption of atomic oxygen leads to oxidation of near-surface uranium atoms. Investigations into possible transition state spin configurations for oxygen interacting with the substrate are discussed. A co-adsorption case for water and oxygen is tested on both UO2 and ThO2 surfaces. On the UO2 (111) surface, the presence of water is found to enhance the oxidation of near-surface uranium atoms, a phenomenon attributed to the semi-conducting nature of UO2 as this trend is not observed on the insulating ThO2 surface. Finally, the distance dependence of this “surface proximity effect” (Rosso and Becker, 2002) is tested as a function of distance between adsorbates on the (110) surface. By using quantum mechanical methods to investigate the very first interactions of oxygen and water with actinde oxide surfaces, we start to develop a mechanistic understanding of processes that ultimately affect oxidation and dissolution rates on the macroscopic scale.

References F.N. Skomurski, R.C. Ewing, A.L. Rohl, J.D. Gale, and U. Becker (2006) Am. Min., 91, 1761-1772. K.M. Rosso and U. Becker (2002) Geochim. Cosmochim. Acta., 67(5), 941-953.

A946

Goldschmidt Conference Abstracts 2007

Garnet growth in the Zermatt-Saas Fee eclogites S. SKORA1, L.P. BAUMGARTNER1, N.J. MAHLEN2 AND C.M. JOHNSON2

Investigation of the interaction between green rust sodium sulfate and aqueous selenium L. L. SKOVBJERG AND S. L. S. STIPP

1

Institut de Minéralogie et Géochimie, Université de Lausanne, l'Anthropole, 1015 Lausanne, Switzerland ([email protected], [email protected]) 2 Department of Geology and Geophysics, University of Wisconsin, 1215 W Dayton St, Madison, WI 53706, USA ([email protected], [email protected]) Different approaches have been used to infer the growthlimiting mechanism. The radius rate relations of e.g. Kretz (1973) rely on major element compositional zoning as time markers assuming that all porphyroblasts are precipitating the same chemical composition at any point in time. Comparison of chemical contour lines in variably sized garnets is used to distinguish between the different rate laws. A second approach is based on the 3-D spatial distribution of porphyroblasts. It assumes that depletion halos developing around early porphyroblast in the diffusion-controlled case will inhibit nucleation within the halos. This is supposed to lead to an ordered distribution while a random distribution is to expect for the interface-controlled case. Application of the radius-rate concept to garnets of the eclogites of the Zermatt-Saas Fee ophiolite (ZSF, Western Alps, Switzerland/Italy) suggests that all garnets precipitated the same amount of radius per time interval, hence were grown limited by an interface-controlled growth mechanism. Approximately bell-shaped crystal size distributions suggest continuous nucleation and growth throughout the garnet growth history. Preliminary 3-D spatial distribution data using the nearest neighbour method point towards a random distribution, consistent with an overall interface-controlled growth mechanism, but tend towards an ordered distribution in rocks with higher modal abundance of garnet. This could be the case because new porphyroblasts cannot nucleate randomly if early crystallized garnets are abundant (e.g. Denison et al. 1997).

References Kretz, R. (1973) Canadian Mineralogist, 12, 1-21. Denison, C., Carlson, W.D., and Ketcham, R.A. (1997) J. Metamorphic Geol., 15, 29-44.

NanoGeoScience, Department of Chemistry, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen Ø, Denmark ([email protected]; [email protected]) Selenium is an important trace nutrient in the body at low concentrations but becomes toxic at high concentrations. It is a natural compound in many rocks and sediments in various concentrations, but selenium is also produced as a decay product in radioactive waste. It exists in various redox states of which the higher, selenate (VI) and selenite (IV) are mobile in the aquatic environment. Inorganic reduction to elemental, insoluble selenium is one pathway to minimize the bioavailibility of selenium in areas of increased levels. Green rust is an Fe(II)-containing compound, known for its capability of reducing a large number of oxidised elements and compounds. It belongs to the family of layered double hydroxides (LDHs) with layers of Fe(II)-Fe(III) hydroxide, separated by interlayers of water, anions and for some types also cations (poster by Christiansen et al.). The reduction of selenate and selenite by green rust has been studied during the last decade. We have focused our investigations at the nanoscale level to understand the mechanism of the reactions. We use transmission electron microscopy (TEM), atomic force microscopy (AFM) and X-ray diffraction (XRD) to observe the particles at various times during the reactions. Our results suggest that the mechanism of reaction is dependent on the type of green rust as well as the oxidation state of selenium. We used a green rust type which had SO42incorporated in the interlayers. When SeO42- was added, we observed signs of intrusion into the interlayers. However, when SeO32-, which has a different steric nature, was added, interlayer exchange was not observed. In both cases, the oxidised species were reduced to insoluble elemental selenium. Previous studies on chromate (Skovbjerg et al., 2006) showed that the reduction mechanism is important for the mobility of the reduced contaminant, but in the case of selenium it appears not to be of relevance.

References Skovbjerg L.L, Stipp S.L.S., Utsunomiya S., Ewing R.C. (2006), GCA 70, 3582-3592.

Goldschmidt Conference Abstracts 2007

REE distribution in volkhovites – New type of the tektite-like glasses

Plešovice zircon – A new natural standard for U-Pb and Hf isotopic microanalysis

S. G. SKUBLOV

J. SLAMA1, J. KOSLER2, J. L. CROWLEY3, A. GERDES4, J. HANCHAR5, M. HORSTWOOD6, G. A. MORRIS7, L. NASDALA8, N. NORBERG8, U. SCHALTEGGER9, M.N. TUBRETT10 AND M. J. WHITEHOUSE11

Institute of Precambrian Geology and Geochronology RAS, St-Petersburg, Russia ([email protected]) Volkhovites are the tektite-like glasses of mafic and ultramafic composition; for the first time we have discovered them among fluvial-glacial sediments of Valday glaciations (10-65 thousand years) at the right side of the river Volkhov (North-West Russia). Volkhovite particles are characterized by small size (0.1-3.0 mm), varied microtectite aerodynamic (spherules, drop-shaped, dumbbell-like) and irregular forms, and the perfect safety indicating their postglacial age (Skublov et al., 2007). Volkhovites can be separated into four groups: Caenriched, Mg-enriched, Mn-enriched and K-enriched. REE distribution was studied in volkhovites locally by ion microprobe Cameca IMS-4f (IMI RAS, Jaroslavl, Russia). REE patterns for K-volkhovites and Australasian microtektites are very similar. We suggest that volkhovites were formed as a result of outburst of the fluidisite slag-stone-melt-mud-gas mixture from the crust deep levels up to surface (Skublov et al., 2007). Figure 1: Chondrite-normalized REE patterns for mean values for different types of volkhovites. Dotted line indicates the mean value for the normal Australasian microtektites (Glass et al., 2004).

Ca-volkhovites

Mg-volkhovites K-volkhovites

Mn-volkhovites

Russian Federation President Grant MD-551.2007.5.

References Glass B.P., Huber H. and Koeberl C., (2004), Gechim. Cosmochim. Acta. 68. 3971-4006. Skublov G.T., Marin Yu.B., Semikolennykh V.M. et al., (2007), Proceed. Rus. Min. Soc. 1. 50-68.

A947

1

Department of Petrology, Charles University in Prague, Czech Republic ([email protected]) 2 Department of Earth Science, University of Bergen, Norway ([email protected]) 3 Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, USA ([email protected]) 4 Department of Geosciences, Johann Wolfgang Goethe University, Frankfurt am Main, Germany ([email protected]) 5 Department of Earth Sciences, Memorial University of Newfoundland, St. John's, Canada ([email protected]) 6 NERC Isotope Geosciences Laboratory, Kingsley Dunham Centre, Keyworth, UK ([email protected]) 7 Stockholm University, Department of Geology and Geochemistry, Sweden ([email protected]) 8 Institute of Mineralogy and Crystallography, University of Vienna, Austria ([email protected]; [email protected]) 9 Department of Earth Sciences, University of Geneva, Geneva, Switzerland ([email protected]) 10 CREAIT Network, Memorial University of Newfoundland, St. John's, Canada ([email protected]) 11 Laboratory for Isotope Geology, Swedish Museum of Natural History, Stockholm, Sweden ([email protected]) Increasing number of geological applications requires the use of well-characterized and widely available reference materials for in situ analysis. We have established a new natural zircon standard (“Plešovice”) for U-Pb dating and Hf isotopic analysis by laser ablation ICP-MS. The ID-TIMS U-Pb age of the Plešovice zircon has been determined to 336.9±0.2 Ma (95% confidence limits; mean 206 Pb/238U age). The U-Pb ages obtained by LA ICP-MS (3 labs) and SIMS techniques show larger spread but within their analytical uncertainties they are consistent with the TIMS age. Hafnium isotopic composition of the Plešovice zircon appears to be homogenous within and between grains. Combined laser ablation and solution MC ICP-MS analyses gave a mean 176 Hf/177Hf value of 0.282481±0.000013 (95% confidence limits, 87 analyses). Cathodoluminescence and BSE imaging and chemical analyses revealed that discrete zones in the Plešovice zircon are enriched in trace elements and especially in U and Th. Raman spectroscopy suggests that these highly radiation-damaged areas have not undergone any annealing. Our LA ICP-MS analyses did not indicate any Pb-loss but the high intensities of measured U signal require that these areas are avoided during routine U-Pb isotopic analysis.

A948

Goldschmidt Conference Abstracts 2007

Constraining carbon sources and growth of microbialites in Pavilion Lake, BC using 14C

Modeling marine Carbon and Phosphorus cycling during Cretaceous Oceanic Anoxic Events

G.F. SLATER 1, A.L. BRADY 1, B. LAVAL 2 3 AND D.S.S. LIM

C.P. SLOMP, P. KRAAL AND I. TSANDEV

1

School of Geography and Earth Sciences, McMaster Univ., 1280 Main Street W. Hamilton, ON, Canada 2 Dept. of Civil Engineering, Univ. of British Columbia, Vancourver, Canada 3 NASA-Ames Research Center, Moffett Field, CA USA The morphologically varied freshwater carbonate microbialite structures in Pavilion Lake, B.C. Canada represent an opportunity to investigate the processes leading to their formation and potential associated biosignatures that will contribute to our understanding of geo-microbe interactions and to our ability to interpret the geologic record. A primary question in such systems is determination of the primary carbon sources and cycling. In some systems, such as Mono Lake, carbonate structures are proposed to be the result of abiotic precipitation due to supersaturation resulting from groundwater-surface water mixing. Alternatively, modern stromatolites such as those in Shark Bay and the Bahamas are proposed to form via significant biological influence using bulk DIC. Determination of the '14C of dissolved inorganic carbon (DIC) sources and mid-depth microbialite carbonate demonstrated that microbialite carbonate was significantly depleted in '14C with respect to bulk surface water indicating either contributions of geologically derived carbon or significant time since precipitation. Assuming surface carbonate was recently precipitated, comparison to local and regional groundwater '14C indicates that regional '14C depleted groundwater DIC sources provide 12% of carbonate carbon. '14C of the detrital wood sample resulted in an estimated constant growth rate of 3 to 6 cm/thousand years, approximately double a previous U/Th based estimate. The '14C of a deep water carbonate sample was highly depleted indicating that either groundwater was making a larger contribution to this carbonate or that this carbonate was precipitated significantly earlier than the mid-depth carbonates.

Faculty of Geosciences, Utrecht University, Budapestlaan 4, 3584 CD, Utrecht, the Netherlands, ([email protected]) Phosphorus (P) is a key nutrient and may control rates of primary productivity and organic carbon (C) burial in the oceans. Changes in phosphorus (P) availability thus may have played an important role in the initiation, formation and termination of Cretaceous oceanic anoxic events (OAEs). Besides redox-dependent changes in the recycling efficiency of sediment P, as deduced from elevated organic C/total P ratios in black shales, the marine P cycle can be affected by variations in sealevel, oceanic circulation and chemical weathering. In this study, we use a model for the coupled marine cycles of P and C to examine the relative role of these various factors in determining changes in P availability and organic C burial during OAEs. We focus on OAE-2 (~94 Myrs BP; 500 kyr duration) and specifically study (1) possible triggers for the OAE, such as enhanced weathering and reduced oceanic circulation, (2) factors leading to its termination and (3) the relative role of the continental shelves and open ocean.

Goldschmidt Conference Abstracts 2007

Abiotic Nitrogen reduction in Hadean hydrothermal systems A. SMIRNOV1,2, D. HAUSNER3, R. LAFFERS1, F. M. MCCUBBIN1, D.R. STRONGIN3, AND M.A.A. SCHOONEN1 1

Department of Geosciences, SUNY Stony Brook, Stony Brook, NY 11794, USA ([email protected]) 2 Geophysical Laboratory, 5251 Broad Branch Road, NW, Washington, DC 20015, USA 3 Department of Chemistry, Temple University, PA 19122, USA One of the outstanding questions in Astrobiology is the source and formation mechanisms of NH4+ which presumably was required for reactions of prebiotic synthesis and origin of life. The uncatalyzed reduction of abundant N2 to NH4+ is prohibitively slow due to the strong triple bond in the molecule. However, NO3- and NO2- present in the Hadean Ocean as result of atmospheric reactions may have been more susceptible to reduction. We have experimentally tested the hypothesis, which suggests that Ni, Fe metals and alloys formed as a result of hydrothermal (HT) serpentinization processes in the Hadean oceanic crust could have acted as catalysts and/or reactants in reactions leading to abiotic NH4+. Our results show NO3- and NO2- were converted into NH4+ more rapidly than N2, and the reduction process had a strong temperature dependence. Metals, especially Ni were found to be more effective than alloys in reducing N2 with yields usually not exceeding few percent. Based on the experimental results we have estimated NH4+ yield of Hadean HT systems from to be approximately 1010-1012 mol.yr-1 which is comparable to values estimated by Brandes et al, 1999 (10101011 mol.yr-1) and Schoonen and Xu (2001) (108-109 mol.yr-1) in HT systems as well as Summers and Chang (1993) for NO2reduction by Fe2+. Our estimate only includes N2 to NH4+ reaction yield and therefore if NO2-/NO3- were present in the advected seawater, the yields would have been proportionally higher considering their high conversion rates to NH4+ in the presence of metals/alloys. We have also considered iron meteorites as possible sources of N since they commonly cointain reduced N species such as nitride (N3-) which could have reacted to form NH4+ during dissolution in the Hadean Ocean. When compared, however, with HT production, the meteoritic NH4+ flux during the Late Heavy Bombardment is approximately 6 orders of magnitude smaller.

References Brandes, J.A., Boctor, N.Z., Cody, G.D., Cooper, B.A., Hazen, R.M. and Yoder, H.S. (1998), Abiotic nitrogen reduction on the early Earth. Nature 395, 365-367. Schoonen, M.A.A. and Xu, Y. (2001), Nitrogen reduction under hydrothermal vent conditions: Implications for the prebiotic synthesis of C-H-O-N compounds. Astrobiology 1, 133-141. Summers, D.P. and Chang, S. (1993), Prebiotic ammonia from reduction of nitrite by iron(II) on the early Earth. Nature 365, 630-632.

A949

Melt segregation and near source fractionation: Examples from small scale basaltic systems IAN E M SMITH Geology programme, School of Geography, Geology and Environmental Sciences, University of Auckland, New Zealand ([email protected]) Basaltic magmatic systems, expressed at the earth’s surface as volcano fields, are characterised by very low rates of magma production (< 1.0-4 km3 /year) over relatively long time scales (up to 107 years); these are tiny igneous provinces contrasting with Large Igneous Provinces that represent the other extreme in the spectrum of mantle derived magmatism. Tiny igneous provinces are the result of very small degrees of partial melting from a mantle source. In the Auckland volcanic field of Northern New Zealand, sampling of stratigraphically defined eruption sequences typically shows compositional trends in small magma batches that cannot be accounted for by fractionation involving low pressure mineral assemblages in a shallow pre-eruption environment. Detailed study of several individual volcanic centres in the Auckland field has defined two styles of compositional variation. 1). Least evolved compositions (as defined by geochemical parameters such as Mg-number and incompatible element content) are erupted first followed by more eveolved compositions. This is interpreted to reflect magma extraction from a source in which a thermal gradient has produced a range of melting proportions with the greater proportion melt leading the extraction process. 2). The eruption sequence is initiated by relatively evolved compositions followed sequentially by progressively less evolved compositions. This trend is interpreted as the result of high pressure fractionation immediately above the source in a part of the conduit where melt is thermally connected to its surroundings and side wall crystallisation controls fractionation. The fact that these well defined compositional variations can be observed in stratigraphic sequence shows that the fractionated magma column rose very rapidly and without mixing once it left the source region. Further, such subtle compositional trends are probably only preserved because of the extremely small volumes of melt involved. In the Auckland volcanic field there is evidence for these extraction/fractionation processes occuring in both garnet and spinel peridotite facies giving rise to a range of alkaline to sub-alkaline basalt compositions. There is also evidence for compositionally distinct small magma batches coexisting independantly and rising simultaneously to the surface demonstrating the inherent instability of small scale mantle based magmatic systems.

A950

Goldschmidt Conference Abstracts 2007

Purification of sterols and alkenones for compound specific hydrogen isotopic analysis using HPLC RIENK H. SMITTENBERG*, VALERIE F. SCHWAB JULIAN P. SACHS

Continental temperatures from the Paleocene-Eocene boundary in the Big Horn Basin, WY from carbonate clumped isotope thermometry

AND

University of Washington, School of Oceanography. Box 355351, Seattle, WA 98195, USA. ([email protected]; [email protected], [email protected]) *moving to ETH Zürich, Geological Institute. Universitätstrasse 16, 8092 Zürich, Switzerland. Here we present two variations of a method for purifying sterols and alkenones out of total lipid extracts using an HPLC-MS coupled to a fraction collector. The presented methods reduce the amount of work needed and simplify the procedure to obtain fractions pure enough for compound specific irm-GC/MS analysis, compared to traditional wet chemical techniques. This allows a higher throughput of samples so that high-resolution paleoclimatic or paleoenvironmental proxy records based on compound-specific isotope measurements can be obtained more efficiently. The presented method was developed for hydrogen isotope analysis, and introduces no isotopic fractionation. The method could also be used in other cases where purification of lipid biomarkers out of total lipid extracts is required.

Figure 1: Gas Chromatograms of: A) Total lipid extract of a Chesapeake Bay sediment and fractions as purified by semi-preparative HPLC. Individual alkenones are separated from notoriously co-eluting alkenoates. B) Neutral fraction of a sediment from Palau (West Pacific) and collected fractions containing various alcohol classes.

K.E. SNELL1, J.M. EILER2, D. DETTMAN3 1 AND P.L. KOCH 1

University of California Santa Cruz, Earth and Planetary Science Dept. ([email protected]) 2 California Institute of Technology, Div. of Geological and Planetary Sciences ([email protected]) 3 University of Arizona, Dept. of Geosciences We used carbonate clumped isotope thermometry to constrain growth temperatures of paleosol carbonates and fossil unionid bivalves collected from the Big Horn Basin (Wyoming) from sections that span the Paleocene-Eocene boundary. Long-term global warming of ~10qC occurred from the Late Paleocene to the Early Eocene. The PaleoceneEocene Thermal Maximum (PETM) is an extreme thermal event of short duration (< 200 ky) superimposed on the longterm warming trend, and has been identified globally in the ocean sediment record and on the continents in sedimentary basins. The Big Horn Basin is one such basin that has been extensively studied with multiple climatic and biotic proxies in an attempt to characterize the PETM. Therefore, it is an ideal case study for the new paleothermometry technique we use here. Temperature estimates for the paleosol carbonates capture the pattern of temperature change through time suggested by other paleotemperature proxies, but are consistently higher than previous estimates. Temperature estimates from the fossil mollusk shells, however, are too high to reflect original climatic conditions and do not mimic the stratigraphic change in temperature seen in other proxies. These samples were buried to > 1 km and subsequently exhumed. Our results suggest the paleosol carbonate samples were not dramatically reset by burial metamorphism, whereas the mollusk fossil carbonate was reset by re-crystallization or other processes. We speculate that carbonate that originally forms as calcite is more resistant to resetting during burial metamorphism than carbonate intially formed as metastable aragonite. Although X-ray diffraction analyses detected primary aragonite and no calcite in these fossil mollusks, trace metal analysis and more detailed SEM and/or XRD studies may be required to identify sufficiently unaltered fossil mollusks, if they exist (Came et al, in revision, 2007). We conclude that the soil carbonate data constrain continental climate across the Paleocene to Eocene transition. In addition, the contrast between soil carbonates and fossil molllusks provides an important first case study of the relative ability of different forms of carbonate to retain primary temperatures as measured by clumped isotope thermometry.

Reference Came, R.E., Eiler, J.M., Veizer, J., Azmy, K., Brand, U., and Weidman, C.R. (2007) In Revision.

Goldschmidt Conference Abstracts 2007

Melt inclusions and host olivines: What do they tell about mantle processes and sources?

Recycled oceanic crust as a source of Siberian flood basalts A.V. SOBOLEV1,2, N.A. KRIVOLUTSKAYA1, D.V. KUZMIN2AND A.W. HOFMANN2

ALEXANDER V. SOBOLEV12 1

Vernadsky Institute of Geochemistry, Moscow, Russia, ([email protected]) 2 Max-Planck-Institut für Chemie, Mainz, Germany

1

Vernadsky Institute of Geochemistry, Moscow, Russia, ([email protected]) 2 Max-Planck-Institut für Chemie, Mainz, Germany

Recent study (Sobolev et al, 2007) has shown that Gudchikhinskaya suit (Gd2) from the base of Siberian flood basalts in Norilsk region likely formed by melting of olivinefree hybrid pyroxenite, produced by reaction of melts from recycled crust and peridotite. Here we present new data on major and trace element compositions and volatile contents in parental melts for Gd2 basaltic suit recovered from the study of homogenised melt inclusions in olivine phenocrysts by EPMA, LA-ICP MS and SIMS. We show that the composition of trapped melt varies from similar to tholeiitic OIB (e.g. Mauna Loa, Hawaii) with notable depletion in Pb, U, Th and Rb to those enriched in these elements (Fig.1). Abundances of these elements correlate with concentrations of Si, K (positively) and Nb, Ti (negatively). This suggests significant contamination of melt by continental crust during magma fractionation. All melts are severely undersaturated by S and contain low water concentrations. The composition of melts unaffected by crustal contamination indicate oceanic crustal component. This suggests that the recycled oceanic crust was a major source of Siberian flood basalts at the initial stage of LIP formation.

100

XC51 4270/13 Continental Crust

Melt/PM

Host olivines and magmatic (melt and fluid) inclusion studies have had significant impact on petrology and geochemistry of mantle-related igneous processes in recent years. This includes better understanding of melting processes, source heterogeneities, and volatile contents of parental melts. At the same time, new results have shown possible processes (such as magma mixing and interaction with crustal materials) which appear to compromise the idea that melt inclusions represent simple primary liquids (e.g. Danyushevsky et al, 2003). Here I present a summary of last 5 years studies of our group concerning compositions of melt inclusions and host olivines from the mantle plume, LIP and MOR environments. Melt inclusions in olivines are likely trapped during relatively fast growth of crystals in the environments of significant temperature and (or) compositional gradients such as magma mixing. Compositional variability of melt inclusions far exceeds variability of lavas representing endmembers commonly present in bulk rock in highly attenuated form. Melt inclusions of variable compositions coexisting in a single olivine phenocryst are often trapped sequentially at different depths. This suggests complicated multistage crystallization and trapping process rather then nearly simultaneous trapping of locally heterogeneous melts. Data suggest that each volcanic plumbing system or particular lava represents dynamic mixing of numerous parental melts and products of their fractionation and (or) interaction with crystal mashes and melts in shallow conduits. Some inclusions approach primary melt compositions much better then any studied rocks. These inclusions show extreme compositional ranges far exceeding those of bulk surface lavas. They are usually trapped in the earliest crystals formed in the deepest parts of plumbing system. The compositional and isotopic ranges of the recovered parental melts suggest highly efficient open system melting; fast melt transport and compositional heterogeneity of mantle sources in all volcanic environments studied so far. In particular the concentrations of Mn, Ni, Ca, Zn and Sc of early olivine phenocrysts from wide range of mantle derived magmas are not consistent with common peridotitic source and suggest significant amount olivine-free hybride pyroxenite source formed by melting and reaction of recycled crustal component in the convecting mantle (Sobolev et al, 2005, 2007).

A951

Mauna Loa

10

1 Rb Ba Th U Nb Ta La Ce Pb Pr Nd Sr Sm Zr Hf Eu Gd Tb Dy Y Yb Lu

Figure 1:. Average compositions of melt inclusions in olivine from Gd2 picrites (2 samples) compared with continental crust (Rudnik, 2003) and typical Mauna Loa melt (Sobolev et al, 2005) normalized to primitive mantle (Hofmann, 1988).

References

References

Danyushevsky L. V. et al., (2003). Contr. Mineral.Petrol. 144, 619-637. Sobolev A.V. et al, (2007). Science 316, 412-417. Sobolev A.V. et al, (2005). Nature 434, 590-597.

Hofmann A.W., (1988). Earth Planet. Sci. Lett. 90, 297-314. Rudnick R.L., (2003). Treatise on Geochemistry, 3, 1-64. Sobolev A.V. et al, (2007). Science 316, 412-417. Sobolev A.V. et al, (2005). Nature 434, 590-597.

