Tailings and Mine Waste Conference

Application of high-resolution microscopy methods to the analysis of fine-grained and amorphous treatment.. Martin, Alan 2011

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


Martin_A_et_al_TMW_2011.pdf [ 1.36MB ]
JSON: 1.0107744.json
JSON-LD: 1.0107744+ld.json
RDF/XML (Pretty): 1.0107744.xml
RDF/JSON: 1.0107744+rdf.json
Turtle: 1.0107744+rdf-turtle.txt
N-Triples: 1.0107744+rdf-ntriples.txt

Full Text

Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 Application of High-Resolution Microscopy Methods to the Analysis of Fine-Grained and Amorphous Treatment Sludges Alan Martin, Skya Fawcett, Ezra Kulczycki Lorax Environmental Services Ltd., Vancouver, Canada Diana Loomer, Tom Al University of New Brunswick, Fredericton, Canada Andrew Rollo Canadian Environmental Assessment Agency, Vancouver, Canada Abstract Lime addition is a common method for the treatment of acid mine drainage (AMD) whereby neutralization promotes a reduction in acidity and the precipitation of metals as voluminous sludges that may contain gypsum, calcite and a spectrum of other phases. Due to the extremely fine-grained and often amorphous (i.e., non crystalline) character of sludge solids, the composition of these materials has been difficult to elucidate. Traditional methods, such as X-Ray Diffraction (XRD) and optical microscopy, have proved largely ineffective. In order to provide further insight into the solid-phase characterization of neutralization sludges, samples from mine sites across Canada were examined by high resolution microscopy techniques, including Scanning Electron Microscopy (SEM), Scanning Transmission Electron Microscopy (STEM), and X-ray Absorption Spectroscopy (XAS). The results revealed sludge-specific host phases, including relatively pure Fe- oxyhydroxide, amorphous Mg-Al-(Fe) hydroxysulphate and amorphous metal hydroxides. The data indicate that the nature of metal phase associations is strongly dependent on AMD influent composition. Introduction At many mine sites, the management of acid rock drainage (ARD) involves neutralization with lime to reduce levels of acidity and trace elements prior to discharge to the environment. In general, treatment is achieved by raising the pH to values greater than 8.5 and separating the resulting precipitates (sludge) from the treated water prior to discharge (MEND, 1994). Within Canada alone, it is estimated that approximately 7 million m3 of neutralization sludge are produced on an annual basis from mining/metallurgical operations (MEND, 1997). In general, trace-element associations (i.e., metal- bearing phases) within lime neutralization sludge are not well understood.  Due to the extremely fine- grained and often amorphous (i.e., non crystalline) character of sludge solids, metal associations have been difficult to elucidate. In particular, traditional methods such as optical microscopy and X-ray diffraction have largely proved ineffective. Given that neutralization sludges represent large inventories of metal-rich material, understanding sludge composition is prerequisite for assessing the long-term chemical stability in various depositional settings (pH and Eh dependent solubility). In this regard, high resolution microscopy methods offer an attractive means to elucidate trace metal associations (phase relationships) in sludge materials. As part of a study jointly funded by industry and the Mine Environment Neutral Drainage (MEND) program, several high-resolution microscopy methods were utilized to characterize the composition of ARD neutralization sludges. In this paper, the utility and application of these methods for the analysis of fine-grained and amorphous materials is discussed. The methods examined include Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 Scanning Electron Microscopy (SEM), Scanning Transmission Electron Microscopy (STEM), and X- ray Absorption Spectroscopy (XAS).  Data for the analysis of ARD neutralization sludges are presented for three mines in Canada. A primary objective of the analysis was to assess the links between sludge composition and ARD influent chemistry. Microscopy Methods Scanning Electron Microscopy Scanning Electron Microscopy (SEM) is used to gain both physical and chemical information of solid materials. Imaging on the SEM provides information on grain size, texture, and mineralogical associations. In addition, the relative greyscale intensity shown in backscatter electron images corresponds to the average atomic number such that bright areas indicate relatively high atomic numbers (e.g., CuS2) while darker areas indicate low atomic numbers (e.g., SiO2). Elemental analysis is conducted using point energy dispersive x-ray spectroscopy (EDS) and by producing 2-D elemental maps. Under optimized EDS experimental setups, the volume of beam interaction can be as small as 1 x 1 x 3 µm. Point EDS analyses produce spectra which are processed to reveal the elements that are present within the volume of beam interaction as well as their relative quantity. While not quantitative, the production of 2-D