Applied Geochemistry (v.24, #4)

12th International Symposium on Water–Rock Interaction (WRI-12) by William C. Evans; Russell S. Harmon; Richard B. Wanty (491-492).

Preliminary measurements were carried out of the solubility of the O2-buffering assemblage bismuth + bismite (Bi2O3) in aqueous liquid–vapor and vapor-only systems at temperatures of 220, 250 and 300 °C. All experiments were carried out in Ti reaction vessels and were designed such that the Bi solids were contained in a silica tube that prevented contact with liquid water at any time during the experiment. Two blank (no Bi solids present) liquid–vapor experiments at 220 °C yielded Bi concentrations (±1σ) in the condensed liquid of 0.22 ± 0.02 mg/L, whereas the solubility measurements at this temperature yielded an average value of approximately 6 ± 9 mg/L, with replicate experiments ranging from 0.3 to 26 mg/L. Although the 6 mg/L value is associated with a considerable degree of uncertainty, the experiments do indicate transport of Bi through the vapor phase. Measured Bi concentrations in the condensed liquid at 250 °C were in the same range as those at 220 °C, whereas those at 300 °C were significantly lower (i.e., all below the blank value). Vapor-only experiments necessarily contained much smaller initial volumes of water, thereby making the results more susceptible to contamination. Single blank runs at 220 and 300 °C yielded Bi concentrations of 82 and 16 mg/L, respectively. Measured concentrations (±1σ) of Bi in the vapor-only solubility experiments at 220 °C were 235 ± 78 mg/L for an initial water volume of 0.5 mL, and at 300 °C were 56 ± 30 mg/L and 33 ± 21 for initial water volumes of 1 and 2 mL, respectively, suggesting strong preferential partitioning of Bi into the vapor. The results indicate a negative dependence of Bi solubility on temperature, but are inconclusive with respect to the dependence of Bi solubility on water density or fugacity.The experiments reported here suggest that significant Bi transport is possible in the vapor phase. Comparison of the liquid–vapor and vapor-only experiments further suggests that, at the conditions studied, the solubility of Bi in the vapor is significantly higher than that in the liquid phase (calculated distribution coefficients, i.e., C Vapor/C Liquid at 220 °C ranged from 81 to 168, where C is concentration in mg/L). The concentrations measured in the vapor-only experiments at 220 °C are approximately 11 orders of magnitude higher than those calculated from available thermodynamic data assuming no interaction between volatile Bi species and water molecules. This finding indicates that hydration of volatile Bi species probably occurs to a significant extent; similar to behavior observed previously in published vapor phase solubility studies of other metals. However, the interpretation of the results was complicated by the formation of Bi silicate phases and the lack of certainty that O2 fugacity buffering was effective throughout the experiments.

The Red Lake greenstone belt is one of the foremost Au mining camps in Canada and hosts the world-class Campbell-Red Lake Au deposit. Belt-scale hydrothermal alteration is characterized by proximal ferroan dolomite zones associated with Au mineralization surrounded by distal calcite zones, both being accompanied by potassic alterations (sericite, muscovite, and biotite). At the Campbell-Red Lake and Cochenour deposits Au mineralization (in particular high-grade ore) is associated with silica and sulfides (especially arsenopyrite) that replace carbonate ± quartz veins and the host rocks. The prevalence of carbonic fluid inclusions and rare occurrence of aqueous-bearing inclusions in carbonate–quartz–Au veins in the Campbell-Red Lake deposit, and the consistency of homogenization temperatures of carbonic inclusions within individual fluid inclusion assemblages (FIA), have been interpreted to indicate that H2O-poor, CO2-dominated fluids were responsible for the carbonate veining and Au mineralization. Further studies of fluid inclusions in carbonate–quartz veins within and outside the deformation zone hosting the Campbell-Red Lake deposit (the Red Lake Mine trend) including the Cochenour Au deposit, the Redcon Au prospect, and outcrops in the distal calcite zone also reveal the dominance of carbonic fluid inclusions in vein minerals. These studies indicate that CO2-dominated fluids were flowing through fractures during carbonate vein formation and Au mineralization both within and outside major structures. The carbonic fluid may have been initially undersaturated with water, or it may have resulted from phase separation of an H2O–CO2–NaCl fluid. In the latter case, phase separation modeling indicates that the initial fluid likely had X CO 2 values larger than 0.8. Calculations based on hydrothermal mineral assemblages indicate X CO 2 values in the host rocks from 0.025 to 0.85, reflecting a change from CO2-dominated fluids in the fractures (veins) to H2O-dominated fluids in the host rocks away from the fractures. The CO2-dominated fluids were likely advected from granulite facies in the deeper crust, whereas the H2O-dominated fluids were derived from the ambient host rocks of amphibole to greenschist facies. Calculations based on CO2 requirements and source constraints indicate that the mineralizing fluids were likely two orders of magnitude more enriched in Au than the commonly assumed values of a few μg/L, which may explain why the Campbell-Red Lake deposit has a very high-grade of Au (average 21 g/t for the whole deposit and 81 g/t for the Goldcorp High-Grade zone). Fluid inclusion data suggest that the carbonate veining and Au mineralization likely took place at depths from 7 to 14 km. The development of crustiform–colloform structures in the carbonate ± quartz veins, which was previously interpreted to indicate relatively shallow environments, may alternatively have been related to extremely high fluid pressures and the CO2-dominated nature of the fluids, which could have enhanced the brittle properties of the rocks due to their high wetting angles.