A952

Goldschmidt Conference Abstracts 2007

Evidence for correlation of late CFBs from East Greenland and the Faeroe Islands (North Atlantic Igneous Province)

Occurrence and origin of igneous fragments in chondritic breccias A.K. SOKOL1, 3, M. CHAUSSIDON2, A. BISCHOFF1 3 AND K. MEZGER 1

The Faeroe Islands formed early as part of the North Atlantic Igneous Province (NAIP) and is composed of a 45km thick basaltic lava sequence erupted before and during continental breakup ~56-55 Ma ago (Storey et al. 2007). The present dataset represents the syn-breakup part of the sequence and comprises both enriched high-Ti plume related basalts and depleted MORB-like low-Ti basalts. The Faroese lavapile has earlier been correlated geochemically with the East Greenland Paleogene lavaseries from Kangerlussuaq and Blosseville Coast by Larsen et al. (1999), where the Faroes Middle and Upper Formations were correlated with the Milne Land Formation (MLF) from East Greenland. New data from the southern island of Sandoy on the Faeroe Islands shows that thin counterparts of the East Greenland Geikie Plateau Formation (GPF) and Rømer Fjord Formation (RFF) are present on the Faeroe Islands in the very top of the sequence. The correlation is based on variations in Zr/Nb and 206Pb/204Pb but is also consistent with petrological observations that GPF and RFF lavas are mainly aphyric and RFF lavas have lower contents of SiO2. This means that volcanism has proceeded longer than previously thought on the Faeroe Islands although in this area with much reduced eruption rates. The isotopic compositions of the Faroes and East Greenland Palaeogene lavas show that the high-Ti basalts from MLF and GPF and their Faroese counterparts are all centered around the Icelandic IE2 end-member from Thirlwall et al. (2004), while the RFF lavas belong to the IE1 endmember. The depleted low-Ti lavas from the riftzone show only few examples of mixing between high and low-Ti magmas/sources and the low-Ti lavas can be modelled as mixtures between a MORB source and a component identical with the NAEM composition of Ellam and Stuart (2000). The proposed Icelandic depleted plume end-member ID1 (Thirlwall et al. 2004) does not seem to be present in the Paleogene lavas which is an argument against its existence.

References Ellam R.M., Stuart F.M. (2000), J. Petr. 41, 7, 919-932. Larsen L.M., Waagstein R., Pedersen A.K., Storey M. (1999), J. Geol. Soc. Lond. 156, 1081-1095. Storey M., Duncan R.A., Tegner C. (2007), Chemical Geology, in press. Thirlwall M.F. Gee M.A.M., Taylor, R.N., Murton B.J. (2004), Geochimica et Cosmochimica acta 68, 2, 361-386.

The ordinary chondrite breccias Adzhi-Bogdo (LL3-6) and Study Butte (H3-6) contain igneous-textured inclusions that are best described as alkali-granitoids in Adzhi-Bogdo (Bischoff et al., 1993) and andesite in Study Butte (Fredriksson et al., 1989). Both the granitoids and the andesite appear to have been formed by melting and magmatic differentiation on a parent body and indicate mixing of achondritic fragments with chondritic components. Al-Mg isotope data for these igneous-textured clasts reveal no evidence for radiogenic 26Mg and indicate that the formation of these igneous clasts, the incorporation into the parent body regolith, and the lithification must have occurred late, after almost all 26Al had decayed (Sokol et al., 2007). Oxygen isotope ratios of plagioclase, quartz and pyroxene in the fragments were measured in situ with the CRPG-CNRS Cameca IMS 1270 ion microprobe. On a three-O isotope diagram all fragments fall in the range of ordinary chondrites (Fig 1). These results imply that the fragments derive from an ordinary chondrite precursor and that the granitic fragments may have formed on the same parent body as the surrounding host rock material. The andesite seems to derive from a LL chondrite in contrast to its host rock (which is H3-6). This in turn indicates that melt formation and extreme differentiation occurred on ordinary chondrite parent bodies in the early stage of solar solar system formation. Alternatively, the fragments may have formed on another parent body but within the same oxygen isotope region of the solar nebula. In this case, they may represent projectile fragments within the chondritic breccias. 8 6

SMOW

Institute of Geography and Geology, University of Copenhagen, Øster Voldgade 10, DK-1350 Copenhagen, Denmark ([email protected])

Institut für Planetologie, Münster, Germany ([email protected]; [email protected]) 2 CRPG, Nancy, France ([email protected]) 3 Institut für Mineralogie, Münster, Germany ([email protected])

į17O

N. SØAGER AND P.M. HOLM

4 2 0

TFL

Adzhi-Bogdo Frag 5 Adzhi-Bogdo Frag 3

-2 CCAM

Adzhi-Bogdo Frag 4 Study Butte andesite

-4 -6 0

2

4

18

6

8

10

18 OSMOW į Od SMOW

References Bischoff A. et al. (1993), Meteoritics 28, 570-578. Fredriksson K. et al. (1989), Z. Naturforschung 44a, 945-962. Sokol A.K. et al. (2007), MAPS, submitted.

Goldschmidt Conference Abstracts 2007

New approaches to geochemical exploration for deep-seated and covered mineral deposits S.V. SOKOLOV, A.G. MARCHENKO, S.S. SHEVCHENKO, YU.V. MAKAROVA AND V.O. ILCHENKO

A953

Nisa granitic massif: SHRIMP zircon U-Pb age and source constraints A.R. SOLÁ1, I.S. WILLIAMS2, A.M.R. NEIVA3 1 AND M.L. RIBEIRO 1

Dep. Geologia, INETI, Portugal, ([email protected]) RSES, ANU, Australia, ([email protected]) 3 Dep. Cências da Terra, Univ Coimbra, Portugal, ([email protected]) 2

A.P.Karpinsky Russian Geological Research Institute (VSEGEI), St.-Petersburg, Russia ([email protected], [email protected], [email protected], [email protected], [email protected]) At present, because of the urgent need to carry out exploration for deep-seated mineral deposits, as well as deposits within areas covered by drifts, it is necessary to create and develop new geochemical techniques which are deeppenetrating ones and enable revelation of such concealed ore deposits. Routine geochemical prospecting surveys are not effective enough in such terrains due to low contrast or absence of geochemical anomalies related to ores. Among the deep-penetrating geochemical methods based on the phenomenon of jet-flow vertical migration of chemical elements from the deep to the surface resulting in superimposed dispersion halos formation, is the Method of Analysis of Superfine Fraction (MASF) developed in VSEGEI that uses extraction and analysis of superfine fraction of soils (<3-10 Pm) where superimposed dispersion halos occur. These halos are predominately created by the process of secondary fixation of mobile forms of elements due to the sorption of metals from the gaseous and water upward flows by clays, Fe and Mn hydroxides, and other natural substances. MASF surveys use sampling of definite horizons of soils and/or stream sediments, extraction of superfine fraction from samples by means of special technology, determination of contents of indicator elements using ICP-AES, ICP-MS, AAA with specific sample preparation, and geochemical data processing and interpretation with the help of original algorithms. Another perspective technique is the geochemical prospecting using water-extractable and weak-acid-extractable forms of chemical elements (mobile ions) from soils and stream sediments. Resultes of our survey carrried out in the Far East region has shown that most reliable prediction of gold mineralization can be distinguished by getting together data obtained by both mentioned deep-penetrating geochemical techniques.

The Nisa granitic massif crops out over an area of 1000 km2 in SW Iberia. It is a zoned batholith dominated by a rim consisting of very coarse-grained porphyritic two mica S-type monzogranite-syenogranite and a discontinuous core of very fine-grained I-type tonalite-granodiorite. To constrain the age relationships and petrologic processes responsible for this zonation, SHRIMP 206Pb/238U zircon ages were obtained for the monzogranite and tonalite. Zircons from the monzogranite are typical of granitic rocks and can be broadly classified into three texturally and chemically distinct types: 1) high-U, low Th/U outermost overgrowths (307.4 ± 4.0 Ma); 2) moderate U and Th/U zircon with concentric zoning occurring both as inner overgrowths and whole grains (305.4 ± 6.2 Ma) and 3) texturally discordant cores (309.0 ± 4.6 Ma and inherited). It was impossible to identify in advance, on any textural basis, which cores were ‘young’ or inherited. Despite textural and compositional contrasts the three “young” zircon types have mutually indistinguishable ages. Zircons 1) and 2) represent different stages of igneous zircon growth and zircon 3) must represent an earlier stage of growth. Either the protolith of the monzogranite contained some zircon slightly older than the monzogranite itself or zircon grew in two stages, separated by a period of zircon undersaturation. The former hypothesis seems to be unrealistic in the regional geological context. The latter would be possible if the magma was reheated soon after cooling to the point of zircon saturation. This is consistent with the dissolution features found in some of both older and younger cores. There is a very marked chemical contrast between zircons 1) and 2), as Th/U in 1) is almost 10x lower than in 2), which is compatible with saturation of monazite at a late stage of crystallization and/or the presence of U-rich fluid soon after the monzogranite was intruded. The inherited old cores fall broadly into Neoproterozoic, near concordant ages (506–661 Ma), and Paleoproterozoic and older, mostly discordant ages (1.85–2.55 Ga). There is a noticeable absence of Mesoproterozoic ages, which is significant in a regional geodynamic context. In contrast, zircons from the tonalite have banded zoning that is typical of zircon from mafic igneous rocks, and inherited cores were not found. Further, their Th/U is generally >1, higher than in zircon from the monzogranite. Their age, 306.2 ± 3.0 Ma, overlaps the ages of the three generations of zircon from the monzogranite, but zircon features suggest different sources for these two granitoids. The tonalite protolith might have been a more refractory level that melted soon after the crystallization of the “young” zircon cores from the monzogranite due to an increase in temperature (causing zircon dissolution). Tonalite in the core of the batholith probably intruded immediately after the dominant monzogranite rim.

A954

Goldschmidt Conference Abstracts 2007

Land ocean interactions in a coastal embayment, Kaneohe Bay, Hawaii: Nutrient dynamics, productivity, and CO2 exchange between seawater and atmosphere R. F. SOLOMON1, C.E. OSTRANDER1, E. HEINEN DE CARLO1 , M.A. MCMANUS1, F.T. MACKENZIE1, K.E. FAGAN, C. SABINE2 2 AND R.A. FEELY 1

Department of Oceanography, University of Hawaii, Honolulu, HI 96822, USA; ([email protected]) 2 NOAA PMEL, 7600 Sand Point Way NE, Seattle, WA 98115 USA Human activities throughout the past two centuries have generated large increases in the atmospheric content of greenhouse gases, leading to higher global mean surface temperatures. The warming of oceans may increase the frequency of storms, which facilitate the transfer of nutrients, sediment, and pollutants from rivers into coastal ecosystems. Increased oceanic CO2 concentrations attributable to anthropogenic input lower the saturation state of seawater with respect to carbonate minerals, can cause “ocean acidification” and have been argued by some to negatively impact calcification (e.g., Kleypas, 1999; Orr, 2005). Coastal areas and estuaries, however, may be either net annual sources or sinks of atmospheric CO2 (-41 to 7.3 Mole C m-2 yr-1, Mackenzie and Lerman, 2006). We present here results from 18 months of observations at CRIMP-CO2, a collaborative effort in Kaneohe Bay, Hawaii between UH Manoa and NOAA/PMEL. This buoy was the first coastal buoy of the NOAA/PMEL-CO2 program. CRIMPCO2 has documented the response of bay waters to pulsed inputs throughout a La Niña winter season (2005-06) and the much drier winter of 2006-07. The evolution of bay waters following storm-derived inputs was studied, along with the impacts of blooms and physical forcing on the air-sea exchange of CO2. Physical forcing strongly influences system response, in particular stratification and mixing, hence controls both the duration of blooms and attendant changes in CO2 concentration. Southern Kaneohe Bay often becomes a CO2 sink following storm inputs (0.2-0.7 m Mole C m-2 hr-1), but remained a net source of CO2 to the atmosphere (-1.06 Mole C m-2 yr-1) throughout our study period. This result is similar to estimates from Hog Reef flat in Bermuda and from the Scheldt Estuary plume (-1.2 and –1.1 to –1.9 Mole C m-2 yr-1, respectively, Mackenzie and Lerman, 2006).

Diamonds, xenoliths and kimberlites: A window into the Earth’s Mantle. UNESCO IGCP 557 HOLGER SOMMER1, KLAUS REGENAUER LIEB2 AND CHRISTOPH HAUZENBERGER3 1

Department of Geology, University of Botswana, Private Bag 0022, Gaborone, Botswana ([email protected]) 2 CSIRO Exploration & Mining, PO Box 1130, Bentley, WA 6102, Australia ([email protected]) 3 Institut für Erdwissenschaften, Universität Graz, Universitätsplatz 2/II, A-8010 Graz, Austria ([email protected]) The principal aim of this study is to develop a dynamic process based understanding for the formation of the e- and ptype diamonds. A joint study of the petrology and geodynamics of diamond bearing rocks is thus understood as a space-time window for unravelling deep processes in the Earth’s Upper Mantle. Our assessment aims particularly at the migration paths of hydrogen in nominally OH-free minerals in xenoliths, their geological/geodynamic setting as well as their petrological evolution. Recently, a 3-D dimensional modelling tool for understanding the geometry of a subduction zones was developed (Morra, et al., 2006). The new method comprises a novel Finite-Element/Boundary Element method (FEM/BEM) coupled to a thermodynamic solver (PERPLEX). The primary target of the code was to give a quantitative tool for modelling mantle tomography and geodynamics using the constraint of phase equilibria. Although the simulations are mainly geared at subduction/collision environments, both methods can be used in a much broader sense and thus provide a unique opportunity to test geodynamic settings for diamond formation in a truly quantitative manner. The strength of the FEM-BEM method of Morra and Regenauer-Lieb (2006), lies in the fact that the physical parameters derived from thermodynamic calculations provide a robust tool for predicting a geodynamic processes. In our approach, density contrasts and mineral chemical compositions are used for driving geodynamic processes occurring in a subduction zone. Material parameters, are equally derived and complemented by the best of our knowledge on rheological properties of rocks from laboratory data. We propose to compare the predicted phases with in situ observations using multi-scale laboratory analyses complemented by a high resolution synchrotron based FT-IR analysis to derive passage of aqueous fluids through the lithosphere. This latter technique has proven to be very powerful to map the formation of mineral inclusions in mantle minerals such olivine. The project is carried out under the auspices of an international UNESCO IGCP557 collaboration involving a large group of international scientists.

References Morra, G., K. Regenauer-Lieb and D. Giardini, 2006, Geology. 34(10), 877-880. Morra and Regenauer-Lieb, 2006, Phil. Mag. 86(21-22), 3307-3323.

Goldschmidt Conference Abstracts 2007

High sea bed methane emission rates at Hikurangi margin (New Zealand) associated with extremely dense populations of ampharetid polychaetes 1

1

1

2

S. SOMMER , P. LINKE , O. PFANNKUCHE , D. BOWDEN , J. GREINERT3 AND M. HAECKEL1 1

Leibniz Institute of Marine Sciences, IFM-GEOMAR, Kiel, Germany, ([email protected], [email protected], [email protected], [email protected]) 2 National Institute of Water & Atmospheric Research, Wellington, New Zealand ([email protected]) 3 Institute of Geological & Nuclear Sciences Ltd, Lower Hutt, New Zealand, ([email protected]) Occurrence of gas hydrates and gas seepage from Hikurangi margin sediments has been inferred from BSR structures, methane derived carbonates, gas flares in the water column, and the presence of chemosynthetic bivalve mollusks. Apart from these geophysical indications and sporadic observations by fishermen and scientific dredge samples, detailed biogeochemical studies are missing. We determined the in situ sea bed methane emission and associated fluxes of oxygen, nitrate, sulfate and sulfide at 3 different locations along the Hikurangi margin in water depths ranging from 662 to 1104 m using GEOMAR lander technology. Highest sea bed methane emissions of up to 203 mmol m-2 d-1 were associated with extremely high abundances of ampharetid polychaetes (10320 ind. m-2), highly elevated total oxygen uptake rates (up to 98 mmol m-2 d-1) and steep pore water gradients of methane, oxygen, sulfate and sulfide. Although members of the family Ampharetidae have been also reported from other cold seep sites and whale falls, this is to the best of our knowledge the first record where these polychaetes constitute key organisms of a cold seep environment. The distribution of the ampharetids was restricted to spatially highly confined patches of darkened sediment at the fringe of extended carbonate concretions. Close to these sites expulsion of free gas from the sea bed has been observed. They live in tubes of organic material that penetrate about 3 cm deep into the sediment. Density and spatial arrangement of their tubes, that extend a few millimeters into the water column, likely affects the current regime of the bottom contact water and therewith physical sediment properties and in consequence the exchange of solutes across the sediment-water interface. Micro gradients of oxygen and sulfide indicate further solute transport along the tubes. Both mechanisms might largely affect methane carbon turnover and release from this particular environment.

A955

Sequential oxidation of arsenite by both permanganate and the reaction byproduct K.S. SONG1,2, G. LEE 1, S.Y. KIM2 AND C.H. EUM2 1

Department of Earth System Science, Yonsei University, Seoul, Korea ([email protected]) 2 KIGAM, Daejeon 305-350, Korea ([email protected], [email protected], [email protected]) Arsenic contamination of soil, surface- and groundwater is of serious concern over the world because of its toxicity and carcinogenicity. One of the most effective ways to remove As from water is to oxidize it to less mobile and less toxic As(V), which can be subsequently immobilized by sorption to various sorbents. This study investigates the effects of permanganate on the oxidation of As(III) at pH 4 and 8 at varying doses of As(III) (10 or 100 PM) and permanganate (1 – 170 PM) in 0.01 M NaNO3. The solution pH was adjusted to the desired values using 10 mM of either monobasic or dibasic phosphate buffer with 10 mM HNO3. The oxidation of As(III) by permanganate at the stoichiometric ratio of As(III) to Mn(VII) (i.e., 3:2 for As(III) oxidation to As(V) with an assumption of Mn(VII) reduction to Mn(IV)-oxide) or at excessive Mn(VII) concentrations is instantaneous at both pH’s. In addition, the ratio of oxidized As(III) to reduced Mn(VII) is ca. 2.1, indicating the reaction product is probably Mn(III)-oxide/oxyhydroxide. When As(III) dose exceeds Mn(VII) concentration at pH 8, the excess As(III) is further oxidized by the primary reaction product, which is thereby reduced to dissolved Mn(II). This subsequent heterogeneous reaction between the excess of As(III) and the manganese solid byproduct is slower than the primary reaction. These results indicate that permanganate may be an effective and efficient reagent for As(III) oxidation in water treatment processes.

Goldschmidt Conference Abstracts 2007

A956

Study on 3-D crustal structure in the area along Yangtze River: The significance to multi-metal mineralization

Historical variations in zinc stable isotope compositions of smelter polluted sediments JEROEN E. SONKE1, YANN SIVRY1, JEROME VIERS1, STEPHANE AUDRY1, LEON DEJONGHE2,, LUC ANDRE3, JUGDEEP AGGARWAL4, JÖRG SCHÄFER5, GERARD BLANC5 AND BERNARD DUPRE1

S.Y. SONG AND E.G. GAO CAS Key Laboratory of Crust-Mantle Materials and Environments, School of Earth and Space Sciences, University Science and Technology of China, Hefei 230026, China ([email protected])

Introduction The complex structure characteristics of the area along the Yangtze River in Anhui Province is located in the convergent collision belt formed by two big tectonic plates, which are North China Plate and Yangtze plate. Plenty of mineral sources were generated in the duration of the frequent magmatic activities in this area. Therefore, it has a significant meaning in the exploring mineral sources to get a comprehensive view of the 3-D crustal structure. Based on the inversion and integrated interpretation of six geophysical profiles across this area, we obtain the 3-D characteristics about the crustal structure in Anhui Province along Yangtze River.

Data Processing and Results Taking the crustal structure of Yangtze Plate as the frame, and combining with the geological strata or tectonic units, the initial model containing the different blocks was designed. Taking the exeisting partially seismic profile interpretation as the constraint condition, gravity and magnetic robust iterative inversion and the 3-D crustal structure integrated interpretation are carried out in the six geophysical profiles across the entire area. Several conclusions were obtained as follows: (1) The crustal structure has the obvious three layers overally, namely the upper crust, the mid-crust and the lower crust. (2) The upper crust, with a violent changing structure, has deposited the relative thicker cap rock. The partial area may hit 10km, with the character of the low velocity, low resistivity, low density and low magnetism (V: 5.7~5.8km/s, ȡs: 5~10ȍm, ı: 2.6×103kg/m3, J: 10-2~10-1 A/m). (3) The mid crust has the top depth generally below 10 km and the bottom depth generally about 20km, with the characteristics of high velocity, high resistivity, high density and high magnetism (V: 6.0~7.0 km/s, ȡs: 1000~1500ȍm, ı: 2.9~3.0×103kg/m3, J: n×A/m), differently from the upper crust. (4) The lower crust, with an average of 12km depth, has the character of high velocity, high resistivity and high density (V: 6.8~7.6km/s, ȡs: 103~104ȍm, ı: 3.0~3.2 kg/m3). The Curie surface is in this layer. This study has siginificance to multimetal mineralization.

Acknowledgment This study are supported by the National Natural Science Foundation of China (40674071) and Foundation of Hubei Provincal Theoretical-Technological Key Lab for Oil-Natural Gas Exploration and Development (YQ2006KF11).

1

LMTG, Toulouse, France ([email protected]) Geological Survey of Belgium, RBINS, Brussels, Belgium 3 Royal Museum of Central Africa , Tervuren, Belgium 4 Department of Earth Sciences, UC Santa Cruz, USA 5 Universitee Bordeaux I, Talence, France 2

Two case studies were carried out to investigate the use of zinc stable isotopes as tracers for industrial smelter sources and processes: 1) two organic sediment cores obtained 800 m from the former pyrometallurgical zinc smelter of Lommel (Belgium), which received exclusively atmospheric inputs, and 2) two dam lake sediment cores taken 15km upstream and 30km downstream of the former pyro/hydro-metallurgical zinc plant of Viviez (France), which received predominantly riverine dissolved Zn and particulate Zn inputs. Zinc isotopic compositions were measured on the LMTG Thermo-Finnigan Neptune MC-ICPMS in Toulouse and expressed as per mille deviations from the JMC 3-0749L standard. In case 1 atmospheric pre-industrial deposits have į66Zn = +0.31±0.09 ‰ (2sd). Deposits dated from 1900-1930 have į66Zn = +0.29 ± 0.06 ‰ (2sd), and sediments dated from 1956-1995 shift to lighter isotopic compositions of į66Zn = +0.13±0.08 ‰ (2sd) in 1968. 42 ZnS minerals from ore import dominating Australian and African mines yield, together with literture ZnS data, a grand average of į66Zn = +0.16±0.07 ‰ 2se, n=83 for ZnS. Emission control since 1955 is a likely cause for the į66Zn sediment shift. In case 2 the polluted riverine sediments dated from 1952-2002 have elevated į66Zn of +0.75 to +1.32 ‰ relative to the geochemical background į66Zn = 0.33±0.06 ‰ (2sd). Mine tailing slag samples also had elevated į66Zn ranging from +0.18 to +1.49 ‰. In summary we show that 1. Bulk ZnS ore minerals have homogeneous į66Zn. 2. Zinc refinery processes fractionate Zn isotopes: slags are enriched in heavy isotopes. 3. near-field (<1km) atmospheric deposition resembles Zns ores, far-field carries isotopically light Zn, and mine tailing drainage carries heavy Zn.

Goldschmidt Conference Abstracts 2007

A957

Diffusion profiles of Li in plagioclase/clinopyroxene and plagioclase/olivine intergrowths

Jadeitite, lawsonite eclogite, and related rocks, Guatemala: Fluid-rock histories from a cold subduction zone

I. SONNTAG1, T. LUDWIG2 AND R. ALTHERR3

S. S. SORENSEN1, V. B. SISSON2, G. E. HARLOW2, H.G. AVÉ LALLEMANT3 AND T. TSUJIMORI4

123

Mineralogisches Institut, Universität Heidelberg, INF 236, ([email protected])

Diffusion in magmatic systems occurs during crystal growth, magma mixing and ascent. In some magmatic crystals diffusion profiles are preserved. These profiles provide information about the timescales of magmatic processes. It is known that Li diffuses very rapidly in plagioclase and clinopyroxene (Giletti and Shanahan 1997, Coogan et al. 2005, Parkinson et al. 2007) and that Li diffuses in plagioclase two to three times faster than in clinopyroxene (Coogan et al. 2005) This study is focused on the diffusion of Li between intergrown plagioclase/clinopyroxene and plagioclase/olivine. These intergrowths are observed in andesitic and dacitic rocks from the volcanic island Nisyros (Greece). Plagioclase phenocrysts in the andesitic lavas are cloudy zoned with slightly higher An content in the cores, while olivine and clinopyroxene phenocrysts are not zoned. In the dacitic lavas plagioclase phenocysts have complexely zoned (‘splotchy’) cores and low An rims, while clinopyroxene phenocrysts show nearly no zoning. Both intergrowths of phenocrysts were analysed for Li concentration. Concentration profiles across the minerals were measured using a Cameca ims3f ionprobe with ~ 5 Pm lateral resolution. Li concentration in the dacitic sample jumps at the border of the intergrowth from 10 μg/g in the plagioclase to 44 μg/g in the clinopyroxene and then drops steadily over a distance of 100 μm to 5 μg/g in the clinopyroxene core. Similar patterns are observed for plagioclase/clinopyroxene pairs in other dacitic and andesitic samples. In intergrowths of plagioclase and olivine in andesitic samples, the same Li increase on the border of the crystals is found ( 7 μg/g in the plagioclase to 13 μg/g in the olivine rim and a slow decrease to 5 Pg/g in the core). Preliminary Li isotope measurements reveal negative G7Li values (~ –26‰) in clinopyroxene rims to zero values in the core. Isotope and concentration profiles are estimated to be caused by diffusion at ~ 1000°C and on timescale of only hours.

