UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An investigation into the construction, excavation, and geochemical history of a waste rock dump and… Dagenais, Paul James 1996

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

Item Metadata

Download

Media
831-ubc_1997-0058.pdf [ 18.21MB ]
Metadata
JSON: 831-1.0081146.json
JSON-LD: 831-1.0081146-ld.json
RDF/XML (Pretty): 831-1.0081146-rdf.xml
RDF/JSON: 831-1.0081146-rdf.json
Turtle: 831-1.0081146-turtle.txt
N-Triples: 831-1.0081146-rdf-ntriples.txt
Original Record: 831-1.0081146-source.json
Full Text
831-1.0081146-fulltext.txt
Citation
831-1.0081146.ris

Full Text

A N I N V E S T I G A T I O N INTO T H E C O N S T R U C T I O N , E X C A V A T I O N , A N D G E O C H E M I C A L H I S T O R Y OF A W A S T E R O C K D U M P A N D IMPLICATIONS F O R L O N G - T E R M W A T E R Q U A L I T Y A T T H E I S L A N D C O P P E R M I N E , P O R T H A R D Y , BRITISH C O L U M B I A by P A U L J A M E S D A G E N A I S B . S c , The University of British Columbia, 1993 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R OF A P P L I E D S C I E N C E in T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Mining and Mineral Process Engineering) We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y OF BRITISH C O L U M B I A November 1996 © Paul Dagenais, 1996 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 hot be allowed without my written permission. Department of ^ V X ^ v ^ \ w c A ? r e c ^ S > ^ V v s v The University of British Columbia Vancouver, Canada Date V\o^. V& \<V\fe DE-6 (2788) ABSTRACT The Island Copper Mine, near Port Hardy, British Columbia, is currently in the process of closure. As part of its closure plans, the mine excavated a waste rock dump (the Northwest Dump) and disposed of it into the open pit. The open pit was flooded with seawater from Rupert Inlet and will become chemically meromictic once a freshwater cap becomes established on top of the seawater. The decision to excavate the waste rock dump and place it into the open pit was made because the dump was generating acid rock drainage. The excavation of the dump provided a unique opportunity to examine the interior of a waste rock pile that had been weathering for over ten years. This thesis examined three main questions: 1) was local groundwater contaminated by the acid rock drainage coming from the Northwest Dump?, 2) has excavation of the dump led to improvements in the quality of the local groundwater and nearby Francis Lake, and 3) will the material from the Northwest Dump have a noticeable effect on water quality in the flooded pit, specifically the freshwater cap? To answer these questions, samples were collected from various parts of the dump and subjected to numerous analyses. Samples were characterized in terms of particle size, mineralogy, and chemical composition. Since the rock in this dump had been weathering for so long, it contained large amounts of reaction products. Before meaningful results could be obtained from kinetic prediction tests, these reaction ii products had to be removed, otherwise the kinetic tests would have had to continue for many months. Shake flask tests were used to determine what type of solution would be most efficient at removing the reaction products from the samples. When the best solution was determined, the samples underwent a batch leach to remove the reaction products. After the samples came out of the batch leach, they were subjected to both static and kinetic prediction tests. These tests were used to characterize the samples in terms of their acid-generating and acid-consuming ability and to determine their oxidation, neutralization, and leaching rates. As a result of this testwork, the following conclusions were made: 1) based on the batch leach tests, significant levels of sulphate and metals would be produced by this material when it was first rinsed (either by precipitation or by seawater during flooding), 2) the static prediction tests indicated that nearly all of the dump material tested had a high potential to generate acid, 3) the kinetic tests indicated that all but three of the samples tested would be acidic from the time they were first exposed, (three of the samples remained neutral throughout kinetic testing with correspondingly low levels of sulphide oxidation and metal leaching), 4) local groundwater has been contaminated and acid rock drainage from the Northwest Dump is probably the primary source of this contamination, 5) excavation of the dump has not yet led to any noticeable improvements in the quality of the local groundwater or Francis Lake, and 6) the material from the dump could have a noticeable impact on the water quality in the flooded pit, specifically the freshwater cap (copper concentrations especially are likely to exceed the water quality guidelines of the Water Management Pond discharge permit within five years). Ul TABLE OF CONTENTS Abstract 1 1 Table of Contents iv List of Figures X 1 V List of Plates * v i i List of Tables xviii Acknowledgements x x 1. I N T R O D U C T I O N 1 1.1 Thesis Project 1 1.1.1 Topic 2 1.1.2 Objectives 2 1.2 Mine Location 3 1.3 Mine Setting 4 1.4 Geology 6 1.4.1 Local Geology 6 1.4.2 Island Copper Geology 10 1.4.3 Geology Related to Acid Rock Drainage 14 1.4.3.1 Acid-Generating Minerals 14 1.4.3.2 Acid-Consuming Minerals 15 1.5 Operating History 1 7 1.6 Closure Plans 18 iv 1.6.1 Open Pit 18 1.6.2 Treatment of Ac id Rock Drainage 19 1.6.3 Waste Rock Dumps 20 1.7 Ac id Rock Drainage 21 1.7.1 Chemistry of Acid Rock Drainage 21 1.7.1.1 Components of Acid Rock Drainage 22 1.7.1.2 Chemical Reactions of Ac id Generation 22 1.7.1.3 Factors Controlling Acid Generation 24 1.7.1.4 Stages in the Development of A c i d Generation 26 1.7.2 Ac id Rock Drainage Prediction Methods 29 1.7.2.1 Static Testing 30 1.7.2.2 Kinetic Testing 34 1.8 Waste Rock Dumps 37 1.8.1 General Characteristics 37 1.8.1.1 Construction Methods 37 1.8.1.2 Solids Transport 39 1.8.1.3 Gas Transport 40 1.8.1.4 Water Transport 43 1.8.1.5 Internal Chemistry 45 1.8.2 The Northwest Dump 46 1.8.2.1 Construction 46 1.8.2.2 Physical and Geochemical Characteristics of the Northwest Dump 53 V 1.8.2.3 Previous Research 55 1.9 Summary 60 2. E X P E R I M E N T A L M E T H O D S 61 2.1 Sampling Program 61 2.1.1 Sample Collection 61 2.1.2 Sample Preparation 68 2.1.2.1 Shake Flask Testing and Acid-Base Accounting 70 2.1.2.2 Batch Leach and Kinetic Testing 71 2.1.3 Sample Characterization 71 2.1.3.1 Particle Size Analysis 72 2.1.3.2 Thin Section Analysis 73 2.1.3.3 X-ray Diffraction 73 2.1.3.4 Multi-Element and Whole Rock Analysis 73 2.2 Investigation of Northwest Dump 74 2.3 Preliminary Testwork 75 2.3.1 Shake Flask Tests 75 2.3.1.1 Objectives 75 2.3.1.2 Methods and Materials 76 2.3.2 Batch Leach Tests 77 2.3.2.1 Objectives 77 2.3.2.2 Methods and Materials 78 2.4 Static Prediction Tests 79 vi 2.4.1 Acid-Generating Potential 79 2.4.1.1 Sulphur Mass Balance 80 2.4.1.2 Standard Acid-Base Accounting 82 2.4.1.3 Modified Acid-Base Accounting 83 2.4.2 Neutralization Potential 83 2.4.2.1 Paste p H 84 2.4.2.2 Standard Acid-Base Accounting 84 2.4.2.3 Modified Acid-Base Accounting 86 2.4.2.4 Carbonate C 0 2 Analysis 88 2.4.2.5 Calculations 90 2.5 Kinetic Prediction Tests 91 2.5.1 Objectives 91 2.5.2 Methods and Materials 91 2.5.2.1 Samples 91 2.5.2.2 Equipment and Procedure 93 2.5.2.3 Leachate Analysis and Calculations 100 2.5.3 Humidity Cell Disassembly 103 2.5.3.1 Objectives 103 2.5.3.2 Methods and Materials 103 2.6 Water Quality Analysis 105 2.6.1 Groundwater Quality 107 2.6.1.1 Objectives 107 vii 2.6.1.2 Methods and Materials 107 2.6.2 Water Quality Comparison 110 2.6.2.1 Objectives 110 2.6.2.2 Methods and Materials 110 2.6.3 Predicted Water Quality 111 2.6.3.1 Objectives I l l 2.6.3.2 Methods and Materials 112 3. R E S U L T S 113 3.1 Sample Characterization 113 3.1.1 Particle Size Analysis 113 3.1.2 Thin Section Analysis 115 3.1.3 X-ray Diffraction Analysis 116 3.1.4 Multi-Element and Whole Rock Analysis 117 3.2 Northwest Dump 120 3.2.1 Reanalysis of Previous Research 120 3.2.2 Excavation 124 3.2.2.1 Visit 1 125 3.2.2.2 Visit 2 125 3.2.2.3 Visit 3 128 3.3 Preliminary Testwork 128 3.3.1 Shake Flask Tests 130 3.3.2 Batch Leach 130 vi i i 3.4 Static Prediction Tests 131 3.4.1 Acid-Generating Potential 132 3.4.1.1 Sulphur Mass Balance 132 3.4.1.2 Standard Acid-Base Accounting 133 3.4.1.3 Modified Acid-Base Accounting 134 3.4.2 Neutralization Potential 136 3.4.2.1 Paste p H 136 3.4.2.2 Neutralization Potential Determination 137 3.4.3 Summary of Static Prediction Tests 140 3.5 Kinetic Prediction Tests 144 3.5.1 General Observations 144 3.5.2 p H 145 3.5.3 Conductivity 145 3.5.4 Alkalinity/Acidity 148 3.5.4.1 Alkalinity 148 3.5.4.2 Acidity 150 3.5.5 Sulphate 150 3.5.5.1 Cumulative Loading 150 3.5.5.2 Sulphide Oxidation Rate 153 3.5.5.3 Percent Sulphide Remaining 155 3.5.5.4 Time to Deplete A P 155 3.5.6 Molar Ratios 158 ix 3.5.6.1 Cumulative NP Produced 158 3.5.6.2 N P Depletion Rate 161 3.5.6.3 Percent NP Remaining 164 3.5.6.4 Time to Deplete N P 167 3.5.7 Zinc I 7 0 3.5.7.1 Cumulative Loading and Metal Leaching Rate 171 3.5.7.2 Percent Metal Remaining 174 3.5.7.3 Time to Deplete Metal 174 3.5.8 Copper 177 3.5.8.1 Cumulative Loading and Metal Leaching Rate 177 3.5.8.2 Percent Metal Remaining 180 3.5.8.3 Time to Deplete Metal 180 3.5.9 Other Relationships 183 3.5.9.1 Conductivity vs. Sulphate 183 3.5.9.2 Sulphate:Alkalinity or Acidity Ratio 184 3.5.9.3 Other Ratios 187 3.5.10 Humidity Cell Disassembly 189 3.5.10.1 Flowpaths.... 189 3.5.10.2 Mass Loss 190 3.6 Water Quality 190 3.6.1 Groundwater Quality 191 3.6.1.1 p H 192 X 3.6.1.2 Conductivity 192 3.6.1.3 Sulphate 194 3.6.1.4 Calcium 194 3.6.1.5 Zinc 196 3.6.1.6 Copper 196 3.6.2 Water Quality Comparison 198 3.6.2.1 Groundwater 198 3.6.2.2 Francis Lake 199 3.6.3 Predicted Water Quality 200 4. DISCUSSION 206 4.1 Sample Collection and Preparation 206 4.2 Sample Characterization 207 4.2.1 Particle Size Analysis 207 4.2.2 Thin Section and X R D Analyses 208 4.2.3 ICP and Whole Rock Analysis 209 4.3 Northwest Dump 209 4.4 Preliminary Testwork 210 4.4.1 Shake Flask Tests 210 4.4.2 Batch Leach 211 4.5 Static Prediction Tests 214 4.5.1 Acid-Generating Potential 214 4.5.1.1 Sulphur Mass Balance 214 xi 4.5.1.2 Standard Acid-Base Accounting 217 4.5.1.3 Modified Acid-Base Accounting 218 4.5.2 Neutralization Potential 218 4.5.2.1 Paste p H 218 4.5.2.2 Acid-Base Accounting 219 4.6 Kinetic Prediction Tests 221 4.6.1 General Observations 222 4.6.1.1 Procedural Problems 222 4.6.2 p H 223 4.6.3 Conductivity 224 4.6.4 Alkalinity/Acidity 225 4.6.4.1 Alkalinity 225 4.6.4.2 Acidity 225 4.6.5 Sulphate 226 4.6.5.1 Cumulative Loading. 226 4.6.5.2 Sulphide Oxidation Rate 228 4.6.5.3 Percent Sulphide Remaining and Time to Deplete Sulphide 231 4.6.6 Molar Ratios 232 4.6.6.1 Cumulative N P Produced 236 4.6.6.2 N P Depletion Rate 237 4.6.6.3 Percent NP Remaining 238 4.6.6.4 Time to Deplete N P 238 xii 4.6.7 Zinc 239 4.6.7.1 Cumulative Loading and Metal Leaching Rate 239 4.6.7.2 Percent Metal Remaining 241 4.6.7.3 Time to Deplete Metal 241 4.6.8 Copper 242 4.6.8.1 Cumulative Loading and Metal Leaching Rate 242 4.6.8.2 Percent Metal Remaining and Time to Deplete Metal 244 4.6.9 Other Parameters 245 4.6.9.1 Conductivity vs. Sulphate 245 4.6.9.2 Sulphate:Alkalinity or Acidity Ratio 245 4.6.10 Humidity Cell Disassembly 246 4.7 Water Quality... 247 4.7.1 Groundwater Quality 247 4.7.2 Water Quality Comparison 251 4.7.2.1 Groundwater 251 4.7.2.2 Francis Lake 252 4.7.3 Predicted Water Quality 253 5. C O N C L U S I O N S A N D R E C O M M E N D A T I O N S 256 5.1 Conclusions 256 5.2 Recommendations 258 R E F E R E N C E S 260 A P P E N D I C E S 268 xii i LIST OF FIGURES Figure 1-1 Location Map 5 Figure 1-2 Mine Site Schematic 7 Figure 1-3 Mine Site Watersheds 9 Figure 1-4 Geology of the Island Copper Deposit 11 Figure 1-5 Geology of the Island Copper Deposit (cross-section) 11 Figure 1-6 Alteration Types of the Island Copper Deposit 13 Figure 1-7 Alteration Types of the Island Copper Deposit (cross-section) 13 Figure 1-8 Oxidation Rate as Affected by Temperature 25 Figure 1-9 Oxidation Rate as Affected by p H 25 Figure 1-10 Effect of Buffering on p H 27 Figure 1-11 Stages and Associated Reactions of Ac id Generation 27 Figure 1-12 Original Ground Surface of Northwest Dump 48 Figure 1-13 Original Watercourses Beneath Northwest Dump 49 Figure 1-14 Sequential Construction of Northwest Dump 51 Figure 1-15 Final Dump Configuration 52 Figure 1-16 Location of Northwest Dump Drillholes 58 Figure 2-1 Northwest Dump - Visit 1 63 Figure 2-2 Northwest Dump - Visit 2 66 Figure 2-3 Flowchart used for Sulphur Mass Balance 81 Figure 2-4 Schematic Diagram of a Humidifier 95 XIV Figure 2-5 Schematic Drawing of a Humidity Cell 97 Figure 2-6 Schematic of a "Bubbler" 98 Figure 2-7 Location of Groundwater Wells 109 Figure 3-1 Cross-Section of Northwest Dump - Previous A B A Results 122 Figure 3-2 Cross-Section of Northwest Dump - Previous A B A Results 123 Figure 3-3 Locations of Active Dumping During Site Visits 126 Figure 3-4 Kinetic Test Data (pH) 146 Figure 3-5 Kinetic Test Data (Conductivity) 147 Figure 3-6 Kinetic Test Data (Alkalinity) 149 Figure 3-7 Kinetic Test Data (Acidity) 151 Figure 3-8a Kinetic Test Data (Cumulative Loading - Sulphate) 152 Figure 3-8b Kinetic Test Data (Sulphide Oxidation Rate) 154 Figure 3-8c Kinetic Test Data (% Sulphide Remaining) 156 Figure 3-8d Kinetic Test Data (Time to Deplete AP) 157 Figure 3-9a Kinetic Test Data (Carbonate Ratio) 159 Figure 3-9b Kinetic Test Data (Silicate Ratio) 160 Figure 3-10a Kinetic Test Data (NP Depletion Rate - Carbonate Ratio) 162 Figure 3-10b Kinetic Test Data (NP Depletion Rate - Silicate Ratio) 163 Figure 3-1 la Kinetic Test Data (% NP Remaining - Carbonate Ratio) 165 Figure 3-1 lb Kinetic Test Data (% N P Remaining - Silicate Ratio) 166 Figure 3-12a Kinetic Test Data (Time to Deplete NP - Carbonate Ratio) 168 Figure 3-12b Kinetic Test Data (Time to Deplete NP - Silicate Ratio) 169 XV Figure 3-13a Kinetic Test Data (Cumulative Loading - Zinc) 172 Figure 3-13b Kinetic Test Data (Leaching Rate - Zinc) 173 Figure 3-13c Kinetic Test Data (% Metal Remaining - Zinc) 175 Figure 3-13d Kinetic Test Data (Time to Deplete Metal - Zinc) 176 Figure 3-14a Kinetic Test Data (Cumulative Loading - Copper) 178 Figure 3-14b Kinetic Test Data (Leaching Rate - Copper) 179 Figure 3-14c Kinetic Test Data (% Metal Remaining - Copper) 181 Figure 3-14d Kinetic Test Data (Time to Deplete Metal - Copper) 182 Figure 3-15 Conductivity versus Sulphate 185 Figure 3-16 Kinetic Test Data (Sulphate:Alkalinity Ratio) 186 Figure 3-17 Sulphate:Acidity Ratio 188 Figure 3-18a Groundwater Data (pH) 193 Figure 3-18b Groundwater Data (Conductivity) 193 Figure 3-18c Groundwater Data (Sulphate) 195 Figure 3-18d Groundwater Data (Calcium) 195 Figure 3-18e Groundwater Data (Zinc) 197 Figure 3-18f Groundwater Data (Copper) 197 Figure 4-1 Batch Leach Data (Calcium:Sulphate Molar Ratio) 212 Figure 4-2 Kinetic Test Data (Sulphate:Calcium Ratio vs Sulphate Concentration) 229 XVI LIST OF PLATES Plate 1-1 Air Photo of Island Copper Mine 8 Plate 1-2 Air Photo of Northwest Dump 47 Plate 1-3 Air Photo of Francis Lake and Northwest Dump 47 Plate 2-1 Exposed Face of Northern Section of Northwest Dump - Visit 1 64 Plate 3-1 Talus Slope in Northwest Corner of Pit 127 Plate 3-2 Oxidation Zone in Northwest Dump 129 Plate 3-3 Talus Slope in Pit 129 Plate 3-4 Northwest Dump Material on Upper Benches of Pit After Flooding 201 XVll L I S T O F T A B L E S Table 1-1 Construction of Northwest Dump (after L i , 1990) 50 Table 1-2 Typical Surface Drainage Quality Prior to Excavation (March-July, 1995) 54 Table 1-3 Typical Groundwater Quality Prior to Excavation (May-June, 1995) 55 Table 2-1 Mass of Sample in Humidity Cell 93 Table 2-2 Name of Well and Probable Source of Groundwater 110 Table 3-1 Particle Size Analysis (% passing 3.35 mm) 114 Table 3-2 Surface Area of Humidity Cell Material 114 Table 3-3 Results of Thin Section Analyses 116 Table 3-4 Summary of X R D Analysis 117 Table 3-5 ICP Analysis (Significant Metals) 118 Table 3-6 Whole Rock Analysis 119 Table 3-7 Sulphur Mass Balance 132 Table 3-8 Acid-generating Potential (Standard Method) 134 Table 3-9 Acid-generating Potential (Modified Method) 135 Table 3-10 Paste p H 137 Table 3-11 Summary of Neutralization Potential Determination 138 Table 3-12 Summary of Static Prediction Tests 141 Table 3-13 Depletion Rate, % NP Remaining and Time to Deplete NP 143 Table 3-14 Sample Mass (Kinetic Testing) 190 Table 3-15 Comparison of Groundwater Data 198 Table 3-16 Comparison of Francis Lake Data 199 Table 3-17 Assumptions of Mass Balance Calculation 203 Table 3-18 Metal Loading of Dump Material 204 Table 3-19 Resulting Water Quality of Freshwater Cap 204 xix ACKNOWLEDGEMENTS I would first like to thank my supervisor, Dr. George Poling, for his guidance and friendship. He was always ready to help and his support is much appreciated. I am also grateful to Dr. Richard Lawrence for helping with the interpretation of the data I collected. I would also like to thank B H P Minerals Canada Ltd., Island Copper Mine for their generous funding and in particular Ian Home, Steve Lacasse, and Al l an Rowden whose assistance made this project possible. I extend my thanks to the former Ministry of Energy, Mines, and Petroleum Resources, particularly John Errington, for a grant which went a long way towards covering the cost of my analytical work. Thanks goes out to the Department of Min ing and Mineral Process Engineering for the support I received from the faculty, staff, and my fellow students. I could always find answers to my questions. I also wish to thank Shannon Shaw from the Department of Earth and Ocean Sciences for performing some X R D analysis. Most of al l , I would like to thank my parents. Their support has made possible everything I have accomplished and everything I hope to accomplish. I could not have done it without them. X X 1. INTRODUCTION Acid rock drainage is a serious environmental issue facing the mining industry worldwide. It needs to be addressed at all stages in mine development, from exploration to closure. The presence or absence of acid rock drainage can alone determine if a mine will be economically viable to operate. The Island Copper Mine has recently closed after having been in operation for the past 25 years. Like all mines in British Columbia, Island Copper is committed to restoring any land disturbed by mining activities to a level of productivity equal to or greater than that which existed prior to mining. Towards this end, the mine prepared for closure by decommissioning the plant site, reclaiming the land and beach dumps and flooding the open pit. As part of its closure plan, the mine must address acid rock drainage that is being generated by some of the waste rock dumps. One relatively small dump, the Northwest Dump, was producing acid rock drainage before it was excavated and placed back into the open pit. This thesis examines the construction, excavation, and geochemical history of the Northwest Dump, the effect the dump material may have on water quality in the flooded open pit and the effect the dump had on local groundwater. •1.1 Thesis Project A brief review of the thesis topic is provided and the objectives outlined for the research program are discussed. A n outline of the project is given at the end of the section. l 1.1.1 Topic The Northwest Dump at the Island Copper Mine has been generating acid rock drainage for a number of years. A decision was made to excavate the dump and place it back into the open pit to protect the Stephens Creek watershed. It was believed that the aqueous cover provided by the flooded pit at closure would prevent any further oxidation of the dump material. The excavation of the dump provided a rare opportunity to examine the inner workings of an acid-generating waste rock dump. This thesis will examine the construction, geochemical history, excavation, and remediation of the Northwest Dump. 1.1.2 Objectives The overall objective of this project is to evaluate the impact of the Northwest Dump on water quality at the Island Copper Mine. Two main areas will be explored: i) the potential impact of dump material on water quality in the flooded pit, and ii) the impact acid rock drainage from the dump had on local groundwater. The research conducted will attempt to provide answers to the following questions: 1) Did acid rock drainage from the Northwest Dump contaminate local groundwater? 