Applied Geochemistry (v.24, #5)

The Pitzer’s interaction parameters, λ N–M, involving the Mth cationic Al species Al3+ or AlOH2+ or AlO+ and the Nth neutral species SiO2(aq) (at temperatures of 25–300 °C) or CO2(aq) (at temperatures of 25–150 °C), have been evaluated through empirical linear relationships between λ N–M and the surface electrostatic field of the ionic species of interest. These relationships have been obtained starting from the known λ N–M for both SiO2(aq) and CO2(aq) with the main dissolved cations. The Pitzer’s interaction parameter thus estimated for the pair CO2(aq)–Al3+ at 25 °C, 0.327, is 20–40% higher than the corresponding values obtained from CO2 solubilities in concentrated solutions of AlCl3, 0.272 ± 0.010 (2σ), and Al2(SO4)3, 0.232 ± 0.002 (2σ), partly corroborating the empirical approach adopted in this study. To test the Pitzer’s interaction coefficients for cationic Al species with aqueous SiO2, the log  K values of the kaolinite dissolution reaction have been computed starting from available experimental data at 23–25 °C and ionic strengths of 0.0001–0.12 mol/kg adopting, alternatively, the Pitzer’s equations and the Debye–Hückel equation. A satisfactory agreement has been found between the log  K values obtained through these two approaches, with maximum deviations of 0.11–0.12 log units. This good convergence of results is encouraging as it represents a necessary condition to prove the reliability of the Pitzer’s interaction coefficients estimated in this work. These results are a first step to take into account specific interactions among solutes in concentrated electrolyte solutions, such as those hosted in sedimentary basins or geothermal waters, for instance through the Pitzer’s equations. However, experimental or field data at higher ionic strengths are absolutely necessary to validate the reliability of the Pitzer’s interaction coefficients determined in this study.

Mineral trapping is one of the safest ways to store CO2 underground as C will be immobilized in a solid phase. Carbon dioxide will be, therefore, sequestered for geological periods of time, helping to diminish greenhouse gas emissions and mitigate global warming. Although mineral trapping is considered a fairly long process, owing to the existence of kinetic barriers for mineral precipitation, it has been demonstrated both experimentally and by numerical modeling. Here the results of experimental and numerical modeling studies performed in sandstones of the saline aquifer of the Rio Bonito Formation, Paraná Basin, are presented. The Rio Bonito Formation consists of paralic sandstones deposited in the intracratonic Paraná Basin, southern Brazil, during the Permian (Artinskian–Kungurian). These rocks have the largest potential for CO2 storage because of their appropriated reservoir quality, depth and proximity to the most important stationary CO2 sources in Brazil. Here it is suggested that CO2 can be permanently stored as carbonates as CO2 reacts with rocks of the Rio Bonito Formation and forms CaCO3 at temperatures and pressures similar to those encountered for CO2 storage in geological formations. Results of this work will be useful for studies of partitioning mechanisms for C trapping in CO2 storage programs.

The objective of this paper is to build a general correction model that takes into account all the different radiocarbon-dilution reactions and resolving the processes that are geochemically “aging” the groundwater in the Neogene Aquifer. For this, δ13C and radiocarbon in groundwater are investigated with their relationship to other chemical components in groundwater. The δ13C values in the Neogene Aquifer are influenced by various geochemical reactions like calcite dissolution, oxidation of organic matter and methanogenesis. Calcite dissolution and CH4 production increase δ13C while the oxidation of organic matter decreases δ13C in the groundwater. The reactions that modify δ13C also influence the 14C activity. Due to the complex geochemical environment, existing correction models are not applicable to this situation. A correction model for initial 14C activity is formulated in which the different C sources that influence 14C activity are taken into account. It is observed that recent dissolved organic matter plays an important role in redox reactions. The corrected 14C ages lie between −0.792 and 6.425 ka representing the maximum age. If a part of the organic matter that oxidises is fossil, the determined age will represent an overestimated age.

Molecular analysis of petroleum derived compounds that adsorb onto gas hydrate surfaces by Anna E. Borgund; Sylvi Høiland; Tanja Barth; Per Fotland; Kjell M. Askvik (777-786).
Field observations have shown that some streams of water, gas and crude oil do not form gas hydrate plugs during petroleum production even when operating within thermodynamic conditions for hydrate formation. Also, when studied under controlled laboratory conditions, some oils are found to form hydrate dispersed systems whereas others form plugs. Oils with low tendency to form hydrate plugs are believed to contain natural hydrate plug inhibiting components (NICs) that adsorb onto the hydrate surface, making them less water-wet and preventing the particles from agglomerating into large hydrate clusters. The molecular structure of the NICs is currently unknown. In this work, hydrate adsorbing components were extracted from crude oils using freon hydrates as an extraction phase. The fractions were found to be enriched in polar material, and more polar material is associated with hydrates generated in biodegraded crude oils than in non-biodegraded oils. Various fractionation schemes and analytical techniques have been applied in the search for molecular characterisation. The average molecular weights were found to be approximately 500 g/mole. GC–MS chromatograms show a large UCM (Unresolved Complex Mixture). Thus, GC–MS has a limited potential for identification of compounds. A commercial biosurfactant was used as a model compound in the search for similar structures in the extracts. The results from analysis of the hydrate adsorbing components suggest that the type and structure are more important for hydrate morphology than the amount of material adsorbed.

