UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Geochemistry of selenium release from the Elk River Valley coal mines Lussier, Christine 2002

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2002-0159.pdf [ 22.09MB ]
Metadata
JSON: 831-1.0081121.json
JSON-LD: 831-1.0081121-ld.json
RDF/XML (Pretty): 831-1.0081121-rdf.xml
RDF/JSON: 831-1.0081121-rdf.json
Turtle: 831-1.0081121-turtle.txt
N-Triples: 831-1.0081121-rdf-ntriples.txt
Original Record: 831-1.0081121-source.json
Full Text
831-1.0081121-fulltext.txt
Citation
831-1.0081121.ris

Full Text

GEOCHEMISTRY OF SELENIUM RELEASE F R O M THE E L K RIVER V A L L E Y COAL MINES by CHRISTINE LUSSIER B.Sc, McGill University, 1998 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Mining and Mineral Process Engineering) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA November 2001 © Christine Lussier, 2001 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my writ ten permission. Department The University of British Columbia Vancouver, Canada Date / I 2 / Q I DE-6 (2/88) 11 ABSTRACT Elevated levels of selenium (Se) were detected downstream from the five open pit coal mines in the Elk River Valley, British Columbia. Se is an essential nutrient but, in excessive amounts, it may cause teratogenic deformities and reproductive failure in fish and birds. To provide mine operators in the Elk River Valley with the information needed to assess the risk of Se release from waste rock and plant refuse a study of Se's modes of occurrence in the strata disturbed by mining and the geochemical mechanisms of its release was conducted. The mineralogical associations of Se were studied in 16 samples representative of the different types of material at the sites. Methods used to characterize sample mineralogy include X-ray diffraction, scanning electron microscopy, sequential extractions and heavy liquid separation. Se had both organic and inorganic associations in all lithologies tested, but sulphides, in particular pyrite, were indicated as the main Se-bearing component in the studied lithologies. The amount of organic matter in the materials appeared to play a role in determining the degree of Se enrichment in sulphides, with materials high in organics containing sulphides with less Se substitution. Humidity cells were used to determine the rate of Se release from coal, interburden, foot wall, parting and coarse refuse. The rate of Se release was not proportional to the total amount of Se in the sample, suggesting that mineralogical factors, such as texture, pyrite liberation and porosity, determine the rate of Se oxidation. A strong positive correlation between the amount of Se and sulphate in leachate from the humidity cells, suggested that sulphide oxidation is likely the source of Se being released into tributaries of the Elk River. Ill TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iii LIST OF TABLES vi LIST OF FIGURES ix ACKNOWLEDGEMENTS xii 1.0 INTRODUCTION 1 1.1 SELENIUM IN THE ELK RIVER V A L L E Y 1 1.2 OBJECTIVES 4 2.0 SELENIUM IN THE ENVIRONMENT 5 2.1 BIOGEOCHEMICAL CYCLE OF SELENIUM 5 2.1.1 Chemical Properties of Selenium 5 2.1.2 Selenium Occurrence 8 2.1.2.1 Selenium Associated with Organics 10 2.1.2.2 Selenium Associated with Sulphides 12 2.1.2.3 Selenium Associated with Clay Minerals 14 2.1.2.4 Selenium Associated with Hydrous Ferric and Manganese Oxides 17 2.1.3 Cycling of Selenium 18 2.1.3.1 Major Global Fluxes 18 2.1.3.2 Biogeochemical Transformations 20 2.1.3.3 Anthropogenic Inputs 23 2.2 SELENIUM IN BIOTA 25 2.2.1 Effects of Selenium 25 2.2.1.1 Humans 25 2.2.1.2 Livestock 26 2.2.1.3 Fish and Wildlife 27 2.2.1.4 Plants 30 2.2.2 Selenium Guidelines and Criteria 30 IV 3.0 ENVIRONMENT OF THE E L K RIVER V A L L E Y 33 3.1 BIOPHYSICAL ENVIRONMENT 33 3.1.1 Geology 33 3.1.2 Environmental Variables 36 3.2 MINING OPERATIONS 37 3.2.1 History 37 3.2.2 Current Mining Activity 37 3.2.2.1 Fording River 38 3.2.2.2 Greenhills 39 3.2.2.3 Coal Mountain 39 3.2.2.4 Elkview 39 3.2.2.5 Line Creek ! 40 3.3 SELENIUM DISTRIBUTION 41 3.3.1 Biota and Sediment of the Elk River and Tributaries 41 3.3.2 Selenium in the Mist Mountain Formation 43 4.0 METHOD DEVELOPMENT AND PROCEDURES 47 4.1 SAMPLING 47 4.2 SAMPLE SELECTION AND PREPARATION 54 4.3 MINERALOGY 56 4.3.1 Qualitative Mineralogy 56 4.3.1.1 X-Ray Diffraction 56 4.3.1.2 Scanning Electron Microscopy 58 4.3.2 Chemical Analyses 59 4.3.3 Quantitative Mineralogy 66 4.4 SEQUENTIAL EXTRACTIONS 68 4.4.1 Extraction Method for Exchangeable Selenium 70 4.4.2 Extraction Method for HFMO Associated Selenium 73 4.4.3 Extraction Method for Sulphide and Organic Matter Associated Selenium 75 V 4.4.4 Extraction Method for Selenium in Silicates 76 4.5 H E A V Y LIQUID SEPARATION 76 4.6 HUMIDITY C E L L TESTS 77 5.0 RESULTS AND DISCUSSION 83 5.1 CORRELATION OF SELENIUM CONCENTRATION WITH S A M P L E M I N E R A L O G Y 83 5.2 SEQUENTIAL EXTRACTIONS 90 5.3 H E A V Y LIQUID SEPARATION 97 5.4 HUMIDITY C E L L TESTS 99 6.0 CONCLUSIONS 114 7.0 RECOMMENDATIONS 118 8.0 ABBREVIATIONS 120 9.0 REFERENCES 121 10.0 APPENDICES 136 10.1 APPENDIX 1 X-Ray Diffractograms of the <2 um and <200 urn Fractions of 5 Samples 136 10.2 APPENDIX 2 Calculation of the Percent of Total Sample Mass Accounted for by Major Mineral Components 146 10.3 APPENDIX 3 Raw Data from the Sequential Extractions 148 10.4 APPENDIX 4 Raw Data from the Humidity Cell Tests 152 10.5 APPENDIX 5 Calculation of Acid-Generation and Neutralization Potential. 172 10.6 APPENDIX 6 Chemistry of Sulphide and Selenide Oxidation 173 V I LIST OF TABLES Table 2.1 Chemical Properties of Sulphur and Selenium 5 Table 2.2 Occurrence of Se in Various Lithologies 9 Table 2.3 Distribution of Se in Coal from the Powder River Basin, Wyoming 11 Table 2.4 Major Inventories of Se 18 Table 3.1 Key Water Quality Parameters 36 Table 3.2 Average Se in Different Lithologies of the Mist Mountain Formation 45 Table 4.1 Samples Included in the Geochemical and Kinetic Tests 55 Table 4.2 Minerals in <200 mesh Fraction Determined by X-Ray Diffraction 58 Table 4.3 Minerals in <2 um Fraction of Samples Selected for 20-Weeek Humidity Cell Tests Determined by X-Ray Diffraction 58 Table 4.4 Interlaboratory Comparison of Se Values Expressed in mg/kg 61 Table 4.5 Selenium, Sulphur, Sulphides, Carbon and Ash Analyses 63 Table 4.6 Oxide Concentrations Expressed as a Percentage of Total Sample Weight 64 Table 4.7 Trace Element Concentrations Expressed in mg/kg 64 Table 4.8 Oxide Concentrations in the <2 um Fraction Expressed as a Percentage of Total Sample Weight 65 Table 4.9 Trace Element Concentrations in the <2 um Fraction Expressed in mg/kg 65 Table 4.10 Se Concentrations in the <2 um Fraction Compared with Se Concentrations in the Whole Sample 65 Table 4.11 Percentages of the Major Mineral Oxides 67 Table 4.12 Summary of Selected Sequential Extraction Procedures Described in Literature .. 69 Table 4.13 Se Extracted from Sample E99-22 (Foot Wall) Expressed in ug/L 72 Table 4.14 Results from Time Trials for the Extraction of the Water Soluble Phase from Sample E99-22 (Foot Wall) 73 Table 4.15 Results from Time Trials for the Extraction of the Sulphide and Organic Matter Associated Se from Sample E99-22 (Foot Wall) with KC103/HC1 and H 2 0 2 76 V l l Table 4.16 Samples Included in the Humidity Cell Tests 78 Table 5.1 Correlation Coefficients for Se versus Various Trace Elements 84 Table 5.2 Amount of Se in Sequential Extraction Leachate and Residue as a Percent of the Total Amount of Se Extracted 91 Table 5.3 Correlation Coefficients Relating Se and Other Elements in Leachate from the Water Soluble, HFMO and Sulfide/Organic Matter Fractions of A99-4, A99-23, B99-16, B99-44, CREF, E99-19, E99-59 and E99-61 93 Table 5.4 Correlation Coefficients Relating TOC, Sulphides and Se in the Solids and Se Solubilized by Water for 16 Samples 94 Table 5.5 Se Concentrations and Percent Distribution in the Different Density Fractions .. 97 Table 5.6 Se Concentrations in mg/L in Humidity Cell Leachate in 20-Week Test 100 Table 5.7 Se to Sulphide Ratios in the Materials Used for the Humidity Cell Tests 107 Table 5.8 Peak Se Release in the First 3 Weeks as a Percent of Total Se Release from Humidity Cells in 20 Weeks 108 Table 5.9 Weekly Se Release after the First 3 Weeks in ug and as a Percent of Total Se in the Head Samples 108 Table 5.10 NP and AP Values and Neutralization Potential Ratios of the Five Materials Included in the Humidity Cell Tests 109 Table 5.11 Percent of Water Soluble Se in the Sequential Extraction Compared with Percent of Total Se Represented by Peak Extraction from the Humidity Cells 112 Table A2.1 Formulas Used to Calculate the Percent of Total Mass Accounted for by Major Mineral Components 146 Table A3.1 Se Concentrations in Leachate and Solid Residue Collected from the Sequential Extractions 147 Table A3.2 Variation Between Se Concentrations in Extractant and Residue Duplicates .. 147 Table A3.3 Concentrations of Trace Elements Associated with the Water Soluble Phase in ug/L 148 Table A3.4 Measured Se Concentrations in Head Samples versus Total Se Calculated from the Sequential Extraction Results 149 Table A3.5 Concentrations of Trace Elements Associated with the HFMO Phase in ug/L .. 150 Table A3.6 Concentrations of Trace Elements Associated with the Sulphide/Organic Phase in (ig/L 151 viii Table A4.1 Analysis of Leachate from the E99-61A (Interburden) Humidity Cell 152 Table A4.2 Analysis of Leachate from the E99-61B (Interburden) Humidity Cell 154 Table A4.3 Analysis of Leachate from the A99-23A (Parting) Humidity Cell 156 Table A4.4 Analysis of Leachate from the A99-23B (Parting) Humidity Cell 158 Table A4.5 Analysis of Leachate from the CREFA (Refuse) Humidity Cell 160 Table A4.6 Analysis of Leachate from the CREFB (Refuse) Humidity Cell 162 Table A4.7 Analysis of Leachate from the B99-44A (Coal) Humidity Cell 164 Table A4.8 Analysis of Leachate from the B99-44B (Coal) Humidity Cell 166 Table A4.9 Analysis of Leachate from the E99-19A (Foot wall) Humidity Cell 168 Table A4.10 Analysis of Leachate from the E99-19B (Foot wall) Humidity Cell 170 Table A5.1 Relative Reactivity of Minerals at pH 5 172 ix LIST OF FIGURES Figure 2.1 Eh-pH Diagram for Se 6 Figure 2.2 Proposed Classification Scheme for Se Occurrence in Coal 10 Figure 2.3 A Schematic Presentation of Cations in the Electric Double Layer 15 Figure 2.4 Inner- and Outer-Sphere Complexes Formed by Se Oxyanions 16 Figure 2.5 pH-Dependent Charge on Broken-Bond Aluminosilicate Surfaces 17 Figure 2.6 Generalized Cycle of Se 19 Figure 2.7 The Biogeochemical Cycling of Se in Aquatic Habitats 22 Figure 2.8 Typical Dose-Response Curve for Micronutrients 26 Figure 3.1 Elk Valley, Crowsnest and Flathead Coalfields of Southeastern British Columbia 34 Figure 3.2 Jurassic-Cretaceous Stratigraphy of the Kootenay Group 35 Figure 3.3 Total Se in Tributaries of the Elk River 42 Figure 3.4 Cross-section of Pit Wall Illustrating the Different Types of Material Sampled. 44 Figure 4.1 Sample Collection from Pit Walls 47 Figure 4.2 Stratigraphic Section of Mine A Illustrating Variations in Se Concentration with Depth 48 Figure 4.3 Stratigraphic Section of Mine B Illustrating Variations in Se Concentration with Depth • 49 Figure 4.4 Stratigraphic Section of Mine C Illustrating Variations in Se Concentration with Depth 50 Figure 4.5 Stratigraphic Section of Mine D Illustrating Variations in Se Concentration with Depth 51 Figure 4.6 Stratigraphic Section of Mine E (East Pit) Illustrating Variations in Se Concentration with Depth 52 Figure 4.7 Stratigraphic Section of Mine E (West Pit) Illustrating Variations in Se Concentration with Depth 53 Figure 4.8 Sample preparation 56 Figure 4.9 Sequential Extraction Procedure 70 X Figure 4.10 Heavy Liquid Separation Procedure 77 Figure 4.11 Humidity Cell Diagram 79 Figure 4.12 Humidity Cells in Operation 79 Figure 5.1 Concentration of Se versus Concentration of TOC 85 Figure 5.2 Concentration of Organic S versus Concentration of TOC 85 Figure 5.3 Concentration of Sulphides versus Concentration of TOC 86 Figure 5.4 Concentration of Total Sulphur versus Concentration of Organic Sulphur 86 Figure 5.5 Concentration of Total Sulphur versus Concentration of Sulphides 88 Figure 5.6 Concentration of Se versus Concentration of Sulphides 88 Figure 5.7 Concentration of Se versus Concentration of Sulphides Normalized for TOC .. 89 Figure 5.8 Percent of Total Se Accounted for by Se in the Water Soluble, HFMO, Sulphides/Organic Matter and Residual Fractions 92 Figure 5.9 Percent of Sulphide/Organic Associated Se versus TOC Concentration in Solids 95 Figure 5.10 Percent of Sulphide/Organic Associated Se versus Sulphide Concentration in Solids 95 Figure 5.11 Incorporation of Sulphur in Coal 96 Figure 5.12 Se Concentration in Humidity Cell Leachate in mg/L in 20-Week Test 101 Figure 5.13 Amount of Se in Humidity Cell Leachate in mg in 20-Week Test 101 Figure 5.14 Peak Se Release from the Humidity Cells versus Sulphide Content 103 Figure 5.15 Se versus Sulphate Concentrations in Leachate from the Five Humidity Cells. 104 Figure 5.16 Se versus Sulphate Concentrations in Leachate from the E99-61 (Interburden) Humidity Cells 104 Figure 5.17 Se versus Sulphate Concentrations in Leachate from the A99-23 (Parting) Humidity Cells 105 Figure 5.18 Se versus Sulphate Concentrations in Leachate from the CREF (Refuse) Humidity Cells 105 Figure 5.19 Se versus Sulphate Concentrations in Leachate from the B99-44 (Coal) Humidity Cells 106 xi Figure 5.20 Se versus Sulphate Concentrations in Leachate from the E99-19 (Foot Wall) Humidity Cells 106 Figure 5.21 Percent of Se Extracted from the 20-week Humidity Cell Tests 110 Figure 5.22 Percent of Total Se Extracted During the 20-week Humidity Cell Tests 110 Figure A 1.1 X-Ray Diffractogram for the <200 um Fraction of Interburden Sample E99-61.136 Figure A1.2 X-Ray Diffractogram for the <2 um Fraction of Interburden Sample E99-61.... 13 7 Figure A1.3 X-Ray Diffractogram for the <200 um Fraction of Parting Sample A99-23 138 Figure A1.4 X-Ray Diffractogram for the <2 um Fraction of Parting Sample A99-23 139 Figure A1.5 X-Ray Diffractogram for the <200 um Fraction of Refuse Sample CREF 140 Figure A1.6 X-Ray Diffractogram for the <2 um Fraction of Refuse Sample CREF 141 Figure A l .7 X-Ray Diffractogram for the <200 pm Fraction of Coal Sample B99-44 142 Figure A l .8 X-Ray Diffractogram for the <2 um Fraction of Coal Sample B99-44 143 Figure A l .9 X-Ray Diffractogram for the <200 um Fraction of Foot Wall Sample E99-19.. 144 Figure ALIO X-Ray Diffractogram for the <2 pm Fraction of Foot Wall Sample E99-19 .... 145 X l l ACKNOWLEDGEMENTS Many thanks are extended to Fording Coal Ltd., Teck Corporation and Luscar Ltd. and the National Science and Engineering Research Council for funding this project. A number of people have played a crucial role in the realization of this project. In particular, I would like to thank my advisors, Marcello Veiga and Susan Baldwin, who generously made time in their busy schedules for insightful discussions throughout the project, as well as Stephen Day and Rob Bowell of SRK for offering their advice on methodology and for critically reviewing reports. I would also like to thank Brenda Dixon, Mark Graham, Roger Berdusco, Billie O'Brien, Matt Cole, Scott Dressier, Ron Jones, Bill Kovach, Bob Logan and Jim Lant, and staff at the five mines. Their assistance with project logistics, sampling and obtaining information about the sites was much appreciated. Barry Ryan of the Ministry of Energy and Mines deserves special thanks for coordinating sample collection, helping with data interpretation and supplying enough coal- and selenium-related information for several theses. I also extend warmest thanks to Maggie Dittrick, for her help with sampling and crushing. I gratefully acknowledge Sally Finora, Pius Lo, Frank Schmidiger and Larry Wong, in the Mining and Mineral Process Engineering Department, as well as Matti Raudsepp and Elizebetta Pani, in the Geology Department, for their tireless support in the labs. Maria Holuszko, Marek Pawlik, Sam Cho and Jennifer Hinton also provided valuable advice and assistance. Last, but not least, I would like to thank my family and friends for their encouragement. 1 1.0 INTRODUCTION 1.1 SELENIUM IN THE E L K RIVER V A L L E Y In 1995, selenium (Se) concentrations as high as 25 ug/L were detected during a water quality assessment for an effluent permit amendment at one of the five coal mines in the Elk River Valley. This was well in excess of the current Canadian water quality guideline of 1 ug/L total Se (CCME, 1999), which includes both dissolved and particulate-bound Se. This prompted a review of water quality data collected between 1984 and 1994 from the mouth of the Elk River, 65 km downstream of the coal mines. Reexamination of the data revealed that total Se had increased from 0.5 ug/L to 2.0 ug/L over that period (Wipperman and Webber, 1997). These values were the highest among those recorded at sampling sites on 30 rivers across British Columbia over the same time period. Se is an essential trace element required for the elimination of tissue damaging free radicals. However, in excessive amounts, Se may cause reproductive failure and teratogenic deformities in fish and birds. A more detailed study of Se in the Elk River and its tributaries was thus undertaken (McDonald and Strosher, 1998). McDonald and Strosher (1998) surveyed Se levels in the water column, sediment and biota. The study indicated that Se concentrations in unfiltered water samples from waterways upstream from the mines were consistently below the 1 ug/L detection limit, while those in the major rivers downstream from the mines ranged between 2 and 20 ug/L. The highest concentrations were recorded at sampling sites closest to the mines. 92 % of the Se was in the dissolved form (McDonald and Strosher, 1998). Se levels remained relatively constant throughout the year despite a tenfold increase in stream flow during the spring, indicating that proportionally greater quantities of Se were mobilized during this period. In sediment, Se concentrations were consistently below the 5 mg/kg provincial criteria for the protection of aquatic wildlife (Nagpal et al., 1995) likely due to the fast flow of water in the Elk River and its tributaries. This limits the extent of Se bioaccumulation as it minimizes opportunities for plants and microorganisms to fix 2 Se from sediment. Se concentrations in algae, aquatic insects and fish tissues downstream from the mines were nonetheless 2 to 5 times greater than at the reference sites. Se levels in some of the westslope cutthroat trout caught below the Fording River mining operations exceeded the published toxic effects threshold of 8 and 12 ug Se/g dry weight for muscle and liver respectively (McDonald and Strosher, 1998; Lemly, 1993). Although McDonald and Strosher's (1998) study revealed no toxicological effects in the Elk River's cutthroat trout population, there is a need to identify the source of the Se and to better understand the biogeochemical cycling of Se in this environment in order to determine the potential hazard to aquatic life. In 1999, the Ministry of Energy and Mines conducted a comprehensive geochemical study of Se levels in all exposed strata in the coal-bearing Mist Mountain Formation at the Elkview, Line Creek, Coal Mountain, Greenhills and Fording mines (Ryan and Dittrick, 2000). Results from this study and data from earlier studies (Goodarzi, 1988; Grieve and Goodarzi, 1994) show that the lithologies closely associated with the coal, i.e. partings, hanging wall and foot wall, contain the highest amounts of Se. The data show that Se is found at levels of approximately 1.1 mg/kg in the sand- and siltstone layers between the coal seams (interburden). Coal seams contain, on average, 1.9 mg/kg Se while material immediately above and below the seams (hanging wall and foot wall) contains 4.2 mg/kg Se and bands of mineral matter in the seams (partings) 3.2 mg/kg. Estimates by Ryan and Dittrick (2000) suggest that 80% of the Se is found in the interburden, 6 to 20% in the coal and 5 to 10% in hanging wall, foot wall and parting materials. The extraction and combustion of coal has been associated with Se mobilization in numerous instances. Specific examples include elevated Se levels detected in groundwater at a surface coal mine in Wyoming (Dreher and Finkelman, 1992) and in runoff from fly ash dumps in Texas and North Carolina (Shepard, 1987). Studies by Naftz and Rice (1988) and Dreher and Finkelman (1992), looking at the modes of occurrence of Se in coal, suggest that the bulk of it appears to be associated with the organic fraction. Se has also been shown to substitute for sulphur in sulphides present in coals (Clarke and Sloss, 1992; White et al., 1989). When assessing Se mobilization by surface- or groundwater from coal mines, the critical factor is Se's mode of occurrence in rocks adjacent to coal seams rather than in coal seams. There have been no studies looking at the modes of occurrence of Se in these lithologies. Information about the rate of Se release from these materials is also lacking. The aim of this thesis was to address these gaps in our understanding of Se mobilization from coal-bearing geological formations in order to identify potential sources of Se to the Elk River. This study had three main components. The first focused on sample mineralogy. Coal, parting, hanging wall, foot wall, interburden and refuse samples were characterized by means of x-ray diffraction and scanning electron microscopy, as well as by chemical analyses. The second section explored the modes of occurrence of Se in the same set of samples using a four-step sequential extraction procedure and heavy liquid separation. The latter was used to determine the relative importance of organic matter- and sulfide-associated Se since this could not be effectively achieved with the sequential extractions. Lastly, humidity cells were used to investigate the rate of Se release from coal, parting, foot wall/hanging wall, interburden and refuse over the course of a twenty-week leaching period. Together, results from these experiments were used to establish the amount of Se in each of the different fractions and to highlight potential trends in geochemical factors influencing Se release. 4 1.2 OBJECTIVES The objectives of this study were to: ; • Identify Se-bearing mineralogical components in the Mist Mountain Formation. • Evaluate the rate of Se release from the different lithologies, so as to pinpoint potential sources of Se to the tributaries of the Elk River. • Suggest possible biogeochemical mechanisms of Se mobilization. • Provide mine operators in the Elk River Valley with information needed to assess the risk of Se release from waste rock and plant refuse. 5 2.0 S E L E N I U M IN T H E E N V I R O N M E N T 2.1 B I O G E O C H E M I C A L C Y C L E O F S E L E N I U M 2.1.1 Chemical Properties of Selenium J J Brezelius discovered Se in 1818 as a residue from the oxidation of chalcopyrite in producing sulfuric acid. Atomic number 34, Se is located between sulfur and tellurium in group VIA of the periodic table and between arsenic and bromine in the fourth period. A list of the key chemical and physical properties of Se and sulphur (S) is provided in Table 2.1. Table 2.1 Chemical Properties of Sulphur and Selenium (Adapted from Kudriavtsev, 1974) S Se Atomic weight 32.06 78.96 Atomic number 16 34 Naturally occurring isotopes 32s, 33s,34s,36s, 67Se, 74Se, 77Se, 78Se, 80Se, 82Se Covalent radius 1.04 1.17 Ionic radius (A) 1.86 1.98 Ionization potential, eV 10.36 9.75 Electron affinity, eV 2.33 4.21 Electronegativity 2.53 2.55 Oxidative states -2, 0, +2,+3, +4, +6 -2, 0, +4, +6 There is considerable overlap of the biogeochemical cycles of S and Se due to the similarity of their physico-chemical properties. Se is readily taken up by organisms via channels intended for S (Milchunas & Lauenroth, 1984) and readily substitutes for S in sulphides. There are, however, a few important differences between Se and S chemistry that affect its occurrence and mobility. Se is less volatile than S and it does not share S's tendency to catenate, meaning that it will not form compounds like S's thiosulphate (S2032~) (Herring, 1990). Se compounds are more easily reduced than their S analogs (Zehr and Oremland, 1987). Hence, S is more readily mobilized than Se in oxidizing environments and the co-occurrence of similar S and Se compounds can lead to the oxidation of S via the reduction of Se. 6 Se has four oxidation states: selenide (-2), elemental Se (0), selenite (+4) and selenate (+6), all four of which can be found in soils and sediment. The partitioning between these species is, in large part, defined by the redox conditions of the depositional environment and, to a certain extent, by microbial activity and the availability of complex-forming species (McNeal and Balistrieri, 1989). Figure 2.1 illustrates the fields of stability of predominating solid and dissolved species of Se. The area enclosed within the dashed line shows the range of Eh-pH conditions commonly found in coal swamps. Under these conditions, Se would be found in sediment as Se° or as Se"2 or in solution as Se+4. 2.0 -l.o -I , 1, , , , , 1 0 2 4 6 8 10 12 14 p H »s /' Common Eh-pH range in coal swamps Figure 2.1 Eh-pH Diagram for Se at 25°C, 1 bar pressure and I = 0 for a dissolved Se activity of 10"6 mol/L (Adapted from Seby et al., 2001) Se"2 is the most reduced form of Se and is found in compounds such as H2Se and Fe2Se. Due to the similarity of the ionic radii of selenide ions (1.98 A) and sulphide ions (1.86 A), Se"2 readily substitutes for S"2 in sulphide minerals (Fischer and Zemann, 1978). Se forms selenides with Ag, 7 As, Au, Bi , Cd, Co, Cu, Fe, Hg, Ni, Pb, Sb, Tl and Zn (Fischer and Zemann, 1978). The amount of substitution varies, occurring in decreasing order in chalcopyrite (CuFeS2), arsenopyrite (FeAsS2), sphalerite (ZnS), pyrite (FeS2), pyrrhotite (Fe,.xS) and galena (PbS) (Lakin, 1973). The Se concentration in sulphides averages 88 mg/kg (Badalov et al., 1970), but ranges from 0 to 15 mg/kg in PbS to more than 2000 mg/kg in some CuFeS2 samples (Fleischer, 1959). Metallic selenides are sparingly soluble (Elrashidi et al., 1987) and are not rapidly oxidized (Masscheleyn et a l , 1990). Elemental Se is allotropic and exists in either an amorphous or crystalline state (Kudriavtsev, 1974). Amorphous selenium is liquid at temperatures above 230°C. Below 31 °C elemental Se is a hard and brittle glass. Naturally occurring amorphous Se is transformed into the hexagonal crystalline form at 70°C, whereas man-made amorphous Se crystallizes at 60°C. The three crystalline forms of Se are alpha-monoclinic, beta-monoclinic, and hexagonal. Both monoclinic forms are red in colour, but the alpha form consists of flat hexagonal and polygonal crystals whereas beta-monoclinic ones are needle-like or prismatic. Hexagonal selenium is grey or black and is composed of spiral Se chains. This is the most stable form of elemental Se with both alpha and beta monoclinic Se converting to the hexagonal form at temperatures above 110°C. Elemental Se is sparingly soluble and both its oxidation and reduction kinetics are slow (Herring, 1990). In its +4 oxidation state, Se exists as crystalline Se dioxide (Se02), selenious acid (H2Se03), or selenite (Se0 3 2) salts. Se+4 is readily reduced to elemental Se in the presence of reducing agents or certain bacteria (Herring, 1990). When amorphous selenium is oxidized in the presence of water, H 2Se0 3 is formed. The latter is a weakly dibasic acid that frequently acts as an oxidizing agent. Se+4 adsorbs readily to hydrous and ferric manganese oxides (HFMO) as Se03"2 (Ballistrieri and Chao, 1987). Naturally occurring selenite salts include those formed with Ag, Ba, Be, Ca, Cd, Cu, Fe, Hg, K, Mg, Na, Ni, Pb and Zn (Kudriavtsev, 1974). 8 At high redox potentials, Se+ 6 predominates. Potassium and sodium selenate (Se042") salts are rare, but do occur in nature (Emerick and DeMarco, 1990). They are appreciably more soluble than the corresponding sulphate compounds (Seby et al., 2001) and do not bind to HFMO as readily as selenites (Ballistrieri and Chao, 1987). Se+6 is thus common in aqueous environments and is readily available for uptake by plants and microorganisms. The conversion of selenates to the less stable selenites is very slow (Bar-Yosef and Meek, 1987). Selenium reacts with halogens to form gaseous halides in which Se+4 or Se+6 are found (i.e., SeF6, SeF4, SeCl4, SeBr4) (Bar-Yosef and Meek, 1987). Selenium halides form acid complexes with the halogen derivatives of acids and with some of their salts. Se occurs in several organic forms, including dimethylselenide and dimethyldiselenide, organic selenides, seleno-carbohydrates, seleno-amino acids and selenopeptides (Herring, 1990). The two methylated forms are volatile. A number of these organic compounds are synthesized in plants, while others are produced as by-products of microbial metabolic processes. Organo-selenium compounds in soil and sediment are readily available to plants (Fleming, 1962). 2.1.2 Selenium Occurrence Taylor (1964) estimated the crustal average of Se to lie between 0.05 and 0.1 mg/kg (Table 2.2). Se concentrations of in sedimentary rocks range from below 0.1 mg/kg in sandstones and limestones to 0.6 mg/kg in shales, though concentrations as high as 100 mg/kg have been observed in some shales (Herring, 1990). Rosenfeld and Beath (1964) found concentration of Se ranging from 1 to 300 mg/kg in phosphatic rocks in the North Western United States, but did not observe a direct correlation between Se and P 2 0 5 concentrations. 9 Table 2.2 Occurrence of Se in Various Lithologies Material Range (mg/kg) Mean (mg/kg) Reference Crustal 0.05-0.1 - 1 Sandstones <0.1-1.7 0.10 3 Limestone <0.1-7.4 0.22 3 Igneous rocks 0.09-1.08 0.35 2 Shales <0.1-12 0.53 3 Mudstones 0.4-0.6 - 4 Oil 0.01-1.4 - 4 Coal (global) <0.1 2.15 4 1 Taylor, 1964 2 Neal, 1995 3 Connor and Shacklette, 1975 4 Coleman et al., 1993 Swaine (1990) includes Se among the trace elements in coal of greatest environmental concern, along with arsenic, boron, cadmium, mercury, molybdenum and lead. It is the most enriched trace element in coal (Coleman et al., 1993; Spears and Zheng, 1999). Se can be found in coal at up to 82 times its crustal concentration (US National Committee for Geochemistry, 1980). The reduction of soluble selenate and selenite to selenide under low oxygen conditions created by the decomposition of organic matter in coal swamps is likely responsible for the high Se levels observed in coal and shales (Dale, 1996). The global average Se content of coal is 2.15 mg/kg (Coleman et al., 1993). Se concentrations in American coals average 3.3 mg/kg (Pillar et al., 1969) while those in the Mist Mountain Formation of the Elk River Valley average 1.9 mg/kg (Ryan and Dittrick, 2000). The highest reported concentration of Se in coal is 8400 mg/kg in a sample from a mine in China (Yang et al., 1983). The mode of occurrence of the Se in this sample was not specified. Se has both organic and inorganic associations in coal (Figure 2.2). Like S, Se can be covalently bound within the molecular structure of organic material or ionically bound to its surface. Covalent bonds, such as those formed between C and Se, involve interatomic sharing of electrons, while ionic bonds, like those between Se and O in hydroxyl or carboxyl groups of organic matter, involve the complete transfer of one or more electrons from one atom to another. 10 Alternatively, Se may substitute for S in sulphide minerals, occur as selenides or, in the form of selenate or selenite, be bound to aluminosilicates or hydrous ferric and manganese oxides (HFMO). Se in Coal Organic/Maceral Association Covalently Bound Ionically Bound Dispersed in Minerals Inorganic/Mineral Association Forming Discrete Mineral Bound to Clays or H F M O Figure 2.2 Proposed Classification Scheme for Se Occurrence in Coal (Adapted from Dale, 1996) 2.1.2.1 Selenium Associated with Organics Combustion experiments by Finkelman et al. (1990) showed that during ashing at 500°C, 80-95% of Se in Argonne Premium Coal samples volatilized. Combining results from high temperature ashing with those from sequential extraction and heavy liquid separation experiments, Dreher and Finkelman (1992) tried to establish the relative importance of inorganic and organic associations of Se in coal. Noting its high volatility and its low susceptibility to leaching by water or weak acids, they concluded that the bulk of the Se was associated with the organic fraction (Table 2.3). Coleman et al. (1993) reviewed data on 9000 U.S. coals and found a positive correlation between Se concentration and total S, pyritic S and organic S. In non-marine coals, Se levels correlated strongly with ash yield, suggesting that inorganic associations were of primary importance. 