Study on the kinetics of iron oxide leaching by oxalic acid

The presence of iron oxides in clay or silica raw materials is detrimental to the manufacturing of high quality ceramics. Although iron has been traditionally removed by physical mineral processing, acid washing has been tested as it is more effective, especially for extremely low iron (of less than 0.1% w/w). However, inorganic acids such as sulphuric or hydrochloric acids easily contaminate the clay products with SO₄²⁻ and Cl⁻, and therefore should be avoided as much as possible. On the other hand, if oxalic acid is used, any acid left behind will be destroyed during the firing of the ceramic products. The characteristics of dissolution of iron oxides were therefore investigated in this study.The dissolution of iron oxides in oxalic acid was found to be very slow at temperatures within the range 25–60 °C, but its rate increases rapidly above 90 °C. The dissolution rate also increases with increasing oxalate concentration at the constant pH values set within the optimum range of pH2.5–3.0. At this optimum pH, the dissolution of fine pure hematite (Fe₂O₃) (105–140 μm) follows a diffusion-controlled shrinking core model. The rate expression expressed as 1 − (2 / 3)x − (1 − x)^2 / 3 where x is a fraction of iron dissolution was found to be proportional to [oxalate]^1.5.The addition of magnetite to the leach liquor at 10% w/w hematite was found to enhance the dissolution rate dramatically. Such addition of magnetite allows coarser hematite in the range 0.5–1.4 mm to be leached at a reasonable rate.     

Geological and anthropogenic factors influencing mercury speciation in mine wastes: an EXAFS spectroscopy study

The speciation of Hg is a critical determinant of its mobility, reactivity, and potential bioavailability in mine-impacted regions. Furthermore, Hg speciation in these complex natural systems is influenced by a number of physical, geological, and anthropogenic variables. In order to investigate the degree to which several of these variables may affect Hg speciation, extended X-ray absorption fine structure (EXAFS) spectroscopy was used to determine the Hg phases and relative proportions of these phases present in Hg-bearing wastes from selected mine-impacted regions in California and Nevada. The geological origin of Hg ore has a significant effect on Hg speciation in mine wastes. Specifically, samples collected from hot-spring Hg deposits were found to contain soluble Hg-chloride phases, while such phases were largely absent in samples from silica-carbonate Hg deposits; in both deposit types, however, Hg-sulfides in the form of cinnabar (HgS, hex.) and metacinnabar (HgS, cub.) dominate. Calcined wastes in which Hg ore was crushed and roasted in excess of 600 °C, contain high proportions of metacinnabar while the main Hg-containing phase in unroasted waste rock samples from the same mines is cinnabar. The calcining process is thought to promote the reconstructive phase transformation of cinnabar to metacinnabar, which typically occurs at 345 °C. The total Hg concentration in calcines is strongly correlated with particle size, with increases of nearly an order of magnitude in total Hg concentration between the 500–2000 μm and <45 μm size fractions (e.g., from 97–810 mg/kg Hg in calcines from the Sulphur Bank Mine, CA). The proportion of Hg-sulfides present also increased by 8–18% as particle size decreased over the same size range. This finding suggests that insoluble yet soft Hg-sulfides are subject to preferential mechanical weathering and become enriched in the fine-grained fraction, while soluble Hg phases are leached out more readily as particle size decreases. The speciation of Hg in mine wastes is similar to that in distributed sediments located downstream from the same waste piles, indicating that the transport of Hg from mine waste piles does not significantly impact Hg speciation. Hg L_III-EXAFS analysis of samples from Au mining regions, where elemental Hg(0) was introduced to aid in the Au recovery process, identified the presence of Hg-sulfides and schuetteite (Hg₃O₂SO₄), which may have formed as a result of long-term Hg(0) burial in reducing high-sulfide sediments.     

Dissolution of cinnabar (HgS) in the presence of natural organic matter

Cinnabar (HgS) dissolution rates were measured in the presence of 12 different natural dissolved organic matter (DOM) isolates including humic, fulvic, and hydrophobic acid fractions. Initial dissolution rates varied by 1.3 orders of magnitude, from 2.31 × 10⁻¹³ to 7.16 × 10⁻¹² mol Hg (mg C)⁻¹ m⁻²s⁻¹. Rates correlate positively with three DOM characteristics: specific ultraviolet absorbance (R² = 0.88), aromaticity (R² = 0.80), and molecular weight (R² = 0.76). Three experimental observations demonstrate that dissolution was controlled by the interaction of DOM with the cinnabar surface: (1) linear rates of Hg release with time, (2) significantly reduced rates when DOM was physically separated from the surface by dialysis membranes, and (3) rates that approached constant values at a specific ratio of DOM concentration to cinnabar surface area, suggesting a maximum surface coverage by dissolution-reactive DOM. Dissolution rates for the hydrophobic acid fractions correlate negatively with sorbed DOM concentrations, indicating the presence of a DOM component that reduced the surface area of cinnabar that can be dissolved. When two hydrophobic acid isolates that enhanced dissolution to different extents were mixed equally, a 20% reduction in rate occurred compared to the rate with the more dissolution-enhancing isolate alone. Rates in the presence of the more dissolution-enhancing isolate were reduced by as much as 60% when cinnabar was prereacted with the isolate that enhanced dissolution to a lesser extent. The data, taken together, imply that the property of DOM that enhances cinnabar dissolution is distinct from the property that causes it to sorb irreversibly to the cinnabar surface.     

