Goldschmidt Conference Abstracts 2005

Goldschmidt Conference Abstracts 2005

Goldschmidt Conference Abstracts 2005 Petroleum Processes from Source to Trap A501 Recognition of oil-condensate mixtures: Implications for basin sc...

2MB Sizes 0 Downloads 222 Views

Goldschmidt Conference Abstracts 2005 Petroleum Processes from Source to Trap


Recognition of oil-condensate mixtures: Implications for basin scale petroleum processes

A new genetic scheme for natural gas formation and isotopic evidence for oil cracking




USGS, Denver, CO 80225, USA ([email protected]) 2 Humble Geochemical Services, Humble, TX 77338, USA ([email protected]) 3 Department of Geological and Environmental Sciences, Stanford University, Stanford CA, 94305, USA ([email protected], [email protected], [email protected]) Oil-condensate mixtures are commonly difficult to recognize, in part due to a limited understanding of the genetic significance of condensate hydrocarbon distribution. Recent geochemical advances in our discernment of shale, carbonate, G. prisca kukersite, and terrigenous source-rock facies based on light hydrocarbon distributions have established criteria for recognizing oil-condensate mixtures, and insight into basinscale processes that control petroleum composition. Here, we demonstrate how the interpretation of petroleum processes is possible through the integration of routine biomarker, light hydrocarbon, diamondoid and compound specific isotope analyses. Biomarker and bulk geochemical analyses of oils allow us to interpret the lithology, redox and salinity conditions, organic matter type, and approximate geologic age of the source rock that generated the petroleum. However, biomarker applications are limited for light oils and condensates that are devoid of biomarkers or contain these compounds in low concentrations. Shale, carbonate, G. prisca kukersite, and terrigenous sourced oils can be clearly differentiated using ternary plots of C7 hydrocarbon isomers and can reveal source facies not apparent from biomarker analysis. Furthermore, light hydrocarbon and biomarker integrated with diamondoid technology and compound specific isotope data allows for interpretation and better understanding of basin scale petroleum generation, migration and mixing processes.


Power Environmental Energy Research Center, California Institute of Technology, Pasadena, CA, USA ([email protected]) 2 GasConsult International Inc, Berkeley, CA, USA ([email protected]) The positive identification of the source rock(s) of natural gas in a sedimentary basin is still mostly based on empiric models. We developed a method based on laboratory simulation of gas formation and isotope compositions of gas components that allows the correlation of gases to their source(s) and an assessment of amount of gas (gas potential) and time of gas formation. From these experiments we developed a new genetic scheme of gas formation. We differentiate three principal types of gases (1) Early Gas: These gases form through low temperature reactions, very likely either biologically or inorganically catalyzed degradation processes of bitumen. Examples are bacterial gases and low-temperature (40 to 80C) shale gases. (e.g. gases from the Wildemere Basin in Western Canada. (2) Gas from Primary Cracking of KANSO: At temperatures between 80 and 120C gases form through primary cracking of asphaltenes and NSO compounds that form early from kerogens as transitional precursors that are converted to gas during primary cracking (3) Gases from Secondary Cracking of oils : One of our findings is that gases from secondary cracking of oils are systematically depleted in 13C in methane and ethane compared to gases from primary cracking of KANSO. Comparing our oil cracking calibrations with natural gas data, we find that the isotope data of the South Pars Field, a giant gas accumulation in the Gulf of Persia and many North Sea gas deposits cannot be explained by primary cracking of kerogen. For example, the South Pars gases would correspond to the very beginning of primary cracking (1% Ro) of gas from type III kerogen with a very low gas potential or alternatively to a high temperature environment of oil cracking at temperatures around 215C which is much more sensible suggesting that the South Pars gases are derived from secondary cracking of oils. Studies of gas isotope fractionation for Northsea will also be discussed.

Goldschmidt Conference Abstracts 2005 Petroleum Processes from Source to Trap


Stable isotope systematic of coalbed methane

The application of thermal simulating experiment in gas-source correlation



([email protected], [email protected], [email protected]) Stable carbon and hydrogen isotopes find useful applications in the exploration activities associated with Coalbed Methane (CBM), in particular the identification of different gas types and generation pathways. Isotopes provide critical information on the degree of gas saturation in coal seams. Kinetic isotope effects offer improved estimations of the “residual gas component”, which is difficult and imprecise using conventional canister methods. The isotope signatures may also improve or replace the common canister desorption method to estimate “total gas”. We have conducted analyses on a base of over 1000 gas samples collected from 7 different coal-bearing basins with ranks ranging from sub-bituminous to anthracite. These samples were taken 1) as time series during conventional desorption experiments and 2) from CBM production wells over a time span of several months. The analyses include gas molecular composition (C1 to C4, CO2) and C-, H-isotope ratios of hydrocarbons. Initial results reveal diagnostic trends in isotopic composition. The desorption experiments reveal consistent and systematic shifts in isotope ratio to increasing 13-C with time. In contrast, methane from production well samples grows isotopically 12-C enriched with time. Time series measurement of the gas composition recognize the point of incipient CO2 desorption. They also provide information on the volume of coal in the subsurface that is influence by desorption. The same principles can be applied to monitoring the effectiveness of lithologic fracturing processes, i.e., the progress and success of the operation.


Department of Earth Science, Nanjing University P. R. China ([email protected]) 2 Wuxi Institute of Experimental Petroleum and Geology, SINOPEC, P. R. China

Experiment 6 samples from Kuche Depression of Tarim basin were collected and thermal simulated with a step of 40℃ and 8 stages from 260℃ to 540℃. The components, and ethane carbon isotopic compositions of simulated gases generated from mudstone, carbargilite and coal are analyzed. The drying coefficients (C1/(C2+C3)) are caculated.

New method

Using drying coefficient of natural gas from well Ku-1 as parameter for interpolation method, relevant temperatures and thermal maturities of simulated samples are figured out (Table 1). According to geological background and the possible maturity range, the study showed that only the gas generated at about 420℃ from sample YM-4 of Jurassic coal measures is consistent with well Ku-1 both in hydrocarbon components and ethane isotopic composition. That means Jurassic coal is the main source rock of natural gas from well Ku-1. Table 1 The interpolation temperature and Ro when the drying coefficent equals to that of gas from well Ku-1.

Samples Kuchehe13 T3t Kuchehe 16 T3t YM-1 T3t YM-2 T3t YM-3 J2kz YM-4 J2kz TS-1 J TS-2 J

Low temperature stages T(℃) Ro(%) 327.4 0.97 325 0.96 294 0.82 334.4 1 317 282 310

0.93 0.76 0.89

High temperature stages Ro(%) T(℃) 483 2.61 503 2.86 493 2.73 469 2.44 477 2.54 422 1.79 504 2.87 578 3.31

Conclusion Combined with ethane carbon isotope, component and maturity data, the simulation gas drying coefficient can be used in gas-source correlation. The interpolation method overcomes the influence caused by organic thermal maturity and provides a new way for study of oil-gas source correlation.

Goldschmidt Conference Abstracts 2005 Petroleum Processes from Source to Trap

Trace-element of calcite cement in reservoir rocks as a useful tool defining hydrocarbon migration pattern, Junggar Basin, China WENXUAN HU1, JIAN CAO1, YIJIE ZHANG2, YUEQIAN ZHANG2 AND XIAOKANG GAO1 1

State Key Laboratory for Mineral Deposits Research, Earth Sciences Department, Nanjing University, Nanjing 210093, China ([email protected]) 2 Xinjiang Oilfield Company, PetroChina, Karamay 834000, China ([email protected]) Hydrocarbon migration patterns are becoming increasingly defined using molecular geochemical tools, as well as fluid inclusions in cements. A case study in Junggar Basin suggests that it might also be identified by analyzing trace-element of reservoir calcite cements. As the Permian source sequences of the Junggar Basin contain many volcanic materials (e.g. basaltic volcanic and volcaniclastic rocks), the hydrocarbon fluid is consequently Mn-enriched. When this Mn-rich fluid traveled from source to trap, it would interact with reservoir hosts, dissolving feldspars and pre-existing calcite cements precipitated from formation fluid, and even resulting in crystallization of new generations of calcite cements. In contrast, marine and formation fluid calcite are often enriched in Mg and Fe respectively. Thus the calcite forming in the hydrocarbon fluid has proportionally high Mn contents relative to Mg and Fe. EPMA analyses of ca. 1, 000 cement samples from migration pathway framework (fault, unconformity, and reservoir rock) for Mn, Mg, and Fe show that the fault calcites is rich in Mn (averagely greater than 1.0 wt.%), the unconformity samples have less Mn (0.1 – 1.0 wt.%), whereas calcites in reservoir rocks attain more Mg and relatively least Mn (generally less than 0.1 wt.%). Fe varies unclearly in these three suits. Therefore, it is inferred that the fault water was more hydrocarbon influenced than in the other two zones, indicating fault is most favorable for hydrocarbon migration. Furthermore, it is also implied the hanging wall of the fault (Mn>2.0 wt.%; Mg<0.5%) takes priority to be charged compared to the footwall (Mn=1.0 – 1.5%; Mg>1.0%). Modern surface seeps at Karamay have been proved to be manifestation of fault-controlled subsurface pathway. Some of such fault calcites were developed with several tens of zoned thin hydrocarbon intergrowths, along which the content of Mn, Mg, and Fe fluctuate cyclingly, implying that fault was an episodically active conduit for oil-bearing fluid.


Goldschmidt Conference Abstracts 2005 Molecular Computer Simulations


Computational molecular modeling of ions in aqueous solutions

Ab initio molecular dynamics of clay mineral surfaces and interfaces



Department of Earth Sciences, University of Bristol, Bristol BS8 1RJ, UK ([email protected]) Nearly all geochemical processes in the Earth’s crust involve the complexation of ions in aqueous fluids. Complexation by ligands such as Cl- and HS- is responsible for the extraction of metals from primary rocks and the formation of ore deposits. Complexation of metals by mineral surfaces limits the aqueous concentrations of biolimiting micronutrients and controls the mobility of toxic metals and radionuclides in soil and groundwater. For the past 50 years, geochemists have sought to develop thermodynamic models of the speciation of ions in complex aqueous solutions. Stability constants and mineral solubilities measured at low P and T can be extrapolated to high PT using models based on the Born theory of ion solvation. However, it is often ambiguous which complexes are necessary to account for measured mineral solubilities. There are a number of reasons why the Born model of aqueous solvation may be unreliable. We now have the ability to explore aqueous solutions at a molecular level using computational quantum chemistry (based on density functional theory) and atomistic simulations (based on classical interatomic potentials) Several examples will be given to illustrate the range of possibilities achievable with a small parallel computing cluster (< 20 nodes). Static calculations on large atomic clusters can be used to predict the structures and energetics of metal complexes on mineral surfaces. Insights from these calculations are used to develop surface complexation models to fit sorption isotherms. Examples include the sorption of As, Cu, U on FeOOH and kaolinite. Ab initio molecular dynamics simulations (100’s of atoms, t < 10 ps) can be used to predict the speciation of metals such as Cu and Sn in hydrothermal fluids as a function of pressure and temperature. Classical molecular dynamics calculations allow very large simulations (1000’s of atoms, t > 100 ps) that can predict phase separation and equations of state of NaCl-H2O mixtures. For practical geochemical applications, we need to be able to predict the activities of aqueous species. Direct calculations of ion chemical potentials are possible but at much greater computational expense. It is anticipated, however, that such calculations will become increasingly practical and routine as computational technology develops.


Austrian Research Centers Seibersdorf, Austria ([email protected]) 2 Institute for Theoretical Chemistry, University of Vienna, Austria ([email protected]) 3 Institute of Soil Research, University of Agricultural Sciences Vienna, Austria ([email protected]) Sorption/desorption processes on surfaces or in interlayer spaces of clay minerals play an important role both in natural environments as well as in industrial production. Owing to the structural complexity of clay mineral surfaces it is often difficult to explain in detail the molecular mechanisms of these processes in spite of the application of various experimental techniques. In this situation, computer simulation methods provide a very powerful tool to model particular scenarios and to study surface complexes at molecular level for providing a detailed insight into the adsorption processes. In this work we discuss ab initio molecular dynamics simulations of interactions between selected clay minerals (namely kaolinite and montmorillonite) and several polar molecules (water, acetic acid and phenoxyacetic-acid derivatives). These interactions are studied with either isolated molecules or entire molecular layers adsorbed on the surfaces or embedded in the interlayer space. Solvent effects are considered by including water molecules explicitly in the simulation. Structural, energetic and dynamic properties are evaluated. The most improtant phenomena in forming the surface complexes with the polar molecules are hydrogen bonds. The kaolinite surface formed from hydroxyl groups is chemically very active and hydroxyl groups are able to act as proton donors or acceptors. On the other hand, the kaolinite surface formed from basal oxygen atoms form only weak hydrogen bonds with polar molecules. In case of a water layer on this surface hydrogen bonds formed among water molecules prevail over hydrogen bonds formed with this surface. In contrast to kaolinite, the surface of the montmorillonite layer is formed only from basal oxygen atoms. Moreover, this layer also posesses an excess negative charge due to isomorphic substitutions within the layer. The models used here are more complex than in the previous cases since the sorption of polar or ionic species can proceed via a cation bridge mechanism. This situation was investigeted and it was found that in the presence of water molecules the surface cation bridge complexes are less stable than the clay+cation complexes in the solution.

Goldschmidt Conference Abstracts 2005 Molecular Computer Simulations

Electron transfer reactions in solution and at interfaces JAMES R RUSTAD1, KEVIN M. ROSSO2 2 AND ANDREW R. FELMY

First-principles simulation of solvation structure and deprotonation reactions in very nonideal solutions ERIC BYLASKA1, MARAT VALIEV1, STUART BOGATKO2 2 AND JOHN WEARE


Department of Geology, University of California, Davis, One Shields, Ave, Davis, CA 95616, USA ([email protected]) 2 Pacific Northwest National Laboratory, MS K8-96 Richland, WA 99352, USA ([email protected], [email protected])

We present a molecular model for ferrous-ferric electron transfer in an aqueous solution and at interfaces that accounts for electronic polarizability and exhibits spontaneous cation hydrolysis, and allows estimation of pH dependence. In solution, the model predicts that the diabatic barrier to electron transfer increases with increasing pH, due to stabilization of the Fe3+ by fluctuations in the number of hydroxide ions in its first coordination sphere, in much the same way as the barrier would increase with increasing dielectric constant in the Marcus theory. As expected, increasing pH reduces the potential of mean force between the ferrous and ferric ions in the model system. The magnitudes of the predicted increase in diabatic transfer barrier and the predicted decrease in the potential of mean force nearly cancel each other at the canonical transfer distance of 0.55 nm. Even though hydrolysis is allowed in our calculations, the distribution of reorganization energies has only one maximum and is Gaussian to an excellent approximation, giving a harmonic free energy surface in the reorganization energy F(∆E) with a single minimum. Evidently, fluctuations in hydrolysis state can be viewed on a continuum with other solvent contributions to the reorganization energy. There appears to be little justification for thinking of the transfer rate as arising from the contributions of different hydrolysis states. We have used the same methods to examine electron transfer rates at interfaces. An important question is whether electron-hopping rates would be faster in the bulk or at the interface. On the one hand the enhanced conformational flexibility and coupling with proton hopping may facilitate surface-mediated electron hopping. On the other hand, the proximity of the high dielectric water layer will act to trap the electrons. Our calculations suggest that, at low pH, the dielectric contribution dominates and reorganization energies are higher near the interface than in the bulk.



Pacific Northwest Laboratories, Richland, WA, 99352, USA ([email protected], [email protected]) 3 University of California San Diego, San Diego, CA, 92093, USA ([email protected], [email protected]) The structure of the water molecules neighboring an M3+ ion in an aqueous solution is strongly perturbed leading to a structured second and a weakly structured third solvation shell and increased acidity of the solution. For simple ions, e.g. Al3+, standard pseudopotential based first principle molecular dynamics methods can be applied. Simulation results agree well with the measured octahedral structure of the 1st solvation shell of Al3+. Less can be determined experimentally about the structure of the 2nd shell. However, the calculated average radius is in good agreement with the measured values. This shell contains roughly 12 water molecules, which are trigonally coordinated to the 1st shell waters. This is also consistent with experimental estimates. The emergence of tetrahedral bulk water coordination as a function of the distance from the ion center occurs in the third shell. While there is no transfer of waters from the first to the 2nd shell, there is picosecond time scale transfer between the 2nd shell and third shell. For high T the transfer of protons in the solvation shells leads to hydrolysis species. For transition metal ions pseudopotential methods are not reliable. For these systems we have implemented an augmented wave method which allows the use of a plane wave basis without the introduction of pseudopotentials. Results for the solvation structure of Fe3+ will be presented. As time allows, the local structure of counter ions in the solvation shells will also be discussed.

Goldschmidt Conference Abstracts 2005 Molecular Computer Simulations


Isotopic effect on phase equilibria of pure fluids and mixtures: Molecular simulation, theory and experiment

Ab initio molecular dynamics simulations of silicate liquids at high pressure




Chemical Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN37831-6110, USA ([email protected], [email protected])

The temperature dependence of vapor – liquid and vapor – solid isotope fractionation factors was predicted by NVTMolecular Dynamics and Gibbs Ensemble Monte Carlo simulations for atomic molecules (36Ar/40Ar, 20Ne/22Ne, 80 Kr/84Kr) to assess the accuracy of ħ2-order KirkwoodWigner free energy expansion for specific Lennard – Jones parameterizations. In addition, the composition dependence of the corresponding fractionation factors for binary Ar–Kr mixtures was also predicted. Our simulation results compare very well with the existing experimental data. Fractionation factors were also predicted for other isotopic pairs, which have not been studied experimentally to date, including 20Ne/21Ne, and 132Xe/136Xe. Some premises behind the microscopic interpretation of the fractionation factors are also tested. Relevant extensions of this study to polyatomic molecular systems of geochemical interest, including O2, N2, CO, and H2O, are also discussed. The advantages of this approach over quantum mechanical calculations and the limitations of the methodology are addressed as well.

Acknowledgements This research was sponsored by the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences under contract number DE-AC0500OR22725 with Oak Ridge National Laboratory, managed and operated by UT-Battelle, LLC.

Reference Chialvo, A.A. and Horita, J. (2003), J. Chem. Phys., 119, 4458.


Department of Geological Sciences, University of Michigan, Ann Arbor, MI, USA ([email protected]) 2 Department of Computer Science, Louisiana State University, Baton Rouge, LA, USA ([email protected]) Silicates liquids are primary agents of mass and heat transport, yet little is known of their properties over much of the vast range of pressures and temperatures that are relevant to planetary evolution. In the last few years, it has become possible to study for the first time silicate liquids with first principles quantum mechanical methods. We have applied density functional theory to the study of silicate liquids via ab initio molecular dynamics. The electronic structure, total energy, stress tensor, and forces acting on the atoms are computed self-consistently at each time step and atomic positions advanced according to Newton's equations of motion. The simulations are performed in the NVT ensemble with a Nose thermostat. Typical simulations consist of 80 atoms and are run for several picoseconds. We present initial results on MgSiO3 liquid that span the entire pressure and temperature range of the deep earth. Preliminary results indicate that the fluid contains dominantly four-fold coordinated silicon and that the average coordination number increases gradually with increasing pressure. A surprise is that isotherms at 3000 K and 6000 K diverge on compression, behavior that is contrary to that of most crystals, and which implies a Grüneisen parameter that increases on compression. We will address issues including liquid-crystal density inversion, comparison with experiment, and analysis of the equation of state in terms of inherent structure.

Goldschmidt Conference Abstracts 2005 Molecular Computer Simulations

Computer modeling of the equations of state of crystals and melts in the CaO-MgO-Al2O3-SiO2 system MASANORI MATSUI Earth Sciences, School of Science, University of Hyogo, Hyogo 678-1297, Japan ([email protected]) Crystals and melts in the CaO-MgO-Al2O3-SiO2 (CMAS) system are important components of the earth’s crust and mantle. Both molecular dynamics (MD) and lattice dynamics (LD) methods are used to calculate the properties of crystals, while the MD technique is applied to simulate melts. The interionic potential is taken to be the sum of pairwise additive Coulomb, van der Waals, and repulsive interactions. In addition, in order to take account of many-body forces in crystals and melts, the breathing shell model (BSM; Matsui, 1998) is developed for simulation, in which the repulsive radii of O ions are allowed to deform isotropically under the effects of other ions in the system concerned. The net charges of the ions are constrained to be q(Ca) = q(Mg) = 2/3q(Al) = 1/2q(Si) = -q(O) to apply the potential to both crystals and melts with any composition in the CMAS system. Required energy parameters were derived empirically to reproduce the observed temperature-pressure-volume equations of state of a wide structural variety of crystals in the CMAS system, as well as the measured volumes of enstatite, wollastonite, diopside, and anorthite melts at high temperatures. The LD and MD simulations with the BSM potential are quite successful in reproducing well these measured properties of both crystals and melts. In enstatite, wollastonite, diopside, and anorthite melts at 1900 K and 0 GPa, we found the simulated cation coordination numbers are four for Si, about four, five, and seven for Al, Mg, and Ca, respectively. The pressure dependences of cation coordination numbers in these melts are also investigated. The MD method is further applied to study the compositional dependences of the volumes of melts in the SiO2-Al2O3, SiO2-Ca3Al2O6, and MgSiO3-CaSiO3 joins at high temperatures.


Molecular dynamics simulations of the structural and kinetic properties of amorphous intergranular films in alumina STEPHEN H. GAROFALINI AND SHANGHONG ZHANG Department of Ceramic and Materials Engineering, Rutgers University, Piscataway, NJ 08854, USA Thin, intergranular films (IGFs) of only one to a few nanometers thickness are prevalent in minerals and, while making up only a small volume percent of the material, often strongly influence material properties. However, the glassy nature and thickness of the IGF have made experimental observations of atomistic descriptions of these phases difficult to obtain until recently and computational techniques offer a complementary approach to understanding these atomistic details. The results of molecular dynamics computer simulations of the structural and kinetic properties of glassy calcium alumino-silicate (CAS) intergranular films (IGFs) in polycrystalline alumina will be presented. Calcia and silica can exist as impurities in alumina and segregate out to the crystal surface at high temperatures, affecting dissolution and grain growth. In experimental studies using liquid phase sintering, composition of the IGF was shown to significantly alter grain growth, although the mechanisms were not understood. The role of composition on the atomistic structure of the IGF and growth behavior of different crystallographic orientations will be presented. The simulations show ordering into the amorphous IGF caused by the presence of the crystal interface. Such structural changes also effect diffusion of species within the IGF. The simulations also show that grain growth along the surface normal of the basal plane is inhibited by preferential segregation of Ca ions at the IGF/crystal interface, although this behavior is affected by composition of the IGF. Dissimilar behavior occurs at the prism surface, where Al (and O) adsorption from the IGF onto the prism surface occurs in a manner consistent with grain growth along the surface normal. These simulation results are consistent with experimental studies regarding anisotropic grain growth in alumina as a function of composition and provide atomistic mechanisms regarding grain growth.