Fluid inclusions and clusters of water molecules at nanometer-to submicron-scale in size have been investigated using transmission electron microscopy (TEM) and Fourier transform infrared spectroscopy (FTIR) in jadeite, quartz and symplectite aegirine–augite, albite, taramite and magnetite corona minerals from ultrahigh-pressure (UHP) jadeite–quartzite at Shuanghe, the Dabie Mountains, China. Fluid inclusions from 0.003 μm to 0.78 μm in size occur in jadeite and quartz crystals, and a small number of fluid inclusions from 0.001 μm to 0.25 μm have also been detected in symplectite-forming minerals. Most of the fluid inclusions have round or negative crystal morphology and contain aqueous fluids, but some contain CO2-rich fluids. They are usually connected to dislocations undetectable at an optical scale. The dislocations represent favorable paths for fluid leakage, accounting for non-decrepitation of most fluid inclusions when external pressure decreased at later stages, although there was partial decrepitation of some fluid inclusions unconnected to defect microstructures resulting from internal overpressure. Non-decrepitation and partial decrepitation of fluid inclusions resulted in changes of original composition and/or density. It is clear that identification of hidden re-equilibration features has significant implications for the petrological interpretation of post-peak metamorphic processes. Micro-FTIR results show that all jadeite and quartz samples contain structural water occurring as hydroxyl ions (OH) and free water (H2O) in the form of clusters of water molecules. The H2O transformed from OH during exhumation and could have triggered and enhanced early retrograde metamorphism of the host rocks and facilitated plastic deformation of jadeite and quartz grains by dislocation movement, and thus the H2O released during decompression might represent early-stage retrograde metamorphic fluid. The nominally anhydrous mineral (NAM) jadeite is able to transport aqueous fluids in concentrations of at least several hundred ppm water along a subduction zone to mantle depths in the form of clusters of water molecules and hydroxyl ions within crystals.

Diffuse gas emissions at the Ukinrek Maars, Alaska: Implications for magmatic degassing and volcanic monitoring by William C. Evans; Deborah Bergfeld; Robert G. McGimsey; Andrew G. Hunt (527-535).
Diffuse CO2 efflux near the Ukinrek Maars, two small volcanic craters that formed in 1977 in a remote part of the Alaska Peninsula, was investigated using accumulation chamber measurements. High CO2 efflux, in many places exceeding 1000 g m−2  d−1, was found in conspicuous zones of plant damage or kill that cover 30,000–50,000 m2 in area. Total diffuse CO2 emission was estimated at 21–44 t d−1. Gas vents 3-km away at The Gas Rocks produce 0.5 t d−1 of CO2 that probably derives from the Ukinrek Maars basalt based on similar δ 13C values (∼−6‰), 3He/4He ratios (5.9–7.2 R A), and CO2/3He ratios (1–2 × 109) in the two areas. A lower 3He/4He ratio (2.7 R A) and much higher CO2/3He ratio (9 × 1010) in gas from the nearest arc-front volcanic center (Mount Peulik/Ugashik) provide a useful comparison. The large diffuse CO2 emission at Ukinrek has important implications for magmatic degassing, subsurface gas transport, and local toxicity hazards. Gas–water–rock interactions play a major role in the location, magnitude and chemistry of the emissions.