Hydrological and geochemical control of metals and arsenic in a Mediterranean river contaminated by acid mine drainage (the Amous River, France); preliminary assessment of impacts on fish (Leuciscus cephalus) by Corinne Casiot; Marion Egal; Françoise Elbaz-Poulichet; Odile Bruneel; Chrystelle Bancon-Montigny; Marie-Ange Cordier; Elena Gomez; Catherine Aliaume (787-799).
Dissolved and particulate concentrations of metals (Fe, Al, Mn, Co, Ni, Cu, Zn, Cd, Tl, Pb) and As were monitored over a 5 year period in the Amous River downstream of its confluence with a creek severely affected by acid mine drainage (AMD) originating from a former Pb–Zn mine. Water pH ranged from 6.5 to 8.8. Metals were predominantly in dissolved form, except Fe and Pb, which were in particulate form. In the particulate phase, metals were generally associated with Al oxides, whereas As was linked to Fe oxides. Metal concentrations in the dissolved and/or particulate phase were generally higher during the wet season due to higher generation of AMD. Average dissolved (size < 0.22 μm) metal concentrations (μg/L) were 1 ± 4 (Fe), 69 ± 49 (Al), 140 ± 118 (Mn), 4 ± 3 Co, 6 ± 4 (Ni), 1.3 ± 0.8 (Cu), 126 ± 81 (Zn), 1.1 ± 0.7 (Cd), 0.9 ± 0.5 (Tl), 2 ± 3 (Pb). Dissolved As concentrations ranged from 5 to 134 μg/L (30 ± 23 μg/L). During the survey, the concentration of colloidal metals (5 kDa < size < 0.22 μm) was less than 25% of dissolved concentrations. Dissolved metal concentrations were generally higher than the maximum concentrations allowed in European surface waters for priority substances (Ni, Cd and Pb) and higher than the environmental quality standards for other compounds. Using Diffusion Gradient in Thin Film (DGT) probes, metals were shown to be in potentially bioavailable form. The concentrations in Leuciscus cephalus were below the maximum Pb and Cd concentrations allowed in fish muscle for human consumption by the European Water Directive. Amongst the elements studied, only As, Pb and Tl were shown to bioaccumulate in liver tissue (As, Pb) or otoliths (Tl). Bioaccumulation of metals or As was not detected in muscle.

Multi-element signatures of stream sediments and sources under moderate to low flow conditions by M.I. Stutter; S.J. Langan; D.G. Lumsdon; L.M. Clark (800-809).
This study assesses a simple sediment source tracing method using major- (Al, Ca, Fe, K, Mg, Mn, Na, P, Si, Ti) and trace-element (Ba, Be, Ce, Co, Cr, Mo, Nd, Pb, Sr, Th, V, Y, Zn) signatures of stream suspended particulate matter (SPM), bed sediments and soils in a small agricultural catchment in NE Scotland. Whilst most erosion studies characterise the large amounts of material mobilised at the highest flows, this study aimed to assess properties of sediments during moderate to low flow periods. These occur more frequent than intense storms and are important in linking stream sediments, near-channel sources and aquatic ecosystem impacts. Data were transformed by multivariate statistical methods to compare elemental signatures of SPM (ranging from 3 to 53 mg L−1 in the stream) and stream bed sediments with a limited number of near-channel source soils. Increased concentrations of Ce, Nd, Th and Y in subsoils contributed to the ability to discriminate between surface fieldslope and stream bank erosion sources. Stream bed sediments showed close matches with compositions of stream bank and arable surface soils, but signatures of SPM differed greatly from any of the sources. Large concentrations of Cr, Pb and Zn in SPM, particularly during summer (677, 177 and 661 mg kg−1, respectively) exceeded water quality standards and were linked to an accumulation of trace elements associated with biological material. The potential for within-stream alteration of SPM in relation to erosion sources was confirmed by changes in the nature of the SPM organic matter observed by IR spectroscopy. Thus the potential is shown for multi-element signatures to give information on catchment sediment sources to aid land management decisions, given careful consideration of the effects of in-stream alteration of eroded material. However, this combined information may be beneficial to process understanding linking land use and stream ecosystems at critical ecological periods.