References Coogan L.A., Kasemann S.A. and Chakraborty S., (2005), Earth Planet. Sci. Lett. 240, 415-424. Giletti B.J. and Shanahan T.M., (1997), Chem. Geol. 139, 320 Parkinson, I.J., Hammond S.J., James R.H. and Rogers N.W. (2007), Earth Planet. Sci. Lett. (in press), doi:10.1016/j.epsl.2007.03.023

1

Department of Mineral Sciences, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. ([email protected]) 2 Department of Earth and Planetary Sciences, American Museum of Natural History, New York, NY 10024, U.S.A. ([email protected], [email protected]) 3 Department of Earth Science, Rice University, Houston, TX 77005, U.S.A. ([email protected]) 4 Institute for Study of the Earth's Interior, Okayama University, Misasa, Tottori 682-0193, Japan ([email protected]) Jadeitite (jadeite rock) occurs rarely in a few serpentinite bodies from worldwide subduction complexes. Geochemical and CL-textural data show that much jadeitite crystallizes directly from multiply sourced Na-, Al, and Si-bearing fluids. Both jadeitite and lawsonite eclogite occur together in a few serpentinite-matrix mélanges of Guatemala. Also a rare rock, lawsonite eclogite forms under wet, extreme P/T conditions. Each rock type shows complex fluid-rock interactions. Relationships of jadeitite-forming fluids and host rocks are all but unknown—such contacts are rarely preserved. In the Sierra de las Minas, a 3 m-wide pit exposes such a contact, with jadeitite and serpentinite respectively altered to albitite and meta-ultramafic rocks. At the paleocontact, Zr, U, Hf, Pb, Ba, Sr, Y and Cs are greatly enriched in albitite relative to the other rocks. Enrichments coincide with the appearance of and/or increase in abundance and/or grain size of zircon, titanite, celsian and REE-rich epidote in albitite. All contain albite inclusions—many appear poikiloblastic—suggesting nucleation/growth of “trace-element-rich grains in albitite. Lawsonite eclogite records another fluid-rock system. For example, sample 2-14 (Jalapa dept.) is LREE-rich, with La 50× and Sm 30× Chondrite. It contains lawsonite grains as: 1) 50-100μm, subidiomorphic inclusions in cores of 3-5 mm garnet1; 2) 200-400 μm, irregular grains in garnet1 rims; 3) 100-300 μm idiomorphic matrix, partly consumed by amoeboid titanite; 4) 100-300 μm, idiomorphic matrix, with 100-200 μm garnet2; and 5) 300-600 μm idiomorphic late veins. Each textural type shows distinct LREE abundances and fractionations, as well as Sr contents and zoning (all by LA-ICPMS). Mass balance suggests only lawsonite (1) reflects “protolith” LREE values. The others manifest LREE redistribution, or deposition of exotic REE. These data show that Guatemalan serpentinite-matrix mélanges yield field examples of the mobility of “immobile” elements under low-T (300-450oC), high to very high-P (~8~23 kbar) conditions, in chemically distinct fluids.

A958

Goldschmidt Conference Abstracts 2007

Improved in situ measurements of lead isotopes in silicate glasses by LA-MC-ICPMS using multiple ion counters A.K. SOUDERS AND P.J. SYLVESTER Micro-Analytical Facility, INCO Innovation Centre and Department of Earth Sciences, Memorial University, St. John’s, NL A1B 3X5, Canada ([email protected]) A new technique that improves the spatial resolution and detection limits of the measurement of lead isotope ratios in silicate glasses with < 15 ppm total Pb by laser ablationmulticollector magnetic sector-inductively coupled plasma mass spectrometry (LA-MC-ICPMS) is presented. The method allows for the concurrent, static measurement of 204Pb, 206 Pb, 207Pb, 208Pb, along with 202Hg and 235U, in six Multi-Ion Counters (MICs) fitted on a Finnigan NEPTUNE MC-ICPMS. Use of a collector array consisting only of MICs eliminates the need for cross calibration between Faraday cups and ion counters, as employed in previous methods reporting 204Pb values by LA-MC-ICPMS. Standard-sample-standard bracketing using BCR-2G as the calibrant is used to correct for instrumental mass bias. Accuracy and precision of the method was evaluated by replicate analyses of various MPIDING reference glasses, with low Pb concentrations (1.38 to 11.6 ppm total Pb) and well-determined isotopic ratios. Typical spot sizes for in situ analyses ranged from 40-69 microns, providing better spatial resolution than previous LAMC-ICPMS reporting 204Pb. Ablations for all analyses were carried out using a 193 nm ArF GeoLas laser at a repetition rate of 10 Hz and a pulse energy of 5 J/cm2. Mercury derived from argon gas is a chronic problem for ICPMS measurements of the minor 204Pb isotope because of the isobaric inteference by 204Hg. Due to the high sensitivity of the MICs, the Hgcorrection of the measured 204-mass was significant, especially for samples with < 5 ppm total Pb. Two different methods were used to correct for Hg on the 204-mass with the results agreeing within error for each method on all lead isotope ratios. Measured lead isotope ratios for the MPIDING reference glasses T1G (11.6 ppm Pb) and ATHO (5.67 ppm Pb) agree within 0.10% and 0.16% respectively of the accpeted values. For MPI-DING KL2G (2.07 ppm Pb) and ML3B (1.38 ppm Pb), measured Pb ratios involving 204Pb agree within 1% of the accepted values with typical precisions of < 2.9 % RSD (2 sigma). Measured 208Pb/206Pb and 207 Pb/206Pb ratios for KL2G and ML3B are within 0.40% of the accepted values and typical precisions are < 0.75% RSD (2 sigma). The results for KL2G and ML3B demonstrate improvement over previous LA-MC-ICP-MS data in terms of both detection limits and spatial resolution, while retaining similar levels of accuracy and precision. The new method provides the capability of making quantitative in situ lead isotope measurements on tiny objects of geologic interest such as mineral growth bands, melt inclusions, and accessory minerals, even where they are lead poor.

Mineral phase identification of coral skeletal microstructure K. SOWA, T. WATANABE, Y. SETO, S. MOTAI AND T. NAGAI Dept. Natural History Sciences, Graduate School of Science, Hokkaido University. ([email protected]) Mineral phase identification of the coral skeletal microstructure is important for reconstructing the paleoclimate. Especially, Porites sp. is widely used a paleo-climatic proxy. To reconstructing the paleo-environment, those chemical compositions have been studied as coral thermometry. For example, Sr/Ca ratio in coral skeletons is sensitive to the paleo-environment variant so that Sr/Ca ratio has the possibility as the accuracy archive. However, there are one assumption is that trace elements in coral aragonite originate from solid solution. Greegor et al. (1997) reported that the as much as 40% of strontium (Sr) in coral aragonite existed as a strontianite which was not the solid solution by using X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS). On the other hand, Finch et al. (2003a, 2003) and Allison et al. (2005) showed no evidence of the presence of strontianite or it’s intermediate state in coral aragonite. They attempted to seek out the strontianite in coral samples using EXAFS. However, XANES and EXAFS could not identify the mineral phase directly. In addition, few studies have referred to mineral phase in coral skeletons with micro scale. X-ray diffraction analysis (XRD) with synchrotron radiation plays a role in detecting the mineral phase identification of coral skeletal microstructure for their strong energy, which enables phase identification with small range of X-ray spots: X-ray spot sizes of this study are 15μm and 40μm. We performed XRD analysis with synchrotron radiation for coral, Porites sp., living/fossil microstructure. The experiment conducted with imaging plate (IP) on BL-18c at Photon factory, KEK (HIGH ENERGY ACCELERATOR RESEARCH ORGANIZATION), Tsukuba, Japan. A monochromatic incident X-ray beam with a wavelength of 0.616Å was used and was collimated to a diameter of 15μm or 40μm. The two-dimensional IP data were integrated and conducted to one dimensional intensity data. The size of the centre of calcification (COC) is approximately 30μm in diameter in the samples of this study. That size correspondents to the X-ray spot sizes of this study in 15 or 40μm in diameter, so we got the different mineral structural information between the COC and fibres. All diffraction peaks of COC and fibres can be explained as aragonite, and no significant difference of cell parameters can be observed. This indicates that COC and fibres consist of aragonite: this study could not detect calcite phase.

Reference Greegor R. B., Pingitore N. E. Jr., and Lytle F. W., (1997). Science. 275, 1452–1454.

Goldschmidt Conference Abstracts 2007

Survival times of anomalous melt inclusions; Constraints from REE diffusion in olivine and chromite

Biogeochemical insight on the origin of carbonaceous matter in metalliferous lowest Cambrian black shale, South China

C. SPANDLER1, H. ST.C. O’NEILL2 AND V. S. KAMENETSKY3 1

Institute of Geological Sciences, University of Bern, Bern 3012, Switzerland: [email protected] 2 Research School of Earth Sciences, Australian National University, Canberra, 0200, Australia 3 ARC Centre of Excellence in Ore Deposits and School of Earth Sciences, University of Tasmania, Hobart, 7001, Australia It is widely assumed that incompatible element diffusion through olivine and chromite is very slow and hence melt inclusions (MI) hosted in these minerals do not experience diffusive re-equilibration with the external magma. However, at present there are no data on diffusion coefficients for most trace elements in olivine or chromite. We have conducted experiments at one atmosphere to determine diffusion coefficients for REE in forsteritic olivine and chromite. Clean samples of olivine and chromite that contain well-characterized MI suites were annealed with REEdoped (Pr, Ho, Tb, Lu) synthetic melts for up to 25 days at 1300-1450ºC and under controlled fO2. Diffusion profiles were measured across sections of the crystal/synthetic melt boundary by laser-ablation ICP-MS and by electron microprobe. Element concentrations were then fitted to the diffusion equation to obtain diffusion coefficients. Calculated diffusivities for REE are relatively fast (D = 10-15 m2/s at 1300ºC), whereas P and Al diffusion appears to be very slow (D <10-18 m2/s at 1300ºC). MI in the olivine and chromite crystals have distinct enrichments in the REE that were doped in the external melts. The systematic degree of enrichment of Lu>Ho>Tb>Pr and correlation between degree of enrichment and MI size and/or experiment duration are all consistent with re-equilibration of these inclusions via lattice diffusion through the host crystal. Applying our diffusion coefficients to the equations of Qin et al. [1], we calculate that the REE compositions of olivineor chromite-hosted MI will completely re-equilibrate with external magma in years to decades. These timescales are consistent with the REE enrichment observed in the experimental MI, and are significantly shorter than the times estimated for magma extraction from the mantle or residence in the lower crust. Therefore, anomalous MI must be trapped in the upper crust shortly before eruption. Our results show that the assumption of chemical isolation of incompatible elements in olivine and chromite-hosted MI is not valid, and hence calls for re-evaluation of the popular interpretation that anomalous MI represent preserved samples of unmodified mantle melts.

Reference [1] Qin Z., Lu F., Anderson Jr. A.T. (1992) Am. Min. 77, 565576.

A959

J.E. SPANGENBERG1 AND H.E. FRIMMEL2 1

Institute of Mineralogy and Geochemistry, University of Lausanne, BFSH-2, CH-1015 Lausanne, Switzerland, ([email protected]) 2 Institute of Mineralogy, University of Würzburg, Germany, ([email protected]) Early Cambrian black shale deposits in the Yangtze Platform in South China are, in places, strongly enriched in Ba (~470 Mt barite and witherite), Ni, Mo, V, Co, Cr, Au, U, As, Pb, Zn, Cu, Re, and PGE. Several models have been put forward for explaining the extreme metal enrichment of the ~0.50-0.52 Ga old black shale. These include enrichment by diagenetic fluids, hydrothermal exhalation, extraterrestrial impact, synsedimentary enrichment from seawater, and discharge of petroleum into the basin. This communication presents the first results of a molecular and isotopic organic geochemical study of the metalliferous carbonaceous shale. The new biogeochemical data provide further insight into the source of the hydrocarbons and the formation of the organic matter-metal association. The TOC content of the analysed samples is between 0.7 and 8.2 wt%. The Rock-Eval parameters were not reliable in most samples because of the low S1 and S2 peaks. The massive aspect of the samples without altered surfaces and coating of metal oxyhydroxides suggest indigeneity of the organic extracts. The G13C and G15N ratios of the kerogen range respectively from –31.5 to –35.4‰ V-PDB and from –1.4 to 0.6‰ V-Air. The G34S values of the organic-solvent soluble molecular sulphur of black shale samples range between 13.9 and 15.8‰ V-CDT, similar to the values obtained for sulphides. The distribution of saturated HC are characterized by: nalkanes in the C11–C31 range (maximum at C16, bimodal distribution with further maxima at C22 or C26 in some samples, slight even-over-odd dominance), moderate to large i-C18 to i-C21 isoprenoid peaks, pronounced unresolved naphthenic humps in the n-C15-22 range, generally pristane and phytane in roughly equal proportion, and alkylcyclohexanes in the range C11 – C28. Some bitumens contain C27-C35 hopanes and C27-C29 steranes, with the latter dominating over the former. Aromatic HC such as alkylbenzenes, naphthalene, and alkylnaphthalenes were detected only in very minor concentrations. The G13C values of n-alkanes and isoprenoids vary between –33 and –25‰. In all samples pristane and phytane are depleted in 13C by up to 6‰ compared to C17 and C18 n-alkanes. All these results are best explained by derivation of hydrocarbons from algal and bacterial mats that were deposited in a saline, anoxic marine environment.

A960

Goldschmidt Conference Abstracts 2007

Chlorine Partitioning: The behavior of Cl in the presence of sulfide silicate melts and aqueous fluid HEATHER A. SPARKS1, JOHN MAVROGENES12, JEREMY WYKES2 AND JÖRG HERMANN1

Planktonic foraminifera: Calcifying microenvironments, diffusive boundary layers and a peek at the event horizon H. J. SPERO

1

Research School of Earth Sciences, Australian National University, Canberra ACT 0200 Australia ([email protected]) 2 Department of Earth and Marine Sciences, Australian National University, Canberra ACT 0200 Australia Can sulfide melts exsolve halogen rich fluids? Several ore deposits around the world (eg. Broken Hill, Stillwater, Sudbury and Bushveld) have occurrences of halogen-rich minerals in association with the ore. To help better understand these occurrences, piston cylinder experiments were performed to investigate the partitioning behavior of Cl between coexisting haplogranite and Pb-Fe-Zn-sulfide melts at 0.5 GPa pressure and 810ºC temperature. In water undersaturated experiments where a Cl doped haplogranite glass was used as the silicate starting material, the Cl was found to partition strongly into the sulfide melt. Here Cl may be acting as a flux within the sulfide melt resulting in lowering its eutectic temperature. Not only is Cl dissolved in the sulfide melt, but as the sulfides crystallize the residual sulfide melt progressively enriches in Cl. In water saturated experiments where Cl was doped into the sulfides, Cl was noticeably absent from both quenched sulfide melt and silicate glass, suggesting strong partitioning into the coexisting aqueous fluid. The results indicate that the partitioning preference of Cl, when in equilibrium with sulfide-silicate melts and aqueous fluid, decreases in the order aqueous fluid sulfide melt - silicate melt. These experimental results can be applied to understand the cooling of sulphur- and chlorine-bearing magmatic systems. Initially, Cl will partition into a sulfide melt that coexists with a hydrous silicate melt. Once an aqueous fluid exsolves from the silicate melt – due to decompression or crystallization at the solidus – Cl will partition into the newly formed fluid causing the sulfides to freeze and crystallize rapidly. The escaping Cl-rich fluid might lead to extensive halogen alteration in the country rocks of the intrusion. The concepts investigated in this study may have significant implications for understanding the evolution of magmas giving rise to copper porphyry deposits and aid in our understanding of halogen-alteration and halogen-rich minerals in major sulfide deposits.

Department of Geology, University of California Davis, Davis California, USA ([email protected]) The planktonic foraminifera, Orbulina universa, should be considered the proverbial white rat for researchers studying biomineralization in the marine realm. This protozoan produces a spherical test in a single brief (several hours) calcifying event, that involves organic matrix secretion, rapid calcite precipitation and spine elongation. Chamber thickening in subsequent days is intimately linked to diurnal physiological oscillations related to symbiont photosynthesis and host assemblage respiration which controls large shifts in microenvironment pH (7.8 – 8.8) and [O2] (~80-220% air saturation) (Rink, et al., 1998). Interestingly, microelectrode measurements of Ca2+ at the chamber surface show concentrations that are 10% higher than ambient seawater, while [CO2] measurements on O. universa under dark and light conditions suggest 6CO2 varies between ambient (~2 mMol/kg) and ~5 mMol/kg respectively in the boundary layer near the calcifying shell (Köhler-Rink and Kühl, 2005). Together these data point to an active calcium and carbon concentrating mechanism in this species, and may help explain how O. universa can calcify in seawater at pH between 7.4 and 8.8 (Bijma, et al., 1999). Laboratory experiments and stable isotope and trace/minor elemental analyses of O. universa demonstrate that shell geochemistry is influenced by physiologically-controlled boundary layer chemistry. Elemental ratio banding, possibly related to the diurnal cycle, as well as observations of nonequilibrium calcite precipitation, provide insight into aspects of calcification that were previously poorly constrained. Together, these observations may provide clues for novel applications of foraminifera geochemistry for paleoenvironmental reconstructions.

References Bijma, J., Spero, H.J. and Lea, D.W. (1999). in Use of Proxies in Paleoceanography: Examples from the South Atlantic, edited by G. Fischer and G. Wefer, pp. 489-512. Köhler-Rink, S., and Kühl, M. (2005). Mar. Biol. Res. 1. 6878. Rink, S., Kühl, M., Bijma, J. and Spero, H.J. (1998). Mar. Biol. 131. 583-595.

Goldschmidt Conference Abstracts 2007

Trace elements in garnets of diamondiferous xenoliths from the Nurbinskaya pipe, Yakutia Z. V. SPETSIUS1, W.L. GRIFFIN2, S.Y. O’REILLY2, 1 AND V.I. BANZERUCK 1

YaNIGP CNIGRI, ALROSA Co. Ltd., Mirny, Yakutia, 678170, Russia, ([email protected]) 2 GEMOC ARC National Key Centre, Macquarie University, NSW 2109, Australia, ([email protected]) Rare earth elements (REE) and other trace elements were analyzed by LAM-ICPMS in garnets of about 150 mafic and ultramafic diamondiferous xenoliths from the Nurbinskaya pipe. Most garnets are homogeneous in terms of major- and trace-element contents. Garnets from ultramafic xenoliths define two groups, one with sinusoidal REEN (chondrite normalised) patterns (10 harzburgites, two lherzolites) and one with flat MREEN (lherzolites, some websterites). Most eclogitic garnets have LREE-depleted patterns (CeN as low as 0.1), and no Eu anomalies. Heavy rare earth elements are variably enriched; most LuN varies 20-50. Most websteritic garnets show REE patterns similar to this but they are typically enriched in LREE with CeN (0.2-0.5). Garnets with nearly flat HREE and small positive Eu anomalies are common in coesite-bearing eclogites and those containing kyanite and/or corundum [1]. Another group of garnets (n=9) from eclogites and websterites have small negative Eu anomalies. These types of pattern commonly are interpreted as evidence of the reaction of plagioclase to garnet and used to support the origin of mantle eclogites by subduction of oceanic crust [1, 2], but may be simply a redox feature. Garnets from highly aluminous eclogites show convex REE patterns enrichment in LREE and strong depletion of in HREE (YbN<5). Garnets of corundum-bearing eclogites commonly have positive slopes within the LREEN, peaking at Sm and then slowly decreasing to about chondritic abundance for Lu. LAM-ICPMS analysis could be used to show how different populations within an eclogite xenolith series can document the heterogeneous evolution of the lithospheric mantle beneath cratonic areas.

References [1] Spetsius Z.V. (2004) Lithos 48, 525–538. [2] Jacob D. E., Foley S. F. (1999) Lithos 48, 317–336.

A961

Latest-stage exhumation history of the Central Alps C. SPIEGEL1, N. DÖRR1, M. RAHN2 AND M. DANIŠÍK1 1

Institute of Geosciences, University of Tübingen, Germany ([email protected], [email protected], [email protected]) 2 Swiss Federal Nuclear Safety Inspectorate, Villingen, Switzerland ([email protected]) The European Alps, like many other Cenozoic orogens, show a pronounced increase in erosion rages since ~5 Ma. In a recent paper, Willett et al. (2006) proposed that this accelerated erosion was due to enhanced precipitation subsequent to the Messinian salinity crisis and marked the transition from orogenic construction to orogenic destruction. In this study, the latest-stage exhumation history of the Central Alps was investigated by using zircon fission track (ZFT), apatite fission track (AFT) and apatite (U-Th)/He (AHe) thermochronology along the eastern margin of the Lepontine Dome. The study area covers basement nappes east and west of the Forcola fault, a major Alpine normal fault bordering the Lepontine Dome to the east. ZFT ages from both east and west of the Forcola fault range between 25 and 16 Ma, with youngest ages occurring close to the fault system. These ages reflect a period of enhanced exhumation related to the onset of orogen-parallel lateral extension of the Alps. The area east of the Forcola yielded AFT ages between 17 and 5 Ma and AHe ages between 7.5 and 6 Ma, whereas the area west of the Forcola yielded AFT ages between 9 and 4 Ma and AHe ages between 6.6 and 3.4 Ma. The age difference between basement nappes east and west of the Forcola normal fault indicates that the fault system has been active (or re-activated) during the Latest Miocene to Pliocene, much later than previously assumed. Late Neogene reactivation, however, has also been observed for the Simplon normal fault west of the Lepontine Dome, which is supposed to be conjugated to the Forcola fault. Ageelevation relationships reveal that the area west of the Forcola fault experienced a period of rapid exhumation between approximately 5.5 to 4 Ma, and that exhumation rates slowed down after ~4 Ma. The period of rapid exhumation is temporally consistent with the increase in deposition rates in the foreland basins, whereas the end of rapid exhumation coincided with the end of deformation in the Jura mountains. These temporal relations support the idea of a Pliocene shift from orogenic construction to orogenic destruction and thus to a decrease of the actively deforming area of the Alps.

Reference Willett S.D., Schlunegger F., and Picotti V. (2006), Geology 34/8 613-616.

Goldschmidt Conference Abstracts 2007

A962

Spatial record of recent anthropogenic changes in the sedimentary soils of the Netherlands; Opportunities for a knowledge-based soil legislation framework 1

J. SPIJKER , G. VAN DER VEER

2

3

AND G. MOL

1

National Institute of Environment and Health, P.O. Box 1, 3720 BA, Bilthoven, The Netherlands, ([email protected]) 2 Department of Earth Sciences, Universiteit Utrecht, P.O. Box 80.021, 3584 CD, Utrecht, The Netherlands, ([email protected]) 3 Soil Science Centre, Alterra, PO Box 47, 6700 AA, Wageningen, The Netherlands, ([email protected]) Much of our knowledge about recent environmental changes – such as the increased anthropogenic emission of various metals – has been inferred from sedimentary records. These paleorecords, however, mainly focus on environmental changes through time. To generalize such changes in a spatial context, geochemical baseline surveys are required. Using the data from the Geochemical Soil Survey of the Netherlands (Van der Veer, 2006), a geochemical baseline model was developed. The baseline model, based on covariability in pristine sediments, revealed a substantial overall enrichment of metals including Cd, Cu, Hg, Pb and Zn. These enrichments reflect the accumulation of metals in the topsoil compartment as a result of ongoing diffuse input from various anthropogenic sources. The enrichment, generally a factor 2-3 above natural concentrations, varies on a local as well as a regional scale. Besides tracking environmental changes in a spatial context, the model furthermore offers unique possibilities to derive soil quality standards for a knowledge-based soil legislation framework. In this presentation we will show how the model can be used to derive soil quality standards, and how to take the natural variation of soils, as well as the diffuse enrichment, into account.

Reference Van der Veer G. (2006). Geochemical soil survey of the Netherlands. Atlas of major and trace elements in topsoil and parent material; assessment of natural an anthropogenic enrichment factors. Netherlands Geographical Studies 347, 248 pp.

Bionergetics of the buried seafloor ARTHUR J SPIVACK Graduate School of Oceanography, URI, Narragansett, RI, USA, ([email protected]) Major questions in subsurface biosphere research are related to bioenergetics. How much energy is required to support a given amount of biomass? And what is the minimum energy yield of reactions that are biologically utilized? Beyond these, there are related issues, such as, the energetic rules that govern the distribution of subsurface life, in particular, the controls on the distribution and rates of the various energy producing metabolic reactions and the possibility that some buried marine ecosystems rely on radiolytic H2 as their principle electron donor. In situ metabolic rates and energy yields of diverse microbial activities in sediments of the eastern equatorial Pacific have been determined based on sedimentary pore fluid chemical profiles. Fe and SO4-2 reduction, and methanogenesis co-occur. These are energetically favorable throughout the sediment column with relatively constant energy yields. Based on this, minimum biologically utilizable energies of reaction can be inferred and it appears that this ecosystem operates as a thermodynamic homeostat. When combined with reaction rate estimates, based on a numerical solution to the diffusion/reaction continuity equation, these data allow the calculation of an average maintenance energy that is orders of magnitude lower than observed in the laboratory. Based on this, it is inferred that in marine sediments, radiolytic H2 could support approximately 105 while approximately 104 could be supported in a watersaturated sediment that has a U, Th, and K similar to that estimated for the Martian crust.

Goldschmidt Conference Abstracts 2007

Plate-tectonic controls on intraplate volcanism in New Zealand P. SPRUNG1, S. SCHUTH 1,2 AND C. MÜNKER1,2

The Podili alkaline complex, Prakasam alkaline province, Andhra Pradesh, southern India M. SRINIVAS1 AND A. G. S. REDDY2

1

Institut fuer Mineralogie, University of Muenster, Germany ([email protected]) 2 Mineralogisch-Petrologisches Institut, University of Bonn, Germany Cretaceous to recent intraplate volcanism in the New Zealand microcontinent provides an ideal case to explore possible factors that exert control on this type of volcanism. Located in the vicinity of the active Australian-Pacific plate boundary, the New Zealand microcontinent exhibits intraplate volcanic activity that is associated with various tectonic regimes: behind an active arc, in domains with predominant strike-slip motion, and in areas that are virtually unaffected by recent plate-tectonic activity. Based on major-, trace-element-, and Hf-Nd-Pb-Sr isotope compositions of a representative set of near-primitive samples, distinct end member compositions corresponding to different tectonic regimes can be recognized: (1) HIMU-like signatures (206Pb/204Pb up to 20.57, “decoupled” Hf-Nd systematics; e.g., Chatham Islands), (2) dilute HIMU-like trace element signatures and depleted, asthenospheric isotope compositions (HHf: + 9.9, HNd: + 7.0; North Island), and (3) compositions having affinities to subducted sediments (87Sr/86Sr: 0.7037, 206 Pb/204Pb: 18.99, 207Pb/204Pb: 15.67; South Island). The HIMU- and sediment-like signatures are interpreted to originate from a veined lithosperic mantle, reflecting a variable overprint by an ancient, possibly Cretaceous mantle plume and by Phanerozoic subduction zone enrichment. Variations in the average melting depths of the magmas are manifested in a variable impact of residual garnet (e.g., GdN/YbN: 1.84 - 4.74). Changes in relative melting depths reflect variations in lithospheric thickness [1]. The geodynamic setting (lithospheric thickness, heat flow, degree of extension) controls the proportions of asthenospheric and lithospheric source components. For the South Island, magma compositions provide “snapshots” of a decreasing lithospheric thickness beneath active volcanic fields since ~20 Ma. This finding is in agreement with recent geophysical data that suggest the presence of thickened, dense lithospheric mantle beneath the Southern Alps (collision zone) but not beneath the intraplate volcanic fields east of the plate boundary [2]. It has been proposed that the regional thickening of the lithospheric mantle is a result of ongoing oblique convergence and strikeslip motion [e.g., 2]. We propose that the same processes also caused regional transtension and thinning of adjoining parts of the lithospheric mantle, thus controlling the composition of the intraplate volcanic rocks.