Formation waters of the 14 km thick late Cretaceous–Cenozoic Beaufort–Mackenzie basin were examined as part of a larger project to better understand the petroleum potential of the region, where early exploration defined petroleum reserves of 744 × 109  bbls recoverable crude oil and 11.74 tcf gas. Historical water analyses (2583 samples from 250 wells drilled up to 5 km depth) were compiled and culled to remove incomplete and poor quality samples. The resultant database shows a broad range of salinity and water chemistry that has no systematic relationship with depth. Three main water types are defined, paleo seawater, and freshwaters related to a Miocene age gravity-driven flow system, and low TDS–high alkalinity waters. High alkalinity waters are isolated in overpressured fault blocks that were rapidly buried by post-Miocene Iperk shale deposition. The high alkalinities (up to 9000 mg/L) are interpreted to be related to in situ CO2 generation through anaerobic methanogenesis in response to freshwater invasion. The dominant control on biogenic gas generation appears to be maximum burial temperature rather than the modern temperature distribution. This is consistent with the paleopasteurization model that suggests once critical burial temperatures are reached, sterilized rocks are inhibited from further biodegradation, even when temperatures subsequently drop back into the habitable zone.

The origin and evolution of formation water from Upper Jurassic to Upper Cretaceous mudstone–packstone–dolomite host rocks at the Jujo–Tecominoacán oil reservoir, located onshore in SE-Mexico at a depth from 5200 to 6200 m.b.s.l., have been investigated, using detailed water geochemistry from 12 producer wells and six closed wells, and related host rock mineralogy. Saline waters of Cl–Na type with total dissolved solids from 10 to 23 g/L are chemically distinct from hypersaline Cl–Ca–Na and Cl–Na–Ca type waters with TDS between 181 and 385 g/L. Bromine/Cl and Br/Na ratios suggest the subaerial evaporation of seawater beyond halite precipitation to explain the extreme hypersaline components, while less saline samples were formed by mixing of high salinity end members with surface-derived, low salinity water components. The dissolution of evaporites from adjacent salt domes has little impact on present formation water composition. Geochemical simulations with Harvie-Mφller-Weare and PHRQPITZ thermodynamic data sets suggest secondary fluid enrichment in Ca, HCO3 and Sr by water–rock interaction. The volumetric mass balance between Ca enrichment and Mg depletion confirms dolomitization as the major alteration process. Potassium/Cl ratios below evaporation trajectory are attributed to minor precipitation of K feldspar and illitization without evidence for albitization at the Jujo–Tecominoacán reservoir. The abundance of secondary dolomite, illite and pyrite in drilling cores from reservoir host rock reconfirms the observed water–rock exchange processes. Sulfate concentrations are controlled by anhydrite solubility as indicated by positive SI-values, although anhydrite deposition is limited throughout the lithological reservoir column. The chemical variety of produced water at the Jujo–Tecominoacán oil field is related to a sequence of primary and secondary processes, including infiltration of evaporated seawater and original meteoric fluids, the subsequent mixing of different water types and the formation of secondary minerals by water–rock interaction. A best fit between measured and calculated reservoir temperatures was obtained with the Mg–Li geothermometer for high salinity formation water (TDS > 180 g/L), whereas Na–K, Na–Ka–Ca and quartz geothermometers are partially applicable for less salinite water (TDS < 23 g/L).

The chemical and isotopic characterization of formation water from 18 oil production wells, extracted from 5200 to 6100 m b.s.l. at the Jujo–Tecominoacán carbonate reservoir in SE-Mexico, and interpretations of historical production records, were undertaken to determine the origin and hydraulic behavior of deep groundwater systems. The infiltration of surface water during Late Pleistocene to Early Holocene time is suggested by 14C-concentrations from 2.15 to 31.86 pmC, and by 87Sr/86Sr-ratios for high-salinity formation water (0.70923–0.70927) that are close to the composition of Holocene to modern seawater. Prior to infiltration, the super-evaporation of seawater reached maximum TDS concentrations of 385 g/L, with lowest δ 18O values characterizing the most hypersaline samples. Minor deviations of formation water and dolomite host rocks from modern and Jurassic 87Sr/86Sr-seawater composition, respectively, suggest ongoing water–rock interaction, and partial isotopic equilibration between both phases. The abundance of 14C in all sampled formation water, 87Sr/86Sr-ratios for high-salinity water close to Holocene – present seawater composition, a water salinity distribution that is independent of historic water-cut, and a total water extraction volume of 2.037 MMm3 (1/83–4/07) excludes a connate, oil-leg origin for the produced water of the Jurassic–Cretaceous mudstone-dolomite sequence. Temporal fluctuations of water chemistry in production intervals, the accelerated migration of water fronts from the reservoir flanks, and isotopic mixing trends between sampled wells confirms the existence of free aquifer water below oil horizons. Vertical and lateral hydraulic mobility has probably been accelerated by petroleum extraction.The combination of interpreting historical fluctuations of salinity and water percentage in production wells with chemical-isotopic analysis of formation water resulted in a successful method to distinguish four groundwater bodies, stratified vertically within the Jujo–Tecominoacán reservoir. Two with low TDS from 10 to 23 g/L are preserved in the upper reservoir section, mainly in Lower Cretaceous and Kimmeridgian strata. For the deeper part of the reservoir, 87Sr/86Sr trends indicate an affiliation of most samples to two independent mixing trends between “intermediate saline” (TDS ∼200 g/L) and hypersaline (>350 g/L) groundwater end-members.

