Rona J. Donahoe

Groundwater chemistry and water-rock interactions at Stripa

Groundwaters from near surface to a depth of 1232 m in the Stripa granite have been sampled and analyzed for major and trace constituents. The groundwater composition consists of two general types: a typical recharge water of Ca-HCO3 type ( 700 m depth) of high pH (8–10) that reaches a maximum of 1250 mg/L in total dissolved solids (TDS). Intermediate depths show mixtures of the two types that are highly fracture-dependent rather than depth-dependent. Any borehole can vary significantly and erratically in TDS for either a horizontal or vertical direction. The general transition from Ca-HCO3 type to Na-Ca-Cl type correlates with the depth profile for hydraulic conductivity that drops from 10−8 m/s to 10−11 m/s or lower. Thermomechanical stress (from heater experiments) clearly shows an effect on the groundwater composition that could be caused by changing flow paths, leakage of fluid inclusions or both. Dissolution and precipitation of calcite, fluorite and barite, aluminosilicate hydrolysis, and addition of a saline source (possibly fluid inclusion leakage) play the major roles in defining the groundwater composition. The low permeability of the Stripa granite has produced a groundwater composition that appears intermediate between the dilute, shallow groundwaters typical of recharge in a crystalline rock terrain and the saline waters and brines typical of cratonic shield areas at depth.

Fluid inclusions in the Stripa granite and their possible influence on the groundwater chemistry

Fluid inclusions in quartz and calcite of the Proterozoic Stripa granite, central Sweden, demonstrate that the rock and its fracture fillings have a complex evolutionary history. The majority of inclusions indicate formation during a hydrothermal stage following emplacement of the Stripa pluton. Total salinities of quartz inclusions range from 0–18 eq.wt% NaCl for unfractured rock and from 0–10 eq.wt% for fractured rock. Vein calcites contain up to 25 eq.wt% NaCl but the inclusion size is larger and the population density is lower. Homogenization temperatures are 100–150°C for unfractured rock and 100–250° for fractured rock. Pressure corrections, assuming immediate post-emplacement conditions of 2 kbar, give temperatures about 160°C higher. Measurements of fluid-inclusion population-densities in quartz range from about 108 inclusions/cm3 in grain quartz to 109 inclusions/cm3 in vein quartz. Residual porosity from inclusion densities has been estimated to be at least 1% which is two orders of magnitude greater than the flow porosity. Breakage and leaching of fluid inclusions is proposed as an hypothesis for the origin of major solutes (Na-Ca-Cl) in the groundwater. Evidence for the hypothesis is based on (1) mass balance—only a small fraction of the inclusions need to leak to account for salt concentrations in the groundwater, (2) chemical signatures—BrCl ratios of fluid inclusion leachates (0.0101) match those ratios for the deep groundwaters (0.0107), (3) leakage mechanisms—micro-stresses from isostatic rebound or mining activities acting on irregular-shaped inclusions could cause breakage and provide connection with the flow porosity, and (4) experimental studies—water forced through low permeability granites leach significant quantities of salt. This hypothesis is consistent with the available data although alternate hypotheses cannot be excluded.

Controlling processes in a CaCO3 precipitating stream in Huanglong Natural Scenic District, Sichuan, China

Abstract Huanglong Scenic District is well known for its unusual and diversified landforms such as travertine pools, travertine falls and travertine flows. These landforms, resulting from high-altitude surface cold-water CaCO 3 precipitation, were chosen by UNESCO in 1994 as an entry in The Worlds Nature Heritage. Huanglong is a pristine region where there are limited human activities. Water analyses and thin section (glass slide) precipitation experiments were conducted to determine the aqueous processes controlling CaCO 3 precipitation and travertine landform formation. Results from the travertine flow indicate that the concentrations of HCO 3 − , Ca +2 , and H + decrease regularly along the flow paths. Chemical equilibrium modeling results demonstrate the importance of CO 2 out-gassing and CaCO 3 precipitation processes. CO 2 out-gassing and CaCO 3 precipitation increase with increasing flow velocities. In the pool area, varying hydrodynamics are the primary factors which determine the extent of processes such as advection and diffusion, and hence also control CaCO 3 precipitation and CO 2 out-gassing. When the pool water circulation is very slow, the pH of water flowing over the travertine dams increases significantly (approximately 0.15 pH units) downstream. When the circulation is relatively fast, the pH of stream water initially decreases followed by an increase of approximately 0.21 pH units as it flows past the travertine pool dams. In both cases, the pH rise is caused by sudden changes in the hydrodynamics of the pools, despite the different initial flow conditions. Pool development is a consequence of spatial variations in pH which provide different conditions for CaCO 3 precipitation inside the travertine dam, where less precipitation or even dissolution occurs, compared to conditions at the top and downstream side of the dams. Precipitation experiments demonstrate that the top and downstream side of travertine dams are the locations of the most active precipitation, particularly for pools having faster circulation. Precipitation experiments also reveal that vaterite, a rare polymorph of CaCO 3 , co-precipitates with calcite in milky opalescent water near the upstream input portion of the pool groups. Thin sections covered by algae at the bottom of pools have 40% less CaCO 3 precipitation than those not covered by algae. SEM photographs of the surface of natural travertine deposits show that biofilms with diatom minimize CaCO 3 precipitation and that diatom-adhered calcite surfaces show signs of etching, suggesting that calcite dissolution may be aided by diatoms.

