Applied Geochemistry (v.23, #4)

An X-ray absorption study of the fate of technetium in reduced and reoxidised sediments and mineral phases by K. Morris; F.R. Livens; J.M. Charnock; I.T. Burke; J.M. McBeth; J.D.C. Begg; C. Boothman; J.R. Lloyd (603-617).
Technetium is a long lived (2.13 × 105  a), beta emitting radionuclide which is a groundwater contaminant at a number of nuclear facilities throughout the world. Its environmental behaviour is primarily governed by its redox state. Under oxic conditions it forms the highly soluble pertechnetate ( TcO 4 - ) ion; under reducing conditions it forms the poorly soluble, reduced forms of Tc, particularly the Tc(IV) ion which is expected to precipitate as hydrous TcO2 above its solubility limit (10−9  mol l−1 at ∼pH 7) or to be strongly sorbed to mineral surfaces at lower concentrations. Thus the redox cycling behaviour of Tc is predicted to be key to its environmental behaviour in the natural and engineered environment. Here the results of a series of X-ray absorption spectroscopy (XAS) experiments which examine the oxidation state and coordination environment of Tc in a range of estuarine, aquifer and freshwater sediment suspensions, and in an environmentally relevant amorphous Fe(II) phase under both reduced and reoxidised biogeochemical conditions are presented. In reduced sediments and the amorphous Fe(II) phase prior to reoxidation, XAS results show that Tc was retained as hydrous TcO2-like phases across all samples. Under air reoxidation, experiments showed significant (up to 80%) remobilisation of Tc to solution as TcO 4 - . In pre-reduced freshwater sediments, aquifer sediments and the amorphous Fe(II) phase oxidised with air, XAS indicated that Tc remained associated with the solids as hydrous TcO2-like phases. By contrast, in air reoxidised estuarine sediment XAS analysis suggested that both hydrous TcO2-like phases and TcO 4 - were retained within the sediment. Finally, when microbially-mediated NO3 reoxidation occurred in estuarine and aquifer sediment slurries, experiments showed comparatively low (<8%) remobilisation of Tc from solids over similar timescales to air reoxidation experiments, whilst XAS again showed that both hydrous TcO2-like phases and TcO 4 - were retained within the sediment. By contrast, in the amorphous Fe(II) phase, although NO3 reoxidation again led to low (<4%) remobilisation of Tc from solids, XAS analysis showed that Tc was retained as hydrous TcO2-like phases alone. These results are discussed in the context of the redox cycling behaviour of Tc in the natural and engineered environment.

Mercury is a contaminant of great concern in the marine environment, particularly in coastal environments where the formation of methylmercury (MeHg) in aquatic sediment, and its subsequent bioaccumulation in edible aquatic organisms (mainly fish), presents a major pathway for human exposure to MeHg. Consequently, it is important to determine the factors controlling MeHg production in sediment, especially in contaminated environments. This study investigates some geochemical factors affecting the speciation and distribution of Hg in estuarine sediment from two highly industrialized macrotidal salt marsh/mudflat systems in the Seine estuary, France, and in the Medway estuary, UK. Obtained data revealed that the entire sediment core from a rapidly accreting mudflat in the Seine estuary (170 cm vertical accretion over the last 10–50 a) was contaminated with Hg at concentrations which are 10–50 times higher than the Hg background for sediments of the Seine basin. In the Medway mudflat, characterized by a slow sedimentation rate (400 cm over approximately the last 800 a), near-surface sediment was significantly more contaminated (10–40 times) than sediment at greater depths, which were characterized by pre-industrial Hg concentrations. Geochemical conditions in the surface sediment of the Medway mudflat are characterized by stable anoxic redox conditions (about −200 mV), which are generally favourable for Hg methylation, whereas near-surface sediment of the Seine mudflat is characterized by more oxidizing redox conditions (about +100 mV), which are generally less conducive to Hg methylation. Consequently, MeHg concentration in the upper 10 cm of the sediment column was about four times higher in the Medway than in the Seine mudflat, in spite of similar total Hg concentrations. In surface sediment, where Hg is actively methylated, MeHg variability was associated with the activity of SO 4 2 - -reducing microorganisms (SRM) and the presence of acid volatile sulphides (AVS). A strong correlation was observed between MeHg and AVS in sediments from these mudflats, which may be a consequence of the common origin of AVS and MeHg (both produced by microorganism activity), but also can be derived from the ability of Fe monosulfides to adsorb, and thus stabilize, solid phase MeHg.