Adsorption of NO, SO2 and light hydrocarbons on activated Greek brown coals

Twenty-eight samples of peat, peaty lignites and lignites (of both matrix and xylite-rich lithotypes) and subbituminous coals have been physically activated by pyrolysis. The results show that the surface area of the activated coal samples increases substantially and the higher the carbon content of the samples the higher the surface area.The adsorption capacity of the activated coals for NO, SO₂, C₃H₆ and a mixture of light hydrocarbons (CH₄, C₂H₆, C₃H₈ and C₄H₁₀) at various temperatures was measured on selected samples. The result shows a positive correlation between the surface area and the gas adsorption. In contrast, the gas adsorption is inversely correlated with the temperature. The maximum recorded adsorption values are: NO = 8.22 × 10^− 5 mol/g at 35 °C; SO₂ = 38.65 × 10^− 5 mol/g at 60 °C; C₃H₆ = 38.9 × 10^− 5 mol/g at 35 °C; and light hydrocarbons = 19.24 × 10^− 5 mol/g at 35 °C. Adsorption of C₃H₆ cannot be correlated with either NO or SO₂. However, there is a significant positive correlation between NO and SO₂ adsorptions. The long chain hydrocarbons are preferentially adsorbed on activated lignites as compared to the short chain hydrocarbons.The results also suggest a positive correlation between surface area and the content of telohuminite maceral sub-group above the level of 45%.     

Hydrochemistry and isotope compositions of groundwater from the Shihongtan sandstone-hosted uranium deposit, Xinjiang, NW China

Groundwaters and surface water in the Shihongtan sandstone-hosted U ore district, Xinjiang, NW China, were sampled and analyzed for their major-, and trace element concentrations and oxygen, hydrogen, boron and strontium isotope compositions in order to assess the possible origins of the waters and water–rock interactions that occurred in the deep aquifer system. The waters in the study district have been grouped into three hydrochemical facies: Facies 1, potable spring-water, is a pH neutral (7.0), Na–Ca–HCO₃ type water with low total dissolved solids (TDS; 0.2 g/l, fresh) and has δ¹⁸O of − 8.3‰, δD of − 48.2‰,δ¹¹B of 1.5‰, and ⁸⁷Sr/⁸⁶Sr of 0.70627. Facies 2 groundwaters are mildly acidic to mildly alkaline (pH of 6.5–8.0, mean 7.3), Na–Ca–Mg–Cl–SO₄ type waters with moderate TDS (8.2 g/l–17.2 g/l, mean 9.3 g/l, brackish) and haveδ¹⁸O values in the − 5.8‰ to − 9.3‰ range (mean − 8.1‰), δD values in the − 20.8‰ to − 85.5‰ range (mean − 47.0‰),δ¹¹B values in the + 9.5‰ to + 39.1‰ range (mean + 17.1‰), and ⁸⁷Sr/⁸⁶Sr values in the 0.70595 to 0.70975 range (mean 0.70826). Facies 3, Aiting Lake water, is a mildly alkaline (pH = 7.4), Na–Ca–Mg–Cl–SO₄ type water with the highest TDS (249.1 g/l, brine) and has δ¹⁸O of − 2.8‰, δD of − 45.8‰,δ¹¹B of 21.2‰, and ⁸⁷Sr/⁸⁶Sr of 0.70840. The waters from the study district show a systematic increase in major, trace element and TDS concentrations and δ¹¹B values along the pathway of groundwater migration which can only be interpreted in terms of water–rock interaction at depth and strong surface evaporation. The hydrochemical and isotopic data presented here confirm that the groundwaters in the Shihongtan ore district are the combined result of migration, water–rock interaction and mixing of meteoric water with connate waters contained in sediments.     