An experimental study of heavy metal attenuation and mobility in sandy loam soils

Abstract Column flow-through experiments reacting wastewater solutions with sandy loam soil samples were performed to study heavy metal attenuation by two soils with different physical and chemical properties. Reacted soil columns were leached with synthetic acid rain to study the mobility of attenuated heavy metals under leaching conditions. This study demonstrates that cation exchange, surface adsorption, chelation with solid organic material, and precipitation were the important attenuation mechanisms for the heavy metals (Cd, Cr, Cu, Mo, Ph, and Zn). Adsorption on soil hydrous oxide surfaces was the primary attenuation mechanism for Cd and Zn in both soils, and for Cu in a soil with low organic matter content. Wastewater solution pH is also an important factor that influences the retention of heavy metals. Cadmium, Cu, Cr, and Zn became mobile after prolonged application of spiked wastewater solution, either through saturation of soil adsorption sites or due to decreasing pH. Only Cr, Pb, and Mo, which are attenuated primarily through precipitation, show significant net retention by soil. Acid rain water removed heavy metals left in the column residual pore solution and weakly sorbed heavy metals in the soils, and has the ability to mobilize some strongly attenuated heavy metals, especially when the soil organic matter content is high. The results have important applications in predicting heavy metal mobility in contaminated soil, the disposal of acid mine drainage, and assessing the risks of landfall leachate leakage.

An experimental study on the process of zeolite formation

Results are reported of an experimental study which examined the effect of solution composition on the composition, structure, and crystallization path of phillipsite and merlinoite in the system Na2OK2O-Al2O3-SiO2H2O at 80°C and pH = 13.34–13.71. At a fixed 3.5 M total dissolved silica concentration, zeolite Si/Al ratio was found to be a linear function of pH within the pH range of the experiments.29Si NMR spectra of the initial solutions show that pH determines the distribution of aqueous aluminosilicate species and, as a result, the precipitated zeolite Si/Al ratio. SEM observations reveal that zeolites may precipitate with or without the presence of an intermediary gel phase, depending on solution composition. The growth rate of the zeolites was found to be dependent upon solution pH and total dissolved aluminum concentration. These observations are discussed in terms of their possible applications to natural zeolite paragenesis and serve to delineate the framework of a comprehensive theory for the mechanism of zeolite crystallization from highly alkaline solutions.

The environmental fate of arsenic in surface soil contaminated by historical herbicide application.

Soils from many industrial sites are contaminated with arsenic because of the historical application of herbicide containing arsenic trioxide. The strong affinity of aqueous arsenic species for soil components has led to the retention of significant amounts of arsenic in surface soils decades after the original source application. Soil collected from a site which received a one-time surficial application of arsenical herbicide in the 1950s was investigated to understand the fate of arsenic under natural leaching conditions. Sequential chemical extraction of the contaminated soil revealed that the majority of the arsenic is in its secondary form. The synthetic acid rain leaching of arsenic from the weathered soil can be divided into two distinct stages. During the first stage, the leachate arsenic concentration underwent a rapid decline which suggests an equilibrium-controlled release event. The second leaching stage was marked by a slow, steady release of arsenic, a signature of a kinetically controlled process. A mathematical approach was employed to identify and describe the two distinct arsenic releasing processes (equilibrium desorption and kinetic desorption). This model considers both desorption processes simultaneously and produces leachate arsenic concentrations in good agreement with the measured data. According to the modeling results, 20% of the arsenic remaining in the soil resides in the herbicide source material after five decades of natural leaching; 25% exists on reversible adsorption sites and 55% is present on irreversible adsorption sites.