References [1] Sprung P., Schuth S., Münker C. and Hoke L., (2007), Contrib. Mineral. Petrol. 153, 669-687. [2] Bourguignon S., Stern T. A. and Savage M. K., (2007), Geophys. J. Int. 168, 681–690.

A963

1

Department of Geology, Post-Graduate College of Science,Osmania University, Saifabad, Hyderabad500004, Andhra Pradesh, India 2 Central Ground Water Board, Southern Region, Hyderabad500 029, Andhra Pradesh, India The Podili alkaline complex in Prakasam alkaline province of Andhra Pradesh, in the southern India, is one among the cluster of alkaline complexes with a near NE-SW disposition pervading the cratonic corridor that is terminated by intracratonic sedimentary Cuddapah Basin in the west and Eastern Ghat Mobile belt in the east. This N-S trending complex (12 km2 ) has a close spatial and temporal association with granites and gabbros that represents the manifestations of basic, acidic and alkaline magmatism. Alkali syenite, and subordinate quartz syenite constitute the important lithounits of the complex. The syenites are leucocratic dominated by microcline mesoperthite, plagioclase feldspar, ±quartz. The mafic minerals are alkali pyriboles and biotite with subordinate magnetite, sphene and apatite. The mafic minerals attribute to the alkaline and hydrous nature of parental liquids. The syenites are of different degrees of silica saturation, and alkali syenites in particular carry normative nepheline and acmite, an indication of the peralkaline trait possibly inherited from the parental alkaline magma.either by prolongoued differentiation or lowest degress of partial melting of an enriched/fertile mantle source.

A964

Goldschmidt Conference Abstracts 2007

Soil contamination due to heavy metals from tannery industries: A case study of Jajmau (Kanpur) and Unnao industrial areas, Uttar Pradesh, India

New constraints on the origin of short-lived radioactive nuclides in the early solar system GOPALAN SRINIVASAN1, MARC CHAUSSIDON2* AND ADDI BISCHOFF3 1

S. SRINIVASA GOWD, N.N. MURTHY AND PRADIP K. GOVIL National Geophysical Research Institute (NGRI), Hyderabad, India Environmental geochemical studies were carried out in and around Jajmau (Kanpur) and Unnao industrial areas, to find out the extent of chemical pollution in soil due to waste disposal from tannery industries. There are more than 2500 tanneries in the country and nearly 80% of them are engaged in the chrome tanning process. In Uttar Pradesh, Jajmau (Kanpur) and Unnao (80o18’ – 80o30’ E longitude and 26o25’ – 26o34’ N latitude) are prominent centers for leather processing and there are two clusters of tannery industries (about 450) along the banks of river Ganga. Geologically the study area is covered by alluvium of Quaternary age consisting of older alluvium of middle to upper Pleistocene and newer alluvium of Holocene and the climate of the study area is semi-arid type. Fifty-three soil samples were collected from Jajmau (Kanpur) and Unnao industrial areas from top 10 cm layer of the soil and were analyzed for heavy metals by using Philips PW 2440 X-ray fluorescence spectrometer. The data reveals that the soil in this area is significantly contaminated, and shows very high concentrations of chromium ranging from 162 to 60819 mg/kg (14535 mg/kg average). Other heavy metals such as Ba ranges 44-781 mg/kg (295.7 mg/kg average), Cu 1.7-126 mg/kg (42.9 mg/kg), Pb 22-68 mg/kg (40.4 mg/kg average), Sr 47-151 mg/kg (105.3 mg/kg average), V 1.3-209 mg/kg (54.4 mg/kg average) and Zn 44688 mg/kg (159.9 mg/kg average). High concentrations of these toxic/heavy metals are contributors for the degradation of human health in the study area and people suffer from occupational diseases such as asthma, chromium ulcers and skin diseases. Distribution and correlation of heavy metals in soil along with possible remedial measures are discussed.

Department of Geology, University of Toronto, Toronto, ONM5S3BA1, Canada; ([email protected]) 2 CRPG-CNRS, BP20, 54501 Vandoeuvre-lès-Nancy, France; ([email protected]) 3 Institut für Planetology, Wilhelm-Klemm-Strass 10, 48149 Münster, Germany; ([email protected]) The presence of several short-lived radioactive nuclides (7Be, 10Be, 26Al, 36Cl, 41Ca, 53Mn, 60Fe) in the early solar system is established from the presence of their decay products in constituents of primitive meteorites such as Ca-, Al-rich inclusions (CAIs). These nuclides are either (i) the products of stellar nucleosynthesis (as demonstrated by the presence of 60Fe) and were injected in the protosolar cloud before or during its collapse or (ii) the result of interactions of energetic particles (as demonstrated by the presence of 10Be) with gas and dust either in the protosolar nebula or in the presolar cloud. As shown by X-ray observations of young stellar objects, one obvious source of an intense flux of accelerated particles in the protosolar nebula is the young active Sun. CAIs being the oldest solids formed in the solar system, they may have formed close to the young Sun and may contain a record of these irradiation processes. Understanding the origin of short-lived radioactive nuclides is thus fundamental not only for early solar system chronology but also for deciphering the astrophysical context of the formation of the first solids in the early solar system. We report Li, B and Mg isotopic analyses by ion microprobe (Cameca ims 1270) of a set of various CAIs from the CH chondrite Acfer 182 and CV3 chondrite Efremovka, including some hibonite-rich CAIs which because of their refractory composition are considered to be among the earliest CAIs. The hibonite-rich CAIs have lower 26Al/27Al ratios (<1.1±0.5x10-5) than the classical type B CAIs from Efremovka (e.g. CAI E66 which has a 26Al/27Al ratio of 5.49±0.15x10-5). The 10Be/9Be ratios are lower by a factor of two in hibonite-rich CAIs compared to Efremovka type B CAIs (e.g. 10Be/9Be=1.2±0.3x10-3 in E65). These data show that 26Al and 10Be are likely decoupled in the early solar system. The low 26Al/27Al ratios in hibonite-rich CAIs may reflect the steady state abundance of 26Al in the local interstellar medium. The hibonite-rich CAIs show a systematic slight but significant 6Li enrichment indicative of the presence of a component produced by spallation. This can be used to put a higher limit on the amount of 10Be which could have been produced in the presolar molecular cloud by trapping of galactic cosmic rays.

Goldschmidt Conference Abstracts 2007

Water structure and dynamics on aqueous barium ion and the {001} barite surface

Continental erosion averaged over space and time JANE W. STAIGER1 BRANDON MCELROY2 3 AND LESLEY PERG

A. G. STACK 1 AND J. R. RUSTAD 2 1

School of Earth and Atmospheric Sciences, Georgia Institute of Technology. ([email protected]) 2 Dept. of Geology, University of California-Davis. ([email protected]) Water structure and exchange kinetics are important to a number of fundamental geochemical processes. For example, the rate of dissolution of isostructural materials often correlates with the rate of water exchange on the aqueous cation. However, thus far most computational studies have focused on structure of the mineral surface. Water structure and kinetics are less often studied, yet are necessary to understand such basic properties of the interface such as adsorption energy. In this work, we use an existing molecular dynamics potential model [1] to estimate the water structure and exchange kinetics of aqueous barium ion the barite {001} surface. Water exchange kinetics were using a correlation function as well as the potential of mean force and a reactive flux. The latter methods involve calculating a transitition state theory rate constant through integrating a free energy of activation estimated from a radial distribution function and a transmission coefficient estimated via a reactive flux. It was found that using only the barium-oxygen distance to constrain the reaction coordinate resulted in unphysical results and a solvent organization parameter is necessary. Using these methods, water structure and exchange kinetics surrounding aqueous barium ion fall within the range of experimental estimates.[2] Water-barium distances for the surface ions on the {001} barite surface are very close to those of the aqueous ion, yet the rate constant is significantly faster. This increase in exchange rate is attributed to a relatively hydrophobic, high energy interface. The calculated water structure of the overall interface is relatively complicated, with up to five distinguishable oxygen positions interacting with both surface sulfates and bariums. This complex structure contrasts to the fit of experimental Xray reflectivity data, which can only justifiably include a single ordered layer of water that matches unsaturated bariumoxygen bonds.[3] It is unclear at this point how to reconcile of this discrepancy, but there are substantial uncertainties in both the experimental and computational water structures. This study is the first time the kinetics of water exchange on aqueous barium ion and barite surfaces have been examined using a reactive flux method and as such, represent a significant expansion of this new and hitherto relatively unexplored method to systems of geochemical interest.

References [1] Jang, Y. H. et al. (2002) J. Phys. Chem. B, 106, 9951. [2] Richens, D. R. (1997) The Chemistry of Aqua Ions, Wiley. [3] Fenter, P. et al. (2001) J. Phys. Chem. B, 105, 8112.

A965

1

National Center for Earth-surface Dynamics, Saint Anthony Falls Laboratory, University of Minnesota, Minneapolis, Minnesota, USA; ([email protected]) 2 Dept. of Geological Sciences, Jackson School of Geosciences, University of Texas, Austin, Texas, USA; ([email protected]) 3 Department of Geology and Geophysics, University of Minnesota, Minneapolis, Minnesota USA; ([email protected]) Matching rates of denudation at all temporal scales have been used to infer erosional steady state in various mountain belts. What variation in erosion rates is significant or expected from one temporal and spatial scale to another and over a range of tectonic regimes? Under ideal conditions, the cosmogenic nuclide flux out of a basin can be used to measure the denudation rates that average over millennia for the entire drainage area. Erosion rates measured over shorter timescales (e.g. sediment yield, reservoir infill) show greater variance due to the stochastic nature of erosion and surface process interactions. Short-term measure-ments miss rare events in a predictable way and because of this property, rates of surficial processes measured over increasingly longer time intervals can incorporate longer intervals of process inactivity, thus producing a different apparent rate of deposition and erosion (Sadler, 1981; Gardner et al., 1987). We use a new (available from the author) compilation of all previously published and new cosmogenic nuclide-derived denudation rates and sediment-flux measurements for the same basins to evaluate this variability. Despite anthropogenic effects, the majority of sediment fluxes are less than long-term rates of erosion, contrary to that predicted by Gardner et al. (1987).

References Gardner, T.W., Jorgensen, D.W. Shuman, C. and Lemieux, C.R. (1987). Geomorphic and tectonic process rates: Effects of measured time interval. Geology 15: 259-261. Sadler, P. M. (1981). Sediment accumulation and the completeness of stratigraphic sections. J. Geology 89: 569-584.

A966

Goldschmidt Conference Abstracts 2007

What controls Sulfur isotope fractionation in modern estuarine sediments? M.C. STAM1 , P.R.D. MASON1, A.M. LAVERMAN1, C. PALLUD2 AND P. VAN CAPPELLEN1 1

Department of Earth Sciences, Utrecht University, The Netherlands; ([email protected], [email protected], [email protected], [email protected]) 2 Department of Geological and Environmental Sciences, Stanford University, USA; ([email protected]) Sulfur isotopes are a promising tool for tracing sulfate reduction in sedimentary rocks, and for providing constraints on one of the oldest metabolic process on Earth. However, the relationship between isotopic fractionation and parameters such as sulfate reduction rate, temperature and availability of organic matter remains unclear, with conflicting results from pure culture and natural population studies [1,2,3]. Here, we use flow through reactors that contain undisturbed slices of sediment to measure sulfate reduction rate under quasi-steady state conditions. Flow through experiments were run to investigate the control of temperature, sediment depth, organic matter content, sulfate concentration and the effects of inhibitors on sulfate reduction rate and sulfur isotope fractionation. Samples were collected at a brackish location of a temperate estuarine sediment (Western Schelde, The Netherlands). Our results indicate an inverse relation between sulfate reduction rate (SRR) and sulphur isotope enrichment factor under optimum temperatures (20 and 30ºC). This trend disappears at low rates (<10 nmol cm-3h-1) and for nonoptimum temperatures (10 and 50ºC). Large fractionations (>20‰) were observed only at low SRR (<10 nmol cm-3 h-1). Sediment depth as well as organic matter content did not significantly affect isotope fractionation. This study demonstrates that the degree of isotope fractionation can be used to infer SRR in natural populations as well as in pure cultures providing that the bacterial population is thriving under optimal conditions.

References [1] Canfield D.E. (2001) GCA 65 (7), 1117-1124. [2] Detmers, J. et al. (2001). Applied and Environmental Microbiology 67(2): 888-894. [3] Habicht, K. S et al (1997) GCA 61(24): 5351-5361.

History of seafloor hydrothermal activity in the SW Pacific Bare Zone using fish teeth strontium isotope dating of metalliferous sediments A.M. STANCIN, J. D. GLEASON, R. M. OWEN, D. K. REA, AND J. D. BLUM Department of Geological Sciences, University of Michigan, Ann Arbor, MI 48109-1005 ([email protected]) A 2 million km2 region virtually devoid of sediment has been identified in the remote SW Pacific Basin (February/March 2005 drill site survey cruise - Rea et al., 2006). This region, informally termed the “South Pacific Bare Zone” comprises ocean floor dating back to the Late Cretaceous. Seismic profiling, piston cores and gravity cores reveal the full extent of barren crust – an area nearly the size of the Mediterranean Sea. Within the Bare Zone, a small (1km2) abyssal valley with 24 m of sediment was identified and sampled with a large diameter piston core, leading to recovery of 8.35 meters of metalliferous sediment at 5082 m water depth. Fish-teeth Sr-isotope stratigraphy reveals a continuous record of sedimentation 31Ma to present, with an average linear sedimentation rate at this site of 0.27 mm/kyr. However, the fish teeth age-depth profile and INAA geochemistry show an exponentially decreasing hydrothermal flux, with sedimentation rates approaching <0.05 mm/kyr between 17 Ma and the present. The origin of the main pulse of hydrothermal activity is uncertain, but may be related to a series of late Eocene/early Oligocene ridge jumps and propagating rifts that accompanied large-scale plate tectonic reorganization of South Pacific seafloor. The fish teeth Sr isotope age-depth profile and pelagic clay geochemistry also reveals that the terrigenous component at this site registers a very low eolian flux, increasing in proportion to the hydrothermal component upcore. Primary dust sources were likely Australia and New Zealand, consistent with Nd-Sr-Pb isotopes of detrital extracts. The unusual conditions of Cenozoic non-deposition that characterize this area of the South Pacific make this the first record of its kind, providing unique insight into hydrothermal activity and eolian sedimentation since the early Oligocene. The utility of the fish teeth Sr isotope method for dating marine hydrothermal cores should be explored further.

References Rea, D., and TUIM-03 Site Survey Scientific Party (2006) Geology 34, 873-876.

Goldschmidt Conference Abstracts 2007

A967

No role for discrete, depleted high 3 He/4He mantle

What controls iron isotope fractionation in an acid mining pile?

N. STARKEY1,2, F.M. STUART2, R.M. ELLAM2, J.G. FITTON1, S. BASU2 AND L.M. LARSEN3

M. STAUBWASSER1, F. VON BLANCKENBURG2 3 AND A. SCHIPPERS

1

Grant Institute of Earth Sciences, University of Edinburgh, Edinburgh EH9 3JW, UK ([email protected], [email protected]). 2 Isotope Geosciences Unit, SUERC, East Kilbride. G75 0QF, UK ([email protected], [email protected], [email protected]). 3 GEUS, Øster Voldgade 10, DK-1350 Copenhagen K, Denmark ([email protected]). Recently, several studies have shown an apparent link between high 3He/4He and depleted mantle (Stuart et al., 2003; Class and Goldstein 2005) which is inconsistent with prevailing orthodoxy where high 3He/4He reflects a lack of mantle degassing and depletion. Here we present 28 new 3 He/4He analyses from Tertiary picrites of Baffin Island and West Greenland and have identified 21 samples with 3He/4He >35Ra (the highest value observed in recently erupted basalt). Whole rock 143Nd/144Nd for these samples show a significant peak at depleted values (mean of 0.513024), as observed in an earlier study (Stuart et al., 2003), but an interesting feature of the new data is the scatter in 143Nd/144Nd down to 0.512876, a value which although not chondritic is certainly less depleted than that observed by Stuart et al., (2003). It is important to establish the effects of crustal contamination that would act to lower both 3He/4He and 143Nd/144Nd. Whole rock Pb isotope data show significant variations which are consistent with incorporation of unradiogenic Pb of crustal origin but critically there is no relationship between Pb and Nd isotopes. Thus, if crustal contamination is the cause of Pb isotope variation, it was not responsible for the range in 143Nd/144Nd. In addition, major and trace element compositions of olivine hosted melt inclusions allow us to compare melt inclusion compositions from olivines of various sizes, that record different stages of magma evolution, to provide better resolution on the effects, if any, and timing of potential crustal contamination. We conclude that variations in 143Nd/144Nd are likely to derive from heterogeneity in the mantle sources tapped by the Baffin Island and West Greenland picrites and that these new data do not support the concept of a discrete depleted-high 3He/4He end-member in the mantle.

References Stuart, F.M., Lass-Evans, S., Fitton, J.G. & Ellam, R.M. (2003). Nature 424, 57-59. Class, C. & Goldstein, S.L. (2005). Nature 436, 1107-1112.

1

Institut of Mineralogy, Leibniz-University Hannover, Germany ([email protected]) 2 Institut of Mineralogy, Leibniz-University Hannover, Germany ([email protected]) 3 BGR, Hannover, Germany ([email protected]) This study explores the potential use of Fe isotope analysis in long-term risk assessment of acid mine drainage generating mining piles. Balci et al. (2006) have demonstrated the enrichment of heavy Fe isotopes in Fe oxidation products with respect to Fe(II)aq during growth of Acidthiobacillus ferrooxidans in batch culture-experiments. They argued that inorganic equilibrium fractionation between Fe(III)aq and Fe(II)aq is the controlling reaction as described earlier by Welch et al. (2003). These authors have shown that in equilibrium the isotopic difference is '56FeFe(III)-Fe(II) ~ 2.9 ‰. Here, we present Fe isotope ratios measured on sequentially leached tailings material sampled from a 25 m drill core into an active pile at Selebi-Phikwe, Botswana, and of an additional analog column bio-leaching experiment in the laboratory. Throughout the drill core, reactive solid Fe(III) is enriched in the heavy isotopes (averageG56Fe = –0.15 ‰) with respect to pyrrhotite (averageG56Fe = –0.40 ‰). This is in general agreement with the batch experiments of Balci et al. (2006). The fraction of exchangeable and soluble Fe shows very low G56Fe values down to –2.4 ‰. Such extreme values can only occur in a small residual pool after the majority of dissolved Fe has been removed by precipitation. Our results can best be explained if the total reaction is broken down as follows. Fe dissolved from pyrrhotite is partitioned by microbial Fe(II)-oxidation into a mixed Fe(III)aq and Fe(II)aq pool which quickly equilibrates isotopically. Precipitation of solid Fe(III) occurs with perhaps additional fractionation from the Fe(III)aq in the mix. Our data suggest that the mixed reservoir initially consisted mostly of Fe(III)aq whose G56Fe would be close to the initial pyrrhotite. The strongly fractionated residual derives from the remaining small pool of Fe(II)aq. However, low G56Fe values cannot be sustained unless resupply of unfractionated, freshly dissolved Fe from pyrrhotite is sluggish. Although showing complex results, the bio-leaching experiments appear to confirm that the ratio of Fe(II)aq to Fe(III)aq in the fluid, and thus the overall metal sulfide oxidation activity in the pile, is the main factor that controls the isotopic composition of soluble Fe.

References Balci, N., Bullen, T.D., Witte-Lien, K., Shanks, W.C., Motelica, M., and Mandernack, K. (2006), Geochim. Cosmochim. Acta 70, 622-639. Welsch, S.A., Beard, B.L., Johnson, C.M., and Bratermann, P.S. (2003). Geochim. Cosmochim. Acta 67, 4231-4250.

A968

Goldschmidt Conference Abstracts 2007

The hydrothermal Wenzel deposit, South Germany: Implications for the formation of Kongsberg-type silver deposits SEBASTIAN STAUDE, THOMAS WAGNER GREGOR MARKL

AND

Institut für Geowissenschaften, Universität Tübingen, D-72074 Germany; ([email protected], [email protected], [email protected]) The post-Variscan Kongsberg-type Wenzel deposit near Wolfach, Schwarzwald, Germany, the type locality of the AgSb alloy dyscrasite, was investigated by ore microscopy, electron microprobe analysis, stable isotope and fluid inclusion analysis. Three mineralization stages could be distinguished in the vein. Whereas the first stage is a typical sulfide mineralization including galena and tetrahedrite, the second and third stage show a sulfide-poor association of AgSb alloys with Fe-, Co- and Ni-arsenides and -sulfarsenides in a calcite matrix. The main ore minerals of this stage are allargentum and dyscrasite. Seven distinct generations of calcite were distinguished. The G13C (V-PDB) and G18O (V-SMOW) values of these generations show a positively correlated trend that evolves from -13.0 to -4.0 ‰ and from 12.3 to 23.6 ‰, respectively. Fluid inclusion data of stage I fluorite and quartz show homogenization temperatures of 100-180 °C at salinities of 17-26 wt.% NaCl eqv. Fluid inclusions in stage II calcite display similar, but more restricted values of 110-150 °C and 25-28 wt.% NaCl eqv., respectively. The stage III fluid inclusions of calcite show similar homogenization temperatures, but different salinities. Earlier Ag-Sb-alloy bearing calcite of this stage contains inclusions with salinities of 27-30 wt.% NaCl eqv., whereas later ore-free calcite crystals show lower salinities of 3-10 wt.% NaCl eqv. The initial ice melting temperatures of most fluid inclusions range between -45 and -60 °C and are typical of an H2O-NaCl-CaCl2 fluid. Based on all available geochemical data and phase equilibrium constraints in the system Ag-Ca-Na-C-Si-Cl-O-H, we favor a model in which basement-derived near-neutral-pH hydrothermal fluids remobilized older Ag-Sb-bearing mineralizations. Mixing of these fluids with more alkaline formation waters from the Mesozoic cover rocks resulted in the precipitation of the silver alloys in an enrichment zone at P-T conditions of 120-150 °C and approximately 200 bars. A significant pH shift from near-neutral to alkaline is able to explain the abundant association of silver alloys with calcite gangue and the general absence of quartz in the enriched ore zone. This conceptual model can be applied to similar ore deposits world-wide, where rich ores of native silver and silver alloys are hosted by calcite-rich and quartz-poor gangue mineral assemblages.

Interpreting reaction rates at the field scale C. STEEFEL AND L. LI Earth Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 ([email protected]) Interpreting reaction rates and associated rate formulations (rate constants, catalytic and inhibitory effects, reaction affinity) at the field typically requires explicit consideration of transport . This is because overall rates may be either locally transport-controlled in the case of heterogeneous systems, or even globally transport-controlled if sufficiently large length scales for reaction are considered. The potential role of physical and chemical heterogeneity has been discussed, but is still not routinely evaluated at the field scale. Heterogeneity may introduce a scale dependence to field-scale rates even in the case where the pore fluids remain far from equilibrium due to either transverse concentration (and therefore rate) gradients, or as a result of longitudinal gradients that develop where the extent of reaction is large. These two effects can be quantified in models of the field-scale system respectively by 1) comparing results from 2D or 3D representations with those from 1D continuum models, and 2) by comparing results from 1D models with those from well-mixed flowthrough reactor models. The analysis indicates that well-mixed reactor models, implicit in the so-called “Inverse Models” to the extent that they are used to determine rate constants, should be used with considerable caution at the field scale. Another effect associated with the presence of heterogeneities that complicates the interpretation of fieldscale rates has to do with the determination of hydrologically accessible reactive surface area. Some regions within the reactive domain may be largely inaccessible because of their low permeability, or mass transfer from the low to high permeability regions (where the bulk of the flow occurs) may be rate-limited. An approach that uses the retardation of a reactive tracer to quantify the hydrologically accessible reactive surface area is presented and is combined with an example involving the weathering of smectite to kaolinite at the Shale Hills site in Pennsylvania, USA. The complexity of many multicomponent reaction networks presents another significant obstacle to the interpretation of rates at the field scale. Problems are usuallly manifested when an imcomplete data set has been collected at the field scale, in which case potentially important pathways may be neglected altogether. In this regard, combining major and minor element aqueous and solid phase chemical analyses with isotopic analyses offers a powerful approach for delineating all of the important pathways within a field-scale reaction network. Successful examples of such an approach will be provided.