elemental mapping is useful in visualizing elemental distribution and elemental associations over a given area. In this study, electron microscopy was carried out at the University of New Brunswick Microscopy and Microanalysis Facility using a JEOL JSM6400 Scanning Electron Microscope equipped with an EDAX Genesis Energy Dispersive X-ray Spectroscopy system (EDS). Scanning Transmission Electron Microscopy Scanning Transmission Electron Microscopy (STEM) provides higher resolution information for both imaging and analysis in comparison to SEM, thereby improving the assessment of discrete trace element-mineral phase associations. It can also provide information concerning the mineral structure of individual grains. However, with the increase in resolution, the amount of sample material analyzed is consequently reduced. For example, the thickness of the STEM sample is only ~0.1 µm, compared to the 3 µm-thick SEM sample. STEM offers the ability for high resolution imaging (texture, grain size), selected-area electron diffraction (SAED), and qualitative EDS. When in STEM mode, a Gatan Annular Dark Field (ADF) detector, which produces High Angle Angular Dark Field (HAADF) images, is used for imaging. In these images, the brightness reveals information with respect to atomic number, sample thickness, and diffraction contrast (related to the crystallinity of the grains). Similar to SEM, point EDS analyses and 2-D elemental maps are possible using STEM. Unique to the STEM setup is the SAED component. SAED produces a pattern that can be used to derive d-spacings and hence mineral identification (see Figure 5). Also, the nature of the pattern (spotty, narrow smooth rings, broad diffuse rings, etc.) provides information on the degree of crystallinity of the mineral phase (i.e., coarsely crystalline, nanocrystalline, amorphous). STEM analysis was conducted at the University of New Brunswick Microscopy and Microanalysis Facility using a JEOL 2011 STEM. X-Ray Absorption Spectroscopy X-ray absorption spectroscopy (XAS) is a spectroscopic technique that uses x-rays to probe the chemical and physical structure around an element. Using this technique, x-rays penetrate the atom and in doing so, cause an electron close to the core of the atom to be ejected. The ejection of this core electron causes a photo-electron wave to emanate from around the atom (Figure 1). The nature of the photoelectron wave reveals information about the chemistry (what element it is) as well as its oxidation state (i.e. As3+ or As5+). Furthermore, as the photoelectron wave progresses outward from the atom it interacts with surrounding atoms and gets reflected. The nature of the photoelectron wave and the Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 reflected portions of the wave are recorded as a spectra. The spectra are processed to reveal the speciation of the element of interest, the distance to the nearest neighbouring elements, and the number of neighbouring elements. This information allows for the distinction between Cu adsorbed onto the surface of an Fe-oxyhydroxide and structurally-incorporated Cu, as in a Cu-oxyhydroxide, for example. Sludge samples are currently being investigated at the Canadian Light Source in Saskatoon, SK. Results were not available at the time of writing.  Figure 1: X-ray absorptions spectroscopy: Illustration (courtesy of the Canadian Light Source) of the interaction of the photoelectron wave with surrounding atoms and the resulting spectra produced. Applications In order to illustrate the applications of SEM and STEM to fine-grained and amorphous mine residues, results for the analysis of ARD neutralization sludges from three mine sites are presented: Equity Mine, British Columbia (Goldcorp Canada Ltd.), Geco Mine (Xstrata Zinc), and Britannia Mine, British Columbia (EPCOR). XAS data were not available at the time of writing. Equity Mine The Equity Mine was operated by Placer Dome Inc. between 1980 and 1994 and is located in the central interior of British Columbia, 35 km southeast of Houston. The site is currently under care and maintenance by Goldcorp Canada Ltd. The Equity Mine ARD is characterized by abundant sulphate, Al, Fe and Mg. The treatment of ARD follows the Heath Steele process which produces a high-density sludge (HDS) (Aubé, 2005). Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 SEM and X-ray element map analysis indicates that the main Cu- and Zn-bearing phase in the Equity Mine sludge is a Mg-Al-(Fe) hydroxysulphate (Figure 2), consistent with the abundant Mg and Al in ARD. In contrast, gypsum shows undetectable levels of Cu and Zn. The STEM analyses are consistent with the SEM results, indicating that Cu and Zn are primarily associated with the Mg-Al-(Fe) hydroxysulfate phase which is present as particles ranging from <1 to approximately 15 µm in size (Figure 3). While SEM indicates that Ca may also be an important component of the bulk sludge (Figure 2), the higher resolution STEM analyses indicate that Ca is not a major component of the Mg- Al-(Fe) hydroxysulfate phase (Figure 3). The relatively high Ca peaks in the SEM spectra are likely a result of the close spatial association of the Mg-Al-(Fe) hydroxysulfate phase with gypsum, rather than from the inclusion of Ca in the former. Such observations highlight the spatial-resolution limitations of SEM analysis. Selected Area Electron Diffraction (SAED) analysis did not produce any discernible diffraction patterns or rings, suggesting that the Mg-Al-(Fe) hydroxysulfate phase is amorphous.  