Magnesium in well water and the spatial variation of acute myocardial infarction incidence in rural Finland by A. Kousa; A.S. Havulinna; E. Moltchanova; O. Taskinen; M. Nikkarinen; V. Salomaa; M. Karvonen (632-640).
The protective role of water hardness and Mg in cardiovascular illness has been suggested in several epidemiological studies. In the present ecological study, the association of Ca, Mg, Al, Cu, F, Fe, Zn and NO3 in local ground water and spatial variation of acute myocardial infarction (AMI) incidence among men and women 35–74 years of age in rural Finland in 1991–2003 were examined. Data on AMI cases, 67,755 men and 25,450 women, were obtained from the Finnish Cardiovascular Disease Register. The statistical analysis was carried out using Bayesian modeling. 10 × 10 km grid cells were used instead of administrative boundaries to partition the study area. On average, 1 mg/L increment in Mg level in local ground water was associated with 2% (95% HDR −0.0391, −0.0028) decrease in incidence of AMI in the rural population. In conclusion, high AMI incidence in eastern Finland is associated with soft ground water, low in Mg.

Cation exchanged Fe(II) and Sr compared to other divalent cations (Ca, Mg) in the bure Callovian–Oxfordian formation: Implications for porewater composition modelling by Christophe Tournassat; Catherine Lerouge; Philippe Blanc; Jocelyne Brendlé; Jean-Marc Greneche; Stéphane Touzelet; Eric C. Gaucher (641-654).
Iron and Sr bearing phases were thoroughly investigated by means of spectrometric and microscopic techniques in Callovian–Oxfordian (COX) samples originating from the ANDRA Underground Research Laboratory (URL) in Bure (France). Strontium was found to be essentially associated with celestite, whereas Fe was found to be distributed over a wide range of mineral phases. Iron was mainly present as Fe(II) in the studied samples (∼93% from Mössbauer results). Most of the Fe(II) was found to be in pyrite, sideroplesite/ankerite and clay minerals. Iron(III), if present, was associated with clay minerals (probably illite, illite-smectite mixed layer minerals and chlorite). No Fe(III) oxy(hydro)xide could be detected in the samples. Strontianite was not observed either. Based on these observations, it is likely that the COX porewater is in equilibrium with the following carbonate minerals, calcite, dolomite and ankerite/sideroplesite, but not with strontianite. It is shown that this equilibrium information can be combined with clay cation exchange composition information in order to give direct estimates or constraints on the solubility products of the carbonate minerals dolomite, siderite and strontianite. As a consequence, an experimental method was developed to retrieve the cation exchanged Fe(II) in very well preserved COX samples.The very homogeneous cation exchange composition of the formation is completely in agreement with a homogeneous presence of calcite and dolomite minerals whose equilibrium reactions control part of the porewater composition. Amongst the broad range of values available for dolomite solubility products in thermodynamic databases, the value of log  K dolomite  = −3.57 is the most appropriate for the one present in the COX formation. With regard to strontianite, it appears that the equilibrium constant tabulated in the Llnl database is not valid for natural clay systems. The value given by Busenberg et al. used by most of the other available thermodynamic databases seems to be far more appropriate. Concerning Fe(II) and siderite/ankerite equilibrium, the measured Fe/Ca ratio on the clay exchanger (∼0.01) could only be considered as a maximum value due to possible experimental bias, leading to the following constraint for the solubility of the sideroplesite phase present in the COX:Fe0.7Mg0.2Ca0.1CO3  + H+ ⇔ 0.7 Fe2+  + 0.2 Mg2+  + 0.1 Ca2+  +  HCO 3 - logK sid_COX  < 0.42This constraint is in agreement with the estimated solubility of this phase based on a solid solution model: log  K sid_COX  = 0.23.

Diffusion of anionic species in Callovo-Oxfordian argillites and Oxfordian limestones (Meuse/Haute–Marne, France) by M. Descostes; V. Blin; F. Bazer-Bachi; P. Meier; B. Grenut; J. Radwan; M.L. Schlegel; S. Buschaert; D. Coelho; E. Tevissen (655-677).
Diffusive parameters of tritiated water (HTO) and several anionic tracers (36Cl, 125I, 35 SO 4 2 - and 75 SeO 3 2 - ) were determined in the Callovo-Oxfordian argillite formation (Meuse/Haute–Marne, France), and also in the Oxfordian limestone formation, located just above. Twenty two drillcore samples, from depths of 150–480 m and from almost each lithofacies along the same borehole EST205, were tested. Three distinct and complementary experimental set-ups were used (batch, column filled with crushed argillite and through-diffusion experiments) associated with a thorough characterisation of the rock, care being taken to maintain redox conditions. Tritiated water behaviour is similar to water which diffuses in the total porosity of the rock. Tritiated water and anionic tracers displayed analogous diffusive behaviour in calcareous samples with ε a(anion)/ε a(HTO) = 1 and D e(anion)/D e(HTO) close to 0.85, i.e. the ratio of their diffusion coefficients in free water. Lower and scattered values were observed in the lower part of the Oxfordian limestones which is characterized by the presence of mineralogical heterogeneities and dolomitic diagenesis. In the Callovo-Oxfordian argillite formation, diffusive fluxes of 36Cl, 125I and 35 SO 4 2 - are similar but lower than that of HTO. The diffusive behaviour analogy between I and Cl was verified, as D e(halide)/D e(HTO) and ε a(halide)/ε a(HTO) ratios were always lower than 1. This reduction in diffusivity is a consequence of the anionic exclusion, which limits the diffusion-accessible porosity to a value lower than the total porosity for anions. A weak sorption of I and SO 4 2 - on argillite samples was always observed whatever the technique used. In the case of I, this sorption appears to be partly reversible and kinetically controlled. Very low values of diffusive parameters were obtained for 75 SeO 3 2 - and are still not well explained. Last, diffusive parameters of each radioactive tracer tested can be plotted according to Archie’s law, which therefore allows an estimation of D e(non-reactive species) from the knowledge of both the sample total porosity and the corresponding exponent m  = 2 for argillite samples.