Strontium isotopes (87Sr/86Sr) are routinely measured in hydrochemical studies to determine sources and mixing relationships. They have proved particularly useful in determining weathering processes and quantifying end-member mixing processes. A number of routine case studies are presented which highlight that Sr isotopes represent a powerful tool in the geochemists toolbox helping to constrain weathering reactions, weathering rates, flow pathways and mixing scenarios. Differences in methodologies for determining the weathering component in natural environments, inherent differences in weathering rates of different minerals, and mineral heterogeneity often cause difficulties in defining the weathering component of different catchments or aquifer systems. Nevertheless, Sr isotopes are useful when combined with other hydrochemical data, to constrain models of water–rock interaction and mixing as well as geochemical processes such as ion-exchange. This paper presents a summary of recent work by the authors in constraining the sources of waters and weathering processes in surface catchments and aquifers, and indicates cases where Sr isotopes alone are insufficient to solve hydrological problems.

High As groundwater is widely distributed in the northwestern Hetao Plain, an arid region with slow groundwater flow. Arsenic concentration in groundwater ranges from 1 to 1000 μg/L. Most water samples have elevated salinities, with Cl and/or HCO3 as the dominant anions and Na as the dominant cation. High concentrations of As in shallow aquifers are associated with strongly reducing conditions, as evidenced by high concentrations of dissolved organic C (DOC), NH4, dissolved sulfide, arsenite and dissolved CH4, and relatively low concentrations of NO 3 - and SO 4 2 - . Results of the hydrochemical, and H and O isotope geochemical studies indicate that evapotranspiration is an important process controlling the enrichment of Na and Cl as well as trace elements such as As, B, F and Br in groundwater. In Na–HCO3-dominated groundwaters, As, B and F were enriched. Decades of irrigation using Yellow River water has resulted in elevation of the groundwater level, which has accelerated salt accumulation in shallow groundwater and surface soil. In addition, irrigation is responsible for the release of some components from aquifer materials and mixing with saline groundwaters, as indicated by minor element and isotope geochemical data. Used to trace groundwater flow paths, Sr isotope composition also indicates that bedrock weathering is one of the primary sources of As in groundwater in the study area.

Geochemistry of surface water in alpine catchments in central Colorado, USA: Resolving host-rock effects at different spatial scales by Richard B. Wanty; Philip L. Verplanck; Carma A. San Juan; Stanley E. Church; Travis S. Schmidt; David L. Fey; Ed H. DeWitt; Terry L. Klein (600-610).
The US Geological Survey is conducting a study of surface-water quality in the Rocky Mountains of central Colorado, an area of approximately 55,000 km2. Using new and existing geologic maps, the more than 200 rock formations represented in the area were arranged into 17 groups based on lithologic similarity. The dominant regional geologic feature affecting water quality in central Colorado is the Colorado mineral belt (CMB), a NE-trending zone hosting many polymetallic vein or replacement deposits, and porphyry Mo deposits, many of which have been mined historically. The influence of the CMB is seen in lower surface-water pH (<5), and higher concentrations of SO 4 2 - (>100 mg/L) and chalcophile metals such as Cu (>10 μg/L), Zn (>100 μg/L), and Cd (>1 μg/L) relative to surface water outside the CMB. Not all streams within the CMB have been affected by mineralization, as there are numerous catchments within the CMB that have no mineralization or alteration exposed at the surface. At the regional-scale, and away from sites affected by mineralization, hydrothermal alteration, or mining, the effects of lithology on water quality can be distinguished using geochemical reaction modeling and principal components analysis. At local scales (100 s of km2), effects of individual rock units on water chemistry are subtle but discernible, as shown by variations in concentrations of major lithophile elements or ratios between them. These results demonstrate the usefulness of regional geochemical sampling of surface waters and process-based interpretations incorporating geologic and geochemical understanding to establish geochemical baselines.