Figure 2: Equity Mine HDS: SEM back scattered electron (BSE) images (left) and associated EDS spectra (right).  The inset spectra have been vertically expanded to show trace element peaks. a) gypsum blade with Al-hydroxide; b, and c) Mg-Al-(Fe) hydroxysulphate, illustrating the Cu and Zn associations.  Figure 3: Equity Mine HDS: STEM High Angle Angular Dark Field (HAADF) image (left) and associated EDS spectra (right) of metal-bearing Mg-Al-(Fe) hydroxysulfate.  The inset Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 spectra have been vertically expanded to show trace element peaks.  EDS spectra are shown for four points of the Mg-Al-(Fe) hydroxysulfate showing variable trace element associations. The Ni peaks result from the Ni support grid used in STEM analysis, and do not represent Ni content in the sample. Geco Mine The Geco Mine operated from  1957 to 1995 and is located 200 km East of Thunder Bay Ontario. Geco produced over 53 Mt of ores bearing Cu, Zn, Ag and Au (Jamieson et al., 1995). The site is currently being reclaimed under care and maintenance of Xstrata Zinc, whose major focus is the collection and treatment of ARD. The Geco deposit is a metamorphosed volcanogenic Cu-Zn-Ag (Au- Pb-Cd) massive sulfide deposit, similar to those in the Noranda district (Petersen, 1986). The HDS process at Geco involves lime neutralization of ARD, air oxidation, precipitation of excess gypsum and dissolved metals, flocculation, and solid/liquid separation. SEM analysis indicates that Zn in the Geco HDS is predominantly associated with Fe-dominant hydroxide (Figure 4). The Fe-oxyhydroxides occur mostly as well rounded grains, ranging in size from less than 1 µm up to approximately 20 µm in diameter. Other phases identified by SEM include dominant gypsum, Al-hydroxides and lime/calcite (Figure 4). Neither the gypsum, Al-hydroxides  nor calcite were shown to be significant repositories for Zn. Higher resolution EDS collected by STEM indicates that the Fe hydroxide phase is essentially pure with only minor amounts of other constituents such as Mg, Al and SO4 (Figure 5). This is consistent with the high Fe content of the ARD, and low Mg and Al. Selected area electron diffraction (SAED) of the Fe oxyhydroxide indicates the presence of broad diffuse rings, suggestive of poorly crystalline to amorphous mineral phases (Figure 6). D-spacings obtained from one SAED pattern are consistent with ferrihydrite [Fe2O3·0.5H2O], while d- spacings obtained from another SAED pattern are suggestive of lepidocrocite [FeO(OH)]. Collectively, these data indicate that while a majority of the Fe-hydroxide material present within the Geco sludge sample is amorphous, crystallization of Fe oxyhydroxides is occurring, potentially resulting from the lack of competition from other co-precipitating and competing metal phases (Aubé and Zinck, 1999). Britannia Mine Britannia Mine operated from 1902 to 1974 at Britannia, 48 km north of Vancouver, Canada. Approximately 47 Mt of predominantly Cu ore was processed during operations. Treatment of drainage water is being managed by EPCOR Utilities Incorporated. Britannia is a Kuroko type, volcanic massive sulfide deposit formed from hydrothermal solutions rich in Cu and Zn, and genetically related to dacitic volcanism. The ARD at Britannia hosts relatively low concentrations of sulphate, Fe, Al and Mg, and high levels of Cu and Zn. ARD is treated using a HDS water treatment system. From the SEM data, two major mineral phases in the Britannia HDS are apparent: 1) a Cu-Zn oxyhydroxide with associated Mg, Al and Si; and 2) a calcium-rich phase, likely calcite or Mg-calcite as indicated by XRD analysis.  The Cu-Zn oxyhydroxide is the dominant metal hosting phase, with the HDS containing 5 wt.% of both Cu and Zn. 2-D elemental maps obtained from EDS show Cu and Zn are present throughout the sample in both calcite (as indicated by the Ca map) and Cu-Zn oxyhydroxide (as indicated by the Cu and Zn maps) (Figure 7). It is apparent that the Cu-Zn oxyhydroxide and calcite phases are closely associated and, at the spatial resolution of the SEM, overlap of the phases is apparent in many of the EDS spectra. Fe oxyhydroxide grains were also identified although their low abundance precludes them from being significant repositories for trace elements. Examples of STEM HAADF images and point EDS analyses show that the Cu-Zn Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 oxyhydroxide forms fibrous grains, ranging in size from 5 to 100 µm (Figure 8). Selected area electron diffraction patterns collected from the Cu-Zn phases display diffuse and spotty rings, consistent with amorphous or nanocrystalline material.  