2) Has excavation of the Northwest Dump led to improvements in the quality of ground-water and Francis Lake? 3) Wil l the generation of acid rock drainage from the Northwest Dump material have a significant effect on water quality in the open pit? If so, for how long and to what degree? 2 A n additional aspect of the project is the evaluation of a dike that separates Twin Lakes from the open pit. It is believed that the dike is generating acid and releasing metals into Twin Lakes. The dike will be characterized in terms of its acid-generating potential and its ability to release heavy metals. In order to achieve the research objective, work was performed in the following three areas: i) static prediction testing (acid-base accounting), ii) kinetic prediction testing (humidity cells), and iii) statistical analysis of water quality data. The thesis has been divided into five main sections. Section 1 provides introductory information on the Island Copper Mine, acid rock drainage, and waste rock dumps. The methodology followed in conducting the research is described in Section 2. Sample collection, preparation, and characterization are explained, the different tests that were performed (shake flask tests, batch leach tests, acid-base accounting, and humidity cell tests) are discussed, and the method used to analyze the water quality data is described. Section 3 presents the results that were obtained over the course of the research. A detailed discussion of these results is provided in Section 4 and conclusions and recommendations are presented in Section 5. 1.2 Mine Location The Island Copper Mine is situated on the north shore of Rupert Inlet ( 5 0 ° 3 6 ' N , 1 2 7 ° 2 8 ' W ) , approximately 16 km south of Port Hardy on Vancouver Island, British 3 Columbia (Figure 1-1). Rupert Inlet is part of the Quatsino Sound fjord system which opens up to the Pacific Ocean on the west side of Vancouver Island. Access to the mine is by paved road from Port Hardy or by sea from the Pacific Ocean through Quatsino Narrows. There is also a municipal airport located approximately 12 km southeast of Port Hardy. 1.3 Mine Setting The mine is situated in the Submontane Very Wet Maritime Coastal Western Hemlock Variant ( C W H v m l ) of the Coastal Western Hemlock Biogeoclimatic Zone (Green and Klinka in BHP, 1996). The local environment is characterized by low hills, up to 150 meters above sea level, which are covered by overburden composed of moss, peat, colluvium, and glacial till. The overburden is from 1 to 20 m in depth, but can occur locally up to 75 m deep (Lister, 1994; Young and Rugg, 1971; Perello et al., 1995). Average annual precipitation in the Port Hardy area is approximately 1780 millimeters. Over 50% of this precipitation occurs between October and January, which have monthly averages of 244 millimeters of precipitation. Only 76 millimeters of the annual precipitation occurs as snowfall which does not accumulate for more than a few weeks per year. The driest months of the year are from May to August, with monthly precipitation of approximately 65 millimeters. The average annual temperature is 8 ° C , with temperatures ranging from -7°C to 2 7 ° C (Lister, 1994). 5 The mine (Figure 1-2, Plate 1-1) is situated within three separate watersheds (Figure 1-3). Two of these watersheds, the End Creek (approximately 3 km 2 ) and Trey Creek (approximately 1.5 km 2 ) watersheds, are considered historic because they have been permanently altered by the mining operation. The third watershed, the Stephens Creek watershed, is much larger and only marginally affected by the mine. The Northwest Dump occupied a small fraction (<1%) of this watershed. This dump was producing acid rock drainage which may have had an impact on Francis Lake, an important salmon fry rearing habitat in the Stephens Creek watershed. 1.4 Geology A general description of the local geology of the Island Copper deposit is provided along with a more in-depth review of the mine's geology that is relevant to acid rock drainage. 1.4.1 Local Geology Rocks of the Bonanza and Vancouver groups dominate the local stratigraphy. The Karmutsen Formation is evenly covered by the Quatsino Formation, which slopes upwards into the Parson Bay Formation. The Parson Bay Formation can be divided into an upper non-calcareous fine-grained siliciclastic sequence and a lower calcareous sequence in the region around the mine (Perello et al., 1995; Koyanagi and Panteleyev, 1992). 6 7 8 Close to the Island Copper mine, basal pyroclastic-epiclastic sequences of the Bonanza Group are more than 1000 m thick and contain interbedded fossiliferous and tuffaceous cherty sediments, dacitic tuffs and basaltic and andesitic flows. The End Creek Fault is an important structure in and around the mine and it trends 3 0 0 ° for over 3 km. In the area of the open pit this fault, consisting of a tabular, steeply northeast-dipping zone, parallels the trend of the rhyodacite porphyries (Perello et al., 1995). 1.4.2 Island Copper Geology The three main rock types found in the Island Copper pit are Bonanza Group volcanics, rhyodacite porphyries, and hydrothermal breccias (Figures 1-4 and 1-5). The Bonanza Group rocks consist of lithic tuff, breccia, and interbedded basaltic and andesitic flows. Flow mineralogy includes augite, amphibole, hypersthene, and plagioclase of labradorite-bytownite composition as well as calc-alkaline, high-alumina basalts. The rhyodacite porphyry exposed in the open pit can be divided into three phases: Main, Intra-mineral and Late-mineral. The Main porphyry is dike-like and contains the most intense alteration and mineralization of the porphyry phases. It is characterized by intense magnetite-quartz stockwork, banded quartz-magnetite veins and lesser quartz-pyri te±chalcopyri te-molybdenite-magnet i te veins. The Intra-mineral porphyry contains abundant quartz-pyri te±chalcopyri te±molybdeni te veinlets and minor quartz-magnetite veins. Late-mineral porphyry lacks the intense alteration, sulphide mineralization and magnetite - quartz stockwork of the Main and/or Intra - mineral 10 Figure 1-4 : Geology of the Island Copper Deposit (after Perello et al., 1995) k ^ l Bedded Tuff in Bonanza Figure 1-5 : Geology of the Island Copper Deposit (after Perello et al., 1995) porphyry bodies. Breccias are volumetrically significant and two types, Marginal and Pyrophyllite, are recognized (Perello et al., 1995). Several types of alteration have affected the host volcanics at Island Copper (Figures 1-6 and 1-7). Young and Rugg (1971) list them in order of decreasing intensity as: 1) silicification, 2) argillization, 3) saussurization, and 4) biotitization. Perello et al. (1995) describe them as early stage alteration, intermediate stage alteration, and late stage alteration. Early stage alteration occurred during the Main porphyry intrusion and formed four alteration zones: stockworked quartz-amphibole-magnetite core, biotite-magnetite, ch lor i te±magnet i te and epidote. Intermediate stage alteration is present in two forms: pervasive and structurally-controlled quartz-sericite-pyrite, and pervasive sericite-clay (illicite)- chlorite pyrite (SCC alteration). This stage is characterized by various combinations of quartz, sericite, kaolinite, illitic clays and chlorite accompanied by pyrite, molybdenite and minor chalcopyrite. The Late stage alteration is restricted to the Pyrophyllite breccia. It is advanced argillic and is dominated by pyrophyllite, quartz, sericite, kaolinite and dumortierite (Perello et a l , 1995). Other alteration stages, which occurred later and at lower temperatures, include ankerite-calcite in fracture-related veins, zeolites, and precipitation of remobilized carbon-bearing organics of gilsonite type (Fleming, 1983). 12 Figure 1-6 : Alteration Types of the Island Copper Deposit (after Perello et al., 1995) SW End Creek N E Proposed Ultimate Pit ^ 4 0 B e n c h _ (380m below sea level) 400 Feet (-120 m) Pyrophyllite ± Dumortierite Quartz - Sericite ± Chlorite Quartz - Amphibole - Magnetite Biotite - Magnetite Chlorite ± Magnetite Epidote Figure 1-7 : Alteration Types of the Island Copper Deposit (after Perello et al., 1995) 13 1.4.3 Geology Related to Acid Rock Drainage In terms of acid rock drainage, there are two main rock types of interest: acid-generating minerals such as sulphides (eg. pyrite, chalcopyrite) and acid-consuming minerals such as carbonates (eg. calcite, dolomite). Acid rock drainage itself is discussed in more detail in Section 1.6. The geology relevant to acid rock drainage at the Island Copper Mine is described below. 1.4.3.1 Acid-Generating Minerals The major acid-generating minerals found at Island Copper are sulphides. In decreasing order of abundance the main sulphides are pyrite, chalcopyrite, and molybdenite with minor amounts of bornite and sphalerite (Cargill, 1975; Perello et al., 1995). Chalcopyrite and molybdenite are the two minerals which were economically recovered during mining operations. Copper introduction occurred during one primary stage, which deposited quartz-chalcopyrite - magnetite - pyrite, quartz - chalcopyrite - pyrite - biotite ± K-feldspar and chalcopyrite-pyrite ± magnetite ± molybdenite veinlets that lacked alteration selvages (Perello et al., 1995). Chalcopyrite occurs as thin (0.1 mm) veinlets and disseminations on fractures and slip surfaces. Minor amounts of chalcopyrite occur with molybdenite on slip surfaces and with sphalerite in late stage carbonate-zeolite veins (Cargill, 1975). Molybdenum introduction followed the copper stage in quartz-molybdenite-pyrite-chalcopyrite veinlets with distinct quartz-sericite haloes and, more importantly, in 14 molybdenite-pyrite-chalcopyrite veinlets. Molybdenite occurs mainly along veinlet margins and on slip surfaces (Perello et al., 1995). The most common sulphide mineral at Island Copper is pyrite. Pyrite is present in both finely disseminated form and as thin, randomly oriented seams. It is located within and adjacent to the orebody with an average concentration of 2-5 %, but occurs locally in concentrations up to 15-20 % (Young and Rugg, 1971). Within the orebody, pyrite is associated with molybdenite and chalcopyrite in veinlets and is also found within chloritized mafic minerals. Within waste rock, fine-grained (1 mm) pyrite is disseminated throughout the porphyry dike. It also occurs as a disseminated secondary alteration mineral with chlorite in both the early stage chlorite-magnetite and intermediate stage sericite-chlorite-clay (SCC) alterations in all lithological units (Cargill, 1975; Perello et al., 1995). Pyrite is also found as relatively coarse grains (2 mm) with sphalerite in late stage carbonate-zeolite veins (Cargill, 1975). 1.4.3.2 Acid-Consuming Minerals Acid-consuming minerals such as carbonates are important with respect to acid rock drainage because they neutralize acid that is generated by sulphide oxidation. Other minerals ofimportance to A R D that occur at Island Copper include silicates (Kwong, 1994; Sherlock et al., 1995) and zeolites (Vos and O'Hearn, 1993). Calcite is the most common carbonate found at Island Copper. It occurs in a wide range of forms, from very thin veinlets less than 0.5 mm in width, up to relatively thick veins of 1-2 cm. It is also found as irregular alteration patches with sericite, and appears 15 to be more prevalent in the less altered rocks. Although uncommon, dolomite and ankerite have also been observed at Island Copper (Lister, 1994). Kwong (1994) indicated that fast-weathering (eg. olivine) and intermediate-weathering (eg. biotite) silicates may provide significant acid neutralization. Sherlock et al. (1995) also report that acid neutralization by silicates could be an important factor with respect to acid rock drainage. O f the minerals mentioned by Kwong, the ones occurring in substantial quantities at Island Copper include epidote, pyroxene minerals, chlorite, and biotite. Kwong (1994) also mentions anorthite, a fast-weathering silicate. Although anorthite has not been identified at Island Copper, both labradorite and bytownite (slightly more sodic feldspars) are relatively common and occur in weakly altered or unaltered Bonanza Group volcanics (Lister, 1994; Perello et al., 1995). Although original feldspar contents are estimated to be 45-55 % in Bonanza Group volcanics and 25-45 % in rhyodacite porphyries (BHP-Utah and Rescan, 1988), much of the primary plagioclase has been replaced by sericite or sericite-clay (Leitch in Lister, 1994). Therefore, because of their replacement by secondary minerals, feldspars are not considered to be significant in terms of acid neutralization (Lister, 1994). Epidote, pyroxenes, and biotite, although locally common, account for less than 10 % of the rock at Island Copper and probably do not contribute to acid neutralization (Lister, 1994). Chlorite minerals are likely the most significant of the silicates described 16 by Kwong (1994) that contribute to acid neutralization at Island Copper. Chlorite minerals are found in most alteration zones and as a replacement in primary mafic minerals (Cargill, 1975). Significant amounts of chlorite (greater than 15 %) occur in the Bonanza Group volcanics and some is also found in the rhyodacite porphyry (Cargill, 1975; Leitch in Lister, 1994). 1.5 Operating History Mining began in the Port Hardy area as far back as 1835 with the extraction of coal at Fort Rupert, just south of Port Hardy, and at Coal Harbour, situated on the north shore of Holberg Inlet. From 1835 to 1965 intermittent exploration took place and a few mines actually went into production. The Coast Copper, Yreka, and Empire Development mines were all operating mines in the region, but have since ceased operations (Young and Rugg, 1971). In 1965, Gordon Milbourne, a local prospector, discovered massive chalcopyrite in the area of Francis Lake. A n agreement between Mr. Milbourne and the Utah Construction and Mining Co. , the original owners of the mine, was signed in January, 1966. Diamond drilling commenced shortly thereafter, and by May, 1969 a total of 128 drillholes totaling 35,595 m had been completed to define the Island Copper deposit (Young and Rugg, 1971; Perello et al., 1995). 17 Production at the Island Copper Mine began in October, 1971 and the mine closed at the end of August, 1996. Over the mine life, more than 1 billion tonnes of ore and waste were extracted. Over 345 million tonnes of mill feed with average head grades of 0.41% Cu, 0.017% M o , 0.19 g/t A u , and 1.4 g/t A g yielded approximately 1300 million kg of copper, 31 million kg of molybdenum, 31.7 million grams of gold, and 336 million grams of silver (BHP, 1996). 1.6 Closure Plans Under the closure plan (BHP, 1994), the plant site was to be dismantled and all buildings leveled to concrete. The land dumps are currently being recontoured and revegetated and acid rock drainage generated by the North Dump will be collected and injected into the bottom of the pit. The pit itself has been flooded with seawater and is being capped with a freshwater layer to produce a meromictic lake. The bottom layer of seawater will become anoxic and the presence of sulphate-reducing bacteria will act to passively treat the acid rock drainage being injected from the land dumps. This section will provide information on the closure activities planned for the pit and the land dumps. 1.6.1 OPEN PIT The open pit at Island Copper is the most visible feature of the mine and its reclamation is the key issue in the mine's closure plans. The pit occupies an area of 215 hectares and has been excavated to a depth of 400 m. This translates into a volume of 287 million cubic meters (m3). The pit was flooded with seawater by excavating a 18 channel from Rupert Inlet to the southwest corner of the open pit. Flooding occurred from June 15 - July 23, 1996. A freshwater cap will develop and be maintained on the surface of the lake through precipitation and runoff from the surrounding watersheds. 1.6.2 Treatment of Ac id Rock Drainage Acid rock drainage (ARD) is one of the most important environmental issues concerning Island Copper. Ac id rock drainage has been produced by the North Dump (76.4 million tonnes) and the Northwest Dump (1.0 million tonnes). Due to its small size, however, the Northwest Dump was excavated and placed back into the pit. Drainage from the North Dump will be collected and injected to the bottom of the pit where it is believed the A R D will be passively treated by sulphate-reducing bacteria (SRB). Fish offal was dumped into the pit prior to flooding to promote the establishment and growth of a SRB colony. Passive treatment of the A R D will be accomplished in three ways: dilution, neutralization, and precipitation. The large volume of water in the pit compared to that of the drainage from the dump will dilute the contaminants. The relatively alkaline seawater will neutralize the A R D as it enters the pit. Acidity will also be neutralized because the SRB produce bicarbonate ions that will react with hydrogen ions from the A R D , thereby increasing the pH. In addition to dilution and neutralization, sulphide ions produced by the SRB will react with metal ions from the A R D producing insoluble metal sulphides which will precipitate out of solution. 