Journal of Conference Abstracts

Volume 4 Number 1


Symposium J10
Aquatic Geochemistry



Session J10:2B

KEYNOTE
J10 : 2B/29 : G6

Micro-Niche Mobilazation of Metals

Bill Davison (Billd@det.csiro.au)

Lancaster Univ, Bailrigg, Lancaster, LA1 4YW, Lancaster, UK

Traditionally solute concentrations in pore waters have been regarded as being horizontally homogeneous, allowing steady state one-dimensional models to be used to describe their variation with depth. The technique of DGT (diffusive gradients in thin-films), whereby ions are accumulated in a binding layer after diffusing through a thin layer of gel, is providing new insights into interactions between solutes and solid phases in sediments. DGT and its sister technique of DET (diffusive equilibration in thin-films) can be used to measure solutes including trace metals and dissolved sulfide in pore-waters at mm and sub-mm spatial resolution in one and two dimensions. Deployments in freshwaters and deep and shallow marine systems have demonstrated lateral as well as vertical concentration gradients of trace metals and highly localized sources of remobilization. Such features can not be explained by the traditional view of steady-state pore-water concentrations. The observed structure has been successfully modeled on both macro (cm) and micro (sub-mm) scales by considering sediments as random arrays of micro-niches of varying density. The results question the simple use of vertical concentration gradients obtained by single measurements or at cm-scale resolution to measure exchange fluxes and highlight the need for the simultaneous measurement of both horizontal and vertical gradients. The implications of these results in terms of the dynamic and temporal nature of pore-water concentrations, the origins of micro-niches and the selection of measurement scales are discussed.

J10 : 2B/31 : G6

Carbon and Nitrogen Isotope Geochemistry of Lacustrine Organic Matter as a Tool for Determining Paleoproductivity and Sediment Transport

Stefano M. Bernasconi (stefano@erdw.ethz.ch)1,

Jane L. Teranes (jteranes@umich.edu)2,

Yvonne Van Lith1 &

Fuchs Frederic1

1 Geologisches Institut, ETH-Zentrum, 8092 Zurich, Switzerland
2 Department of Geological Sciences, University of Michigan, Ann Arbor, MI48109-1063, USA

Organic matter is an important component of settling particles and sediments in lakes and is the most important factor controlling the oxygen budget of the bottom waters and the cycling nutrients and trace metals. The quantification of fluxes and accumulation rates of organic matter are, therefore, important parameters for biogeochemical lake models. Secondary sedimentation processes such as resuspension of bottom sediments, sediment focusing or transport at depth of riverine detrital matter however, often impair the precise determination of sediment accumulation rates with sediment traps. Therefore tracers for sediment transport are needed to improve the understanding of the organic matter cycling. The carbon and nitrogen isotopic composition of sedimentary organic matter have been used as proxy indicators of past changes in primary productivity but the exact controls on the isotopic composition of lacustrine organic matter are still poorly understood.

Towards these goals, particulate organic matter (POM) and its carbon and nitrogen isotopic composition were measured with sediment traps in Lake Lugano and lake Baldegg (Switzerland) to evaluate their usefulness as tracers for sediment transport in the water column and as paleoproductivity proxies. In addition, the we studied the nitrogen and carbon isotope composition of sediment cores to compare them with the historical records of productivity changes. In both lakes 13C and 15N of POM show large seasonal changes. In Lake Lugano 13C ranges between -40‰ in winter and -22‰ (PDB) in summer whereas in Baldeggersee 13C varies between -36‰ and -30‰. Nitrogen isotopes show an opposed trend with heavy compositions in winter and lower in the summer. In Lake Lugano 15N of POM vary between 4‰ and 16‰ (AIR) and in Lake Baldegg between 11‰ and 18‰. The variations in 13C can be attributed to variations in primary productivity levels, changes in the carbonate chemistry, and isotope discrimination during photosynthesis. The variations in 15N, however cannot be explained by a Raleigh distillation process related to nitrate availability as observed for marine environments, and are not a direct indicator of lacustrine productivity. The variations in carbon isotope composition of organic matter could also be used to show that sediment resuspension processes are not responsible for the increased accumulation rates of organic matter observed in the deep sediment traps in Lake Lugano.

Large variations in carbon and nitrogen isotope compositions are also observed in the sediment cores and their variations can be correlated with the trophic evolution of the two lakes. However the lateral variations observed by the comparison of multiple cores suggest that care must be used in the interpretation of sedimentary records of carbon an nitrogen isotopic compositions of lacustrine organic matter.

J10 : 2B/32 : G6

Association of Cd, Cu, Pb and Zn with Colloids in Lake Geneva Surface Water

Celine Gueguen (gueguen@sc2a.unige.ch)1 &

Janusz Dominik (dominik@sc2a.unige.ch)2

1 Institut F.-A. Forel, 10 Route de Suisse, CH-1290 Versoix, Switzerland
2 Centre d'Etudes Naturelles en Sciences de l'Environnement, Universite de Geneve, 10 Route de Suisse, CH-1290 Versoix, Switzerland

Despite a relatively high input of metals to the pre-alpine lakes, their concentrations in surface water remain low which suggests a relatively short residence of metals in the epilimnion. Colloids play an important role in removal of metals from estuaries and coastal ocean but virtually nothing is known on colloid-metal association in lakes. This paper presents first data on repartition of metals between colloidal and dissolved fractions in surface water from the central part of Lake Geneva. Water samples were filtered at 1.2 µ m and ultra-filtered using tangential flow filtration device with a regenerated cellulose membrane (nominal cut-off 1 kD). After colloid preconcentration (concentration factor < 62), retentate was ultracentrifuged to separate < 20 nm fraction. Measurements of trace elements in water requires extreme care during sampling, preparation and analysis. All samples were handled under ultra-clean conditions. Water samples were exposed to UV-irradiation to improve metal retention on resin during the preconcentration step. The possible variations of efficiency during irradiation and/or preconcentration on the resin were corrected using the isotopic dilution technique. Metal concentrations were measured with a High Resolution Inductively Coupled Plasma Mass Spectrometry. Chlorophyll, nitrates, pH were also measured.Distribution of Cd, Cu, Pb and Zn between colloidal and ultrafiltered fractions was determined at two periods, in March during first plankton bloom, and in July, when chlorophyll concentration was higher. The concentration of colloids was 7 times higher in July than in March. Cu and Pb were predominantly found in the ultrafiltered fraction in spring, while at the end of the high plankton productivity in July, up to 50% of these metals was associated with colloidal fraction. Actually, Cu and Pb are known for their affinity to organic components, principal constituent of natural colloids. The metal partitioning between colloids and ultrafiltered fraction (Kc) was also calculated. Pb exhibits the highest distribution coefficient, synonymous of a stronger affinity for colloidal fraction, whereas Cd the lowest. These preliminary results indicate that even at the relatively constant total metal concentrations their availability to phytoplankton may evaluate with time depending on the degree of association with colloids.

J10 : 2B/33 : G6

The Chemical Coupling Between Suspended Matter and Sediment in Lake Kutsasjärvi, Northern Sweden

Elsa Peinerud (Elsa.Peinerud@sb.luth.se)

Div. of Applied Geology, Lulea University of Technology, 971 87 Lulea, Sweden

In Lake Kutsasjärvi, an unpolluted lake in northern Sweden, the chemistry of the sediment was compared with a time series of suspended matter (particles > 0.45 µm). The suspended matter is largely influenced by biogenic particles and secondary Fe and Mn oxyhydroxides, and therefore its composition varies over the year.

Major elements as well as trace elements are affected by the non-detrital Fe and Mn oxyhydroxides. Adsorption of e.g. Ca and Mg on suspended non-detrital Fe oxyhydroxides is controlled by distribution equilibria between dissolved and solid concentrations.

The concentrations of trace elements are partly controlled by dilution with trace-element poor phases and association to non-detrital phases. The concentrations of the REE are between 3 and 4 times higher in the sediment than in the suspended matter. Dilution alone is not sufficient to explain this difference, since the concentration of the diluting phases, mainly organic matter and diatom silica, is only about 1.5 times higher in the suspended matter. An additional explanation may be the generally more detrital character of the sediment, possibly because detrital particles settle out of the water column much sooner than other types of particles.

