Goldschmidt Conference Abstracts 2005

Goldschmidt Conference Abstracts 2005

Goldschmidt Conference Abstracts 2005 Interaction along Mineral Grain Boundaries A351 Hem-switching, chemically induced grain boundary migration, an...

3MB Sizes 0 Downloads 182 Views

Goldschmidt Conference Abstracts 2005 Interaction along Mineral Grain Boundaries


Hem-switching, chemically induced grain boundary migration, and rocks

A new thin film approach to study grain boundary transport in an incompatible matrix



Dept. EAPS, 54-718,Mass. Inst. Tech., Cambridge MA 02139, USA ([email protected])

Institut für Geologie, Mineralogie und Geophysik, RuhrUniversität Bochum, Universitätsstr. 150, 44780 Bochum ([email protected])

Based on observations of the microstructure of Bavarian granites, Voll (1960) described a process that he called Wechselsäume¸ i.e., hem-switching, in which mixing and unmixing could occur via intergrowth of two mineral phases. The geometry, microstructure and chemistry that he described are equivalent to those formed during chemically induced grain boundary migration, a process recognized nearly 20 years later during laboratory experiments in metals with (Yoon and Huppman 1979) and without partial melts (Hillert and Purdy, 1978) and, later still, in ceramics and minerals. In the laboratory, the process is favored under conditions where the distance of boundary migration is greater than that associated with lattice diffusion of the reactants. During solidsolution reactions, the kinetic driving force appears to be constrained by elastic coherency forces along the migrating interface. As boundaries migrate though the solvent phase, the solute concentrations incorporated in the zone are smaller than that of a homogeneous solid-solution. In order to understand the kinetics and driving forces for this process we have been studying the solid solution of divalent cations into calcium carbonates. In bicrystals, the coherency strain hypothesis explains many, but not all, of the aspects of mixing (Hay and Evans, 1992). More recently, we conducted experiments in which calcite and dolomite at 800°C react to form high magnesian calcite. After an early induction period, grains of both high Mg-calcites and Ca-rich dolomites are nucleated and grow slowly. This growth is then followed by rapid replacement of the early nucleation products by more stoichiometric dolomites. Thus, the overall dolomitization reaction occurs by at least three elementary reactions: nucleation of reactive intermediates, growth of the metastable phases, and replacement by dolomites with thermodynamically optimized stoichiometry. Thus far, CIGM has been produced in the laboratory only in the carbonate system, but it seems likely that such a process is possible during the formation of solid solutions in any mineral phase, and that CIGM be a wide-spread natural process.

References Hay, R. S., and B. Evans (1992) Acta metall. mater. 40, 2581. Hillert, M., and G.R. Purdy (1978) Acta metall., 26, 333. Voll, G. (1960) Geol. Jahrb. Geih., 42, 1-382. Yoon, D., and W. Huppman (1979) Acta metall., 27, 973.

Recently, Dohmen and Chakraborty [1] have developed a general kinetic model for an exchange reaction between two solids mediated by an “intergranular fluid phase”. One application of the model is that it allows to evaluate the integrated transport properties of an interganular fluid phase by modeling of the compositional profiles in the solid reactants. As a test of the model and as an alternative way to study grain boundary transport in e.g, metals, oxides or silicates, a new experimental setup was developed. The general idea is that single crystals (e.g. olivine) are deposited with inert polycrystalline thin films (e.g. ZrO2) of defined geometry and chemistry, such that any exchange with the covered surface of the single crystal has to pass this inert layer. Single crystals of San Carlos olivine (Fo90) were deposited with 100 to 500 nm thick layers of ZrO2 by Pulsed Laser deposition [2]. Two sets of diffusion experiments were performed with these samples: (1) Diffusion anneals in the temperature range 900 – 1100°C in which the samples have an additional olivine thin film with Fo30 composition on top of the ZrO2 layer. (2) Diffusion anneals at 1200°C, where olivine samples were packed into olivine powder with Fo30 composition. The analyses of the final chemical zoning with either Rutherford Backscattering or Electron Microprobe show that in the lower temperature range (<1100°C) the ZrO2-film is not any kinetic barrier for the exchange between the olivines. However, in the experiments at 1200°C (experimental set 2) we observe non-equilibrium phenomena partly related to the ZrO2 film and partly related to the insufficient contact with the surrounding powder. According to [1] this reaction system is controlled by solid state + fluid diffusion. Further experiments (e.g. a time series) at these conditions are going to be performed and the analytical observations as well as a quantitative modeling of the chemical zonings will be presented at the meeting.

References [1] Dohmen, R, and Chakraborty, S. (2003) Am. Min., 88, 1251-1270. [2] Dohmen, R., Becker, H.-W., Meißner E., Etzel T. and Chakraborty S. (2002), Eur. J. Mineral., 14, 1155-1168.

Goldschmidt Conference Abstracts 2005 Interaction along Mineral Grain Boundaries


Grain-boundary and intra-crystal dissolution-reprecipitation reactions in alkali feldspars I. PARSONS





Grant Institute of Earth Science, University of Edinburgh, UK ([email protected]) 2 Centre for Geosciences, University of Glasgow, UK Most alkali feldspars exhibit exsolution or replacement microtextures (perthite). Regular, µm-scale ‘strain-controlled’ perthites form by continuous processes involving only volume diffusion of Na+ and K+ (plus Ca2+ and coupled Al3+ in ternary feldspars) through an Al-Si-O framework which remains continuous (coherent). The different ionic radii of Na+ and K+ lead to elastic strain and exsolution lamellae adopt crystallographic orientations which minimize strain energy. Irregular, much coarser, discontinuous (incoherent) ‘deuteric perthites’ form by dissolution-reprecipitation ‘unzipping’ reactions driven by release of coherency strain. These affect entire >1 cm crystals without modification of crystal shapes. In the Klokken intrusion layers of impermeable finegrained syenite (feldspar bulk composition ~Ab60Or40) are interleaved with layers of compositionally similar, permeable coarse grained syenite. Sub-µm cryptoperthites in the former have a ‘braid’ configuration which is modified by forces acting between adjacent crystals, so that the textures form a ‘strain-map’. The textures coarsen to ~20µm near crystal boundaries leading to ‘pleated rims’ in which Na- and K-rich volumes alternate on opposing crystal surfaces. The local change in bulk composition leads to change in the lamellar texture to parallel films with edge dislocations (semicoherent). The dislocations allow ingress of water into crystals and local development of patch perthites. Patch perthites (up to 250 µm) dominate in the coarse syenites although relics of braid microtexture persist. TEM shows that individual patches are mosaics of incoherent subgrains which CL shows have well developed, 1-10s of µm oscillatory zoning at blue-UV wavelengths. Adjacent albite or microcline subgrains may display the same zoning pattern, with individual zones traceable over 100s of µm, implying dissolution in laterally extensive fluid films. Laser ICPMS shows that Ga, Rb, Sr, Ba, La, Eu, Pb and Cs partitioned isochemically in bulk from braid into Ab- and Or-rich patches, whereas Fe, Ti, Ni, La, Ce, Pr and Nd were lost from crystals during unzipping. After unzipping, Or-rich patches underwent a further phase of strain-controlled exsolution leading to straight lamellar cryptoperthites with dislocations. Solvus relationships show that this occurred at ≤~350 °C, implying that the deuteric fluids had ceased to play a role below this T.

Importance of Ar, He transport and partitioning in grain boundaries ETHAN F. BAXTER Department of Earth Sciences, Boston University, 685 Comonwealth Ave. Boston, MA 02215 ([email protected]) Analysis of noble gases, most notably Ar and He, in rocks and minerals are used in a variety of geological applications including geo- and thermochronology, tracing crustal recyling, and reconstructing Earth’s degassing history. Interpretations of such data are effected by the manner in which the gases are partitioned between system phases (or reservoirs) and therefore reflect the noble gas content of the environment from which the samples were derived. Partitioning also effects the net rate by which the gases are transported in a given system. Mineral-mineral grain boundaries represent one reservoir common to all “dry” rocks into which noble gases may be partitionined and stored, and through which noble gases may be efficiently transported. An experimental method has been developed for the measurement of grain boundary partition coefficients for noble gases. 37Ar and 4He are introduced into solid diopside composition glass samples via neutron irradiation. Samples are crystallized in sub-solidus conditions in a piston cylinder at 1350-1550 C and 2-3 GPa. Noble gases simultaneously equilibrate between the evolving crystal and grain boundary reservoirs. After equilibration, G.B. Ar and He is differentiated from that within the crystals by means of bulk step heating analysis. Results suggest an expected trend of decreasing bulk G.B. noble gas content with increasing grain size and yield a value of effective grain boundary surface partitioning, Ksurf, in units of (mol Ar/m3 of solid)/(mol Ar/m2 of GB) of 6.8 x 103 – 2.4 x 104 m-1. Values for He are less well constrained but lie in a similar range. These data suggest that grain boundaries constitute a significant, but not infinite, reservoir, and therefore bulk transport pathway, for noble gases in nominally dry systems (i.e. free of a fully wetting fluid phase). For example, this parameter may be used in a physical model of the local development of excess Ar (or He) (Baxter 2003) to help predict the amount of excess Ar (or He) that would be sequestered in the phases of a nominally dry system such as many ultra-high-pressure crustal settings or portions of the mantle. Also, partitioning and transport of noble gases into melts will be influenced initially by mineral-mineral grain boundaries where up to 30% of the bulk rock noble gas could potentially be stored before incipient melting.

Reference Baxter E.F., (2003) EPSL 216, 619-634.

Goldschmidt Conference Abstracts 2005 Interaction along Mineral Grain Boundaries


On the process of dihedral angle change

Quartz-H2O dihedral angles and crystal misorientation



Dept. Earth Sciences, University of Cambridge, Cambridge, CB2 3EQ, UK. ([email protected]) Interstitial phases in igneous rocks commonly pseudomorph the residual porosity, with shapes inherited from melt-filled pores. The inherited angle at pore junctions may reflect either an impingement texture (e.g. Elliot et al., 1997) or melt-present textural equilibrium. Since these are lower (median < 60˚) than solid-state equilibrium values (~ 120˚), such pseudomorphs are out of textural equilibrium. The solidstate texture will thus move towards one of larger dihedral angle. Interstitial clinopyroxene in cumulates from the Rum Layered Series shows variable approach to solid-state textural equilibrium from an initial inherited state. In contrast to the currently accepted model of dihedral angle change, which assumes an instantaneous establishment of the new angle at the pore corner with subsequent outwards propagation of the new surface curvature (Mullins, 1957), textural equilibration at pore corners actually occurs as a continuous process, with a gradual movement of the entire dihedral angle population towards the equilibrium final state At any instant during this process, the static dihedral angle equation (Herring, 1951) does not hold.

References Elliot, M., Cheadle, M. & Jerram, D. (1997) Geology, 25, 355358. Herring, C. (1951) Physics of Powder Metallurgy, pp. 143179. (Kingston, W.E., ed.) McGraw-Hill. Mullins, W.W. (1957) J. App. Physics, 28, 333-339.


Dept. of Earth and Environmental Sciences, Rensselaer Polytechnic Institute, Troy, NY, USA ([email protected]) 2 Dept. of Geology and Geophysics, Yale University, New Haven, CT, USA The physical and chemical properties of deep-seated rocks are strongly influenced by the presence and intergranular geometry of fluids. At equilibrium, the latter is strongly determined by solid-solid interfacial energies. In order to better understand the role of interfacial energies in determining fluid topology, we developed a new experimental technique for characterizing the 3-D geometry of individual pores. The technique involves introduction of H2Oalong synthetic grain boundaries produced in a pistoncylinder apparatus by juxtaposing polished quartz disks at high P and T. H2O trapped along the interface forms lenticular pores during the experiments as the grains weld together to form a grain boundary. After the experiment, grain boundaries were separated to expose half of each fluid-filled pore on complementary disks. Two advantages of this technique are that the: (1) 3-D pore geometry can be directly measured using atomic force microscopy thus removing effects caused by random sectioning of pores, and (2) solidsolid interfacial energy can be controlled by varying the amount of crystal lattice misorientation (determined by electron backscatter diffraction) between adjacent disks, and thus complementary dihedral angles can be measured as a function of variable lattice coincidence. Ten synthetic quartz grain boundaries were exposed to pure H2O over a range of induced crystal misorientations ranging from ~1° to 120° by rotations about the c-axis. It was expected that low-energy grain boundary configurations (e.g. 0° misorientation) would yield high dihedral angles. It was impossible to separate grain boundaries with misorientations <5° because the adjacent disks of quartz grew together to form a single crystal. Surfaces of quartz disks that were misorientated by values >5° yielded a narrow, normal distribution of dihedral angles ranging from 46° to 68° with a median value of 56°±4 (n=246). In natural quartzose rocks it is expected that there exists a large range of misorientations across grain boundaries. However, these results indicate that misorientation values >5° do not contribute to variable grain boundary energy or the spread of observed dihedral angles.


Goldschmidt Conference Abstracts 2005 Interaction along Mineral Grain Boundaries

Oxygen diffusion "fast-paths" in titanite single crystals

Effects on a basaltic surface of an impact-derived hot fluid bed (Kirbet-el-Umbachi, Syria)

X.Y. ZHANG, E.B. WATSON AND D.J. CHERNIAK Department of Earth and Environmental Sciences, Rensselaer Polytechnic Institute, Troy, NY 12180 ([email protected], [email protected], [email protected]) Natural titanites are far from perfect ideal crystals. In addition to point defects, which affect oxygen lattice diffusion, there might be line- or planar defects along which fast diffusion could occur. While experimentally measuring oxygen lattice diffusion in titanites, we found that almost all of the oxygen diffusion profiles for natural titanites departed from the complementary error function solution expected for simple lattice diffusion, instead having a “tail” reaching deeper into the samples. For both dry and hydrothermal experiments, 18O was used as the diffusant. In dry experiments, the source material was 18O-enriched SiO2 powder, while 18O-enriched water was used for the hydrothermal experiments. Diffusive uptake profiles of 18O were measured in all cases by nuclear reaction analysis (NRA) using the 18O (p,α)15N reaction. The diffusion “tails” can be explained by either parallel planar defect or one-dimensional “pipe” models. In our experiments, different sizes of “tails” (with varying 18O concentrations) were observed. Under the same temperature and pressure conditions, the sizes of tails were affected by two factors: the diffusion duration and the defect density. For the same experiment duration, the higher the defect density, the larger the “tail”; for the same defect densities, the longer the diffusion duration, the larger the “tail”. The oxygen diffusion rates in the fast-paths were obtained by traditional graphical analysis methods, using the WhippleLe Claire equation (for 2-D defects) assuming that the width of the fast path is 1nm. Two Arrhenius relations were obtained for the fast-path diffusion, one for experiments under dry conditions, the other for hydrothermal conditions: Ddry = 4.03 × 10-2 (m2/sec)exp( -313± 22)(kJ/mol)/RT Dwet= 3.48 × 10-7 (m2/sec)exp( -219± 39)(kJ/mol)/RT AFM imaging of HF etched titanite surfaces suggest that the diffusion fast-paths might be either parallel planar defects or parallel pipe defects. In addition to the lattice diffusivity, the presence and 3-D distribution of any fast paths⎯and the diffusivity in these paths⎯is important to the bulk closure properties of single crystals. For titanites, AFM imaging showed that the fastpaths may not be interconnected at a length-scale comparable with the crystal dimension, so they may not have a dramatic effect on bulk closure properties.


UMR 5198, CNRS-IPH, CERP. Av. Léon-Jean Grégory, 66720 Tautavel, France ([email protected]) 2 ENSCP. 11 rue Pierre et Marie Curie, 75231 Paris, France ([email protected]) 3 LEPMI-ENSEEG, Dom. Univ., BP75, 38042 Saint-Martin d’Hères, France ([email protected]) 4 MHN, Laboratoire de Minéralogie, 61 rue Buffon, 75005 Paris, France ([email protected]) Evidence for widespread dispersion of impact ejecta micro-debris at ca. 4-kyr-BP was recently investigated from archives in lands and seas. Regional diversity of surface effects linked to the ejecta fallout that range from moderate heating to localized melting raised questions about the nature and composition of the impact cloud when reaching the soil surface. The most extensive melting caused by the ejecta fallout has been reported as a unique bone-rich basaltic breccia over ten square metres at Kirbet-el-Umbachi (Syrian desert). The aims of the present study are to elucidate composition of the ejecta, its interaction with the host materials, and its significance in terms of impact-related processes. This is achieved by in situ analytical characterization based on high resolution SEM, WDS microprobe, and Raman spectrometry. The basaltic breccia displays a complex imbrication of anomalous petrographic facies that are distinctive from the local basalts. The unique suite of carbonaceous polymorphs (nano-sized diamonds, graphitic carbon, PAH species and amorphous carbon) in the recrystallised materials trace the carbonaceous component of the 4-kyr BP impact ejecta. Heterogeneous tear-dropped clasts that are embedded in the recrystallised basalts consist of Ca-rich silicate glass with heterogeneities indicating silicate-carbonate immisci-bility, diaplectic quartz, barium sulphate, and diverse re-crystallized phases (silico-phosphate, Ca-phosphates, silicates). They trace solidified debris from the ejecta melt that derived from partial melting of carbonates, silicates, phosphates and soluble salts. Occurrence of a cm-thick crust with abundant domains of heated to melted bone fragments and a flow acicular facies is explained to result from high temperature (1200-1400°C) interaction between CO2-rich hot intrusive fluids and materials present at the surface: animals, basalts, and calcareous soils. Heating and selective melting over a few square meters would trace the exceptional pulverisation of a large mass of hot liquid melt with solid debris, in contrast to the widespread spray of hot fine debris in other regions.

Goldschmidt Conference Abstracts 2005 Interaction along Mineral Grain Boundaries

Dissolution of oil well cement in presence of CO2/H2S under HTHP J. CENTENO1, A. RAMIREZ1, A. COLINA2 2 AND A. BLANCO 1

Instituto de Ciencias de la Tierra, UCV, Venezuela. ([email protected]) 2 PDVSA –INTEVEP, Los Teques, Venezuela ([email protected]; [email protected])

Introducction The cement material used in the construction of deep oil well reservoirs is frequently exposed to high temperatures and pressures (HTHP) and to the corrosive action of gases like CO2 and H2S, presents in the media, that can alter its properties. According to the criterion of Nickel (1995), the cementing material can be considered as a mineral mixture. In this study the interactions between the cement material and CO2/H2S are investigated to establish a method for optimal selection of cylindrical cement samples of two different dimensions (1” x 2” and 1/2” x 1”).

Methods Samples were exposed to the action of a mixture of CO2/H2S, dissolved in water, in a Parr reactor for HTHP (T= 160 °C; PCO2= 37.77 atm; PH2S= 16.33 atm; PT= 102.07 atm) to compare their dissolution rates on laboratory scale and chose the most suitable geometry. The reaction times were 1, 3 and 7 days. The test solution was analyzed by Inductive Coupled Plasma -ICP- (elements) and Ionic Chromatography (anions). Dissolution rates were calculated and the Na/Ca and K/Ca relations were compared.

Results and Conclusions After performing laboratory tests, it was observed severed damage (leached) in the samples exposed to the mixture of CO2/H2S within different time scales. The results obtained for Na, K, Ca, suggested that ½” x 1” samples show important benefits in terms of time laboratory constraints compared with the 1” x 2” dimensions. This statement is strongly supported by a chemical study. These results will be used in further studies of mechanical properties and mineralogical composition of cement material in aggressive conditions.

Reference Nickel, E. H. (1995) The Definition of a Mineral. The Canadian Mineralogist. Vol. 33, pp. 689-690.


Geochemical feature of chlorites in No. 201 and No. 361 uranium deposit, South China RENMIN HUA, ZHANSHI ZHANG AND JUNFENG JI State Key Laboratory for Mineral Deposit Research, Department of Earth Sciences, Nanjing University, Nanjing 210093, China. ([email protected]) The No.201 and No.361 uranium deposits are genetically related with the Zhuguang granite pluton. Chloritization was one of major hydrothermal alterations and some chlorites were very closely associated with uranium mineralization. Chemical compositions of the chlorites analyzed from EPMA are: 28.81~22.20%, Al2O3 = 22.38~15.79%, SiO2 = FeO = 39.74~26.18%, and MgO = 15.45~3.47%, attributed mostly as prochlorits or ferromagnesian chlorite. They are Ferich species and formed under a relatively reductive environment. However, the wider range of FeO and MgO concentrations indicated that some of the chlorites might formed from an Fe- and Mg-enriched fluid. In the n(Mg)/n(Fe+Mg) vs. n(Al)/n(Al+Mg+Fe) diagram (Laird, 1988), most chlorites have a source of argillaceous rocks, which was the source of the granites in that area. But less chlorites seem to have sources similar to mafic rocks or related fluids. The forming temperatures of chlorites were calculated from the T-d001 equation of Battaglia (1999). Chlorites in granite formed at 276~220ºC, while those related to later diabase dyke 241~169ºC. It is postulated that there are two forming mechanisms of chlorites in the studied area. One was resolving-precipitating, and the other was resolving-migrating-precipitating. The later was represented by the chlorite veinlets occurred in the fissures. The close spatial relation between these chlorite veinlets and uranium ore revealed that the chloritization in these two deposits not only reactivated and transported uranium from granite, but can also adsorb and precipitate the uranium to form ore-bodies. This study is financially supported by the National Natural Science Foundation of China (Grant No. 40132010) and the State Key Fundamental Research Project (Grant No. 1999CB043209).

References Battaglia., (1999). Clay and Clay Minerals, 47(1): 54-63 Laird J., (1988). In: Bailey (ed), Reviews in Mineralogy, V19: 405-453

Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes


Sorption and surface precipitation as controls on the reactivity and bioavailability of sorbates and sorbents JANET G. HERING California Institute of Technology, Environmental Science & Engineering, 1200 E. California Blvd. (138-78), Pasadena, CA 91125, USA ([email protected]) The partitioning of chemical species between solid and dissolved phases determines their mobility in aquatic systems and can strongly influence their reactivity and bioavailability. The interfacial processes of sorption and surface precipitation can affect the reactivity not only of the surface species but also of the underlying solid.

Constraints on the structure of surface species Insights into the structure of surface species are gained through macroscopic sorption experiments, spectroscopic interrogation, and modeling. These methods provide information that is often complementary and it is critical to examine the level of constraints imposed by each type of information. For example, the protonation level of surface complexes is often inferred from the pH dependence of sorption, but this can be shown to be only a weak constraint. Spectroscopic techniques provide evidence for binuclear surface complexes, but formation of such complexes can be inconsistent with sorption densities observed in macroscopic studies. Further, it is important to consider how the structure of surface species may influence the reactivity of the underlying solid.

Surface reactivity The effect of surface speciation on the reactivity of the underlying solid is examined in the case of the oxidative dissolution of chromium(III) hydroxide. This reaction exhibits inhibition by the product chromium(VI) at pH 2 but not at pH 9. This pH dependence suggests that the inhibition is due to interaction of chromium(VI) with the surface of the dissolving solid. The reactivity of both sorbates and sorbents toward microbial reduction will be examined for arsenic(V) sorbed on iron(III) oxyhydroxides. In both field and laboratory studies, reduction of arsenic(V) to arsenic(III) can be observed without any detectable release of arsenic from the iron(III) oxyhydroxide surface into solution. Arsenic(V) reduction is observed to precede reduction of iron(III). Steric, kinetic, and equilibrium constraints to microbial transformations of surface species are considered.