19 1.6.3 Waste Rock Dumps Island Copper excavated approximately 1.09 billion tonnes of rock out of the open pit. The majority of this material was waste rock which was placed either into Rupert Inlet or on land. There is one ocean dump, the Beach Dump and three land dumps: the North, West, and South Dumps. Due to the production of acid rock drainage and its potential impact on Francis Lake, the Northwest Dump was excavated and placed back into the pit prior to flooding. Excavation of the Northwest Dump started in August, 1995 and was completed by December, 1995. The three remaining land dumps at I C M (North, West, and South) contain approximately 14% of the waste produced during operation and they cover approximately 193 hectares of land. Reclamation of the land dumps has been ongoing since the late 1970s. By the end of 1994, a total of 170 hectares of the land dumps had been reclaimed (recontoured, seeded/planted, and fertilized). Five more hectares were reclaimed in 1995 and a grand total of 187 of the 193 hectares will have been reclaimed by 1996. Dump surfaces that have not yet been reclaimed will be recontoured to safe working slopes (30°) and topped with 0.5 m of glacial till to act as a growth medium. The plan is to establish grass and legume species to reduce soil erosion from the dumps. Hand and hydroseeding have been the usual way for seeding although broadcast seeding and harrowing were used on the wide open, flat surfaces of the dump tops. Once a healthy and robust ground vegetation community has been established, trees will be planted as the final step in land dump reclamation. Local colonizing species such as red alder and lodgepole pine will be planted on the reclaimed sites. Climax species such as Western hemlock and Western red cedar will become established naturally. Based on observations of previously reclaimed areas at the mine, spruce, hemlock and 20 natural ground cover (eg. salal, huckleberry, salmonberry) will also become established on these sites. 1.7 Acid Rock Drainage Acid rock drainage (ARD) is the "contaminated drainage resulting from the oxidation and leaching of sulphide-bearing rocks when exposed to air and water" (Lawrence and Robertson, 1994). It is a serious environmental concern which can have significant repercussions, both economically and environmentally, for any mining operation. This section will examine the chemistry involved in acid generation and the common methods used for predicting acid rock drainage. 1.7.1 Chemistry of Acid Rock Drainage Acid generation, and its subsequent transport, involves very complex chemistry. The chemical reactions and interactions of A R D have been described in great detail in a large number of studies (Sherlock, 1995; Sherlock et al., 1995; Morin and Hutt, 1994; Morin et al., 1995; Morin et al., 1991; SRK, 1989). A detailed description of A R D is beyond the scope of this thesis, but a brief discussion of the components, reactions, and stages of acid generation will be provided. This review will provide the necessary background to allow a more rigorous discussion of the results obtained from the testwork carried out over the course of this thesis. 1.7.1.1 Components of Ac id Rock Drainage There are four components required for the generation of acid rock drainage: • Sulphide-bearing minerals - eg. pyrite, pyrrhotite • Air - specifically oxygen • Water - from infiltration or groundwater - both a component and a transport mechanism • Bacteria - primarily Thiobacillus ferrooxidans A l l four of these components must be present before environmentally significant quantities of acid will be generated. The presence of all four components does not necessarily mean that acid generation will occur, however. For example, the sulphide may be in a non-reactive form thus preventing the generation of acid. 1.7.1.2 Chemical Reactions of Ac id Generation The main oxidizing reactions of acid generation, using pyrite as the primary sulphide, are as follows (SRK, 1992): FeS 2 + 7/20 2 + H 2 0 = F e 2 + + 2S0 4 2 " + 2 H + (1.1) 2Fe' + + l / 2 0 2 + 2 F f = 2Fe 3 + + H 2 0 (1.2) 14Fe 3 + + FeS 2 + 8 H 2 0 = 15Fe 2 + + 2S0 4 2 " + 16H + (1.3) 22 Ferric iron (Fe ), generated in the second reaction, is an oxidant and will oxidize FeS 2 , producing F e 2 + (ferrous iron) which is then available to continue the cycle. These reactions are exothermic and so much heat is produced by them that waste rock dumps can stay free of snow over the winter. Thiobacillus ferrooxidans catalyzes all three of these reactions at all p H levels and is considered the most important species. Other bacteria, such as Sulfolobus and T. thioperus, catalyze only one type of reaction (eg. sulphur or iron oxidation), and therefore are not usually considered as important as T\ ferrooxidans. Once acid has been generated, a number of other chemical reactions can occur as the acid moves along flowpaths (Lawrence and Robertson, 1994). These reactions include (Morin et al., 1991; Ritchie, 1994a,b; SRK, 1989; SRK, 1992; Stumm and Morgan, 1981; Alpers et al., 1994; Blowes and Ptacek, 1994): • ACID NEUTRALIZATION H + + C a C 0 3 + S0 4 2 ~ + H 2 0 = C a S 0 4 » 2 H 2 0 + H C 0 3 ~ (1.4) Buffering of acid that occurs as the acid encounters neutralizing minerals, such as calcite (CaC0 3 ) , can result in the formation of gypsum ( C a S 0 4 » 2 H 2 0 ) which precipitates out of solution. • METAL LEACHING M S + 0 2 + H + ( o r F e 3 + ) = metals in solution (1.5) Metals in the surrounding rock are leached into solution and can be carried away by infiltrating rainwater or groundwater into the environment. 23 • Iron hydrolysis F e 3 + + 3 H 2 0 = Fe(OH) 3 + 3 H + (above p H 4) (1.6) 3Fe 2 (S0 4 ) 3 +14H 2 0 = 2(H 30)Fe 3(S04) 2(OH) 6+5H 2S04 (below p H 4) (1.7) The iron hydroxide (Fe(OH) 3) that is produced precipitates out and results in the iron staining that is so indicative of A R D . 1.7.1.3 Factors Controlling Acid Generation The rate of acid generation is influenced by a number of factors. The two most important ones are temperature and p H (Lawrence and Robertson, 1994), but other factors include oxygen availability, carbon dioxide availability (used as a carbon source for the bacteria), nutrient availability (nitrogen and phosphorus), and the surface area of exposed sulphide minerals (SRK, 1992). Temperature and p H are considered the two most important factors, and are examined in more detail. Temperature At extremely high temperatures, chemical oxidation is usually the fastest type of reaction. At natural temperatures, biological oxidation is usually the predominant form. Biological oxidation has an optimum temperature at which the oxidation rate is the highest. Above and below this optimum temperature, oxidation will occur at slower rates (Figure 1-8). The optimum temperature for T. ferrooxidans is approximately 35 °C. 24 1 / Biological X ^ ^ / ^ Jxdlc _ N / Chemical 0 0 Temperature (C) 70 Figure 1-8: Oxidation Rate as Affected by Temperature (after Lawrence and Robertson, 1994) 0 p H 7 Figure 1-9: Oxidation Rate as Affected by pH (after Lawrence and Robertson, 1994) pH Sulphide oxidation and leaching occurs both chemically and biologically. The chemical reactions are very slow and do not show any significant differences in rate with changes in pH. At neutral pH, the biological reactions occur at very slow rates. As the p H drops, however, the biological reactions occur much faster and at low p H (<3) it is the biological reactions that dominate the kinetics of acid generation. At high p H (>6), chemical oxidation occurs at a faster rate than biological oxidation. As the p H drops to approximately 4.5, the oxidation of ferrous iron becomes very slow, so slow in fact that biological oxidation becomes the predominant form. As p H continues to decrease, biological rates of oxidation continue to increase, until they reach an optimum pH. Above and below this pH, oxidation occurs at slower rates (Figure 1-9). The optimum p H for T. ferrooxidans is approximately 2. The smooth curve seen in Figure 1-9 above represents the reaction kinetics of biological oxidation as it is affected by changes in pH. The curve does not relate to actual p H changes in nature. During acid generation, pH will not drop at a steady rate over time. Different minerals (eg. carbonates, silicates) have different neutralizing capacities and will act to buffer the p H at various levels until the minerals have been depleted (Figure 1-10). 1.7.1.4 Stages in the Development of Acid Generation Acid is generated in a series of steps that involve both chemical and biological oxidation reactions. In general, acid generation is considered to be a 3 - stage process 26 p H plateaus result from buffering by different minerals Time Figure 1-10: Effect of Buffering on pH (after Lawrence and Robertson, 1994) 8-r Reactions in Stages I and II FeS2(s)+7/202+H20 — • Fe2++2S042"+2Ff ¥e2++l/402+W — • Fe3 ++1/2H20 Fe 3 ++3H 20 Fe(OH)3(s)+3Ff Time Figure 1-11: Stages and Associated Reactions of Acid Generation (SRK, 1992) (Figure 1-11) and is controlled by the p H of the water around the sulphide minerals (SRK, 1992). The p H plateaus are a result of the acid getting buffered when it encounters minerals with a neutralizing capacity. Stage I This stage covers the initial exposure of the sulphide minerals when the p H of the drainage is still neutral to the time when the primary neutralizing minerals (eg. calcite) are depleted. This is the period when chemical oxidation is dominant. The overall drainage at this stage is determined by the buffering capacity of the surrounding rock. The drainage usually contains raised levels of sulphate and the acid that is produced gets buffered while ferric iron precipitates out as a hydroxide. The drainage is usually at near neutral p H until the neutralizing minerals are consumed and the p H begins to decrease. Stage II This stage covers the time when the drainage is partially buffered by secondary neutralizing minerals (eg. gibbsite) to the time that these minerals are consumed and the p H again decreases. At this pH, both chemical and biological oxidation occur. The overall drainage at this stage is close to neutral (slightly below) and has high levels of ferrous iron and sulphate. The acidity is relatively high, but there are low concentrations of metal in solution. 28 Stage III At this stage of the process, either the neutralizing minerals have all been consumed or acidity is being generated faster than alkalinity. The dominant reactions change from chemical oxidation to biological oxidation. Ferrous iron is produced from sulphide oxidation, biologically oxidized to ferric iron, and ferric iron then becomes the primary oxidant in the reaction replacing oxygen. The drainage at this stage is usually acidic, has increased levels of sulphate and metals in solution, and iron occurs mainly in the ferric state. 1.7.2 ACID ROCK DRAINAGE PREDICTION METHODS The purpose of A R D prediction is to determine the overall acid-generating potential of specific geologic material, usually waste rock or tailings, and to determine the effects, both short-term and long-term, this material may have on drainage quality. Prediction of A R D occurs during all stages of a mining operation, from exploration through to closure. A wide variety of methods exist for the purpose of predicting A R D , including (SRK, 1989): • geographical/environmental comparisons - comparing mines with similar locations or climates • geological comparisons - comparing mines with similar geology • static tests • kinetic tests • mathematical modelling 29 The most commonly used methods are the static and kinetic tests. Both of these methods were employed in this thesis and are described in more detail below. 1.7.2.1 Static Testing Static methods are laboratory tests used to determine the difference between the overall acid-generating potential and acid-neutralizing potential of a sample. A detailed description of the various tests is provided in Coastech Research Inc. (1991). The most common static tests include: • Standard acid-base accounting • Modified acid-base accounting • B C Research Initial test • Alkaline Production Potential:Sulphur (APP:S) ratio • Net acid production test (NAP) Sobek et al. (1978) are usually credited with developing the first widely used static test, the standard acid-base accounting (ABA) . The other tests listed above were developed to account for perceived and/or identified shortcomings of the standard A B A (Sherlock, 1995). Standard A B A and modified A B A are the two most commonly used static tests in British Columbia. 30 Standard and Modified Acid-Base Accounting In general, the above tests attempt to determine the overall acid-generating potential (AP) and the overall neutralizing potential (NP) of a sample. The A P is typically calculated by assaying the sample for sulphur. The sulphur concentration is then multiplied by 31.25 to convert the value into equivalent kilograms of calcite per tonne of material (kg CaC0 3 / t ) . This determination of A P is one of two areas where the standard and modified A B A methods differ. The standard A B A uses the total sulphur content and assumes that (Sobek et al., 1978): • all sulphur is present as sulphide • iron sulphide completely oxidizes to sulphate and ferric iron • ferric iron precipitates as iron hydroxide The modified A B A method takes into account the fact that sulphate may already be present in the sample. Sulphate will not contribute to the acid-generating potential of the sample and therefore, the A P is calculated using the sulphide content of the sample rather than the total sulphur content. The modified method assumes that the sulphide content of the sample can be adequately determined by subtracting sulphate from total sulphur. If barite is present in the sample, however, it is necessary to estimate its content because barite is insoluble in hydrochloric acid meaning barium sulphate will not be accounted for in the analysis. Barium sulphate does not contribute to the A P and if it is not accounted for, the total sulphur content will be higher than it should be, resulting in a higher AP. 31 The other area where the standard and modified A B A methods differ is in the determination of the neutralization potential (NP). Both methods involve dissolving the sample in excess acid and back-titrating to a specified end-point. Where the methods differ is that in the standard A B A , the sample is dissolved in strong, excess HC1, boiled for one minute, and then back-titrated to p H 7.0 (Sobek et al., 1978). In the modified A B A , however, the sample is dissolved in less HC1, placed on a shaking table for 24 hours, and then back-titrated to p H 8.3 (Coastech Research Inc., 1991). The modified method was developed because it was believed that the standard A B A produces an unrealistically high NP. The standard A B A produces a higher N P than the modified method because boiling the sample results in the dissolution of neutralizing minerals, such as slow-weathering silicates (eg. muscovite), which would not otherwise dissolve. The modified method uses a less rigorous procedure for dissolving the sample to better emulate natural conditions, thus leading to the dissolution of only those neutralizing minerals which provide relatively rapid neutralization (eg. carbonates, fast-weathering silicates). Interpretation of Acid-Base Accounting Interpreting the results of static tests is a somewhat subjective process and interpretation can vary significantly depending on the values used. The two common methods of interpretation involve calculating the net neutralization potential (NNP) and calculating the neutralizing potentiakacid-generating potential ratio (NP:AP) (SRK, 1992). In most cases, both methods are utilized to provide a higher level of confidence to the interpretation than using the methods independently. 32 Net Neutralization Potential Method The net neutralization potential (NNP) is calculated by subtracting the acid-generating potential (AP) from the neutralization potential (NP): N N P = N P - A P (1.13) A n N N P of -20 kg C a C 0 3 / t or less indicates there is a high potential that the sample will generate acidity. A n N N P of +20 kg C a C 0 3 / t or greater indicates a low potential for the sample to generate acidity. A n N N P between these two values represents uncertain overall acid-generating potential and further testing using kinetic methods should be performed. There is a possibility that incorrect conclusions can be drawn by only looking at the arithmetic difference between the N P and AP. For example, a sample with an NP of 15 kg C a C 0 3 / t and an A P of 5 kg C a C 0 3 / t is less likely to produce acidic drainage than a sample with an N P of 150 and an A P of 140. This is because even though both samples have an N N P of 10 kg CaC0 3 / t , the first sample has three times as much neutralization potential as acid-generating potential while the second sample has almost equal amounts of N P and AP. It is for this reason that the ratio method was proposed. Ratio Method The ratio method uses the proportions of N P and A P in the sample to interpret the likelihood of acid generation. A n N P . A P ratio of 1 or less indicates there is a high potential that the sample will generate acidity. A n NP:AP ratio of 3 or greater indicates a low potential for the sample to generate acidity. A n N P . A P ratio between these two 33 values represents - an uncertain potential for the sample to generate acidity and kinetic testing should be carried out. It is believed that the ratio method provides a more conservative interpretation of A B A results and a more realistic estimate of the sample's overall acid-generating potential. In British Columbia, an NP:AP ratio of 4:1 has been proposed for the initial classification of waste material and to determine i f kinetic testing is warranted (Price and Errington, 1995). Other jurisdictions (eg. Nevada, California, Idaho) have suggested ratios ranging from 1.5:1 to 3:1 (EPA, 1994). 1.7.2.2 Kinetic Testing Kinetic tests are usually conducted on samples identified through static testing as having an uncertain overall acid-generating potential. Kinetic testing can be used to confirm static test results, determine rates of acid generation, neutralization, sulphide oxidation, metal depletion/release, or neutralization potential depletion and to determine appropriate control/treatment methods (SRK, 1989; Morin and Hutt, 1994). In addition to their review of static prediction methods, Coastech Research Inc. (1991) also reviewed commonly used kinetic prediction methods. Kinetic methods are used both in the laboratory and in the field and include (Coastech Research Inc., 1991; E P A , 1994): • humidity cells • columns/lysimeters • B C Research confirmation tests • soxhlet extraction • batch reactors (shake flasks) • field scale tests Humidity cells and columns are the most commonly used kinetic methods in British Columbia. Humidity cells were extensively used during this thesis and will be described in more detail below. Humidity Cells The purpose of the humidity cell is to accelerate the rate of weathering of a sample in a controlled environment. The laboratory setting allows factors such as airflow, temperature, and leachate volume to be manipulated to produce the desired degree of weathering in the sample. Weathering is achieved through the standard operation of the cell which typically involves a 3-stage, 7-day cycle (Coastech Research Inc., 1991). In Stage 1, dry air is passed through the sample (for waste rock) or over the sample (for tailings) for three days. In Stage 2, humid air is passed through or over the sample for three days. The third stage involves leaching the sample on the seventh day with a measured volume of solution (usually distilled water). The solution is normally applied by letting it drip into the cell over a number of hours (Coastech Research Inc., 1991; White et al., 1993), although new methods have suggested submerging the sample in the solution and shaking the cell prior to collecting the leachate. The above procedure produces a weekly leachate that is collected and analyzed as described below. 