Applied Geochemistry (v.19, #10)

Modeling diffusion of an alkaline plume in a clay barrier by Eric C Gaucher; Philippe Blanc; Jean-Michel Matray; Nicolas Michau (1505-1515).
The design of clay plugs used for sealing access galleries to a radioactive waste repository built with concrete structures in a deep clayey formation must take into consideration their chemical evolution over time. Diffusion of an alkaline plume from concrete into bentonite was therefore modeled over a 100 ka period with the PHREEQC geochemical code in order to determine, as a function of time, modifications to mineral surfaces, dissolution of existing minerals and precipitation of new mineral phases. The modeled system consisted of an OPC (Ordinary Portland Cement) barrier, a MX80 bentonite clay barrier and the corresponding equilibrated pore waters. A specific database including aqueous complexes, mineral-phase solubilities and ion-exchange parameters for Na+, K+, Ca2+, Mg2+ and H+ for an MX80 bentonite was created. Only the mineral phases capable of precipitating in this system were considered in the model. The width of the clay barrier was taken as 8 m. Simulations were carried out at 25 °C and 1 bar. Transport modeling was based on a one-dimensional diffusion model, assuming thermodynamic equilibrium. Constant dissolved concentrations were assumed for the concrete pore fluid. The clay barrier was modeled as a semi-infinite medium with a single diffusion coefficient of 10−11 m2/s. The simulation revealed a sequence of mineralogical transformations after 100 ka. From the host clayey rock to the concrete, the transformations begin with ion exchange reactions changing Na-montmorillonite into a more potassic and calcic phase. Then illitization of the montmorillonite occurs. Between the illitized zone and the concrete interface, zeolite phases are precipitated. Finally, cement phases replace zeolites at the concrete interface. The cementation of the concrete interface leads to a large increase in the total clay volume whereas illitization of montmorillonite produces a decrease in this volume. The sensitivity of the calculations to exchange reactions and the diffusion coefficient was tested. Calculations were done linking and not linking the cation exchange capacity to the amount of montmorillonite in order to evaluate the influence of the exchange reactions. This produced only very minor differences, indicating a limited influence of exchange reactions in the mineralogical evolution. Simulations for sensitivity to the value of the diffusion coefficient enabled the authors to develop a phenomenological law indicating that the extent of the mineral transformations is proportional to the square root of the diffusion time and the diffusion coefficient. The simulations also demonstrate the efficiency of pH buffering by a mineralogical assemblage that controls the CO2 partial pressure.

An attempt to identify the hydrogeochemical processes that accompany current or past intrusion of seawater in the Castell de Ferro coastal aquifer (SE Spain) was made using ionic deltas, ionic ratios and saturation indices. Alluvial and coastal Quaternary sediments and Triassic marbles constitute this complex aquifer. Cation exchange, dolomitization, and calcite dissolution were identified as operating in the aquifer. The cation exchange processes in the detrital materials are different from those in the marbles. In addition, it was determined that while the processes of marine intrusion associated with periods of low groundwater levels still persist in the detrital aquifer, in the carbonate sector the flushing process characteristic of wetter conditions has begun.

Protective mineral layers formed from granite–dry steam interaction by N Yamasaki; I.R Korablova; S.F Korablov (1529-1535).
This work focuses on interaction between granite–stainless steel (SUS) pipe–dry steam in the presence of Cu. Three types of hydrothermal experiments were conducted: (1) SUS–Cu–granite, (2) SUS–Cu, (3) SUS–granite. It was found that a high protective amorphous AlSi1,6O4 layer (thickness about 5 μm) was formed on the supporting pipe surface only in the case of SUS–Cu–granite interaction. The Al silicate layer formed during the experiment was characterized by X-ray diffraction (XRD) techniques and scanning electron microscopy (SEM) with EDX. According to kinetic data this layer has high protective properties.