Use of B isotopes as a tracer of anthropogenic emissions in the atmosphere of Paris, France by Benjamin Chetelat; Jérôme Gaillardet; Rémi Freydier (810-820).
In order to test the potential of B isotopes as a tracer of contamination of the atmosphere, the B isotopic composition of rainwater samples monitored over a year in the centre of Paris, France were determined. Boron concentrations range from 19 nmol/L to 500 nmol/L and δ 11B range from 0‰ to +38‰. Mean annual values are 148 nmol/L and +25‰, respectively. The results suggest that variability in B isotopic compositions is mainly caused by mixing of two main sources, although isotopic fractionation during the evaporation–condensation processes may also be important. One source is a marine component, which exhibits a heavy B isotopic composition. The decrease of δ 11B in rainwater with increasing NO3/B and SO4/B molar ratios suggests that a second source may be anthropogenic emissions. To constrain this end-member, B was determined in urban particulates, which were enriched in the light isotope and the lowest values were consistent with a B-rich fossil fuel composition. These results confirm the great sensitivity of B to anthropogenic sources and the ability of B isotopic ratios to decipher the origin of B in the atmosphere.

A compilation and comparison of fracture mineral studies from the Canadian and Fennoscandian Shields and the French Massif Central shows many similarities indicating larger external control over fracture mineral deposition, with different rock types exerting local controls. The sites investigated represent a wide range of geological settings, and host rock types ranging from felsic intrusive and extrusives to ultramafic intrusives and volcanics that span an age range from 2.5 to 0.36 Ga. Typical fracture minerals found at Canadian Shield sites include calcite, quartz, chlorite and clays, and these do not appear to be dependant on age, erosional depth or geological environment. The Fennoscandian Shield has a much larger variety of fracture filling minerals with epidote, zeolites, prehnite, fluorite, pyrrhotite, Fe oxides, serpentine, graphite, magnesite and barite in addition to the minerals typically found at Canadian Shield sites. The major control on fracture mineral type is most likely variations in rock type, and fluid chemistry and temperature.The C and O isotopic range of calcite is very similar among sites. Late-stage hydrothermal calcite, with strongly depleted δ 18O values, is common at many sites. All of the sites have calcite with δ 18O isotopic values in the range of −5 to −20‰ PDB, indicative of formation from meteoric water or basinal brines that have undergone varying degrees of water/rock interaction. One Canadian and a few Swedish sites have calcite in the shallower portion of the rock that shows isotopic evidence of dissolution and re-precipitation in equilibrium with the present-day waters. There are some striking similarities in fluid inclusion data among sites. Most sites have an elevated temperature (100–300 °C), low salinity group of fluid inclusions within the NaCl–H2O system, and a lower temperature (50–150 °C), higher salinity group of fluid inclusions within the NaCl–CaCl2–H2O system. Fluid inclusion density plots show some evidence of simple cooling, but most sites show two or more fluids were responsible for calcite formation. The origin of most of these fluids was magmatic/hydrothermal or meteoric water that had undergone varying degrees of water/rock interaction, but basinal brines and seawater were also possible sources.Several techniques and methods have been used to further characterize calcites. Strontium isotopes and rare earth elements can be useful to recognize different families of calcite. Uranium–Th dating has found many old calcites beyond the useful range of the technique, but also some relatively young calcites that may be related to interglacial periods. Where fluid inclusion data exists, formation temperatures were not consistent with a glacial water origin. Crush and leach experiments (with ion and gas chromatography and thermal ion mass spectrometry) have characterized inclusion fluids, but special care must be used to ensure only one generation is sampled at a time. Cathodoluminescence and scanning electron microscopy with energy-dispersive spectrometry has been useful in identifying multiple fluid generations within single calcite samples. Laser ablation and Raman spectrometry are additional techniques that are useful in determining individual fluid inclusion chemistry and isotopes.

Controlled electrochemical dissolution of hydrothermal and sedimentary pyrite by Ran Liu; Amy L. Wolfe; David A. Dzombak; Colin P. Horwitz; Brian W. Stewart; Rosemary C. Capo (836-842).
Electrochemically controlled pyrite dissolution was performed with three pyrite materials from different geological origins under mixed potential and high overpotential conditions. Both solid electrodes and C paste electrodes of powdered pyrite were used. The rate of pyrite dissolution increased with applied positive potential and was strongly affected by temperature. Current density measurements over the applied potential range successfully described the rate of pyrite dissolution of each pyrite electrode. Controlled dissolution performed under mixed potential conditions on the solid electrodes successfully reflected the same pyrite reactivity and dissolution rate order as in batch reactor dissolution studies with the same pyrite materials. Therefore, the relative reactivity of different pyrite materials can be determined through current density measurements on their solid electrodes under mixed potential conditions. This technique could be a useful tool to compare rapidly the relative reactivity for different pyrite materials. In contrast, electrochemically controlled dissolution studies with C paste electrodes constructed with fine-grained pyrite and paraffin/graphite mixture did not result in accurate ranking of pyrite samples by dissolution rate.

