Applied Geochemistry (v.15, #9)

A laboratory evaluation of metal release and transport in flooded pre-oxidized mine tailings by Paul H. Simms; Ernest K. Yanful; Luc St-Arnaud; Bernard Aubé (1245-1263).
A potential problem in implementing water covers over preoxidized tailings impoundments is the dissolution of oxidation products following flooding, which may result in high concentrations of metals in both tailings pore water and the water cover. To examine metal release phenomena under controlled conditions, a laboratory study consisting of four column experiments of flooded partially oxidized tailings was performed over 2 years. In two of the columns enough water was added to establish a 1-m water cover, while the other two columns were only filled to the surface of the tailings. The columns were kept stagnant for the first half of the study (“static phase”) and flushed under constant flow during the second half (“kinetic phase”). Samples of pore water and water cover were collected and analyzed for metals, SO4, acidity, pH, redox, and dissolved O2. Mass balance calculations were performed to assess metal release rates. At the end of the experiments the quality of the pore-water (pH 6) within the water covered tailings was better than that of the water cover itself (pH 3). While mineral dissolution released large amounts of Fe during the static phase in all experiments, no substantial dissolution or removal of Fe was detected after 300 days. In the water-covered tailings, Zn was removed from solution during both the static and kinetic phases, apparently through co-precipitation or adsorption with Fe(III) hydroxides. Metal and SO4 release rates measured in the laboratory were similar to those observed at full-scale field sites. The laboratory observations suggest that while implementing a water cover over pre-oxidized tailings does substantially reduce metal release, long-term treatment of pond effluent and seepage may still be required in some cases. In the field, however, metal concentrations would be influenced by other phenomena, such as dilution due to surface water flow and sediment resuspension through wind, which were not simulated in the laboratory. Thus, the laboratory results cannot be directly extrapolated to the field.

Massive sulphide deposits of the northern Iberian Pyrite Belt (IPB) are mainly hosted by felsic volcanic rocks of rhyolitic to dacitic composition. Beneath most of the massive ores of this area (e.g., Concepción, San Miguel, Aguas Teñidas Este or San Telmo deposits) there is usually a wide hydrothermal alteration halo associated with stockwork-type mineralization. Within these alteration envelopes there are two principal rock types: (1) chlorite-rich rocks, linked to the inner and more intensely altered zones and dominantly comprising chlorite+pyrite+quartz+sericite (+carbonate+rutile+zircon+chalcopyrite), and (2) sericite-rich rocks, more common in the peripheral zones and showing a dominant paragenesis of sericite+quartz+pyrite+chlorite (+carbonate+rutile+zircon+sphalerite). Mass-balance calculations comparing altered and least-altered felsic volcanic rocks suggest that sericitization was accompanied by moderate enrichment in Mg, Fe and H2O, with depletion in Si, Na and K, and a slight net mass loss of about 3%. Chloritization shows an overall pattern which is similar to that of the sericitic alteration, but with large gains in Fe, Mg and H2O (and minor enrichment in Si, S and Mn), and a significant loss of Na and K and a minor loss of Ca and Rb. However, chloritization has involved a much larger net mass change (mass gain of about 28%). Only a few elements such as Nb, Y, Zr, Ti, P and LREE appear to have remained inert during hydrothermal alteration, whilst Ti and Al have undergone very minor mobilization. The results point to the severity of the physico-chemical conditions that prevailed during the waxing stage of the ore-forming hydrothermal systems. Further, mineralogical and geochemical studies of the altered footwall rocks in the studied deposits indicate that hydrothermal ore-bearing fluids reacted with host rocks in a multi-stage process which produced a succession of mineralogical and chemical changes as the temperature increased.