35 Interpretation of Kinetic Test Data Most kinetic tests involve the collection of a leachate on a regular basis. This leachate is usually analyzed for, among other things, pH, conductivity, alkalinity/acidity, sulphate concentration, and metal concentrations. The data obtained enable a number of values to be calculated. The rates of metal depletion/release and sulphide oxidation can be determined, assuming that the sulphate present in the leachate is the result of sulphide oxidation. Both of these rates are typically expressed in terms of mass of metal/sulphate released per unit mass or surface area of sample per unit time (eg. mg/kg/week). One can determine i f the sulphate in solution is caused by sulphide oxidation or some other mechanism (eg. gypsum dissolution) by calculating the calcium+magnesiunr.sulphate molar ratio (Morin et al., 1995a). A ratio less than 1 indicates that the neutralization potential has been depleted, or acid is being generated faster than it can be neutralized. A ratio close to 1 indicates gypsum dissolution is occurring, because calcium and sulphate are in a 1:1 ratio in gypsum. A ratio greater than 1 indicates that carbonate (eg. calcite, dolomite) dissolution is occurring in response to acid generation. Other useful values that can be calculated from the leachate data include rates of N P depletion by examining oxidation rates and molar ratios (Morin et al., 1995a, 1995b; Morin and Hutt, 1994) and rates of metal leaching (White et al., 1994). Metals of interest usually include toxic, heavy metals such as copper and zinc, as well as base cations such as calcium and magnesium. 36 1.8 Waste Rock Dumps The focus of this research is the impact of acid generation from the Northwest Dump on water quality at the Island Copper Mine. Therefore, general characteristics of waste rock dumps that are relevant to acid rock drainage will be reviewed, followed by a specific examination of the Northwest Dump. The dump's construction will be discussed, its physical and chemical characteristics will be presented, and a summary of previous research conducted on the dump will be provided. 1.8.1 GENERAL CHARACTERISTICS A number of reports have examined waste rock dumps in relation to A R D (Morin et al., 1991; Ritchie, 1994a; Smith et al., 1995). A brief summary of the findings of these reports will help place the Northwest Dump into context for further discussion of acid rock drainage from the dump and its effect on water quality. 1.8.1.1 Construction Methods The manner in which a dump is built can have a significant influence on the way it behaves in regard to solids, gas, and water transport. Typical methods used for dump construction include: end-dumping, push-dumping, free dumping, and drag-line spoiling (Morin et al., 1991). Only the first three are used to a large degree in British Columbia. Each of the methods listed above produces waste rock dumps with different physical characteristics, which in turn can affect the acid-generating behaviour of a 37 dump. End-dumping, when the waste-rock is dumped directly over the crest of the dump from a truck, causes particle segregation, with fine particles accumulating near the crest of the dump and larger particles collecting at the toe (Morin et al., 1991; Ritchie, 1994a; Whiting, 1985; Smith et al., 1995). Push-dumping, when the waste-rock is dumped near the dump crest and then pushed over with a bulldozer, also results in the large particles gathering at the toe, although to a lesser degree. Unlike end-dumping, however, there does not appear to be any segregation of the fine particles at the dump crest. Free-dumping, when the waste-rock is dumped in individual piles across the surface of the dump, results in little particle segregation, but significantly more compaction than the previous methods. Each of these methods will affect the permeability, hydraulic conductivity, and other hydrogeological properties of the dump in different ways. As will be discussed below, these properties play an important part in the generation of A R D within a waste rock dump. Compaction is a key factor in the development of a waste rock dump because it can dramatically influence the volume of air and water that infiltrates a dump (Broughton and Robertson, 1991). The most common cause of compaction is vehicle traffic on the top of the dump. The breakdown of dump material into smaller particles and the repositioning of particles, due to compaction, will affect both flow paths and infiltration in the dump. Compaction can be considered an indirect form of solids transport which is discussed below. 38 1.8.1.2 Solids Transport Morin et al. (1991) list two primary methods of solids transport in a waste rock dump: internal migration and geotechnical failure. Internal migration occurs when small particles move past larger particles and is a result of the physical and geochemical processes that occur within the dump. Geotechnical failure, on the other hand, is when large portions of the dump move towards the toe and is a result of internal forces, slope erosion, and foundation failure. The significance of internal migration is unclear, as some studies say it is rare while others say it is common. A common form that probably occurs in most dumps is the migration of fine particles from the surface during periods of precipitation. This process can lead to particles being carried away by runoff, infiltration of particles into the dump, and obstruction of flowpaths and channels. Although its significance is uncertain, when it does occur, internal migration can be an important process because it can seriously alter the hydrological properties of a dump. Dump permeability, porosity, and hydraulic conductivity can all be affected by particle migration (Morin et al., 1991; Whiting, 1985; Smith et al., 1995). Geotechnical failure is of a much more serious nature, but since it did not occur in the Northwest Dump, it will not be discussed. 1.8.1.3 Gas Transport The concentration of oxygen is usually considered the rate-limiting step in sulphide oxidation (Ritchie, 1994b; Nicholson, 1994; Morin, 1993; Broughton and Healey, 39 1992; Harries and Ritchie, 1983; Kleinmann et al., 1981). Therefore, the supply of oxygen to a waste rock dump is very important in terms of acid generation. The transport of gas (air, hydrogen sulphide, etc.) can have dramatic effects on the geochemistry of a waste rock dump. Morin et al. (1991) identify four modes of gas transport: gaseous diffusion, gaseous advection, aqueous diffusion, and aqueous advection. Gaseous diffusion and advection are applicable to all gases that can be found within waste rock dumps, while aqueous diffusion and advection are only relevant for those gases which can dissolve in water to significant concentrations. Ritchie (1994a) specifically examines the mechanisms of oxygen transport in waste rock dumps and he also identifies four primary modes: oxygen dissolved in rainwater, diffusion through the pore spaces in the dump, convection, and advection. In terms of Morin et al.'s modes, oxygen dissolved in rainwater would fall under aqueous advection, diffusion through the pore spaces would be covered by gaseous diffusion, and convection and advection would be included under gaseous advection. Gaseous diffusion involves the random movement of gas particles which continues until there are equal concentrations of the gas throughout a given volume (Morin et a l , 1991). During construction, the concentration of gases within the dump will be equivalent to that in the atmosphere. As the dump weathers, however, different amounts of the gases will be consumed through various geochemical processes and concentration gradients will develop, usually with lower concentrations inside the dump 40 (Ritchie, 1994a; Morin et a l , 1991). These gradients will drive the diffusion of gas from the outside of the dump to the inside and this process will continue as long as these gradients exist. Due to the presence of channels, large pore spaces, and typically low/no water table in a waste rock dump, the effect of diffusion on gas transport is usually negligible. Gaseous advection involves the movement of gas through the pore spaces within a dump and is due to a pressure gradient (as opposed to the concentration gradient which drives gaseous diffusion). The two primary factors affecting gaseous advection are dump permeability and the existing pressure gradient (Morin et al., 1991). Permeability is the ability of liquids to flow through a medium and is controlled by properties of the fluid (Whiting, 1985; Fetter, 1994). Permeability is comparable to hydraulic conductivity, which is a function of both the properties of the fluid and the characteristics of the medium, in this case the dump. As described above, permeability can be affected by, among other factors, solids migrations through the dump and void spaces. Permeability values in waste rock dumps range anywhere from 50-200 darcies for air and 0.10-10 darcies for water (Whiting, 1985), where a darcy equals 9.87 x 10"9 cm 2 . Ritchie (1994a) provides permeability values for air through dumps in the range of 1x10 9-9xl0" 1 3 m 2 (1-1000 darcies). There are three mechanisms that affect the pressure gradient within a dump (Ritchie, 1994a; Morin et al., 1991): wind current around the dump, heat generation within the dump, and barometric changes in and around the dump. Wind blowing against the side of a dump will lead to higher pressures on that side of the dump, causing air to move into the dump. It is unclear as to the significance of wind advection, 41 but because this effect lasts in the order of hours and the transport time of air through a dump under these conditions is also probably hours, there may not be enough time for significant changes to occur (Ritchie, 1994a). Heat generation within the dump is probably the most significant form of gaseous advection (Ritchie, 1994a; Morin et al., 1991). Sulphide oxidation is a very exothermic reaction leading to temperatures of 5 0 - 6 0 ° C inside waste rock dumps. The temperature gradient that is generated leads to convection currents within the dump. These currents cause heated air to move upwards from the inside of the dump and the heated air gets replaced by cooler air from the outside. Gaseous advection is most significant in winter due to the extreme differences in temperature between the dump interior and the atmosphere. Barometric pumping is a result of the time required for the interior of the dump to respond to changes in outside atmospheric pressure (Morin et al., 1991). When the exterior pressure increases, gas moves into the dump and when the pressure drops, gas moves out of the dump. Barometric pumping is considered to have only minor effects on gaseous advection within a dump, although in some cases (eg. Rum Jungle; Harries and Ritchie, 1981) it can be significant. Aqueous diffusion is important for those gases which can dissolve in water to significant concentrations, but because the diffusion coefficients of gases in water are usually orders of magnitude lower than they are in air, this process is usually considered a negligible form of transport (Morin et al., 1991). The only significant form of aqueous diffusion is likely the diffusion of oxygen through the water film surrounding rock particles where it can react with acid-generating minerals. 42 Aqueous advection is only a factor after the gases have dissolved into water. The amount of gas moving into a waste rock dump will be directly related to the dissolution rate of the gas into water and the amount of water that moves through the dump (Morin et al., 1991). Dissolution rates of gas into water vary, but at 2 5 ° C , the concentration of oxygen is approximately 8 mg/L. This concentration will increase as temperature decreases until the solubility limit is reached, in this case 12 mg/L at 10°C. Water movement, as will be discussed in Section 1.7.1.4, is influenced by a number of factors, including hydraulic conductivity, size of pore spaces, and the void ratio of the dump. The importance of aqueous advection on gas transport in waste rock dumps is unclear. Morin et al. (1991) indicate it can be an important mode of transport in highly permeable dumps with a significant source of oxygenated water, while Ritchie (1994a) suggests that oxygen saturated rainwater is not a significant source of oxygen in the context of acid rock drainage. 1.8.1.4 Water Transport Water flow into, through, and out of a waste rock dump is of great importance to A R D for three reasons: 1) water is a primary ingredient in sulphide oxidation, 2) water contains dissolved gases, such as oxygen, which are used in various geochemical reactions, and 3) water acts as a medium for the transport of A R D from its source into the environment. 43 Whiting (1985) lists a number of factors, both physical and chemical, that can affect the hydrology of a waste rock dump. Physical factors include stratification, channelling, segregation, and permeability, while chemical factors include precipitation/hydrolysis, temperature, and oxidation. Morin et al. (1991) also added hydraulic conductivity, hydraulic gradient, and groundwater flow to Whiting's list. A l l of the physical factors listed above are influenced by the construction method used in building the dump. In addition, some of the chemical factors, such as precipitation and sorption, can affect channelling, stratification, and permeability. Typical values for permeability of water through waste rock dumps are 0.10-10 darcies (Whiting, 1985), and for hydraulic conductivity are 10"2-10"9 m/s (Morin et al., 1991). Smith et al. (1995) report saturated hydraulic conductivity values in B . C . spoil piles (both coal and metal mines) in the range of lO^-lO" 6 m/s, while Ritchie (1994a) suggests that the higher values (10~4 m/s) are more reasonable. In general terms, one can look at three different scenarios of a waste rock dump and describe it in terms of water flow and acid drainage generation (Smith et al., 1995). The first type of dump is a non-segregated, coarse grained pile full of channels in which the water passes through very quickly and the finer grained material next to the channels could be the source of long-term A R D as flushing occurs within the channels. The second type of dump is a non-segregated, fine grained pile which is likely to form a water table near the base. In this type of dump, sulphide oxidation and neutralization reactions can occur throughout the unsaturated zone. Water flow will transport the soluble 44 oxidation and neutralization products from the source points to the saturated zone at the base of the dump, and from there to discharge points. The last type of dump is a segregated pile, where the fine grained particles have accumulated near the dump crest. This fine grained zone will act as a barrier to infiltrating water and may not allow significant quantities of water into the dump. Most infiltration will occur along the slopes of the dump. This preferential infiltration will produce two distinct regions within the dump: 1) areas that are flushed frequently, removing any reaction products away from the source, and 2) areas where water flow is minimal, resulting in the storage of large amounts of reaction products which will only be transported away from the source during periods of uncommonly heavy infiltration. The Northwest Dump, as will be described below, could have been considered a combination of Types 1 and 3 due to the methods used during its construction. 1.8.1.5 Internal Chemistry The chemical reactions that occur inside a waste rock dump are the cause of acid rock drainage and are the main focus of most A R D studies. The most important of these reactions are sulphide oxidation (both abiotic and biotic) and acid neutralization. Other reactions include metal leaching and iron precipitation. These reactions have been described above in Section 1.7 and are covered in great detail by others (Morin et al., 1991; Ritchie, 1994b; SRK, 1989; SRK, 1992; Stumm and Morgan, 1981; Alpers et al., 1994; Blowes and Ptacek, 1994), therefore they will not be reexamined here. 45 1.8.2 The Northwest Dump At approximately 1 million tonnes, the Northwest Dump (Plates 1-2, 1-3) was the smallest dump at the Island Copper Mine. As part of the mine's closure plans, the dump was excavated and placed back into the open pit prior to its flooding because it had been observed that the dump was producing acid rock drainage. Contamination of Francis Lake is thought to have been caused by this acidic drainage (Home, pers. comm., 1995). Water quality data indicated that zinc levels had begun to increase in Francis Lake in 1994 and there was concern that leaving the Northwest Dump where it was could result in even higher zinc concentrations in the lake. This section will examine the construction, previous research, and physical and geochemical characteristics of the Northwest Dump and provides the background to the research described in later sections. Excavation of the dump and reanalysis of the previous data are examined later on in Section 3 (Results). 1.8.2.1 Construction The Northwest Dump was constructed adjacent to the northwest corner of the open pit, between the pit and Francis Lake. The original topography of the ground beneath the dump is shown in Figure 1-12 and the original watercourses in this area can be seen in Figure 1-13. The dump was constructed between December, 1982 and January, 1984, although dumping only actively took place during four months (Table 1-1). Waste rock deposition was primarily by push-dumping but free dumping was occasionally used as well (Lister, 1994). The different methods of dumping and their 46 Plate 1-2: Air Photo of Northwest Dump (Looking Southwest to Northeast) photo courtesy of I. Home Plate 1-3: Francis Lake and Northwest Dump (Looking Northwest to Southeast) photo courtesy of I. Home 47 F i g u r e 1 - 1 2 : O r i g i n a l G r o u n d S u r f a c e ( a f t e r L i , 1 9 9 1 ) 48 49 TABLE 1-1 CONSTRUCTION OF NORTHWEST DUMP (AFTER LI, 1990) DATE BENCH LOCATION IN PIT ROCK TYPE #OF LOADS TRUCK FACTOR TONNAGE (TONS) %OF DUMP Jan/83 1240 N W , W Andesite, Pyrophyllite 2433 140* 340,620 32.38 June/83 1200 N W Andesite 1173 137.6 161,405 15.34 July/83 1200 N W Andesite 3895 140.1 545,690 51.87 Jan/84 1160 N W - 39 108.3 4,244 0.41 TOTAL 7540 1,051,939 100.00 Notes: * : assumed factor - : information unavailable tonnage = (truck factor) x (# of loads) effects on dump hydrology were discussed above. The vast majority of the dump material was deposited in the first three months and only a small fraction of the material was dumped in the fourth month. The sequential deposition of the Northwest Dump is shown in Figure 1-14, while the ultimate dump configuration can be seen in Figure 1-15. It is interesting to note that no glacial till was deposited in the Northwest Dump. This is significant because local glacial till is net acid-consuming (Lister, 1994). 