11 Table 2.3 Distribution of Se in Coal from the Powder River Basin, Wyoming (From Coleman et al., 1993, adapted from Dreher and Finkelman, 1992) Phase with which Se is associated %ofTotalSeinCoal Water soluble 0-15 Ion exchangeable 0-15 Pyrite 5-10 Other sulfides and selenides 1-5 Organics >60 It should be noted that when using heavy liquid separation, the term "organically associated" does not necessarily imply that a chemical bond with organic matter exists. A portion of the trace elements found in the light (specific gravity <1.5 mg/cm3) organic fraction may be bound to micron-sized inorganic particles entrapped in the organic matter (Swaine, 1990; Palmer et al., 1990). In the context of. this study though, the term "organically associated" will be used to designate elements or compounds covalently or ionically bound to organic matter, excluding any mineral matter that may be intimately associated with it. Before considering how Se is associated with coal, a basic understanding of the composition of coal itself is required. The process of coalification involves a series of bio- and physico-chemical transformations that result in the progressive decrease of moisture, volatiles, oxygen and hydrogen content of organic matter deposited in the coal swamp. Coal rank is defined by the degree of coalification, increasing from peat, lignite, sub-bituminous, bituminous through to anthracite (Stach et al., 1982). Biochemical transformations, often bacterially-mediated, end at the rank of sub-bituminous coal when humic colloids have polymerized. Increased pressure and heat caused by deep post-depositional subsidence and/or tectonic activity further increases the aromaticity of the organic molecules. Coal organic matter thus consists of a complex matrix of aromatic structures with a variable number of aliphatic groups attached as side chains or as links between aromatic nuclei (Stach et al., 1982). Macerals are microscopic units in coal analogous to minerals in rocks. They lack the ordered structure of minerals and their physico-chemical 12 properties vary with coal rank. They are divided into three groups based on their parent material. Vitrinite generally stems from woody plant material and liptinite from spores, cuticles and resins, while inertinite forms from vegetation that has been oxidized prior to coalification. Demir and Harvey (1991) found that both S and Se tend to be enriched in vitrinite macerals. Ryan and Dittrick (2000) hypothesized that this might be due to the volatilization of Se from inertinite macerals during charring of the vegetation. Noting Se's tendency to substitute for S, much can be deduced about the occurrence of Se in coals by looking at the information available on S. Considering first organic associations, S can be incorporated into phenolic sulphates and sulphated polysaccharides by ester bonds and into amino acids, cysteine and methionine, by carbon bonds (Casagrande and Siefert, 1977). Ester sulphate bonds are readily reduced to H 2S, but mercapto (-SH) groups persist to the bituminous stage of coalification (Saxby, 1973). A second type of organic complexing involves the formation of ionic bonds between selenate or selenite molecules and carboxylic acid groups on the surface of organic matter. For such bonds to form, the pH must be low enough that carboxyl groups will be protonated and bear a positive charge. A more detailed explanation of pH-dependent charge is provided in section 2.1.2.3. Both Se incorporated into the organic matrix and Se bound to its surface contribute to total Se during the gelification-humification stage of coalification. The importance of surface-bound Se decreases with increasing rank, since carboxylic acid groups do not persist much beyond the lignite stage (Schafer, 1984). 2.1.2.2 Selenium Associated with Sulphides Sulphides in coals are found in a range of forms and sizes and may be of syngenetic or epigenetic origin. Studying coal samples from mines in the United Kingdom, Spears and Zheng (1999) found a strong positive correlation between pyrite content and Se, As, Mo, Sb and Tl concentrations. Pyrite is the most common sulphide in coal, but marcasite is also frequently found in them (Spears et al., 1994). The amount of pyrite in coal depends on the availability of S and Fe 13 and the intensity of sulphur-reducing bacterial activity. Fe is introduced to the system through the weathering of silicate minerals and the influx of Fe 2 + and Fe 3 + ions with groundwater. In most coals, the occurrence of siderite (Fe2C03) provides evidence of iron availability (Spears et al., 1994). Casagrande and Nug (1979) suggest that the breakdown of plant and animal protein during coalification supplies the S needed for pyrite formation. Se has been detected in pyrite grains using a number of different methods, including X-ray fluorescence (White et al., 1989), proton-induced gamma ray/X-ray emission (PLXE) microanalysis (Hickmott and Baldridge, 1995), instrumental neutron activation analysis (INAA) (Zodrow and Goodarzi, 1993) and scanning electron microscopy (SEM) (Galbreath and Brekke, 1994). Leutwein (1978) notes that sedimentary sulphides tend to contain less Se than those of magmatic origin. Magmatic sulphides form at high temperatures that favour lattice expansion and isomorphous substitution. He indicates that sedimentary sulphides contain between 10 and 20 mg/kg Se and sulphides of magmatic origin between 50 and 100 mg/kg. Wandless (1957), however, found concentrations as high as 70 mg/kg Se in British coals and Rosenfeld and Beath (1964) reported a Se concentration of 548 mg/kg in Wyoming coal pyrite concentrates, suggesting that the range proposed by Leutwein (1978) may not adequately reflect the heterogeneity of sedimentary depositional environments. A number of studies have identified discrete Se minerals in coal, though these tend to be quite rare as Se substitutes readily for S in sulphides. Wandless (1959) and Minkin et al. (1984) found minute selenide inclusions in pyrite concentrates extracted from British and Indiana coals respectively. Pyrite samples from bituminous coal studied by Minkin et al. (1984) contained approximately 2 mg/kg Se, but electron microprobe revealed isolated points containing as much as 3000 mg/kg Se. Finkelman (1981) found clausthalite (PbSe) in Appalachian coals, while Goodarzi and Swaine (1993) documented the occurrence of both PbSe and ferroselite (FeSe2) in coals from Western Canada. A sequential extraction experiment by Dreher and Finkelman (1992) 14 indicates that these Se minerals make only a minor contribution to total Se in coals. The oxidation of Se-bearing pyrite and selenides generates selenite, which is soluble and binds readily to clay minerals (Frost and Griffin, 1977) and HFMO (Balistrieri and Chao, 1987). 2.1.2.3 Selenium Associated with Clay Minerals Clay minerals are the most abundant minerals in coals and associated sedimentary rock (Dale, 1996). Clay minerals are introduced into coal swamps by surface water. Those commonly found in coal include kaolinite, illite, montmorrilonite, vermiculite and chlorite. They are derived from the weathering of feldspars, micas and ferromagnesian minerals and consist of different combinations of tetrahedral silica sheets and octahedral alumina sheets bound by shared O2" ions occupying the vertices of the tetrahedra and octahedra (McBride, 1994). The 1:1 clays, like kaolinite, are composed of octahedral and tetrahedral sheets stacked one upon the other. In illite and montmorillonite, both 2:1 clays, an octahedral sheet is sandwiched between two tetrahedral sheets. Illite has K + ions trapped between the silicate layers preventing water from penetrating into the matrix and causing swelling. In montmorillonite and vermiculite the interlayer K + ions are replaced by hydrated M g 2 + ions. Both have a high specific surface area and cation exchange capacity (CEC) relative to kaolinite and illite. CEC, defined as the quantity of cations that can be reversibly adsorbed per unit weight of mineral, is generally expressed in terms of meq/100 g (McBride, 1994). Chlorite has a hydroxide sheet between the 2:1 silicate layers and is thus termed a 2:1:1 clay. The hydroxide sheet stabilizes chlorite structure impeding any swelling. Clay minerals have two types of charge, one permanent and the other pH-dependent. The permanent charge comes from the substitution of cations of similar ionic radii but different charge. A l 3 + substitutes for S i 4 + in the tetrahedral layers while M g 2 + or Fe 2 + replace Fe 3 + in the octahedral layers (McBride, 1994). Though common in 2:1 clays, relatively little isomorphous substitution occurs in 1:1 clays. This type of substitution confers a negative surface charge to clay minerals. This attracts counterions and repels co-ions leading to the formation of an electric 15 double layer (Gouy, 1910; Chapman, 1913). According to Stern (1924), the electric double layer is composed of the Stern layer, where ions oscillate about fixed adsorption sites, and the Gouy-Chapman layer, a diffuse band of cations extending several hundred nanometers from the mineral surface. The diffuse nature of this layer results from the competition of electrostatic and diffusion forces (Figure 2.3). - Y - ^ - r l ^ J(* Clay Stern Gouy-Chapman Bulk Solution Diffuse Layer Figure 2.3 A Schematic Presentation of Cations in the Electric Double Layer (Adapted from Yariv and Cross, 1979) The thickness of the Gouy-Chapman diffuse layer depends on the magnitude of the surface charge and the concentration of electrolytes in solution. Though cations are attracted to the surface, there is also a tendency for them to diffuse away from the clay surface into the bulk solution where the cation concentration is lower. The greater the ion concentration in solution, the lower the tendency of cations to diffuse away from the clay surface. Ions in the Stern layer may form inner- or outer-sphere complexes. In inner-sphere complexes, ions form covalent bonds with hydroxyl groups on the surface of the clay mineral. Outer-sphere complexes, on the other hand, consist of hydrated ions held to the mineral surface by ionic bonds. 16 Selenite and selenate form different types of bonds at clay surfaces (Figure 2.4). Geering et al. (1968) suggest that selenite forms inner sphere complexes like those formed by ortho-phosphate. Binding to ligands at the surface, it becomes essentially nonexchangeable. Selenate does not adsorb specifically. Like sulfate and nitrate, selenate tends to form weak outer-sphere complexes. It is therefore easily replaced by other exchangeable anions and is more likely to be found in solution (Hayes et al., 1987). Oxygen 1 Central atom OUTER-SPHERE C O M P L E X INNER-SPHERE C O M P L E X S e = 0 C X ^ _ W ) cQ Water molecule r y © - C 5 o = s c - 0 -i j Hydrogen Mineral Surface Figure 2.4 Inner- and Outer-Sphere Complexes Formed by Se Oxyanions The second type of charge, pH-dependent charge, is of minor importance relative to the charge conferred by isomorphous substitution (Taylor, 1987). At the surface of clay minerals, unsatisfied charges on broken bonds are balanced by the adsorption of cations or anions. The point of zero charge (PZC) is the pH at which the net surface charge is zero. Below the PZC, the surface bears a positive charge and is capable of attracting anions, above it, the surface bears a negative charge (Figure 2.5). Clay minerals have PZC values between pH 2.5 and 4.5 (Forstner and Wittmann, 1979). Charge on acidic functional groups in organic matter is also pH-dependent. 17 OH Si \ O...H+ A l - O H 2 + O...H+ Si In aqid solution ANI&^EXCHANGER OH / Si OH A l - O H OH / At the point of zero charge O. . .HOH / Si O / A1-O. . .HOH \ O / Si In alkaline solution CATION EXCHANGER Figure 2.5 pH-Dependent Charge on Broken-Bond Aluminosilicate Surfaces (Adapted from Yariv and Cross, 1979) 2.1.2.4 Selenium Associated with Hydrous Ferric and Manganese Oxides Selenite adsorbs readily to hydrous ferric and manganese oxides (HFMO) (Balistrieri and Chao, 1987). HFMO are secondary oxides that generally occur as coatings on mineral surfaces or as fine discrete particles. Crystalline HFMO consist of hexagonal or cubic structures with Fe 3 +, Mn 4 + or Mn 2 + in the central position surrounded by O2" and/or OH' anions. Their structure is determined by the degree to which they share corners, edges or faces. Common crystalline HFMO include goethite (a-FeOOH), lepidocrocite (y-FeOOH), manganite (MnOOH) and birnessite (Na 4Mn 1 40 2 7- 9H20). Due to the variability of the chemical environment in which HFMO form and their ability to incorporate foreign ions into their structure, these minerals frequently differ both in composition, crystallinity and morphology from pure specimen minerals. HFMO lacking a well-defined crystalline structure are termed "amorphous" and account for much of the HFMO in soils and sediment. They play an important role in the retention and release of contaminants such as heavy metal ions and soil nutrients due to their relatively large surface area. Many ions form covalent bonds with ligands on HFMO surfaces (Veiga et al., 1991). The surfaces of HFMO develop a limited positive or negative charge depending on variations in pH. Fe oxides have PZC values in the range of pH 7 to 9. Mn oxides tend to have a more complex mineralogical structure 18 than Fe oxides and have PZC values between 1.5 and 4.6 (McBride, 1995). Redox conditions, the amount of the oxide present, its degree of crystallinity and the presence of organic matter will also affect the relative importance of Mn and Fe oxides as trace element scavengers. 2.1.3 Cycling of Selenium 2.1.3.1 Major Global Fluxes Estimates of major Se inventories from Nriagu (1990) are listed in Table 2.4. Fluxes between these compartments are illustrated in Figure 2.6. The bulk of Se is found in the lithosphere (Nriagu, 1990). Se from rocks is transferred to soil through weathering and uplift. In hot and arid areas underlain by seleniferous rocks, high Se soils are common since losses of Se through leaching are minimal (Herring, 1990). Transfers of Se from soil and sediment to water are governed by a number of factors including redox conditions and microbial activity, as well as the type of mineral and the amount of organic matter present. Table 2.4 Major Inventories of Se (Adapted from Nriagu, 1990) Reservoir Se (kilotonnes) Lithosphere 3xl0 y Soils 10s Fossil fuel deposits 1.4xl05 Terrestrial biomass 70 Atmosphere (gaseous and particulate) 2 Oceans (dissolved) 2x10" Rivers (dissolved and particulate) 12 Shallow groundwater 0.8 Polar ice 400 19 ATMOSPHERE A N T H R O P O G E N I C S O U R C E S roasting o f sulfide ores industrial use o f Se coal combustion V O L C A N I C E R U P T I O N S S U B L I M A T I O N ¥ V O L A T I L I Z A T I O N A V | & M E T H Y L A T I O N t % V O L A T I L I Z A T I O N | BIOTA | I \ Y P R E C I P I T A T I O N FRESHWATER _ J RESOURCES | — » | SURFACE O C E A N T I P R E C I P I T A T I O N SOIL& SEDIMENT U P W E L L I N G S E T T L I N G DEEP O C E A N t W E A T H E R I N G A D S O R P T I O N V LITHOSPHERE MARINE SEDIMENT Figure 2.6 Generalized Cycle of Se (Adapted from Herring, 1990) Se concentrations in freshwater generally range between 0.1 and 3.0 ug/L (Lemly, 1985). Selenate and selenite account for the bulk of river-borne Se and are either dissolved or adsorbed to particulate matter (Cutter, 1989). The biogeochemical cycle of Se in soil and freshwater environments will be discussed in greater detail in Section 2.1.3.2 as it is highly relevant in identifying possible mechanisms of Se mobilization from the Mist Mountain Formation. The average concentration of Se in sea water and freshwater is 0.2 ug/L (Drever, 1982). Se has a short residence time in seawater as it adsorbs rapidly to sediment. Upwelling of deep water renews the supply of dissolved Se to surface waters. Selenate is the dominant form of Se in surface waters of the ocean and is assimilated by marine organisms and incorporated into particulate organic matter (Herring, 1990). The amount of Se introduced into the marine 20 environment through riverine inputs and rainfall is very small and consequently has virtually no effect on the overall S balance in the oceans (Herring, 1990). Natural sources introduce between 7000 and 18 000 tonnes of Se to the atmosphere annually (Nriagu and Pacyna, 1988). About 60 to 80% of this is said to be gaseous Se of marine biogenic origin. Other natural sources of Se to the atmosphere include volcanic activity, volatilization from land plants, suspension of sea salts, sublimation and wind erosion of rocks or soils. Anthropogenic sources contribute between 1700 and 5800 tonnes of Se to the atmosphere (Nriagu and Pacyna, 1988). Se is estimated to have only a 45-day residence time in the atmosphere such that little inter-hemispheric transfer occurs (Nriagu, 1990). Removal of gaseous forms of Se from the atmosphere likely occurs by condensation onto precipitation or, in the case of particulate Se, by scavenging. Continental rain averages 0.3 to 1.1 ug/L Se (Mosher and Duce, 1989). Haygarth et al. (1993) determined that the amount of Se deposited by precipitation is highly dependent on distance from the source. 2.1.3.2 Biogeochemical Transformations Although Se may occur as selenate (+6), selenite (+4), elemental Se (0) and organic Se in solution, only the first two forms are found under standard conditions. Most soils are fairly well aerated, with Eh values lying between +0.3 and +0.5 volts and pH values between 3 and 9 (Sparks, 1995). It is important to note that soil is not a homogeneous media. It contains organic and mineral matter, water, air and microorganisms, all of which affect the physico-chemical properties of the soil. This, in turn, determines the speciation and sorptive capacity of metals and trace elements and thus, their mobility. Se adsorption to soil and sediment is positively correlated with organic carbon and clay content and negatively correlated with salt concentration, alkalinity and pH (Sillanpaa and Jansson, 1992). Se bound to particulate matter can be transported into waterways, but when referring to mobile Se, one is generally making reference to Se in solution 21 (Kabata-Pendias and Pendias, 1984). Neal and Sposito (1989) identify selenite as the predominant form of Se in soils. Based on redox potentials, one would also expect selenate to be quite common in well-aerated alkaline soils, but due to its poor affinity for soil particles it is readily leached and constitutes only a minor portion of total Se. In soil, most transformations of Se appear to be microbially mediated (Neal, 1995). There are three categories of microbially mediated transformations of Se: oxidation and reduction, immobilization and mineralization, and methylation (Doran, 1982). Most research on microbial transformations of Se has focused on microorganisms capable of rendering Se less mobile through reduction. Many fungi, bacteria and actinomycetes in soil can reduce Se oxyanions to elemental Se or to volatile or non-volatile organic compounds (Thompson-Eagle and Frankenberger, 1992). Both Pseudomonas stutzeri and Thaurea selenatis have been found to reduce selenate to selenite (Macy et al., 1992). Altschuler et al. (1983) found that Desulfovibrio desulfuricans and Clostridium desulfuricans can generate FeS2 through the reduction of organic S compounds. These species are likely able to reduce analogous Se compounds. Nelson et al. (1996) found that co-cultures of Desulfovibrio desulfuricans and Chromatium vinosum could reduce selenate to elemental Se in a two-step process where D. desulfuricans reduces selenate to selenide which C. vinosum then oxidizes to elemental Se. Other Bacterial species able to produce elemental S and Se, are Begiatoa, Chlorobium and Ectothiorhodospira (Nelson et al., 1996). In acidic, poorly aerated soils, rich in organic matter, Se tends to be immobilized in sulphides or bound to clay minerals, HFMO or organic matter (Elrashidi et al., 1987). Oxidation of selenides proceeds slowly in most soils (Rosenfeld and Beath, 1964). In alkaline soils, Se is readily transformed by chemical or bacterial oxidation into plant-available selenate. In soils, selenite will only be reduced to elemental Se after soluble Pb+ 2, Cu + 1 , Cu + 2 , Sn + 2 and Cd + 2 have been reduced (Elrashidi etal., 1987). 22 There are four processes that may participate in the transformation of elemental Se: 1. Oxidation and methylation of Se by plant roots and microorganisms 2. Oxidation of sediment through burrowing and feeding activity of benthic invertebrates 3. Chemical oxidation associated with water circulation 4. Plant photosynthesis Once transformed into selenate or selenite, Se may be taken up by aquatic organisms or by rooted plants or it may adsorb to clay and organic carbon phase particulates. In general, as much as 90% of total Se in aquatic systems is found in the upper few centimetres of sediment (Bowie et al, 1991). The cycle is illustrated in Figure 2.7. StqgtiUrtd <wg»nlc, mintril, t l tmtnl i l , ind/or idtorbtd S« Figure 2.7 The Biogeochemical Cycle of Se in Aquatic Habitats (From Lemly and Smith, 1990) In the marine environment, Se may be found as elemental Se, selenate, selenite or organic Se. The latter is found in particulate form near the surface and as it sinks, it breaks down and undergoes 23 oxidation by opportunist species yielding selenate and selenite. Only these two forms are present in deeper waters (Cutter and Cutter, 1995). Se has a short residence time in oceans, adsorbing readily to clay sediment (Geering et al., 1968). The intensity of upwelling strongly influences the ratio of organic to inorganic Se in surface waters. Lemly and Smith (1990) found that Se, like Hg, biomagnifies i.e. it accumulates at progressively higher concentrations in successive trophic levels of a food chain placing piscivorous fish and aquatic birds at greatest risk in Se contaminated environments. They determined that Se biomagnification generally ranges from 2 to 6 times between algae and plants and invertebrates or foraging fish species. 2.3.1.3 Anthropogenic Inputs Approximately 1500 tonnes of Se are extracted annually for commercial use through the electrolytic refining of copper, zinc and nickel ores (Herring, 1990). Se is used in the manufacture of electronic and photocopier components, glass, plastics, ceramics, paints and lubricants. Nriagu and Pacyna (1988) studied the effects of human activity on the cycles of a number of trace elements including Se. Major sources of anthropogenic emissions to the atmosphere include coal combustion (2800 tonnes), oil combustion (800 tonnes), pyrometallurgical processes, such as Cu-Ni refining, (2100 tonnes) and municipal refuse and sewage sludge incineration (100 tonnes). Anthropogenic sources introduce between 10 and 72 tonnes of Se to aquatic ecosystems annually. Smelting, refining and steam from power generation are the primary sources of Se to aquatic ecosystems, releasing up to 20 and 30 tonnes of Se per year respectively. Manufacturing processes, sewage sludge and domestic wastewater also release Se into aquatic ecosystems. Human activity contributes approximately 73 000 tonnes of Se to terrestrial environments annually. This total excludes Se from atmospheric fallout which deposits between 1300 and 2600 tonnes of Se annually and be anthropogenic or biogenic in origin. Fly ash, the primary anthropogenic source of Se to terrestrial ecosystems, introduces up to 60 000 tonnes of Se to soils 24 per year. Agricultural wastes, wood wastes, urban refuse, fertilizer use and mine tailings account for the remainder of Se inputs to the environment. Several instances of Se contamination resulting from mining activity have been documented. Se is widely distributed in the area around the Red Dog zinc mine in Alaska. Se concentrations of 9 ug/L were found in discharge water from the mill (Brienne et al., 2000). This is almost two times greater than the EPA chronic freshwater criterion of 5 ug/L (USEPA, 1987). High Se concentrations are also common in uranium and coal deposits in Wyoming (Naftz and Rice, 1989) and in the Western Phosphate Resource Area in Idaho (Munkers et al., 2000). Se concentrations as high as 1000 |J.g/L were detected in groundwater at a uranium mine in Wyoming's Powder River basin (Naftz and Rice, 1989). Coal mining and combustion have been particularly problematic. Groundwater samples from backfill at one of the Powder River Basin coal mines had Se concentrations between 600 and 900 ug/L Se were detected (Dreher and Finkelman, 1992). Coals in the Powder River Basin contain, on average, 12.6 mg/kg Se (Naftz and Rice, 1989). Dreher and Finkelman (1992) attributed the elevated Se levels to the oxidation of pyrite present in the backfill of mines in the Powder River Basin. Sampling conducted in river basins of West-Central Alberta showed that Se concentrations in water samples collected immediately downstream of the Cardinal River, Gregg River and Coal Valley coal mines were up to an order of magnitude above the 1 ug/L Canadian Council of Ministers of the Environment (CCME) guideline (Casey and Siwik, 2000). While the impact of Se on biota at many of these sites has yet to be determined, Se leached from fly ash landfills has been conclusively linked to fish kills in Martin Lake, Texas and Belews Lake, North Carolina (Shepard, 1987). More details on Se toxicity are provided in Section 2.2.1.3. 25 2.2 SELENIUM IN BIOTA 2.2.1 Effects of Selenium While this study focuses on Se geochemistry the ultimate question is whether the final concentration of Se in waterways constitutes a hazard to aquatic wildlife. There is much debate, both in Canada and in the United States, as to what the fresh water criteria should be set at to adequately protect fish and bird populations. A review of the information available about Se toxicity will provide a practical framework in which to consider the implications of the results of this study. 2.2.1.1 Humans Se was discovered to be an essential nutrient in 1957 (Schwartz and Foltz, 1957). However, its role in glutathione peroxidase, an enzyme essential for protecting cellular membranes from destruction by free radicals, was not established until 1973 (Rotruck et al., 1973). In humans, there is a very limited range over which selenium (Se) is neither toxic nor deficient, with the recommended dietary allowance for adults set at 55 ug per day (National Institute of Health, 2000). Figure 2.8 illustrates a typical dose response curve for a micronutrient with a narrow range between critically high or low levels such as Se. Se intake occurs primarily through diet. Insufficient Se may result in nutritional muscular dystrophy, hepatic necrosis and impaired immunity. Severe nutritional Se-deficiency has been observed in two areas in China (Ge and Yang, 1993). In one case, it resulted in endemic juvenile cardiomyopathy typical of Keshan disease, and in the other it caused Kashin-Beck disease, characterized by chondrodystrophy, a disorder affecting caitillage formation. Other symptoms of Se deficiency include hemolytic anemia, hypertension, ischemic heart disease, cirrhosis and arthritis. Se deficiency has also been observed in Serbia and Croatia (Maskimovic et al., 1992) and Tibet (Moreno-Reyes et al., 1998). 26 Increase in Micronutrient Supply Figure 2.8 Typical Dose-Response Curve for Micronutrients (From Thornton, 1995) Se's toxic effects were first documented by Franke and Painter (1936). Se toxicity results from its substitution for S in cysteine and methionine, S-containing amino-acids, which affects the functioning of the proteins in which they are found. Se has also been shown to reduce growth hormone production (Thorlacius-Ussing et al., 1987). Consumption of crops grown on seleniferous soils or of livestock that have grazed on high Se fodder areas may lead to selenosis, a toxic response to Se manifested by hair loss, skin lesions and nerve damage (Zheng et al., 1999). The inhalation of hydrogen selenide in occupational settings like copper refineries and self-medication of dietary supplements are, however, the primary causes of Se toxicity in humans (World Health Organization, 1986). 2.2.1.2 Livestock Se deficient soils are far more common than Se enriched soils (Neal, 1995). In instances when livestock does not obtain the minimal daily Se requirement of 0.01 mg/kg of body weight, they should be provided with Se supplements, salt licks or injections (Wuyi et al., 1987). Failure to do so can lead to a disorder known as "white muscle disease", where calcium salts are deposited 27 among muscle fibers impeding muscle function. Laboratory tests and field studies indicate that decreased appetite, growth and fertility may also characterize Se deficiency. Records of selenium toxicity in livestock date back as far back as the thirteenth century. Marco Polo described a necrotic hoof disease affecting his horses when they consumed plants generally avoided by local animals. A similar ailment observed in domestic animals in Colombia during the 16th century was also attributed to excess Se (Oldfield, 1992). Outcrops of Se-rich saline seeps in the northern Great Plains of the United States have caused problems for ranchers (Rosenfeld and Beath, 1964). In addition to hoof loss and reduced growth and reproductive performance, chronic Se intoxication in livestock may also cause hair and weight loss (Combs and Combs, 1985). Se toxicity results from the consumption of forage containing 5 to 40 mg/kg Se. High protein intake can counter the effects of excess Se by transforming it into easily excreted methylated metabolites. 2.2.1.3 Fish and Wildlife Excess Se can pose an especially serious problem in aquatic environments as evidenced by incidents in the United States at Belews Lake, North Carolina in the late 1970s and at the Kesterson National Wildlife Refuge in California in the mid-1980s. At Belews Lake, average , Se concentrations were approximately 5 times greater than the background concentration of Se in freshwater. In some water samples, Se concentrations as high as 200 ug/L were detected (Lemly, 1985). The source of the Se was identified as fly ash from a nearby thermal power plant. The elevated Se concentrations lead to severe reproductive failure in bluegill (Lepomis macrochirus), green sunfish (L. cyanellus), largemouth bass {Micropterus salmoides) and flat bullhead (Ictalurus platycephalus) and a rapid collapse of these fish populations (Sorensen et al., 1984). Congenital malformations observed in these fish species included missing fins, protruding eyes and deformed spines and heads. Not all species were 28 equally affected by the high Se concentrations. Forage species, including red shiners, fathead minnows and mosquitofish, all showed a higher degree of tolerance (Lemly, 1985). Although Se levels in water eventually dropped below 1 pg/L, consumption advisories remained in place as elevated Se concentrations and teratogenic effects were still being observed up to 10 years after Se inputs from the fly ash had ceased (Lemly, 1997). Kesterson Reservoir was constructed between 1968 and 1975 as a part of an agricultural drainage facility for the San Joaquin Valley, California. Soils in this area formed through the weathering of seleniferous Cretaceous marine sedimentary strata (Presser, 1994). The reservoir consisted of a series of 12 ponds that were to be managed as wetlands and were to be supplied with water subsurface drainage from the irrigated agricultural lands via the San Luis Drain. Problems arose in 1983 when aquatic birds nesting around the reservoir were found to have high rates of embryo deformities and mortality (Ohlendorf, et al, 1986; Ohlendorf, 1989). Water quality tests showed that water entering the ponds had Se concentrations as high as 1400 ug/L (Presser and Barnes, 1984). By 1985, laboratory and field studies had conclusively identified Se as the agent responsible for the fish and bird deaths. Reclamation efforts were promptly undertaken; discharge to the ponds was halted in 1986 and in 1988 the ponds were dewatered and filled. Mallards were missing eyes, beaks, wings, legs and feet and showed brain, liver, heart and skeletal abnormalities. Fish populations were also affected. Mosquitofish contained up to 100 times more Se than fish of the same species in the nearby Volta Wildlife Area which received no drainage water. Invertebrates, amphibians, reptiles and mammals were also shown to have bioaccumulated Se (Ohlendorf, 1989). Daphnia were found to contain up to 12.4 mg Se/kg dry weight, meaning 6 to 10 times the normal background level for these organisms. Although tissues from raccoons autopsied contained from 10 to 30 times more Se than those in the Volta Wildlife Area no effects on growth or reproduction were observed. Nor were effects noted on kit foxes, coyotes, voles and shrews despite the elevated concentrations of Se in their diet. 29 Food is the primary exposure pathway for fish to Se, but they can also take up Se directly from the water column via their gills (Lemly, 1993). Specific manifestations of Se toxicity in fish include blindness, popeye, osteological deformities and reduced hatchability of eggs (Clark et al., 1986; Skorupa et al., 1996). In the field, reproductive failure and deformity of larvae have been reported in bluegill (Lepomis macrochirus) exposed to Se concentrations of 9 ug/L (Gillespie and Baumann, 1986) and malformation of late juvenile and early adult fathead minnow {Pimephales promelas) at concentrations of 10 ug/L (Hermanutz, 1992). In a comprehensive review of studies of Se toxicity in freshwater fish, Nagpal and Howell (2001) found L C 5 0 values to range from 5 to 126 600 ug/L. Factors found to affect Se toxicity include fish species and life stage and the form and concentration of waterborne Se. Several studies have indicated that selenite is more toxic than selenate (Niimi and LaHam, 1976; Brooke et al., 1987; Hamilton and Buhl, 1990). The lowest 96-hour L C 5 0 values for selenite and selenate reported by Nagpal and Howell (2001) were 620 ug/L and 2300 ug/L respectively. The combined effect of selenite and selenate is strictly additive (Hamilton and Buhl, 1990). The methylated forms of Se are much less toxic for the organism than selenite and selenate. However, the methylated Se derivatives have strong synergistic toxicity with other minerals such as arsenic (Jonnalagada and Prasada Rao, 1993). Waterfowl feeding on fish from lakes, streams or wetlands with high Se concentrations are at risk of bioaccumulating Se. Lesions resulting from excess Se detected in birds include necrosis of pancreas and liver cells and atrophy of lymphoid organs, feather follicles and fat (Green and Albers, 1997). Brix et al. (2000) identify teratogenicity (embryo deformities) and chick mortality as useful endpoints for gauging the ecotoxicological effects of Se. They note that Se toxicity can also result in failure to breed and in reduced hatchability of fertile eggs and that sensitivity to Se in waterfowl varies, from relatively tolerant avocets and snowy plovers to highly sensitive mallards. Stilts and kildeer show intermediate levels of sensitivity (Brix et al., 2000). 