Diffusion of HTO, Br, I, Cs+, 85Sr2+ and 60Co2+ in a clay formation: Results and modelling from an in situ experiment in Opalinus Clay by P. Wersin; J.M. Soler; L. Van Loon; J. Eikenberg; B. Baeyens; D. Grolimund; T. Gimmi; S. Dewonck (678-691).
The migration of radioactive and chemical contaminants in clay materials and argillaceous host rocks is characterised by diffusion and retention processes. Valuable information on such processes can be gained by combining diffusion studies at laboratory scale with field migration tests. In this work, the outcome of a multi-tracer in situ migration test performed in the Opalinus Clay formation in the Mont Terri underground rock laboratory (Switzerland) is presented. Thus, 1.16 × 105  Bq/L of HTO, 3.96 × 103  Bq/L of 85Sr, 6.29 × 102  Bq/L of 60Co, 2.01 × 10−3  mol/L Cs, 9.10 × 10−4  mol/L I and 1.04 × 10−3  mol/L Br were injected into the borehole. The decrease of the radioisotope concentrations in the borehole was monitored using in situ γ-spectrometry. The other tracers were analyzed with state-of-the-art laboratory procedures after sampling of small water aliquots from the reservoir. The diffusion experiment was carried out over a period of one year after which the interval section was overcored and analyzed. Based on the experimental data from the tracer evolution in the borehole and the tracer profiles in the rock, the diffusion of tracers was modelled with the numerical code CRUNCH.The results obtained for HTO (3H), I and Br confirm previous lab and in situ diffusion data. Anionic fluxes into the formation were smaller compared to HTO because of anion exclusion effects. The migration of the cations 85Sr2+, Cs+ and 60Co2+ was found to be governed by both diffusion and sorption processes. For 85Sr2+, the slightly higher diffusivity relative to HTO and the low sorption value are consistent with laboratory diffusion measurements on small-scale samples. In the case of Cs+, the numerically deduced high diffusivity and the Freundlich-type sorption behaviour is also supported by ongoing laboratory data. For Co, no laboratory diffusion data were yet available for comparison; however, the modelled data suggests that 60Co2+ sorption was weaker than would be expected from available batch sorption data. Overall, the results demonstrate the feasibility of the experimental setup for obtaining high-quality diffusion data for conservative and sorbing tracers.

Analysis of arsenic speciation in mine contaminated lacustrine sediment using selective sequential extraction, HR-ICPMS and TEM by Kelly L. Haus; Robert L. Hooper; Laura A. Strumness; J. Brian Mahoney (692-704).
In order to determine how As speciation in lacustrine sediment changes as a function of local conditions, sediment cores were taken from three lakes with differing hydrologic regimes and subjected to extensive chemical and TEM analysis. The lakes (Killarney, Thompson and Swan Lakes) are located within the Coeur d’ Alene River system (northern Idaho, USA), which has been contaminated with trace metals and As, from over 100 a of sulfide mining. Previous analyses of these lakebed sediments have shown an extensive amount of contaminant metals and As associated with sub-μm grains, making them extremely difficult to analyze using standard methods (scanning electron microscopy, X-ray diffraction). Transmission electron microscopy offers great advantages in spatial resolution and can be invaluable in determining As speciation when combined with other techniques. Data indicate that because of differences in local redox conditions, As speciation and stability is dramatically different in these lakes. Killarney and Thompson Lakes experience seasonal water-level fluctuations due to drawdown on a downstream dam, causing changes in O2 content in sediment exposed during drawdown. Swan Lake has relatively constant water levels as its only inlet is dammed. Consequently, Killarney and Thompson Lakes show an increase in labile As-bearing phases with depth, while Swan Lake data indicate stable As hosts throughout the sediment profile. Based on these observations it can be stated that As in lakebed sediments is much less mobile, and therefore less bioavailable, when water is kept at a constant level.