Goldschmidt Conference Abstracts 2007

A969

Magma sources in the Icelandic Western Rift Zone (WRZ): Crustal and mantle input

The hydrolysis and cloro complexation of iron(III) in hydrothermal solutions

R. C. J. STEELE, M. F. THIRLWALL, C. J. MANNING, M.A.M. GEE, M. REGELOUS AND D. LOWRY

A. STEFANSSON1, T.M. SEWARD2 AND I. GUNNARSSON1

Department of Geology, Royal Holloway, University of London, Egham, Surrey, TW20 0EX, UK Low G18O Icelandic rift basalts have been interpreted as a result of contamination by low G18O crust, or as derived from an unusual, low G18O mantle source. In the Reykjanes Peninsula (RP) G18OOL values • +4.2 ‰ have been interpreted as mantle derived, while lower values were thought to reflect contamination [1]. (230Th/238U) disequilibrium has also been used as an indicator of crustal contamination in both rift, and off-rift zones in Iceland [2]. Since the magnitude of (230Th/238U) disequilibrium varies radially with proximity to the proposed centre of the Iceland plume [3], variations in (230Th/238U) may however, reflect melting parameters rather than crustal contamination. The WRZ is an ideal area to study the extent to which low G18O and (230Th/238U) disequilibrium indicate crustal contamination, due to their eruption during post/last-glacial time minimizing the need for an age correction. Further, the rift zone is orientated tangentially to the assumed location of the centre of the plume, and thus variations in extent of disequilibrium cannot be strongly controlled by distance to the plume. A study of new O-Sr-Nd-Pb data from the WRZ demonstrates G18O as low as +3.79 ‰ in primitive lavas (9.5 % MgO, 143Nd/144Nd 0.513046). The WRZ data appear compatible with varying degrees of crustal contamination in terms of G18O and 143Nd/144Nd. However, assuming the WRZ assimilant has a G18OOL of +1.2 ‰ [1] then 35 % bulk assimilation is required to produce the lowest G18OOL sample of 3.79 ‰ from the mantle mean of 5.2 G18OOL. This is difficult to reconcile with the relatively low degrees of fractional crystallization, shown by the 9.52 % MgO found in this sample. Isotopic data from Hengill central volcano, located at the triple junction between the WRZ, RP and the south Iceland seismic zone (SISZ), plot at the enriched end of WRZ Nd-SrPb trends, showing the sources are isotopically linked. G18OOL from Hengill is 3.83 - 4.5 ‰, with 5.91 - 8.73 % MgO. This range is present in rocks with homogenous Nd-Sr isotope ratios, showing that Hengill is much more consistent with assimilation producing the low G18O signature.

References [1] Thirlwall et al., (2006). Geochim. Cosmochim. Acta 70, 993-1019. [2] Sigmarsson et al., (1992) E.P.S.L. 110, 149-162. [3] Kokfelt et al. (2003). E.P.S.L. 214, 167-186.

1

Institute of Earth Sciences, University of Iceland, Sturlugata 7, 101 Reykjavík, Iceland 2 Institut für Mineralogie und Petrographie, ETH-Zentrum, Universitätstrasse 16, 8092 Zürich, Switzerland A knowledge of the stability of iron(III) hydroxide and chloride complexes in hydrothermal solutions is important for quantitative interpretation of the transport and precipitation of iron by crustal fluids. It is generally accepted that iron is predominantly transported as iron(II) species in reduced hydrothermal fluids in the Earth’s crust. However, boiling (phase separation) occurs ubiquitously in hydrothermal systems with the partitioning of hydrogen into the volatile (less dense) phase. The redox state of the residual liquid (or denser phase) and pH changes, such that iron(III) may become the predominant oxidation state of iron. A knowledge of the complex equilibria involving Fe3+ is therefore important in understanding the transport and precipitation chemistry of iron by fluids in the crust at high temperatures and pressure. The complexation of iron(III) with hydroxo and chloro ligands has been studied as a function of solution composition (salinity and pH) in hydrothermal solutions to 300°C at saturated water vapour pressure. Two experimental methods have been applied, uv-vis spectrophotometry using a hightemperature, flow-through gold-lined optical cell and hematite solubility using flow-though and static autoclave systems. The strong ligand-to-metal charge transitions of the uv-vis spectra of iron(III) hydroxo and cloro complexes at wavelengths below 400 nm were used to obtain molar absorptivities, İ, and equilibrium formation constants using principle component analysis and non-linear least squares treatment of the hematite solubility measurements were used to obtain solubility constants. Based on the experimental result iron(III) was found to hydrolyse to form FeOH2+, Fe(OH)2+, Fe(OH)3(aq) and Fe(OH)4- with increasing pH and FeCl2+, FeCl2+, FeCl3(aq) with increasing chloride concentration in acid solutions. Iron(III) hydroxide complexes were found to predominate in dilute and alkaline hydrothermal solutions whereas with increasing chloride concentration and temperature iron(III) chloride complexes become increasingly important species in oxidised acid solutions.

Goldschmidt Conference Abstracts 2007

A970

A 3.3 Ga Mo-Cu porphyry-style deposit at Spinifex Ridge, East Pilbara, Western Australia: Re-Os dating of Paleoarchean molybdenite H.J. STEIN1,2, M.E. BARLEY3, A. ZIMMERMAN1 4 AND B. CUMMINS

Atlantic cold-water spells into the Mediterranean caused the abrupt changes in the Levant’s post-Glacial hydrology and human-culture development M. STEIN AND Y. YECHIELI

1

AIRIE Program, Department of Geosciences, Colorado State University, USA; ([email protected]) 2 Geological Survey of Norway, 7491 Trondheim, Norway 3 University of Western Australia, Crawley WA 6009 4 Moly Mines Ltd, 46-50 Kings Park Rd, W. Perth WA 6005 Here we show that the Re-Os chronometer is robust for Paleoarchean molybdenites. Prior to this study, AIRIE’s oldest dated molybdenite was from the 3128 ± 13 Ma Sergeevske porphyry Au-Cu-Mo deposit, Ukraine. The youngest is from the 2.120 ± 0.007 Ma Boyongan porphyry Cu-Au deposit, Philippines. Thus, molybdenite chronology can be readily used across the full span of geologic time [1]. The Spinifex Ridge (Coppins Gap) Mo-Cu deposit is located immediately north of the variably deformed Mount Edgar batholith in the 3.52-2.85 Ga East Pilbara granitegreenstone terrane. Mo-Cu mineralization, estimated at 481 million tons carrying 0.06% Mo and 0.08% Cu, is associated with high-level quartz-plagioclase porphyry intruded into Warrawoona Group basalts and rhyolites, and porphyritic granodiorite bodies [2]. Quartz veins with molybdenite and chalcopyrite are most abundant where both granodiorite and quartz-plagioclase porphyry are present. Two molybdenite samples ascertain the timing of porphyry-style Mo-Cu mineralization at Spinifex Ridge. The drill core samples (SRD053, 227.2 and 227.6 m) represent main-stage stockwork ore hosted in potassically altered porphyry. Molybdenite was analyzed by NTIMS using a Carius tube digestion and double Os spike. Re-Os ages are 3298 ± 11 for a 1.5 cm quartz vein with irregular molybdenite selvages and 3284 ± 11 Ma for a 0.2 cm molybdenite clot adjacent to a similar vein. Ages are indistinguishable within their 2-sigma uncertainties, and agree with SHRIMP U-Pb zircon ages [3] for the Mount Edgar batholith (3314 r 13 Ma, Coppin Gap suite; 3304 r 10 Ma, Boodallana suite). Ore zone geometry inclusive of silicified, brecciated borders to the quartz-plagioclase porphyry suggest that the entire intrusive-mineralization system was strongly tilted by regional listric faulting that accompanied uplift of the batholith. The Re-Os dates present a maximum age for that faulting and a minimum age for the hosting quartz-plagioclase porphyry. The geology supports formation of the Spinifex Ridge Mo-Cu deposit in a weakly extending brittle regime at ~3.3 Ga, similar to magmatic-tectonic conditions that produce Mo-Cu porphyry-style deposits today.

References [1] Stein, H.J. et al. (2001) Terra Nova 13, 479-486. [2] Barley, M.E. (1982) Econ Geol 77, 1230-1236. [3] Williams, I.S. and Collins, W.J. (1990) EPSL 97, 41-53.

Geological Survey of Israel, Israel ([email protected]) Intrusion of cold Atlantic-water to the east Mediterranean at the 14th millennium BP (e.g “melt water pulse MWP1-A”) caused the abrupt drop of Lake Lisan (the last glacial precursor of the Dead Sea) from its maximum MIS2 stand of ~160 m (below mean sea level) to its lowest level (< 500 m bmsl), marking the severest catastrophic aridity that prevailed in the late Quaternary Levant. Regional rains resumed and lake level rose during the Younger Dryas (13 millennium BP) but dropped sharply again during the 11th millennium BP (reflecting “melt water pulse MWP1-B”). While the “melt-water pulses” amplified the post-glacial warming trend in the Levant, causing extreme aridities, the NA-cooling of the YD imposed a strong deviation from this trend. It seems that the YD cooling lags after the melt water pulses by causing shifts in the Polar fronts and Westerlies that brought more rain to the Levant. The abrupt changes in the Levant climate led the major developments in the regional cultural evolution - the collapse of the Natufian culture and the rise of the Pre-Pottery Neolithic (PPN) culture and agriculture society upon the transition to the milder Holocene. The rapidity of the response of the regional hydrological systems to the global climate changes and the sensitivity of past human cultures to these changes are certainly important lessons and alarming signals for our human society.

Goldschmidt Conference Abstracts 2007

UV femtosecond laser ablation applied to stable Fe isotopes in BIFs

Potassium partitioning in the lowermost mantle from ab-initio computations

G. STEINHOEFEL, I. HORN AND F. VON BLANCKENBURG Institut für Mineralogie, Leibniz Universität Hannover, Callinstr. 3, 30167 Hannover, Germany ([email protected] [email protected] [email protected]) Here we present in situ stable Fe isotope measurements in Precambrian banded iron formations (BIFs) using our in-house built laser ablation system which consists of a frequencyquadrupled fs laser operating at a wavelength of 196 nm and a multicollector-ICP-MS. The short pulse length turns the ablation process away from thermal pathways preventing artificial fractionation and minimizing matrix dependency (Horn and von Blanckenburg, 2007). The accuracy of this method has been verified for different types of matrices giving a reproducibility of 0.1‰ (2SD) for the 56Fe/54Fe ratio (Horn et al., 2006). BIFs are fine-grained chemical sediments which are the product of initial precipitation from seawater and subsequent diagenetic and metamorphic processes. All of these processes involve redox reactions, dissolution and precipitation of Fe resulting in fractionation of Fe isotopes. In order to investigate these processes, we have determined the Fe isotope composition of single Fe-oxide and Fe-carbonate crystals by spot analyses at 30 microns resolution in different thin sections of low-grade metamorphosed BIFs. Small-scale isotopic variations of up to 0.9‰ in G56Fe within single layers exist for hematite as well as for magnetite. Furthermore we detect isotopic zonation in magnetite crystals as small as 30 microns which becomes heavier in their Fe isotope composition towards the rim. These heterogeneities suggest variable relocation of Fe on a sub-millimeter scale during diagenesis and metamorphism. Fe-carbonates have also been investigated. Although crystal sizes are often less than 25 microns and Fe contents are low giving only low Fe signal intensities, we have found that variations in chemical composition have little influence on the Fe isotopic composition. Both, Fe-oxides and Fe-carbonates show constant average Fe compositions over all layers within a thin section. Since also magnetite and Fe-carbonate exhibit a constant relative difference of ca. 0.9‰, either a diagenetic process with the same precursor material must have established these isotope equilibrium fractionations or, alternatively, fluid sources were distinct but pathways to the Fe-carbonate and Fe-oxide were steady with time. These results illustrate that the study of stable Fe isotopes at high spatial-resolution have the potential to gain a better understanding on the mechanism of BIF formation and the Precambrian Fe cycle.

References Horn I., v. Blanckenburg F., Schoenberg R., Steinhoefel G. and Markl, G., (2006), Geochim. Cosmochim. Acta 70. 3677–3688. Horn I. and von Blanckenburg F., (2007), Spectrochim. Acta Part B, in press.

A971

G. STEINLE-NEUMANN1, K.K.M. LEE2 3 AND S. AKBER-KNUTSON 1

Bayerisches Geoinstitut, University of Bayreuth, Germany, ([email protected]) 2 Department for Physics, New Mexico State University, Las Cruces, NM, USA and Bayerisches Geoinstitut, University of Bayreuth, Germany, ([email protected]) 3 Scripps Institution of Oceanography, University of California at San Diego, La Jolla, CA, USA, ([email protected]) Partitioning of radioactive isotopes in the Earth’s interior is of great importance to the thermal and dynamic state and evolution of the Earth as radioactive decay provides an important source of energy for mantle dynamics. As a consequence, enriched (or depleted) reservoirs in the mantle can influence the energy balance in geodynamics. The recently discovered phase transition in MgSiO3 from perovskite (pv) to post-perovskite (ppv) in the lowermost mantle provides the possibility for an enriched or depleted zone at the base of the mantle. Therefore, the partitioning of radiogenic isotopes among the phases of the lower mantle is of central importance in geochemistry and geodynamics of the deep Earth. Here we take a first step in addressing this issue by considering the partitioning of potassium between pv and ppv by performing ab-initio computations. We have set up computations for a coupled substitutions of K and Al or Fe3+ (M ion) for 2 Mg on the A site, and evaluate the energetics of a (Mg30,K,M)Si32O96 composition in the pv and ppv structure. We compare the energetics of the following reactions for both pv and ppv to compute the enthalpy of formation: (Mg30,K,M)Si32O96 +2MgO Æ Mg32Si32O96 + 1/2 K2O + 1/2 M2O3.

The computations are performed with the VASP package using the projector augmented wave method for the static lattices of the high pressure phases in the reaction above. We use the generalized gradient approximation to the exchange and correlation potential. Structures are optimized for internal and external degrees of freedom at constant volume, and we assume that the M and K ions are at A positions directly adjacent to one another. Computations are performed for a wide compression range, reaching pressures of the CMB. Slightly above 100 GPa we predict that DK<1, implying that K partitions preferably into the ppv phase. Increased temperatures would decrease the magnitude of preferred differentiation. This makes a K enriched layer at the base of the mantle a possibility. It is clear, however, that the effect of other phases stable in the lower mantle (Ca-pv and mw) must be considered in a full assessment of K distribution in the lower mantle.

A972

Goldschmidt Conference Abstracts 2007

The mobility of actinides and 90Sr from bomb test fallout in a karstic area, Jura Mountains (Switzerland)

Biogeochemical cycling of rare earth elements in surface soils M. STEINMANN1, P. STILLE2 AND M.C. PIERRET2

PHILIPP STEINMANN1, LAURENT POURCELOT2 AND PASCAL FROIDEVAUX1

1

EA 2642 Géosciences, Université de Franche-Comté, F-25030 Besançon, France ([email protected]) 2 CGS, CNRS, UMR 7517, EOST, F-67084 Strasbourg, France ([email protected], [email protected]) Stille et al. (2006) have suggested in a study on the Strengbach catchment in the Vosges mountains (eastern France, http://ohge.u-strasbg.fr) that preferential absorption of light REE (LREE) by vegetation may contribute to LREE depletion of stream water. New REE data on soil solutions recovered on the same site from 10 to 70 cm depth show similar to vegetation an enrichment of the LREE. Strontium and neodymium isotope data from these solutions indicate that up to 90 % of Sr and Nd are derived from vegetation litter. This origin of the REE in the surface soil is in strong contrast to the underlying regolith, where the mobile REE essentially originate from the dissolution of apatite derived from the granitic bedrock (Aubert et al.,2001). The Sr isotope record from tree rings of 3 spruce specimens planted 90-100 years ago on the study site yields valuable information on the formation of these different REE pools. The data demonstrate that the spruce mainly absorbed Sr from granite-derived minerals during the first years after plantation. But rapidly the isotopic compositions of the tree rings are shifted to values typical for atmospheric deposits. We interpret this evolution by transformation of the initial mineral soil into a surface soil rich in organics issued from the decomposition of vegetation litter. This new surface soil contained in the beginning mainly regolith-derived Sr inherited from vegetation. This initial Sr was then continuously removed by soil water runoff and replaced by atmospheric Sr. We are actually analyzing Nd isotopic compositions on the same spruce samples in order to confirm this hypothesis for the REE. This scenario thus suggests that vegetation and surface soil form an almost closed biogeochemical cycle for the REE, accumulating preferentially LREE issued from vegetation litter. This cycle is modified by inputs from atmospheric sources and leaching of remaining soil minerals, and outputs by surface and soil water runoff. Local stream water is in contrast to surface soil and vegetation depleted in LREE indicating that surface soil water is not the dominant source for stream water. This is in agreement with Nd-Sr isotope data from Stille et al. (2006) showing that stream water REE are mainly derived from alteration of apatite within the mineral soil. We suggest that the formation of a LREE enriched reservoir in the surface soil has together with preferential scavenging of the LREE by adsorption and precipitation during groundwater flow contributed to the formation of a LREE depleted groundwater pool within the underlying regolith. This groundwater is the main source for stream water at low stream flow.

References Aubert et al. (2001). GCA 65(3): 387-406. Stille et al. (2006). GCA 70(13): 3217-3230.

1

University Institute of Radiation Physics, University Hospital Center and University of Lausanne, Grand Pré 1, CH-1007 Lausanne, Switzerland; ([email protected]; [email protected]) 2 Institut de Radioprotection et de Sûreté Nucléaire IRSN, CEA Cadarache Bat 153, 13108 St Paul Lez Durance, France; ([email protected]) The presence of artificial radioisotopes of plutonium, americium and strontium is mainly due to the atmospheric tests of nuclear bombs before 1964. Present day inventories of these radionuclides in soils of the Jura Mountains (Switzerland) often are less than expected. Apparently, there has been leakage from the soils (especially the thin soils) into the karst of this region of carbonate rocks. With the goal to assess this possible long-range migration in a natural setting we analyzed the radioisotopes - in upland soils (soil profiles and soil solution profiles near Col de Mollendruz, Switzerland, 1200 a.s.l.), - in a karstic source at a distance of ca. 5 km (water and aquatic mosses, 660 m a.s.l.), and - in cave deposits from the same area. The mean soil inventories (0-25 cm) for thick soils are 1.3 kBq/m2 and 0.12 kBq/m2 for 90Sr and Pu, respectively. The radioisotopes have significantly diffused towards deeper soil layers (40 cm depth or more) and they are also present in measurable amounts in the soil solution (< 0.22 μm, i.e. dissolved and colloidal). These findings highlight the potential of long-range transport in colloidal form. The calculated distribution coefficients Kd are on the order of 10'000; 3'000; and 1'500 for Pu, Am, and Sr, respectively. The radioisotopes appear in the waters of the karstic source and in aquatic mosses collected close to the source. The plutonium activity found in the filtered (<0.45 μm) source water is 4 and 22 μBq l-1 (U95 = ±25 %) at high-stand and lowstand, respectively. The Sr-90 activity of these waters is approximately 1000 times higher. In the mosses we find up to 20 mBq/g or more of Sr-90. However, most of this activity is associated with carbonates, which have precipitated on the mosses after degassing of CO2 from the source waters. The Pu activities of the moss samples vary between 0.1-0.2 mBq/g. Taken together our results demonstrate the long-range transport (km-scale) of Pu, Am, and 90Sr in a natural karstic environment with no radioisotope contamination other than "global fallout".

Goldschmidt Conference Abstracts 2007

Solute exchange across the sediment water interface in an acidic pit lake R. STELLMACHER AND M. KOSCHORRECK Helmholtz Centre for Environmental Research – UFZ, Brückstraße 3a, D-39114 Magdeburg, Germany ([email protected]; [email protected]) Solute exchange across the sediment-water interface affects the water quality in mining pit lakes. Fluxes are supposed to be affected by advective groundwater flow. However, benthic solute fluxes in such lakes have mostly been determined by sediment core incubations in the laboratory or by calculation from pore water profiles. In the present study direct solute flux measurements were carried out in situ in acidic Mining Lake 111 in the Lausitz lignite mining area, Germany. Three sites were chosen to represent A) groundwater inflow conditions B) the lake profundal and C) a reference littoral site. A volume of lake water and the sediment below were enclosed using an opaque benthic flux chamber. Oxygen decrease in the chamber was directly measured using an optode, whereas parameters such as total inorganic carbon (TIC) and dissolved Fe2+ were analyzed in a series of water samples withdrawn from the chamber. Sites B and C showed oxygen consumption rates between 5 and 15 mmol m-2 d-1 while oxygen loss from the chamber at site A equaled rates between 19 and 35 mmol m-2 d-1. TIC fluxes from the sediment varied closely around 5 mmol m-2d-1 at sites B and C, but reached up to 100 mmol m-2d-1 at site A. Further a distinct increase of ferrous iron in the chamber was detected only at site A, i.e. in the groundwater inflow area. Our results reveal significant differences in benthic solute fluxes depending on groundwater inflow and setting in the lake. The groundwater inflow area is characterized by high oxygen consumption and high inflow of TIC and dissolved iron. To our knowledge these are the first measurements of this kind, providing valuable practical experience for the use of benthic flux chambers in pit lake research.

A973

FTIR water observation in minerals from diamond inclusions and matrix of diamondiferous eclogite A.S. STEPANOV, D.A. ZEDGENIZOV AND V.S. SHATSKY Russian Federation, Novosibirsk, Institute of Geology and Mineralogy. ([email protected]) The series of recent observations have recovered that nominally anhydrous minerals (NAM) in the Earth’s mantle may contain significant amount of water. There are some evidences that water plays an important role in diamond formation. Here we present the first data on water content in clinopyroxenes (Cpx) in the matrix and diamond inclusions (DI) from diamondiferous eclogite xenolith from Udachnaya kimberlite pipe (Yakutia, Russia, description in [1]). The water content has been estimated from unpolarized FTIR spectra of slices of the rock and polished plates of diamonds with DI. Spectra from individual DI have been obtained by subtraction to diamond absorption in the point near inclusions. The thickness of rock slices was measured by micrometer and thickness of inclusions was determined by up and bottom focusing under microscope with reference to refractive index. The strong band at 3450 cm-1 and two weaker bands at 3620 and 3740 cm-1 are observed in FTIR spectra of Cpx from the matrix of eclogite. Most Cpx inclusions in diamonds show single absorption band at 3450 cm-1 and only one inclusion has additional band with the maximum at 3600 cm-1 (may be attributed to OH band in chlorite [2]). Water content in Cpx has been estimated by using calibrations from [3]. Water concentrations are 60-90 ppm in Cpx from the matrix and 80150 ppm in Cpx from diamond inclusions. Neither Grt DI nor Grt from the matrix do not show any water absorption in their FTIR spectra. Specific results of this study are: (i) the major phase containing water in the eclogite is Cpx; (ii) water content in the matrix and DI from eclogite is not significantly different. It may testify that water content in eclogite did not change during period between diamond formation and ascent of xenolith to the surface by kimberlite magma. This work was supported by the Siberian branch of the Russian Academy of Science.

References [1] Stepanov A.S., Shatsky V.S., Zedgenizov D.A., et al (2007) Russ. Geol. and Geoph., 9 (in press.). [2] Koch-Muller M., Matsyuk S. and Wirth R. (2004) Amer. Mineral., 89, 921-931. [3] Bell D.R., Ihinger P.H. and Rossman, G.R. (1995). Amer. Mineral., 80, 465í474.

A974

Goldschmidt Conference Abstracts 2007

Towards improved accuracy of SHRIMP zircon 207Pb-206Pb measurements

Crustal anatexis in the early Archean: Geochemical and isotopic evidence from the ca. 3.66 Ga Nuvvuagittuq Tonalite Suite

R. A. STERN1, S.L KAMO2, S. BODORKOS3, A. HICKMAN3, AND F. CORFU4 1

Geochronology Laboratory, Onshore Energy and Minerals Division, Geoscience Australia, P.O. Box 378 Canberra 2601, Australia ([email protected]) 2 Jack Satterly Geochronology Laboratory, Department of Geology, University of Toronto, 22 Russell St., Toronto, Ont., M5S 3B1, Canada ([email protected]) 3 Geological Survey of Western Australia, Department of Industry and Resources, 100 Plain St., East Perth, Western Australia 6004, Australia ([email protected]; [email protected]) 4 Department of Geosciences, University of Oslo, Postbox 1047, Blindern N-0316, Oslo, Norway ([email protected]) SHRIMP and other ion probes are routinely utilized to determine 207Pb/206Pb ratios of zircons, but only in special cases can the accuracy of the measurements be claimed to be better than a few ‰. A great deal of attention is placed on calibrating 206Pb/238U ratios, yet for 207Pb/206Pb this is seldom the case. Any systematic bias in 207Pb/206Pb of zircon is considered insignificant relative to the low precision of the individual spot analyses, a reasonable assumption based upon the generally well-known characteristics of the large ion probe. Nevertheless, as several individual 207Pb*/206Pb* measurements are typically pooled to improve overall precision to the ±1-5 ‰ level for Paleoproterozoic or older zircons, the accuracy of the composite ratio should be a concern. There are several potential sources of bias in ion probe 207 Pb*/206Pb* measurements of zircon, including instrumental mass fractionation, peak shape, detector performance, isobaric interferences, common Pb correction, and the method of data processing. Although it is important to understand and potentially control these individual factors, it is critical to be able to measure the overall bias. Unfortunately, there appear to be no ion probe zircon reference materials developed expressly for 207Pb*/206Pb* calibration. Our data obtained from analyzing the Proterozoic U-Pb zircon QGNG suggest that systematic errors of several ‰ may exist between sessions on the same instrument and between different instruments. However, this zircon is not a suitable Pb-isotope standard for several reasons, and the validity of the results is uncertain. We have developed an Archean (ca. 3.5 Ga) 207Pb*/206Pb* zircon reference material, and TIMS data indicate homogeneity. The goal is to incorporate analyses of this material as a routine part of SHRIMP zircon sessions in order to monitor and potentially correct for any systematic error in the 207Pb*/206Pb* ages.