The distribution of several trace elements in different aqueous fractions has been studied in running waters from Sardinia (Italy). Trace elements and major components were determined in water samples collected at high- and low-discharge from rivers (90 samples) and streams (70 samples). At selected sites, total (non filtered samples) and dissolved (0.4 μm and 0.015 μm pore-size filtered samples) amounts of trace elements were determined, and the composition of the solid matter retained on the filters was investigated for estimating the eventual interrelationship. The elements B, Li, Rb, Sr, Ba, As, Sb, Mo, Tl and U in the studied waters showed small differences between total and dissolved amounts; dissolved concentrations were higher under low flow conditions, when the contribution of rainwater to the rivers was minimum; their concentrations were often correlated with total dissolved solids (TDS), and appeared to be related to the intensity of water-rock interaction processes. The elements Al, Fe, Mn, Pb, Zn, Cd, Cu, Co, Ni, Cs, Y, REE and Th were not related to TDS and/or major ions; they showed higher concentrations under high flow conditions; marked differences occurred between total and dissolved amounts; much lower concentrations were generally observed in the water filtered through 0.015 μm than in the water filtered through 0.4 μm, especially when sampling was carried out after heavy rain events that enhanced the load of solid matter in the water. These observations indicate an aqueous transport via sorption processes on very fine particles, such as Fe-oxide/hydroxide and clay mineral particles, which have been inferred by SEM-EDX analyses of the matter retained on the filters.Results on modelled aqueous speciation of trace elements, aimed to assess the role of sorption processes, suggest that: (i) those elements speciated as oxyanions (e.g. As(V), Mo(VI)) or as negatively charged complexes (e.g. UO 2 ( CO 3 ) 3 4 - and UO 2 ( CO 3 ) 2 2 - ) are either poorly or not adsorbed onto oxide/hydroxide minerals under neutral to alkaline conditions as observed in the studied waters; (ii) dissolved concentrations of metals prevalently speciated as free, positively charged ions (e.g. Zn, Cd, Ni) are strongly affected by the amount of Fe-oxide/hydroxide particles retained on 0.015 μm filters; (iii) trace metals occurring as neutral carbonate complexes (e.g. PbCO 3 ° and CuCO 3 ° ) are affected (Pb) or not affected (Cu) depending on the extent of adsorptive affinity of these aqueous species to form strong ternary complexes to oxide/hydroxide surfaces. The results of this study could be used for understanding the aqueous dispersion of trace elements and the seasonal variability in concentrations, which in turn could allow estimation of the risk of exposure to the toxic and harmful elements.

Geochemistry of four tropical montane watersheds, Central Panama by Russell S. Harmon; W. Berry Lyons; David T. Long; Fred L. Ogden; Helena Mitasova; Christopher B. Gardner; Kathleen A. Welch; Rebecca A. Witherow (624-640).
The major element chemistry was determined for surface waters from four watersheds in Central Panama during the 2005 dry season to ascertain geochemical patterns resulting from differing geology and human influences as well to estimate chemical denudation rates for this montane region of tropical rain forest. The Upper Rio Chagres (580 km2), Rio Pequini (281 km2) and Rio Cuango (175 km2) watersheds are formed on a geologically mixed terrain that consists of strongly hydrothermally altered andesite and volumetrically subordinate mafic-intermediate volcanic rocks and felsic intrusive lithologies, whereas the Rio Pacora watershed (374 km2) is developed largely on gabbroic and dioritic lithologies. The headwater areas of all four river basins lie in pristine tropical rainforest, with the Rio Cuango, Rio Pequini and Rio Pacora subject to varying degrees of different land uses in their middle to lower reaches. Values of pH for the four watersheds are near neutral to slightly alkaline (7.0–8.5), DO saturation is high (typically >90%) and dissolved solute contents of the rivers and tributary streams are low (SPC = 130 ± 31 μS/cm), documenting the overall pristine quality of the waters in all four basins. Cluster analysis, supported by a comparison of elemental variations, indicates a broad geochemical similarity of rivers and streams in the four watersheds, but also reveals subtle differences that can be attributed to lithologic control rather than anthropogenic influences. Low-order streams in the Pacora watershed have distinctly higher TDS values plus silica and Ca2+ concentrations than those forming in the mixed lithology terrain. Streams and rivers developed on mafic terrain are also slightly more enriched in total dissolved cations (TZ+) and HCO 3 - , relative to silica, than streams and rivers developed in the mixed lithology terrain. Potassium concentrations are uniformly low, and like Mg2+ and Na+, are similar in both terrains. Calcium/Mg ratios for all watersheds are lower than the world river average, indicating the importance of the weathering of Mg-rich minerals. The Ca/Na, HCO3/Na and Mg/Na ratios for the Rio Pacora streams and rivers fall within the mid-range of what has been observed globally for other streams/rivers draining mafic rocks. The chemical weathering rate calculated is 108 tons/km2/a, which is about 40% of the physical denudation rate for the Upper Rio Chagres watershed of 275–289 tons/km2/a. The results of this study document that both chemical and physical erosion rates within tropical montane watersheds in central Panama are significant in a global context.