Bacterial adsorption controls on mineral solubility K.J. JOHNSON AND J.B. FEIN University of Notre Dame, Civil Engineering and Geological Sciences, 156 Fitzpatrick, Notre Dame, IN 46556, USA ([email protected], [email protected]) Bacterial adsorption reactions can dominate the speciation of metal cations under a range of conditions of geologic interest, and hence bacterial adsorption can strongly influence the saturation state of water-rock systems. Although bacterial effects on the rate of mineral dissolution are well-established, the effects of bacteria on mineral solubilities, or the extent of dissolution, have not been studied. In this study, we measure the effect of the common soil bacteria, Bacillus subtilis, on the solubility of the mineral cerrusite (PbCO3). We compare solubility estimates from surface complexation modeling to the observed experimental solubilities, providing a rigorous test of the ability of thermodynamic modeling to account for the observed solubility enhancement. B. subtilis cells were suspended in 0.1 M NaClO4, and powdered PbCO3 was sealed in dialysis tubing prior to being placed in the bacterial suspension. This approach enabled chemical contact between all components during the experiment, but allowed separation of the mineral from the bacteria after the experiment so that the distribution of released Pb could be determined. Surface complexation models were also developed based on previously determined bacterial site concentrations and stability constants in order to independently estimate the expected effect of bacterial adsorption of Pb on the mineral solubility. We observed significantly enhanced cerrusite solubility in the bacteria-bearing systems, with Pb-bacterial surface complexes dominating the Pb speciation in the biotic systems. Results from this study suggest that adsorption of mineralforming cations onto bacterial cell walls can lower the mineral saturation state, causing an increase in mineral dissolution. Surface complexation models can be used successfully to quantitatively estimate the effect of bacterial adsorption on mineral solubilities, thereby providing a better understanding of the role that bacteria play in weathering and porosity evolution in the subsurface.

Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes


Effect of sorbed arsenic species on bacterial reduction of HFO

A general rate law for bacterial Fe(III) oxide reduction




California Institute of Technology ([email protected]) 2 University of Califonia, Santa Cruz The geochemistry that controls the partitioning of arsenic species between solid and aqueous phases in sedimentporewater systems can be strongly influenced by bacterial processes. Particulary, bacteria can control redox cycling of both arsenic and soil minerals, which can in turn strongly affect the ultimate mobility of arsenic. The release of arsenic into the aqueous phase has created a serious health problem in several countries thoughout the world. One proposed mechanism for the mobilization of arsenic from the surface of an iron (oxy)hydroxide mineral is through the reductive dissolution of the iron mineral, which has been linked to bacterial respiration. However, arsenic mobility also depends on its redox state (mainly inorganic As(III) or As(V)), which can also be microbially controlled. Bacterial respiration is potentially an important route for the reduction of As(V) to As(III) in the environment [1,2]. This study investigates the rates of reductive dissolution of hydrous ferric oxide (HFO) by Shewanella sp. ANA-3 in laboratory inclubations with either As(V), As(III), or a mixture of As(V) and P(V). Iron and arsenic speciation were measured in both the solid and dissolved phases. In addition, ambient bacteria from Haiwee Reservoir, a field site known to have a population capable of iron reduction and arsenate respiration [3,4], were inoculated into a similar As(V)/HFO slurry experiment. In both Shewanella ANA-3 and Haiwee incubations, As(V) was initially reduced before HFO. A fraction of the total HFO was also reduced over the course of the experiment. The relative rates of reductive dissolution will be presented.

References [1] Oremland R.S., and Stolz J.F., (2003) Science 300, 939943 [2] Newman D.K., Ahman, D., and Morel F.M.M., (1998) Geomicrobiology 15, 255-268. [3] Kneebone, P.E., O’Day P.A., Jones N., and Hering J.G. (2002) Environ. Sci. Technol. 36, 381-386. [4] Malasarn D, Saltikov C.W., Campbell K.M., Santini J.M. Hering J.G., and Newman D.K., (2004) Science 306, 455.

Department of Geology and Geophysics, University of Wisconsin, 1215 W. Dayton St., Madison, WI 53706, USA ([email protected]) Recent experimental studies of synthetic and natural Fe(III) oxide reduction permit development of a general reaction-based rate law for bacterial Fe(III) oxide reduction at circumneutral pH. The findings collectively support a rate law that differs fundamentally from those applied to abiotic reductive dissolution as a result of two basic phenomena: (1) the relatively minor influence of oxide mineralogical and thermodynamic properties on surface area-normalized rates of enzymatic reduction compared to abiotic reductive dissolution; and (2) the major limitation which sorption and/or surface precipitation of biogenic Fe(II) on residual oxide and Fe(III)-reducing bacterial (FeRB) cell surfaces poses to enzymatic electron transfer in the presence of excess electron donor. Parallel studies with two major FeRB genera (Shewanella and Geobacter) lead to common conclusions regarding the importance of these phenomena in regulating the rate and long-term extent of Fe(III) oxide reduction. Models in which rates of enzymatic reduction are limited kinetically by the abundance of “available” oxide surface sites (as controlled by oxide surface area and the abundance of surfacebound Fe(II)) and by FeRB cell density provide an adequate macroscopic description of controls on the initial rate and long-term extent of oxide reduction. Although thermodynamic limitations posed by accumulation of aqueous reaction end-products (e.g. Fe(II) and alkalinity) cannot generally account for the slow-down and cessation of Fe(III) oxide reduction over time in batch reaction systems, a thermodynamic term (e.g. a TST function) or some other kind of switch must nevertheless be incorporated into the general rate law in order accurately simulate long-term patterns of reduction.


Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

Direct electrochemistry of cytochrome c on oxide electrodes C.M. EGGLESTON, N. KHARE, D. LOVELACE Department of Geology and Geophysics, University of Wyoming, Laramie, Wyoming, USA ([email protected], [email protected], [email protected]) Redox metalloproteins involved in biochemical electron transfer can interact with mineral surfaces. Cytochromes, and in particular cytochrome c (mitochondrial), have been used in the construction of biosensors based on semiconducting iron oxide; photovoltaic devices have been built based on cytochrome c sensitized titanium oxide nanoparticles, and cytochromes have been used as environmental remediation catalysts for reductive dehalogenation of contaminants (e.g., TCE). It has been proposed that c-type cytochromes anchored to the outer membrane of dissimilatory iron-reducing bacteria effect electron transfer to solid ferric minerals. Electron transfer between an oxide and the cytochrome is important in all of these settings. A key properties of cytochromes is “redox-linked conformation change” in which conformational changes of the protein affect redox potentials. The interaction of cytochromes with surfaces can induce conformational changes that trigger electron transfer. Here, we use cytochrome c a model cytochrome in an investigation of the effects of conformation change on reduction potential in the presence of oxide surfaces. In this presentation, direct electrochemistry of cytochrome c using iron oxide electrodes is demonstrated. The results are compared and contrasted to those for Indium-doped tin oxide electrodes, which have been used previously. Cyclic voltammetry shows that the reduction potential of cytochrome c on hematite electrdoes is similar to that of the native protein, suggesting that there is little or no denaturation of the protein at the electrode surface. This is in agreement with results for other hydrophilic electrodes, and in contrast with results for certain hydrophobic and metal electrodes. We therefore used tin oxide electrodes in subsequent studies of denaturation. Following biochemical studies of guanidine-induced cytochrome c denaturation, we observed the direct electrochemistry of cytochrome c on tin oxide electrodes in the presence of varying concentrations of guanidine. The “native” reduction potential disappeared with increasing guanidine concentration, and two new reduction waves at more negative potential appeared (with higher current than the native reduction peak). These two peaks can be attributed to a “molten globule” intermediate unfolding state at intermdiate reduction potential, and to a fully unfolded state at the most reducing potentials.

Transformation of hematite into magnetite – How do bacteria contribute? T. BEHRENDS AND P. VAN CAPPELLEN Department of Earth Sciences, Utrecht University, 3508 TA Utrecht, The Netherlands ([email protected]) Formation of secondary iron minerals frequently accompanies dissimilatory iron reduction. In particular, the formation of magnetite has received much attention as it affects the magnetic characteristics of the medium and because of the high capacity of biogenic magnetite to reduce environmental contaminants. However, the precise role of bacteria in magnetite formation during iron reduction is still enigmatic. The abiotic transformation of highly reactive iron oxyhydroxides, such as ferrihydrite and lepidocrocite, into magnetite under environmentally relevant conditions is well documented. For iron oxyhydroxides of low reactivity, such as goethite and hematite, no such transformation has so far been reported. In incubations with the bacteria Shewanella putrefaciens we observed the formation of magnetite as a product of bacterial reduction of nanoparticulate hematite. This model system thus appears to be a very suitable to unravel the role of bacteria in the formation of magnetite. We performed incubation experiments to constrain the optimal conditions for magnetite formation. Magnetite formation was enhanced by relatively high dissolved carbonate concentrations, while relatively low concentrations of phosphate or arsenate inhibited magnetite formation. Experiments were also run in which a fraction of the hematite suspension was separated from the rest of the suspension in a dialysis bag. Bacteria were present outside, but not inside the bag. Surprisingly, significant magnetite formation was observed inside the dialysis bag, but not outside the bag. These results suggest that transformation of hematite into magnetite is an abiotic process, with the role of bacteria restricted to the supplying of reduced iron. However, all attempts to induce magnetite formation from hematite in abiotic controls have not been successful so far. In the presentation, we will discuss possible mechanisms, which may explain how bacteria regulate the transformation of hematite to magnetite in the experimental systems. Furthermore, we will discuss the environmental significance of our observations.

Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

Fe(II) adsorption at the oxide-water interface: From macrosopic observations to spectroscopic measurements M.M. SCHERER, P. LARESE-CASANOVA AND A.G.B. WILLIAMS Civil and Environmental Engineering, The University of Iowa, Iowa City, IA, USA ([email protected]) The remarkable reactivity of Fe oxides, and the discovery of microbial Fe respiration has led to intense speculation about the surface species that form when Fe(II) ions adsorb on mineral surfaces. Spectroscopic measurements of Fe(II) adsorbed at the Fe-mineral water interface, however, have been difficult to obtain because of the very small quantity of Fe at the interface and interference from the bulk structural Fe(III). We are using the isotope specificity of 57Fe Mössbauer spectroscopy to characterize Fe(II) adsorbed on commonly occurring Fe minerals, such as hematite (α-Fe2O3), as well non-Fe containing oxides, such as aluminum oxide (Al2O3). Using this technique, we have previously observed that electron transfer occurs between adsorbed Fe(II) and structural Fe(III) in ferrihydrite [1]. These experiments, however, were somewhat limited in that only a narrow range of geochemical conditions was explored (pH 7.4, KBr electrolyte, and low Fe(II) concentrations). Here, we expand on these experiments to spectroscopically characterize adsorbed Fe(II) over an environmentally relevant range of Fe(II) concentrations and pH values (i.e., adsorption isotherms and pH edges). Initial results suggest that electron transfer between adsorbed Fe(II) and structural Fe(III) in hematite occurs in the presence of high aqueous Fe(II) concentrations (up to 1 mM) and over a pH range of 6.8 to 8.2. At a pH value of 8.2, however, an Fe(II) doublet is observed in the room temperature Mossbauer spectra, suggesting that at least some portion of the Fe(II) is not oxidized. The spectral parameters of the Fe(II) doublet are not consistent with a ferrous hydroxide precipitate and may represent an adsorbed Fe(II) species.

Reference Williams A. G. B. and Scherer M. M. (2004). Environ. Sci. & Technol 38(18), 4782-4790.


Does surface acidity of ferric (oxy)hydroxides affect reactivity towards H2S? S. PEIFFER AND W. GADE Chair of Hydrology, University of Bayreuth, Germany ([email protected])

Material and Methods The reactivity of hematite (Hem), goethite (Gt), 2-line (2LFh) and 6-line ferrihydrite (6LFh) with respect to H2S has been studied at steady-state conditions in a pH range between 2 and 5 using a fluidized-bed reactor which is supplied with a constant flow of electrochemically generated H2S.

Results and Discussion

Exept for Hem, the experimental reaction rates Robs showed a strong pH dependency and also varied between minerals. The pH dependency can be explained by a surface speciation model according to which the rate is proportional to the concentration of a surface complex >FeHS that forms upon reaction with >FeOH. Considering this model an empirical rate law can be derived Robs = k*c(H2S)tot c(H+)-n , where the empirical rate coefficient k* accounts for surface and solution chemical properties..

k* significantly varies between minerals and increases in the order 6LFh, 2LFh, Gt, Hem. Assuming that the different mineral phases also have different surface acidities, the pH pattern of the reaction rate could be satisfactorily fitted by an intrinsic pKa of 4.9 for hematite and 6.9 for goethite.


There are two major implications from this study: 1) A low intrinsic pKa1 provides a higher concentration of the reactive surface species >FeOH under acidic conditons and therefore the rate will be higher for this mineral; 2) the higher the intrinsic pKa1, the more the rate will depend on pH.

References Dos Santos Afonso M & Stumm W. Langmuir, 8, 1671-1675 Peiffer S et al, Environ. Sci. Technol., 26, 2408-2413


Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

Rates of uranium electron transfer: A theoretical perspective K.M. ROSSO, Z. WANG, C.C. AINSWORTH AND J.K. FREDRICKSON Pacific Northwest National Laboratory, P.O. Box 999, MSIN K8-96, Richland, WA 99352, USA ([email protected], [email protected], [email protected], [email protected]) Redox transformation of U(VI) to U(IV) by metal reducing bacteria is a potentially important technological strategy for controlling uranium mobility in aqueous subsurface environments. Despite the well-known thermodynamic relationship between uranium oxidation state and its solubility, molecular-scale controls on the kinetics of uranium electron transfer (ET) have not been examined in detail. For example, equatorial ligands have the capacity to serve as ET bridges, steric hinderances, or to stabilize various oxidation states.

Figure 1. Ball and stick model of a calculated U(VI)-U(V) electron transfer encounter complex including solvation. One of the hallmarks of Marcus’ ET theory is the crossrelation, which describes the ET rate (k12) for A + B → A+ + B− in terms of k11 and k22, the separate selfexchange rates of the reactants (i.e., A + A+ → A+ + A and B + B− → B− + B). Hence, analysis of self-exchange ET is akin to accessing the intrinsic ET behavior of a species. In this study, we present the results of ab initio calculations on the kinetics of UO22+ − UO2+ self-exchange ET, including a discussion of the roles of equatorial water, hydroxide, CO32−, SO42−, and PO43− ligands. The quantities that enter into Marcus’ theory have been computed. Increasing the numbers of equatorial water ligands increases the reorganization energy. The electronic coupling matrix element is highly dependent on the structure of the encounter complex. Calculated second-order ET rates are slow (~ 1 M−1 s−1), in accord with the limited available experimental data.

Effects of metal cation binding on the colloidal stability of kaolinite-fulvic acid particles RUBEN KRETZSCHMAR, ILONA HEIDMANN AND ISO CHRISTL Institute of Terrestrial Ecology, ETH-Zurich, Switzerland. ([email protected]) Adsorbed natural organic matter has a strong influence on trace metal sorption to mineral surfaces and the colloidal stability of clay particles. Mobile clay-organic colloids play an important role in trace element cycling in rivers, lakes, soils, and groundwater aquifers [1]. We investigated the effects of Cu(II) and Pb(II) on the colloidal stability of kaolinite colloids in the presence of adsorbed fulvic acid, an important component of natural organic matter. The effects of Cu and Pb on the aggregation rate and electrophoretic mobility of kaolinite-fulvic acid colloids were compared with that of Ca, a major divalent metal cation which is less strongly adsorbed. Kaolinite-fulvic acid suspensions (at pH 4 and 6) were spiked with solutions containing Cu, Pb, or Ca to give total divalent cation concentrations between 10-5 and 8×10-3 M. The mass ratio of kaolinite to fulvic acid was 500:3 in all experiments. Starting with a well-dispersed suspension, the increase in average hydrodynamic radius was monitored by dynamic laser light scattering (DLS). Relative aggregation rates were determined from the slopes of the initial linear increase in average hydrodynamic radius with time. Addition of fulvic acid to the suspensions completely inhibited kaolinite aggregation at pH 4 and 6. Additions of Cu, Pb, and Ca resulted in strongly increased aggregation rates of kaolinitefulvic acid colloids. Based on total concentrations, the potential of the three cations to enhance aggregation was much higher for Cu and Pb than for Ca. However, the relationship between electrophoretic mobility and aggregation rate was the same for all three divalent metal cations at pH 4. In the presence of Ca, an increase in pH from 4 to 6 resulted in decreased aggregation rates, while the opposite trend was observed for Cu and Pb. The effects of Cu, Pb, and Ca on the aggregation rates of kaolinite-fulvic acid colloids are explained by the sorption behavior of the three divalent metal cations to kaolinite-fulvic acid colloids [2].

References [1] Kretzschmar, R., Borkovec, M., Grolimund, D., and Elimelech, M. (1999). Adv. Agron. 66: 121-194. [2] Heidmann, I., Christl, I., and Kretzschmar, R. (2005) Environ. Sci. Technol. (in press).

Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes


Citrate impairs the diffusion of phosphate into goethite

Nucleation and growth of manganese oxide films



Berlin University of Technology, Department of Soil Science, Salzufer 12, 10587 Berlin, Germany ([email protected])

Division of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138, USA ([email protected], [email protected])

Objective and Methodologies

Manganese oxide in surface and ground waters dissolves and precipitates in response to natural and anthropogenic cycling of aqueous pE and pH conditions. When precipitating, the manganese oxides form thin coatings on mineral surfaces. Although the thin coatings affect the overall rates of manganese dissolution and precipitation, the current understanding of the active mechanisms is limited. In the current work, we use atomic force microscopy to investigate how surface morphology, manganese concentration, and substrate chemistry affect the manganese oxide film’s characteristics such as orientation, height, and shape. The (10 1 4) surfaces of MnCO3, MgCO3, and CaCO3 are used as substrates. Aqueous Mn2+, which serves as the source material for the manganese oxide film, derives in some experiments from MnCO3 dissolution and in others from external addition of Mn2+(aq). We find that the growth of manganese oxide films depends on substrate morphology, i.e., terraces versus highly stepped surfaces. Rhombohedral two-dimensional manganese oxide islands grow on terraces of MnCO3. The islands are rotated 90° relative to the crystallographic axis of the underlying carbonate. Island heights self-limit between 2 and 3 nm depending on reaction conditions. In comparison to island height, the lateral dimensions observed begin at 60 nm and grow steadily to several microns before the islands begin to collide and coalesce. The two-dimensional islands do not grow over substrate steps. In contrast to terraces, highly stepped surfaces result in the growth of manganese oxide film along steps. The areal growth rate in regions of highly stepped surfaces is three times faster than that on terraces. Comparison studies done with MgCO3 and CaCO3 show that the former also promotes heteroepitaxial growth whereas the latter does not. This difference is explained by the relative bond length mismatch between the structures of the carbonate substrates and the atomic structure of the Mn oxide film. A free energy model is also presented to explain why the heights of the Mn oxide islands self-limit. These results provide an improved basis both for the development of predictive models of contaminant fate and transport and for the modelling of hydraulic flow through carbonate aquifers.

Root excreted citrate has been shown to mobilise phosphate (PO4) in the rhizosphere by competition for sorption sites of Fe- and Al (oxy)hydoxides. We hypothesize that citrate additionally impairs the diffusion of PO4 into micropores (<2 nm) of goethite in the rhizosphere. We studied the PO4 sorption kinetics of a synthetic goethite and, to be closer to soil conditions, a goethite that was coated with dissolved organic matter (DOM). Sorption kinetics were determined in batch systems (three weeks, pH5). We varied the order of addition of citrate and PO4 at equimolar concentrations (500 µM). In addition, we studied the PO4 sorption kinetics in the absence of citrate. We analyzed our samples for specific surface area, porosity and ζpotential prior and after PO4 and citrate sorption.

Results and Discussion

Both, PO4 and citrate diffused into micropores of pure and DOM-coated goethite, causing a partial clogging of these pores. While the clogging of micropores by PO4 has recently been observed (Makris et al., 2004), no evindence for citrateinduced micropore clogging exists in literature yet. The diffusion of PO4 into the micropores of pure and DOM-coated goethite during three weeks of sorption was impaired in the presence of citrate by up to 100%. The competition of PO4 with citrate for micropore diffusion depends on the order of addition of both ions. A greater diffusion limitation of PO4 occurred when citrate was added before PO4.


Our results indicate that PO4 and citrate not only compete for sorption sites but also for diffusion into goethite particles. This mechanism might reduce the PO4 fixation in micropores of Fe (oxy)hydroxides in the rhizosphere and thus enhance the availability of PO4 to plants following the exudation of low molecular weight organic acids.

Reference Makris, K.C. et al., (2004), Environ. Sci. Technol. 38, 65906596.


Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

Self-consistent self-interaction corrected DFT studies of annite

Siderophore-metal interactions on oxide surfaces




Pacific Northwest Laboratories, Richland, WA, 99352, USA ([email protected], [email protected], [email protected]) 2 University of Washington, Seattle, WA, 98195, USA ([email protected]) One of the more persistent failures of standard DFT methods has been their failure to yield localized charge states such as polarons, excitons and solitons in solid-state and extended systems. It has been suggested that standard DFT functionals which are not self-interaction free tend to favor delocalized electronic states since self-interaction creates a Coulomb barrier to charge localization. Pragmatic approaches in which the exchange correlation functionals are augmented with small amount of exact exchange have shown promise (i.e. B3LYP and PBE0) in localizing charge states. However, a large amount of exact exchange must be added in order for a these methods to yield localized charge states, which results in band-gaps and reaction barriers being overestimated. We have recently developed a framework for implementing selfconsistent self-interaction corrections (SIC) into pseudopotential plane-wave density functional theory (PSPW). This technique has shown great promise predicting localized charge states as well as accurate band-gaps and reaction barriers. Furthermore, this technique is applicable to both confined and extended systems, as well as to CarParrinello ab initio molecular dynamic simulations. We present here results from a recent study of the oxidation of annite in air. Some evidence exists that oxidation of annite will result in a charge states inside the single-particle band gap, associated with localized Fe3+ ion state. Such a state is typically the result of local distortion of the lattice leading to localization of the electronic wave function. So far, only small cluster models within Hartree-Fock approximation have yielded such a localized Fe3+ state. It is well known that the Hartree-Fock approximation unnecessarily favors localized states. In contrast, large-scale periodic DFT calculations using standard DFT functionals do not localize these types of states. Using our newly developed DFT+SIC method we have been able to obtain a significant degree of charge localization. Various aspects of this application and implications of the DFT+SIC theory to modeling charge transfer in extended systems will be discussed.


Sierra Nevada Research Institute, University of California, Merced, California 95344, USA 2 University of Texas, Austin, Texas 78712, USA Desferoxamine B (DFB) is a representative oxamic acid siderophore. These natural chelates are produced by a number of microorganisms and play a significant role in Feavailability to these organisms. Siderophores have recently received considerable attention because of their potential to react with lanthanides, actinides and an array of other toxic metals. Indeed, the fate and transport of a number of metals in soils and sediments may be greatly influenced by reactions with siderophores. The present study examines the effects of on DFB of Eu, Nd, Pb and Zn desorption from hematite and maghemite. Macroscopic measurements show that siderophore-induced metal desorption is incomplete removing only a fraction of the surface bound sorptive. Dissolution of the mineral substrate (indicated by release of Fe to solution) appears to be subsequent, rather than simultaneous to the release of Eu, Nd, Zn and Pb. In all cases, DFB-induced release of Fe is less in the presence of adsorbed Eu, Nd, Zn and Pb than their absence. Europium-, Nd-, Pb- and ZnEXFAS collected prior to reaction of the mineral-sorptive adducts with DFB indicate a prevalence of mono-dentate surface species at the mineral/water interface. In contrast EXAFS spectra collected after reaction of DFB with these same mineral-metal complexes shows a prevalence of bidentate surface bound metals, suggesting that these species are less reactive and therefore less susceptible to removal by the complexing ligand. The implications of this work to contaminant chemistry and elemental cycling will be discussed.

Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

Formation of manganesedesferrioaxime B complexes by dissolution of manganese oxides OWEN W. DUCKWORTH AND GARRISON SPOSITO Divison of Ecosystem Sciences, University of California, Berkeley, CA 94720, USA ([email protected], [email protected]) Recent laboratory and field studies suggest that Mn(III) forms persistent aqueous complexes with high-affinity ligands. Aqueous Mn(III) species thus may play a significant, as-yet largely unexplored role in biogeochemical processes. To that end, we have studied the interaction of desferrioxamine B (DFOB), a common tri-hydroxamate siderophore, with manganese. DFOB stabilizes Mn(III) for the pH range 7.011.3, forming a Mn(III)HDFOB+ complex. The Mn(III)HDFOB+ complex may be formed by DFOBpromoted dissolution of solid Mn-oxides (Figure 1). At pH > 6.5, the dissolution of manganite (γ-MnOOH) in the presence of DFOB is predominantly a non-reductive ligand-promoted reaction whose rate (RL) is proportional to the adsorbed surface concentration of DFOB. At pH < 6.5, Mn2+ is the dominant species resulting from manganite dissolution, thus implicating a reductive dissolution pathway (RR).