The enrichment of trace elements in the suspended phase due to adsorption onto non-detrital Fe oxyhydroxides is in some cases overshadowed by dilution. Adsorption on non-detrital Mn oxyhydroxides, as for Hg and Ni, is more likely to render higher concentrations in the suspended matter than in the sediment, since the average concentration of non-detrital Mn in the suspended phase is about 15 times higher than in the sediment. The importance of uptake by biota of e.g. Cu and Zn is also discussed in this context. When the carrier phases, non-detrital Fe and Mn oxyhydroxides and organic matter, become buried in the sediment, they partly dissolve or decompose, resulting in a release of trace elements. Diagenetic processes, including e.g. secondary enrichment, influence different trace elements differently. Diagenetic behaviour is thus a third mechanism controlling the distribution of trace elements between sediment and suspended matter.

J10 : 2B/34 : G6

Different Behaviour of Fe and Mn in Anoxic Sediments of Eutrophic Monomictic Lake (Aydat, France) Evidenced by Sequential Chemical Extractions

Sylvie Ogier (Sylvie.Ogier@univ-orleans.fr),

Bernard Guillet (Bernard.Guillet@univ-orlens.fr) &

Jean-Robert Disnar

(Jean-Robert.Disnar@univ-orleans.fr)

UMR 6531 du CNRS Sédimentation & Diagenèse de la Matière Organique, Université d'Orléans, BP 6759, 45067 Orléans cedex 02, France

The mobility of elements in aquatic systems depends on their chemical forms which is controlled by physico-chemical characteristics of the environment (pH, Eh, ionic strength) and their selective affinity for organic and mineral surfaces. In order to study the speciation of Si, Al, Fe, Mn, Mg, Ca, Na, K, P, Ba, V and Sr, three successives chemical extractions were performed on 23 samples from 35 cm long sediment core. The extractions were carried out at ambient temperature on humid sediments with an air dried material weight/solution ratio of 8 g/100 cc. The first treatment with NH4NO3 M releases soluble and exchangeable elements. The second with EDTA-Na2 0.05 M (pH=4.6) extracts elements complexed by organic and mineral ligands. Finally, a mixed solution of hydroxylamine-HCl 0.1 M and HNO3 0.01 M reduces FeIII and MnIV mineral surfaces and solubilizes the associated elements.

Sediments consist predominantly of biogenic fractions composed of amorphous silica skeletons of diatoms (50-65%) and organic material (TOC=10.7 to 5.7%). The other components are authigenic minerals (amorphous Fe-Mn-hydroxides) and terrigenous minerals of volcanic origin (10-22% of bulk sediments) like clay minerals and Fe-Ti oxides.

The NH4NO3 extraction mostly releases Mn with an average value of 0.02 meq g-1. Contrary to Mn, Fe is not extracted. The different behaviour of these two elements in anoxic bottom sediments can be explained by pH-Eh conditions under which Mn exists under Mn2+ soluble form whereas Fe is under FeIII hydroxide form. The other released elements are in decreasing order of concentration Mg, Ca, K, Na, Ba and Sr. The vertical distribution of these elements is comparable to that of TOC. Their higher contents in the five upper centimetres suggest that organic compounds highly contribute to the cation exchange capacity of the sediment.

The EDTA-Na2 extraction releases Fe (20 mg.g-1), Si (2 mg.g-1), Al (1 mg.g-1) followed by P and V. All these elements present similar concentration profiles. P and V are strongly correlated with Fe (R2P-Fe=0.91, R2V-Fe=0.95) that is in agreement with element speciation in Aydat lake where these anions are adsorbed on Fe-hydroxides. The relatively minor proportions of Si and Al released with Fe may probably be explained by an association with poorly crystallized Fe-hydroxides which can develop on mineral surfaces or within organic aggregates.

The final Hydroxylamine-HCl and HNO3 treatment entails a strong release of Fe with very few amounts of P and V, but also Al, Ca, Ba and Sr which present similar concentration profiles. The former result suggests alteration of detrital Fe-oxides while the latter indicates the sensitivity of plagioclase feldspars and clay minerals to acid extraction.

Session J10:3A

J10 : 3A/01 : G6

Geochemical Migration of Elements in a Lake Under Influence of Mining and Metallurgical Activities

Vladimir Dauvalter (vladimir@inep.ksc.ru),

Tatyana Moiseenko &

Ilia Rodyushkin

Institute of North Industrial Ecology Problems, Kola Science Centre, Russian Academy of Sciences, 14 Fersman St., Apatity, Murmansk region, 184200, Russia

The Kola Peninsula is rich in mineral deposits, containing for instance copper-nickel, apatite-nepheline, rare earth elements (REE), iron, platinum and chromium. This has led to the development of an important mining and metallurgical industry, with resulting release of a number of elements, more or less toxic for biota. Lake Imandra is the largest in Kola North. It is oligotrophic and ultrafresh, but is now polluted by wastes and emissions from the Apatite, Severonickel and Olenegorsk plants. Detailed studies of water parameters and element concentrations (in both dissolved and suspended phase) in depth profiles of the water column of Lake Imandra were performed during 1995. Sediments and pore waters in the sediments were also studied.

The first up-to-date concentration profiles of trace elements such as Hg, Ni, Cu, Pb, Ga, Mo, Ba, U, Th, REE etc. were obtained. Imandra Lake water enrichments in Ni, Mo, Rb, Cu and Sr. Their concentrations are more than 10, 8, 5, 3 and 2.5 times higher than average background values for the Kola Peninsula, respectively. The suspended fraction caries a very low part of the total concentration of alkali elements, but is important for Fe, Al and REE. The redox cycle of Mn controls the concentrations of elements such as Ba, Mo, Ni and Ga in the water column.

The industrial activities changed the geochemistry of the sediments. Sediment composition is the result of several complex processes in the water column such as 1) accumulation of sedimented solid suspended particles (Al, Ti, Ca, Mg, K, Na, P, Rb, Sr, Ta, U, V, rare-earth elements play the basic role), 2) uptake by living organisms and adsorption by other high-molecular organic compounds (Ti, Mn, P, As, Ba, Cd, Cu, Hg, Mo, Ni, Pb, Rb, Sr, Ta are involved into this process), 3) redox processes of Fe, Mn and S in the water column and in the sediments (mainly Mn, Fe, As, Ba, Be, Cd, Cu, Hg, Mo, Ni, Co, Zn). Onset of alteration of sediment geochemistry was recorded at the depth of 10 cm. The highest enrichment in the upper sediment layers in comparison with background values was found for Ni, Pb, Hg, Cu, Nb, Ta, Cd, As, Mo, Co, REE (30, 7.5, 7, 5.1, 4.1, 3.6, 2.8, 2.5, 2.3, 2.3, 2, 1.5-3, respectively). More pronounced changes are observed from a depth of 7 cm, which correlates with the inception and growth of the mining and metallurgical activities in the Barents region.

J10 : 3A/02 : G6

Rare Earth Elements Speciation in River Sediments

Lydia Leleyter (leleyter@illite.u-strasbg.fr)1,

Jean-Luc Probst (jlprobst@illite.u-strasbg.fr)1,

Pedro Depetris (pdepetris@powernet.com.ar)2,

Souad Haida (souad_hs_haida@yahoo.com)3,

Jefferson Mortatti (jmortatt@pintado.ciagri.usp.br)4,

Robert Rouault (rouault@illite.u-strasbg.fr)1 &

Jean Samuel (jsamuel@illite.u-strasbg.fr)1

1 Centre de Géochimie de la Surface, EOST, 1 rue Blessig, Strasbourg, France
2 Université Nationale de Cordoba, Département de Chimie Analytique Minérale, Avenida Velez Sarsfield 299, Cordoba 5000, Argentina.
3 Département de Géologie, Université Ibn Tofail, Faculté des Sciences, B.P.133, Kénitra, Morocco
4 Centro de Energia Nuclear na Agricultura (CENA), Universidade de Sao Paulo, Campus de Piracicaba, Avenida Centenario 303, CP 96, 13 400, Piracicaba, SP, Brazil.

In aquatic systems, rare earth elements (REE) are mainly transported in the suspended load. The sequential extraction procedure (Leleyter and Probst, in press), applied on 37 river suspended matter or bed sediments from Patagonian rivers (Argentina), Piracicaba river (Brazil), Ill and Garonne rivers (France) and Sebou river (Morocco), allowed to identify the particulate fractions which contribute to REE fluvial transport. The results show that REE from the leached fractions originate from carbonates, Fe-oxides and organic matter. The dominating fraction essentially depends on the initial composition of the sediment. For the Patagonian river sediments, rich in clay minerals, REE mainly originate from the residual fraction. The leached REE from the Piracicaba river sediments, rich in iron oxides, are mainly associated with iron oxides, whereas the leached REE from Ill river sediments, rich in organic matter, are mainly linked with organic matter and the leached REE from Sebou river samples, rich in carbonates, are mainly supply by carbonates dissolution.