The shallow aquifer beneath the Western Snake River Plain (Idaho, USA) exhibits widespread elevated arsenic concentrations (up to 120 μg L−1). While semi-arid, crop irrigation has increased annual recharge to the aquifer from approximately 1 cm prior to a current rate of >50 cm year−1. The highest aqueous arsenic concentrations are found in proximity to the water table (all values >50 μg L−1 within 50 m) and concentrations decline with depth. Despite strong vertical redox stratification within the aquifer, spatial distribution of aqueous species indicates that redox processes are not primary drivers of arsenic mobilization. Arsenic release and transport occur under oxidizing conditions; groundwater wells containing dissolved arsenic at >50 μg L−1 exhibit elevated concentrations of O2 (average 4 mg L−1) and NO3 (average 8 mg L−1) and low concentrations of dissolved Fe (<20 μg L−1). Sequential extractions and spectroscopic analysis of surficial soils and sediments indicate solid phase arsenic is primarily arsenate and is present at elevated concentrations (4–45 mg kg−1, average: 17 mg kg−1) relative to global sedimentary abundances. The highest concentrations of easily mobilized arsenic (up to 7 mg kg−1) are associated with surficial soils and sediments visibly stained with iron oxides. Batch leaching experiments on these materials using irrigation waters produce pore water arsenic concentrations approximating those observed in the shallow aquifer (up to 152 μg L−1). While As:Cl aqueous phase relationships suggest minor evaporative enrichment, this appears to be a relic of the pre-irrigation environment. Collectively, these data indicate that infiltrating irrigation waters leach arsenic from surficial sediments to the underlying aquifer.

Carbon dioxide degassing and thermal energy release in the Monte Amiata volcanic-geothermal area (Italy) by Francesco Frondini; Stefano Caliro; Carlo Cardellini; Giovanni Chiodini; Nicola Morgantini (860-875).
The quaternary volcanic complex of Mount Amiata is located in southern Tuscany (Italy) and represents the most recent manifestation of the Tuscan Magmatic Province. The region is characterised by a large thermal anomaly and by the presence of numerous CO2-rich gas emissions and geothermal features, mainly located at the periphery of the volcanic complex. Two geothermal systems are located, at increasing depths, in the carbonate and metamorphic formations beneath the volcanic complex. The shallow volcanic aquifer is separated from the deep geothermal systems by a low permeability unit (Ligurian Unit). A measured CO2 discharge through soils of 1.8 × 109  mol a−1 shows that large amounts of CO2 move from the deep reservoir to the surface. A large range in δ 13CTDIC (−21.07 to +3.65) characterises the waters circulating in the aquifers of the region and the mass and isotopic balance of TDIC allows distinguishing a discharge of 0.3 × 109  mol a−1 of deeply sourced CO2 in spring waters. The total natural CO2 discharge (2.1 × 109  mol a−1) is slightly less than minimum CO2 output estimated by an indirect method (2.8 × 109  mol a−1), but present-day release of 5.8 × 109  mol a−1 CO2 from deep geothermal wells may have reduced natural CO2 discharge. The heat transported by groundwater, computed considering the increase in temperature from the infiltration area to the discharge from springs, is of the same order of magnitude, or higher, than the regional conductive heat flow (>200 mW m−2) and reaches extremely high values (up to 2700 mW m−2) in the north-eastern part of the study area. Heat transfer occurs mainly by conductive heating in the volcanic aquifer and by uprising gas and vapor along fault zones and in those areas where low permeability cover is lacking. The comparison of CO2 flux, heat flow and geological setting shows that near surface geology and hydrogeological setting play a central role in determining CO2 degassing and heat transfer patterns.

The molecular and stable isotope compositions of coalbed gases from the Upper Carboniferous strata and natural gases accumulated within the autochthonous Upper Miocene Skawina Formation of the Dębowiec-Simoradz gas deposit were determined, as well as the chemical and stable isotope compositions of waters from the Skawina Formation and waters at the top of the Upper Carboniferous strata of the Kaczyce Ridge (the abandoned “Morcinek” coal mine) in the South-Western part of the Upper Silesian Coal Basin. Two genetic types of natural gases within the Upper Carboniferous coal-bearing strata were identified: thermogenic (CH4, small amounts of higher gaseous hydrocarbons, and CO2) and microbial (CH4, very small amounts of ethane, and CO2). Thermogenic gases were generated during the bituminous stage of coalification and completed at the end of the Variscan orogeny. Degassing (desorption) of thermogenic gases began at the end of late Carboniferous until the late Miocene time-period and extended to the present-day. This process took place in the Upper Carboniferous strata up to a depth of about 550 m under the sealing Upper Miocene cover. A primary accumulation zone of indigenous, thermogenic gases is present below the degassing zone. Up to 200 m depth from the top of the Upper Carboniferous strata, within the weathered complex, an accumulation zone of secondary, microbial gas occurs. Waters within these strata are mainly of meteoric origin of the infiltration period just before the last sea transgression in the late Miocene and partly of marine origin having migrated from the Upper Miocene strata. Then, both methanogenic archaebacteria and their nutrients were transported by meteoric water into the near-surface Carboniferous strata where the generated microbial CH4 saturated coal seams. Waters within the Miocene strata of the Dębowiec-Simoradz and Zabłocie are of marine origin, and natural gases accumulated within autochthonous Miocene strata of the Dębowiec-Simoradz gas deposit were most probably generated by microbial processes of on organic matter dispersed within the strata, though some contribution of gases migrating from the Carboniferous coal-bearing strata cannot be excluded.