Figure 1. DFOB-Promoted dissolution of in 0.1 M NaCl and 10 mM buffer.

-MnOOH at 25ºC

The results of this study suggest that Mn(III)-siderophore complexes may be readily produced by Mn-oxide dissolution. The formation of these complexes may have implications for the biogeochemical cycling of manganese, redox-active elements, and siderophores in natural environments.


Effects of siderophores on Pb adsorption to kaolinite PATRICIA MAURICE1, BHOOPESH MISHRA2, ELIZABETH HAACK1 AND BRUCE BUNKER2 1

Dept. of Civil Engineering & Geological Sciences, University of Notre Dame, Notre Dame, IN 46556, USA ([email protected]) 2 Dept. of Physics, University of Notre Dame, Notre Dame, IN 46556, USA Siderophores are low molecular weight organic ligands synthesized by aerobic microorganisms to acquire Fe under conditions of Fe stress. In addition to Fe (III), siderophores may complex other metals such as Pb. This study compared the effects of trihydroxamate siderophores desferrioxamine-B (DFO-B), desferrioxamine-D1 (DFO-D1), desferrioxamine-E (DFO-E), and monohydroxamate siderophore-like ligand acetohydroxamic acid (aHA) on Pb adsorption to kaolinite (KGa-1b) at pH 4.5 to 9, in 0.1 M NaClO4, at 22oC, in the dark. At pH > 6, all of the studied ligands decreased Pb adsorption to kaolinite: aHA by 5 to 40% and DFO-B, DFOD1 and DFO-E by 30 to 75%. However, DFO-B enhanced adsorption at pH ~5 to 6.5, which we hypothesized to be due to adsorption of the doubly positively charged PbH3(DFOB)2+complex, which occurs in this pH range. EXAFS results confirmed that this is indeed the case. EXAFS results also showed that Pb adsorption to kaolinite in the presence of DFO-B is dominated by Pb-DFO-B species at pH > 5. Overall, our results demonstrate that siderophores may have different effects on Pb adsorption and mobility through porous media depending upon the siderophore structure and the pH; comparison with other studies suggests sensitivity to mineral structure as well.

Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

Pb speciation in the presence of siderophores and clay surfaces – XAFS study 1





Dept. of Physics, University of Notre Dame, IN 46556, USA ([email protected], Corresponding author: [email protected]) 2 Dept. of Civil Engineering and Geological Sciences, IN 46556, USA Although siderophores are low molecular weight organic ligands produced by microorganisms to acquire Fe, the ligands also have high binding affinities for other metals, such as Pb. This work is part of a larger project by our group that has examined the impact of the trihydroxamate siderophore DFOB on Pb sorption to kaolinite. Results indicate that the presence of DFO-B affects the partioning of Pb from the aqueous to the solid phase; Pb adsorption is enhanced in the presence of DFO-B around pH 5-6 and is inhibited at pH values > 6.5. To better understand these results, we measured Pb speciation at the molecular level, using x-ray absorption spectroscopy (XAFS), in the mixed Pb-DFO-B (aqueous phase), Pb-kaolinite (solid phase), and Pb-kaolinite-DFO-B (solid phase) systems at pH 4, 6 and 7.5. XAFS results confirm that pH affects the coordination environment of Pb complexed by DFO-B in solution, and XANES suggests a change in chemical state between complexes. At the kaolinite surface, Pb is sorbed dominantly as a Pb-DFO-B complex at pH > 5. Further, the extent of Pb sorption at pH 6 exceeded that at pH 7.3 in the mixed Pb-kaolinite-DFO-B solid phase as measured by X-ray fuoresence, in agreement with the results of the bulk adsorption study. Comparison of the three different phases suggests that the mixed Pb-kaolinite-DFO-B solid phase is not a simple linear combination of the solid Pbkaolinite and Pb-DFO-B aqueous phase at any of the pH values examined. Taken together, bulk adsorption measurements and XAFS experiments represent a powerful approach for determining metal speciation and refining existing models of metal distribution at water-rock interfaces.

Effect of synthetic and biogenic surfactants on iron oxide dissolution NARAYA CARRASCO, RUBEN KRETZSCHMAR AND STEPHAN M. KRAEMER Institute of Terrestrial Ecology, ETH Zurich, Grabenstr. 3, 8952 Schlieren, Switzerland ([email protected]) Bio-surfactants are released by microorganisms and plants into soils and sediments. They modify the properties of solid/water interfaces and facilitate microbial attachment (biofilm formation) or nutrient acquisition. We have investigated the effect of synthetic and biogenic surfactants on the physico-chemical properties of iron oxide (goethite) surfaces, on the adsorption of organic ligands, and on ligand controlled weathering. The ligands chosen for this investigation included microbial siderophores and other common biogenic low molecular weight organic ligands. They were selected for their significance in biogeochemical processes in natural systems and for their contrasting properties regarding charge and hydrophobicity. without SDS 15 µM SDS

60 Dissolved Fe [µM]


50 40 30 20 10 0


1000 2000 3000 4000 5000 6000 Time [min]

Figure 1: Dissolution of goethite in presence of 80 µM DFOB with and without SDS. We observed slow adsorption kinetics of surfactants on goethite surfaces. Adsorption rates increased with increasing surface charge and with increasing surfactant concentrations. Adsorption of surfactants had a strong effect on the mineral surface charge as indicated by electrophoretic mobility measurements. Surface charge reversal at pH 6 was observed in adsorption equilibrium with very low surfactant concentrations in the micromolar range. In this low concentration range, we observed a significant effect of surfactants on the adsorption of the microbial siderophore DFO-B and on the rates of ligand (i.e. DFO-B) controlled dissolution (figure 1). These observations indicate that even low concentrations of surfactants can have a significant effect on important biogeochemical processes including nutrient acquisition and mineral weathering.

Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

Metal binding by a novel biogenic chalkophore, methanobactin, and the effect on microbial activity JEREMY D. SEMRAU1, ALAN A. DISPIRITO2 3 AND W.E. ANTHOLINE 1

Department of Civil and Environmental Engineering, The University of Michigan, USA ([email protected]) 2 Department of Biochemistry, Biophysics, and Molecular Biology, Iowa State University, USA ([email protected]) 3 Department of Biophysics, Medical College of Wisconsin, USA ([email protected]) One of the persistent and substantial problems in remediation of hazardous waste sites is the mobilization and uncontrollable transport of radionuclides and heavy metals from these sites to surrounding areas. Some microbiallymediated processes can at least temporarily immobilize and reduce the toxicity of these materials through dissimilatory reduction that leads to precipitation and sorption under anaerobic conditions. As such, microbial-mediated processes can limit the dispersal of these materials and thus also reduce the risk of exposure to surrounding areas. One must realize, however, that micro-organisms have effective and ubiquitous mechanisms to solubilize different metals and that nonspecific binding of radionuclides and heavy metals by these biogenic metal chelators may increase their solubility, mobility, and bioavailability. A group of cells ubiquitous in the environment, methanotrophs, require substantial amounts of copper as it is an integral part of one of their key enzymes, the particulate methane monooxygenase. These cells thus have strong mechanisms to sequester copper, and have recently been discovered to produce a novel biogenic metal chelator, what we term methanobactin. This compound, composed of 7 amino acids and two 2-thionyl-5-hydroxy-imidzole chromophores along with a pyrrolidine that confers a strong bend in the overall chain, binds copper quite well. The metal coordination environment is comprised of a dual N,S-donating system derived from the imidazolyl moieties and forms a distorted tetrahedral geometry. Of key interest here is that methanobactin will bind a variety of metals other than copper, including cadmium, cobalt, iron, mercury, manganese, zinc and uranium. Furthermore, cells expressing methanobactin can scavenge metals from other chelating agents as well as metals bound to soils. In this presentation, spectroscopic (UV/VIS and EPR) data along with the thermodynamics of metal binding by methanobactin will be discussed as well as the effects of such binding on methanotrophic activity. Finally, the diversity of methanotrophs that can express methanobactin as determined using PCR methodologies will be discussed.


Microbial diversity and geochemical heterogeneity within siliceous sinters SAMIR ABD EL-FATAH1, ZIYA S CETINER2, THOMAS J. WILLIAMS2 AND SUSAN E. CHILDERS2 1

Environmental Health and Safety, University of Idaho, Moscow, ID 83844-3022, USA ([email protected]) 2 Department of Geological Sciences, University of Idaho, Moscow, ID 83844-3022, USA ([email protected], [email protected], [email protected]) The formation of sinters is of biological and geological interest because silica-rich geothermal environments are most similar to environments thought to harbor life on early Earth, and because deposits of precious minerals (ie, gold) are often associated with siliceous sinters. Recent studies have demonstrated a variety of thermophiles are found in association with siliceous sinters from a diversity of environments yet the role of microorganisms in sinter formation is not apparent. The purpose of the study is to compare the microbial and geochemical diversity within siliceous sinter taken from spatially separate thermal springs of the Alvord Basin (OR, USA). Previous studies have indicated that the overall geochemistry of thermal waters emanating from the >200 thermal springs within the Alvord Basin is remarkably similar. As the role of microorganisms to the formation of siliceous sinter within any geothermal system is not clear, a comparison of the microbial diversity of the sinters formed within geochemically similar waters could lead to the delineation of a core group of microbes essential for sinter formation. Denaturing gradient gel electrophoresis was used to compare communities of ten different sinter samples from springs with temperatures ranging from 70-95°C. To determine sinter geochemistry, EDX and X-ray diffraction analyses were performed on the sinter samples. The springs within the Alvord Basin provide a unique opportunity to determine if functionally similar groups of microorganisms are necessary for sinter formation, and the correlation of certain microorganisms to the precipitation of precious metals.


Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

Effects of maghemite (γ-Fe2O3) nanoparticles on the toxicity of arsenic within cultured human fibroblasts M. AUFFAN, J. ROSE AND J.Y. BOTTERO CEREGE, Interfacial Physical-chemistry Group and IFR PMSE, Europôle Méditerranéen de l'Arbois BP80 13545 Aix en Pce Cedex4 France ([email protected]) Arsenic is a toxic metalloid largely studied these last years. Its toxicity is known to be related to its oxidation state and speciation. In cellular media, arsenic shows a strong affinity with thiol functions, which play an important role on arsenic biocellular transformations and cancerogenic effects. Serious health problems due to high rates of arsenic in drinking water, amplified the interest in minerals such as iron oxide which can immobilize arsenic species on their surfaces and so decrease their toxicity. Recently, Nano-Maghemites (iron oxide nanoparticles with a diameter <10nm), are studied for biomedical purposes (MRI contrast agents, drug delivery or cell engineering). These nanoparticles are very attractive because around 50% of their atoms are near the surface, that increases significantly their surface energy, reactivity and affinity with adsorbates. The aim of this work is to assess if Nano-Maghemites could be used, in cellular media, as an effective tool of As immobilisation and detoxification. Interactions between Nano-Maghemites and As(III) in cultured human fibroblasts were studied by X-ray Absorption Spectroscopy (XAS) at the As K-edge on the XAS CRG french beamline “FAME”. Preliminary experiments led in deionised water indicate that the adsorption capacity of NanoMaghemites is high (4.7 As atoms per nm2). Arsenic is sorbed through double corner sharing between two Fe surface atoms, with a mean distance As-Fe of 3.38ű0.2Å. However, in cultured human fibroblasts, As adsorption is reduced due to competition effects, at nanoparticles surface, between As species and other organic or inorganic compounds. Another XAS results indicate that Nano-Maghemites cause a 40% reduce of the total number of thiolated arsenic species, in the extracellular medium. Arsenic is then more available to interact with cells. Thus Nano-Maghemites seems to modify the As biocellular transformations and to increase its toxicological effects.

Rapid ectomycorrhizal channel development on biotite in liquid culture experiments ZS. BALOGH1, C.K. KELLER1, F. STEVENS2 2 AND J.T. DICKINSON 1

Geology, Washington State University, Pullman, WA 991642812, USA ([email protected], [email protected]) 2 Physics, Washington State University, Pullman, WA 991642814, USA ([email protected], [email protected]) Liquid culture experiments were carried out to investigate the alteration of biotite by an ectomycorrhizal fungus, Suillus tomentosus. 1cm by 1cm biotite flakes were incubated on a shaker table at room temperature in 8% glucose solution inoculated with the fungus. The pH of the glucose solution was lowered from 6 to 4 over the course of the experiment by the fungus, whereas the control pH increased to 6.5. After 12 weeks the flakes were examined by Scanning Electron Microscopy (SEM). Fungal hyphal development was documented growing from the edges inward, but no significant surface changes were detectable by SEM (Fig.1). Figure 1: SEM images of the hyphal development on a biotite

The flakes were examined by Atomic Force Microscopy (AFM) using contact mode in air. When compared to freshly cleaved material, the samples incubated with fungi exhibited rougher surfaces with well defined dissolution channels (Fig. 2). The AFM observed dissolution channel network is similar to the hyphal network documented by SEM. Figure.2: AFM images of dissolution channels on a cleaved biotite surface

Our results suggest that ectomycorrhizal fungus is responsible for producing dissolution channels in a short time interval. Such fungi can therefore play an important role in biogeochemical cycling of nutrients in biotite.

Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

Reduction of U by adsorbed vs. surface-precipitated Fe(II) at model cell surfaces MAXIM I. BOYANOV1, EDWARD J. O’LOUGHLIN1, SHELLY D. KELLY1, ERIC E. RODEN2, JEREMY B. FEIN3 1 AND KENNETH M. KEMNER 1

Argonne National Laboratory, Argonne, IL, USA ([email protected], [email protected], [email protected], [email protected]) 2 University of Alabama, Tuscaloosa, AL, USA ([email protected]) 3 University of Notre Dame, Notre Dame, IN, USA ([email protected]) We investigated the effect of Fe-Fe proximity on the reduction of U(VI), using aqueous suspensions containing Fe(II) and carboxyl-functionalized microspheres as an abiotic model of bacterial cell-surface associated Fe(II). Fe K-edge and U L-edge X-ray absorption fine-structure spectroscopy (XAFS) was used to determine the valence state and atomic environment of the adsorbed species, while acid-base titrations and batch-sorption experiments provided information on proton release and metal uptake as a function of pH. In the binary (Fe + carboxyl surface) system we observe two regimes of Fe(II) sorption: mononuclear at circumneutral pH and 1 mM Fe and polynuclear at conditions just below saturation with respect to bulk Fe(OH)2(s) formation. In the ternary system (U + Fe + carboxyl surface) we did not observe U(VI) reduction when Fe(II) was sorbed as mononuclear species; however, U(VI) was reduced to uraninite nanoparticles when Fe(II) was adsorbed as a polynuclear species. Our results suggest that Fe-Fe coordination is necessary for U(VI) reduction by Fe(II) in the absence of electron-shuttling agents or conductive surfaces. This abiotic study also elucidates the physico-chemical processes that may be occurring at and near the charged cell surface. The submitted manuscript has been created by the University of Chicago as operator of Argonne National Laboratory under Contract No. W-31-109ENG-38 with the U.S. Department of Energy. The U.S. government retains for itself, and others acting on its behalf, a paid-up, nonexclusive, irrevocable worldwide license in said article to reproduce, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the government.

Acknowledgements Work supported by the U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research, under contract W-31-109-Eng-38.


The geomicrobiology of surficial geochemical anomalies L.D. GOODHUE1, S. HAMILTON2 AND G. SOUTHAM1 1

Dept. of Earth Sciences, The University of Western Ontario, London, Ontario, N6A 5B8, Canada ([email protected], [email protected]) 2 Ontario Geological Survey, Sudbury, Ontario, P3E 6B5, Canada ([email protected]) Dispersion halos occur in soil horizons above buried ore bodies that possess geochemical signals originating from the buried ore (Cameron et al., 2004). The model by Hamilton et al. (2004) describe a reduced column that forms above buried sulfides, facilitating the transport of reduced iron (and base metals) to the soil surface where it is oxidized and precipitates (immobilising the base metals). Aerobic heterotrophic bacteria, anaerobic heterotrophic bacteria, dissimilatory sulfate reducing bacteria (SRB) and acidophilic iron oxidizing bacteria were enumerated in samples collected from the Cross Lake dispersion halo environment. The heterotrophic bacterial counts were generally enhanced within this environment; with the anaerobic heterotrophic bacteria counts exceeding those of the aerobic heterotrophic bacteria. Also, low populations of SRB were enriched over mineralization suggesting that low concentrations of sulfate are being transported to the surface from the ore body. Acidophilic iron oxidising bacteria were not recovered from these samples, suggesting that chemical oxidation may be responsible for the mobilisation of the geochemical signatures from the ore. The enhanced bacterial biosphere suggests that one of the roles for bacteria in these systems is to maintain the reduced column above the buried sulphide deposit, which then facilitates the transport of dissolved metal species to the surface. The relationship between bacteria, oxidation-reduction potential (ORP), spontaneous potential (SP), selective leach extractions, and soil gas hydrocarbons (SGH) then, appears to be a contemporary process that could be examined under laboratory conditions to improve understanding of these systems.

References Cameron E.M., Hamilton S.M., Leybourne M.I., Hall G.E.M. and McClenaghan M.B, (2004), Geochem.: Explor., Environ., Anal. 4:7-32. Hamilton S.M., Cameron E.M., McClenaghan M.B. and Hall G.E.M., (2004), Geochem.: Explor., Environ., Anal. 4:3344.


Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

Examining the nature of siderophoremontmorillonite interactions E.A. HAACK AND P.A. MAURICE Dept of Civil Engineering and Geological Sciences, University of Notre Dame, IN 46556, USA (Corresponding author: [email protected]) Whereas microbial siderophores have been shown to affect clay mineral dissolution and metal adsorption to clays, our understanding of the mechanisms of interactions between siderophores and the clay structure itself, including adsorption, are more limited. To better understand the nature of the interaction between siderophores and montmorillonite clay, we measured sorption of two hydroxamate desferrioxamine siderophores, DFO-B and DFO-D1, to Na-saturated montmorillonite in 0.1 M NaClO4 at 22oC. These two siderophores have the same fundamental structures, but different terminal groups lead to DFO-B being monovalently charged to pH ~8 (pKa of deprotonation reaction is 8.4), whereas DFO-D1 is uncharged. Sorption of DFO-B was significant at pH between 4 and 6 (195 µmol/g), decreased to 170 µmol/g at pH 8, and to < 20 µmol/g at pH 10. DFO-D1 followed the same sorption trend with pH as DFO-B, but was sorbed to a lesser extent. DFO-B sorption densities were approximately 85 µmol/g, 45 µmol/g, and 5 µmol/g at pH 4-6, 8 and 10, respectively. The observed adsorption of the neutral molecule DFO-D1 to montmorillonite over a wide pH range suggests that electrostatic interactions are at most only partly responsible for the sorption of trihydroxamate siderophores to montmorillonite. Additional research on sorption kinetics and mechanisms, including application of spectroscopic and diffraction techniques, is ongoing.

Abiotic Tc(VII) reduction by Fe(II) BYONG-HUN JEON, JOHN M. ZACHARA, CHONGXUAN LIU, RAVI KUKKADAPU AND ALICE DOHNALKOVA Pacific Northwest National Laboratory P. O. Box 999, MS K8-96, Richland, WA 99352, USA The subsurface behavior of 99Tc, a contaminant resulting from nuclear fuels reprocessing, is strongly dependent on its valence (e.g., IV or VII). We investigated the abiotic reduction of soluble Tc(VII) by Fe(II) in pH 6-8 solutions under strictly anoxic conditions using an oxygen trap (<7.5 ×10-9atm O2). Complete and rapid reduction of Tc(VII) to precipitated Tc(IV) [Tc(IV)O2·nH2O(s)] was observed when 11µM of Tc(VII) was reacted with 0.4mM Fe(II) at pH 7.0 and 8.0, while no significant reduction was observed over 1 month at pH 6.0. The reduction kinetics were strongly pH dependent. Experiments conducted at pH 7.0 with Fe(II) = 0.05-0.8mM further revealed that Tc(VII) reduction was a combination of homogeneous and heterogeneous reaction (Figure 1). The kinetics of homogeneous reduction correllated with the concentration of Fe(OH)+. Fe(II) loss from solution was higher than the stoichiometric demand for complete reduction to Tc(IV) indicating Fe(II) sorption or precipitation. Preliminary TEM and Mössbauer spectroscopy measurements indicated that the reaction product was a magnetite/Tc(IV) coprecipitate.

Figure 1. Abiotic reduction of Tc(VII) by Fe(II) (A) and recovery of added Fe(II) (B) from the 10mM PIPES buffer solutions for different added Fe(II) concentrations at pH 7.0. [Tc(VII)]added = 11µM, and [Fe(II)]added = 0.05-0.8mM.

Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

The effect of bacterial and plant siderophores on UIVoxide dissolution

Citrate adsorption at the watergoethite interface: A spectroscopic evaluation of surface complexes



Institute of Terrestrial Ecology, ETH Zurich, Grabenstr. 3, 8952 Schlieren, Switzerland ([email protected]) IV


1 IV

Tetravalent actinides such as Plutonium or Uranium form oxide phases that have very low solubilities. Therefore, tetravalent actinides are often considered immobile in stable reducing environments. However, it was demonstrated that tetravalent actinides are coordinated by siderphores forming exceptionally stable soluble somplexes. Siderophores are organic ligands that are exuded by plants and microorganism to increase the bioavailability of iron, an essential nutrient. They are considered iron specific due to their high affinity for iron relative to other common constituents of ground- and surface water. In this context we investigated the effect of plant and bacterial siderophores on the solubility of uranium oxide, and on the rates and mechanisms of uranium oxide dissolution under reducing conditions. We also investigated the effect of the presence of Fe(III) on the dissolution process. The uranium oxide was synthesized by reduction of dissolved uranyl nitrate and precipitation, followed by drying, washing, and further reduction in a pure hydrogen atmosphere at 850 degree C. Dissolution experiments were performed in batch and in continuous flow stirred tank reactors. Siderophore adsorption as a function of soluble siderophore concentrations or of pH were conducted in batch experiments. Plant siderophores (desoxymugineic acid) were purified form root exudates of wheat. Bacterial siderohpores (desferrioxamine B) were purchased. Proton promoted dissolution rates of uranium oxide decreased with increasing pH. In the presence of siderophores, dissolution rates were several times greater than the proton or hydroxide promoted dissolution rates alone. A linear correlation was observed between dissolution rates and surface concentrations of adsorbed siderophores. This is consistent with a ligand controlled dissolution mechanism. However, dissolution rate coefficients had a minimum at pH 7.5 and increased toward lower and higher pH. The presence of Fe(III) had only a minor effect on uranium oxide dissolution rates at molar ratios Fe(III)/siderophore < 0.9. This is consistent with the low redoxs potential of Fe-siderophore complexes below the redox potential of the U(IV)/U(VI) pair which inhibits oxidation of uranium(IV) by Fe(III). In summary, our observations demonstrate that biogenic ligands can could have a significant effect on the mobility of tetravalent actinides in the environment.

Dep. of Chemistry, Inorganic Chemistry, Umeå University, 901 87 Umeå, Sweden ([email protected]) 2 Idaho Natl Engn & Environm Lab, Idaho Falls, Idaho, USA The mobility and bioavailability of organic and inorganic ligands in soils and aquifers are dependent on interactions with mineral surfaces. Carboxylic acids are a group of ligands that are abundant in most soil systems. As complexing agents they are involved in detoxification of metals, chemical weathering of minerals and acquisition of nutrients by plants. The present work focuses on interactions between citrate and goethite (α-FeOOH) surfaces. Citrate is a hydroxyl-tri-carboxylic acid that has been shown to participate in all of the above mentioned processes. There are disparities in the literature regarding the types and structures of citrate-goethite surface complexes [e.g. 1-2], and the present work was an attempt to resolve this controversy. In order to meet our research objective, both quantitative and qualitative data were collected. In-situ ATRFTIR spectroscopy was used to evaluate the surface speciation of citrate, and scintillation counting and atomic absorption spectroscopy were used to gain quantitative adsorption and dissolution data. The work was conducted in series of batch experiments and a simultaneous infrared and potentiometric titration, measuring ligand adsorption as a function of pH, time and ligand concentration. With the methods used, we find evidence for at least three different surface complexes. Two of these are outer sphere complexes of which one is protonated and one is deprotonated. The protonated form transforms to the unprotonated one at around pH 3.5. The third species is an inner sphere complex which exists over the whole pH range investigated (3-9), persisting to unusually high pH values compared to other mono-, di-, tri- and tetracarboxylic acids. This increased stability is ascribed to a polydentate coordination between carboxylic group(s) and the hydroxyl group of citrate and surface Fe(III). No dissolution was detected above pH 5 at any citrate concentration and only up to 0.15% of an estimated monolayer below that pH.