Goldschmidt Conference Abstracts 2005 Molecular Computer Simulations


MgSiO3 post-perovskite at D" conditions

Development of transferable interatomic potentials for oxides and silicates using DFT calculations

TAKU TSUCHIYA, JUN TSUCHIYA AND RENATA M. WENTZCOVITCH Department of Chemical Engineering and Materials Science, Minnesota Supercomputing Institute for Digital Technology and Advanced Computations, University of Minnesota, Minneapolis, USA (612 625 8597, [email protected]) The thermoelastic properties of the new polymorph of MgSiO3 with the CaIrO3 structure and more stable than the Pbnm-perovskite phase at conditions close to those expected in the D" region has been investigated by first-principles computations and contrasted with those of the perovskite phase. Although we investigate only single and pure phases, the elasticity of aggregates containing predominantly these phases is expected to differ similarly, although in smaller magnitude. We therefore predict the major trends in seismic properties expected to occur in the presence of such phase change, such as velocity discontinuities, ratios of velocities and density anomalies, and anisotropy in aggregates with preferred orientation. Consequences of this model mineralogy for the D" region will be discussed.

Acknowledgements J. Tsuchiya and T. Tsuchiya thank JSPS for research fellowships. Research supported by NSF/EAR 0135533 (COMPRES), 0230319.


GeoForschungsZentrum Potsdam, Department 4, Telegrafenberg, 14473 Potsdam, Germany ([email protected]) 2 Departamento de Fisica Teorica, Universidad de Valladolid, 47011 Valladolid, Spain ([email protected]) 3 Chemistry Department, University of Edinburgh, Edinburgh EH9 3JJ, United Kingdom Atomistic modeling of physical properties of minerals and melts under extreme conditions of pressure and temperature requires a reliable and transferable representation of the interaction potential. Electronic structure calculations from first principles have proven the most accurate techniques, but their computational cost is still too high for systems that require a large number of atoms and long simulation times. It is, therefore, highly desirable to obtain interatomic potentials that include the essential interactions and that are transferable in a wide range of p/T and between different compositions. Here we present an ionic interaction model that includes explicitly ionic polarization effects up to the quadrupolar level, and size and shape deformations of the ions that depend on the ionic environment of the individual ions. In the spirit of Born-Oppenheimer dynamics the energy due to these terms has to be minimized before the forces on the ions are calculated. The potential parameters are optimized by fitting forces and multipoles of individual ions and the stress tensor of different reference configurations to corresponding properties obtained by planewave-DFT calculations. The resulting interaction potentials are shown to be transferable between different coordination environments and in a wide range of p/T. This is demonstrated for simple oxides, like MgO or Al2O3. Examples include phase stability of crystal polymorphs, thermoelastic properties, solid-melt interfaces, melt structure and transport properties. Finally, first results of new potentials for silica and silicates will be presented and discussed.

References Aguado A., Bernasconi L., Jahn S. and Madden P.A., (2003), Faraday Discuss. 124, 171-184. Aguado A. and Madden P.A., (2004), Phys. Rev. B 70, 245103. Jahn S., Madden P.A. and Wilson M., (2004), Phys. Rev. B 69, 020106.

Goldschmidt Conference Abstracts 2005 Molecular Computer Simulations

O triclusters revisited: Classical MD and quantum cluster results for glasses of composition (Al2O3)2(SiO2) 1





Department of Chemistry and Biochemistry, Univ. of Maryland ,College Park, MD 20742, USA ([email protected]) 2 Institut für Physik, Johannes Gutenberg-Universität, Staudinger Weg 7, D-55099 Mainz, Germany ([email protected]) The 17O NMR spectrum of CaAl2Si2O8 glass shows two types of O sites which are not present in the crystalline material. One of these, with 17O NMR parameters CQ = 2.3 MHz and δ = +20 ppm, was been assigned to a "tricluster" O, a local geometry in which the O is coordinated to three tetrahedrally coordinated atoms, either Al or Si. However, several different quantum chemical cluster calculations employing energy-optimized geometries for various tricluster species have given CQ values considerably larger than that seen experimentally in the CaAl2Si2O8 glass. We have recently shown that for edge-sharing geometries, in which the tricluster O atoms participate in "two-membered rings" of composition Al2O2, the calculated CQ values are considerably lower, in the range identified in the glass. A recent classical MD simulation of the structure of glassy aluminum silicate AS2, (Al2O3)2(SiO2) gave a predominance of O triclusters within 2membered rings. We have now calculated 17O nuclear quadrupole coupling constants and NMR shielding values for clusters extracted from these simulations. The calculated CQ values for these O triclusters are now in the range observed experimentally in the CaAl2Si2O8 glass (around 2.3 - 2.6 MHz) when the tricluster O is surrounded by three Al, two of which are part of an Al2O2 ring.


Molecular orbital study on dissolution of allophane with dilute alkali solution Z. ABIDIN, N. MATSUE AND T. HENMI Faculty of Agriculture, Ehime University, 3-5-7 Tarumi, Matsuyama 790-8566 Japan Allophane is a hydrous aluminum silicate of hollow spherical morphology with some pores, and wall of the sphere is composed of gibbsite sheet with orthosilicate attached to inside of it. Allophane is known to susceptible to alkaline condition than layer silicates, but detailed dissolution mechanism has not been known. Here we present molecular orbital calculation results for the interaction between allophane and dilute alkali solution, with focusing nucleophilic attack of hydroxyl ion on Si atom exposed at the pore region of allophane. Model cluster of allophane simulated the pore region, in which each Si atom is attached to gibbsite sheet via three SiO-Al bondings with one Si-OH exposed. Molecular orbital calculation was done with semiempirical MOPAC-PM3 and AM1 method in WinMOPAC program (ver. 3.9, Fujitsu). We assumed that the hydroxyl ion first attacked Si atom at the pore, and formed five-coordinated Si as a intermediate product. The nucleophilic attack of hydroxyl ion was indicated to be promoted by adsorption of Na+ which added as NaOH, because calculated positive charge of Si atom was greater for Si-O-Na than for Si-O-. Si-O bond length was also greater for the Na+ exchanged model.

By forming five-coordinated Si, bond order of not only the original four Si-O bondings, but adjacent Al-O and O-Al inside bondings also decreased. This suggests that one Si atom dissolution may be followed by three Al atoms dissolution, and is in good agreement with our previous experimental results carried out at about pH 12. Molecular orbital calculation was also done with some water molecules added around Na+ to simulate reactions in water, but the obtained result was similar to that in vacuo. The overall reaction of allophane dissolution with dilute NaOH solution is proposed as: fast cation exchange reaction to produce Si-ONa, followed by nucleophilic attack of hydroxys ion on Si atom of increased positive charge, then dissolution of Si and Al through bond weakening.


Goldschmidt Conference Abstracts 2005 Molecular Computer Simulations

The effect of the sizes of alkali cations on structural variations in layered silicate materials N.H. DE LEEUW1,2 AND Z. DU1 1

School of Crystallography, Birkbeck College, University of London, Malet Street, London WC1E 7HX, UK ([email protected]) 2 Department of Chemistry, University College London, 20 Gordon Street, London WC1H 0AJ, UK ([email protected]) Computational techniques were used to investigate the structures and stabilities of a series of solid solutions of the phyllosilicate material Li2(1-x)M2xSi2O5 (M = Na, K, Rb, Cs and x = 0, 0.25, 0.5, 0.75, 1) as well as solid solutions of the mixed alkali phyllosilicate KLiSi2O5 with Na, K, Rb and Cs replacing the Li and K ions. To eliminate unnecessary duplication of calculations, a program based on symmetry arguments is employed to identify equivalent configurations. Even so, hundreds of calculations were still needed to sample the complete set of inequivalent configurations of a wide range of solid solutions. Our simulations show that in the wide range of composition studied, solid solutions of the mixed KLiSi2O5 phyllosilicate with Na, Rb and Cs, retain the basic features of the KLiSi2O5 structure, e.g. a three- dimensional channel structure of six-membered rings made up of corner-sharing silica tetrahedral. However, the K ions rather than the Li ions are preferentially replaced by the guest ions, especially Rb and Cs. In the mono-cationic phyllosilicate Li2Si2O5, the main features are a two-dimensional six-ring structure and symmetric chair-like conformation of the silicate groups. We found that this structure became significantly distorted when the bigger Rb and Cs cations replaced lithium ions. Although solid solutions of all guest ions in silicates are energetically feasible to some extent, only ion exchange of K ions for Na ions from aqueous solution is calculated to be an exothermic process.

Molecular dynamics simulation of the water/α-quartz interface A.G. KALINICHEV1,2 AND R.J. KIRKPATRICK1,2 1

Department of Geology, University of Illinois, Urbana, IL 61801, USA ([email protected], [email protected]) 2 NSF Water CAMPwS, University of Illinois, Urbana, IL 61801, USA ([email protected], [email protected]) Interaction of water with oxide surfaces affects both the surface reactivity and functionality and the structure and dynamics of the near-surface aqueous phase. Quantitative understanding of these interactions is of great interest in many geochemical and technological systems. Silica is very important in this regard, because of its natural abundance and ubiquitous practical applications. To better understand the molecular-level structural and dynamical properties of water/α-quartz (0001) interfaces under different protonation states, we have performed MD computer simulations using the recently developed CLAYFF force field (Cygan et al., 2004). The effect of pH on the interfacial properties was emulated by varying the degree of surface protonation for the simulated models. For a fully protonated quartz surface (low pH), about 50% of surface OH groups are oriented parallel to the surface, form H-bonds to other surface OHs, and accept H-bonds from H2O molecules. The other 50% of OHs are oriented perpendicular to the surface and donate H-bonds to the interfacial H2O molecules. On average, all surface water molecules donate or accept H-bonds to/from the surface OH groups with equal probablility, and together they form a well interconnected Hbonding network similar to the ones observed for hydroxide interfaces (Kalinichev and Kirkpatrick, 2002; Wang et al., 2004). The structure and composition of the quartz surface imposes significant positional and orientational ordering on the H2O molecules nearest to the surface. This is qualitatively consistent with the results of recent sum-frequency vibrational spectroscopic measurements for the same system, where some observed spectral features were interpreted as being “ice-like” (Ostroverkhov et al., 2004).

References Cygan R. T., Liang J.J., and Kalinichev A. G. (2004), J. Phys. Chem. B 108, 1255-1266. Kalinichev A. G. and Kirkpatrick R. J. (2002), Chem. Materials 14, 3539-3549. Ostroverkhov V., Waychunas G. A. and Shen Y. R. (2004), Chem. Phys. Lett. 386, 144-148. Wang J., Kalinichev A. G., and Kirkpatrick R. J. (2004), Geochim. Cosmochim. Acta 68, 3351-3365.

Goldschmidt Conference Abstracts 2005 Molecular Computer Simulations

New developments of fast computational methods for first principles geochemical and geophysical simulations MARAT VALIEV1, ERIC BYLASKA1, KIRIL TSEMEKMAN2, STUART BOGATKO3 AND JOHN WEARE3 1

Pacific Northwest Laboratories, Richland, WA, 99352, USA ([email protected], [email protected]) 2 University of Washington, Seattle, WA, 98195, USA ([email protected]) 3 University of California, San Diego, La Jolla, CA 92093, USA ([email protected], [email protected]) Conventional methods of directly simulating the behavior of complex strongly interacting atomic systems (molecular dynamics, Monte Carlo) have provided important insight into the behavior of gases, fluids, and solids of geochemical and geophysical interest. The even wider application of these methods is limited by the difficulty of developing molecular level representations of potential interactions to capture complex chemistry commonly encountered in these systems (reactions, polarization, etc.). Static quantum chemistry methods have provided a means to calculate reactive mechanisms in cluster approximations to mineral systems. These methods are limited to small atomic sizes and generally cannot be applied to problems in which dynamics play a role. In this talk new developments in the implementations of methods to simultaneously simulate the electronic structure and molecular dynamics of nanoscale materials will be described (ab-initio molecular dynamics, AIMD). These methods, implemented into NWChem software, calculate of inter-atomic forces directly from the fast solution of DFT equations for very large systems and, therefore, avoid problems of force development limiting the application of MD. This talk will focus on the development and application in three areas: (1) the implementation the PAW method an all-electron plane-wave method; application to the solvation structure of transition metal ions in solution; (2) the development of a QM/MM method for simulations of large systems; application: simulation of enzyme reactions; (3) the development of a plane wave implementation of exact exchange; application: the localization of charge in hematite. The talk will focus on the fundamentals of these methods and the realities in terms of system size, computational requirements and simulation times that are required for their application.


Goldschmidt Conference Abstracts 2005 Nanogeoscience

Accurate multidisciplinary identification of nanophase iron minerals in simulated pedogenic environment 1




Nanoparticle-mediated processes and the Ostwald step rule ALEXANDRA NAVROTSKY 3


Department of Geology & Geophysics and the Institute for Rock Magnetism, University of Minnesota-Twin Cities, Minneapolis, MN, U.S.A. ([email protected]) 2 Department of Chemistry, University of Minnesota-Twin Cities, Minneapolis, MN, U.S.A. ([email protected]) 3 Laboratoire des Sciences du Climat et de l'Environnement (UMR 1572, CEA/CNRS), Gif-sur-Yvette, France ([email protected]) 4 Department of Marine Biology, University of Southern California, Los Angeles, CA, U.S.A. ([email protected]) We have begun an inter-disciplinary investigation (solid state chemistry, low temperature magnetism, Mössbauer effect and microbiology) to test chemical reactivity and environmental controls in simulated pedogenic transformations of nanophase (1-100 nm) iron oxyhydroxides and oxides of controlled size, crystallinity and impurity, in the presence or absence of iron-reducing bacteria (IRB). Pedogenic nanophase magnetite/maghemite end-products found in modern and ancient soils developed over loess (windblown silt deposits) parent materials in temperate climates are currently assumed to be proxies of annual rainfall (alone). We are testing and calibrating this abiotic model as well as comparing it with IRB-mediated magnetite formation since the latter (microbial type and concentration) could represent length of summer season and hence provide a new parameter, seasonality, for numerical climate model benchmarks. Our research to date has led to a comparison of known synthesis techniques to produce monodisperse ferrihydrite (Fh) of different crystallinity, nanogoethite (Gt) and magnetite (Mt).High resolution transmission electron microscopy (HRTEM), x-ray absorption fine structure (XAFS) and Mössbauer spectra were used for sample characterization. Magnetometry and Mössbauer spectra at low temperatures (to 2.5K) and high field (to 5Tesla) have allowed cross-checked grain-size determinations. These two latter methods, used in conjunction, have proven to be rapid and reliable tools for granulometry, and for discovery of oriented aggregation in grain growth, adsorbed species on the surface and for core/shell distinction in these nanophase materials. Examples of data from iron oxyhydroxide precursors and their microbial conversion product of nanophase magnetite will be presented.

NEAT ORU & Thermochemistry Facility, University of California, Davis, CA 95616, USA ([email protected]) There is increasing recognition that many processes which involve crystallization or phase separation do not occur by classical nucleation and growth but rather by the aggregation of pre-existing nanoparticles of 2-3 nm size. Such particles may, because of a competition between the energetics of polymorphism and surface energetics, have structures of phases which are metastable when coarse-grained. Furthermore, within a given composition, it appears generally true that the more metastable a structure in the bulk, the lower its surface energy. The classical Ostwald step rule observes that, during crystallization, a series of metastable phases is often visited on the way down to the final appearance of the most stable crystals. As the figure shows, the correlation between metastability and decreasing surface energy may offer a rationalization of the Ostwald step rule. Nanoparticles of metastable phases may be more stable thermodynamically, as well as having lower energy barriers and smaller critical nuclei, than nanoparticles of the final stable phase. The crossover in stability between polymorphs may occur at much larger particle sizes than the “critical nucleus” for crystallization. nanoclusters

Free energy (schematic)


critical nucleus or cluster for assembly

nanoparticles bulk phases species in solution or melt

metastable polymorph stable polymorph

Particle radius

Goldschmidt Conference Abstracts 2005 Nanogeoscience

Antiferromagnetic nanoparticles 1,2






Department of Physics, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark 2 School of Conservation, DK-1263 Copenhagen, Denmark Nanoparticles of antiferromagnetic materials such as αFe2O3 (hematite), α-FeOOH (goethite) and ferrihydrite, commonly found in geological environments, have attracted limited attention in magnetic studies, because the sublattice magnetizations are aligned antiparallel such that the particles may have negligible magnetic moments. A number of studies have, however, revealed that antiferromagnetic nanoparticles have a wealth of fascinating magnetic properties, e.g. a magnetization due to uncompensated spins with implications for their rock magnetic signature [1]. Recently we proposed that the thermal energy may excite the magnetic structure of antiferromagnetic nanoparticles such that a thermoinduced magnetic moment occurs [2]. This is a novel type of nanomagnetism where the magnetization increases with temperature. At room temperature the thermoinduced magnetic moment can be similar in magnitude to that originating from uncompensated spins. Magnetic dipole interactions between antiferromagnetic nanoparticles can be considered negligible [3-5] despite magnetic moments from e.g. uncompensated spins, but strong exchange interactions can be established between surface atoms of neighboring particles by drying aqueous suspensions of particles [3-5]. This interparticle exchange interaction significantly influences the properties of the individual particles [3-5] e.g. it suppresses superparamagnetic relaxation. In samples with strong interactions, there is a tendency for oriented attachment of the particles [4]. It is possible that exchange interaction can act as a driving force for the attachment. The agglomeration process is reversible, in the sense that ultrasonic treatment or grinding can separate the particles and reinduce fast superparamagnetic relaxation without reducing particle sizes [4,6]. Thus the nanoscopic coupling between particles is sensitive to simple macroscopic treatments. The results stress that the properties of nanoparticles, in addition to differing from bulk properties, have to be described in terms of interactions.

References [1] Robinson P. et al., (2002) Nature 438, 517. [2] Mørup S. and Frandsen C., (2004), Phys. Rev. Lett. 92, 217201. [3] Frandsen C. and Mørup S., (2003), J. Magn. Magn. Mater. 266, 36. [4] Frandsen C. et al., in prep. [5] Frandsen C. and Mørup S., Phys. Rev. Lett. (accepted). [6] Xu M. et al.. (2004), J. Colloid Interface Sci. 279, 132.


In situ observation of thermodynamic size effects on melting of natural gold nanoparticles M. REICH1, S. UTSUNOMIYA1, U. BECKER1, L.M. WANG2,3 AND R.C. EWING1,2,3 1

Dept. of Geological Sciences, University of Michigan, Ann Arbor, MI, USA ([email protected]) 2 Dept. of Materials Science and Engineering, University of Michigan, Ann Arbor, MI, USA 3 Dept. of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, USA Thermodynamic and quantum size effects, i.e. the change in physical and electronic properties as size is decreased to the nanoscale, are widely documented for synthetic nanomaterials. However, there is little information on how natural nanocrystalline materials behave under a wide range of geological temperatures. In order to evaluate size effects on melting of metal nanoparticles, we have performed controlled heating experiments of natural gold nanoparticles in As-rich pyrite (formed at ~150-200°C) under HAADF-STEM observation. In-situ heating in the TEM revealed no changes in the initial size distribution until ~350°C, where the smallest nanoparticles (<2 nm) start to melt. The most dramatic changes in nanoparticle size and distribution occur between 400-500°C; nanoparticles with diameters <2 nm melt completely at ~440°C, while at temperatures above 500°C, only nanoparticles >8 nm in size are stable. Computer-aided HAADF image analysis of size distributions as a function of temperature reveal that the initial average diameter (~4 nm) of Au nanoparticles is constant until ~350°C, above which it starts to increase gradually. The increase in average size is coupled with a significant decrease in the number of nanoparticles as temperature is raised above 400°C. At 600°C, the upper temperature bound of the experiment, only 3 nanoparticles (~25 nm) have survived, replacing the initial 115 of average size ~4 nm. During heating, larger nanoparticles (~9 nm) grow at the expense of the smaller ones (<2 nm) although no coalescence is observed. Analytical results reveal that a significant fraction of the initial size distribution of Au nanoparticles (~2-8 nm) is unstable at temperatures above ~350°C, confirming thermodynamic size effects on melting in natural samples. As a result, the preservation of metallic gold nanoparticles in the geologic record is size-dependent, and restricted to lower temperature (<400°C) hydrothermal systems.

Goldschmidt Conference Abstracts 2005 Nanogeoscience


Reactivity of iron oxyhydroxide nanoparticles with heavy metals as a function of particle size 1





Department of Physical Sciences, Chapman University, Orange, CA, USA ([email protected]) 2 Earth Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA ([email protected]) Nanoparticles of inorganic mineral phases are widespread in aqueous environmental systems and can play a significant role in the natural cycling of heavy metals and semi-metals. In particular, iron oxyhydroxides are present in many natural environments as dispersed and/or aggregated nanoparticles covering a wide range of sizes. However, the effects of particle size on heavy metal uptake, especially during nanoparticle aggregation and growth, are not well understood yet hold significant implications for contaminant sequestration and mobility. To study the effects of particle size on metal uptake, a series of iron oxyhydroxide batches ranging from 5-80 nm in average diameter was synthesized using a rapid microwave technique followed by aging in suspension at 90°C. Selected iron oxyhydroxide nanoparticle suspensions of 5, 25, and 75 nm in effective diameter were then exposed at pH 6 to 0.5 mM concentrations of As(V), Cu(II), Hg(II), and Zn(II), metal(loid) contaminants frequently associated with acid mine drainage systems. EXAFS spectroscopy of the resulting solids shows that while metal speciation on the 25- and 75-nm particles is identical (suggesting direct inner-sphere bonding), uptake to the 5-nm particles displays slightly longer (0.1-0.3 Å) secondneighbor metal-iron distances. This suggests distortion and/or disordering of the Fe(O,OH)6 octahedra which comprise the structure of iron oxyhydroxides. It is also consistent with the oblong morphology and greater degree of surface curvature observed in the 5-nm particles through TEM microscopy, with particles becoming more tabular/acicular as particle size increases (25-, 75-nm). Macroscopic uptake curves were generated using the same nanoparticle batches and metal contaminants over a concentration range of 5-1000 µM. Results show that the smallest particles exhibit greater overall uptake (due to higher surface areas) yet have reduced surface loadings (expressed in µmol/m2) relative to larger particles. Thus there are observable differences in both the extent and mode of metal uptake onto iron oxyhydroxides which appear to be largely dependent on size and morphology at the nanoscale.