Fluid geochemistry in the Ivigtut cryolite deposit, South Greenland

The 1.27 Ga old Ivigtut (Ivittuut) intrusion in South Greenland is world-famous for its hydrothermal cryolite deposit [Na₃AlF₆] situated within a strongly metasomatised A-type granite stock. This detailed fluid inclusion study characterises the fluid present during the formation of the cryolite deposit and thermodynamic modelling allows to constrain its formation conditions.Microthermometry revealed three different types of inclusions: (1) pure CO₂, (2) aqueous-carbonic and (3) saline-aqueous inclusions. Melting temperatures range between − 23 and − 15 °C for type 2 and from − 15 to − 10 °C for type 3 inclusions. Most inclusions homogenise between 110 and 150 °C into the liquid.Stable isotope compositions of CO₂ and H₂O were measured from crushed inclusions in quartz, cryolite, fluorite and siderite. The δ¹³C values of about − 5‰ PDB are typical of mantle-derived magmas. The differences between δ¹⁸O of CO₂ (+ 21 to + 42‰ VSMOW) and δ¹⁸O of H₂O (− 1 to − 21.7‰ VSMOW) suggest low-temperature isotope exchange. δD (H₂O) ranges from − 19 to − 144‰ VSMOW. The isotopic composition of inclusion water closely follows the meteoric water line and is comparable to Canadian Shield brines. Ion chromatography revealed the fluid's predominance in Na, Cl and F. Cl/Br ratios range between 56 and 110 and may imply intensive fluid–rock interaction with the host granite.Isochores deduced from microthermometry in conjunction with estimates for the solidification of the Ivigtut granite suggest a formation pressure of approximately 1–1.5 kbar for the fluid inclusions. Formation temperatures of different types of fluid inclusions vary between 100 and 400 °C. Thermodynamic modelling of phase assemblages and the extraordinary high concentration in F (and Na) may indicate that the cryolite body and its associated fluid inclusions could have formed during the continuous transition from a volatile-rich melt to a solute-rich fluid.     

High-field cantilever magnetometry as a tool for the determination of the magnetocrystalline anisotropy of single crystals

Cantilever torque magnetometry is utilized widely in physics and material science for the determination of magnetic properties of thin films and semiconductors. Here, we report on its first application in rock magnetism, namely the determination of K₁ and K₂ of single crystal octahedra of natural magnetite. The design of cantilever magnetometers allows optimization for the specific research question at hand. For the present study, a cantilever magnetometer was used that enables measurement of samples with a volume up to 64 mm³. It can be inserted into an electromagnet with a maximum field of 2 T. The cantilever spring is suitable for torque values ranging from 7.5 × 10^− 7 N·m to 5 × 10^− 6 N·m. The torque is detected capacitively; the measured capacitance is converted into torque by using a calibrated feedback coil. The magnetometer allows in-situ rotation of the sample in both directions and is, therefore, also suitable to analyze rotational hysteresis effects.The evaluation of the magnetite anisotropy constants involved Fourier analysis of the torque signal on the magnetite crystals' (001) and (110) planes. The absolute anisotropy constant has been computed using the extrapolation-to-infinite-field method. The value of K₁ at room temperature is determined at − 1.28 × 10⁴ [J m^− 3] (± 0.13, i.e. 10%) and that of K₂ at − 2.8 × 10³ [J m^− 3] (± 0.1, i.e. 2%). These values concur with earlier determinations that could not provide an instrumental error, in contrast with this work.The cantilever magnetometer performs four times faster than other torque magnetometers used for rock magnetic studies. This makes the instrument also suitable for magnetic fabric analysis.     

11 Å tobermorite from cement bypass dust and waste container glass: A feasibility study

A novel one-step hydrothermal synthesis of 11 Å tobermorite, a cation exchanger, from a unique combination of waste materials is reported. 11 Å tobermorite was prepared from stoicheiometric quantities of cement bypass dust and waste container glass at 100 °C in water. The product also comprised 10 wt.% calcite and trace quartz as residual parent phases from the cement bypass dust. In a batch sorption study at 20 °C the uptakes of Cd²⁺ and Pb²⁺ by the waste-derived tobermorite product were found to be 171 mg g^− 1 and 467 mg g^− 1, respectively, and in both cases the removal process could be described using a simple pseudo-second-order rate model (k₂ = 2.30 × 10^− 5 g mg^− 1 min^− 1 and 5.09 × 10^− 5 g mg^− 1 min^− 1, respectively). The sorption characteristics of the 11 Å tobermorite are compared with those of other waste-derived sorbents and potential applications are discussed.     

Comparison of methods for the estimation of pyrite oxidation rate in a waste rock pile at Mine Doyon site, Quebec, Canada