30 2.2.1.4 Plants Plants vary considerably in their capacity to survive in high Se environments. Rosenfeld and Beath (1964) divided plant species into three groups based on their capacity to accumulate Se. The first, which includes the genera Astragalus, Machaeranther a, Haplopappus and Stanleya, may contain hundreds to thousands of milligrams of Se per kilogram. In these plants, Se is generally present as water soluble methylselenocysteine, rather than as protein-bound selenomethionine (Mayland et al., 1989). In other words, these plants have metabolic pathways which ensure that Se is not incorporated into proteins. Plants in the second group accumulate between 50 and 100 mg/kg when grown on seliniferous soils (Rosenfeld and Beath, 1964), whereas those in the third group (grains, grasses and forbs) rarely accumulate more than 50 mg Se/kg (Rosenfeld and Beath, 1964). Plant tissue containing more than 5 mg/kg dry weight is considered toxic to grazing animals (Levander, 1985). Although both selenate and selenite are available for plant uptake, selenate is the primary form taken up by plants since selenite adsorbs more strongly to clay minerals and HFMO (Brown, 1990). High soil pH decreases cation exchange capacity as hydroxide ions bind to the available sites releasing Se into the soil-solution and making it available to plants. Selenate and selenite compete with other anions, such as sulfate, phosphate, molybdate and oxalate in the soil-solution for adsorption sites on plant roots. Effects of Se on plants include reduced dry-matter yield, loss of leaf pigmentation and potentially death if concentrations in the soil-solution are very high (Girling, 1990). 2.2.2 Se Guidelines and Criteria The EPA updated its freshwater chronic criterion for Se in 1987. In the wake of the Belews Lake and Kesterton incidents the acute criterion was set at 20 ug/L while the long-term exposure 31 criterion was set at 5 ug/L (USEPA, 1987). Both the acute and chronic criteria are expressed in terms of total Se and, as such, fail to identify the fraction accounted for by bioavailable species. In British Columbia, the Ministry of Environment Lands and Parks refers to guidelines rather than to criteria, the difference being that the former only has legal ramifications once it has been used to set effluent quality standards in permits. The Canadian Council of Ministers of the Environment (CCME, 1999) set the freshwater guideline at 1 pg/L, which is five times lower than the U.S. criterion. Recently, these figures have come under scrutiny with some arguing that they may not adequately protect sensitive aquatic organisms and others that they may be overprotective, especially in lotic environments such as the Elk River Valley. The criterion for Se in drinking water in the United States is 50 pg/L (USEPA, 2001), while the Canadian guideline for drinking water is 10 pg/L (Health and Welfare Canada, 1993). There are no limits set by the USEPA, World Health Organization or Canadian government on Se intake from the consumption of fish. Van Derveer and Canton (1997) suggested that site-specific guidelines should be developed, noting that bioaccumulation in fish may be a factor of 10 times greater in lentic systems than in lotic systems. The static nature of lentic systems allows Se to accumulate in sediment (Lemly, 1999) where one frequently finds anoxic zones rich in organic matter favouring microbial production of organic Se species. Presser et al. (1994) determined that organoselenium compounds, such as selenocysteine and selenomethionine, are much more readily bioaccumulated than inorganic forms of Se. The fact that many organisms in lentic systems have a restricted feeding range increases exposure and thus the amount of Se that bioaccumulates. Some studies stress the importance of site-specific environmental factors, such as pH and redox conditions, in determining Se speciation and its bioavailability (Bowie and Grieb, 1991; Porcella et al., 1991). It is precisely this complexity that will render it difficult to establish a framework for setting site-specific guidelines. It is necessary to consider not only the physical environment, but also how 32 different species utilize different niches present in the ecosystem. The USEPA is currently considering the viability of site-specific criteria related to Se concentrations in sediment and in the water column, but plans to address the issue of dietary exposure are conspicuously absent (Sappington, 1998). The generic bioaccumulation model developed by Adams et al. (2000) is perhaps more useful as it produces a criterion based on tissue values which can then be modified according to site-specific conditions. Such a model would recommend less restrictive criteria in instances where Se concentrations are high but little bioaccumualtion is occurring. Brix et al. (2000) argue that tissue based criteria provide the most effective method of assessing the risk of Se toxicity. Using logit and probit models to analyse existing laboratory and field data, they derived dose-response relationships and proposed thresholds for whole body and ovary tissue of fish of 6 to 9 and 17 mg/kg dry weight respectively. The logit and probit models are qualitative variable model estimators. In the. simple probit model, the dependent variable is usually binary while in the logit model, there is a discrete choice among a small set of alternatives. Using the same approach, Brix et al. (2000) determined a bird egg threshold of 16 mg/kg dry weight and teratogenesis threshold of 26 mg/kg dry weight. This bird egg threshold is more than two times greater than the concentration found to cause chick mortality in field studies (Skorupa et al., 1996). Brix et al. (2000) suggest that this discrepancy may.be due to factors other than Se bioaccumulation that were not controlled for in the field studies. There are also problems associated with tissue-based criteria; the primary one being that fish or birds of the same species living in different areas may have different levels of tolerance to Se (Van Derveer and Canton, 1997). 33 3.0 ENVIRONMENT OF THE E L K RIVER V A L L E Y 3.1 BIOPHYSICAL ENVIRONMENT 3.1.1 Geology The Southeast Coalfields of British Columbia, located between 114° 30' and 115° 30' longitude and 49° 00' and 51° 00' latitude, are grouped into three basins: the Elk Valley, Crowsnest and Flathead coalfields. The elongated Elk Valley Coalfield lies just north of the Crowsnest Coalfield, while the Flathead Coalfield is located immediately north of the Canada-US border (Figure 3.1). Coal deposits in the three basins are located in the Mist Mountain Formation of the Jurassic-Cretaceous Kootenay Group formed between 150 and 130 million years ago. In the Flathead Coalfield, the Mist Mountain Formation is heavily eroded. The marine Fernie Formation underlies the Kootenay Group. The lower sections of this formation consist of a series of shales, ranging from dark gray shales to brown silty shales with limestone beds while calcerous sandstone with limestone and glauconitic sandstone form the upper units. The Passage Beds, composed of interbedded shale and sandstone coarsening upwards, are the last unit deposited in this formation (Gibson, 1977). The Morrissey Formation, with an average thickness of 40 metres, forms the base of the Kootenay Group. It is easily recognized across the Elk River Valley and contains two distinct members. The Weary Ridge member, lying immediately above the Passage Beds, is composed of quartzose, argillaceous, calcerous and ferruginous sandstone. The quartz-chert sandstone Moose Mountain member is coarser grained than that of the lower member. Minor carbonaceous shale and coal deposits occur sporadically within the Morrissey Formation. Material from tectonically active uplands was likely deposited on lower delta coastal plains draining eastward into the inland 34 Figure 3.1 Elk Valley, Crowsnest & Flathead Coalfields of Southeastern British Columbia (Ryan and Dittrick, 2000) 35 Fernie Sea (Gibson, 1977). The plains were protected from direct marine influence by sand dunes that now form the Passage Beds of the Morrissey Formation (Vesey and Bustin, 2000). The Mist Mountain Formation conformably overlies the Morrissey Formation. It averages 500 metres in thickness, but ranges from 240 to almost 1000 metres. Depositional patterns typical of levee, splay, flood-basin, swamp and marsh settings suggest that material was deposited in alluvial channels and flood plains (Gibson and Hughes, 1981). s o w o a — u CADOMIN FORMATION P O u < W ei u a z < 2 Pa g > z w H O E L K FORMATION MIST MOUNTAIN FORMATION MORRISSEY FORMATION MOOSE MOUNTAIN M E M B E R W E A R Y RIDGE M E M B E R z o H u O b tt. PASSAGE BEDS Figure 3.2 Jurassic-Cretaceous Stratigraphy of the Kootenay Group (Gibson, 1977) Coal comprises 8 to 12% of the stratigraphic thickness of the Mist Mountain Formation (Grieve, 1985). It ranks from high- to low-volatile bituminous (Smith, 1989) and has applications in coking and power generation. The main silicate minerals present in the coal are kaolinite and quartz, as in most non-marine influenced deposits (Pearson, 1980). Coals in the Mist Mountain 36 Formation contain between 6.5 and 33.1% ash and generally have S contents below 0.5% (Vessey and Bustin, 2000). There is a gradual shift from inertinite rich to vitrinite rich coals from the base to the top of the formation (Grieve, 1985). Siltstone, mudstone and shale of non-marine origin are interbedded with the coal. Apart from a basal coal zone within the lowermost 25 metres of the formation, there are no coal seam clusters in the stratigraphy (Gibson, 1977). Extensive tectonic activity during the Laramide Orogeny has lead to considerable deformation of the seams making it difficult to correlate seams from one mine to another (Gibson, 1979). Many coal seams are highly sheared and thickened due to folding. Thrust faults are not uncommon. Clastic sediments and thin, discontinuous humic and sapropelic coal seams in the Elk Formation overlie the Mist Mountain Formation. This constitutes the uppermost formation of the Kootenay Group. Like the Mist Mountain Formation, it is non-marine in origin (Gibson and Hughes, 1981). 3.1.2 Environmental Variables The Elk River Valley is located in a temperate climate zone, with average highs between 19 and 25 °C during the summer months and average lows between -13 and -9 °C during the winter months. The valley receives, on average, 1095 mm of precipitation per year. Peak flow rates in the Elk River and its tributaries are generally observed in March and April as a result of snowmelt. Base flow at the 08NK005 monitoring station at Phillips Bridge on the Elk River is in the order of 20 m3/s (Wipperman and Webber, 1997). Peak flows are generally between 329 and 384 m3/s, but reached 683 m3/s in 1986. Table 3.1 lists the values of key water quality parameters measured at Phillips Bridge. Table 3.1 Key Parameters of Elk River Water Quality (Data from Wipperman and Webber, 1997) Parameter Range pH 7.5-8.5 Alkalinity 100- 150 mg/L CaC03 Water temperature 0 - 15 °C Non-filterable residue 0-1130 mg/L 37 MacDonald and Strosher (1998) noted that Se concentrations in the Elk River and its tributaries remain relatively constant throughout the year. Significantly more Se is mobilized during the spring when runoff rates increase due to snowmelt. The river system can be generally characterized as fast flowing and well oxygenated. 3.2 MINING O P E R A T I O N S 3.2.1 History There are currently five coal mines in operation in the Elk River Valley. Their locations are illustrated in Figure 3.1. The first commercial coal mining operation in the East Kootenays went into production in 1897 at Coal Creek. By 1910, The Crow's Nest Pass Coal Company had opened mines at Michel and Morrissey and the Canadian Pacific Railroad had opened the Hosmer mine. The amount of coal produced by the mines in the Elk River Valley has varied over the years. Production peaked prior to World War I around 1,387,000 tonnes per year and decreased notably during the 1920s and early 1930s. Production began to rise again in the mid 1930s and averaged about a million tonnes until 1960. Large-scale coal mining in the Valley began in the 1960s at Elkview. Fording, Greenhills and Coal Mountain came into operation in the 1970s and Line Creek at the beginning of the 1980s. Approximately 25,000,000 tonnes of coal are produced annually in British Columbia, contributing roughly $900,000,000 to the province's economy (Ryan, 2001). The East Kootenay coalfields are the most important coalfields in the province. 3.2.2 Current Mining Activity All five mines are open-pit operations. Once a section has been blasted, waste is loaded into haul trucks by electric shovels and disposed of at the site's waste dumps. Cumulatively, the five mines generate approximately 140 x 106 tonnes of waste rock pear year (MacDonald and Strosher, 1998). Waste rock consists primarily of musdstone and siltstone interburden but also includes 38 coal from seams that are too thin or not of saleable quality. The bulk of the coal at the sites is, however, destined for sale and is transported to the crusher station to be crushed to an appropriate size for processing. At the processing plants coal is crushed and screened. It is then washed with dense-medium cyclones, spirals and flotation and dried for shipping. Due to the friable nature of coals collected from highly sheared seams, a considerable amount of fines is generated in this process. Coal Mountain and Line Creek mix coarse and fine refuse from this process, while Fording River, Greenhills and Elkview dispose of the fine refuse in tailing ponds and the coarse material in crusher reject dumps. 3.2.2.1 Fording River Fording Coal Limited owns and operates the Fording River, Greenhills and Coal Mountain operations. Mining at Fording River began in 1971. Early mining activity at the Fording River site took place at Eagle Mountain and at a section of the property called the Greenhills Range. Henretta Ridge and Valley were mined during the 1990s. Fording River produced 8.3 Mt of coal in 1999 most of which was destined for steel production. The mine has the widest range of bituminous coals in Canada. Much of the activity planned for the next 20 years at the site focuses on Eagle Mountain, as it contains over 65% of Fording River's 515 Mt reserves but mining will continue at Henretta Ridge until 2007. In terms of waste disposal, there are spoils on both sides on Henretta Creek, around the Greenhills Range pits and on the south, east and west sides of Eagle Mountain. There are two tailings ponds on the site, with the north one currently active. Crusher reject is currently dumped at the Kilmarnock reject spoil north east of the south tailings pond. Al l streams on the site are directed through settling ponds before discharge into the Fording River. 39 3.2.2.2 Greenhills Fording acquired Greenhills in 1992. Pits active in the 1980s include the Hawk, Falcon and Cougar pits. Cougar Pit is still active. Raven Pit was mined in the 1980s and again in the late 1990s. Greenhills has both its tailings pond and coarse refuse piles at the south end of the site. Water from the site drains into the Elk River via a series of creeks along the western side of the site. Some drains in Greenhills Creek, a tributary of the Fording River. Greenhills produced 5.0 Mt of coal in 1999. The mine has approximately 128 Mt of reserves and generates both metallurgical and thermal coals. 3.2.2.3 Coal Mountain Small-scale mining started at Coal Mountain in 1908, largely as underground operations. Open pit mining on Coal Mountain started in the 1940s in Pit 34. Most of the waste from both Pit 3, mined in the 1970s, and Pit 51, mined in the 1980s, was dumped over the mountain's western flank. Waste from the currently active Pits 7, 34 and 37 is dumped to the east in the Corbin Creek drainage. A rock drain was created over Corbin Creek in the 1990s. The site has no tailing impoundment. Fines are dried, mixed with coarse refuse and dumped on Middle Mountain, northwest of the crushing plant. Water from the site drains into Corbin Creek, which connects with Michel Creek, a tributary of the Elk River. The mine produced 2.1 Mt of medium-volatile thermal and soft coking coals in 1999. Reserves on this site are estimated to be in the order of 35 Mt of cleaned coal and will be extracted over the course of the next 14 years. This will come from one main coal horizon. This seam, known as the Mammoth seam, varies in thickness from 1 m up to 200 m due to its extreme structural character. 3.2.2.4 Elkview The Elkview mine, owned by Teck-Cominco Corporation, has been in operation since 1969. The first pits dug were the Harmer Knob and Adit 29 pits. Mining in the Harmer, Baldy, Camp 8, A40 40 and A40C pits started in the 1970s. Waste was trucked to the Erickson Dump, which now forms a rock drain over Erickson Creek, and into mined out areas. The South, A40FW and Adit 29 East pits were started in the 1980s. In the 1990s, mining continued in some of these areas and started in the Elk, Natal West and Indigo pits. The waste rock from Elk pit has been trucked to the Bodie dump since 1996. Currently, waste from Natal West is hauled to the Erickson dump and waste from the Indigo pit to another area draining into Erickson Creek. The mine has a tailing impoundment located on the western edge of the property where it borders the Elk River. The coarse refuse dumps are found just north of the tailings pond. Some water drains directly into the Elk River via creeks along the western side of the property, but some first flows into Erickson or Michel creeks. Teck acquired the property from Westar Mining Limited in 1992. The mine now produces approximately 5 million tonnes of metallurgical coal annually. At current extraction rates, mine life is estimated to be more than 40 years. 3.2.2.5 Line Creek Line Creek is a 50:50 joint venture owned by Luscar Limited and Consol Energy Canada Limited. It has been operation since 1981. The mine has an annual production capacity of 3.5 million tonnes of coal. Thermal coal accounts for approximately one-fifth of the total production. Seams vary from 2 to 15 meters in thickness. The first section of the property to be mined was the Line Creek Ridge (South Pit/North Pit). Having trended northward over the last two decades, mining in this area continues. Work in the Mine Services Areas (MSA) North, MSA West Pits and North Horseshoe Ridge Pits started in 1992, 1993 and 1994 respectively. There are rock dumps adjacent to the pits. Line Creek has no tailings impoundment. Instead, fines are dried, mixed with the coarse refuse. Al l water from the site drains via Line Creek that flows into the Fording River. 41 Land reclamation at the coal mines is carried out towards a designated land use objective. In the Elk Valley emphasis is place on reclaiming mined lands primarily for moderate yield forestry and wildlife habitat. Reclamation is carried out in a progressive manner as the mines develop. When possible, pits are backfilled with waste from adjacent pits. Prior to revegetation, waste rock dumps are resloped. Topsoil is generally not used at the coal mines since the waste rock is broken down to a fine, soil-like texture during recontouring. Most areas are seeded with a grass-legume forage mix, but selected areas are planted with deciduous trees and conifers. 3.3 S E L E N I U M DISTRIBUTION 3.3.1 Biota and Sediment of the Elk River and Tributaries Attention was first drawn to the Se issue in the Elk River Valley in 1995 during an effluent permit amendment when Se levels of 25 ug/L were detected in surface water downstream from one of the coal mines. Further sampling (Figure 3.3) showed that Se concentrations in water of some of the mine affected tributaries were 5 to 10 times greater than the provincial guideline, set at 2 ug/L as of August 2001 (Nagpal, 2001). Seepage from waste dumps contained up to 542 ug/L Se. Remarkably, concentrations in sediment, algae, insects and fish were only 2 to 5 times higher than at reference sites (MacDonald and Strosher, 1998). The situation nonetheless warranted further investigation as Se levels in Westslope cutthroat trout tissues were as much as two times greater than the respective toxic effects thresholds for muscle and liver of 8 and 12 mg/kg Se dry weight (MacDonald and Strosher, 1998). Though the study produced no evidence of deformities or increased mortality in cutthroat trout caught downstream of the mines, serious teratogenic effects could have occurred (Lemly and Smith, 1987, Lemly, 1997). Two of the 17 fish analyzed in the MacDonald and Strosher study had very high Se levels (64 and 81 mg/kg Se dry weight) in their eggs. 42 Analytical deflection l imi i - I ug/1, A.M.A.D • Apr, May, Aug, Dec 0 O 4 ' m i n c w a * ' * r o c l c dumps kilometers 0 5 10 15 20 Figure 3.3 Total Se in Tributaries of the Elk River (from McDonald and Strosher, 1998) 43 The remaining eggs contained between 8 and 25 mg/kg Se dry weight. Lemly (1997) determined that that the occurrence of teratogenic effects increases significantly at Se concentrations greater than 10 mg/kg dry weight in eggs. However, McDonald and Strosher's (1998) study showed that even at Se concentrations of 80 mg/kg in egg, no decrease in survivorship was observed suggesting that the Elk River's Westslope cutthroat trout may have a relatively elevated tolerance to Se. In order to conclusively establish if Se has any adverse effect on biota in the Elk River watershed, a comprehensive study has been undertaken by the five coal mines. This study, conducted by EVS Environment Consultants, will include long term monitoring of the cutthroat trout populations and an investigation of possible Se bioaccumulation in aquatic birds. Studies focussing on Se mobility will complement those addressing potential ecotoxicity. They will help clarify whether species in the Elk River Valley have a naturally high tolerance for Se or if environmental conditions, such as temperature, pH, flow rate, sediment composition and the amount and type of particulate matter suspended in the water column, affect complexation and speciation and render the Se less bioavailable. 3.3.2 Selenium in the Mist Mountain Formation Until the recent study by Ryan and Dittrick (2000) from the British Columbia Ministry of Energy and Mines, relatively little data on Se occurrence in the province's coals were available. Coal samples collected from the Mist Mountain Formation at the Fording River and Coal Mountain operations (Goodarzi, 1987; Goodarzi 1988, Goodarzi and Swaine, 1993) had an average Se content of 1.36 mg/kg. Ryan and Dittrick (2000) sampled all accessible strata of the Mist Mountain Formation at the five mines in the Elk River Valley. Samples from waste piles and tailings ponds were also collected. Samples of 4 to 5 kg were obtained for each stratum by chipping material from the pit walls at regular intervals. Partings in coal seams, consisting generally of carbonaceous mudstone, were sampled separately, as were the hanging wall and foot wall of seams where there was a distinct band of carbonaceous mudstone between the seam and 44 the rock between the seams, termed interburden. Figure 3.4 illustrates the position of hanging wall, foot wall, parting material and interburden relative to the coal seams. There was no major stratigraphic control of Se concentration in coal seams (Ryan and Dittrick, 2000). A weak trend of increasing Se concentration was observed moving upward in the Mist Mountain Formation. Due to the extensive folding it is not possible to correlate Se levels in strata at the different mines. Average concentrations of Se in the different lithologies are presented in Table 3.2. The average Se content of Mist Mountain coals (1.9 mg/kg) is below the global average of 2.15 mg/kg for coals. Average Se concentrations in interburden ranged from 1.1 mg/kg for coarser grained material to 3.2 mg/kg for mudstones with coal stringers. The highest concentrations were observed in samples collected from transitional zone between coal and interburden (Ryan and Dittirck, 2000). k— INTERBURDEN HANGING WALL COAL SEAM FOOT WALL V INTERBURDEN -l, PARTING y- COAL SEAM 4 — INTERBURDEN Figure 3.4 Cross-section of Pit Wall Illustrating the Different Types of Material Sampled 45 Table 3.2 Average Se in Different Lithologies of the Mist Mountain Formation (From Ryan and Dittrick, 2000) Lithothology Se concentration (mg/kg) Number of samples Coal 1.9 107 Hanging wall 4.2 21 Foot wall 4.2 21 Partings 3.2 23 Coarse breaker refuse 2.8 24 Interburden 1.1 130 Se concentrations in the hanging wall and footwall were more than twice the levels found in the coal seams. Partings also contained relatively elevated Se levels. It is interesting to note that refuse material contains less Se than the hanging wall, foot wall and parting samples. Based on the Se content of these materials and the relative amount of them present at the sites, Ryan and Dittrick (2000) estimated that 80% of the Se is found in the interburden, while 6 to 20% is in the coal and 5 to 10% in the hanging wall, foot wall and parting material. Ryan and Dittrick (2000) also collected samples from the Fernie Shale, which underlies the Mist Mountain Formation and is exposed at a number of locations. It was thought that it might contain relatively high concentrations of Se since it formed at the margins of the Fernie Sea, however, Se concentrations in samples from the Fernie Shale averaged only 1.1 mg/kg Se. Ryan and Dittrick (2000) analyzed incremental samples from two coal seams to determine to what extent Se concentrations varies within seams. Se and ash concentrations were closely correlated in the cross-section of these seams. Float sink data obtained by Ryan and Dittrick (2000) for hanging wall and footwall samples indicate that Se concentrations in ash, consisting of sulphides, clays and HFMO, are slightly higher than those in organic matter. The same authors suggest that adsorption to clays is not the primary mode of occurrence of Se in coal mineral matter since a poor correlation between Se concentrations and the Al 2 0 3 /S i0 2 ratio was obtained. 46 Se content also correlates weakly with organic S. At higher ash concentrations, the Se/organic S ratio increases, highlighting the importance of Se associated with the inorganic fraction. To date, no study has been conducted to determine the main Se-bearing minerals in the Mist Mountain Formation or the geochemical mechanisms by which it is mobilized. 47 4.0 METHOD DEVELOPMENT AND PROCEDURES 4.1 SAMPLING Sampling for this study was conducted in conjunction with a team of geologists from the British Columbia Ministry of Energy and Mines who have been studying Se occurrence in the Mist Mountain Formation (Ryan and Dittrick, 2000). Samples were collected from the pit walls of the five coal mines in the Elk River Valley. An effort was made to sample the entire cross-section of the Mist Mountain Formation at each of the sites, however some strata were not accessible. Chipping material from the pit walls at regular intervals in the cross-section of each stratum 1 to 5 kg samples were obtained (Figure 4.1). • • • • Denotes line of sampling Note: Small coal seam sampled separately Figure 4.1 Sample Collection from Pit Walls 48 "ISO + I-425 --400 m 3so -r 325 4 300 4 275 2SO + 225 4 rsx>o H F 175 4 ISO 125 4" MOO 75 1 50 -I— e — coal seam * A99-4 * A99-7 * A99-12 * A99-23 - i 3 S e l e n i u m p p m 7 * Samples used in this study Figure 4.2 Stratigraphic Section of the Mist Mountain Formation at Mine A Illustrating Variations in Se Concentration with Depth (From Ryan and Dittrick, 2000) 49 »«J50 1*25 4 LOO + 525 575 A ISO c o o M 7 5 H50 4 H25 4 h o o 4 coal seam 50 • f U ' * B99-44 * B99-16 —i 9 - 1 3 S e l e n i u m p p m 7 * Samples used in this study NOTE: B99-62 is located above the top stratum in this figure Figure 4.3 Stratigraphic Section of the Mist Mountain Formation at Mine B Illustrating Variations in Se Concentration with Depth (From Ryan and Dittrick, 2000) Samples used in this study Figure 4.4 Sratigraphic Section of the Mist Mountain Formation at Mine C Illustrating Variations in Se Concentration with Depth (From Ryan and Dittrick, 2000) 51 3 6 0 -r 3 * 0 4 • 3 2 0 3 0 0 p i s o 2-40 +. 220 4 E i 0 0, a o -f e o 4 4 0 4 . J20 4 -100 4 a o 4 G O 4 20 4 • coal seam * D99-8 - 1 3 Selenium ppm 7 Samples used in this study Figure 4.5 Sratigraphic Section of the Mist Mountain Formation at Mine D Illustrating Variations in Se Concentration with Depth (From Ryan and Dittrick, 2000) 52 1 3 0 + I 2 0 ho -F ho© + coal seam oo + I; 0 2 TO +_ ©O + \ *E99-6l * E99-63 3 0 -L_ • 2 0 | 1 0 * E99-59 1 _1 3 Selenium ppm « 4 Figure 4.6 Sratigraphic Section of the Mist Mountain Formation at Mine E (East Pit) Illustrating Variations in Se Concentration with Depth (From Ryan and Dittrick, 2000) 53 Figure 4.7 Sratigraphic Section of the Mist Mountain Formation at Mine E (West Pit) Illustrating Variations in Se Concentration with Depth (From Ryan and Dittrick, 2000) 54 Hanging wall and foot wall of seams where there was a distinct band of carbonaceous mudstone between the seam and the interburden were sampled separately.. Partings in coal seams were also sampled separately. All 375 samples were collected and labeled using an A to E code to designate from which of the five mines the samples originated. The mines are not identified by their names to respect the confidentiality agreement signed at the outset of this work. The samples were then crushed in a cone crusher to a top size of approximately 50 mm and split so as to provide material for both this study and Ryan and Dittrick's (2000). Results from the latter study are graphically represented in Figures 4.2 to 4.7. The depth is reported in meters from the base of the Mist Mountain Formation. Samples in which Se concentrations were below the analytical detection limit are represented in the graphic at the limit value that ranged from 0.2 to 0.8 mg/kg, depending on interferences during analyses. 4.2 SAMPLE SELECTION AND PREPARATION It was not financially feasible to run a complete element analysis on all of the samples collected before selecting material for the static and kinetic tests. Instead, 16 pit wall samples representing the different lithologies were chosen based on the Se concentrations determined by Ryan and Dittrick (2000) (Table 4.1). Materials with relatively high Se concentrations were selected to ensure that the amount of Se in leachate would exceed the analytical detection limit and allow for a comparison between the different materials. Mineralogical analyses, sequential extractions, heavy liquid separation, and scanning electron microscopy (SEM) were used to investigate the mineralogical associations of Se in these samples. Sample preparation is outlined in Figure 4.8. Material pulverized below 200 mesh (0.074 mm) was used for all tests other than SEM and the humidity cell tests. Humidity cells were used to model the rate of Se release from 1 kg samples following procedures developed to predict acid rock drainage rates (Lawrence, 1990). 