Nanoscience meets geochemistry: Size-dependent reactivity of hematite A.S. MADDEN AND M.F. HOCHELLA, JR. Department of Geosciences, Virginia Tech, Blacksburg, VA, 24061, USA ([email protected], [email protected]) Geological materials at the nanoscale, typically considered to be approximately 1-100 nm, are extremely common. Nanoscience explores how the properties of nanoscale materials change as a function of size. Consequences of these size-dependent property changes for the geochemical reactivity of nanoscale materials in the environment remain almost entirely unexplored. Results will be presented involving the size-dependent reactivity of synthetic (7 nm, 9 nm, and 40 nm average diameter), commercial (NANOCAT 3 nm, MACH-1, Inc.), and natural (150-250 nm ground fraction) hematite and hematite-like materials. In this study, the reactivites of these particles is tracked using heterogeneous manganese oxidation rates and photochemical reactivity. Dramatic size-dependent reactivity is observed and rationalized in both systems. Initial rates of heterogeneous manganese oxidation on synthetic 7 nm, 9 nm, and 40 nm hematite in the presence of oxygen decrease by approximately 1-2 orders of magnitude as the average particle diameter increases, when normalized to surface area. This size-dependent rate change is hypothesized to result from changes in metal coordination environment and surface chemistry. In additional experiments, photochemical release of ferrous iron has been measured experimentally upon UV illumination in pH 4.0 oxalate (0.005 M) solutions. Surface area normalized release of ferrous iron from synthetic 40 nm and natural ground 150-250 nm hematite are very similar. Release was approximately two orders of magnitude greater for 3 nm NANOCAT material and approximately one order of magnitude greater for 7 nm and 9 nm synthetic hematite. Previous researchers have found photochemical reduction of bulk hematite to be relatively inefficient; electron-hole pairs recombine after an average diffusive length of a few nanometers. As the dimensions of the particles approach this length, trapping of electrons at the particle surface (as Fe2+) is expected to compete significantly with recombination.

Goldschmidt Conference Abstracts 2005 Nanogeoscience

Reactivity of ferrihydrite nanoparticles prepared with and without added carbonate, arsenate, and other oxoanions R. LEE PENN, AMY J. ANSCHUTZ, TERESA JENTZSCH AND JASMINE ERBS University of Minnesota, Department of Chemistry, 207 Pleasant St. SE, Minneapolis, MN 55406, USA ([email protected], [email protected], [email protected], [email protected]) Iron oxides and oxyhydroxides are common and important materials in the environment. These materials commonly occur as nanoparticles in the 3-10 nm size range and strongly impact the biogeochemical cycle of iron and other species at the Earth's surface. Surface-area normalized rates of reduction of ferrihydrite nanoparticles using hydroquinone in batch experiments enables comparisons of reactivity as a function of preparation conditions. We have used this system to characterize the reactivity of several different ferrihydrite samples. In general, ferrihydrite was synthesized by precipitation from homogeneous solution. Surprisingly, ferrihydrite nanoparticles prepared using sodium bicarbonate are substantially more reactive than ferrihydrite nanoparticles prepared using sodium hydroxide. Furthermore, the reactivity of microwave-heat treated ferrihydrite nanoparticles (after the method of Knight and Sylva, 1974) is greater than untreated ferrihydrite nanoparticles. Ferrihydrite nanoparticles prepared by coprecipitation with sodium arsenate are substantially more reactive than ferrihydrite nanoparticles that are equilibrated in solutions containing sodium arsenate, and both arsenate containing samples are substantially less reactive than arsenate-free ferrihydrite nanoparticles.


The structural chemistry of hydroxyl moieties in ferric polymers LAURA F. HARRINGTON1 AND SATISH C.B. MYNENI1,2 1

Department of Geosciences, Princeton University, Princeton, NJ, USA ([email protected]) 2 Earth Sciences Divison, Lawrence Berkeley National Laboratory, Berkeley, CA, USA ([email protected])

Introduction and Objectives Ferric (oxy)hydroxy polymers are ubiquitous in surface environments and play an important role in several geochemical processes. They are short-lived intermediates produced as aqueous Fe(III) hydrolyzes to form stable crystalline Fe(III)-(oxy)hydroxides. Our goal is to examine the short range order in ferric polymers, the chemical variables that influence their structure and formation, and the relationship between structure and reactivity. The bridging and terminal hydroxyls in ferric polymers hold the key to their reactivity in the environment. Because it is highly sensitive to small changes in hydroxyl coordination, infrared (IR) spectroscopy was used to interpret the structure of hydroxyls in ferric polymers based on well characterized crystalline Fe(III)-(oxy)hydroxides.

Discussion of Results Our investigation focused on the hydroxyl stretching and bending vibrations and their variations as a function of pH, ligand type, and reaction time. Since ripening of ferric polymers into nano-crystalline, and ultimately into macrocrystalline phases, is slower at room temperatures, initial studies were conducted at elevated temperatures (50-90oC). Freshly prepared ferric polymers from the hydrolysis of Fe(III) in the pH range of 3-7.5 exhibited a broad band around 900 cm-1, corresponding to the Fe-OH bending vibrations. In time (0-144 hrs), this band gradually evolved into two distinct bands centered around 900 and 800 cm-1, similar to those observed for goethite. Ferric polymers aged in the presence of NO3- and Cl- exhibited the greatest maturation rate at high pH. Conversely, the presence of SO42- not only accelerated polymer aging relative to the NO3- and Cl- systems, but also maximized the maturation rate at low pH. Ligand concentration associated with the polymers also changed with gel maturation. A mechanistic explanation for this disparity will be the focus of future research.

References Bigham, J.M. et al. (2002). SSSA Book Series 7, 323-366. Brinker, C.J. and Scherer, G.W. (1990). Sol-Gel Science Academic Press: New York.

Goldschmidt Conference Abstracts 2005 Nanogeoscience



Computational subcolloidal mineralogy

Controls of step length and direction on crystal solubility




Department of Geology, and Land, Air, and Water Resources, University of California, Davis, One Shields Avenue, Davis, CA 95616, USA ([email protected], [email protected])

Computation is now well established as a primary means of discovering and understanding chemical reactivity in aqueous and interfacial systems. Progress has not moved as quickly in the Earth sciences as it has in molecular biology because low-temperature Earth materials are so difficult to characterize. An important aspect of nano-Earth science is that while “small” is not necessarily different, “small” may be more precisely characterized. Recent advances in research on nanometer-sized aqueous poly(hydr)oxocations provide a particularly well defined path toward understanding surface reactivity in aqueous oxide systems. It has been possible to crystallize such ions as salts and determine their structure with single-crystal x-ray diffraction. These salts may then be redissolved in solution while preserving the complex ion for reactivity studies. These ions can form an unusually tight link with computer simulation. This synergy is discussed in two case studies. The first of these involves water and hydroxide ligand exchange reactions in the ε-(Al, Ga, Ge)A112 “Keggin” ion system. We find that the exchange mechanism is highly cooperative and not at all similar to mechanisms derived from the studies of octahedral substitution in aquo ions, which are often used in conceptualization of ligand exchange processes on oxide surfaces. For example, activation volumemechanism relationships are completely opposite to those known for aquo ions. The second case study involves “in silico” acidometric titrations of model poly(hydr)oxocations in aqueous solutions. Here the reactivity of the Keggin MO4(OH)24(H2O)127+ structure is contrasted against that of the larger M30O8(OH)56(H2O)2618+ structure. The M30 structure has six triply-bridging hydroxide functional groups which might be expected to make the M30 ion more acidic than the M13 ion. The enhanced acidity of the M30 ion is indeed predicted by the model, but the contributing functional groups are unexpected and do not correlate at all with gas-phase acidity.

Department of Chemistry, The George Washington University, Washington, DC 20006, USA ([email protected]) Solubility’s direction dependence is a fundamental property of anisotropic materials. For example, thermodynamics dictates that chemical potential necessary to maintain the growth of an individual crystal differs amongst faces that have different surface energies. This leads to the conclusion that multiple solubilities exist on anisotropic crystals where faces unrelated by any symmetry operations are present. Consider a step in the ith direction, the directiondependent solubility can be expressed by (1) lnKi=lnKsp+(2Mw/3RTρ)[γi/l] where Ki and Ksp are step solubility and standard state solubility product, Mw and ρ are molecular weight and density, γi the surface energy of the step riser, l the step length, and RT have their usual thermodynamic meaning. Eq. (1) states that Ki is always greater than Ksp. However, a closer examination of the ln(Ki/Ksp) ~ l relationship reveals that all direction-dependent solubility converges to Ksp when l is infinitely large (in reality, a length of a few µm is sufficient). While this warrants the rationality of approximating solubility by Ksp in traditional studies of crystal growth and dissolution for larger crystals, it implies that direction-dependence cannot be ignored when dealing with nano-crystal growth/dissolution. In situ AFM experiments conducted on the cleavage face of calcite show that the < 441 > ± steps indeed exhibit direction and length dependent behaviour at nm to sub-µm scales. Whereas advance is observed in steps of µm-lengths in both directions at near equilibrium conditions defined by Ksp, nano steps (10-100 nm) do not grow until the solution saturation is up to 50% higher than that defined by Ksp. Furthermore, simultaneous growth and dissolution at the two directions are observed in a narrow range of saturation, providing a direct proof for the existence of direction dependent solubility in calcite.

Goldschmidt Conference Abstracts 2005 Nanogeoscience

Nanogeochemistry: Geochemical reactions in nanopores YIFENG WANG1 AND HUIFANG XU2 1

Sandia National Laboratories, P.O. Box 5800, Albuquerque, New Mexico 87185-0776, USA 2 Department of Geology and Geophysics, University of Wisconsin, Madison, WI 53706, USA Nanopores are ubiquitous in porous geologic media and constitute an integral part of total porosity of rocks. Existing data indicate that the contribution of nanopores to the total surface area in geologic materials can be very high, probably over 90 percents. To clarify the effect of nanopore confinement, acid-base titration and metal adsorption experiments were performed on both nanoporous alumina and alumina particles under various chemical conditions. The experiments have demonstrated that the nano-scale confinement has a significant effect, most likely via the overlap of the electric double layer, on ion sorption onto nanopore surfaces. Under the same chemical conditions, the surface charge per mass on nanoporous alumina was as much as 45 times higher than that on alumina particles. The nanopore confinement leads to a shift of ion sorption edges and enhances ion sorption on nanopore surfaces for both cations and anions. As a result, trace elements in natural systems tend to be preferentially enriched in nanopores. This effect cannot be adequately modeled by existing surface complexation models. To understand the state of water in nanopores, the experiment of water adsorption under various relative-humidity conditions were performed in combination with H magic-angle spinning nuclear magnetic resonance (MAS NMR) and Fourier transform infrared (FTIR) spectroscopic analyses. The experimental results show that water molecules in nanopores form more rigid structures than those in bulk solutions. The observed nanopore confinement effects will shed a light on many important geochemical phenomena.


Relationship between Freundlich equation constants for zinc sorption on nanocrystalline calcite PETER BALÁŽ AND ANDREA ALÁČOVÁ Institute of Geotechnics, Slovak Academy of Sciences, Watsonova 45, 043 53 Košice, Slovakia The solid state properties of nanocrystalline particles obtained by high-energy milling with a high surface area has led recently to their utilization as reactive minerals and heavy metal sorbents [1]. The nanoparticles have been shown to possess a much greater number of defect sites per unit surface area, which are believed to be responsible for the observed chemistry [2,3]. The aim of this work was to examine the sorption behaviour of zinc ions on calcite whose solid state properties were modified by high-energy milling. Freundlich equation has been applied for the description of sorption process. A good relationship has been found between the Freundlich equation constants logk and 1/n for zinc sorption on 16 samples of nanocrystalline calcite modified by highenergy milling (Fig.1). Figure 1 Freundlich isotherm

The different solutes applied for the adsorption on one sample of carbon described in paper [4] and the different calcites (as for origin and solid state properties) applied for the adsorption of one solute (zinc ions) in our work has led to the same character of relationship between constants of Freundlich equation which demostrates its universal character.

References [1] Baláž, P. (2000) Extractive Metallurgy of Activated Minerals, Elsevier, Amsterdam [2] Banfield, J.F., Navrotsky, A., Eds. (2001) Nanoparticles and the Environment, Reviews in Mineralogy and Geochemistry, Vol. 44, Washington [3] Klabunde, K. J.: (2001) Nanoscale Materials in Chemistry, Wiley-Interscience, New York [4] Abe, I., Hayashi, k., Hirashima, T.: Am. Chem. Soc. 1982, 104, 6452-6453

Goldschmidt Conference Abstracts 2005 Nanogeoscience


Iron-iron oxide core shell nanoparticles for contaminant underground water treatment

Thermodynamics of high temperature iron oxide nanoparticles obtained by laser pyrolysis



([email protected], [email protected], 3 [email protected], [email protected], 5 [email protected], [email protected], 7 [email protected], [email protected], 9 [email protected]) Metallic iron chemically reduces contaminants such as chlorinated hydrocarbons for underground water cleanup. The use of nano-sized particles of iron is of interest because of enhanced possibilities for distribution, a high rate of reactivity and the possibility enhancing environmentally friendly reaction paths. An oxide shell or other protective layer plays an important role along with the metallic iron core in chemical reactions. Examination of the chemical properties of monodispersed metallic iron nanoparticles with a well-defined clean oxide shells is important to understand the chemistry of nanoparticles. However, production of these particles by conventional methods is difficult. Therefore, we are using a cluster deposition system, which prepare the iron nanoparticles and iron-iron oxide core shell nanoparticles at room temperature by a method that is a combination of highpressure sputtering and aggregation techniques. The outer oxide layer acts as a passivation layer of these particles, preventing further oxidation of the cores upon continued exposure to the atmosphere. Size of the iron-iron oxide coreshell nanoparticless can be varied with the rate of He and Ar gas flow, the chamber pressure, the sputtering power and the growth distance. These films are characterized by XPS, XRD, HRTEM and voltametry measurements. Fig. 1TEM images of Fe/FeO core-shell nanoparticles

References M.L. Támara et al., (2004). Environ. Sci. Technol. 38,18661876 J. Nurmi et al., (2004).Environ. Sci. Technol. ASAP Article. T.L. Johnson et al., (1998). J. cont hydro. 29, 379-398


Thermochemistry Facility and NEAT ORU, University of California at Davis, One Shield Avenue, Davis, CA 95616-8779, USA ([email protected]) 2 Materials Science Institute of Madrid. C/ Sor Juana Inés de la Cruz SN, 28049, Cantoblanco, Madrid, Spain ([email protected]) 3 Thermochemistry Facility and NEAT ORU, University of California at Davis, One Shield Avenue, Davis, CA 95616-8779, USA ([email protected]) Pure and uniform γ-Fe2O3 nanoparticles particles between 2 and 15 nm have been prepared at high temperature by a continuous process based on ethylene sensitised laser pyrolysis of gaseous Fe(CO)5 in an oxidizing atmosphere, suggesting that γ-Fe2O3, rather than α-Fe2O3, is the thermodynamically stable phase for ultra small iron oxide particles. The samples were characterised by X-ray diffraction (XRD), transmission electron microscopy (TEM), BrunauerEmmet-Teller gas adsorption (BET), thermogravimetric analysis (TG), elemental analysis and Mössbauer spectroscopy. The thermodynamic properties of the iron oxide samples were determined by transposed temperature drop calorimetry and high temperature drop solution calorimetry using sodium molybdate solvent at 700º C. The crystal structure of the iron oxide ultra fine particles potentially depends not only on the particle size, but also on the thermal history of the particle formation. In order to explore the effect of the thermal history of the particles, the thermodynamic properties of the obtained materials were compared with the properties of γ-Fe2O3 nanoparticles obtained by chemical solution techniques.

Goldschmidt Conference Abstracts 2005 Nanogeoscience

Migration of geogas-carrying gold nanoparticles in Quaternary sediments CAO JIANJIN






Department of Earth Sciences, Sun Yat-sen University, Guangzhou 510275, China ([email protected]) 2 Key Laboratary of Ore Deposit Geochemistry, Institute of Geochemistry, the Chinese Academy of Sciences, Guiyang 550002, China 3 Graduate School of Chinese Academy of Sciences, Beijing 100039, China 4 Institute mineral Resources, Chinese Academy of Geological Sciences, Beijing 100037, China The ascending geogas from the Earth’interior can capture nanoparticles from the ores during passing through ore bodies and deliver them to the surface. Hidden ore bodies may be detected by examing the geogas-carrying nanoparticles in the top of Quaternary sediments [1]. In this study, we investigated the migration behaviors of gold nanoparticles in the Quaternary sediments by means of indoor simulation experiments and field tracer tests. The gold nanoparticles at 20 nm were used in the study. The indoor simulation experiments were designed in which the ascending gas flows with gold nanoparticles flowed through the sediments. The field tracer tests were performed in Shuntian town, Heyuan county, northeast Guangdong Province, China. The gold nanoparticles were laid at the bottom of the sediments and the tests ran for five months. The samples of both the indoor simulation experiments and the field tracer tests, including sediments and particles collected from the gas in the sediments, were analysed by means of instrumental nuclear activation analysis (INAA) and transmission electron microscopy (TEM). The results show that: 1 In the geogas, gold nanoparticles migrate by absorbing on the surface of other particles, or in the form of groups; 2 The gold nanoparticles in the geogas are absorbed by kaolinites, illites, montmorillonites, hematites, goethites, gibbsites, etc., during passing the Quaternary sediments; 3 Gold nanoparticles exist in both gas and solid phases of the Quaternary sediments, and their distribution between the two phases is related to mineral commpositions and pHs of the Quaternary sediments.

Acknowledgement The project supported by National Natural Science Foundation of China, Grant No. 40273027.

Reference [1] Wang Xueqiu, et al., (1999) J. Geoche. Explor. 1-2, 85–97.


Effects of cobalt on oxide film formation on manganese carbonate YOUNG-SHIN JUN AND SCOT T. MARTIN Division of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138, USA ([email protected], [email protected]) The fate and the transport of metal contaminants in natural waters are linked with manganese oxide coatings. In the current work, cobalt adsorption and co-precipitation with manganese oxides on manganese carbonate is studied by atomic force microscopy at circumneutral pH. We also investigate how the cobalt ion affects the dissolution of manganese oxide film and MnCO3. In the presence of O2(aq), crystallization begins as islands that expand laterally to grow into a film across the surface. The islands have flat tops and are rhombohedral with a uniform thickness of 2.4 nm in the absence of added Co2+(aq). In contrast, a multilayer structure with apparently unrestrained z-directional growth forms in the presence of Co2+(aq). The net macroscopic dissolution rate slows by a factor of two in the presence of Co2+(aq). The normally rhombohedral shape of the dissolution pits of MnCO3 is distorted in the presence of cobalt, and Co2+(aq) preferably associates with the obtuse steps. Cobalt also affects the dissolution rate of the Mn oxide film. Dissolution of the Mn oxide layer is observed simultaneous to the formation of a second oxide film. The observations are explained by the parallel oxidation of Co2+(aq) and reduction of MnIII/IV on the surface. Under this model, a MnxCoyOz composition is suggested for the second oxide film. The cobalt adsorption rate (mol m-2 s-1) on the surface increases with the initial [Co2+](aq) and with increasing Mn oxide film coverage. The timing of cobalt addition affects film growth. When no Co2+(aq) is added initially, rhombohedral Mn oxide islands nucleate by heterogeneous oxidation on the MnCO3 substrate and film growth continues until the entire surface is covered. Adding Co2+(aq) disrupts growth of the Mn oxide film and results in the formation of a second MnxCoyOz phase. When Co2+(aq) is added at the beginning of the experiment, growth of the MnxCoyOz phase is predominant, and no evidence of the Mn oxide film is observed. Direct microscopic observations of the interactions of cobalt with manganese oxide coatings provide new mechanistic insights that are important in the quantitative modeling of the fate and the transport of toxic metals in the environment.

Goldschmidt Conference Abstracts 2005 Nanogeoscience


Reduction of halogenated groundwater contaminants by nano-sized magnetite Department of Civil and Environmental Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USA

Introduction Permeable reactive barriers have been shown to be a viable technology for remediation of halogenated solvent contaminated groundwaters. When the zero-valent iron used in these barriers corrodes under anoxic conditions, the corrosion film that forms at the iron-water interface is often composed of nanocrystalline corrosion products such as magnetite (Fe3O4). This magnetite has the potential to participate in the reduction of halogenated contaminants. The ubiquity and small size of nanoscale magnetite makes this a potentially important reactive phase in granular iron systems. Many nanosized materials exhibit reactivities that differ significantly from larger sized bulk particles. The objective of this study was to quantify the reactivity of nanoscale magnetite particles under anoxic conditions towards halogenated solvents and to determine if a relationship exists between reactivity and particle size.

Experimental Monodisperse suspensions of Fe3O4 were synthesized under anoxic conditions. Several batches of magnetite were synthesized under different solution conditions (pH, ionic strength, co-solutes) such that precise control of particle size was possible. Batch reactors of buffered magnetite (both nanosized as well as bulk magnetite) were spiked with halogenated solvents (carbon tetrachloride, trichloroethene) and monitored over periods of one week.

Results and Discussion The reduction of halogenated solvents occurred more rapidly when exposed to nanoparticulate (10-20 nm diameter) magnetite as compared to “bulk” magnetite particles (>100 nm). In addition, daughter products were more fully reduced by nanoparticulate magnetite than by bulk magnetite. This presentation will address the relationship between magnetite particle size and halogenated contaminant reduction capability. Additionally, this presentation will discuss the effects of oxidation on particle aggregation in nano-sized magnetite suspensions.