Several methods were evaluated and compared for the estimation of pyrite oxidation rates (POR) in waste rock at Mine Doyon, Quebec, Canada. Methods based on data collected in situ, such as the interpretation of temperature and oxygen concentration profiles (TOP) measured in the waste rock pile and pyrite mass balance (PMB) on solid phase samples were compared with the oxygen consumption measurements (OCM) in closed chamber in the laboratory. A 1-D analytical solution to a gas and heat transport equation used temperature and oxygen profiles (TOP) measured in the pile for the preliminary POR estimates at a site close to the slope of the pile (Site 6) and in the core of the pile (Site 7). Resulting POR values were 1.1 × 10^− 9 mol(O₂) kg^− 1 s^− 1 and 1.0 × 10^− 10 mol(O₂) kg^− 1 s^− 1 for the slope site and the core site, respectively. Oxidation rates based on pyrite mass balance (PMB) calculations for solid samples were 2.21 × 10^− 9 mol(O₂) kg^− 1 s^− 1 and 2.03 × 10^− 9 mol(O₂) kg^− 1 s^− 1, respectively, for the same slope and core sites, but the difference between sites was within the error margin. The OCM measurements in the laboratory on fresh waste rock samples yielded higher POR values than field methods, with average oxidation rate of 6.7 × 10^− 8 mol(O₂) kg^− 1 s^− 1. However, the OCM results on weathered and decomposed material from the rock stockpile (average oxidation rate 3.4 × 10^− 9 mol(O₂) kg^− 1 s^− 1) were consistent with results from the field-based estimates. When POR values based on fresh material are excluded, the remaining POR values for all methods range from 1.0 × 10^− 10 to 3.4 × 10^− 9 mol(O₂) kg^− 1 s^− 1. The lowest estimated value (1.0 × 10^− 10 mol(O₂) kg^− 1 s^− 1) was based on TOP estimates in the interior of the pile where oxygen transport was limited by diffusion from the surface. These results suggest that small-scale OCM laboratory experiments may provide relatively representative values of POR in the zones of waste rock piles in which oxygen transport is not dominated by diffusion.     

Quantification of water exchange between a hill reservoir and groundwater using hydrological and isotopic modelling (El Gouazine, Tunisia)

Stable isotope compositions (¹⁸O and ²H), determined for underground and surface waters from the watershed of a hill reservoir and downstream from the reservoir, suggest that (i) the reservoir water, which is more or less evaporated, represents a mixture between surface waters (rainfall, runoff) and the upstream alluvial groundwater meteoric in origin; (ii) the downstream alluvial groundwater have a stable isotope composition of a previously infiltrated reservoir water. The ¹⁸O isotope enrichment modelling of the mixed reservoir water shows that an input flux of 50 m³ day⁻¹ is balanced by an output flux of 300 m³ day⁻¹ when the reservoir water level is above 4.5 m, and by an output flux of 170 m³ day⁻¹ when water levels are lower. The contribution of hill reservoirs to local groundwater movements must be considered in any regional scale hydrological investigations of areas that contain hill reservoirs. To cite this article: O. Grünberger et al., C. R. Geoscience 336 (2004).

Résumé

La composition isotopique (¹⁸O et ²H) d'eaux souterraines et superficielles, déterminée pour le bassin versant d'une retenue collinaire et sa partie aval, suggère (i) que l'eau de la retenue, qui est plus ou moins évaporée, représente un mélange entre les eaux de surface (pluie, ruissellement) et la nappe alluviale amont d'origine météorique et (ii) que la nappe alluviale aval a la composition isotopique d'une eau de la retenue anciennement infiltrée. La modélisation de l'enrichissement en isotope ¹⁸O pour l'eau mélangée de la retenue montre qu'un flux entrant de 50 m³ j⁻¹ est compensé par un flux sortant de 300 m³ j⁻¹, lorsque le niveau d'eau de la retenue est supérieur à 4,5 m, et par un flux sortant de 170 m³ j⁻¹ en dessous. La contribution des lacs collinaires aux écoulements souterrains doit être prise en compte pour de futures études hydrologiques à l'échelle régionale. Pour citer cet article : O. Grünberger et al., C. R. Geoscience 336 (2004).     

Geology and geochemistry of the adjacent Changkeng gold and Fuwang silver deposits, Guangdong Province, South China