Lead concentrations and isotopic compositions were determined on both bulk sediments deposited in the Thau lake in southern France during the last 200 years, and leachates derived from a series of sequential leachings of the sediments, making it possible to identify the sources, natural (i.e. indigenous lithologic) or anthropogenic, and to quantify the different inputs of Pb.Two distinct inputs of Pb could be distinguished. One of these corresponds to the terrigenous material entering the basin, representative of the local natural Pb ‘background’. Its supply remained steady most of the time with 206Pb/207Pb ratios of 1.200±0.003, except at the time of heavy storms producing voluminous and sudden depositions, such as that of September 1875. This Pb supply is mainly hosted by the detrital silicate fraction of the sediments. The second Pb input is a direct consequence of anthropogenic activities of various industrial and domestic emissions in the region, particularly due to the city of Sète and, to a lesser extent, to the villages in the watershed. The 206Pb/207Pb ratios of this input are of 1.142–1.162. The Pb added to gasoline could also be identified in the uppermost sediments, because of its specific 206Pb/207Pb ratios of 1.069–1.094. The leaching experiments also showed that the anthropogenic Pb is mainly hosted by the oxi-hydroxides of the sediments and to a lesser extent by the carbonates. It may also be adsorbed on particle surfaces, while only limited amounts are bound to organic matter.

Interaction between aqueous chromium solutions and layer silicates by Maria Franca Brigatti; Giancarlo Franchini; Cristina Lugli; Luca Medici; Luciano Poppi; Elisa Turci (1307-1316).
The interactions between Cr in aqueous solutions and phyllosilicates were studied to determine: (a) the amount of Cr(VI) to Cr(III) reduction in aqueous solutions by Fe(II)-bearing phyllosilicates; (b) the removal of the Cr species from solution by interaction with phyllosilicates, as a function of Cr(III) concentration and anionic environment. Chlorite, corrensite and montmorillonite were reacted with solutions containing Cr(VI) (1.62×10−3 N, 5.77×10−3 N and 1.32×10−1 N, respectively). The sorption/desorption of Cr(III) by saponite was investigated in different anionic environments (Cl, NO 3 and CH3COO) and at different initial Cr(III) concentrations (3.21×10−3 N, 5.49×10−3 N and 8.49×10−3 N).The extent of Cr(VI) reduction and the amount of Cr removed by phyllosilicates were measured by analysis of the liquid portion separated by centrifugation after controled periods of exposure. The minerals were studied by chemical, thermal and X-ray powder diffraction analyses. The results show that: (i) Fe(II)-bearing phyllosilicates sorb Cr and reduce Cr(VI) to Cr(III); (ii) the extent of reduction depends on the solution concentration and on mineral crystal chemistry; (iii) Cr(III) sorption isotherms show that the degree of uptake depends both on the initial concentration of metal in solution and on the anionic environment, the order of effectiveness being Cl≅NO 3>CH3COO; (iii) Cr(III) is retained in the mineral substrate and its release is difficult.

Production of 210Pb from a Slochteren Sandstone gas reservoir by Arthur P. Schmidt; F.A. Hartog; B.J.H. van Os; R.D. Schuiling (1317-1329).
Recently, precipitates of metallic Pb and galena in gas production facilities have been reported to contain 210Pb. In the North Sea area, U-bearing Kupferschiefer or Carboniferous coal measures have been suggested as a possible source of 210Pb. Through coproduction of formation water with natural gas, 210Pb may enter production facilities together with non-radioactive Pb, and precipitate as metallic Pb or galena. In this study, bituminous sandstones with up to 2.8 wt% organic C from a 210Pb producing Permian, Slochteren Formation gas reservoir have been found to contain up to 330 ppm U. The sandstones show a complex diagenetic history, during which migrating oil was trapped in secondary pores. Continuing enlargement of pore space and the wetting characteristics of the reservoir fluids lead to a preferred distribution of oil rims around quartz grains, creating fluid-filled voids between the grains. Syngenetic precipitation of finely grained U oxide occurred along the bitumen–fluid boundaries. U-Pb chemical age dating of U oxide grains provides an age of 246 Ma for the influx of the oil, and the close association of abundant anhydrite with the bitumen impregnations points to a Zechstein source for both bitumen and U. One of the bituminous sandstones shows a marked depletion of 210Pb with respect to 238U and 226Ra, probably caused by the production of natural gas from the reservoir. Production started at the end of the 1960s and induced fluid flow throughout the reservoir. This enabled transport of 222Rn and its decay product 210Pb away from the U oxide grains. The observed U concentrations and permeabilities make the bituminous sandstones in this reservoir probably a more important source of 210Pb than the overlying Kupferschiefer or underlying Carboniferous coal measures.