Thermochemistry Facility, University of California at Davis, One Sheilds Avenue, Davis, CA 95616, USA ([email protected]) Enthalpy of formation of akaganeite, ß-FeOOH, as a function of particle size is reported for the first time. Fourteen samples with the particle size range of 5-100 nm and surface area 30-280 m2/g were prepared. Charachterisation was performed using XRD, FTIR, BET, TEM, SEM and TGA. Solution enthalpy in hydrochloric acid at room temperature and in sodyum molybdate at 700 ºC was measured. Refinement of XRD data showed that at higher surface areas akaganeite has a monoclinic polymorph. Calorimetric data were corrected for excess water assuming this loosely adsorbed water has the same energetics as bulk liquid water. The plot of enthalpy of solution for monoclinic akaganeite vs surface area (see figure) gives a surface enthalpy of 0.33 ± 0.05 J/m2 and enthalpy of formation of bulk akaganeite of -557.3 ± 1.2 kJ/mol. Although surface energy of akaganeite is found to be lower than for lepidocrocite γ-FeOOH and goethite α-FeOOH, the amount of adsorbed water is higher especially for lower surface areas (see -39 figure). The reason is -42 probably the tunnel structure of akaganeite, -45 wth excess water enthalpy of solution -48 accumulating in the enthalpy of formation -549 tunnels. Preliminary calorimetric data -552 showed that the surface -555 energy of akaganeites -558 with lower surface 5000 10000 15000 20000 25000 areas (and probably a 1.0 different polymorph) is higher, but additional 0.8 work is required. The 0.6 energetics of adsorbed water on the akaganeite 0.4 surface is also under goethite investigation. lepidocrocite 0.2 akaganeite




excess water, x



Thermochemistry of bulk and nano akaganeite

0.0 0

5000 10000 15000 20000 25000 2

surface area, m /mol

Goldschmidt Conference Abstracts 2005 Nanogeoscience


Nano-crystalline osbornite from Carbonados: Spectroscopic studies

Incorporation of guest-molecules into natural zeolite mordenite




National Geophysical Research Institute, Hyderabad,India 2 Flourida International University, Miami, FL-33199, USA 3 Indian Institute of Chemical Technology, Hyderabad, India Carbonado has been recognised as a polycrystalline aggregate of diamond since 1840 and was mined as a placer mineral in Sincoro County in Brazil. Carbonado has been recovered only from alluvial deposits in Brazil and the Central African Republic (CAR). The origin of the unusual type of polycrystalline diamond is not yet understood. In order to understand the origin of this exotic diamond aggregates, we have carried out powder-X-ray diffraction, Nuclear Magnetic Resonance and FT-IR, and TEM studies on the carbonados from Brazil as well as CAR. While our data on the chemical and trace elements analyses agree with the earlier data, there is a minor difference in the observation of presence minor minerals. We have identified few grains of osbornite (TiN) with 50 to 100 nm size, and presence of lonsdaleite- a high – pressure polymorph of diamond. The presence of these two minerals has been independently and unambiguously confirmed by XRD. The NMR spectrum also show a shockinduced broadening of the diamond line centred at 34.8 ppm which is more than the chemical shift 34.5ppm of pure diamond crystal. Our observation of presence of osbornite in carbonado shows the possible extraterrestrial origin of carbonado, as this mineral is known to exist only in enstatite chondrites and achondrites.


Thermochemistry Facility, Department of Chemical Engineering and Materials Science, University of California at Davis, USA ([email protected]) 2 Laboratory for chem. and mineral. Crystallography, University of Bern, Switzerland ([email protected]) Microporous materials such as zeolites are more and more applied for the design of advanced nano-materials, e.g. miniaturized electronic devices. They are also a group of minerals with geologic and environmental importance. Despite kinetic and spectroscopic studies of guest incorporation, little is known about the exact positions, geometrical arrangement and disorder phenomena of encapsulated guest-compounds. Large synthetic mordenite single-crystals of good quality and suitable morphology were used for incorporation of semiconductor material (selenium) and organic dye molecules (thionin blue (C12H9N3S+), methylene blue (C16H18N3S+)). Single-crystal X-ray diffraction experiments were performed with a conventional in-house X-ray source as well as using synchrotron radiation at the Swiss Norwegian Beamline at ESRF, France. Complementary methods such as polarization and fluorescence microscopy were applied to analyze the orientation of the dye molecule transition dipole-moment. Guest-molecule arrangement in all three guest-host systems is highly influenced by electrostatic interaction with the mordenite framework and the present extraframework occupants (Na+, H2O). Elemental Se in mordenite builds chains, which show a highly variable geometrical arrangement and differ clearly from the regular chain geometry as occurring in elemental, trigonal selenium. The dye molecules show distinct occupational disorder along the channel axis and prefer an inclined arrangement within the channel crosssection resulting in short CH...O and NH...O contacts to the framework.

References Ozin G.A., Kuperman, and Stein A., (1989) Adv. Mater. 101, 373-390. Simoncic P., (2004), PhD thesis, University of Bern, Switzerland.

Goldschmidt Conference Abstracts 2005 Nanogeoscience


Genesis and mineralogical characteristics of hematite in

Mechanical, chemical, magnetic, transport, and electronic properties changes at the nanometer scale



loess-paleosol sequences of China 1

School of Natural Resource and Environment Engineering, Hefei University of Technology, China ([email protected]) 2 Department of geology and Geophysics, The University of Wisconsin, Madison, USA ([email protected] 3 Department of Geoscience, Nanjing University, China ([email protected]) Hematite is an important iron oxide mineral in loesspaleosol sequences of China, and it is benefit reconstruction of paleoclimate of Chinese inner terrene investigation and understanding mineral characteristics, forming process of hematite as well as relationship with other iron oxides and containing iron minerals. For example if we know on grain size distribution, exactly quantitative testing of hematite will be obtain by diffuse reflectance spectroscopy. Employed optic microscopy, X ray diffraction, scanning electron microscopy, high resolution transmission microscopy for original samples and magnetic extracted samples, obtaining results as follows: There are four kinds of mechanisms for hematite occurring in loess-paleosol sequences of China during pedogenic. (1) Weathering from containing iron silicate, example for chlorite, etc, which released free iron ion, hydrolyzed, and dehydration; (2) Oxidation from magnetite which is micrometers and wind dust origin; (3)Phase transformation from wind dust goethite to hematite by topomorphology; (4) Hematite forming on the edge of maghemite. The hematite forming from the first mechanism is the best important occurrence and size from several to tens nanometers, irregular morphology, low crystallinity, as well as characteristic mesoporous because of dehydration from iron hydroxide (example for ferrihydrite), which results in redness in paleosol.

Acknowledgements The work was supported by the Outstanding Overseas Chinese Scholars Fund of Chinese Academy of Sciences (2003-1-7), National Science Foundation of China (40331001, 40472026)


The University of Wisconsin, Madison, WI 53706, USA ([email protected]) 2 Sandia National Laboratories, Albuquerque, NM 87185, USA ([email protected]) Systems (such as materials, minerals, and fluids) at nanometer scale behave differently from their bulk systems. Nanoscience and nanotechnology are making big progress now because it is possible to purposefully manipulate materials and theoretically simulate / compute the properties of small systems at nano-scale. Deformation mechanisms for nanocrystalline metals will be dominated by twinning, grain boundary sliding, wide stacking faults, and partial dislocation emission from grain boundaries. Brittle oxide minerals will behave like elastic when their sizes reach the nanometer scale. Reactivity and stability of nanocrystals needs to be modified according to their crystal sizes, shapes, and textures. Geochemical reactions in nanopores and nanotube environments will be greatly different from those in bulk solution systems because of electric double layer overlap and water property changes (such as activity and dielectric constant). Especially the change in dielectric constant of water will modify Born solvation energies of ions, which will affect many aspects of geochemical processes, such as chemical weathering, ore deposit formation, replacement reactions, and fate of toxic metals in ground water aquifer. Nanocrystalline and nano-structured magnetite and maghemite that are superparamagnetic will enhance magnetic susceptibility of rocks and sediments dramatically. Darcy’s law for the transport behaviour of water in nanoporous media will deviate. The band gap between valence band and conduction band of nano-crystalline semiconductor minerals (such as CdS, and CdSe) and materials will be a function of their crystal dimensions. The electron densities of states of nanocrystalline semiconductors will be greatly different from their bulk system. This property has been used in many opto-electronic materials and devices.

Acknowledgement This research was supported by NSF (EAR-0210820).

Goldschmidt Conference Abstracts 2005 Nearly Nano-Compositional Imaging

Isotopic analysis of presolar dust grains with the NanoSIMS ERNST ZINNER Laboratory for Space Sciences and the Physics Department, Washington University, One Brookings Drive, St. Louis, MO 63130, USA ([email protected]) Although presolar grains isolated from primitive meteorites have been studied in detail for their isotopic compositions, most analyses have been limited to grains ≥1µm [e.g., 1, 2]. The NanoSIMS with its high sensitivity and high spatial resolution [3] makes it possible to analyze sub-micron grains, charasteristic of the size of interstellar dust [4]. This capability has resulted in several isotopic studies of grains that previously would not have been possible. One example is the analysis of presolar spinel grains from the Murray carbonaceous chondrite [5]. Oxygen isotopic measurements showed that the abundance of presolar grains ≤0.5µm is much higher (~2% of all spinel grains) than that of ≥1µm grains (~0.1%). Magnesium isotopic measurements of these presolar spinels revealed large 26Mg excesses [6]. Inferred initial 26Al/27Al ratios are much higher than expected from shell H burning in asymptotic giant branch (AGB) stars and require extra mixing (cool bottom processing). Another example is the discovery of presolar silicates in interplanetary dust particles (IDPs) [7, 8] and in primitive meteorites [9, 10]. These grains are ≤1µm and although their abundances are higher than those of most other presolar grain types identified so far (~180ppm in a primitive meteorite and ~890ppm in IDPs), only the analysis of many thousands of grains by high-spatial-resolution isotopic imaging has made their identification among an overwhelming majority of isotopically normal silicate grains of solar-system origin possible.

References [1] Nittler L. R. (2003) EPSL 209, 259-273. [2] Zinner E. (2004) in: Treatise on Geochemistry 1,17-39. [3] Stadermann F. J. et al. (2000) in: Beyond 2000-New Frontiers in Isotope Geoscience. [4] Mathis J. S. (1990) Ann. Rev. A&A 28, 37-70. [5] Zinner E. et al. (2003) GCA 67, 5083-5095. [6] Zinner E. et al. (2004) LPS XXXV, Abstract #1337. [7] Messenger S. et al. (2003) Science 300, 105-108. [8] Floss C. and Stadermann F. J. (2004) LPS XXXV, Abstract #1281. [9] Nguyen A. N. and Zinner E. (2004) Science 303, 14961499. [10] Mostefaoui S. and Hoppe P. (2004) Ap. J. 613, L149L152.


O- and S-isotope imaging of primitive solar system materials with the Mainz NanoSIMS P. HOPPE1, S. MOSTEFAOUI1 AND T. STEPHAN2 1

Max-Planck-Institut für Chemie, 55020 Mainz, Germany ([email protected]) 2 Institut für Planetologie, Universität Münster, 48149 Münster, Germany ([email protected])

Introduction Primitive meteorites and IDPs contain small quantities of nm- to µm-sized presolar grains that formed around evolved stars [1]. The invention of the NanoSIMS with its superior lateral resolution (<100 nm) and capability for the search of in-situ presolar dust in slices of IDPs [2] and meteorites [3] has opened a new window in this field of astrophysical research. Here, we will present results from NanoSIMS imaging surveys of O- and S-isotopic compositions in the Acfer 094 and Bishunpur meteorites and in two IDPs.

Acfer 094 and Bishunpur O-isotope mapping with lateral resolution of 100 nm was performed on the matrix in polished thin sections of Acfer 094 and Bishunpur. Based on large O-isotopic anomalies we identified 17 presolar silicate and 3 presolar spinel grains, 150 to 600 nm in size. The O-isotopic compositions (and Si- and Fe-isotopic compositions, measured on a subset of those grains) point to origins from RGB and AGB stars. Matrixnormalized abundances are estimated to be 15 ppm (Bishunpur) and 130 ppm (Acfer 094) for silicates and 50 ppm for spinel (Acfer 094). S-isotope mapping was done on selected areas of the matrix in Acfer 094. Automatic particle recognition revealed some 400 S-rich grains, two of which exhibit large negative 36 32 S/ S anomalies of ~3σ. Whether these anomalies are just statistical outliers or the signature of real presolar matter remains to be seen.

IDPs U2071J2 and U2071C9 O-isotopic mapping was performed on four microtome sections of IDP U2071J2 (9×5 µm2) and on seven microtome sections of IDP U2071C9 (12×8 µm2). No presolar grains were found. Upper limits for presolar grains in these two IDPs of several 100 ppm are inferred.

References [1] Nittler L. R. (2003) EPSL 209, 259. [2] Messenger S. et al. (2003) Science 300, 105. [3] Mostefaoui S. and Hoppe P. (2004) ApJ 613, L149.

Goldschmidt Conference Abstracts 2005 Nearly Nano-Compositional Imaging


Distinguishing solar and extrasolar origins of submicrometer grains in IDPs

NanoSIMS Mg isotope analyses of refractory inclusions in metal-rich CB chondrites



Robert M Walker Laboratory for Space Science, NASA Johnson Space Center, Code KR, 2101 NASA Parkway, Houston, TX 77058, USA ([email protected], [email protected]) Interplanetary dust particles (IDPs) are primitive solar system objects from comets and asteroids that contain abundant interstellar materials [1]. These materials include Dand 15N-rich organic matter formed in a cold molecular cloud environment and grains of stardust [2]. The presolar (stardust) grains are identified by their exotic isotopic compositions that significantly differ from solar isotopic compositions [3]. Silicates are the most abundant type of stardust, but because of their small size (< 1 µm) were only recently discovered. The Cameca NanoSIMS ion microprobe was key to this discovery. The NanoSIMS uses an ultra fine beam of (50 nm) Cs+ or (200 nm) O- and detects five mass lines simultaneously in imaging mode with high transmission at high mass resolving power. In practice, presolar grains as small as 200 nm can be distinguished in situ from solar system materials. Silicates are more abundant in IDPs (450 – 5,500 ppm) than in meteorites (< 130 ppm), possibly because they are from comets [3,4]. A key issue to be resolved is the true proportion of presolar grains and solar system materials. Much of the material in IDPs is too fine grained (< 200 nm) to distinguish between solar and extrasolar origin. Some presolar silcates are amorphous (GEMS; 5), and may be mixtures of stardust and deposited material. The problem is further compounded by recent compositional mapping of GEMS grains by field emission TEM, showing that many GEMS grains are composite objects [6]. The isotopic measurement of GEMS grains may thus represent an average of still smaller subgrains. Resolving this issue may require higher precision isotopic measurements as the NanoSIMS is near the limiting size scale accessible to SIMS.

References [1] J. P. Bradley, S. A. Sandford, R. M. Walker, in Meteorites and the Early Solar System, J. Kerridge and M.S. Matthews, Eds. (Univ. Arizona Press, Tucson, AZ, 1988), pp. 861-895 [2] Messenger S. Nature 404, 968-971 [3] S. Messenger, L. P. Keller, F. J. Stadermann, R. M. Walker, E. Zinner Science 300, 105 (2003). [4] A. Nguyen, E. Zinner, Science 303, 1496 (2004) [5] Bradley, J.P. (1994) Science 265, 925 [6] L. P. Keller and S. Messenger (2005) Lunar Planetary Sci., submitted.


Lawrence Livermore National Laboratory, Livermore, CA 94551, USA ([email protected]) 2 SOEST, Univ. of Hawaii, Honolulu, HI 96822, USA Hammadah al Hamra (HH) 237 is a metal-rich (CB) chondrite with unusual mineralogical and chemical signatures. HH237 appears to be a remarkably pristine relic of large-scale high temperature processes in the very early solar nebula [1]. We present Mg isotope analyses of refractory inclusions in HH237, collected with the LLNL NanoSIMS, demonstrating large mass-dependent isotope fractionation and excesses of radiogenic 26Mg. Mg isotope compositions were measured by static multicollection at a mass resolving power of ~4000, using a 20 pA, 250 nm 16O- primary ion beam rastered over a 4x4 mm area. Isotope ratios were calculated using the sample-standard bracketing approach [2]. Ca-Al-rich inclusion (CAI) HH237-s #1 is a compact CAI composed of grossite (CaAl4O7) surrounded by a melilite-pyroxene rim with enclosed metal blebs; HH237-MW #4 is a fine-grained, hibonite-rich CAI containing minor spinel, melilite and perovskite, surrounded by layers of melilite and Al-diopside; HH237-104 CAI#1 is a spinel-hibonite spherule containing hibonite laths surrounded by a spinel mantle and a melilite-pyroxene rim. NanoSIMS analyses show large mass-dependent fractionation in all three CAIs favoring the lighter Mg isotopes with FMg values ranging from -20 to -12 ‰/amu; 2σ is ~4‰/amu. Both hibonite-rich CAIs exhibit large 26Mg excesses, with initial 26Al/27Al ratios of (7.5±3.3) x 10-5, while the grossite-rich inclusion contains no excess 26Mg with 26Al/27Al<2x10-6. The large enrichment of the lighter Mg isotopes is unusual for igneous CAIs and, together with group II REE patterns [1], indicate many CAIs in CB meteorites preserve primary condensation signatures inherited from the nebular gas. The isotope fractionation effects appear decoupled from incorporation of 26Al. Additional refinement of initial 26Al abundances will elucidate the time scales of condensation and melting and test the hypopthesis grossite-rich CAIs formed before injection of 26Al into the solar nebula.

References [1] Krot, A.N. et al. (2002) MAPS 37, 1451-1490. [2] Galy, A. et al. (2003) J. Anal. At. Spectr. 18, 1352-1356.

Goldschmidt Conference Abstracts 2005 Nearly Nano-Compositional Imaging

Microscale D/H and C/H imaging of meteorites and IDPs – Calibration of ion microprobe data with terrestrial analogues and meteoritic residues H. BUSEMANN, C.M. O’D ALEXANDER AND L. NITTLER Department of Terrestrial Magnetism, Carnegie Institution of Washington, 5241 Broad Branch Road NW, Washington DC 20015, USA ([email protected]) Chondrites and interplanetary dust particles (IDPs) show large C contents and inhomogeneously distributed, strongly elevated D/H ratios compared to Earth and the solar system. The anomalies are mainly carried by organic matter that was formed in the interstellar medium and subsequently altered by parent body processing (Alexander et al., 1998). The carriers of the isotope anomalies have not unequivocally been identified, but µm-scale isotopic and elemental imaging by ion microprobe provides useful information. For example, quantitative D/H, C/H and 15N/14N imaging has been used to identify at least three distinct types of carbonaceous material in IDPs (Aléon et al., 2001, 2003). In general, the highest D/H ratios are associated with C, but high D/H ratios have also been observed in some C-poor regions of IDPs (Mukhopadhyay and Nittler, 2003). Precise C/H ratios are hence essential to distinguish the phases and ultimately identify the carriers of the various D/H signatures in IDPs. However, element ratios are difficult to determine with ion microprobes due to matrix effects depending on secondary yields. We have begun a systematic examination of the isotope and element fractionation of our Cameca 6f imaging system (Nittler and Messenger, 1998). Various well-characterized extraterrestrial and terrestrial organic samples with a large range of C/H and D/H ratios such as coals, aromatic and aliphatic hydrocarbons, hydrated minerals and organic residues of primitive meteorites will allow us to precisely establish calibration ratios for C/H and D/H, to assess possible matrix effects and to determine realistic uncertainties. Furthermore, imaging data for µm-sized matrix particles from a range of meteorite types will be presented in order to enlarge the database for both weathered and relatively unaltered primitive chondrites allowing for a comparison with IDPs and returned STARDUST samples.

References Aléon J. et al. (2001) GCA 65, 4399-4412. Aléon J. et al. (2003) GCA 67, 3773-3783. Alexander C.M.O’D. et al. (1998) M&PS 33, 603-622. Nittler L.R. and Messenger S. (1998) LPS XXIX, #1380. Mukhopadhyay S. and Nittler L.R. (2003) LPS XXXIV, #1941.


TOF-SIMS – A tool for sub-micrometer analysis in geo- and cosmochemistry THOMAS STEPHAN Institut für Planetologie, Wilhelm-Klemm-Str. 10, 48149 Münster, Germany ([email protected])

Introduction The analysis of micrometer and even sub-micrometersized samples has become more and more important in geoand cosmochemistry. Typical examples are individual interplanetary dust particles (IDPs), ~10 µm in size, that often consist of hundreds of smaller grains. Some of these phases are isotopically highly anomalous, presolar grains embedded in isotopically normal, solar matter. These grains provide us with information about stellar and interstellar processes that formed the building blocks of our solar system.

TOF-SIMS Time-of-flight secondary ion mass spectrometry (TOFSIMS) is one of the techniques that are ideally suited for these tiny, complex samples (Stephan, 2001). With TOF-SIMS major, minor, and trace elements, their isotopes, and molecules can be measured at a lateral resolution of ~200 nm. The main advantage of TOF-SIMS compared to doublefocusing ion microprobes is its detection of all secondary ions with one, selectable polarity in a single experiment. This is due to the time-of-flight principle where even the heaviest molecular ion eventually reaches the detector. Therefore, hydrogen, e.g., can be measured simultaneously not only with all elements that form ions with the respective polarity but also with the heaviest bio-molecules. Further major differences to other SIMS instruments are little sample consumption (typically only a few atomic monolayers are sputtered during a measurement) and high transmission (20– 80 %, depending on the ion species). However, count rates for individual ion species are rather low and statistical errors are often limiting the measurement accuracy, especially for isotope analysis.

Conclusion Whenever high lateral resolution is required, and preselection of ion species to be measured has to be avoided, especially for samples with unknown composition, TOF-SIMS is an extremely valuable tool that can provide a huge range of information with a minimum of sample destruction.

Reference Stephan T. (2001), Planet. Space Sci. 49, 859–906.

Goldschmidt Conference Abstracts 2005 Nearly Nano-Compositional Imaging


Hyperspectral data analysis of martian meteorite alteration: A tool for constraining surface processes on Mars? E.P. VICENZI AND D. ROST Department of Mineral Sciences, National Museum of Natural History, Smithsonian Institution, Washington, DC 205600119, USA ([email protected], [email protected])

Introduction Our information-base regarding in situ analysis of surface materials stems from a handful of spacecraft missions. Large areas of Mars are apparently coated with a layer of surface fines (dust) that are quite similar in composition, suggesting this material has been distributed globally. Knowledge of aqueous processes beneath the surface of Mars is severely restricted, however, preterestrially altered portions of the Martian meteorites offer an opportunity to directly examine low temperature precipitates. Some have pointed-out the similarities between the major element chemistry of micrometer scale alteration products and Martian dust [1]. In this study, we have analyzed Time-of-Flight Secondary Ion Mass Spectrometry (ToF-SIMS) data using multivariate statistical analysis (MVA) to compute the “local bulk composition” of secondary mineral veinlets to rigorously evaluate a chemical linkage to Martian dust. In doing so, we can also evaluate the possibility that Martian dust was processed aqueously.

Approach and Results Because the kinetics of low temperature reactions in the Martian crust are slow, the products have small grain sizes (µm to sub µm), or are finely zoned amorphous phases [2]. As a result, ToF-SIMS and a pulsed Ga+ ion beam (~300 nm) were used to characterize these delicate secondary mineral assemblages. One advantage of ToF-SIMS is that all secondary ions are collected (effectively in parallel) at each pixel, producing a 3D hyperspectral data cube. We have processed several such mass-resolved data cubes using MVA (e.g. multivariate curve resolution) to compute the major, minor, and trace element compositions of “bulk” alteration. Our initial findings, based upon Si/Fe for example, suggest that aqueously altered silicates in meteorite alteration are distinct from soils analyzed by MER rover instrumentation. These findings suggest that Martian dust likely formed by a process distinct from low-T aqueous alteration.