References [1] Filius J. et al (1997) J. Col. Int. Sci 195, 368-380 [2] Lackovic K. et al. (2003) J. Col. Int. Sci. 267, 49-59

Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes


Effects of the interaction between vermiculite and the bacterium Pseudomonas fluorescens strain CHA0 and its genetic derivatives BARBARA MÜLLER





Institute for Geotechnical Engineering, ETH Hönggerberg, Zürich, Switzerland ([email protected]) 2 Institute of Plant Sciences, ETH Zentrum, Zürich, Switzerland ([email protected]) It has been known for a long time that bacteria can alter various properties of clay minerals considerably. In an ongoing study we have analyzed the influence of Pseudomonas fluorescens strain CHA0 (wildtype) and its genetic derivatives strains CHA631, CHA77, CHA89, CHA400 and CHA661 (derivatives modified in the production of various metabolites) on the chemical (e. g. minor and trace element content), mineralogical (e. g. X-ray diffraction pattern, water content, grain size distribution, cation exchange capacity, layer charge, specific surface) and mechanical (e. g. rheology) properties of the clay mineral vermiculite affected by microbial activity. All strains change the BET surface as well as the grain size considerably. All of the mentioned strains with the exception of strain CHA661 take up Fe from vermiculite. The wildtype CHA0 moreover incorporates V, whereas strains CHA631 and CHA400 incorporate Mn. Zn is used by strains CHA77, CHA89, CHA400 as well as by CHA661 and Co by strain CHA89.

Electrochemical analysis at the microbe/mineral interface C.E. TURICK1, A.A. EKECHUKWU1 AND D.A. LOWY 2 1

Savannah River National Laboratory, Aiken, SC. 29808, USA ([email protected], [email protected]) 2 Nova Research, Inc., Alexandria, VA 22308, USA ([email protected]) Dissimilatory metal reducing bacteria (DMRB) grow by transferring electrons directly to solid, insoluble metal oxides, thereby contributing to the biogeochemical cycling of metals. By examining how DMRB transfer electrons from the cell to solid minerals, we can better understand, predict, and control this phenomenon as it relates to the biogeochemical cycling of metals, including the bioremediation of metal contaminated environments. We have developed novel analytical methods that address electron transfer (ET) from whole cells and their reactivity with specific solid terminal electron acceptors, including metal oxides, through the use of cyclic voltammetry (CV). Shewanella oneidensis MR-1, a model DMRB, is capable of coupling H2 oxidation to the reduction of solid terminal electron acceptors. To date we have shown that living cells of S. oneidensis MR-1 demonstrate ET to several types of working electrodes (including Fe(III) oxide/carbon paste electrodes), under anaerobic conditions in the presence of H2. Differences in ET were recorded using bacterial mutants with altered metal reduction capabilities as well as with different electrode surface composition. In order to verify CV studies, current response was monitored over time at a poised potential of 550mV vs. Ag/AgCl, as determined from CV analyses. Bacterial cells were immobilized onto a Pt working electrode with a dialysis membrane (10kDa MW cut-off) and were exposed to either N2 or H2 saturated deionized water containing 170mM NaCl. A sustained current response over time was greater for cells exposed to H2 as compared to N2.

Goldschmidt Conference Abstracts 2005 Interfacial Biogeochemical Processes

Siderophore and oxalate mediated desorption of uranyl from goethite D. WOLFF-BOENISCH AND S. TRAINA Sierra Nevada Research Institute, University of California Merced, 4225 N Hospital Rd., Atwater, CA 95340, USA Uranium is the major contaminant in soils and groundwaters at sites associated with the processing and disposal of materials during nuclear energy and nuclear weapons production. The immobilization and removal of radionuclide contaminants in soils by microbial transformations, sorption and mineralization show the remediation potential some natural microbes may have. In this study uranyl (UO22+) batch sorption and desorption experiments on goethite surfaces were performed in the presence and absence of organic ligands to estimate their effect on the mobility of U in soils. The ligands were oxalate and desferrioxamine B (DFO-B). The latter is a strong iron chelator produced by bacteria and other microbes to overcome low iron availability but which also complexes actinides. As model soil constituent a synthetic goethite with a BET specific surface area of 34.4 m2/g was employed. Experiments were performed at pH 6 (buffered with MES) in a NaNO3 background electrolyte solution of 10mM. The goethite concentration was set at 0.5 g/L and the adsorbate concentration ranged between 24 ppb and 12 ppm. Sorption of uranyl under these circumstances is nearly 100% and the surface concentration of U(VI) is a function of the amount of adsorbent in the solution. Desorption of uranyl by DFO-B versus time and ligand concentration reveals desorption to be immediate and the U recovery rate reaches a plateau at ~60% of the initial U(VI) concentration irrespective of any further increase in the aqueous DFO-B concentration. Comparing the desorptive effect of DFO-B with oxalate on an equimolar basis (240 µM), oxalate is a more effective desorbent than the siderophore. However, it might be rather dissolution of the underlying goethite surface than desorption that causes detachment of uranyl molecules. Dissolution experiments of goethite in the presence of oxalate or DFO-B anions reveal that oxalate shows hardly any dissolution promoting effect at low concentrations (< 200 µM) whereas DFO-B significantly enhances dissolution of goethite at any concentration (13-1200 µM). At high ligand concentrations (>2000 µM), however, oxalate out-competes DFO-B in the dissolution of goethite and this observation might explain the higher uranyl detachment with oxalate. U(VI) desorption with DFO-B is not related to an increase in the aqueous Fe content indicating that goethite dissolution takes place at uranyl-free sites.



Goldschmidt Conference Abstracts 2005 Microbeam Isotopic Ratio Measurement

Characterization of laser-induced aerosol for quantitative analysis of solids using LA-ICP-MS

Laser ablation MC-ICP-MS: Shedding new light on in-situ isotope ratio measurement




Laboratorium für Anorganische Chemie, Eidgenössische Technische Hochschule Zürich, 8093 Zürich, Switzerland ([email protected],[email protected]) 2 Paul Scherrer Institute, OBAA/A1,PSI Ost, CH-5232 Villigen, Switzerland ([email protected]) Over the last 20 years of existence of LA-ICP-MS a lot of fundamental insights into the ablation process, aerosol transport and the atomization, vaporization and ionization within the ICP have been gained. However, the source of elemental fractionation as the most significant limitation of this technique and their influence on the quantification and isotope ratio measurements is still in controversial discussion. Different methods to measure the particle size distribution were applied to characterize the structure of aerosols produced from various silicate matrices. Results show that the nanoparticles generated within the ablation cell already agglomerate to nano-particles clusters up to a size of 1 micron before leaving the ablation cell (193 nm) whereas the aerosols generated at 266 nm contain agglomerates and a significant portion of single particles in the order of 1 micron. Furthermore, the total composition of the aerosol entering the ICP was collected and analysed using solution nebulizationICP-MS. The collected aerosol generated at 266 nm and 193 nm in helium and argon represents the stoichiometric composition of the original sample (except Be, Fe and Cd). However, the size dependent composition of the aerosol (generated using 266 nm) entering the ICP-MS shows significant enrichment of volatile elements within the particle fraction below 340 nm and 125 nm, which indicates that the ion formation within a broader particle size distribution with highly variable composition can lead to matrix effects, especially when the absorptivity between the external calibration standard and the sample of interest differs. To validate that 193 nm generated aerosols are less influenced by thee matrix effects liquid calibration standards were applied to quantify solids. It will be shown that 193 nm laser aerosols and desolvated liquids behave similar within the ICP and can therefore be used for quantification.

GEMOC Key Centre, Department of Earth and Planetary Sciences, Macquarie University, NSW 2109, Australia The development of the multi-collector ICP-MS has greatly enhanced the capability for in-situ high precision isotope ratio analysis. Although we can build on experience from laser ablation trace element analysis on single-collector ICPMS, there are additional and complex issues to be considered in LAM-MC-ICPMS. Some are common to solution analysis (eg mass bias) but others are a consequence of the complex matrix of natural samples. Accurate analysis requires rigorous correction procedures for mass bias and isobaric interferences. Instrument sensitivity, elemental concentration and grain size are the main controls in determining laser conditions and ablation times, and thus analytical precision. Signals produced by laser ablation are transient, but operating conditions for the laser can be chosen to achieve near steady-state signals. Frequency, laser fluence and spot size settings all contribute to signal stability and intensity. In the analysis of major elements in common rock-forming minerals (eg Mg in olivine; Fe in pyrite) sensitivity is not a problem and isobaric interferences are negligible. In analyses of the Hf isotopic composition of zircon, corrections for REE interferences are critical. The measurement of isotopic ratios of trace elements presents greater problems; successful applications include Os in mantle sulfide; Sr in cpx, plagioclase and carbonate; Pb in feldspar. In some cases the best precision has been obtained at the expense of spatial resolution or in minerals with very low parent/daughter ratios. Further improvements will come with wider use of multi-ion counting systems; this will require the development of more stable detectors and more complex calibration routines. Laserinduced isotopic fractionation remains poorly documented and understood. It is unclear if fractionation occurs during ablation, during transport or within the plasma due to particle size distribution and/or matrix effects. Compared with SIMS, a much wider range of isotopic systems has been developed on the LA-MC-ICPMS. There is a critical need for isotopically homogeneous reference materials that can be used to verify accuracy and for standardsample bracketing in the measurement of mass-dependent isotopic fractionation.

Goldschmidt Conference Abstracts 2005 Microbeam Isotopic Ratio Measurement


Spatial resolution and the analysis of complex geometries in LA-MC-ICPMS

In-situ single spot analysis of B isotope ratios by laser ablation multiple ion counting ICPMS



School of Earth Sciences, The University of Melbourne, VIC 3010, Australia ([email protected]) In typical LA-MC-ICPMS applications the requirement for large signals (ideally many volts, measured on Faraday cups), necessitates ablation using relatively large spot sizes, and comparatively rapid drill rates. Unfortunately, however, the size and/or geometry of many analytical materials precludes the use of such conditions. In such circumstances a number of alternative analytical approaches can be contemplated using a combination of appropriate ablation systems and time-resolved analysis (TRA) software. Since ablation pits are generally many tens of microns in diameter and yet individual pulses may ablate only a few tenths of microns at a time, theoretically, depth profiling is capable of providing the best resolution and appears to be a viable technique as long as pit aspect ratios do not greatly exceed 2:1 (depth:width). In many circumstances, however, (e.g. relatively large yet thin samples) depth profiling is not feasible and, in these cases line scans must be considered as an alternative analytical strategy. In this case excimer lasers offer the potential for ablation of a ‘slit’ rather than a simple spot, thus maintaining high resolution while still allowing ablation of sufficient material for analysis. Finally, for samples with complex zonation, one further solution is to lower the spot size, increase the repetition rate, and ablate along a pre-digitised path using an appropriate translation speed to achieve optimum signal intensity. In all these cases, adequate TRA software is essential to the task and yet, unfortunately, remains an area where individual users are often required to implement their own solutions. Examples will be shown of all these approaches in the analysis of a range of common geological and zoological materials.


Thermo Electron Bremen GmbH, Germany ([email protected], [email protected]) 2 C.N.R.- Istituto di Geoscienze e Georisorse- Sezione di Pavia, Italy ([email protected]) 3 Dipartimento di Scienze della Terra, Università di Pavia, Pavia, Italy ([email protected]) We have developed a method for the in-situ single spot B isotopic analysis of geological materials using laser ablation multicollector ICPMS. A New Wave UP213 laser was coupled to a Finnigan Neptune equipped with both Faraday and ion counting detectors. Samples with different B contents and isotopic compositions have been analysed, including B4 tourmaline and three MPI-DING glasses (StHs6/80-G, GOR132-G and GOR128-G). Before firing the laser, the mass spectrometer was tuned and the ion counters were cross-calibrated by a peak jumping routine, using a very diluted B solution. Spot sizes varied between 60 and 80 µm and the laser energy ranged between 5 to 20 J/cm2. The analysis run consisted of 30-40 cycles (each 1 s). The B signals were corrected for the B background (typically ~900 cps 11B), measured before the start of laser firing. To correct for fractionation effects the standard-sample bracketing approach was applied using NIST SRM 610 as external standard. The corrected 11B/10B is finally referenced to NIST SRM 951) in order to obtain the delta notation. The B4 tourmaline, containing up to 31400 ppm B, was measured using Faraday detectors, with internal precisions (on single spot analyses) better than 0.1‰ (1σ). The weighted average δ11B is -8.3±0.15‰ (1σ). The MPI-DING glasses, with B contents between 11 and 23 ppm, were measured on multiple ion counters. They have δ11B values of -4.3±2.4‰ (StHs6/80-G), +6.8±3.0‰ (GOR132-G) and +13.5±1.6‰ (GOR128-G). Within-run precisions are between 1.6 and 3.2‰, which are very close to the theoretically expected uncertainties based on counting statistics (~2‰).


Goldschmidt Conference Abstracts 2005 Microbeam Isotopic Ratio Measurement

Microanalysis of oxygen isotopes JOHN W. VALLEY AND NORIKO KITA Dept. of Geology, Univ. Wisconsin, Madison, WI, 53706 USA ([email protected]) The preferred techniques for microanalysis of oxygen isotope ratios are laser fluorination/gas-source massspectrometry and ion microprobe/SIMS. These techniques are complementary. IR-laser fluorination of mg-size chips or powder provide the most accurate and precise analyses (~±0.05‰, 1SD). From 2 to 5 analyses/hour are routine and, unlike with Ni-reaction vessels, all components of the reaction line are evaluated by analysis of standards. Analysis at 50 µg scale requires transfer of O2 to avoid fractionation during conversion to CO2 and has less precision. In situ analysis by UV-laser carries the same trade-off of precision to sample size. Spatial resolution of 300 µm (~50 µg) is possible at 0.1-0.2‰. However, in situ analysis requires that the entire sample is exposed to reagent, and room temperature fluorination of grain boundary alteration and other reactive phases that are not in the laser spot can be a problem. Stable isotope analysis by ion microprobe attains the smallest sample size and best spatial resolution. Instrumental mass-fractionation varies with instrument and mineral. Careful use of appropriate standards, bracketing unknown analyses, allows accuracy to approach precision. Analytical precision is greatly improved by use of multiple Faraday cup or electron multiplier detectors. Under favorable conditions, reproducible precision of 0.1-0.2‰ (1SD, δ18O, spot to spot) has been attained for 10-20 µm spots (~1 ng) in many silicate minerals (5-20 min./analysis) in at least four labs by Cameca ims-1270 with dual Faradays. Linear spatial resolution can be better than 1 µm in depth profile mode. Spot size of 0.3 µm is possible and, even at poorer precision, is useful for isotopically labelled experiments. Design improvements to the Wisconsin Cameca ims-1280 are intended to enhance reliability and precision of analysis, including: digital electronics; PC control; continuous monitoring of primary beam intensity; automatic focusing of the secondary beam in the field aperture and entrance slit, and of the objective lens; and better control of stray magnetic fields in the sample chamber and secondary column. Ion microprobe analysis is required for samples that are zoned, precious, or rare. Published studies include coesite inclusions in diamond, zoned zircons, meteorites, quartz overgrowths, and speleothems. Future studies will explore new applications including biological targets. Most traditional and non-traditional stable isotope systems will enjoy the same advantages of small spot size and enhanced accuracy by multiple collector ion microprobe.

In situ-Fe isotope determination using femtosecond LA-MC-ICP-MS G. STEINHOEFEL, I. HORN, R. SCHOENBERG, AND F. VON BLANCKENBURG Univ. of Hannover, Inst. for Mineralogy, 30167 Hannover, Germany ([email protected]) The in situ determination of stable iron isotopes by laser ablation coupled to multiple collector ICP-MS requires a precision of better than 100 ppm if the small variations expected in minerals shall be resolved. However, nanosecond laser systems that are common in geochemical laboratories are likely to produce an instrumental mass bias that is notoriously difficult to control and correct for. We expect that femtosecond lasers operated in the UV range produce ablation conditions that are by far superior in terms of aerosol size, stability, and ionisation efficiency. Stable Fe isotope data of a variety of different matrices measured with our in-house build femtosecond laser ablation system are compared to standard solution ICP-MS measurements after chemical purification (Schoenberg & von Blanckenburg 2005) in order to evaluate the potential and advantages over traditionally employed nanosecond laser ablation. A pure Fe-metal standard (Puratronic) has been calibrated versus IRMM-14 yielding a δ56Fe of +0.085 ± 0.03‰ in solution. Using laser ablation we obtained a δ56Fe of –0.06 ± 0.04‰ (2σ) for this pair. We used this metal as a bracketing laser standard for all types of matrices including the iron meteorite phases taenite and kemacite, and the mineral phases hematite, siderite, goethite and biotite. A short term precison of 0.057‰ has been achieved on iron meteorites when bracketed against this metal. Without a matrix match (e.g. bracketing an oxide, carbonate, or silicate against a metal) no significant degradation of either precison nor accurracy could be observed. Goethite yields an average δ56Fe of -0.515‰ (LA) which compares well to the average value of -0.510‰ determined conventially from solution. Siderite shows a δ56Fe of -0.838‰ (LA) which again compares well to the δ56Fe value of –0.854‰ determined conventially. However many of the low-temperature minerals analysed so far show significant zonations or large-scale inhomogeneities which can now be resolved by using femtosecond laser ablation. This provides us with a spatial resolution that adds a further dimension to our interpretation of stable metal isotope fractionation.

Reference Schoenberg, R. and von Blanckenburg, F, (2005) Int. J. Mass Spec. in press.

Goldschmidt Conference Abstracts 2005 Microbeam Isotopic Ratio Measurement


Low-level 187Os/188Os analysis by laser ablation, multi-ion-counting ICPMS

In-situ analysis of Os and Pb isotope ratios using laser ablation and collision-cell quadrupole ICP-MS




WHOI, Woods Hole, MA 02543, USA, ([email protected]) 2 WHOI, Woods Hole, MA 02543, USA, ([email protected]) 3 Thermo Electron (Bremen) GmbH, Germany ([email protected]) 4 Thermo Electron (Bremen) GmbH, Germany (j[email protected]) Precise 187Os/188Os analysis of <0.1 ng Os traditionally has been achieved by negative thermal ionization mass spectrometry (N-TIMS). Though N-TIMS remains the most precise method for Os isotope analyses, its time-consuming nature makes generating large numbers of precise Os isotope data at low concentration levels expensive and timeconsuming. Some applications, particularly in lowtemperature geochemistry, require large quantities of highquality data. For instance, the short marine residence time of Os causes variations in 187Os/188Os on time-scales of a few thousand years. Consequently, >10,000 data points are required to fully characterize the marine Os isotope record of the Cenozoic alone. We present a new method for analyzing small quantities (>5 pg) of Os by multiple ion counting ICPMS. In a multidynamic measurement procedure Os isotopes are simultaneously detected using continuous dynode electron multipliers on a Finnigan NEPTUNE multicollector ICPMS. Osmium is separated and purified by conventional methods (sparging or micro-distillation) and loaded with a micropipette in small (sub-mm) depressions onto a carrier substrate. After drying on a hotplate at low temperature, the substrate is placed in a laser ablation cell and ablated with a 213 nm laser beam operated in pre-programmed raster mode. Each analyte spot is ablated in less than one minute, causing a transient signal in the ion counters of several hundred cps on isotope 187 Os. The use of a transient signal is advantageous for very small amounts of analyte as it improves the signal to noise ratio and allows for an internal precision in 187Os/188Os of 12%, even at total analyte amounts of ~5 pg. This is more than an order of magnitude better than with conventional sparging using a single-collector Finnigan ELEMENT2. The method also does not suffer from washout problems. We are currently working towards refining this method for the routine analysis of low-level samples (e.g., natural waters).

Science de la Terre, Universite du Quebec a Chicoutimi, Chicoutimi, Quebec, Canada. ([email protected])

Introduction The use of a conventional (quadrupol) ICP-MS with laser ablation for precise, in-situ Os and Pb isotope measurements have been investigated. A hexapol collision cell, with a small gas flow (<2ml/min) has been used in order to improve accuracy and precision. The collision cell produces effects known as collisional focusing, which improves sensitivty, and collisional damping, which improves isotope ratio precision. By using both collisional focusing and damping, in-situ variations in Os and Pb isotope ratios in magmatic minerals have been measured.

Method Description By introducing a low gas flow into the hexapol collision cell of an ICP-MS, ion energy spread is reduced. More ions are focused through the quadrupol leading to higher transmission and thus, improved sensitivity (collisional focusing). Collisions between ions and the collision cell gas reduces the kinetic energy of the ions. Each ion packet spends longer in trasmission (collisional damping). Adjusting the hexapol and quadrupol settings to approximately match the rate of ion transmittion leads to improved isotope ratio precision. Laser conditions were optimized by adjusting laser frequecy, power, spot size and using various raster patterns or static spots. Sulfides were calibated using the NiS reference material PGE-A for Os-isotopes. NiS beads, spiked with NIST SRM 981 and 982 Pb-isotope reference materials, were used to calibrate for Pb-isotopes. For silicates, the Pb-isotope ratios of NIST 610 and 612 glasses were used for calibration.

Results and Applications

Precision for 187Os/188Os and Pb-isotope ratios in sulfides varies with Os- and Pb-contents. Errors range from 0.25%-2% (2σ). Errors for Pb-isotopes in silicates are also related to Pbcontents, ranging from 0.3%-2.5% (2σ). Mineral-scale records of crustal contamination and assimiliation in sulfide and silicate phases were examined from the Duluth, Bushveld and Muskox Intrusive complexes. Sulfides from the Duluth Complex show a wide range of Os- and Pb-isotope ratios. The variations recorded are consistent with a model of sulfur sequestration from crustal sources during magma emplacement.


Goldschmidt Conference Abstracts 2005 Microbeam Isotopic Ratio Measurement

In-situ analysis of Pb isotope ratios by LA-MC-ICP-MS: Applications to ore genesis and igneous petrogenesis ADAM J.R. KENT AND JOHN H. DILLES Department of Geosciences, Oregon State University, Corvallis, OR 97330. ([email protected]) UV laser ablation systems and sector field ICP-MS instruments, particularly those with multi-collection capacity, allow rapid in-situ measurement of isotopic ratios in small sample volumes. We report here details of in-situ Pb isotope measurement technique using a 193 nm ArF Excimer laser and NuPlasma MC-ICP-MS instrument and discuss applications to studies of igneous petrogenesis and ore genesis. Pb isotope ratios can be measured in materials with Pb contents as low as a few ppm. Limitations on analytical precision and accuracy derive from the mass bias correction, instrumental drift and isobaric interference of 204Hg on 204Pb, however the primary limit of precision is the Pb ion beam intensity. This is largely a function of Pb contents and the analytical volume available for analysis and thus may vary considerably between applications. Mass bias is monitored via external normalization based on analysis of standard glasses, and instrumental drift rates require analysis of standards only every ~20 minutes. 204Hg background is relatively low, typically contributing < 1% of the total 204Pb signal. Overall this protocol enables accurate and reproducible analyses of basaltic and rhyolitic glasses, silicate groundmass, plagioclase and alkali feldspar, and sulfide phases. Ratios of 208Pb-207Pb206 Pb can be measured to internal and external precisions of better than 0.1-0.5% at total Pb intensities as low as 5-10 mV, however ratios involving 204Pb require either use of on an electron multiplier, or if using a Faraday cup, total Pb beam intensities of >100 mV. Applications of this technique to Mount St Helens dacite and the Bushveldt intrusion are described in other presentations. Analysis of hydrothermal and igneous feldspars from the Butte mining district, Montana constrain sources of Pb and other ore elements. Pb isotope compositions of hydrothermal K-feldspar from main stage ore veins suggest that both the porphyry and main stage hydrothermal systems introduced hydrothermal Pb that cannot be derived entirely from the host Butte quartz-monzonite. This interpretation is consistent with the presence of abundant inherited zircons of Proterozoic age, also found in the mineralizing porphyries.