The Changkeng Au and Fuwang Ag deposits represent an economically significant and distinct member of the Au–Ag deposit association in China. The two deposits are immediately adjacent, but the Au and Ag orebodies separated from each other. Ores in the Au deposit, located at the upper stratigraphic section and in the southern parts of the orefield, contain low Ag contents (< 11 ppm); the Ag orebodies, in the lower stratigraphic section, are Au-poor (< 0.2 ppm). Changkeng is hosted in brecciated cherts and jasperoidal quartz and is characterized by disseminated ore minerals. Fuwang, hosted in the Lower Carboniferous Zimenqiao group bioclastic limestone, has vein and veinlet mineralization associated with alteration comprised of quartz, carbonate, sericite, and sulfides. Homogenization temperatures of fluid inclusions from quartz veinlets in the Changkeng and Fuwang deposits are in the range of 210 ± 80 °C and 230 ± 50 °C, respectively. Salinities of fluid inclusions from the two deposits range from 1.6 to 7.3 wt.% and 1.6 to 2.6 wt.% equiv. NaCl, respectively. The δD_H2O, δ¹⁸O_H2O, δ¹³C_CO2 and ³He/⁴He values of the fluid inclusions from the Changkeng deposit range from − 80‰ to − 30‰, − 7.8‰ to − 3.0‰, − 16.6‰ to − 17.0‰ and 0.0100 to 0.0054 Ra, respectively. The δD_H2O, δ¹⁸O_H2O, δ¹³C_CO2 and ³He/⁴He values of fluid inclusions from the Fuwang deposit range from − 59‰ to − 45‰, − 0.9‰ to 4.1‰, − 6.7‰ to − 0.6‰ and 0.5930 to 0.8357 Ra, respectively. The δD_H2O, δ¹⁸O_H2O, δ¹³C_CO2 and ³He/⁴He values of the fluid inclusions suggest the ore fluids of the Changkeng Au-ore come from the meteoric water and the ore fluids of the Fuwang Ag-ore are derived from mixing of magmatic water and meteoric water. The two deposits also show different Pb-isotopic signatures. The Changkeng deposit has Pb isotope ratios (²⁰⁶Pb/²⁰⁴Pb: 18.580 to 19.251, ²⁰⁷Pb/²⁰⁴Pb: 15.672 to 15.801, ²⁰⁸Pb/²⁰⁴Pb: 38.700 to 39.104) similar to those (²⁰⁶Pb/²⁰⁴Pb: 18.578 to 19.433, ²⁰⁷Pb/²⁰⁴Pb: 15.640 to 15.775, ²⁰⁸Pb/²⁰⁴Pb: 38.925 to 39.920) of its host rocks and different from those (²⁰⁶Pb/²⁰⁴Pb: 18.820 to 18.891, ²⁰⁷Pb/²⁰⁴Pb: 15.848 to 15.914, ²⁰⁸Pb/²⁰⁴Pb: 39.579 to 39.786) of the Fuwang deposit. The different signatures indicate different sources of ore-forming material. Rb–Sr isochron age (68 ± 6 Ma) and ⁴⁰Ar–³⁹Ar age (64.3 ± 0.1 Ma) of the ore-related quartz veins from the Ag deposit indicate that the Fuwang deposit formed during the Cenozoic Himalayan tectonomagmatic event. Crosscutting relationships suggests that Au-ore predates Ag-ore. The adjacent Changkeng and Fuwang deposits could, however, represent a single evolved hydrothermal system. The ore fluids initially deposited Au in the brecciated siliceous rocks, and then mixing with the magmatic water resulted in Ag deposition within fracture zones in the limestone. The deposits are alternatively the product of the superposition of two different geological events. Age evidence for the Fuwang deposit, together with the Xiqiaoshan Tertiary volcanic-hosted Ag deposit in the same area, indicates that the Pacific Coastal Volcanic Belt in the South China Fold Belt has greater potential for Himalayan precious metal mineralization than previous realized.     

Kinetics of leaching of oxidized and reduced ilmenite in dilute hydrochloric acid solutions

Ilmenite separated from beach sands of Bangladesh was oxidized for 1 h at 950 °C and then reduced in charcoal for 4 h at 1050 °C. This was followed by leaching in 5% to 15% hydrochloric acid solution in temperature range of 30 to 75 °C for periods of up to 2 h. The results were compared with those obtained by leaching of ilmenite reduced without oxidizing. Oxidation prior to reduction of the ilmenite was found to increase both the extent and the rate of leaching. The residual iron contents after leaching were also found to be lower than that obtained for non-oxidized samples. The kinetic data of leaching of ilmenite reduced after oxidation was found to follow first order reaction model, i.e., G(α) = − ln(1 − α) up to an α value of 0.5 (i.e. up to 50% reduction) and then changed to spherical model, i.e., G(α) = [1 − (1 − α)]³. On the other hand, leaching of ilmenite reduced without oxidizing was found to follow the Ginstling-Brounshtein reaction, i.e., G(α) = 1 − (2/3)α − (1 − α)^2/3 throughout the leaching process. Oxidation of ilmenite prior to reduction was also found to have decreased the activation energy of leaching from 43 kJ/mol, found for samples leached after reduction without oxidizing, to 30 kJ/mol.     

A pilot-scale field experiment for the microbial neutralization of a holomictic acidic pit lake

A strategy to neutralize acidic pit lakes was tested in an upscaling process using field mesocosms of 26 to ca. 4500 m³ volume in the acidic pit Mining Lake 111 in Germany. After addition of the substrates Carbokalk and straw a neutral sediment layer formed, in which microbial sulfate and iron reduction as well as sulfide precipitation occurred. The net rate of neutralization was limited by the precipitation of iron sulfides rather than by microbial reactions. Oxidation of H₂S by ferric iron in the anoxic sediment lowered the net sulfate reduction rate. Seasonal fluctuations of iron sulfides in the sediment showed that the reaction products were not necessarily stable. The long-term success of the approach depends on the net partition of the precipitated iron-(mono-/di-) sulfide that is permanently buried in the anoxic sediment. It could be shown by field experiments that the long-term success of the neutralization depends on the spatial scale and duration of the experiments. Volumes from 26 to 4500 m³, exposition times from 4 months to 5 years, and increasingly thick coverings of the sediments with straw, from zero to 40 cm, were used. Net neutralization rates decreased from 41 meq m^− 2 d^− 1 in laboratory microcosms to a mean rate of 2.3 meq m^− 2 d^− 1 in the 4500 m³ field experiment. The results show that the success of the microbial treatment of acid pit lakes lastly depends on the limnological conditions in the lake that cannot be simulated by upscaling of simple laboratory experiments.     