These results show that the leached REE don't behave as a coherent group: they don't have the same preferential scavenged sites. Indeed carbonates adsorb mainly the middle REE and particularly europium. Thus the abundance of carbonates in a river sediment can even cause a positive europium anomaly whereas the residual fraction of the sediment presents a negative europium anomaly. This affinity of europium towards carbonates is due to the similitude of the ionic radius of Eu2+ with Sr2+ which can provoke some substitutions in carbonates. Organic matter is mainly associated with middle and light REE, except cerium. The organic matter of the sediment can cause a light and middle REE enrichment and sometimes a negative cerium anomaly in the total sample whereas the residual fraction of this sample present a positive cerium anomaly. On the contrary, oxides adsorb always preferentially the heavy REE and cerium. Thus the abundance of oxides can explain the positive cerium anomaly of the total sample whose residual fraction present a negative cerium anomaly. The affinity of cerium towards oxides is due to the oxidation of Ce(III) in Ce(IV)O2 which is a solid phase. These results imply that the REE signature of river sediments is linked to the relative abundance of carbonates, oxides and organic matter in the sediment, which is also controlled by the physico-chemical composition of the aqueous phases (specially pH and redox potential). These results are fundamental to study the signature of the continental crust erosion using the river transports of dissolved and suspended matters and the signature of river inputs to the oceans. Most of previous results obtained in the literature could be revised in the light of these results.

Leleyter Land Probst JL, Int. J. Environ. Anal. Chem, in press.

J10 : 3A/03 : G6

Determination of Interactions between Ground and River Water: the Ra and Sr Isotope Approach

Jost Eikenberg (eikenberg@psi.ch)1 &

Aude Tricca2

1 Paul Scherrer Institute, Switzerland
2 California Institute, of Technology, Pasadena, CA 91125, USA

Ground and river waters of the Upper Rhine Valley (Alsace France) were investigated for chemical composition of the major elements, Sr isotopes and radionuclides from the U and Th series. In particular, the isotope ratios and concentrations of Ra and Sr were used as geochemical tracers to distinguish between different types of waters and their replacements. Usingthe combination of the radiogenic isotopes 224Ra, 226Ra and 228Ra on the one side and stableisotopes of Sr (i.e. 87Sr, 86Sr) on the other, it was possible to clearly distinguish between three end member water components from different hydrogeological origin, i.e. plain groundwater, Rhine river water and highly radiogenic, less mineralized spring waters from the Vosges mountains. Strong interactions between these three components were obtained by investigation of further surface waters in the plain (river Ill and smaller feeder rivers) before their confluence with the river Rhine at Strasbourg. The displacement of these plain waters by groundwater along the flow path of the Ill could be precisely determined using Sr istope relationships (i.e. 87Sr/86Sr vs. 1/Sr) and the mixing fractions obtained there agreed well with those results obtained by 228Ra/226Ra vs. 224Ra/226Ra. Since with respect to 87Sr/86Sr the components Rhine water and groundwater are almost identical it was interesting to observe that the Ra isotope system could be successfully applied to study more closely the displacements between the Rhine water, groundwater andthe other surface waters. The combination of Ra and Sr isotope data provides hence a valuablekey to determine interactions between ground and surface water in hydrologically complex regions.

J10 : 3A/04 : G6

The Organic Matter from Fresh Water: A Potential Carrier of Trace Metals in the Tropical Regions

Stephane Mounier (mounier@univ-tln.fr),

Nathalie Patel (patel@univ-tln.fr) &

Jean-Yves Benaim (benaim@univ-tln.fr)

Université de Toulon - BP 132, Laboratoire RCMO, Bat R, 83957 La Garde Cedex, France

The Amazon basin is the largest river catchment area in the world. It extends over 6x10-6 km2. As well as being the largest drainage basin with the highest discharge volume, 209,000 m3/s, it contributes to 20% of the world's fluvial hydrological input. Sampling was carried out during low water levels (March 1995). Sequential tangential ultrafiltration (STUF) was used to fractionate the organic matter. The cut-off values were at 0.22 µm, 100 kD and 5 kD, giving the organic pool separated in particulate, large colloidal, fine colloidal and dissolved material. Total organic carbon (TOC) analysis was measured for each group with a carbon analyser, using a wet oxidation method performed by the action of persulfate combined with UV irradiation. By fluorescence quenching, copper complexing capacities (CL) measurements were done. Results show that particulate and colloidal organic matters are the predominant contributions to black water rivers TOC (73%). For the white waters, the dissolved fraction is the predominant one (61%). The dissolved molecules are responsible for the complexing capacity for the Rio Negro waters (59%), with little contribution from the particulate and colloidal fractions. Similar behaviour was observed for the white river water. The CL/TOC ratio defined as the quantity of available complexation sites (Benaim, 98), is also a parameter used to compare different samples and fractions. The results indicate that the sites density show a different behaviour for black (123.5 µmol of Cu per µmol of C) and white waters (2880.0 µmol of Cu per µmol of C). In fact, this parameter is higher in the dissolved fraction of the black water body and, in contrast, higher in the large colloidal fraction of the white water body. These types of investigations give a good estimation of metal transport abilities of dissolved organic matter in fresh water.

Benaim JY & Mounier S, Croatica Chemica Acta, 71, 405-419, (1998).

J10 : 3A/05 : G6

Characterization of an Aquifer System ­ Stable Isotopes and Microbiology

Ulrike Schulte (U.Schulte@ghj.de)1,

Harald Strauss

(Harald.Strauss@ruhr-uni-bochum.de)1,

Jan Detmers (jdetmers@mpi-bremen.de)2 &

Jan Küver (jkuever@mpi-bremen.de)2

1 Ruhr-Universität Bochum, Institut für Geologie, Universitätsstrasse 150, D-44801 Bochum, Germany
2 Max-Planck-Institut für Marine Mikrobiologie, Celsiusstrasse 1, D-28359 Bremen, Germany

Results from a combined isotopic and microbiological investigation clearly characterize a succession of biologically mediated redox processes in a setting of multiple aquifers. These are situated in marine sands of Tertiary age and overlying Quaternary gravel deposits. Intercalated lignite seams and a major clay horizon define the aquitards, separating the groundwater horizons.

Waters in all aquifers originate from recharge of local precipitation as evident from their stable hydrogen and oxygen isotopic composition in comparison to the local meteoric waterline. Tritium concentrations up to 30 TU have been determined for the uppermost aquifer while all lower aquifers yielded concentrations below detection limit (< 2 TU).

Depth profiles of redox-sensitive parameters document a change from aerobic to anaerobic conditions and a sequence of microbial processes which includes aerobic respiration, nitrate, iron and sulfate reduction, and methanogenesis. Bacterial reworking of organic matter is evident from dissolved inorganic carbon characterized by carbon isotope values around -17‰. Strongly positive sulfur isotope values (up to +50‰) indicate sulfate reduction under closed system conditions with respect to sulfate availability. Additional support comes from the sulfate oxygen isotopic composition. Dissolved hydrogen sulfide in concentrations up to 180 µg/l resulting from sulfate reduction displays sulfur isotope values between -28 and +10‰. Methane carbon isotope values range from -110 to -65‰ and indicate its biological origin. Correlation between decreasing methane abundance and increasing isotopic composition suggest partial oxidation.

Microbiological investigations (enrichments, MPN) reveal the full spectrum of bacteria which are relevant for all processes mentioned above. Sulfate reducing bacteria were abundant up to 1500 cells/gram dry weight. Spore-forming bacteria seem to dominate the SRB population. Low sulfate reduction rates were determined by 35S-radiotracer method. A detailed assessment indicates an increase in reduction rate in close proximity to the lignite seam(s), suggesting that brown coal and/or its degradation products represent the substrate for sulfate reduction.