In situ Sr isotopes measured by LA-MC-ICPMS: Utility for the average Joe FRANK C. RAMOS1 AND JOHN A. WOLFF2 1

Dept. of Geological Sciences, Central Washington University, Ellensburg, WA, 98926, USA ([email protected]) 2 Dept. of Geology, Washington State University, Pullman, WA, 99164, USA ([email protected]) In situ analysis of Sr isotopes measured by laser ablation multi-collector inductively coupled plasma mass spectrometry (LA-MC-ICPMS) holds great promise for a broad range of scientific fields from igneous petrogenesis to fish ecology. Although the potential for such analyses has been apparent from at least the early 1990’s, it has only been in the last few years that this technique has become refined to the point to have wider applicability, mainly as a result of technological advances in both inductively coupled mass spectrometry and lasers. To date however, most applications have focused on minerals with abnormally high Sr concentrations, giving researchers the inaccurate impression that common minerals with normal Sr contents cannot be successfully analyzed. In fact, minerals from typical alkalic and tholeiitic basalts can be successfully targeted, usually with adequate measurement precisions to evaluate and constrain open-system effects and mineral residence times. For Sr isotopes, accuracy of analyses has proven to be critically dependent on successfully accounting for the interfering effects of both elements and molecules such as Ca-dimers/Ca-argides, Fe-oxides, Rb, Kr, doubly positive Er and Yb, and potentially others. Current literature suggests the presence of such interfering elements are not only material dependent but also machine dependent in regard to both manufacturer and individual machines. These traits combine to make accurate measurement of Sr isotopes very difficult to obtain in situ. Great strides have been made however, and we can now accurately measure Sr isotopes on a variety of complex materials. We present an overview of current successes, analytical challenges, and future directions for Sr isotopes obtained by LA-MC-ICPMS. We incorporate results for carbonate fish otoliths and typical volcanic rock phenocrysts including basaltic and dacitic plagioclase, alkalic and tholeiitic clinopyroxene, and basaltic, basaltic andesite, and andesitic groundmass analyzed by LA-MC-ICPMS.

Goldschmidt Conference Abstracts 2005 Microbeam Isotopic Ratio Measurement

Ion microprobe U-series dating and cathodoluminescence of secondary opal at Yucca Mountain, Nevada J.B. PACES1, L.A. NEYMARK1 AND J.L. WOODEN2 1

U.S. Geological Survey, Denver, CO, USA ([email protected]), ([email protected]) 2 U.S. Geological Survey, Menlo Park, CA, USA ([email protected]) Understanding past unsaturated-zone (UZ) hydrology in the proposed high-level radioactive waste repository at Yucca Mountain, Nevada, is important for understanding responses to future climate change. Secondary opal and calcite on floors of natural cavities record water flow through fractured tuff over the past 12.8-m.y. Because of the very slow mineral growth rates (µm/k.y.), sampling and dating methods must use the finest possible spatial resolution. Opal with tens to hundreds of µg/g U and negligible Th was analyzed by sensitive high-resolution ion microprobe (SHRIMP) using a 30-µm-diameter primary O- beam. Precision for 230Th/238U and 234U/238U data depends on U content and age, and ranges from 5 to 25% (2σ) due to small 230Th16O and 234U ion-beam intensities. Despite these analytical errors, 230Th/U and model 234 U/238U ages more accurately represent timing of deposition than ages obtained from milligram-sized samples by highprecision thermal ionization mass spectrometry. U-series ages for profiles across individual opal hemispheres increase from 20 to 50 ka for outermost layers to >1,200 ka for inner layers. Depth-age relations yield growth rates from 0.2 to 3 µm/k.y. for different samples. Although growth rates change abruptly in some profiles, these changes are not cyclic as expected for climate-induced variations in fluid flux. Instead, growth rates are remarkably uniform for the outer 200 to 400 µm of individual hemispheres, which spans the last several climate cycles. Depth-age relations for opal intergrown in the outer parts of calcite-rich coatings indicate similar µm/k.y. growth rates for U-poor calcite. Oscillatory zoning in opal is present in cathodoluminescence (CL) images. CL intensity is positively correlated (R2=0.95) with 100-fold differences in U concentration between lightest and darkest zones. Patterns of CL cyclicity and intensity correspond to ~100-k.y. climate cycles in some profiles; however, evidence for correlations in CL patterns between different samples is not obvious. Causes for cyclic U variation within single profiles may be related to variable U concentration in infiltration or differences in the degree of subsurface evaporative concentration at different times. These CL data, along with the age framework and growth rates provided by opal SHRIMP dating, help elucidate the hydrologic response to climate change at Yucca Mountain.


In situ U-series dating by laser-ablation MC-ICPMS S. EGGINS1, R. GRÜN1, M.T. MCCULLOCH1, A.W.G. PIKE2, J. CHAPPELL1 AND L. KINSLEY1 1

Research School of Earth Sciences, The Australian National University, Canberra ACT 0200, Australia ([email protected], [email protected], [email protected], [email protected], [email protected]) 2 Dept. of Archaeology and Anthropology, University of Bristol, 43 Woodland Road Bristol BS8 1UU, UK ([email protected]) The capabilities and potential applications of in-situ dating of Quaternary materials using laser ablation-MC-ICPMS are explored. 234U/238U and 230Th/234U can be measured with precision sufficient for dating at a spatial resolution of better than 100 µm in samples that contain as a little as 1 ppm uranium. Moreover, U and Th concentrations and U-series isotope ratios can be continuously profiled to determine changes in age that occur with sample growth (e.g. in speleothems). These capabilities further permit the dating of bones, teeth and possibly molluscs, which are subject to postmortem open-system behaviour of U-series isotopes (cf. Pike et al., 2002). They can also be employed to elucidate processes of U-series migration during weathering and diagenesis. A drawback of laser ablation-MC-ICPMS is that it cannot in general provide U-series age estimates with precision equivalent to conventional TIMS or solution MC-ICPMS methods. This reflects, in part, the inability to reproduce with very high precision the elemental Th/U ratio due to fractionation processes that occur at the site of ablation and within the Ar-ICP. However, sample preparation is straightforward, the amount of sample consumed negligible, and the technique may be used to rapidly characterise and/or screen and select samples from which more precise and accurate dates can be obtained using conventional methods. Given further instrumental developments and the establishment of suitable matrix-matched standards for carbonates and other materials, laser ablation-MC-ICPMS will play an increasingly important role in Quaternary dating and research.

Reference Pike, A.W.G., Hedges, R.E.M., and Van Calsteren, P., 2002. Geochim Cosmochim Acta 66, 4273-4286.


Goldschmidt Conference Abstracts 2005 Microbeam Isotopic Ratio Measurement

In situ U-Pb zircon dating using LA-MC-ICPMS and a multi-ion counting system A. COCHERIE, M. ROBERT AND C. GUERROT BRGM, Orléans, France ([email protected]) LA-MC-ICPMS allows the rapid and precise in situ determination of isotope ratios in minerals. The aim of this work was to date zircons using the U-Pb isotope system. In order to achieve this, it was necessary to obtain 207 Pb/206Pb and 206Pb/238U ratios with a precision of around 2% and 5% respectively (95% confidence limit). This rapid method allows a relatively large number of individual data to be collected, which enables statistics on ages suitable for sound geochronological investigation. The currently available 213 nm UV laser, which uses ablation in He, provides a more controlled ablation and smaller particles than lasers of longer wavelength. Because the laser ablation process produces rather erratic transient signals, it is necessary to collect the data over a short period of time and in static mode. In addition, a soft ablation at high spatial resolution (20 µm) produces a limited amount of ionised particles. For these reasons, multi-collection is required. Furthermore, for young zircons containing a low amount of Pb (e.g. 10 ppm), Faraday detectors are insufficient and a multi-ion counting system is thus required. The dispersion of the variable multi-collector array is specifically designed to simultaneously measure 204Pb, 206 Pb, 207Pb, 208Pb masses on four ion counters, while 238U ions can be collected either with a Faraday cup or an attached ion counter. In addition, at low mass levels, another ion counter is dedicated to simultaneous 202Hg measurement. We corrected for 204Hg originated from gas and mass discrimination of the actual mass spectrometer. Then, after common-Pb correction, the resulting radiogenic 207Pb*/206Pb* ratio is quite close to the expected value for the zircon standard. This means that the actual laser system does not significantly fractionate Pb/Pb ratios. In contrast, the precision on U/Pb determination is constrained by the contrasting volatile properties of U and Pb and any subsequent ablationrelated fractionation. An analytical procedure is proposed that provides reliable ages and results on reference zircons already dated using ID-TIMS or SIMS.

Characterization of Nd, Te and U isotope ratios in UO2 using SIMS M. FAYEK1, C.S. PALENIK2 AND R.C. EWING2 1

Oak Ridge National Laboratory, PO Box 2008, Oak Ridge TN 37831, USA ([email protected]) 2 Dept. of Geological Science, Univ. of Michigan, Ann Arbor, MI 48109, USA ([email protected], [email protected]) Isotopes provided the initial evidence for the 2 Ga naturally occurring fission reactors discovered at Oklo and Okelobondo in Gabon, Africa and have been subsequently applied to characterize reactor operation conditions and the migration of actinide and fission product elements. However, the application of SIMS has been largely limited due to analytical challenges. Uraninite (UO2), which forms the core of these natural reactor zones represents an extremely complex matrix, containing measurable concentrations of most elements between 70
Reactor Zone Range 0.00543-0.00722 0.859-0.978 0.591-0.658 0.516-0.589 0.107-0.244 3.965-5.387

Here we present the use of SIMS to measure isotopic ratios of Nd, Te and U isotopes (Table). Elements and isotopes were selected based on instrumental considerations (e.g., interferences and relative abundances) and for their sensitivity to reactor operation conditions. To minimize compound ion interferences, extreme energy filtering (100 - 200V) was used with O- (for U and Nd) and Cs+ (for Te) primary beams. Instrumental mass fractionation was corrected by measuring reactor zone uraninite previously characterized by TIMS or ICP-MS. Analysis of samples distributed across the Okelobondo reactor zone provided insights into the heterogeneity of isotopic ratios at various length scales (µm to m). In conjunction with computer simulations of reactor operation, these isotopic ratios were used to constrain the reactor operating conditions.

Goldschmidt Conference Abstracts 2005 Microbeam Isotopic Ratio Measurement


Laser ablation ICP-MS dating of complex magmatic and metamorphic zircon

Time of flight secondary ion mass spectrometry (ToF-SIMS) use in sedimentary geochemistry



Department of Mineralogy, JW Goethe-University, Frankfurt, Germany ([email protected]) Zircon belongs to the key minerals for unravelling many processes during earth history. The U-Th-Pb systematic in most zircon grains is complex due to alteration processes, such as dissolution, recystallization and new zircon growth. A spatial resolution of 10 to 40 µm are commonly needed to resolve the different age pattern in complex grains. Only few LA-ICP-MS studies have shown so far that they are capable to analyse 30 µm areas and precisely detect low Pb contents, which are relative common e.g. for younger and detrital zircon grains. The U-Th-Pb zircon dating technique was developed in our lab using a 213 Merkantek laser systems attached to a Thermo-Finnigan Element II. A spot-size of 25–30 µm is used as the best compromise between spatial resolution and acceptable internal precision of the 207Pb/206Pb and 206Pb/238U ratios. A teardrop-shaped, low volume laser cell with a washout time below 1 s (Horstwood et al., 2003) enables precise detection of a sampled material, which is heterogeneous with time. A relative simple approach is used to correct for within-run U-Pb fractionation before normalisation to a repeatedly analysed reference zircon. The typical internal precision of 206Pb/238U is 0.6 % (1 s.e.). Frequent analyses of zircon with known (GJ-1, 91500, CS3 and Temora) and unknown U–Pb ages using 91500 or GJ-1 zircon as references show that an external precision (1σ) and an accuracy of 1–2 % can be achieved routinely for crack-less grains with a homogeneous U–Pb composition. This is comparable to or even better as precision reported by other laboratories. All errors are propagated with the reproducibility of the standard over the session taken into account. A common lead correction is usually not possible or necessary as total counts of 204(Hg + Pb) were similar (± 10 %) to the 204Hg counts. Applying such a correction to individual ratios would have increased the 207Pb/206Pb error 2–10 times. The main controlling factor for the error is counting statistic in case of the 207Pb/206Pb and the reproducibility of the ablation process and sample heterogeneity for the 206Pb/238U. We will present data of different studies currently completed. Analysed grains spanning a wide range of magmatic, metamorphic and detrital zircon, demonstrating their natural complexity. The results of the studies are encouraging and illustrate the ability but also the limits for LA ICP-MS dating of complex zircon.


CAMCOR, University of Oregon, Eugene, OR 97403 Dept. of Geological Sciences, University of Oregon, Eugene, OR 97403 3 Kaaterskill Exploration, 735 Northwood Loop, Prescott, AZ 86303 ([email protected]) 4 CAMCOR, University of Oregon, Eugene, OR 97403 5 Dept. of Geological Sciences, Indiana University, 1001 E. 10th St., Bloomington, IN 47405 2

A relatively new instrument, Time of Flight Secondary Ion Mass Spectrometry (ToF-SIMS) has recently become available to the sedimentary geochemical community. ToFSIMS is basically a mass spectrometer capable of acquiring microanalytical data on solid surface samples at the micron or submicron levels. It provides simultaneous detection of all elemental, isotopic and chemical element abundances via molecular and cluster ions. The device has high lateral resolution (<50nm), high surface sensitivity (>1 nm) and high depth resolution (> 1 nm). Detection limits are ppm of a monolayer for elements and sub-fmol for molecules. We studied elements and isotopes across a black shale with layers composed mainly of FeS2. About 30 elements and isotopes were observed, and we were able to ratio S34/S32. This ratio produced an ion picture, and an intensity vs distance graph; we used a backscattered electron microscope photograph to pinpoint the area being analyzed. This permitted us to follow the ratio across a 50x50 micron area. With further refinement and the growing need for very high resolution in situ chemical analyses of geological materials, this may become the instrument of choice to address many geological problems where textural contex and two and or three dimensional chemical information is critical.


Goldschmidt Conference Abstracts 2005 Microbeam Isotopic Ratio Measurement

Sr isotope studies of melt inclusions by TIMS

Hf isotopes by laser ablation multi-collector ICP-MS: Progress, pitfalls and prognosis




Department of Earth Sciences, University of Durham, Durham, UK. ([email protected], [email protected], g.m.nowell@, and [email protected]) 2 Danish Lithosphere Centre, Copenhagen, DK 3 Department of Geosciences, Oregon State University, Corvallis, OR. ([email protected]) In-situ measurements by SIMS have revealed considerable variation in the Pb isotope compositions of melt inclusions from single magmas (Saal et al. 1998). Variations among such inclusions have important implications for melt aggregation processes and source variations. To examine further the potential for isotopic studies on melt inclusions we focus on Sr isotope compositions of olivine hosted melt inclusions found in the extreme (high) 3He/4He picrites of Vestfirdir (NW Iceland). Our previous study of the major, trace, and volatile element systematics of these melt inclusions and their picritic host lavas suggests that melt inclusions and lavas are related by a combination of accumulation and fractionation of olivine and clinopyroxene. Furthermore, the variations in incompatible trace element ratios within each melt inclusion population reflect either (1) variations in the degree and depth of melting or (2) originate from a heterogeneous mantle source. We aim to test these alternative hypotheses using high precision Sr isotope analyses on the melt inclusions. Melt inclusions hosted in the olivine phenocrysts range in size between 50 and 150 µm and have Sr concentrations (by LA-ICPMS) of 200 to 700 ppm. Such melt inclusions should have total Sr contents of 0.04 to 4ng. Preliminary experiments carried out on whole olivine grains containing visible melt inclusions show the potential for analyzing single grain and possibly single melt inclusions for high precision 87Sr/86Sr (50 to 100 ppm 2-sigma internal errors). Using miniaturized micro-Sr chemistry techniques based around Sr spec resin (Charlier et al., in prep.), this technique gives total procedural blanks as low as 3 pg, enabling the analysis of sub- 0.5 ng samples. Samples are run using a TaF5 activator using a Triton TIMS.

Reference A.E. Saal, S.R. Hart, N. Shimizu, E.H. Hauri, and G.D. Layne (1998), Science Vol. 282 pages 1481-1484.


Department of Earth Sciences, University of Bristol, Bristol BS8 1 RJ, UK ([email protected]) 2 FALW, Vrije Universiteit, 1081HV Amsterdam, The Netherlands ([email protected]) Hafnium isotopes in the accessory mineral zircon are becoming an increasingly popular tool for deciphering the processes of magmatic and crustal evolution. Much of this utility stems from the physical resilience of the host phase, and its capacity to retain chemical and isotopic information that predate a given magmatic or metamorphic episode. The rapidity of analysis that reflects the combination of new generation lasers and plasma-source multi-collector mass spectrometers is another major attraction. Hf has six naturally occurring isotopes, of which five are non-radiogenic, making correction for instrument-induced mass fractionation (IMF) routine. The isotope ratio of geological interest is 176Hf/177Hf, with 176Hf produced by the β- decay of 176Lu with a half-life of 35.9 Gyr. However, the most critical factor in obtaining meaningful 176Hf/177Hf ratios by laser ablation concerns the ability to accurately correct for the isobaric interference of Yb and Lu on 176Hf. The former is the most severe, and may exceed 300 000 ppm, depending on the REE budget of the analysed zircon. Given the chemical complexity of the matrix, molecular species (e.g. oxides) are other potential interferers, and may affect the non-radiogenic Hf isotopes and IMF factors calculated therefrom. This contribution aims to examine the limitations and resolution of the Hf isotope in zircon technique, with particular reference to the veracity of the various methods for interference correction. Data from both homogeneous zircon standards and strongly zoned (age and compositional) detrital and igneous zircons will be reported for this purpose. Some new avenues for extracting isotopic information from zircon by micro-beam methods shall be explored, including concurrent measurement of Lu-Hf and Pb-Pb or U-Pb isotopes during laser ablation, and the correlation of these with other isotopic and chemical indices.

Goldschmidt Conference Abstracts 2005 Microbeam Isotopic Ratio Measurement


In-situ petrographic thin section U-Pb dating of zircon and titanite by laser ablation-MC-ICP-MS

In-situ Fe Isotopic Fractionation Determination by Laser Ablation MC-ICP-MS




Earth & Atmospheric Sciences, Univ. Alberta, Edmonton, AB, Canada (antonio.simonetti; [email protected]; [email protected]; Tom.Chacko@; [email protected]) 2 Alberta Energy and Utilities Board, Alberta Geological Survey, Edmonton, AB, Canada ([email protected]) U-Pb geochronological studies of accessory minerals (zircon, monazite, titanite) using laser ablation-MC-ICP-MS have typically involved mounting carefully selected grains subsequent a labour-intensive crushing and heavy mineral separation procedure. Apart from the time-consuming aspects involved with this mounting procedure, more importantly the textural context of the grains is lost. This in-turn significantly limits the ability to place the geochronological data within the broader framework of the petrogenetic and deformational history of a rock. In contrast, the capability of acquiring in-situ U-Pb dates in petrographic thin section provides the opportunity to directly link age information, deformational fabrics, and pressure-temperature data derived from analysis of mineral grains. The laser ablation (UP213 nm)-MC-ICP-MS (NuPlasma) instrument configuration at the University of Alberta is equipped with a collector block that includes three ion counters and twelve Faraday buckets. This configuration allows for the simultaneous detection of ion signals ranging from mass 238U to 203Tl. Advantages of the multiple ion counting system include the capability to readily measure low Pb ion signals (<1 x 106 counts per second), conducting ablation runs using relatively low laser energy output (~2 J/cm2), and high spatial resolution using predominantly single spot analysis of ≤40 µm. During a typical 30 second analysis, the depth of the laser pit is ~15 µm and corresponds to a small amount of total ablated mass (~90 nanograms) for zircon. An analytical protocol and data reduction scheme were developed that typically yield external reproducibility (2σ) for 206 Pb/238U (and 207Pb/235U) and 207Pb/206Pb values of ca. 3% and ~0.7%, respectively. We have dated and titanite from several petrographic thin sections representing samples of lower crystal xenoliths and plutonic intrusions; for example, a thin section study of the Wathaman Batholith (northern Saskatchewan/Manitoba) yields a concordant age of 1845.4 ± 5 Ma (2σ; n=5 grains) and corroborates available (ID-TIMS) age determinations (~1850 to 1865 Ma).

Laboratory for Planetary Sciences, Tokyo Institute of Technology, O-okayama 2-12-1, Meguro, Tokyo, Japan 152-8551 ([email protected], [email protected]) Stable isotope geochemistry using heavy elements such as Ca, Zn or Fe has become a strong tool to understand the origin or formation sequences of the sample. Among the heavy elements, iron has been receiving a great interest in the past few years to reveal geochemical and biological cycling because of its ubiquitous presence as well as various in chemical form and oxidation state in nature (Anbar 2004 and Beard et al 1999). In the case of isotopic analysis of Fe, multicollector-ICP-MS coupled with solution introduction technique has been widely adopted. However, sample decomposition or chemical separation precedures prior to the analysis were required and therefore, we can not achieve insitu isotopic analysis of Fe based on the solution sample introduction technique. In this study, laser ablation (LA) sample introduction in stead of solution nebulization was used for the isotopic analysis of Fe. Although we first reported the in-situ isotopic data on Fe obtained with LA-MC-ICP-MS technique. However, resulting precision on 56Fe/54Fe was 2-3 ‰ level, which was not high enough to detect the natural variation in Fe isotopic ratios (Hirata and Ohno 2001). In this study, LA system utilizing shorter wavelength (193 nm) was applied to improve the precision of the Fe isotopic ratio measurements. With a stabilizer device, aiming at stabilizing the signal intensity (Tunheng and Hirata 2004), resulting precision of the 56Fe/54Fe and 57Fe/54Fe ratio measurements as 0.3‰ and 0.5‰ (2SD), respectively. The results obtained here clearly demonstrate that LA-MC-ICP-MS technique can promote further application of the stable isotope geochemistry using heavy elements.

References Anbar, A. D. (2004) Earth Planet. Sci. Lett. 217, 223-236. Beard, B. L., Johnson, C. M., Cox, L. Sun, H., Nealson, K. H. Aguilar, C. (1999) Science 285, 1889-1892. Hirata, T. and Ohno, T. (2001) J. Anal. At. Spectrom. 16, 487491. Tunheng, A. and Hirata, T. (2004) J. Anal. At. Spectrom. 19, 932-934.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution


Earth formation: Combining physical models with isotopic and elemental constraints DAVID J. STEVENSON Caltech 150-21, Pasadena, CA 91125, USA ([email protected]) Models of Earth formation have undergone substantial refinement in recent years, although their basic characteristics differ little from the “classic” views of Safronov and Wetherill. In these models, Earth accumulation can be mostly completed in 10 Ma but may extend out over tens of millions of years and the source material for Earth is derived from a fairly wide range of circumsolar distances of condensation or first accumulation. Both of these conclusions may still undergo revision as dynamicists seek a better understanding of the role of small bodies and dissipation in the formation process. But it seems unavoidable that giant impacts occurred and these events are likely to set the stage for lunar formation and the initial state of Earth, including much of the physics that governs core formation, initial core temperature, siderophile abundances in the mantle, possible differentiation of the mantle into a layered structure and formation of Earth’s initial atmosphere and hydrosphere. I will focus here on some recent efforts to understand Hf-W and what his may tell us about core formation (both timing and process). I will also talk about oxygen isotope puzzle: Why is Earth and Moon so similar. I will argue that the Hf-W story does not require early (~10Ma) earth formation because it tells core formation time and this process may have taken place in large part in precursor bodies rather than Earth itself. I will also argue that the oxygen isotopic similarity may be an out come of a giant impact scenario in which the lunar-forming disk equilibrates isotopes with Earth on a timescale ~100 years, even though the Mars-mass projectile might have had a Mars-like isotopic character.