Genesis of arsenic/fluoride-enriched soda water: A case study at Datong, northern China by Yanxin Wang; Stepan L. Shvartsev; Chunli Su (641-649).
The high As and F groundwaters from Datong Basin are mostly soda waters with a Na/(Cl+SO4) (meq) ratio greater than unity, As and F up to 1550 μg/L and 10.4 mg/L, respectively, and with pH between 7.6 and 9.1. Geochemical modeling indicates that the waters are oversaturated with respect to calcite and clay minerals such as kaolinite, and undersaturated with respect to primary rock-forming minerals such as anorthite and albite. The water chemistry also is affected by evapotranspiration. The degree of evaporative enrichment is up to 85 in terms of Cl. Results of the hydrogeochemical studies indicate that the occurrence of soda water at Datong is the result of incongruent dissolution of aluminosilicates at one stage of their interaction with groundwater when the water is oversaturated with respect to calcite and evapotranspiration-related salt accumulation is not too strong. Studying the genesis of soda waters provides new insights into mechanism of As and F enrichment in the aquifer system. Due to CaF2 solubility control and OH–F exchange reactions, F can be enriched in soda water. And the high pH condition of soda water favors As desorption from oxyhydroxide surfaces, thereby increasing the concentration of As in the aqueous phase.

Experimental simulation of soil contamination by arsenolite by Ziming Yue; Rona J. Donahoe (650-656).
Column experiments were conducted to experimentally simulate the initial conditions at a field site where legacy As contamination was caused by application of arsenolite as a herbicide to the soil. The experiments were designed to investigate the influence of herbicide loading and carbonate gravel cover on the release of As to, and the As adsorption capacity of, the background soil. The results showed that the As release from the doubly spiked column was three times higher than for the singly spiked columns. The addition of gravel above the herbicide layer halved the leaching time needed for peak As release from the soil. Extrapolation of the experimental data predicted that the columns with carbonate gravel should retain more As (15.6 g for column FWAsG and 12.3 g for column F2AsG) than the column without gravel (0.13 g for column FWAs). The results also indicated the thickness of the arsenolite layer was not important in terms of the As retention capacity of the soil columns. Arsenate was detected and quantified in the effluent solutions by ion chromatography. The data indicated As(III) dominated in the effluents for the first 180 pore volumes (PVs). After 180 PVs, As(III) almost disappeared and As(V) dominated in the effluents. The unexpected high As peak release from the doubly spiked column FW2AsG indicated that solutions having higher As(III) concentrations had a higher capability to dissolve arsenolite, possibly through solute–solute interactions or polymerization. Effluent solution Eh–pH values indicated that As(V) was the thermodynamically stable form of As throughout the experiments. Therefore, As(III) dominance in the initial column effluent solutions was kinetically controlled.

Batch and column experiments were conducted on As adsorption from aqueous solution by natural solids to test the feasibility of these materials to act as adsorbents for As removal from groundwater and drinking water. The solids considered are natural hematite and natural siderite. The As species studied are As(V), As(III) and dimethylarsinic acid (DMA). Arsenic(III), As(V) and DMA were removed to different extents by the solids studied from water solutions containing these three As species, with the highest efficiency for As(V). In aqueous solutions with a mixture of As species, adsorption kinetics depend on the species. On both materials, As(V) was preferentially adsorbed in the batches and first reached equilibrium, followed by DMA and As(III). The As adsorption took place more slowly on natural hematite and natural siderite compared with ferrihydrite. The results demonstrate that the amount of As removed from As(III) batches was greater than that from As(V) batches due to a surface alteration of the solids caused by As(III) oxidation. Although the highest efficiency for As retention was observed on hematite HIO1 in the batch experiments, siderite used as column filling was more efficient in removing As from water containing the As species studied in comparison with hematite. The coating of fresh Fe(III)-oxides was much more intensive in the siderite-packed column than in the hematite-packed column. The combination of siderite and hematite would promote the column filling performance in removing As from aqueous solution.