Acid groundwater in an anoxic aquifer: Reactive transport modelling of buffering processes by Gudrun Franken; Dieke Postma; Wilhelmus H.M. Duijnisveld; Jürgen Böttcher; John Molson (890-899).
The acidification of groundwater, due to acid rain, was investigated in a Quaternary sandy aquifer in the Fuhrberger Feld, near Hannover, Germany. The groundwater, recharged through an area covered by a coniferous forest, had a pH in the range 4–5 down to a depth of 5 m. The evolution in groundwater chemistry along the flow path was investigated in a transect of multisamplers. A 2D groundwater flow model was established delineating the groundwater flow field and a groundwater flow velocity of around 80 m/a along the flow path was derived. Speciation calculations showed the groundwater to be close to equilibrium with the mineral jurbanite (AlOHSO4) over the pH range 4.0–6.5. This suggests an accumulation of acid rain derived SO 4 2 - in the aquifer sediment during the decades with high atmospheric S deposition. The groundwater has a pH of around 4.5 in the upstream part of the flow path increasing to near 6 further downstream. 1D reactive transport modelling, using PHREEQC, was used to analyze different combinations of buffering processes. The first model contains ion exchange in combination with jurbanite dissolution. At the ion exchange front Al3+ is adsorbed leading to the dissolution of jurbanite and an increase in pH. Comparison with field data showed that the simulated increases in pH and alkalinity are much lower than observed in the field. The second model includes organic matter degradation. In addition to ion exchange and jurbanite dissolution, the model included the reduction of SO 4 2 - and Fe-oxides as well as the precipitation of Fe sulfide. This model matches the field data well and illustrates the importance of redox processes for pH buffering in the Fuhrberg aquifer. The current progress of the acidification front is about 4 m/a. This corresponds to an average value of 150 a of acid input, which covers large historical variations. Remediation is expected to take the same time span because it requires desorption and neutralization of adsorbed Al3+ from the aquifer sediment.

This paper presents the results of a study on the geochemistry of waters circulating in the mineralised area of the south-eastern sector of Mt. Peloritani (north-eastern Sicily, Italy), aimed at basic understanding of the geochemical processes influencing their chemical composition. Chemico-physical parameters and data on 26 major and minor chemical elements are reported for 103 water samples. Water chemistry is mainly dominated by dissolution of carbonates and hydrolysis of aluminosilicate minerals. Total dissolved salts (TDS) range from 80 to 1398 mg/L. All the waters exhibit E H characteristic of an oxygenated environment. Excluding two samples, which show very high H+ activity (pH = 3.0 and 2.7), all the waters have pH values in the range 6.2–8.6. Cluster analysis based on major ion contents defined three main chemical water types, reflecting different hydrochemical processes. The first, group I, has low salinity (average TDS = 118 ± 30 mg/L) and abundance orders (meq/L) Na > Ca ≈ Mg > K and Cl ≈ HCO3  > SO4. With increased water–rock interaction, waters in groups II and III become more saline, changing composition towards SO4–Cl-alkaline earth and HCO3-alkaline earth types. Weathering of carbonate minerals causes waters to become saturated with respect to calcite and dolomite, whereas the incongruent dissolution of aluminosilicate minerals causes the solution to reach equilibrium with kaolinite and to form smectites. Trace element geochemistry in the analysed waters reflects interactions between waters and existing mineralisation, with elemental concentrations showing highly variable values, and higher concentrations of As, Pb, Sb and Zn near known mineralisation. Lead–Zn and As–Sb statistical associations, probably distinguishing interactions with different mineralogical phase paragenesis, were revealed by factor analysis. The main aqueous chemical forms of trace elements predicted by chemical speciation calculations are also reported. As most of the analysed spring waters provide the main source of freshwater for domestic purposes, attention should be given to As and Sb, whose concentrations exceed the recommended limits.