The biosphere: A homogeniser of Pb-isotope signals by Clemens Reimann; Belinda Flem; Arnold Arnoldussen; Peter Englmaier; Tor Erik Finne; Friedrich Koller; Øystein Nordgulen (705-722).
Rock, soil, and plant (terrestrial moss, European mountain ash leaves, mountain birch leaves, bark and wood, and spruce needles and wood) samples, collected at 3 km intervals along a 120 km long transect (40 sites) cutting the city of Oslo, Norway, were analysed for their Pb concentration and Pb-isotope ratios. A general decrease in 206Pb/207Pb, 208Pb/207Pb and 206Pb/208Pb ratios, with a characteristic low variability in all plant materials and the plant-derived O-horizon of soil profiles, compared to rocks and mineral soils, is observed along the transect. It is demonstrated that minerogenic and biogenic sample materials belong to two different spheres, the lithosphere and biosphere, and that geochemical processes determining their chemical and isotopic compositions differ widely. Background variation for both sample materials needs to be established and documented at the continental and global scale before the anthropogenic influence on the geochemistry of the earth’s surface can be reliably estimated.

The concentration and distribution of metals were studied in metallophytes, growing on and in the vicinity of Pb–Zn gossans, NW Queensland. The study investigated the accumulation of metals in plant species and assessed their potential use as indicators in geobotanical and biogeochemical prospecting and as metal excluders in mine site rehabilitation. Plant species colonising the gossans tolerate high concentrations of metals. Total mean metal concentrations of soils ranged from minima of 14 ppm Cu, 28 ppm Pb and 34 ppm Zn in background areas to maxima of 660 ppm Cu, 12000 ppm Pb and 2100 ppm Zn over mineralised soils. Over the gossans, the grass species Eriachne mucronata forma, Enneapogon lindleyanus and Paraneurachne muelleri replace the characteristic grass Triodia molesta where the soils have high Pb and Zn concentrations. Of the 16 plant species identified, 3 of them, Hybanthus aurantiacus, Clerodendrum tomentosum and Bulbostylis barbata, were confined to the gossan sites. B. barbata appears to be of particular use in geobotanical prospecting as it indicates base metal mineralisation in the region.The biogeochemical analyses indicate significant enrichment of Cd, Pb and Zn in the tissue of all plant species, with the abundance of Cd, Pb and Zn in dried vegetation from the gossans being up to one order of magnitude above background. In particular, the uptake of Cd, Pb and Zn by Sida sp., T. molesta, Cleome viscosa and Eriachne mucronata forma increases linearly with DTPA-extractable soil metal concentrations. The biogeochemistry of these plants provides the best anomaly definition of the exposed gossans. Furthermore, the analysis of roots demonstrate that the plant species T. molesta, Eriachne mucronata forma and P. muelleri allow the transport of Cd, Pb and Zn from the roots to the above-ground biomass. Hence, the species best suited for biogeochemical prospecting for base metals in the region and semi-arid inland northern Australia are Eriachne mucronata and the genus Triodia. The analyses also reveal differences in the ability of each species to accumulate metals. Among all gossan plants, Eremophila latrobei displays distinctly low Pb concentrations and low correlations with soil DTPA extractable Pb. This plant is the most efficient in excluding Pb from its biomass, making it most suitable for the revegetation of Pb contaminated soils. This study demonstrates that biogeochemical examinations of gossans can reveal indicator and excluder plants, which are of potential use in mineral exploration as well as mined land reclamation.

Dissolved inorganic carbon (DIC) constitutes a significant fraction of a stream’s carbon budget, yet the role of acid mine drainage (AMD) in DIC dynamics in receiving streams remains poorly understood. The objective of this study was to evaluate spatial and temporal effects of AMD and its chemical evolution on DIC and stable isotope ratio of DIC (δ 13CDIC) in receiving streams. We examined spatial and seasonal variations in physical and chemical parameters, DIC, and δ 13CDIC in a stream receiving AMD. In addition, we mixed different proportions of AMD and tap water in a laboratory experiment to investigate AMD dilution and variable bicarbonate concentrations to simulate downstream and seasonal hydrologic conditions in the stream. Field and laboratory samples showed variable pH, overall decreases in Fe2+, alkalinity, and DIC, and variable increase in δ 13CDIC. We attribute the decrease in alkalinity, DIC loss, and enrichment of 13C of DIC in stream water to protons produced from oxidation of Fe2+ followed by Fe3+ hydrolysis and precipitation of Fe(OH)3(s). The extent of DIC decrease and 13C enrichment of DIC was related to the amount of HCO 3 - dehydrated by protons. The laboratory experiment showed that lower 13C enrichment occurred in unmixed AMD (2.7‰) when the amount of protons produced was in excess of HCO 3 - or in tap water (3.2‰) where no protons were produced from Fe3+ hydrolysis for HCO 3 - dehydration. The 13C enrichment increased and was highest for AMD-tap water mixture (8.0‰) where Fe2+ was proportional to HCO 3 - concentration. Thus, the variable downstream and seasonal 13C enrichment in stream water was due in part to: (1) variations in the volume of stream water initially mixed with AMD and (2) to HCO 3 - input from groundwater and seepage in the downstream direction. Protons produced during the chemical evolution of AMD caused seasonal losses of 50 to >98% of stream water DIC. This loss of DIC in AMD impacted streams may have implications for CO2 transfer to the atmosphere and watershed DIC export.