Partitioning geochemistry of arsenic and antimony, El Tatio Geyser Field, Chile by J.T. Landrum; P.C. Bennett; A.S. Engel; M.A. Alsina; P.A. Pastén; K. Milliken (664-676).
The abundance of As and Sb in aqueous, mineral and biological reservoirs was examined at El Tatio Geyser Field, a unique hydrothermal basin located in the Atacama Desert region of Chile. Here the concentration of total As and Sb in hydrothermal springs and discharge streams are the highest reported for a natural surface water, and the geyser basin represents a significant source of toxic elements for downstream users across Region II, Chile. The geyser waters are near neutral Na:Cl type with ∼0.45 and 0.021 mmol L−1 total As and Sb, respectively, primarily in the reduced (III) redox state at the discharge with progressive oxidation downstream. The ferric oxyhydroxides associated with the microbial mats and some mineral precipitates accumulate substantial As that was identified as arsenate by XAS analysis (>10 wt% in the mats). This As is easily mobilized by anion exchange or mild dissolution of the HFO, and the ubiquitous microbial mats represent a significant reservoir of As in this system. Antimony, in contrast, is not associated with the mineral ferric oxides or the biomats, but is substantially enriched in the silica matrix of the geyserite precipitates, up to 2 wt% as Sb2O3. Understanding the mobility and partitioning behavior of these metalloids is critical for understanding their eventual impact on regional water management.

Reduction of chromate by granular iron in the presence of dissolved CaCO3 by Lai Gui; YanQi Yang; Sung-Wook Jeen; Robert W. Gillham; David W. Blowes (677-686).
Column experiments were conducted to determine the effects of dissolved CaCO3 on the reactivity of Fe towards Cr(VI) reduction and hydraulic conductivity. To provide mechanistic descriptions of reaction progress, Fe corrosion potential was measured continuously, and surface film composition was determined by Raman spectroscopy. Results showed that in the absence of CaCO3, Cr(VI) was removed primarily by precipitation of chromite (Fe n Cr m O4 spinel) and Fe(III)–Cr(III) mixed (oxy)hydroxides. However, the precipitates caused positive shifts in Fe corrosion potential. The reactivity of Fe towards Cr(VI) reduction progressively decreased, resulting in a gradual migration of the Cr removal front. In the presence of CaCO3, the increase in pH due to Fe corrosion and Cr(VI) reduction was buffered, and relatively low corrosion potentials were maintained; therefore, the reactivity of the Fe towards Cr(VI) reduction was not significantly affected. A multi-component reactive transport model that incorporates reactivity loss of Fe due to accumulation of various mineral precipitates reasonably reproduced the experimental data. The simulation results suggest that the migration of Cr(VI) profiles was governed by the formation of Fe(III)–Cr(III) precipitates rather than carbonate minerals. The formation of carbonate minerals, however, significantly affected other geochemical parameters and caused porosity changes. In addition, the model was used to estimate the longevity of Fe PRBs in the absence and in the presence of dissolved CaCO3. The simulations showed that Cr(VI) can be treated effectively for substantial periods of time. However, the loss of porosity may compromise the long-term performance in situations where dissolved CaCO3 concentrations and Fe corrosion rates are high.

An innovative setup of a permeable reactive barrier (PRB) was installed in Willisau, Switzerland to remediate chromate contaminated groundwater. Instead of a conventional continuous barrier, this PRB consists of cylinders installed in rows: a single row for lower expected CrVI-concentrations and an offset double row for higher expected CrVI-concentrations. The cylinders are filled with reactive grey cast-Fe shavings mixed with gravel to prevent extensive precipitation of secondary phases in the pore space. The treatment of the contaminants takes place both within the cylinders and in the dissolved FeII plume generated downstream of the barrier. Monitoring of the contamination situation over a period of 3 a provided evidence of the mobilization, transport and behavior of the contaminants in the aquifer. Groundwater and reactive material were sampled upstream, within and downstream of the barrier by a Multi-Port Sampling System (MPSS) that revealed the geochemical processes as a function of time and space. Comprehensive chemical analyses included sensitive parameters such as CrVI, FeII/FeIII, redox potential, dissolved O2 and pH. Several campaigns using multiple optical tracers revealed a rather complex hydrological regime at different scales, thereby complicating the barrier performance.Results from the large 3D hydrogeochemical dataset show that the double row of cylinders successfully treated the chromate contamination. Remediation by the single row was not effective enough due to insufficient lateral overlap of the cylinders and their FeII-plumes. The low amount of precipitated secondary phases observed in the pore space of the reactive material reduced the risk of clogging the system and suggested a favorable longevity of the barrier. Limiting factors for the long-term operation are inferred to be the availability and accessibility of FeII within the cylinders and the concentration within the generated FeII-plume.