Hydrogeological investigations conducted by the Geological Survey of Canada in the Lake Saint-Martin region of Manitoba have confirmed earlier reports of naturally elevated F and B concentrations in local groundwaters. Fluoride and B concentrations are highly correlated (r 2  = 0.905) and reach 15.1 mg/L and 8.5 mg/L, respectively. Virtually all groundwaters with F concentrations greater than the drinking water limit of 1.5 mg/L are from wells within the Lake Saint-Martin impact structure, a 208 Ma complex crater 23 km in diameter underlying a large part of the study area. The high-F groundwaters can be classified into two groups according to their anionic and isotopic compositions. Group I samples consist of Na-mixed anion groundwaters, with Cl greater than 100 mg/L and highly depleted 18O compositions indicative of recharge under much cooler climatic conditions than at present. Samples belonging to this group exhibit a striking relationship to crater morphology, and are found in an arcuate belt within the southern rim of the impact structure. Group II high-F samples consist of Na–HCO3–SO4 groundwaters, with little Cl, and less depleted 18O compositions. Samples belonging to this group are associated with groundwaters recharged locally, on a low ridge within the impact structure. This paper traces the probable source of high-F groundwaters to phosphatic pellets in shales of the Winnipeg Formation, a regional basal clastic unit which sub-crops at shallow depth beneath the crater rim as a result of more than 200 m of structural uplift associated with the impact event. This extensive aquifer is known elsewhere in southern Manitoba for its naturally-softened groundwaters and locally elevated F concentrations. Group I groundwaters are interpreted as discharge from the Winnipeg Formation where it abuts against crater-fill deposits. Group II high-F groundwaters are interpreted as modern recharge from within the impact structure, displacing Group I groundwaters. Thus, elevated F and B concentrations observed in groundwaters of the Lake Saint-Martin area represent the geochemical signature of upwelling from a deep regional aquifer. The previously unsuspected discharge zone occurs within an isolated sub-crop of the aquifer formed as a result of structural uplift caused by the impact event.

Arsenic mobilization from mine tailings in the presence of a biosurfactant by Suiling Wang; Catherine N. Mulligan (928-935).
Batch experiments were conducted to investigate As mobilization from mine tailings in the presence of a biosurfactant (JBR425, mixed rhamnolipids) and to evaluate the feasibility of using biosurfactant in remediating As contaminated mine tailings/soils. Introduction of the biosurfactant increased As mobilization greatly. When the mass ratio was 10 mg biosurfactant/g mine tailings at pH 11, As mobilization by the biosurfactant was greatest after 24 h, with a corresponding concentration ratio (the ratio of As mobilization by the biosurfactant to that by distilled water at same adjusted pH, wt/wt) of 21.6. Selective sequential extraction indicated that As was easily mobilized from the weakly bound and relatively more mobile fractions by washing with the biosurfactant. A mobilization isotherm was developed to predict As mobilization from the mine tailings in the presence of biosurfactant. It was shown that biosurfactant sorption to the mine tailings is essential for As mobilization. Arsenic mobilization was found to be positively correlated with the mobilization of Fe and other metals (i.e., Cu, Pb and Zn), which might further enhance As mobilization by helping incorporate it into soluble complexes or micelles. Capillary electrophoresis analyses indicated that As redox or methylation reactions had insignificant effect on As mobilization. Biosurfactants might be used potentially to remove bulk As from mine tailings or contaminated soils under alkaline conditions.

Estimating selenium removal by sedimentation from the Great Salt Lake, Utah by Wade Oliver; Christopher Fuller; David L. Naftz; William P. Johnson; Ximena Diaz (936-949).
The mass of Se deposited annually to sediment in the Great Salt Lake (GSL) was estimated to determine the significance of sedimentation as a permanent Se removal mechanism. Lake sediment cores were used to qualitatively delineate sedimentation regions (very high to very low), estimate mass accumulation rates (MARs) and determine sediment Se concentrations. Sedimentation regions were defined by comparison of isopach contours of Holocene sediment thicknesses to linear sedimentation rates determined via analysis of 210Pb, 226Ra, 7Be and 137Cs activity in 20 short cores (<5 cm), yielding quantifiable results in 13 cores. MARs were developed via analysis of the same radioisotopes in eight long cores (>10 cm). These MARs in the upper 1–2 cm of each long core ranged from 0.019 to 0.105 gsed/cm2/a. Surface sediment Se concentrations in the upper 1 or 2 cm of each long core ranged from 0.79 to 2.47 mg/kg. Representative MARs and Se concentrations were used to develop mean annual Se removal by sedimentation in the corresponding sedimentation region. The spatially integrated Se sedimentation rate was estimated to be 624 kg/a within a range of uncertainty between 285 and 960 kg/a. Comparison to annual Se loading and other potential removal processes suggests burial by sedimentation is not the primary removal process for Se from the GSL.

Samples of surface water, soil, sediment and plants from the Xunyang Hg mining area in Shaanxi Province, China, were analyzed to assess the effects of the Hg mining activities on the environment. The results show that: (1) the mining activities are sources of Hg to the environment surrounding the mine; (2) the environment, especially riverbeds in the Xunyang area, is contaminated with Hg and (3) Hg accumulation in cabbage leaves, the main vegetable for the local residents, is high, reaching concentrations that are 6–540 times higher than the maximum Hg concentration recommended for edible plants in China. Action should be taken to remediate contaminated sites, manage waste discharges and reduce the exposure of Hg to local residents by issuing advisories against consuming agricultural products grown in the area, which are contaminated with Hg.

