Goldschmidt Abstracts 2007- N

Goldschmidt Abstracts 2007- N

Goldschmidt Conference Abstracts 2007 A698 A field-scale study on in situ measurement of microbial activities below the sea floor, Ikeshima, Japan ...

571KB Sizes 1 Downloads 181 Views

Goldschmidt Conference Abstracts 2007

A698

A field-scale study on in situ measurement of microbial activities below the sea floor, Ikeshima, Japan

D/H of kerogen across the Permian/Triassic (P/Tr) boundary B. NABBEFELD1, K. GRICE1, R. TWITCHETT2, A. SCHIMMELMANN3 AND P. SAUER3 1

Curtin University of Technology, Perth, Australia ([email protected]; [email protected]) 2 University of Plymouth, Plymouth, UK 3 Indiana University, Bloomington, USA One of the most severe mass extinctions of Earth’s history occurred near to the Permian/Triassic (P/Tr) boundary approximately 251 million years ago. Nearly 90% of all marine species and 70% of all terrestrial life became extinct (Benton & Twitchett, 2003). Several stable isotope excursions (G13Cbiomarkers, G13Ckerogen, G34Spyrite) occur near this boundary and suggest a global disruption to carbon and sulfur cycles (e.g. Grice et al., 2005). Stable hydrogen isotope (GD) data from stratigraphic sequences of kerogen can provide information about changes in palaeoclimate if the kerogens are of comparable type and thermal maturity (e.g. Lis et al., 2006). We present new GD data of stratigraphic series of kerogens from 3 different localities across the P/Tr boundary from Western Australia, East Greenland and Spitsbergen. The insoluble nature of kerogen guarantees that this material reflects in situ geochemical conditions. A dramatic change in GD of kerogen (35 ‰) is evident in several of the sections. The significance of this change along with GD of biomarkers will be discussed in terms of source/facies, paleoclimate and the extinction event. The kerogen preparation method included demineralisation of solvent-extracted and decarbonated rock samples, heavy liquid separation, a second solvent extraction and final equilibration of the kerogen with isotopically heavy and light water vapours ahead of the on-line TC/EA measurements (Robl & Davis, 1993; Schimmelmann et al., 1999).

References Benton, M.J., Twitchett, R.J. (2003), Trends in Ecology and Evolution 18/7 358-365 Grice, K., Cao, C., Love, G.D., Boettcher, M.E., Twitchett, R.J., Grosjean, E., Summons, R.E., Turgeon, S.C., Dunning, W., Jin, Y. (2005), Science 307 706-709 Lis, G.P., Schimmelmann, A., Mastalerz, M. (2006) Org. Geochem. 37 342-353 Robl, T.L., Davis, B.H. (1993), Org. Geochem. 20/2 249-255 Schimmelmann, A., Lewan, M.D., Wintsch, R.P. (1999), Geochim. Cosmochim Acta, 63/22 3751-3766

T. NAGAOKA1 AND E. NAKATA2 1

Biotechnology Sector, Environmental Science Research Laboratory ([email protected]) 2 Geosphere Science Sector, Civil Engineering Research Laboratory ([email protected]) Central Research Institute of Electric Power Industry (CRIEPI) 1646 Abiko, 270-1194, Japan We employed the push-pull test (Istock et al., Groundwater. 35(4), 619 (1997)) to determine the possibility of developing a measurement system for studying in situ microbial activities (organic matter degradation, iron and sulfate reducing activities, etc) that result in a change in groundwater geochemistry below the sea floor, Ikeshima, JAPAN. The test solution was prepared from ca. 3000 ml of groundwater with the addition of ca. 10 mmol/l KBr as a tracer, 10 mmol/l Na-lactate, and 10 mmol/l NaNO3. The test solution was injected at a pressure of 0.6 MPa into a borehole ca. 130 m below sea surface level on the Ikeshima island, Japan. The borehole was 2-m deep and 86 mm in diameter, and the working volume was ca. 12000 ml. After the injection, the groundwater in the borehole was mixed with the test solution and extracted at a specified time. During extraction, the pH, redox potential (ORP vs Ag/AgCl), and dissolved oxygen (DO) were measured, and the concentrations of lactate , Br–, and NO3– were determined by an ion chromatogram. Prior to the injection of the test solution, the properties of the groundwater were as follows: pH, 7.8; ORP, –98 mV; DO, 0.6 mg/l; EC, 4.92 S/m; and 25.8°C. After 24 h, the concentrations of lactate and NO3– had decreased drastically, and DO had reduced slightly from 0.6 to 0.2 mg/l. These results reveal that there is a possibility of a microbial effect on the geochemistry of groundwater and the measurement of in situ microbial activities.

Goldschmidt Conference Abstracts 2007

A699

In situ observation of radial pyroxene chondrule formation from levitated melts

Crystal chemistry of pumpellyite: Chromium distribution between the octahedral sites

K. NAGASHIMA1, Y. MORIUCHI1, K. K. TANAKA2 AND K. TSUKAMOTO1

MARIKO NAGASHIMA1, MASAHIDE AKASAKA2, ATSUSHI KYONO3, KUNIAKI MAKINO4 AND KO IKEDA5

1

Graduate School of Sci., Tohoku Univ., Japan ([email protected]) 2 Institute of Low Temperature Sci., Hokkaido Univ., Japan Chondrules were crystallized from levitated melts to study the rate of crystallization during cooling. Levitation is known to reduce the heterogeneous nucleation rate considerably and thus highly supersaturated melt become stable for a long time. The levitation experiments were done by using a gas-jet levitator (Nagashima et al., 2006). MgSiO3 glass spherule (diameter < 3 mm) was set at the top of nozzle of the gas jet levitator and the melt temperature was kept at a given temperature by adjusting the power of CO2 laser power. To investigate the seeding effect, enstatite (MgSiO3) particles, ~ 5 Pm in diameter, in the gas flow were forced to be impinged to the surface of the supercooled melt droplet, the nucleation process of which was observed by a high speed CCD camera. Although we had expected easy heterogeneous nucleation from the melt droplets by this seeding, this seeing experiments failed over a wide range of supercooling. Crystallization was found to be possible only at a very large supercooling temperature regime ('T ~ 260-860K). When the 'T is less than 260K, neither homogeneous nucleation nor heterogeneous nucleation could be detected. If 'T is more than 860K, crystallization was impossible because of glass formation. Typical radial textures from levitated melts (fig. 1) was obtained at 'T ~ 800K, in which a single nucleation center at surface of the melt was observed. The fact that natural radial textures in chondrule starts from the surface might suggest the impingement of a few cosmic dust particles during the cooling.

A

crystal

B

melt 500µm Figure 1: (A) In-situ observation of the levitated melt during crystallization. Radial texture was formed at 'T ~ 800 K. (B) Polarized image of (A).

Reference [1] Nagashima K., Tsukamoto K., Satoh H., Kobatake H. and Dold P., (2006). Journal of Crystal Growth 293, 193-197.

1

Institut für Geowissenschaften, Christian Albrechts Universität zu Kiel, Germany ([email protected]) 2 Department of Geoscience, Shimane University, Japan ([email protected]) 3 Division of Earth Evolution Sciences, University of Tsukuba, Japan 4 Department of Geology, Shinshu University, Japan 5 Department of Advanced Materials Science and Engineering, Yamaguchi University, Japan The crystal structure of a chromian pumpellyite of general formula VIIW8VIX4VIY8IVZ12O56-n(OH)n from Sarani, Urals, Russia, was refined using X-ray single-crystal data to investigate the crystal-chemical behavior of Cr in pumpellyite. The chemical composition of the crystal is CaO 21.9, MgO 3.6, total Fe2O3 0.2, V2O3 0.2, Cr2O3 14.6, Al2O3 16.3, and SiO2 35.4 wt.%. Optical spectroscopic analysis showed the oxidation state of Cr to be trivalent. The crystal structure [a=8.807(6), b=5.943(4), c=19.18(1) Å, E=97.44(2)º] was refined using 1297 unique reflections that converged to a R factor of 2.8%. The site populations at the X and Y sites were determined by using the observed site-scattering values. Since Mg is only located at the X site, as confirmed by TOF neutron powder diffraction analysis (Nagashima and Akasaka, in press), it was assigned to the X site. The crystal chemical formula Ca8(Mg1.81Cr1.45Al0.74)Ȉ4.00(Al5.58Cr2.42)Ȉ8.00Si12O42.19 (OH)13.81 results. The Cr amount, derived from the sitescattering values (3.87 apfu), is consistent with that obtained by EPMA analysis (3.85 apfu). Although Cr3+ is located at both the X and Y sites, the distribution coefficient ((Cr/Al)X/(Cr/Al)Y=4.46) indicates a stronger preference of Cr for the X site than the Y site, contrary to the proposal of Ivanov et al. (1981) that Cr3+ prefers the Y site. The volume of the YO6 octahedra and the mean Y–O distance increase with increasing Cr3+ substitution for Al at the Y site, which also causes an increase in the a, b, and c unit-cell dimensions. However, the volume of the XO6 octahedra and the mean X–O distance are not correlated with the site populations of cations at the X site. Distortion parameters for the Y site tend to decrease with an increasing mean Y–O distance and volume of the YO6 octahedra. This indicates that the YO6 octahedra become less distorted with the expansion of the octahedra through substitution of cations larger than Al.

References Ivanov, O.K., Arkhangel’skaya, V.A., Miroshinikova, L.O. and Shikova, T.A. (1981) Zap. Vses. Mineral. Obshchest., 110, 508-512. Nagashima, M. and Akasaka, M. (2007) Can. Min. (in press)

Goldschmidt Conference Abstracts 2007

A700

The crystal chemistry of macfallite 1

1

Geochemistry of basalts from intratransform spreading centers: Implications for melt migration models

1

M. NAGASHIMA , N-S. RAHMOUN , E.V. ALEKSEEV , C.A. GEIGER1 AND M. AKASAKA2 1

Institut für Geowissenschaften, Christian Albrechts Universität zu Kiel, Germany ([email protected]) 2 Department of Geoscience, Shimane University, Japan ([email protected])

A.N. NAGLE1, R.C. PICKLE1, A.E. SAAL1, E.H. HAURI2 1 AND D.W. FORSYTH 1

The crystal structure of macfallite (ideallyCa2Mn3+3Si3O11(OH)3) from the Keweenaw Peninsula, USA, was refined by single crystal X-ray diffraction. The distribution of cations between the three independent octahedral sites M1, M2 and M3 were determined in order to investigate the nature of the cation substitutions and the crystalchemical properties of macfallite and related minerals. The chemical composition of the macfallite crystal in this study in terms of oxide wt.% is 33.14 SiO2, 2.49 Al2O3, 0.09 V2O3, 35.80 total Mn2O3, 0.05 total Fe2O3, 0.45 CuO, 0.69 MgO, 20.53 CaO, 0.07 SrO, and 0.03 Na2O. The structure [a = 10.218(4), b = 6.072(2), c = 8.959(3) Å, E = 110.75(3)º] was refined using 1717 unique reflections that converged to a R factor of 4.2%. The site-populations at the M1, M2 and M3 sites were determined from their site-scattering values. The structural formula Ca2W(Mn3+0.77Al0.23)M1(Mn3+0.84Al0.04Mg0.09Cu0.03)M2(Mn3+0.96A l0.04)M3Si3O10.88(OH)3.12 results. Divalent cations are located at the M2 site. This is consistent with the fact that the volume of the M2O6 octahedron is the largest among the three octahedral sites, and that the bond valence sum for the M2 site is about 2.75 v.u., which is less than those for M1 and M3. Aluminum is essentially partitioned onto the M1 site, which is consistent with the proposal of Moore et al. (1985) ((Mn3+0.7Al0.3)M2). However, the present study also shows that small amounts of Al are located at the M2 and M3 sites. The calculated bond valence sums of 1.69, 1.26, 0.99 and 1.18 v.u. for OH1, OH2, OH3 and O5, respectively, suggest that hydroxyl groups are located not only at the positions of OH1, OH2 and OH3 (Moore et al., 1985) but also at O5. The values for OH3 and O5 are close to 1.0, while those for the OH1 and OH2 are greater than 1.0. The substitution of M2+M2 + H+OH1 + O2-OH1 ļ M3+M2 + O2-OH1 may act to reduce the number of hydrogen atoms in the structure. This substitution has been suggested for pumpellyite (Yoshiasa and Matsumoto, 1985).

References Moore, P.B., Shen, J. and Araki, T., (1985), Amer. Min. 70, 171-181. Hawthorne, F.C., Ungaretti, L. and Oberti, R., (1995) Can. Min. 33, 907–911. Yoshiasa, A. and Matsumoto, T., (1985), Amer. Min. 70, 1011-1019.

Brown University Department of Geological Sciences, Providence, RI, USA ([email protected]) 2 Department of Terrestrial Magnetism, Carnegie Institution of Washington, Washington, DC, USA The advantage of studying basalts from intra-transform spreading centers is that they provide well-defined locations of melt delivery and crustal formation. Also, magmatism in fracture zones is not supported by long-lived magma chambers or along-axis transport of melt from other parts of the spreading system. Thus lavas from fracture zones represent pre-aggregated melts and their compositions provide insight into models of melt generation and transport processes beneath mid-ocean ridges. The Quebrada/Discovery/Gofar (QDG) transform fault system offsets the fast-spreading East Pacific Rise (3q-5qS) by approximately 400km and is composed of 7 active intratransform spreading centers ranging in length from 5 to 70km. Forty-seven dredges of young intra-transform basalts were collected from this area and analyzed for volatiles, major and trace elements. QDG basalts exhibit varying degrees of differentiation, which correlate with the estimates of crustal thickness of each ridge segment derived from gravity models. The incompatible trace element ratios (e.g., Th/La) of these lavas range significantly from ultra-depleted to extremely enriched compositions. Furthermore, the level of enrichment correlates well with both ratios of volatiles to similarly incompatible refractory elements (e.g., H2O/Ce) and indicators of depth of melt segregation (e.g., Sm/Yb). Overall, the chemical variation of these basalts is greater than that previously found in fracture zones (such as Siqueiros and Garrett FZ) and is similar to the compositional range defined by northern EPR seamounts. Models of mantle flow and melt delivery in the QDG study area predict significantly different compositions and crustal thicknesses at the various intra-transform spreading centers. We will test the validity of the different models by comparing the observed and predicted lava compositions.

Goldschmidt Conference Abstracts 2007

Effects of self-consistently-calculated thermodynamic properties in thermochemical multiphase mantle convection in a 3D spherical shell TAKASHI NAKAGAWA1, PAUL J. TACKLEY2, FREDERIC DESCHAMPS3 AND JAMES A.D. CONNOLLY4 1

Department of Earth and Planetary Sciences, Kyushu University, Fukuoka, Japan ([email protected]) 2 Institute of Geophysics, ETH Zürich, Switzerland ([email protected]) 3 Institute of Geophysics, ETH Zürich, Switzerland ([email protected]) 4 Institute of Mineralogy and Petrology, ETH Zürich, Switzerland ([email protected]) High pressure and temperature experiments and calculations of the properties of mantle minerals show that many different mineral phases exist as a function of pressure, temperature and composition (e.g. Irifune and Ringwood, 1987), and that these have a first-order influence on properties such as density and elastic moduli (hence seismic velocity). Numerical models of thermo-chemical mantle convection have typically used a simple approximation to treat these complex variations in material properties, such as the extended Boussinesq approximation. Some numerical models have attempted to implement multiple, composition-dependent phases into thermo-chemical mantle convection (e.g. Tackley and Xie, 2003; Nakagawa and Tackley, 2005) and to calculate seismic anomalies from mantle convection simulations based on polynominal fitting for temperature, composition and mineral phase (Nakagawa and Tackley, 2006). However, their linearised treatments are still approximations and may not adequately represent properties including effect of composition on phase transitions. In order to get closer to a realistic mineralogy, we here calculate composition-dependent mineral assemblages and their physical properties using the code PERPLEX, which minimizes free energy for a given combination of oxides as a function of temperature and pressure, and use this in a numerical model of thermochemical mantle convection in a three-dimensional spherical shell, to calculate three-dimensionally-varying physical proporties. In this presentation we compare the results obtained with this new, self-consistently-calculated treatment, with results using the old, approximate treatment, focusing particularly on thermo-chemical-phase structures and seismic anomalies in the CMB region and the transition zone.

References Irifune, T., and Ringwood, A. E., (1987), Earth. Planet. Sci. Lett., 86, 365-376. Nakagawa T., and Tackley, P.J., (2006), Geophys. Res. Lett., 33, L12S11, doi:10.1029/2006GL025719. Nakagawa, T., and Tackley, P.J., (2005), Proc. 3rd MIT conference. Tackley, P.J., and Xie, S., (2003), Proc. 2nd MIT conference.