References [1] Lodders K. and Fegley B.J. (1997), Icarus 126, 373. [2] Vicenzi E.P. and Heaney (2000) MAPS. 35, A164-165.

Carbonate-silicate assemblages in the Lafayette martian meteorite D. ROST AND E.P. VICENZI Dept. of Mineral Sciences, National Museum of Natural History, Smithsonian Institution, Washington, DC 205600119, USA ([email protected], [email protected])

Introduction Some secondary mineral assemblages in Martian meteorites have long been interpreted as preterrestrial [1]. Poorly crystalline clay veinlets within nakhlitic olivines [1,2] and carbonate veinlets in Lafayette olivine [3-5] are probably the result of aqueous alteration in a near surface setting within the last 700 Ma [6]. Carbonate veinlets contain central silicates, zoned from the center outward, in the following way: amorphous Si-rich phase, amorphous Fe-rich phase, followed by poorly crystalline clays. Two microanalytical tools including full-spectrum X-ray mapping and time-of-flight secondary ion mass spectrometry (ToF-SIMS) were used to reveal the distribution of major, minor, and trace elements within these assemblages. ToF-SIMS has the unique capability of detecting a large suite of trace elements (e.g. Li, Be, B, Sc, V, Cr, Co, Ni, Cu, Rb, Sr, and Ba) in parallel on the sub micrometer scale, allowing one to evaluate different depositional models.

Results We examined five different carbonate-silicate assemblages in Nakhla olivine. The minor and trace element composition of different silicate phases is complex. Poorly crystalline clays: Most element abundances differ only within a factor of two, suggesting clays are similar in all carbonate veinlets. However, Be, Ti, Cr, Cu, and Ba show distinctively greater variability. Fe-rich amorphous material: Compositional variation is more restricted relative to clays, including Ti abundances. Si-rich amorphous material: This phase exhibits the greatest variability between veinlets. These findings are best explained by low and decreasing water/rock ratios during veinlet formation. Thus, the local mineralogy (100s of µm) seemingly was increasingly important in determining veinlet fluid chemistry.

References [1] Gooding J.L. et al. (1991), Meteoritics 26, 135–143. [2] Treiman A.H. et al. (1993), MAPS 28, 86–97. [3] Vicenzi E.P. and Heaney P.J.. (2000), LPS XXX, #2005. [4] Bridges J.C. and Grady M.M.. (2000), EPSL 176, 267– 279. [5] Rost D. and Vicenzi E.P. (2004), MAPS 39, A92. [6] Swindle T.D. et al. (2000), MAPS 35, 107–115.

Goldschmidt Conference Abstracts 2005 Nearly Nano-Compositional Imaging

Microbial C and N assimilation in soils and model systems as revealed by ToF-SIMS 1



Pacific Northwest National Laboratory, Richland, WA 99352, USA ([email protected], [email protected]) 2 Oregon State University, Corvallis, OR 97331, USA ([email protected], [email protected]) Our research has utilized sub-µm ion probes in conjunction with time of flight secondary ion mass spectrometry (ToF-SIMS) for studying C and N assimilation in soil microbes. Early proof of concept studies using 13C and 15 N labelled substrates clearly showed that C and N assimilation are detectable in single microbes using secondary ions of CN– isotopes. Because of practical limitations, high spatial resolution analyses are typically performed at low mass resolution. We were able to use a simple subtraction algorithm to remove 13C14N– form the total 27CN– peak and quantify organic 15N isotope ratios. We applied this algorithm to region of interest analyses of microbes adhering to Si contact slides from soils and model soil systems. These studies have shown that microsite heterogeneities consisting of concurrent assimilation of inorganic N and rapidly mineralizing-organic-N may exist separated by only tens to hundreds of microns.

Acknowledgements A portion of the research described in this abstract was performed at the W. R. Wiley Environmental Molecular Sciences Laboratory, a national scientific user facility sponsored by the U.S. Department of Energy's Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory. PNNL is operated for the Department of Energy by Battelle.


Applications of SIMS Microanalysis at NIST ALBERT J. FAHEY, CHRISTINE MAHONEY AND GREG GILLEN National Institute of Standards and Technology, Gaithersburg, MD, USA ([email protected], [email protected], [email protected]) Three Secondary Ion Mass Spectrometry (SIMS) instruments have been used for various microanalytical measurements applied to, or have the potential to be applied to geochemically interesting problems. Two of the instruments are dynamic SIMS magentic sector instruments, a Cameca ims-4f and an ims-1270, and one is an IonTof Tof-SIMS IV. In addition to the measurement of sub-micron particles and structures, various cluster primary ion beams have been explored to enhance signals from organic materials as well as to limit damage of sub-surface layers to enhance our ability to obtain ultra-shallow depth profiles and depth profiles of organic materials. Applications and measurements made with each of the instruments at NIST will be shown and discussed. Measurements of depth profiles, trace element abundances will be addressed, as well as isotopic ratios of natural minerals and anthropogenic materials. The importance of these measurements not only to geochemistry but to industry and international matters will be touched upon. Comparisons and limitations of each of the instruments will be addressed, specifically with respect to spacial resolution, both laterally and as a function of depth. Implications of signal intenstity for hyperspectral analysis from limitations of secondary ionization, increased lateral resolution, and depth resolution will also be discussed.


Goldschmidt Conference Abstracts 2005 The Robotic Exploration of Mars and Titan

The astrobiological aspects of Titan: A new vision from Cassini-Huygens

Biologically enhanced energy and carbon cycling on Titan?




LISA, CNRS & Université Paris 7 and 12, France ([email protected]) 2 Service d’Aéronomie du CNRS, Verrières-le-Buisson, France, ([email protected]) 3 NASA-GSFC, Greenbelt, USA, ([email protected]) 4 IFA, University of Hawaii, Honolulu, USA ([email protected]) Since the discovery of the presence of an active organic chemistry in its atmosphere, Titan is considered as a planetary body of prime astrobiological importance. The very first flybys of Titan by the Cassini spacecraft a few months ago, and the in situ exploration of its atmosphere and surface thanks to the Huygens probe, on January 14th 2005, are providing us new and spectacular data on Titan’s environment. Some of the astrobiological consequences of these new data will be presented and discussed.


Dept. of Geological Sciences, WSU, [email protected] Dept. of Space Sciences, Southwest Research Institute, [email protected]


With the Cassini-Huygens Mission in orbit around Saturn, the large moon Titan, with its reducing atmosphere, rich organic chemistry, and heterogeneous surface, moves into the astrobiological spotlight. Environmental conditions on Titan and Earth were similar in many respects 4 billion years ago, the approximate time when life originated on Earth. Life may have originated on Titan during its warmer early history and then developed adaptation strategies to cope with the increasingly cold conditions. If organisms originated and persisted, metabolic strategies could exist that would provide sufficient energy for life to persist, even today. Metabolic reactions might include the catalytic hydrogenation of photochemically produced acetylene, or involve the recombination of radicals created in the atmosphere by UV radiation. Metabolic activity may even contribute to the apparent youth, smoothness, and high activity of Titan’s surface via biothermal energy. Our calculations indicate that biothermal melting would be a possible explanation for the smooth surfaces observed by the Cassini-Huygens mission. Given the low temperatures, the biological effect on Titan, if it exists, should be larger than on Earth. In conditions where the ability to sustain liquid microenvironments is a key limitation on survival, then adaptive pressures could lead to a larger percentage of the free energy of exothermic metabolic reactions going towards heating the immediate environments of organisms living close to the freezing point. On the other hand, much energy has to be expended to reach the liquid state. If volcanic activity or other energy sources are present and significant, it would increase the chances for life on Titan by elevating temperatures and providing potentially habitable geothermal areas and gases that could be used for metabolism. Any liquid water-ammonia mixture is lighter than the surrounding ice and will float if produced at depth. Given the current sample size of one biosphere upon which astrobiologists must base their theories and speculations, our ideas about life elsewhere must remain fluid and not too heavily based upon the specific metabolisms, strategies and structures of terrestrial organisms. The basic requirements of life, as they are understood today, are all present on Titan, including organic molecules, energy sources and liquid media.

Goldschmidt Conference Abstracts 2005 The Robotic Exploration of Mars and Titan

Mars and Titan: Assessing the plausibility of life on two worlds with similar features and exotic differences 1





Department of Biological Sciences, University of Texas at El Paso, Texas, USA [[email protected]] 2 Department of Geology, Washington State University, Pullman, Washington, USA [[email protected]] Mars has long been considered conducive to the possibility of life because of its proximity to Earth, evidence for geophysical processes similar to those on Earth, and episodic temperatures above the melting point of water. The abundance of organic molecules on Titan has stimulated speculation about the possibility of life there as well. Recent data from robotic missions to Mars and Titan have opened the door to a reassessment of the plausibility of life (POL) on both. The abundance of subsurface water now evident on Mars has enabled the reconstruction of a planetary history highly conducive to the origin and possible persistence of at least microbial life (Schulze-Makuch, et al., 2005), strengthening the case for a POL of II on Mars (Irwin & Schulze-Makuch, 2001). Images from the Huygens probe to Titan reveal a heterogeneous surface with superficial similarities to Martian topography, indicative at least of chemical and energy cycles capable of reshaping what appears to be a young surface. Coupled with the abundance of organic matter, some of the requirements for living systems appear to be met. The nature of solvent availability remains a question. Subterranian water liquefied by internal heating or gravitational flexing could support water-based cellular life beneath the surface as known on Earth. The much colder temperature and consequent liquid state of organic compounds near the surface, if they are the prevalent solvents, would require a very different cellular boundary (membrane) composition and metabolic systems substantially divergent from anything known on Earth. Titan thus presents the possibility of harboring either relatively familiar or totally exotic forms of life. Ongoing analysis of data from Huygens should eventually support one possibility over the other, and suggest whether the original POL rating of III for Titan (Irwin & Schulze-Makuch, 2001) deserves to be elevated to II (Schulze-Makuch & Irwin, 2004).

References Irwin L.N. and Schulze-Makuch D., (2001) Astrobiology 1, 143-160. Schulze-Makuch D. and Irwin L.N., (2004) Life in the Universe: Expectations and Constraints. Schulze-Makuch D, Irwin L.N., Lipps J., LeMone D. and Dohm J. (2005) JGR-Planets., in press


Mars and Earth: Results of recent Mars missions V. R. BAKER1,2 1

Dept. of Hydrology and Water Resources, Univ. Arizona, Tucson, AZ 85721-0011 ([email protected]) 2 Lunar and Planetary Laboratory, Univ. of Arizona Current news releases to the contrary, many geological investigators have known for thirty years that Mars had an early history with aqueous activity on its surface. However, new mission results are revealing the extensive sedimentary and geochemical evidence for that history. The new results strongly corroborate the long-standing geological and geomorphological inferences that early Mars had extensive lakes and probably transient seas (“oceans”) that were associated with a climate capable of generating the precipitation and runoff to sculpt its landscape and transport materials to sedimentary basins. High-resolution images from the Mars Orbiter Camera (MOC) of the Mars Global Surveyor (MGS) Mission reveal a diverse suite of exceptionally young, globally distributed landforms of aqueous origin, including glacial, periglacial, fluvial, lacustrine, mass moverment, and phreatovolcanic features. These landforms are uncratered or exceedingly lightly cratered, implying an age of less than a few million years. If observed on Earth, most of these landforms would be ascribed to processes requiring a relatively dense atmosphere and extensive movement of water through precipitation from that atmosphere. Because such conditions do not currently prevail on Mars, these observations imply ongoing climate change on Mars Results from the Mars Odyssey Mission are consistent with an early phase of plate tectonics, which could have produced the Martian highland crust by continental accretion. By concentrating volatiles in a local region of the Martian mantle the early plate-tectonic phase of Mars would have led to a superplume at Tharsis. The resulting concentration of volcanism at Tharsis would iinfluence climate change by initiating immense megafloods. The observed persistence of this volcanism through later Martian history provides the mechanism for the episodic, short-duration aqueous phases. Earth’s early history of megaglaciations has some broad similarities to the newly understood history of Mars. Earth’s late Proterozoic glaciation is particularly interesting since there is considerable geological evidence that Earth may have temporarily switched to Mars-like icehouse conditions by freezing of the surface of the global ocean. Both Mars and Earth may be subject to major endogenetically driven shifts in climatic states.

Goldschmidt Conference Abstracts 2005 The Robotic Exploration of Mars and Titan


The past geochemical environment of Meridiani Planum, Mars, and its implications for astrobiology

Distribution of some elemental abundances on Mars: Results from the Mars 2001 Odyssey gamma ray spectrometer

BRIAN HYNEK AND TOM MCCOLLOM Laboratory for Atmospheric and Space Physics, Univ. of Colorado [[email protected]]

Motivation The recent chemical and inferred mineralogical data returned from the Mars Exploration Rover Opportunity are unlike anything previously seen or anticipated on Mars. The data, combined with textural information and geologic context, suggest deposition in a shallow body of water and subsequent chemical weathering [1-2]. Aqueous geochemical modeling offers a way to explore the conditions required for emplacement and alteration of the mineral assemblages. This is crucial for assessing the potential habitability for life in the paleoenvironment. Our approach has been to use the known alteration products at the Opportunity site to evaluate potential geochemical pathways and range of environmental conditions that can produce the observed chemistry and mineralogy. From the likely chemical pathways we can calculate the energy yield available for chemosynthetic organisms and place constraints on duration of the fluid-rock interactions as well as assess habitability.

Example Results: One typical simulation, run in Eq3/6, involves the reaction of “Martian” groundwater (pure water equilibrated with the current Mars atmosphere) with iron-bearing sediments in the form of pyrite (FeS2). Pyrite is a common product from the hydrothermal alteration of basalt as seen from terrestrial examples. As the fluid reacts with pyrite the pH drops rapidly through the production of sulfuric acid from oxidation of sulfide from pyrite, and dissolved Fe increases also. As a result, the fluid becomes saturated with hematite and jarosite and these minerals precipitate (the amount of jarosite precipitated is largely controlled by the assumed concentration of K+ in the initial fluid). The model suggests it is feasible that known minerals at the Opportunity site (namely hematite and jarosite) could arise from interaction of pyrite-bearing sediments with an aqueous fluid in contact with a moderately oxidizing atmosphere like that currently on Mars. Using typical assumptions, a microbe could potentially obtain 10 kJ of energy from each gram of pyrite reacted under such conditions. This is just one pathway we are exploring to understand the probable geochemical history of Meridiani Planum. Additional pathways will be presented as well as chemical energy yields and their implications for the suitability for the development of Martian life.

References [1] Squyres S. W. et al. (2004), Science, 306, 1709-1714. [2] Squyres S. W. et al., (2004), Science, 306, 1698-1703.


Lunar & Planetary Laboratury, Univ. of Arizona, [email protected]

The 2001 Mars Odyssey Spacecraft has recently completed its first martian year in orbit. Data collection with full boom deployment began in June, 2002 and has continued to the present with only short interruptions due to Solar Particle Events, instrument annealings and related activities. Gamma rays produced by individual elements can be the result of any of three processes: inelastic scatter, capture of thermal neutrons, or radioactive decay, each of which requires a separate method of processing and normalization. Our data collection rate for individual gamma rays is measured in counts per thousand seconds. Nevertheless, sufficient data has now been gathered to begin to examine some elemental distributions for the mid-latitudes on Mars. While we continue to analyze the data and determine the exact normalization procedures to use for each process leading to the observed gamma rays, we can now map out relative abundances for several elements. These elements include hydrogen (mapped as its water equivalent), chlorine, iron, silicon, potassium and thorium. The elements that we have been able to map to date show modest global differences in their distribution and these variations in elemental distribution are clearly associated with previously mapped geologic and morphologic regions on Mars. Silicon has limited variability (less than a factor of 2) over the planet but shows a modest enrichment in the northern lowlands and a significant decrease over the Tharsis region. Iron varies by a factor of 2 globally and also shows enrichment over the northern lowlands. (It should be noted that all of the in situ measurements to date have come from landers in the northern lowlands.) Chlorine shows a larger variation, on the order of a factor of 3, and is particularly high in the Medusae Fossae formation west of Tharsis. The range in distribution of potassium and thorium is relatively modest, especially when compared to that of the Moon. In addition to elements whch can be mapped in the mid latitudes, other elements can be averaged over large, geologically defined regions. Comparisons and correlations of these elements will be presented along with the maps.

Goldschmidt Conference Abstracts 2005 The Robotic Exploration of Mars and Titan

Subsurface Sounding in “Mars Advanced Radar for Subsurface and Ionosphere Sounding” (MARSIS) G. PICARDI 1,R. SEU1, A. FRIGERI2, P.T.MELACCI2 1

Infocom Dept. - “La Sapienza” Univ. of Rome ([email protected]) 2 Universty of Perugia According to the Mars Express mission, the MARSIS primary scientific objectives are to map the distribution of water, both liquid and solid, in the upper portions of the crust of Mars. Detection of such reservoirs of water will address key issues in the hydrologic, geologic, climatic and possible biologic evolution of Mars, including the current and past global inventory of water, mechanisms of transport and storage of water. According to the previous scientific objectives, this paper provides a description of the design approach and expected performances of the MARSIS, taking into account of Mars Orbital Laser Altimeter (MOLA) data. The principle of operation of MARSIS the following: the transmitted radar pulse will reach the top of the Mars surface producing a first reflection echo which propagates backward to the radar. However, due to the long wavelengths employed, a significant fraction of the e.m. energy impinging on the surface is transmitted into the crust and propagates downward. Additional reflections, due to subsurface dielectric discontinuities, will occur and the relevant echoes will propagate backward to the radar. As consequence time domain analysis of the strong surface return, eventually after multilook non-coherent integration, will allow estimation of surface roughness, reflectivity and mean distance, just like in classical pulse limited surface radar altimeters. The presence of weaker signals after the first strong surface return will enable the detection of subsurface interfaces, while the estimation of their time delay from the first surface signal will allow the measurement of the depth of the detected interfaces.The detection of these subsurface echoes is limited by the surface echoes (especially if surfaces are rough), for this reason three different methods will be implemented in MARSIS: Doppler Beam Sharpening, Secondary Monopole Antenna, and Dual Frequency Processing. Finally, the Marsis frequency-agile design will allow to tune the sounding parameters in response to changes in sun illumination condition, latitude etc.. According to the previous scientific objectives and the operation planning, this paper provides a description of the approach referred to the data inversion and expected performances of the MARSIS


New results from the robotic exploration of Mars R.C. ANDERSON1 AND ATHENA SCIENCE TEAM 1

Jet Propulsion Laboratory, California Institute of Technology, [email protected].

The Mars Exploration Rover Spirit landed successfully in Gusev Crater on January 4, 2004 (UTC), followed three weeks later with the successful landing in Meridiani Planum of its twin, Opportunity. Guesev Crater: The landing site at Gusev Crater lies on a densely populated rock-strewn plan. Rocks identified around the lander range in a variety of sizes and angular shapes. Preliminary results of the rock textures show that a majority of the rocks consist of fine-grained volcanic and several (Adirondack) appear to contain some sort of surface coating.. Preliminary results are that the concentrations of presumably dust-borne elements like sulfur and chlorine decrease and you go deeper into the rock. A majority of the rock observations from the plains rocks being classified as an unweathered olivine, magnetite-bearing, low silica basalt. At the present time, three distinct rock types have been identified in the Columbia Hills. Since the initial landing, Spirit has traveled over 4 km and is presently heading up to the top of the Columbia Hills. Meridiani Planum: The Opportunity landing site lies inside a 20 m diameter impact crater. The lander came to rest near an exposed layer (roughly 12 m long; 0.5 m high) of bedrock in the crater wall. Initial results from microscopic images (MI) data suggest this unit consist a fine-grained rock with a variety of sedimentary structures consisting of crossbedded, thin layer of sediments. Alpha Particle X-ray Spectrometer (APXS) suggest a high concentration of sulfur. Embedded within the outcrop and weathering out are highly spherical granules. Opportunity spent 90 sols examining Endurance crater and is presently outside the crater examining its own heatshield which landing approximately 2 km from its initial landing spot.

Goldschmidt Conference Abstracts 2005 The Robotic Exploration of Mars and Titan


The Tharsis and Elysium corridor: A marker for an internally active Mars? J.M. DOHM Department of Hydrology and Water Resources, University of Arizona, Tucson, AZ, 85721 ([email protected]) The paradigm of an ancient warm, wet, and dynamically active Mars, transitioning into a cold, dry, and internally dead planet has persisted up until recent in spite of published Viking-based geologic maps, which indicate geologic and hydrologic activity into the youngest part of the history of Mars, the Late Amazonian epoch. This paradigm is shifting to a water-enriched, still possibly internally active planet, based on a collection of geologic, hydrologic, topographic, and elemental evidences obtained by the Viking, Mars Global Surveyor (MGS), Odyssey, and Mars Express missions. A collection of diverse information unfolds this possibility, including: (1) stratigraphically young rock materials such as pristine lava flows with few, if any, superposed impact craters, (2) tectonic features that cut stratigraphically young materials, such as fractures, faults, graben, and pit crater chains, (3) features with possible aqueous origin such as structurallycontrolled channels that dissect stratigraphically young materials and anastomosing-patterned slope streaks on hillslopes, (4) elevated elemental abundances, such as hydrogen and chlorine, and (5) methane, all of which occur in regions that reportedly record ancient, middle planet, and geologically recent magmatic, tectonic, and hydrologic activity. Specifically, parts of Tharsis, Elysium, and the region that straddles the two volcanic provinces, collectively referred to here as the Elysium/Tharsis corridor, unfolds a potentially internally active Mars.