This work is aimed at investigating the weathering processes of the granodiorites cropping out in a small catchment of the Sila Massif. The mineral constituents in this granodiorite are plagioclase, often zoned with a Ca-rich core and a Na-rich rim, quartz, chlorite, K-feldspar, white mica and epidote. During this study, dolomite was discovered in local stream sediments, as separate monomineralic grains, probably resulting from erosion of veins cutting the crystalline rocks. Prevailing dissolution of foreign dolomite and a Ca-rich plagioclase is suggested by the Ca–Mg–HCO3 chemical composition of local groundwaters and stream waters, which is rather unexpected for waters interacting with granitoid rocks. These qualitative observations are quantitatively confirmed by reaction path modelling of the weathering processes occurring in the study area, which was carried out using the EQ3/6 software package, version 8.0, and the Double Solid Reactant Method. Indeed, it was possible to ascertain that the release of both major dissolved constituents and several trace elements (Ba, Co, Cr, Fe, Mn, Ni, Pb, Sr, V and Zn), from rocks to waters, is chiefly controlled by the dissolution of foreign dolomite and the Ca-rich core of zoned plagioclases.

Increasing attention is being focused on the rapid rise of CO2 levels in the atmosphere, which many believe to be the major contributing factor to global climate change. Sequestering CO2 in deep geological formations has been proposed as a long-term solution to help stabilize CO2 levels. However, before such technology can be developed and implemented, a basic understanding of H2O–CO2 systems and the chemical interactions of these fluids with the host formation must be obtained. Important issues concerning mineral stability, reaction rates, and carbonate formation are all controlled or at least significantly impacted by the kinetics of rock–water reactions in mildly acidic, CO2-saturated solutions. Basalt has recently been identified as a potentially important host formation for geological sequestration. Dissolution kinetics of the Columbia River Basalt (CRB) were measured for a range of temperatures (25–90 °C) under mildly acidic to neutral pH conditions using the single-pass flow-through test method. Under anaerobic conditions, the normalized dissolution rates for CRB decrease with increasing pH (3 ⩽ pH ⩽ 7) with a slope, η, of −0.15 ± 0.01. Activation energy, E a, has been estimated at 32.0 ± 2.4 kJ mol−1. Dissolution kinetics measurements like these are essential for modeling the rate at which CO2-saturated fluids react with basalt and ultimately drive conversion rates to carbonate minerals in situ.

Can pelagic net heterotrophy account for carbon fluxes from eastern Canadian lakes? by Kristal Dubois; Richard Carignan; Ján Veizer (988-998).
Lakes worldwide are commonly oversaturated with CO2, however the source of this CO2 oversaturation is not well understood. To examine the magnitude of the C flux to the atmosphere and determine if an excess of respiration (R) over gross primary production (GPP) is sufficient to account for this C flux, metabolic parameters and stable isotopes of dissolved O2 and C were measured in 23 Québec lakes. All of the lakes sampled were oversaturated with CO2 over the sampling period, on average 221 ± 25%. However, little evidence was found to conclude that this CO2 oversaturation was the result of an excess of pelagic R over GPP. In lakes Croche and à l’Ours, where CO2 flux, R and GPP were measured weekly, the annual difference between pelagic GPP and R, or net primary production (NPP), was not sufficient to account for the size of the CO2 flux to the atmosphere. In Lac Croche average annual NPP was 14.4 mg C m−2 d−1 while the average annual flux of CO2 to the atmosphere was 34 mg C m−2 d−1. In Lac à l’Ours average annual NPP was −9.1 mg C m−2  d−1 while the average annual flux of CO2 to the atmosphere was 55 mg C m−2  d−1. In all of the lakes sampled, O2 saturation averaged 104.0 ± 1.7% during the ice-free season and the isotopic composition of dissolved O218ODO) was 22.9 ± 0.3‰, lower than atmospheric values and indicative of net autotrophy. Carbon evasion was not a function of R, nor did the isotopic signature of dissolved CO2 in the lakes present evidence of excess R over GPP. External inputs of C must therefore subsidize the lake to explain the continued CO2 oversaturation. The isotopic composition of dissolved inorganic C (δ13CDIC) indicates that the CO2 oversaturation cannot be attributed to in situ aerobic respiration. δ13CDIC reveals a source of excess C enriched in 13C, which may be accounted for by anaerobic sediment respiration or groundwater inputs followed by kinetic isotope fractionation during degassing under open system conditions.