Laboratory simulation of flocculation processes in a flooded tailings impoundment at the Kristineberg Zn–Cu mine, northern Sweden by Anders Widerlund; Elena Shcherbakova; Jerry Forsberg; Henning Holmström; Björn Öhlander (1537-1551).
A laboratory mixing experiment was performed to simulate the flocculation processes that can be expected when natural stream water mixes with the saline water in flooded tailings impoundments. Mixing plots where dissolved (<0.22 μm) Ca, Mg, Na, K, S and Si were plotted vs. a conservative mixing index revealed a conservative mixing behaviour for these elements. Similar plots for dissolved Fe, Mn, Al and total organic C (TOC) showed that dissolved Fe and Al flocculated within 24 h after the mixing of the waters. Dissolved Mn was removed from solution 13–98 days after mixing, presumably due to the oxidation of Mn2+. Removal of TOC could not be detected in the mixing bottles. However, the flocculants that settled in the mixing bottles (1.7–3.4 mg flocs L−1 of stream water) contained 18 wt% C and 14–19 wt% acid-leacheable Fe. Organic C and Fe oxyhydroxides thus appear to form substantial fractions of the flocculants. If stream water is diverted through tailings ponds as part of a remediation programme, flocculation processes are likely to contribute to natural sedimentation in the pond. However, the relative contribution from flocculated particulate matter may be significant only when the stream-water transport of suspended matter into the impoundment is low (<10–15 mg L−1). Trace metal uptake in the flocculants that settled in the mixing bottles resulted in removal of Cd (0.024–0.028 μg L−1), Co (0.15–0.17 μg L−1), Cu (1.8–3.5 μg L−1) and Zn (15–29 μg L−1) from the dissolved phase. Relative to the dissolved trace metal concentrations in the tailings pond water used in the experiment (Cd=0.435–0.438 μg L−1; Co=0.738–0.763 μg L−1; Cu=3.16–5.05 μg L−1; Zn=26.6–32.6 μg L−1), the trace metal uptake exceeded 50% of these concentrations only for Cu and Zn.

Recent research has shown that phytoextraction approaches often require soil amendments, such as the application of EDTA, to increase the bioavailability of heavy metals in soils. However, EDTA and EDTA–heavy metal complexes can be toxic to plants and soil microorganisms and may leach into groundwater, causing further environmental pollution. In the present study, vetiver grass (Vetiveria zizanioides) was studied for its potential use in the phytoremediation of soils contaminated with heavy metals. In the pot experiment, the uptake and transport of Pb by vetiver from Pb-contaminated soils under EDTA application was investigated. The results showed that vetiver had the capacity to tolerate high Pb concentrations in soils. With the application of EDTA, the translocation ratio of Pb from vetiver roots to shoots was significantly increased. On the 14th day after 5.0 mmol EDTA kg−1 of soil application, the shoot Pb concentration reached 42, 160, 243 mg kg−1 DW and the root Pb concentrations were 266, 951, and 2280 mg kg−1 DW in the 500, 2500 and 5000 mg Pb kg−1 soils, respectively. In the short soil leaching column (9.0-cm diameter, 20-cm height) experiment, about 3.7%, 15.6%, 14.3% and 22.2% of the soil Pb, Cu, Zn and Cd were leached from the artificially contaminated soil profile after 5.0 mmol EDTA kg−1 of soil application and nearly 126 mm of rainfall irrigation. In the long soil leaching experiment, soil columns (9.0-cm diameter, 60-cm height) were packed with uncontaminated soils (mimicking the subsoil under contaminated upper layers) and planted with vetiver. Heavy metal leachate from the short column experiment was applied to the surface of the long soil column, the artificial rainwater was percolated, and the final leachate was collected at the bottom of the soil columns. The results showed that soil matrix with planted vetiver, could re-adsorb 98%, 54%, 41%, and 88% of the initially applied Pb, Cu, Zn, and Cd, respectively, which may reduce the risk of heavy metals flowing downwards and entering the groundwater.