The oxygen isotope composition of the sun T.R. IRELAND, P. HOLDEN AND M.D. NORMAN Planetary Science Institute and Research School of Earth Sciences, The Australian National University, Canberra ACT 0200, Australia ([email protected], [email protected], [email protected]) The oxygen isotope composition of the Sun is one of the outstanding issues in cosmochemistry. Refractory inclusions show up to 6 % enrichment in 16O relative to terrestrial which was originally ascribed to a nucleosynthetic input to the solar system. With the discovery of non-linear fractionation in the Earth’s atmosphere it appears more likely that a chemical mechanism is responsible. Clayton (2002) has proposed that photochemical predissociation and self shielding lead to preferential enrichment of 16O in CO gas, and 17O, 18O-rich water ice. Recently Yurimoto and Kuramoto (2004) have proposed that the 16O anomaly is inherited from the molecular cloud precursor. In both of these models, the Sun will have a composition that is 16O-rich, in this case, at least as 16O-rich as refractory inclusions. We have measured the oxygen isotopic composition of solar wind in lunar metal grains (Ireland et al., 2004). The oxygen implanted in the grains is depleted in 16O by 54 ± 5 ‰ relative to terrestrial oxygen. Rather than having the 16Oenriched composition of refractory inclusions, the Sun is depleted in 16O, and does not match any prior prediction. If the Sun is depleted in 16O, it cannot be reconciled with a solar nebula predissociation model, but can be accomodated in a model of molecular cloud inheritance. Specifically, removal or partial removal of the C16O gas in the star-forming region will move the nebula to a heavier oxygen isotope composition. If molecular cloud inheritance is the source of the oxygen isotopic systematics observed, it has profound implications for our view of the early solar system. Thermal processing in the early solar system simply unmixes the refractory 16O-rich components from the molecular cloud dust. Refractory inclusions cannot be direct solar condensates, rather they must be predominantly remelted residues of dust from the molecular cloud, potentially with recondensation of evaporated material. Fractionation of volatile elements may take place in the star-forming region leading to abundance differences between dust in the accretion disk and in the Sun.

References Clayton R. N. (2002) Nature 415, 860. Yurimoto H. and Kuramoto K. (2004) Science 305, 1763. Ireland T. R. et al. (2004) Lunar Planet. Sci. XXXV, #1448.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution

CO self-shielding and oxygen isotopes in the solar nebula J.R. LYONS IGPP Center for Astrobiology, U. California, Los Angeles; ([email protected]) 1

by Clayton that CO The recent suggestion photodissociation is the source of the oxygen isotope anomaly measured in meteorites2 offers an opportunity to relate a longstanding and fundamental problem in meteoritics to the farultraviolet (FUV) environment in which the solar nebula formed. Presently, three locations for CO photodissociation are being considered in models: the X-point region of the solar nebula1, the surface of the nebula3, and the parent cloud from which the nebula was formed4. The surface disk and parent cloud models require a greatly enhanced FUV radiation field, ~ 1000 times the local interstellar medium (ISM) field, consistent with solar system formation in a star-forming region. A key test of CO photodissociation as the source of anomalous oxygen isotopes in CAIs is that nebular H2O (produced from product oxygen atoms) has the δ17O/δ18O value measured in CAIs. Figure 1 demonstrates that CO photodissociation in the presence of abundant H2 produces nebular H2O with δ17O/δ18O ~ 1.0 (solid curve), similar to measured values, whereas photodissociation of pure CO produces a ratio ~ 1.1 (dotted curve). This suggests that CO self-shielding in an H2-rich environment is in fact a viable mechanism for explaining oxygen isotopes in CAIs. Fig.1 Model δ-values for nebular H2O


no H2 absorption

1000*ISM FUV midplane 30 AU

H2 absorption 0


[δ O - 1.0*δ O](H2O) (per mil)




δ O=1.0*δ O

nebular H2O inferred from meteorites





δ O=.94*δ O -10 -60










δ O(H2O) (per mil)

References [1] Clayton R.N. (2002) LPSC 33rd, #1326. [2] Clayton R.N. et al. (1973), Science 182, 485. [3] Lyons J.R. and Young E.D. (2004), LPSC 35th, #1970. [4] Yurimoto H. and Kuramoto K. (2004), Science 305,1763.


Early solar system timescales FRANK A. PODOSEK Earth & Planetary Scences, Washington University, St. Lois, MO 63130, USA ([email protected]) From our present perspective, in many respects it is useful to think of the formation of the solar system as an event, a singular occurrence that happened a long time ago. The current best estimate for the absolute age of that event is usually taken to be 4.567 Ga [1], but in fact the absolute age of the formation of the solar system has been known for more than a generation with sufficient precision and robustness to satisfy any legtimate curiosity. There is greater interest in viewing the formation of the solar system as a process, a series of events linked by cause and effect, extending over some finite time. The events of interest are chiefly dynamical, but dynamical theory is not yet able to predict a robust complete history. Appeal must therefore be made to an empirical history based on geochemical events caused by the dynamic processes, datable by geochronological techniques based on radioactive decay. For quite some time, the class of refractory-element-rich objects called CAIs has been esteemed the first solid material to have formed in the solar system (e.g. [2]), on the basis of chemical characteristics, stable isotope anomalies that suggest primitiveness, the initial presence of some short-lived (now extinct) radionuclides, and of course on the absence of greater known ages than for any other objects. Typically, the “age of the solar system” means the isotopic closure age of CAIs. There are two other kinds of geochemical events often thought to date solar nebula processes – the formation of chondrules, and the chemical differentiation by which planetary materials were extracted from a source of cosmic composition. Chondrules (like CAIs) were once thought to have formed in the very earliest stages of nebular history, but are now thought to have formed later and over an exended timescale, a few Ma after CAIs. The chemical compositions of the differentiated meteorite parent bodies are also thought to have been formed a few to several Ma after CAIs. Recent results [3] for the 182Hf-182W system are reported to indicate chemical events a few Ma before formation of CAIs. If this conclusion is substantiated, much of the current picture of what happened in the early solar system will have to be rethought.

References [1] Amelin Y. A. et al. (2002) Science 297, 1678-1683. [2] Wood J. A. (2004) GCA 68, 4007-4021. [3] Kleine T. et al. (2005) GCA (submitted)

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution


U-Pb dating of meteoritic perovskite 1



Pb-Pb geochronology of the early solar system





WASM, Curtin University, Western Australia ([email protected]) 2 Geological Survey of Canada, Ottawa ([email protected]) Perovskite (CaTiO3) is a minor but important oxide in CaAl-rich inclusions (CAIs) and is a particularly useful mineral for precise geochronology as it can contain significant concentrations of U and Th. Ireland et al. (1990) investigated U-Pb systematics of perovskite from Allende and Murchison CAIs using SHRIMP and obtained a 207Pb/206Pb age of 4565 ± 34 Ma for Allende. Here we report new TIMS and ion-probe data with significantly improved precision on perovskite grains from an Allende CAI. The CAI, referred to as 134-1, is a large elongate (1.8 cm x 0.9 cm) Compact Type-A inclusion composed of melilite with minor amounts of spinel and perovskite. It also has a very thick (>1 mm) multi-layered rim sequence surrounding the coarse grained interior. Perovskite occurs as small (<50 µm) rounded grains scattered throughout, but is more abundant towards the core. It occurs as both interstitial grains and as inclusions in melilite. Nine (~1µg) fractions of perovskite were separated and analysed via TIMS. Fractions were washed in acid to reduce surface contamination. The residues yielded low 206Pb/204Pb ratios between 43.5 to 167.5. Seven points define a Pb-Pb isochron indicating a 207Pb/206Pb age of 4568.3 ± 2.4 Ma (MSWD = 1.3). This age is consistent with other studies on Allende CAIs. Two analysis are lower than the line determined by the points defining the isochron. One of these two points has the highest proportion of common Pb. U-Pb SHRIMP analysis of in-situ perovskite was investigated using a polished thick section of the same inclusion. U concentration in perovskite was between 3 and 23 ppm. Perovskite analyses yielded raw 206Pb/204Pb ratios between 437 and 6340. The data define a 16-point isochron indicating a 207Pb/206Pb age of 4562.2 ± 5.5 Ma (MSWD = 1.09) and within error of the more precise TIMS results. There is a significant difference in the 206Pb/204Pb ratios between SHRIMP and TIMS analysis indicating that most of the common Pb in the TIMS data is from surface contamination and that there may be no initial Pb present. The data also suggests that there is some scatter in the 207Pb/206Pb ratios although no apparent age differences are resolvable at this level.

Reference Ireland T.R., Compston W., Williams I.S., and Wendt I. (1990), Earth and Planet. Sci. Lett. 101. 379-387.


ENS-Lyon, 46 Allée d'Italie, 69364 LYON Cedex 07, France ([email protected]) 2 Washington State University, Pullman 99164, WA, USA ([email protected]) 3 University of Idaho, Moscow 83844, ID, USA ([email protected]) We report high-precision Pb isotope data by MC-ICP-MS on whole-rocks and chondrule separates from ordinary chondrites (OC). Compared with literature CAI and mineral separate data [1-3], the ages obtained for OC throw new light on the timescale for the formation and cooling of planetary bodies in the early Solar System. All age calculations were done in 207Pb/206Pb-204Pb/206Pb space, which minimizes noise-induced correlations between variables. We then assumed that the sample is a mixture of radiogenic (*) and common lead and used the intercept of the residue-leachate array in the 207Pb/206Pb-204Pb/206Pb plot to estimate the 207Pb*/206Pb* ratios. Full error propagation inclusive of blank contribution (0.01-0.9% of the total Pb contents) required the development of a specific correction scheme. We find that whole-rocks yield ages clearly less reliable than those of chondrules. Chondrules from Ste Marguerite (H4) give an age of 4563.4 ± 0.6 Ma indistinguishable from the phosphate age of 4562.7± 0.7 Ma [2]. These values are also indistinguishable from the age of the chondrules from Nadiabondi (H5) (4562.5 ± 0.9 Ma), whereas the phosphates give a younger age of 4555.6 ± 3.4 Ma [2]. The oldest age is obtained for Forest City (H5, 4567.8± 0.7 Ma). The chondrules from Tuxtuac (LL5) are 4559.9 ± 0.5 Ma old, again older than the phosphates (4543.6 ± 2.1 Ma) [2]. Overall, the ages of L OC are younger than those of H and LL OC. Cooling rates may be estimated using the data of [4] on Pb diffusion in rock-forming minerals: a fast cooling rate (~300°C/My) is inferred for Ste Marguerite, whereas Nadiabondi (~40°C/My) and Ausson (L5; 10°C/My) appear to have cooled much more slowly. The age of Ste Marguerite is therefore the closest to the temperature peak. Assuming that the dominant heat source was 60Fe (T1/2 = 1.5 My), this places the age of the protoplanet at 4567 ± 1 Ma, which is similar to the age of the CAI of [1].

References [1] Amelin et al. (2002) Science 297, 1678-1683 [2] Göpel et al. (1994) EPSL 121, 153-171 [3] Amelin (2000) LPSC 1201 [4] Cherniak et al. (1991) GCA 55, 1663-1673.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution


From dust to planets: Time scales of accretion and differentiation in the early solar system

The formation of the solar system: New constraints from the 247Cm-235U chronometer



Department of Geology, The Field Museum, Chicago, IL 60605, USA ([email protected]) In recent years, several chronometers based on the decay of short-lived radionuclides have been applied towards obtaining high-resolution time constraints for events in the early solar system (e.g. [1]). However, it is still a matter of debate whether many of these extinct radionuclides may be applied as chronometers since their origin and initial distribution in the early solar system are not yet well constrained. Neverthless, several recent studies have demonstrated that high precision absolute (U-Pb) ages obtained for some meteoritic materials are concordant with relative high-resolution ages obtained from the application of the 26Al-26Mg (half life ~0.72 My) and the 53Mn-53Cr (half life ~3.7 My) chronometers [2,3,4]. Specifically, it has been shown that the earliest solids in the solar nebula (i.e., Ca-Alrich inclusions) were formed at 4567.2 ± 0.6 Ma [3]. Furthermore, our recent work on U-Pb, 26Al-26Mg and 53Mn53 Cr systematics in the eucrite Asuka 881394 suggests that some asteroidal bodies had accreted and differentiated within ~3 My of CAIs [4]. We have recently also applied the 182Hf-182W (half life ~9 My) and 146Sm-142Nd (half life ~103 My) systems to the shergottite-nakhlite-chassignite (SNC/martian) meteorites to constrain the timing of metal segregation and silicate differentiation on Mars [5]. This work demonstrates that major silicate differentiation on Mars occurred well within ~50 My of solar system formation. Specifically, the mantle source reservoir of the shergottites was established at 4525 ± 20 Ma, while that of the nakhlites was likely established prior to ~4542 Ma.

References [1] Wadhwa M. and Russell S. S. (2000) In Protostars and Planets IV, pp. 995-1018; and references therein. [2] Zinner E. and Göpel C. (2002) Metoritics Planet. Sci., 37, 1001. [3] Amelin Y., Krot A. N., Hutcheon I. D., and Ulyanov A. A. (2002) Science, 297, 1678-1683. [4] Wadhwa M., Amelin Y., Bogdanovski O., Shukolyukov A., Lugmair G. W. and Janney P. (2005) Lunar Planet. Sci. Conf. XXXVI, #2126. [5] Foley N., Wadhwa M., Borg L., Janney P., Hines R. and Grove T. L. (2004) Geochim. Cosmochim. Acta, submitted.


ETH Zürich, Switzerland ([email protected]) University of Oxford, U.K. ([email protected])


The r-process only nuclide 247Cm decays to 235U with a characteristic half-life of ~16 million years. 247Cm is presently extinct, but offers potential as a short-lived r-process chronometer, providing constraints on the time interval between the last r-process nucleosynthetic event and the formation of the solar system. The existence of “live” 247Cm in the early solar system should be observed today as variations in 238U/235U, provided Cm/U fractionation occurred. The Cm-U system also has a direct bearing on the U-Pb cosmochronometer, which currently assumes no Cm effects in early solar system material. Using a Nu Instruments NuPlasma and new techniques in MC-ICPMS, we are able to resolve variations in 238U/235U at the two epsilon level (2σM) on sample sizes consisting of <20 pg of 235U. Because no long-lived isotope of Cm exists, our study uses Nd as a chemical proxy for Cm. Thus, additional concentration data for 144Nd, 147Sm and 238U were acquired by MC-ICPMS using techniques in isotope dilution. Uranium isotopic measurements and Nd/U values for a suite of bulk meteorites show no well-resolved excursions in 235 U/238U from the terrestrial value at the ~2 epsilon level. These data provide an upper limit on 247Cm/235U at the start of the solar system of 1 x 10-4, assuming Nd is a suitable proxy for Cm and the Nd/U ratios have not been significantly modified (Stirling et al., in press, GCA). We have extended the search for “live” 247Cm in the early solar system to small samples from mineral phases in primitive objects that are likely to display strong Cm-U fractionations. Uranium isotopic measurements have been acquired on acid-etched leachates for a suite of chondritic meteorites, and for a suite of minerals separated from earlyformed carbonaceous chondrites and angrites. Some of these phases show significant 235U excesses with respect to the bulk chondritic value, although no correlation with Nd/U is observed. These data may indeed reflect 247Cm effects. The suitability of Nd as a chemical proxy for Cm, however, may require revision. These new results have important implications for the 247Cm-235U cosmochronometer and the timing of r-process nucleosynthesis relative to the formation of the first solar system materials.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution


The Hf-W isotopic system and the origin of the Earth and Moon

Magnesium isotope composition of chondrites, achondrites and the Earth-Moon system

S.B. JACOBSEN Department of Earth and Planetary Sciences, Harvard University, 20 Oxford St., Cambridge, MA 02138, USA ([email protected]) The Earth has a radiogenic W-isotopic composition compared to chondrites, demonstrating that it formed while 182 Hf (half-life 9 Myr) was extant in the Earth and decaying to 182 W. This implies that the Earth underwent early and rapid accretion and core formation, with most of the accumulation occurring in ~10 Myr, and concluding about 30 Myr after the origin of the solar system. The Hf-W data for lunar samples can be reconciled with a major Moon-forming impact which terminated the terrestrial accretion process ~30 Myr after the origin of the solar system. The suggestion that the proto-Earth to impactor mass ratio was 7:3 and occurred during accretion is inconsistent with the W isotope data. The W isotope data is satisfactorily modeled with a Mars-sized impactor on protoEarth (proto-Earth to impactor ratio of 9:1) to form the Moon at ~30 Myr.


Eidnenössische Technische Hochschule, ETH-Zentrum, NO, 8092 Zürich, Switzerland ([email protected]) 2 Dept. of Earth Sciences, University of Oxford, Parks Road, Oxford, OX1 3PR, UK ([email protected]) We have measured 24Mg, 25Mg, and 26Mg abundances of chondrites, achondrites, lunar and terrestrial rocks and minerals using Nu1700 a high mass resolution MC-ICPMS. Mg-isotopes of terrestrial and meteorite samples fall on a single mass dependent fractionation line. This provides evidence that inner solar system objects were derived from a well mixed reservoir. In contrast to Fe isotopes [1], Mg isotopes show very limited mass-dependent variation. The Mg isotopic composition (δ26Mg) of mantle olivines, enstatites and Cr-diopsides average at -0.07±0.02‰, -0.08 ±0.11‰, and +0.05±0.02‰ (2 s.d.) relative to DSM3, respectively. Based on these data we estimate a δ26Mg between -0.1‰ and 0.0‰ for Earth. Terrestrial volcanic rocks average at δ26Mg=0.15±0.14‰. Eucrites (n=9) and Martian meteorites (n=4) have a mean of −0.06±0.14‰ and -0.12±0.11‰ respectively, overlapping with terrestrial mantle minerals and volcanic rocks. In contrast, 12 analyses of 7 carbonaceous and ordinary chondrites yield an average of -0.35±0.08‰ (2s.d.). Preliminary data for lunar rocks and minerals vary between chondritic values and DSM3 and are therefore slightly heavy relative to terrestrial values. The overall range of δ26Mg for planetary bodies in the inner solar system is <0.5‰. The homogeneity among chondritic meteorites is surprising because large isotopic variations have been reported for Ca,Al-rich inclusions and to a lesser extent chondrules e.g. [2,3]. The small δ26Mg range requires that, whatever produces the oxygen isotope heterogeneity in chondrites involves processes that cause little mass-dependent Mg isotope fractionation. This is consistent with a recently published model of mass-independent fractionation of oxygen at mineral surfaces during condensation of the solar nebula [4] but seems to be in conflict with the self-shielding model [5].

References [1] Zhu et al. (2001) Nature 412, 311. [2] Young and Galy (2003) Rev. Min. 55, 197. [3] Bizzarro et al. (2004) Nature 431, 275. [4] Marcus (2004) J chem Phys 121, 8201. [5] Clayton (2002) Nature 415, 860.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution

Ag isotope variations in the Earth M. SCHÖNBÄCHLER, E.H. HAURI, R.W. CARLSON AND M.F. HORAN Dept. of Terrestrial Magnetism, Carnegie Institution of Washington, 5241 Broad Branch Road, NW, Washington, DC 20015, USA ([email protected]) The short-lived radionuclide 107Pd decays to 107Ag with a half-life of 6.5 Myrs. Pd is more siderophile than Ag and therefore partitions more strongly into the core thereby allowing this system to place chronological constraints on core formation processes. Preliminary results on terrestrial samples revealed Ag isotopic variations on the order of –1.1ε to 5.3ε (relative to the SRM 978a standard) for ore samples and slightly smaller variations for basalts from the Hawaiian volcanoes Kilauea and Mauna Loa (Hauri et al., 1999). Basalts from Mauna Loa have been shown to have radiogenic 186Os, from the decay of 190 Pt, which may be a signature of core-mantle interaction on the Earth (Brandon et al., 1998). These basalts are therefore promising samples to search for Ag isotopic anomalies due to the extinct nuclide 107Pd. However, this task requires precise and accurate Ag isotopic measurements, which has proven difficult because Ag has only two naturally occurring isotopes that can easily experience mass fractionation during chemical processing and analysis (Carlson et al., 2001, Woodland et al. 2003). We have improved the chemical separation procedure in order to obtain a better separation of Ag from Ti. This is essential for basalt analyses because of matrix effects related to Ti during ICP analysis. First results for the Steens Mtn. flood basalt CH83-55 yield an average Ag isotopic composition of -1.6 ± 1.6 ε, indistinguishable from the SRM 978a standard. Analyses of a more comprehensive set of Hawaiian basalts and chondrites are underway. We also reanalyzed the native Ag metals derived from globally distributed ores and previously investigated with a VG Plasma 54-30. The new measurements, performed with an Axiom, confirm the previously reported Ag isotopic variations in the range of –1.1ε to 5.3ε.

References Carlson R.W., Hauri E.H. (2001) Geochim. Cosmochim. Acta 65, 1839-1848 Hauri E.H., Carlson R.W., Bauer J.(1999) LPSC 23, 1812 Woodland S.J. et. al. (2003) In Plasma Source Mass Spectrometry: Applications and Emerging Technologies, pp. 338-350, Royal Society of Chemistry. Brandon A.D., Walker R.J., Morgan, J.W., Norman, M.D., Prichard H.M. (1998) Science 280, 1570-1573.


Timing, mechanisms and conditions of terrestrial planet accretion and early differentiation ALEX N. HALLIDAY Dept Earth Sciences, Oxford University, Parks Road, Oxford, UK ([email protected]) Over the past few years there have been major advances in our understanding of both the timescales and processes of terrestrial planet accretion, primarily as a result of 182Hf-182W chronometry. The accretion of Mars would appear to have been extraordinarily fast, providing evidence of runaway growth. The current best estimates for some martian reservoirs would imply that they formed within the first 1 or 2 million years of the solar system. Mars differentiation appears incomplete and lacks the uniform and well-mixed W depletion found on Earth. Although accretion may have been rapid the timescales for early metal segregation in the source for Nakhla for example could be more than 20 Myrs. In contrast, the final stage of Earth accretion – the Moon-forming collision (or Giant Impact) between the Earth and another planet “Theia”, took place 40 to 50 million years after the start of the solar system. Comparisons between the results of 182Hf-182W chronometry and more traditional systems like 235/238U207/206 Pb indicate that the models used to define these timescales are over-simplified. With the age of the Moon well-constrained the W isotopic data for the Earth provide evidence for minor amounts of incomplete mixing during accretion. For example, the degree of equilibration of W between the core of Theia and the silicate Earth during the Moon-forming Giant Impact could have been as little as ~30%. This cannot explain the discrepancy between Hf-W and U-Pb timescales. A change in U/Pb in the silicate Earth during accretion and differentiation is the best explanation. However, the mechanism for this is unclear.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution


The implications of the Hf and Nd isotopic records for the early history of the silicate Earth J.D. VERVOORT Department of Geology, Washington State University, Pullman, WA, USA 99164 ([email protected]) The Sm-Nd and Lu-Hf isotope systems have been valuable tools in trying to understand the differentiation and evolution of bulk silicate Earth (BSE). Together these systems help constrain evolution of the crust and mantle and provide a basis for quantifying bulk Earth isotopic mass balance. Linking these systems is especially important for constraining the history of the early silicate Earth as they provide independent records in rocks with long and complex tectonothermal histories. Ultimately, the utility of these systems will depend on how well we know the 147Sm and 176Lu decay constants as well as the Lu-Hf and Sm-Nd isotopic composition of BSE. Recent work on the 176Lu decay constant by crosscalibration of U-Pb and Lu-Hf isotope systems on mineral isochrons in terrestrial rocks [1,2] have determined values (1.867 x 10-11y-1) 3-4% lower than the values in use for the last two decades and ~6% lower than the value determined from meteorites [3]. The terrestrial value, if true, would result in predominantly negative initial εHf values for the oldest terrestrial rocks and zircons, indicating they were derived from a source with a prior crustal history. In contrast, a faster decay constant would result in more positive εHf values and indicate derivation from a depleted mantle source. The chondritic LuHf parameters, while still a matter of debate [4], will have little effect on εHf values in the early Earth because plausible increases in the 176Hf/177Hf BSE value are linked to higher 176 Lu/177Hf values, based on Lu-Hf systematics in chondrites. The Nd isotope record for the early Earth is broadly characterized by positive εNd values. Thus, if the terrestrial 176 Lu decay constant is correct, there appears to be a fundamental conflict between the Nd and Hf isotopic records for the Earth’s oldest rocks with Nd indicating a depleted mantle source and Hf a crustal source. In light of large amounts of new Hf data emerging for early Archean rocks and zircons, and the implications of that data for early crust on the Earth, this apparent conflict between the Nd and Hf isotopic records needs to be resolved.