A701

Development of chelate resin column preconcentration method for precise isotope analysis of Mo in seawater Y. NAKAGAWA1, M. LUTFI FIRDAUS1, K. NORISUYE1, Y. SOHRIN1, K. IRISAWA2 AND T. HIRATA2 1

Institute for Chemical Research, Kyoto University, Uji, Kyoto, 611-0011, Japan ([email protected]) 2 Tokyo Institute of Technology, 2-12-1 Oookayama, Meguroku, Tokyo, 152-8550, Japan Mo is present nearly conservatively in the modern ocean (~10 ppb). However, Mo is depleted in reducing environments and hydrothermal systems and so on, and such fractionation shoule be reflected in the Mo isotope composition. In addition, because Mo is an essential element for organisms, biological activity may change the Mo isotope composition. Therefore, precise isotope analysis of Mo in seawater is expected to provide new valuable information, and to be a good tool in geochemical oceanography. Only a few data have been reported on the isotopic composition of Mo in seawater. More than 50-fold concentration of Mo in seawater is required for isotopic measurement. An ion exchange resin method was used to purify Mo, but this method requires concentrated acid for elution, and may not be able to separate major elements of seawater completely. So, we have newly developed a chelate resin column preconcentration method. 8-hydroxyquinoline (8HQ) bonded covalently to a vinyl polymer resin, TSK-8HQ (Dierssen et al., 2001), was used in a column to concentrate Mo. The sample solution that has been adjusted to pH 2.0 by HCl was passed through the column, and then the adsorbed Mo was eluted by the back flushing of 2 M NH3. To obtain a high concentration factor, the eluate was evaporated to dryness, and Mo was then redissolved in 0.05% tetramethylammonium hydroxide. Mo isotope ratio was measured by MC-ICP-MS (Nu Plasma 500). The mass discrimination effect on Mo was internally corrected by 88Sr/86Sr to be 0.1194 using the exponential low. Individual sample uncertainties (2V) of this element spike method are 0.03‰ per amu for 1 ppm Mo. Mo in 500 mL seawater sample was collected quantitatively. The procedure blank was less than 10-3 times of Mo concentration in seawater. The residual ratio of major elements of seawater in the eluate was less than 10-5, and their matrix effects on isotope analysis of Mo were negligible. G98/95Mo in seawater collected from 697 m depth in Suruga Bay, Japan was 2.49 ± 0.10 (2V) relative to JMC standard. We will report the vertical profiles of Mo isotope ratio from stations in the western North Pacific Ocean at the presentation.

References Dierssen, H., Balzer, W., Landing, W.M., (2001), Mar. Chem. 73, 173-192.

A702

Goldschmidt Conference Abstracts 2007

Effects of hydrogen on limits of radiative emission from a planet with a saturated-water-vapor atmosphere MIKI NAKAJIMA1, MASAHIRO IKOMA1, HIDENORI GENDA2 AND SHIGERU IDA1

Chemical composition and behavior of atmospheric aerosols in the mountainous and plain regions of Toyama, Central Japan T. NAKAMURA, T. YAMAZAKI AND K. KAWASAKI

1

Department of Earth and Planetary Sciences, Tokyo Institute of Technology, Tokyo, Japan ([email protected]) 2 Research Center for the Evolving Earth and Planets, Tokyo Institute of Technology, Tokyo, Japan It is known that there are upper limits of radiative emission from a planet covered with sufficient water (i.e., an ocean). If incoming thermal flux exceeds the limits, extra thermal energy is used to evaporate the ocean, which means that an ocean is unable to exist under such a circumstance. These limits have close relationships with the early evolution and habitability of planets. Since the energy flux from the planet’s surface decreases with time because of cooling of the planetary interior, the limits constrain the timing of ocean formation. Moreover, because the stellar radiative flux decreases with distance from the star, this limits constrain the inner edge of the habitable zone in a planetary system. Several studies investigated these limits (Komabayashi 1967; Ingersoll 1969; Abe and Matsui 1988; Kasting et al. 1988; Nakajima et al. 1992). However, no study has evaluated the effects of hydrogen on these limits. According to recent studies of the accumulation of the Earth’s atmosphere, the initial Earth is likely to have been rich in hydrogen (e.g., Ikoma and Genda 2006). Furthermore, in extrasolar planetary systems, there might exist hydrogen-rich terrestrial planets. In this research, we have investigated the effects of hydrogen (e.g., changes in mean molecular weight and opacity) and gravity on the limits by simulating the structure of the atmosphere in the radiative-convective equilibrium. Then we discuss the early evolution of the Earth and the location of the habitable zone in extrasolar planetary system.

References Abe Y., and Matsui T., (1988) J. Atmos. Sci., 45, 3081-3101. Ikoma M. and Genda H., (2006), ApJ. 648, 696-706. Ingersoll A.P., (1969), J. Atmos. Sci., 26, 1191-1198. Kasting J. F., (1988). Icarus, 74, 472-494. Komabayashi M., (1967), J. Meteor. Soc. Japan, 45, 137139.

Toyama Prefectural Environmental Science Research Center, Toyama, Japan ([email protected])

Introduction East Asia is one of the most troubled regions in terms of atmospheric environment due to its huge population and industrial development. The emission of anthropogenic substances in East Asia has been on the rise due to the use of fossil fuels, industrialization and due to the lack of waste control practices. In particular, it is predicted that NOx emission will increase approximately 5-fold from 1990 to 2020 (Akimoto,2003). The excessive N loading may have several adverse effects on both ecosystem health and water quality. The western North Pacific recieves a large influx of mineral dust and pollutants from East Asia through atmospheric long-range transport. The Japan Sea is a marginal sea of the western Pacific Ocean and located in a transit zone for the continental air masses traveling to the open North Pacific. The coastal area of Japan Sea is an important area to study the transport mechanisms and modifications to chemical properties of aerosols.

Method Chemical composition of aerosols and suspendes particle matter (SPM) measurements were conducted at Mt. Tateyama (2450 m a.s.l.) and Imizu (10 m a.s.l.) in Toyama, central Japan, simultaneously from 18 May to 8 June 2006. The sampling period was about 24 hours and the sampled air volume was from 27.7 to 39.1 m3.

Results During the observation period, anthropogenic plume were observed. One of these episodes were almost equal nss-SO42levels to the urban atmosphere, even in Tateyama. The mean concentrations of nss-SO42- and NO3- in Tateyama were 5.37 and 0.59 Pg m-3, respectively. The concentration of nss-SO42- was relatively lower than that of Imizu 8.68 Pg m-3. However, the concentration of NO3- was much lower than that of Imizu (2.93 Pg m-3). The NO3-/nssSO42- ratio in Tateyama (0.11) was lower than that of in Imizu (0.34). This is probably because NO3- combined with nss-Ca2+ removed from the atmosphere faster than that of nss-SO42-. New particulate nss-SO42- is expected to form by gas to particle conversation with time. These pollutants will influence air quality and radiative forcing. To understand the effect of nitrogen deposition to the ecosystems, both wet and dry including gaseous matter measurements are required.

Reference Akimoto, H. (2003) Science 302, 1716-1719.

Goldschmidt Conference Abstracts 2007

The mineralogy of Comet Wild-2 nucleus samples – What we think we know and what we do not know KEIKO NAKAMURA-MESSENGER1,2, MICHAEL ZOLENSKY1 AND THE STARDUST MINERALOGY/PETROLOGY SUBTEAM

A703

Rb-Sr single bed isochron dating with evidence of isotope equilibrium T. NAKAO1 AND T. TANAKA AND S. KOJIMA2 1

Dept. Earth and Environmental Sci, Nagoya Univ, Nagoya 464-8601 JAPAN ([email protected]) 2 Dept. Civil Engineering, Gifu Univ, Gifu 501-1193 JAPAN

1

NASA Johnson Space Center, Houston, TX 77058 USA ESCG/ Jacobs Engineering, Houston, TX 77058 USA

2

The sample return capsule of the Stardust spacecraft was successfully recovered in northern Utah on January 15, 2006, and its cargo of coma grains from Comet Wild-2 captured in aerogel tiles has now been the subject of intense investigation by approximately 200 scientists scattered across five continents. We have now performed mineralogical and petrographic analyses of particles derived directly from the Jupiter-family Comet Wild-2. Numerous particles were collected by impact into low density aerogel. Dozens to thousands of fragments of the impacting particles are distributed along 100 m- 1cm-length tracks. So far 52 of these tracks have been harvested, and samples of 26 tracks have been studied in enough detail to give first order views of their mineralogy [1]. The recovered Wild-2 samples are mixtures of crystalline and amorphous materials. However, it is unclear how much of the amorphous material was produced from the aerogel during the particle capture. Crystalline materials are abundant in Comet Wild-2 and many are “coarse-grained” (1-10 Pm) relative to the submicrometer scales characteristic of anhydrous IDPs and interstellar dust. Of the best studied 26 tracks, eight are dominated by olivine, seven by low-Ca pyroxene, three by similar amounts of olivine and pyroxene, and the remaining eight are dominated by other minerals, mainly Fe-Ni sulfides. Our emerging model is that many of the impacting particles were very fine-grained, loosely-bound aggregates with a bulk chondritic composition. Most also contain large individual crystals (most commonly) of olivine, pyroxene and Fe-Ni sulfides, similar to chondritic IDPs. Olivine and Low-Ca Pyroxene is present in most Wild-2 particles, with grain sizes ranging from sub-Pm to over 10 Pm. Wild-2 olivine has an extremely wide compositional range, from Fo4-100, with a pronounced peak at Fo99. The wide Mg-Fe compositional range of Wild-2 olivine is similar to both anhydrous chondritic IDPs, and also to matrix minerals of the chondrites Murchison (CM2), and Orgueil (CI1), which have experienced significant-to-pervasive aqueous alteration. Both low- and high-Ca pyroxenes are present among the Wild-2 grains, with the former being dominant. The compositional range displayed by the low-calcium Ca pyroxene is also very extensive, from En52-100, with a significant frequency peak centered at En95.

Reference [1] Zolensky M. et al. (2006) Mineralogy and Petrology of Comet Wild 2 Nucleus Samples. Science 314, 1735-1740

Introduction Sedimentary rocks consist of various components, isotope ratios of which are considered to indicate different ages. If isotopic age determinations of the sedimentary rocks can be carried out, the isotopes had to equilibrate, after sources of them had deposited. It is difficult to verify the isotope equilibrium. Shibata and Mizutani (1982) reported Rb-Sr isochron age of 211 ± 4.7 Ma with initial 87Sr/86Sr ratio of 0.71325 ± 20 for the Triassic chert, which is younger than the fossil age. They regarded the Rb-Sr age as a diagenesis age. They collected the samples ranging 10 m in thickness. The strata of this thickness is hard to regard to be formed by a single sedimentary event. They didn’t discuss equilibrium of Sr isotopes. In this study, we used a single layer of the Triassic bedded chert in the Mino belt, central Japan to determine a Rb-Sr isochron age. Conodont fossils, one of the components of the chert, were also separeted in order to measure the 87Sr/86Sr ratio.

Analytical method The sample was cut parallel to the bedding plain into twelve slices, each of which was used as a whole rock sample. Concentration of Rb and Sr of the slices were determined by a thermal ionization quadrupole mass spectrometer, Finngan MAT Thermoquad THQ. Sr isotope ratio was measured with a VG Sector 54-30 thermal ionization mass spectrometer at Nagoya University. In case of conodonts, a Ta emitter solution (Brick, 1986) was used to load Sr onto single Re filament with 2M-H3PO4.

Result and Discussion Rb-Sr isochron age of the chert was 219 ± 22 Ma with the initial 87Sr/86Sr ratio of 0.7125 ± 9. The Rb-Sr age is indistingushable from the depositional age (Norian: Sugiyama, 1997) indicated by the radiolarian fossils Conodonts in the chert consist mainly of apatite. Apatite contains extrmely low content of Rb but rich in Sr. This indicates that the 87Sr/86Sr ratio of conodonts would not have evolved since the conodonts were deposited. Since conodonts are part of marine organisms, the 87Sr/86Sr ratio must be same as that of seawater at the time of deposition. The 87Sr/86Sr ratio of the Triassic conodonts, however, is 0.712211 ± 22 (87Rb/86Sr=0.10) and higher than that of Triassic seaweater (=0.708), while it is indisutingushable from the initial Sr ratio (=0.7125 ± 9) of the chert. This indicates that the chert layer including conodont fossils are equilibrated in term of Sr isotopes after its deposition. Because the Rb-Sr age is consistent with the radiolarian age, the Sr isotopes were equilibrated just after the sources of chert had deposited.

A704

Goldschmidt Conference Abstracts 2007

Noble gases and nitrogen in the Isheyevo CH/CB chondrite

Physicochemical speciation of trace metals during the mesoscale iron enrichment (SEEDS II) in the western North Pacific

D. NAKASHIMA1,2, S. P. SCHWENZER1 AND U. OTT1 1

Max-Planck-Institut für Chemie, Mainz, Germany ([email protected]) 2 University of Tokyo, Tokyo, Japan ([email protected]) Metal-rich CH and CB carbonaceous chondrites are known to show characteristic noble gas signatures and nitrogen isotopic compositions [e.g., 1-2]. Noble gases in CHs and CBs are characterized by Ar-rich gases and low 136 Xe/132Xe ratios (solar wind-Xe like) [1]. In addition, these chondrites have the isotopically heaviest N in chondrites (G15N ~ +1000‰; [2]). In this study, we analyzed noble gases and nitrogen in magnetic fraction (MF) and non-magnetic fraction (NMF) from Isheyevo, which is an unique meteorite consisting of CH-like and CB-like lithologies [3]. Isotopic ratios of He and Ne show that light noble gases in both fractions are dominated by solar and cosmogenic noble gases. Trapped 4He/20Ne ratios are about 400 for MF and 170 for NMF, indicative of high retentivity of solar noble gases in MF. Trapped 20Ne/36Ar ratio of MF (~5) is lower than that of NMF (28). This can not be explained by preferential loss of solar 20Ne, because of higher retentivity of solar noble gases in MF. It is likely that the low 20Ne/36Ar ratio is due to enrichment of 36Ar, i.e., Ar-rich noble gases, which are considered to be fractionated solar wind noble gases [e.g., 4]. In the case of Ishyevo, Ar-rich gases appear to be concentrated in metallic phases (MF). Xe isotopic ratios in NMF are almost identical to Xe-Q, whereas Xe in MF seems to be the mixture of Q-Xe and solar wind-Xe (SW-Xe). This indicates that SW-Xe is contained in MF. Since MF contains both solar and Ar-rich gases, it is not clear whether SW-Xe is associated with solar or Ar-rich gases. Nitrogen in MF (G15N = +1230‰) is isotopically heavier than that in NMF (+310‰). Nitrogen in NMF may be diluted with isotopically “normal” nitrogen (0s50‰; [5]).

Acknowledgement We thank M. A. Ivanova for providing magnetic and nonmagnetic fractions of the Isheyevo meteorite.

S. NAKATSUKA1 *, Y. SOHRIN1, K. NORISUYE1, K. OKAMURA2, S. TAKEDA3 AND J. NISHIOKA4 1

Institute for Chemical Research, Kyoto University, Kyoto, Japan ([email protected]) 2 Center for Advanced Core Research, Kochi University, Kochi, Japan 3 Graduate School of Agricultural and Life Sciences, The University of Tokyo, Tokyo, Japan 4 Institute of Low temperature Science, Hokkaido University, Hokkaido, Japan The mesoscale iron-enrichment experiment SEEDS II was conducted near the edge of the subarctic Western Notrh Pacific gyre, where SEEDS I had been carried out [1, 2]. We investigated dissolved and particulate Co, Ni, Cu, Zn, Cd and Pb in seawater from both field observation for 26 days and shipboard incubation with a natural phytoplankton assemblage for 10 days. Discrete seawater samples were filtered through 0.2 μm filter and acidified to pH 2.2 for the determination of the dissolved species by ICP-MS. The filter was used to measure the particulate species by FI-ICP-MS. Before the iron enrichment, the average concentrations for dissolved Co, Ni, Cu, Zn, Cd and Pb in the surface mixed layer (0-20 m) were 70 pM, 4.9 nM, 2.1 nM, 1.6 nM, 0.48 nM and 52 pM, respectively, and those for the particulate species were 1.7 pM, 0.052 nM, 0.094 nM, 0.46 nM, 0.037 nM and 5.2 pM, respectively. After the enrichment, there was a threefold increase in chlorophyll a (~3 μg/L) by day 12. However, there was no detectable difference in the dissolved and particulate trace metals between inside and outside the patch. In the shipboard incubation, addition of 1 nM Fe caused a 30-fold increase in chlorophyll a (~9 μg/L) dominated by Pseudo-nitzschia sp. and increases in the particulate trace metal fraction up to 3-45%. These results suggest that Fe was a limiting factor for the growth of phytoplankton. In addition, enhanced-grazing by mesozooplankton presumably limited the growth of phytoplankton and the transformation of trace metal speciation during the mesoscale Fe enrichment.