Iron isotopes in acid mine waters and iron-rich solids from the Tinto–Odiel Basin (Iberian Pyrite Belt, Southwest Spain)

The isotopic composition of Fe was determined in water, Fe-oxides and sulfides from the Tinto and Odiel Basins (South West Spain). As a consequence of sulfide oxidation in mine tailings both rivers are acidic (1.45 < pH < 3.85) and display high concentrations of dissolved Fe (up to 420 mmol l^− 1) and sulphates (up to 1190 mmol l^− 1).The δ⁵⁶Fe of pyrite-rich samples from the Rio Tinto and from the Tharsis mine ranged from − 0.56 ± 0.08‰ to + 0.25 ± 0.1‰. δ⁵⁶Fe values for Fe-oxides precipitates that currently form in the riverbed varied from − 1.98 ± 0.10‰ to 1.57 ± 0.08‰. Comparatively narrower ranges of values (− 0.18 ± 0.08‰ and + 0.21 ± 0.14‰) were observed in their fossil analogues from the Pliocene–Pleistocene and in samples from the Gossan (the oxidized layer that formed through exposure to oxygen of the massive sulfide deposits) (− 0.36 ± 0.12‰ to 0.82 ± 0.07‰). In water, δ⁵⁶Fe values ranged from − 1.76 ± 0.10‰ to + 0.43 ± 0.05‰.At the source of the Tinto River, fractionation between aqueous Fe(III) and pyrite from the tailings was less than would be expected from a simple pyrite oxidation process. Similarly, the isotopic composition of Gossan oxides and that of pyrite was different from what would be expected from pyrite oxidation. In rivers, the precipitation of Fe-oxides (mainly jarosite and schwertmannite and lesser amounts of goethite) from water containing mainly (more than 99%) Fe(III) with concentrations up to 372 mmol l^− 1 causes variable fractionation between the solid and the aqueous phase (− 0.98‰ < Δ⁵⁶Fe_{solid–water} < 2.25‰). The significant magnitude of the positive fractionation factor observed in several Fe(III) dominated water may be related to the precipitation of Fe(III) sulphates containing phases.     

Geochemistry and stable isotopic composition of tufa waters and precipitates from the Interlake Region, Manitoba, Canada: Constraints on groundwater origin, calcitization, and tufa formation

New major, trace and isotopic geochemical results from a regional study of springs discharging from the major carbonate rock aquifer in the Interlake Region of Manitoba, Canada, are used to understand water–rock reactions, timing of recharge/discharge, tufa formation processes, and as baseline data. Spring waters are fresh with total dissolved solids (TDS) concentrations ranging from 150 to 880 mg/L. Waters discharging in the northern part of the study area have lower TDS, are dominantly Ca–Mg–HCO₃ waters with low SO₄ concentrations (<< 50 mg/L), and appear to have interacted primarily with Silurian carbonate lithologies. In contrast, waters in the southeastern part of the study area have higher TDS and have elevated SO₄ concentrations (up to 210 mg/L). Spring waters have elevated Mg/Ca_molar (1.23 ± 0.23), typically greater than congruent dissolution of dolomite. Ca and Mg concentrations and Mg/Ca_molar indicate that groundwater residence times were sufficient to allow equilibration with bedrock dolomite lithologies; elevated tritium in northern waters indicates a significant recharge component in the 1960's and 1970's. Tufa precipitates that have formed from many of the spring waters are low-Mg calcite (MgO = 1.70 to 5.80 wt.%). Sr concentrations are variable (57 to 657 ppm) and tufa Sr/Ca_molar ratios appear to be entirely controlled by spring water Sr/Ca_molar. Empirically determined Sr distribution coefficients (D_Sr = 0.389 ± 0.083) indicate rapid crystallization following CO₂ degassing, consistent with heavier δ¹³C_VPDB compared to spring waters. Sulfate concentrations are generally too low for calcitization (dedolomitization) reactions driven by anhydrite dissolution to be the dominant control on the elevated groundwater Mg/Ca_molar, implying either extensive sulfate reduction along the flow paths (however, δ¹³C_DIC suggests the elevated SO₄ is more consistent with Fe-sulfide oxidation), or that other processes are involved. Major ion ratios suggest that the waters in the southern part of the study area are more consistent with interaction with siliciclastic rocks than with anhydrite dissolution. We suggest that calcitization (dedolomitization) reactions driven by anhydrite dissolution may not dominate all carbonate aquifers and that mixing of waters in karst conduits combined with ion exchange reactions are important controls on water chemistry in these systems.     

