The application of mineralogical characterization to mine waste has the potential to improve risk assessment, guide appropriate mine planning for planned and active mines and optimize remediation design at closed or abandoned mines. Characterization of minerals, especially sulphide and carbonate phases, is particularly important for predicting the potential for acidic drainage and metal(loid) leaching. Another valuable outcome from mineralogical studies of mine waste is an understanding of the stability of reactive and metal(loid)-bearing minerals under various redox conditions. This paper reviews analytical methods that have been used to study mine waste mineralogy, including conventional methods such as X-ray diffraction and scanning electron microscopy, and advanced methods such as synchrotron-based microanalysis and automated mineralogy. We recommend direct collaboration between researchers and mining companies to choose the optimal mineralogical techniques to solve complex problems, to co-publish the results, and to ensure that mineralogical knowledge is used to inform mine waste management at all stages of the mining life cycle. A case study of arsenic-bearing gold mine tailings from Nova Scotia is presented to demonstrate the application of mineralogical techniques to improve human health risk assessment and the long-term management of historical mine wastes. (C) 2015 Elsevier Ltd. All rights reserved.

ABSTRACT A comprehensive understanding of the risk associated with metal-rich soils and other materials includes identification of the solid phases hosting the metals. Synchrotron microanalysis provides a powerful diagnostic tool to characterize metal-bearing particles in mine tailings, soils, lake sediments, windblown dust, and household dust. A near simultaneous combination of X-ray fluorescence, diffraction, and absorption experiments using a microfocused beam can provide information on elemental concentrations, crystal structure, and oxidation state of individual particles. This approach can distinguish multiple metal-hosting minerals and industrial compounds in a single sample. Our objective is to provide examples of the application of this technique to a range of materials representing potential risk to human or ecosystem health. These examples include arsenic-contaminated materials and metal-rich household dust. We have identified grains of scorodite and other arsenate minerals in mine tailings and associated airborne dust, arsenic trioxide in organic soils near an ore roaster, metallurgical products dispersed to the environment, and various metal-rich particles in household dust. A comparison of chemical analysis of individual particles using electron microprobe analysis and synchrotron-based X-ray fluorescence analysis is provided.

Background Arsenic is toxic to most living cells. The two soluble inorganic forms of arsenic are arsenite (+3) and arsenate (+5), with arsenite the more toxic. Prokaryotic metabolism of arsenic has been reported in both thermal and moderate environments and has been shown to be involved in the redox cycling of arsenic. No arsenic metabolism (either dissimilatory arsenate reduction or arsenite oxidation) has ever been reported in cold environments (i.e. < 10°C). Results Our study site is located 512 kilometres south of the Arctic Circle in the Northwest Territories, Canada in an inactive gold mine which contains mine waste water in excess of 50 mM arsenic. Several thousand tonnes of arsenic trioxide dust are stored in underground chambers and microbial biofilms grow on the chamber walls below seepage points rich in arsenite-containing solutions. We compared the arsenite oxidisers in two subsamples (which differed in arsenite concentration) collected from one biofilm. 'Species' (sequence) richness did not differ between subsamples, but the relative importance of the three identifiable clades did. An arsenite-oxidising bacterium (designated GM1) was isolated, and was shown to oxidise arsenite in the early exponential growth phase and to grow at a broad range of temperatures (4-25°C). Its arsenite oxidase was constitutively expressed and functioned over a broad temperature range. Conclusions The diversity of arsenite oxidisers does not significantly differ from two subsamples of a microbial biofilm that vary in arsenite concentrations. GM1 is the first psychrotolerant arsenite oxidiser to be isolated with the ability to grow below 10°C. This ability to grow at low temperatures could be harnessed for arsenic bioremediation in moderate to cold climates.