The provenance of Australian uranium ore concentrates by elemental and isotopic analysis by Elizabeth Keegan; Stephan Richter; Ian Kelly; Henri Wong; Patricia Gadd; Heinz Kuehn; Adolfo Alonso-Munoz (765-777).
Elemental and isotopic ratio analyses of U ore concentrate samples, from the 3 operating U mining facilities in Australia, were carried out to determine if significant variations exist between their products, thereby allowing the U ore concentrate’s origin to be identified. Elemental analyses were conducted using inductively coupled plasma mass spectrometry (ICP-MS) and X-ray fluorescence spectrometry (XRF). Lead isotope ratios were measured using ICP-MS and U isotope analyses were conducted using thermal ionisation mass spectrometry (TIMS). Minute quantities of sample, such as that obtained from a swipe, were also examined for elemental concentrations using secondary ion mass spectrometry (SIMS). The results of multivariate statistical analysis show clear patterns in the trace elemental composition of the processed U ores, indicating that it is possible to use this feature as a unique identifier of an Australian U ore concentrate’s source. Secondary ion mass spectrometry analyses also allow individual particles to be differentiated using this ‘fingerprinting’ technique. Isotope ratios determined using TIMS reveal that there is a significant difference in the n(234U)/n(238U) isotope ratio between the U ore concentrate from each mine.

Trace element associations with Fe- and Mn-oxides in soil nodules: Comparison of selective dissolution with electron probe microanalysis by Alexander Neaman; Carmen Enid Martínez; Fabienne Trolard; Guilhem Bourrié (778-782).
Selective dissolution methods have been largely used to get insight on trace element association with solid phases. Modern instrumental techniques offer many tools to test the validity of selective dissolution methods and should be systematically used to this end. The association of trace elements with Fe- and Mn-oxides in soil nodules has been studied here by electron probe microanalysis. The results were compared with findings from an earlier study on selective dissolution of the same nodules by hydroxylamine hydrochloride, acidified hydrogen peroxide, and Na-citrate-bicarbonate-dithionite. Electron probe microanalysis results were consistent with previous findings using selective dissolution and showed that P, As and Cr were mainly present in Fe-oxides, while Co was mainly associated with Mn-oxide phases. These results support the applicability of the studied selective dissolution methods for fractionation of trace elements in soils and sediments containing appreciable amounts of Fe and Mn-oxide phases.
Keywords: Hydroxylamine-hydrochloride; Hydrogen-peroxide; Citrate-bicarbonate-dithionite; Tropical-soil; EPMA; Iron-oxides; Manganese-oxides; Fractionation;

Iron-complexes on the surface of minerals may play an important role in Fe dissolution in acidic cloud water containing certain organic ligands, and dissolved Fe serves as a critical nutrient in biogeochemical cycles in certain aquatic systems. As the first step to explore this issue, laboratory experiments were conducted to investigate the effects of oxalate on the dissolution of hematite, with leaching of Fe from oxalate-coated hematite in comparison with pure hematite in oxalate solution. The dissolution of oxalate-coated hematite was measured as a function of dissolution time, loading amount of oxalate and pH. The amount of oxalate adsorbed on hematite at pH 2.4 is greater than that at pH 5, while the amount of adsorption increases with increasing oxalate equilibrium concentration in solution. Adsorption of oxalate on hematite follows the Freundlich adsorption model. The amount of Fe dissolved at pH 2.4 is much more than that at pH 5. In low concentration oxalate solution, the amount of Fe dissolution from hematite is independent of oxalate loading on the surface of hematite. In high concentration oxalate solution, however, a relatively high oxalate loading on the hematite surface releases more Fe relative to low oxalate adsorption density on the surface of hematite when the system reaches equilibrium, suggesting that the high content of Fe(III)–oxalate complexation promotes Fe dissolution. Ferric ion dissolution and adsorbed oxalate leaching in solution as a function of pH are two co-existing processes, and pH ∼2.5 is a critical turning point relating the two processes that occur simultaneously. The fitting of the experimental data from this work to a model indicates that Fe–(oxalate)+ and Fe – ( oxalate ) 2 - are predominant species in solution in the pH range of 1.5–4.5 during oxalate-coated hematite dissolution in background electrolyte.

