Eric J. Reardon

The formation and potential importance of cemented layers in inactive sulfide mine tailings

Abstract Investigations of inactive sulfide-rich tailings impoundments at the Heath Steele (New Brunswick) and Waite Amulet (Quebec) minesites have revealed two distinct types of cemented layers or “hardpans.” That at Heath Steele is 10–15 cm thick, occurs 20–30 cm below the depth of active oxidation, is continuous throughout the tailings impoundment, and is characterized by cementation of tailings by gypsum and Fe(II) solid phases, principally melanterite. Hardpan at the Waite Amulet site is only 1–5 cm thick, is laterally discontinuous (10–100 cm), occurs at the depth of active oxidation, and is characterized by cementation of tailings by Fe(III) minerals, principally goethite, lepidocrocite, ferrihydrite, and jarosite. At Heath Steele, an accumulation of gas-phase CO2, of up to 60% of the pore gas, occurs below the hardpan. The calculated diffusivity of the hardpan layer is only about 1 100 that of the overlying, uncemented tailings. The pore-water chemistry at Heath Steele has changed little over a 10-year period, suggesting that the cemented layer restricts the movement of dissolved metals through the tailings and also acts as a zone of metal accumulation. Generation of a cemented layer therefore has significant environmental and economic implications. It is likely that, in sulfide-rich tailings impoundments, the addition of carbonate-rich buffering material during the late stages of tailings deposition would enhance the formation of hardpan layers.

Anaerobic Corrosion of Granular Iron: Measurement and Interpretation of Hydrogen Evolution Rates

Anaerobic corrosion of iron metal produces Fe 2+ , OH - , and H 2 (g). Growing interest in the use of granular iron in groundwater remediation demands accurate corrosion rates to assess impacts on groundwater chemical composition. In this study, corrosion rates are measured by monitoring the hydrogen pressure increase in sealed cells containing iron granules and water. The principal interference is hydrogen entry and entrapment by the iron. The entry rate is described by Sieverts low (R = kP H2 0.5 ), and the rate constant, k, is evaluated by reducing the cell pressure once during a test. For the 10-32 mesh iron used in this study, k initially was 0.015 but decreased to 0.009 mmol kg -1 d -1 kPa -0.5 in 150 d. The corrosion rate in a saline groundwater was 0.7 ± 0.05 mmol of Fe kg -1 d -1 at 25 °C-identical under water-saturated or fully-drained conditions. The rate decreased by 50% in 150 d due to alteration product buildup. The first 40-200 h of a corrosion test are characterized by progressively increasing rates of pressure increase. The time before steady-state rates develop depends on the solution composition. Data from this period should be discarded in calculating corrosion rates. Tests on pure sodium salt solutions at identical equivalent concentrations (0.02 equiv/L) show the following anion effect on corrosion rate : HCO 3 > SO 4 2- > Cl - . For NaCl solutions, corrosion rates decrease from 0.02 to 3.0 m.

Pyrite oxidation in carbonate-buffered solution: 2. Rate control by oxide coatings

Abstract The kinetic behaviour of pyrite oxidation in the laboratory was studied over a period of about 10,000 hours in reactors through which a carbonate-buffered solution and air (20% O2) flowed continuously. Three grain size fractions were monitored. The concentration of sulphate and the mass of the effluent solution were measured periodically to calculate oxidation rates. The results indicate that the rates of reaction decreased significantly with time. The rates initially exhibited an inverse dependence on grain size (within 400 h) then became more linear with the square of the inverse grain size at later times (after 8000 h) suggesting a surface-layer control of the reaction with time. Surface analysis by X-ray Photoelectron Spectroscopy revealed the presence of ferric oxide on the pyrite surfaces, and ion boring with auger electron spectroscopy indicated a layer thickness on the order of 0.6 microns on the 215 micron grains. The data are represented by a shrinking core model which includes the effects of the surface rate constant plus the diffusive resistance to oxygen transfer through the accumulating reaction layer as oxidation proceeds. The three grain sizes (representing different specific surface areas) exhibited consistent estimates of the surface rate constant (Ks = 3.07 × 10−6mh−1 ± 46%) and the diffusion coefficient for oxygen through the oxide layer (Ds = 1.08 × 10−12m2h−1 ± 30%). The estimated thickness of the oxide layer at the end of the experiment agreed well with the measured value. Oxide accumulation on the pyrite surfaces under neutral pH conditions results in a significant reduction in oxidation rates over time. This behaviour has important implications for the reduction of the rate of release of oxidation products, including hydrogen ions, to environments where sulphide mineral wastes are exposed to the atmosphere.