J10 : 3A/06 : G6

Non-Stoichiometric Pyrite Oxidation by Thiobacillus Ferrooxidans

Jae-Young Yu (jyu@cc.kangwon.ac.kr)1,

Terry McGenity (t.j.mcgenity@reading.ac.uk)2 &

Max Coleman2

1 The Department of Geology, Kangwon National University, Chuncheon, Kangwon-Do 200-701, Korea
2 PRIS, The University of Reading, Whiteknights, Reading RG6 6AB, United Kingdom

Pyrite was oxidised by growth of Thiobacillus ferrooxidans aerobically at 32°C by shaking at 100 r.p.m. in the laboratory. The experimental solutions were sampled regularly and analysed for their chemical compositions and the number of microbes. The reactant pyrite was also regularly collected and observed by scanning electron microscopy. The analyses and observations showed a very long period of adaptation (lag-phase) before the microbes actively oxidised pyrite. The purpose of this study was to determine chemical and microbial changes during lag-phase compared with changes during the exponential growth phase. Observation of pyrite under the scanning electron microscope indicated that microbial attachment on the pyrite surface could only be detected after the lag-phase. During the exponential phase, cells of Thiobacillus ferrooxidans could be seen attached to the surface of pyrite in an irregular arrangement, and, where cells had been sloughed off, pits in the shape of the bacterial cells were observed. Lag-phase lasted for approximately 400 hours. During this period, the dissolved iron and sulphur content increased very slowly compared to a very rapid rise during the exponential phase. There was a parallel increase in the number of microbes in solution. The dissolved iron in the solution during lag-phase was mostly in the form of ferrous iron, while during the exponential phase it was mostly ferric, indicating that the microbes obtained energy mainly from the oxidation of sulphur rather than iron during this lag-phase. The molar ratio of the dissolved Fe to S in solution decreased from more than 2 to approximately 1 during the lag phase. The molar ratio continued to decrease during the exponential phase and reached approximately 0.5, which is the ratio defined by the stoichiometry of pyrite. All the dissolved sulphur was in the form of sulphate, and neither elemental sulphur nor any other secondarily precipitated form of sulphur was found in the experiment. This indicates that the dissolution of pyrite through oxidation by Thiobacillus ferrooxidans is non-stoichiometric. These observations are relevant to understanding the mechanism of initiation of pyrite oxidation.

J10 : 3A/09 : G6

Osmium Isotopes in Hydrothermal Fluids from the Juan de Fuca Ridge

Mukul Sharma1,

G. J. Wasserburg2 &

A. W. Hofmann1

1 Max-Planck-Institut für Chemie, Abteilung Geochemie, D-55020 Mainz, Germany
2 Lunatic Asylum, Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA 91125, USA

We report the first Os concentration and isotopic composition data for hydrothermal fluid samples collected from the Monolith vent on the southern Juan de Fuca Ridge. These data represent the results of an improved technique for high Os yields and tracer equilibration following the report of Levasseur et al. (1998). Samples A [2811D6 (T ~270°C, Mg = 2.36 mmol/kg,)], and B [2429-10 (T = 308°C, Mg = 2.38 mmol/kg)] were collected in 1994 and 1991, respectively using Ti-samplers deployed from the DSV Alvin (D. Butterfield, pers. comm.). For sample A 187Os/188Os = 0.130 ± 0.002 and COs = 55 ± 2 fmol/kg; for sample B 187Os/188Os = 0.388 ± 0.002 and COs = 202 ± 1 fmol/kg. The 187Os/188Os ratios of these fluids are far lower than the seawater value of 1.05 (Sharma et al., 1997; Levasseur et al., 1998) and point to a dominance of mantle-derived Os. Indeed, 187Os/188Os in sample A is close to that estimated for the upper mantle (= 0.125). We note that COs of this sample is close to that determined for seawater (= 57 fmol/kg; Levasseur et al., 1998). In contrast, sample B has a higher COs and yet it is much more radiogenic in comparison to A. Important issues are: 1) why is the Os concentration of A the same as that of seawater, and 2) what are the mechanisms that explain the high COs in B with a 187Os/188Os intermediate between mantle and seawater.

We must explain these observations given that the ultimate source of hydrothermal fluids is seawater circulating through hot rocks at the ridge. The 187Os/188Os of A implies complete overprinting of the seawater Os signature in the fluid by the mantle signature. If there is no Os loss during the downwelling of seawater, a binary mixing calculation indicates that COs of the fluid should be about 150 times that of seawater in order to swamp the seawater signature. This is not observed. If the match in concentrations between seawater and sample A is not accidental, this suggests that there may be a major ion buffering mechanism that controls the COs in both seawater and hydrothermal fluids. This could play a major role in regulating the mantle Os input into seawater. The extent to which mantle Os is dominant compared to cosmic dust remains to be resolved. Concerning sample B, we note that the sediments associated with the TAG hydrothermal plume (Ravizza et al., 1996) display high COs and intermediate 187Os/188Os values. This suggests that Os in sample B may be the result of a small amount of suspended particles in the fluid and not reflect the true dissolved Os load.

We are grateful to David Butterfield, NOAA for providing the hydrothermal fluids and the temperature and Mg concentration data. CIT-GPS Contribution No. 8525(1010).

J10 : 3A/10 : G6

U Isotope Ratio in Rainwaters: Implication for its use as a Hydrological Tracer

François Chabaux (fchabaux@illite.u-strasbg.fr),

Jean Riotte (jriotte@illite.u-strasbg.fr) &

Anne-Désirée Schmitt

CGS-EOST, 1,rue Blessig, 67084 Strasbourg Cedex, France

It is classically admitted that rainwater contains virtually no U and consequently that U isotopes in the river waters are not sensitive to atmospheric inputs but only influenced by U supplies from rock weathering. To test these asumptions we have begun one of the first TIMS study of U isotopes in rainwaters and compared the data with U isotope variations in rivers waters during rain periods. This study is focused on two different geographical areas: 1- the Upper Rhine Basin with analyses of rainwaters from Strasbourg and from the Strengbach hydro-geochemical observatory in the Vosges mountains, 2- the Himalaya with rain waters collected in the watersheds of two himalayan rivers, namely the Kali Gandaki and the Trisuli rivers. The U concentrations are measured by ICPMS with a precision of about 10%. The U isotope ratios are analysed by thermo-ionisation mass spectrometry with a precision of 1-2% for 2-3 L water samples. Procedural blanks are low and do not significantly affect the isotope determinations.

The present results confirm that U concentrations in rainwaters are very low, lower than 20 ppt in both studied areas. These results also indicate that U and Sr in rainwaters of the Upper Rhine Basin have a seawater signature with (234U/238U) activity ratios and 87Sr/86Sr isotope ratios close to 1.14 and 0.708, respectively. On the other hand, the himalayan rainwaters have (234U/238U) activity ratios between 1.03 and 1.06 and 87Sr/86Sr isotope ratios higher than 0.712, which clearly indicate a contribution of a non-seawater component for U and Sr in these rainwaters. It was observed that the U activity ratio in the Strengbach waters at the outlet of the experimental watershed decreases from 1.02 to 0.95 when the discharge of the stream increases (Riotte and Chabaux, submitted). Similarly, the U activity ratio in the Trisuli river decreases from values close to secular equilibrium during non-monsoon period, to values of 0.95 during the monsoon. These trends are interpreted in terms of mixing between waters in secular equilibrium or slightly enriched in 234U which represent waters having weathered the bed-rock, and waters with (234U/238U) < 1. The contribution of this end-member increases during rain periods, but its U activity ratio clearly indicates, that the U cannot be supplied by rainwater. It has to be mobilized from material that has already been weathered, i.e. from alterites and/or soils.

These results confirm that rainwater does not significantly affect the U budget of river waters. They also highlight the potential of the U activity ratio to identify the contribution of weathered material and bedrock to the chemical composition of river waters.

J10 : 3A/11 : G6

Sources of U in the Himalayan Rivers: Evidence from the Narayani Watershed

Jean Riotte (jriotte@illite.u-strasbg.fr)1,

François Chabaux (fchabaux@illite.u-strasbg.fr)1,

Albert Galy (agaly@crpg.cnrs-nancy.fr)2 &

Christian France-Lanord

(cfl@crpg.cnrs-nancy.fr)2

1 CGS-EOST, 1, rue Blessig, 67 000 Strasbourg, France
2 CRPG, 15, rue N.D. des pauvres, 54 501 Vandoeuvre-les-Nancy, France

The Ganga-Brahmaputra river system significantly contributes to the dissolved riverine flux to the oceans, with a major effect on the global budget of elements such as Sr and U. The Himalayan rivers have been recognised to be a significant contributor to these fluxes even though the precise origin of these latter remains to be better constrained. The aim of this work is to evaluate the potential of U isotope ratios in order to discuss the origin of the U Himalayan fluxes. This work focuses on the Narayani watershed in the Central Nepal. This area has been the subject of a detailed geochemical investigation for several years. The stream flows through the main Himalayan formations from the Tethian Sedimentary Series (TSS) to the Gangetic plain, including the High Himalayan Crystalline Series (HHCS), the Low Himalayan Series (LHS) and the Siwaliks Series (SS). Water samples from the main stream of the Narayani basin have been collected up from the springs in Tibet down to the Gangetic plain, at different periods of the year including monsoon. In order to characterise the waters coming from each formation, several streams flowing on smaller monolithological watersheds have also been sampled. The U isotope compositions were measured on a VG sector mass spectrometer with a precision of about 0.5%.