References [1] Lewis R. S. (1985) Meteoritics 20, 689. [2] Prombo C. A. and Clayton R. N. (1985) Science 230, 935937. [3] Ivanova M. A. (2006) LPSXXXVII, Abstract#1100. [4] Nakamura T. et al. (2003) MAPS 38, 243-250. [5] Grady M. M. and Pillinger C. T. (1990) EPSL 97, 29-40.

References [1] Tsuda et al., (2003), Science, 300, 958. [2] Kinugasa et al., (2005), Prog. Oceanogr., 64, 129.

Goldschmidt Conference Abstracts 2007

The Sept Iles mafic layered intrusion: An example of ferrobasaltic differentiation

Zircon M257 – A new standard for SHRIMP U-Pb geochronology L. NASDALA1, W. HOFMEISTER2, J. MATTINSON3, W. DÖRR4, F. CORFU5, P.W. REINERS6, A. KRONZ7, N. NORBERG1 AND A. KRÖNER2

O. NAMUR1, B. CHARLIER1, M.D. HIGGINS2 AND J. VANDER AUWERA1 1

Department of Geology, University of Liège, Belgium 2 Earth Sciences, University of Chicoutimi, Canada ([email protected]) The Sept Iles Mafic Intrusion (SIMI) (Quebec, Canada), part of the Sept Iles Intrusive Suite (SIIS), is a 6 km-thick layered troctolite-gabbro intrusion with a funnel shape of 80 km in diameter. The SIIS has been dated at 564±4 Ma [1] and supposedly results from the activity of a mantle plume related to the formation of the St-Laurent rift system. Flood basalts may have erupted before its emplacement [2]. A detailed petrologic study of the SIMI has been undertaken with the aim of better constraining the differentiation of ferrobasaltic liquids [3]. We present here the first results of this study which focus on the possible composition of the SIMI parent magma and on the main structure of this magma chamber. The exposed part of the SIIS is surrounded by a Border zone of variable thickness made of massive and locally fine grained gabbros. These fine-grained gabbros have high FeOt (up to 15 wt. %) and TiO2 (3 wt. %) contents, moderate P2O5 content (0.35 wt. %) and no Eu anomaly (Eu/Eu* = 0.95). They are typical ferrobasalts very similar to the estimated parent magmas of the Skaergaard [4] and Newark Island intrusions [5]. The composition of liquidus phases obtained with the MELTS algorithm (plagioclase An67, olivine Fo75) perfectly matches the most primitive compositions of plagioclase (An68) and olivine (Fo75) observed in the SIMI. Moreover, the sequence of crystallization experimentally obtained by [5, 6] on samples which are very similar to the ferrobasalts from Sept Iles is in agreement with the sequence of crystallization of the SIMI (plagioclase and olivine, + Carich clinopyroxene, + magnetite and ilmenite - olivine, + olivine + apatite). The fine-grained gabbros from the Border Intrusion are thus plausible compositions for the parent magma of the SIMI. From base to top, the plagioclase composition evolves continuously from An68 to An40, then displays a significant reversal to An60 followed by a new decrease down to An53. The magnetite Cr content displays a parallel evolution. This reversal is interpreted as a new magma influx which is marked by an intraplutonic quench horizon. Other small reversals are observed throughout the Series.

References: [1] Higgins, M.D. and van Breemen, O. (1998) J. Geol. 106, 421-431. [2] Higgins, M.D. (2005) Lithos. 83, 199-213. [3] Thy, P. et al. (2006) Lithos. 92, 154-180. [4] Hoover, J.D. (1989) J. Petrol. 30, 441-476. [5] Snyder, D.A. et al. (1993) Contrib. Mineral. Petrol. 113, 73-86. [6] Toplis, M.J. and Carroll, M.R. (1996) J. Petrol. 37, 837858.

A705

1

Institut für Mineralogie und Kristallographie, Universität Wien, Austria ([email protected]; [email protected]) 2 Institut für Geowissenschaften, Johannes GutenbergUniversität Mainz, Germany ([email protected]; [email protected]) 3 Department of Earth Science, University of California, Santa Barbara, CA, U.S.A. ([email protected]) 4 Institut für Geowissenschaften, Universität Giessen, Germany ([email protected]) 5 Department of Geosciences, University of Oslo, Norway ([email protected]) 6 Department of Geosciences, University of Arizona, Tucson, AZ, U.S.A. ([email protected]) 7 Geowissenschaftliches Zentrum, Georg August-Universität Göttingen, Germany ([email protected]) The possibility to determine U-Pb ages from micron-sized areas within single zircon crystals using the SHRIMP technique has opened up new possibilities for detailed studies of the history of crustal rocks and constraining their ages. This technique, however, depends on the availability of wellcharacterized, natural reference materials (Kennedy, 2000). We propose a gem-quality zircon (Univ. Mainz sample no. M257) as a future SHRIMP U-Pb zircon standard. Zircon M257 is a pale brown, clear stone with a weight of 25.7 ct [for comparison, SHRIMP standard CZ3 (Pidgeon, 1997), which has been used for more than ten years, had a weight of only 4.5 ct]. M257 is remarkably homogeneous; no internal zoning at all was found using CL imaging, transmitted light microscopy and Raman line-scanning. Its comparably high actinide concentrations (U, 810 ppm; Th, 224 ppm) will account for high U, Th and Pb count rates. Preliminary TIMS analyses (done in Giessen, Santa Barbara and Oslo) indicate a concordant U-Pb age of a562 Ma. Unit cell dimensions (a0, 6.626 Å; c0, 6.030 Å) and Raman parameters {e.g., FWHM[Q3(SiO4)], 11.8 cm–1} correlate very well with the calculated time-integrated alpha dose of 1.66 u 1018 D/g (compare Nasdala et al., 2004). This and the (U+Th)/He age (419 r 9 Ma), which is consistent with previous helium ages for zircon from Sri Lanka, allow us to exclude any heattreatment. Further characterization of the isotopic composition of sample M257 is in progress.

References Kennedy A.K., (2000), In: Woodhead J.D., Hergt J.M. and Noble W.P. (Eds.), Beyond 2000: New Frontiers in Isotope Geoscience, Lorne, Abstr. & Proc., pp. 109–111. Nasdala L., Reiners P.W., Garver J.I., Kennedy A.K., Stern R.A., Balan E. and Wirth R., (2004), Am. Mineral. 89, 219–231. Pidgeon R.T., (1997), Z. Dt. Gemmol. Ges. 46/1, 21–28.

A706

Goldschmidt Conference Abstracts 2007

Deficient electron microprobe totals: Discussion of potential causes

The first record of allochthonous kimberlite within the Batain Nappes, Eastern Oman

L. NASDALA1, T. VÁCZI1, A. KRONZ2, R. WIRTH3, C. PÉREZ-SOBA AGUILAR4 AND A. WILLNER5

S. NASIR1., S. AL-KHIRBASH1, A. AL-SAYIGH1, A. ALHARTHY1, A. MUBAREK1, H. ROLLINSON1, A. LAZKI1, E. BELOUSOVA2, W. GRIFFIN2 AND F. KAMINSKY3

1

Institut für Mineralogie und Kristallographie, Universität Wien, Austria ([email protected]; [email protected]) 2 Geowissenschaftliches Zentrum, Georg August-Universität Göttingen, Germany ([email protected]) 3 GeoForschungsZentrum Potsdam, Germany ([email protected]) 4 Departamento de Petrología y Geoquímica, Universidad Complutense, Madrid, Spain ([email protected]) 5 Institut für Geologie, Mineralogie und Geophysik, RuhrUniversität Bochum, Germany ([email protected]) Radiation-damaged minerals, including zircon, coffinite, thorite and monazite, are occasionally found to yield electron microprobe results whose totals (oxide sums) are deficient, i.e. appreciably below 100 wt%. Causes of the apparently too low totals are controversial; most authors have assigned the phenomenon to enhanced contents of hydrous species, other non-analyzed elements, textural peculiarities such as voids, or instability and degradation under the electron beam. It has also been observed that micro-areas yielding low totals are typically recognized from very low BSE intensities (Kempe et al., 2000), which is in apparent contradiction to their usually high degree of radiation damage and the positive correlation between radiation damage and BSE intensiy (Nasdala et al., 2006). We have studied low-totals zircon from three localities. Deficient totals and accompanying very low BSE intensities are only detected in areas affected by secondary alteration whereas primary zircon never shows these phenomena. The alteration has often, but not always, led to enhanced actinide content, which is why low-totals areas are mostly radiationdamaged. Low-totals areas were found to contain up to 5 wt% hydrous species, which explains both the deficient analysis sums and the low BSE intensity (due to decrease of the average atomic number). In the TEM, low-totals areas show a sponge-like texture with numerous sub-micron sized voids, as it was suggested by Pointer et al. (1988). We explain the formation of this texture and the water uptake by secondary, fluid-driven alteration of previously radiation-damaged and thus volume-expanded zircon.

References Kempe U., Gruner T., Nasdala L. and Wolf D., (2000), In: Pagel M., Barbin V., Blanc P. and Ohnenstetter D. (Eds.), Cathodoluminescence in Geosciences. Springer, Berlin, Heidelberg, New York, pp. 415455. Nasdala L., Kronz A., Hanchar J.M., Tichomirowa M., Davis D.D. and Hofmeister W., (2006), Am. Mineral. 91, 17381746. Pointer C.M., Ashworth J.R. and Ixer R.A., (1988), Mineral. Petrol. 39, 2137.

1

Department of Earth Sciences, Sultan Qaboos University ([email protected]) 2 GEMOC Laboratory, Macquarie University, Australia 3 KM Diamond Exploration Ltd. Vancouver, Canada Kimberlites, carbonatites, alkaline and ultramafic lamprophyres and other alkalic ultrabasic rocks have been recently discovered within the Batain Nappes in the eastern Oman Mountains. The kimberlite area comprises several allochthonous bodies. Most of these are carbonatite and carbonate kimberlite which contains mantle derived (altered peridotitic) xenoliths and xenocrysts. The kimberlite occurs either as breccia pipes and/or as long dykes ~ 6 km long. The kimberlites contain abundaunt macrocrysts of mica (phlogopite and/or biotite), chromite, chrome diopside, pelletal lapilli and autolithic fragments in a calcite + serpentine matrix. The kimberlites are dominantly composed of ‘hapabyssal and diatreme facies’ volcaniclastic rocks. These include pyroclastic lapilli-, carbonate-dominated tuffs, and volcaniclastic kimberlite, all of which intruding late Jurassic to Lower Cretaceous marine sedimentary rocks (cherts and shales) of the Wahra Formation, within the Allochthonous Batain Mel'ange. Major and trace elements and isotopic composition indicate that the Omani kimberlites in this study form a compositionally cohesive group of rocks more akin geochemically to the Koidu type kimberlites of West Africa than the Group I kimberlites from South Africa. The kimberlite contained zircon, G9 garnet and chromite grains with typical kimberlitic morphologies and chemical properties similar to diamond inclusion chromite. However, there were no micro-diamonds observed. Fifteen pinkish (‘kimberlitic’) zircon grains, 0.5 to 1.5 mm in size, were picked from the kimberlite tuff and were analyzed at GEMOC. Trace-element patterns are typical of kimberlitic to carbonatitic zircons. Their mean age of 137.5 ± 1 Ma (95 % confidence,MSWD = 0.49) is consistent with intrusion into Lower Cretaceous rocks . Their 176Hf/177Hf (0.28286±1, HHf = 6.2) is typical of kimberlitic zircons of this age, and may represent the subcontinental lithospheric mantle.

Goldschmidt Conference Abstracts 2007

Is the isotope composition of Reunion plume really homogeneous?

Influence of porosity on basalt weathering rates from the clast to watershed scale

F. NAURET, M. ARIGOT, P. LOUVAT AND M. MOREIRA Equipe de Geochimie-Cosmochimie. Institut de Physique du Glode de Paris. Tour 14, 3 Etage. ([email protected]) Réunion Island, Indian Ocean, is the present signature on the Earth surface of the Réunion hotspot. The island is composed by two volcanoes: the inactive Piton des Neiges and the active Piton de la Fournaise. The isotope compositions of Réunion lavas are usually described as homogeneous. This homogeneity is remarkable compared to other hotspot volcanism. However, a fine study of the Réunion database reveals that the large majority of the available data comes from analyzed lavas produced by Piton de la Fournaise. Furthermore, the few available Pb isotope data have been published before 1972. Consequently, the apparent homogeneity of Reunion lavas is potentially due to analytical or sample bias. With the intention of testing this homogeneity, we have recently sampled the Piton des Neiges. The sample collection (n=47) covers temporally and geographically this volcano. Taking advantage of recent analytical development, Pb isotope compositions were measured using a MC-ICP-MS (Neptune) and mass fractionation is corrected using a Tl spike. The Pb isotope composition range from 206Pb/204Pb: 18.7918.90, 207Pb/204Pb: 15.56-15.60 and 208Pb/204Pb: 38.79-39.04. These ranges overlap these given for the Piton de la Fournaise (206Pb/204Pb: 18.55-18.91, 207Pb/204Pb: 15.58-15.65 and 208 Pb/204Pb: 38.74-39.01 from 71 samples). Based on these preliminary results, we suggest that lavas from both volcanoes plot on same lines in Pb isotope spaces. However, it seems that Piton des Neiges lavas might extend Pb isotopes ranges given for Reunion plume. Pb isotopes and trace element concentration analyses currently in progress will help further constrain the geochemical composition of the Piton des Neiges, the chemical structure of the Réunion hotspot and the influence of the Reunion hotspot on the regional magmatism.

A707

A. NAVARRE-SITCHLER1, C. STEEFEL2, E. HAUSRATH1 3 AND S. BRANTLEY 1

Department of Geosciences, The Pennsylvania State University, ([email protected]) 2 Lawrence Berkeley National Laboratory, Berkeley CA 3 Earth and Environmental Systems Institute and Center for Environmental Kinetics Analysis, The Pennsylvania State University Production of saprolite from bedrock is an important source of both mineral detritus and dissolved ions. The rates at which saprolite is produced have far reaching impacts on important processes such as global flux of dissolved material to oceans, nutrient cycling in soils, sedimentation that ultimately leads to oil and gas generation, and the fate of atmospheric CO2 over geologic time scales. Despite their importance, saprolite production rates cannot currently be predicted. In low porosity rocks, like basalt, rates of saprolite production are dependent upon the ability of water to infiltrate the parent rock. Fracturing of these rocks by either physical or chemical mechanisms is one way of increasing available water. However, not all rocks fracture during weathering and in these systems porosity can be enhanced through the dissolution of primary phases. We are currently using weathering rinds developed on unfractured basalt clasts as a natural system where we can study processes and rates of saprolite formation over long time periods (35-250 ka). Across a weathering interface approximately 2 mm wide parent basalt weathers to form saprolite that is completely depleted in Ca, Na, Si and Mg. Associated with the weathering is an increase in porosity from 3-50%. Theoretical calculations of diffusion limited weathering predict that the weathering rind will grow as a function of the square-root of time (t0.5). However, these calculations are for constant porosity systems and many studies of weathering rinds indicate a t0.8 dependence. By including dissolution enhanced porosity in reactive transport models of the basalt clasts, we are able to predict both the geometry of the weathering interface and time dependence of rind formation. Given similar lithology and climate, the knowledge we have gained by studying saprolite production on basalt clasts should provide insight into weathering of basalt watersheds. However, making comparisions between watershed denudation rates and weathering rind advance rates is complicated due to the difference in methods used to measure surface area. We compare weathering advance rates (mm yr-1) calculated from laboratory, clast, soil profile and watershed studies using fractal theory in an attempt to reconcile basalt weathering rates across scales.

Goldschmidt Conference Abstracts 2007

A708

The impact of subducting sediment on the HFSE budget of arc-related igneous rocks, Banda arc, Indonesia

Tracing Earth’s first crust with Hf isotopes in zircons from the Narryer Gneiss Complex, Australia

O. NEBEL1, M.L.A. MOREL1, P.Z. VROON1, G.R. DAVIES1 AND M.J. VAN BERGEN2

Y. NEBEL-JACOBSEN1, C. MÜNKER1,2, K. MEZGER1, O. NEBEL3, A. GERDES4 AND D. NELSON5

1

Department of Petrology, Vrije Universiteit Amsterdam, The Netherlands 2 Department of Petrology, Universiteit Utrecht, The Netherlands Subduction processes and associated arc volcanism play a key role in determining the chemical composition of continental crust. The contribution of subducting sediment to arc magmatism is still poorly constrained but potentially accounts to a significant extend for the HFSE budget of arcrelated igneous rocks. The Banda arc in Indonesia has a unique geologic setting that allows HFSE behaviour in a subduction zone to be studied. The amount of subducting sediment being dumped into the subduction regime increases from NE-SW from 0.1 to 5 wt.% [1]. Samples collected along the arc front, well defined for their Nd-Pb-Sr-O isotopes and trace elements [2,3], are analyzed for their Hf isotopes and HFS element concentrations. Hafnium isotope data, ranging from –3.4 HHf to +13.0 HHf, and negative-correlated Hf-O isotope co-variations support the increase in sedimentary influx. The Hf isotope data are systematically decoupled from Nd isotopes, most likely due to partially enhanced fluid-melt mobility of LREE. Zr/Hf exhibit no variation with Th/Yb, which points to no significant decoupling of Zr from Hf with elevated amounts of subducting sediment. In contrast, Nb/Yb is significantly elevated with Th/Yb, pointing to enhanced Nb concentrations with increasing sediment subduction. High precision Nb/Ta data will be used to monitor variations in HFSE caused by sediment subduction and potential residual mineral phases.