Experiments utilizing meteorologically normalized sampling conditions were used to illustrate the role and function of urban pavement, bare soil and turf grass surface properties with respect to the air-surface exchange of total gaseous Hg (TGM). After ensuring uniform meteorological effects to each surface, resultant TGM fluxes from turf grass, bare soil and pavement were specifically representative of their diverse physical and biogeochemical properties. Results spanning the entire sampling year show distinct TGM flux signatures for each surface (5.69 ± 5.79 (ng/m2  h) for bare soil, 0.53 ± 1.25 for turf grass, 0.26 ± 0.41 for pavement). Based on medians, the surface limitations of pavement decreased TGM flux by a factor of 22 compared to bare soil and by a factor of 2 compared to turf grass. Turf surface limitations decreased TGM flux by a factor of 11 compared to bare soil. By comparing these results to a parallel study, meteorological effects were found to develop 24% of the TGM flux signature for pavement, 53% for turf and 60% for bare soil. The remaining percentage contributions to each TGM flux signature were from the cumulative surface property effects of each surface. These results suggest that the greater the TGM flux magnitude for a particular surface, the more measurements are needed under a wide variety of meteorological conditions to develop a broad understanding of its TGM flux characteristics. Seasonal observation allowed closer investigation of a large shift to TGM deposition for the turf surface during the fall season. The large shift toward deposition was suspected to be linked to the formation of a thatch layer on the unexposed soil surface just beneath the turf layer.

The “Mina Fe” U deposit (Salamanca, Spain) has been studied in the context of Enresa’s programme for U-mine sites restoration and also as a natural analogue for processes in high-level nuclear waste (HLNW) geological disposal. The investigations encompassed an array of geoscience disciplines, such as structural geology, mineralogy, hydrogeology and elemental and isotopic geochemistry and hydrogeochemistry of the site. Based on the obtained results, a conceptual mineralogical and geochemical model was performed integrating the main geochemical processes occurring at the site: the interaction between oxidised and slightly acidic water with pyrite, pitchblende, calcite and dolomite, as essential minerals of the U fracture-filling mineralisation, and hydroxyapatite from the host rock, as the main source of P. This conceptual model has been tested in a systematic numerical model, which includes the main kinetic (pyrite and pitchblende dissolution) and equilibrium processes (carbonate mineral dissolution, and goethite, schoepite and autunite secondary precipitation). The results obtained from the reactive-transport model satisfactorily agree with the conceptual model previously established. The assumption of the precipitation of coffinite as a secondary mineral in the system cannot be correctly evaluated due to the lack of hydrochemical data from the reducing zone of the site and valid thermodynamic and kinetic data for this hydrated U(IV)-silicate. This precipitation can also be hampered by the probable existence of dissolved U(IV)-organic matter and/or uranyl carbonate complexes, which are thermodynamically stable under the alkaline and reducing conditions that prevail in the reducing zone of the system. Finally, the intense downwards oxic and acidic alteration in the upper part of the system is of no relevance for the performance assessment of a HLNW disposal. However, the acidic and oxidised conditions are quickly buffered to neutral–alkaline and reducing at very shallow depths, of relevance for the performance assessment of a HLNW repository, even in a natural or artificially perturbed geological environment as “Mina Fe”.

High levels of Cd and Zn in Jamaican soils observed in geochemical surveys are related to the presence of phosphorites of possible Late-Miocene or Pliocene age. The trace element and REE geochemistry of the phosphorites, together with SEM studies, indicate a guano origin for the phosphorites. No specific host minerals for Cd could be identified in the fossiliferous phosphorite which is characterized by uniquely high levels of Cd, Zn, Ag, Be, U and Y. However, in the soil Cd is present in lithiophorite and a complex history of pedological development is preserved in the aluminous–goethite present in the soil. The unique guano signature is preserved in the soil despite the fact that guanos themselves have either not been observed or have been destroyed by continuing karst and soil development. The phosphorite geochemical signature can be traced in the data of a 1988 island-wide soil geochemical survey, identifying areas where the Palaeo-environment that supported bird ‘rookeries’ existed in the Late-Miocene or Pliocene.