References [1] Scherer, et al., (2001), Science 293, 683-686. [2] Söderlund et al., (2004), EPSL 219, 311-324. [3] Bizzarro et al, (2003), Nature 421, 931-933. [4] Patchett et al., (2004), EPSL, 222, 29-41.

Early terrestrial mantle dynamics from the 143Nd isotopic record of 3600 Ma to >3850 Ma mafic and felsic rocks V.C. BENNETT AND A.P. NUTMAN Research School of Earth Sciences, The Australian National University, Canberra, ACT 0200, Australia Mantle isotopic evolution for 143Nd/144Nd is frequently depicted as linearly increasing from a chondritic primitive mantle composition at 4.56 Ga to modern MORB compositions. This implies a progressive depletion of the mantle and a continuity of rates and processes from ancient to modern Earth. The test of these concepts relies on clarifying and extending the isotopic record by finding the oldest, least altered mantle-derived rocks, providing them with age and geologic context and determining accurate initial compositions. Our continuing field and laboratory investigations have identified the islands near Nuuk, SW Greenland, including Qilangarssuit and Innersuartuut, as having the most extensive record of early (3800-3850 Ma) Earth chemistry. Six oldest orthogneisses, with SHRIMP zircon dates of 3850-3840 Ma provide minimum ages for varied mantle derived mafic/ultramafic rock sequences. Initial εNd(3850Ma) from >30 mafic samples are in a narrow range of +2 to +4 (±0.5), providing a robust control on early mantle depletion. No negative εNd values have been determined. Assuming depletion occurred at 4.56 Ga, these data require a minimum average 147Sm/144Nd=0.224 in the pre-3850 Ma upper mantle. The apparent Nd isotopic homogeneity at 3850 Ma contrasts with the more extreme depletion and large range of positive and negative εHf measured in >4.0 Ga detrital zircons from Western Australia (Harrison et al., this volume). The oldest felsic suites from SW Greenland provide no evidence for pre-existing felsic crust in the form of either inherited zircons, or Nd and Hf isotopic compositions. The emerging global Nd and Hf isotopic patterns reveal a increasingly detailed picture of rapid and extreme early planetary differentiation, likely associated with accretionary processes, followed by partial re-homogenisation of the mantle before 3.9 Ga. For at least the next billion years (3.9 Ga to <2.7 Ga) εNd and εHf isotopic compositions were near constant, indicating rapid crustal recycling timescales (100300 Myr) and/or reservoir mixing. Long held views of simple, linearly evolving mantle depletion need to be revised in recognition of the isotopic record of punctuated mantle evolution.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution


Early differentiation of terrestrial reservoirs and extinct radioactivities

Remains of an enriched Hadean protocrust in modern mantle



Laboratoire de Géochimie et Cosmochimie, Institut de Physique du Globe de Paris, 4 Place Jussieu, 75252 Paris Cedex 05, France ([email protected], [email protected])

Department of Earth Sciences, 6105 Fairchild Hall, Dartmouth College, Hanover NH 03755. ([email protected], [email protected])

The formation of the core, the atmosphere and the isolation of the first proto-crust in the Earth seem to have taken place in the first hundred million years after the beginning of the Solar System. Yet, these timescales of these events are probably overlap and hence, the isotopic systems that are used for the respective chronology are not independent. For example, it has been argued that core formation requires a partially molten mantle and crust formation could result from the late stage of magma ocean crystallization The timescale of the first crustal extraction (4,460±115 Ma) as evidenced by recent measurements of 142Nd anomalies in Isua rocks (Caro et al. 2003), is not very precisely defined but if it predates 4,48 Ga, then this event could have affected Hf-W systematics in the silicate Earth and should be taken into account for assessing the age of core formation. Crustal extraction is also likely to affect I-Xe systematics since I is efficiently extracted during in melts. In this contribution, we examine with a three-box model including accretion (with equilibration) how an early crustal growth would affect Hf-W systematics. An early crustal formation could result in Hf/W fractionation in the crustmantle system and W anomalies detectable in Archean rocks.

Applications of short-lived 146Sm-142Nd (half-life of 103 Ma) and long-lived 147Sm-143Nd (half-life of 106 Ga) coupled isotope systems in early Archaean rocks have indicated that mantle differentiation occurred 50 to 200 Ma after the formation of the earth and produced depleted silicate reservoirs with high Sm/Nd ratios. The nature and fate of the protocrust, with complementary low Sm/Nd ratios, is unknown. Two relevant models are: (1) a stagnant lithospheric lid of the magma ocean consisting dominantly of depleted mafic-ultramafic lithologies. This model proposes that the low Sm/Nd protocrust did not form and therefore no terrestrial sample should display a deficit in 142Nd. (2) Burial of early enriched (low Sm/Nd) crust into the deep mantle. This model suggests that early crust was foundered into the lower mantle leaving behind a depleted upper mantle. If protocrust is preserved in the lower mantle some of the lower mantle derived plume material should possess negative 142Nd anomalies. Here, we present ultraprecise measurements of Nd isotopes in 65 Ma old Deccan Traps lavas, which sample the deep mantle, the asthenospheric mantle and the subcontinental lithospheric mantle. We show that samples representing magma from deep mantle plume give no evidence of the existence of the protocrust at the core-mantle boundary and that the samples with lithospheric component give a negative 142 Nd anomaly with µ142Nd (=[(142Nd/144Nd)meas / 142 ( Nd/144Nd)nNd-β -1)] ×106) values ranging from -10 to -12 ppm. It follows that if the Deccan starting plume was isotopically homogeneous, a portion of Hadean protocrust resided in the ancient Indian subcontinental lithospheric mantle.

Reference G. Caro, B. Bourdor, JL Birck, S. Moorbath, (2003) Nature 6938, 428-431.


Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution

Heterogeneous Hadean hafnium: Evidence of continental crust by 4.5 Ga?

Crystallization temperatures of Hadean zircons: Plate tectonics at 4.35 Ga?




Research School of Earth Sciences, Australian National University, Canberra, A.C.T. 2601, Australia 2 CNRS UMR 5570, Ecole Normale Superieure, 69364 Lyon Cedex 7, France 3 Dept. of Geological Sciences, University of Colorado, Boulder, CO 80309, USA The paradigm for continent formation long favoured by isotope geochemists is that growth began at ~4 Ga and was ~80% of its present mass by 2.5 Ga. This view reflects the absence of a >4 Ga rock record and the systematic post-4 Ga evolution of depleted mantle 143Nd/144Nd and 176Hf/177Hf. The observations of some early Nd and Hf isotopic heterogeneities leave open the possibility of earlier global fractionations and a minority view has persisted that continental crust was widespread in the early Hadean (e.g., Armstrong, 1981; Reymer and Schubert, 1984). In this regard, the relative lack of evidence of earlier depletions (from a magma ocean or continent formation) reflects remixing of these heterogeneities. Detrital zircons from Jack Hills, Australia, with 4.0-4.4 Ga U-Pb ages transcend this ambiguity as they represent pieces of crust that have been sequestered for up to ~4.4 Ga. Zircons have very low Lu/Hf and thus record near initial 176Hf/177Hf at the time given by the U-Pb age. Amelin et al. (1999) used Jack Hills zircons as old as 4.14 Ga to investigate early crustal evolution and inferred the existence of re-melted Hadean crust. We have carried out 176Hf/177Hf analyses by both solution and laser ablation MC-ICPMS on over 70 Jack Hills zircons ranging in age from 3.96 to 4.35 Ga. These results indicate very large positive and negative εHf(T) deviations from CHUR (λ176 = 0.01867/Ga; 176Hf/177Hf = 0.282772; 176Lu/177Hf = 0.0332). Negative values of εHf(T) equivalent to (176Hf/177Hf)4.5 Ga observed between 4.35 and 4.2 Ga are consistent with development of a Lu/Hf = 0 reservoir by 4.5 Ga. Positive εHf(T) deviations in the same age interval imply a depleted reservoir with Lu/Hf up to 0.08. We interpret these results as indicating either 1) that the remains of original mantle heterogeneities from accretion were not yet well mixed by mantle convection by 4.35 Ga, or 2) that significant continental crust had formed by ~4.5 Ga. The latter interpretation is consistent with inferences from ∆142Nd systematics (Caro et al., 2003; Boyet et al., 2003).


Dept. of Earth and Environmental Sciences, Rensselaer Polytechnic Institute, Troy, NY 12180, USA 2 Research School of Earth Sciences, Australian National University, Canberra, A.C.T. 2601, Australia As the sole know survivors from the Hadean Eon, the detrital zircons of Western Australia's Narryer Gneiss Complex have already provided intriguing glimpses into the state of the very early Earth. Given the significance of the chemical and isotopic information these crystals contain, it is crucial to understand the circumstances under which they grew. Knowledge of their crystallization temperatures, in particular, would place useful constraints on the types of magmas being produced during the Hadean. Toward this goal⎯and to provide information on zircons of all ages⎯we have calibrated a new thermometer based on the concentration of Ti in zircons coexisting with rutile, using both high P-T experiments and analyses of natural zircons of wellconstrained provenance. Zircon Ti content varies by three orders of magnitude over the temperature range ~600°– 1450°C, and is insensitive to changes in pressure. In applying this new thermometer to zircons removed from their original context (such as detrital Hadean zircons), coexistence with rutile cannot be assumed. However, thermodynamic considerations show that TiO2 activities generally range from 0.5 to 1.0 in igneous and metamorphic systems capable of crystallizing zircon. That range of TiO2 activities affects calculated temperatures by only a few 10’s of °C. Sixty-eight analyses of 55 Jack Hills zircons from 4.0 to 4.35 Ga in age yield temperatures ranging from 791° to 659°C (710±32°). This range is indistinguishable from that expected of granitoid zircon formation today, and strongly suggests a limited range of regulated mechanisms producing zircon-bearing rocks during the Hadean. Combined with the suite of mineral inclusions observed by previous workers, these temperatures substantiate the existence of wet, minimum-melting conditions within 200 m.y. of solar system formation. They further suggest that by 4.35 Ga, the Earth had settled into a pattern of crust formation, erosion, and sediment recycling similar to that produced during the known era of plate tectonics.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution

A lower age limit for the Archean based on δ18O of detrital zircons

Hf and Nd isotope evolution of lithologies from the 3.8 Ga Nuvvuagittuq Sequence, northern Superior Province, Canada


Dept. of Geology, Univ. of Wisconsin, Madison, WI, USA ([email protected]) 2 Dept. of Applied Geology, Curtin Univ., Perth, Australia A lower boundary for the Archean Eon is proposed based on changes in magma chemistry due to interaction with liquid water at the Earth’s surface. The δ18O of detrital igneous zircons from Jack Hills, Western Australia, record the changing surface environment during the onset of ‘cool early Earth’ conditions. Mildly elevated δ18O for igneous zircon occurs as far back as 4325 Ma, however a later rise of ~1‰ in δ18O to 7.3‰ provides strong evidence for the recycling of supracrustal rocks into magmas, and the first evidence for the presence of liquid water on Earth (Cavosie et al., 2005). This boundary may also signify the stabilization of continental crust, as evidenced by the formation of supracrustal rocks. Rare and scattered geologic occurrences (e.g. 3850 Ma metasediments, 4030 Ma gneisses, or a 4400 Ma detrital zircon) are merely ‘snapshots’ in a fragmentary rock record, and do not define boundary conditions, as some have speculated. Such occurrences, no matter how significant, should not be used to define eon boundaries. The earliest eon (before Cool Earth conditions) encompasses accretion, differentiation, and the early meteorite bombardment of Earth (including Moon formation). The naming of the earliest eon (e.g. Hadean, Priscoan, etc.) requires further international discussion, as chronostratigraphic principles cannot be applied where superposition is unknown due to the lack of a preserved intact rock record. The published δ18O of dated zircons suggest that the onset of cool Earth conditions was at ca. 4200 Ma, which is here proposed as the beginning of the Archean Eon.

References Cavosie et al. (2005) Earth Planet. Sci. Lett. (in review). Cavosie et al. (2004) Precam. Res. 135, 251-279. Gradstein et al. (2004) Episodes, 27(2), 83-100. Valley et al. (2002) Geology, 30(4), 351-354.



GEOTOP, UQAM, P.O. Box 8888, Station Centre-Ville, Montreal, Quebec, Canada, H3C 3P8 ([email protected]) 2 MC-ICP-MS Laboratory, Geological institute, University of Copenhagen, Øster Voldgade 10, DK 1350 Copenhagen K, Denmark ([email protected]) The 3.8 Ga Nuvvuagittuq volcano-sedimentary sequence in the Inukjuak Domain of the Superior Province in northern Quebec lies on the coast of Hudson Bay. The Nuvvuagittuq sequence consists of volcanic and sedimentary units as well as tonalitic gneiss and pegmatites. The volcanic units include mafic amphibolites with associated intermediate volcanic horizons. The sedimentary rocks consist largely of iron formation that vary from iron oxide dominated facies to silicate (quartz-amphibole) dominated facie. One particular silicate facies forms a dominant and persistent horizon that can be followed throughout much of the sequence. The volcanosedimentary units are intruded by layered ultramafic sills that vary from dunite and pyroxenite to gabbro compositions. The bulk of the surpacrustal sequence forms a semi-oval structure that is cored and enveloped by ca 3.6 Ga tonalitic gneisses. The tonalitic gneisses range from granodiorite to tonalite and granite in composition and interleave with amphibolite and ultramafic units along the outer margin of the sequence. The TTG suite is characterised by heavy rare earth element (REE) depletion compared to light REE. Hf and Nd isotope analyses of the tonalites yield negative initial epsilon values suggesting that the tonalites have recycled significant amounts of older ca 3.8 Ga crust. The volcanic units range from basaltic to dacitic in composition with the amphibolites yielding flat to slightly light REE enriched profiles and the dacitic compositions yielding light REE enriched and heavy REE depleted profiles. Hf and Nd isotope data from the volcanic lithologies range from slightly depleted initial epsilon values (+2 for Nd, +3 for Hf) to slightly negative values (-1 for Nd and –4 for Hf). Comparison with geochemical data suggests that the Hf isotope system has been perturbed in the volcanic lithologies whereas Nd has been more robust. Comparison of the Nuvvuagittuq Nd and Hf isotope data with that of West Greenland localities suggests that the mantle source for the Nuvvuagittuq lithologies was less depleted than that of West Greenland.


Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution

Platinum-Osmium isotope evolution of the Earth's mantle

Pt-Re-Os isotope and HSE systematics of 2.8 Ga komatiites




NASA Johnson Space Center, Mail Code KR, Building 31, Houston, TX 77058, USA ([email protected]) 2 Department of Geology, University of Maryland, College Park, MD 20742, USA 3 NHMFL & Dept. of Geological Sciences, Florida State University, Tallahassee, FL 32310, USA Extraction of metal-rich core likely resulted in strong depletion of the silicate portion of Earth in highly siderophile elements (HSE), including Re, Os, and Pt. Replenishment of HSE through a ‘late veneer’, or high-pressure equilibrium between metal and silicate have been proposed to explain the excess of HSE relative to that expected from core extraction at lower pressures. To address these issues on core extraction, late accretion and terrestrial mantle evolution, 30 Os-rich alloys from upper mantle peridotites, and 13 samples of carbonaceous, enstatite and ordinary chondrites, were measured for high-precision Os isotopic compositions on the JSC Triton. Two of the Os-rich alloys with present-day 187 Os/188Os of 0.1095 and 0.1096 have 186Os/188Os of 0.1198320 and 0.1198329, respectively. They have 187Os model ages of 2.8 Ga. The measured 186Os/188Os for these two samples are within uncertainty of the initial ratio of 0.1198323±09 for 2.7 Ga Pyke Hill komatiites and constrains the Pt-Os upper mantle evolution curve. The remaining 28 Os-rich alloys with 187Os/188Os from 0.1167 to 0.1596 have a range in 186Os/188Os from 0.1198352 to 0.1198408, with an average of 0.1198382±29 (2σ). Carbonaceous chondrites have 186Os/188Os of ≤ 0.1198352 (n=3). Ordinary chondrites have 186Os/188Os from 0.1198345 to 0.1198408 (n=7), and enstatite chondrites from 0.1198335 to 0.1198401 (n=3), with averages of 0.1198387±12, and 0.1198374±40, respectively. The enstatite and ordinary chondrite time evolution curves match those for the Pyke Hill komatiites and the 2.8 Ga Osrich alloy mantle sources. These 186Os/188Os relationships between Os-rich alloys and enstatite and ordinary chondrites are consistent with previous arguments based on 187Os/188Os systematics for a late veneer of similar types of materials controlling the HSE budget of the upper mantle. Unless liquid metal/silicate melt partition coefficients for Pt and Os are within about 10% of each other, a high pressure core extraction model will not alone explain these 186Os/188Os compositions.


Department of Geology, University of Maryland, College Park, MD 20742, USA ([email protected]) 2 NASA Johnson Space Center, Mail Code SR, Building 31, Houston, TX 77058, USA 3 NHMFL & Department of Geological Sciences, Florida State University, Tallahassee, FL 32310, USA If coupled 186Os-187Os enrichments in Archean komatiites reflect addition of a core component, the Os isotopic compositions could be used to infer the timing of the onset and the rate of inner core crystallization. Here, Pt-Re-Os isotope and highly siderophile element (HSE) abundance data for 2.8 Ga komatiites atKostomuksha are presented. The Pt-Os data for seven komatiitic samples define an isochron with an age of 2816±190 Ma and an initial 186Os/188Os of 0.1198340±8. Corresponding Re-Os data yield an isochron age of 2880±83 Ma and an initial 187Os/188Os of 0.10916±0.00067. These Os isotopic compositions characterize those of the komatiite mantle source and are 64±17 ppm and 1.8±0.6% more radiogenic, respectively, than those of the contemporary convecting upper mantle or chondritic references. The calculated komatiite source had absolute HSE abundances similar to those of the Abitibi komatiite and an average depleted spinel lherzolite. The coupled 186Os-187Os enrichments in the komatiite source are best explained via derivation of the Os largely from the outer core. If this interpretation is correct, the komatiite data provide minimum constraints on the Os isotopic composition of the outer core at 2.8 Ga. The existing models of core crystallization based on experimentally determined partition coefficients for Pt, Re, and Os between solid metal and liquid metal can adequately explain the Os isotopic composition of the Kostomuksha komatiite source, although require the onset of inner core crystallization at 3.5 Ga at the latest. The results of this study indicate that core-mantle interaction might occur in the form of isotopic exchange without significant mass transfer from the core to the mantle, and that at least some komatiites originated from mantle plumes that rose from the core-mantle boundary.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution

δ37Cl values of the solar system 1


Uncoupled C and S biogeochemical cycling in the Neoproterozoic from the Huqf Supergroup, Oman



Department of Earth & Planetary Sciences, University of New Mexico, Albuquerque, NM ([email protected]) 2 CNRS, CRPG, Vandoeuvre-les-Nancy, France Multiple sodalite grains within altered CAIs in Allende (CV3) were analyzed for their chlorine isotopic composition using in situ ion microprobe analysis. δ37Cl values range from –0.39 to –2.09‰, averaging to –1.33 ± 0.55‰ (n = 10). Isotopic variation within individual CAIs and among the inclusions is within external reproducibility over the analytical session (±0.25‰), indicating that sodalite formed from a homogeneous reservoir. Published δ37Cl values fall in two distinct clusters (Fig. 1).

Fig. 1. δ37Cl values of meteorites. Sodalite in Allende (black box; this study); Water-soluble extracts (circles; Bridges et al., 2004); Bulk carbonaceous chondrites (triangles; Magenheim et al., 1994, 1995); Bulk meteorites (dashed box, Bonifacie et al., 2004); terresterial seawater is 0‰. These clusters have been used to argue for two distinct reservoirs in the early solar system: an isotopically light parent body brine and heavier chondrite silicate solid (Bridges et al., 2004). If a significant portion of the Cl in Allende is hosted in sodalite, then our data contradict the earlier reported ~+4‰ value, a conclusion supported by Bonifacie et al. (2004). We conclude that the solar system average, and by extension, the bulk Earth, have δ37Cl values ≤ –1‰.

References Bridges et al., 2004, Meteor. & Planet. Sci, 657-666. Bonifacie et al., 2004, 14th Goldschmidt Conf., A49. Magenheim et al., 1994, GCA, 3117-3121. Magenheim et al., 1995, EPSL, 427-432.



Dept. Earth, Atmospheric, & Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA, USA ([email protected]) 2 Dept. Geological Sciences, Indiana University, Bloomington, IN, USA Huqf Supergroup strate [~715 – 540 Ma] have been analyzed for organic and carbonate δ13C, as well as δ34S from anhydrite and carbonate-associated sulfate (CAS). The Huqf contains a negative excursion in carbonate δ13C of ~13‰ in the Shuram Formation. This “Shuram” excursion extends through >500m of section, suggesting a long term perturbation to the C cycle. Sulfate δ34S shows little variability throughout the period of the Shuram excursion. There is no indication of a correlation between δ34S and carbonate δ13C during the Shuram excursion, indicating that the oxidation of depleted carbon responsible for the Shuram δ13C excursion was not accompanied by significant oxidation of sulfide. Together, these data are best explained by the oxidation of a vast reservoir of dissolved organic carbon in the deep ocean. Following the Shuram excursion, sulfate δ34S climbs sharply over 6 Myr, reaching a maximum of ~41‰ just before the Precambrian/Cambrian boundary. This is followed by a slight decrease in δ34S in the overlying earliest Cambrian strata. There is no appreciable change in carbonate δ13C coincident with or subsequent to the ~20‰ increase in sulfate δ34S. This observed rate of change in sulfate δ34S (~3‰/Myr) indicates low Neoproterozoic sulfate concentrations. The decline in sulfate δ34S following the Precambrian/Cambrian boundary suggests a progressive increase in the concentration of seawater sulfate. These data indicate a massive Neoproterozoic reorganization of the global C-S biogeochemical cycles, likely due to oxygenation of the deep ocean.


Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution

Was there voluminous ancient (>4.0 Ga) sialic crust? Implications from the Hf composition of detrital zircons R.P. HARTLAUB1, L.M. HEAMAN1, A. SIMONETTI1 2 AND C.O. BÖHM 1

Department of Earth and Atmospheric Sciences, 1-26 ESB, University of Alberta, Edmonton, AB, Canada, T6G 2E3 ([email protected]) 2 Manitoba Industry, Economic Development and Mines, 3601395 Ellice Ave., Winnipeg, MB, Canada, R3G 3P2 We report LA-MC-ICPMS Hf isotope results of ancient detrital zircon from two locations in the Canadian Shield. Quartzite, deposited prior to 2.3 Ga in the Beaverlodge Belt, Rae Province, contains zircon entirely >3.6 Ga old. Metagreywacke from the Assean Lake Area, western Superior Province, was deposited ca. 3.2 Ga and contains predominantly Paleoarchean zircon. The U-Pb and Lu-Hf isotope characteristics from these locations indicate that significant Paleoarchean crust was exposed at the time of sediment deposition, and crustal growth/reworking occurred between 3.7 and 3.86 Ga (Figure 1). Do to the extremely negative εHf of much of the zircon, some of the reworked crust must have been ≥4.0 Ga. However, the only presently known location of ≥4.0 Ga crust, the Acasta Gneiss, Canada, does not have a significant 3.7 to 3.86 Ga component.