55 Table 4.1 Samples Included in the Geochemical and Kinetic Tests Sample Number Type of Material Description Se (mg/kg) A99-4 Coal Coal 5.2 A99-7 Parting Multiple partings 4.9 A99-12 Hanging wall Carbonaceous mudstone 8.0 A99-23* Parting Carbonaceous mudstone 7.0 B99-16 Foot wall Mudstone with coal 5.8 B99-44* Coal Coal 7.5 B99-62 Hanging wall Coal, silstone, mudstone 2.2 CREF* Refuse Plant refuse 3.5 C99-21 Hanging wall Coal (w/ elemental S) 1.8 C99-25 Foot wall Mudstone <0.5 D99-8 Parting Parting 2.9 E99-19* Foot wall Mixed coal & mudstone 8.4 E99-45 Hanging wall Mudstone & coal 7.1 E99-59 Coal Coal 2.6 E99-61* Interburden Mudstone 3.1 E99-63 Parting Carbonaceous mudstone 4.9 * Samples selected for the humidity cell tests ** Se values obtained from Ryan and Dittrick (2000) Samples selected for the humidity cell tests represent the main categories of material in the waste piles or exposed along the pit walls. These include coal, interburden, carbonaceous mudrocks closely associated with the coal seams (partings, hanging wall, foot wall) and plant refuse. All materials used in the humidity cell tests were crushed with a disc crusher and passed through a 3 mesh sieve (6.3 mm). In general, approximately 1 kg of each of these samples remained after the two 1 kg humidity cells were set up in duplicate. This material, once pulverized to below 200 mesh in a planetary pulverizer, was used for mineralogical analyses, sequential extractions, heavy liquid separation, and SEM. These tests were run on the eight samples used in the humidity cell experiments and on eight additional samples selected based on lithological characteristics and Se content. About 500 g of each of the additional eight samples were pulverized to below 200 mesh using the planetary pulverizer. Samples were stored at room temperature in polyethylene bags. 56 Samples from Pit Wall 4 kg Ministry of Energy & Mines ~ l k g Cone Crusher < 5 cm J ~ 3 kg University of British Columbia Sequential Extractions Elemental and Oxide Analyses 2g 1-2 g 2 kg Disc Crusher < 3 mesh |^~1 kg Pulverizer < 200 mesh 40 g ! g I — 1 l g Settling Columns < 2 um Scanning Electron Microscocv Humidity Cells Heavy Liquid Separation X-Ray Diffraction Figure 4.8 Sample preparation 4.3 MINERALOGY 4.3.1 Qualitative Mineralogy 4.3.1.1 X-Ray Diffraction The X-ray diffraction spectra of the 16 selected samples were analyzed using a D5000 Diffractometer at the Department of Earth and Ocean Sciences of the University of British Columbia. Approximately 2 g of each of the <200 mesh samples were ground with a mortar and pestle to ensure that the grains would be finer than 2 pm. Approximately 0.4 g of sample were then mixed with 2 to 3 mL of distilled water in a glass test tube and placed in an ultrasonic bath 57 for 2 to 3 minutes for dispersion of the suspension. The mixture was then pipetted onto a glass slide, coating it evenly. The same procedure was used to prepare slides with the <2 um fraction extracted from the <200 mesh fraction. The intention of was to concentrate the clay minerals. The <2 um particles were separated from the bulk of the sample using sedimentation. This procedure was repeated five times to guarantee efficient removal of the <2 pm fraction. The settling velocity of the <2 um particles in 500 mL graduated cylinders fdled with distilled water was calculated by equation 4.1 (Stake's Law). V d = D2( p - P n ) * g Equation 4.1 18n V d : velocity of sedimentation p : particle density (g/cm3) p0 : fluid density (g/cm3) n : viscosity of water (g/cm/s) g : gravitational acceleration constant (980 cm/s) D : particle diameter (cm) The diffracted x-ray detector attached to the goniometer scanned at a speed of 2°(20)/min from 2 to 33 degrees of 20. This range of 20 was broad enough to detect the main diffracted peaks of minerals typically found in coals i.e. quartz, clay minerals and carbonates. Tables 4.2 and 4.3 summarize results from the x-ray diffraction analyses. The diffractograms for the <2 um and <200 um fractions of the 5 samples included in Table 4.3 are found in Appendix 1 (Figures A 1.1 through ALIO). A strong quartz peak was observed in all samples except in the <2 pm fraction of B99-44, a high Se coal with the highest kaolinite content. In the <2 pm fraction, kaolinite contents were higher than those in the coarse fraction. Illite was enriched only in the <2 pm fraction of the mudstone interburden sample (E99-61). Carbonates were detected in only three samples. E99-61 was the only sample in which carbonates were detected in the <2 pm fraction. Calcite, ankerite and siderite, the most common carbonates in coals, were found in trace quantities in the three samples. 58 Table 4.2 Minerals in <200 mesh Fraction Determined by X-Ray Diffraction Type of Material Quartz Kaolinite Illite Calcite Ankerite Siderite A99-4 Coal +++ + + ND ND ND A99-7 Parting +++ + + ND ND ND A99-12 Hanging wall +++ + + ND ND ND A99-23 Parting +++ + + ND ND ND B99-16 Foot wall +++ + + ND ND ND B99-44 Coal +++ +++ + ND ND ND B99-62 Hanging wall +++ + + ND ND ND CREF Refuse +++ + + Tr Tr Tr C99-21 Hanging wall +++ + + Tr Tr Tr C99-25 Foot wall +++ + + ND ND ND D99-8 Parting +++ + + ND ND ND E99-19 Foot wall +++ + + ND ND ND E99-45 Hanging wall +++ + + ND ND ND E99-59 Coal +++ + + ND ND ND E99-61 Interburden +++ + + Tr Tr Tr E99-63 Parting +++ + + ND ND ND +: Amount present (estimated by relative X-ray diffractogram peak height) ND: Not detected Tr: Trace Table 4.3 Minerals in <2 um Fraction of Samples Selected for 20-Week Humidity Cell Tests Determined by X-Ray Diffraction Type of Material Quartz Kaolinite Illite Calcite Ankerite Siderite A99-23 Parting +++ + + ND ND ND B99-44 Coal ++ +++ + ND ND ND CREF Refuse +++ ++ + ND ND ND E99-19 Foot Wall +++ ++ + ND ND ND E99-61 Interburden +++ ++ ++ Tr Tr Tr +: Amount present (estimated by relative X-ray diffractogram peak height) ND: Not detected Tr: Trace 4.3.1.2 Scanning Electron Microscopy Scanning electron microscopy (SEM) was used to investigate the occurrence of selenides and sulphides. A SEM image is produced by scanning an electron beam across the surface of a sample. Characteristic fluorescent X-rays produced by a Philips XS 30 SEM were detected using 59 a Princeton Gamma-Tech energy-dispersive spectrometer (EDS) equipped with a Li-Si detector. The X-rays were generated from a volume of 1 pm3. Four pieces of A99-11, B99-44 and E99-19 weighing approximately 2 g each were selected for SEM analysis. They were mounted in an epoxy resin, allowed to harden under ambient conditions and polished. The polished sections were coated with a conductive layer of carbon. Sample A99-11, with a Se content of 10 mg/kg, was from a massive pyrite inclusion in a coal seam at Mine A. B99-44 was a coal with 8.8 mg/kg Se and E99-19 was a foot wall sample with 8.4 mg/kg Se and both have pyrite visible on their surfaces. It was assumed that samples consisting of or containing pyrite would be most likely to host selenides or Se in solid solution with sulphides given the similarity between S and Se geochemistry. These samples have the highest Se concentrations among those collected for this study, but the probability of detecting Se at 10 mg/kg is low. Reed (1995) notes that detection limit for SEM-EDS is generally around 0.1% i.e. two orders of magnitude greater than the concentrations found in the selected samples. Taking this into account, two methods were tested. For A99-11, a 10 minute line scan with 15 000 counts per second was used. A typical count time is 100 seconds (Reed, 1995). A longer count time was chosen to scan as large an area as possible and integrating the signal thereby maximizing the chance of detecting Se. For the other two samples, 8 to 10 sections of pyrite (5 pm by 6 pm) were scanned in an attempt to obtain an average for those areas. This was repeated for two mounts of each sample. No selenium was detected in the three materials analyzed. 4.3.2 Chemical Analyses The 16 selected samples were pulverized prior to analysis for Se, total S, sulphides, total organic carbon (TOC) and trace elements by ACME Analytical Laboratories in Vancouver, British Columbia. 60 ACME conducted Se analyses using ICP-MS, which has a detection limit of 0.1 mg/kg. An aqua regia digestion i.e. a 3:1 ratio of nitric to hydrochloric acid, was used to solubilize the sample. Results from the ICP-MS analysis of 1 g samples were compared with those obtained by Ryan and Dittrick (2000). Since there was poor agreement between the two sets of results (Table 4.4), a second set of samples was sent to ACME for analysis. To minimize sampling error, 15 g of each sample were then analyzed. This time, Se values differed from Ryan and Dittrick's (2000) by less than 35% (1 mg/kg) for all but 3 samples (Table 4.4). To provide a third measure of the accuracy of ACME's analyses, a set of eight samples was sent to Analytical Laboratories Services (ALS) Limited, Vancouver, British Columbia, where Se analysis was conducted using hydride generation atomic absorption spectrometry (HG-AAS) following EPA Method 7742 (USEPA, 1994). Samples were digested using a 1:1 ratio of nitric acid and hydrochloric acid as specified in EPA Method 3051 (USEPA, 1986). Se values obtained using the different analytical methods varied by more than 70% for B99-44 and CREF (Table 4.4). The difference between the Se values was below 15% for A99-7, B99-62, E99-45 and E99-63. The percent difference ranged from 20% to 50% for the remainder of the samples. Dale (1996) reviewed analytical techniques for trace elements in coal. For Se, he recommended using INAA or HG-AAS to obtain accurate results. Without interference, the detection limit of LNAA is 0.1 mg/kg and its precision is estimated to be 5% (Coleman et al., 1993). Disadvantages of LNAA include its high cost and the length of time required to obtain results. HG-AAS also has a detection limit of 0.1 mg/kg for Se in solids. The main drawback associated with HG-AAS is the need to digest the sample, which may result in serious losses of Se through volatilization (Raptis et al., 1983; Verlinden, 1982; Welz, et al., 1984). 61 Table 4.4 Interlaboratory Comparison of Se Values Expressed in mg/kg ACME ICP-MS 1 g sample ACME ICP-MS 15 g sample ALS HG-AAS 1 g sample INAA* 10 g sample Average Standard Deviation A99-4 5.9 5.2 3.4 5.2 4.9 1.1 A99-7 4.1 4.5 n/a 4.9 4.5 0.4 A99-12 6.4 6.7 n/a 8.0 7.0 0.9 A99-23 5.8 6;1 4.8 7.0 5.9 0.9 B99-16 5.4 5.0 3.4 5.8 4.9 1.1 B99-44 4.0 16.6 7.0 7.5 8.8 5.4 B99-62 2.2 2.3 n/a 2.2 2.2 0.1 CREF 11.8" 3.6 3.2 3.5 3.4 0.2 C99-21 2.4 2.5 n/a 1.8 2.2 0.4 C99-25 0.7 0.9 n/a <0.5 0.8 0.3 D99-8 2.2 2.2 n/a 2.9 2.2 0.0 E99-19 10.0 10.4 5.5 7.7 8.4 2.3 E99-59 1.5 1.7 1.4 2.6 1.8 0.5 E99-45 7.2 7.7 n/a 7.1 7.3 0.3 E99-61 3.3 3.2 2.3" 3.1 3.0 0.1 E99-63 4.2 4.1 n/a 4.9 4.4 0.4 n/a: not analyzed * INAA data from Ryan and Dittrick (2000) ** These anomolous results were excluded from the average based on criteria specified by Dean and Dixon (1951) described in the section below In recent years, ICP-MS has gained acceptance as an analytical method for Se (Ting et al., 1989; McCurdy et al., 1993; Menegario and Gine, 2000). When using ICP-MS to measure Se concentrations, signal interference from Cu, Ni, Co, precious metals and hydride-forming elements may pose a problem (Dale, 1996; D'Ulivo, 1997) and like HG-AAS, ICP-MS requires sample digestion. Ryan and Dittrick (2000) analyzed 19 samples using both ICP-MS and INAA and found that the two methods yielded similar results. Ryan and Dittrick (2000) calculated the average Se value for the 19 samples tested and found it to be slightly higher when using ICP-MS (3.4 mg/kg) rather than INAA (2.7 mg/kg) indicating that losses to volatilization were minimal. Average Se values obtained by the different analytical methods are compiled in Table 4.4. In order to eliminate the effect of anomolously high or low values on the estimate of Se content, a Quotient Test, as described by Dean and Dixon (1951), was used. The difference between a 62 questionable value and its nearest neighbour is calculated and then divided by the difference between the highest and lowest values of the set. If the ratio, Q, exceeds the Rejection Quotient, Q0.90, the value may be rejected with 90% confidence i.e. the value has a 10 % likelihood of rejection in a normally distributed population. The Q0.90 values used are those published by Dean and Dixon (1951) of 0.94 and 0.76 for n=3 and n=4 respectively (where n = number of samples analyzed). Based on this criteria three values were excluded and the average recalculated using the remaining values. Al l four available Se values were included in the average calculated for B99-44 despite a difference of as much as 74 % (8 mg/kg) between the highest and next highest values. The wide range of Se values for B99-44 is taken as an indication of greater overall variability making it less likely that a notably high or low value occurs only as a product of chance. Total S, sulphide, total C, TOC and ash concentrations were determined by ACME Analyical Laboratories, Vancouver and results are recorded in Table 4.5. Total S was determined with a CS-244 LECO analyser by ASTM Method D4239 (ASTM, 2001). A 0.1 g sample was leached with 15% HCI at 70°C for 1 hour, washed and then solution dried in a furnace at 200°C for 3 hours. Sulphide content was determined by placing evaporation residues in a furnace at 800°C for 1 hour and using a LECO analyser to measure the S remaining (Sobek et al., 1978). This method was selected over the Sobek method, which calculates sulphides as the difference between S remaining after leaching the sample with 40% HCI and organic S, eliminated by leaching the sample with 12.5% HN0 3 . While the organic S measurements obtained by the 15% HCI and 12.5% HNO3 leaches are fairly similar, the Sobek method's 1 hour leach with 40% HCI does not extract as much of the organic S as sample ignition at 800°C (Leong, 2001, pers. com.). The ASTM method used thus provides a more conservative estimate of sulphides. 63 CO 73 a < JS < •a a os a o £> >-« U "2 a. s a 0/3 ON CS in 0 0 rs rs ro vq o cs ro ^H VO p ON 0 0 r~ ON oo ON oi cs ^ H t~ ON ON r~ VO VO oo VO ro ro in VO rf t - Tf oo oo ro K NO <: b Tf ON O oo OO ro O in ro p ON rt ON m ON ro TT VO OO d vd cs T T CN Tf vd ON oo o ^ m CS CS T T - H in ,—i m cs Tf - H Tf ro vd o ^ t-; ON ro CO ON p CS vq oo ^ H p ro OO ON vo oo r 1 ON T T T T oo iri vd od TT ^ H od in Tf vd vo p in CS CN in CM r> in l— in Tt cs Tf — 1 Tf Tf o ' H 00 »—< r- CN m TT oo ON ro VO Tf cs CS o O o rr o - H *—H o o o © r-H o o © © d d d d d d d d d d d d d d 3 Cd 0 s 60 ' IH O Tf CS ,_ ro m cs ro OO Tf ro m oo Tr o ro m o ro - H cs Tf • ro O i—i d © d d d d d d d d d d d d d d 0 0 in CS ro rs) to o CS oo oo oo ro o ON O VO © ro T r T T VD VO Tf CS VO T-H Tf ^ H j—* 0 0 © d d d d d d d d d d d d d d O w H ON in p OS ON oo CS T T in 0 0 cs Tf ro oo cs Tf T f in od CS ro cs d CS 0 0 " H ro Tf u 0 0 60 <4-» ^ 'a 73 'a 73 ° -s u 60 00 60 1 1 60 *0 i-t .3 .3 1> a 60 .3 3 .3 tr, §* .3 tr O 60 a a •*-» o o So "cd •e CU 'I o cd to ed O o o cd o O cd u PU PH UH U « P< w UH OH U PH CS ro VO TT CN ^ H m 00 ON in ON ^ H ro 3 S Tf r- CS »-H Tj- VO PH cs cs i—i m VO v  cd 3 ON ON 0\ ok ON ON ON s ON ON ON ON ON CP\ ON ON oo £ ON ON ON ON ON ON ON PS ON ON ON ON ON ON ON ON < < PQ m pa O o U Q W w W W w T f c o •a I a o o o * DC = a TS O eu >-a w a o w a o U E S O 3 H U N Table 4.8 Oxide Concentrations in the <2 urn Fraction Expressed as a Percentage of Total Sample Weight A99-23 B99-44 CREF E99-19 E99-61 Si0 2 52.9 9.47 44.8 29.0 53.6 A1 20 3 19.9 4.91 17.9 10.6 19.6 Fe 2 0 3 1.51 0.48 1.63 1.55 2.53 MgO 0.88 0.34 0.66 0.44 1.40 CaO 0.44 1.13 0.58 0.34 1.72 Na 20 0.08 0.04 0.08 0.09 0.11 K 2 0 3.15 0.44 2.20 1.54 3.54 T i0 2 0.71 0.19 0.77 0.42 0.76 P 2 O 5 0.51 0.09 0.24 0.19 0.40 MnO 0.01 O.01 0.01 0.01 0.02 LOP 19.5 83.0 31.0 55.0 16.0 Total 99.6 100 99.8 99.1 99.7 * LOI: Loss on ignition Table 4.9 Trace Element Concentrations in the <2 pm Fraction Expressed in mg/kg A99-23 B99-44 CREF E99-19 E99-61 Cr 82 14 75 62 96 Ba 1730 158 673 378 711 Ni 55 <20 160 <20 <20 Sr 378 148 152 107 71 Zr 116 28 118 133 159 Y 30 <10 26 14 195 Nb <10 <10 <10 <10 <10 Sc 12 3 11 7 30 Table 4.10 Se Concentrations in the <2 urn Fraction Compared with Se Concentrations in the Whole Sample Se in <2 (im Fraction (mg/kg) Se in Whole Sample (mg/kg) % Total Weight in <2 nm Fraction % D S e in <2 um Fraction A99-23 7.3 5.9 2.2 2.7 B99-44 20.6 8.8 2.3 5.4 CREF 4.2 3.4 4.1 5.1 E99-19 11.9 8.4 2.8 3.9 E99-61 2.8 3.2 3.7 3.2 66 Total carbon in the sample was measured using the CS-244 LECO analyser after transforming all carbon species in the sample into C 0 2 by combustion in an induced furnace at 2000°C. Inorganic carbon was also determined using the CS-244 LECO analyser, but in this case, the sample was first leached in a boiling water bath for 1 hour with 5 mL HN0 3 and 5 mL of 50% HF, washed and dried at 200°C for 3 hours. TOC was calculated by subtracting inorganic carbon from total carbon as specified by ASTM Method D5291 (ASTM, 2001). Ash was determined according to the ASTM D3174-00 Standard Test Method for Ash in the Analysis Sample of Coal and Coke from Coal (ASTM, 2001). A sample of 1 g was heated from room temperature to 750°C over an hour and kept at this temperature for two additional hours. The combustion residue was then weighed. Trace element and oxide concentrations were determined using Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES). The 0.2 g samples were melted with L i B 0 2 to transform all silicates into glass and digested with aqua regia. The concentrations of major oxides (Si0 2, A1 20 3, Fe 20 3, CaO, MgO, Na 20, K 2 0 , MnO, Ti0 2 , P 2 0 5 , Cr 2O s) are listed in Table 4.6 for the whole samples and in Table 4.8 for the <2 pm fraction. Detection limits for these oxides ranged from 0.001 to 0.04%. Trace element concentrations are compiled in Tables 4.7 and 4.9 for the whole samples and <2 pm fractions. Se concentrations in the <2 pm fractions are compared with those in the whole samples in Table 4.10 . As and Pb concentrations were measured by ALS Limited, Vancouver using ICP-MS and were found to be below their respective detection limits of 200 and 100 mg/kg in all 16 samples. 4.3.3 Quantitative Mineralogy Percentages of major mineral components were calculated using data from the chemical analyses of major oxides (Table 4.11) and qualitative informartion on sample mineralogy obtained by XRD (Tables 4.2 and 4.3). The most common mineralogical formulas for the minerals were obtained from Deer (1993) (Appendix 1). Assuming that illite is the only K-bearing phase, K 2 0 Table 4.11 Percentages of the Major Mineral Components Type of % % % % % % Material Illite Kaolinite Quartz Carbonate Sulphide Total A99-4 Coal 5.71 11.3 4.02 0.71 0.84 22.6 A99-7 Parting 14.6 10.6 12.3 0.46 0.26 38.3 A99-12 Hanging wall 22.5 3.96 11.3 0.89 0.42 39.0 A99-23 Parting 21.1 6.97 20.4 1.28 0.07 49.8 B99-16 Foot wall 17.5 3.15 16.9 1.34 0.31 39.2 B99-44 Coal 3.96 1.84 1.30 2.30 0.12 9.5 B99-62 Hanging wall 6.55 5.30 5.98 1.23 0.53 19.6 CREF Refuse 13.5 17.9 14.6 2.03 0.05 48.1 C99-21 Hanging wall 7.46 6.37 6.81 2.07 0.37 23.1 C99-25 Foot wall 6.24 5.35 17.4 0.20 0.17 29.4 D99-8 Parting 13.5 20.8 9.77 0.39 0.22 44.6 E99-19 Foot wall 10.5 5.65 9.36 0.57 0.43 26.5 E99-45 Hanging wall 18.3 12.0 16.6 0.62 0.18 47.6 E99-59 Coal 9.14 14.2 6.92 0.34 0.34 31.0 E99-61 Interburden 18.0 3.01 22.3 5.85 0.03 49.1 E99-63 Parting 22.5 9.64 17.5 0.25 0.15 50.0 concentrations were used to calculate the illite content of the samples, that ranged from 3.9 % (B99-44) to 22.5% (A99-12) (Table 4.8). Kaolinte content was calculated based on the difference between total A1 20 3 in the sample and A1 20 3 in illite. The percent of kaolinite in the 16 selected samples ranges from 1.8 to 20.8% (Table 4.11). Quartz content was calculated by subtracting Si0 2 in illite and kaolinite from total Si0 2 . The results show that the quartz content ranged from 1.3 % (B99-44) to 22.3 % (E99-61) (Table 4.11). To obtain a rough estimate of the carbonate content of the sample, all CaO was assumed to be present in calcite (CaC03). Some CaO was likely present as ankerite (Ca1.05(Feo.48,Mgo.45,Mno.o4)(C03)2) and/or siderite (FeC03). Subtracting Fe in sulphides from total Fe in the samples shows that iron sources other than pyrite, on average, account for 0.7% of total Fe mass. These include siderite, HFMO, Fe complexed by organic matter and Fe in the silicates. 68 4.4 SEQUENTIAL EXTRACTIONS Single and sequential extractions are usually used to establish which mineralogical component or components bear heavy metals or trace elements in soils, sediment or rocks. This can also provide indirect information about metal bioavailability and the potential for mobilization by surface- or groundwater. As no universally accepted methodology currently exists, the Standards, Measurements and Testing Program of the European Commission has tried to establish a definitive procedure and create a database allowing for quality control and comparison between different materials (Ure et al., 1993; Quevauviller et al., 1994). There is, however, still much debate concerning the efficiency and specificity of extractants. The fractions are thus defined operationally by the reagent employed, the extraction sequence, the time, temperature and nature of contact and the sample to extractant ratio. Table 4.12 provides an overview of commonly used sequential extraction methods for most trace elements. Only the procedure developed by Chao and Sanzolone (1989) was designed specifically for Se. Figure 4.9 illustrates the four-part sequential extraction procedure used in this study. Details on extractant selection are provided in Sections 4.4.1 through 4.4.4. Al l solutions were prepared using de-ionized water obtained from a MILLI-Q water purification system. Extractions were performed in 250 ml Erlenmeyer flasks washed with 10% HN0 3 and rinsed with de-ionized water After each extraction, samples were filtered through 0.45 pm millipore filters. A volume of 2 ml of 1:1 FfN03 was added to the filtered solution to prevent the formation of insoluble metal complexes. Solids collected on filter paper were air dried and weighed after each extraction. Filtrates were sent to ALS Limited for Se and metal analysis by HG-AAS and ICP-MS respectively. Residues collected after the KCIO3/HCI extraction were sent to ACME Limited for Se analysis by ICP-MS. 69 tu 73 H © C U - M a cu M 3 • M « hi CU a •o cu • M a cu W3 cu l a PM CA cu In 3 •o cu cu o u • cu c cu 3 O" cu CO •a cu cu "3 C O C M o >-> u C3 a a 3 C O r-i H T t C U 2 H Reference Chao and Theobald (1976) Tessier et al. (1979) Hoffman and Fletcher (1979) Filipek and Theobald (1981) Bogle and Nichol(1981) Cardoso Fonseca and Martin (1986) Chao and Sanzolone (1989) Ure et al. (1993) Hall et al. (1996) Residual fi O i P H X Tf O u X 1 P H a fl O Tf o u X t ro O P H U a a fl O t fl O P H U a a I Tf o y « a o P H Z a a Tf o y •* a o P H Z a a Sulphides & Organics u X 1 2 O ^ X fl o 1 O X m O O Prt ea MH ">< O NO a § . s u a i fi o u o a ro u a i fi „ 2 d <-> t § fc4 a Crystalline Fe oxides Citrate-dithionate buffer Citrate-dithionate buffer 1.0 M NH2OH- HClin 25% HOAc d o s q a a ~ d o © 4MHC1 0.25 M NH2OH- HClin 25% HCl Amorphous Fe oxides 0.25 M NH2OH- HCl/ 0.25 M HCl 0.04 M NH2OH- HCl in 25% HOAc 3 ' * ° o u £ g S d o © a o • a TJ- a S P « O »|« © Z ' . S 0.25 M NH2OH- HCl 0.25 M HC10 •!o o u s q a £ g S d o © 4MHC1 U a a o a u s g s © Z -3 © Soluble Organics NaOCl pH9.5 NaOCl 0.25 M Mn Oxides .3 u d X Z 2 o 2 ^ ^ ^ H - H X © 6 Z d U s q S © Z P H g ' 5 d K i a ^ —i a © © Z © U a a s o ^ CN © Z u a a - a aa © Z DH Adsorbed, Carbonates 1 M NaOAc atpH5 atpH5 Exchangeable <N U S a ^ H 1 M HOAc 1 M HOAc atpH4.5 Tf o OH . - H © 1 M HOAc at pH 4.5 1 M NaOAc 70 2.0 g sample Distilled Water filter I shake 1 hr room temperature solution 1.0MNH,OH- HCI filter solution stir 2 hrs 90°C - • KClOj/HCl 1 hr 90°C filter HCl-HNO, solution 1 hr 90°C solution Water Soluble/ Exchangeable HFMO Sulphides and Organics Residual (silicates) Figure 4.9 Sequential Extraction Procedure Se concentrations were measured by HG-AAS by ALS Limited using a Perkin Elmer model 2380 HVAK atomic absorption spectrophotometer. This instrument generally has a detection limit of 0.5 pg/L for Se in solution. Se volatiliation and signal interference from matrix components, such as chlorides, sulfates, phosphates and iron may increase the detection limit. One sample from each batch was selected for quality control of the analytical procedure. In instances when results from the replicates differed from the original analyses, they did so by less than 2%. The Se concentration in the distilled water with which the cells were leached was below the detection limit of 0.5 pg/L. Metal concentrations in solution were measured by ALS Limited using Thermo Janell Ash Model ICAP61 and Thermo Janell Ash IRIS ICP-OES instruments. The procedure followed corresponds to EPA Method 6010B (USEPA, 1996). 4.4.1 Extraction Method for Exchangeable Selenium The first set of extractions targeted exchangeable Se, or more precisely, Se oxyanions held to the 71 surface of clays, HFMO or organic matter through electrostatic attraction. Four extractants were tested to determine their relative efficiency in removing exchangeable Se from E99-22, a foot wall sample containing 7.1 mg/kg Se. The extractants tested were: • 0.05 M ethylenediamine-tetraacetic acid (EDTA) • 1.0 M magnesium chloride (MgCl2) • 1.0 M sodium acetate (CH 3COONH 4, abbreviated NaOAc) • 0.1 M potassium dihydrogenase phosphate (KH 2P0 4) • Distilled water To test the extractants, 2 g of pulverized sample were placed in an Erlenmeyer flask with 100 ml of solution. The mixture was shaken for 1 hour and passed through a 0.45 pm millipore filter. The filtrate was sent to ALS Limited for Se analysis. Se concentrations in the leachate were consistently above the detection limit (Table 4.13). The sample to solution ratio used was recommended by Chao and Sanzolone (1989) and Hall et al. (1996). A low solid to extractant ratio could yield a solution with a Se concentration below the detection limit, while a high ratio could lead to incomplete extraction. In order to determine the effect of an increase in the solid solution ratio on the amount of Se extracted, 4 g were mixed with 100 ml of KH 2 P0 4 . The percent of total Se extracted from the 2 g and the 4 g samples using K H 2 P 0 4 were identical (Table 4.13), confirming that there was enough K H 2 P 0 4 present in 100 ml of the K H 2 P 0 4 solution to extract all of the exchangeable Se in 2 g of this sample. Most procedures reviewed used a centrifuge to separate the supernatant from the solid portion of the sample. However, since filtration is simpler and has been shown to be as effective as centrifugal methods (Quevauviller et al., 1996) this approach was used for these experiments. EDTA, a chelating agent, was of interest because it has been used to determine the amount of bioavailable metals in soils, sediment and biological tissues (Quevauviller et al., 1996). EDTA 72 extracted 20% of total Se from E99-22 i.e. 3 to 4 times more Se than the other solutions (Table 4.13), suggesting that it may have been removing Se from other mineral phases. MgCl 2 and NaOAc, on the other hand, are frequently used in sequential extraction procedures (Table 4.12). Extracting 6.1 and 7.8% of total Se respectively, MgCl 2 and NaOAc provided a more conservative estimate of exchangeable Se than EDTA. KH 2 P0 4 , an extractant used in several Se-specific extraction procedures (Ballistrieri and Chao, 1987; Chao and Sanzolone, 1989; Martens and Suarez, 1997), removed approximately the same amount of Se as distilled water (Table 4.13). Table 4.13 Se Extracted from Sample E99-22 (Foot Wall) Expressed in pg/L Se Extracted Blank •(ug/L) E99-22 (Ug/L) E99-22 Replicate (ug/L) Average % of Total MgCl2(1.0mol/l) 1.9 8.2 9.2 6.1 EDTA (0.05 mol/1) 0.7 36.0* 20.0* 19.7 NaOAc (1.0 mol/1) 0.7 10.6 11.6 7.8 KH2PO4(0.1 mom) 0.7 7.9 7.6 5.5 KH2PO4(0.1 mom)** 0.7 16.3 15.3 5.5 Distilled water <0.5 6.1 7.8 4.9 * Total Se was measured as a precipitate had formed prior to arrival of the samples at ASL ** 4 g of sample were used instead of 2 g as for the other samples Reagents such as NaOAc, MgCl 2 and K H 2 P 0 4 release adsorbed Se anions into solution replacing them with acetate, chloride and phosphate anions. Distilled water extracts only marginally less than the other three solutions. This suggests that the Se being removed by these extractants is primarily in the form of salts, possibly the products of sulphide oxidation. Distilled water was selected as it enabled a comparison between the sequential extraction and humidity cell results. Lacking significant ionic potential, distilled water does not remove ions through an exchange process. The fraction is therefore more accurately termed water soluble and will be refered to as such in the text. 73 In the initial round of tests, the goal was to assess the relative strength of the different extractants rather than to determine an optimal extraction time so a 1 hour extraction time was used (Dold, 1999). Extractions periods of 0.5, 1, 2 and 4 hours were tested to select a final extraction time that would maximize Se removal and minimize Se re-adsorption. Maximum Se removal occured after 2 hours (Table 4.14). However, the 2 hour extraction removed less Se than the 1 hour extractions in the initial round of tests (Table 4.13). The difference between the three 1 hour extractions (Tables 4.13 and 4.14) was greater than the difference between the least and most effective extractions in the time trial, suggesting that a 1 hour extraction would provide as accurate a measure of water soluble Se as the 2 hour extraction. Table 4.14 Results from Time Trials for the Extraction of the Water Soluble Phase from Sample E99-22 (Foot Wall) Extraction Time (hours) Se Extracted ug/L % of Total Se 0.5 4.3 2.4 1 4.8 2.7 2 5.9 3.3 4 5.1 2.9 4.4.2 Extraction Method for H F M O Associated Selenium Reagents that have been used to extract metals and trace elements associated with amorphous and crystalline HFMO include: citrate-dithionate buffer (CDB), ammonium oxalate in oxalic acid (NH 4) 2C c0 4 in H 2 C 2 0 4 ) , HCI and hydroxyl ammonium chloride (NH2OH- HCI) in 25% acetic acid (Table 4.9). Hall et al. (1996) note that CDB is often highly contaminated and may yield sulphides precipitates to which trace elements may adsorb. They also note though it dissolves Fe-rich layer silicates, it does not effectively dissolve hematite or goethite. Ammonium oxalate in oxalic acid, known as Tamm's reagent, is not effective on samples with a high organic matter content (Hall et al., 1996), making it unsuitable for use with the hanging wall, foot wall and coal samples in this study. HCI effectively dissolves HFMO, but it may partially attack sulphides (Hall 74 et al., 1996). For the reasons listed above, CDB, Tamm's reagent and HCI were not given further consideration. To roughly assess the effectiveness of NH 2OH- HCI, a 2 g sample of limonite sample (from the Carajan Region of the Brazilian Amazon) was dissolved in 100 ml of a 1.0 M NH 2OH- HCI in an Erlenmeyer flask in a 90°C water bath. Approximately 80% of the sample was dissolved. Tessier et al. (1979) investigated the specificity of NH 2OH- HCI measuring both the sulphur and carbon content of the solids before and after the extraction. Both remained constant indicating that neither sulphur in sulphides or in organic material was affected by the extractant. Taking the limonite test and Tessier et al.'s (1979) results as grounds to assume that NH 2OH- HCI was sufficiently selective, the concentration and contact time were then selected. Hall et al. (1996) removed first amorphous HFMO from soil and till samples, using an extraction with 0.25 M NH 2OH- HCI lasting 2 hours, and then crystalline HFMO, with a 3 hour extraction with 1.0 M NH 2OH- HCI. Tessier et al. (1979) employed a single 6 hour long extraction with 0.04 M NH 2OH- HCI. Since we sought to remove Se from both amorphous and crystalline HFMO with a single extraction, a 1.0 M NH 2OH- HCI solution was used. 2 g of E99-22 crushed to <200 mesh and mixed with 100 ml of solution for time trials of 1, 2, 4 and 8 hours. Matrix interference made it impossible to detect Se in filtrate from these extractions. As much as 5% of total Se could have been extracted from the samples without being detected if the concentration were just below the 10 pg/L detection limit. If all Fe in the samples not accounted for by pyrite were in the form of HFMO, the mass of iron and manganese oxides contained in the 16 selected samples would range from 0 to 6% of the total. Making this assumption, HFMO account for on average only 1%. In order to determine whether signal interference would pose a problem for all analyses of Se in NH 2OH- HCI or if no Se was associated with HFMO in E99-22, three other samples, B99-25, B99-44 and CREF, were tested and the contact time set at 3 hours following the procedure developed by Hall et al. (1996). These samples were selected on the basis of their Fe 20 3 contents 75 which ranged from 0.27 to 2.24 %. There was less signal interference and 6.5, 4.4 and 4.8% of total Se was found to be associated with HFMO in B99-25 (a coal sample containing 3.1 mg/kg Se), B99-44 and CREF respectively. Based on the results of the exploratory tests, 100 mL of 1.0 M NH 2OH- HCI in 25% acetic acid were used to extract the Se associated with amorphous and crystalline HFMO from the 16 selected samples. For the NH 2OH- HCI extraction, the flasks were sealed with a piece of paraffin film, placed in a 90°C water bath and stirred constantly with a magnetic stirring rod for 3 hours. 4.4.3 Extraction Method for Sulphide and Organic Matter Associated Se Chao and Sanzolone (1977) found KC10 3 combined with HCI to be the most effective method to extract trace elements associated with sulphides and organic matter. Chao and Sanzolone (1989) use KCIO3/HCI in their procedure developed to extract Se. Hydrogen peroxide (H 20 2) has also been used to extract trace elements associated with the sulphide/organic fraction (Tessier et al., 1979; Filipek and Theobald, 1981; Ure et al., 1993). A pyrite sample from Zacatelas, Mexico (Ward's Natural Science Establishment Incorporated, catalogue reference number 46 E 6445) and the organic fraction of B99-44 collected via heavy liquid separation (see Section 4.5) were used to test the relative efficiency of KC103/HC1 and H 2 0 2 . A 100 ml volume of KC103/HC1 digested 95% of the 0.1 g pyrite sample in an hour at room temperature, while the same volume of 35% H 2 0 2 dissolved only 75% of the pyrite under the same conditions. KCIO3/HCI and 35% H 2 0 2 displayed similar efficiency in destroying organic matter, with the former reducing the mass of the 2 g sample of organic matter from B99-44 by 69% and the latter by 76%. 10% H 2 0 2 dissolved 39% of a 0.1 g of pyrite suggesting that one might be able to first use a low concentration H 2 0 2 wash to remove organics, however, this proved unfeasible when 10%> H 2 0 2 failed to digest more than 37% of a 2 g sample of organic matter from B99-44. 