References: [1] Vroon et al, (1995), GCA, 59, 2573-2598; [2] Vroon et al. (1993), JGR, 98;B12, 22349-22366; [3] Vroon et al., (2001), GCA, 65, 589-609

1

Universität Münster, Germany ([email protected]) Universität Bonn, Germany 3 Vrije Universiteit Amsterdam, The Netherlands 4 Universität Frankfurt/Main, Germany 5 Curtin University, Perth, Australia 2

Witnesses of the infant crust on Earth are rare because no rocks that formed in the fist 500 Myrs are preserved and the only remnants of Hadean material are zircon crystals that survived reworking of their host rocks. In order to unravel the evolution of the Earth’s oldest crust, Archean and Hadean detrital zircons from the Mt. Narryer Gneiss Complex were investigated for their U-Pb and Lu-Hf isotope systematics. The U-Pb systematics of six zircons from the Meeberrie Gneiss define an upper concordia intercept age of 3300 Ma and two lower intercepts at ~ 500 Ma and 700 Ma, which we interpret as the formation ages of the gneiss protolith and later thermal overprints, respectively. Corresponding HHfT range from -8.6 to -11.2. Eight zircons from a metasediment from Mt. Narryer show ages from 3.2 to 4.2 Ga with corresponding eps Hf of +3.4 to -7.1. Two zircons from a metasediment from Jack Hills are 4200 Ma old and have HHfT of -0.5 and -2.9. The age distribution and corresponding initial Hf isotopes indicate similar source regions for the Mt. Narryer metasediments and Meeberrie Gneiss. Older (>3.8 old) grains from Jack Hills and the Mt. Narryer metasediments origin from sources that already formed in the Hadean.. The combination of these new data with published zircon data, indicate that crustal growth during the first 500 Myrs occurred more or less continuously by formation of small crustal domains.

Goldschmidt Conference Abstracts 2007

Trace element partitioning in the granulite facies F. NEHRING1, S. FOLEY1 AND P. HÖLTTÄ2 1

Institute of Geosciences, Johannes Gutenberg-University Mainz ([email protected]) 2 Geological Survey of Finland, Espoo Granulites from Central Finland experienced pTconditions of 800-950°C at 9-11 kbar during a major collisional event in the late Archean. Anatexis was induced by dehydration melting of amphibole leading to the formation of cpx, opx, garnet and tonalitic melts. We constrained granulite facies trace element distribution by in situ LA-ICP-MS studies of the principal mineral phases. Our analyses indicate that equilibrium was attained between most phases. Garnets grew in microdomains and only equilibrated with their immediate surrounding. Hence, garnet exerts limited control on the bulk partition coefficient. Accessory phases are important hosts of REE (apatite), HFSE (ilmenite) and transition elements (ilmenite, magnetite). Using predictive models, comparisons with published data and relationships between minerals and observed melt compositions, we formulate Dmineral/melt values that are applicable for trace element modelling under lower crustal conditions; these are broadly similar to magmatic values for intermediate melt compositions. Uniform Damph/cpx for the REE were integrated into the set of D-values so that Dmineral/melt = 3uDamph/cpx We provide the first self-consistent set of D-values for Sc, V, Cr and Ni between cpx, amph, grt, opx, bt and ilm:

Sc 1 cpx 0.8-1.2 grt amph 0.8-1.2 opx 0.15-0.25 0.15 bt 0.15-0.3 il

D ( mineral / cpx ) V Cr 1 1 0.4-0.6 1-2 1.6-2.5 1-3 0.2-0.5 0.2-0.5 2.2 2.5 1 1-1.7

Ni 1 0.04 1.6-2.5 1 3.6 0.2-0.3

D ( mineral / melt ) Sc V Cr Ni 11 7.3 2.5 1.7 14.5 2.6 2.5 0.06 11 12 6 3.2 2.7 2 1 1.7 1.6 13 6.3 6.1 4 12 3 0.5

Ilmenite was found to strongly influence the distribution of Nb and Ta and has partition coefficients an order of magnitude higher than amphibole. Thus, ilmenite has to be considered in future discussions about the role of melting of amphibole-bearing source rocks as the driving mechanism for Archean crust generation.

A709

Geochemistry of I-type granodiorite and tin-bearing S-type granites from Gouveia area, central Portugal A.M.R. NEIVA Department of Earth Sciences, University of Coimbra, 3000-272 Coimbra, Portugal ([email protected]) The Gouveia area is located in the Iberian Massif, which is a large segment of the European Variscan Belt. Granitic rocks intruded a Cambrian schist-metagraywacke complex and predominate in the area. SHRIMP U-Pb zircon age for a peraluminous biotite I-type granodiorite is 481.8r5.9 Ma. So, it was emplaced during the Early Ordovician. The granodiorite has (87Sr/86Sr)482=0.7036, HNd482=2.5 and G18O=8.84‰, which indicate that it contains a mantle component. Seven peraluminous two-mica S-type granites are Variscan, as SHRIMP U-Th-Pb monazite ages range from 288 to 304 Ma and record granite emplacement. They have average Sn contents of 16-40 ppm. Most granites define individual fractionation trends, have REE patterns that intersect each other and distinct initial 87Sr/86Sr ratios ranging from 0.7086 to 0.7129 and G18O values of 10.34–13.34 ‰ and represent different pulses of granite magma. Their zircon core ages range from a300 to 2100 Ma suggesting that these granites were derived by partial melting of sediments having a range of sources. Granites contain zircon cores similar in age to the granite. One of these granites and another granite define a sequence shown by fractionation trends for major and trace elements of granites and their micas, subparallel REE patterns of granites and a common whole-rock Rb-Sr isochron. Least square analysis of major elements and modelling of trace elements indicate that the younger granite is derived from the older granite by fractional crystallization of quartz, plagioclase, biotite and ilmenite. Initial 87Sr/86Sr ratio and G18O values increase from the older to the younger granite suggesting that fractionational crystallization was accompanied by assimilation of metasedimentary materials. Fractional crystallization was the mechanism responsible for the increase in Sn content in individual granites, sequence and their micas, but it is crucial that the granite melt contains at least 8 ppm Sn. Muscovite retains a higher granite tin content than biotite does. Cooling 39Ar-40Ar plateau ages of micas through the Ar closure temperature range from 285 to 293 Ma. There is no significant difference between these ages if errors are taken into account. When the four youngest S-type granites were emplaced at 288-290 Ma, I-type granodiorite emplaced at 482 Ma and the three oldest S-type granites emplaced at 301-304 Ma were affected by the Saalian Variscan movements and their mica 39Ar-40Ar ages record recrystallization or Ar loss. Trace elements of biotite from the I-type granodiorite were affected, while trace elements of micas from the three oldest S-type granites define individual fractionation trends.

Goldschmidt Conference Abstracts 2007

A710

Improved U-(Th)-Pb dating of monazite by ion microprobe: Correcting for an isobaric interference of PrP04 on 204Pb DAVID R. NELSON

1

AND ANDREAS MÖLLER

2

1

Department of Applied Physics, Curtin University, Bentley 6102, Australia ([email protected]) 2 Institut f. Geowissenschaften, Univ. Potsdam, D-14776 Potsdam-Golm, Germany ([email protected]) High U and Th abundances, resistivity to diffusion and resilience to weathering make monazite one of the most important minerals for U-Th-Pb dating. Chemical characteristics (e.g. Y content, REE pattern) can be used as indicators of monazite growth reactions, and allow correlation of age data with metamorphic, magmatic or hydrothermal processes. It has been noted that ion microprobe analysis of Th-rich monazite can yield discordant Pb-U and Pb-Th dates (e.g. Stern & Berman 2000; Zhu & O'Nions 1998), or may show excess scatter on the 207Pb/206Pb dates. Both effects compromise the analytical accuracy and the geological interpretation. These effects may be attributable to matrix effects, i.e. subtle crystallographic differences may result in different ablation and Pb/U characteristics of unknowns and the monazite inter-element fractionation standard used during SIMS analysis. In comparison to zircon, commonly occurring monazite cover a much wider range of chemical compositions. To evaluate the effects, we have systematically studied a suite of monazite reference materials with a range of chemical compositions, from the brabantite (CaTh(PO4)2) as well as the huttonite (ThSiO4) solid solution series, and with varying Yttrium content. Independently determined age data exist for all these materials from TIMS measurements. We propose that the observed excess scatter on the 207 Pb/206Pb ratios results from an isobaric interference of PrP04 on 204Pb that has so far been ignored in ion microprobe data reduction and correction procedures. Using a correction based on the La/Ce counts, routinely measured in monazite analytical session on SHRIMP, and extrapolating to Pr, a correction can be applied before further corrections for Threlated “excess 204Pb are applied. Our correction routine has been built into the recent version of CONCH, a visual Basic program for processing of ion-microprobe analytical data developed by Nelson (2006). 

Acknowledgements Thanks to N. Rayner, D. Upadhyay, J.W. Boyce, A. Kennedy and P.D. Kinny for donating material, to D. Rhede and E. Thern & A. Frew expert expert help on the EPMA and SHRIMP, respectively.

References Stern R.A. and Berman R.G. (2000) Ch. Geol. 172, 113-130. Zhu X.K. and O'Nions R.K. (1998) Ch. Geol. 144, 305-312. Nelson D.R. (2006) Comp. & Geosc. 32, 1479-1498.

Surface complexation modeling of the adsorption of both arsenate and copper to the surface of goethite H. NELSON, S. SJÖBERG AND L. LÖVGREN Department of Chemistry, Umeå University, SE-901 87 Umeå, Sweden ([email protected]) To understand the migration and mobility of pollutants in soils, it is important to have knowledge about the sorption processes at the mineral-water interface. Adsorption of metals and ligands to mineral particles in aqueous solution can be described as the formation of surface complexes at specific surface sites. In complex, competitive, multicomponent systems, surface complexation modeling can be a good tool to predict ion adsorption and complex formation. The models are valid over a broad range of pH values and reactant concentrations, provided that they are established using an extensive set of experimental data. Adsorption processes in ternary systems with a mineral phase, a metal ion, and an inorganic anion have seldom been treated using surface complexation modeling. The focus in this study is the ternary arsenate-coppergoethite system. Goethite (Į-FeOOH) is an iron oxyhydroxide that is common in soils and plays a major role in the speciation of metals and (in)organic anions.. Arsenate and copper are components of a wood preserving liquid that has contaminated soils at many sites. Since the two pollutants are emitted simultaneously it is important to study how their coadsorption onto soil minerals impacts their speciation. Surface complexation models for the ternary arsenatecopper-goethite system and the copper-goethite and arsenategoethite subsystems will be presented. The surface complexation modeling is based on results from potentiometric titrations, adsorption experiments and spectroscopic data. The Basic Stern Model, together with the concept of charge distribution, is used to adjust for the distribution of charge between different planes at the goethite surface.

Goldschmidt Conference Abstracts 2007

Heterogeneity of MORB composition along the eastern part of the Southwest Indian Ridge NATSUKI NEO AND SUMIO MIYASHITA Graduate School of Science & Technology, Niigata University, Niigata, 950-2181 Japan ([email protected]; [email protected]) Regional variations of basalt compositions from the eastern part of the Southwest Indian Ridge (SWIR) were shown by Natland (1991), Meyzen et al. (2003) and Robinson et al. (1996). However, the definitions for the regional variation along the SWIR are different from each other in above studies. We have analyzed major and minor element compositions (including REE) of basalts obtained from 55E to 70E along the SWIR. In this study, We discuss for the regional variation based on our result and PETDB. We defined from 55E to Rodriguez Triple Junction (RTJ) into 4 areas; Area I: 68E to RTJ (70E)᧨Area II: 60E to 68E᧨Area III: 57E to Melville F. Z. (60E)᧨ Area IV: 55E to Atlantis II F. Z. (57E). The basaltic rocks from Areas I&IV, II and III show different trends, indicating that they were not derived by simple fractional crystallization, but also by a different degree of partial melting. The basalts of Areas I &IV show a N-MORB-like REE pattern, while those of II show enriched pattern in LREE. The Area II basalts are also characterized by high Na2O contents (3.5~4.5 wt%; Mg# 57~67) nevertheless with less-evolved composition. Area III basalts are intermediate~both in REE pattern and Na2O contents. We conclude that enriched source in terms of LREE and Na2O exists under Area II. Whereas Area I & IV are underlain by N-MORB-like source. The intermediate composition in Area III may be explained by mixing between enrich source and N-MORB source. It is noted that the enriched source of Area II is completely distinct from E-MORB signature.

References Natland J. H. (1991) Oceanic Basalts (ed. Floyd P. A.) 288310 Meyzen. M. C. et al., (2003) Nature, 421, 731-733 Robinson, C. J. et al., (1996) Geological Society Special Publication, 118, 131-141

A711

Boron influence on biochemical reactions in natural zeolites A. NEUBECK, M. IVARSSON AND N.G. HOLM Department of Geology and Geochemistry, Stockholm University, 106 91 Stockholm, Sweden, ([email protected], [email protected], [email protected]) The special zeolitic characteristics such as cation exchange, molecular sieve capacity, open crystal framework and reversible dehydration make them interesting in many aspects. They may be used as absorbents for pollutants in the agricultural and environmental industry, fertilizers and in the treatment of nuclear wastewaters. Another use is to investigate their role as catalysts or adsorbents of organic compounds in their crystal channels, which provide a protected microenvironment for chemical reactions. Adsorbance of labile biomolecules onto the zeolite crystal tunnel walls may stabilize and therefore trigger their readiness to further react, possibly in polymerization reactions. It has also recently been shown that borate minerals stabilize ribose, a highly thermo-labile pentose, by forming complexes1. Even synthesis of pentoses is possible in the presence of borate. Thus, boron containing minerals may have been important in the development of larger biomolecules on the early Earth. In our project, we are going to do gaseous and aqueous adsorption tests on natural zeolites, with and without boron in their crystal structure. In some of our material, boron has substituted for beryllium and possibly silica in the zeolitic crystal framework. The combination of the microenvironment given from the zeolite and the stabilizing properties of boron may be ideal for some organic reactions. Until now, most adsorption tests have been made on synthetic zeolites for industrial usage. Natural zeolites will provide a better analogue to actual earthly systems than synthetic ones.

Reference Ricardo A., Carrigan M.A., Olcott A.N., Benner S.A., (2004), Science 303, 196

A712

Goldschmidt Conference Abstracts 2007

Mo isotopes in modern euxinic environments: Water column and sediment data

Extremely refractory oceanic lithospheric mantle and its implications for geochemical mass balance

N. NEUBERT1, T.F. NÄGLER1 AND M.E. BÖTTCHER2

E.-R. NEUMANN1, N.S.C. SIMON1, C. BONADIMAN2, M. COLTORTI2, G. DELPECH3 AND M. GRÉGOIRE4

1

Institute of Geological Sciences, Isotope Geology Group, University of Bern, Switzerland ([email protected], [email protected]) 2 Leibniz Institut for Baltic Sea Research, Warnemünde, Germany ([email protected]) The present study investigates the pathways of molybdenum (Mo) scavenging under euxinic conditions following the fractionation of stable Mo isotopes in three modern settings: The Black Sea, the largest permanently euxinic basin, and two anoxic troughs of the Baltic Sea. Water column profiles, as well as surface sediment samples, were recovered from different water depths and analyzed for the concentration and isotopic composition of Mo, besides other geochemical parameters. Mo is a redox-sensitive trace metal which is soluble as the molybdate oxyanion in oxic seawater with a residence time of about 800 ka. The isotope signature of Mo is a relatively new proxy used to reconstruct the paleo-redox conditions of the Earth’s atmosphere and the oceanic system. The Mo isotope composition in seawater is homogeneous, as shown by [1]. Scavenging of Mo under euxinic conditions is related to the amount of free H2S. Near total removal of Mo from the water column is reached at a H2Saq concentration of 11±3 μM [2]. In the Black Sea this corresponds to a water depth of about 400 m. Sediment samples of the Black Sea from more then 400 m water depth show seawater isotopic composition, agreeing completely with the assumption of bulk Mo removal. However, shallower sediments deposited under lower H2S concentrations show significant Mo isotope fractionation. Brackish Baltic Sea surface sediment samples are taken from two separate basins, the Gotland Deep and the Landsort Deep which have maximum water depths of 248 m and 459 m, respectively. The Baltic Sea oceanographic conditions, including temporary bottom water oxygenation due to sporadic North Sea water inflows, are more complex than in the Black Sea. The dissolved sulfide concentrations in the water column are less than 50 μM and the salinity is less than 22 PSU. In the anoxic part, Mo isotope signatures are shifted towards heavier values indicating in-situ fractionation upon scavenging under euxinic conditions. The surface sediments are the sink for reduced Mo and show Mo fractionation similar to the oxic to slightly euxinic sediments of the Black Sea. Intra-basinal differences and downcore variations can be explained by changes in the bottom water redox conditions due to episodes with different inflow intensities and reflect varying concentrations of dissolved sulfide in the water column.