Results from recent Mars missions and their implication to possible life JEFFREY S. KARGEL U.S. Geological Survey, 2255 N. Gemini Dr., Flagstaff, AZ 86001, USA ([email protected]) The Mars Opportunity and Spirit rover missions, Mars Express, Mars Odyssey, and Mars Global Surveyor have revolutionaized our view of Mars. No scientist has faced these data with prior ideas well intact. While the orbiting missions have provided global and regional views of ice distribution in the shallow crust, topography,geology, and geochemistry, as well as aspects of local geology and geomorphology, the rover missions have provided critical ground truth that lend confidence to the global observations and add critical details not discernable from orbit. Most critically, the rovers have provided close-up chemical and petrologic data that elucidate important enviornmental conditions at two points in geologic time and space. It is evident from these orbital and rover data that Mars, like Earth, is a "water world," but it is one where most of the water has been locked as solid ice for most of Martian history. There were periods and places on the surface where liquid water was abundant for geologically significant periods of time. We see it in the landforms and in the chemistry and mineralogy. Limited rover-based ground truth points toward highly acidic, highly saline, and probably very cold brines as the depositional agent in Meridiani Planum and as a rock-altering agent in Gusev Crater. The rover data are consistent with a an interpretation that Mars is analogous to a global acid-mine waste site. Many aspects of Martian geology, geochemistry, and mineralogy--right down to the red color of Mars-- can be explained within the context of this aqueous acid model. Orbiter observations of very young water-related landforms indicate a surprising geologic recency of some aqueous activity. However, in retrospect, considerations of multicomponent solid-liquid-gas phase equilibria suggest that aqueous acid brines still are active at shallow levels and potentially even at the surface of Mars. Whether life ever could have originated or evolved there is an important unanswered question, but one which should bear very strongly on our exploration strategy. Recent Mars Express observations of methane are consistent with the past or present existence of methanogenic life; however, abiogenic formation of methane also is possible. In sum, we have answered long-standing questions about whether Mars was icy or wet; it was and still is both.

Goldschmidt Conference Abstracts 2005 The Robotic Exploration of Mars and Titan

Sulfate minerals as targets for biomolecule detection on Mars

Next-generation robotic planetary reconnaissance missions: A paradigm shift

A. D. AUBREY1, H. J. CLEAVES2, J.H. CHALMERS3 4 AND J. L. BADA Scripps Institution of Oceanography, University of California at San Diego, CA 92093-0212 1 [email protected] 2 [email protected] 3 [email protected] 4 [email protected] A first step in the search for life on Mars must focus on the detection of specific organic molecules associated with biology as we know it. Amino acids are prime target molecules in this search strategy. Advanced fluorometric analytical techniques offer high detection sensitivities for amino acids, and amino acid chirality (handedness) can be used to distinguish biological from abiotic origins. One important consideration with respect to amino acids as target molecules is whether they are preserved over long periods of time at the cold temperatures characteristic of Mars. One way to address this issue is to investigate terrestrial mineral analogs to determine which types offer the best matrix for amino acid preservation. Strong evidence for evaporitic sulfate minerals such as jarosite, gypsum, and anhydrite has recently been found on Mars by the Spirit and Opportunity rovers. As these minerals are deposited in terrestrial evaporitic environments, any organic molecules from extant or extinct microorganisms should be co-deposited. Thus, we have investigated concentrations of organic matter along with amino acids in natural terrestrial sulfate mineral samples. We have found that sulfate minerals contain between 0.03 to 0.69 % organic carbon as well as high ppb to low ppm abundances of amino acids and their degradation products in samples ranging from 30 million years old to contemporary. From our data, it appears that amino acids and their amine decarboxylation products are well preserved over long geological time in the sulfate mineral matrices on Earth. This preliminary evidence indicates that sulfate minerals should be prime targets in the search for organic compounds, including those of biological origin, on Mars. Suitable in situ instrumentation is now available to detect amino acids at subppb levels while also providing chiral resolution. We thus conclude that amino acids in sulfate mineral matrices are strong candidates in the search for organic molecules of possible biological origin on Mars.



California Institute of Technology, Pasadena, CA, USA ([email protected]) 2 Department of Hydrology and Water Resources, University of Arizona, Tucson, AZ, USA 3 United States Geologic Survey, Flagstaff, AZ, USA We introduce a fundamentally new scientific mission concept for remote planetary surface and subsurface reconnaissance that will soon replace the engineering and safety constrained mission designs of the past, allowing for optimal acquisition of geologic, paleohydrologic, paleoclimatic, and possible astrobiologic information of Mars and other extraterrestrial targets. Traditional missions have performed local ground-level reconnaissance through immobile landers and rovers, or global mapping performed by an orbiter. The former is safety and engineering constrained, affording limited detailed reconnaissance of a single site at the expense of a regional understanding, while the latter returns immense datasets, often overlooking detailed information of local and regional significance. A “tier-scalable” paradigm (Fig. 1) integrates multi-tier (orbitÙatmosphereÙground) and multi-agent (orbiterÙblimpsÙrovers) hierarchical mission architectures, not only introducing mission redundancy and safety, but enabling and optimizing intelligent, unconstrained, and distributed science-driven exploration of prime locations on Mars and elsewhere, allowing for increased science return, and paving the way towards fully autonomous robotic missions.

Fig. 1: Tier-scalable multi-tier and multi-agent hierarchical mission architecture

References Fink W., Dohm J., Tarbell M., Hare T., and Baker V. (2004) “Next-Generation Robotic Planetary Reconnaissance Missions: A Paradigm Shift”, submitted to Planetary and Space Science

Goldschmidt Conference Abstracts 2005 The Robotic Exploration of Mars and Titan


Enigmatic linear patterns of hydrogen concentration on Mars

Columbia Plateau Basalt as an analog to the basalt of the Martian Northern Plains

J.R. CLEVY1 AND S. A. KATTENHORN 2 Department of Geological Sciences, University of Idaho, Moscow, ID 83844-3022, USA ([email protected], [email protected]) Chemically or physically bound hydrogen within a meter of the Martian surface has been mapped using neutron spectroscopy [1]. The Neutron Spectrometer, part of the Gamma-Ray Spectrometer on board Mars Odyssey, is able to detect thermal, epithermal and fast neutron fluxes. Each of these have specific energy ranges with epithermal neutron energy ranging from 0.4 – 500 keV. This band is the most sensitive for hydrogen mapping purposes. Regions with high hydrogen concentrations have a low epithermal energy flux. These concentrations are believed to indicate locations of subsurface water ice. As such the flux maps pinpoint locations where small quantities of liquid water may intermittently form today or where liquid water may have pooled in the past. The possibility of life existing on Mars – either in the distant past or at present – depends on the availability of liquid water. Epithermal neutron flux maps of the equatorial region east of Schiaparelli Crater in Mars’ eastern hemisphere indicate hydrogen ion concentrations in the shallow subsurface with a hydrogen water equivalent of just over 10 percent [1]. Published maps [2] reveal anomalous linear concentrations of hydrogen with a northeast to southwest trend. The width and trend of these linear anomalies match those of the graben between Scylla Scopulus and Charybdis Scopulus, west of Hellas Basin. These linear ion concentrations suggest structural control of the hydrogen. Structural control of fluids can be attributed to fault activity or structural topography. Terrestrial faults are known to exert a strong control on groundwater flow immediately after earthquake events. Subsurface faults may also act as a barrier to fluid flow, creating a confined channel or aquifer within the width of the graben. Alternatively, the graben may have acted as a topographic control on surface water accumulation in the past. Any seepage of this water into the subsurface may have resulted in a hydrogen ion fingerprint in graben valleys, resulting in the linear patterns observed.

References [1] Feldman, W.C. et al. (2002) Science 297, 75-78. [2] Boynton, W.V. et al. (2002) Science 297, 81-85.


Dept. of Geological Sciences, WSU, (cf[email protected]) Dept. of Geological Sciences, WSU, ([email protected])


The basalt of the northern plains on Mars is more andesitic and weathered than the basalt of the southern highlands. It appears to be well represented by the Bounce Rock at the Meridiani Site, which is dominated by pyroxene (clinopyroxene ~55%, orthpyroxene ~5%) and plagioclase (~20%), and is poor in olivine (~5%). Oxides are acounting for ~10%. The chemical composition of Bounce Rock is more evolved than the basalts in the Gusev crater. It has a high P2O5 content of 0.95wt%, a Fe/Mg ratio of 36, a low Mg number (molar MgO/ MgO+FeO) of 0.42 and a high Ca/Al ratio of 1.7, a lower FeO (15.6%), and a higher CaO (12.5%) content. The basalt in the northern plains is in general rich in sulfur and variably enriched in bromine relative to chlorine, indicating a past interaction with water. The Columbia Plateau Basalt (CPB) is a typical continental flood basalt, composed of four basalt formations made up of more than 300 individual basalt flows. The wrinkle ridges and low viscosity of these basalt flows are typical features of CPB. On a normative cpx-ol-pl-qz projection, the Grande Ronde Formation (~ 87% of CPB), is very close to the 1 atmosphere pseudo-cotextic, indicating that the erupted melts underwent fractionation and mixing processes at a very shallow level. CPBs are dominated by relatively low Mg tholeiite and basaltic andesite, with SiO2 at 52-58%, FeO 10.0-13.5%, CaO 8.6-10.4%, a Mg number of 0.55, and a higher P2O5 content of 0.68%. Both, CPB and Mars basalt were formed at low pressure conditions with low-vicosity basaltic lava spreading over wide areas. Basaltic rocks in the northern part of Mars and the CPB were subject to sporadic interactions with water bodies, weathering, and glacial processes. The northern Martian basalts and CPB are similar geochemically and mineralogically being both evolved basalts with some signatures of andesite. Also, the higher P content of Bounce rock and the wrinkle ridges of Martian basalt are similar to that of CPB. Life based entirely on chemoautotrophic energy sources has been reported from a deep basalt aquifer of the CPB. Since Martian surface conditions are extremely hostile for life as we know it, the primary energy source for putative life on Mars is likely chemical energy rather than light energy. Thus, CPB and the chemotrophic organisms in the CPB serve as a suitable analog to Mars.

Goldschmidt Conference Abstracts 2005 The Robotic Exploration of Mars and Titan

AGFA: (Airborne) Automated Geologic Field Analyzer

Paleoenvironmental study of Doushantuo Formation: Insights of trace element and carbon isotope




California Institute of Technology, Pasadena, CA, USA ([email protected]) 2 Robotics Institute, Carnegie Mellon University, Pittsburgh, PA, USA 3 Department of Hydrology and Water Resources, University of Arizona, Tucson, AZ, USA This work reports on an integrated software system (AGFA) that automatically and unbiasedly characterizes rocks/soil in an imaged scene (in various wavelengths). This technology enables automated science analysis for robotic spacecraft, further expanding the possibilities for future intelligent and autonomous robotic exploration of remote planetary surfaces. This can be accomplished by merging the expertise of a field geologist with traditional exploration spacecraft to form a science craft. Its main benefit will be the ability to yield high science returns at a significant savings in cost and time. Although reconnaissance field geologists can become astronauts, the initial forays to Mars and other planetary bodies will be done by robotic craft. Numerous steps are necessary in order for a science craft to map, analyze, and characterize a geologic field site, and effectively formulate working hypotheses. We present and discuss a tool for automated science analysis of geologic field sites: the Automated Geologic Field Analyzer (AGFA). AGFA, using various wavelengths, maps and characterizes rock/soil materials; this is done, both on the ground and from the air, by extracting features such as size, color, albedo, vesicularity, and angularity. Based on the extracted features, AGFA summarizes the field site numerically and flags targets of interest. It is our vision that this step will lead to autonomous robotic space exploration of remote planetary surfaces.

References Fink W. et al. (2004) “Next-Generation Robotic Planetary Reconnaissance Missions: A Paradigm Shift”, submitted to Planetary and Space Science Schulze-Makuch D. et al. (2004) “Comparative Planetology of the Inner Planets of the Solar System: Geologic Setting, Astrobiological Assessment and Implications for Mission Design”, Journal Astrobiology in press



The State Key Laboratory for Mineral Deposits Research, Nanjing University, Nanjing, 10093, China; 2 Resource and Environmental Enginering Center, East China Institute of Technology, Fuzhou, 344000, China 3 Institute of Mineralogy and Mineral Resources, Technical University of Clausthal, 38678 Clausthal-Zellerfeld, Germany The upper phosphorite bed of Neoproterozoic Doushantuo Formation, Weng’an, South China, preserves a unique assemblage of what are propally the earliest metazoan fossils in the world that could contribute to a better understanding of early faunal evolution on Earth. However, no animal forms have been found in the lower phosphorite bed. Trace element geochemical characteristics show that the upper ore bed with clear negative Ce anomalies (Ceanom ranging from -0.34 to -0.17) and lower redox element contents (Mn ranging from 125ppm to 452ppm, Mo ranging from 0.23ppm to 1.35ppm, U ranging from 3.79ppm to 7.05ppm and V ranging from 13.85ppm to 24.46ppm) compared to the lower ore bed with less negative Ce anomalies (Ceanom ranging from -0.05 to -0.02) and higher redox element contents (Mn ranging from 897ppm to 1524ppm, Mo ranging from 1.52ppm to 13.95ppm, U ranging from 5.53ppm to 22.07ppm and V ranging from 9.70ppm to 52.68ppm), indicating the marine depositional environment changed from anoxic in the lower ore bed to oxic in the the upper ore bed. We might infer that increase of oxygen content in paleo-ocean probably caused the emergence of the earliest diverse eukaryote in the upper ore bed of Duoshantou Formation. δ13C value trend generally increase from lower ore bed to upper ore bed although they are with several oscillations, indicating higher organic productivity and higher burial rate of organic carbon in the upper ore bed than in the lower upper ore bed.

Acknowledgements This study was financially supported by the National Natural Science of China (Grant Nos. 40272080 and 40232020). The authors wish to thank Prof. Zhu Maoyan and Prof. Chen Junyuan for help in the field work.

Goldschmidt Conference Abstracts 2005 NOM-Metal Complexation


Humic ion-binding modelling and its application to field processes and ecotoxicology 1




Metal binding to NOM from database to field systems M.F. BENEDETTI Laboratoire de Géochimie des Eaux – Université Denis Diderot, Paris, France ([email protected])


Centre for Ecology & Hydrology, Lancaster, UK ([email protected], [email protected], [email protected]) 2 Department of Geography, University of Leeds, UK ([email protected]) WHAM/Model VI combines an ion-binding model for humic substances with an inorganic speciation code. Model VI uses a structured formulation of discrete, chemicallyplausible, binding sites for protons, to allow the creation of regular arrays of bidentate and tridentate binding sites for metals, including relatively rare sites with high affinity. The (competitive) binding of each metal, and its first hydrolysis product, is characterised by a single intrinsic equilibrium constant. Account is taken of electrostatic interactions. The catchment model CHUM-AM includes process descriptions of N and S uptake and release by the soil-plant system, percolation and evaporation of water, solid-solution partitioning of solutes (described with WHAM/Model VI), chemical interactions in solution (also described with WHAM/Model VI), and chemical weathering. The model operates on a yearly time-step and is driven by inputs of wet and dry deposition. CHUM-AM has been applied to moorland catchments in northern England, to simulate acidification and its reversal, and heavy metal behaviour. By 1998, the catchment soils had retained 89-98% of previously deposited anthropogenic Cu, and 95-100% of Pb. Retention of the other metals, which sorb relatively weakly to the soil, depended on soil pH and varied from 5%, for Ni in the most acid soil, to 62%, for Zn in the least acid soil. Simulations of future metal behaviour suggest that weakly-sorbing metals (Ni, Zn, Cd) will respond on timescales of decades to centuries to changes in metal inputs or acidification status. More strongly-sorbing metals (Cu, Pb) will respond over centuries to millennia. WHAM/Model VI provides estimates of free metal ion concentrations in soil solution or in surface water, and these can be combined with information on ecotoxicity to gauge toxic effects. Soil metal toxicity data have been analysed, using multiple regression equations describing free ion concentrations, to derive pH-dependent free ion Critical Limit Functions. These CLFs are used in a proposed method to determine steady-state Critical Loads for heavy metals in Europe. They can also be combined with CHUM-AM outputs to explore temporal variations in potential toxicity.

In the past decade advances in modelling metal ion binding to natural organic matter have been made. In parallel an extensive data set has been produce to help to validate and calibrate the modelling effort. Today robust models are available to describe complex geochemical systems such as soils and rivers. The validity of the model calculation is however strongly dependant on the reliability of the database used for calibration. This is the case for elements like Cu, Cd, Ca, Eu for which numerous experimentals data are available and can be compared to each other. However for important elements like Al, Fe and in general tri and tetravalent elements the range covered by the available experiments is not sufficient for reliable model predictions. In the present contribution I will report new data for trace elements interactions with various types of natural organic matter at different pH and metal to ligand ratios. The model parameters like affinity constants, heterogeneity and site densities derived within the NICA-Donan model framework are compared to existing values. Based on those new parameters complex geochemical systems will be simulated. The case of iron in tropical rivers and trace elements in contaminated soil solutions will be discussed.

Goldschmidt Conference Abstracts 2005 NOM-Metal Complexation

Fundamental aspects of interaction between metals and humics in the environment W.H. VAN RIEMSDIJK1 AND L.P. WENG2

Influence of SOM on aluminium mobility in a forested brown acidic soil: A view from soil solutions held at different matrix potentials F. GÉRARD1, J. JAFFRAIN1, J.-P. BOUDOT2 1 AND J. RANGER


Metal ion interaction with humic substances is pH dependent, and the pH dependence has to obey the thermodynmamic consistency principle. One of the simplest models in this respect is the pH dependent Freundlich model. Such models have their limitation, because they can not easily be extended to include metal-metal competition and moreover the basic proton buffering behavior of humics is not icluded in the simplified models. The NICA-Donnan model and model V/VI do link the basic charging behavior of the humics with competitive metal ion binding. The NICA-Donnan model has been applied to model metal ion binding and mobility in soils with quite some succes. Especially the possibility to measure the free metal ion at very low concentration with the recently improved DMT technique is very helpful to test the quality of the models. The mobility of metal ions will be affected by the binding to DOC and by the binding to the immobile phase. The DOC concentration can be influenced by the binding of the DOC to the immobile mineral phase. Metal ions may also interact with humics that are in close association with mineral particles. The DOC-mineral-metal interaction process is in our opinion the next frontier for a better understanding of the mobility of metal ions in the environment. Minerals have a rather complex surface chemistry, and to combine this type of chemistry with the also rather complex chemistry of humics is an enormous challenge. A first attempt in this direction has been the development of the LCD model, which combines the NICA-Donnan model for humics with the CD-MUSIC model for metal(hydr)oxides. This model has been applied to fulvic acid binding to goethite. Extension of this model to humic acid binding has so far been an unresolved challenge. We will show a new theoretical approach that will probably allow to solve this challenging problem.


INRA Nancy, BEF, 54280 Champenoux, France CNRS Nancy, LIMOS, 54501 Vandoeuvre, France


We studied a number of soil water samples held at different matrix potential from the surface horizons of a brown acidic forested soil. The matrix potential can be converted into an equivalent pore diameter. Solutions have been analysed for total chemistry in major elements, pH and total inorganic monomeric Al. We used the WHAM VI speciation model to calculate Al speciation, by fitting results to the mesured pools of inorganic monomeric species. Results are shown in the figure below. Black symbols represent solutions collected beneath the litter layer (0 cm), white symbols stand for solutions collected at 15 and 30 cm depth by mean of tension lysimeters (matrix potential; |ψ| < 600 hPa), and grey symbols are for solutions held by centrifugation (1800 hPa < |ψ| < 16000 hPa) at the 0-15 cm and 15-30 cm depth intervals. Well crystallized gibbsite

10 Interlayer Al(OH) 3

-log[A l3+ ]

Dreienplein 10 Department Soil Quality, Wageningen University, The Netherlands ([email protected]) 2 Dreienplein 10 Department Soil Quality, Wageningen University, The Netherlands ([email protected])



logAl = 0.8pH + 2 2

R = 0.73

Amorphous Al(OH) 3

logAl = 1.4pH - 1.1

logAl = 2.6pH - 7.6


R = 0.77


R = 0.95

2 3




5 pH





We found that Al mobility in solutions collected in the larger soil pores (tension lysimeters) are controlled by the neoformation of poorly crystallized Al-hydroxides, likely to be interlayer Al. In smaller pores (centrifugation) and especially those from the top soils, which contain more acidic water, a control by Al3+-H+ exchange reactions with SOM is found. The maximum control by SOM is consistently found beneath the litter layer.


Goldschmidt Conference Abstracts 2005 NOM-Metal Complexation

The binding of cadmium, copper and iron by fractions of dissolved organic matter and humic substances originating from compost Y. CHEN1, A. KASCHL2 AND P. GAT3 1,3

The Hebrew University of Jerusalem, Rehovot, Israel ([email protected], [email protected]) 2 European Commission, Brussels, Belgium ([email protected]) Complexation by soluble organic ligands originating from composted municipal solid waste (MSW) and biosolids (BS) amended to the soil, were studied. Dissolved organic matter (DOM) from these composts were separated into six fractions and their complexation of Cd, Cu and Fe was quantified by performing titrations with an ion-selective electrode (ISE) at pH 5 (Cd, Cu) or by ligand exchange (Fe). The complexing capacity (CC) for Cd was highest for HoA, a fulvic-type, easily soluble fraction, at 1042 µmol Cd g-1C of ligand. The other DOM fractions exhibited distinctively lower CCs. The highest stability constants (logKint) measured were: 7.74 (HoA), 7.69 (HoN), 7.02 (HiA), 6.93 (HiN), and 8.11 (HiB); logKint for the HA was 10.05 and that for the FA>1000, was 7.98. Using a continuous distribution model to evaluate the titration data, the MSW fractions HiA and HiN demonstrated a rather narrow range of binding site strengths, as opposed to the broader distribution curves exhibited by HoA and HiB. The HoN distribution curve was markedly different from that of the other DOM fractions and was shifted towards higher binding strength. Using the same distribution model for Cu binding by compost derived humic substances (HS), it was shown that the HoA, HoN, and HiB exhibited a broader range of logKint (9 to 10.5). Stability constants with Fe of the DOM and its fractions exhibited values for logK ranging from 7-8. A specific ligand exchange technique, which was developed in our laboratory, was used for the determination of the logKs for Fe with HS.

Quantifing uranium complexation in groundwater DOC using coupled detection methods THORSTEN N. RESZAT1, M. JIM HENDRY1 2 AND JAMES F. RANVILLE 1

Department of Geological Sciences, University of Saskatchewan, Saskatoon, SK, Canada. ([email protected], [email protected]) 2 Department of Chemistry and Geochemistry, Colorado School of Mines, Golden, CO, USA ([email protected]) The long-term mobility of actinides in groundwaters is important for citing nuclear waste facilities and managing waste-rock piles at uranium mines. Dissolved organic carbon (DOC) may influence the mobility of uranium, but few fieldbased studies have been undertaken to examine this in typical groundwaters. In addition, few techniques are available to isolate DOC and directly quantify the metals complexed to it. Determination of U-DOC association constants from analysis of field-collected samples compliments laboratory measurements, and these constants are needed for accurate transport calculations. The partitioning of U to DOC was tested on both laboratory prepared samples and samples (n=8) collected from a high DOC clay-rich aquitard. Laboratory tests were conducted to determine the partitioning of U(VI) onto a standard fulvic acid. In both cases, the association of U and DOC was examined directly using on-line coupling of Asymmetrical Flow Field-Flow Fractionation (AsFlFFF) with UV absorbance (UVA), DOC, and inductively coupled plasma-mass spectrometer (ICP-MS) detectors. This method has the advantages of utilizing very small sample volumes (20-50 µl) as well as yielding molecular weight information on U-DOC complexes. UVA and DOC detection results were used to determine percentages of total uranium associated with DOC (and other colloidal materials). In all cases, AsFlFFFICP-MS suggested that only a small percentage of the U, likely present as U(VI), was complexed with the DOC. This result was in good agreement with U speciation modeling performed on the groundwater samples. This agreement suggests that the AsFlFFF-ICP-MS method may be very useful for determining U-DOC association in small volume samples. Because the pH (7.0-8.1) and carbonate concentrations of these waters are typical of many groundwaters, these data suggested that facilitated transport of U by DOC may be limited in its importance in many groundwater systems.