Activation and regeneration of a soil sorbent for defluoridation of drinking water

Geomaterials can be cost-effective sorbents for use in water treatment. In this study, a heavily-weathered Tertiary soil from Xinzhou, China was used as a sorbent for defluoridation of high-fluoride drinking water. The soil is composed of quartz, feldspar, illite and goethite, with an Fe oxide content of 6.75%. Batch and column experiments were done to characterize the F− removal properties and to develop an optimal activation and regeneration procedure. The soil can be regenerated following a simple base-acid rinsing procedure. This can be performed in situ, i.e., by passing the rinsing solutions directly through the treatment column. The same regeneration procedure can be used to activate the pristine soil. Fluoride sorption is described by a Freundlich isotherm model and the bulk of the uptake occurs within 1.5 h. Iron oxide coatings on soil particles and perhaps ≡FeOH surface groups at particle edges of illite grains are likely responsible for the soils F-sorption property. As collected in the field, the soil has a low permeability and is thus unsuitable for direct use in a flow-through column. Heat-treatment at 400–500°C for 2 h, however, produces a granular and permeable sorbent. Although the soils sorption capacity (150 μg/g ) is about a quarter of the low end range of values reported for commercially-available activated alumina, the sorption for F− is specific. A batch sorption experiment in the presence of Cl−, SO42− and HCO3− shows little or no competition from these other anions.

Migration of contaminants in groundwater at a landfill: A case study 6. Hydrogeochemistry

Multilevel point-samplers located on a longitudinal cross-section through the plume of contaminated groundwater in the sand aquifer at the abandoned Borden landfill were used to obtain a suite of small-volume samples for analysis of major ions and of minor and trace-level inorganic constituents. Calcium is the dominant cation and sulphate and bicarbonate are the dominant anions in the contaminant plume, with maximum concentrations of 400, 2000 and 1200 mg l−1, respectively. Beneath the landfill the most highly contaminated water in the aquifer has total dissolved solids of ∼4000 mg l−1. Sulphate and iron are the only inorganic constituents that exceed the recommended limits for drinking water. The plume contains above-background concentrations of dissolved zinc and manganese. No heavy metals or other hazardous inorganic trace elements occur at levels above the maximum limits for drinking water. The plume contains levels of total dissolved organic carbon that are above background levels, ranging from 30 mg l−1 beneath the landfill to 5 mg l−1 near the front of the plume. The landfill, which is almost entirely above the water table, contains high partial pressures of carbon dioxide and methane. The high carbon dioxide levels induce substantial dissolution of calcite, which occurs in sand layers in the landfill, and in the aquifer immediately below the landfill. Calcite dissolution is the origin of much of the dissolved inorganic carbon and some of the calcium in the plume. Most of the sulphate and much of the calcium in the plume appears to be derived from gypsum in construction debris in the landfill. Groundwater beneath the landfill is saturated with respect to gypsum and slightly supersaturated with respect to calcite. Some of the dissolved iron in the plume results from the release under reducing conditions of iron from ferric iron coatings on sand grains located in the aquifer beneath the landfill and in the layers in the refuse. Dissolved iron concentrations appear to be limited by the solubility of siderite. Dissolved sulphide is present at only very low concentrations and is probably controlled by ferrous sulphide solubility. Exchange of cations between the aquifer material and the leachate-contaminated groundwater is an apparent cause of calcium release and subsequent precipitation of calcite in the plume. As much of 80% of the K+, 20% of the Mg2+ and 15% of the Na+ in the plume exists on the exchange sites. The effects of dispersive and geochemical attenuation were evaluated from the ratios of chloride to other species along a central flow path from the landfill to the front of the plume. Platinum electrode measurements and electrochemical potentials calculated from equilibrium speciation of analysed constituents varied over a wide range of values and provided poor agreement among methods for the same sample. Only potentials derived from the iron redox couples and Pt-electrode measurements exhibited fair agreement for any one sample. The rather narrow range of Pt-electrode potential values along the central flow line through the plume suggest that the electrochemical potentials are buffered, apparently as a result of Fe(OH)3(s) /ag FeCO3(s) equilibria.