The U signature of 'monolithological' rivers is different in each Himalayan formation: the TSS rivers have high U concentrations with (234U/238U) <1. The HHC and LH rivers have low U concentrations with quite limited 234U enrichments, i.e. (234U/238U) = 1.06, whereas the Siwaliks streams and the plain have intermediate U contents, and (234U/238U) activity ratios between 1.2 and 1.33. The comparison of the Narayani water samples with the monolithological end-members in the 234U/238U -87Sr/86Sr diagram reveals that the dissolved U in the Narayani river is a mixture between the TSS and the HHC-LC end-members, with a high predominance of U from the TSS. On the other hand the comparison of these data with the U and Sr results obtained from the waters in the Ganga river could suggest that the Siwaliks series as well as the Gangetic plain formations do not significantly contribute to the dissolved flux of U in the Ganga-Brahmaputra river system.

This study highlights the interest of the (234U/238U) activity ratios for tracing the origin of the elements during weathering and more specifically reveals the utility of the U-Sr combined studies.

J10 : 3A/12 : G6

Sources of Boron in Rain Waters of Nepal and Bangladesh

Estelle Rose (erose@crpg.cnrs-nancy.fr),

Jean Carignan (carignan@crpg.cnrs-nancy.fr) &

Albert Galy (agaly@crpg.cnrs-nancy.fr)

CRPG-CNRS, 15 N.D. des Pauvres, 54501 Vandoeuvre-les-Nancy, France

Boron is present in the atmosphere either as gas phase, for about 90%, or as aerosol particles, for the leaving 10%. Sources are sea water, volcanoes, mineral continental aerosols and anthropic emissions. Major unknowns concerning geochemistry of atmospheric B are: 1) the budget of particles and gaseous B in fonction of geography and 2) the contribution and the extend over the continents of marine B. In a previous communication to EUG (1997) we reported boron contents of lichens collected along a ca. 500 km north-south transect in the boreal forest of Québec, from Hudson Bay towards the continent. The fact that the B/I and B/Cl ratios measured in lichens remain constant, independently of their concentrations, indicates that these elements originate from a homogeneous source (gas + particles), and that they have similar atmospheric residence time. The results suggest that marine B can penetrate more than 500 km inland.

Over the last three years, rain waters were collected in Bangladesh and Nepal (Kathmandu Valley and higher Himalaya Mountains). These yielded boron concentrations varying from 0.01 and 4 micromole/L (filtered to 0.45 micron), which is similar or lower than available data reported in the literature for rain sampled in coastal areas of the world. Boron concentrations are positivelly correlated with Ca, Sr and sulphate contents and moderatly correlated with Cl contents. The correlation with sulphates may indicate that part of the B in rain may originate from anthropogenic activities, although no correlation is observed between B and metals such as Cd, Pb and V. In opposition to what was found for lichens in northern Québec, the Cl/B (and Na/B) of rain waters in Nepal and Bangladesh vary systematically with their B concentrations. This is not a systematic difference between lichen and rain data because similar trends are observed between rain and lichens from Himalaya Mountains. The distribution of the data in diagrams such as Cl/B or Na/B vs B content may be interpreted in terms of mixing between sea salts and carbonate aerosols. The highest Cl/B and Na/B ratios measured confirm that these three elements in rain events originate from unfractionated sea salts. The other end-member is characterised by low Na/B and Cl/B ratios and high Ca content. Considering a congruent dissolution of B in carbonate aerosols, less than 1-3% of carbonate-B is needed to explain the B content measured in the rain samples having the lowest Na/B ratios. This contribution is a maximum estimate considering a possible incorporation of gaseous B in the rain along with a contribution from silicate aerosols. As the Na/B do not vary systematically with the distance from the coast, it is suggested that the carbonate aerosols have a local origin.

J10 : 3A/13 : G6

Boron Continental Input to the Ocean: Global Flux and Isotopic Assessment

Damien Lemarchand (lemarcha@ipgp.jussieu.fr),

Jerome Gaillardet &

Claude Jean Allegre

4, Place Jussieu, Laboratoire de Geochimie et Cosmochimie - Tour 14/24 3eme Etage, 75252 Paris Cedex 05, France

The geochemical cycle of boron and its secular evolution in the ocean is still poorly constrained. Although no consensus has been reached, the river boron flux seems to be the major input to the ocean (You et al., 1993, Rose et al., 1998). Due to analytical difficulties, the systematic of boron concentration and isotopic composition in the world largest rivers has not been addressed so far.

We have improved the sensitivity for boron isotope measurement using the classical PTIMS technique and developed a new chemical procedure to separate boron from organic rich media. With an external reproducibility for boron isotopic composition close to 0.15 per mil, we determined both concentrations and boron isotopic compositions in several large rivers. Our results remain preliminary, but boron concentrations range from 3 ppb for the Congo river to 201 ppb for the Huanghe river. We estimate the global boron discharge to the ocean at about 3.1011 g/an, which is 10 times higher than the flux given by You et al (1993) and consistent with the estimate of Rose et al. (1998).

Boron isotopic compositions of studied rivers range from 3.9 per mil to 17.5 per mil. This allows to calculate a preliminary discharge weighted isotopic composition for the river input at about 9 per mil. The data obtained at this time do not show any obvious correlation between boron concentration and isotopic composition at a world scale.

These results will be discussed both in term of oceanic isotopic budget and continental weathering processes.

You CF, Spivack AJ, Smith JH & Gieskes JM, Geology, 21, 207-210, (1993).

Rose EF, Chaussidon M & France-Lanord C, Goldschmidt. Conf. Abs, (1998)

J10 : 3A/14 : G6

7Be as a Tracer for a Rapid Boundary Transport in Lake Geneva

Janusz Dominik (dominik@sc2a.unige.ch)1,

Jean-Luc Loizeau2 &

M. Dulinski3

1 Centre d'Etudes en Sciences Naturelles de l'Environnement, Universite de Geneve, 10 Route de Suisse, CH-1290 Versoix, Switzerland
2 Institut F.-A. Forel, 10 Route de Suisse, CH-1290 Versoix, Switzerland
2 University of Mining and Metallurgy, al. Mickiewicza 30, Krakow, Poland

Lake Geneva is a mesotrophic lake which periodically suffers from oxygen deficit in the near bottom water. This results from prolonged periods of incomplete winter overturn and a relatively high biomass production in surface waters. A rough oxygen budget suggests a supply of oxygen to the deep water by mechanisms other that complete mixing or vertical eddy diffusion. Beryllium-7 is a cosmogenic, short-lived radionuclide (T1/2= 53 days) which can be used for tracing dynamics of particle transport in lakes (particle settling velocity, colloid coagulation rate, surface sediment resuspension and erosion). Here, we present data suggesting that 7Be sedimentary flux are useful in tracing a rapid transport from surface to bottom water along the water/sediment interface.Monthly measurements of fluxes of settling particles and radionuclides have been carried out using sediment traps deployed at 5 depths at a station located in the central part of Lake Geneva. The 210Pb flux follows variations of sediment flux with water depth, however, the comparison with the atmospheric input suggests that a part of the 210Pb-bearing particles is delivered to the near-bottom zone by sediment focusing. In principle, bottom sediment resuspension can be traced using of a pair of radionuclides with very different half lives (for example 7Be -210Pb pair). Short lived 7Be is impoverished in the top layer of bottom sediments as compared to the settling material. Thus a contribution of resuspended material to total sediments collected in traps results in a lower 7Be/210Pb ratio. This allows the discrimination between resuspension of old sediments and a focusing of material delivered from the lake surface waters. Although the mechanisms of such focusing is not precisely known, several evidences point out to the occasional renewal of bottom waters (oxygen anomaly, mean deep water age). Among probable processes delivering 7Be excess to the bottom water are boundary currents displacing rapidly cooled water masses in winter from coastal areas or from mixed shallower basin (Petit Lac), and density currents originating from the Rhone River.

Session J10:3P

J10 : 3P/01 : PO

Stable Isotope Investigation of a BTEX-Contaminated Shallow Aquifer

Paul Eckert (peckert@swd-ag.de)1,

Frank Wisotzky

(wisotzky@geol3.ruhr-uni-bochum.de)2,

Oliver Kracht

(Oliver.Kracht@ruhr-uni-bochum.de)2 &

Harald Strauss

(Harald.Strauss@ruhr-uni-bochum.de)2

1 Stadtwerke Düsseldorf, Luisenstrasse 105, D-40215 Düsseldorf, Germany
2 Ruhr-Universität Bochum, Institut für Geologie, Universitätsstrasse 150, D-44801 Bochum, Germany

Substantial contamination of soil and shallow groundwater with up to 100 mg/l BTEX as a consequence of a former gas-works was investigated over a period of three years from detection and characterization through the current remediation process. The aquifer consists of sand and gravel with the water table at 6 m below surface. Hydrochemical data indicate the natural biodegradation of organic contaminants by oxygen, nitrate, manganese(IV)oxide, ferric hydroxide and sulfate. Thereby, sulfate reduction is the most important degradation process.