Solid phase iron–sulfur geochemistry of a reactive barrier for treatment of mine drainage by Roger B. Herbert; Shawn G. Benner; David W. Blowes (1331-1343).
This paper discusses the solid phase Fe–S geochemistry of a reactive barrier at the Nickel Rim mine site (Ontario, Canada). The barrier, designed to treat groundwater contaminated by acid mine drainage, is composed of leaf and municipal compost and wood chips. This study shows that S is accumulating in the organic material as primarily acid volatile sulfides, at concentrations up to 195 μmol S g−1 d.w. (0.63 wt% S) in a zone of preferential flow. Pyrite and/or S0 account for only a small fraction of the total reduced inorganic S, as oxidants are probably present at only low concentrations in the barrier system. The results of the solid phase analyses, the formation of disordered mackinawite (Fe1+x S) on piezometer tubing, and thermodynamic calculations indicate that the precipitation of poorly crystalline Fe monosulfides is the primary sink for Fe and S in the barrier. Siderite (FeCO3) formation is proposed as an additional Fe sink in areas of high Fe flux. Minor accumulations of acid-soluble, organically-bound S in the reactive barrier occur in zones of low aqueous Fe concentration. After 23 months of operation, the average rate of S accumulation at the up-gradient edge of the barrier is calculated to be 87 μmol S g−1 a−1 (d.w. organic material), or 47 mol S m−2 a−1 in the direction of groundwater flow. Solid phase analyses from samples collected 3 and 14 months after installation indicate that the S accumulation rate declined by a factor of 3 over that time period.

Rare earth elements, neodymium and strontium isotopic systematics in mineral waters: evidence from the Massif Central, France by Ph Négrel; C Guerrot; A Cocherie; M Azaroual; M Brach; Ch Fouillac (1345-1367).
Rare Earth Elements (REEs), and Sr and Nd isotope distributions, have been studied in mineralized waters from the Massif Central (France). The CO2-rich springs are characterized by a neutral pH (6–7) associated with total dissolved solids (TDS) from 1 to 7 g l−1. The waters result from the mixing of very mineralized water pools, thought to have equilibrated at a temperature of around 200°C with superficial waters. These two mineral water pools evidenced by Sr isotopes and dissolved REEs could reflect 2 different stages of water–rock interaction and an equilibrium with different mineral assemblages.The concentrations of individual dissolved REEs and total dissolved REEs (ΣREE), in the mineral waters examined, vary over several orders of magnitude but are not dependent on the main parameters of the waters (TDS, T°C, pH, Total Organic C). The dissolved REE concentrations presented as upper continental crust normalized patterns show HREE enrichment in most of the samples. The time evolution of REE patterns does not show significant fluctuations except in 1 borehole, located in the Limagne d’Allier area, which was sampled on 16 occasions over an 18 month period. Ten samples are HREE-enriched, whereas 6 samples show flat patterns.The aqueous speciation of REEs shows that CO2− 3 complexes dominate (>80%) over the free metal, F, SO2− 4 and HCO 3 complexes. The detailed speciation demonstrates that the fractionation of REEs (i.e. the HREE enrichment) in CO2-rich and pH neutral fluids is due essentially to the predominance of the CO2− 3 complexes.The Sr isotopic composition of the mineral waters in the Massif Central shows different mixing processes; in the Cézallier area at least 3 end-member water types exist. The most dilute end-member is likely to originate as poorly mineralized waters with minimal groundwater circulation. Two other mineralized end-members are identified, although the link between the geographical location of spring outflow and the mixing proportion between the 2 end-members is not systematic. The range in ϵ Nd(0) for mineralized waters in the Massif Central correlates well with that of the known parent rocks except for 4 springs. One way to explain the ϵ Nd(0) in these instances is a contribution from drainage of volcanic rocks. The isotopic systematics help to constrain the hydrogeological models for this area.