76 The time trials confirmed that KCIO3/HCI was more effective than H 2 0 2 in targetting the combined sulphide/organic fraction, with optimal Se extraction occurring after 1 hour (Table 4.15). For the KC103/HC1 extraction, a modified version of Hall et al.'s (1996) procedure was used. A volume of 40 mL of concentrated HCl was gradually added to Erlenmeyer flasks containing 2 g of sample and 2 g of KC10 3. The flasks were covered and the mixture stirred continuously with a magnetic stirring rod. At the end of the extraction, the samples were topped up to 100 ml with de-ionized water. Table 4.15 Results from Time Trials for the Extraction of the Sulphide and Organic Matter Associated Se from Sample E99-22 (Foot Wall) with KCIO3/HCI and H 2 0 2 Extractant Extraction Time (hours) Se Extracted % of Total Se 35% H 2 0 2 1 250 65.6 2 270 73.7 4 240 60.5 8 350 50.3 KCIO3/HCI 0.25 150 84.3 0.5 140 78.7 1 150 84.3 1.5 140 78.7 4.4.4 Extraction Method for Selenium in Silicates The final digestion, extracting Se from the silicate lattice, was conducted by ACME Analytical. The solid residue from the sequential extractions was fused using L i B 0 2 and digested using aqua regia. Most procedures also use HF (Table 4.12) to ensure complete dissolution of organic material. The sum of the Se extracted from the four fractions differed little from the Se totals (see Section 5.2) indicating that the aqua regia digestion was sufficiently rigorous. 4.5 H E A V Y LIQUID S E P A R A T I O N Heavy liquid separation was used to determine if the Se was concentrated in the organics or in the sulphides. Eight samples were split into three density fractions (Figure 4.10). A mixture of 77 perchlorethylene and methylene bromide with a density of 1.6 g/cm3 was used for the first separation. About 20 g of sample ground to <200 mesh were mixed with 300 ml of the heavy liquid in a glass separator and left over night. 20 g sample I Perchlorethylene d = 1.6 g/cm3 FLOAT Se Analysis 1 SINK Methylene iodide d = 3.3 g/cm3 FLOAT Se Analysis 1 SINK Se Analysis Figure 4.10 Heavy Liquid Separation Procedure The sink fraction was collected, rinsed with methanol and mixed with methylene iodide with a density of 3.3 g/cm3. The float fraction from the first separation, as well as the float and sink fractions from the methylene iodide separation were air dried and weighed. All samples were sent to ACME Analytical Laboratories for Se analysis (refer to section 4.3 for details concerning the analytical method). Replicates were run for each of the samples. Due to the fine particle size, in a few cases distinct float/sink fractions were not generated. In such instances, approximately half of the material was categorized as light and the remainder as heavy. 4.7 HUMIDITY CELLS Two bench-top cells were mounted for each of the five materials selected for the 20-week humidity cell tests in the Surface Chemistry Laboratory in the Department of Mining and Mineral 78 Process Engineering. Samples included in the humidity cell tests are listed in Table 4.16 These samples were selected to represent the major classes of material being handled at the sites. Materials relatively high in Se were selected to ensure that Se concentrations in leachate would be above the detection limit. Table 4.16 Samples Included in the Humidity Cell Tests Designation in Tables Se Content (mg/kg) E99-61 Interburden 3.2 CREF Refuse 3.4 A99-23 Parting 5.9 B99-44 Coal 8.8 E99-19 Foot wall 8.4 The humidity cells tests were run following ASTM procedure D5744-96 (White and Sorini, 1997). The 30 cm high and 10 cm wide cells were made of Plexiglass pipe. A hole for air supply was drilled into the side of the pipe, 1 cm from the base of the cells as illustrated in Figure 4.11. A perforated acrylic plate supported the sample above the air feed. Another hole was drilled into the center of the humidity cell cover through which to add distilled water for the leaches and to allow air to escape. A final hole was drilled into the bottom of the cell to drain the leachate. A plastic connector piece, to which tubing was connected, was screwed into each of these holes. Three pieces of plastic coated window mesh were placed at the bottom of each of the cells to minimize the loss of fines during the weekly leaches. 1 kg of sample crushed to below 6.3 mm (1/4 inch) was loaded into each of the cells. The material was thoroughly mixed before being placed in the cell to ensure that the fines would be even distributed throughout the cells. Before putting the covers on the cells, a circular piece of absorbent cloth was placed on top of the material to distribute water evenly during the leaching step. The humidity cells were then placed on a metal stand (Figure 4.12) and operated following a cycle of 3 days of moist air, 3 days of dry air and a 1-hour leach on the seventh day using 500 mL distilled water. Distilled water input J f Air released via bubblers , • . Air supply Plexiglass tube Perforated support Leachate collection Figure 4.11 Humidity Cell Diagram Figure 4.12 Humidity Cells in Operation 80 (1/4 inch) was loaded into each of the cells. The material was thoroughly mixed before being placed in the cell to ensure that the fines would be even distributed throughout the cells. Before putting the covers on the cells, a circular piece of absorbent cloth was placed on top of the material to distribute water evenly during the leaching step. The humidity cells were then placed on a metal stand (Figure 4.12) and operated following a cycle of 3 days of moist air, 3 days of dry air and a 1-hour leach on the seventh day using 500 mL distilled water. After each of the weekly leaches, the connector on top of the cell was screwed back in and the stopper on the tube connected to the air supply was re-opened. Moist air from a humidifier was passed through the cells for the first three days following each leach and dry air for the next three. Water in the humidifier was kept between 26 and 28°C. A single hose supplied air to the cells. Airflow through the cells was maintained at approximately 0.5 litres per minute using an airflow regulator. The cells were connected to the air hose by plastic T-connectors. The uniformity of airflow was monitored using "bubblers". These consisted of 100 mL vials containing 50 mL of water capped with a rubber stopper fitted with two glass tubes, one long and the other short. The rubber outflow tubes connected to the top of the humidity cells were attached to the glass tube extending into the water forcing the air to escape through the vial as bubbles. As noted by Frostad et al. (2000), material in the cells connected closest to the tube supplying air to the system tended to dry faster than material in the cells attached at the end of the air supply hose. Less carbonaceous materials also tended to dry faster. Since dry materials received increased airflow, screw clamps were installed on the inlet hose of all the cells to allow airflow to each cell to be regulated independently. The ASTM Procedure (White and Sorini, 1997) requires a minimum test duration of 20 weeks, while Price (1997) recommends a minimum of 40 weeks. In this case, Se concentrations in the leachate from these columns reached relatively constant levels as early as the fourth week, with 81 the notable exception of the parting material, such that it was feasible to stop the tests after 20 weeks. The pH and Eh of the leachate were measured in all samples immediately after filtering. The potential meter was equipped with Ag-AgCl electrodes filled with saturated KC1. In order to transform the potential obtained by the calomel electrode into E S H . E (standard hydrogen electrode), 199.0 mV were added to the potential readings (Light, 1972). Leachate from the humidity cell tests was analyzed by ALS Limited, Vancouver for Se every week and for dissolved metals sulphates, total inorganic carbon (TIC), total organic carbon (TOC), alkalinity and acidity every second week. As with samples from the sequential extractions, 2 ml of 1:1 FTN03 were added to the solutions to prevent the formation of insoluble metal complexes during transport of the solution to the analytical lab. To prevent the growth of bacteria, 2 ml of 1:1 H 2 S0 4 were added to samples sent for TIC and TOC analysis. Se in humidity cell leachate was determined using HG-AAS and total metals by ICP-OES. Total alkalinity was measured using a colorimetric method adapted from EPA Method 310.2 (USEPA, 1979). Methyl orange was mixed with a weak buffer solution acid at pH 3.1 and added to the sample in a fixed amount. Since methylorange has a pH range similar to that of the equivalence point for total alkalinity, the loss of colour corresponds to the amount of alkaline species present. The change in colour is measured with a COBAS FARA II spectrophotometer and compared against a standard curve. Acidity was determined following American Public Health Association (APHA) Method 2310 (APHA, 1998) that called for potentiometric titration of a sample aliquot with a standard solution of NaOH to pH 8.3. Though it is expressed in milliequivalents CaC0 3 per litre, all acidic species are measured. 82 ALS Limited analyzed TIC and TOC with a Shimazu TOC Analyzer model TOC-5000A. The analysis was carried out using procedures adapted from APHA Method 5310 (APHA, 1998). Total carbon was measured using a LECO analyzer that converted organic carbon to C 0 2 by combustion at 980°C. A second carbon analyzer was used to convert inorganic carbon to C 0 2 by acidification with phosphoric acid at 200°C. The amount of C 0 2 produced, proportional to the amount of carbon in the sample, was measured with an infrared detector. TOC, representing the carbon fraction covalently bonded in organic molecules was obtained by subtracting TIC from total carbon. Sulphate concentrations were determined by APHA Method 4500-SO4 (APHA, 1998). Sulphate ions are converted to aluminum sulphate suspension in which turbidity was measured using a COBAS FARA U spectrophotometer. The reference method for Sulphite is 4500-SO3 B Iodometric Method (APHA, 1998). An acidified sample containing sulphite is titrated with a standardized potassium iodide-iodate titrant. Free iodine, liberated by the iodide-iodate reagent, reacts with S0 3 . The titration endpoint is signaled by the blue color obtained when the first excess iodine reacts with the starch indicator. At the end of the 20-week period, the material from the cells was crushed to <200 mesh and sent to ACME Limited, Vancouver for Se analysis. 83 5.0 R E S U L T S A N D DISCUSSION Based on the methodology applied to 16 samples representative of the different lithologies of the Mist Mountain Formation this chapter aims to answer three main questions: 1. In which mineralogical associations is Se found? 2. From which materials is Se being released and how fast is this occurring? 3. What geochemical mechanisms are involved in Se mobilization? 5.1 C O R R E L A T I O N O F S E L E N I U M C O N C E N T R A T I O N S W I T H S A M P L E M I N E R A L O G Y In order to establish the predominant Se-bearing phases in waste rock from the mining operations, the correlation between concentrations of Se and TOC, organic S, sulphides and trace elements reported in Table 4.5 was assessed. Pearson's correlation coefficient, r, was used as a measure of the strength of the linear relationship between variables. It is calculated using equation 5.1. r = S(Xi-X)(Yj - Y) Equation 5.1 y / s p C i - X ) 2 Z ( Y i - Y) 2 The statistical significance of the correlations was determined using the Student t-test with the level of significance set at ct=0.05. The relationship was said to be significant when P<0.05. P is the probability of obtaining, by chance, a value equal to or more extreme than that observed. None of the trace elements found in the 16 samples studied (Table 4.5) are well correlated with Se (Table 5.1). These elements are not expected to form Se-bearing minerals, but based on knowledge of their modes of occurrence correlations with Se might have indicated whether Se was found primarily in organic or inorganic associations. As and Pb were not included in this table as concentrations were below the detection limit for all samples tested. Se concentrations also correlate poorly with carbonates (r = -0.018), kaolinite (r = -0.340) and illite (r = 0.228). 84 Table 5.1 Correlation Coefficients for Se versus Various Trace Elements * r Ba 0.160 Ni -0.014 Sr 0.109 Zr -0.206 Y -0.369 Nb -0.118 Sc -0.180 * In cases where concentrations were below the detection limit, values half that of the limit were assigned A correlation between Se and TOC would be expected if organic matter were the main Se-bearing component. However, no correlation was observed between Se and TOC (Figure 5.1), suggesting that organic associations are not of primary importance in the 16 selected samples. An equally poor correlation is noted between organic S and TOC (Figure 5.2). Coleman et al. (1993) and Chou (1990) report a positive correlation between organic S and TOC in U.S. coals. The relationship of organic S with TOC may be less consistent in materials other than coal. Looking at only the 7 high TOC samples (TOC >40%) in Figures 5.1 and 5.2, one finds that both Se and organic S correlate well with TOC (r = 0.832 and 0.814 respectively). But because there is a positive correlation between sulphides and TOC (Figure 5.3), the correlation between organic S and TOC may simply reflect an increase in Se associated with sulphides rather than an increase in the amount of Se substituting for organic S. Indeed, studies of coals from British Columbia and the U.S. have shown a positive correlation between TOC and pyritic S (Wandless, 1959; Coleman, 1993), likely resulting from the microbial reduction of sulphate to sulphides occurring in low 0 2 environments like coal swamps (Spears et al., 1994). 85 10 ox -£ CU 09 6 4 -2 -0 r = 0.002 P = 0.993 0 1 1 — 20 40 TOC (%) 60 80 Figure 5.1 Concentration of Se versus Concentration of TOC 0 2 U '2 ed ox -C TOC (%) Figure 5.2 Concentration of Organic S versus Concentration of TOC • refuse • coal A parting • interburden hanging wall/foot wall 86 eu - G •£ "5 20 40 TOC (%) 60 80 Figure 5.3 Concentration of Sulphides versus Concentration of T O C 1.2 -j 1.0 -2 0.6 4 a 3 0.4 H 0.2 4 1 0.0 r = 0.540 P = 0.875 0.0 0.2 0.4 Organic S(%) 0.6 Figure 5.4 Concentration of Total Sulphur versus Concentration of Organic Sulphur • refuse I coal A parting # interburden hanging wall/foot wall 87 Holuszko et al. (1992) report an average total S content of 0.57 % for metallurgical and thermal coals of southeastern British Columbia. S contents of materials included in this study range from 0.40 to 0.83 %. While organic S shows no evident association with total S (Figure 5.4), sulphides and total S in the 16 samples are highly correlated (Figure 5.5) because sulphides account for, on average, 70% of total S (Table 4.5). In three samples, B99-44 (coal), C-REF (refuse) and E99-61 (interburden), sulphide content is below 30% of total S (Table 4.3). Pyritic S was identified as a major contributor to total S in lignite from Spain (Monterroso and Macias, 1998) and in bituminous coals from the U.K. (Spears et al., 1994) suggesting that the sulphide/total S ratio in B99-44, C-REF and E99-61 is abnormally low. Sulphides are known to be important Se-bearing minerals (Fleischer, 1959; Lakin, 1973; Wedepohl, 1978; White et al., 1989; Hickmott and Baldridge, 1995). Figure 5.6 though, shows no link between Se and sulphides. Organic matter can influence the Se content of sulphides in two ways: 1. By drawing from the pool labile Se in the coal swamp (Price and Shieh, 1979) 2. By decreasing the redox potential in the depositional environment (Brown et al., 2000) In the first case, organic matter competes directly with sulphides for available Se. In the second, it creates a more reducing environment. This, in turn, leads to an increase in the ratio of Se to S incorporated into diagenetic pyrite as Se is more readily reduced than S (Zehr and Oremland, 1987). In order to distinguish between the effects of sulphides and TOC on Se levels, Se was plotted against sulphides/TOC ratio (Figure 5.7). By dividing sulphides by TOC, all 16 samples can be compared as if they contain a fixed amount of TOC. In this way, a sample with a high Se concentration related to its high TOC content can be distinguished from a sample with high Se, but low TOC. Se levels in 11 of the 16 samples showed a high degree of correlation (r = 0.916) 88 0.2 0.4 0. Sulfides (%) 1.0 Figure 5.5 Concentration of Total Sulphur versus Concentration of Sulphides "wo E 10 8 6 4 2 0 A A r = 0.021 P = 0.937 0.0 0.2 0.4 0.6 Sulfides (%) 0.8 —i 1.0 Figure 5.6 Concentration of Se versus Concentration of Sulphides • refuse • coal A parting • interburden hanging wall/foot wall 89 Sulfides/TOC ' Ni Note: Excluded from linear regression and calculation of the correlation coefficient • refuse I coal A parting • interburden hanging wall/foot wall Figure 5.7 Concentration of Se versus Concentration of Sulphides Normalized for TOC with the sulphide/TOC ratio. Only the interburden sample and four samples with Se contents greater than 6 mg/kg Se did not fit this trend. The interburden sample was excluded from the linear regression on the grounds that it was the only sample in this study deposited in an environment geochemically distinct from a coal swamp. Eh and pH in this environment were likely higher than those found in coal swamps or transitional environments (Casagrande et al., 1977) which would affect Se speciation and binding. Four other samples were excluded from the linear regression: A99-12 and E99-45, hanging wall samples with 7.0 and 7.3 mg/kg Se, E99-19, a foot wall sample with 8.4 mg/kg Se and B99-44, a coal with 8.8 mg/kg Se. Se concentrations in A99-12, E99-19 and E99-45 were almost two times greater than the average values for hanging wall and foot wall material in the Mist Mountain Formation (Table 4.3). B99-44 contained almost four times as much Se as the average coal in the formation. Se in these samples likely has an atypical mode of occurrence, suggesting the presence of selenides or high Se sulphides, and for 90 atypical mode of occurrence, suggesting the presence of selenides or high Se sulphides, and for this reason do not provide a fair linear correlation between Se and the sulphide/TOC ratio. No selenides were detected in these samples by microanalysis with SEM, but this cannot be considered conclusive as samples with Se concentrations as low as 10 mg/kg are usually unsuitable for SEM-EDS analysis. Galbreath and Brekke (1994) and Katrinak and Benson (1995) were able to detect Se in coal pyrite using SEM, but proton-induced X-ray emission (PLXE) may be more effective for samples with low Se concentrations (Hickmott and Baldridge, 1995). Clauthalite and ferroselite have been found in coals from Western Canada (Goodarzi and Swaine, 1993). The study of correlations between Se and other elements shows that Se is not found in a single, dominant geochemical association. Rather, Se concentrations seem to be controlled by an interaction between sulphides and organic matter resulting from direct competition for Se or from indirect effects on redox conditions in the depositional environment. 5.2 SEQUENTIAL EXTRACTIONS Results from the four-step sequential extractions are presented in Table 5.2. The amount of Se in leachate and residue is reported as a percentage of the total amount of Se extracted from the 16 samples. Actual Se concentrations in the leachate and in the solid residue are found in Table A3.1 of Appendix 3. Percentages were calculated using average Se concentrations from the duplicates. Concentrations were below the 3 pg/L detection limit in the leachate collected from half of the NH 2OH- HCI extractions targeting HFMO-associated Se. For these samples, the Se concentration was assumed to be half the detection limit i.e. 1.5 pg/L when calculating the percentages. The magnitude of the variation in the Se concentrations found in the duplicates is presented in Table A.3.2 in Appendix 3. The sum of the Se extracted from the four fractions correlates well with the concentrations of Se measured in the head samples (r = 0.954) (see Table A.3.3 in Appendix 3). 91 Figure 5.8, a graphical representation of the results in Table 5.2, illustrates that the bulk of the Se in the 16 samples is associated with sulphides and organics. This fraction contains between 60.7% and 83.7% of total Se. Water soluble Se accounts for between 2.4 % and 21.3 % of total Se. Between 1.0 % and 10.6 % of total Se is associated with HFMO. The residual fraction, consisting essentially of silicates, contributes between 5.9% and 24.7 % to the total Se. No lithology-related trend in Se distribution was observed. For instance, in the case of water soluble Se both the maximum and minimum values are accounted for by foot wall samples. On average, the water soluble, HFMO-associated, sulphide/organic-associated and residual Se account for 8 %, 4 %, 73 % and 15 % of total Se respectively. Table 5.2 Amount of Se in Sequential Extraction Leachate and Residue as a Percent of the Total Amount of Se Extracted Amount of Se Extracted (% of Total Se in Sample) Water HFMO Sulphides & Organics Residual A99-4 Coal 7.9 3.0 68.0 21.1 A99-7 Parting 2.5 3.1 77.3 17.1 A99-12 Hanging wall 5.8 3.2 79.1 11.9 A99-23 Parting 6.0 2.6 83.7 7.7 B99-16 Foot wall 21.3 3.3 59.6 15.8 B99-44 Coal 4.7 1.4 79.6 14.3 B99-62 Hanging wall 6.7 7.9 60.7 24.7 CREF Refuse 4.8 3.9 81.3 10.0 C99-21 Hanging wall 8.4 3.4 74.5 13.7 C99-25 Foot wall 4.8 8.9 74.4 11.9 D99-8 Parting 9.8 3.6 67.6 19.0 E99-19 Foot wall 2.4 1.0 79.3 17.3 E99-45 Hanging wall 18.1 1.0 72.9 8.0 E99-59 Coal 2.9 10.6 66.9 19.6 E99-61 Interburden 8.9 7.5 62.1 21.5 E99-63 Parting 10.8 4.5 78.8 5.9 ICP-MS analyses were conducted to determine the concentrations of 30 elements in the water soluble, HFMO and organic/sulphide fractions of 8 of the 16 samples. Results from these analyses are compiled in Tables A3.3 through A3.5 in Appendix 3. Of the 30 elements, less than half were present in concentrations above the detection limit. Those with concentrations above 93 Table 5.3 Correlation Coefficients Relating Se and Other Elements in Leachate from the Water Soluble, H F M O and Sulfide/Organic Matter Fractions of A99-4, A99-23, B99-16, B99-44, C R E F , E99-19, E99-59 and E99-61 Water HFMO Sulphides & Organics Aluminium * 0.15 -0.42 Barium 0.04 -0.04 -0.31 Calcium -0.03 -0.28 -0.0003 Chromium * -0.22 * Copper * 0.47 -0.40 Iron * -0.14 -0.37 Magnesium 0.07 -0.27 -0.08 Manganese * -0.09 * Phosphorous * 0.20 * Potassium 0.17 0.40 -0.09 Silicon 0.08 0.18 -0.48 Strontium -0.12 -0.27 * Titanium * * 0.89 Zinc * 0.65 -0.63 * Concentrations consistently below the detection limit the detection limit were compared with Se (Table 5.3). Among these, only Ti in the sulphide/organic fraction showed a strong correlation with Se. Organic associations are the dominant mode of Ti occurrence in coals (Mclntyre et al., 1985; Miller and Given, 1987). The strong correlation between Se and Ti is thus due to their respective modes of occurrence i.e. associated with organic matter or sulphides and not a geochemical interaction between them. Correlations between Zn and Se in leachate from the HFMO and Sulphide/Organic fraction also seem to be relevant, but they are not statistically significant. Data in Table A3.2 in Appendix 3 shows that a few extreme values bias the r value in a situation where there is otherwise little correlation between the two sets of values. The correlations between Se solubilized by the water extraction and TOC, sulphide and Se concentrations in the solids and are weak (Table 5.4). Were a significant amount of Se ionically bound to the surface of organic matter, there should have been a correlation between Se in the 94 water soluble fraction and the amount TOC in the material. This was not the case in the 16 samples included in this study. A link between water soluble Se and sulphide concentrations Table 5.4 Correlation Coefficients Relating T O C , Sulphides and Se in the Solids and Se Solubilized by Water for 16 Samples r TOC - 0.345 Sulphide 0.056 Se 0.075 seemed more probable given that Chang and Berner (1999) determined sulphide oxidation to be the primary source of S released in the process of coal weathering. The absence of such a trend could be accounted for by variations in sulphide liberation or the attenuation of sulphide oxidation by the accumulation of HFMO on exposed surfaces (Goldhaber, 1983; Nicholson et al., 1990) particularly in the case of selenite which adsorbs readily to HFMO (Ballistrieri and Chao, 1987). Alternatively, the release of selenide/sulphide oxidation products may be affected by interactions with organic ligands. The lack of a correlation between total Se and water soluble Se confirms that the rate of oxidation varies as a function of the form in which Se occurs in the material or of other mineralogical factors. Se has been shown to adsorb onto HFMO, particularly in the form of selenite (Balistrieri and Chao, 1989). With PZC values between pH 7 and 9 (Forstner and Wittmann, 1979), HFMO could potentially release Se adsorbed to pH-dependent charges since the pH measured in surface water in the Elk River falls within that range. On average though, HFMO account for only 1.3 % of sample mass in the 16 samples herein studied, such that HFMO-associated Se would not be expected to account for a large portion of total Se in the lithologies of the Mist Mountain Formation or of the Se being released into the Elk River. 95 Cu •a 3 •J-J «8 C cu TOC (%) Figure 5.9 Percent of Sulphide/Organic Associated Se versus TOC Concentration in Solids cu IE c. "a (/} «8 0 cu 90 85 80 -75 -70 65 60 55 0 0 r =0.325 P = 0.219 — i 1 1 1— 0.2 0.4 0.6 0.8 Sulphides (%) Figure 5.10 Percent of Sulphide/Organic Associated Se versus Sulphide Concentration in Solids • refuse • coal A parting • interburden hanging wall/foot wall 96 The KCIO3/HCI extraction showed that the Sulphide/Organic fraction accounted for the bulk of the Se. Se in the Sulphide/Organic fraction, while not immediately available, constitutes a reservoir for long term Se release. Since KC103/HC1 did not selectively attack sulphides or organic matter, it was necessary to use indirect means of estimating the relative contribution of these two fractions. To evaluate the relative contributions of sulphide- and organic-associated Se to the total amount of Se in the Sulphide/Organic fraction, the percent of total Se extracted by KCIO3/HCI was plotted against the sulphide and TOC contents of the solids (Figures 5.9 and 5.10). Neither of these plots yielded a strong correlation, suggesting that the amount of Se associated with organics and sulphides varies. Price and Shieh (1979) and Casagrande and Nug (1979) studied the relationship between organic and inorganic forms of S and found that H2S from the microbial reduction of S0 4 2" could react with Fe to form pyrite or with organic matter to be incorporated into coal (Figure 5.11). Se04"2 S O 4 - 2 SeO/ 2 S04"2 Bacterial Plant Reduction - Assimilation i I H 2S 1 Reacts with Reacts with Iron Organic Matter •• Organic S Pyrite Figure 5.11 Incorporation of Sulphur in Coal (From Price and Shieh, 1979) Organic matter and sulphides would thus be competing for labile Se. A brief look at S biogeochemistry in coal swamps provides some insight into the variability of Se associations. In early coal diagenesis, ferric iron is reduced to ferrous iron, which reacts with hydrogen sulfide produced by sulphur reducing bacteria to form iron monosulfide (Casagrande et al., 1977; Given 97 and Miller, 1985). Iron monosulfide is later transformed by reaction with elemental sulfur into pyrite. Organic S, on the other hand, is incorporated into coal by the reaction of reduced sulfur species with the premaceral humic substances formed by bacterial decomposition of the peat (Chou, 1990; Brueechert and Pratt, 1996). Organic S compounds formed in peat are mostly thiols and sulfides, which gradually convert to thiophenes with increasing coal maturation. Given the similarity between the biogeochemical properties of S and Se, this is expected to substitute for S in such compounds and undergoing analogous transformations. The residual fraction is greater than expected given that the literature cites organic and sulphidic associations as being of primary importance. On average, Se in the residual fraction accounts for 15% of total Se. The high Se content of the residual fraction may also be a result of incomplete extraction of Se in the organic fraction. Se in the silicate matrix is highly resistant to weathering. 5.3 H E A V Y LIQUID S E P A R A T I O N Heavy liquid separation was used to determine the amount of Se associated with material of low, medium and high density. Se concentrations in the three density fractions for each sample and the percent distribution are listed in Table 5.6. Table 5.5 Se Concentrations and Percent Distribution in the Different Density Fractions Light (<1.6g/cm3) Medium (1.6-2.9 g/cm3) Heavy (>2.9 g/cm3) [Se] (mg/kg) % Distribution [Se] (mg/kg) % Distribution [Se] (mg/kg) % Distribution A99-4 4.2 39 5.3 43 19 18 A99-23 * 0 4.8 75 4.9 25 B99-16 * 0 4.1 44 8.6 56 B99-44 11 62 8.1 17 12 21 CREF * 0 3.1 100 * 0 E99-19 7.3 39 7.2 50 12 12 E99-59 1.2 43 0.8 15 2.9 41 E99-61 * 0 2.7 100 * 0 *No material in these fractions 98 The light fraction consists mainly of organic material. In samples A99-23, B99-16, CREF and E99-61 no light fraction (<1.6 g/cm3) floated in the heavy liquid. These samples contain relatively little TOC. The other four samples all contain high TOC. Neither B99-44, nor E99-19, produced a distinct light fraction in the separation funnel during the separation. Instead, it was possible only to split the material into two equal parts by volume, designating one light and the other heavy. Noting that all three fractions contain similar amounts of Se, it is possible that the organic and mineral particles were not liberated and remained mixed to a greater or lesser extent in all three fractions. However, calculated total Se concentrations from the summation of the amount of Se in the different fractions do correlate well with the total Se concentrations analyzed (r = 0.925). The average Se content of the intermediate density fraction (1.6 to 2.9 g/cm3) density fraction was 4.5 mg/kg. This fraction is rich in clays. The relatively high concentration of Se in the intermediate density fractions suggest that clay minerals, as well as carbonates might still have pyrite associated with them, given that little Se was found to be associated with these minerals in the sequential extractions. Fine pyrite (<20 pm) has been found to be associated even with clean coal (Frankie and Hower, 1985). The <2 pm fraction contains, on average 42% more Se. The enrichment of Se in the <2 pm fraction could be indicative of an association with fine pyrite or with clay minerals. The heavy fraction is enriched in sulphides and other heavy minerals such as HFMO and siderite. A clean separation of sulphide was not possible due to lack of liberation. However, all heavy products do show higher Se than the other fractions, indicating that part of Se is associated with heavy minerals. The correlation between sulphide and Se concentrations in the >2.9 g/cm3 density fraction was calculated. The correlation coefficient of 0.69 indicates that sulphides account for a significant portion of Se in.this fraction. A99-4, the sample containing the highest level of sulphides (0.84%) also had the'highest concentration of Se in the heavy fraction. In this instance, a clean split was obtained for the duplicates. Assuming that all the Se in the heavy fraction is 99 found in sulphides, the Se content of sulphide in this material is approximately 19 mg/kg. This falls within the 10 to 30 mg/kg range of Se concentrations in pyrite in coals reported by Leutwein (1978). The heavy liquid separations demonstrate that both organic and inorganic modes of occurrence are important for Se. The relative amount of Se in these fractions varies from one lithology to another. However, the heavy fraction contains up to four times more Se than the light fraction, suggesting that sulphide-associated Se may make an important contribution to total Se. 5.4 H U M I D I T Y C E L L T E S T S The humidity cells tests were used to assess the rate of Se release from five different types of material: interburden (E99-61), parting (A99-23), refuse (CREF), coal (B99-44) and foot wall (E99-19). Duplicate cells were mounted for each of the materials. Once a week for 20 weeks the cells were leached with distilled water. The leachate was analysed for Se, metals, sulphate, alkalinity, TOC and TIC. Results from these analyses are compiled in Tables A4.1 through A4.10 of Appendix 4. The volume of leachate collected and its Eh and pH values are also recorded in these tables. Leachate pH varied from 4.99 to 8.57 and averaged 6.61. No particular trend was noted in pH variation. Average leachate Eh was 425.6 mV, but it ranged from 309 to 540 mV indicating highly oxidizing conditions. Under these conditions, most of the Se would be found as Se (IV) or Se (VI) (Figure 2.1). Se concentrations in leachate from the 10 cells can be compared in Table 5.6 or in Figure 5.12. Although the volume of leachate collected from the different cells remained fairly constant, in some instances the amount varied by as much as 10%. To facilitate a comparison of Se release from the five materials, a graph of the mass of weekly Se release (in pg) was prepared (Figure 5.13). In the first leach (week 0), the cells containing coal released most Se, 80 pg, and the 100 (A CU H cu cu o « u CU U on a. a o a a> u a o U cu t/3 VO vi cu 3 CQ CU VO OV 101 Figure 5.13 Amount of Se in Humidity Cell Leachate in ug in 20-Week Test 102 interburden containing cells the least amount of Se, 10 pg. The footwall released just over half as much Se as the coal, while the parting and refuse released less than a quarter of that amount. Se release from the coal, foot wall and refuse samples rose after the first week, but by the third week decreased to around 0.01 mg and stabilized. Se levels in leachate from the parting cells also peaked before the third week, but did not stabilize. Amounts of Se leached from the parting material appeared to taper off somewhat after week 12, but remained higher than those from other cells. Se release from the interburden samples remained constant around 0.01 mg through the 20-week period. The peaks observed in the first three weeks result from either: 1. The release of Se oxyanions adsorbed to clay minerals. 