Hydrochemical characteristics and seasonal influence on the pollution by acid mine drainage in the Odiel river Basin (SW Spain) by Aguasanta M. Sarmiento; José Miguel Nieto; Manuel Olías; Carlos R. Cánovas (697-714).
The Odiel river Basin is heavily affected by acid mine drainage (AMD) from the sulphide mining areas in the Iberian Pyrite Belt (IPB). A thorough study has been conducted along this fluvial system, monitoring the seasonal influence on the pollution level and its hydrochemical characteristics. From 2002 to 2006, surface water samples were collected at 91 different points throughout the Odiel river Basin and analyzed by field and laboratory methods for dissolved metals and metalloids. Acid mine drainage affects 37% of the length of the drainage network, which shows a great diversity of geochemical conditions as well as significant variations through the hydrological year. Unaffected streams show different water types depending on the lithological substrate and the marine aerosol influence. Mean concentrations in the contaminated streams are very high: 231 mg/L of Fe, 135 mg/L of Al, 56 mg/L of Zn, 16 mg/L of Cu, etc. Four types of contaminated streams were recognized based on hydrochemical and physicochemical characteristics. There are important seasonal variations depending on the precipitation regimen, level of pollution and proximity to the AMD sources. In the more contaminated samples the M/Fe ratio (M = metals other than Fe) decreases during the summer season. Slightly contaminated samples show an inverse evolution as this ratio increases in spring and summer due to substantial Fe precipitation. A recomparison of contaminant loads suggests that the Odiel river Basin (including the Tinto river) accounts for 15% of the global gross flux of dissolved Zn and 3% of the global gross flux of dissolved Cu transported by rivers into the ocean.

Fracture minerals calcite, pyrite, gypsum, barite and quartz, formed during several events have been analysed for δ 13C, δ 18O, δ 34S, 87Sr/86Sr, trace element chemistry and fluid inclusions in order to gain knowledge of the paleohydrogeological evolution of the Simpevarp area, south-eastern Sweden. This area is dominated by Proterozoic crystalline rocks and is currently being investigated by the Swedish Nuclear Fuel and Waste Management Co. (SKB) in order to find a suitable location for a deep-seated repository for spent nuclear fuel. Knowledge of the paleohydrogeological evolution is essential to understand the stability or evolution of the groundwater system over a time scale relevant to the performance assessment for a spent nuclear fuel repository. The ages of the minerals analysed range from the Proterozoic to possibly the Quaternary. The Proterozoic calcite and pyrite show inorganic and hydrothermal-magmatic stable isotope signatures and were probably formed during a long time period as indicated by the large span in temperatures (c. 200–360 °C) and salinities (0–24 wt.% eq. CaCl2), obtained from fluid inclusion analyses. The Paleozoic minerals were formed from organically influenced brine-type fluids at temperatures of 80–145 °C. The isotopic results indicate that low temperature calcite and pyrite may have formed during different events ranging in time possibly from the end of the Paleozoic until the Quaternary. Formation conditions ranging from fresh to brackish and saline waters have been distinguished based on calcite crystal morphologies. The combination of δ 18O and crystal morphologies show that the fresh–saline water interface has changed considerably over time, and water similar to the present meteoric water and brackish seawater at the site, have most probably earlier been residing in the bedrock. Organic influence and closed system in situ microbial activity causing disequilibrium are indicated by extremely low δ 13C (down to −99.7‰), extreme variation in δ 34S (−42.5‰ to +60.8‰) and trace element compositions. The frequency of calcite low in δ 13C and high in Mn, as well as pyrite with biogenically modified δ 34S decreases with depth. Strontium isotopes have been useful to separate the different generations and the Sr isotope ratios in the groundwaters have been determined mainly by in situ water–rock interaction processes. The difficulty of separating late Paleozoic calcite from possibly recent calcite, and the fact that these calcites are usually found in the same fracture systems indicate that water conducting structures have been intermittently conductive from the Paleozoic and onwards. The methodology used has been successful in separating the different generations and characterising their formation conditions.

Fossil bones and teeth from terrestrial environments encode unique rare earth and trace element (REE and TE) signatures as a function of redox conditions, pH, concentrations of complexing ligands, and water-colloid interactions. This signature is set early in the fossilization process and serves as a paleoenvironmental and paleoclimatic proxy. These signatures can also be used to interpret temporal and spatial averaging within vertebrate accumulations, and can help relocate displaced fossil bones back into stratigraphic context. Rare earth elements in vertebrate fossils from upper Eocene and Oligocene strata of Toadstool Geologic Park, northwestern Nebraska, record mixing and evolution of Paleogene vadose or groundwaters and variations in paleoenvironments. REE signatures indicate that HREE-enriched alkaline groundwater reacted with LREE- and MREE-enriched sediments to produce 3-component mixtures. REE signatures become increasingly LREE- and MREE-enriched toward the top of the studied section as the paleoenvironment became cooler and drier, suggesting that REE signatures may be climate proxies. Time series analysis suggests that REE ratios are influenced by cycles of ca. 1050, 800, 570, 440, and 225 ka, similar to some previously determined Milankovitch astronomical and climate periodicities.