50 51 9600 23350 F i g u r e 1 - 1 5 : F i n a l D u m p C o n f i g u r a t i o n ( a f t e r L i , 1 9 9 1 ) 1.8.2.2 Physical and Geochemical Characteristics of the Northwest Dump Physical Characteristics The base of the dump had an areal extent of approximately 5.3 hectares while the top of the dump was approximately 1.6 ha in area. The dump had an overall mass of 954, 319 tonnes and a volume of 516, 582 m 3 . Thus, the overall bulk density of the dump material was 1847.4 kg/m 3 . If one assumes that the solid waste material from the Northwest Dump had a specific gravity of 2650 kg/m 3 , then the average dump porosity was 30.3% (Li, 1990). As mentioned earlier, the dump was constructed by both push-dumping and free dumping. These methods of construction likely had some important effects as described in Section 1.8.1.1 above. Push-dumping would have resulted in partial segregation with large particles gathering at the toe of the dump, but little segregation of the fine particles at the dump crest. Free-dumping would have resulted in little particle segregation, but significantly more compaction. Together these methods would have affected the permeability, hydraulic conductivity, and other hydrogeological properties of the dump. Due to particle segregation, permeability is probably higher at the base of the dump than at the top. Compaction from free-dumping probably resulted in low permeability layers, wherever this method was used. These low permeability layers would tend to conduct water laterally and not allow water to flow to the bottom. 53 Geochemical Characteristics This section reviews water quality data gathered from the dump and local groundwater and provides a brief description of the A R D testwork that was carried out as part of the previous research (Section 1.8.2.3). The review of water quality will provide a basis for evaluating the effects of dump excavation on local water quality around the dump site. Northwest Dump Water Quality Seepage, characteristic of acid rock drainage, emanated from this dump for 11 years of its 13 year history. This seepage was collected in a series of perimeter ditches that were constructed soon after contaminated seepage was first detected. Drainage typical of that produced by the Northwest Dump is shown in Table 1-2. This drainage data is a composite produced from water quality data gathered for the last four months prior to the start of excavation. Table 1-2 Typical Surface Drainage Quality Prior to Excavation (March-July, 1995) p H Conductivity (mS/cm) Sulphate (mg/L) Acidity (mg/L) C u (mg/L) 3.65 2.39 1708 296 0.39 Z n (mg/L) C a (mg/L) M g (mg/L) A l (mg/L) C d (mg/L) 50 458 94 32 0.23 54 The production of acid in the Northwest Dump and the resulting contaminated drainage is significant because the dump was inside the Stephens Creek watershed (Figure 1-3). This means that groundwater from the dump flowed towards Francis Lake not the open pit, although surface drainage was pumped into the North Dump drainage system. Typical groundwater quality compiled from samples taken between the dump and Francis Lake is shown in Table 1-3. These data are a four-week average from May to June, 1995 prior to the start of dump excavation. 1.8.2.3 Previous Research A detailed investigation of the Northwest Dump was conducted from September, 1989 to April , 1990. The work performed included a drilling program, static and kinetic prediction testing, and monitoring (Li, 1990) The conclusions presented as part of this investigation are provided at the end of this section. TABLE 1-3 TYPICAL GROUNDWATER QUALITY PRIOR TO EXCAVATION (MAY-JUNE, 1995) p H Conductivity (mS/cm) Sulphate (mg/L) Acidity (mg/L) C u (mg/L) 4.52 1.57 1105 55 0.06 Zn (mg/L) Ca (mg/L) M g (mg/L) A l (mg/L) C d (mg/L) 16 309 44 3 0.06 55 Drilling Program A drilling program initiated in September, 1989 provided samples for acid-base accounting (ABA) analysis and allowed access to the interior of the dump for temperature and gas measurements. There were four main objectives of the N W D drilling program (Li, 1990): • to collect waste rock samples from the dump for analysis • to install gas content - temperature monitoring probes • to install piezometers to allow measurement of the water table • to install P V C pipes to house gas pressure sensors Seven holes were drilled to the original ground surface using a Becker drill. The holes were drilled along two lines that bisected the dump and intersected at an angle of approximately 7 4 ° (Li, 1990). Monitoring wells were installed as soon as drilling had reached the original ground surface, while the drill casing was still in place. Installation began by sealing off the bottom of the drill hole with moistened bentonite pellets. A one-inch inside diameter (I.D.) P V C pipe was then inserted for the purpose of housing temperature probes. To the outside of this pipe were attached 3/16 inch plastic tubes, the first terminating at six inches above the bottom and the rest attached at two meter intervals to within two to four meters of the surface. These tubes were used for gas sampling at different depth intervals within the dump. The space between the P V C pipe and the wall of the drill hole was filled with silica sand and/or pea gravel. Bentonite seals were put in place between the 56 openings of the plastic tubes to prevent vertical migration of dump gases during sampling. The seven holes were designated N W D #1, #2, #4, #6, #7, #8, and #9 (see Figure 1-16). Three of these holes ( N W D #1, #4, and #6) were drilled in duplicate and these additional holes were designated N W D #1A, #4A, and #6A respectively. The duplicate holes were completed due to space limitations in the original drillholes. It was found that once a temperature probe and piezometer were installed, there was little room remaining for the installation of the pressure probes. It was decided that separate holes drilled in close proximity to the originals would be used to hold the pressure probes. The locations of holes N W D #2, #4, and #6 were chosen in an attempt to intersect the original watercourses that existed beneath the dump (see Figure 1-13). Previous Acid Rock Drainage Study Samples from the Northwest Dump were collected throughout the drilling program described above. Samples were collected from rock cuttings at eight-foot intervals from the surface of the dump to the original ground surface. These samples were analyzed for A R D potential using the standard acid-base accounting and humidity cell methods described in Section 1.7. The results of these analyses are included in Appendix C and a reanalysis of these data is provided in Section 3.2. 57 Monitoring Conducted During Previous Work In addition to the A B A analyses that were performed, the drillholes were monitored for temperature, C 0 2 concentration, and 0 2 concentration. Monitoring took place on three occasions over a five-month period (October, 1989; November, 1989; February, 1990). The data gathered during these times were analyzed to determine i f oxidation was occurring within the dump. Due to the nature of the reaction, sulphide oxidation would be indicated by the results when there was a significant increase in temperature, an increase in C 0 2 concentration, and/or a drop in 0 2 concentration. Conclusions from Previous Work The conclusions presented in the work conducted in 1989/90 include (Li, 1990): • there was an inverse correlation between C 0 2 and 0 2 concentration - as C 0 2 concentration increased, 0 2 concentration decreased and vice-versa • C 0 2 and 0 2 concentrations did not change significantly with time, therefore it is possible the dump had reached some sort of steady state • there was a partial relationship between gas concentrations and the results of the acid-base accounting - an interval that exhibited a significant decrease in 0 2 tended to have a high A P , but the reverse relationship was not true. That is, an interval with a high A P did not necessarily exhibit a significant decrease in 0 2 . Therefore, no significant correlation between the two parameters was calculated. 59 • temperatures at or near the dump surface mimicked the ambient air temperature (ie. an increase in dump temperature corresponded with an increase in the ambient air temperature), while the temperatures within the dump body proper were generally elevated with respect to the ambient temperature. - this suggests that oxidation was occurring within the dump, causing an increase in the internal temperature of the dump. • no correlation existed between temperature and gas concentrations • no relationship existed between temperature and the A B A analyses 1.9 Summary Section 1 has presented the background information necessary for understanding the work conducted in the following sections. The thesis was introduced, information regarding the mine was provided, and a general description of waste rock dumps was presented. The remainder of the thesis is as follows: the experimental methodology will be presented (Section 2), followed by the results (Section 3), then a discussion of these results (Section 4), and finally conclusions and recommendations arrived at as a result of the research will be provided (Section 5). 60 2. EXPERIMENTAL METHODS The following chapter provides descriptions of the various methods used to carry out the research program. In each section, objectives are stated and the procedure followed to achieve these objectives is described. Where possible, references for standard procedures are provided in addition to a brief synopsis of the methodology employed. Where new or uncommon methods were used, a more detailed description of the procedure is given. 2.1 Sampling Program The sampling program for this research consisted of three phases: collection, preparation, and characterization. Each of these phases is discussed in more detail below. 2.1.1 Sample Collect ion Samples were collected during site visits in September/October and October/November of 1995. Due to the inherent heterogeneity of waste rock dumps, it would have been very difficult to collect representative samples. According to the Draft Acid Rock Drainage Technical Guide, Volume I (SRK, 1989), because the Northwest Dump was approximately 1.0 million tonnes in size, 20-25 samples should have been collected to adequately characterize the dump. Due to the nature of the dump and the 61 resources available, a sampling program that would have sufficiently represented the dump, according to the above guidelines, was not possible. After reviewing the previous work that was conducted on the Northwest Dump (see Section 1.8.2), it was determined that the portion of the dump being excavated in the first lift contained some of the least reactive material in the dump and that a large number of samples of this material would not be necessary. Examining the previous work it was believed that the second lift to be excavated would expose more significant acid-generating material than the first lift and should therefore be sampled more intensively. Excavation of the third and final lift of the dump was completed before another site visit could be arranged, thus no samples from the lower portion of the dump were collected. At the time of the first site visit (September 29-October 1, 1995) the middle two-thirds of the first lift had already been excavated (Figure 2-1). A small portion at the south end of the dump was being actively excavated during the visit while the northern quarter was relatively intact. It is from this northern section that samples were first collected. Due to the nature of the dump, a talus slope formed almost immediately upon disturbance of the rock during excavation (Plate 2-1). Based on earlier work conducted on the North Dump (Lister, 1994), it was thought the Northwest Dump material would be quite competent and that excavation would produce sheer vertical faces. These faces would have allowed comprehensive and accurate sampling to be conducted and detailed vertical profiles of the dump to be constructed. The formation of the talus slope was 62 unexpected and the sampling program had to be altered accordingly. Grab samples were collected along a vertical transect of the exposed face which was approximately 10 m in height. The exposed face (Figure 2-1, Plate 2-1) was located approximately 10 m north-east of Drillhole NWD#8 . Five samples were collected at 2 m intervals starting 1 m below the original dump surface. Approximately 5 kg of rock were collected at each location and stored in heavy-duty plastic sample bags for transport back to U B C . Due to the formation of the talus slope, the sample origin could only be determined for the rock collected from the first location. Collection of the other four samples was to provide material that could be used as a composite to represent the rest of the rock in the first lift. The second site visit was conducted from October 29 to November 1, 1995 at which time excavation of the second lift was already in progress. Excavation was proceeding from the northwest corner of the dump to the northeast corner. The northwest corner had already been removed resulting in an exposed face running northeast to southwest along the long axis of the dump. Active excavation of the dump was occurring during sample collection. This meant surveying was not possible, so the precise location of the sampling sites is not known. The approximate locations of the active face and the sampling sites are shown in Figure 2-2. Samples were collected from a total of eight locations along the active face, starting in the vicinity of Drillhole NWD#9 and proceeding to the southwest to a position approximately 20 m due west of Drillhole NWD#7. Samples were collected at approximately 15 m intervals along the working surface starting at the northeast end of the active face and proceeding towards 65 66 the southwest end. Five to ten kilograms of material were collected at each location and stored in heavy-duty plastic sample bags for transport back to the university. In addition to the samples collected from the active face of the Northwest Dump during the second site visit, samples were also collected from dump material already in the open pit as well as from a dike that separates Lower Twin Lake from the northwest corner of the open pit. As before, 5-10 kg of sample were collected from each location and placed into heavy-duty, plastic sample bags for transport. Material from the open pit was collected in two locations. Samples were obtained from coarse material that had reached the first bench of the open pit and from finer material that was located approximately 3 m below the crest of the open pit. Due to the unstable nature of the material caught on the upper benches and the possible safety hazard, these were the only samples obtained from the material in the pit. Three samples were collected from the dike. The samples were taken along a transect paralleling the lake shore. The transect was approximately 3 m above the water surface and ran for approximately 40 m along the shoreline. Five to ten kilograms of sample were collected at roughly 10 m intervals. At the end of each site visit, the sample bags were placed in large ( l m x 0 . 5 m x 0.5m) plastic shipping crates for transport back to the Department of Mining and Mineral Process Engineering. The crates were shipped by Greyhound Bus Lines and reached the Centre for Coal and Mineral Processing (CMP) at the University of British Columbia within three days of their delivery to the Port Hardy station. 2.1.2 Sample Preparation At the C M P laboratory, samples were removed from their bags, air-dried and transferred into clean, heavy-duty plastic sample bags for storage. Samples were stored in the C M P for 2-3 months before being prepared for testing. A total of eight samples were selected for testing. Two of these samples were material collected from the exposed face of the first lift during the first site visit, four of the samples were material collected from the active face of the second lift during the second visit, one sample was material collected from the open pit, and the eighth sample was material collected from the dike separating Lower Twin Lake from the pit. The sample identified as Sample 1 was taken from material 1 m below the original dump surface. The sample identified as Sample 2/3 was a composite of material collected from the talus slope. It was decided to composite this material because its origin was unknown. It was believed that a composite would be more representative of material in this portion of the dump than individual samples. The sample labeled Pit 1/2 was a composite of the material collected from the two locations in the open pit. The origin in the dump of this material was also unknown so it was decided to combine the two samples to produce a more representative sample of material from the pit. Sample Dike 1/2/3 was a composite of the material collected from the dike. The composite was produced based on a visual assessment of the material in the exposed surface of the dike. Surficial appearance of this material suggested that it was all very similar and therefore the material from the three locations was combined to provide a single sample 68 representative of the dike material. Sample Bulk 1 was material collected in the vicinity of Drillhole NWD#9 at the working surface of the second lift. The samples identified as Bulk 3-2 were replicates of material collected from an area with visible evidence of sulphide oxidation (heavy iron precipitation). This material was chosen because it was the most oxidized material encountered within the Northwest Dump. This material was collected approximately 2 m from the working surface of the second lift at the location indicated in Figure 2-2. The sample was divided into four portions to provide duplicates for the humidity cell testing and to provide some idea as to the effect of batch leaching on the dump material. Samples Bulk 3-21 and Bulk 3-22 were duplicates of the original material that were subjected to batch leaching, while samples Bulk 3-23 and Bulk 3-24 were duplicates of the original material which did not undergo batch leaching. Bulk 5 was material collected at the working surface approximately halfway between samples Bulk 1 and Bulk 8 and approximately 10-15 m west of Sample 3-2. Sample Bulk 8 was material collected approximately 20 m west of Drillhole NWD#7 and was the southern-most sample collected from the active face of the second lift. The samples were divided, using a Gilson Riffle Splitter (2-inch openings), into portions of various mass depending on the test being conducted. Sub-samples of approximately 1 kg were produced for shake flask testing and acid-base accounting, while sub-samples of approximately 2 kg were produced for batch leaching and kinetic testing. Approximately eight kilograms of sample were divided into two 4 kg sub-samples. Only a few particles did not fit through the openings in the Gilson Riffle Splitter. Any oversized particles were divided equally between the sub-samples. One of 69 these sub-samples was put aside for storage and the other sub-sample was split twice more to produce one 2 kg sub-sample and two 1 kg sub-samples. One of the 1 kg sub-samples was used for shake flask testing and acid-base accounting, while the 2 kg sub-sample was used for batch leaching and kinetic testing. The remaining 1 kg sub-sample was added to the original 4 kg sub-sample and both were placed back into the plastic bag for storage. 