References [1] Siebert, C., Nägler, T.F., von Blanckenburg, F., Kramers, J.D. (2003) EPSL. , 211(1-2):159-171. [2] Erickson, B.E., Helz, G.R., (2000). GCA, 64 (7)

1

Physics of Geological Processes, University of Oslo, P.O.Box 1048 Blindern, N-0316 Oslo, Norway 2 Earth Science Department, University of Ferrara, Via Saragat, 1 44100 Ferrara, Italy 3 UMR CNRS 8148 IDES, "Interactions et Dynamique des Environnements de Surface", Faculté des Sciences, Université Orsay-Paris Sud, Bât.504, 91405 ORSAY Cedex 4 Observatoire Midi-Pyrénées, UMR 5562-DTP, OMP, Université Toulouse III, Avenue E. Belin, 31400 Toulouse, France In several ocean islands (e.g. the Canary Islands, Kerguelen, Cape Verde and Savai’i) the mantle xenoliths population is dominated by extremely refractory spinel harzburgites. These harzburgites (OI1) show no petrographic evidence of metasomatism, and carry no primary clinopyroxene. However, minor metasomatism may be reflected in the incompatible trace elements and radiogenic isotopes. The major element and modal compositions of the OI1 harzburgites correspond to residual mantle formed from a primordial source after 25-30% partial melting, leaving a clinopyroxene- and garnet-free residue. These peridotites are, on average, significantly more refractory than MOR peridotites. Estimated P-T conditions (850-1200ºC and 0.7-1.3 GPa) indicate that the OI1 peridotites last equilibrated within the abyssal mantle lithosphere. Their textures and high solidus temperatures imply that their strongly refractory nature cannot be the result of processes (“additional” partial melting or melt – wall-rock reactions) associated with the present plumes. Although interaction with plume melts has caused minor enrichment in radiogenic isotopic ratios and the most strongly incompatible elements, their original (pre-ocean-island) compositions are preserved in the most abundant major elements and in the modal relationships. Highly refractory OI1 spinel-harzburgites have higher solidi than more fertile, clinopyroxene-bearing peridotites, and are likely to be resistant to further partial melting. We interpret the OI1 peridotites as fragments of recycled, “sterile” asthenospheric material that have been trapped in the abyssal mantle lithosphere. These peridotites imply that the convecting mantle includes material that is significantly more refractory than MOR peridotites. The OI1 peridotites are buoyant relative to less refractory, denser mantle material and may preferentially accumulate at the top of the convecting mantle, where they freeze to the base of newly formed abyssal lithosphere.

Goldschmidt Conference Abstracts 2007

A713

Distribution and redox status of Arsenic in framboidal pyrite from estuarine sediments

Degassing of oceanic H2S and its delivery to terrestrial ecosystems during the Permo-Triassic extinction

T. NEUMANN1, F. SCHOLZ1, U. KRAMAR1, R. SIMON2 2 AND S. MANGOLD

R.J. NEWTON1, P.B. WIGNALL1, S.H. BOTTRELL1 AND I. METCALFE2

1

1

Synchrotron μXRF measurements of 106 μm-sized pyrite framboids from anoxic sediments of the Achterwasser lagoon, a part of the estuarine system of the river Oder, SW Baltic Sea, show highly variable As concentrations ranging from 6 to 1140 μg/g within the uppermost 50 cm of the sediment column. Even within a single depth layer, the As concentration of different framboids vary greatly with no clear depth trend visible. Median values calculated from μXRF data for each depth layer are similar to the chemical extraction data (28 to 143 μg As per g pyrite), suggesting that the applied extraction procedure yields realistic average As concentrations in pyrite from bulk sediment. Pyrite can account for 9 to 55% (average 22%) of the total As budget of the sediments indicating that authigenic pyrite plays an important role in the geochemical cycling of As in coastal sediments. High-resolution μXRF mapping of single pyrite grains shows that As is distributed inhomogeneously within larger framboids, and more regularly in the smaller pyrite grains. This may be due to changing pore water composition during pyrite growth, with larger and older framboids being subjected to more frequent changes in the pore water composition as compared to the smaller and younger framboids from the same depth layer. XANES spectra indicate that a considerably part of As is present in higher valence state (+III/+V) in samples from near the sediment/water interface. This can be explained by frequently occurring resuspension of the surficial sediments to the oxic water column due to wave action and subsequent redeposition. This process might be responsible for adsorption of As(V) oxyanions onto pyrite. Reduced As(-I) species become more important in the deeper samples, reflecting decreasing redox potential and increased time since deposition. This suggests that adsorbed As(V) and As(III) species will be reduced by the sulfidic pore waters prior incorporation during pyrite growth.

Evidence is growing that drastic changes in the oceanic sulfur cycle were intimately linked with the greatest extinction of the Phanerozoic at the Permo-Triassic boundary. Isotopic changes in proxies for seawater-sulfate suggest the transformation of a significant portion of the marine sulfate reservoir to dissolved sulfide by bacterial sulfate reduction during the late Permian (Newton et al., 2004). Rapid return to 34 S depleted sulfate at the extinction level can be explained by ocean overturn and re-oxidation of this sulfide, providing a kill mechanism in the form of H2S poisoning. Modelling using end-Permian boundary conditions permits the escape of this sulfide from the surface of the ocean (Kump et al., 2005), so ocean overturn and sulfide release could therefore link the marine and terrestrial extinctions which are closely spaced in time. Here we present new results from lacustrine sediments from north-west China, and re-interpret existing results from South Africa (Maruoka et al., 2003). Sulfate concentrations in freshwater lake systems are low when compared to supplies of reactive Fe and organic matter. Hence, sulfate is often quantitavley reduced and preserved as pyrite in lake sediments, recording the isotope composition of the original sulfate. Isotope data from both localities record relatively stable backgrounds (į34S between -4 to +5‰, China, and 0 to +5‰, South Africa), within the normal range for weathered sulfate from large geologically diverse catchments. Close to the P-Tr boundary these stable records are interupted by rapid negative shifts to around -12‰ į34SVCDT (both locations). These excursions could not be caused by inputs of volcanogenic sulfur (į34S -5 to +5‰), and provide the first direct evidence of transfers of oceanic derived H2S to terrestrial ecosystems. The exact role of H2S in the terrestrial extinction is unclear. Possible mechanisms include direct poisoning, acidification, or as part of an ozone depletion scenario, leading to radiation exposure (Lamarque et al., 2007).

Institut für Mineralogie und Geochemie, Universität Karlsruhe, 76131 Karlsruhe, Germany ([email protected]) 2 Institut für Synchrotronstrahlung, Forschungszentrum Karlsruhe, 76028 Karlsruhe, Germany

School of Earth and Environment, University of Leeds, UK, LS2 9JT ([email protected]) 2 Asia Centre, University of New England, Armidale NSW 2351, Australia

References Lamarque, J.-F., Kiehl, J.T. and Orlando, J.J., Geophys. Res. Lett., 34, doi:10.1029/2006GL028384 Kump, L.R., Pavlov, A. and Arthur, M.A., (2005), Geology, 33: 397-400. Newton, R.J., Pevitt, E.L., Wignall, P.B. and Bottrell, S.H., (2004), EPSL, 218: 331-345. Maruoka, T., Koeberl, C., Hancox, P.J. and Reimold, W.U., (2003), EPSL, 206: 101-117.

Goldschmidt Conference Abstracts 2007

A714

Isotopic evidence for Uranium retardation in zeolitic rocks at Yucca Mountain, Nevada

Weathering of Ca- and P-bearing minerals by fungi in a northern hardwood forest

LEONID A. NEYMARK

C.A. NEZAT1, J.D. BLUM2 AND D.R. ZAK3

U.S. Geological Survey, DFC Box 25046, MS 963, Denver, CO 80225 ([email protected]) Retardation of radionuclides by sorption in the nonwelded tuffs below the proposed high-level nuclear waste repository in the unsaturated zone at Yucca Mountain, Nevada, is a positive attribute of the natural barrier. Alteration of volcanic glass in these tuffs produced thick, widespread zones of zeolite- and clay-rich rocks with high sorptive capacities. The alteration of glass to zeolites, however, was accompanied by reduction in the matrix porosity and permeability causing most flow to occur through fractures, which may have decreased the overall effectiveness of radionuclide retardation in the zeolitic rock matrix. Chemical and uranium-series isotopic compositions were measured in samples of unfractured and rubble rock from core, surfaces of natural fractures in the core, and in rock leachates and pore water extracted from these core samples. Uranium concentrations in sodium acetate leachates indicate that the mobile 238U is 0.3 to 1.7 percent of total 238U in rock samples and allow estimates of the time-integrated in situ U distribution coefficient Kd (Kd=Csolid/Cwater, where C is concentration). Use of median U concentrations in pore water (0.005 Pg/mL) and sodium acetate leachates (0.035 Pg/g rock) yields an estimate of the 238U Kd value of 7 mL/g. Samples of rock from unfractured core, rubble core, and fracture surfaces have similar 234U/238U activity ratios (AR) ranging from 0.92 to 1.16, indicating both enrichments and depletions in the daughter 234U relative to the parent 238U. In contrast to the rock, all pore water and rock leachate samples have elevated 234U/238U AR ranging from 1.1 to 5.2. The chemical and isotopic data indicate that the matrix in zeoliteand clay-rich rocks is capable of exchanging uranium with 234 U-enriched percolating water and that retardation of radionuclides can occur in altered rocks below the proposed repository.

1

Department of Geological Sciences, University of Michigan, Ann Arbor, Michigan, USA ([email protected]) 2 Department of Geological Sciences, University of Michigan, Ann Arbor, Michigan, USA ([email protected]) 3 School of Natural Resources & Environment, University of Michigan, Ann Arbor, Michigan, USA ([email protected]) In many ecosystems, mineral weathering is the initial source of calcium and phosphorus, both of which are essential nutrients for plants. Mycorrhizal fungi are symbiotically associated with tree roots and exude organic acids that enhance mineral dissolution and transfer nutrients to trees (e.g., Jongman et al., 1997). Understanding this process in forests of the northerneastern USA is important because acid deposition has increased Ca leaching from Ca-poor soils such as those with granitic parent material. To test the role of fungi in weathering of Ca- and/or Pbearing minerals, mesh bags containing quartz (as a control), quartz plus 1% wollastonite (CaSiO3), or quartz plus 1% apatite (Ca5(PO4)3F) were buried in mineral soil beneath American beech (Fagus grandifolia Ehrh.), sugar maple (Acer saccharum Marsh.), and mixed spruce and basalm fir stands (Picea rubens Sarg. and Abies balsamea L.) at the Hubbard Brook Experimental Forest, New Hampshire, USA. A 50-Pm mesh size was chosen to exclude roots but allow fungal hyphae to enter the bags. Microbial community composition and biomass in the mesh bags and surrounding soil were characterized and quantified using phospholipid fatty acid (PLFA) analysis. Fungal biomass (estimated as moles of fungi-specific PLFA 18:2ǔ6 and 18:1ǔ9c) in the soil and control bags did not differ significantly among stand types. In contrast, the degree of fungal colonization in apatite- and wollastonite-amended bags varied significantly with stand type. In the beech stands, fungal biomass was significantly greater in the apatiteamended bags, suggesting that apatite dissolution stimulated fungal colonization. In the spruce-fir stands, the fungal biomass did not vary as a function of the mineral assemblage, suggesting that the natural supply of Ca and P in these stands is high enough to meet nutrient demands. In the sugar maple stands, fungal biomass was significantly lower in the wollastonite-amended bags relative to the control and apatiteamended bags. The different response in each stand type appears to be related to the type of fungi present as well as the nutrient status of the soils. These results are important for evaluating nutrient demands of vegetation, and the effects of microbial community composition on the mineral dissolution of Ca and P-bearing minerals.

Reference Jongmans A.G. et al., (1997), Nature 389, 683-683.

Goldschmidt Conference Abstracts 2007

The accuracy of G11B measurements of foraminifera

H2O diffusion in rhyolite at 1-2 GPa HUAIWEI NI AND YOUXUE ZHANG Department of Geological Sciences, The University of Michigan, Ann Arbor, MI 48109-1005, USA ([email protected]; [email protected]) Water diffusion in silicate melts is crucial for understanding various volcanic processes, including bubble formation and growth, magma degassing, and magma fragmentation. H2O diffusivity in rhyolite were previously measured at 0.1-810 MPa (Zhang et al., 1991; Zhang and Behrens, 2000). To better constrain the pressure effect on H2O diffusion, we conducted diffusion-couple experiments at 407-1629°C and 1-2 GPa in piston-cylinder apparatus. After experiments, samples are doubly polished to ~200 μm thickness, and H2O concentration profile along the centerline is measured using FTIR microscope. Compared to Zhang and Behrens (2000), H2O diffusivity at 1-2 GPa increases less rapidly with water cotent. H2O diffusivity in rhyolite melt decreases from one atmosphere to 2 GPa, and such negative pressure effect is more remarkable as the temperature decreases, which leads to a larger activation energy at higher pressure. If the model of Zhang and Behrens (2000) at <2 wt.% water content is extrapolated to 2 GPa, the prediction results in roughly the same activation energy, but the predicted diffusivity is systematically lower than our measurements by a factor of about 3, which suggests a smaller pressure effect and therefore a smaller activation volume. By combining previous data and our measurements and assuming total H2O diffusivity is proportional to total H2O concentration, a new H2O diffusivity model at <2 wt.% water content has been constructed, which better constrains the pressure dependence. We are also developing a general H2O diffusivity model applicable to higher water contents and 0.12000 MPa.

References Zhang Y., Stolper E. M., and Wasserburg G. J. (1991) Diffusion of water in rhyolitic glasses. Geochim. Cosmochim. Acta 55, 441-456. Zhang Y. and Behrens H. (2000) H2O diffusion in rhyolic melts and glasses. Chem. Geol. 169, 243-262.

A715

Y. NI1,2, G.L. FOSTER2 AND T. ELLIOTT2 1

Research Institute of Petroleum Exploration & Development, Beijing, China, 100083 ([email protected]) 2 Bristol Isotope Group, Department of Earth Sciences, Universtiy of Bristol, Bristol, BS8 1RJ, UK The pH of seawater reflects the speciation of dissolved carbon and hence also, in part, the extent of the partitioning of carbon dioxide between the oceans and atmosphere. Reconstructions of past seawater pH therefore allow investigations of, for example, the concentration of CO2 in the past and the mechanisms of glacial-interglacial CO2 change. A number of contributions have demonstrated that the boron isotopic composition of marine carbonates, such as foraminifera, record the pH of the seawater in which they grew. Sample size requirements have led to most workers measuring the small amounts of boron in foraminifera by negative-ion thermal ionisation mass spectrometry (NTIMS), which has very high ion yields. What is more, this technique uses the Ca (plus trace metals) of the dissolved marine carbonate as an activator, requiring no additional purification and/or concentration of boron, which greatly simplifies the approach. We show here, by a comparison between the G11B of samples and solutions measured by both total evaporation NTIMS and multicollector inductively coupled plasma mass spectrometry (MC-ICPMS), that this latter feature, although seemingly advantageous, results in a significant bias in the G11B measurement by NTIMS. We conclude that organic carbon, probably hosted within the foraminiferal shell and hence inadequately oxidised during cleaning, has a detrimental and variable effect on the thermal ionisation of boron. This effect does not necessarily relate to the presence of CN- molecules and results in the TE-NTIMS G11B measurements being shifted to heavier not lighter values as would be expected from CNO- interference at mass 42. This matrix effect may go some way to explaining the reported large G11B differences between laboratories for similar foraminiferal samples (up to 6 ‰ disagreement by NTIMS). With careful treatment, reliable G11B data can be generated by NTIMS. However, we hope that this realisation, along with a modified loading technique outlined here, makes G11B isotope measurements by negative thermal ionisation techniques more routine and accurate.