Goldschmidt Conference Abstracts 2005 NOM-Metal Complexation

A potentiometric and 113Cd NMR study of cadmium complexation by natural organic matter at two different magnetic field strengths

Lead transport and speciation in organic horizons of forest soils J.M. KASTE1, A.J. FRIEDLAND2, B.C. BOSTICK1 1 AND A.W. SCHROTH 1

Dartmouth College Earth Sciences 6105, Hanover, NH 03755, USA ([email protected], [email protected], [email protected]) 2 Dartmouth College Environmental Studies 6182, Hanover, NH 03755, USA ([email protected]) Anthropogenic emissions during the 20th century resulted in global lead (Pb) contamination of soils. Remote forests in New England received 1 to 4 g Pb m-2, with higher amounts deposited at montane sites. We present isotopic and soil timeseries data demonstrating that forest floor horizons (e.g., the upper 5 to 9 cm of organic material) retain atmosphericallydelivered Pb in excess of 50 years. Despite their role in retaining lead, the precise nature of Pb partitioning within this organic-rich matrix remains elusive. Selective chemical extractions and x-ray absorption spectroscopy suggest that Pb is frequently bound to organics; however, a significant proportion of the Pb appears to be associated with inorganic phases. Extractions with sodium hydroxide and Na2P4O7, two commonly used treatments for targeting organically-bound metals released 50 to 100% of the Pb, but the specificity of these reagents to organic matter are suspect. Selective chemical extractions with the reducing agent hydroxylamine hydrochloride released up to 40% of the Pb, but liberated negligible organic matter. Lead released during chemical extractions coincided with iron release. We conclude that despite very high organic matter contents (>80%) a significant proportion of the lead in forest floor horizons may associated with inorganic and possibly “reducible” mineral phases.



School of Earth and Atmospheric Sciences, Georgia Institute of Technology, Atlanta, GA 30332, USA 2 Institute for Ecological Chemistry, GSF Research Center for Environment and Health, D-85758 Oberschleißheim, Germany The binding of cadmium(II) to Suwannee River natural organic matter (NOM) has been investigated across a broad range of molar Cd/C ratios (0.00056 – 0.0056) and pH values (3.5 – 11) by 113Cd NMR spectroscopy at two magnetic field strengths (B0 = 9.4 and 11.2 T). As a result of the very peculiar and highly complex characteristics of the Cd-NOM exchanging system, these 113Cd NMR spectra are characterized by a pH- and concentration-dependent superposition of slow, intermediate and fast chemical exchange. Detailed insight into the complex interplay of solution chemistry and chemical exchange is provided by a thorough mapping of this Cd-NOM chemically exchanging system through acquisition of more than 100 113Cd NMR spectra at two magnetic field strengths and with systematic variation of Cd/C ratios and pH values. The interpretation of 113 Cd NMR spectra is greatly facilitated and constrained by simultaneous measurements of pH and pCd, which allows a model-independent calculation of organically bound Cd(II) under all experimental conditions. Under the conditions applied in this study, Cd(II) has been found to bind to oxygen, nitrogen, and, to a much lesser extent, sulfur ligands of NOM.


Goldschmidt Conference Abstracts 2005 NOM-Metal Complexation

Interactions between dissolved organic matter and mercury in aquatic environments G.R. AIKEN1, J.N. RYAN2 AND K.L. NAGY3 1

US Geological Survey, 3215 Marine Street, Boulder, CO 80303, USA ([email protected]) 2 Department of Civil, Environmental, and Architectural Engineering, University of Colorado, Boulder, CO 80309, USA 3 Department of Earth and Environmental Sciences, University of Illinois at Chicago (MC-186), Chicago, IL 60607-7059, USA Interactions of mercury (Hg) with dissolved organic matter (DOM) and aquatic humic substances play important roles in controlling the reactivity, bioavailability and transport of Hg in aquatic systems. Our research has shown that DOM influences the transport and reactivity of Hg by strong HgDOM binding and colloidal stabilization of mercuric sulfide (HgS) under all redox conditions. Strong interactions (KDOM' = 1023.2±0.5 L kg-1 at pH = 7.0 and I = 0.1), indicative of Hgthiol bonds, were observed at Hg/DOM ratios below approximately 1 µg Hg per mg DOM. Only a small fraction (approximately 2%) of the reduced-S groups, as determined by X-ray absorption near edge spectroscopy, was involved with these stronge interactions between Hg and DOM. These results suggest that the binding of Hg to DOM under natural conditions (low Hg/DOM ratios ranging from 0.01 to 10 ng of Hg/mg of DOM) is controlled by a small fraction of DOM molecules containing reactive thiol functional groups. In the case of fully oxygenated water (sulfide-free), the binding of Hg+2 by DOM should dominate dissolved inorganic mercury speciation. Where measurable total sulfide concentrations are present in surface water and pore water, Hg-sulfide complexes predominate. In these cases, which are common in sulfate reducing environments, DOM interacts strongly with HgS (log Ksp = -52.4) to inhibit aggregation and precipitation of HgS by colloidal stabilization. In experiments designed to define this interaction, precipitation of HgS was strongly inhibited in the presence of low concentrations (<3 mg C/L) of DOM, and organic matter rich in aromatic moieties was more reactive with HgS than less aromatic fractions. Aquatic humic substances were observed to inhibit the precipitation of HgS to a greater degree than low molecular weight ligands and other fractions of DOM. These results suggest that DOM can influence the geochemistry and bioavailability of HgS in aquatic environments by maintaining higher dissolved total Hg concentrations than predicted by speciation models.

The influence of dissolved organic matter on cinnabar dissolution K.L. NAGY1 J.S. WAPLES2, G.R. AIKEN3 4 AND J.N. RYAN 1

Department of Earth and Environmental Sciences, University of Illinois at Chicago (MC-186), Chicago, IL 60607-7059, USA 2 Golder Associates Inc., 44 Union Blvd Suite 300, Lakewood, CO 80228, USA 3 United States Geological Survey, Water Resources Division, 3215 Marine Street, Boulder, CO 80303, USA 4 Department of Civil, Environmental, and Architectural Engineering, University of Colorado, Boulder, CO 80309, USA Cinnabar (HgS) dissolution rates were measured in the presence of twelve different natural dissolved organic matter (DOM) isolates, including humic, fulvic, and hydrophobic acid fractions. Initial dissolution rates ranged over 1.3 orders of magnitude, from 2.31 × 10-13 to 7.16 × 10-12 mol Hg (mg C)-1 m-2 s-1, and produced dissolved Hg concentrations orders of magnitude above those expected in purely inorganic solutions. Rates were positively correlated with three characteristics of the DOM: specific ultraviolet absorbance at 280 nm (R2 = 0.88), aromaticity (R2 = 0.80), and molecular weight (R2 = 0.76). Three observations demonstrate that direct interaction of DOM was necessary to cause dissolution of the cinnabar surface: (1) rates of Hg release were linear with time, (2) rates were significantly reduced when the DOM was physically separated from the surface by 1000 Dalton dialysis membranes, and (3) rates approached maximum constant values at a specific ratio of DOM concentration to cinnabar surface area, suggesting that rates are a function of surface coverage by dissolution-reactive DOM. Dissolution rates for the hydrophobic acid fractions correlate negatively with sorbed DOM concentrations, indicating the presence of a DOM component or components that reduced the surface area of cinnabar that can be dissolved. When two hydrophobic acid isolates that enhanced dissolution to different extents were mixed equally, a 20% reduction in rate occurred compared to the rate with the more dissolution-enhancing isolate alone. Rates in the presence of the more dissolutionenhancing isolate were reduced by up to 60% when cinnabar was pre-reacted with the isolate that enhanced dissolution to a lesser extent. All the data together imply that the property of DOM that enhances cinnabar dissolution is distinct from the property that causes it to sorb irreversibly to the cinnabar surface and lower the dissolution rate.

Goldschmidt Conference Abstracts 2005 NOM-Metal Complexation


Role of humic substances in metal compexation, reduction and reoxidation processes

Reduction of inorganic arsenic with humic materials


University of Idaho ([email protected], [email protected])

Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831. Humic substances are known to be complex in nature with varying structural and functional characteristics, and their effect on biological reduction and oxidation of metals such as uranium(VI) and Fe(III) is not fully understood.. In this study, various humic materials and/or subfractions were charatcterized by both wet-chemical and spectroscopic analyses such as 13C-nuclear magnetic resonance (NMR) and Fourier-transform infrared (FTIR) spectroscopy. The role of humic materials in enhancing the biological reduction of Fe(III) and U(VI) was evalauted in both laboratory batch kinetic studies and field push-pull tests. Results indicate that, in the presence of the dissimilatory metal reducing bacterium Shewanella putrefaciens CN32, all humic mateirals were able to enhance the reduction of Fe(III) and U(VI) under anaerobic, circumneutral pH conditions. The presence of humic materials enhanced the U(VI) reduction rates (up to ~10 fold) and alleviate the toxicity effect of Ni2+ to microorganisms. Such an enhancement effect is attributed to the ability of these humics in facilitating electron transfer reactions and in complexing toxic Ni2+ ions. Among various humic materials, soil humic acid was found to be particularly effective in mediating the biological reduction of Fe(III) and U(VI) as compared with the fulvic acid and other humic fractions with relatively low aromatic contents. These results suggest that, depending on chemical and structural properties, different NOM components may play different roles in enhancing the biological reduction of Fe(III) and U(VI). Polycondensed aromatic humic acids may be particularly useful in mediating the biological reduction of these metals in soil. Of particular interest of these findings is the possibility that humic-mediated bioreduction of contaminants such as U(VI) can lead to immobilization or remediation of the contaminated site because reduced forms of U(IV) are sparingly soluble and can thus be strongly retained in soil. However, further studies indicate that humics also formed complexes with reduced U(IV) and prevented it from precipitation. In addition, humics increased the re-oxidation of reduced U(IV) (when exposed to oxygen) with a re-oxidation half-life on the order of a few minutes. Both of these processes render uranium soluble and potentially mobile in soil. These observations point out important challenges on the use of the bioreduction technology for immobilizing uranium thereby remediating contaminated sites in the field because of the wide occurrence of humic substances in soil and groundwater. Therefore, future studies must address the stability and retention of reduced or immobilized U(IV) as a means of longterm stewardship of contaminated sites.


In natural systems, arsenate (As-V) can be reduced to arsenite (As-III) by a number of different processes. The goal of the present research is to study the abiotic reduction of arsenate through the intervention of different humic materials. Results indicate that arsenate can be reduced to varying degrees depending on the humic composition, concentration and solution pH. The reduction potentials of several humic acids were studied, and these materials were used as reducing agents for inorganic arsenic. A novel ion chromatographic method of monitoring arsenate reduction by humate solutions was developed. Reduction of arsenate to arsenite was found to proceed in the 20 to 50% range. The kinetics appeared to be quite slow, with the reactions often taking up to 48 hours to reach completion.

Goldschmidt Conference Abstracts 2005 NOM-Metal Complexation


Hg2+ bonding in soil humic acid and equilibrium partitioning in suspension

Effects of natural organic matter on the speciation of uranium ANTHONY J. BEDNAR, VICTOR F. MEDINA AND STEVEN L. LARSON US Army Engineer Research and Development Center, US Army Corps of Engineers, Vicksburg, MS 39180, USA

Introduction Organic matter is known to complex metals and affects properties, such as sorption, mobility, and bioavailability. This work focuses on interactions of uranium with model organic compounds and complex plant extracts. Molecular weight cut-off filtration and size exclusion chromatography (SEC) with online UV absorption coupled to inductively coupled plasma mass spectrometry were used to qualitatively identify organouranium species in soil and plant materials.

Results and Discussion An example size exclusion chromatogram (Figure 1) below shows that nearly half of the uranium extracted from the plant tissue grown in uranium containing soil is bound to organic ligands, as indicated by the concurrent ICP-MS and UV absorbance peaks.

Figure 1: Size exclusion chromatogram of a plant extract. Sorption studies indicate that humic acid can reduce uranium sorption to soils and can promote subsequent desorption of uranium. These experiments suggest organic compounds may be an important influence on the mobility and chemistry of uranium in the environment.

Acknowledgements The tests described and the resulting data presented herein were obtained from research conducted under the Environmental Quality Technology Program of the United States Army Corps of Engineers by the USAERDC. Permission was granted by the Chief of Engineers to publish.


Univ. of Minnesota, St. Paul, MN, USA ([email protected]) GE Infrastructure and Process Technologies, Trevose, PA, USA ([email protected]) 3 Univ. of MN., Minneapolis MN, USA ([email protected]) 2

Hg2+ bonds very strongly with natural organic matter (NOM) and NOM is important in the fate and transport of Hg. At environmentally realistic Hg/C ratios, Hg2+ is bonded to thiol S groups. X-ray absorption spectroscopy suggests that most of the Hg2+ is in 2-coordinate thiol S sites when Hg2+ is added to H+ saturated soil humic acid (HA) at less than 0.3 moles of Hg per mole of low oxidation state S (Sred). With more Hg2+, O (or N) sites also participate in the bonding. Because of the very strong bonding to thiol, it is not possible to investigate partitioning of Hg2+ into solution without using a competitive ligand. DTPA and EGTA do not bond Hg2+ with sufficient strength to compete with the S sites in soil HA but dl-penicillamine, a thiol ligand, can desorb Hg2+ from HA to attain equilibrium in 6 to 8 days. We conducted equilibrium experiments with IHSS Pahokee Peat HA varying the ratio of Hg to Sred from 1:1370 to 1:120, the ligand concentration from 0.25 to 5 mM, and the pH from 2 to 6. We determined Koc ([HA-Hg/[Hg2+]) and the apparent formation constant, K, for RS-Hg-SR sites. Koc was affected by all of the variables but pH had the greatest impact and Koc ranged from1022 at pH 2 to 10 33 at pH 6. Variation in Hg loading and penicillamine concentration had little effect on K but the “constant” did increase linearly with pH from 1041 at pH 2 to 1044 at pH 6. The slope of the plot of K vs. pH was 0.65. This suggests that at these low loadings, much less than in the XAFS study, 3-fold thiol sites might also participate in bonding with Hg2+. Results for six soil HA samples from a forest in Northern Minnesota and an associated SOM sample were consistent with the Pahokee Peat results.

Goldschmidt Conference Abstracts 2005 NOM-Metal Complexation


Multielement and rare earth element profiles in an ombrotrophic peat from Germany

Impacts of iron and aluminum ions on solubility of phosphates associated with natural organic matter



Institute of Interdisciplinary Isotope Research, Leipzig, Germany ([email protected])

Introduction Ombrotrophic bogs are suggested to preserve records of enrichments of trace elements. Due to the consistent relationship between the vertical profiles and the time of deposition insights in the historical trends of atmospheric metal depositions can be obtained. Aside the pure deposition of inorganic matter the interaction of the trace elements with the natural organic matter influences their enrichment and fixation in the peat. In this presentation the results of the investigation of two cores of the ombrotrophic bog Kleiner Kranichsee, Ore Mountains, Saxony, Germany, are shown with respect to stratigraphic records of trace elements and in particular to the vertical rare earth element profiles.

Results and discussion The vertical profiles of metals such as K, Mg, Ca, Fe, V, Cr, Mn, Zn, Cd, Pb and U and REE were measured as well as other parameters like ash content, C, H, N, P and S. Due to the comparison of the two cores enrichment processes in various profiles, reflecting changes of the atmospheric impact, could be identified. For REE we found similar trends in all horizons of the vertical profiles, characterized by an enrichment of the HREE vs. the LREE in the chondrite normalized REE pattern. Nevertheless, the profiles of the redoxsensitive REE Ce and Eu exhibit the influence of redox barriers and the influence of the underlying material respectively (Figure 1). Figure 1: Vertical profile of the chondrite normalized amounts of Ce and Eu. 0






Depth [cm]




300 Ce 400

References Shotyk, W., Krachler, M., Martinez-Cortizas, A., Cheburkin, A.K. and Emons, H. (2002) Earth Plan. Sci. Lett. 199, 2137. Weiss, D., Shotyk, W., Schäfer, H., Loyall, U., Grollimund, E. and Gloor, M. (1999) Fresenius J. Anal. Chem. 363, 300305.


USDA-ARS, New England Plant, Soil, and Water Laboratory, Orono, ME 04469, USA ([email protected]) 2 Dept. of Plant, Soil, and Environmental Sciences, Univ. of Maine, Orono, ME 04469, USA The interactions among metal ions (Fe and Al) and phosphates in natural organic matter were investigated. IHSS Elliott Soil humic acid standard, Elliott soil fulvic acid standard II, Waskish peat humic acid reference, Waskish peat fulvic acid reference, phytate and inorganic orthophosphate (all in 0.322 mM P) were separately incubated with control, AlCl3 or FeCl3 (3.22 mM) in 100 mM Na acetate buffer (pH 5.0) at 22oC for 20 h. These mixtures were then equally divided into two aliquots. 3-Phytase (0.1 U/ml) or water was added into these aliquots. After 20-h incubation at 37oC, soluble inorganic P in these mixtures was determined by a modified molybdenum blue method. Less than 1% of phytate P was detected in the absence of the enzyme. About 24% of P in Elliott soil humic acid and fulvic acid, or 5% of Waskish peat humic acid and fulvic acid P was molybdenum reactive P. Enzymatic hydrolysis released 75% of phytate P, 17% of additional P associated with Elliott soil humic acid and fulvic acid, and 2% of additional P associated with Waskish peat humic acid and fulvic acid. Ferric ions precipitated 80% of soluble inorganic orthophosphate. Aluminum ions precipitated only 17% of soluble inorganic orthophosphate. Inclusion of metal additives lowered the detectable P in Elliott soil humic acid to 17%. Both metal ions totally inhibited the enzymatic release of phytate P. In contrast, the inhibitory influence of metal additives on enzymatic release of humic- or fulvicassociated P was not significant. These results suggest that these two metals may have preferential interactions with other functional groups to phosphoesters in humic and fulvic acids.


Goldschmidt Conference Abstracts 2005 NOM-Metal Complexation

Complexation of Pb and Zn by humic substances in contamined soils G. MÜHLBACHOVA1, M. CONTIN2 AND M. DE NOBILI2 1

Research institute of Crop Production, Drnovska 507, 161 06 Prague 6 – Ruzyne, Czech Republic ([email protected]) 2 Dipartimento di Scienze Agrarie ed Ambientali, University of Udine, Via delle Scienze 208, 33100 Udine, Italy ([email protected]) The complexation capability of HA extracted from longterm contaminated and non-contaminated sites close to a lead smelter operating for over 200 hundred years in Pribram (Czech Republic) was tested by coupled ICP-AES-SEC in order investigate the effect of HA on metal mobility and bioavailability. HA from fossile deposits (Leonardite) were also considered. Molecular weights fractions important for metalcomplexing were found in the range of 2500-5000 Da where HA of Leonardite exhibited the major binding capacity. Native HA from contaminated soils demonstrated different capacities of metal complexation. Whereas HA of grassland soil moderately complexed Zn in their structure, the HA of the arable contaminated soil showed only small ability to complex Zn and the major part of the Zn added remained as free ion in the solution.

Goldschmidt Conference Abstracts 2005 Tracing Paleooceanographic Processes

Controls on paleo-alkenone δ13C MARK PAGANI Yale University, Department of Geology and Geophysics, New Haven, CT, USA ([email protected]) The carbon isotopic fractionation that occurs during marine photosynthetic carbon fixation (εp) is primarily a function of surface-water [CO2aq], growth rate, and cell geometry. Although modern data suggest that haptophyte growth rates exert a dominant control on the value of εp37:2, some patterns of paleo-εp37:2 change are contrary to changes in nutrients inferred from foraminiferal trace element concentrations. Further, comparison of εp37:2 values for the past ~45 million years with εp37:2 values from modern growth environments, spanning oligotrophic to eutrophic sites, indicate that changes in algal growth rate is not the first-order control on the long-term trend. If changes in cell geometry of alkenone-producers were minimal, then long-term trends in εp37:2 qualitatively reflect a decrease in [CO2aq] from the middle Eocene to the early Oligocene. Atmospheric carbon dioxide concentrations can be estimated using the modern calibration for εp37:2 as a function of surface-water [PO43-] and [CO2aq], assuming the range of paleo-[PO43-] for each site was similar to modern distributions. This approach yields middle Eocene pCO2 levels ~3 to 5 times that of modern levels. pCO2 rapidly declined following the Eocene/Oligocene boundary reaching modern concentrations near the end of the Oligocene. Recent measurements of coccolith geometries suggest changes in algal geometries of alkenone-producing algae did occur. These results and their impact on carbon dioxide estimates will be discussed.


The importance of a vital effect on the Ca isotopic composition of foraminiferal tests A. GALY, N.G. SIME AND E.T. TIPPER Department of Earth Sciences, University of Cambridge, Downing Street, Cambridge, United Kingdom ([email protected], [email protected], [email protected]) The influence of temperature on Ca isotope fractionation during biomineralisation was investigated through the paired analyses of δ44/42Ca (via MC-ICP-MS) and δ18O on the calcite tests of twelve species of planktonic foraminifera from coretop sediments [1]. No significant correlation between temperature and Ca isotopes was observed in any of the twelve species of foraminifera investigated. The results suggest that the theoretically-expected relationship [2-3] between Ca isotopes and temperature can be obscured by, as yet, unquantified metabolic and physiological processes in nature. Variable growth-rate could be a reason for this vital effect [3] but cannot explain inter-species variations. Vital effects on Ca-isotopes are particularly relevant to the globorotaliid species and G. bulloides in core-top studies but could also explain the discrepancy between laboratorydetermined temperature calibrations and core-top data for G. sacculifer [4]. It is doubtful that the effects of metabolic and physiological processes remained constant through time. This could complicate models of the temporal evolution of the Ca isotopic composition of seawater by introducing a variable fractionation factor between seawater and the carbonate sink.