R.K. STEVENSON 1 AND M. BIZZARRO 2 1

GEOTOP and Département des sciences de la terre et de l’atmosphère, Université du Québec à Montréal, PO BOX 8888, Station Centre-Ville, Montréal, QC, H3C 3P8 (stevenson.ross @uqam.ca) 2 Geological Museum, Øster Voldgade 5-7, 1350 Copenhagen K, Denmark ([email protected]) Geochemical and Nd isotope data are presented for a suite of ca. 3.66 Ga tonalites from the 3.8 Ga Nuvvuagittuq supracrustal sequence in the Inukjuak Domain in the western Minto Block, northeast Canadian Superior Province. The Nuvvuagittuq volcano-sedimentary sequence consists of mafic amphibolites with ultramafic lenses, and intermediate and felsic schists dated at ca. 3.8 Ga and interpreted to be of volcanic origin. The sedimentary rocks consist of conglomerate, banded iron formation and quartzitic iron formation. The supracrustal assemblage has a semi-oval form and the tonalite suite both mantles the exterior of the oval and forms the inner core of the assemblage. The exterior sheath of the tonalite is mylonitized and in tectonic contact with Neoarchean (2.7-2.8 Ga) tonalite suites of the Inukjuak Domain. The tonalite core is less mylonitized but both the marginal and core portions of the suite vary from tonalite to granodiorite and granite in composition. A Lu-Hf isochron mineral-whole rock isochron from the tonalite core yields an age of 3645±26 Ma which is the same age, within error, of the 3.66 Ga U-b zircon age from the tonalite sheath. The tonalite suite is characterized by light Rare Earth element (REE) enriched profiles and variable heavy REE depletion as well as high La/Yb (n) and moderate Sr/Y ratios at low Yb and Y concentrations. Initial Nd and Hf isotopic values are largely negative suggesting that the suite was formed by melting of an older crustal component. Comparison with the isotopic evolution of the Nuvvuagittuq supracrustal sequence suggests that the tonalites may have formed by crustal anatexis of the 3.8 Ga supracrustal sequence. The tonalite suite represents the oldest tonalite suite in the Superior Province and compares in age with anatectic melting events described in the Eoarchean gneisses of West Greenland (e.g. Amitsoq gneisses) and Labrador (e.g. Uivak Gneisses) of the North Atlantic Craton (NAC). The commonality of the ca 3.66 Ga anatectic event may imply a widespread Early Archean migmatization event.

Goldschmidt Conference Abstracts 2007

A975

Fungal transformation of lignite in overburden dumps

Isotopic fractionation of Uranium in low-temperature environments

E. STIEBITZ AND A. KASSAHUN

C.H. STIRLING1,2, M.B. ANDERSEN3,2, R. WARTHMANN2 4,2 AND A.N. HALLIDAY

Groundwater Research Institute Dresden, Meraner Strasse 10, 01217 Dresden, Germany ([email protected]) Acidity in overburden dumps of open cast lignite mines poses a considerable environmental threat. Despite the persistence of the dump organic material (lignified organic matter), microbial acidity attenuation processes such as iron sulphide formation by microbial sulphate reduction were observed within several dumps (Storch et al.; 2007). Fungi are known for their ablility to liquefy lignite (Ward; 1985). Therefore, autochthonous acidotolerant fungi may provide bacterial substrates by transformation of lignite into water soluble organic matter.

Methods Autochthonous fungal strains were isolated from dump material of the Plessa field site (Lusatia, Gemany). The sediment was placed on Sabouraud glucose agar and incubated at 20°C. After formation of a thick hyphae mat identical dump material was placed on this mat. A few days later bioliquefaction of lignite particles started.

Results The autochthonous fungies (Mucor hiemalis) transformed lignite particles into black and shiny droplets of very high organic carbon contents (~15 g/L). The liquid product was analysed by gel chromatography to consist of humic substances (a), building blocks (b), low molecular weight substances (c), neutral substances (d), and polysaccharides (e). The liquefied organic matter is highly polar and water soluble. HPLC analysis (Schmalz et al.; 2002) of the water dissolved droplets revealed the predominance of carbonylic and carboxylic functional groups. Carbonic acids are well known substrates for sulphate reducing bacteria (Widdel; 1988). 16

concentration [g/L]

14 12 10

d c b

8 6

a

4 2

e

0

fig. 1: liquefied lignite

fig. 2: DOC fractions

1

Dept. Chemistry, Univ. Otago, New Zealand ([email protected]) 2 Dept. Earth Sciences, ETH Zürich, Switzerland ([email protected]) 3 Dept. Earth Sciences, Univ. Bristol, UK ([email protected]) 4 Dept. Earth Sciences, Univ. Oxford, UK ([email protected]) Uranium is the heaviest naturally occurring element and mass-dependent isotopic fractionation between 235U and 238U, which scales with GM/M2, is not normally considered significant given the small ~1% difference in mass. It is therefore usual to assume that 238U/235U is constant in the terrestrial environment and equal everywhere to 137.88 at the present day. Importantly, isotopic fractionation of the very heavy elements has recently been investigated for mercury and thallium in the context of mass-independent nuclear field shift effects1, which do not scale with GM/M2, and are predicted to have permil-level effects on the heavy masses, including uranium. We have developed experimental protocols for the precise measurement of 238U/235U and 238U/234U by multiple-collector ICPMS (MC-ICPMS) to investigate potential isotopic fractionation in uranium. Using multiple-Faraday protocols and a high-purity 233U-236U double spike to internally monitor instrumental mass bias effects, we are able to resolve variations in 238U/235U and 238U/234U at the 0.4 and 0.3 epsilon level (2V; 1 epsilon = 1 part in 10,000), respectively. Measurements for samples formed in a range of lowtemperature environments reveal sizeable, permil-level natural variability in 238U/235U2. Present experiments are focussed on the isotopic fractionation of uranium during the biologicallymediated reduction of U(VI) to U(IV). Our new observations indicate that uranium isotopic fractionation of 238U/235U may offer the potential to monitor biological pathways and redox processes occurring during mineralization, weathering, and the transition between the U(IV) and U(VI) oxidation states, offering new insight into the processes at work. Moreover, variability in 238U/235U will have a direct bearing on the U-series and U-Th-Pb chronometers, when applied to samples formed in lowtemperature environments, as these chronometers currently assume an invariant 238U/235U equal to 137.88.

References Storch, A., Hoth, N., Tröger, K., Knöller,K. (2007) 17th Internat. Biohydrometallurgy Symp. (IBS 2007) Frankfurt Ward, B., (1985). System. Appl. Microbiol. 6. 236-238 Schmalz, L.; Rossner, U.; Grupe S., (2002). Vom Wasser 99. 63-78 Widdel, F. (1988) in Zehnder, A.J. Biology of anaerobic microorganisms, Wiley and sons New York. 469-585

References Schauble E.A. (2007). Geochim. Cosmochim. Acta 71, 217089 Stirling C.H., Andersen M.B., Potter E.-K and Halliday A.N. (2007). Earth Planet. Sci. Lett. accepted for publication.

A976

Goldschmidt Conference Abstracts 2007

Rivers of North Rhine Westphalia – Revisited A. STÖGBAUER1, H. STRAUSS1, J. ARNDT1, V. MAREK1, F. EINSIEDL2 AND R. V. GELDERN3

Cosmogenic nuclide calibration – A progress report from the CRONUS project JOHN STONE

1

Geological and Palaeontological Institute, Westfälische Wilhelms-Universität Münster, Corrensstr. 24, 48149 Münster, Germany; ([email protected]; [email protected]) 2 GSF-National Research Center for Environment and Health, Institute of Groundwater Ecology, Ingolstaedter Landstr. 1, 85764 Neuherberg, Germany ([email protected]) 3 Leibniz Institute for Applied Geosciences, Section3: Geochronology and Isotopehydrology, Stilleweg 2, 30655 Hannover, Germany ([email protected]) In the early nineties of the last century rivers of NorthRhine Westphalia were examined by Veizer and members of his working group (Flintrop et al., 1996). Their motivation for having a close look on these – and on other more eminent – rivers was primarily the search for the missing carbon-dioxide sink. In addition, they provided a documentation of the pollution situation. We revisited some of these rivers, looking for possible changes that occurred during the past 15 years, and visited additional rivers with the focus not on the carbon budget but more on the pollution state. In order to isolate occuring in-river processes and to identify sources of river constituents we analysed the isotopic compositions of G2H and G18O of water, G13C of dissolved inorganic carbon, G34S and G18O of sulphate, and G15N and G18O of nitrate. Isotopic work was supplemented with measurements of common physical parameters and concentrations of major anions and cations. The different geographic settings of the rivers are mirrored by their isotopic composition of water. Respective regional trends are affected by processes as evaporation and mixing of different water masses. Compared to previous work, salt pollution (i.e. ions commonly attributed to an anthropogenic origin) has decreased. This holds true especially for potassium and nitrate, but also for sodium and chloride. The isotopic composition of nitrate in the river Ruhr identifies organic fertilizers and sewage as the main sources, with some evidence for denitrification processes. Compared to 1991, sulphate concentration has increased in the downstream part of the Lippe. Low G34S values suggest a higher contribution from pyrite oxidation as the likely cause. This could indicate changes in mining activity and related groundwater hydrology. For dissolved inorganic carbon we found higher concentrations characterized by lower G13C-values in all rivers compared to results from the early 1990s, caused by differing meteorological situations and in-river carbon cycling.

Reference Flintrop, C., Hohlmann, B., Jasper, T., Korte, C., Podlaha, E.G., Scheele, S., and Veizer, J. (1996), American Journal of Science 296 58-98.

Department of Earth and Space Sciences, University of Washington, Box 351310, Seattle, WA, 98195 ([email protected]) One goal of the overall CRONUS project (CRONUSEarth, CRONUS-EU and contributors from countries outside the US and EU) is to derive cosmogenic nuclide production rates from well-characterised geologic sites. We aim to generate a network of natural calibration data that span a wide range of altitudes, latitudes and exposure periods, including as many nuclides and target minerals as possible at each site. This talk will provide a status report on geological calibration of 10Be, 26Al, 36Cl and 3He production rates, reviewing published data and presenting initial CRONUS results from sites in North America, Scotland and Antarctica. The geological calibration effort is closely integrated with work on altitude-latitude scaling schemes, which provide the framework for comparing production rates between different latitudes, altitudes and exposure periods. All published scaling schemes can reconcile existing 10Be calibration data to within approximately 10% (1 sd). Assigning appropriate scaling errors is more complicated, however, because examination of the misfit between scaled production rate estimates and calibration data shows that there are systematic effects (biases in latitude and/or altitude) for all calibration schemes, as well as random error. The systematic component is minimal close to calibration sites, but may be large (and is unknown) at latitudes and altitudes far from calibration data. Additional calibration sites are needed to provide denser coverage, especially in latitude and exposure duration. Low latitudes are poorly represented in the data so far, and old sites are required to test the predictions of time-dependent scaling schemes which attempt to correct for paleomagnetic variations.

Goldschmidt Conference Abstracts 2007

Age of magnetite–apatite deposits and geochemistry of host rocks, Bafq District, Central Iran H.-G. STOSCH1, R.L. ROMER2 AND F. DALIRAN1 1

Universität Karlsruhe, Institut für Mineralogie und Geochemie ([email protected], [email protected]) 2 GeoForschungsZentrum Potsdam ([email protected]) Major hydrothermal/metasomatic Kiruna-type REEapatite-magnetite ores (2x109t) of the Bafq district in East Central Iran are hosted by Early Cambrian rhyolitic rocks and occur locally also in late spilitic basalts. This intimate association within a 150 km N–S striking structural zone, bordered by crustal faults, suggests a genetic link between magma generation and ore formation. The exact age of the mineralization, however, has not yet been determined. The tectonic setting of granitoids and rhyolites has been controversially discussed (rift setting versus magmatic arc environment) with Ramezani and Tucker (2003) favoring a collisional setting based on the chemical signature of the felsic magmatic rocks and paleogeographic reconstructions. Our current study deals with the age of the mineralization and the geochemistry of the host rocks. Petrographic investigations indicate that the rhyolites have been subjected to a broad-scale alkali metasomatism and an ore-related metasomatism that have modified their original composition, thus complicating the interpretation of geochemical data. Indeed, our major and trace element data do not yet permit an unambiguous interpretation of the geotectonic environment. However, we note that volcanism in the southern sector of the Bafq district is essentially bimodal with intermediate (andesitic) rocks being rarely encountered and mafic rocks being subordinate to felsic ones. Clear apatite crystals were separated from three apatite– magnetite deposits. The nine analyzed samples yield apparent 206 Pb/238U ages between about 527 and 539 Ma and thus fall entirely within the age range of the felsic magmatic rocks dated by Ramezani and Tucker (2003) [525–547 Ma]. This confirms field evidence that the ore formation was closely related to the Early Cambrian magmatic event (Daliran, 2002).

References Ramezani, J. & Tucker, R. D. (2003) Am. J. Sci. 303, 622–665 Daliran, F. (2002) In: Hydrothermal Iron Oxide Copper–Gold and related deposits, Volume 2, T. M. Porter [Editor], PGC Publishing, Linden Park, South Australia, 303–320

A977

Tracing the mineralogy of oceanic basalt sources A. STRACKE1 AND M. WILLBOLD2 1

ETH Zürich, Institut für Isotopengeologie, Clausiusstr. 25, 8092 Zürich, Switzerland ([email protected]). 2 University of Bristol, Department of Earth Sciences, Bristol, BS7 1RJ, United Kingdom ([email protected]) Subduction of oceanic crust has introduced considerable quantities of basalt into the Earth’s mantle, which, in the form of eclogite or pyroxenite, have often been proposed to constitute part of oceanic basalt sources. Isotopic evidence for mafic consitutents in oceanic basalt sources remains ambi-guous because their inferred isotopic composition is highly model dependent. Identification of pyroxenite derived melts on the basis of their major element composition is compli-cated by second order processes, i.e. fractional crystallization. Incompatible trace element compositions, however, are less affected by small extents of fractional crystallization. In addition, the higher modal abundance and different composition of garnet and clinopyroxene in pyroxenite compared to peridotite can lead to different partitioning behavior of the incompatible trace elements, which may be a useful tracer of pyroxenite versus peridotite derived melts. On the basis of recent partitioning experiments, bulk D values for pyroxenite are, on average, about factor of 3-6 higher than for garnet peridotite. The relative compatibility of the lithophile trace elements, however, is broadly similar. The main differences are an order of magnitude higher Ba/(Th, U, Nb, La), and about a factor of 2-3 lower Nb/(Th, U, La) and Sr/(Nd, Hf, Zr, Pb) bulk D ratios in pyroxenite compared to peridotite. Owing to the extremely low bulk D values of the most incompatible elements Ba, Th, U, Nb and La, however, even large differences in their bulk D ratios result in no resolvable differences in derivative melt compositions at melt fractions • 5%. For the more compatible elements Sr, Nd, Hf, and Zr, partition coefficient induced differences (higher Sr/(Nd, Hf, Zr) and Nd/Hf ratios) persist up to melt fractions of 20-30%. Whether differences in melt composition owing to the different partitioning behavior of garnet-peridotite and pyroxenite can be identified therefore depends largely on the style of melting and melt aggregation, i.e. how do melts form, what degree of melting is reached before separation from the solid (melt extraction) and how and to what extent do melts pool to form the aggregate melts erupted on the surface. Owing to its lower solidus temperature, pyroxenite starts melting deeper than peridotite. Key parameters for being able to resolve partitioning induced differences in melt compositions are therefore the extent of melting of pyroxenite before the onset of peridotite melting and the extent to which early pyroxenite melts can be extracted, if at all, without being significantly mixed with melts from the ambient peridotite. Regardless of the exact melt extraction scenario, the large extents of melting expected for pyroxenite (>20%) suggest that any partition coefficient induced differences between peridotite and pyroxenite melts are likely to be subordinate to initial compositional differences.

A978

Goldschmidt Conference Abstracts 2007

Mineralogical and ore-petrographic investigation of the iron ore occurrence of Ano Valsamonero, Rethymno (Crete)

Multiple sulphur isotopes reflecting compositional changes in Earth’s early atmosphere H. STRAUSS1, M. PETERS1, J. FARQUHAR2, D. JOHNSTON2, U. WIECHERT3 AND A. STÖGBAUER1

A. STRATAKIS AND G. ALEVIZOS 1

Dept. of Mineral Resources Engineering, Technical University of Crete, Polytechnioupoli, Gr-73100 Chania, Greece ([email protected]) Numerous iron ore occurrences appear within the PhylliteQuartzite-Series (PQS) of Crete, some of which were occasionally mined. The PQS forms is the lower part of Phyllite nappe, which with Gypsum-Rauwacke-Formation, composes the Phyllite nappe of Crete that includes the metamorphic rocks between the Plattenkalk-Series underneath and the Tripolitza-Series above. The PQS contains mainly phyllites and quartzites, in addition to metaconglomerates, marbles, calcareous phyllites and metabasalts. The examined occurrence is located about 15 km southwest from Rethymno next to Ano Valsamonero and occurs in the form of lenses within the phyllite and quartzite of the PQS. Sampling took place along a 10 m thick profile, in distances approximately 1 m in vertical arrangement. X-ray diffraction and ore microscopy were used for determination of the mineralogy and structure of the iron ore, while the chemical composition was determined by X-ray fluorescence. The iron ore consists predominantly of hematite, goethite and quartz, and subordinately of chlorite and muscovite. Hematite occurs in the form of radial, or more rarely in punctate aggregates between the quartz grains of ferruginized quartzite, especially in those samples, which originate from the above layers of the profile. In the samples from the middle and lower parts of the profile, hematite and goethite form a ferruginized front, which replaced largest part of the groundmass of quartzite. The Fe-rich solutions penetrated quartzite and precipitated as goethite, which forms concentric textures. The goethite changes gradually by dehydroxylation into hematite. The replacement by the Fe-rich solutions takes place from hair-cracks of the existing quartz grains. The composition of the iron ore along the profile varies regarding the Fe2O3 and SiO2 contents within a wide range. According to the microscopic investigation of the ore, it is concluded that the processes, which have caused the massive replacing ferrugination of the ore occurence of Ano Valsamonero are of epigenetic origin and can be classified to the continental ones.

Geologisch-Paläontologisches Institut, Westfälische Wilhelms-Universität Münster, Münster, Germany ([email protected]) 2 Department of Geology, University of Maryland, College Park, Maryland, U.S.A. 3 Institut für Geowissenschaften, Freie Universität Berlin, Berlin, Germany Pyrite extracted from more than 120 samples collected from Archean and Paleoproterozoic siliciclastic sedimentary units in southern Africa, western Australia, south-central Canada as well as in southwest Greenland and ranging in age from 3.8 to 2.1 Ga display nonzero '33S and '36S values. Based on empirical and experimental data as well as respective modeling results, non mass-dependent sulphur isotope fractionations are considered to result from photochemical reactions of different sulphur-bearing compounds in the atmosphere under low atmospheric oxygen abundances (e.g., Farquhar et al., 2000; Farquhar and Wing, 2003; Pavlov and Kasting, 2002). The record of newly obtained non mass-dependent sulphur isotope results displays distinct temporal variations in their magnitude. High-magnitude '33S values for the Paleoarchean are followed by a somewhat attenuated '33S signal in the Mesoarchean, while the Neoarchean and early Paleoproterozoic show extremely variable '33S values (total range of 11.5‰). Post-2.3 Ga old sediments do not show a non mass-dependent sulphur isotope signal. In addition to the temporal change in '33S, distinct temporal differences also exist considering relationships between G34S and '33S as well as between '33S and '36S. This suggests variations in the respective photochemical reactions (e.g., different sulphur-bearing compounds involved) and/or reflect differences in atmospheric composition with respect to effective UV-shielding. However, we clearly rule out transient oxygen abundances as possible cause. Instead, potential alternatives include variations in the atmospheric abundance of methane and/or in the ratio of carbon dioxide to methane.

References Farquhar, J., Bao, H.M., Thiemens, M., (2000), Science 289, 756-758. Farquhar, J., Wing, B., (2003), Earth Planet. Sci. Lett. 213, 1-13. Pavlov, A.A., Kasting, J.F., (2002), Astrobiology 2, 27-41.

Goldschmidt Conference Abstracts 2007

A979

Molecular characterization of Selenium in the environment

Origin of sulfur rich apatite in silicic, calc-alkaline magmas

DANIEL G. STRAWN, LIBBIE ORAM AND GREGORY MÖLLER

M. J. STRECK1, C. A. BRODERICK1 AND W. E. HALTER2

University of Idaho, Moscow, ID 83844-2339 ([email protected], [email protected], [email protected]) Selenium in the environment is an important element for ecosytem nutrition, but also can be toxic when present at slightly elevated concentrations. Selenium has similar chemistry as sulfur, and thus has multiple oxidation states and molecular forms in the environment. Availability of selenium for uptake by organisms or leaching to ground and surface waters is a function of its speciation. Thus understanding Se speciation and biogeochemistry in the environment is critical for evaluating risks and developing best management practices. In this presentation results on speciation of Se in rocks, soils, plants, and stream waters located in an active mining region in southeastern Idaho will be presented. Phosphorusore in the region comprises the Western Phosphate Resource Area. The affected samples have elevated concentrations of Se relative to background levels, resulting from distribution of an ore interbed-shale (middle-waste shale) throughout the surficial environment during mine-site reclaimation. Speciation of Se in the samples was investigated using microscopically focused X-ray absorption spectroscopy. This technique utilizes focused synchrotron-generated radiation to excite core electrons in Se atoms within a sample, and fluorescence or transmission can be monitored either spatially or as a function of impinging energy. It is ideally suited for speciation in natural samples because it is element specific, and has a resolution of a few microns, thus allowing for speciation in heterogeneous samples to be investigated. In the middle-waste shale, Se existed as reduced Se(0) or Se(II-) species. Three end-members were identified in the shale: Se-substituted pyrite, an iron selenide mineral, and an organic Se phase. Within the soil, both reduced and oxidized Se phases were detected, with the oxidized phases primarily Se(IV) (selenite); very little Se(VI) (selenate) was detected. In the plant materials both reduced organic Se and Se(VI) were identified. Finally, in the stream sediments, reduced Se, and selenite and selenate were identified. Using the speciation information together with known ecological and biogeochemical processes, we are developing a better understanding of reaction processes, sources, and sinks for Se in the Western Phosphate Resource Area. This detailed information will facilitate a better understanding of the biogeochemical cycling in the system.

1

Department of Geoloy, Portland State University, Portland USA ([email protected], [email protected]) 2 Isotope Geochemistry & Mineral Resources, Department of Earth Sciences, ETH, Zürich, Switzerland ([email protected]) We have investigated volcanic apatite from a well-known center of calc-alkaline magmatism – the Oligocene Central San Juan Caldera Complex (CSJCC), Colorado, USA – to constrain the origin of sulfur rich (> ~0.6 wt.% SO3) apatite found in dacitic to rhyolitic, calc-alkaline magmas. All apatites are small (typically 20-100 μm) and apatite from most units investigated (Fish Canyon Tuff, tuffs and lavas of the San Luis caldera complex) yield SO3 contents ranging from ”0.2 wt.% to values of 0.8–2.0 wt.%; the bulk of apatite, however, indicate SO3 contents of 0.3–0.5 wt.%. We performed laser-ablation ICP-MS analyses on apatite with a range in sulfur concentrations from selected units to correlate variations observed in sulfur with variations in trace elements (e.g REE) to find evidence for melt compositional changes during crystallization of S-rich vs. S-poor apatites. Apatite among units indicates characteristic compositional changes but within single units, apatite tends to form tight compositional clusters in parameters like Eu/Eu*, La/Yb and Sr contents, while REE concentrations may vary by a factor of two. Exceptions are a few distinct apatites. REE concentrations are typically •50-60x of bulk rock and/or of interstitial glass in keeping with other natural systems (e.g. Dempster et al., 2003) and suggest partition coefficients several times the ones of experimental studies. To constrain actual melt composition from which apatites grew, we find the combination of REE, Sr concentrations and Eu/Eu* most useful. Based on the trend of decreasing Sr towards more silicic compositions of bulk rock & glass for the CSJCC (this study and Lipman, 2004) and a DSr • 3, all apatite grew from melts more silicic than ~68 wt.% SiO2 at the exclusion of the most evolved melt composition (~6 ppm Sr, 77 wt.% SiO2) that is too depleted in Eu to yield a DEu consistent with those of neighboring REE. Our study suggests that apatite with low to high S concentrations grew in rhyodacitic/rhyolitic melt requiring either 100’s of ppm of sulfur in the melt or alternative mechanisms to explain upper end of S range in apatite. One such alternative explanation may involve some sort of interraction of S-rich fluids (which could be largely derived from an underplated, degassing mafic magma) with crystallizing apatite.

References Dempster, T.J., Jolivet, M., Tubrett, M.N., Braithwaite, C.J.R., (2003), CMP 145: 568-577. Lipman, P.W. (2004) USGS Open File Rep. 2004-1194.

A980

Goldschmidt Conference Abstracts 2007

Climate change in the southern central Andes at 8 Ma

Gabbroic bodies in the Trinity Ophiolite

M.R. STRECKER1, A. MULCH2, C. UBA1, A. SCHMITT3 4 AND P. CHAMBERLAIN

K. STREMMEL AND G. SUHR Dept. Geol. Min., Uni Köln ([email protected])

1

Institut fuer Geowissenschaften, Universitaet Potsdam, 14476 Potsdam, Germany ([email protected]) 2 Institut für Geologie, Leibniz Universität Hannover, 30167 Hannover, Germany ([email protected]) 3 Dept. of Earth and Space Sciences, UCLA, Los Angeles, CA 90095-1567, U.S.A. ([email protected]) 4 Dept. of Geol. and Environ. Sciences, Stanford University, Stanford, CA 94301, U.S.A. ([email protected]) Being situated in a subtropical high-pressure region with atmospheric subsidence and cold upwelling along the western coast of the continent, the southern central Andes are extremely arid between about 15°S and 27°S. About 20mm/yr rainfall and low erosion rates characterize the Atacama Desert on the western flank of the orogen. With hyper-aridity in this region and less than 200mm/yr rainfall on the adjacent intraAndean Puna-Altiplano plateau and in the intermontane basins E of the Puna, the southern central Andes thus comprise the most arid sector of the orogen. Despite this inherently arid character, the eastern flanks of the ranges intercept moistureladen, easterly winds that result in up to 3000mm/yr precipitation. The cause and the timing for the establishment of this pronounced asymmetry is not known, however. Here, we report on stable C-isotope data obtained from paleosols developed in foreland-basin strata of S Bolivia that serve as proxies for paleoclimatic conditions along the eastern flanks of the Andes. Tephras in these sediments provide the chronostratigraphy for the paleoenvironmental evolution. In two sections our preliminary data show delta G13C values that both decrease from -7 to -8‰ to -11 and -13‰ between about 12.5 and 8 Ma, before they become less negative with an average of -10‰ in the remainder of the profiles. The relatively rapid ~4‰ change in G13C is attributed to a greater availabilty of moisture and increased climatic variability in this inherently dry area of the Andes. Our data suggest that the landscape was characterized by a mixed C3/C4 vegetation cover since approximately 8 Ma. Moisture availability must have been similar to the present humid conditions with a dry winter season. Holocene G13C values are between -9 and 10‰, thus indicating the dominance of a moisture stressed C3/C4 vegetative cover. Thus, present-day atmospheric circulation patterns and the distribution of rainfall are similar to the conditions during late Miocene time. Based on these observations we suggest that precipitation in the southern central Andes was associated with the South American Monsoon and increased at about 8 Ma. We suggest that enhanced precipitation in this region was closely linked to uplift of the Puna-Altiplano and its adjacent eastern orographic barriers that forced the southward displacement of easterly moisture-bearing winds via the Low Level Andean Jet.