Goldschmidt Conference Abstracts 2007

A716

Generation and emplacement of granitic magmas at the Paleocene Rum Igneous Complex, northwest Scotland

Root zone of sheeted dike complex in Oman ophiolite-dynamical model A. NICOLAS AND F. BOUDIER Université Montpellier II, France ([email protected]; [email protected]) As in many other ophiolites, there is in Oman an horizon called the root zone of the sheeted dike complex (RZSDC) located below the volcanics and the diabase sheeted dike complex and above the gabbro unit. This RZSDC is described here based on structural and petrological observations carried during the extensive mapping of this ophiolite (Nicolas et al., 2000) and on a new fine-scale mapping in a selected area. The Oman ophiolite is derived from a fast spreading ridge where a melt lens is located above the main gabbro unit and just below the RZSDC. With a few exceptions, the RZSDC thickness is rather constant, in the 100 m range, with a crude internal stratigraphy. The base of RZSDC is a level of isotropic doleritic gabbros interpreted as a thermal boundary layer transitional between the magmatic system of the melt lens convecting at 1200°C, and a high temperature (HT) hydrothermal system convecting within the RZSDC at <1100°C. In this field of very sharp thermal gradient (~7°C/m), increasing seawater flux locally generates near total hydrous melting at the expense of these isotropic gabbros, at ~1100°C (companion abstract). The upper part of the RZSDC is richer in varitextured doleritic and pegmatitic gabbros, with trondjhemitic intrusions, formed by a similar process at temperatures below 1000°C. The latter formations constitute screens between the lower dikes of the sheeted dike complex. At the very base of this complex a new boundary layer, with HT diabase dikes, separates the preceding HT convective system from the well known LT (T<450°C) hydrothermal system operating throughout the sheeted dike complex, to the seafloor. Intruding the isotropic gabbros near the base of the RZSDC, protodikes with a distinctive microgranular border and a doleritic center testify that the RZSDC was generated by melt conduits issued from the melt lens. Each dike of the sheeted dike complex is thus fed by one protodike. However, protodike swarms are exceptional because, crystallizing ~1100°C, they are largely destroyed by dike-in-dike intrusions and by hydrous remelting

Reference Nicolas A., Boudier F., Ildefonse B.and Ball E., 2000. Marine Geophys. Res., 21/2-3, 147-179

G.R. NICOLL1, V.R. TROLL1, C.H. DONALDSON2, R.M. ELLAM3 AND C.H. EMELEUS4 1

Dept. of Geology, Trinity College Dublin, Dublin 2, Ireland ([email protected]) 2 School of Geosciences, University of St Andrews, St Andrews KY16 9AL, Scotland 3 S.U.E.R.C., Rankine Avenue, East Kilbride, G75 0QF, Scotland 4 Dept. of Earth Sciences, University of Durham, Durham DHI 3LE, England The Isle of Rum is one of several igneous complexes within the British-Irish Paleocene Igneous Province that formed prior to the opening of the North Atlantic Ocean during a period of crustal stretching and thinning. An elliptical ring fault, 12km across, bounds the igneous centre enclosing remnants of an early granitic phase of intrusive and eruptive activity that is cross-cut by basic and ultrabasic intrusions. This activity consists of intra-caldera rhyodacitic ignimbrites and microgranitic to granitic shallow-level intrusive bodies that are all located within, and in close proximity to the Main Ring Fault. These early granitic magmas show a strong involvement and interaction with crustal materials. Isotopic data implies 40-90% contamination of mantle-derived magmas by upper crustal Lewisian amphibolite-facies gneiss. Some of these felsic intrusions are intimately associated with mafic magmas. 87 Sr/86Sr isotope data from basaltic margins and lobate inclusions, to and within the granitic rocks, indicate that very few of these mafic components have uncontaminated mantlelike values. Most of them also show varying degrees of crustal contamination (87Sr/86Sr 0.7029 - 0.7140) by both Lewisian amphibolite-facies material and a component that is similar in isotopic composition to Lewisian granulite-facies gneiss. This isotopic information provides a unique window into the crustal structure beneath the Rum Igneous Complex and combined with detailed field investigations allow for an integrated approach to interpreting the generation, emplacement and eruption of anorogenic granitic magmas. We propose that basic and ultrabasic intrusions at depth, migrating towards the surface, heated and partially melted the more felsic components of the surrounding crust which gave rise to a voluminous, but ultimately short-lived, episode of granitic magmatism on Rum.

Goldschmidt Conference Abstracts 2007

Assessing the relative production rates of cosmogenic 3He and 21Ne in olivine, pyroxene and quartz

Hydrothermal alkali and oxygenisotope exchange in alkali feldspars controlled by dissolution reprecipitation mechanisms

S. NIEDERMANN1, P. PILZ1 AND M. GOETHALS1,2

D.R.D. NIEDERMEIER1, A. PUTNIS2 AND T. GEISLER2

1

GeoForschungsZentrum Potsdam, Germany ([email protected]) 2 Westfälische Wilhelms-Universität Münster, Germany ([email protected]) Currently used methods to derive cosmogenic 3He and Ne production rates (P3 and P21) for pyroxenes based on their major element composition yield inconsistent exposure ages by up to 40% (Niedermann et al., 2007), indicating a substantial flaw in our knowledge on production systematics of the noble gas isotopes, in spite of a general agreement of the models with experimental determinations of P3 in olivine and P21 in quartz. However, P3(ol) has never been directly compared to P21(qz), probably because these minerals do not coexist in mineralogical equilibrium. A rock of basaltic-andesitic composition sampled from a Pleistocene lava flow on the Puna plateau in NW Argentina, south of the Salar de Antofalla at 3660 m elevation, contained a quartz xenocryst from crustal contamination ~3 cm in size as well as olivine and pyroxene phenocrysts. Although the sample is not from a well-defined surface location (the exposure geometry may have changed during degradation of a lava tube), it provides the opportunity to study the ratios of P3 and P21 in olivine, pyroxene and quartz. The resulting P3(ol)/P21(qz)=7.80r0.55 and P3(px)/P21(qz)=8.00r0.62 are ~30% higher than the value obtained by comparing the experimental P3 in olivine (~116 at g–1 a–1) to the experimental P21 in quartz (~19 at g–1 a–1). The difference is similar to that resulting from different methods to calculate P3 and P21 from major element composition. A difference of 10% at most might be assigned to unfavorable irradiation conditions, if the quartz xenocryst was located at the very bottom of the sampled rock during most of the exposure. But still the results would imply P3(ol) ~20% higher or P21(qz) ~20% lower than the values usually assumed. The determinations of cosmogenic 3He and 21Ne in our sample are robust since they are based on a value of 7.9 RA for the magmatic 3He/4He ratio and an atmospheric trapped Ne composition. Any reasonable different assumption could only increase the P3/P21 ratios. The comparison of cosmogenic 21Ne in olivine (Mg-rich) and pyroxene (Mg-poor) may, in principle, be used for an experimental determination of the P21(Mg)/P21(Si) ratio. Our present data indicate a value of ~3.4, but with a large uncertainty of 30-50% which critically depends on the uncertainty of the 21Ne(ol)/21Ne(px) ratio. We hope to obtain a more accurate result by analyzing Ne in olivine and pyroxene from a well-defined surface sample. 21

Reference Niedermann S., Schaefer J.M., Wieler R. and Naumann R., (2007), Earth Planet. Sci. Lett., doi: 10.1016/ j.epsl.2007.03.020.

A717

1

Institut für Petrologie, Universität Bremen, Germany Institut für Mineralogie, Westfälische Wilhelms-Universität Münster, Germany

2

The re-equilibration mechanisms of solid phases in hydrothermal fluids are only poorly understood. Depending on the prevailing conditions a sequence of fluid-induced dissolution-reprecipitation or solid state processes may take place. In the present study, we have investigated the reequilibration of albite in the presence of a fluid phase. For this, we have experimentally treated a natural albite crystal in an aqueous KCl solution at 600°C and 2 kbars. To trace the exchange between mineral and fluid, the aqueous KCl solution was enriched with 18O (95 atom %). The major aim of this study was the characterization of the chemical and physical properties of the K-feldspar reaction rim that is known to form during the interaction of albite with an aqueous KCl-rich fluid under the chosen conditions (Labotka et al. 2004). Transmission electron microcopic investigations have shown that the interface between the pristine albite and the K-feldspar product phase is sharp on a nano-scale, with a strong diffraction contrast difference between both phases, suggesting a defect-rich K-feldspar product. A porous zone also evolved near the interface. Existing crystallographic orientation as well as twinning in the albite is preserved in the K-feldspar reaction rim. Powder X-ray diffraction analyses revealed a change from triclinic Al,Si ordered albite structure to an Al,Si disordered monoclinic structure in the K-feldspar. Furthermore, Raman spectroscopy has been applied to map the mass frequency shift of the Si-O-Si bending vibration near 476 cm-1 in the K-feldspar due to the incorporation of 18O. The frequency of this mode shifts to about 457 cm-1 in K-feldspar formed in 18O-enriched solution, reflecting a high enrichment of 18O in the K-feldspar lattice. In the K-feldspar formed in 18 O-enriched solution the highest frequency shift was found close to the interface, which coincides with an enrichment of the orthoclase component close to the interface, as revelaed by electron microprobe measurements. This indicates that the higher frequency shift close to the interface results from chemical changes rather than from a higher 18O content. An important observation is that there is no gradient in the frequency shift that could be related to diffusion of 18O. We propose that the new data are fully consistent with an interface-controlled dissolution-reprecipitation mechanism that operates at an inward moving replacement front.

Reference Labotka, T.C., Cole, D.R., Fayek, M., Riciputi, L.R., and Stadermann, F.J. (2004) Coupled cation and oxygenisotope exchange between alkali feldspar and aqueous chloride solution. American Mineralogist, 89, 1822-1825.

Goldschmidt Conference Abstracts 2007

A718

Systematic thallium isotope variation in Fe-Mn crusts: A proxy for changes in ocean chemistry? SUNE G. NIELSEN1, VERONIKA KLEMM2, DOUG LAROWE3, ALEX N. HALLIDAY1 AND JAMES R. HEIN4

Geochemical analysis of obsidian and the pattern recognition of sites spatial distribution in the chalcolithic of the Eastern Lake Urmia, Northwestern, Iran KAMAL A. NIKNAMI1 AND AHMAD C. AMIRKHIZ2

1

University of Oxford, Dept. of Earth Science, Parks Road OX1 3PR, UK 2 ETH-Zurich, Dept. of Earth Science, 8092 Zurich, Switzerland 3 Utrecht University, Dept. of Earth Science, 3508 TA Utrecht, Netherlands 4 USGS, 345 Middleton Road, Menlo Park, CA 94025, USA We have measured the thallium (Tl) isotope composition of samples from the Pacific Fe-Mn crust CD29-2, which have recently been dated using Os isotope stratigraphy [1]. The samples have a spatial resolution of about 1mm which equates to about 0.5 Ma. Thus far we have investigated the time interval spanning 60 to 30 Ma and large systematic Tl isotope variations ranging between H205Tl = +6.8 and H205Tl = +11.4 are apparent (where H205Tl = 104 x (205Tl/203Tlsample205 Tl/203TlNIST 997)/ 205Tl/203TlNIST 997). The most striking feature is a systematic increase from H205Tl = +6.5 to +10.5 occurring in the time interval 58 to 50 Ma. A small number of previous analyses at various depths in CD29-2 [2] are fully consistent with our more detailed investigation. Rehkämper et al [2] argued that these variations most likely reflected a change in the Tl isotope composition of seawater and this remains a firm possibility. However, recent theoretical calculations of the mechanism controlling Tl isotope fractionation have shown that there is a large equilibrium isotope fractionation between Tl1+ and Tl3+ [3]. Published Tl isotope compositions of modern Fe-Mn crusts [4] and altered mid ocean ridge basalts [5] are consistent with these two reservoirs exclusively incorporating Tl3+ and Tl1+, respectively. Hence, changes in the speciation of Tl in seawater may also be responsible for the observed Tl isotope variation recorded in Fe-Mn crusts. This proposition appears to be in agreement with the fact that the timing of Tl isotope change in CD29-2 coincides exactly with the massive change in atmospheric CO2 inferred from boron isotope compositions of planktic foraminifera [6]. We are currently performing thermodynamic calculations aimed at determining the speciation of Tl in seawater in order to investigate the potential for Tl isotopes to be used as a proxy for changes in the chemical and physical conditions of seawater.

References [1] Klemm et al, EPSL, 2005 [2] Rehkämper et al EPSL, 2004 [3] Schauble, GCA, 2007 [4] Rehkämper et al, EPSL, 2002 [5] Nielsen et al, EPSL, 2006 [6] Pearson and Palmer, Nature, 2000

1

Department of Archaeology, University of Tehran, Enghlab Street, Tehran, Iran ([email protected]) 2 Department of Archaeology, University of Tehran, Enghlab Street, Tehran, Iran ([email protected]) Results presented here are the chemical analysis of significant numbers of obsidians from several Chalcolithic sites in the eastern shores of Lake Urmia, Northwestern Iran, as part of a large and most comprehensive study of obsidian sources in this region. In this study, XRF is used to generate trace element data for some elements. A number of analytical methods such as Bivariate plots of the incompatible trace elements were used to separate the obsidian sources in study area and source separation was confirmed by using linear Discriminant analysis. The existence of multiple sites containing obsidian artifacts in the area of Urmia and on some of the other nearby regions, however, has enabled the study of specific spatial patterns of source exploitation and the trade mechanisms which resulted in the distribution of obsidian hundreds of kilometers away during the Chalcolithic period (ca. 5000-3000 BC).

Goldschmidt Conference Abstracts 2007

A719

Extreme Pb-isotope diversity in the sources of K-rich magmas in Italy: Evidence from melt inclusions

Volcanic arc development due to intraoceanic subduction: Numerical model

I.K. NIKOGOSIAN1, M.J. VAN BERGEN1, J.C. DE HOOG2, M.J. WHITEHOUSE3 AND S.H.J.M. VAN DEN BOORN1

KSENIA NIKOLAEVA, TARAS V. GERYA AND JAMES A.D. CONNOLLY

Department of Earth Sciences, Utrecht University, Utrecht, The Netherlands ([email protected]; [email protected]) 2 Department of Earth Sciences, University of Oxford, Oxford, United Kingdom ([email protected]) 3 Lab. for Isotope Geology, Swedish Museum of Nat. Hist., Stockholm, Sweden ([email protected])

Department of Earth Sciences, ETH - Zürich, CH-8092 Zurich, Switzerland ([email protected], [email protected], [email protected])

1

We present 170 SIMS (Cameca 1270) analyses of Pbisotope ratios of homogenized melt inclusions in Fo-rich (Fo9288) olivines from primitive lavas in Central-Southern Italy. The samples cover the entire compositional spectrum of potassium-rich magma series erupted from major PlioceneQuaternary volcanic centres in peninsular Italy. The inclusions contain 1-140 ppm Pb. For inclusions with > 5 ppm Pb (80% of the population) in-run precision was 0.50.05% (2ı) for 207,208Pb/206Pb, and 1.5-0.15% (2ı) for 206,207,208 Pb/204Pb. Our results significantly extend the existing bulk-rock data towards more extreme compositions. A second key observation is the strong isotopic variability that appears to exist within individual volcanic centres, within magma series, and even within individual lava samples, testifying that widespread mixing in magmatic systems involved melts derived from mantle columns that are isotopically heterogeneous on small volume scales. The Pb-isotope systematics confirm that magmas were derived from mantle sources that are largely binary mixtures between components with low- and high-radiogenic 206,207,208Pb/204Pb ratios, respectively. However, extrapolation of mixing trends points to isotopic variability in each of these end-members, both in the original mantle component and in the metasomatic component that is inferred to be of upper crustal/sedimentary origin. The nature of the ‘pre-metasomatic’ mantle of the Roman and Campanian Provinces is distinct, the former being MORB-type, the latter showing a modest contribution of the FOZO-HIMU signature seen at Vulture and Etna. The Pbisotopic signatures further indicate that in a transition zone (Roccamonfina-Ernici) the High-K-series melts are derived from the ‘Roman’ mantle, but K-series melts from the ‘Campanian’ mantle. Variations in the metasomatic agent are inferred to be related to (former) subduction, and reflect along-strike isotopic changes between Adriatic and Ionian domains of the plate. In conjunction with trace-element signatures of the melt inclusions, source mixing scenarios and implications for the regional geodynamic controls will be discussed.