Depth profiles of BTEX (predominantly benzene), dissolved sulfate and sulfide concentrations define a distinct depth interval which is characterized by active sulfate reduction. The decrease in sulfate is paralleled by a depletion of benzene and an increase of dissolved sulfide. In the presence of Fe(II), the latter is fixed in the sediment as iron sulfide.

Bacterial sulfate reduction is associated with a distinct change in the isotopic composition of sulfur and oxygen in the residual, unreacted sulfate towards more positive values. Residual sulfate concentrations around 50 mg/l are characterized by sulfur and oxygen isotope values as high as +41‰ and +15‰, respectively. This indicates an advanced level of sulfate reduction under closed system conditions with respect to sulfate availability. In contrast, aquifer sulfate which is unaffected by bacterial reduction displays an average concentration of 175 mg/l and isotope values around +6‰ (sulfur) and +9‰ (oxygen), respectively.

Remediation activities include removal of the contamination source, thermal treatment of contaminated sediment material and in-situ bioremediation through infiltration of nitrate-containing water. These processes have also been monitored by hydrochemical and isotopic analyses. Results indicate the termination of sulfate reduction likely due to more aerobic conditions and, besides BTEX degradation, a mobilization of previously precipitated iron sulfides. This is expressed, e.g. by an increase in dissolved sulfate concentration and a depletion in its sulfur isotopic composition.

J10 : 3P/02 : PO

Amazing Spatial Variation of Stable Carbon Isotopes in DIC of Pore Waters from Two Freshwater Intertidal Areas Along the Schelde Estuary, Belgium

Koen Van den Driessche (kvddries@vub.ac.be)1,

Liesbet Hellings (scheldac@vub.ac.be)2,

Frank Dehairs (fdehairs@vub.ac.be)1 &

Eddie Keppens (ekeppens@vub.ac.be)1

1 Vrije Universiteit Brussel, WE-GISO, Pleinlaan 2 1050 Brussel, BELGIUM
2 Vrije Universiteit Brussel, WE-ANCH

Diagentic reactions affecting the C-cycle were studied in freshwater intertidal systems using stable carbon isotopes combined with stable oxygen isotopes as a tracer for water fluxes.The Schelde estuary is one of the few European estuaries with extensive freshwater tidal mudflats and tidal marshes mainly covered with willow and reed.The concentration of dissolved inorganic carbon (DIC) and the 13CDIC values in pore waters of barren mudflats were significantly different from plant covered marshes. In the mudflats DIC and 13CDIC were high (40 - 80 mM and +12‰) compared to river water (5 to 6 mM and -12‰). Marshes had lower DIC (5 mM) and much lower 13CDIC (-20‰). Moreover, in the strong reducing mudflats DIC becomes isotopically heavier with depth. This suggests that in the sediments fermentation occurs producing isotopically light methane and 13C-enriched CO2. This induces a positive correlation between 13CDIC and DIC. On the contrary, in the tidal marshes, 13CDIC is controlled mainly by heterothrophic respiration on terrestrial organic carbon (-27‰) resulting in isotopically lighter DIC. Here soil structure and an extensive root system enables oxygen diffusion preventing strong anoxicity.Water dynamics, which exert some control on 13CDIC-variability, were characterised by following the propagation of 18O-variations of precipitation through river and intertidal groundwater. Rainwater-O displays singular peaks superimposed on seasonal variation (mean -6.8‰ versus SMOW in a range of 7‰). This variability is mimicked in a attenuated mode and with a delay of weeks by river water (mean - 6.2‰ versus SMOW in a range of 3.5‰). This variability is futher propagated, in an even more attenuated mode and with another delay of weeks through the intertidal sediments (with a range of 1.5‰). The pore water of barren mudflats, that are frequently flooded and have an almost impermeable saturated structureless top layer, is significantly 18O-enriched (mean -5.6‰ versus SMOW) compared to river water (-6.2‰ versus SMOW) by evaporative fractionation. This is stronger for the much longer exposed higher parts of the mudflats and is also stronger in summer. Both 13CDIC and O in barren mudflats increase with longer exposure time, but 13CDIC increases with depth while O decreases, so that fermentation processes rather than evaporative fractionation seem to control 13CDIC. On the contrary, the marshes (mean -6.1‰), that have a much higher infiltration capacity due to soil structure, do not show 18O-enrichment despite very strong evapotranspiration. We conclude that water has a rather high but variabel residence time, controlled by variable permeability. Low permeability explains the "closed system" 13CDIC-behaviour.The observed extreme 13CDIC spatial variation (30‰ in ten meters) controlled by soil structure, plant cover and hydrological conditions opens perspectives to the paleo-environmental interpretation of 13C-variations in intertidal sediments. (This research was financed by the Flemish Community: OMES-project.)

J10 : 3P/03 : PO

Nucleation and Growth of Calcium Carbonate in Monolayers at the Air/Water Interface Using Brewster Angle Microscopy

Susanne Hacke (shacke@gwdg.de)1 &

Dietmar Moebius (dmoebiu@gwdg.de)

Max-Planck-Institute f. Biophys. Chemistry, Am Fassberg 11, 37077 Goettingen, Germany

From natural systems it is well known that natural biofilms could conduct CaCO3 precipitation processes. The subject of our special interest is the site were nucleation starts and the time dependence of the crystallization processes.In a first step to achieve a better understanding of these processes we have investigated the crystallization of CaCO3 beneath monolayer films at the air/water interface. Pure and mixed monolayers of stearic acid (C18) and stearic acid methyl ester (SME) were spread on subphases of pure water and supersaturated CaCO3 solutions at 20°C. The premise of the experiment is to induce the nucleation and growth of CaCO3 through favorable electrostatic interactions with monolayers at the air/water interface. By controlling the monolayer composition and subphase conditions the three modifications of CaCO3 (calcite, aragonite and vaterite) can be selected. One important aspect of the investigation is to imitate the different crystal faces by modifying the composition and packing density of the monolayers. Varying ratios of neutral (SME) to charged (C18) headgroups in the monolayers were choosed to create motifs of controlled charge density. Crystal nucleation and growth were characterized using Brewster angle microscopy (BAM), surface pressure/area (p-A) and surface potential/area (V-A) isotherms. During the BAM experiments, micron-sized crystals were observed immediately after spreading the C18 monolayer on the CaCO3 subphase. The C18 p-A isotherms on pure water and CaCO3 subphases were identical, suggesting the crystals were beneath the film, interacting only with the monolayer head-groups. The C18 V-A isotherm on pure water subphase increased with decreasing area as expected. However, the C18 surface potential on CaCO3 subphases was invariant with area. The constant surface potential on CaCO3 subphase might result from a strong interaction between the charged monolayer head-groups and calcium ions in the subphase.In future work more experiments for a specific control of the calcification processes will be done. Different structures of new formed crystals will be determined by HR-TEM measurements.

J10 : 3P/04 : PO

Crystallyzation of Apatite in Modified Seawater Solutions

Sergey Golubev (o.chaplygin@mailcity.com)

117463, Moscow, Golubinskaya street, 25-2-298, Russia

The precipitation of authigenic calcium phosphate (apatite) is generally believed to be major sink of oceanic phosphorus (P) in marine sediments. This study addresses the solubility of Ca and Mg phosphates precipitated from modified seawater solutions and the transformation of metastable solids in apatite.

Methods: Supersaturated solutions were made by a mixing of equal volumes of artificial 70‰ phosphate-free seawater with a Na2HPO4 solution (0.5-3.0 mM) at different pHs. To test the effect of magnesium on the solubility, the [Mg2+]/[Ca2+] ratio in artificial seawater varied from 0 to 10 at a fixed ionic strength of 0.7 M. The first portion of the precipitated solid was immediately separated from solution by filtration through a membrane filter, rinsed in H2O, dried at 50°C, and analysed for Ca, Mg, and P content. Solution was analysed for P and F content immediately after precipitation and in the process of aging of precipitated phases (2, 4, 6 months after precipitation).