The Trinity Ophiolite (N. California) displays discrete silurian to devonian gabbroic bodies hosted by mantle rocks. It might be a good on-land analogue to test the applicability of current accretionary models for slow spreading systems. We report here first results from a mapping campaign in a northern (China Mtn.), central (Bear Creek) and southern gabbroic body (Bonanza King). The central body contains the most regular vertical lithological sequence: mantle peridotite; a shallow dipping transition zone with thick pyroxenitic layers; foliated gabbro; vary-textured gabbro. The northern body displays a disrupted transition zone separated from varytextured gabbro by plastically deformed amphibolized gabbro. The strain is potentially related to extensional ridge tectonics. The southern body exhibits small exposures of wehrlitic and pyroxenitic rocks overlying mantle peridotite; gabbro and gabbronorite with a mineral foliation subparallel to the local dykes; vary-textured gabbro; doleritic dikes and sills. Two kinds of lateral contacts are observed. (1) Xenolithic margins, described by Cannat et al. (1991), demonstrating brittle behavior during emplacement into a cold lithosphere. (2) Dyke-like pyroxenites intruded between mantle peridotite and gabbro. No magmatic strain is obvious. We interprete this as a reactivated contact because of the sharp, sheet-like marginal zone. An inhomogeneous succession of rare peridotite; pyroxenite; doleritic dykes and vary-textured gabbro occurs at the topographic highest parts of two bodies and might represent a roof position. The disrupted character of the transition zones in the northern and southern body suggests the presence of multiple intrusive events which probably caused displacement of existing rock units. Specifically, it is our impression that the abundance of pyroxenites and primitive gabbros is too low relative to the exposed volume of evolved gabbro if a regular, mantle-derived magma is assumed.

Reference Cannat, M., and C. Lecuyer, 1991: Tectonophysics, v. 186, p. 313-328.

Goldschmidt Conference Abstracts 2007

Helium and neon isotopes as mantle tracers N. A. STRONCIK1, S. NIEDERMANN1 AND K. M. HAASE2 1

GeoForschungsZentrum Potsdam, Telegrafenberg, D-14473 Potsdam, Germany ([email protected], [email protected]) 2 Department of Earth Sciences, University of Aarhus, HøeghGuldbergs Gade 2, DK-8000 Aarhus C, Denmark ([email protected]) Noble gas isotopes, especially He isotopes, are widely used tracers for the formation and evolution of the Earth’s mantle and atmosphere. Basic concepts on mantle structure and evolution are primarily based on the interpretation of mantle 3He as reflecting primordial, undegassed mantle material. In general, the relative enrichment of 3He observed in MORBs and OIBs, compared to atmospheric values, is interpreted in terms of the retention of primordial He by the mantle throughout the history of the Earth. Primordial He isotopic ratios in terrestrial matter are largely thought to be solar-like, with the deviation from those solar-like ratios increasing during Earth’s history caused by the production of radiogenic 4He. Thus high 3He/4He ratios are interpreted to represent deep mantle material, whereas ratios around 8 RA (RA stands for the atmospheric 3He/4He ratio of 1.39 x 10-6) are thought to be representative for the upper mantle. Based on He, Ne and Ar fusion data of fresh, submarine volcanic glasses of a number of Mid-Atlantic Ridge off-axis seamounts we show that melt formation and evolution can have a larger impact on He than on e.g. Ne resulting in a decoupled behavior of He from other elements, such as e.g. Ne or Pb. All obtained He data are indistinguishable from the MORB range. In contrast, Ne isotopic compositions are much more primitive than MORB. Combined He, Ne and Ar systematics show that the source region of these seamounts experienced a preferential loss of He compared to Ne and Ar. This He loss, combined with subsequent 4He production, resulted in the decoupling of the He isotope systematics from Ne and Pb. Thus, among He and Ne only Ne has preserved the evidence that a primitive mantle component contributed to the formation of the investigated seamounts. As these seamounts are not fed from a mantle plume being derived from the deep mantle, the primitive Ne component resides within the upper mantle, implying that primitive noble gases are not necessarily indicative for deep mantle material. Our studies point out the necessity of obtaining Ne data in addition to He for the modeling of mantle formation and evolution and correct source characterization.

A981

He-Ne-Ar isotope constraints on the nature and origin of high 3He/4He mantle F.M. STUART AND S. BASU Isotope Geosciences Unit, SUERC, East Kilbride G75 0QF, UK ([email protected]) Olivine phenocrysts from early Tertiary picrites from Baffin Island (BI) and West Greenland have 3He/4He = 38-50 Ra (n=24). The high 3He/4He are consistent with derivation from a mantle reservoir that is relatively undegassed compared to the depleted upper (MORB-source) mantle. Although Ne and Ar concentrations are typically 2 orders of magnitude lower than in basaltic glasses a low blank crusher is allowing Ne and Ar isotope determinations of high-3He/4He olivines. On a conventional 3-isotope plot, the BI picrites are indistinguishable from the Iceland and solar trends; the highest 20 Ne/22Ne is 11.3. 40Ar/36Ar are typically less than 1,000. 38 Ar/36Ar are indistinguishable from air values providing no evidence for solar Ar in the high-3He/4He mantle. Most samples define a trend in 20Ne/22Ne-40Ar/36Ar space that is consistent with mantle end-member with 40Ar/36Ar of 6,0008,000. 4He*/21Ne* and air-corrected 3He/22Ne imply the magmatic noble gases have suffered intense fractionation. This is supported by co-variation of 4He*/40Ar* and 3He/36Ar. The elemental fractionation is consistent with recent magmatic degassing and provides no evidence for an ancient degassing event necessary if the high-3He/4He mantle was a residue of early Earth depletion. The BI picrites plot on a trend in 4 He*/21Ne*-3He/22Ne space defined by basaltic glasses from Iceland. This is distinct from samples of Kola intrusives and we tentatively propose different degassing/depletion histories for the high-3He/4He mantle domains.

A982

Goldschmidt Conference Abstracts 2007

Structural incorporation of Eu(III) into calcite: Process understanding on a molecular level T. STUMPF1, M. MARQUES-FERNANDES2, C. WALTHER1, M. SCHMIDT1, K. DARDENNE1, D. BOSBACH1 AND TH. FANGHÄNEL3 1

Institut für Nukleare Entsorgung, Forschungszentrum Karlsruhe, P.O. Box 3640, 76021 Karlsruhe, Germany ([email protected]) 2 Laboratory for waste management, Paul-Scherrer Institute, 5232 Villigen PSI, Switzerland 3 European Commission, Joint Research Center, Institute for Transuranium Elements, P.O. Box 2340, 76125 Karlsruhe, Germany Demonstrating the geochemical aspects of the long term safety of a nuclear waste repository can significantly be improved by a molecular level understanding of the actinides behavior in the geosphere. In particular the interaction of radionuclides with minerals (adsorption, structural incorporation) strongly affects their mobility and retardation. In this presentation we will focus on the interaction of trivalent actinides and lanthanides with calcite, with special focus on the structural incorporation. Calcite (CaCO3) is an omnipresent mineral in many rocks which are discussed as potential host for a nuclear waste repository. Furthermore, many waste repository designs include cement based components. Calcite is one of the major secondary alteration products formed during the degradation of cement over geological timescales. Actinide and lanthanide partition data derived from co-precipitation experiments indicate a high sorption affinity of these elements to calcite but a comprehensive understanding of actinide and REE uptake by calcite is not yet available. From a geochmical perspective the molecular level substitution mechanism is of key interest. Trivalent actinides and lanthanides have a similar ionic radius compared to Ca, the charge compensation mechanism upon substitution is unclear despite various recent studies. We have studied synthetically doped calcite crystals with Time Resolved Laser Fluorescence Spectroscopy and Extended X-Ray Absorption Fine Structure Spectroscopy. The focus has primarily been on Eu(III), due to its fluorescence properties. The structural parameters of Am(III) doped calcites obtained by EXAFS confirm the substitution of Ca2+ within the calcite structure. Site-selective TRLFS measurements at temperatures < 20 K show the presence of various incorporated molecular species. The incorporation into calcite involves several structurally distinguishable “sites”. These sites may be due to local lattice distortions/relaxations around the metal ion in the calcite structure. These investigations give a mechanistic understanding of the incorporation process and show that the incorporation of trivalent metal ions is not a simple coupled substitution mechanism but involves complex substitution mechanisms.

Bugs in stress: Microbial control of surface reactivity in a stress field C.A. STURM1, M.S. WATERS2, R.S. ARVIDSON1, M. EL-NAGGAR3, S.D. GOODMAN2, K.H. NEALSON3, A. LÜTTGE1 AND F.E. UDWADIA4 1

Dept. of Earth Science, Rice Univ., Houston TX 77005 USA ([email protected], [email protected]) 2 Div. of Diag. Sci., USC, Los Angeles, CA 90089 USA 3 Dept. of Earth Sciences, USC, Los Angeles, CA 90089 USA 4 Dept. of Aerospace and Mechanical Engineering, USC, Los Angeles, CA 90089 USA The reactivity of a solid surface is strongly affected by the imposition of non-hydrostatic stress: the resulting strain rate is a driving force for mass transfer during pressure solution. Previous work [1,2] has shown that bacteria (e.g., Shewanella oneidensis MR-1) recognize and modify crystal surfaces through apparent recognition of surface energy, and are thus potentially sensitive to the distribution of surface stress as well. Here we present the results of a novel integration of vertical scanning interferometry (VSI) with the means to control and measure stress distribution on a solid surface, through controlled deflection of a cantilever beam. These deflection data, measured with VSI at (sub-)nanometer vertical and sub-micron lateral resolution, yield a high resolution map of surface deformation. These data can be compared with Euler-Bernoulli beam bending theory and elastic constants to yield a quantitative prediction of material response. This controlled system can be immersed in a fluid cell innoculated with MR-1 (or other microorganisms) and thus permits the study of bacterial interaction with a stressed surface. This new technique can be used for abiotic systems as well and has a large potential for applications in earth, environmental, and material sciences.

References [1] Lower SK et al. (2001) Science 292:1360 [2] Luttge A Conrad PG (2004) Appl Env Microbiol 70:1627

Goldschmidt Conference Abstracts 2007

Duration of metamorphism in the eclogite type locality K. STÜWE, V. TENCZER, M. PUTZ

Fractional crystallization of monosulfide solid solution from sulfide liquids lead to the PGE enrichment in the Jinchuan Ni-Cu sulfide deposit, western China

AND C. BIERMEIER

Department of Earth Sciences, University of Graz, Heinrichstr. 26; A-8010 Graz; Austria; email: ([email protected]) The eclogite type locality is located in the Kor- and Saualpe region of the Eastern European Alps. The region also hosts one of the largest shear zones of the orogen: The Plattengneiss shear zone. Peak metamorphism occurred around 15 kbar and 700°C and the duration of its metamorphic cycle is loosely constrained by peak ages around 80 Ma and fission track ages around 50 Ma. Despite the high temperatures of peak metamorphism, equilibration of both, major elements and radiogenic isotopes is extremely heterogeneous so that metamorphic conditions and the absolute timing thereof are not very well constrained. The lack of pervasive equilibration suggests that the highest grade metamorphic conditions were only achieved for a very short time and/or at very dry conditions. In a series of projects over the last 10 years we have attempted to constrain both, the water content of the rocks and the time scales of metamorphism using petrological rather than geochronological methods. Although we generally obtain very short time scales and dry conditions, our studies are plagued by the inherent problems of petrological methods. Nevertheless, on the largest scale, we have determined a metamorphic field gradient from north to south across the Koralpe range that shows that metamorphic conditions increase by less than 20°C per kilobar. We suggest that this shallow field gradient may be interpreted in terms of a nonlithostatic pressure gradient. This in turn implies that these conditions prvailed only very briefly.

A983

SHANGGUO SU1, 2, CHUISI LI1, 2, 3, MEIFU ZHOU4, EDWARD M. RIPLEY 3 AND LIANG QI4 1

State Key Laboratory of Geological processes and Mineral Resources, China University of Geosciences, Beijing 100083, China 2 Department of Geology, China University of Geosciences, Beijing 3 Department of Geological Sciences, Indiana University, Bloomington, Indiana, USA 4 Department of Geological Sciences, University of Hong Kong, Hong Kong, China Discordant lenses of Pt-Pd enriched zones (ores bearing up to 1.0 ppm of Pt or Pd) have recently been identified in the sulfide-bearing peridotite of the Jinchuan Cu-Ni-PGE (Platinum group element) sulfide deposit, China. Chalcopyrite, pyrrhotite, and pentlandite occur in both Pt-Pd enriched zones and normal ores, but Cu-bearing minerals such as cubanite and Bi-, Te-, and As-bearing minerals are more abundant in the PtPd enriched zones. Sperrylite is the major Pt-host minerals in the Pt-Pd enriched zones interstinally and occurs mainly as euhedral grains within base-metal sulfides which occur among the cumulates of olivine. PGE-enrichment is found only in sulfide-bearing samples. In orebody  1 and orebody 24, Rh, Ru, and Ir are positively correlated, but a negative Ir-Pd and Ir-Pt correlation. However in orebody 2 Rh, Ru, Pt, Pd and Ir are positively correlated. Taken together, the elemental correlations and mineralogical data support a model for the origin of Pt-Pd rich ores in Orebody 1 and Orebody 24 of the Jinchuan deposit are consistent with fractional crystallization of monosulfide solid solution from sulfide liquids on cooling; The origin of Orebody 2 involves variable magma/sulfide liquid mass ratios (R-factors).

A984

Goldschmidt Conference Abstracts 2007

New 40Ar/39Ar and K-Ar ages from Macolod Corridor, SW Luzon, Philippines: Implication for its volcanic history M. SUDO1, E.L. LISTANCO2, N. ISHIKAWA3, T. TAGAMI4, H. KAMATA3 AND Y. TATSUMI5 1

University of Potsdam, Institute of Geosciences, Germany ([email protected]) 2 University of Philippines, NIGS, Philippines ([email protected]) 3 Kyoto University, HES, Japan ([email protected], [email protected]) 4 Kyoto University, Graduate school of science, Japan ([email protected],) 5 JAMSTEC IFREE, Japan ([email protected]) Macolod Corridor, southwestern Luzon, Philippines, is an extensive Quaternary volcanic field including Taal caldera and many monogenetic/polygenetic volcanoes. Recently reported K-Ar or 14C ages (Listanco, 1994; Sudo et al., 2000) suggest that several volcanoes have younger ages than e.g., 0.14 Ma. However, the history of the possible caldera, located at the central lobe of the lake Laguma de Bay, at the north of Macolod Corridor has not been well understood. Radiometric dates remain few other than the K-Ar age, 1.84±0.07 Ma (error: 1V), from an essential scoria (sample name 013006; Sudo et al., 2000) with a size of ~30 cm in the pyloclastic (scoria) flow at the northeastern rim, or the 14C ages of 5000 to 47000 yBP, from pyroclastic materials between Laguna de Bay and Metropolitan Manila. In this study, the essential scoria (sample name P4-2), smaller than 10 cm, included in the pyroclactic (scoria) flow near Teresa city located at the north of Laguna de Bay, and the scoria, 013006, have been dated at the new 40Ar/39Ar chronology laboratory in the University of Potsdam. The system used consists of a continuous CO2 laser, Micromass 5400 noble gas mass spectrometer and the ultra-high vacuum tubes adopting SAES getters and a cold trap. The prepared samples were irradiated for 96 hours at the reactor in the Geesthacht Neutron Facility (fast neutron flux: 1x1012 n/cm2/s) together with the Fish Canyon Tuff sanidine and crystals of CaF2 and K2SO4. The obtained 40Ar/39Ar plateau ages are 1.85±0.01 Ma and 1.83±0.02 Ma from P4-2, while 1.92±0.03 Ma from 013006. With the K-Ar age, 1.82±0.05 Ma (unpublished), obtained from the scoria in the same outcrop for 013006, all the ages show agreement within 2 sigma error. These ages would be the constraints for solving the history of the caldera at Laguna de Bay.

References Listanco, E. (1994) Dr. Thesis, University of Tokyo Sudo, M., E. L. Listanco, N. Ishikawa, T. Tagami, H. Kamata and Y. Tatsumi (2000) Journal of the Geological Society of the Philippines, 54, 89-104

Gabbros drilled by IODP Leg 305, 30°N, Mid-Atlantic Ridge G. SUHR1, E. HELLEBRAND², K.P JOCHUM 3. AND LEGS 304/305 SCIENTIFIC PARTIES 1

Geol.-Mineral. Inst., Uni-Koeln, ([email protected]) Dept. Geol. Geophy., Univ. Hawaii, ([email protected]) 3 Max Planck Inst. Mainz, ([email protected]) 2

IODP Hole 1309D (Legs 304/305) penetrated 1415 m into the core-complex of the Atlantis Massif at 30°N, Mid-Atlantic Ridge. Of the 75% recovered rock, 96% is gabbroic. Possibly, a larger-scale magmatic cycle is preserved between 600 und 1240 mbsf, as marked by olivine-rich troctolites at the base and more common gabbronorites near the top. In detail, however, internal magmatic contacts are numerous with more evolved rocks intruding into less evolved ones. Here we present an overview of the 800-1200 mbsf interval and a detailed study of the transition from evolved rocks below 1235 mbsf to overlying primitive gabbros. Geochemically, there is a good correlation between the REE in cpx, the Mg# in cpx, and lithological evolution. The data can be modeled as increments of batch fractionation ranging from 5 to >80% from a primitive N-MORB. There is no immediate need for magma replenishment. The rims of nearly all clinopyroxenes were overprinted by an evolved melt. A three stage model is required to explain the data set: (1) a cumulus phase; (2) displacement of residual melt and formation of new cumulate bodies from it. These two processes can explain the presence of well-equilibrated, but differently evolved cores of clinopyroxene. (3) Compaction of residual melt and differentation of it as it migrates through the cumulus pile. This stage explains the late overprint of the clinopyroxene rims.

Goldschmidt Conference Abstracts 2007

A985

‘Green’ geosequestration: Secure carbon sequestration via plant silica biomineralisation

Subducted noble gas and halogen preserved in wedge mantle peridotite from the Sanbagawa belt, SW Japan

L.A. SULLIVAN1,2 AND J.F. PARR 1,2

H. SUMINO1, 2, R. BURGESS2, T. MIZUKAMI3, S. R. WALLIS3, AND C. J. BALLENTINE2

1

Southern Cross University, PO Box 157, Lismore, NSW 2480, Australia ([email protected]) 2 Plantstone Pty Ltd, 90 Zouch Road, Stoney Chute, 2480, NSW, Australia Phytoliths form via biomineralization processes as microscopic grains of silica in plants, especially grasses. During plant growth a proportion of the organic carbon produced by plants is encapsulated by silica within the microscopic phytoliths. This phytolith occluded carbon (PhytOC) usually comprises much less than 1% of the dry weight of plants, and less than 10% of the total carbon pool in grassland topsoils (with ages of < 200 years). However, data from soil chronosequences shows that the high resistance of the PhytOC fraction (relative to the other soil carbon fractions) against decomposition processes results in PhytOC comprising ~70% of the total carbon pool in grassland topsoils that have been isolated for >3,000 years from fresh plant material addition by burial. Therefore, unlike most plant matter, which readily decomposes returning CO2 to the atmosphere, the organic carbon occluded in phytoliths effectively sequestrates carbon in soils and sediments in a very secure manner. This ‘green’ geosequestration process (i.e. occlusion of carbon in phytoliths) is currently responsible for the secure sequestering within soil of ~300 million tonnes of CO2 equivalent per year globally. More over different plant types biomineralise silica and yield PhytOC at greatly different rates. Some major crops produce over 100 times more PhytOC than other major crop types. Furthermore, varieties within a single major crop type, such as sugar-cane and sorghum, have been found to produce widely differing quantities of PhytOC. This suggests that crop/cultivar choice decisions by farmers and foresters etc have a considerable impact on the amount of soil carbon sequestered and are a significant contributing factor affecting the global carbon cycle. It follows that the management of PhytOC in crops, pastures, forests etc has the potential to greatly enhance the current rates of secure terrestrial carbon sequestration.

1

Laboratory for Earthquake Chemistry, Graduate School of Science, University of Tokyo, Tokyo, Japan ([email protected]) 2 SEAES, The University of Manchester, Manchester, UK 3 Department of Earth and Planetary Sciences, Graduate School of Environmental Studies, Nagoya University, Nagoya, Japan Water-rich fluids released from subducting slabs play an important role in arc volcanism. Indeed, subduction volcanism is thought to efficiently return volatiles contained in subducting material back to the Earth’s surface. Less than 100% removal of the volatile component may result in volatile recycling into the deep mantle [1]. The volatile composition of subducting fluids is however, not well characterised. The Higashi-akaishi peridotite body in the Sanbagawa metamorphic belt, southwest Japan, is possibly a unique example of a km-scale sliver of a former mantle wedge exhumed from depths of at least 100 km. Serpentine dominated micro-inclusions in olivine grains in the peridotite are regarded as relics of former water-rich inclusions developed in the wedge mantle above a subducting slab. Thus, it is expected that these micro-inclusions should preserve characteristics of slab-derived fluids. Determination of their compositions could provide important geochemical constraints on subduction zone processes. Noble gas and halogen determination of the microinclusions has been carried out using noble gas isotope analysis of both neutron-irradiated and unirradiated samples. The following isotopic characteristics have been determined: (1) 3He/4He ratios of 1.4-1.8 Ra represent a mixture of mantle and radiogenic He; (2) 40Ar/36Ar ratios up to 470 are close to the atmospheric ratio with a small contribution of mantle and/or radiogenic Ar; (3) Seawater-like noble gas elemental ratios enriched in heavy noble gases; and (4) halogen (Cl, Br, and I) composition is similar to marine pore fluids and brines. These characteristics imply that noble gases and halogens with compositions little different to marine pore fluids are injected into the mantle wedge just above the subducting slab. The subducted halogen and noble gas elemental ratios are clearly distinct from those of arc volcanic gases. This implies that the Higashi-akaishi peridotite body has frozen in and preserved an inferred but previously unseen part of the volatile recycling process. Return of these volatiles to the atmosphere via arc volcanism requires the addition of a mantle component and fractionation during degassing. A small proportion preserved in the downgoing slab can explain the heavy noble gases observed in the convecting mantle.

Reference [1] Holland G. and Ballentine C. J. (2006), Seawater subduction controls the heavy noble gas composition of the mantle, Nature 441, 186-191.

A986

Goldschmidt Conference Abstracts 2007

Using reflectance spectroscopy for the reconstruction of penguin palaeoecological process in Antarctic ornithogenic sediments JING SUN1, XIAODONG LIU1, LIGUANG SUN1 2 AND WENQI LIU 1

Institute of Polar Environment, University of Science and Technology of China, Hefei, Anhui 230026, P. R. China ([email protected]; [email protected]) 2 Structure Research Laboratory, University of Science and Technology of China, Hefei, Anhui 230026, P. R. China The present study explored the application of the rapid, cost-effective, non-destructive and simultaneous technique of reflectance spectroscopy within visible-near-infrared region to infer penguin palaeoecological records in the maritime Antarctic. A total of 106 samples taken from four sediment cores (Y2, Y2-4, Y4, AD3) on the Ardley Island were measured by both chemical and spectral methods. These cores were previously reported to be influenced by penguin guano, and nine elements including sulfur (S), phosphorus (as P2O5), calcium (as CaO), copper (Cu), zinc (Zn), selenium (Se), strontium (Sr), barium (Ba), and fluorine (F) were identified as bio-elements; their concentrations could be used as inorganic geochemical indicators for tracking historical penguin population change. The reflectance (r) and its derived indexes were employed to develop calibrations for predicting nine bioelement concentrations, using stepwise multiple linear regression (s-MLR) and principal component regression (PCR) approaches. R between optimal spectra-predicted and chemically analyzed concentrations were Ba: 0.894, all the other eight elements >0.954 for s-MLR; Ba: 0.926, all the other eight elements >0.963 for PCR. Furthermore, principal component analysis (PCA) was performed on all the reflectance spectra data and the results showed that the first two factors were able to account for 98.9% of the variance of the data. The first PCA factor(PC1), accounting for 95.8% of the total variance, could be explained to bear the information of the content of penguin guano, and thus the PC1 score against depth of the samples (curve A) indicated the fluctuation of historical penguin population. In addition, by using spectra of pure guano and pure soil, these absorbance spectra data (log1/r) of ornithogenic sediments were linearly separated to guano part and soil part. The penguin population change inferred from the separated guano proportion (curve B) was consistent with curve A, and both curves showed similar historical change trends as inferences from inorganic elements and isotopes. Overall, this study demonstrated that using reflectance spectroscopy to infer palaeoecological information recorded in Antarctic ornithogenic sediments is feasible. Acknowledgment This research was supported by the NNSF (No. 40606003).