Dissolved air flotation (DAF) of fine quartz particles using an amine as collector

Experimental studies concerning the dissolved air flotation (DAF) of fine (d_p < 100 μm) quartz particles, using two different flotation cells (setups), are presented. Pure and well characterised quartz samples were treated with a commercial amine as collector prior to flotation and bubbles were characterised by the LTM-BSizer technique. Bubble size distribution showed 71% (by volume) and 94% (by number) of the bubbles having sizes (d_b) lower than 100 μm (i.e. microbubbles). The Sauter and arithmetic mean diameters were 79 μm and 56 μm, respectively, for the bubbles generated at 300 kPa (gauge) saturation pressure (after 30 minute saturation time). Quartz particle size distribution (obtained by laser diffraction) showed a volume-moment diameter of 13 μm. The Rosin–Rammler–Bennett, Gates–Gaudin–Schumann and log-normal distribution functions were well fitted (R² > 0.96) to the bubble size distribution and quartz particle size distribution data. Values of total quartz recovery ranging from 6% to 53% (by mass) were obtained for the DAF experiments under different collector concentrations (up to 2 mg g^− 1), with an optimal collector concentration found at 1 mg g^− 1. These results are significant considering that 27% (by volume) of the quartz particles are ultrafine (d_p < 5 μm), demonstrating the widely-known efficiency of DAF to remove small particles when applied in the field of water and wastewater treatment. The true flotation behaviour, as a function of particle diameter (d_p), exhibits a local minimum when particles are approximately 3–5 μm in size. The results contribute to the discussion in the literature about the existence of such a minimum, which is generally interpreted as a change in the mechanism of particle collection from convection (collision) to diffusion at lower particle sizes.     

Mercury in itabirite-hosted soft hematite ore in the Quadrilátero Ferrífero of Minas Gerais

Mercury contents in Precambrian banded iron formation-hosted hematite ores are virtually unknown. In an attempt to provide information on the abundance and distribution of Hg in Fe ore, we present analyses for Hg in samples of high-grade soft hematite ore from Gongo Soco, Minas Gerais, Brazil. Bulk samples contain from <  5 to 25  ppb Hg without obvious correlation with major elements. Granulometric fractions of follow-up samples have amounts of Hg from 6 to 48  ppb and display positive linear correlations with total Mn as MnO (r = 0.87), LOI (r = 0.87) and SiO₂ (r = 0.76), as well as a negative linear correlation with total Fe as Fe₂O₃ (r = −  0.87). The correlations suggest that Hg is associated with a hydrated ferruginous groundmass bearing residual Mn, Al and Si, which replaced gangue minerals in itabirite in the process of formation of the Gongo Soco soft hematite ore.     

Distribution of copper, zinc, lead and cadmium concentrations in stream sediments from the Mapocho River in Santiago, Chile

The Mapocho river, which crosses downtown Santiago, is one of the most important rivers in contact with a population of about six million inhabitants. Anthropogenic activities, industrialization, farming activities, transport, urbanization, animal and human excretions, domestic wastes and copper mining have affected the river, contaminating it and its sediments with heavy metals. Concentration and distribution of Cu, Zn, Pb and Cd were studied with the purpose of determining their bioavailability and their relation with the characteristics of the sediments. Freshly deposited seasonal sediments were collected from 0–8 cm depths from 6 locations (S₁ to S₆) along the 30-km long channel length, in the four seasons of year on the following dates: May 2001 (D₁, autumn); August 2001 (D₂, winter); October 2001 (D₃, spring) and January 2002 (D₄, summer). The dried samples were sifted to obtain the < 63-μm sediment fraction, since it has been shown that large amounts of heavy metals are bound in the fine-grained fraction of the sediment. Cu and Zn were analyzed by atomic absorption spectrophotometry and Pb and Cd by square wave anodic stripping voltammetry. The highest concentrations of Cu (2850 μg g^− 1) were found in the northern part of the river (S₁, average D₁–D₄), near the mountains and a copper mine, and then decreased downstream to 209 μg g^− 1 (S₆). Total Zn showed an irregular variation, with higher values at S₁ (1290 μg g^− 1) and high values in some winter sampling (1384 μg g^− 1 S₄, S₅–D₂). Pb showed different trends, increasing from S₁ to S₆ (17 to 61 μg g^− 1), with the highest values in the summer samples (83 μg g^− 1, S₄–S₆, D₄), and total Cd increased slightly from mean values of 0.2 and 0.5 μg g^− 1. Partition into five fractions was made using Tessier's analytical sequential extraction technique; the residue was treated with aqua regia for recovery studies, although this step is not part of the Tessier procedure. The results show that Cu, Zn and Pb in the sediments were dependent on the sampling places along the river, and variation in two years was low (D₁–D₄). The highest values of total organic matter, carbonate and conductivity were found in S₆, which has the smallest size particles, while at S₁ the sediments were predominantly sand and contain larger amounts of silica. Cu associated with carbonate decreased gradually from 58% (1771 μg g^− 1, S₁) to 16% (32 μg g^− 1, S₆); Cu bonded to reducible fraction was almost constant (33% to 37%), and Cu associated with oxidizable fraction increased from 7% (S₁) to 34% (S₆), but copper content was lower (214 to 68 μg g^− 1). Zn had a similar fractionation profile. However, Pb bound to oxidizable fraction did not show significant percent variation along the river (20% to 19%), but the amount bounded was 4 to 12 μg g^− 1. The residual fraction increased from 24% to 41% (5 to 25 μg g^− 1, S₁ to S₆). The distribution of Cd in the sediment was almost independent of the sampling stations and was bound to carbonate, reducible and residual fraction in similar proportion. Cu and Zn at S₁ were mainly bound to carbonates and reducible phases with 91% and 73% (2779 and 965 μg g^− 1, respectively), and with a change in the pH and/or the redox potential of the sediment–water system, these contaminants could easily enter the food chain. In S₆ the amount of Cu and Zn in these phases was 50% and 53% (100 to 313 μg g^− 1, respectively).     