Bioaccessibility tests and mineralogical analyses were performed on arsenic-contaminated tailings and soils from gold mine districts of Nova Scotia, Canada, to examine the links between soil composition, mineralogy, and arsenic bioaccessibility. Arsenic bioaccessibility ranges from 0.1% to 49%. A weak correlation was observed between total and bioaccessible arsenic concentrations, and the arsenic bioaccessibility was not correlated with other elements. Bulk X-ray absorption near-edge structure analysis shows arsenic in these near-surface samples is mainly in the pentavalent form, indicating that most of the arsenopyrite (As1-) originally present in the tailings and soils has been oxidized during weathering reactions. Detailed mineralogical analyses of individual samples have identified up to seven arsenic species, the relative proportions of which appear to affect arsenic bioaccessibility. The highest arsenic bioaccessibility (up to 49%) is associated with the presence of calcium iron arsenate. Samples containing arsenic predominantly as arsenopyrite or scorodite have the lowest bioaccessibility (<1%). Other arsenic species identified (predominantly amorphous iron arsenates and arsenic-bearing iron(oxy)hydroxides) are associated with intermediate bioaccessibility (1 to 10%). The presence of a more soluble arsenic phase, even at low concentrations, results in increased arsenic bioaccessibility from the mixed arsenic phases associated with tailings and mine-impacted soils.

The mineral form, grain size and texture of As-bearing particles are important factors influencing the risk to human health associated with exposure to As-contaminated soils, sediments and mine wastes. Mining of arsenopyrite-bearing gold ores in Nova Scotia in the late 1800s and early 1900s has left a legacy of weathered, As-rich tailings deposits in more than 60 gold districts across the province. Fourteen samples of near-surface tailings and one of soil from several former gold mines frequented by the public were sieved to < 150 mu m and characterized using conventional mineralogical techniques (XRD, microscopy and EPMA) and synchrotron micro-analysis (mu-X-ray diffraction, mu-X-ray fluorescence and mu-X-ray absorption spectroscopy). Two high-As (> 20% As) mill concentrates exposed at the surface within the tailings deposits are dominated by a single As mineral, fine-grained scorodite (FeAsO4 center dot 2H(2)O) in one case, and massive unweathered arsenopyrite in the other. In the tailings (0.7 to 7% As), scorodite and amorphous hydrous ferric arsenate (HFA) are the most common As-bearing major components, occurring as discrete grains or grain coatings on gangue minerals. Other major As phases identified in the tailings include As-bearing amorphous hydrous ferric oxyhydroxides (HFO), kankite (FeAsO4 center dot 3.5H(2)O), pharmacosiderite [KFe4(AsO4)(3)(OH)(4)center dot 6-7H(2)O], yukonite [Ca7Fe12(AsO4)(10)(OH)(20)center dot 15H(2)O], amorphous Ca-Fe arsenates, and arsenopyrite. Minor or trace constituents include: As-bearing ferric oxyhydroxides with up to 10% As ( HFO, goethite, lepidocrocite and akaganeite), As-bearing sulfates (jarosite [(K,Na,H3O)Fe-3(SO4)(2)(OH)(6)], tooeleite [Fe-6(AsO3)(4)(SO4)(OH)(4)center dot 4H(2)O]) and realgar (As4S4). Arsenic-bearing HFO (2.5% As) and goethite (0.08% As) were identified in the single B-horizon soil sample. This study is part of a broader coordinated effort by a multi-department federal and provincial advisory committee formed to coordinate the study of ecosystem and human health risks associated with historical gold mine sites in Nova Scotia. Our study shows that (i) the mineralogy of As in weathered tailings is highly variable, with aggregates of more than one As-bearing phase common in a given sample, and (ii) major differences in As mineralogy in the tailings are mainly controlled by factors that influence the weathering history (e.g., presence or absence of mill concentrates, degree of water saturation, and abundance of relict carbonate minerals). The variable solubility of these primary and secondary As-bearing minerals influences both the environmental mobility and the bioaccessibility of As in near-surface tailings and soil samples.