Thirty-eight samples of stream sediments draining high-grade metamorphic rocks in the Walawe Ganga (river) Basin, Sri Lanka, were analysed for their REE contents, together with samples of metamorphic suites from the source region. The metamorphic rocks are enriched in light REE (LREE) compared to heavy REE (HREE) and are characterised by high La/Lu ratios and negative Eu anomalies. The chondrite-normalised patterns for these granulite-grade rocks are similar to that of the average post-Archaean upper crust, but they are slightly enriched with La and Ce. The REE contents of the <63-μm fraction of the stream sediments are similar to the probable source rocks, but the other grain size fractions show more enriched patterns. The <63-μm stream sediments fraction contains lower total REE, more pronouncd negative Eu anomalies, higher EuN/SmN and lower La N/LuN ratios relative to other fractions. The lower La N/LuN ratio is related to the depletion of heavy minerals in the <63-μm fraction. The 63–125-μm and 125–177-μm grain size fractions of sediments are particularly enriched in LREE (average ΣLREE=2990 μg/g and 3410 μg/g, respectively). The total HREE contents are surprisingly uniform in all size fractions. However, the REE contents in the Walawe Ganga sediments are not comparable with those of the granulite-grade rocks from the source region of the sediments. The enrichment of REE is accounted for by the presence of REE containing accessory mineral phases such as zircon, monazite, apatite and garnet. These minerals are derived from an unknown source, presumably from scattered bodies of granitic pegmatites.

Water pollution arising from base metal sulphide mines is problematic in many countries, yet the hydrogeology of the subsurface contaminant sources is rarely well-characterized. Drainage water pumped from an active F–Pb mine in northern England has unusual chemistry (alkaline with up to 40 mg.l−1 Zn) which profoundly impacts the ecology of the receiving watercourse. Detailed in-mine surveys of the quantity and quality of all ground water inflows to the mine were made. These revealed major, temporally persistent heterogeneities in ground water quality, with three broad types of water identified as being associated with distinct hydrostratigraphic units. Type I waters (associated with the Firestone Sill aquifer) are cool (<10°C), Ca–HCO3–SO4 waters, moderately mineralized (specific electrical conductance (SEC)≤410 μ−1) with <4 mg.l−1 Zn. Type II waters (associated with the Great Limestone aquifer) are warmer (≈15°C), of Ca–SO4 facies, highly mineralized (SEC≤1500 μ−1) with ≤40 mg.l−1 Zn. Type III waters (in the deepest workings) are tepid (>18°C), of Ca–HCO3–SO4 facies, intermediately mineralized (SEC≤900 μ−1) with ≤13 mg.l−1 Zn, and with significant Fe (≤12 mg.l−1) and Pb (≤8 mg/l). Monotonic increases in temperature and Cl concentration with depth contrast with peaks in total mineralization, SO4 and Zn at medium depth (in Type II waters). Sulphate, Pb and Zn are apparently sourced via oxidation of galena and sphalerite, which would release each metal in stoichiometric equality with SO4. However, molal SO4 concentrations typically exceed those of Pb and Zn by 2–3 orders of magnitude, which mineral equilibria suggest is due to precipitation of carbonate “sinks” for these metals. Contaminant loading budgets demonstrate that, although Type II waters amount to only 25% of the total ground water inflow to the mine, they account for almost 60% of the total Zn loading. This observation has important management implications for both the operational and post-abandonment phases of the mine life cycle.

The kinetics and mechanisms of simulated British Magnox waste glass dissolution as a function of pH, silicic acid activity and time in low temperature aqueous systems by P.K. Abraitis; F.R. Livens; J.E. Monteith; J.S. Small; D.P. Trivedi; D.J. Vaughan; R.A. Wogelius (1399-1416).
Dissolution of a simulated British Magnox waste glass is governed by two pH-dependent processes. At low pH, dissolution is governed by reactions occurring predominantly at non-Si sites and residual Si-rich gels develop at the glass surface as B, Al and modifier cations are selectively leached. Here, extensive proton promoted hydrolysis of BO and AlO bonds is coupled with hydration and ion exchange processes. Hydrolysis of siloxane bonds governs the rate of dissolution at high pH and the glass dissolves congruently as the silicate network breaks down extensively. Differences in the surface chemistries and morphologies of glass samples reacted in strongly acidic and highly alkaline media reflect the net effects of these processes. The rate of the congruent dissolution process is influenced by the activity of silicic acid. The results are compared with published data for other glass formulations and are discussed in the context of proposed kinetic dissolution models.