We have created a new 2D coupled geochemicalpetrological-thermomechanical numerical model of retreating intraoceanic subduction associated with volcanic arc development. The model includes spontaneous slab bending, subducted crust dehydration, aqueous fluid transport, mantle wedge melting and melt extraction resulting in crustal growth. This model allows us to study influence of melt extraction intensity on the dynamics of subduction, mantle wedge plumes development and magmatic arc growth and displacement. In our numerical experiments subduction nucleates across the weak transform fault separating two oceanic plates different age. Subduction rate strongly varies with time. In all studied cases there is a deceleration period of a few Myr after the beginning of subduction, during this period subduction rates decrease from ~ 7 cm/yr to ~ 4 cm/yr. Subsequently, two scenarios can be distinguished: (1) decay and, ultimately, the cessation of subduction, (2) increase in subduction rate (to up to ~12 cm/yr) and stabilization of subduction. In scenario 1 the magmatic arc crust includes large amounts of rocks formed by melting of subducted crust atop the thermally relaxing slab. In contrast, in case of stable subduction, magmatic rocks produced by partial melting of hydrated mantle wedge clearly dominate the crust. In several numerical experiments an intraarc extension is observed during subduction. This process results in splitting of previously formed magmatic arc crust by a newly formed spreading center. The loci of magmatic activity and intensity of crustal growth is strongly dependent on the dynamics of hydrous partially molten upwellings (cold plumes) rising from the slab.

Goldschmidt Conference Abstracts 2007

A720

2a and 4a Polytypes of (Ge, Si)-Wollastonite F. NISHI1, T. KURIBAYASHI2, Y. KUDOH2 AND R. MIYAWAKI3

Temporal change of a layer sequence in reaction zones in the system dolomite – quartz – H2O T. NISHIYAMA1, T. TANOUE1, A. TOMINAGA2 AND H. ISOBE1

1

Saitama Institute of Technology, Saitama, Japan ([email protected]) 2 Mineralogy, Petrology and Economic Geology, Faculty of Science, Tohoku University, Miyagi, Japan 3 Dept. of Geology and Paleontology, Shinjuku Branch of National Science Museum, Tokyo, Japan Three kinds of superstructures of wollastonite whose compositions were Ca(Ge0.65, Si0.35)O3, Ca(Ge0.50, Si0.50)O3 and Ca(Ge0.15, Si0.85)O3 were synthesized. In this abstract, they are called 65Wo, 50Wo and 15Wo. The lattice constants of 65Wo, 50Wo and 15Wo are: (angstrom, degree) a=2X8.038(4), 2X7.995(2), 4X7.949(5), b=7.451(1), 7.400(1), 7.352(1), c=7.194(2), 7.148(2), 7.093(1), alpha= 89.93(2), 90.05(2), 90.06(2), beta=94.85(2), 94.97(2), 95.11(1), gamma=103.34(2), 103.43(1), 103.39(1), respectively. 65Wo consists of the neighbouring two units of the basicwollastonite along the a-axis. The X-ray diffraction pattern showed a pseudo-C lattice. Therefore, the stacking sequence of 65Wo can be represented as AB, where A is the unit cell of the basic-wollastonite and B is the unit cell of the basicwollastonite with b/2 displacement. The final R-value was 12%. 50Wo consists of the neighbouring two units of the basicwollastonite along the a-axis, too. However, the X-ray diffraction pattern was different from 65Wo. Judging from the characteristics of the X-ray diffraction, the stacking sequence of 50Wo is represented as AA. The final R-value was 13%. 15Wo consists of the four units of the basic-wollastonite along the a-axis. There were four possible stacking sequences described as AAAA, ABAB, AABB and AAAB. The structure having the sequence AAAA will show the strong intensity on the h=4n diffraction. The structure ABAB will show the strong intensity on the h=2n diffraction. The structure AABB will show the extinction rule of the pseudo-C lattice. The last structure AAAB will not show any characteristic rules on the X-ray diffraction. As the result of the observation of the Xray diffraction of 15Wo, it was clear that the stacking sequence AAAA was most reasonable. The final R-value was 14%. In common with 65Wo, 50Wo and 15Wo, each tetrahedral site has the statistical distribution of Ge and Si atoms.

References Buerger, M.J. and Prewitt, C.T., (1961). Pro . Natl. Acad. Sci.USA, 47, 1884-1888. Ito,T., (1950). In X-ray Study on Polymorphism. Maruzen. Peacor D.R. and Prewitt, C.T. (1963). Amer.Miner. 48, 588596.

1

Dept. of Earth and Environment, Kumamoto University, ([email protected]; [email protected]) 2 Dept. of Earth and Environment, Kumamoto University

From an initial transient state to a steady state Hydrothermal experiments at 0.1GPa and 600o were carried out with two kinds of configrations for starting materials: one is dolomite single crystal + quartz powder + H2O and the other is quartz single crystal + dolomite powder + H2O. We got the following results. (i) Reactions were incomplete in all runs and the reaction zone developed only locally on the dolomite crystal. (ii) In short duration (45 – 71h) runs metastable layer sequences involving wollastonite and talc occur in the reaction zones, whereas they disappear in longer duration (168 – 336 h) runs. (iii) The layer sequence of the reaction zones in short duration runs differ from place to place on the dolomite crystal even in the same run. (iv) The diversity of layer sequences in the short duration runs merges into a unique layer sequence of Qtz / Di / Fo + Cal / Dol + Cal / Dol in the longer duration runs. (v) No reaction zone was observed on quartz crystals. These lines of evidence show that the system evolves from an initial transient state to a steady state and that the kinetic effect is important in the development of reaction zones. This part of the work has been recently published (Nishiyama et al., 2007).

Temporal change of a layer sequence at 800oC We carried out additional experiments in the same system at 0.1GPa and 800oC with run durations of 48, 109 and 357 h. In these higher temperature runs reactions were complete in all runs such that the dolomite crystal was completely surrounded by a reaction zone with a definite layer sequence. Observed conspicuous features are as follows. (i) The layer sequence Dol / Dol + Cal / Fo + Cal / Fo / Mon (Monticellite) / Di / Wo / Qtz is common in all runs. (ii) The thickest layer is Di in 48 h run, Mon in 109 h run and Fo (or Fo + Cal) in 357 run. (iii) A void layer formed between Dol + Cal and Fo + Cal layers in all runs, and it becomes thicker as the duration. (iv) Reaction zones have never formed on the quartz crystal, although the quartz shows resorption after the run. These results show little change of the layer sequence but a considerable temporal change in the growth mode of the reaction zone. A steady diffusion modelling was applied to the representative layer sequence to discuss this temporal change.

Reference Nishiyama, T., Tominaga, A. and Isobe, H. (2007) Island Arc,16, 16-27.

Goldschmidt Conference Abstracts 2007

Varying Ni in OIB olivines – Product of process not source Y. NIU1 AND M.J. O’HARA2 1

Department of Earth Sciences, Durham University, Durham DH1 3LE, UK ([email protected]) 2 Institute of Geography and Earth Sciences, University of Wales, Aberystwyth SY23 3DB, UK ([email protected]) The source of ocean island basalts (OIB) has been widely accepted as resulting from ancient recycled oceanic crust (ROC) [1] although there are many more difficulties than certainties in this model [2]. Sobolev et al. [3] concluded that Ni content in olivines of Hawaiian shield basalts is too high, precluding basalt origin by partial melting of mantle peridotite, asserting that this can only be explained by melting of an olivine-free pyroxenite. This pyroxenite originated by reacting mantle peridotite with a SiO2-rich melt derived from partial melting of recycled oceanic crust in the form of “SiO2oversaturated eclogite”. This is a revised version of the ROC model, but the complex behavior of Ni makes the interpretation non-unique. Sobolev et al. [4] show that while Ni content in olivines of basalts varies widely, it is conspicuously higher in basalts erupted on thick (> 70 km) lithosphere (THICK, including Hawaii) than on thin (< 70 km) lithosphere (THIN, including Iceland), and is still higher than in MORB. For primitive olivines with Fo > 89, NiTHICK (3417r452ppm, 1937; meanr1V n) > NiTHIN (2477r263ppm, 746) > NiMORB (2324r296ppm, 1700). This lithospheric thickness control (lid effect) poses the question why recycled oceanic crust prefers to exist and participate in magmatism beneath thickened lithosphere relative to beneath thin lithosphere and ocean ridges. Melting beneath thick (> 70km) lithosphere is largely in the garnet peridotite facies: aCpx + bGnt + cOl = 1.0Melt + dOpx, where olivine, the primary Ni host, contributes to the melt. Melting beneath thin lithosphere occurs mostly in the spinel peridotite facies: aCpx + bOpx + cSpl = 1.0Melt + dOl, where olivine crystallizes and demands Ni from the melt. A common peridotite source Ni = 1900 ppm, and ~ 10% melting, can yield ~ 400 ppm and > 560 ppm Ni in melts parental to MORB and those erupted on thick lithosphere respectively. With these results and realistic magma chamber process models [5], the observed NiTHICK, NiTHIN and NiMORB in olivines can be explained without invoking the revised ROC model which still has to address the difficulties noted in [2].

References [1] Hofmann, A.W., and White, W.M., (1982), Earth Planet. Sci. Lett., 57, 421-436. [2] Niu, Y., and O’Hara, M.J., (2003), J. Geophys. Res., 108, 2209, doi:10.1029/2002JB002048. [3] Sobolev A.V. et al., (2005), Nature, 434, 590-597. [4] Sobolev A.V. et al., (2007), Scienceexpress, 29 March 2007 (10.1126/science.1138113). [5] O’Hara, M.J., and Mathews, R.E., (1981), J. Geol. Soc. London, 138, 237-277.

A721

Ninetyeast Ridge, Indian Ocean: Constraining its origin and relation with the Kerguelen, Amsterdam and St. Paul hotspots INÊS G. NOBRE SILVA, DOMINIQUE WEIS, LISA SWINNARD AND JAMES S. SCOATES PCIGR, Dept. of Earth and Ocean Sciences, University of British Columbia, 6339 Stores Road, Vancouver, V6T 1Z4 BC, Canada ([email protected]) With a length of ~5000 km in the eastern Indian Ocean, the Ninetyeast Ridge (NER) is the longest linear feature on Earth, corresponding to ~44 Myr of evolution of the Indian Ocean basin and of the Kerguelen mantle plume. Despite being the subject of several geochemical studies during the late 1970’s to early 1990’s, debate regarding its mantle sources still persists. Since ODP Leg 121 in 1988, where the NER basement was last successfully drilled (~82 m on Site 756; ~52 m on Site 757; and ~178 m on Site 758) and subject of a thorough geochemical study, analytical techniques and sample processing have greatly improved. In this study, we present new high-precision isotopic compositions by MC-ICP-MS (Pb, Hf) and TIMS (Sr, Nd) obtained on samples recovered during ODP Leg 121. Our extended multi-isotopic analyses demonstrate that components comparable to the Kerguelen, Amsterdam and St. Paul mantle plumes contribute to the generation of NER basalts, but not Indian MORB mantle, which supports previous Hf and Pb isotopic studies [1]. In all isotopic spaces, the age-corrected NER isotopic compositions define distinct fields for each site that plot within a triangle defined by Kerguelen, Amsterdam and St. Paul islands. Site 756 (~43 Ma) overlaps with the isotopic compositional field of St. Paul Island basalts, reflecting a greater contribution from this component in its generation. In both Hf-Sr and Hf-Nd isotopic spaces, Sites 757 (~56 Ma) and 758 (~82 Ma) show isotopic compositions intermediate between those represented by Amsterdam Island and Kerguelen. Site 758 has Pb isotopic compositions that plot closer to the field proposed for the Kerguelen plume head [2] as defined by basalts from Kerguelen Plateau Site 1138 (~100 Ma), whereas Site 757 shows much more radiogenic Pb isotopic compositions, plotting between the proposed Kerguelen plume tail component [2] and Amsterdam Island basalts. Hf isotopic compositions further support the absence of a continental crust contribution for the origin and evolution of the NER.

References [1] Barry, T.L. et al. (2002), GCA, 66, spl.1, A53. [2] Ingle, S. et al. (2003), G3, 4, 1-28.

Goldschmidt Conference Abstracts 2007

A722

Dissolution, nucleation and growth of mineral phases in aqueous solutions C. NOGUERA1, B. FRITZ2 AND A. CLEMENT2 1

INSP, CNRS-UPMC, Paris, France ([email protected]) 2 CGS, CNRS-ULP, Strasbourg, France ([email protected]) In order to take into account the initial stages of formation of mineral nanoparticles in water-rock interaction models, we have developed a theoretical approach of nucleation, growth and ageing processes, relevant for precipitation in solution. It is based on the classical nucleation theory, on a sizedependent (algebraic) growth law allowing growth, resorption and ripening of particles simultaneously, and on conservation laws akin to a thermodynamically closed system. We will apply it to the formation of silicate phases, and particularly kaolinite-type minerals, from aqueous solutions resulting from rock alteration processes, through the geochemical code NANOKIN that we have elaborated. The nanoparticle formation is simulated versus time in terms of populations of particles appearing successively, surviving and growing or disappearing depending on the evolution of the resulting saturation state of the solution with respect to the corresponding mineral phase. We will show that two very different scenario may take place, depending upon whether an initial supersaturation is provoked in the solution (i.e. in an experimental approach), or whether it progressively appears as a consequence of rock dissolution. The time dependence of the supersaturation, the size evolution of the particles and the crystal size distribution functions present very specific features in these two cases. The simultaneous precipitation of competitive secondary phases has also been tested taking into account their specific solubilities, surface energies, geometric shapes and growth rates. This allows to predict the surviving phase in long-term processes Work is currently under progress to include chemical variability of the produced particles in the model. Support of the French ANR-PNANO 2006 (project “SIMINOX” 0039) is acknowledged in the last phase of this work.

References C. Noguera, B. Fritz, A. Clément and A. Baronnet, (2006) J. Cryst. Growth 297 (2006) 180 C. Noguera, B. Fritz, A. Clément and A. Baronnet, (2006) J. Cryst. Growth 297 (2006) 187 B. Fritz, A. Clément, Y. Amal, and C. Noguera, (2007) “NANOKIN: a geochemical computer model for dissolution, nucleation and growth in aqueous solutions” (in preparation)

Stabilisation of continental crust by dehydration melting: An example from the Västervik area, SE-Sweden NICOLE NOLTE, WIEBKE BAERO, BENT T. HANSEN AND ILKA C. KLEINHANNS GZG-Isotope Geology, Georg-August University of Göttingen, Goldschmidtstr. 3, 37077 Göttingen, ([email protected]) The Paleoproterozoic Västervik area (Baltic shield, SESweden) is located at one major terrane boundary on the Baltic Shield. To the south lies the Transscandinavian igneous belt and to the north follows the Svecofennian domain. In addition, the Västervik area is known for lp-hT amphibolite metamorphism in combination with intensive metasomatism. From field relationships at least three different granitoid generations can be identified that discriminate in their geochemical and isotope signatures as well. Primitive granitoides (QMD) of quartzmonzodioritic and granodioritic composition, metagranitoides (MG) of quartzmonzonitic (MGnf = non fluid) and alkaligranitic composition (MGf = fluid) as well as anatectic monzogranites (AG). QMD and MGnf seem to represent “normal” crust-building granitoides with geochemical signatures of rather typical continental crust. The above mentioned metasomatic fluids seem to have affected the MGf and AG as for example initial Sr isotope ratios are as high as 1.5, respectively. Such increased ratios are observed in the surrounding metasedimentary VästervikFormation as well indicating a genetic relation with the anatectically created AG. Further, geochemical signatures and fluid classifications show that both MG-groups probably share the same petrogenetic evolution with the sole difference of a fluidchemical imprint in the MGf granitoides. Based on the results, a geotectonic model for the evolution of the different granitoides was developed. We postulate an active subduction zone beneath the svecofennian continental margin. This subduction zone was blocked leading to the evolution of a new subduction zone further southwest building up an island arc of which rocks of the Oskarshamn-Jönköping belt belong to. Our geotectonic model relates the evolution of the QMD and MGnf to this maturing island arc. The accretion of this mature island arc on the continental margin of the Svecofennian terrane caused compression and folding leading to crustal thickening. Subsequently, intruding melts show higher SiO2 contents and are higher differentiated (MGf). Pressure increase due to crustal thickening and heat input by the ascending melts finally resulted in mica breakdown. Resulting dehydration melting is thus the trigger for the crustal anatexis event that produced the AG and led to fluids that percolated through some of the MG.