The sorption of Cs on crushed and intact (non-crushed) Opalinus Clay (OPA) was studied. A Cs sorption isotherm was measured in a synthetic OPA pore water covering an equilibrium concentration (C Cs,eq) range between 10−2  M and 10−9  M. The isotherms measured on crushed (<63 μm) and intact OPA were in excellent agreement indicating that all sorption sites for Cs in intact OPA are available. The isotherms were modelled assuming that illite is the dominating clay mineral responsible for Cs sorption. The model described the results very well at low Cs concentrations, over-predicted the results in the mid-concentration range and under-predicted the data at the highest Cs equilibrium concentrations. Improvements in the mid-concentration range were obtained by lowering the log K Cs K c selectivity coefficient of the Type II sites by 0.4 log units. At the highest Cs concentration an improvement can be achieved by considering sorption of Cs on other clay minerals present in OPA. The main result from this study is that in the case of cation exchange processes the exchange sites in the intact material are all available. This corroborates the common practice in performance assessment (PA) of applying batch sorption data obtained on dispersed systems to the intact rock for predicting the retention behaviour of radionuclides.

Geochemistry of trace elements in surface waters of the Arno River Basin, northern Tuscany, Italy by Gianni Cortecci; Tiziano Boschetti; Enrico Dinelli; Rosa Cidu; Francesca Podda; Marco Doveri (1005-1022).
Trace element geochemistry of the Arno River and its main tributaries was investigated on the basis of two sampling campaigns carried out in November 1996 and June 1997. By analyzing filtered and unfiltered water samples, Fe and Al are found in solution mainly as colloidal particles of size lower than 0.45 μm. In June (lower flow rate), Fe and Al are enriched in the filtered waters from the main river, and this feature was interpreted in terms of higher water temperature promoting the formation of smaller particles, thus reducing their aggregation properties. Iron and Al show perfectly synchronous downstream profiles along the Arno River, correlate quite well each to other, and display abrupt concentration increases near to Florence, where the lithology of the catchment changes from siliciclastic dominated to clay-sand (lacustrine-marine)-dominated. The same behaviour is shown by most of the other trace elements in the river, thus supporting a general lithological control. Trace elements in the final part of the Arno River are influenced by flocculation processes in addition to mixing. Adsorption phenomena on oxy-hydroxides are denoted by good elemental correlations with Fe (and Al). Sporadic anomalous concentration values, possibly related to anthropogenic contributions, may prevent such correlations. Referring to the quality of waters for potable use and fish life, toxic elements are below the acceptable limits of current European regulations, with few exceptions for Hg exceeding guideline values. Multivariate analysis groups trace elements according to geochemical affinities and natural or anthropogenic sources, thus distinguishing contaminated from uncontaminated samples. The results achieved in this work will help regional and national Authorities for compliance with the EU water policy, especially in assessing the water quality at the river basin scale and its vulnerability to human activities.

Oxidizing conditions normally prevail in surface waters and near-surface groundwaters, but there is usually a change to reducing conditions in groundwater at greater depth. Dissolved O2 originally present is consumed through biogenic and inorganic reactions along the flow paths. Fracture minerals participate in these reactions and the fracture mineralogy and geochemistry can be used to trace the redox front. An important task in the safety assessment of a potential repository for the disposal of nuclear waste in crystalline bedrock, at an approximate depth of 500 m in Sweden, is to demonstrate that reducing conditions can be maintained for a long period of time. Oxygen may damage the Cu canisters that host nuclear waste; additionally, in the event of a canister failure, oxidizing conditions may increase the mobility of some radionuclides. The present study of the near-surface redox front is based on mineralogical (redox-sensitive minerals), geochemical (redox-sensitive elements) and U-series disequilibrium investigations of mineral coatings along open fractures. The fractures have been sampled along drill cores from closely spaced, 100 m deep boreholes, which were drilled during the site investigation work in the Laxemar area, south-eastern Sweden, carried out by the Swedish Nuclear Fuel and Waste Management Co. (SKB). The distribution of the redox-sensitive minerals pyrite and goethite in open fractures shows that the redox front (switch from mainly goethite to mainly pyrite in the fractures) generally occurs at about 15–20 m depth. Calcite leaching by recharging water is indicated in the upper 20–30 m and positive Ce-anomalies suggest oxidation of Ce down to 20 m depth. The U-series radionuclides show disequilibrium in most of the samples, indicating mobility of U during the last 1 Ma. In the upper 20 m, U is mainly removed (due to oxidation) or has experienced complex removal and/or deposition. At depths of 35–55 m, both deposition and removal of U are indicated. Below 55 m, recent deposition of U is generally indicated which suggests removal of U near surface (oxidation) and deposition of U below the redox front. Scattered goethite occurrences below the general redox front (down to ca 80 m) and signs of U removal at 35–55 m mostly correlate with sections of high transmissivity (and/or high fracture frequencies). This shows that highly transmissive fractures are generally required to allow oxygenated groundwaters at depth greater than ca 30 m. Removal of U (oxidation) below 55 m within the last 300 ka is not observed. Although penetration of glacial waters to great depths has been confirmed in the study area, their potential O2 load seems to have been reduced near the surface.