Problems and approaches to the prediction of the chemical composition in cement/water systems

Abstract The slow diffusion of elements to the porewater from unreacted cores of cement minerals ensures that the solution and solid phases of a cement/water system are in a transient rather than in an equilibrium state. Nevertheless, solution/mineral equilibrium is closely attained soon after the contact of water with cementitious material even though the composition of both phases are dynamically evolving. Chemical equilibrium models can thus have an important role in representing the chemical composition of cement/water systems. There are obstacles, however, in the application of equilibria programs towards this purpose. Although there are kinetic models to predict the release of alkalies from hydrating cement particles, there is no ion sorption or solid solution model to describe the uptake of alkalies from the pore solution by CSH. Because Na and K are the principal cations in cement pore solutions, their concentrations exert the most control on the porewater pH. Since the solubility of the principal hydrated cement minerals are pH dependent, an inability to model alkali levels translates into an inability to model the concentration of most other dissolved constituents. More experimental data are needed to construct a thermodynamic model for alkali association with CSH because the data that are available are not in agreement. In the meantime, an empirical approach to the problem has been offered. Thermodynamic data for the aluminum sulphate minerals, ettringite and monsulphate, indicate that ettringite should be the stable phase in cement porewater solutions, yet monosulphate is the phase generally present in mature cement pastes. The reasons for this are not clear, but it may be that OH − substitution for SO 4 2− in monosulphate at high pHs inverts its thermodynamic stability with respect to ettringite. Solubility data of the C 4 AH 13 -monosulphate solid solution series are needed to address this problem. Equilibria modelling and experimental data of CSH compositions in alkali solutions indicate that CSH in cement porewater solutions should have a Ca/Si ratio of around 1.1, much lower than values usually reported from SEM measurements on actual pastes (avg. 1.6). However, SEM also yields results over a wide range (1.0–2.0) and it may be that the CSH and Ca(OH) 2 reaction products of C 2 S and C 3 S hydration in a cement paste are so intimately intermixed that some amount of Ca(OH) 2 invariably gets included in analyses of CSH under the electron beam. This might explain both the large variation and higher values obtained in SEM measurements on pastes compared to experimental data on CSH/solution equilibrations.

Reaction paths and equilibrium end-points in solid-solution aqueous-solution systems

Equations are presented describing equilibrium in binary solid-solution aqueous-solution (SSAS) systems after a dissolution, precipitation, or recrystallization process, as a function of the composition and relative proportion of the initial phases. Equilibrium phase diagrams incorporating the concept of stoichiometric saturation are used to interpret possible reaction paths and to demonstrate relations between stoichiometric saturation, primary saturation, and thermodynamic equilibrium states. The concept of stoichiometric saturation is found useful in interpreting and putting limits on dissolution pathways, but there currently is no basis for possible application of this concept to the prediction and/ or understanding of precipitation processes. Previously published dissolution experiments for (Ba, Sr)SO4 and (Sr, Ca)CO3orth. solids are interpreted using equilibrium phase diagrams. These studies show that stoichiometric saturation can control, or at least influence, initial congruent dissolution pathways. The results for (Sr, Ca)CO3orth. solids reveal that stoichiometric saturation can also control the initial stages of incongruent dissolution, despite the intrinsic instability of some of the initial solids. In contrast, recrystallisation experiments in the highly soluble KCl-KBr-H2O system demonstrate equilibrium. The excess free energy of mixing calculated for K(Cl, Br) solids is closely modeled by the relation GE = χKBrχKClRT[a0 + a1(2χKBr−1)], where a0 is 1.40 ± 0.02, a1, is −0.08 ± 0.03 at 25°C, and χKBr and χKCl are the mole fractions of KBr and KCl in the solids. The phase diagram constructed using this fit reveals an alyotropic maximum located at χKBr = 0.676 and at a total solubility product, ΣΠ = [K+]([Cl−] + [Br−]) = 15.35.