Particulate-associated potentially harmful elements in urban road dusts in Xi’an, China by Yongming Han; Junji Cao; Eric S. Posmentier; Kochy Fung; Hui Tian; Zhisheng An (835-845).
Sixty five urban road dust samples were collected from different land use areas of ∼240 km2 in Xi’an, China. The concentrations of Ag, As, Cr, Cu, Hg, Pb, Sb and Zn were determined to investigate potentially harmful element (PHE) contamination, distribution and possible sources. In addition, the concentrations in different size fractions were measured to assess their potential impact on human health. The highest concentrations were found in the fraction with particle diameters between 80 μm and 101 μm, the finest particles (<63 μm) were not the most important carriers for Ag, As, Cd, Cr, Cu, Hg, Pb and Zn. The percentages of these elements in particles with diameters less than 63 μm (PM63) and less than 101 μm (PM101) were in the range of 7–15%, and 30–55%, respectively. Three main factors influencing element distributions have been identified: (a) industrial activities; (b) prior agricultural land use; and (c) other activities commonly found in urban areas, such as traffic, coal combustion, waste dumping, and building construction/renovation. The highest concentrations were found in industrial areas for As (20 mg kg−1), Cr (853 mg kg−1), Cu (1071 mg kg−1), Pb (3060 mg kg−1) and Zn (2112 mg kg−1), and in previous agricultural areas for Ag and Hg, indicating significant contributions from industrial activities and prior agricultural activities.

Recharge sources and hydrogeochemical evolution of groundwater in semiarid and karstic environments: A field study in the Granada Basin (Southern Spain) by Claus Kohfahl; Christoph Sprenger; Jose Benavente Herrera; Hanno Meyer; Franzisca Fernández Chacón; Asaf Pekdeger (846-862).
The objective of this study is to refine the understanding of recharge processes in watersheds representative for karstic semiarid areas by means of stable isotope analysis and hydrogeochemistry. The study focuses on the Granada aquifer system which is located in an intramontane basin bounded by high mountain ranges providing elevation differences of almost 2900 m. These altitude gradients lead to important temperature and precipitation gradients and provide excellent conditions for the application of stable isotopes of water whose composition depends mainly on temperature. Samples of rain, snow, surface water and groundwater were collected at 154 locations for stable isotope studies (δ 18O, D) and, in the case of ground- and surface waters, also for major and minor ion analysis. Thirty-seven springs were sampled between 2 and 5 times from October 2004 to March 2005 along an altitudinal gradient from 552 masl in the Granada basin to 2156 masl in Sierra Nevada. Nine groundwater samples were taken from the discharge of operating wells in the Granada basin which are all located between 540 and 728 masl. The two main rivers were monitored every 2–3 weeks at three different altitudes. Rainfall being scarce during the sampling period, precipitation could only be sampled during four rainfall events. Calculated recharge altitudes of springs showed that source areas of mainly snowmelt recharge are generally located between 1600 and 2000 masl. The isotope compositions of spring water indicate water sources from the western Mediterranean as well as from the Atlantic without indicating a seasonal trend. The isotope pattern of the Quaternary aquifer reflects the spatial separation of different sources of recharge which occur mainly by bankfiltration of the main rivers. Isotopic signatures in the southeastern part of the aquifer indicate a considerable recharge contribution by subsurface flow discharged from the adjacent carbonate aquifer. No evaporation effects due to agricultural irrigation were detected.

A regional scale hydrogeochemical study of a ∼21,000-km2 area in the western Bengal basin shows the presence of hydrochemically distinct water bodies in the main semiconfined aquifer and deeper isolated aquifers. Spatial trends of solutes and geochemical modeling indicate that carbonate dissolution, silicate weathering, and cation exchange control the major-ion chemistry of groundwater and river water. The main aquifer water has also evolved by mixing with seawater from the Bay of Bengal and connate water. The isolated aquifers contain diagenetically altered water of probable marine origin. The postoxic main aquifer water exhibits overlapping redox zones (metal-reducing, sulfidic and methanogenic), indicative of partial redox equilibrium, with the possibility of oxidation in micro-scale environments. The redox processes are depth-dependent and hydrostratigraphically variable. Elevated dissolved As in the groundwater is possibly related to Fe(III) reduction, but is strongly influenced by coupled Fe–S–C redox cycles. Arsenic does not show good correlations with most solutes, suggesting involvement of multiple processes in As mobilization. The main river in the area, the Bhagirathi–Hoogly, is chemically distinctive from other streams in the vicinity and probably has little or no influence on deep groundwater chemistry. Arsenic in water of smaller streams (Jalangi and Ichamati) is probably introduced by groundwater discharge during the dry season.