CO2 concentrations in perennially ice-covered lakes of Taylor Valley, Antarctica

Lakes in Taylor Valley, southern Victoria Land,Antarctica, are unusual in that they areperennially covered by a 3–5 m thick ice layer.Previous work on gas concentrations in theselakes has shown that the surface waters aresupersaturated with respect to O2,N2O, as well as the noble gases. Our datashow that the dissolved CO2 (CO2(aq))concentrations, calculated from pH andΣCO2, can be highly undersaturatedat shallow depths of the lakes. CO2partial pressure values (pCO2) are as lowas 10−4.3 atm and 10−4.2 atm in theeast and west lobes of Lake Bonney,respectively, and 10−3.8 atm in LakeHoare. CO2(aq) depletion occurred only inthe uppermost part of the water column, inassociation with elevated primary productivity(PPR). The upward diffusion of CO2(aq)from the aphotic zone, and the annual input ofCO2 via glacial meltwater can notreplenish the amount of CO2(aq) annuallylost to primary productivity in the uppermostmeters of the water column. Calcification is alimited source of CO2(aq), since the lakesare undersaturated with respect to calcitethrough portions of the austral summer.Preliminary respiration rates have been used toobtain an annual inorganic carbon balance.Further down in the water column, at the sitesof the deep-water maximum in primary production(PPRmax), which in Lakes Bonney andFryxell is associated with nutrient gradients,CO2(aq) is not undersaturated. A largeupward flux from CO2-supersaturatedaphotic waters provides a surplus ofCO2(aq) at the PPRmax. Lake Fryxell,unlike the other lakes, is supersaturated withCO2(aq) throughout the entire water column.

Attenuation of trace elements in coal fly ash leachates by surfactant-modified zeolite.

Potential leaching of trace elements from older, unlined fly ash disposal facilities is a serious threat to groundwater and surface water contamination. Therefore, effective methods for containing the pollutant elements within the unlined coal combustion products (CCPs) disposal facilities are required to minimize any potential impact of leachate emanating from such facilities into the nearby environment. Because surfactant-modified zeolite (SMZ) has the potential to sequester both cationic and anionic trace elements from aqueous solutions, bench-scale batch and column experiments were performed to test its ability to remediate trace elements in leachates generated from both alkaline and acidic fly ash samples. Fly ash leachate treatment results showed the potential application of SMZ as an effective permeable reactive barrier (PRB) material to control the dispersion of heavy metals and metalloids from ash disposal sites. Quantitative comparison of the elemental composition of SMZ-treated and untreated leachates indicated that SMZ was effective in decreasing the concentrations of trace elements in fly ash leachates. Similarly, SMZ treatment column experiments showed the delayed peak leaching events and overall reductions in leachate concentrations of trace elements. The effectiveness of SMZ column treatments, however, decreased with time potentially due to the saturation of sorption sites.

Modeling arsenic desorption from herbicide-contaminated soils.

The application of arsenical herbicides has created legacy environmental problems by contaminating soil in some agricultural areas and at various industrial sites. Numerous previous studies have suggested that the adsorption of arsenic by common soil components is largely controlled by kinetic factors. Four arsenic-contaminated soil samples collected from industrial sites were characterized and subjected to sequential leaching using a synthetic acid rain solution in order to study the release of arsenic. A dual-site numerical sorption-desorption model was constructed that describes arsenic desorption from these soils in terms of two different release mechanisms: Release from type I (equilibrium) and type II (kinetic) sorption sites. Arsenic held on both type I and II sorption sites is accessible through extensive acid rain leaching. Arsenic desorption from these sites follows a linear Kd model; the manner of approaching the Kd model, however, differs. Arsenic desorption from type I sites reached equilibrium with the aqueous phase under the physical environment provided by the experiment (shaking for 24 h at 25 degrees C), while desorption from type II sites followed a first-order kinetic pattern when approaching equilibrium. During synthetic acid rain sequential leaching of the soils, type I sites released their sorbed arsenic rapidly and subsequent desorption was dominated by the kinetic release of arsenic from type II sites. This shift in desorption mechanism dominance generated data corresponding to two intersecting straight lines in the n-logC dimension for all four soils. The dual-site desorption model was solved analytically and proven to be successful in simulating sorption processes where two different mechanisms are simultaneously controlling the aqueous concentration of a trace element.

Modeling Arsenic Sorption in the Subsurface with a Dual‐Site Model

Arsenic is a well-known groundwater contaminant that causes toxicological and carcinogenic effects in humans. Predicting the transport of arsenic in the subsurface is often problematic because of its complex sorption characteristics. Numerous researchers have reported that arsenic sorption on soil material is initially fast and then subsequently slow. A dual-site numerical sorption model was previously developed to describe arsenic desorption from arsenic-contaminated soils in batch experiments in terms of two different release mechanisms. Experiments involving synthetic acid rain leaching of four arsenic-contaminated soil columns were performed to verify the dual-site numerical sorption model in the context of one-dimensional vertical transport. The fitted models successfully simulated the signature long tailings and the two-stage arsenic leaching patterns for all four soil columns. The dual-site sorption model was incorporated within the general solute transport simulation code Modular Three-Dimensional Multispecies (MT3DMS), version 5.10. The resulting version was named MT3DDS and is available for public access. This experimental study has shown that MT3DDS is capable of simulating phase redistribution during transport, and thus provides a new numerical tool for simulating arsenic transport in the subsurface.