2.1.2.1 Shake Flask Testing and Acid-Base Accounting Both shake flask testing and acid-base accounting require fine samples, so the 1 kg sub-sample set aside for this purpose was ground, crushed, and pulverized. For both of these procedures, material that passed 80% - 200 mesh was used. Comminution of the samples took place in the C M P laboratory. Gyratory crushers, cone crushers and a pulverizer were used to reduce the sample to the desired particle size. Samples were first placed in a Traylor 22" gyratory crusher which reduced the samples down to -2+1 inch. This material was then placed in a Massco 10" cone crusher which produced an end-product of -1+1/2 inch. A Massco 6" cone crusher further reduced the material to -1/4+1/8 inch. At this particle size, the material was small enough to be pulverized. A Braumpulverizer (Type U A ) was used, and the grinding discs were adjusted to produce a 100% -100 mesh end-product. 70 2.1.2.2 Batch Leach and Kinetic Testing Material approximately 6 mm in size was required to conduct the batch leach and kinetic testing. Therefore the 2 kg sub-sample was reduced until the majority of the sample passed through a 4 mesh (Tyler series) sieve. As with the shake flask testing and acid-base accounting, the samples were reduced using the equipment in the C M P . The samples were passed through the Traylor gyratory crusher and the Massco 10" cone crusher to produce a sample in the size range stated above. The reduced sample was passed through a Gilson Riffle Splitter to produce a sub-sample approximately 1.2 kg in mass which was used for the batch leach and kinetic testing. The remainder of the sample was placed in heavy-duty, plastic sample bags for storage. 2.1.3 SAMPLE CHARACTERIZATION The samples collected were characterized both physically and chemically. Physical characterization included particle size analysis and thin section analysis, while chemical characterization included x-ray diffraction (XRD), multi-element ICP analysis, and whole rock analysis. A l l of these procedures provide different information about the sample and together they provide an overall picture as to the size, mineralogy, and chemical composition of the sample. 71 2.1.3.1 Particle Size Analysis The original samples and the post-batch leach material were subjected to particle size analysis so that it could be determined what proportion and size fraction of the original material was represented by the material used in the humidity cells. The post-batch leach material was also analyzed so that the surface area of the sample used in the humidity cells could be calculated. The exception to this procedure was for samples Bulk 3-23 and 3-24. For these samples, the two 1 kg sub-splits of the original material were analyzed, rather than the post-batch leach material because these samples did not undergo a batch leach. A standard V2 sieve series analysis using Tyler sieves was performed. Sieves with openings ranging from 53 mm to 0.075 mm (200 mesh) were used for the original samples and sieves with openings ranging from 9.51 mm (2 mesh) to 0.075 mm (200 mesh) were used for the post-batch leach material. The surface area of the rock used in humidity cells can be calculated using the following formula (SRK, 1989): S = I (6 Mj/di) (2.1) g Where: S = surface area of material Mj = mass of material in fraction i d! = diameter of material in fraction i - taken as the mid-point between the two sieves g = representative density of solids The density of the rocks in the dump was derived from L i (1990) and the range of densities provided in the Draft Ac id Rock Drainage Technical Guide, Volume I (SRK, 72 1989). This equation assumes that all of the particles are spheres. Although this is not always true, a fairly accurate determination of surface area is still obtained. It is important to know the surface area of the sample because this can affect reaction rates of the material. 2.1.3.2 Thin Section Analysi s Thin sections were prepared in the Department of Earth and Ocean Sciences at the University of British Columbia. What were believed to be representative samples of the original material were chosen with the help of a member of the Earth and Ocean Sciences Department and polished thin sections were made. Thin section analysis was conducted by Vancouver Petrographies Ltd. 2.1.3.3 X-ray Diffraction X-ray diffraction (XRD) analysis was performed by the Department of Earth and Ocean Sciences at the University of British Columbia. A Siemens D5000 powder diffractometer using C u K a radiation (40 k V , 40 mA) was employed. The spectra were collected from 3 to 6 0 ° 20, with a step size of 0 . 0 2 ° 20 and a count time of eight seconds. 2.1.3.4 Multi-Element and Whole Rock Analysis Multi-element analysis was conducted by Chemex Labs of North Vancouver, British Columbia. Inductively-coupled plasma atomic-emission spectroscopy (ICP-AES) 73 was used after total sample digestion in a triple acid solution of hydrochloric-perchloric-nitric acid. Chemex Labs also conducted whole rock analysis of the samples. Wavelength dispersive x-ray fluorescence (XRF) was used to determine the concentration of major oxides present in the sample. Quality assurance/quality control checks were performed on the ICP and whole rock analyses to confirm their accuracy. These checks were conducted by plotting the results of the whole rock analysis against the results of the ICP analysis for a particular element (Downing, pers. comm. 1996). A linear relationship indicates that the two analyses correspond with each other and the results can be used with a high degree of confidence. Scattered data indicates that there is some discrepancy between the analyses and caution should be used when interpreting the results. 2.2 Investigation of Northwest Dump The Northwest Dump was described in some detail in Section 1.8 of the introduction. Two aspects of the dump, however, required further attention than what was provided in the introduction: reanalysis of the A R D work conducted during the previous research and an account of the excavation of the dump. 74 The data collected from the A R D study in the previous work were converted into the same units (kg CaCCVt) and additional terms (eg. N P . A P ratio) were calculated. These were then interpreted using the methods outlined in Section 1.7. The excavation of the dump was the basis of this research. Anecdotal evidence from mine personnel and observations made during site visits will be used to provide a description of the excavation. 2.3 Preliminary Testwork Before kinetic prediction tests could be performed, it was determined that any readily soluble weathering products should be removed from the samples. To do this, shake flask tests were conducted to determine the optimum leach solution and then a batch leach was performed. The methodology followed to conduct the shake flask tests and perform the batch leach is provided below. 2.3.1 Shake Flask Tests 2.3.1.1 Objectives The primary objective of the shake flask tests was to determine the solution most effective at removing accumulated weathering products from the Northwest Dump waste rock. The number of solution exchanges required to provide optimal washing of the sample was also determined. 75 2.3.1.2 Methods and Materials The following procedure was developed for the purpose of conducting small-scale shake flask testing: 1. Sample ground to approximately 80% - 200 mesh and 10 g sub-splits of the sample were used for testing 2. Sample placed into 250 ml Erlenmeyer flask 3. 200 ml of solution added - used solutions of p H 2, p H 4, and distilled water 4. Flask placed on shaking table for 24 hours 5. Flask removed from shaking table and allowed to settle for 2 hours 6. Supernatant drained from sample (measured volume) and replaced 7. Supernatant centrifuged and analyzed for sulphate 8. Procedure repeated if sulphate concentration decreased by more than 10% from last rinse Only four of the eight samples were chosen for shake flask testing (Pit 1/2, Dike 1/2/3, Bulk 3-2, and Bulk 5) because the objective of the testing was to determine the most effective solution for removing soluble sulphate from the samples, not to determine the sulphate loading of each sample. Sulphuric acid was used to prepare the acidic solutions for the shake flask testing and subsequent batch leach because acid rock drainage is a sulphate-based system (acid generation produces sulphate). The leachate from the shake flask tests was analyzed using a Hach D R 100 Portable Colorimeter and Hach Permachem Reagents Sulfaver 4 Sulfate Reagent Powder. The reagent powder was added to the sample and the sample was set aside for 5 minutes. This incubation period 76 was to allow for the formation of barium sulphate. The sample was then inserted into the colorimeter and the sulphate concentration was read off a log scale. Standard solutions of 10 mg SO4/L and 40 mg SO4/L were used, as well as blanks, to test the accuracy of the colorimeter. 2.3.2 Batch Leach Tests The Northwest Dump existed for approximately 13 years. During this time weathering products formed and some were stored within the dump while others were leached into seeps. Before accurate data could be obtained from kinetic testing, these stored products had to be washed out of the samples. In standard humidity cell testwork, this flushing period can take from 20-30 weeks depending on the age of the samples (Sherlock, 1995). Due to time constraints, it was decided to batch leach the samples prior to kinetic testing to remove the majority of these stored products. The results of the shake flask tests were used to determine the optimal leach solution and the necessary number of solution exchanges. 2.3.2.1 Objectives The purpose behind the batch leach was to remove accumulated weathering products from the surfaces of the rock samples before kinetic testing was initiated. It was hoped that this pre-kinetic test wash would reduce the time required to reach equilibrium conditions in the humidity cells. 77 2.3.2.2 Methods and Materials The batch leach of the waste rock samples was performed according to the following procedure: 1. Sample crushed to -6 mm and 1.2 kg were used for leaching 2. Sample placed into washed 20 L plastic containers 3. 10 L of leach solution added (solution determined from shake flask testing) 4. Containers agitated three times a day (vigorous shaking and swirling) for three days 5. Supernatant removed (measured volume by weight) and replaced 6. Leachate analyzed for sulphate and metals 7. Steps 4-6 repeated three times After the batch leach had been completed, the sample was removed from the plastic container and placed into a sample tray. The sample was allowed to air dry overnight in the C M P laboratory. The next day, the samples were placed in low temperature ovens ( 3 0 - 4 0 ° C ) and allowed to dry. The samples were weighed at 2 hour intervals and placed back into the oven i f the weight was still decreasing. The sample was then split, using the Gilson Riffle Splitter as above, into two portions of approximately 200 g and 1 kg. The 200 g portion was rinsed and subjected to acid-base accounting (Section 2.4). The 1 kg portion was used in the humidity cells (Section 2.5) after particle size analysis (see above) had been conducted. 78 2.4 Static Prediction Tests Static prediction tests in the form of acid-base accounting and carbonate analysis were performed on sub-splits of original material, post-batch leach material, and post-humidity cell material. The procedure used to create these sub-splits was described earlier. The analyses were performed on all three types of material to provide the most reliable estimate of the neutralization potential of the rock and allow N P depletion to be calculated. The post-batch leach material was analyzed because it was thought that the leaching solution used in the batch leach may have removed some of the neutralization potential from the sample. As outlined in Section 1.7, acid-base accounting requires the determination of two values: the acid-generating potential and the neutralization potential. The acid-generating potential (AP) determination will be discussed in Section 2.4.1 and the neutralization potential (NP) determination will be discussed in Section 2.4.2. Carbonate analysis, also used to determine the NP, is described in Section 2.4.2.4. 2.4.1 ACID-GENERATING POTENTIAL The two procedures that are commonly followed to determine the acid-generating potential of a sample are part of the standard acid-base accounting and the modified acid-base accounting methods described in Section 1.7. Both of these methods determine the sulphur content of the sample and this value is then entered into an equation to calculate the acid-generating potential. Before the acid-generating potential can be calculated, 79 however, a mass balance should be conducted. The mass balance is performed to account for the sulphur content in the samples as they progressed from their original state, through batch leaching, and into the kinetic testing. 2.4.1.1 Sulphur Mass Balance A flowchart that shows each stage the sample went through during testing, and the components that account for the sulphur content of each stage, is presented in Figure 2-3. The following assumptions regarding sulphur content were made in order to perform the mass balance: S original Snatch leach residue + ^lost material + S i e a c n solutions " Sacid added (2-2) Sbatch leach residue ~~ ^washed residue + S^nse solutions S y n e t i c head (2.3) ^kinetic head ^kinetic residue + ^weekly leachate (2.4) Where S ; = the mass of sulphur in stage i The sulphur content of the batch leach residue was unknown because the sulphur analysis performed after the batch leaching was conducted on the 200 g sub-split which had been rinsed, not on the batch leach residue itself. Therefore, the sulphur content of the batch leach residue was calculated through manipulation of Equation 2.2, where the sulphur content of the batch leach residue equals the sulphur content of the original sample plus the sulphur in the acid added minus the sulphur in the batch leach solutions 80 ORIGINAL S A M P L E (known mass, known assay) I A C I D (known mass) • A D D E D BATCH LEACH L O S T M A T E R I A L B A T C H L E A C H R E S I D U E L E A C H S O L U T I O N S (known mass, assay) (known mass, unknown assay) (known volume, assay) W A S H E D R E S I D U E RINSE S O L U T I O N S K I N E T I C H E A D (known mass, assay) (known volume, assay) (known mass, unknown assay) I WEEKLY LEACH ^ \ K I N E T I C W E E K L Y R E S I D U E L E A C H A T E (known mass, assay)(known volume, assay) FIGURE 2-3 FLOWCHART USED FOR SULPHUR MASS BALANCE 81 minus the sulphur content of the lost material. The sulphur content of the batch leach residue was also calculated in three other ways. The first way was using Equation 2.3, where the sulphur content was back-calculated from the washed residue and its rinse solutions, the kinetic residue and the weekly leachate. The second way was by assuming that the sulphur content of the washed residue plus the sulphur contained in the rinse solutions would equal the sulphur content of an equivalent mass of the batch leach residue. The third way was by assuming that the sulphur content of the kinetic residue plus the sulphur contained in the weekly leachate would equal the sulphur content of an equivalent mass of batch leach residue. From Figure 2-3 it can be seen that the batch leach residue and the kinetic head material are equivalent. Thus calculating the sulphur content of one provides the sulphur content of the other. 2.4.1.2 Standard Acid-Base Accounting Determining the acid-generating potential using the standard method simply involves determining the total sulphur content of the sample and then entering this value into the following equation (Coastech Research Inc., 1991): A P = % sulphur x 31.25 (2.5) Where : A P = acid-generating potential of the sample in tonnes C a C 0 3 equivalent per 1000 tonnes of material The above equation assumes that sulphur is converted entirely into sulphate and that 4 moles of H + are produced per mole of pyrite oxidized. The total sulphur content was determined using gravimetric analysis (see Analytical Methods, Appendix A). 82 2.4.1.3 Modified Acid-Base Accounting The modified method is slightly more complicated than the standard method because the sulphide content is determined rather than the total sulphur content. To calculate the sulphide content, both total sulphur and sulphate sulphur are determined and the sulphide content is assumed to be the difference between the two values. The other assumptions of this method are that sulphide is converted entirely into sulphate and that 4 moles of H + are produced per mole of pyrite oxidized. The equation used to calculate the A P for the modified method is the same as that used for the standard method (Equation 2.5), except that the sulphide content is used rather than the total sulphur content. The total sulphur and sulphate sulphur contents were determined using gravimetric analysis (Analytical Methods, Appendix A). 2.4.2 NEUTRALIZATION POTENTIAL There are a number of different ways that the neutralization potential of a sample can be determined. Among the most common methods are the standard A B A , devised by Sobek et al. (1978), the modified A B A , developed by Coastech Research Inc. (1991), and the carbonate C 0 2 method. Prior to any of these tests being conducted, however, a paste p H is typically measured to determine i f acid generation has already occurred in the sample. 8 3 2.4.2.1 Paste p H The paste p H is a preliminary test used to determine if a sample has generated acid prior to static testing. A paste p H above 7 indicates that reactive carbonate is still present in the sample. If the paste p H is below 5, however, then this indicates that the sample contains acidity from previous acid generation. Objectives The objectives of the paste p H analysis are to determine the p H of a paste of the sample made by mixing the finely ground sample in distilled water and to determine i f acid generation has already occurred in the sample prior to static testing (Coastech Research Inc., 1991). Methods and Materials The equipment required for this analysis and the procedure that was followed are described in the Ac id Rock Drainage Prediction Manual (Coastech Research Inc., 1991). No changes were made to the methodology outlined in this document during the performance of these analyses. 