Occurrence of a 4.2 Gyr old zircon in the Acasta Gneiss Complex of northwestern Canada TSUYOSHI IIZUKA1, KENJI HORIE2, TSUYOSHI KOMIYA1, SHIGENORI MARUYAMA1, TAKAFUMI HIRATA1, HIROSHI HIDAKA2 AND BRIAN F. WINDLEY3 1

Department of Earth and Planetary Sciences, Tokyo Institute of Technology, Japan ([email protected]) 2 Department of Earth and Planetary Systems Science, Hiroshima University, Japan 3 Department of Geology, The University of Leicester, UK The recognized terrestrial materials more than 4.1 Gyr old have been restricted to detrital zircons within metasediments and a xenocrystic zircon within granitic gneiss in the Narryer Gneiss Complex of Western Australia. Consequently, knowledge of crustal evolution and surface environment in the early Earth is based on mainly the chemical and isotopic signatures of the zircons. However, it is essential for better understanding of them to obtain additional information from other regions. Here we report a further occurrence of a very old zircon grain in the Acasta Gneiss Complex of northwestern Canada, a locality which contains the oldest crustal rocks as old as 4.03 Gyr. The zircon grain was separated from 3.9-Gyr-old tonalitic gneiss and has a xenocrystic core with an U-Pb age of 4,203 ± 58 Myr (Fig. 1). The trace element composition of the core shows that it crystallized from granitoid magma. These data therefore indicates wide presence of continental crust on the Earth by 4.2 Gyr ago and subsequent reworking of them in the early crustal evolution.

Figure 1: 176Hf/177Hf vs Age plot of ancient detrital zircons. Fig.1 Cathodoluminescence image of zircon crystal AC012/ 1-12.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution

A 15N-enriched Archean atmosphere Y. JIA





CSIRO Exploration and Mining, School of Geosciences, Monash University, P.O. Box 28, Victoria 3800 Australia ([email protected]) 2 Department of Geological Sciences, University of Saskatchewan, 114 Science Place, Saskatoon, SK, S7N 5E2 Canada ([email protected])

Introduction The origin and evoltion of nitrogen in Earth’s atmosphere is controversial. Earlier results on Archean cherts and BIF show a range of δ15N from –6 to 30‰. Given temporal and spatial association of chert-BIF with volcanic sequences erupted from mantle plumes, we selected blacks shales distal from chert-BIF to obtain a marine biogenic signature, together with hydrothermal micas from gold provinces, to constrain the secular varistion of δ15N through time.

Results Kerogen in metashales from the 2.7 Ga Sandur belt, E. Dharwar craton, is characterized by δ15N 13.1 ± 1.3‰, and C/N 303 ± 93. Kerogen from 1.7 Ga carbonaceous shales of the Cuddapah basin average 5.0 ± 1.2‰, close to the mode at 3 to 4‰ for kerogen and bulk rock of Phanerozoic sediments. Hydrothermal micas from late-metamorphic quartz-vein systems of the 2.6 Ga Kolar gold province, E. Dharwar craton, that proxy for average crust, are enriched at 14 to 21‰, which are also consistent with other late Archean gold provinces, confirming that the N–budget of the hydrothermal fluids is dominated by sedimentary rocks.

Discussion and Implication Enriched values in Precambrian rocks cannot be caused either by N-isotopic shifts due to metamorphism or Rayleigh fractionation, or by long-term preferential diffusional loss of 14 N. It is possible that the 15N-enriched values stem from a different N-cycle in the Archean, with large biologically mediated fractionations, yet the magnitude of the fractionations between atmospheric N2 and organic nitrogen observed exceeds any presently known. We attribute the 15Nenriched reservoir to a secondary atmosphere derived from CIchondrite-like material and comets with δ15N of +30 to +42‰. Shifts of δ15N to its present atmospheric value of 0‰ can be accounted for by a combination of early growth of the continents with sequestration of atmospheric N2 into crustal rocks, and degassing of mantle N ~ -5‰.

References Beaumont, V., and Robert, F., (1999), Precamb. Res. 96. 6382. Condie, K.C. et al., (2001), Precamb. Res. 106. 239-260. Jia, Y., and Kerrich, R., (2004), Geochem. Geophys. Geosyst. 5. Q07005, dio:10.1029/2004GC000716.


Experimental study of sulfur isotope fractionation associated with pyrite oxidation by H2O2 LILIANA LEFTICARIU, LISA M. PRATT, EDWARD M. RIPLEY AND DAVID L. BISH Department of Geological Sciences, Indiana University, Bloomington, IN, USA ([email protected], [email protected]) Radiolysis of water can accelerate water/rock interaction . . . through production of radicals (e.g., H , HO2 , OH ) and reactive molecules (e.g., H2, H2O2, O2). Radiolytic oxidation can be observed in modern groundwater associated with uranium ore bodies and can be inferred for ancient groundwater. Prior to development of an O2-rich atmosphere on Earth, radiolytically generated oxidants could have reacted with pyrite and provided local sources of partially to fully oxidized sulfur species suitable for microbial metabolism. We evaluated sulfur isotope effects associated with reactions between pyrite and radiolytic oxidants using a series of sealed-quartz-tube experiments run with 60 mg of acidcleaned pyrite, 10 ml of deoxygenated water, and concentrations of H2O2 at micromolar levels. Experiments ranged in temperatures from 4 to 150ºC and had durations from 1 to 10 days. In initial experiments, primary oxidation products were dissolved sulfate, elemental sulfur, iron sulfate minerals, and iron oxyhydroxide minerals. X-ray diffraction patterns and images from scanning electron microscopy reveal anhedral to subhedral hydrated iron sulfates in globular clusters of about 10-30 µm in diameter forming on pyrite surfaces. δ34S remained unchanged for pyrite but showed distinct enrichment of 34S in produced sulphate and elemental sulphur. ∆sulfate-pyrite was 0.5-1 permil and ∆elemental sulphur-pyrite was 1-2 permil. Our results indicate that pyrite oxidation by H2O2 induces greater fractionation that has been recognized in previous studies. Although ∆ values for sulphates and elemental sulphur are not large, compensating depletion of 34S in undetected products could be substantial if the proportional yields are small. Preliminary isotope results from high-temperature experiments indicate that the 34S-depleted fraction might be held in iron sulfate and/or iron oxyhydroxides.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution


Carbon isotope variations of carbon deposits synthesized in the laboratory by arc discharge

Lu-Hf systematics of the earliest crust in Antarctica: The Napier Complex of Enderby Land




Department of Earth and Space Science, Graduate School of Science, Osaka University, Toyonaka, Osaka 560-0043, Japan ([email protected], [email protected]) 2 Department of Geosciences, Osaka City University, 3-3-138, Sugimoto, Sumiyoshi-ku, Osaka 558-8585, Japan ([email protected])

Carbon has many allotropes such as diamond, graphite, fullerene, etc. Such carbon allotropes are important in the noble gas study because they could be good noble gas carriers in meteorites. In fact, the presolar diamond (host phase of HL noble gas component) and Q (host phase of normal noble gas component) are carbon phases. Heymann (1986) also suggested that fullerene could be a noble gas carrieer in carbonaceous chondrites. Thus, it is very important to study the trapping effiency of noble gases in carbon at the syntheses. The isotope effect of carbon itself should be also very important. Thus we newly set up a new apparatus to synsesize the carbon deposit including fullerene under noble gas atmosphere. The method is so-called “arc method” which is widely used to make fullerene. We use graphite as electrodes and apply 0-30V and 0-170A electric power to have an arc dissharge. The graphite of anode is vaporized by the bombarment of electrons, and deposits on the cathode and the surrounding wall as soot under noble gas atmosphere. We collect cabon soot from the wall for our experiment. We made these syntheses under various experimental conditions, and measued carbon isotopic ratios and the trapping efficiency of noble gases in the carbon soots. The interesting result is obtained in carbon isotopes. Our preliminally results suggest that there were about several permil difference of carbon isotopes between the carbon soots deposited at the different place inside the apparatus. We had thought that there should be no isotope difference for the sample that was vaporized at such high temperatures. There might be some isotopic effect at the deposition or during the frying of vaporized carbon. It is likeky that light carbon flies in the long distance inside the apperatus.

Reference Heymann D. (1995) Meteoritics, 30, 436-438.


Dept Geological Sciences, University of Michigan, Ann Arbor, MI 48109-1063, USA ([email protected]) 2 Department of Environmental Changes, Kyushu University 3 School of GeoSciences, University of Edinburgh, Edinburgh EH9 3JW, UK We have conducted Lu-Hf systematics of mineral separates and whole-rock samples from the Napier Complex of the East Antarctic Shield in order to constrain Archean crustal and mantle evolution and to assess the robustness of Lu-Hf isochrons during UHT metamorphism. Evidence that at least portions of the Earth’s mantle were already chemically depleted before formation of the oldest surviving crust comesfrom Sm–Nd isotopic data However, doubts have arisen because the Sm-Nd system can be perturbed by younger events. Our measurements on individual zircon grains have yielded a remarkably uniform range of initial 176Hf/177Hf values between 0.280399 ± 5 and 0.280469 ± 7. Because the samples were collected from different localities and in different rock types, it had been assumed that they would have a wide range in isotopic compositions, indicating a complex history of mantle melting and crustal recycling. Instead, these results now indicate that the source of the crustal materials that formed the Napier Complex at 3.8 Ga were depleted relative to the chondritic uniform reservoir, or CHUR. Moreover, the results demonstrate that even the silicic rocks in the Napier Complex are juvenile products of mantle melting or are remobilized crustal materials recycled on a very short timescale. Measurement of Lu-Hf isotopic compositions for garnet, opx, sapphirine, osmulite, rutile and whole rock samples from several localities in the Napier Complex have yielded isochrons with ages between 2459 ± 23 Ma and 2173 ± 37 Ma with rather uniform initial ratios between 0.280876 and 0.280884. Although the calculated errors are larger than those often obtained from zircon U-Pb dating, it should be noted that they are only slightly large than one percent. Besides producing isochrons, one major contribution of this work has been demonstration of what happens to the Lu-Hf system at the mineral and whole-rock scale during ultra high temperature (UHT) metamorphism. The main rock-forming minerals record an initial Hf ratio acquired during metamorphism while zircons “see through” this event and record the initial ratio of the rocks acquired at the time of crystallization.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution

Rare earth elements in the core?

NANO-SIMS U-Pb dating of monazite

YU PUSHKAREV Institute of Precambrian Geology and Geochronology, Emb. Makarova 2, St-Petersburg, 199034, Russia ([email protected])

Introduction It was traditionally assumed that all REE are in the silicate geosphere only and that in Sm-Nd system the Bulk Earth corresponding to chondrites is identical to the Bulk Silicate Earth (BSE). But these assumption have never been proved. The objective of this study was to check them and find out if the Earth core can contain REE.

Results The mantle xenoliths which judging by HREE concentrations, Al2O3/MgO and CaO/Al2O3 in them are identical to chondrites, and hence to a primitive mantle, were studied. Model Pb-Pb isotope ages (TCHUR) of such xenoliths correspond to 4510±30 Ma. At the same time ENd and 87Sr/86Sr in these xenoliths are typical of the MORB source.

Discussion The results evidence that the MORB source composition in not really identical to DM but to BSE or, which is the same, to perochemically primitive mantle. It means that deficit of LREE in the MORB source displayed by the value εNd=+10 arose at the earliest stage of the planet evolution and is not connected with formation of the crust or EM, as it was traditionally assumed. Accordingly, all kinds of the mantle material having εNd <+8 have to be considered as an EM. The DM (εNd>+12) is petrogenetically fruitless, has no volcanic derivatives and is represented only with xenoliths. Moreover, if the ratios of refractory elements such as REE, in initial planetary material in reality corresponded to those in chondrite one, just kipping the balance requires that deficient LREE in BSE be concentrated in the Earth core as a complementary reservoir. It is supported by the presence of phosphates and phosphides enriched with REE in irons [1]. So, the most probably the core is not an inert reservoir for REE.

Acknowledgement The study was supported by RFBR (grant 04-05-64462).

Reference [1] Davis, Olsen (1996) Meteorit. Planet. Sci., 31, A34.



Ocean Research Institute, The University of Tokyo, Tokyo, Japan ([email protected], [email protected]) 2 Department of Geology, The National Science Museum, Tokyo, Japan ([email protected]) We have developed 238U-206Pb and 207Pb-206Pb dating method of monazite by using a NanoSIMS NS 50 ion microprobe installed at Ocean Research Institute, The University of Tokyo. A ~4 nA O- primary beam was used to sputter a 5-µm-diameter crater and secondary positive ions were extracted for mass analysis using a Mattauch-Herzog Geometry. Multi-collector system was modified to detect 140 Ce+, 204Pb+, 206Pb+, 238U16O+, and 238U16O2+ ions at the same time. A mass resolution of ~5500 at 10% peak height was attained with a flat peak top, while the sensitivity of Pb was about 4 cps/1nA/ppm. A monazite from North-Central Madagascar with a U-Th-Pb chemical age [1] of 525.2±8.2 Ma (2σ) obtained by EPMA was used as the standard for Pb+/UO+ - UO2+/UO+ calibration. There is a positive correlation between the Pb+/UO+ and UO2+/UO+ ratios of the standard. A simple linear regression was more appropriate than the quadratic relation to fit the trend. The 207Pb/206Pb ratios were measured by a magnet scanning. U-Pb ages of 44 monazite grains extracted from a sedimentary rock in Western foothills of Taiwan were analyzed. Observed ages were compared with the U-Th-Pb chemical ages of the same sample [2]. 238U-206Pb ages agree well with those of the chemical except for a few samples. The discrepancy may be due to common Pb effect by an overestimation of radiogenic Pb by the chemical age. The 207Pb206 Pb ages also agree with the chemical age while there are a few discordant samples in addition to several samples with common Pb signature. Taking into account of concordant samples, there are three main age groups, 230Ma, 440Ma and 1850Ma. The age distribution suggests that the provenance of detrital monazites is possibly North China Craton [3,4].

References [1] Suzuki & Adachi (1991) Geochem. J. 25, 357-376. [2] Tsutsumi et al. (2004) Annual Meeting of Mineral. Soc. Japan Abstract G6 P-13, 255. [3] Zhao et al. (2002) Amer. J. Sci. 302, 191-226. [4] Guan et al. (2002) Precam. Res. 113, 1-18.

Goldschmidt Conference Abstracts 2005 Isotopic Records of Early Planetary Evolution


Comparative stable Fe isotope systematics of terrestrial and meteoritic materials R. SCHOENBERG1, B.S. KAMBER2 1 AND F. VON BLANCKENBURG 1

Institut für Mineralogie, Universität Hannover, Germany ([email protected]) 2 ACQUIRE, University of Queensland, Australia ([email protected]) Thus far, variations in δ56Fe values reported for chondrules from carbonaceous and ordinary chondrites (Zhu et al., 2001) and between bulk samples of HED meteorites, chondrites, Mars, Moon and Earth (Poitrasson et al., 2004) were interpreted by a number of different processes. These include Fe isotope fractionation through partial Fe evaporation and condensation, physical metal-silicate differentiation, and low-temperature aqueous alteration on the parent bodies. The aim of this study was to investigate (1) to what extent intraplanetary and inter-planetary processes may be responsible for the variability of Fe isotope signatures of planets and planetesimals and (2) whether trace element patterns of chondrules and CAIs that are related to condensation correlate with their Fe isotope variabilities. A comprehensive set of new high-precision Fe isotope data (Schoenberg et al., 2005) of 15 bulk chondrites, 15 iron meteorites, 8 eucrites, 20 terrestrial rocks as well as chondrules and CAIs from carbonaceous and ordinary chondrites will be presented. Within student-t test limits the three chondrite groups remain indistinguishable from each other and have an average δ56Fe value of –0.023±0.104‰ (2 S.D.) relative to the IRMM-014 Fe standard. However, student-t statistics also reveal that chondrites, eucrites (-0.002±0.024‰), iron meteorites (+0.035±0.078‰), and terrestrial silicate rocks (+0.085±0.080‰) analysed here have discernible, different Fe isotope compositions at a high level of confidence. The meteorite data suggest that preferentially heavier Fe isotopes were incorporated into metal cores during metal-silicate differentiation. Therefore, this process cannot be the cause for the slightly elevated δ56Fe values of terrestrial silicate rocks compared to bulk chondrites. We reproduce the wide Fe isotope fractionation in CAI and also show that chondrules form different chondrites are variably fractionated in Fe. Trace element data for chondrules and CAI show that those with the most anomalous refractory element abundances tend to also be strongest fractionated in Fe.

References Zhu, X.K. et al., (2001), Nature 412, 311-313. Poitrasson, F. et al., (2004), EPSL 223, 253-266. Schoenberg, R. et al., (2005) Int J Mass Spectrom (in press).

Goldschmidt Conference Abstracts 2005 Kinetics and Metamorphic Processes

A brief retrospective of Bill Carlson’s work on metamorphic disequilibrium and kinetics DAVID M. HIRSCH 516 High St., Geology Dept., Western Washington University, Bellingham, WA 98229, USA ([email protected]) Bill Carlson’s work has spanned a wide range of metamorphic and mineralogic topics. One overarching theme has been the extension of metamorphic petrology to occurrences dominated by disequilibrium, rather than equilibrium textures. Beginning with his graduate work on calcite-aragonite transition kinetics, continuing with coronal reaction textures, and, for the past 15 years, focusing on the quantitative analysis of porphyroblast textures, Bill has driven the field forward. His advances have been both in the theoretical realm, building on the work of folks like Ralph Kretz, and in the technical realm, in which he pioneered the use of high-resolution computed tomography for the analysis of porphyroblast textures in three-dimensions. While he has not been alone in this effort, his work, along with that of his students and colleagues, has been instrumental in advancing our science from the clean ideality of equilibrium towards the messy but more accurate world of disequilibrium.


Oxygen isotope speedometry in the Biwabik iron-formation ELIZABETH P VALAAS AND JOHN W VALLEY Dept. of Geology, University of Wisconsin, Madison, USA ([email protected], [email protected]) The Biwabik banded iron-formation (BIF) of northern Minnesota (1.9 Ga), underwent contact metamorphism by intrusion of the Duluth Complex (1.1 Ga). The igneous contact is sub-parallel and intersects the BIF at Dunka Pit. Apparent temperatures, calculated from ∆18O (QuartzMagnetite) (Clayton and Kieffer, 1991), are precise and decrease smoothly from 700ºC at the contact to 375ºC at greater than 2.6 km (3-D, normal to the contact). However, measured temperatures are similar to closure temperatures (Tc; Dodson, 1973) for oxygen diffusion in magnetite at a cooling rate of ~2000°C/Ma over a gradient in observed grain size of magnetite of 2 mm at the contact to 10 µm at 2.6 km. In the absence of recrystallization, resetting of ∆18O (Qt-Mt) is limited by diffusion in magnetite inside the grunerite isograd (2.6 km from the contact). At 1.25 km from the contact, the measured “temperature” is 490ºC and Tc is 468ºC; at 2.6 km from the contact the temperatures are 375ºC and 404ºC respectively. If Tc at the contact equals the measured apparent temperature, the cooling rate is defined. To test whether ∆18O (Qt-Mt) is a thermometer or a speedometer in these rocks, additional quartz magnetite pairs were analyzed from two outcrops in Dunka Pit within 10 m of the contact for two grain sizes: 500-350 µm and 150-105 µm diameter. The apparent temperatures are: 669°C, 702°C, and 716°C for 03BIW43B and 601°C, 644°C, and 732°C for 03BIW18C; for 150-105 µm, bulk, and 500-350 µm diameter magnetite grains respectively. Thus magnetites have been reset in δ 18O as a function of grain size and presumably are zoned in δ 18O due to diffusion. Inverted pigeonite in BIF at Dunka Pit indicates temperature > 775°C which further supports the interpretation that ∆18O (Qt-Mt) records retrograde exchange. Thus, measured ∆18O (Qt-Mt) values record the rate of retrograde exchange. This suggests that the Biwabik ironformation at Dunka Pit cooled from ~800°C to Tc (700°C) in 50 kyr. Measured apparent ∆18O temperatures are lower than peak temperatures and should not be used to estimate depths of intrusion or peak metamorphic temperatures.

References Clayton and Kieffer (1991) The Geochemical Society Special Publication. 3, 3-10. Dodson (1973) Contrib. to Min. and Pet.. 40, 259-274.

Goldschmidt Conference Abstracts 2005 Kinetics and Metamorphic Processes


Effects of reaction kinetics on mixed volatile (CO2-H2O) decarbonation reactions in contact aureoles JOHN R. BOWMAN





Dept. of Geology and Geophysics, University of Utah, Salt Lake City, UT 84112, USA ([email protected]) 2 Veritas Geoservices, 10300 Town Park Drive, Houston, TX 77072, USA ([email protected]) One-dimensional models of heat and mass (CO2-H2O) transport are used to evaluate the roles of fluid flux, diffusion, and reaction rate during infiltration-driven metamorphism by modeling the reaction: calcite + quartz = wollastonite + CO2. Incorporation of reaction kinetics (rate constants based on experimental data) produces little change in the rate of advance of a reaction front in up-T flow, but causes reductions of < 20 to 25% in the rate of advance of a reaction front in down-T flow. In contrast, diffusion has no significant impact on the rate of advance or width of a reaction zone in either upor down-T flow at significant fluid flux (10-9 m3 m-2 sec-1) and for limited timescales ( < 50,000 yrs) of flow and reaction. A significant impact of reaction kinetics in down-temperature flow is to expand reaction fronts to reaction zones. At reaction rates based on experimental studies, reaction zones of > 40 m are quickly developed at significant fluid flux (10-9 m3 m-2 sec-1) and high porosity (φ = 0.1). If fluid flux decreases by decreasing porosity, the width of the reaction zone narrows considerably; at low porosity (φ < 0.001) reaction zones are less than 5 m in width. However if reaction rates in contact metamorphic environments are one to two orders of magnitude less than experimental results, then reaction zones of significant width (> 8-10 m) could be developed in contact aureoles at porosity of > 0.001 and fluid flux > 10-11 m3 m-2 sec-1. Isograds formed in contact aureoles during down-T flow are typically narrow (<< 10 m; many << 1 m). Our modeling results indicate that these narrow reaction zones require significant fluid infiltration rates (>1 x 10-8 m sec-1), and either reaction rates faster than experimental values or low porosity (< 0.001). Rapid reaction rates appear at odds with an increasing number of reported occurrences of mineralogic and oxygen isotope disequilibrium in metamorphic systems. Alternatively, the narrow reaction zones may indicate that infiltration-driven metamorphism occurs at low porosity (< 0.001). If so, the modeling results indicate that equilibrium reaction surfaces may be significantly overstepped for both up- and down-T flow. Such overstepping leads to the possibility that index minerals could be produced in contact aureoles by metastable reactions.

Unreactivity of garnet in low pressure metapelites D.R.M. PATTISON, D.K. TINKHAM AND P. YANG Dept. Geology & Geophysics, U Calgary, Calgary, Alberta T2N 1N4, Canada ([email protected], [email protected], [email protected]) Garnet is one of the most widespread and petrologically important metamorphic minerals in metapelites. Several lines of evidence suggest, however, that in low pressure metamorphism at least, garnet forms at low grade but may not participate modally in subsequent mineral reactions until considerably higher grade is reached: even though it is physically present, it is to a large degree chemically inert due to sluggish kinetics of reaction. The evidence includes: distribution of garnet in individual rocks that is independent of the distribution of later-formed porphyroblasts; euhedral shape of garnet in rocks in which phase equilibrium considerations suggest it should have dissolved significantly (e.g, staurolitebearing rocks); chemical zoning patterns of garnet in such rocks that show no evidence for reaction; and phase equilibrium systematics of low pressure metapelites that work fine when garnet is ignored. To test the implications of these observations, we have performed phase equilibrium modelling for isobaric low pressure (3.8 kbar) prograde metamorphism of average metapelite for two end member situations: perfect equilibrium crystallization and perfect fractional crystallization with regard to garnet. The predicted sequence and position of reaction isograds is immaterially different between the two situations, possibly accounting for the success of low pressure phase equilibrium systematics that ignore garnet. These results show that care is required in using garnet in mineral assemblage-based petrogenetic grids, such as for bathozone determinations. Staurolite appears to show some of the same behaviour as garnet, which we have also modelled.