Lead concentrations and isotopic compositions measured along 80 km of the Debed River in the Republic of Armenia provide new insights into the geochemical and physical controls on riparian Pb transport by allowing comparison of the long-range, downstream impacts of acid mine drainage with runoff from an industrialized city. The modern background Pb concentration in Armenian surface waters is estimated to be ∼0.01 μg/L, based on analyses of remote alpine rivers in Armenia. The lead concentration in the Debed River is 8 μg/L (800 times background) after passing through Vanadzor, the second largest industrial city in Armenia; it then decreases to 1 μg/L before the Debed River flows into the Alaverdi mining district. There, the Debed River receives waters from two mining drainage streams with Pb concentrations >3000 μg/L, but those concentrations decrease 3 orders of magnitude to ∼3 μg/L by the time the river exits Armenia and flows into the Republic of Georgia.Isotope mixing plots show shifts in Pb isotope composition as the river flows out of Vanadzor, evidencing the mixture of an average terrestrial Pb composition (206Pb/207Pb ≈ 1.17; 208Pb/207Pb ≈ 2.45) with past leaded gasoline and other industrial Pb emissions retained in the river's sediments within that region (208Pb/207Pb ⩽ 2.45). The isotopic composition again shifts (e.g., 208Pb/207Pb ⩾ 2.46) as the river passes through the Alaverdi mining district, where isotopic ratios in the water are characteristic of Pb in the area's massive sulfide deposits.Modeling both downstream elemental concentrations and Pb isotopic compositions further resolves the physical and chemical behavior of the contaminants in the river system. A multi-element model of concentration gradients in the acid mine drainage streams indicates Pb is attenuated by Al(OH)3 precipitation (54% of the loss) and by adsorption onto other particles settling out of the water column (46% of the loss). Modeling of Pb transport in the Debed River indicates the natural outflow of Pb is ∼10 kg/a, while the contributions from industrial runoff and acid mine drainage are each ∼103 kg/a, two orders of magnitude greater than the natural flux. Thus, the total annual flux of Pb from the Debed River as it leaves Armenia and flows into Georgia is calculated to be 2500 kg/a. This value is consistent with the average annual Pb flux of 2360 kg/a estimated for the last 20 a. The predicted effects of local climate change over the next 100 a may reduce annual Pb flux by 10–15%, slowing the removal of contaminant Pb and maintaining the two order of magnitude increase over the natural flux.

Association of dissolved radionuclides released by the Chernobyl accident with colloidal materials in surface water by Takeshi Matsunaga; Seiya Nagao; Takashi Ueno; Seiji Takeda; Hikaru Amano; Yu. Tkachenko (1581-1599).
The association of dissolved 90Sr, 239,240Pu and 241Am with natural colloids was investigated in surface waters in the Chernobyl nuclear accident area. A 4-step ultrafiltration (UF) study (<1 kilodaltons (Da), 1–10 kDa, 10–100 kDa, 100 kDa<) showed that 49–83% of 239,240Pu and 76% of 241Am are distributed in colloids of the two size fractions larger than 10 kDa (nominal molecular weight limit of the filter, NMWL), while 90Sr was found exclusively (85–88%) in the lowest molecular size fraction below 1 kDa (NMWL) for the Sahan River water at the highly contaminated area close to the Chernobyl Nuclear Power Plant (ChNPP). Consistent results were obtained by 2-step fractionation (larger than and smaller than 10 kDa (NMWL)) for river and lake waters including other locations within about 30 km away from ChNPP. It is likely that Pu and Am isotopes were preferentially associated with dissolved organic matter of high molecular size, as suggested by the fact that (i) only a few inorganic elements (Mg, Ca, Sr, Si, Mn, Al) were found in the colloidal size ranges, and (ii) the positive correlation between dissolved organic C (DOC) concentrations and UV absorbance at 280 nm, a broad absorption peak characteristic of humic substances (HS) was found. A model calculation on the complexation of Pu and Am with HS as an organic ligand suggests that the complexed form could be dominant at a low DOC concentration of 1 mgC L−1, that is commonly encountered as a lower limit in fresh surface water. The present results suggest the general importance of natural organic colloids in dictating the chemical form of actinides in the surface aquatic environment.

The phase distribution of trace metals and oxyanions, including U and As, in 2 surface water bodies was investigated within a South Texas watershed hosting a high density of surface U mine pits and tailings. The objectives of the study were to evaluate the environmental legacy of U mining, with particular emphasis on the spatial and temporal variability of water quality in Lake Corpus Christi, a downstream reservoir that serves as the major water resource to a population of ∼350,000 people in the region. Lyssy Pond, a livestock pond bordered by U mine tailings, was used as a model case-study site to evaluate the cycling of U mine-derived oxyanions under changing redox conditions. Although the pond showed seasonal thermal and chemical stratification, geochemical cycling of metals was limited to Co and Pb, which was correlated with redox cycling of Mn mineral phases, and U, which suggested reductive precipitation in the pond's hypolimnion. Uranium levels, however, were too low to support strong inputs from the tailings into the water column of the pond. The strong relationships observed between particulate Cr, Cs, V, and Fe suggest that these metals are associated with a stable particulate phase (probably allochthonous alumino-silicates) enriched in unreactive Fe. This observation is supported by a parallel relationship in sediments collected across a broad range of sediment depositional processes (and histories) in the basin. Arsenic, though selectively enriched in the pond's water column, was dominated by dissolved species throughout the depth of the profile and showed no sign of geochemical cycling or interaction with Fe-rich particles. Arsenic (and other oxyanions) in the water columns of Lake Corpus Christi and Lyssy pond were not affected by the abundant presence of Fe-rich particles but instead behaved conservatively. No evidence was found of anthropogenic impacts of U mines beyond the purely local scale. Arsenic's presence within the Nueces drainage basin is related to interactions between surface and groundwaters with U- and As-rich geological formations rather than large-scale transport of contaminants downstream of the U mine pits and tailings. A quantitative mass balance model, constructed using monthly hydrological data for the reservoir, provides quantitative evidence of seasonal evaporative concentration of As in surface waters demonstrating the predominance of hydrodynamic over geochemical constraints, on the cycling of this element.