2.4.2.2 Standard Acid-Base Accounting As mentioned earlier, standard acid-base accounting ( A B A ) developed by Sobek in 1978 (Sobek et al., 1978) is one of the most widely used static prediction tests in 84 British Columbia. Some possible limitations of this test were described in Section 1.7, but even with these problems the test can still provide some important and interesting information. Objectives The overall objective of this part of the standard A B A is to determine the neutralization potential of the sample. Methods and Materials The analysis consists of two steps: a fizz test and the acid-base accounting. The fizz test is used to determine the volume and concentration of acid that is to be added to the sample during the acid-base accounting. A few drops of concentrated HC1 are added to 2 grams of ground sample and the resulting reaction is rated as none, slight, moderate, or strong. The neutralization potential (NP) is determined by conducting an A B A . This is accomplished by adding excessive hydrochloric acid to the sample, heating the mixture to ensure complete reaction, and then titrating the solution with sodium hydroxide to an end-point of p H 7. The NP is calculated from the calcium carbonate equivalent of the acid that was consumed during the reaction, according to Equation 2.6 (Section 2.5.2.5). The equipment required and the procedure followed for this A B A analysis are described in both the Field and Laboratory Methods Applicable to Overburdens and Minesoils (Sobek et al., 1978) and the Acid Rock Drainage Prediction Manual (Coastech Research Inc., 1991). The method provided in the Acid Rock Drainage Prediction 85 Manual was followed for these analyses. The only difference between the methodology described in the above manual and that used in the actual analyses was the use of 80% -200 mesh sample instead of-60 mesh sample. This change was made because -200 mesh is the size fraction that was used in the modified A B A analyses. At least two of the samples in each run were tested in triplicate to provide an estimate of the accuracy and precision of the procedure. 2.4.2.3 Modified Acid-Base Accounting The other commonly used static prediction test is the modified A B A . This test was developed to overcome perceived shortcomings of the standard A B A . The main differences between the two tests were described earlier (see Section 1.7). Objectives As with the standard A B A , the primary objective of this part of the modified A B A is to calculate the neutralization potential of the sample. Methods and Materials The A c i d Rock Drainage Prediction Manual (Coastech Research Inc., 1991) provides a description of the equipment required and the method followed to perform a modified A B A . A revised procedure has been developed by Lawrence and Wang (1996) and it was this procedure that was followed for conducting the modified A B A on the Island Copper samples. As with the standard A B A , the modified procedure consists of 86 two parts: a fizz test and the acid-base accounting. The fizz test was described earlier for the standard A B A . The acid-base accounting was performed as follows: 1. Measure 2.0 g of pulverized sample (80% - 200 mesh) into a 250 ml conical flask and add approximately 90 ml of distilled water. 2. Add certified HC1 according to the following table (based on the fizz rating): Fizz Rating Volume of I N HC1 none slight moderate strong 2 ml (time = 0 hr.) + 1 ml (t=2 hr.) 2 ml + 1 ml 2 ml + 2 ml 3 ml + 2 ml 3. Agitate contents of the flask for 24 hours by placing it on a shaking table. After two hours, add the remaining volume of HC1 as indicated in the above table. Check the pulp p H after 22 hours. If the p H is greater than 2.5, add enough HC1 to bring the p H to within the range of 2.0-2.5. If the p H is less than 2.0, too much acid was added in Step 2. Repeat the test using the volume of HC1 indicated by the next lower fizz rating. 4. At the end of 24 hours, add distilled water to the flask to make a total volume of 125 ml. Check the p H to make sure it is still within the specified range. 5. If the pulp p H is in the correct range, titrate the contents of the flask using certified 0.5N N a O H to an end-point p H of 8.3. 87 Numerous tests led to a slight modification of the above procedure. When the fizz test indicated a rating of "none", only 2 ml of HC1 was added at the start of the test. The additional 1 ml of acid to be added after two hours was left out because it frequently caused the p H to drop below the 2.0 - 2.5 range stated in Step 3. The neutralization potential is calculated using the same equation as was used to calculate the NP for the standard A B A (see Equation 2.6 in Section 2.5.2.5). 2.4.2.4 Carbonate C 0 2 Analysis The other method used to determine the neutralization potential of the samples was the carbonate C 0 2 method. This method is used to calculate the carbonate content of the sample. Some researchers believe that this is the only portion of the sample that provides any significant neutralization of acid that is generated during sulphide oxidation (Lapakko, 1994). By performing this analysis in conjunction with the above A B A analyses, one can determine i f the sample has sufficient carbonate to fully neutralize any infiltrating acid or i f the NP of the sample comes from some other neutralizing minerals, such as fast-weathering silicates. Carbonate is important because it will provide the initial neutralization of any acid that comes into contact with the sample. If there is little NP supplied by carbonate in the rock, then it is likely that acid coming into contact with the rock will not be fully neutralized and acidic drainage will result. Objectives The purpose of conducting a carbonate ( C 0 2 ) analysis is to determine the carbonate content of the sample. Methods and Materials The carbonate carbon content of the Island Copper samples was determined using a Coulemetrics' Incorporated (CI) 5030 Carbonate Carbon ( C 0 2 ) Apparatus. The carbonate content of the sample is determined by adding acid to the sample and transferring the C 0 2 that is evolved into a CI 5010 C 0 2 Coulometer where it is titrated with a base to provide a value for total carbon present in the sample. The method employed for carrying out these analyses is presented below (Coulometrics Incorporated, 1984): 1. Weigh sample and put into sample tube (sample size is selected to evolve one to three milligrams of C 0 2 ) . For these tests, 40 mg of sample were used which typically evolved 0.05 - 0.2 mg C 0 2 . It was found that using more than 40 mg of sample resulted in destruction of the KI scrubber solution. 2. Allow system to purge C 0 2 which entered during sample charging 3. Inject 2 ml of perchloric acid (HC10 4) using automatic dispenser 4. Rotate sample tube onto heater ( 5 0 ° C ) 5. Allow sample to react until coulometer provides a steady reading (for these tests, each sample was allowed to react for exactly 15 minutes) 89 Four to five blanks were run prior to sample testing to provide a value for the background carbon. These background concentrations were used to adjust the values obtained from the samples. Blanks were run after every five samples and at least two of the samples in each run were tested in triplicate to provide an estimate of the accuracy and precision of the procedure. 2.4.2.5 Calculations The results of static prediction tests can be analyzed in a number of ways. The following is a list of terms and their associated equations that are commonly used for analyzing static test data. The equation used to calculate the acid-generating potential was provided above in Section 2.5.1. NEUTRALIZATION POTENTIAL (NP)= 50 a fx - (b/a) vl (kgCaC0 3 / t ) (2.6) (standard or modified) c Where : a = normality of HC1 used b = normality of N a O H used c = sample weight in grams x = volume of HC1 added (mis) y = volume of N a O H added (mis) NEUTRALIZATION POTENTIAL (NP) = u g C x 8 . 3 4 (kgCaC0 3 / t ) (2.7) (carbonate) mg of sample NET NEUTRALIZATION POTENTIAL (NNP) = NP - A P (kg CaC0 3 / t ) (2.8) NEUTRALIZATION POTENTIAL RATIO (NPR) = NP/AP (unitless) (2.9) The above terms will be used for the analysis of the static test data provided in Section 3.5. 90 2.5 Kinetic Prediction Tests Kinetic testing has become one of the most common methods of assessing the impact of acid rock drainage in Canada. As explained briefly in Section 1.7, the purpose of conducting kinetic tests is to simulate the weathering of mine wastes (tailings or waste rock) to determine if these wastes will generate contaminated drainage. The most commonly used kinetic tests utilize humidity cells or columns. The kinetic tests conducted as part of this research employed humidity cells. The procedure followed to perform the kinetic tests and analyze the weekly leachate is outlined below. 2.5.1 OBJECTIVES The objectives of conducting humidity cell tests on the Northwest Dump material were: 1) to determine the rate of acid generation, 2) to determine the rate of N P depletion, 3) to determine the rate of metal loading/leaching, and 4) to determine leachate quality. 2.5.2 METHODS AND MATERIALS 2.5.2.1 Samples The origin and preparation of the material in the cells were described above in Section 2.1. A Gilson Riffle Splitter was used to obtain an approximately 1 kg sub-split of the material from the batch leach residue. The actual mass of sample in each of the humidity cells is shown in Table 2-1. A l l of the samples except Bulk 3-23 and 3-24 were 91 subjected to batch leaching. This was done in an attempt to determine the effect of batch leaching on the time required to reach equilibrium in the weekly leachate parameters. Bulk 3-2 was run in duplicates of batch leached and non-batch leached samples to indicate variance caused by sampling and testing and to test the accuracy and precision of the procedure. The sample material was placed into the cell in such a way as to reduce particle segregation as much as possible. The sample was spread out on utility paper, mixed, and then placed into the cell in layers using a scoop. Each time the scoop was run through the sample it was at 9 0 ° to the previous pass. After four passes the sample was remixed and the process repeated until the entire sample had been placed into the cell. TABLE 2-1 MASS OF SAMPLE IN HUMIDITY CELL Sample Mass (g) 1 900 2/3 970 Pit 1/2 920 Dike 1/2/3 975 Bulk 1 975 Bulk 3-2! 905 Bulk 3-22 865 Bulk 3-23 1010 Bulk 3-24 1010 Bulk 5 940 Bulk 8 895 92 2.5.2.2 Equipment and Procedure The equipment required for conducting the humidity cell tests included a compressed air source, a humidifier, and humidity cells. Compressed air for the humidity cells was provided through a six-port compressed air manifold located on the fifth floor of the Frank Forward Building. A Parker condensation trap removed excess moisture from the air. Air flow was controlled at the cells using a Dwyer air regulator. Air supply to the cells was maintained at approximately 11 L/min to provide an average air flow to each cell of 1 L/min. During the dry air cycle, the air supply was divided between two feeder lines each of which had junctions leading off to the individual cells. This arrangement meant the cells were receiving air in series with the result that the cells nearest the air supply were subjected to a higher flow rate than the cells at the end of the feeder lines. To provide a more balanced supply of air to all of the cells, tube clamps were placed between the feeder line and the cell and adjusted to produce a flow rate of approximately 1 L/min. The precise flow rate through the cells was adjusted using "bubblers" which are described below. During the humid air cycle, air was supplied to a humidifier which provided a more balanced flow of air to each individual cell. As with the dry air cycle, the specific flow rate through each cell was regulated with a "bubbler". The humidifier was built using specifications provided by Dr. R.W. Lawrence from the Department of Mining and Mineral Process Engineering at the University of British Columbia (Figure 2-4). It was constructed using a 1 m length of 15 cm outside 93 M -M 1 o diameter (O.D.) Plexiglas tubing that was sealed at both ends. The tube was placed parallel to the counter surface and filled approximately half full with distilled water. Inside the tube were a submersible aquarium heater and two aquarium aerators (25 cm and 15 cm in length). The aerators were connected with plastic tubing to the compressed air source during the humid air cycle. Placed at equal intervals in two rows of six along the top of the humidifier were outlet ports fitted with 0.5 cm O.D. Teflon nipples. These ports were connected to the humidity cells with latex tubing. Another port on top of the humidifier was used to hold a thermometer. The thermometer was positioned so that its humidifier. The temperature of the humidifier was maintained at 3 0 ° C to achieve 100 per cent humidity in the air flowing to the cells. To save on time and costs, humidity cells built for a previous research project were used for kinetic testing. The design of these cells is shown in Figure 2-5. The cells were constructed with Plexiglas tubing sealed at the bottom and open at the top. The cells had an overall length of 20 cm and an outside diameter of 10 cm. The bottom of the cells had a drainage hole fitted with a 0.5 cm O.D. Teflon nipple to allow the collection of leachate. Located 2.5 cm from the bottom of the cell was a perforated plate on which the sample charge was placed. On top of the plate, four layers of fine mesh cloth were placed to prevent the loss of sample fines during the weekly leach. On the side of the cell, between the cell bottom and the perforated support plate, another nipple was inserted to allow air injection. Square lids with 12.5 cm sides were placed on top of the cells and held in place with 12-inch (30 cm) F-clamps. Each lid had a nipple in the center to allow air to escape and allow application of the leachate. A slight modification 95 10 cm 20.5 cm T 2.5 i cm 1 distilled water in air out Q Q Q Q Q) leachate out lid filter cloth sample charge (~1 kg) perforated support plate dry/humid air in Figure 2-5 : Schematic Drawing of a Humidity Cell 96 to the standard humidity cell design was the incorporation of a "bubbler" on top of the cells to which the air outlet nipple was attached by means of latex tubing. The purpose of the "bubbler" was to provide a more discrete method of controlling air flow through the cells (White U l and Sorini, 1993). The "bubbler" (Figure 2-6) was constructed using a small glass beaker (9.5 cm in height, 3.5 cm diameter) fitted with a rubber cork which held two glass tubes (12 cm and 5 cm in length). The beaker was filled with 40 ml of distilled water and placed on the lid of the humidity cell. The longer tube was fitted with latex tubing and attached to the air outlet nipple of the humidity cell. The shorter tube acted as an outlet port. Air flow through the cells could be regulated by adjusting the position of the long glass tube in relation to the bottom of the beaker. Moving the tube closer to the bottom increased the head above the outlet of the tube, reducing air flow through it and thereby reducing flow through the humidity cell. Conversely, moving the tube farther from the bottom of the beaker reduced the head above the outlet of the tube, thereby increasing air flow through it and in turn the humidity cell. Airflow in the cells was maintained at approximately 0.75 L/min ( ± 0 . 1 5 L/min) for the duration of the tests. Airflow was measured using a Gilmont No. 12 Standard Compact Flowmeter (Model GF-2200) with an airflow capacity of 0 - 2000 ml/min. The cells were placed on slotted, wooden platforms, approximately 20 cm in height, underneath which were placed 500 ml Erlenmeyer flasks to collect the weekly leachate. Humidity cells were operated on a one-week schedule, whereby dry air was passed through the sample for three days, then humid air was passed through the sample for three days, and on the seventh day, the samples were rinsed with 500 ml of distilled 97 Figure 2-6 : Schematic of a "Bubbler" water. At the end of each stage, the cells were disconnected and weighed. Weighing the cells after each stage provided a measure of the moisture content of the cell. If more than 10 mis of condensate formed in the bottom of the cell during the humid air cycle it was measured and added to the leachate collected after rinsing. The distilled water was applied using 500 ml separatory funnels placed above the cells in clamps and attached to the cells with latex tubing. The water was applied at a rate of approximately 3 ml/min, producing a total rinse time of 2.5 to 3 hours. A layer of coarse filter cloth was placed on top of the sample before rinsing occurred. This cloth was used as a diffusive layer to spread the solution over the entire surface area of the sample while rinsing. It was hoped that this diffusive layer would reduce the formation of preferential flow paths directly below the inlet port. Preferential flow paths could cause short-circuiting of the solution, preventing it from reaching all of the sample, thereby affecting the oxidation rate of the sample. The cells were rinsed with the bottom port open and allowed to drain into flasks for 24 hours before the leachate was collected. At the end of 24 hours the cells were manipulated (tilted) to ensure all of the leachate had been collected in the flasks. Leachate volume was determined by weight using pre-weighed flasks. The above procedure was followed each week except for the first rinse when 500 ml of distilled water was poured into the cells, allowed to sit for 3 hours and then drained. This mode of rinsing was employed the first time because it was necessary to remove the last remnants of the batch leaching solution and it was felt that this method provided the most effective rinse. The cells were run for a total of fifteen weeks after the first rinse, from March 8 - June 22, which provided a total of 16 weeks worth of leachate data. 99 2.5.2.3 Leachate Analysis and Calculations Leachate Analysis The weekly leachate that was collected from the humidity cells was analyzed for pH, conductivity, acidity/alkalinity, sulphate, and metals. The analytical procedures followed for measuring these parameters are provided in Appendix A (Analytical Methods). Calculations A number of terms can be calculated for the analysis of kinetic test data. The calculations used in this thesis are provided below: Related to Acid-Generating Potential (Sherlock. 1995) CUMULATIVE S04,PRODUCED = total of: (SO4 concentration per cycle)(leachate volume) sample weight (expressed as mg S0 4/kg) (2.10) SULPHIDE OXIDATION RATE = (cumulative S 0 4 produced during one cycle -cumulative S G i produced during a previous cycle) time difference between cycles (expressed as mg S0 4/kg/week) (2.11) - typically use a 5-week moving average %Ssujphide Remaining = [(original %S s ui p h l d e)(sample weight)(106)-(cumulative SO4¥32/96)1 x 100 (original %S s uip h lde)(sample weight)(106) (expressed as %) (2.12) 100 TIME TO DEPLETE AP = (%S s u l p h l d e remaining/100)(original %S s u , p h l d e ) (sample weight)(10 ) (sulphide oxidation rate)(32/96)(52) (expressed as years) (2.13) Related to Neutralization Potential (Sherlock. 1995^ ) CALCIUM TO SULPHATE RATIO CARBONATE RATIO = SILICATE RATIO = CUMULATIVE NP PRODUCED NP DEPLETION RATE = % NP REMAINING (Ca concentration/40.08) ( S 0 4 concentration/96) (Ca conc./40.08)+(Mg conc./24.3n ( S 0 4 concentration/96) [(Ca concentration/40.08) + (0.5)(Na concentration/22.99) + (0.5)(K concentration/39.09)l ( S 0 4 concentration/96) (2.14) (2.15) (2.16) TIME TO DEPLETE NP = (cumulative S 0 4 produced)(100/96)(molar ratio) (expressed as mg CaC0 3 /kg) (2.17) (molar ratio)(sulphide oxidation rate)( 100/96) (expressed as mg CaC0 3/kg/week) (2.18) (original N P - cumulative NP) x 100 (2.19) original N P (expressed as %) (% N P remaining/100)(sample w e i g h t ¥ 1 0 4 ) (2.20) (NP depletion rate)(52) (expressed as years) 101 Related to Metal Leaching (Sherlock. 