Results and discussion: 1 day after precipitation the [Mg2+]/[Ca2+] ratio in artificial seawater did not influence the solubility of phosphates (13.3 mg P per liter), but 7 days after precipitation in Mg-free seawater the solubility decreased to 1 mg P per liter when it did not change for solids precipitated in seawater with the [Mg2+]/[Ca2+] ratio from 2 to 10. In the process of aging of precipitated phases the concentration of fluoride (0 - 2*10-4 M) and carbonate (0 - 3.6*10-3 M) did not influence the solubility of phosphates. The solubility of solids in normal seawater is almost 100 times higher than it is Mg-free seawater. XRD analysis of the precipitated phases shows the presence of amorphous phases in solutions having a [Mg2+]/[Ca2+] ratio from 1 to 10, same results have been obtained 2, 4, 6 months after precipitation. Mg-free solution produces a highly crystalline precipitate of hydroxylapatite. Only after heating normal seawater and precipitate at about 90°C for about 1 month XRD analysis shows the appearance of crystalline apatite and aragonite. Analysis of phosphorus in the solids revealed a F/P molar ratio close to 0.10, a (Ca+Mg)/P molar ratio close to 1.5-1.6 in Mg-free seawater and 1.3 in normal seawater. No precipitation of a solid phase was observed in Ca-free "seawater" solutions within the time scale needed for Ca-Mg phosphates precipitation in normal seawater. Based on these data, we suggest that the solid phases precipitated homogeneously from solutions are presented by metastable mixed (interlayed) amorphous phosphates of Ca and Mg such as octa calcium-magnesium phosphates with a stoichiometry of (Ca,Mg)4H(PO4)3*xH2O. The transformation of precipitated phases to crystalline apatite is inhibited by Mg ions.

J10 : 3P/05 : PO

Trace Elements Speciation in River Sediments

Lydia Leleyter (leleyter@illite.u-strasbg.fr)1,

Jean-Luc Probst (jlprobst@illite.u-strasbg.fr)1,

Pedro Depetris (pdepetris@powernet.com.ar)2,

Souad Haida (souad_hs_haida@yahoo.com)3,

Jefferson Mortatti (jmortatt@pintado.ciagri.usp.br)4,

Robert Rouault (rouault@illite.u-strasbg.fr)1 &

Jean Samuel (jsamuel@illite.u-strasbg.fr)1

1 CGS/EOST, 1 rue Blessig, 67000 Strasbourg, France
2 Université Nationale de Cordoba, Département de Chimie Analytique Minérale, Avenida Velez Sarsfield 299, Cordoba 5000, Argentina.
3 Département de Géologie, Université Ibn Tofail, Faculté des Sciences, B.P.133, Kénitra, Morocco
4 Centro de Energia Nuclear na Agricultura (CENA), Universidade de Sao Paulo, Campus de Piracicaba, Avenida Centenario 303, CP 96, 13 400, Piracicaba, SP, Brazil.

In aquatic systems, river sediments are important sinks for a wide variety of chemicals. Particulate trace elements, such as alkalines, metals, alkaline earth or actinides are not necessarily permanently sorbed onto solid phases and they can be released in aqueous phase with changes in physico-chemical conditions such as salinity, pH, redox potential, and concentration of chelators. As the trace element speciation can only be obtained by chemical fractionation, we have developed a new sequential extraction procedure (Leleyter and Probst, in press) which is designed to dissolve selectively and efficiently the different chemical phases of river sediments, likely to be affected by changes in physico-chemical conditions, in the following order: elements dissolved with water, really exchangeable elements, bound to carbonates, manganese oxides, amorphous iron oxides, crystalline iron oxides and organic matter. This new sequential extraction procedure has been applied to 37 bottom and suspended river sediments which were collected from 4 different countries: Argentina, Brazil, France and Morocco.

Predictably, the proportion of each element which could be leached, varies according to the different sediment samples. Nevertheless sediments from a same river or geographic area behave as a coherent group, there is no important spatio-temporal variation. Moreover, whatever the origin of the sample, the elements which belong to the same chemical family present similar leachate patterns for the different rivers / areas.

For example, alkalines (Rb, Cs) are essentially linked to the residual fraction. On the contrary, metals (Fe, Mn, Co and Pb) are very leachable and mainly controlled by oxides but they are sometimes present in the carbonate or organic fractions too. The knowledge of sorption sites for actinides is fundamental in order to predict accurately the retention or migration of radioelements in the geosphere. The actinides (U and Th) present here often two kinds of profiles: Th is mainly associated to the oxides whereas U is linked at the same time to the oxides and to the carbonates. The alkaline earth (Sr and Ca) are mainly associated with exchangeable fractions and with carbonates. Because of their scavenged fractions, alkaline earth are really sensitive to pH and salinity variations. Moreover the leachable strontium has the same radiogenic composition than the dissolved phases. This result implies that the study of the 87Sr/86Sr isotopic signatures of river sediments is strongly linked to the physico-chemical characteristics of the aquatic environment. Indeed river water acidification (natural or anthropogenic) will provokes an increase of the 87Sr/86Sr isotopic ratio of the total river sediment.

Leleyter L & Probst JL, Int. J. Environ. Anal. Chem, in press

J10 : 3P/06 : PO

Phosphate Coprecipitation with CaCO3 from Fresh Waters

Alla Savenko (o.chaplygin@mailcity.com)

Moscow,119899,Vorobyovy Gory,, Moscow University,Geogr.faculty,, department of hydrology, Russia

Phosphorus is an important element limiting primary production of many aquatic systems and controlling their eutrophication. Therefore, phosphorus behaviour has been studied extensively in surface waters. One of the important and poorly investigated processes of dissolved inorganic phosphorus removal is phosphate coprecipitation with CaCO3. This process is widespread in lakes and reservoirs during the vegetation period. The increase of pH by photosynthesis gives rise to favoriable conditions for CaCO3 formation. Field observations demonstrate that the value of pH in the surface layer reaches 9.5 - 10.0. Whereas the pH value in solutions surrounding phytoplankton cells may be even higher. The phosphate coprecipitation with CaCO3 under these conditions was the aim of this study. The experiments were carried out with natural filtrated water from Mozhaisk reservoir (Moscow district) and phosphate concentration from 0.1 to 0.3 mg P / l. Initial concentrations of dissolved carbonates and calcium were equal to 24.4 mg C / l and 30.7 mg Ca / l, respectively. The CaCO3 supersaturation was achieved by addition of a small aliquotes of 1 M NaOH in the flasks with 250 ml of solution. The experiments demonstrated that the value of P/Ca weight ratios in precipitated CaCO3 at pH range from 9.9 to 10.7 is a function of initial phosphate concentration and pH:(P/Ca) =(0.24-0.0174*pH)*[P] ,where [P] is the initial concentration of dissolved phosphates, mg P/l. An increase of pH from 10.7 to 11.2 at the high concentrations of dissolved phosphate (>0.2 mg P/l) causes the strong increase of phosphate coprecipitation that may be explained by the hydroxyapatite formation. The phosphate concentration in CaCO3 increases from 1000 - 2000 ppm at the initial concentration of dissolved phosphate of 0.1 mg P/l to 6000 - 8000 ppm at the dissolved phosphate concentration of 0.3 mg P/l.

J10 : 3P/07 : PO

Metals in Stream Mouth Zones of the Areas of Present Volcanic Activity

Sergei Fazlulin (plof2@hotmail.com)1 &

Elizaveta Bortnikova (kolon@uiggm.nsc.ru)2

1 Institute of Volcanology Russia Academy of Sciences, 9 Piip boulevardi, 638024 Petropavlovsk-Kamchtski, Russia
2 United Institute Geology, Geophysics and Mineralogy, pr. Koptuga 3, 630090 Novosibirsk, Russia

According to the results of carried out works devoted to the study of metals migration with acid water flows in area of Kuril island, following basic scientific results have been got.

There is the fact of deep transformation for waters of surface water flows: its composition becomes of sulfate type, pH decreases, metals ability to migrate, as well as its concentration increases.

During the interaction between acid water flows and sea water, the contrast alkali geochemical barrier is formed. Parameters depend on several characteristics (aluminum and iron concentration, pH, hydrometeorological factors, sea water alkalinity). Those factors also control the velocity of carbon dioxide accent from the sea water and change ionic equilibration in a process of aluminum and iron hydrolysis. Iron and aluminum practically fully turn to suspension that becomes a substrata for micro elements sorption.