The coupling between plate subduction and intraplate evolution in eastern China W.D. SUN1,2, X. DING1, Y.H. HU1 AND X.H. LI1 1

Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Wushan, Guangzhou 510640, P.R. China. ([email protected]) 2 School of Earth and Space Sciences, University of Science and Technology of China, Hefei, 230026 Plate interactions along subduction zones usually cause deformations of overlaying crusts as shown in the Andes. Numerical modelling shows that plate interaction during plate subduction is of critical importance for intraplate tectonic evolutions of the American continent [1]. Eastern China was an active continental margin related to the subduction of the paleo-Pacific plate under Eurasia from Jurassic to Cretaceous. This continent is well known for the removal of subcontinental lithosphere mantle with complicated geological evolutions, which leads to different models, ranging from extension to subduction-related compression and crust delamination etc. We find that the Cretaceous tectonic evolution in eastern China matches remarkably well with the drifting history of the paleo-Pacific plate. The most pronounced phenomenon is that the eastern China large-scale orogenic lode gold mineralisation occurred at about 125 Ma [2], concurrent with the major shift in the drifting direction of the subducting paleo-Pacific Plate [3] and the formation of the Ontong Java Plateau. Given lode gold deposits usually formed onset of compressional or transpressional deformations, the lode gold deposits dated the major tectonic change from extension to transpression in eastern China [4], consistent with the subduction regime and other geological records in the region. The Early Cretaceous drifting history of the paleo-Pacific also matches with other tectonic and magmatic evolutions in eastern China, suggesting that the major geological events in eastern China in the Early Cretaceous have been mainly controlled by the subduction of the paleo-Pacific plate, and that plate interactions are important driving forces for intraplate tectonic evolution in general. This provides a new angle of view to understand the tectonic evolution of the eastern Euroasian continents, e.g., the mechanism behind lithosphere thinning in eastern China as well as the evolution of the Tan-Lu Fault.

References [1] Sobolev S.V. and Babeyko A.Y. (2005) Geology 33, 617620. [2] Yang J.H. and Zhou X.H., (2001) Geology 29, 711-714. [3] Koppers A.A.P., Morgan J.P., Morgan J.W. and Staudigel H., (2001) EPSL 185, 237-252. [4] Zhang Y.Q., Dong S.W. and Shi W., (2003) Tectonophysics 363, 243-258.

Goldschmidt Conference Abstracts 2007

A987

Ar/39Ar dating of muscovite from the Maofeng Granite, N-Guangdong Province, China

Oxygen isotopic zonality at the Iultin Sn-W Deposit (Chukotka, North-East of Russia)

Z.X. SUN, W. ZHANG, B. GAO

T. SUSHCHEVSKAYA 1, M. SPASENNYKH 1, A. IGNATIEV 2, 2 AND T. VELIVETSKAYA

40

East China Institute of Technology, Fuzhou, Jiangxi 344000, China ([email protected]) The Maofeng Granite, located in the Guidong granite composite, N-Guangdong Province, China, has been paid much attention to by geoscientists because of its hosting several big uranium deposits. Two muscovite concentrates have been prepared from representative samples collected from the outcrop of the granite body. They were analysed using incremental-release 40Ar/39Ar method. The 40Ar/39Ar apparent ages of the two samples are shown in Fig. 1.

Figure 1. 40Ar/39Ar apparent ages of muscovite concentrates from the Maofeng Granite. Plateau ages are listed on each spectrum. The plateau ages ranging from 131.2r0.5 to 136r0.9 Ma are much different from the single grain ziron U-Pb ages varying from 207.6r3.2 to 219.6r0.9 Ma of the Maofeng Granite reported by previous researchers. In the area, one of the major uranium mineralization periods was dated back to 122 to 138 Ma ago. These evidences suggest that the muscovite was formed from the hydrothermal fluid which resulted in uranium mineralization, and that the plateau ages did not reflect the age of the Maofeng granite itself but the age of the U ore-forming hydrothermal activity. This work is funded by China National Foundation of Natural Sciences under the Project No. 40472147.

1

Vernadsky Institute of Geochemistry and Analytical Chemistry RAS, Moscow, Russia ([email protected]) 2 Far East Geological Institute RAS, Vladivostok, Russia, ([email protected]) The oxygen isotope composition of host rocks and main vein minerals was studied at the large well known Sn-W Iultin deposit with application to the role of mixing fluids of different origins in the deposition of Sn-W ores in graniterelated hydrothermal systems. The deposit is located in the exo- and endocontact of leucogranite stock (K2) and is formed by the seria of the proximate quartz veins in hornfelsized sandstone-schist rocks (T1-2). The veins and greisens are largely composed of the productive mineral assemblage (quartz, muscovite, cassiterite, wolframite, arsenopyrite, subordinate beryl, scheelite), formed from sodium chloride boiling solutions enriched in CO2 and CH4 at T 270-350ºC, P 0,5-1,0 kbar as it followed from fluid inclusion data. The postore sulfide and fluorite-carbonate assemblages are poorly developed. The oxygen isotope composition of the rocks was examined in: a) metamorphosed sedimentary rocks of outer part of hydrothermal system, b) the contact part of the leucogranite cupola, c) the altered wall rocks in the central part of the deposit across the large ore body (62/50) and d) the wall rocks of the veins (7,10), poor in Sn-W ores. It was found that large (hundred meters) low-18O zones have been formed in the central part of the Iultin hydrothermal system. The į18O values decreased monotonously from the initial 12‰ (1km from the contact) to 3-5‰ at the contact with granite. Substantial 18O decrease in the contact zone, formed by greisenized granites, is the result of active movement of hydrothermal solutions and is explained by higher permeability in comparison with metamorphosed sedimentary rocks (Spasennykh et al., 2002). During the development of vein ore bodies the wall rocks were altered insignificantly in oxygen isotope composition. The high į18O values of minerals of productive association bore witness to magmatogenic source of the fluids. The notable depletion of the wall rocks in 18 O took place during the final stage, when meteoric waters dominated in the hydrothermal system. That time no significant ascending fluid flows were focused within the veins. The initial picture of oxygen isotope zonality, connected with ore deposition, had been wiped away by active interaction of exogenic waters with host rocks. This study was supported by the Russian Foundation for Basic Research (project nos. 03-05-65036, 07-05-00432)

Goldschmidt Conference Abstracts 2007

A988

Synchrotron-based studies of fluids, mineral-water interfaces and glasses

Raman spectroscopy of organics in Antarctic micrometeorites

S. R. SUTTON1,2, M. NEWVILLE2, P. ENG2, M. L. RIVERS1,2, AND S. K. GHOSE2

A. SUZUKI1, Y. YAMANOI1, S. NAKASHIMA1, T. NAKAMURA2 AND E. KATAFUCHI2

1

Department of Geophysical Sciences CARS, University of Chicago, Chicago, IL ([email protected])

2

Synchrotron-based analytical methods have proved to be valuable in a wide variety of studies of fluids, glasses and mineral-water interfaces. This presentation is an overview of experiments of this nature being conducted at the GSECARS facility at the Advanced Photon Source (USA). X-ray absorption fine structure (XAFS) spectroscopy is one of the principal methods used in these studies. One such microbeam application involves the determination of valence states for multivalent elements in igneous glasses, which in turn are used as proxies for oxygen fugacity inferences for both terrestrial and extraterrestrial magmas (e.g., Sutton et al. 2005). MicroXAFS is also used in ore-relevant studies to establish the speciation of metals in hydrothermal fluids. These applications involve measurements on fluid inclusions (both natural and synthetic) where the samples are analyzed in situ above their homogenization temperature using a heating stage (e.g., Berry et al. 2006). Mineral-water interface studies aim to establish the atomic-scale structures of mineral surfaces, the structural and reactivity changes that occur during hydration and the resultant metal sorption properties of these surfaces. These in situ experiments typically involve combinations of x-ray reflectivity, crystal truncation rod, x-ray standing waves and grazing-incidence XAFS methods. A focus of this work is the determination of structural and reactivity changes of hydrated metal oxide surfaces (e.g., Eng et al. 2000). Fluids can be imaged within objects using x-ray computed microtomography (CMT). One aspect involves the determination of the distribution and flow properties of multiphase fluids in soil columns (e.g., Culligan et al. 2006). CMT methods can also reveal transport paths of metals in plants (e.g., McNear et al. 2005). References Sutton S.R., Karner J., Papike J.J., Delaney J.S., Shearer C.K., Newville M., Eng P., Rivers M., and Dyar M.D., (2005), Geochim. Cosmochim. Acta 69. 2333-2348. Berry A.J., Hack A.C., Mavrogenes J.A., Newville M., and Sutton S. R., (2006), Am. Mineral. 91. 1773-1782. Eng, P.J., T.P. Trainor, G.E. Brown, Jr., G.A. Waychunas, M. Newville, S.R. Sutton, and M. Rivers, (2000), Science 288. 1029-1033. Culligan K.A., Wildenschild D., Christensen B.S.B., Gray W.G., Rivers M.L. (2006), Adv. Water Res. 29. 227-238. McNear D.H., Peltier E., Everhart J., Chaney R.L., Sutton S., Newville M., Rivers M., and Sparks D.L., (2005), Environ. Sci. Technol. 39. 2210-2218.

1

Department of Earth and Space Science, Osaka University, 1-1 Machikaneyama-cho, Toyonaka-shi, Osaka 560-0043, Japan ([email protected]) 2 Department of Earth and Planetary Sciences, Kyushu University, Hakozaki, Fukuoka 812-8581, Japan

Introduction Micrometeorites are expected to be the major source of the C flux from extraterrestrial materials delivered to the earth and might represent a major contribution to the pre-biotic organic matter on the early earth (Maurette 2006). On the other hand, organics from carbobaceous chondrites have been characterized non-destructively by means of Raman microspectroscopies (Raynal, 2003; Matrajt et al., 2004).

Methods Raman spectra have been measured on 73 points of 11 unmelted Antarctic micrometeorite (AMM) grains pressed on aluminum foils under a Raman microscope. Peak positions, full widths at half maximum (FWHM), band intensity ratios (ID/IG) and band area ratios (AD/A(D+G)(%)) for D (defect: ~1360 cm-1) and G (graphite: ~1600 cm-1) bands were determined for the AMMs and compared with the literature data on carbonaceous chondrites and other cosmic materials.

Results and Discussion The peak position and FWHM of G band of the AMMs are ranging from 1581 to 1593 cm-1 and from 87 to 133 cm-1, respectively. These value ranges are almost the same as those for CM2 carbonaceous chondrites. The peak position and FWHM of D band for the AMMs are from 1357 to 1379 cm-1 and from 166 to 271 cm-1, respectively. This region for the AMMs mostly overlaps with the region for CI1s. The D peak position range of the AMMs is similar to those for CM2 and CR2 chondrites. The ID/IG and AD/A(D+G)(%) are ranging from 1.1 to 1.7 and from 65 to 78, respectively. These distributions of the AMMs are close to Orgueil carbonaceous chondrite (CI1). AD/A(D+G) band area ratios (%) of the AMMs are in the similar range to those for CR2 and Tagish Lake carbonaceous chondrites. These Raman features of macromolecular carbonaceous materials in the AMMs are similar to those in C1 and C2 chondrites. The AMMs studied here might have the aqueous alteration level higher than 2 and have possibly delivered useful organic components to the early earth.

References Matrajt G., Borg J., Raynal P. I., Djouadi Z., D’Hendecourt L., Flynn G. and Deboffle D. (2004), A&A 416 983-990. Maurette M. (2006), Micrometeorites and the Mysteries of Our Origins. (Springer) Raynal P. I. (2003), Ph. D. Thesis, Université Peirre et Marie Curie, Paris, France

Goldschmidt Conference Abstracts 2007

Osmium behavior in a subduction zone setting elucidated from Cr-spinel sands of boninites and tholeiite in Bonin islands K. SUZUKI, R. SENDA AND K.SHIMIZU

A989

Reduction of U(VI) by Shewanella putrefaciens in the presence of organic acids Y. SUZUKI1, T. NANKAWA1, T. OZAKI1, T. OHNUKI1 AND A.J. FRANCIS2 1

Institute for Reseach on Earth Evolution (IFREE), Japan Agency for Marine-Earth Science and Technology (JAMSTEC), Yokosuka, Japan ([email protected]) Osmium (Os) isotope is a sensitive tracer of crust and sediment as its eroding form, because of significant contrast between a high Os isotope ratio of crust and sediments and a low ratio of mantle (187Os/188Os>0.5) (187Os/188Os<0.13). Therefore, Os isotope provides information on slab components contribution to the mantle source of island arc lavas (Brandon et al., 1999; Borg et al., 2000; Alves et al., 2002). Radiogenic Os isotopic compositions are commonly found in volcanic lavas and peridotite xenoliths in a subduction zone setting. Though most of the authors had attributed these high 187Os/188Os to the input of the slab component to the mantle source, some pointed out that assimilation of crustal materials during magma ascent possibly reproduce the elevated Os isotope ratios of the arc lavas (Lassiter and Luhr, 2001; Woodhead and Brauns, 2004). Since then, whether the high Os isotope ratios of arc lavas are caused by contribution of slab component or by assimilation of the overlying crust has long been highly debated. Here we report the unradiogenic Os isotopic ratios of Crspinel sands from Chichi-Jima and Yome-Shima boninites and the significantly high 187Os/188Os of Cr-spinels of the Mukoojima tholeiite, Izu-Bonin arc. As Cr-spinel is resistant to later alteration and weathering and, more importantly, is the early stage crystal in the fractional crystallization, it preserves the chamical and isotopic compositions of very primitive magma in its melt inclusion and spinel itself without any later stage crustal contamination. Therefore, extremely high 187Os/188Os of 0.1429 of Mukoo-jima tholeiite is most likely caused by inputs from slab components, not by assimilation of the overlying crust. Unradiogenic 187Os/188Os of Cr-spinel of Yome-shima and Chichi-jima boninites (0.1232 and 0.1242, respectively) indicate that Os is not mobile under a boninite formation condition. As a result, we demonstrate that Crspinel is the most useful mineral to decode the Os isotopic compositions of primitive arc magmas.

References Alves S., Schiano P., Capmas F. & Allegre C. J., (2002) Earth Planet. Sci. Lett. 198, 355-369. Borg L. E., Brandon A. D. Clynne M. A. & Walker, R. J., (2000) Earth Planet. Sci. Lett. 177, 301-317. Brandon A. D., Becker, H., Carlson, R. W. & Shirey S. B., (1999) Chem. Geol. 160, 387-407. Lassiter J. C. & Luhr, J. F., (2001) Geochem. Geophys. Geosyst. 2, 2000GC000116. Woodhead J. & Brauns M., (2004), Earth Planet. Sci. Lett. 221, 309-323

Research Group for Hevey Element Biogeochemistry, Japan Atmic Energy Agency, 2-4, Shirakatashirane, Tokai, Ibaraki 319-1194, Japan ([email protected]) 2 Brookhaven National Laboratory, Upton, New York 11973, USA ([email protected])

Introduction The oxdation state of uranium is one of the important factors that affect its migration in the environment. Although microbial reduction of U(VI) has been extensivly investigated, limitted information is available of the effects of organic acids. We examined the reduction behavior of U(VI) by Shewanella putrefaciens in the organic acid solutions. Experimental Shewanella putrefaciens was incubated in anarobic medium at pH 7 that contained 1mM UO22+, 50 mM lactic acid and 100 mM organic acids (acetic, oxalic, malonic, succinic, adipic, malic, tartaric, citric acid or EDTA). Aliquots of medium were periodically withdrawn, and filtered. Dissolved uranium concentrations and UV-vis spectra of the aliquots were measured. Precipitates generated were analyzed by XANES and SEM. The medium without organic acid (except for lactic acid) was examined as a control. Results and Disccussion In the control medium and the media containing acetic or adipic acid, dissolved uranium decreased with time, and precipitates were obserbed. The XANES spectra of the precipitates showed that the precipitates contained U(IV). The SEM analysis showed that the precipitates was uraninite (UO2). In the other media, dissolved uranium was almost constant. The UV-vis spectra showed that dissolved U(VI) was reduced to U(IV), and it was present as U(IV) organic complex in the medium containing oxalic, tartaric, citric acid or EDTA. No evidence of U(VI) reduction was observed in the media with succinic or malic acid. These results suggest that the reduction behaviors of U(VI) by S. putrefaciens are categorized into three cases depending on organic acid.

Goldschmidt Conference Abstracts 2007

A990

Provenance and post-sedimentary low-temperature evolution of the James Ross Basin sediments (Antarctic Peninsula) based on zircon and apatite fission-track analysis 1

1

2

M. SVOJTKA , M. MURAKAMI , D. NÝVLT , J. MACÁKOVÁ3, J. FILIP1 AND P. MIXA2 1

Institute of Geology, Academy of Sciences, Rozvojová 135, Praha, Czech Republic ([email protected], [email protected], [email protected] 2 Czech Geological Survey, Klárov 3, Czech Republic, ([email protected], [email protected]) 3 Department of Mineralogy and Geochemistry, Charles University, Albertov 6, Praha, Czech Republic, ([email protected]) James Ross Basin contains one of the thickest and most complete Jurassic-Paleogene age sedimentary successions anywhere in the Southern Hemisphere. The basin is filled by a sequence of arc-derived clastic and volcaniclastic marine sediments through the Late Jurassic to Late Eocene. In order to reconstruct thermal history for basin lowtemperature evolution of potential sedimentary resources and post-sedimentations succession, we dated detrital zircons and apatites from the sequence of sandstones collected from the James Ross and Seymour Islands using fission-track (FT) thermochronology. All zircon FT ages are older than apatite FT ages provided in the identical individual rocks. Provenance of individual FT zircons and apatites ages varies in wide spread of Carboniferous to Early Paleogene ages between ~60 to ~350 Ma. Jurassic-Cretaceous ages of northwest James Ross Island are probably compatible with derivation of sediment from western lying Mt. Reece and Mt. Bradley region, where the rocks of the Antarctic Peninsula batholith appears. Sediments from Seymour Island are probably originating from Trinity Peninsula Group and Antarctic Peninsula Volcanic Group. Shortening of tracks was due to subsequent volcanic/magmatic activity before transport of rocks and deposition into the James Ross Basin or alternatively, due to volcanic reheating after deposition. Time-temperature modelling of the apatite fission-track samples from Seymour Island (Marambio and Seymour Island Groups) shows a similar thermal history style, involving a period of total thermal annealing and subsequent cooling (erosion/denudation). The samples were above 120ºC until about the Upper Triassic (220 – 210 Ma) age and then followed by the period of relatively quick Neogene and Quaternary exhumation (since ~30 – 40 Ma) to the present erosion surface.

Isotopic and trace element evidence for groundwater discharge in the coastal zone P.K. SWART 1, J STALKER 2 AND R.M. PRICE3 1

MGG/RSMAS, University of Miami, Fl 33149 ([email protected]) 2 Department of Earth Sciences, Florida International University, Fl 33199 ([email protected]) 3 Department of Earth Sciences and SERC, Florida International University, Fl 33199 ([email protected]) Determination of the relative importance of freshwater discharge from surface and groundwater sources into coastal estuaries and lagoons is important in making assessments regarding water control policies. In this study, variations in GD, G18O, G13C, and concentrations of Sr, Ba, and Mg relative to Ca have been used to disinguish the input of fresh groundwater from rainwater and surface discharge into coastal ecosystems of South Florida (Biscayne Bay and Florida Bay). These tracers are able to distinguish between these various sources, because rainwater has relatively negative G18O and GD values and low concentrations of cations. Groundwater also has depleted G18O and GD, but low Sr/Ca, Mg/Ca, and Ba/Ca ratios. In contrast, surface waters have relatively positive G18O and GD values, intermediate concentrations of Ca, high concentrations of Ba, and negative G13C values (SWART and PRICE, 2002). The differences in the geochemical parameters arise because (i) surface waters in South Florida are highly evaporated thereby enriching GD and G18O values, (ii) surface waters are highly influenced by surface organic activitity which produces waters depleted in G13C and elevated in Ba, (iii) groundwaters dissolve the local carbonate rocks which contain high concentration of Ca, but relatively low concentrations of Ba, Sr, and Mg; these waters also contain relatively positive G13C values. Our results indicate that within Florida Bay, the input of fresh groundwater contributes an insignificant amount to the hydrological balance. Instead the salinity is controlled by runoff and precipitation. In contrast in Biscayne Bay, groundwater, surface water, and direct precipitation all contribute equally to the salinity balance.

Reference Swart P. K. and Price K. (2002) Oceanography 47, 1234-1241.

Limnology

and

Goldschmidt Conference Abstracts 2007

In situ U/Pb geochronology of baddeleyite by LA-ICPMS P.J. SYLVESTER, M.N. TUBRETT AND A.K. SOUDERS MicroAnalysis Facility, INCO Innovation Centre and Department of Earth Sciences, Memorial University, St. John’s, NL A1B 3X5, Canada ([email protected]) Baddeleyite (ZrO2) is a useful U-Pb chronometer for determining magmatic crystallization ages of silica-poor rocks that failed to form zircon. In situ baddeleyite dating has been plagued by analytical problems, however, limiting its application. Crystal orientation effects produce unacceptably large variations in 206Pb/238U ratios measured by ion microprobe so that ages must be determined from 207Pb/206Pb ratios alone. Laser ablation-inductively coupled plasma mass spectrometry (LA-ICPMS) analyses have produced 206Pb/238U and 207Pb/235U ratios that plot in arrays extending far above concordia, suggesting severe laser-induced Pb/U fractionation. Correction procedures for instrumental mass bias, in particular the extent of matrix effects between baddeleyite and common zircon standard reference materials (SRMs) used for calibration of measured isotope ratios, are poorly defined. We have dated two baddeleyites of known (TIMS) age using a Finnigan ELEMENT-XR magnetic sector ICPMS coupled to a GeoLas 193 nm ArF excimer laser ablation system. The baddeleyites are from carbonatite in the Phalaborwa Complex in South Africa (PHB; 2059.8± 0.8 Ma) and gabbroic anorthosite of the Duluth gabbro complex at Forest Center, Minnesota (FC-1; 1099.0± 0.6 Ma). Two methods commonly used for LA-ICPMS U-Pb dating were tested. The first involved spot analysis (40 micron beam, 5 Hz, 3 J/cm2) and calibration of isotopic ratios by standard-sample-standard bracketing with zircon 91500 as the calibrant. The second employed raster analysis (10 micron beam, 10 Hz, 5 J/cm2, making a 40 by 40 micron box pattern) and Pb/U determinations by the “intercept” method. Mass bias corrections are made by reference to a tracer solution of Tl-Bi233 U-237Np (with a composition defined by reference to zircon SRMs) aspirated into the argon plasma at the same time as sample ablation. Both spot and raster analyses exhibited laserinduced U/Pb fractionation during analysis, but the degree of fractionation was much more severe for the spot analyses. PHB baddeleyite exhibited much more fractionation than FC-1 baddeleyite. The reason is unclear but may be related to polysynthetic twinning on {100} being less well-developed in FC-1. Pb and U are heterogeneously distributed in the PHB baddeleyite: there is much less scatter on Pb/U ratios when it is drilled perpendicular to the {100} planes than when drilled parallel to them, suggesting that Pb/U ratios are more homogeneous at the scale of laser sampling in this orientation. Both zircon SRM and tracer solution calibration methods produced inadequate mass bias corrections for measured Pb/U and 207Pb/206Pb ratios and erroneous ages for analysed baddeleyites. LA-ICPMS U-Pb ages with accuracies comparable to those determined for zircon can be attained using raster analyses, intercept corrections, and a tracer solution calibrated for baddeleyite.

A991

Efficiency of Cl recycling during subduction of oceanic crust: Constraints from melt inclusions in HIMU lavas LINDSAY ANN SZRAMEK

AND JOHN C. LASSITER

The University of Texas at Austin, Geol Science Dept, 1 University Station C1100, Austin TX, 78712-0254, ([email protected]; [email protected]) The amount of chlorine that is recycled into the deep mantle through subduction processes is poorly constrained. Estimates of the chlorine content in recycled, dehydrated oceanic crust range from <50 ppm [1] up to ~200 ppm [2,3]. In order to better constrain the Cl content in subducted crust, we have examined Cl concentrations as well as Cl/K2O and Cl/Nb ratios in olivine-hosted melt inclusions in HIMU lavas from the island of Raivavae, Austral Islands. Raivavae lavas span a wide range of lead isotopic values, with 206 Pb/204Pb ranging from ~19.3 to 21.3. Previous isotopic and trace element studies suggest that Raivavae lavas derive from a mantle source containing ancient dehydrated oceanic crust. Chlorine and K2O concentrations range from 40-1070 ppm and 0.22-3.5 wt.% respectively. The majority of Cl/K2O ratios range from 0.01-0.12 and Cl/Nb ratios range from 5-25. Cl/K2O and Cl/Nb ratios are positively correlated with 206 Pb/204Pb and negatively correlated with 207Pb/206Pb. In addition, a positive correlation exists between Cl/K2O and Nb/Zr. No correlation exists between host olivine forsterite content and Cl concentration or Cl/K2O ratios. The lack of correlation between host forsterite content and inclusion Cl/K2O and the correlation between Cl/K2O and Pbisotopes suggest that Cl contents and Cl/K2O ratios in HIMU lavas are not controlled by shallow assimilation processes. This contradicts the previous suggestion of [1], who argued that elevated Cl/K2O ratios in some Raivavae lavas resulted from assimilation of Cl-rich brine. Instead, the high Cl/Nb and Cl/K2O ratios observed in the HIMU samples suggest that the HIMU source is preferentially enriched in Cl. Because Nb should be largely retained in subducted oceanic crust during slab dehydration, we can use the average Cl/Nb ratio observed in inclusions from HIMU lavas (~18) to estimate a Cl concentration in recycled oceanic crust. For an average N-MORB Nb content of 5.6 ppm, we estimate ~100 ppm Cl in subducted oceanic crust after slab dehydration, intermediate between previous estimates of [1] and [3]. Given an estimated Cl content in altered oceanic crust of ~157-322 ppm [2], we estimate ~30-65% of the Cl in altered crust is retained in the slab after subduction-induced dehydration and eventually returned to the deep mantle.

References [1] Lassiter et al., Earth Planet. Sci. Lett. 202 (2002) 525-540. [2] Philippot, et al., Earth Planet. Sci. Lett. 161 (1998) 33-44. [3] Stroncik and Haase, Geology 32 (2004) 945-948.