Metal mobilization from base-metal smelting slag dumps in Sierra Almagrera (Almería, Spain) by Andrés Navarro; Esteve Cardellach; José L. Mendoza; Mercé Corbella; Luis M. Domènech (895-913).
Smelting slags associated with base-metal vein deposits of the Sierra Almagrera area (SE Spain) show high concentrations of Ag (<5–180 ppm), As (12–750 ppm), Cu (45–183 ppm), Fe (3.2–29.8%), Pb (511–2150 ppm), Sb (22–620 ppm) and Zn (639–8600 ppm). The slags are mainly composed of quartz, fayalite, barite, melilite, celsian, pyrrhotite, magnetite, galena and Zn–Pb–Fe alloys. No glassy phases were detected. The following weathering-related secondary phases were found: jarosite–natrojarosite, cotunnite, cerussite, goethite, ferrihydrite, chalcanthite, copiapite, goslarite, halotrichite and szomolnokite. The weathering of slag dumps near the Mediterranean shoreline has contaminated the soils and groundwater, which has caused concentrations in groundwater to increase to 0.64 mg/L Cu, 40 mg/L Fe, 0.6 mg/L Mn, 7.6 mg/L Zn, 5.1 mg/L Pb and 19 μg/L As. The results of laboratory leach tests showed major solubilization of Al (0.89–12.6 mg/L), Cu (>2.0 mg/L), Fe (0.22–9.8 mg/L), Mn (0.85–40.2 mg/L), Ni (0.092–2.7 mg/L), Pb (>2.0 mg/L) and Zn (>2.5 mg/L), and mobilization of Ag (0.2–31 μg/L), As (5.2–31 μg/L), Cd (1.3–36.8 μg/L) and Hg (0.2–7 μg/L). The leachates were modeled using the numerical code PHREEQC. The results suggested the dissolution of fayalite, ferrihydrite, jarosite, pyrrhotite, goethite, anglesite, goslarite, chalcanthite and cotunnite. The presence of secondary phases in the slag dumps and contaminated soils may indicate the mobilization of metals and metalloids, and help to explain the sources of groundwater contamination.

Geochemical characterization of surface waters and groundwater resources in the Managua area (Nicaragua, Central America) by F. Parello; A. Aiuppa; H. Calderon; F. Calvi; D. Cellura; V. Martinez; M. Militello; K. Vammen; D. Vinti (914-931).
This paper reports new geochemical data on dissolved major and minor constituents in surface waters and ground waters collected in the Managua region (Nicaragua), and provides a preliminary characterization of the hydrogeochemical processes governing the natural water evolution in this area. The peculiar geological features of the study site, an active tectonic region (Nicaragua Depression) characterized by active volcanism and thermalism, combined with significant anthropogenic pressure, contribute to a complex evolution of water chemistry, which results from the simultaneous action of several geochemical processes such as evaporation, rock leaching, mixing with saline brines of natural or anthropogenic origin. The effect of active thermalism on both surface waters (e.g., saline volcanic lakes) and groundwaters, as a result of mixing with variable proportions of hyper-saline geothermal Na–Cl brines (e.g., Momotombo geothermal plant), accounts for the high salinities and high concentrations of many environmentally-relevant trace elements (As, B, Fe and Mn) in the waters. At the same time the active extensional tectonics of the Managua area favour the interaction with acidic, reduced thermal fluids, followed by extensive leaching of the host rock and the groundwater release of toxic metals (e.g., Ni, Cu). The significant pollution in the area, deriving principally from urban and industrial waste-water, probably also contributes to the aquatic cycling of many trace elements, which attain concentrations above the WHO recommended limits for the elements Ni (∼40 μg/l) and Cu (∼10 μg/l) limiting the potential utilisation of Lake Xolotlan for nearby Managua.

Final disposal of high-level radioactive waste in deep repositories in clay formations is being considered by several countries. Repository safety assessment requires the use of numerical models of groundwater flow, solute transport and chemical processes. These models are being developed from data and knowledge gained from in situ experiments such as the CERBERUS experiment carried out at the HADES facility excavated in the Boom clay formation at Mol (Belgium). This long-term experiment is aimed at evaluating the effect of heating and radiation in Boom clay. The test was performed in a cased well drilled at 223 m depth and lasted from 1989 to 1994. A 60Co source of 400 TBq and two heaters were emplaced inside the well. Dose rate, temperature, porewater pressure and pH/Eh were measured in situ during the experiment and gas and porewater samples were taken for chemical analyses. Here a coupled thermo-hydro-geochemical (THC) model of the CERBERUS experiment is presented which accounts simultaneously for heating, radiation, solute diffusion and a suite of geochemical reactions including: aqueous complexation, acid–base, redox, mineral dissolution/precipitation, cation exchange and gas dissolution/ex-solution. Computed results indicate that heating and radiation causes a slight oxidation, a decrease in pH, slight changes in porewater chemistry and pyrite dissolution near the well. THC model results follow the general evolution of chemical data, but cannot fit SO4 data. Model discrepancies are partly overcome when microbially-mediated Fe and SO4 reduction are taken into account in a coupled thermo-hydro-bio-geochemical (THBC) model. This THBC model captures the trends of geochemical data, improves the fit to dissolved SO4 and predicts pyrite precipitation, a process observed near the CERBERUS well. The ability of the THBC numerical model to reproduce the overall trends of geochemical data of the CERBERUS experiment provides confidence in such a model as a suitable tool for the long-term prediction of geochemistry in the near field of a HLW repository in clay. However, the small number of available chemical data throughout the experiment and the lack of DOC and microbial data allow only a partial validation of the THBC model.