No abstract

From 1861 to the mid-1940s, stamp milling at orogenic lode gold mines in Nova Scotia generated more than 3,000,000 tonnes of tailings. Most of the mined gold was recovered using mercury (Hg) amalgamation, and an estimated 10 - 25% of the Hg used was lost to the tailings and to the atmosphere. Arsenic (As) also occurs naturally in the ore, and is present at high concentrations in the mine wastes. Tailings from these operations were generally slurried into local rivers, swamps, lakes and the ocean. Recent land-use changes (e.g. residential development, recreational activities, shellfish harvesting) in some historical mining districts are increasing the likelihood of human exposure to these tailings. This Open File Report presents the results of a multi-disciplinary investigation of the dispersion, speciation and fate of metal(loid)s in terrestrial and shallow marine environments surrounding 14 abandoned gold mines in Nova Scotia. From 2003 to 2006, samples of tailings, sediment, and water were collected at 14 former gold mines. Field studies reveal that most mine sites contain large volumes of unconfined tailings, and in several districts these have been transported significant distances (>2 km) offsite by streams and rivers. Chemical analyses of 482 tailings and sediment samples show high concentrations of As (10 mg/kg to 31 wt.%; median 2550 mg/kg) and Hg (<5 ug/kg to 350 mg/kg; median 1640 ug/kg). Arsenic is hosted in arsenopyrite and a variety of secondary phases including scorodite (FeAsO4·2H2O), amorphous Fe arsenate, and As bound to Fe oxyhydroxides. Mercury is present in elemental form, amalgam (AuxHgx), and in secondary phases. Results from this study led to the formation of a Provincial-Federal Historic Gold Mines Advisory Committee in 2005, which has evaluated the ecological and human health risks associated with gold mines throughout Nova Scotia and developed recommendations for management of these tailings sites. This Open File Report provides the most comprehensive summary available of the history, distribution, and geochemistry of tailings at gold mines throughout Nova Scotia. The geographic coordinates provided for each district can be used to quickly explore the tailings deposits via most web-based mapping services. The results can be used to help minimize the environmental impacts associated with past, present, and future gold extraction and to inform land-use decisions.

No abstract

The final step in a sequential extraction procedure for Hg in geological samples usually involves a strong acid attack such as aqua regia. It is assumed that the Hg component measured in this step represents HgS (cinnabar), the common and highly insoluble form of Hg in nature. This paper describes the optimization of an HNO 3 -based dissolution of all non-sulphide forms of Hg while minimizing the solubilization of HgS. Previous work indicated that 12M (75% v/v) HNO 3 would fulfil that objective but its application in this work to samples containing very fine-grained HgS showed that this concentration was too strong as it partially dissolved HgS. Thus, HgS control samples were used to study the effect of HNO 3 concentration, duration of extraction and sample-to-volume ratio on the chemical breakdown of HgS. The recommended procedure to separate HgS from other forms of Hg is a two-hour extraction of 1 g of sample in 20 ml of 40% HNO 3 with constant agitation followed by a 10-ml rinse with 40% HNO 3 . This then ensures that all the cinnabar, fine- and coarse-grained, reports to the subsequent aqua regia step.

The objective of the work carried out was to recommend protocols for the collection, filtration (0.45 microm) and preservation of surface water samples for the subsequent determination of total 'dissolved' Hg. Cold vapour (CV) ICP-MS was employed to determine Hg; samples were acidified to a strength of 4 mol l(-1) HCl and 1% NaBH4 was used as the reducing agent in-line. Four types of 125 ml bottles were studied (Teflon, fluorinated ethene propene copolymer, FEP; high density polyethylene, HDPE; polyethylene terephthalate copolyester, PET; polypropylene, PP), together with three cleaning methods (EPA Methods 1631, 1638 and a rinse with reverse osmosis deionised water, 'MilliQ'). The transmission properties of the four materials were also studied to evaluate the potential for contamination from atmospheric Hg0. Results of this bottle study (n = 195), all below the detection limit of 0.5 ng l(-1), indicate that the bottles of choice, from an economic and time-saving perspective, are HDPE and PP, the latter being preferable if the sample is to be stored in a contaminated atmosphere. The bottles would be used on a once-only basis, negating the need for labourious and costly cleaning on repeat use. A simple rinse with MilliQ water would suffice prior to use. Twelve 0.45 microm filter systems (mostly Millipore and Gelman) were studied for (a) their potential Hg contamination properties and (b) their retention of Hg, possibly in colloidal form, during filtration. Ottawa River water, spiked at 50 ng l(-1) Hg, was used as a control sample. Again blank values were all negative, indicating contamination was not a concern but different recoveries of Hg were obtained across the different systems. The optimum systems to use, in that they provided maximum recovery (ca 80%) of Hg, are the Millipore Sterivex capsule and the Millipore Millex disc, both based on the hydrophilic Durapore membrane. The lowest recoveries (23-36%) were found with the Gelman AquaPrep systems and the Millipore HN nylon filter. Four control samples, Ottawa, Rideau and Gatineau Rivers, and a MilliQ blank, were used to study three preservation approaches, in the media: 0.5% BrCl, 2% HCl and 0.04% K2Cr2O7 in 1% HNO3. Mercury was stable for 28 days in a medium of 0.5% BrCl in all four samples but the other two media showed a loss of up to ca 20% Hg over this time frame, the loss being sample and time dependent. This may be a species transformation which is not detected by CV-ICP-MS from an HCl medium but this requires further investigation. Mercury was stable over the 28 days in the spiked blank MilliQ sample for all three preservation media.