Goldschmidt Conference Abstracts 2007

A723

Radiation damage, internal textures and post-growth history of the Plešovice zircon standard

Geochemical meddling, microbing, Marsing, and mitigating mine drainage

N. NORBERG1, L. NASDALA1, J. KOSLER2, J. SLAMA2,3, J. GÖTZE4 AND N. GROSCHOPF5

D. KIRK NORDSTROM

1

Institut für Mineralogie und Kristallographie, Universität Wien, Austria ([email protected]; [email protected]) 2 Dept. of Earth Science, University of Bergen, Norway ([email protected]) 3 Dept. of Geochemistry, Charles University, Prague, Czech Republic ([email protected]) 4 Institut für Mineralogie, TU Bergakademie Freiberg, Germany ([email protected]) 5 Institut für Geowissenschaften, Johannes GutenbergUniversität Mainz, Germany ([email protected]) We have studied internal textures and the post-growth history of accessory zircon crystals from a granulite near Plešovice, southern Czech Republic. This zircon has recently been proposed as a new reference material for determination of U-Pb ages using the laser-ablation inductively coupled plasma mass-spectrometry (LA-ICP-MS) technique (Slama et al., 2006). Extraordinary properties of the Plešovice zircon include its uniform, concordant U-Pb age, relatively high U concentrations (>600 ppm), homogeneous Hf isotopic composition, and its availability in a large quantity. In spite of its isotopic homogeneity, the Plešovice zircon shows heterogeneous incorporation of trace elements. Typically there is strong primary zonation, including both oscillatory growth zoning and sector zoning. About 10% of the grains show sectors (corresponding to the growth of pyramid faces) that are particularly rich in actinides; here U content may be as high as 3000 ppm. Those sectors are easily recognized, e.g., from particularly high BSE intensities. The Plešovice zircon is in general moderately radiation-damaged whereas the high-U sectors are strongly metamict. All areas are roughly as radiation-damaged as it would correspond to U and Th concentrations and an assumed accumulation of radiation damage since the time of Variscan zircon growth (cf. Nasdala et al., 2001), which excludes any major thermal annealing event. Heterogeneous metamictization has resulted in zoned fracturing; fractures have been filled with secondary phases such as Fe hydroxides. First LA-ICP-MS U-Pb analyses of high-actinoide micro-areas turned out to be most difficult and unreliable. Consequently, these areas need to be avoided in using the Plešovice zircon as a U-Pb standard material.

References Nasdala L., Wenzel M., Vavra G., Irmer G., Wenzel T. and Kober B., (2001), Contrib. Mineral. Petrol. 141, 125144. Slama J., Kosler J., Schaltegger U., Tubrett M., Gutjahr M., (2006), Winter Conference on Plasma Spectrochemistry, Tucson, abstract WP05, 187188.

U.S. Geological Survey, 3215 Marine St., Boulder, CO 80303 USA Acid mine drainage (AMD) and acid pit lakes, for all their ugliness and environmental damage, provide opportunities for scientists and engineers to test their skills, push the frontiers of their research, and make a contribution to the remediation. Extreme environments such as the Rio Tinto mining district, Spain, have become favorite sites for both researchers and tourists because of the presumed analogy to the Martian surface. Rio Tinto and Iron Mountain, California, USA have also been a major focus of microbial research. The production of AMD is a complex hydrobiogeochemical process that is much better understood today because of sustained research in the fields of geology, hydrology, geochemistry, and microbiology. Mine-site remediation cannot proceed effectively without the combined expertise from these fields. Further, the most cost-effective remediation benefits from an independent technical oversight committee with such expertise. We now know that pyrite oxidation can create waters of negative pH, that mine plugging is often an ineffective remediation method, that microbial catalysis of pyrite oxidation involves a community structure including, but not limited to, Acidithiobacillus, Leptospirillum, Ferroplasma, and heterotrophs, that surface-water contaminants are attenuated and can be modeled, that dynamic rainstorm events can cause increases in concentration destroying aquatic life, that remediation scenarios can be evaluated before executing, and that the collaboration of scientists and engineers working together on remediation is far better than just engineers working alone. Research papers on sulfide mineral oxidation and mine-site characterization have proliferated over the last two decades. We also know much more about the processes (dissolution rate, solubility, sorption) controlling concentrations of Ca, Mg, Fe, Al, Mn, Cu, Zn, Pb, and Cd in surface and ground waters. A technical issue that still needs attention is the hydrogeological characterization of mined sites. However, our scientific understanding of acid mine drainage systems has advanced to the point where incorporation of applicable research into mine-site remediation as standard practice, communication of useful research to policy makers and project site managers, and communication of information that bears on risk assessment to project managers and the affected public, can make a substantial contribution to mine-waste management.

A724

Goldschmidt Conference Abstracts 2007

Differentiating fluid boiling from condensation

Quantifying accelerated surface denudation as a result of external forcing

DAVID I NORMAN1 AND JOSEPH MOORE2

K.P. NORTON1, F. VON BLANCKENBURG1, F. SCHLUNEGGGER2, M. SCHWAB2 AND P.W. KUBIK3

1

Dept of Earth and Env. Scienec, New Mexico Tech, Socorro, NM 87801 ([email protected]) 2 EGI, U of Utah, Saly Lake City, UT 84105 ([email protected]) Fluid inclusion petrography cannot differentiate boiling from condensation. Inclusions formed during boiling and condensation may have liquid- and vapour-dominant types that show approximately the same Th. During boiling a vapour phase develops in response to changes in liquid P or T. Condensation occurs when a liquid is subject to a vapour flux. This may occur at any depth in geothermal systems. At shallow depths this process generates steam-heated waters. At greater depths vapour from magmatic bodies can flux through circulating meteor waters. Modelling demonstrates that boiling and condensation changes the concentrations of gaseous species in different manors. Boiling strips dissolved gaseous species from liquids and low solubility species such as H2, CH4 and N2 more strongly partition into the vapour phase. Hence a ratio of soluble species like CO2 to N2 increases during boiling. Condensation increases liquid concentration of soluble gaseous species (see fig.). Gas ratios from the USGS Cerro Prieto data set show positive correlations between total gas and CO2/N2 ratios. Sequential fluid inclusion gas analyses shows negative correlations at Guanajuato where there is other evidence of boiling and positive correlations in geothermal systems were condensation is predicted. We know that gaseous components and liquids commonly decouple and recombine in geothermal systems. We find evidence in active geothermal systems and in fluid inclusions that both processes operate. Condensation may be an import process in the production and localization of high grade sulphide ore bodies by locally producing high fluid concentrations of H2S.

1

Institut für Mineralogie, Universität Hannover, Hannover, Germany ([email protected], [email protected]) 2 Institut für Geologie, Universität Bern, Bern, Switzerland ([email protected], [email protected]) 3 Paul Scherrer Institute, ETH Hönggerberg, Zürich, Switzerland ([email protected]) The evolution of mountain belts is a balance between forces that build topography and those that destroy topography. This study focuses on the destructive forces, weathering and mass wasting, fluvial or eolian erosion, glaciers, etc., that are inherently surfacial processes. In order to understand how orogens change through time, we need to identify and quantify the processes responsible for denudation. We apply a combination of novel and proven techniques to address the question of how these forces modify the landsurface. Morphometric analysis based on high resolution LIDAR data is used to identify transient landscapes. These geomorphic data are then combined with cosmogenic nuclides, 10Be from quartz in sediment specifically, in order to quantify process rates. In a case study in the Alpine Foreland, we identified two adjacent drainage basins which have responded differently to the same forcing event, a fall in stream base-level at 16,000±3,000 yrs BP. One of the watersheds has a coupled hillslope-channel system, while the other is decoupled. Denudation rates in the coupled system, 380±50 mm ky-1, are equal to those in the headwaters of the decoupled system, 380±50 mm ky-1. However denudation rates in the decoupled system increase downstream to 540±100 mm ky-1, with denudation in the incised region occurring at extremely high rates, ~1800 mm ky-1. The denudation rates show that within 16,000 yrs, only the channels themselves have responded, and spatially extensive erosion occurs only after a significantly longer lag time. The advantage of this approach is that the geomorphic response of a landscape to external forces can be identified by morphometric analysis, and the process rates that are measured with cosmogenic nuclides can be used to quantify landscape response times on the time scale of climate change or fault movement.

Goldschmidt Conference Abstracts 2007

A725

Sulfur in tree rings in Central Europe: A negative G34S shift relative to regional pollution sources

Quantification of stable strontium isotope variability in nature by MCICP-MS

M. NOVAK1, S. PEIFFER2, K. ZAK3, I JACKOVA1, F. BUZEK1, L. ERBANOVA1 AND E. PRECHOVA1

G.M. NOWELL1, B.L.A. CHARLIER2, D.G. PEARSON1, K. W. BURTON2

1

Czech Geological Survey, Geologicka 6, 152 00 Prague 5, Czech Republic ([email protected]) 2 University of Bayreuth, Bayreuth D-95440, Germany ([email protected]) 3 Institute of Geology, Czech Academy of Sciences, Rozvojova 136, Prague, Czech Republic ([email protected])

1

Sulfur isotope composition in tree rings of 50-year old Norway spruce was compared with G34S values of local coal and atmospheric deposition in the Black Triangle region of Central Europe. For this purpose we have analyzed soft coal from two East German Basins (Espenheim near Leipzig, and Lausitzer region near the German-Polish border) and two Bohemian basins (Sokolov and Most, northern Czech Republic). In all, 150 soft coal samples from all stratigraphic levels of these Tertiary deposits were analyzed for G34S. Sulfur isotope composition of air-borne SO2 was based on 170 monthly samples (Novak et al., 2001), G34S of rainfall sulfate was based on 200 monthly samples (Novak et al., 2000). The studied spruce was one of the few surviving trees in an area affected by pollution-related spruce die-back of the Erzgebirge Mts. The tree was harvested in 2005, while most other trees at the research site died back between 1975 and 1995 due to acidification. Bedrock orthogneiss was S-difficient, and its G34S was 5.8 ‰. The G34S values systematically decreased in the order Espenheim coal (mean of 12.0 ‰) > Sokolov coal (11.7 ‰) > Lausitzer coal (9.6 ‰) > rainfall sulfate (5.4 ‰) > air-borne SO2 (2.5 ‰) > Most coal (1.6 ‰) > Ezgebirge tree rings (0.4 ‰). Three of the coal basins had isotopically relatively heavy sulfur, one coal basin (Most, Czech Republic), had isotopically light sulfur. Most of the emissions from the coal-fired power plants were in the form of SO2, which was partly oxidized to sulfate in the atmosphere before reaching the receptor site. The resulting sulfur became isotopically somewhat heavier than the residual SO2. The spruce tree rings contained the isotopically lightest S of all studied S reservoirs. Our data indicate that (i) German coal S did not significantly influence trees on the Czech side of the Erzgebirge Mts., and (ii) assimilation of S by Norway spruce is associated with a negative isotope shift relative to ambient S.

There is growing evidence that high-temperature processes, such as diffusion, condensation and evaporation can generate significant stable-isotope fractionation. Recent observations of stable isotope fractionation in elements as heavy as Tl (Z=81) raises the possibility that there may also be natural Sr isotope variation. With conventional 87Sr/86Sr measurements, the 86Sr/88Sr ratio is used to for correct instrumental fractionation, and although the absolute value used for this correction is unimportant, implicit in its use is the assumption that that 86Sr/88Sr ratio remains constant in all analysed materials. This ‘internal’ correction for instrumental mass fractionation thereby masks the existence of any natural mass dependent fractionation of 86Sr/88Sr. In this study, we present a new technique that allows simultaneous measurement of both radiogenic Sr isotope variations (87Sr/86Sr) and possible small-scale mass dependent isotopic fractionation of the 86Sr/88Sr ratio. Radiogenic and stable Sr isotope ratios are measured on a Thermo-Finnigan Neptune MC-ICP-MS and are corrected for mass bias ‘externally’ using admixed Zr. We present 86Sr/88Sr data for NBS 987, a range of natural rock standards and extraterrestrial samples (for which 86Sr/88Sr fractionation has already been documented using double spike TIMS). Incomplete recovery of Sr during column separation can significantly fractionate Sr isotopes, which implies that excluding overlapping Rb and Sr tails during column procedures will introduce inaccuracies. We demonstrate that NBS987 is fractionated relative to high-T magmatic rocks (possibly resulting from the manufacturing process), and establish a precise “high-T” reference line. Due to documented variation in low-T and biological samples we suggest that this high-T reference should be adopted for Sr isotope work.

References Novak M., Jackova I. and Prechova E. (2001). Environ. Sci. Tech. 35. 255-260. Novak M. et al. (2000). Geochimica et Cosmochimica Acta. 64(3). 367-383.

Dept. Earth Sciences, University of Durham, Durham, DH1 3LE, UK ([email protected]; [email protected]) 2 Dept. Earth Sciences, The Open University, Milton Keynes, MK7 6AA, UK ([email protected]; [email protected])

A726

Goldschmidt Conference Abstracts 2007

3-D colloidal crystals of magnetite in the Tagish Lake carbonaceous chondrite

The nature of the Arabian lithospheric mantle beneath Aritain Volcano, NE Jordan

JUN NOZAWA1, KATSUO TSUKAMOTO1, HISAO SATOH2, KEN NAGASHIMA1 AND KENTA YAMADA1

TH. NTAFLOS1, G. KURAT1 AND S. SWOBODA2

1

Graduate School of Science, Tohoku University, Sendai, Japan ([email protected]) 2 Mitsubishi Materials Co. Tagish Lake carbonaceous chondrite (fallen on 2000) is known as a primitive meteorite in chemical composition but the meteorite heavily suffered aqueous alteration. In the matrix, magnetite (Fe3O4) particles were reported to be present in between tiny forsterite crystals [1]. However no detailed observation of morphology from these particles has been done, though the morphology of crystals possess the growth conditions precisely in which these crystals have been growing [2]. The chondrite with a few mm, in diameter, were cleaved, so that fine crystalline particles would expose on the fractured surface. These crystal surfaces were investigated employing field emission scanning electron microscopy (FE-SEM) and transmission electron microscopy (TEM). Magnetite crystals (Fd3m, cubic) usually exhibit an octahedron shape surrounded by the {111} faces. In this meteorite, however, magnetite fine particles, which are result of reaction of other silicate minerals with water or ice in the early stage of solar system, exhibit varieties of shapes, e.g. octahedron, rhombic dodecahedron with {110} face (Fig. 1), trisoctahedron with {211} face, 100-1000 nm in diameter (peak size, 200 nm). It is surprising to find that these fine magnetite particles form 3D colloidal crystals with varieties of structures, f.c.c. (Fig. 1), b.c.c, and h.c.p. In this colloidal system, not only the size but also the morphology of these particles determined the lattice structure of the colloidal crystal. It is very interesting to note that only the rhombic dodecahedron particles exhibit multiple twinning.

1

Dept. of Lithospheric Sciences, University of Vienna, Austria ([email protected]) 2 Dept. of Chemistry, VIRIS Laboratory, University of Natural Resources and Life Sciences, Vienna, Austria

Figure 1: FE-SEM image, magnetite colloidal crystals formed by multiple-twin particles. Twin-boundaries are indicated by white arrows.

The mantle xenolith-bearing Jabel Aritain volcano in NE Jordan belongs to the As Shamah Cenozoic lava field, which is the largest in the W part of the Arabian plate extending from S Syria through Jordan into NW Saudi Arabia. Most Jabel Aritain mantle xenoliths are spinel lherzolites but herzburgites, dunites, websterites and wehrlites are also present. Calculated equilibrium P,T conditions range from 900 to 1000 °C and 12-17 kbar, respectively, and are consistent with estimations made for peridotites from other localities in this area of the Arabian platform, indicating that xenoliths were extracted from shallow depths of 40 to 60 km (Medaris and Syada, 1998). Glasses from melt pockets present in a number of xenoliths have an unusually high mg# of 75 and secondary olivines with Fo94.5, which suggests that the metasomatic melt was probably picritic. LA-ICP-MS analyses of cpx show that xenoliths experienced cryptic metasomatism. Three groups of xenoliths have been recognized according to REE and other incompatible trace element patterrns in cpx: group 1 has depleted LREE abundances, group 2 is highly enriched in LREE (80 x chondritic) and group 3 has moderate LREE enrichments. The REE patterns in group 3 indicate that for some reason equilibrium conditions for trace elements has non been achieved as some cpx have higher LREE abundances than others within the same sample. In this group, the common Zr depletion relative to Nd and Sm implies that it is not associated to partial melting processes but rather to metasomatic processes and that the cpx was originally not depleted in Zr. Evidently, the metasomatic agent was a H2Orich fluid (high LREE and Sr) with carbonatitic components (depletion of Zr and Ti). The Jabel Aritain mantle xenoliths are in terms of PT equilibrium conditions similar to xenoliths from Jabel El Arab (ca. 100 km N of Aritain) and other Arabian plate peridotites suggesting a common thermal regime. The metasomatic fluids/melts affected variably the trace element but not the major element abundances. Apparently, the metasomatic processes were interrupted before equlibration could be achieved indicating a pre-eruption metasomatic front rising from lower levels of the local upper mantle.

References

Reference

[1] M.E. Zolensky (2002) Meteorit. Planet. Sci. 37, 737. [2] I. Sunagawa (1975) North-Holland, Amsterdam p.347

Medaris L.G and Syada G. (1998), Int. Geol. Review 40, 305324