Natural observations and experimental studies showed, that only chrome fully fall out together with Fe hydroxide and Al hydroxide at pH > 4.5. During the increase of pH, the transition to the solid phase for Zn, Cu, Ni and Pb increases too. But when pH is near 8, about 40% of these elements are still present in solution. Ag and Hg practically do not transit to the suspended phase. Mn and Co falls out insignificantly (up to 5%). Maximum falling out for Sb (75%) is noted in diapason of pH 4-5, otherwise in alkali and acidic conditions the part of suspended Sb can be decreased down to 10%.

Computer modeling of waters mixture process displayed that this process can lead to the concentrated falling out of Si, Al, Fe and also of heavy metals even in isothermal conditions. In particular when the proportion of mixture is 4/6, all Fe is in suspended phase and when the proportion is 3/7, Al concentration becomes near the same as in sea water. First to fall out from the solution is Fe, for some minerals - Hematite, Hyotite, Fe(OH)2Cl3 - the oversaturation starts in acid river water; for the majority of other Fe-containing minerals, oversaturatin of solution starts at pH from 2.2 to 3.5. Relatively to the main quantity of minerals Al, oversaturation of solutions starts at pH > 3.5. It must be noted that Allophane (several types) and Yurbanite are stable at pH < 2, but when the acidity decreases, solutions becomes unsaturated relatively to these minerals. Results of modeling are satisfactorily correlated with the factual data and results of laboratory experiments but, there observed a divergence in the behavior of trace elements.

J10 : 3P/08 : PO

Properties of Humic Acids Extracted from Bottom Sediments of Lakes of Different Catchment Management

Slawomir S. Gonet (gonet@rol.atr.bydgoszcz.pl) &

Joanna Cieslewicz

University of Technology and Agriculture, Department of Environmental Chemistry, Bernardynska 6, PL-85-029 Bydgoszcz, Poland

Sediment humic substances play a fundamental role in accumulation and exchange processes of chemical compounds (metals and organic pollutants) and energy at the interface water-sediment. They originate mainly from aquatic organisms and from organic matter that is flowing into the lake from surrounding environments (Ishiwatari, 1985). The aim of this work was to determine the effect of lake catchment management on properties of sediment humic acids (HA). Bottom sediments were sampled from littoral zone of lakes with forested catchments and with catchments under agricultural use. For extracted humic acids elemental composition, IR-spectra and differential thermal analysis (DTA) were made. The analysis of elemental composition showed that HAs of forestry lake sediments were characterized by higher content of carbon and lower content of hydrogen and nitrogen than HAs of infield lake sediments, and also the higher value of H:C atomic ratio. Usually it accepts that the value of this atomic ratio is inversly proportionel to the content of aromatic structures in HAs molecules. It was confirmed by IR-spectra (intensity of absorption bands in the region of 1640-1620 and 1520 cm) and the results of differential thermal analysis. On thermograms, one endothermic and two exothermic effects were noted on the DTA curves. The area under DTA curve is proportional to the heat emitted during the exothermic reaction (Gonet, 1989). For HAs of infield lake sediments, the calorific value of organic material decomposed during second exothermic reaction was higher than value of first reaction. The inversly dependence was found for HAs of forestry lakes. The obtained results show that HAs of infield lake sediments were characterized by lower values of internal oxidation degree than humic acids of forestry lake sediments. It is connected rather with lower content of carbon and higher content of nitrogen, than with content of oxygen and oxygen containing functional groups. This interpretations is certified by IR-spectra, in which the content of -COOH groups (1720 cm) and -OH groups were higher for HAs of infield lake sediments. The absece of higher absorption in the region of 1100-1000 cm in the IR-spectra of HAs from forestry lake sediments is very characterictic. Absorption in this region is due to polysaccharides. The existence of strong absorption band in this region was observed for IR-spectra of humic acids extracted from manured soils (Gonet, 1989).

Ishiwatari R, Geochemistry of Humic Substances in Lake Sediments. In: Humic Substances in Soil, Sediment and Water Geochemistry, Isolation and Characterization. GR Aiken, DM McKnight, RL Wershaw and P MacCarthy (Eds). A Wiley-Intersc. Publ. New York, (1985).

Gonet SS, Roczn. Glebozn, 40, 27-38, (1989).

J10 : 3P/09 : PO

Stable Isotopes Study of Lakes in High Tatra Mountains, Slovakia

Katarína Marhevková (marhevko@fns.uniba.sk)1,

Dusan Bodis (bodis@gssr.sk)2,

Juraj Michalko (michalko@gssr.sk)2,

Peter Malík (malik@gssr.sk)2 &

Svetozár Sherer (scherer@gssr.sk)1

1 Department of geochemistry, Comenius University, Mlynská dolina G, 842 15 Bratislava, Slovak Rep.
2 Geological Survey of Slovak Republic, Mlynská dolina 1, 817 04 Bratislava, Slovak Rep

Data of oxygen, hydrogen and sulphate sulphur stable isotopes from the lakes of the Vysoké Tatry Mts., acquired in the framework of International Atomic Energy Agency No. 8675/RB project "Stable isotopes in lakes of High Tatra mountains, Western Carpathians, Slovakia" represent the first set of this kind information on the territory of Slovakia. Moreover a complete set of chemical composition of waters was taken and compared to preceding results and contemporary state of acidification was evaluated. Values of studied chemical compounds respond to their source - initial precipitation waters and bedrock character with influence of biochemical processes. Acidification is due to inactive geological background (granites), high contribution of SOx and NOx from atmospheric deposition and low buffering ability of soil. Acidification of lakes is lowered from 1980 and is approaching to level before acidification, probably as consequence of industry crisis in Eastern Europe after 1989. Water samples from the Furkotská dolina valley water syste follow the MWL with natural trend of increasing content of heavy isotopes with lowering altitude. In depth profiles of single lakes isotope composition does not change - due to natural conditions during sampling campaigns (spring and fall homothermy) - all waters are meteoric in origin. Water of the Strbské pleso lake is enriched to heavy isotopes, all samples fit an evaporation line. This could be explained by longer residence time of water or by recharging is from last phases of snowmelt. Sulphate sulphur isotope ratios from lakes are the same and they are identical with these of snow packs.

J10 : 3P/10 : PO

Distribution of 129I in Central Arctic Ocean

N. Buraglio (nadia.buraglio@material.uu.se)1,

A. Aldahan (ala.aldahan@geo.uu.se)2 &

G. Possnert (possnert@material.uu.se)1

1 Tandem Laboratory, Uppsala University, S-751 21 Uppsala, Sweden
2 Institute of Earth Sciences, Uppsala University, S-752 36, Uppsala, Sweden

The distribution of 129I was measured in 15 water samples from three depth profiles, namely Lomonosov Ridge (lat. 86-25.0, long. 144-29.6), Makarov Basin (lat. 86-10.3, long.155-00.3) and Eurasian Basin (lat. 82-29.9, long. 134-36.3), sampled on August 13-19, 1996. After chemical extraction, the 129I/127I was measured by accelerator mass spectrometry (AMS) at the Uppsala EN-tandem accelerator, with a machine and statistical error at < 10% and background of < 10-13. The results (Fig. 1) indicate > 7x108 atoms/liter in the cold, low salinity, 100-m-deep surface layer and 0.06x108 atoms/liter at depth below 3 km. The elevated concentration of 129I in the surface layer agrees with results form other parts of the Arctic Ocean (e.g. Kilius et al., 1995). The source of iodine is mainly related to discharges from nuclear reprocessing facilities at Sellafield and La Hague. The 129I concentration at depth > 3 km may reflect contribution from the anthropogenic sources. The total inventory of 129I in the central Arctic is estimated at 2.7x1027 atoms (representing about 34% of the total discharge from Sellafield and La Hague until 1996). Advected warm Atlantic water forms the dominant transport mechanism of iodine from the sources to the central Arctic. Fresh water (ice melting, and rivers) provided ˜ 3% of the total inventory. An estimate, based on our inventory and on marine discharges from La Hague and Sellafield (Yiou et al., 1995), suggests = 11 years marine transport time for the 129I from the sources to the central Arctic.

The Measured Distribution of 129I and Major Ocean Layers in Central Arctic. Polar Mixed Layer (PML). Atlantic Warm Water Layer (AWL). Transition Layer (TL) and the Deep Ocean Layer (DOL).

Kilius LR, et al, Environmental Radioactivity in the Arctic (Strand P Cooke A eds), 117-120, (1995).

Yiou F, et al, Environmental Radioactivity in the Arctic (Strand P Cooke A eds), 113-116, (1995).



EUG 10
28th March - 1st April, 1999
Strasbourg, France

Index of EUG 10 Volume
Further EUG 10 Information
Index of the Journal of Conference Abstracts
Cambridge Publications Home Page

Last Updated on Wednesday, March 17, 1999.
© 1997 Cambridge Publications