2. The removal of sulphide or selenide oxidation products. The distilled water used to leach the cells had an average pH of 6.5. The dominant clay minerals in the Mist Mountain Formation, illite and kaolinite, have PZC values of 2.8 and 4.6 respectively (Forstner and Wittrnann, 1979) such that neither selenate nor selenite will remain adsorbed to their surfaces. If this were the source of the Se mobilized during the first three weeks, there should be a positive correlation between Se release and clay mineral content, but no such correlation is observed. A positive correlation is observed between peak Se release and sulphide content (Figure 5.14), suggesting that sulphides are the source of the Se. Peak release from the foot wall sample is low considering the fact that it contains almost four times as much sulphides as any of the other materials. Mineralogical factors such as grain size, degree of liberation, type of sulphide, neutralizing potential and trace element content affect the rate of sulphide oxidation (Lawrence and Day, 1997) and likely account for the differences observed in the rate of Se mobilization. 103 3. 250 200 4 150 100 A 50 4 0.1 0.2 0.3 0.4 Sulphides (%) 0.5 • Interburden A Parting • Refuse • Coal Foot wall Figure 5.14 Peak Se Release from the Humidity Cells versus Sulphide Content S04"2 concentrations in the leachate were measured every second week, providing a larger data set with which to assess the role of sulphide oxidation in Se mobilization. Average S0 4 2 concentrations were calculated for the duplicates. There is no overall correlation between Se and S04~2 concentrations when average S04~2 concentrations from all five materials were plotted against average respective Se concentrations in a single graph (Figure 5.15). A significant correlation between the concentrations Se and S04"2 results though, when the different materials are considered individually (Figures 5.16 through 5.20). This is a critical finding since it provides a much stronger case for sulphides as the main source of water soluble Se, than the peak release data. The correlation is strongest for the coal, interburden and foot wall samples. With r = 0.559 and a bi-modal distribution, the correlation between Se and SO4" for the parting sample is considerably weaker than for the other types of material. The lack of a correlation between Se and S04"2 concentrations in leachate from this sample and the erratic release of Se from the parting humidity cells may be due to any one of the factors affecting oxidation rates, such as texture, porosity and the degree of liberation. 104 120 100 80 60 40 H 20 0 4 .4 . 0 1 1 1 1 1 10000 20000 30000 40000 50000 • Interburden! • Parting • Refuse • Coal Footwall Sulphate (ug/L) Figure 5.15 Se versus Sulphate Concentrations in Leachate from the Five Humidity Cells 20 15 y = 0.0004x + 2.6718 r = 0.903 P = 3.45x 10"4 I 10-1 10000 20000 Sulphate (ug/L) 30000 Figure 5.16 Se versus Sulphate Concentrations in Leachate from the E99-61 (Interburden) Humidity Cells 105 120 -. 100 -80 -1 60 -(/> 40 -20 -y = 0.0032x + 5.6099 r =0.559 P = 0.093 0 5000 10000 15000 20000 Sulphate (ng/L) Figure 5.17 Se versus Sulphate Concentrations in Leachate from the A99-23 (Parting) Humidity Cells Figure 5.18 Se versus Sulphate Concentrations in Leachate from the CREF (Refuse) Humidity Cells 106 o -I 1 1 1 1 0 30000 60000 90000 120000 Sulphate (ug/L) Figure 5.19 Se versus Sulphate Concentrations in Leachate from the B99-44 (Coal) Humidity Cells Figure 5.20 Se versus Sulphate Concentrations in Leachate from the E99-19 (Foot Wall) Humidity Cells 107 Sample to sample variation in the ratio of Se to S04"2 release appears to stem from the ratio of Se to sulphides in the solids (Table 5.7). Samples with a high Se to sulphides ratio tend to show a more pronounced decline in the Se to S04"2 ratio in leachate i.e. they consistently release more Se. Given the low sulphide concentration in the interburden sample, its Se/Sulphide could be drastically inflated by even a slight error in the measurement of sulphides. Table 5.7 Se to Sulphide Ratios in the Materials Used for the Humidity Cell Tests Type of Material Se/Sulphide Ratio in Solids Slope of Se versus Sulphate in Humidity Cell Leachate Plots A99-23 Parting 0.008 0.0032 B99-44 Coal 0.007 0.0026 CREF Refuse 0.007 0.0018 E99-19 Foot Wall 0.002 0.0002 E99-61 Interburden 0.011 0.0004 * Calculated from data in Table 4.5 The multi-element analyses of the leachate provided no evidence of other mineralogical associations contributing to or affecting Se release. Two thirds of the elements included in the ICP-OES scan were consistently below the detection limit (Table A4.1 to A4.10 in Appendix 4). K, L i , Zn, Cu and Fe were above the detection limit in some of the samples. Elements consistently above the detection limit in all the samples were Ba, Ca, Mg, Se and Si. Of these, only Ca concentrations showed any correlation with Se and this was more.likely the result of the gradual dissolution of CaC0 3 than of the release of Se associated with Ca. Se showed no correlation with TIC or TOC concentrations in leachate. Dibenzothiophene, a polycyclic aromatic compound, is frequently used as a model of organic S-containing molecules the macromolecular coal matrix (Van Afferden et al., 1993). When oxidized, this compound releases S as sulfinic acid (Kargi, 1987), meaning that if S were being released primarily as a result of the oxidation of organic matter a correlation between S and TOC concentration should 108 be observed. Noting that the rate of sulphide oxidation is up to three orders of magnitude greater than that of organic matter in coal (Chang and Bemer, 1999), the lack of a correlation between TOC and Se in leachate is not unexpected. Having identified sulphides as a probable source of labile Se, due note should be given to the chemistry involved. The oxidation of sulphides and analogous Se compounds would occur as described in Appendix 6. At neutral pH. the reaction generates water soluble selenate. Selenite, if present, would be expected to adsorb readily to the Fe(OH)3 produced (Ballistrieri and Chao. 1987). The next issue to address is the rate of Se release. A significant amount (20 to 76 %) of the total amount of Se released from the humidity cells during the 20-week period is released in the first three weeks (Table 5.8). After this. Se release occurs at a regular rate (Figure 5.13) with all samples releasing between 0.04 and 0.36 % of their total Se per week (Figure 5.21). Table 5.9 presents these results in both net and relative terms. Table 5.8 Peak Se Release in the First 3 Weeks as a Percent of Total Se Release from the Humidity Cells in 20 Weeks Type of Material Peak Se/Total Se Extracted (%) A99-23 Parting 26 B99-44 Coal 59 CREF Refuse 34 E99-19 Foot wall 76 E99-61 Interburden 20 Table 5.9 Weekly Se Release after the First 3 Weeks in pg and as a Percent of Total Se in the Head Samples Weekly Se Release ug/kg Percent of Total Se in the Sample A99-23 21 0.36 B99-44 9 0.10 CREF 7 0.20 E99-19 3 0.04 E99-61 4 | 0.13 109 Acid-generation potential (AP) and neutralization potential (NP) are commonly used to predict the risk of acid rock drainage (Lawrence and Day, 1997). Although coals of the Mist Mountain Formation have low sulphide contents (Holuszko et al., 1992), the neutralization potential ratio (NP/AP) could conceivably affect the rate of S and Se mobilization. The AP and NP of the five materials used in the humidity cell tests were calculated using the formulas presented in Lawrence and Scheske (1997) (Appendix 5). For the purpose of the NP calculations, calcite was assumed to be the only carbonate mineral present. Given that siderite does not contribute to NP, the calculated value may be somewhat inflated due to the assumption about all CaO being accounted for by calcite. Relative reactivity values used in determining NP were obtained from Kwong (1993). AP was calculated based on the amount of sulphidic S in each of the materials. Table 5.10 NP and A P Values and Neutralization Potential Ratios of the Five Materials Included in the Humidity Cell Tests NP (kg CaC03/tonne) AP (kg CaC03/tonne) NP/AP A99-23 15.2 2.2 6.9 B99-44 23.3 3.8 6.2 CREF 22.9 1.6 14.7 E99-19 7.0 13.4 0.5 E99-61 60.6 0.9 64.7 There is poor correlation between the neutralization potential ratios and the net weekly Se release from the humidity cells (r = -0.335). If the ratio of neutralizing to acid-producing material exercised a dominant control on the rate of Se release, low NP/AP ratios should be associated with high Se release rates. However, the foot wall (E99-19) had the lowest NP/AP ratio and the lowest weekly Se release rate and the parting (A99-23) released more than twice as much Se per week as the coal (B99-44), though they had very similar NP/AP ratios. The parting (A99-23) released most Se over the course of the 20-week period (Figure 5.22). It contains less Se than the coal (B99-44) and the foot wall (E99-19) and only a sixth of the sulphides found in the foot wall. This suggests that the degree of liberation of sulphides from. 110 wo 8.5 % 4.3 % 2.0 % 2.8 % 5 .1% c -3 -CU s -CS CL Cu SB Cu o u C3 o e Percent of total Se extracted Figure 5.22 Percent of Total Se Extracted During the 20-week Humidity Cell Tests I l l other mineralogical components may play a role in determining long-term rates of Se release. Chen et al. (1999) demonstrated that a coating of organic compounds may passivate sulphide oxidation. Organic compounds could be responsible for slowing Se release from the foot wall and the coal (B99-44) which both contain significantly more TOC than the parting (A99-23). Goldhaber (1983) and Nicholson et al., (1990) showed that the build up HFMO on sulphide surfaces may have a similar effect, while Sharmasarkar and Vance (2001) suggest that HFMO may slow Se release, adsorbing the oxyanions. This does not appear to be occurring in this instance as A99-23 has more HFMO-associated Se than E99-19 and B99-44. Although the humidity cell tests have provided useful information, it is important to understand the limitations of this type of test. Correlation between humidity cells and run off from test plots in the field tends to be weak (Frostad et al., 2000). Factors that may contribute to this include: 1. Artificial grain size distribution Reducing grain size to below 6.3 mm for the humidity cell tests exposes a substantially greater surface area of the material to oxidative weathering than would be exposed in waste rock. Price (1997) notes that grinding and crushing may make the minerals more susceptible to oxidation as this process may expose soluble base cations and hydroxides. 2. Flushing frequency The dry air/moist air regime used in humidity cell tests is far more regular and intense than drying and wetting cycles in mine spoils (Lapakko, 1994). 3. Degree of wetting The water flush is meant to remove all of the oxidation products accumulated during the week. Due to channelling of runoff, complete wetting and flushing is not achieved in waste rock piles which generates further discrepancy between laboratory and field tests. 112 3. Degree of wetting The water flush is meant to remove all of the oxidation products accumulated during the week. Due to channelling of runoff, complete wetting and flushing is not achieved in waste rock piles which generates further discrepancy between laboratory and field tests. 4. Operating humidity cell tests at room temperature Temperatures in the laboratory are generally greater than those in the field, which may lead to increased rates of chemical and biological reactions. Though it is clear that humidity cells accelerate the weathering process, it is not clear by how much. Using results from such tests to model geochemical processes in mine spoils is difficult due to the heterogeneous nature of waste rock and the high degree of variability in field conditions (Paktunc, 1999). Differences between the percent of total Se extracted using a 1-hour shake flask extraction in the sequential extraction with distilled water with peak Se extraction from the humidity cells (Table 5.11) emphasize the importance of particle size on leaching rates and illustrate the problems of "scaling up" results even in a controlled environment with relatively homogeneous samples. In the case of the sequential extractions, material crushed to below 200 mesh was used while material below 3 mesh was used for the humidity cell tests. The shake flask leach removes as much as 15 times more Se than the humidity cells do because of the relatively large amount of exposed surface area which increases reactivity. Table 5.11 Percent of Water Soluble Se in the Sequential Extraction Compared with Percent of Total Se Represented by Peak Extraction from the Humidity Cells Water Soluble Se in Sequential Extraction (%) Peak Se Extraction from Humidity Cells (%) A99-23 6.0 2.2 B99-44 4.7 2.6 CREF 4.8 1.7 E99-19 2.4 1.7 E99-61 8.9 0.6 113 Lastly, the relatively short duration of humidity cell tests was of concern since waste rock at the mine is continuously subject to oxidative weathering. It was particularly problematic in this instance because Se release rates showed no sign of slowing by the end of the 20-week period. However, due to both time and budget limitations it was not possible to extend the duration of the tests. In summary, the humidity cell tests show that: 1. Se is being released from all materials, with the parting (A99-23) exhibiting the highest weekly release rate at 21 pg/week and the interburden (E99-61) the lowest at 4 pg/week. 2. Se release, as a function of total Se content of the head samples, is within the same order of magnitude for all five lithologies tested. 3. Sulphide oxidation appears to be the main geochemical mechanism driving Se release. 4. Se release is poorly correlated with the neutralization potential ratios of the five lithologies tested. Organic material may decrease the rate of sulphide oxidation thereby reducing Se mobilization. Further tests will be required to evaluate the role of TOC in regulating Se release and to determine to what extent the rate of Se release is affected by other mineralogical factors, such as grain size, degree of liberation, type of sulphide, and trace element content. 114 6.0 CONCLUSIONS This research project was undertaken to determine the source or sources of Se being mobilized by surface- or groundwater from the five open-pit coal-mining operations in the Elk River Valley, as well as the rate and geochemistry of its release. The following conclusions were made regarding mineralogical associations of Se in the Mist Mountain Formation: • Se is associated with both organic and inorganic components in the 16 samples representing the main lithologies in the coal-bearing stratigraphic section. • Water soluble Se accounts for approximately 10% of total Se, though values range from 3% to 21%). Less than 5% of the Se is found in association with HFMO. Se incorporated into the silicate structure accounts for, on average, 15% of total Se and between 60% and 84% the Se is found in association with sulphides and organic material. • The oxide analyses reveal no correlation between Se and clay mineral concentrations. Se is, however, the medium density fraction contained between 0.8 and 8.1 mg/kg Se, and the <2 pm fraction contained, on average, 42% more Se. These findings suggest that clay minerals contain Se but do not make a consistent contribution to the amount of total Se. • The amount of Se associated with the organics and sulphides varies, reflecting a high degree of heterogeneity in the depositional environments. • The ratio of sulphides to TOC is well correlated with the amount of Se in materials closely associated with coal seams containing less than 6 mg/kg Se. The amount of organic matter present may affect the incorporation of Se into sulphides by modifying redox conditions or it may compete with sulphides for available Se. 115 In seeking to pinpoint potential sources of Se to the tributaries of the Elk River, the magnitude and rate of Se release from different lithologies were investigated. Listed below are the most important findings with regard to this topic: • Se was released, in order of decreasing magnitude, from parting, coal, refuse, foot wall interburden. • Al l samples, other than the interburden, exhibited a peak in Se release in the first three leaches likely resulting from the removal of oxidation products. Peak release accounted for between 20 % and 76 % of total Se leached. It correlates poorly with the amount of Se in the water soluble phase of the sequential extractions highlighting difficulties of scaling-up results even within a controlled laboratory environment. • After the first three weeks, weekly Se release rates stabilized between 3 and 21 pg/L. • The amount of Se leached from each of the different materials on a weekly basis does not correlate well with total Se content. The parting material, containing 5.9 mg/kg Se, released most Se, 21 pg/kg/week. The coal, which contained 8.8 mg/kg Se, released 9 pg/kg/week and the foot wall, with 8.4 mg/kg Se, released only 3 pg/kg/week. The interburden, which contained 3.0 mg/kg Se, released 4 pg/kg/week while the refuse, which contained 3.4 mg/kg Se, released 7 pg/kg/week. • The rate of Se release from these materials was also poorly correlated with sulphide content. The foot wall sample contained most sulphides, but released least Se suggesting that mineralogical factors play an important role in determining the magnitude of Se release. • Se release from the humidity cells with parting material fluctuated more than Se release from the other cells. A larger number of parting samples would have to be analyzed to 116 determine if this is a function of the form in which Se is present in this particular sample or Se is as readily released from other parting samples with similar mineralogical characteristics. • The rate of Se release from the humidity cells did not level off at the end of the 20-week test period. It is unlikely that Se release would continue at the same rate indefinitely since a portion of the Se in all of the materials is found in unliberated sulphides, within the silicate structure, and in slow-weathering organics. Humidity cell tests would have to be run over a much longer period to confirm this. • The percent of total Se leached in the 20-week period, varied from 2.2 % for the foot wall to 8.3 % for the parting. The main conclusions regarding possible biogeochemical mechanisms of Se mobilization were: • No correlation was found between Se and TOC or concentrations of the other 30 elements measured in the humidity cell leachate indicating that the Se being removed did not stem from the organic matrix or a mineral association with any one of those elements. • The only statistically significant correlation was observed between Se and sulphate concentrations in leachate from the humidity cells containing interburden, foot wall, coal and refuse. This strongly suggests that sulphides are the primary source of Se being mobilized by runoff from the mine sites. • The correlation between peak Se release from the humidity cells and sulphide content provides further evidence of sulphide oxidation as a source of Se. Selenides have slower oxidation kinetics than sulphides, but when present, they would likely make a contribution to Se being leached from the samples. However, no selenides were identified in this study. 117 • Given that the rate of Se release was poorly correlated with sulphide content, mineralogical factors, such as texture, porosity and sulphide liberation, clearly play an important role in governing the access of air and water to the Se in the material and hence the rate of Se mobilization. Since the 1970s, an estimated 2.5 billion tonnes of rock have been extracted in the Elk River Valley. Exposing this amount of rock to air and water, creates a significant potential for continued Se mobilization though the coals and associated materials, on average, contain only 0.57% total S. Further study will be required to determine the magnitude of Se release under field conditions. In order to assess the risk presented by Se release from waste rock and plant refuse, it is necessary to evaluate information on both Se geochemistry and bioavailability. If the high Se concentrations in the Elk River and its tributaries prove to have an adverse effect on biota, information on Se occurrence and mobilization would play an important role in adopting an effective remediation strategy. Given that Se is so widely distributed in the Mist Mountain Formation and that it is being mobilized from all types of rock, it would not be practical to try to isolate material from which Se is leaching. Instead, abatement measures would have to focus on diverting water away from the spoil piles and pits or on treating runoff. 118 7.0 RECOMMENDATIONS This project addresses some of the fundamental questions about the modes of occurrence of Se in coal bearing geological units and its mobilization. To improve the understanding of the geochemistry of Se release from lithologies in the Mist Mountain Formation and to better evaluate the risks, the following work is recommended: • Conduct longer term humidity cell tests or set up field plots to establish if Se rates eventually level off or if Se release continues until most or all of it has been released. • Perform a Se mass balance to establish if some spoils are releasing more Se than others. This would involve estimating the amount of Se in waste rock based on results from Ryan and Dittrick (2000) and determining if there is any correlation with the amount of Se in runoff, calculated using data from Se monitoring stations on the rivers and streams running from the mines to the Elk River and flow rates in these waterways. • Study the form in which sulphides occur in the different materials at the sites to determine how it affects reactivity. • Obtain more infonnation on the effect of Se speciation and complexation with dissolved organic matter and sediment on its bioavailability and the process of bioaccumulation. • Test the reliability of the correlation between Se content and the sulphide/TOC ratio to determine whether it could be used to rapidly estimate Se concentrations in lithologies of similar coal-bearing depositional environments. • Investigate the importance of selenides in the Mist Mountain Formation using microprobe analysis. 119 Study Se abatement methods and their applicability. These include: 1. Chemical or wetland-based remediation systems designed to precipitate Se oxyanions with sulphides (Lin et al., 2000; Labrenz et al., 2000) 2. The use of ferrous iron amendments to immobilize Se with hydrous ferric oxides (Manning and Burau, 1995; Qui et al., 2000) 3. The addition of gypsum to precipitate calcium selenate dihydrate (Brienne et al., 2000) 4. Phytoremediation through volatilization (Lin et al., 2000) or harvesting (Banuelos et al., 1998). 8.0 A B B R E V I A T I O N S AP Acid-Producing Potential APHA American Public Health Association ASTM American Society for Testing and Materials BEF Biospheric Enrichment Factor CCME Canadian Council for Metals in the Environment CDB Citrate Dithionate Buffer EDS Energy-Dispersive Spectrometer EDTA Ethylenediamine-Tetraacetic Acid HG-AAS Hydride Generation Atomic Absorption Spectroscopy HFMO Hydrous Ferric and Manganese Oxides ICP-MS Inductively Coupled Plasma Mass Spectroscopy ICP-OES Inductively Coupled Plasma Optical Emission Spectroscopy INAA Instrumental Neutron Activation Analysis NP Neutralization Potential PTXE Proton-Induced Gamma-ray/X-ray Emission PZC Point of Zero Charge SEM Scanning Electron Microscopy TIC Total Inorganic Carbon TOC Total Organic Carbon USEP A/EPA Unites States Environmental Protection Agency XRD X-Ray Diffraction 121 9.0 REFERENCES Adams, W.J., Toll, J.E., Brix, K.V., Tear, L .M. and DeForest, D.K., 2000. Site-specific approach for setting water qual;ity criteria for selenium: differences between lotic and lentic systems. In Proceedings of the 24th Annual British Columbia Mine Reclamation Symposium, The British Columbia Technical and Research Committee on Reclamation, pp. 231-240. Altschuler, Z.S., Schnepfe, M.M., Silber, C C . and Simon, F.O., 1983. Sulfur diagenesis in Everglades peat and the origin of pyrite in coal. Science 221: 221-227. APHA, 1998. Standard Methods for the Examination of Water and Wastewater 20th Edition, Clesceri, L.S., Greenberg, A.E. and Eaton, A.D., (eds.). American Public Health Association, Philadelphia, Pennsylvania, 1220 p. ASTM, 2001. Annual book of ASTM standards. American Society for Testing and Materials, West Conshohocken, Pennsylvania, Vol. 05.06. Badalov, S.T., Belopolskaya, T.J., Prikhidok, P.L., Turebsekov, A., 1970. Geochemistry of selenium in sulphate-sulphide mineral paragenesis. Geochemistry International, 7: 799. Ballistrieri, L.S. and Chao, T.T., 1987. Selenium adsorption by goethite. Soil Science Society of America Journal, 51: 1145-1151. Banuelos, G.S., Aiwa, H.A., Wu, L. and Zambrzuski, S., 1998. Selenium accumulation by Brassica napus grown in Se-laden soil from different depths of Kesterson Reservoir. Journal of Soil Contamination, 7: 4: 481-496. Bar-Yosef, B. and Meek, D., 1987. Selenium sorption by kaolinite and montmorillonite. Soil Science, 144: 11-19. Bogle, E.W. and Nichol, I., 1981. Metal transfer, partition and fixation in drainage waters and sediments in carbonate terrain in southeastern Ontario. Journal of Geochemical Exploration, 15: 405-422. Bowie, G.L. and Grieb, T.M., 1991. A model framework for assessing the effects of selenium in aquatic ecosystems. Water Air and Soil Pollution, 57-58: 13-22. Brienne, S.H., Falutsu, M. , Weakley, J.O. and Kulas, J.E., Kuit, W.J., Geist, D.J., Gustafson, J.A., Wood, S.A., Baker, L.L., Rosenzweig, R.F., Ramamoorfhy, S., Crawford, D.L., Prisbrey, K.A., Moller, G. and George, J., 2000. In: Planning for end land uses in mine reclamation, Proceedings of the 24th Annual British Columbia Mine Reclamation Symposium, The British Columbia Technical and Research Committee on Reclamation, pp. 194-203. Brix, K.V., DeForest, D.K., Fairbrother, A. and Adams, W.J., 2000. Critical review of tissue-based selenium toxicity thresholds for fish and birds. In: Planning for end land uses in mine reclamation, Proceedings of the 24th Annual British Columbia Mine Reclamation Symposium, The British Columbia Technical and Research Committee on Reclamation, pp.220-230. 122 Brooke, L.T., Call, D.J., Harting, S.L., Lindberg, C.A., Markee, T.P., McCauley, D.J. and Poirier, S.H., 1985. Acute toxicity of selenium (IV) and selenium (VI) to freshwater organisms. Center for Lake Superior Environmental Studies, University of Wisconsin-Superior (Cited from USEPA, 1987). Brown, C.J., Schoonen, M.A.A. and Candela, J.L., 2000. Geochemical modeling of iron, sulfur, oxygen and carbon in a coastal plain aquifer. Journal of Hydrology, 237: 3-4: 147-168. Brown, T.H., 1990. Solubility, sorption, and redox relationships for selenium in reclaimed environments-a review. In Proceedings of the 1990 Billings Land Reclamation Symposium on selenium in arid and semiarid environments, Western United States, Severson, R .C, Fisher Jr., S.E. and Gough, L.P. (eds.),U.S. Geological Survey Circular No. 1064. pp. 27-33. Brueechert, V. and Pratt, L.M. , 1996. Contemporaneous early diagenetic formation of organic and inorganic sulfur in estuarine sediments from St. Andrew Bay, Florida, USA. Geochimica et Cosmichimica Acta, 60: 13: 2325-2332. Cardoso Fonseca, E. and Martin, H., 1986. The sequential extraction of Pb and Zn in selected mineral and soil sample, application in geochemical exploration (Portugal). Journal of Geochemical Exploration, 26: 231-248. Casagrande, D.J. and Nug, L., 1979. Incorporation of elemental sulphur in coal as organic sulfur. Nature, 282: 598-599. Casagrande, D. and Siefert, K., 1977. Origin of sulfur in coal: importance of the ester sulfate content of coal. Science 195: 675-676. Casagrande, D., Siefert, K., Berschinski, C. and Sutton, N., 1977. Sulfur in peat-forming systems of the Okefenokee Swamp and Florida Everglades: origins of sulfur in coal. Geochimica et Cosmichimica Acta 41: 161-167. Casey, R. and Siwik, P., 2000. Overview of selenium in surface waters, sediment and biota in river basins of west-central Alberta. In: Planning for end land uses in mine reclamation, Proceedings of the 24th Annual British Columbia Mine Reclamation Symposium, The British Columbia Technical and Research Committee on Reclamation, pp. 184-193. Chang, S. and Berner, R.A., 1999. Coal weathering and the geochemical carbon cycle. Geochimica et Cosmichimica Acta, 63: 19-20: 3301-3310. Chao, T.T. and Sanzolone, R.F., 1977. Chemical dissolution of sulphide minerals. Journal Reserves of the U.S. Geological Survey, 5: 409-412. Chao, T.T. and Sanzolone, R.F., 1989. Fractionation of soil selenium by sequential partial dissolution. Soil Science of America Journal, 31: 21-26. Chao, T.T. and Theobald, P.K., 1976. Chemical dissolution of sulfide minerals, Journal of Resources of the U.S. Geological Survey, 5: 409-412. Chapman, L., 1913. A contribution to the theory of electrocapillarity. Philosophical Magazine, 6: 25: 148:475-481. 123 Chen, Y., Belzile, N . and Goldsack, D.E., 1999. Passivation of pyrite oxidation by organic compounds. In Mining and the Environment n, Sudbury '99 Conference Proceedings, Vol. 3, Goldsack, D.E., Belzile, N. , Yearwood, P. and Hall, G J . (eds.), pp. 1063-1071. Chou, C.L., 1990. Formation of pyrite and organic sulfur compounds in coal; a review. Geological Society of America, 22; 7: 202. CCME, 1999. Canadian environmental quality guidelines. Canadian Council of Ministers of the Environment, Winnipeg, Manitoba. Clark, Jr., D.R., Ogasawara, P.A., Smith, G.J., and Ohlendorf, H.M., 1986. Selenium accumulation by raccoons exposed to irrigation drainwater at Kesterson National Wildlife Refuge, California. Archives of Environmental Contamination and Toxicology, 19: 374. Clarke, L.B. and Sloss, L.S., 1992. Trace element emissions from coal combustion and gasification. IEA Coal Research, London, England, IEACR/49, 109 p. Coleman, R.G., 1959. The natural occurrence of galena-clausthalite solid solution series. American Mineralogist 44: 166. Coleman, L., Bragg, L.J and Finkelman, R.B., 1993. Distribution and mode of occurrence of selenium in US coals. Environmental Geochemistry and Health, 15: 4: 215-226. Combs, Jr., G.F. and Combs, S.B., 1986. The role of selenium in nutrition. Academic, Orlando, Florida. 532 p. Cutter, G.A., 1989. Freshwater systems. In Occurrence and distribution of selenium, Dinat, M . (ed.), CRC Press, Boca Raton, Florida, pp.243-262. Cutter, G.A. and L.S. Cutter. 1995. Behavior of dissolved antimony, arsenic, and selenium in the Atlantic Ocean. Marine Chemistry, 49: 295-306. Dale, L., 1996. Trace elements in coal. IEA Coal Research Perspectives. 61 p. Dean, R.B. and Dixon, W.J., 1951. Simplified statistics for small numbers of observations. Analytical Chemistry, 23: 4: 636-638. Demir, I. and Harvey, R., 1991. Variation of organic sulphur in macerals of selected Illinois Basin Coals. Organic Geochemistry, Pergamon Press, Exiter, pp.525-533. Dold, B., 1999. Mineralogical and geochemical cha nges of copper flotation tailings in relation to their original composition and climate setting- Implications for AMD and element mobility. Terre et Environnement, Vol. 18, 230 p. Doran, J.W., 1982. Microorganisms and the biological cycling of selenium. Advances in Microbial Ecology, 6: 1-32. Dreher, G.B. and Finkelman, R.B., 1992. Selenium mobilization in a surface coal mine, Powder River Basin, Wyoming, U.S.A., Environmental Geology and Water Science, 19:3:155-167. 