Fractionation of sulfur and oxygen isotopes in sulfate by soil sorption

Both field and laboratory data indicate that there is no significant isotope fractionation of sulfate during sorption in upland forest Podzols. The dominant sulfate sorption process in these soils is adsorption onto mineral surfaces. In the Plastic Lake watershed, Dorset, Ontario, Canada, fractions of sulfate from Podzol B-horizons have the following mean isotope (%.) compositions: water soluble sulfate, δ34S = +6.4; δ18O = −5.3; bicarbonate-exchanged sulfate by two methods,δ34S = + 4.5 and + 3.4; δ18O =−6.2 and −5.6; dissolved sulfate in B-horizon soilwater seepage,δ34S = + 4.8; δ18O = −5.4. These data indicate that soil sorption enriches dissolved sulfate in 34S by approximately 1 ± 1%. and in 18O by 0 +- 1 %. relative to sorbed sulfate. Similar results were obtained by laboratory sorption of sulfate by prepared goethite, which is a mineral representative of soil sorption sites in acidic Podzols like the one at Plastic Lake. The mean fractionation between sorbed and dissolved sulfate was found to be − 0.3%. for34S and 0.1 %. for 18O. Earlier literature has confused the term adsorption; in many cases the more general term sorption, or retention, should be used. Pronounced fractionation of S and O isotopes in sulfate by lake and ocean sediments has been attributed to “adsorption” or “retention” but is more likely the result of sulfate reduction. Apparently, at Earth-surface conditions the only substantial isotope shifts in sulfate occur during microbial processes.

Modelling chemical equilibria of acid mine-drainage: The FeSO4-H2SO4-H2O system

Chemical equilibria in the FeSO4-H2SO4-H2O system have been modelled using the Pitzer formulation for the determination of activity coefficients over the temperature range from 10 to 60°C. The available experimental activity coefficient, enthalpy, heat capacity and mineral solubility data have been analysed to determine the temperature dependencies of the various ion interaction parameters necessary to describe chemical equilibria in this system. In this analysis, the previously published ion interaction parameters in the H2SO4-H2O system have been reevaluated incorporating the higher-order electrostatic terms in the Pitzer formulation to account for asymmetric mixing of ions of dissimilar valence. The combined set of interaction parameters yield an accurate description of equilibria in the FeSO4-H2O system over the temperature range of 10 to 90°C and concentration conditions up to the solubility limit of ferrous sulphate hydrate. From an appraisal of solubility data for ferrous sulphate hydrates in water using these parameter expressions, the following relations for the temperature dependence of the solubility products for the heptahydrate (melanterite) and the monohydrate (szomolnokite) phases have been derived: log Ksp = 1.447 − 0.004153T − 214949T2Melanterite(FeSO4·7H2O)log Ksp = 250211 − 0.063405T − 641220T2Szomolnokite(FeSO4·7H2O) A comprehensive evaluation of ferrous sulphate hydrate solubility data in sulphuric acid solutions enabled the determination of the various additional parameters needed to describe the mixed system, FeSO4-H2SO4-H2O. The final model yields an excellent representation of all available mineral solubility data for the combined FeSO4-H2SO4-H2O system over a temperature range from 10 to 60°C, sulphuric acid concentrations from 0 to 6 molal and for iron sulphate concentrations up to the solubility limit.

Celestite (SrSO4(s)) solubility in water, seawater and NaCl solution

Celestite solubility measurements have been conducted in pure water at temperatures from 10 to 90°C. Equilibrium was achieved with respect to a crystalline solid phase from both undersaturated and supersaturated solutions. The measurements show that the solubility undergoes a maximum near 20°C. LogK values for the solubility reaction are adequately described by the following expression over the temperature range 283.15 to 363.15 K: −logK= −35.3106+0.00422837T+318312/T2+14.99586 logT. The following thennodynamic values for the dissolution reaction of SrSO4(s), at 25°C have been derived: ΔGR0 = 37852 ± 30 Jmol−1ΔHR0 = −1668±920Jmol−1ΔSR0= −132.6±3.2JK−1mol−1 Celestite solubility measurements were also determined in NaCl solutions up to 5 m concentration and from 10 to 40°C. These data are in good agreement with the work of StrUbel (1966), who reports solubility measurements to temperatures of 100°C. The application of the Pitzer relations and the solubility constants determined in this study to calculate celestite solubility in NaCl solutions yields excellent agreement between predicted values and experimental measurements over the entire range of temperature and NaCl concentration conditions. For the limited number of solubility measurements in seawater-type solutions and mixed-salt brines, the agreement using the Pitzer relations is within three percent of the measured solubility.