References [1] Sime N.G., De La Rocha C.L. and Galy, A., (In Press). Earth Planet. Sci. Lett. [2] Marriott C.S., Henderson G.M., Belshaw N.S. and Tudhope A.W. (2004) Earth Planet. Sci. Lett., 222, 615624. [3] Lemarchand D., Wasserburg G.J. and Papanastassiou D.A. (2004) Geochim. Cosmochim. Acta, 68, 4665-4678. [4] Nägler T., Eisenhauer A., Müller A., Hemleben C. and Kramers J. (2000) Geochem. Geophys. Geosyst., 1(2000GC000091).


Goldschmidt Conference Abstracts 2005 Tracing Paleooceanographic Processes

Seawater calcium isotopes from marine barite: A potential record of carbonate deposition in the oceans E.J. MORRIS1, A. PAYTAN1 AND T.D. BULLEN2 1

Department of Geological and Environmental Sciences, Stanford University, Stanford, CA 94305, USA ([email protected], [email protected]) 2 U.S. Geoogical Survey, MS 420, 345 Middlefield Rd., Menlo Park, CA 94025, USA ([email protected]) Carbonate deposition and dissolution significantly affect oceanic alkalinity, atmospheric CO2 and ultimately Earth's climate. Despite over a century’s worth of research on various aspects of the carbonate system, we are just beginning to understand the complexity of the system and elucidate the feedbacks among the processes that control it. The biological precipitation of calcium carbonate largely controls the calcium (Ca) isotope ratio in seawater as a result of the discrimination against heavy isotopes associated with this process. This isotopic fractionation causes seawater to be enriched in the heavier isotopes of Ca relative to its input sources to the ocean, primarily terrigenous and hydrothermal. Reconstructing the seawater Ca isotope ratio over time could be used to quantify the fluctuations in the amount of calcium carbonate deposited in the oceans at any given time assuming some knowledge of the isotopic composition of the sources and sinks. Previously published paleo-seawater Ca isotopic records used bulk marine carbonates, select species of foraminifera, and phosphorites as their source for reconstruction of seawater Ca isotopic composition. However, multiple factors (temperature, precipitation rate, species-specific vital effects and diagenetic alteration) affect the Ca isotope ratio in each of these phases and its preservation of the seawater signal, thus results are complicated to interpret. We present Ca isotope data from marine barite, a phase extracted from marine sediments. Barite forms inorganically in the water column and is not directly associated with specific organisms. Assuming that marine barite forms in isotopic equilibrium with contemporaneous seawater and that Ca is incorporated into its crystal structure, it should record the fluctuations in the δ44Ca of seawater. Preliminary results from coretop samples indicate marine barite does form in isotopic equilibrium with contemporaneous seawater with an equilibrium fractionation factor larger than that interpreted from precipitation experiments for inorganic calcite. When used in conjunction with other phases, marine barite may eliminate complications inherent in previously published data and result in a more coherent paleo-seawater curve.

Zinc isotope variations in phytoplankton and seawater S.G. JOHN1,2, B.A. BERGQUIST1, M.A. SAITO3 1 AND E.A. BOYLE 1

Massachusetts Institute of Technology, Department of Earth Atmos. And Planet Sciences, E34-266, 77 Massachusetts Ave., Cambridge, MA 02139, USA ([email protected]) 3 MS #25, Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA The distribution of Zn in the ocean is largely controlled by biological uptake and remineralization, being drawn down from several nanomolar concentrations in the deep ocean to picomolar concentrations in the surface ocean. Zn isotopes may record this biological activity based on the preferential uptake of lighter Zn isotopes by phytoplankton. We have investigated the Zn isotope composition of modern seawater and plankton tows (Fig. 1), and studied Zn isotope fractionation by phytoplankton in culture, to gain a better understanding of how Zn isotopes may be used to trace biological processes in both the modern and ancient oceans. 66

Sample type/location δ Zn (‰) Seawater Northwest Pacific +.46 Plankton Tows Western S. Atlantic +0.17 to +0.32 Central N. Pacific +.44 Northwest Pacific -0.22 to -0.02 Bering Sea -0.08 to +0.00 Alaskan Shelf -0.44 to -0.39 Figure 1: The isotopic compositions of natural seawater and plankton tows. Isotope measurements were made on an IsoProbe multicollector ICP-MS, with analytical precision between 0.02 and 0.04‰. Seawater was collected from the Northwest Pacific and Zn was extracted by co-precipitation with Mg(OH)2. Our measured value is similar to values measured for continental materials. In-situ plankton were collected by trace metal clean plankton tows from both the Atlantic and Pacific Oceans. Our samples show a range of nearly one permil between the heaviest and lightest plankton tows, which we interpret as evidence of fractionation during biological uptake. The diatom T. pseudonana grown in culture was also observed to fractionate Zn, preferentially taking up lighter Zn isotopes with an ε = +0.2 to +0.4 ‰. Additional seawater and plankton tow samples are being measured in order to better understand the distribution of Zn isotopes in the oceans. Addition species of marine phytoplankton are being studied to compare the isotope effects associated with biological uptake.

Goldschmidt Conference Abstracts 2005 Tracing Paleooceanographic Processes

Iron isotopes in the marine system 12



Boron isotope variation and its environmental implication in Wuquan River Estuary, Hainan Island, China



Massachussetts Institute of Technology, Dept. of Earth, Atmospheric and Planetary Sciences, E34-266, 77 Massachussetts Ave., Cambridge, MA 02139 2 ([email protected])


Iron is an essential micronutrient and is proposed to play a role in climate change by influencing primary production in the ocean. Although the importance of Fe in the ocean has been recognized for the past decade, it is difficult to study because of its complex chemistry and behavior. Fractionation of Fe isotopes could be an effective tool to investigate and quantify the marine geochemistry of Fe. Initial studies of stable Fe isotopes show measurable fractionation in both natural samples and laboratory studies spanning 4‰ (in the 56/54 ratio, δ56Fe). However, most natural samples measured are in the solid phase (high Fe) with only a few studies of natural aqueous or biological samples (typically lower concentrations of Fe). This study addresses questions about the modern Fe cycle using direct measurements of stable Fe isotopes in a variety of natural samples including trace metal clean plankton tows, river samples, aerosol leachates, marine sediment trap samples, and marine porewaters. Fe isotopic composition was measured on a GV Instruments IsoProbe Multi-collector ICPMS. External precision for the δ56Fe measurement is typically better than ± 0.2‰ (2σ) for natural low-level Fe samples using sample-standard bracketing. The δ56Fe of the marine samples varied by over 4‰ with plankton tows showing a large range (-3.87‰ to +0.36‰). The range in the δ56Fe of the plankton tow samples demonstrates that significant and potentially useful fractionation is associated with cycling of Fe in the upper ocean. The Fe in the plankton tow samples in this study was a mixture of intracellular and extracellular Fe. For plankton samples with Fe:C ratios greater than 70 µmol/mol, the δ56Fe values were more variable and became isotopically heavier with increasing Fe:C ratios suggesting that extracellular Fe is isotopically heavier than the intracellular Fe. Plankton samples from the Atlantic scatter around a hypothetical mixing line between a planktonic intracellular δ56Fe of approximately -1.5‰ and an extracellular component of Fe that is isotopically similar to igneous rocks (0‰). The North Pacific plankton tow samples were isotopically lighter in δ56Fe than the Atlantic plankton samples.

MOE Key Laboratory of Coast and Island Development, Nanjing University, Nanjing, China ([email protected]) 2 State Key Laboratory of Mineral Deposit Research, Nanjing University, Nanjing, China The boron-isotope composition of authigenic carbonate skeletons may provide a useful tool to record secular boronisotope variations in seawater at various times in the geological record. The potential use of boron-isotope geochemistry in skeletons can be as a tracer for palaeoenvironments (Vengosh et al., 1991). Gastropoda skeletons for this study were isolated from Core K4 collected in Samei Lagoon located in Wuanquan River Estuary, Hainan Island, China. Their isotopic compositions of boron were measured by TMIS with the method of negative thermal ionization producing BO2- ions. The results are shown in the table below. Sample Depth (cm) 14-15 34-35 125-126

Group Gastropoda Gastropoda Gastropoda

δ11B(‰) 6.3 10.0 10.4

δ11B values of Gastropoda skeletons range from 6.310.4‰. Below the depth of 34-35 cm in the core, the values are about 10‰, which are similar to those in modern marine biogenic carbonates (13.3-32.2‰) (Barth, 1993). In the depth of 14-15 cm (about 1900 AD, Ge et a.l, 2003), the value is relatively lower (6.3‰), which may reflect a more “terrestrial” boron-isotope signature of the water (Vengosh et al., 1991). δ11B record implicates that Shamei Lagoon was formerly an embayment and since 1900, the water salinity has become much lower, which is consistent with those of organic carbon isotope values and C/S ratios (Ge et al., 2003).

References Barth S. (1993) Geol Rundsch. 82, 640-651. Ge C.D., Slaymaker O. and Pedersen T.F. (2003) Chinese Sci Bull. 48, 2357-2361. Vengosh A., Kolodny Y., Starinsky A. et al (1991) Geochim Cosmochim Acta. 55, 1689-1695.


Goldschmidt Conference Abstracts 2005 Tracing Paleooceanographic Processes

Sulfur multiple isotope constraints on the Cenozoic-Cretaceous sulfur cycle

Using the multiple isotopes of sulfur to constrain microbial processes in the Proterozoic ocean




ESSIC and Department of Geology, University of Maryland, College Park, MD 20742, USA ([email protected], [email protected], [email protected]) 2 Geophysical Laboratory, Carnegie Institution of Washington, Washington DC 20015, USA ([email protected]) 3 Department of Geological and Environmental Sciences, Stanford University, Stanford CA 94305, USA ([email protected]) Marine barite provides a robust record of the isotopic evolution of seawater sulfate (Paytan et al., 1998, 2004). We are extending the existing data set of δ34S values with highprecision sulfur multiple isotope measurements of marine barite from the last ~120 Ma. Measurements of 33S and 36S abundances provide information that is complementary to 34S abundances for two reasons, both of which require highprecision measurements to be meaningfully applied. First, individual processes can fractionate sulfur multiple isotopes along trajectories other than the reference mass-dependent fractionation lines that are defined by ∆'33S = ln(δ33S/1000+1) – 0.515 × ln(δ34S/1000+1) = 0, and ∆'36S = ln(δ36S/1000+1) – 1.9 × ln(δ34S/1000+1) = 0. Second, the trace abundance approximation is valid for the sulfur multiple isotope system, which results in linear isotope mass-balance equations. Isotope fractionation follows an exponential relationship, however, and the co-operation of fractionation and mass balance can produce nonzero ∆'33S and ∆'36S. Recent calibrations of sulfur multiple isotope effects due to different bacterial metabolisms (e.g., Johnston et al., 2005) allow for a rigorous interpretation of the sulfur multiple isotope evolution of seawater sulfate. We use these calibrations in simple isotope mass-balance models to constrain relative mass fluxes through different sulfur conversion pathways. External reproducibility in sulfur multiple isotope measurements are <0.01‰ for ∆'33S and <0.1‰ for ∆'36S. At their present range of variability (e.g., ∆'33Smax-∆'33Smin ~ 0.08; ∆'36Smax-∆'36Smin ~ 1.0), our measurements are producing an information-rich record (∆'33S signal/noise > 8, ∆'36S signal/noise > 10) of the CenozoicCretaceous sulfur cycle.

References Johnston, D. T., Farquhar, J., Wing, B. A., Kaufman, A. J., Canfield, D. E., and Habicht, K. S. (in press) Am. Jour. Sci. Paytan, A., Kastner, M., Campbell, D., and Thiemens, M. H. (1998) Science, v. 282, p. 1459-1462 Paytan, A., Kastner, M., Campbell, D., and Thiemens, M. H. (2004) Science, v. 304, p. 1663-1665


Dept. of Geology and ESSIC, Univ. of Maryland ([email protected] ) 2 Dept. of Earth Sciences, Univ. of California, Riverside 3 Dept. of Earth and Planetary Sciences, Univ. of Tennessee 4 Geologisch-Paläontologisches Inst, Univ. Münster 5 Inst of Biology, Univ. of Southern Denmark It has been argued that widespread sulfidic deep ocean conditions during the Proterozoic resulted from a predominance of sulfate reducing bacteria (SRB) and their effects on the biogeochemical sulphur cycle [1]. It has further been suggested that sulfur disproportionation (SDB) metabolisms did not begin to play a significant role in the cycling of sulfur until early in the Neoproterozoic [1]. Differences in the metabolic style of SRB and SDB cause measurable differences in the ∆'33Ssulfate that can be used to evaluate these hypotheses [2, 3]. Here we present measurements of the four stable isotopes of sulfur (32S, 33S, 34S, 36S) for proxies of seawater sulfate (CAS, evaporite sulfate, and barite) that we use to evaluate these hypotheses and to test our earlier proposals. Reproducibility of these measurements are 0.01 ‰ or better, clearly demonstrating a small negative excursion of ∆'33Ssulfate from samples older than ~1 Ga and a small positive excursion for ∆'33Ssulfate from samples younger than ~1 Ga (Fig 1.).

Figure 1: ∆’33S of oceanic sulfate proxies versus time. We interpret negative ∆'33Ssulfate to imply a SRB dominated biogeochemical sulfur cycle while small positive ∆'33Ssulfate may reflect the added influence of SDB, both have implications for the oxidation state of the Proterozoic ocean.

References [1] Canfield, D.E. (1998) Nature 396, 450-453. [2] Farquhar, J. et al. (2003) Geobiology, 1, 27-35. [3] Johnston, D.T. et al., (in press), AJS.

Goldschmidt Conference Abstracts 2005 Tracing Paleooceanographic Processes


Dynamic ocean chemistry around the Marinoan glaciation – Isotopic evidence from cap carbonates

Iron isotope constraints on the Archean and Paleoproterozoic ocean redox state




Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing, China ([email protected]) 2 Geologisch-Paläontologisches Institut, Westfälische Wihelms-Universität Münster, Münster, Germany ([email protected]) Hurtgen et al. (2002) suggested that the oceanic sulfate pool would become strongly enriched in 34S during the Neoproterozoic snowball Earth due to continuous bacterial sulfate reduction and lack of supply of S from continental weathering. Overturn of this deep ocean with high δ34Ssulfate would have caused a large positive excursion in the δ34S of CAS (carbonate-associated sulfate) in cap carbonates. Comparable large positive excursions in δ34Ssulfate, more than 25‰, occur in the Doushantuo cap carbonates following the Marinoan glaciation as shallow-water platform facies on the Yangtze Block. Yet, δ34S for CAS from the deep-water cap carbonates of the Jinjiadong Formation do not support the Hurtgen et al. (2002) model. Significantly different from the Doushantuo Formation, CAS from carbonate of the Jinjiadong section has lower δ34S and δ18O values: 11.2 to 19.7‰ with an average of 14.9 ‰ for S and 5.0 to 8.0‰ with an average 6.6‰ for O, whereas the δ34S and δ18O values from the Doushantuo cap carbonates fall mostly between 25 to 45‰ and 12 to 18‰ respectively. In the Jinjiadong cap carbonates, the δ34S values for pyrite range from 6.1 to 22.9‰ with an average of 14.8‰. Comparable δ34S values for CAS and pyrite suggest two possibilities: 1) the pyrite was formed through bacterial sulfate reduction but at very low sulfate concentration; or/and 2) the sulfate originated from reoxidation of bacterially derived H2S. A negative correlation between carbonate δ13C and CASδ34S supports that BSR probably happened in a stagnant ocean. However, the relatively uniform S and O isotope data for CAS imply that the deep-water, representing coeval ocean water, may have had of δ34S values at 15‰ and δ18O values around 7‰, rather than the high δ34S values described by Hurtgen et al. (2002). Therefore, a dynamic model is proposed to interpret short-term variations in S, C, and O isotopes of global ocean water following the Marinoan glaciation.

Reference Hurtgen M.T., Arthur M.A., Suits N.S., Kaufmann A.J., (2002), Earth Planet. Sci. Lett. 203, 413-429.


Marine Chemistry & Geochemistry Dept., Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA ([email protected]) 2 Geophysical Laboratory, Carnegie Institution of Washington, 5251 Broad Branch Road, NW, Washington DC 20015, USA ([email protected]) 3 Geomicrobiology Group; Marine Chemistry & Geochemistry Dept., Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA ([email protected]) The response of the ocean redox state to the rise of atmospheric oxygen by ca. 2.3 Ga ago is poorly constrained although deposition of BIFs until ca. 1.8 Ga suggests at least episodical deep ocean anoxia during the Paleoproterozoic. Since Fe, along with C and S, are coupled with and maintain the redox state of the surface environment, Fe seawater concentration and isotopic composition were likely affected by the change in the redox state of the atmosphere. We will present a study of Fe isotope composition of sedimentary sulfides over geological time and provide evidence for a change in the ocean Fe cycle at the same time as atmospheric redox state changed. We analyzed Fe isotope compositions of ~ 150 pyrites from 20 black shale units, specifically focusing on Late Archean to Paleoproterozoic time. δ56Fe values of handpicked sulfides were obtained using a Neptune MC-ICPMS at WHOI and are reported relative to IRMM-14 with an external precision of 0.1‰ at 2σ level. The emerged general pattern of Fe isotope record allows dividing the Earth's history into three stages which are strikingly similar to the stages defined by the δ34S and ∆33S as well as other indicators of the redox state of the atmosphere and ocean. (1) Stage 1 (>2.8 to 2.3 Ga) is characterized by highly variable and negative δ56Fe values of pyrite (down to – 3.5‰) that are interpreted to reflect the reservoir effects during partial oxidation of hydrothermally-derived Fe(II) and precipitation of Fe-oxides. (2) Stage 2 (2.3 to ~1.6 Ga) is characterized by δ56Fe values ranging from -0.3 to 1.2‰ that might be related to the increased effect of sulfide precipitation in a redox-stratified ocean. (3) Stage 3, from 1.6 Ga through the Phanerozoic, is characterized by sedimentary pyrite having a limited range of δ56Fe variations (less than 0.5‰ around igneous value at ~0‰) reflecting the establishment of an Fepoor oxygenated ocean.


Goldschmidt Conference Abstracts 2005 Tracing Paleooceanographic Processes

Ion microprobe carbon isotope analysis of Archean microfossils? ALAN JAY KAUFMAN1, SHUHAI XIAO2 3 AND LEIMING YIN 1

Geology Department, University of Maryland, College Park, MD 20742-4211, USA ([email protected]) 2 Department of Geosciences, Virginia Polytechnic Institution, Blacksburg, VA 24061, USA ([email protected]) 3 Nanjing Institute of Geology and Paleontology, Nanjing 210008, China Metapelites of the Archean Wutai Group (>2.5 Ga) of North China have been found to contain abundant graphite discs. These discs are confirmed to be composed of carbon by elemental mapping through electron and Raman microprobe analyses. Application of the graphite Raman geothermometer suggests these discs experienced metamorphic temperatures of ca. 500oC, consistent with the amphibolite grade of the host rock. We investigated these circular discs by ion microprobe techniques to determine whether they are metamorphic or biological in origin. It has been suggested that these simple discs, which are 20-220 µm in diameter, were originally spherical vesicles of acritarchs (potentially eukaryotic photoautotrophs) with a thick resistant organic wall, and that these were deflated and compresed during compaction and diagenesis. Some of the discs show evidence of folding, breakage, and medial split, all of which are seen in younger acritarchs with coherent and robust walls. If correct, the discovery of these acritarchs would extend the range of eukaryotic fossils back into the Archean, consistent with recent biomarker data. Ion microprobe analyses of five different discs were conducted with the Cameca 6f ion microprobe at high mass resolution at the Carnegie Institution of Washington. The magnitude of instrumental mass fractionation (IMF) inherent to surface ionization mass spectrometry was quantified by the repeated (n = 7) analysis of the standard Mao diamond (δ13C = -6.5‰, IMF = 53.1 ± 0.4‰). The primary Cs+ beam intensity was 0.5 nA and was focused down to a 20-25 µm spot, which allowed for multiple analyses of the same individual. In most cases the variability of δ13C values between spots on the same individual was small (< 2‰), but the range of values between different individuals spanned from -7.3 to -35.8‰. The overall average of all analyses is close to the -21.3‰ value determined by standard techniques for bulk kerogen. The wide variability between individuals may be tentatively interpreted as biological in origin, insofar as metamorphic graphite measured by the same ion microprobe technique shows much smaller magnitude of variation in primary grains lacking secondary overgrowths.

Molybdenum isotope prospects A.D. ANBAR1 AND G.A. WILLIAMS2 1

Dept. of Geological Sciences and Dept. of Chemistry & Biochemistry, Arizona State University, Tempe, AZ, USA ([email protected]) 2 Dept. of Geological Sciences, Arizona State University, Tempe, AZ, USA ([email protected]) The oxygenation of the oceans has varied through time but the timing and extent of these variations are not well understood, nor do we have a good understanding of the connections between changes in atmospheric and ocean oxygenation. The development and refinement of ocean paleoredox proxies is therefore a high priority. The molybdenum (Mo) stable isotope system is emerging as a valuable tool for investigating ocean paleoredox. To first order, Mo enters the oceans via rivers and is removed in oxic environments by adsorption to Mn oxides and in sulfidic settings via scavenging of Mo oxythiomolybdates. The isotopic contrast between these sinks is ~ 2 ‰. Hence, the steady-state Mo isotope composition of the oceans should reflect the balance between Mo removal to Mn-oxides vs. removal in sulfidic settings. Fractionation occurs during removal to Mn-oxides. Today, this process generates a steadystate isotopic offset between average crustal Mo (~0‰) and seawater (~ 1.6‰). This offset would have been smaller - i.e., seawater isotopically lighter - during extended periods of expanded ocean anoxia because a smaller fraction of isotopically light Mo would have been buried in association with Mn-oxide sediments. This system is particularly valuable because it may constrain regional or global ocean redox, rather than only local redox, as a result of the long ocean residence time of this element. Mo isotope systematics are already being used to obtain qualitative and semi-quantitative information about ancient ocean redox (e.g., Arnold et al, 2004). Quantitative applications require better understanding of a number of parameters, including: The importance of suboxic (O2 < 5 µM) settings for Mo removal and isotope fractionation; the magnitude of isotope fractionation during adsorption to Mn oxides and the sensitivity of this fractionation to temperature and other variables; and Mo isotope variability due to weathering and riverine transport. The current status of research in these areas will be reviewed.

Reference Arnold G.L., Anbar A.D., Barling J. and Lyons T.W., (2004), Science 304, 87-90.