124 Drever, J.L., 1982. The Geochemistry of natural waters. Prentice-Hall Inc, Upper Saddle River, New Jersey, 388 p. D'Ulivo, A., 1997. Determination of selenium and tellurium in environmental samples. Analyst, 22: 117R-144R. Elrashidi, M.A., Adriano, S.M., Workman, S.M. and Lindsay, W.L., 1987. Chemical equilibria of selenium in soils: A theoretical development. Soil Science, 144: 141-152. Emerick, J.C. and DeMarco, L.S., 1990. Geobotany of selenium. In Proceedings of the 1990 Billing Land Reclamation Symposium on Selenium in Arid and Semiarid Environments, Western United States, Severson, R .C, Fisher Jr., S.E. and Gough, L.P. (eds.), U.S. Geological Survey Circular 1064, pp. 37-41. Filipek, L.H. and Theobald, P.K., 1981. Sequential extraction techniques applied to a porphyry copper deposit in the basin and range province. Journal of Exploration Geochemistry, 14: 155-174. Finkelman, R.B., Palmer, C.A., Krasnow, M.R., Aruscavage, P.J., Sellers, G.A. and Dulong, F.T., 1990. Combustion and leaching behaviour of elements in the Argonne Premium Coal Samples. Energy and Fuels, 4: 6: 755-766. Finkelman, R.B., 1981. Recognition of authigenic and detrital minerals in coal. Geological Society of America Abstract Program 13: 7: 451. Finkelman, R.B., Palmer, C.A., Krasnow, M.R., Aruscavage, P.J., Sellers, G.A. and Dulong, F.T., 1990. Combustion and leaching behaviour of elements in the Argonne Premium Coal Samples. Energy and Fuels 4: 6: 755-766. Fleischer, M. , 1959. Minor elements in some sulphide minerals. Economic Geology 50: 970. Fleming, G.A., 1962. Selenium in Irish soils and plants. Soil Science 94:28-35. Franke, K.W., and Painter, E.P., 1936. Selenium in proteins from toxic foodstuffs. 1. Remarks on the occurrence and nature of the selenium present in a number of foodstuffs or their derived products. Cereal Chemistry, 13:67. Frankie, K.A. and Hower, J., 1985. Pyrite/Markasite size, form, and microlithotype association in Western Kentucky prepared coals. Fuel Processing technology, 10: 269-283. Frost, R.R. and Griffin, R.A., 1977. Effect of pH on adsorption of arsenic and selenium from landfill leachate by clay minerals. Soil Science Society of America Journal 41: 53-57. Frostad, S, Klein, B. and Lawrence, R.W., 2000. Kinetic testing 2. Scaling up laboratory data to predict field rates of weathering. Proceedings from the 5 t h International Conference on Acid Rock Drainage, vol. 2, pp. 651-660. Forstner, U. and Wittmann, G.T.W., 1979. Metal pollution in the aquatic environment. Springer-Verlag, Berlin, Germany, 486 p. 125 Galbreath, K.C. and Brekke, D.W., 1994. Feasibility of combined wave-length/energy-dispersive computer-controlled scanning electron microscopy for determining trace metal distribution. Fuel Processing Technology 39: 1-3:63-72. Ge, K. and Yang, G., 1993. The epidemiology of selenium deficiency in the etiological study of endemic diseases in China. American Journal of Clinical Nutrition, 57: 259S-263S. Geering, H.R., Cary, E.E., Jones, L.H.P. and Allaway, W.H., 1968. Solubility and redox criteria for the possible forms of selenium in soils. Soil Science Society of America Proceedings, 32: 35-40. Gibson, D.W., 1977, Sedimentary facies in the Jura-Cretaceous Kootenay Formation, Crowsnest Pass area, southwestern Alberta and southeastern British Columbia, Bulletin of Canadian Petrographic Geology, 25: 4: 767-791. Gibson, D.W., 1979, The Morrissey and Mist Mountain Formations- newly defined lithostratigraphic units of the Jura-Cretaceous Kootenay Group, Alberta and British Columbia, Bulletin of Canadian Petrographic Geology, 27: 2: 183- 208. Gibson, D.W. and Hughes, J:D., 1981, Structure, stratigraphy, sedimentary environments and coal deposits of the Jura-Cretaceous Kootenay-Group, Crowsnest Pass area, Alberta and British Columbia, Field Guides to Geology and Mineral Deposits, Geological Association of Canada Annual Meeting, Calgary, 39 p. Gillespie, R.B., and Baumann, P.C, 1986. Effects of high tissue concentrations of selenium on reproduction by bluegills. Transactions of the American Fisheries Society, 115: 208-213. Girling, C.A., 1984. Selenium in agriculture and the environment. Agricultural Ecosystems and Environments, 11:27. Given, P.H. and Miller, R.N., 1985. Distribution of forms of sulfur in peats from saline environments in the Florida Everglades. International Journal of Coal Geology, 5: 4: 397-409. Goldhaber, M.B, 1983. Experimental study of metastable sulfur oyanion formation during pyrite oxidation at pH 6-9 and 30°C American Journal of Soil Science, 283: 193-217. Goodarzi, F., 1987. Elemental concentrations in Canadian coals 2. Byron Creek Collieries, British Columbia, Fuel, 66: 250-254. Goodarzi, F., 1988. Elemental distribution in coal seams at the Fording Coal Mine, British Columbia. Chemical Geology, 68: 129-154. Goodarzi, F. and Swaine, D.J., 1993. Chacophile elements in western Canadian coals. International Journal of Coal Geology, 24: 1-4: 281-292. Gouy, G., 1910. Sur la constitutionde la charge electrique a la surface d'un electrolyte. Journal de PhysiqueTheorique et Appliquee, 4: 9: 457-468. Green, D.E. and Albers, P.H., 1997. Diagnostic criteria for selenium toxicosis in aquatic birds: Histologic lesions. Journal of Wildlife Diseases, 33: 3: 385-404. 126 Grieve, D.A., 1985. Rank variation, coalification pattern and coal quality in the Crowsnest Coalfield, British Columbia, Candian Institute of Mining and Metallurgy, Bulletin, 78: 881: 39-46. Grieve, D.A. and Goodarzi, F., 1994, Trace elements in coals of the east Kootenay and Peace River coalfields, B.C. Ministry of Energy, Mines and Petroleum Resources, Victoria, British Columbia, 29 p. Gruebel, K.A., Davis, J.A. and Leckie, J.O., 1988. The feasibility of using sequential extraction techniques for arsenic and selenium in soils and sediments, Soil Science Society of America Journal, 52: 390-397. Hall, G.E.M., Vaive, J.E., Beer, R. and Hoashi, M. , 1996. Selective leaches with emphasis on the amorphous Fe oxyhydroxide phase extraction, Journal of Geochemical Exploration, 56: 59-78. Hamilton, S.J. and Buhl, K.J., 1990. Acute toxicity of boron, molybdenum, and selenium to fry of Chinook salmon and coho salmon. Archives of Environmental Contamination and Toxicology. 19: 366-375. Hayes, K.F., Roe, A.L., Brown Jr., G.E., Hodgson, K.O., Leckie, J.O. and Parks, G.A., 1987. In-situ X-ray absorption study of surface complexes- Selenium oxyanions on y-FeOOH. Science, 238:738. Haygarth, P.M., Harrisson, A.F. and Jones, K . C , 1993. Geographical and seasonal variation in deposition of selenium to vegetation. Environmental Science and Technology 27: 13: 2878-2884. Health and Welfare Canada, 1993. Guidelines for Canadian Drinking Water Quality, 5 t h Edition. Minister of Supply and Services Canada, Cat. No. H48-10/1993E. Hermanutz, R.O., 1992. Malformation of the fathead minnow (Pimephales promelas) in an ecosystem with elevated selenium concentrations. Bulletin of Environmental Contamination and Toxicolology, 49: 290-294. Herring, J.R., 1990. Selenium geochemistry- a conspectus. In Proceedings of the 1990 Billing Land Reclamation Symposium on Selenium in Arid and Semiarid Environments, Western United States, Severson, R.C., Fisher Jr., S.E. and Gough, L.P. (eds.), U.S. Geological Survey Circular 1064, pp. 7-24. Hickmott, D.D. and Baldridge, W.S., 1995. Application of PLXE micro-analysis to macerals and sulfides from the Lower Kittaning coal of western Pensylvania. Economic Geology, 90: 2: 246-254. Hoffman, S.J. and Fletcher, W.K., 1979. Selective sequential extraction of Cu, Zn, Fe, Mn and Mo from soils and sedimens. In Geochemical Exploration, Watterson, J.R. and Theobald, P.K. (eds.), Association of Exploration Geochemists, pp.289-299. Holuszko, M.E., Grieve, D.A. and Matheson, A., 1992. Sulphur in British Columbia coals and its implication for cleaning. International Journal of Environmental Issues in Minerals and Energy Industry, 2: 83-93. 127 Jonnalagada, S.B. and Prasada Rao, P.V.V., 1993. Toxicity, bioavailability and metal speciation. Comparative Biochemistry and Physiology, 106: 3: 185-195. Kabata-Pendias, A. and Pendias, H. 1984. Trace Elements in Soils and Plants. CRC Press, Boca Raton, Florida, 315 p. Kargi, F., 1987. Biological oxidation of thianthrene, thioxanthene and dibenzothiophene by the thermophilic organism Sulfolobus acidocaldarius. Biotechnology Letters, 9: 7: 478-482. Katrinak, K.A. and Benson, S.A., 1995. Trace metal content of coal and ash as determined using wavelength-dispersive spectrometry. American Chemical Society, Division of Fuel Chemistry, Preprints, 40: 4: 798-802. Kharkar, D.P., Turekian, K.K. and Bertine, K.K., 1968. Stream supply of dissolved silver, molybdenum, antimony, selenium, chromium, cobalt, rubidium and cesium to the ocean. Geochimica et Cosmichimica Acta, 32: 285-293. Kudriavtsev, A.A., 1974. The chemistry and technology of selenium and tellurium. Collet's Ltd., London, England, 278 p. Kwong, Y.T.J., 1993. Prediction and prevention of acid rock drainage from a geological and mineralogical perspective. MEND Report 1.32.1, CANMET, Natural Resource Canada, Ottawa, Canada, 47 p. Labrenz, M. , Druschel, G.K., Thomsen-Ebert, T., Gilbert, B., Welch, S.A., Kemmer, K.M. , Logan, G.A., Summons, R.E., De Stasio, G., Bond, P.L., Lai, B., Kelly, S.D. and Banfield, J.F., 2000. Formation of sphalerite (ZnS) deposits in natural biofilms of sulfate-reducing bacteria. Science, 290: 5497: 1744-1747. Lakin, H.W. 1973. Selenium in our environment. Advances in Chemistry Series, 123: 96-111. Lapakko, K.A., 1994. Comparison of Duluth complex rock dissolution in the laboratory and in the field. In Proceedings of the International lan Reclamation and Mine Drainage Conference and the Third Jnternationl Conference on the Abatement of Acidic Drainage, Pittsburgh, Pennsylvania, vol. 1, pp. 419-428. Lawrence, R.W., 1990. Prediction of the behaviour of mining and processing wastes in the environment. Proceedings of the Western Regional Symposium on Mining and Mineral Processing Wastes, Berkeley, California, Society for Mineral Engineering, Littleton, Colorado, pp. 115-121. Lawrence, R.W. and Day, S., 1997. Chemical prediction techniques for ARD. 4 t h International Conference on Acid Rock Drainage, May 31st, 1997. 77 p. Lawrence, R.W. and Scheske, M. , 1997. A method to calculate the neutralization potential of mining wastes. Environmental Geology, 32: 2: 100-106. Lemly, A.D., 1985. Toxicology of selenium in a freshwater reservoir: Implications for environmental hazard evaluation and safety. Ecotoxicology and Environmental Safety, 10: 314-338. 128 Lemly, A.D., 1993. Guidelines for evaluating selenium data from aquatic monitoring and assessment studies. Environmental Monitoring and Assessment. 28: 83-100. Lemly, A.D., 1997. A teratogenic deformity index for evaluating impacts of selenium on fish populations. Ecotoxicology and Enviornmental Safety, 37: 259-266. Lemly, A.D., 1999. Selenium transport and bioaccumulation in aquatic systems: a proposal for water quality criteria based on hydrological units. Ecotoxicology and Environmental Safety, 42: 150-156. Lemly A.D. and Smith, G., 1990. Aquatic cycling of selenium: implications for fish and wildlife. In Proceedings of the 1990 Billings Land Reclamation Symposium on selenium in arid and semiarid environments, Western United States, Severson, R.C, Fisher Jr., S.E. and Gough, L.P. (eds.),U.S. Geological Survey Circular No. 1064. pp. 45-49. Levander, O.A., 1985. Considerations on the assessment of selenium status. Proceedings of the American Society of Experimental Biology, 44: 9: 2579-2583. Light, T.S., 1972. Standard solution for redox potential measurements. Analytical Chemistry, 44: 6: 1038-1039. Lin, Z-Q, Schemenauer, R.S., Cervinka, V., Zayed, A., Lee, A. and Terry, N. , 2000. Selenium volatilization from a soil-plant system for the remediation of contaminated water and soil in the San Joaquin Valley. Journal of Environmental Quality, 29: 4: 1048-1056. Macy, J.M., Lawson, S., DeMoll, H. and Rech, S., 1992. 1991-92 Technical Progress Report University of California Task Force, UC Davis, California, 66 p. Manning, B.A. and Burau, R.G., 1995. Selenium immobilization in evaporation pond sediments by in situ precipitation of ferric oxyhydroxide. Environmental Science and Technology, 29: 10: 2639-2646. Martens, D.A. and.Suarez, D.L., 1997. Selenium speciation of soil sediment determined with sequential extractions and hydride generation atomic absorption spectrophotometry, 31: 1: 133-139. Maskimovic, Z., Jovic, V., Djujic, S. and Rsumovic, M. , 1992. Selenium deficiency in Yugoslavia and possible effects on health. Environmental Geochemistry and Health, 14: 107-111. Masscheleyn, P.H., Delaune, R.D., Patrick, W.H., 1990. Transformations of selenium as affected by sediment oxidation-reduction potential and pH. Environmental Science and Technology 24: 1: 91-96. Mayland, H.F., James, L.F., Panter, K.E. and Sonderegger, J.L., 1989. Selenium in seleniferous environments, In Selenium in agriculture and the environment, Jacobs, L.W. (ed.), Soil Science Society of America Special Publication No. 23, pp. 15-50. McBride, M.B., 1994. Environmental Chemistry of Soils. Oxford University Press, Oxford, 406p. McCurdy, Lange, J.D. and Haygarth, P.M., 1993. The determination of selenium in sediment using hydride generation ICP-MS. Science of the Total Environment, 135: 1-3: 131-136. 129 McDonald, L.E. and Strosher, M.S., 1998. Selenium mobilization from surface coal mining in the Elk River Basin, British Columbia: A survey of water, sediment and biota. British Columbia Ministry of the Environment Lands and Parks, Pollution Prevention Branch, Cranbrook, British Columbia. Mclntyre, N.S., Martin, R.B., Chauvin, W.J., Winder, C.G., Brown, J.R. and MacPhee, J.A., 1985. Studies of elemental distributions within discrete coal macerals. Use of secondary ion mass spectrometry and X-ray photoelectron spectroscopy, Fuel: 64: 1705-1712. McNeal, M.J. and Balistrieri, L.S., 1989. Geochemistry and occurrence of selenium: an overview. In Selenium in agriculture and the environment, Jacobs, L.W. (ed.), Soil Science Society of America Special Publication No. 23, pp. 1-13. Menegario, A.A. and Gine, M.F., 2000. Rapid sequential determination of arsenic and selenium in waters and plant digests by hydride generation inductively coupled plasma-mass spectrometrySpectrochimica Acta , 55: 4: 355-362. Milchunas, D.G. and Lauenroth, W.K., 1984, Sulfur deposition, cycling and accumulation. In Effects of S02 on a Grassland: A Case Study in the Northern Great Plains of the United States, Lauenroth W.K. and Preston, E.M. (eds.), Ecological Studies 45. Springer-Verlag, New York. Miller, R.N. and Given, P.H., 1987. The association of major, minor and trace elements with lignites. II Minerals, and major and minor element profiles, in four seams. Greochimica et Cosmichimica Acta, 51: 1843-1853. Minkin, J.A., Finkelman, R.B., Thompson, C.L., Chao, E.C.T., Ruppert, L.F., Blank, H. and Cecil, C.B., 1984. Microcharacterization of arsenic- and selenium bearing pyrite in Upper Freeport coal, Indiana County, Pensylvania. Scanning Electron Microscopy, SEM Inc., AMF O'Hare, Chicago, pp. 1515-1524. Monterroso, C. and Marcias, F., 1998. Prediction of the acid generating potential of coal mining spoils. International Journal of Surface Mining, Reclamation and Environment, 12: 1: 5-9. Moore, J.N., Ficklin, W.H. and Johns, C , 1988. Partitioning of arsenic and metals in reducing sulfidic sediments. Environmental Science and Technology, 22: 432-437. Moreno-Reyes, R., Suetens, C , Matthieu, F., Begaux, F., Zhu, D., Rivera, T.Boelaert, M. , Neve, J., Perlmutter, N . and Vanderpas, J., 1998. Kashin-Beck osteoarthropathy in rural Tibet in relation to selenium and iodine status. New England Journal of Medicine, 339: 1112-1120. Mosher, B.W. and Duce, R.A., 1989. The atmosphere. In Occurrence and distribution of selenium, Ihnat, M . (ed.), CRC Press, Boca Raton, Florida, pp. 295-326. Munkers, J., Bond, M. , Brackney, K. and Moller, G., 2000. The biogeochemistry of selenium in the Western US Phosphate Resource Area: sources, pathways, receptors and controls. In: Planning for end land uses in mine reclamation, Proceedings of the 24 th Annual British Columbia Mine Reclamation Symposium, The British Columbia Technical and Research Committee on Reclamation, pp. 204-219. Naftz, D.L. and Rice, J.A., 1989. Geochemical processes controlling selenium in ground water after mining, Powder River Basin, Wyoming, U.S.A., Applied Geochemistry, 4: 565-575. 130 Nagpal, N.K., 2001. Ambient water quality guidelines for selenium: overview. Ministry of Water, Land and Air Protection, Water Protection Branch, Victoria, British Columbia. Nagpal, N.K., Pommen, L.W. and Swaine, L.G., 1995. Approved and working criteria for water quality- 1995. Ministry of Environment, Lands and Parks, Water Quality Branch, Victoria, British Columbia., 45 p. National Institute of Health, 2000. Selenium in nutrition. Report posted at: www.cc. nih.gov/ccc/supplements/selen. html#food viewed 09/07/01. Neal, R.H., 1995. Selenium. In Heavy metals: problems and solutions, Salomons, W., Forstner, U. and Mader, P. (eds.), Springer-Verlag, Berlin, 412 p. Neal, RH, and G. Sposito. 1989. Selenate adsorption on alluvial soils. Soil Science Society of America Journal, 53: 70-74. Nelson, D.C, Casey, W.H., Sison, J.D., Mack, E.E., Ahmad, A. and Pollack, J.S., 1996. Selenium uptake by sulfur-accumulating bacteria. Geochimica et Cosmichimica, 60: 18: 3531-3539. Nicholson, R.V., Gillham, R.W. and Reardon, E.J., 1990. Pyrite oxidation in carbonate buffered solution. 2. Rate control by oxide coatings. Geochimica Cosmichimica Acta, 54: 395-402. Niimi, A.J., and LaHam, Q.N., 1976. Relative toxicity of organic and inorganic compounds of selenium to newly hatched zebrafish (Brachydanio rerio). Canadian Journal of Zoology, 54: 501-509. Nriagu, J.O, 1990. Global Metal Pollution: poisoning the biosphere? Environment, 32: 7: 7-33. Nriagu, J.O. and Pacyna, J.M., 1988. Quantitative assessment of worldwide contamination of air, water, and soils by trace metals. Nature, 333:134-139. Oldfield, J.E., 1992. Risks and Benefits in Agricultural Uses of Selenium. Environmental Geochemistry and Health, 14: 3: 81. Ohlendorf, H.M., 1989. Bioaccumulation and the effects of selenium in wildlife. In Selenium in Agriculture and the Environment, Jacobs, L.W. (ed.), Soil Science Society of America Special Publication No.23, pp. 133-177. Ohlendorf, H.M., Hoffman, D.J., Saiki, MK. and Aldrich, T.W., 1986. Embryonic Mortality and Abnormalities of Aquatic Birds: Apparent Impacts of Selenium from Irrigation Drainwater. Science of the Total Environment, 52: 49. Paktunc, A.D., 1999. Characterization of mine wastes for the prediction of acid mine drainage. In Environmental impacts of mining, Azcue, J.M. (ed.), Springer, Berlin, pp.19-40. Palmer, C.A., Lyons, P.C, Brown, Z.A. and Mee, J.S., 1990. The use of rare earth and trace element concentrations in vitirinite concentrates and companion whole coals (hvA bituminous) to determine organic and inorganic associations. In Recent advances in coal geochemistry, Chyu, L.L., Chou, CL. (eds.), Geological Society of America Special Paper 248, pp. 55-62. 131 Pearson, D.E., 1980. The quality of Western Canadian coking coal, C1M Bulletin, 73: 813: 70-84. Pickering, S.A. and Batts, B.D., 1992. Persistence patterns of aqueous leachable carboxylic acids in immature to peak mature coals: implications for carboxylic acid patterns in hydrocarbon reservoir waters. Organic Geochemistry, 18: 695-700. Pillar, K.K.S., Thomas, CC. Jr. and Kaminski, J.W., 1969. Neutron activation analysis of the Se content of fossil fuels. Nuclear applications and Technology, 7: 5: 478-483. Porcella, D.B., Bowie, G.L., Sanders, J.G. and Cutter, G.A., 1991. Assessing Se Cycling and Toxicity in Aquatic Ecosystems. Water, Air and Soil Pollution, 57-58: 3-11. Price, W.A., 1997. Guidelines and recommended methods for the prediction of metal leaching and acid rock drainage at minesites in British Columbia. Ministry of Employment and Investment, Energy and Minerals Division, Smithers, British Columbia, 159 p. Price, F.T. and Shieh, 1979. The distribution and isotopic composition of sulfur in coals from the Illinois Basin. Economic Geology, 74: 1445-1461. Presser, T.S., 1994. Geologic origin and pathways of selenium from the California Coast Ranges to the west-central San Joaquin Valley. In, Selenium in the environment. Frankenberger, W.T. and Benson, S. (eds.). Marcel Dekker, New York, pp. 139-155. Presser, T.S., Sylvester, M.A. and Low, W.H., 1994. Bioaccumulation of selenium from natural geological sources in Western states and its potential consequences. Environmental Management, 18: 423-436. Presser, T.S. and Barnes, 1., 1984. Selenium Concentrations in Waters Tributary to and in the Vicinity of the Kesterson National Wildlife Refuge, Fresno and Merced Counties, California. U.S. Geological Survey Water Resources Investigation Report No. 84-4122, California, 26 p. Quevauviller, P., Lachica, M., Barahona, E., Rauret, G., Ure, A., Gomez, A. and Muntau, H., 1996. Interlaboratory comparison of EDTA and DTPA procedures prior to certification of extractable trace elements in soil, Science of the Total Environment, 178: 127-132. Quevauviller, P., Rauret, G., Muntau, H., Ure, A., Rubio, R., Lopez-Sanchez, J.F., Fielder, H.D. and Griepink, B., 1994. Evaluation of sequential extraction procedure for the determination of extractable trace metal contents in sediments. Fresnius Journal of Analytical Chemistry, 349: 808-814. Qui, S.R., Lai, H-F, Robertson, M.J., Hunt, M.L., Amrhein, C, Giancarlo, L.C., Flynn, G.W., and Yarmoff, J.A., 2000. Removal of contaminants from aqueous solution by reaction with iron surfaces. Langmuir, 16: 5: 2230-2236. Raptis, S.E., Kaiser, G. and Tolg, G., 1983. A survey of selenium in the environment and a critical review of its determination at trace levels. Fresnius Journal of Analytical Chemistry, 316: 105. Reed, S.J.B., 1995. Electron probe microanalysis. In Microprobe techniques in the earth sciences, Potts, P.J., Bowles, F.W., Reed, S.J.B. and Cave, M.R. (eds.), Chapman and Hall, London, 419 p. 132 Rosenfeld, 1. And Beath, O.A., 1964. Selenium: geobotany, biochemistry, toxicity and nutrition, Academic, New York. 411 p. Rotruck, J.T., Pope, A.L., Ganther, H.E., Swanson, A.B., Hafeman, D.G. and Howkstra, W.G., 1973. Selenium: biochemical role as a component of glutathione peroxidase. Science, 179: 588. Ryan, B. and Dittrick, M., 2000. Selenium in the Mist Mountain Formation of southeast British Columbia. British Columbia Ministry of Energy and Mines, Geological Fieldwork 2000, Paper 2001-1, pp.337-362. Ryan, B., 2001. Coal production statistics. Report posted at: www. em. gov. bc/Mining/MiningStats/3 lcoalsold.htm viewed 21/08/01 Sappington, K., 1998. Revision of freshwater aquatic life criteria for selenium: overview and status. Peer Consultation Workshop on Selenium Aquatic Toxicity and Bioaccumulation, May 27-28, 1998, Washington, D.C. Saxby, J.D., 1973. Diagenesis of metal-organic complexes in sediments: formation of metal sulphides from cystine complexes. Chemical Geology, 12: 241-248. Schafer, H.N.S., 1984. Determination of carboxyl groups in low-rank coals. Fuel, 63: 723-726. Seby, F., Potin-Gauthier, M., Giffaut, E., Borge, G. and Donard, O.F.X., 2001. A critical review of thermodynamic data for selenium species at 25°C. Chemical Geology, 171: 173-194. Skorupa, J.P., Morman, S.P. and Sefchick-Edwards, J.S., 1996. Guidelines for interpreting selenium exposure of biota associated with nonmarine aquatic habitats. U.S. Fish and Wildlife Services National Irrigation Water Quality Program, Sacramento Field Office, March 1996, 74 pp. Sharmasarkar, S. and Vance, G.F., 2001. Selenite-selenate sorption in surface coal mine environment. Advances in Environmental Research, (in press). Shepard, M., 1987. Toxic research and the real world. EPR1 Journal, September, pp. 17-21. Sillanpaa, M. and Jansson, H., 1992. Status of cadmium, lead, cobalt and selenium in soils and plants of thirty countries. FAO Soild Bulletin, Forssa, Finland, pp. 34-44. Smith, G.G., 1989, Coal resources of the Rocky Mountain front ranges and foothills, Geological Survey ofCanada, 89: 4: 50-68. Sobek, A.A., Schuller, W.A., Freeman, J.R. and Smith, R.M., 1978. Field and laboratory methods applicable to overburden and minesoils. EPA-600/2-78-054, National Technical Information Service. Springfield, Virginia. 495 p. Sorensen, E.M.B., Cumbie, P.M., Bauer, T.L, Bell, J.S. and Harlan, C.W, 1984. Histopathological, Hematological, Condition-Factor, and Organ Weight Changes Associated with Selenium Accumulation in Fish from Belews Lake, North Carolina. Archives of Environmental Contamination and Toxicology 13: 2: 153-162. 1 ~> 1 1J J Sparks, D.L, 1995. Environmental Soil Chemistry. Academic Press, San Diego, 191 p. Spears, A, Martinez Tarazona, M. and Lee, S, 1994. Pyrite in UK coals: environmental significance. Fuel, 73: 7. Spears, D.A. and Zheng, Y, 1999. Geochemistry and origin of elements in some UK coals. International Geology of Coal Geology, 38: 161-179. Stach, E, Mackowsky, M.Th, Teichmuller, M, Taylor, G.H., Chandra, D. and Teichmuller, R, 1982. Stach's textbook of coal petrology. Gebriider Borntraeger. Berlin, 535 p. Swaine, D.J, 1990. Trace elements in coal. Butterworth and Company, London, England. 278 p. Schwartz, K. and C. M. Foltz. 1957. Selenium as an integral part of Factor 3 against dietary necrotic liver degeneration. Journal of the American Chemical Society 79: 3292. Stern, O.Z, 1924. Zur Theorie der elektrolytischen Doppelschist. Z. Elektrokemi, 30: 508. Taylor, R.M, 1987. Non-silicate oxides and hydroxides. In Chemistry of clays and clay minerals, Newman, A.CD. (ed.), Longman Scientific and Technical, Essex, England, pp. 129-202. Taylor, S.R, 1964. Abundance of chemical elements in the continental crust. Goechimica Cosmichimica Acta, 28: 1273-1285. Tessier, A, Campbell, P.G.C. and Bisson, M, 1979. Sequential extraction procedure for the speciation of particulate trace metals, Analytical Chemistry, 51:7: 844- 850. Thompson-Eagle, E.T. and Frankenberger Jr., W.T, 1992. Bioremediation of soils contaminated with selenium. Advances in Soil Science, 17: 261-310. Tholacius-Ussing, O, Flyvbjerg, A. and Esmann, J, 1987. Evidence that selenium induces growth retardation through reduced growth hormone and somatomedin C production. Endocrinology, 120: 2: 659-663. Thornton, 1, 1995. Metals in the global environment: facts and misconceptions. International Council on Metals in the Environment, Ottawa, Canada, 103 p. Ting, B.T.G, Mooers, C S , Janhorbani, M, 1989. Isotopic determination of selenium in biological materials with inductively coupled plasma mass spectrometry. Analyst, 114: 6: 667-674. Ure, A.M., Quevauviller, P, Muntau, H. and Griepink, B, 1993. Speciation of heavy metals in soils and sediments. An account of the improvement and harmonization of extraction techniques undertaken under the auspices of the BCR of the Commission of the European Communities. International Journal of Environmental Analytical Chemistry, 51: 135-151. USEPA, 1979. Methyl orange alkalinity. Handbook for analytical quality control in water and wastewater, EPA-310.1. Laboratories, Environmental Monitoring and Support Laboratory. Office of Research and Development, Cincinnati, Ohio. 134 USEPA, 1986. Test methods for evaluating solid waste, Volume IA: 3rd Edition, EPA/SW-846. National Technical Information Service. Springfield, Virginia. USEPA, 1987. Ambient water quality criteria- 1987. PB88-142. U.S. Environmental Protection Agency, Washington, D.C. USEPA, 1994. Selenium (atomic absorption, borohydride reduction), EPA-7742. U.S. Environmental Protection Agency, Washington, D.C. USEPA, 1996. Inductively coupled plasma atomic emission spectroscopy, EPA Method 601 OB. U.S. Environmental Protection Agency, Washington, D.C. USEPA, 2001. Maximum contaminant levels. Criteria posted at: www.epa.gov/safewater/mcl.html viewed 09/07/01. U.S. National Committee for Geochemistry, 1980. Trace element geochemistry of coal resource development related to environmental quality and health. National Academy Press, Washington D.C, 132 p. Van Afferden, M., Tappe, D., Beyer, M., Truepre, H.G. and Klein, J., 1993. Biochemical mechanisms for the desulfurization of coal-relevant organic sulfur compounds, Fuel, 72: 12: 1635-1643. Van Derveer, W.D. and Canton, S.P., 1997. Selenium sediment toxicity thresholds and derivation of water quality criteria for freshwater biota of western streams, Environmental Toxicology and Chemistry, 16: 6: 1260-1268. Veiga, M.M., Schroscher, H.D. and Fyfe, W.S., 1991. Relationship of copper with hydrous ferric oxides: Salobo, Carajas, PA, Brazil. Ore Geology Reviews 6: 245-255. Verlinden, M., 1982. Acid decomposition of human blood and plasma for determination of selenium, Talanta, 29: 875. Vessey, S.J. and Bustin, R.M., 2000. Sedimentology of the coal-bearing Mist Mountain Formation, Line Creek, Southern Canadian Cordillera: relation to coal quality. International Journal of Coal Geology, 42: 129-158. Wandless, A.M., 1959. The occurrence of sulphur in British Coals. Journal of the Institute of Fuel 32: 258-266. Wedepohl, K.H., 1978. Handbook of Geochemistry. Vol. 2, Part 5, Springer-Verlag, Berlin. Welz, B., Melcher, M. and Neve, J., 1984. Determination of selenium in human body fluids by hydride generation atomic absorption spectroscopy. Optimization of sample decomposition. Journal of Analytical Chemica Acta 165: 131-140. White, R.N., Smith, J.V., Spears, D.A., Rivers, M.L. and Sutton, S.R., 1989. Analysis of iron sulphides from UK coal by synchrotron radiation X-ray fluorescence. Fuel: 68: 1480. 135 White, W.W. Ill and Sorini, S.S, 1997. Standard Test Method for Accelerated Weathering of Solid Materials Using a modified Humidity Cell. ASTM Standards, v. 11.04, Method 5744-96, pp.259-271. Wipperman, B. and Webber, T.N, 1997. State of water quality of Elk River at Highway 93 (Phillips Bridge) 1984-1995. Water Quality Section, Ministry of Environment, Lands and Parks, Victoria, B.C., 62 p. World Health Organization, 1986. Review of potentially harmful substances- arsenic, mercury and selenium, Reports and Studies No. 28, USEPA, Geneva. Wuyi, W, van Dorst, S. and Thornton, I, 1987. The selenium status of sheep in Britain as indicated by wool selenium concentration. Environmental Geochemistry and Health, 9: 48-51. Yang, G, Wang, S, Zhou, R. and Sun, S, 1983. Endemic selenium intoxication in humans in China. American Journal of Clinical Nutrition 37: 872-881. Yariv, S. and Cross, H, 1979. Geochemistry of colloid systems for earth scientists. Springer-Verlag: Berlin, 450 p. Zehr, J.P. and Oremland, R.S, 1987. Reduction of selenate to selenide by sulfate-respiring bac-teria; Experiments with cell suspensions and estuarine sediments. Applied Environmental Microbiology, 53: 1365-1369. Zheng, B, Ding, Z, Huang, R, Zhu, J, Yu, X, Wang, A, Zhou, D, Mau, D. and Su, H, 1999. Issues of health and disease relating to coal use in Southwestern China, International Journal of Coal Geology, 40: 119-132. Zodrow, E.L. and Goodarzi, F, 1993. Environmental implications associated with pyrite concentrates from coal in the Sydney coalfield (Upper Carboniferous), Nova Scotia, Canada, Energy Sources, 15: 4: 639-652. 136 137 138 140 142 144 146 10.2 APPENDIX 2 Calculation of the Percent of Total Sample Mass Accounted for by Major Mineral Components Table A2.1 Formulas Used to Calculate the Percent of Total Mass Accounted for by Major Mineral Components Mineral Chemical Formula* Molecular Mass (g/mol) Calcite CaCOj 100.1 Ankerite Ca1.o5(Fe0.48,Mgo.4s, Mno.04)(C03)2 202.0 Siderite FeC0 3 115.8 Kaolinite Al 2Si 20 5(OH)4 270.2 Illite K1.o.1.5Al4(Si6.5.7.oAl1.o.,.502o)(OH)4 448.3 Quartz Si0 2 60.1 * From Deer (1993) % K san]pie = _n • atomic weight K . % Illite molecular weight Illite % Al sample ' n- atomic weight Al . % Illitj; +f molecular weight Illite n • atomic weight A l . % Kaolinilje molecular weight Kaolinite r % S i sample ' r n • atomic weight Si . % Illitq + molecular weight illite v . n • atomic weight Si . % Kaolinite molecular weight Kaolinite + n • atomic weight Si . % Quartz molecular weight Quartz % Ca sampie = n • atomic weight Ca . % Calcite molecular weight Calcite Although XRD detected the presence of calcite, ankerite and siderite in only 4 of the 16 samples it is likely that most of the samples contain at least a small fraction of these carbonate minerals. Due the fact that the Fe in pyrite and HFMO could not be distinguished from the Fe in siderite or ankerite it was not possible to calculate the content of carbonates using the same approach as the one used to determine the amount of quartz, illite and kaolinite present in the sample. In order to obtain a rough quantitative estimate of the contribution of carbonates to the samples, it was assumed that all of the carbonates present were in the form of calcite. 