Figure 4: Geco HDS: SEM BSE image (left) and associated EDS spectra (right).  The inset spectra have been vertically expanded to show trace element peaks.  These data illustrate that Zn is associated with Fe hydroxides (b) and not with gypsum (a).  Figure 5: Geco Mine HDS: STEM High Angle Angular Dark Field (HAADF) image (left) and associated EDS spectra (right) for points on Zn-bearing Fe oxyhydroxide.  The inset spectra have been vertically expanded to show trace element peaks.  Note that points (a) and (b) have almost identical Zn compositions. The Ni peaks result from the Ni support grid and do not reflect Ni content in the sample.  Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 Figure 6: Geco HDS: Selected area electron diffraction (SAED) pattern of Fe oxide.  The diffuse rings of the SAED pattern suggest the presence of poorly crystalline ferrihydrite or lepidocrocite.  Figure 7: Britannia Mine HDS: SEM BSE image (above left) and selected EDS 2-D elemental maps (elements indicated on the corresponding maps). Ca map shows the calcite distribution while Cu and Zn are indicative of Cu-Zn oxyhydroxide.  Note large particle of Cu- Zn oxyhydroxide at centre-bottom showing contrasting distributions of Cu (concentrated in particle rim) and Zn (dispersed evenly). Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011  Figure 8: STEM High Angle Angular Dark Field (HAADF)  image of Cu oxyhydroxide showing compositional zoning (left) and selected EDS spectra corresponding to points indicated (right).  The upper spectrum represents the exterior rim (Point V) while the low spectrum represents the particle interior (Point IV). The Ni peaks result from the Ni-TEM support grid and do not reflect Ni content in the sample. Conclusions High resolution microscopy methods such as SEM, STEM and XAS provide an effective means to characterize mine waste materials that show amorphous and fine-grained character. For the ARD neutralization sludges examined, all metal-bearing phases were shown to be amorphous (or poorly crystalline) and variable in composition (relatively pure to heterogeneous). The composition of the ARD is a dominant variable in governing sludge metal associations, with the proportions of Fe, Mg, Al and S being particularly relevant. Different metal-hosting phases can be expected to show contrasting chemical stabilities with respect to pH and Eh conditions. Results therefore highlight the need to understand controls governing long-term chemical stability of various sludge phases. Acknowledgements The authors wish to acknowledge the funding provided by the Mine Environmental Neutral Drainage (MEND) Program.  Particular thanks go to Gilles Tremblay and Charlene Hogan (Natural Resources Canada/MEND) for their coordination and support. Completion of this study would not have been possible without the support and cooperation of the participating mines. In this regard, we would like to acknowledge all supporters of this program, including: Mike Aziz (Goldcorp Inc., Equity Division); Manon Richard, Rick Schwenger, James Cormier and Robert Prairie (Xstrata Zinc); Christian Madsen (EPCOR Utilities Inc.); Walter Kuit, Dave Van Dieren and Bruce Dawson (Teck); Stephen West and James Dauk (Hudson Bay Mining and Smelting); and Brent Hamblin (Inmet). References Aubé, B., 2005. The Science of Treating Acid Mine Drainage and Smelter Effluents. Retrieved May 21, 2009 from:  http://www.robertsongeoconsultants.com/rgc_enviromine/publicat/treatment%20science.pdf Aubé B. and Zinck J.M., 2009. Comparison of AMD Treatment Processes and Their Impact on Sludge Characteristics. In: Goldsack D, Belzile N, Yearwood P, Hall G (eds) Sudbury '99, Mining and the Environment II., Sudbury, ON, Canada 13-17 Sept. 1999, pp 261-270. Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 Jamieson, H.E., Shaw, S.C., and Clark, A.H., 1995. Mineralogical Factors Controlling Metal Release From Tailings at Geco, Manitouwadge, Ontario. Paper presented at Sudbury ’95, Conference on Mining and the Environment, Sudbury, Ontario, May 28th – June 1, 1995. MEND, 1994. Acid Mine Drainage - Status of Chemical Treatment and Sludge Management Practices. Prepared for the Mine Environment Neutral Drainage (MEND) Program and the Canadian Centre for Mineral and Energy Technology (CANMET). Prepared by SENES Consultants Ltd. MEND Report 3.32.1. June 1994. MEND, 1997. Characterization and Stability of Acid Mine Drainage Treatment Sludges. Prepared for the Mine Environment Neutral Drainage (MEND) Program. Prepared by Zinck, J.M., Wilson, L.J., Chen, T.T., Griffith, W., Mikhail, S., and Turcotte. Mining and Mineral Science Laboratories. MEND Report 3.42.2a. May 1997. Petersen, E.U., 1986.  Tin in Volcanogenic massive Sulphide deposits: An Example from the Geco Mine, Manitouwadge District, Ontario, Canada. Economic Geology, 81, 323-342. 


Citation Scheme:


Usage Statistics

Country Views Downloads
United States 2 0
China 2 2
City Views Downloads
Beijing 2 2
Redmond 1 0
Sunnyvale 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}


Share to:


Related Items