The B-content of coals from sedimentary basins is a function of the primary depositional environment and secondary enrichments. Boron has been used to infer paleo-environments of coal deposits, and environmental impact of coal burning power plants. To improve our understanding of B in organic matter, this study used secondary ion mass spectrometry to measure the B-isotopic composition of 25 coals and three kerogens (Type I, II and III).Results show that coal is 10B-enriched compared to most terrestrial waters. The δ11B values measured show large variability within a single `homogenized' coal sample indicating that B is heterogeneously bound in various organic macerals. Of greater importance is the observation that all coal δ11B values are negative. The lowest δ11B measured was −70‰ representing the lightest B-isotopic ratios ever reported for terrestrial materials. Bulk analyses of coals may not reveal isotope ratios this low because of the averaging of values from different organic macerals having different δ11B values.This discovery may be important for tracing organic contaminants in fluids. When organic matter is heated during burial, the isotopically light B partitions into fluid phases. The very low δ11B of organic products compared to most groundwaters may be useful in tracing fluids derived from organic sources.

Water mass balances on an isolated embayment (Hoople Bay) in the St. Lawrence River revealed a small stream (Hoople Creek), local groundwater and the St. Lawrence Main Channel as the 3 principal water sources. The latter had an average evaporative isotope signal that was inherited from the Great Lakes ( δ 18 O H 2 O =−7.0‰) and an average Cl content of 0.55 mmol/l. Hoople Creek and Hoople Bay waters were more variable in their isotopic composition and Cl contents, while local groundwater was assumed to have a homogeneous composition year around. These parameters constituted an equation system that was solved with matrix operations to yield monthly contributions of the 3 endmembers. Influx of groundwater and Hoople Creek dominated the embayment only after higher snowmelt discharges, while the Main Channel contributed more than 50% during the remainder of the year. Preliminary results suggest that potential pollution in the Main Channel would strongly affect Hoople Bay and similar ecosystems along the river. Nevertheless, more detailed data are needed for a better water balance over longer time periods. The 3-component mixing technique serves as a good tool to evaluate seasonal water fluxes and may also become useful in other mass balances.

Comparison of in situ uranium K D values with a laboratory determined surface complexation model by Gary P Curtis; Patricia Fox; Matthias Kohler; James A Davis (1643-1653).
Reactive solute transport simulations in groundwater require a large number of parameters to describe hydrologic and chemical reaction processes. Appropriate methods for determining chemical reaction parameters required for reactive solute transport simulations are still under investigation. This work compares U(VI) distribution coefficients (i.e. K D values) measured under field conditions with K D values calculated from a surface complexation model developed in the laboratory. Field studies were conducted in an alluvial aquifer at a former U mill tailings site near the town of Naturita, CO, USA, by suspending approximately 10 g samples of Naturita aquifer background sediments (NABS) in 17-5.1-cm diameter wells for periods of 3 to 15 months. Adsorbed U(VI) on these samples was determined by extraction with a pH 9.45 NaHCO3/Na2CO3 solution. In wells where the chemical conditions in groundwater were nearly constant, adsorbed U concentrations for samples taken after 3 months of exposure to groundwater were indistinguishable from samples taken after 15 months. Measured in situ K D values calculated from the measurements of adsorbed and dissolved U(VI) ranged from 0.50 to 10.6 mL/g and the K D values decreased with increasing groundwater alkalinity, consistent with increased formation of soluble U(VI)-carbonate complexes at higher alkalinities. The in situ K D values were compared with K D values predicted from a surface complexation model (SCM) developed under laboratory conditions in a separate study. A good agreement between the predicted and measured in situ K D values was observed. The demonstration that the laboratory derived SCM can predict U(VI) adsorption in the field provides a critical independent test of a submodel used in a reactive transport model.