147 10.3 APPENDIX 3 Raw Data from the Sequential Extractions Table A3.1 Se Concentrations in Leachate and Solid Residue Collected from the Sequential Extractions Se Extracted (ug/L) Se Remaining (mg/kg) Water Soluble HFMO Sulphides and Organics Residual A99-4A 6.0 3.0 49.0 0.8 A99-4B 5.9 <3.0 54.0 0.8 A99-7A 2.1 <3.0 68.0 0.7 A99-7B 2.3 <3.0 68.0 0.8 A99-12A 7.1 4.0 98.0 0.7 A99-12B 7.4 4.0 101.0 0.8 A99-23A 6.4 <3.0 89.0 0.4 A99-23B 6.1 4.0 86.0 0.4 B99-16A 17.0 3.0 55.0 0.3 B99-16B 18.0 4.0 43.0 1.0 B99-44A 11.5 <3.0 180.0 1.4 B99-44B 12.0 <3.0 220.0 2.2 B99-62A 3.2 <3.0 27.0 0.6 B99-62B 2.8 <3.0 27.0 0.5 CREFA 3.2 <3.0 59.0 0.3 CREFB 3.5 <3.0 55.0 0.4 C99-21A 4.4 5.0 32.0 0.3 C99-21B 2.9 4.0 33.0 0.3 C99-25A 0.7 <3.0 12.0 0.1 C99-25B 0.9 <3.0 13.0 0.1 D99-8A 4.4 <3.0 27.0 0.4 D99-8B 3.9 <3.0 30.0 0.4 E99-19A 3.8 <3.0 122.0 1.2 E99-19B 3.6 4.0 125.0 1.5 E99-45A 27.0 6.0 108.0 0.6 E99-45B 27.0 6.0 110.0 0.6 E99-59A 0.8 <3.0 17.0 0.2 E99-59B 1.0 <3.0 24.0 0.4 E99-61A 5.3 <3.0 36.0 0.3 E99-61B 5.4 <3.0 39.0 1.0 E99-63A 10.2 4.0 126.0 0.5 E99-63B 19.0 8.0 86.0 0.3 Table A3.2 Variation Between Se Concentrations in Extractant and Residue Duplicates Average Variation (% of mean) Range of Variation (% of mean) Water Soluble 4 0-60 HFMO 12 0-91 Sulphides and Organics 21 0-100 Residual 2 0-34 Table A3.3 Measured Se Concentrations in Head Samples versus Total Se Calculated From the Sequential Extraction Results Measured Se Concentration Average Calculated Se (mg/kg) Concentration (mg/kg) A99-4 4.9 3.8 A99-7 4.5 4.3 A99-12 7.0 6.3 A99-23 5.9 5.3 B99-16 4.9 4.2 B99-44 8.8 12.4 B99-62 2.2 2.1 CREF 3.4 3.4 C99-21 2.2 2.3 C99-25 0.8 0.8 D99-8 2.2 2.0 E99-19 8.4 7.9 E99-59 1.8 7.7 E99-45 7.3 1.4 E99-61 3.0 2.8 E99-63 4.4 6.7 149 CQ CQ CQ CQ CQ CQ CQ CQ CQ CQ U 150 CQ CO m a JS PH HH H CU •a CU .2 "3 o CU E cu s cu CJ cs u H a o M CU u d o U J U 3 CQ CQ CQ CQ CQ CQ CQ CQ CQ N 151 o o o o o o o o o o o o o o o o i n o o o so o> o o o o o o o o O o o o o o o o o o o o o o o o V o o o O co o o o i n o o o o CS CQ o o o Tt V o © o m o m i n r o i n © o V o CO r o VO o v CS c s V c s c s V —* V V r - X C M ^ CS V V V r o V V V V c s m CS V vd V UJ o o o o o o o o o o o o © o o © i n O o o VO o o o o o o o O O o o o o i n o o o o o o o o o o o r o o o o © r o o o o i n o o o O CS < o o o r o V o o Ov r o m m — o r o i n o c s <—1 o V o CO —* r o o o o CS CS V c s V c s V V V V v o X c s V CS V V r o V V 7 V 00 o q vd <2( V o o o o o o o o o o o o o o o o r - O o o >o Tt o o o o o © o O o o o o o i n o o o o o o o o o o © o o O © o O CS o o o i n o © o O Tt CQ o o o V o o r ~ m c s m —^ c s r o i n o o o V o r o 1—1 r o OO VO c s c s 7 CO 7 V o o V V V V <3 X c s V c s V V o \ V V V V <3 c s CS V "n r ~ V CT\ UJ o o o o o o © o o o o o o o o o o O o o o o o o o o o O o o o o o m o o o © o o o o o o o v o O o o O —— o o o m o o o O r o <C o o o V o o i n r o i n c s r o i n r o o o o V o r o r o OV r-~ <2 <2 V V V i n V V Tt V V V V V X o o r ~ V <20 <2 V V o o o o o o o o o o o o o o o o o O o O o m o o o o o o o O o o o o o m o o o o o o © o o o i n m o o o © c s o o o m o o o O o o CQ o o o c s V o o CO i n Tt m V r o i n r o X — — o V o r o r o r o o o r o CS c s 7 CO 7 7 c s V V V V V V CS V V ON V V 7 7 c s © c s V i n V ON O UJ O o o o o o © o o o o o o o o o O O o O o c s o o o o o o o O o o o o o m o o o o o o o o o o o i n O o O © c s o o o i n o o o o r -< o o o V o o Tf Tf Tt m —^ V r o i n r o — CO o V o r o r o r o VO r o CS c s 7 o v 7 V VO V V V V V V X V f—i c s V V V V 7 7 o o r ~ <2( V o o o o o o © o o o o o o o o o o O o o o O i n o o o o o o o O o o o o o i n o © o o o o o o o o o o o O o o i n o o o m o o o O r o ca o o o c s V o o CO o o o m m r o r o i n r o " x c s o V o r o r o r o o o CO CS c s V VO V V V V V V V r o vd Tt V .—1 c s V V CS ti. UJ V V 7 V Tt <2( V 2 o o o o o o o o o o o o o o o o O O o O o OV o o o o o o O O o o o o o m o o o o o o o o o o o Tt o o © i n o o o i n o o o O OV < o o o c s V o o VO VO CO i n CO CS r o i n r o v o o V o r o r o r o r o c o CS c s V o V V V V V V c s V ^ c s V V CS ~ V V 7 V c s VO r ~ c s V V o o o o o © o © o o o o o o o o o O o O o O o o o o o O O o o o o o o i n o o o o o o © o o o o m O o O o c s o o o i n o O O O —* CQ o o o V o o CO i n i n V r o i n r o o c s o o o V o r o CS r o CO CS c s c s V V V Tt V V V V V V c s V c s V V V V 7 V V <2( V B99 o o o o o o o o o o o o o o o o o O o O o o o o o o o O o o o B99 o o o o i n o o o o o o o o o o o i n O o o o OO o o o i n o O o O Tt < © o o V o o m r o r ~ i n V r o m r o o VO o V o r o o v r o CS <s <s 7 7 CO 7 V V V V V V V CS V (—s CS V V V V V 7 7 V <2( V o o o o o o o o o o o o o o o o o O o O o o CO o o o o o O o O o o o o o m o o o o o o o o o o o m o o O —-* Tt o o o m o O o O 0 3 o o o V o o i n Tf c s i n V r o i n r o X o V o r o c s r o — Tt CS c s 7 c s 7 7 00 V V V V V r o V (—i CS V V c s VO V V V 7 vd <2( V o\ ca o o o o o o o o o o o o o o o o o O o O •o o m o o o o o O o O o o o o o m o o o o o o o o o o o m © o O i n o o o m o O o O o o < o o o V o o CS — . Tt i n i n 1—1 V r o m r o X OV o V o CO c s r o <2 <2 7 7 V OV V V r - V V V V V V 00 vd V <20 <2 V V c s o o o o o o o o o o o o o o o o o o o O v o o o o o o O o O o o o o o i n o o o i n o o o o o o o O o O O 00 o o o m o o o O —~ CQ o o o CO V o o V Tt Tf VO r o r o m CO 1mm* r o o V o r o r o i n o o CS CS 7 7 7 OV V V V V X c s V o CS V V r o m V V 7 V t-; c s V CN i n V o \ o o o o o o o o © o o o © © o o o o o o o v o o o © o O o O o <- o o o o i n o o o o o o o o o o o m o o o o o o o o o i n o o o O o < © o o CO V o o 00 VD m —^ V r o i n r o 7.8x1 o V o r o — r o V r o t-» <2 <2 V V 7 r o 7 7 VO V V V V V 7.8x1 V <20 <2 V r o o o o o o o o o o o o o o o o o o O o O o Tt o o o o o O o O O o o o o m o o o o o o o o o o o O o O © i n o o o m o O o O Tt CQ o o o c s V o o m Tt o o i n r o i n r o o o V o r o r o Tt r o c s c s V t-~ V V Ov V V V V V V c s « V c s V V ' — ' V V V 7 CO V CS V 99-4 V 99-4 < o o o o o o o o o o o o o o o o o O o O o OV o o o o o O o O o o o o o i n o o © © © o o o o o o VO O o © o Tt o o o m o O o O 00 o o o CS V o o m VO i n —» r o i n r o o o v o V o r o — r o CS < r o <2 <2 7 7 7 r ~ 7 V Tt r o V V V V V V <2 V <20 <2 V V < Sb As Ba Be 5 CQ Cd Ca Cr Co Cu Fe Pb —i OO s Mn 1 Mo 1 Z 0 - i4 Se M < 1 Na | in 1 Th Sn H > | Zn 152 10.4 APPENDIX 4 Raw Data from the Humidity Cell Tests Table A4.1 Analysis of Leachate from the E99<61 A (Interburden) Humidity Cell wash 1 wash 2 wash 3 wash 4 wash 5 wash 6 wash 7 wash 8 wash 9 wash 10 wash 11 Week 0 1 2 3 4 5 6 7 8 9 10 Eh 382.6 334.5 382.4 376.2 340.1 412.0 383.3 428.6 377.3 390.2 390.2 pH 6.80 6.89 7.20 6.13 6.00 6.23 5.98 5.75 5.87 6.17 6.17 Vol. (mL) 346 508 505 477 486 487 488 484 516 497 497 Dissolved Anions Acid." 1000 1000 1000 1000 1000 6000 Alk." 25000 34000 23000 28000 30000 30000 sor 28000 23000 17000 12000 14000 13000 Dissolved Metals Al <200 <200 <200 <200 ooo ooo Sb <200 <200 <200 <200 ooo ooo As <200 <200 <200 <200 ooo ooo Ba 400 190 130 110 110 80 Be <5 <5 <5 <5 <5 <5 Bi <100 <100 <100 ooo ooo ooo Bo ooo <100 <100 ooo ooo ooo Cd <10 <10 <10 oo oo oo Ca 11 200 16 000 8210 20 500 20 900 14 000 Cr <10 <10 <10 oo oo OO Co <10 <10 <10 oo oo oo Cu <10 10 <10 oo oo oo Fe <30 <30 <30 oo oo oo Pb <50 <50 <50 OO oo oo Li 50 50 30 30 20 20 Mg 4200 6400 3700 8000 8500 6000 Mn 11 15 7 16 16 12 Mo <30 40 40 oo oo oo Ni 70 <50 <50 oo oo oo P <300 ooo ooo ooo ooo ooo K 7000 33000 4000 3000 3000 3000 Se 16 10 9 7 8 7 7 7 7 7 8 Si 790 1070 890 880 1080 980 Ag <10 20 <10 oo oo OO Na 2000 <2000 <2000 oooo oooo oooo Sr 79 83 48 95 92 59 Th <200 <200 <200 ooo ooo ooo Sn <30 <30 oo oo oo OO Ti <10 <10 <10 oo oo oo V <30 <30 oo oo oo oo Zn <5 8 <5 5 9 <5 Organic Parameters TIC 600 3100 1600 1000 1100 1600 TOC 7500 4100 3600 1500 1500 5300 NOTE: All concentrations expressed in ug/L 153 Table A4.1 Continued wash 12 wash 13 wash 14 wash 15 wash 16 wash 17 wash 18 wash 19 wash 20 wash 21 Week 11 12 13 14 15 16 17 18 19 20 Eh 453.4 396.3 400.1 490.4 466.8 399.6 415.9 521.4 456.8 444.1 pH 7.02 7.27 5.94 8.47 7.10 6.76 8.17 8.27 8.06 7.14 Vol. (mL) 517 480 450 512 483 490 494 491 506 442 Dissolved Anions Acid. <1000 6000 oooo 3000 Alk. 23 000 23 000 24 000 31 000 S0 4 2 < 15 000 15 000 13 000 12 000 Dissolved Metals Al <200 ooo ooo ooo Sb <200 ooo ooo ooo As <200 ooo ooo ooo Ba 80 90 80 90 Be <5 <5 <5 o Bi <100 ooo ooo ooo Bo <100 ooo ooo ooo Cd <10 oo oo oo Ca 9740 9240 9280 9900 Cr <10 oo OO oo Co <10 oo oo oo Cu <10 oo oo oo Fe <30 oo oo OO Pb <50 oo oo oo Li 10 10 10 10 Mg 4000 4700 4100 4300 Mn 7 6 7 7 Mo <30 oo OO oo Ni <50 oo oo oo P ooo ooo ooo ooo K 2000 2000 oooo 2000 Se 8 8 10 8 9 11 9 9 7 6 Si 850 1030 1050 1030 Ag <10 oo oo oo Na <2000 oooo oooo oooo Sr 46 47 50 51 Th <200 ooo ooo ooo Sn oo oo oo oo Ti <10 oo OO oo V oo oo oo oo Zn <5 <5 <5 <5 Organic Parameters TIC 1700 1500 2100 ooo TOC 1200 1600 1200 1000 NOTE: Al l concentrations expressed in ug/L 154 Table A4.2 Analysis of Leachate from the E99<61B (Interburden) Humidity Cell wash 1 wash 2 wash 3 wash 4 wash 5 wash 6 wash 7 wash 8 wash 9 wash 10 wash 11 Week 0 1 2 3 4 5 6 7 8 9 10 Eh 341.9 470.2 470.9 340.1 418.4 394.5 404.7 408.0 386.4 410.1 504.1 PH 6.76 6.95 7.35 6.28 6.65 6.45 6.12 5.97 5.51 6.30 7.61 Vol. (mL) 365 477 480 482 505 457 508 494 470 498 496 Dissolved Anions Acid." 2000 2000 1000 1000 1000 6000 Alk." 23 000 26 000 30 000 28 000 30 000 37 000 S0 4 2 < 20 000 15 000 13 000 11 000 10 000 11 000 Dissolved Metals Al <200 <200 <200 <200 <200 <200 Sb <200 <200 <200 <200 <200 <200 As <200 <200 <200 <200 <200 <200 Ba 310 240 150 150 190 140 Be <5 <5 <5 <5 <5 <5 Bi <100 <100 <100 <100 <100 <100 Bo <100 <100 <100 <100 <100 <100 Cd <10 <10 <10 <10 <10 <10 Ca 10 500 15 700 9470 13 000 23 000 16 600 Cr <10 <10 <10 <10 <10 <10 Co <10 <10 <10 <10 <10 <10 Cu 10 30 <10 <10 <10 <10 Fe 70 170 <30 <30 <30 <30 Pb <50 <50 <50 <50 <50 <50 Li 40 40 30 30 30 20 Mg 3200 5900 4000 5400 9100 6900 Mn 12 12 5 8 14 13 Mo <30 50 50 30 <30 <30 Ni <50 <50 <50 <50 <50 <50 P <300 <300 <300 <300 <300 <300 K 7000 5000 5000 4000 3000 3000 Se 10 12 11 9 9 8 8 7 8 7 10 Si 690 1040 1090 990 1190 1170 Ag <1() <10 <10 <10 <10 <10 Na 3000 <2000 <2000 <2000 <2000 <2000 Sr 65 82 60 70 110 78 Th <200 <20() <200 <200 <200 <200 Sn <30 <30 <30 <30 <30 <30 Ti <10 <10 <10 <10 <10 <10 V <30 <30 <30 <30 <30 <30 Zn 7 215 <5 <5 <5 47 Organic Parameters TIC 1200 <500 1900 1600 900 1200 TOC 5700 5500 2100 1400 1600 5300 | NOTE: All concentrations expressed in ug/L 155 Table A4.2 Continued wash 12 wash 13 wash 14 wash 15 wash 16 wash 17 wash 18 wash 19 wash 20 wash 21 Week 11 12 13 14 15 16 17 18 19 20 Eh 411.4 398.8 410.5 450.6 445.7 414.7 445.2 455.6 480.7 438.2 pH 7.33 8.27 7.70 8.43 7.60 7.80 8.17 8.15 6.76 7.39 Vol. (mL) 496 483 469 480 494 497 490 479 490 496 Dissolved Anions Acid. <1000 4000 1000 1000 Alk. 30 000 25 000 30 000 30 000 S04l< 15 000 14 000 14 000 12 000 Dissolved Metals Al <200 <200 200 ooo Sb <200 <200 <200 ooo As <200 <200 <200 ooo Ba 120 110 140 130 Be <5 <5 <5 <5 Bi <100 <100 <100 ooo Bo <100 <100 <100 ooo Cd <10 <10 <10 oo Ca 12800 9990 10700 10200 Cr <10 <10 <10 OO Co <10 <10 <10 oo Cu <10 <10 10 oo Fe <30 <30 <30 oo Pb <50 <50 <50 oo Li 20 10 20 oo Mg 5200 4800 4600 4400 Mn 9 7 6 6 Mo 30 <30 <30 oo Ni <50 <50 <50 O O P <300 <300 ooo ooo K 3000 2000 2000 oooo Se 10 9 11 9 10 11 8 14 7 7 Si 1040 1090 1290 1070 Ag <10 <10 <10 oo Na <2000 <2000 oooo oooo Sr 62 51 61 52 Th <200 <200 ooo ooo Sn <30 <30 oo oo Ti <10 <10 <10 oo V <30 <30 oo oo Zn <5 <5 <5 <5 Organic Parameters TIC 2000 2000 900 2200 TOC 1200 1500 1600 900 NOTE: All concentrations expressed in ug/L 156 Table A4.3 Analysis of Leachate from the A99<23A (Parting) Humidity Cell wash 1 wash 2 wash 3 wash 4 wash 5 wash 6 wash 7 wash 8 wash 9 wash 10 wash 11 Week 0 1 2 3 4 5 6 7 8 9 10 Eh 456.2 332.6 350.2 399.0 347.0 374.3 397.5 417.5 382.1 423.8 436.6 pH 6.40 6.58 6.82 6.20 5.87 6.87 7.01 5.40 5.51 5.51 5.74 Vol. (mL) 322 495 510 490 475 452 476 512 495 492 484 Dissolved Anions Acid." 2000 2000 2000 2000 2000 3000 Alk." 11 000 9000 10 000 9000 9000 3000 sor 18 000 16 000 17 000 10 000 10 000 9000 Dissolved Metals Al <200 ooo ooo ooo ooo ooo Sb <200 ooo ooo ooo ooo ooo As <200 ooo ooo ooo ooo ooo Ba 310 100 80 80 100 70 Be <5 <5 <5 <5 <5 <5 Bi <100 ooo ooo ooo ooo ooo Bo <100 ooo ooo ooo ooo ooo Cd <10 oo oo oo oo oo Ca 5780 5330 4600 8960 12 400 13 000 Cr <10 oo oo oo oo OO Co 10 oo oo oo oo 10 Cu <10 oo OO oo 20 20 Fe <30 30 oo oo oo oo Pb <50 oo oo oo oo oo Li 70 50 40 30 40 40 Mg 1400 1400 1100 2200 3300 3400 Mn 12 <5 o <5 <5 6 Mo 80 240 150 30 oo oo Ni <50 oo oo oo OO oo P ooo ooo ooo ooo ooo ooo K 8000 17 000 5000 6000 5000 4000 Se 48 93 100 46 38 34 35 33 19 34 49 Si 800 1330 1660 1550 2620 2880 Ag <10 oo oo oo oo oo Na 4000 oooo oooo oooo oooo oooo Sr 16 14 11 18 24 23 Th ooo ooo ooo ooo ooo ooo Sn oo oo oo oo oo oo Ti <10 oo oo oo oo OO V OO oo oo oo oo oo Zn 50 23 8 21 41 47 Organic Parameters TIC 700 600 1000 ooo ooo ooo TOC 6300 5500 1800 1900 1700 2500 NOTE: All concentrations expressed in ug/L 157 Table A4.3 Continued wash 12 wash 13 wash 14 wash 15 wash 16 wash 17 wash 18 wash 19 wash 20 wash 21 Week 11 12 13 14 15 16 17 18 19 20 Eh 423.3 423.0 417.5 450.8 474.1 448.2 509.5 487.0 540.5 439.9 PH 6.37 6.19 6.24 7.40 6.43 6.37 6.18 6.50 6.08 5.59 Vol. (mL) 486 483 445 470 481 484 507 475 495 493 Dissolved Anions Acid. 2000 2000 2000 2000 Alk. 2000 1000 2000 1000 S0 4 2 < 18 000 18 000 20 000 12 000 Dissolved Metals Al <200 <200 <200 <200 Sb <200 <200 <200 <200 As <200 <200 <200 <200 Ba 70 70 60 50 Be <5 <5 <5 <5 Bi <100 <100 <100 ooo Bo <100 <100 <100 ooo Cd <10 <10 <10 oo Ca 13 800 9640 9330 5570 Cr <10 <10 <10 oo Co 20 20 20 10 Cu <10 10 20 20 Fe <30 <30 <30 oo Pb <50 <50 <50 oo Li 40 20 30 20 Mg 3700 2900 2600 1600 Mn 7 7 8 6 Mo <30 <30 <30 OO Ni <50 <50 <50 oo P <300 ooo ooo ooo K 4000 4000 4000 oooo Se 59 46 59 46 66 49 39 42 41 32 Si 3330 3210 3940 2820 Ag <10 <10 <10 OO Na <2000 <2000 <2000 oooo Sr 27 20 21 12 Th <200 <200 <200 ooo Sn <30 <30 oo oo Ti <10 <10 <10 oo V <30 <30 oo oo Zn 95 80 89 72 Organic Parameters TIC <500 2100 ooo ooo TOC 1400 1500 1400 1100 NOTE: All concentrations expressed in ug/L 158 Table A4.4 Analysis of Leachate from the A99<23B (Parting) Humidity Cell wash 1 wash 2 wash 3 wash 4 wash 5 wash 6 wash 7 wash 8 wash 9 wash 10 wash 11 Week 0 1 2 3 4 5 6 7 8 9 10 Eh 460.5 408.7 344.7 368.5 352.7 415.7 398.9 425.8 378.2 393.4 489.2 PH 7.02 6.51 6.60 6.14 6.05 6.55 6.82 5.70 5.26 5.10 6.12 Vol. (mL) 313 491 495 405 484 461 493 490 480 496 494 Dissolved Anions Acid.* 2000 2000 1000 2000 3000 <1000 Alk.** 10 000 6000 7000 7000 8000 1000 S0 4 : 18 000 16 000 18 000 11 000 12 000 8000 Dissolved Metals Al <200 <200 <200 <200 <200 <200 Sb <200 <200 <200 <200 <200 <200 As <200 <200 <200 <200 <200 <2()0 Ba 340 130 90 60 80 100 Be <5 <5 <5 <5 <5 <5 Bi <100 <100 <100 <100 <100 <100 Bo <100 <100 <100 <100 <100 <100 Cd <10 <10 <10 <10 <10 <10 Ca 8070 7730 4750 3790 6370 10600 Cr <10 <10 <10 <10 <10 <10 Co <10 <10 <10 <10 <10 <10 Cu 10 <10 <10 <10 <10 <10 Fe <30 <30 <30 <30 <30 <30 Pb <50 <50 <50 <50 <50 <50 Li 80 70 40 20 30 30 Mg 1900 1900 1100 900 1700 2700 Mn 24 6 <5 <5 <5 <5 Mo 90 280 140 90 50 <30 Ni <50 <50 <50 <50 <50 <50 P <300 <300 <300 <300 <300 <300 K 8000 6000 5000 5000 5000 5000 Se 55 110 90 46 42 39.7 41 32 32 31 40 Si 1010 1220 1340 1610 2320 2300 Ag <10 <10 <10 <10 <10 <10 Na 6000 3000 <2000 <2000 <2000 <2000 Sr 21 19 11 8 14 21 Th <200 <200 <200 <200 <200 <200 Sn <30 <30 <30 <30 <30 <30 Ti <10 <10 <10 <10 <10 <10 V <30 <30 <30 <30 <30 <30 Zn 82 45 19 9 20 47 Organic Parameters TIC <500 <500 700 800 500 <500 TOC 7100 3600 2500 1400 1500 2400 NOTE: A l l concentrations expressed in ug/L 159 Table A4.4 Continued wash 12 wash 13 wash 14 wash 15 wash 16 wash 17 wash 18 wash 19 wash 20 wash 21 Week 11 12 13 14 15 16 17 18 19 20 Eh 428.3 405.6 418.1 465.7 480.7 442.1 548.9 497.0 529.0 442.7 PH 6.75 6.84 5.96 7.58 6.40 6.22 6.41 6.10 6.45 5.62 Vol. (mL) 508 480 453 511 479 500 478 488 501 489 Dissolved Anions Acid. 2000 4000 1000 2000 Alk. 2000 2000 2000 2000 S0 4 2 < 14 000 16 000 13 000 10 000 Dissolved Metals Al <200 ooo ooo ooo Sb <200 ooo ooo ooo As <200 ooo ooo ooo Ba 70 80 70 60 Be <5 o <5 o Bi <100 ooo ooo ooo Bo <100 ooo ooo ooo Cd <10 oo oo oo Ca 8470 10000 7570 6840 Cr <10 oo oo OO Co <10 oo oo O O Cu 10 oo 10 oo Fe <30 oo OO oo Pb <50 oo oo oo Li 20 20 20 20 Mg 2100 2800 2000 1900 Mn <5 5 <5 <5 Mo <30 oo oo oo Ni <50 oo oo OO P ooo ooo ooo OOO K 4000 5000 4000 3000 Se 46 41 57 41 45 46 44 42 37 33 Si 2030 3440 2780 2560 Ag <10 oo oo oo Na oooo oooo oooo oooo Sr 18 22 18 16 Th ooo ooo ooo ooo Sn oo oo oo oo Ti <10 oo oo oo V OO oo oo oo Zn 39 54 46 55 Organic Parameters TIC ooo ooo ooo ooo TOC 1400 1400 1400 1000 NOTE: All concentrations expressed in ug/L 160 Table A4.5 Analysis of Leachate from the CREFA (Refuse) Humidity Cell wash 1 wash 2 wash 3 wash 4 wash 5 wash 6 wash 7 wash 8 wash 9 wash 10 wash 11 Week 0 1 2 3 4 5 6 7 8 9 10 Eh 441.0 328.8 383.3 354.8 409.9 361.4 371.2 393.3 370.9 310.1 415.0 PH 6.81 6.96 7.03 5.98 6.33 6.88 7.27 6.35 6.01 6.98 7.44 Vol. (mL) 336 469 497 482 490 479 500 499 493 481 492 Dissolved Anions Acid.* 000 2000 1000 oooo 2000 4000 A l t " 26 000 25 000 27 000 35 000 39 000 26 000 S0 4 2 < 24 000 32 000 16 000 18 000 15 000 11 000 Dissolved Metals Al <200 <200 ooo ooo ooo ooo Sb <200 <200 ooo ooo ooo ooo As <200 <200 ooo ooo ooo ooo Ba 160 80 130 90 120 130 Be <5 <5 <5 o o o Bi <100 <100 <100 ooo ooo ooo Bo <100 <100 ooo ooo ooo ooo Cd <10 <10 oo oo oo oo Ca 12 200 21 800 17 600 14 300 15 100 17 600 Cr <10 <10 oo oo oo OO Co <10 <10 oo oo oo oo Cu <10 <10 oo oo oo oo Fe <30 <30 OO oo oo oo Pb <50 <50 oo oo oo oo Li <10 <10 oo oo oo oo Me 3400 5500 5100 4600 5300 5900 Mn 7 7 o <5 o <5 Mo <30 50 oo oo oo oo Ni <50 <50 oo oo oo oo P <300 ooo ooo ooo ooo ooo K 3000 3000 oooo oooo oooo oooo Se 39 55 34 11 15 15 15 13 13 11 34 Si 490 580 530 540 610 530 Ag <10 <10 oo oo oo OO Na <2000 oooo oooo oooo oooo oooo Sr 39 40 42 37 41 46 Th <200 ooo ooo ooo ooo ooo Sn <30 oo oo oo oo oo Ti <10 <10 oo oo oo OO V <30 oo oo oo oo oo Zn <5 6 o <5 o 5 Organic Parameters TIC 1000 1900 1800 2200 ooo 3000 TOC 5800 4900 1200 1300 4300 2500 NOTE: All concentrations expressed in ug/L 161 Table A4.5 Continued wash 12 wash 13 wash 14 wash 15 wash 16 wash 17 wash 18 wash 19 wash 20 wash 21 Week 11 12 13 14 15 16 17 18 19 20 Eh 403.1 397.3 401.5 433.6 447.4 415.8 515.7 482.7 492.7 467.0 PH 6.92 7.97 7.69 8.57 7.31 7.71 7.74 7.63 7.69 7.27 Vol. (mL) 500 502 442 466 491 489 490 478 494 464 Dissolved Anions Acid. < 1000 3000 oooo 2000 Alk. 36 000 26 000 44 000 34 000 sor 16 000 14 000 14 000 10 000 Dissolved Metals Al <200 ooo ooo ooo Sb <200 ooo ooo ooo As <200 ooo ooo ooo Ba 110 90 140 130 Be <5 <5 <5 <5 Bi <100 ooo ooo ooo Bo ooo ooo ooo ooo Cd oo oo oo oo Ca 13400 9870 14000 10000 Cr oo oo oo oo Co oo oo oo oo Cu oo oo OO oo Fe oo oo oo oo Pb oo oo oo oo Li OO oo 10 oo Mg 4700 4400 6100 4100 Mn o <5 <5 <5 Mo OO oo OO oo Ni OO oo OO oo P ooo ooo ooo ooo K oooo oooo oooo oooo Se 16 17 16 170 17 19 13 15 14 9 Si 450 420 600 380 Ag oo oo oo O O Na oooo oooo oooo oooo Sr 38 29 46 32 Th ooo ooo ooo ooo Sn OO oo oo oo Ti oo oo oo oo V oo oo oo oo Zn <5 o <5 <5 Organic Parameters TIC 2500 1200 2800 2100 TOC 1200 1400 1300 1200 NOTE: All concentrations expressed in ug/L 162 Table A4.6 Analysis of Leachate from the CREFB (Refuse) Humidity Cell wash 1 wash 2 wash 3 wash 4 wash 5 wash 6 wash 7 wash 8 wash 9 wash 10 wash 11 Week 0 1 2 3 4 5 6 7 8 9 10 Eh 453.6 322.0 366.0 361.4 393.2 381.5 370.9 399.1 369.5 308.9 453.2 pH 6.87 7.01 6.90 6.18 6.73 7.09 7.24 6.43 6.20 6.80 7.54 Vol. (mL) 305 510 499 480 491 499 490 493 494 488 500 Dissolved A .nions Acid." 2000 1000 1000 <1000 2000 3000 Alk" 26 000 30 000 25 000 32 000 53 000 50 000 S 0 4 2 < 26 000 40 000 18 000 18 000 18 000 12 000 Dissolved N Petals Al <200 <200 <200 <200 o o o o o o Sb <200 <200 <200 <200 o o o o o o As <200 <200 <200 <200 o o o o o o Ba 200 100 140 80 140 120 Be <5 <5 <5 <5 <5 <5 Bi <100 <100 <100 <100 o o o o o o Bo <100 <100 <100 <100 o o o o o o Cd <10 <10 <10 <10 o o o o Ca 15 400 17 100 17 400 13 100 21 700 17 700 Cr <10 <10 <10 <10 o o o o Co <10 <10 <10 <10 o o o o Cu 10 <10 <10 <10 o o OO Fe <30 <30 <30 <30 o o o o Pb <50 <50 <50 <50 OO o o Li <10 <10 <10 <10 10 10 Mg 3500 5000 5000 4300 7000 6200 Mn 13 <5 <5 <5 <5 <5 Mo <30 50 <30 <30 o o o o Ni <50 <50 <50 <50 OO o o P <300 <300 <300 o o o OOO o o o K 4000 2000 <2000 <2000 oooo oooo Se 35 55 25 14 17 13 14 13 11 10 32 Si 500 530 490 470 690 480 Ag <10 <10 <10 o o o o o o —Lis Na <2000 <2000 <2000 oooo oooo oooo Sr 38 46 38 30 47 41 Th <200 <200 <200 o o o o o o o o o Sn <30 <30 <30 o o o o o o Ti <10 <10 <10 o o o o o o V <30 <30 <30 o o o o o o Zn <5 <5 <5 <5 <5 10 Organic P arameters TIC 1400 2600 . 2800 o o o 800 TOC 7200 3500 1400 1200 2100 2800 NOTE: Al l concentrations expressed in ug/L 163 Table A4.6 Continued wash 12 wash 13 wash 14 wash 15 wash 16 wash 17 wash 18 wash 19 wash 20 wash 21 Week 11 12 13 14 15 16 17 18 19 20 Eh 397.5 392.5 393.8 421.1 445.5 501.1 492.3 478.9 470.2 449.1 PH 6.62 7.96 7.52 8.26 7.44 7.68 7.69 7.92 7.88 7.48 Vol. (mL) 502 498 490 492 502 487 501 494 482 499 Dissolved Anions Acid. <1000 2000 1000 1000 Alk. 32 000 33 000 30 000 31 000 S0 4 2 < 16 000 13 000 12 000 10 000 Dissolved Metals Al <200 <200 <200 <200 Sb <200 <200 <200 <200 As <200 <200 <200 <200 Ba 80 90 100 90 Be <5 <5 <5 <5 Bi <100 <100 <100 <100 Bo <100 <100 <100 <100 Cd <10 <10 <10 <10 Ca 12 100 11 400 9680 10 800 Cr <10 <10 <10 <10 Co <10 <10 <10 <10 Cu <10 10 <10 <10 Fe <30 <30 <30 <30 Pb <50 <50 <50 <50 Li <10 <10 <10 <10 Mg 4400 4900 4200 4700 Mn <5 <5 <5 <5 Mo <30 <30 <30 <30 Ni <50 <50 <50 <50 P <300 <300 <300 ooo K <2000 <2000 <2000 oooo Se 16 15 14 150 13 16 13 12 14 9 Si 380 420 420 390 Ag <10 <10 <10 <10 Na <2000 <2000 <2000 <2000 Sr 31 30 30 31 Th <200 <200 <200 ooo Sn <30 <30 <30 oo Ti <10 <10 <10 <10 V <30 <30 <30 oo Zn <5 <5 <5 <5 Organic Parameters TIC 2800 2700 2300 2500 TOC 2000 1900 1900 1100 NOTE: All concentrations expressed in ug/L 164 Table A4.7 Analysis of Leachate from the B99<44A (Coal) Humidity Cell wash 1 wash 2 wash 3 wash 4 wash 5 wash 6 wash 7 wash 8 wash 9 wash 10 wash 11 Week 0 1 2 3 4 5 6 7 8 9 10 Eh 475.2 334.4 495.4 426.7 385.7 381.7 403.3 412.2 375.0 403.8 429.1 pH 7.01 6.05 3.83 5.42 6.62 7.31 4.99 5.49 6.07 5.35 7.06 Vol. (mL) 246 433 445 475 492 498 425 361 490 495 485 Dissolved Anions Acid." 3000 2000 2000 2000 2000 1000 Alk." 6000 3000 2000 2000 2000 1000 S0 4 2 < 95 000 77 000 23 000 17 000 12 000 10 000 Dissolved Metals Al <200 ooo ooo ooo ooo ooo Sb <200 ooo ooo ooo ooo ooo As <200 ooo ooo ooo ooo ooo Ba 30 10 OO oo OO 0 0 Be <5 <5 <5 <5 <5 o Bi <100 ooo ooo ooo ooo ooo Bo <100 ooo ooo ooo ooo ooo Cd <10 oo oo oo oo OO Ca 28 500 23 900 6060 4380 3690 3250 Cr <10 oo oo OO oo 0 0 Co <10 oo oo oo oo oo Cu 20 oo 10 10 10 oo Fe 70 oo oo 50 30 40 Pb <50 oo oo oo oo oo Li <10 oo OO OO oo oo Mg 13 700 11 700 3000 2100 1900 1600 Mn 19 7 o o O <5 Mo <30 oo oo oo OO oo Ni <50 OO oo oo oo oo P ooo ooo ooo ooo ooo ooo K. oooo oooo oooo oooo oooo oooo Se 270 190 110 40 32 31 28 25 20 23 56 Si 740 1290 1120 990 560 650 Ag <10 OO OO OO OO oo Na oooo oooo oooo oooo oooo oooo Sr 40 31 9 6 5 o Th ooo ooo ooo ooo ooo ooo Sn oo OO oo oo oo oo Ti <10 oo oo oo OO oo V oo oo oo oo oo oo Zn 52 11 o 6 <5 6 Organic Pi rameters TIC ooo ooo ooo ooo 500 1300 TOC 12 000 8300 6100 7700 6400 6500 NOTE: A l l concentrations expressed in ug/L 165 Table A4.7 Continued wash 12 wash 13 wash 14 wash 15 wash 16 wash 17 wash 18 wash 19 wash 20 wash 21 Week 11 12 13 14 15 16 17 18 19 20 Eh 437.2 390.6 500.4 496.8 496.8 465.7 449.5 415.8 461.5 445.9 PH 6.68 6.47 7.09 6.81 6.81 6.46 6.11 6.76 6.96 5.30 Vol. 436 478 478 480 499 465 479 466 472 480 Dissolved Anions Acid. 1000 2000 1000 2000 Alk. 2000 oooo 1000 2000 S0 4 2 < 14 000 11 000 11 000 8000 Dissolved Metals Al <200 o o o o o o o o o Sb <200 o o o o o o o o o As <200 o o o o o o o o o Ba <10 OO OO o o Be <5 <5 o <5 Bi <100 o o o o o o o o o Bo <100 o o o o o o o o o Cd <10 o o o o o o Ca 3310 2910 2610 2420 Cr <10 o o 0 0 OO Co <10 OO OO o o Cu <10 o o 10 50 Fe <30 30 o o o o Pb <50 OO o o o o Li <10 o o o o o o Mg 1600 1600 1300 1200 Mn <5 <5 <5 <5 Mo <30 o o o o o o Ni <50 o o OO o o P o o o o o o o o o o o o K <2000 oooo oooo oooo Se 20 21 23 21 17 21 17 14 14 11 Si 490 500 450 370 Ag 10 10 o o o o Na oooo oooo oooo oooo Sr <5 <5 <5 <5 Th o o o o o o o o o o o o Sn o o o o o o o o Ti o o o o o o o o V o o o o o o o o Zn <5 <5 <5 8 Organic Parameters TIC o o o o o o o o o o o o TOC 5000 5600 5300 3900 NOTE: All concentrations expressed in ug/L 166 Table A4.8 Analysis of Leachate from the B99<44B (Coal) Humidity Cell wash 1 wash 2 wash 3 wash 4 wash 5 wash 6 wash 7 wash 8 wash 9 wash 10 wash 11 Week 0 1 2 3 4 5 6 7 8 9 10 Eh 475.8 483.4 410.8 385.7 345.1 394.0 388.4 463.6 388.6 468.5 448.9 PH 6.97 5.43 5.58 6.62 3.42 7.12 5.38 5.81 6.02 6.26 6.62 Vol. (mL) 240 440 505 445 482 505 494 475 502 481 487 Dissolved Anions Acid." 2000 2000 3000 2000 2000 5000 Alk." 6000 3000 2000 1000 1000 2000 S0 4 2 < 134 000 20 000 12 000 16 000 13 000 10 000 Dissolved Metals Al <200 <200 ooo ooo ooo ooo Sb <200 <200 ooo ooo ooo ooo As <200 <200 ooo ooo ooo ooo Ba 50 10 oo oo oo oo Bc <5 <5 <5 <5 <5 <5 Bi <100 <100 ooo ooo ooo ooo Bo <100 <100 ooo ooo ooo ooo Cd <10 <10 oo OO oo oo Ca 37 400 22 500 3710 3890 3510 3140 Cr <10 <10 oo oo OO OO Co 0 0 <10 OO OO oo oo Cu 30 10 oo 10 10 20 Fe 210 <30 60 40 80 50 Pb <50 <50 oo OO OO oo Li <10 <10 oo OO oo oo Mg 17 300 10 900 1800 1900 1800 1500 Mn 16 7 <5 <5 <5 <5 Mo <30 <30 oo oo oo oo Ni <50 <50 OO OO OO OO P <300 ooo ooo OOO ooo ooo K 2000 <2000 oooo oooo oooo oooo Se 400 220 80.1 26 29 21 25 14 15 19 46 Si 1010 1830 1190 860 720 530 Ag <10 oo OO OO oo oo Na <2000 oooo oooo oooo oooo oooo Sr 49 29 6 6 <5 <5 Th <200 ooo ooo ooo ooo ooo Sn <30 oo oo oo oo oo Ti <10 oo oo oo oo oo V <30 oo oo oo oo oo Zn 174 19 6 <5 <5 6 Organic Parameters TIC <500 ooo ooo ooo ooo ooo TOC 15 400 9000 9100 6800 8800 5200 NOTE: All concentrations expressed in pg/L 167 Table A4.8 Continued wash 12 wash 13 wash 14 wash 15 wash 16 wash 17 Wash 18 wash 19 wash 20 wash 21 Week 11 12 13 14 15 16 17 18 19 20 Eh 483.5 436.5 509.8 509.8 495.6 446.4 451.7 476.0 499.8 465.8 PH 6.45 5.70 6.67 6.67 5.95 6.62 6.55 6.37 6.69 5.67 Vol. (mL) 479 540 470 490 500 494 506 496 464 492 Dissolved Anions Acid. 1000 2000 2000 2000 Alk. 2000 1000 2000 2000 sor 15 000 10 000 11 000 8000 Dissolved Metals Al <200 ooo ooo ooo Sb <200 ooo ooo ooo As <200 ooo ooo ooo Ba <10 <10 o o o o Be <5 o o O Bi <100 ooo ooo ooo Bo <100 ooo ooo ooo Cd <10 o o o o o o Ca 3580 2450 2580 2050 Cr <10 o o o o o o Co <10 o o o o o o Cu <10 o o 10 20 Fe 40 o o 30 o o Pb <50 o o OO o o Li <10 o o o o OO Mg 1700 1300 1200 1000 Mn <5 o o o Mo <30 o o o o o o Ni <50 o o o o o o P ooo ooo ooo ooo K 2000 oooo oooo OOOO Se 21 20 15 20 19 19 14 12 12 11 Si 580 480 540 420 Ag <10 OO o o o o Na oooo oooo oooo oooo Sr 5 o o <5 Th ooo ooo ooo ooo Sn o o o o o o o o Ti <10 o o o o o o V o o o o o o o o Zn <5 <5 <5 11 Organic Parameters TIC ooo ooo ooo ooo TOC 4900 4500 5300 4000 NOTE: Al l concentrations expressed in ug/L 168 Table A4.9 Analysis of Leachate from the E99<19A (Foot Wall) Humidity Cell wash 1 wash 2 wash 3 wash 4 wash 5 wash 6 wash 7 wash 8 wash 9 wash 10 wash 11 Week 0 1 2 3 4 5 6 7 8 9 10 Eh 478.0 367.2 447.8 460.6 377.3 394.0 417.2 412.0 387.7 354.2 469.5 PH 6.68 5.51 6.96 5.01 5.75 5.95 5.80 6.10 6.30 6.02 6.00 Vol. (mL) 272 445 425 500 427 480 473 495 504 493 503 Dissolved Anions Acid.* 3000 2000 3000 1000 1000 4000 A l k " 6000 7000 <1000 2000 1000 2000 S O / 368 000 577 000 129 000 44 000 24 000 20 000 Dissolved Metals A l <200 <200 <200 ooo ooo ooo Sb <200 <200 <200 ooo ooo ooo As <200 <200 <200 ooo ooo ooo Ba 80 30 40 40 30 20 Be <5 <5 <5 <5 o o Bi <100 <100 <100 ooo ooo ooo Bo <100 <100 <100 ooo ooo ooo Cd 10 <10 <10 0 0 oo oo Ca 110 000 144 000 36 800 11 800 6760 5570 Cr <10 <10 <10 0 0 oo oo Co 10 <10 <10 oo oo oo Cu 30 <10 10 oo oo oo Fc 280 <30 230 oo oo oo Pb <50 <50 <50 oo O O oo Li 10 20 10 10 oo oo Mg 29 000 47 800 10 200 4200 2600 2100 Mn 112 131 26 10 6 o Mo <30 <30 <30 oo O O oo Ni <50 <50 <50 oo oo oo P <300 <300 ooo ooo ooo ooo K 3000 3000 2000 oooo oooo oooo Se 270 150 90 17 13 8 8 7 6 6 6 Si 740 2410 2700 2210 1920 1580 Ag <10 <10 <10 oo oo oo Na <2000 <2000 <2000 oooo oooo oooo Sr 145 219 64 25 15 12 Th <200 <200 ooo ooo ooo ooo Sn <30 <30 oo oo oo O O Ti <10 <10 <10 oo oo oo V <30 <30 oo oo oo oo Zn 653 329 87 34 21 18 Organic Parameters TIC <500 <500 ooo ooo ooo ooo T O C 6600 4600 2200 2200 2600 2400 NOTE: All concentrations expressed in ug/L 169 Table A4.9 Continued wash 12 wash 13 wash 14 wash 15 wash 16 wash 17 wash 18 wash 19 wash 20 wash 21 Week 11 12 13 14 15 16 17 18 19 20 Eh 411.3 414.6 417.6 456.7 458.9 454.3 510.7 505.8 477.8 455.5 PH 6.84 6.45 6.44 7.68 6.61 6.75 6.16 6.37 6.25 6.50 Vol. (mL) 504 490 486 478 500 493 477 490 487 495 Dissolved Anions Acid. 2000 2000 2000 2000 Alk. 2000 1000 2000 1000 S0 4 2 < 23 000 16 000 15 000 10 000 Dissolved Metals Al <200 ooo ooo ooo Sb <200 ooo ooo ooo As <200 ooo ooo ooo Ba 20 20 20 oo Be <5 <5 <5 o Bi <100 ooo ooo ooo Bo <100 <100 ooo ooo Cd <10 oo oo oo Ca 5510 3620 3630 2800 Cr <10 oo oo oo Co <10 oo oo oo Cu <10 oo oo oo Fe <30 oo oo oo Pb <50 OO oo oo Li <10 oo oo oo Mg 2000 1400 1300 1000 Mn 6 7 o <5 Mo <30 OO oo oo Ni <50 oo oo oo P ooo ooo ooo ooo K oooo oooo oooo oooo Se 7.6 7 5 8 6.5 7 7 5 4 5 Si 1410 1290 1540 1190 Ag OO OO OO oo Na oooo oooo oooo oooo Sr 13 9 10 9 Th ooo ooo ooo ooo Sn oo oo oo oo Ti OO <10 oo OO V oo oo oo oo Zn 19 14 13 13 Organic Parameters TIC ooo ooo ooo ooo TOC 1700 1700 1800 1500 NOTE: All concentrations expressed in pg/L 170 Table A4.10 Analysis of Leachate from the E99<19B (Foot Wall) Humidity Cell wash 1 wash 2 wash 3 wash 4 wash 5 wash 6 wash 7 wash 8 wash 9 wash 10 wash 11 Week 0 1 2 3 4 5 6 7 8 9 10 Eh 376.8 363.0 454.9 412.3 367.3 390.0 389.4 407.4 393.7 351.9 461.5 pH 6.77 5.41 6.83 5.20 5.86 5.93 5.29 5.87 6.20 5.72 6.23 Vol. (mL) 280 488 490 510 484 484 494 495 499 495 494 Dissolved Anions Acid." 2000 2000 2000 1000 1000 7000 Alk." 6000 7000 2000 <1000 1000 1000 S0 4 2 < 395000 525000 114 000 42 000 24 000 -Dissolved Metals Al <200 <200 <200 <200 <200 <200 Sb <200 <200 <200 <200 <200 <200 As <200 <200 <200 <200 <200 <200 Ba 60 30 30 30 20 20 Be <5 <5 <5 <5 <5 <5 Bi <100 <100 <100 <100 <100 <100 Bo <100 <100 <100 <100 <100 <100 Cd <10 <10 <10 <10 <10 <10 Ca 107000 148000 29500 11100 6330 5320 Cr <10 <10 <10 <10 <10 <10 Co <10 <10 <10 <10 <10 <10 Cu 30 <10 <10 <10 <10 <10 Fe 110 <30 <30 <30 <30 <30 Pb <50 <50 <50 <50 <50 <50 Li 10 20 10 10 <10 <10 Mg 28800 52200 10100 4000 2400 1900 Mn 105 115 31 12 7 6 Mo <30 <30 <30 <30 <30 <30 Ni <50 <50 <50 <50 <50 <50 P <300 <300 <300 <300 <300 <300 K 3000 3000 <2000 2000 <2000 3000 Se 88 150 70 25 17 10 8 8 6 6 7 Si 600 2500 2220 2010 1610 1430 Ag <10 <10 <10 <10 <10 <10 Na <2000 <2000 <2000 <2000 <2000 <2000 Sr 148 223 52 23 14 11 Th <200 <200 <200 <200 <200 <200 Sn <30 <30 <30 <30 <30 <30 Ti <10 <10 <10 <10 <10 <10 V <30 <30 <30 <30 <30 <30 Zn 616 345 79 33 20 18 Organic Parameters TIC <500 <500 <500 <500 <500 <500 TOC 6300 2800 1500 2200 2100 <500 NOTE: Al l concentrations expressed in pg/L 171 Table A4.10 Continued wash 12 wash 13 wash 14 wash 15 wash 16 wash 17 wash 18 wash 19 wash 20 wash 21 Week 11 12 13 14 15 16 17 18 19 20 Eh 413.0 410.0 410.6 440.1 485.5 434.3 532.2 508.8 446.0 461.7 D H 6.59 6.37 6.08 7.55 6.46 7.12 6.30 6.39 6.11 5.91 K 1 1 Vol. (mL) 496 500 407 492 490 489 486 489 482 499 Dissolved A nions Acid. 1000 1000 1000 2000 Alk. 2000 1000 2000 2000 S0 4 2 < 22 000 23 000 14 000 9000 Dissolved N letals Al <200 <200 o o o o o o Sb <200 <200 o o o o o o As <200 <200 o o o o o o Ba 20 20 20 OO Be <5 <5 <5 o Bi <100 <100 o o o o o o Bo <100 <100 o o o o o o Cd <10 <10 o o o o Ca 5170 6510 3310 2440 Cr <10 <10 OO OO Co <10 <10 OO OO Cu <10 <10 OO OO Fe <30 <30 o o o o Pb <50 <50 o o o o Li <10 <10 o o o o Me 1800 1900 1100 900 Mn 6 7 <5 o Mo <30 <30 o o o o Ni <50 <50 o o o o P <300 o o o o o o o o o K <2000 oooo oooo oooo Se 8 8 8 8 7 7 6 6 4 4 Si 1290 1260 1430 1030 Aa <10 OO OO OO Na <2000 oooo oooo oooo Sr 12 15 9 <5 Th <200 o o o o o o o o o Sn <30 OO o o o o Ti <10 o o o o o o V <30 o o o o o o Zn 17 19 12 11 Organic P arameters TIC <500 o o o o o o o o o TOC 1800 2400 1900 1400 NOTE: Al l concentrations expressed in pg/L 172 10.5 APPENDIX 5 Calculation Acid-Generation and Neutralization Potential Acid-generation potential (AP) and neutralization potential (NP) are commonly used to predict the risk of acid rock drainage (Lawrence and Day, 1997). AP is a function of the amount of sulphides present in the material, NP is determined by the amount of acid-consuming minerals present in the material. Both AP and NP are expressed in terms of CaC0 3 equivalents/tonne of material. Each mole of S produces 2H +, which can be neutralized by 1 mole of CaC0 3. 1 g of S is therefore equivalent to 3.125 g of CaC0 3 and the acid generating potential of a given material can be determined using the following equation (Lawrence and Day, 1997): AP = % sulphide • 31.25 The contributions of the main mineral components (calcite, kaolinite, illite, quartz) to NP were calculated using the following formula from Lawrence and Scheske (1997): NP contribution = % weight of mineral . 1000 kg . molecular weight of calcite . relative reactivity 100 1 tonne molecular weight of mineral Relative reactivity rates were obtained from Kwong (1993) (Table A10.1), while mineral molecular weights were calculated based on chemical formulas from Deer (1993) (Table A2.1). All carbonates were assumed to be in the form of calcite. NP was calculated by summing the contributions of calcite, kaolinite, illite and quartz. Table A 5 .1 Relative Reactivity of Minerals at p H 5 (From Lawrence and Scheske, 1997) Mineral Group Typical Minerals Relative Reactivity at pH5 Dissolving calcite, dolomite, magnesite, aragonite, brucite 1.00 Fast weathering anorthite, olivine, garnet, diopside, wollastonite, jadeite, nepheline, leucite, spodumene 0.60 Intermediate weathering epidote, zoisite, enstatite, hedenbergite, augite, hypersthene, hornblende, tremolite, actinolite, serpentine, chrysotile, talc, chlorite, biotite 0.40 Slow weathering plagioclase feldspars, kaolinite, vermiculite, montmorillonite, gibbsite 0.02 Very slow weathering K-feldspars, illite 0.01 Inert quartz, rutile, zircon 0.004 A NP/AP ratio greater than 4:1 is indicative of an extremely low risk of acid-generation (Price, 1997). Materials with a NP/AP ratio between 2 and 4 are unlikely to generate acid, but could do so if sulphides were preferentially exposed along fracture planes or if they contained extremely reactive sulphides with insufficiently reactive NP minerals. The likelihood of acid-generation increases as the NP/AP ratio decreases, with a ratio of 1:1 or lower signalling a high risk of acid-generation. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0081121/manifest

Comment

Related Items