Geochemistry of S-type granitic rocks from the reversely zoned Castelo Branco pluton (central Portugal)

The zoned pluton from Castelo Branco consists of Variscan peraluminous S-type granitic rocks. A muscovite>biotite granite in the pluton's core is surrounded successively by biotite>muscovite granodiorite, porphyritic biotite>muscovite granodiorite grading to biotite=muscovite granite, and finally by muscovite>biotite granite. ID-TIMS U–Pb ages for zircon and monazite indicate that all phases of the pluton formed at 310 ± 1 Ma. Whole-rock analyses show slight variation in ⁸⁷Sr/⁸⁶Sr_310 Ma between 0.708 and 0.712, Nd_310 Ma values between − 1 and − 4 and δ¹⁸O values between 12.2 and 13.6. These geological, mineralogical, geochemical and isotopic data indicate a crustal origin of the suite, probably from partial melting of heterogeneous Early Paleozoic pelitic country rock. In detail there is evidence for derivation from different sources, but also fractional crystallization linking some of internal plutonic phases. Least-squares analysis of major elements and modelling of trace elements indicate that the porphyritic granodiorite and biotite=muscovite granite were derived from the granodiorite magma by fractional crystallization of plagioclase, quartz, biotite and ilmenite. By contrast variation diagrams of major and trace elements in biotite and muscovite, the behaviours of Ba in microcline and whole-rock δ¹⁸O, the REE patterns of rocks and isotopic data indicate that both muscovite-dominant granites were probably originated by two distinct pulses of granite magma.     

Intrinsic and scattering attenuation in western India from aftershocks of the 26 January, 2001 Kachchh earthquake

176 vertical-component, short period observations from aftershocks of the M_w 7.7, 26 January, 2001 Kachchh earthquake are used to estimate seismic wave attenuation in western India using uniform and two layer models. The magnitudes (M_w) of the earthquakes are less than 4.5, with depths less than 46 km and hypocentral distances up to 110 km. The studied frequencies are between 1 and 30 Hz. Two seismic wave attenuation factors, intrinsic absorption (Q_i^− 1) and scattering attenuation (Q_s^− 1) are estimated using the Multiple Lapse Time Window method which compares time integrated seismic wave energies with synthetic coda wave envelopes for a multiple isotropic scattering model. We first assume spatial uniformity of Q_i^− 1, Q_s^− 1 and S wave velocity (β). A second approach extends the multiple scattering hypothesis to media consisting of several layers characterized by vertically varying scattering coefficient (g), intrinsic absorption strength (h), density of the media (ρ) and shear wave velocity structure. The predicted coda envelopes are computed using Monte Carlo simulation. Results show that, under the assumption of spatial uniformity, scattering attenuation is greater than intrinsic absorption only for the lowest frequency band (1 to 2 Hz), whereas intrinsic absorption is predominant in the attenuation process at higher frequencies (2 to 30 Hz). The values of Q obtained range from Q_t = 118, Q_i = 246 and Q_s = 227 at 1.5 Hz to Q_t ≈ 4000, Q_i ≈ 4600 and Q_s ≈ 33,300 at 28 Hz center frequencies, being Q_t^− 1 a measure of total attenuation. Results also show that Q_i^− 1, Q_s^− 1 and Q_t^− 1 decrease proportional to f^−ν. Two rates of decay are clearly observed for the low (1 to 6 Hz) and high (6 to 30 Hz) frequency ranges. Values of ν are estimated as 2.07 ± 0.05 and 0.44 ± 0.09 for total attenuation, 1.52 ± 0.21 and 0.48 ± 0.09 for intrinsic absorption and 3.63 ± 0.07 and 0.06 ± 0.08 for scattering attenuation for the low and high frequency ranges, respectively. Despite the lower resolution in deriving the attenuation parameters for a two layered crust, we find that scattering attenuation is comparable to or smaller than the intrinsic absorption in the crust whereas intrinsic absorption dominates in the mantle. Also, for a crustal layer of thickness 42 km, intrinsic absorption and scattering estimates in the crust are lower and greater than those of the mantle, respectively.