1995) CUMULATIVE METAL PRODUCED = total of: (metal concentration per cycle)(leachate volume) sample weight (expressed as mg metal/kg) (2.21) METAL LEACHING RATE % METAL REMAINING = (cumulative metal leached during one cycle -cumulative metal leached during a previous cycle) time difference between cycles (expressed as mg metal/kg/week) (2.22) - typically use a 5-week moving average [(original % metal)(sample weight)(106) -(cumulative metal produced)] x 100 (original % metal)(sample weight)(106) TIME TO DEPLETE METAL (expressed as %) (2.23) (% metal remaining/100)(original % metal) (sample weighty 104) (metal leaching rate)(52) (expressed as years) (2.24) Other Terms (Sherlock. 1995) SULPHATE TO ALKALINITY OR ACIDITY RATIO (SO4 concentration/96) (alkalinity or acidity/100) CALCIUM TO SODIUM,MAGNESIUM RATIO = (Ca concentration/40.08) either (Na concentration/22.99) or (Mg concentration/24.31) CALCIUM, MAGNESIUM, SODIUM OR POTASSIUM TO SULPHATE RATIO = either (Ca concentration/40.08) or (Mg concentration/24.31) or (Na concentration/22.99) or (K concentration/39.09) ( S 0 4 concentration/96) (2.25) (2.26) (2.27) 102 The above terms will be used in the analysis of the kinetic test data presented in Section 3.5. 2.5.3 Humidity Cel l Disassembly At the conclusion of kinetic testing, the humidity cells were carefully taken apart and the contents examined. This was done in an attempt to see what visible effect accelerated weathering may have had on the material and how much material had been lost from the cells. 2.5.3.1 Objectives Humidity cell disassembly was conducted with the following objectives in mind: 1) to determine if visible flowpaths had developed in the sample and 2) to determine the mass of sample lost during kinetic testing. 2.5.3.2 Methods and Material s Humidity cell disassembly was conducted more on a qualitative basis than a quantitative basis. Thus, only a very basic examination was conducted. Flowpaths To determine i f flowpaths had developed in the sample, the material in the humidity cell was extracted and dissected. The following procedure was used: 103 1. A l m x l m piece of heavy-duty utility paper was spread out on countertop 2. Humidity cell, without lid, was inverted at far end of the paper 3. Cell was slowly lowered and pulled towards other end of paper 4. Material spread out in a line approximately 15-20 cm long and width of cell 5. Split material in two along the long axis and slowly spread apart 6. Again using the spatula, one of the newly created halves was split in two along the long axis and slowly spread apart 7. Perform Step 6 on the other half 8. Examine sub-splits for evidence of gypsum or hydroxide precipitation 9. Take photographs It was thought that the presence of precipitates (eg. gypsum, goethite) in a vertical band running through the humidity cell material would provide evidence that flowpaths had formed. Any flowpaths that had formed during kinetic testing would be apparent from the above dissection of the humidity cell material. Mass Loss The material that came out of the cells was reweighed to determine if any mass had been lost as a result of the kinetic testing. The following procedure was used: 1. After flowpath examination, allow humidity cell material to air dry overnight 2. Weigh sample tray and record 3. Place humidity cell material into sample tray, weigh and record 104 4. Place material and tray into a drying oven ( 3 0 - 4 0 ° C ) 5. After two hours, take tray out, weigh and record 6. After another two hours, take tray out, weigh and record 7. If weight in Steps 5 and 6 is same, record weight from step 6 as final weight; otherwise repeat Steps 5 and 6 In order to get an accurate measure of the mass that was lost during kinetic testing, the procedure followed to obtain the sample mass prior to kinetic testing had to be followed during cell disassembly. Therefore the samples were oven-dried according to the above procedure to remove any residual moisture that would cause an inaccurately high weight to be measured. 2.6 Water Quality Analysis Parametric statistics are frequently used to analyze water quality data even though most water quality data are not suitable for these techniques (Helsel, 1987). Water quality data are typically non-normally distributed (a primary assumption of parametric methods) and frequently contain values that are less than detection limits. In order to use parametric techniques, these less than values are commonly deleted (producing a positive bias) or substituted with fabricated values. In addition, water quality data often contain a limited number of values. If fewer than 30 observations have been recorded, the Central Limit Theorem cannot be invoked and parametric methods cannot be used (Helsel, 1987). For the reasons listed above it was decided not to use parametric methods for analyzing the Island Copper data. 105 Another area of statistics that is attracting attention for the purpose of water quality data analysis is non-parametric statistics (Helsel, 1987). Non-parametric techniques have a number of advantages over parametric methods in the field of water quality data analysis, including: 1) transformations prior to analysis are not necessary, 2) data that are non-normally distributed can still be analyzed, 3) greater power is achieved for skewed distributions common to water quality data, 4) comparisons are made between central values (eg. median not mean), and 5) values below the detection limit can be used without having to fabricate values or delete them. As with parametric methods, however, non-parametric methods are only suitable for comparing different groups of samples or two different time periods of the same sample, not for examining multiple observations of the same sample through time. For this reason, time-series analysis was examined as a possible means of analyzing the water quality data collected at Island Copper. Some researchers have suggested that water quality data are well-suited for time-series analysis (Lohani and Wang, 1987; Jayawardena and Lai , 1989). This is an attractive analytical method due to its powerful forecasting and predictive capabilities. A major assumption of time-series analysis is that data were collected at regular time intervals (ie. hourly, daily, weekly, etc.). Island Copper data, however, were collected on a random basis. Samples were sometimes taken daily, sometimes weekly, and in some cases monthly. Therefore it was decided that time series analysis could not be applied to the Island Copper data. Three things had to be done in order to determine the effect or potential effect of acid rock drainage from the Northwest Dump on water quality in Francis Lake and the flooded pit: 1) analysis of current groundwater quality data, 2) comparison of pre-excavation and post-excavation water quality data, and 3) predictions of future water quality. The way in which these objectives were achieved is described below. 2.6.1 Groundwater Quality It is believed that acid rock drainage from the Northwest Dump may have contaminated the groundwater flowing from the dump to Francis Lake. To determine i f this had occurred, monitoring wells had to be installed in order to obtain samples of groundwater. The procedure followed to install these wells and obtain groundwater samples is described below. 2.6.1.1 Objectives Two objectives were set for the groundwater monitoring program: 1) to install monitoring wells to allow the collection of groundwater samples and 2) to determine i f groundwater in the region around the Northwest Dump was contaminated. 2.6.1.2 Methods and Materials Monitoring wells were installed around the North and Northwest Dumps in late January and early February of 1996. The five wells were located to intercept groundwater entering Twin Lakes from the North Dump and Francis Lake from the 107 Northwest Dump (Figure 2-7). Groundwater from an existing stand-pipe, located on the north side of Francis Lake, was used to provide background levels to which the monitoring well data could be compared. The identifying labels of the six wells, the origin of these labels, and the source of the groundwater each well is believed to intercept are provided in Table 2-2. The wells were installed using a gas-powered hand auger with a 4' long, 8" diameter bit. Slotted P V C piping, 3" in diameter, was used as the well casing. The P V C piping was sealed off at the bottom to prevent excessive sediment from entering the well. Holes were drilled to the full extent of the auger bit, the P V C piping was inserted into the hole and the material removed during drilling was used to secure the well in place. The top of the piping was covered to prevent debris and precipitation from entering the well. TABLE 2-2 NAME OF WELL AND PROBABLE SOURCE OF GROUNDWATER Label Origin Source B L R Bay Lake Road Background (north of Francis Lake) G W W Groundwater Well Northwest Dump L G W Lakeside Groundwater Northwest Dump, possibly Twin Lakes P G W Upper Groundwater Northwest Dump P W W Upper Water Well Northwest Dump U T P Upper Twin Lakes Pipe North Dump 108 109 Water samples were collected using a 5'-long piece of flexible plastic tubing with a check valve at the bottom. The tube was inserted into the well, lowered until the bottom of the well was reached, and then raised. As the tube was raised, the check valve would close, trapping the water inside. The water sample thus obtained was transferred to a 1L polyethylene bottle for transport back to the Island Copper environmental laboratory and subsequent atomic absorption analysis. Distilled water was used to rinse out the polyethylene sample bottles and the sampling tube. Samples were collected every week for the first month following installation and then every month thereafter for as long as there was water in the well. 2.6.2 Water Quality Comparison Past water quality data relevant to Francis Lake and the Northwest Dump will be compared to current data (both surface water and groundwater) to determine i f dump excavation has had any impact on water quality. 2.6.2.1 Objectives The objective of this section is to determine i f the excavation of the Northwest Dump had any impact on the water quality of the local groundwater and Francis Lake. 2.6.2.2 Methods and Materials Water quality data collected prior to the excavation of the Northwest Dump will be compared to water quality data collected after excavation. Both groundwater data and no data collected from Francis Lake will be compared. The pre-excavation groundwater data is based on samples collected from a pit that was excavated using a back-hoe. The pit is approximately l m x 6m in area and l m deep and it was completed in an attempt to intercept groundwater flowing from the dump to Francis Lake. The post-excavation groundwater data is based on the analysis of samples collected from the monitoring wells described above. The pre-excavation Francis Lake data is based on the data collected in 1994 while post-excavation data is based on water samples collected in 1996. 2.6.3 Predicted Water Quality The water quality of the freshwater cap that will become established in the flooded pit is very important. The water in this cap will eventually discharge out of the pit and the quality of this effluent must meet the discharge limits that were issued for the Water Management Pond. The kinetic test data collected from the Northwest Dump samples will be used to make predictions regarding water quality in the freshwater cap of the flooded pit. 2.6.3.1 Objectives The purpose of this exercise is to predict the future water quality of the freshwater cap of the flooded pit. i n 2.6.3.2 Methods and Materials To predict the effect of Northwest Dump material on water quality in the flooded pit will be very difficult as the information needed to make accurate predictions is unavailable. The amount of material from the Northwest Dump that is caught on the upper benches of the pit is unknown and neither is its origin within the dump. This makes it almost impossible to determine the amount of metal loading that will occur. Therefore, in order to provide some idea as to the magnitude of the loading that could occur, an example will be presented with assumptions made regarding the amount and origin of the dump material on the upper benches. The loading/leaching rates calculated for this material from the kinetic test data will then be used to determine the potential metal loading (zinc and copper) that could be contributed by this material. 112 3. RESULTS Section 3 presents the results obtained from each stage of the research program. Statistical analysis was performed on the data whenever possible, but due to the small sample size in some cases, the rigour is quite low. 3.1 Sample Characterization Samples were characterized by particle size analysis, thin section analysis, x-ray diffraction (XRD) analysis, multi-element inductively coupled plasma (ICP) analysis, and whole rock x-ray fluorescence (XRF) analysis. The results of these various analyses are presented below. 3.1.1 Particle Size Analysis Particle size analyses were performed on both the original material and the material that was placed into the humidity cells for kinetic testing. The analysis on the humidity cell material can be found in Appendix B - l and a brief summary is presented in Table 3-1. Looking at the data in Table 3-1, it is obvious that the humidity cell material consisted of smaller particles than did the original material. On average, only 19% of the original material passed through a 3.35 mm sieve, while 46% of the material used in the humidity cells passed through the same size sieve. 113 TABLE 3-1 PARTICLE SIZE ANALYSIS (% PASSING 3.35 MM) Sample Original Bulk Sample Humidity Cell Sample 1 20 53 2-3 19 50 Pit 1/2 21 45 Dike 1/2/3 32 48 Bulk 1 14 42 Bulk 3-2! 26 40 Bulk 3-22 - 42 Bulk 3-23 - 44 Bulk 3-24 - 44 Bulk 5 20 50 Bulk 8 20 45 Using the results of the above particle size analyses, the surface area of the waste rock in each humidity cell was calculated using the formula provided in Section 2.1.3.1 (Table 3-2). The representative density of the solids was assumed to be 2.7 g/cm (2700kg/m3). TABLE 3-2 SURFACE AREA OF HUMIDITY CELL MATERIAL Sample 2" Surface Area (m ) Surface Area (m2/kg) 1 0.80 0.89 2/3 0.80 0.82 Pit 1/2 0.68 0.74 Dike 1/2/3 0.58 0.60 Bulk 1 0.59 0.60 Bulk 3-2 j 0.63 0.70 Bulk 3-22 0.64 0.74 Bulk 3-23 0.77 0.76 Bulk 3-24 0.85 0.84 Bulk 5 0.80 0.85 Bulk 8 0.67 0.75 114 The surface area of the material in the humidity cells ranged from 0.60-0.90 m 2 /kg of material. 3.1.2 Thin Section Analysis The report on the thin section analysis performed by Vancouver Petrographies Ltd. is presented in Appendix B-2 and a summary of this information is provided in Table 3-3. Rocks for thin section analysis were chosen on visual appearance, but due to the size of the sample (kilograms of waste rock) compared with the size of the thin section (one slice of one rock from the sample), the probability that the thin section is completely representative of the sample it was taken from is low. The analysis does, however, provide some information on the mineralogy of the sample. Due to the uncertainty in sample representivity care must be taken in interpreting the results and any conclusions regarding sample mineralogy drawn from the data below must be considered carefully. Plagioclase was the primary component of the samples, visually comprising at least 50% of most of the samples (Pit 1/2-20%). Epidote and/or sphene, associated with leucoxene or rutile, were also found in all of the samples and comprised 1-5% of the overall content. Pyrite was another component found in all of the samples, ranging in content from 0.5% (Bulk 5) to 2.5% (Dike 1/2/3). Another common constituent was chlorite (10-25%), found in all of the samples except Sample 2/3 and Pit 1/2. Only three of the samples had significant amounts of carbonate; Sample 1, Sample 2/3, and Bulk 5 115 TABLE 3-3 RESULTS OF THIN SECTION ANALYSES Sample Mineral and estimated content (°/ 1 felsitic plagioclase (62%), chlorite/altered mafic glass (25%), epidote (3%), sphene/ leucoxene (5%), carbonate (3%), pyrite (2%), chalcopyrite (trace) 2/3 plagioclase (50%), altered mafics (33%), leucoxene (5%), epidote (1%), carbonate (10%), pyrite (1%) Pit Vi cryptocrystalline matrix (70%), plagioclase (20%), epidote (7.5%), carbonate (trace), zeolite (trace), pyrite (2%), chalcopyrite (trace) Dike 1/2/3 plagioclase (84%), chlorite (10%), leucoxene (2%), epidote (trace), Fe-Ti oxides (1%), pyrite (2.5%), chalcopyrite (0.5%) Bulk 1 plagioclase (75%), chlorite (20%), epidote (1%), spheneAeucoxene (3%), carbonate (trace), zeolite (trace), pyrite (1%) Bulk 3-2 plagioclase (55%), sericite (5%), chlorite (25%), epidote (10%), carbonate (trace), sphene/rutile (3.5%), pyrite (1.5%) Bulk 5 plagioclase (65%), chlorite (15%), quartz (1%), carbonate (17%), sphene (2%), pyrite (0.5%), chalcopyrite (trace) Bulk 8 plagioclase (60%), sericite (6%), chlorite (25%), epidote (3%), rutile (2%), sphene/ leucoxene (2%), quartz (trace), pyrite (2%) (3%, 10% and 17% respectively). Pit 1/2, Bulk 1 and Bulk 3-2 contained trace amounts of carbonate, while the remaining samples contained no detectable carbonate. Other occasional components occurring in trace quantities were quartz (Bulk 5 and Bulk 8), zeolite (Pit 1/2 and Bulk 1), and chalcopyrite (Sample 1, Pit 1/2, Dike 1/2/3, and Bulk 5). 3.1.3 X-RAY DIFFRACTION ANALYSIS The computer printouts of the X R D analysis can be found in Appendix B-3 and a summary of the results is provided in Table 3-4. 116 TABLE 3-4 SUMMARY OF XRD ANALYSIS Sample Minerals Present 1 2/3 Pit 1/2 Dike 1/2/3 Bulk 1 Bulk 3-2 Bulk 5 Bulk 8 quartz,albite,anorthite,clinochlore,calcite,pyrite,laumontite,gypsum quartz,albite,anorthite,clinochlore,calcite,pyrite,laumontite,gypsum quartz,albite,anorthite,clinochlore,pyrite,laumontite,gypsum,muscovite quartz,albite,clinochlore,pyrite,laumontite,muscovite quartz,albite,anorthite,clinochlore,pyrite,laumontite,gypsum,muscovite quartz,albite,clinochlore,pyrite,laumontite,pyrophyllite,gypsumjarosite quartz,albite,anorthite,clinochlore,calcite,laumontite,gypsum,clinozoisite quartz,albite,anorthite,clinochlore,laumontite,gypsum,muscovite Four minerals were found in all of the samples: quartz, albite, clinochlore, and laumontite, while gypsum was found in all of the samples except Dike 1/2/3. Anorthite was found in all but two of the samples (Dike 1/2/3 and Bulk 3-2), as was pyrite (Bulk 5 and Bulk 8). Calcite was found in Sample 1, Sample 2/3, and Bulk 5, while muscovite was found in Pit 1/2, Dike 1/2/3 and Bulk 8. Less common minerals were pyrophyllite and jarosite in Bulk 3-2, and clinozoisite in Bulk 5. 3.1.4 MULTI-ELEMENT AND WHOLE ROCK ANALYSIS Chemex Labs of North Vancouver performed multi-element ICP analysis and whole rock X R F analysis on the samples. The metals of significance regarding A R D are provided in Table 3-5, while the results of the complete ICP analysis can be found in Appendix B-4. The results of the whole rock analysis are presented in Table 3-6. 117 Table 3-5 ICP Analysis (Significant Metals) Sample A l (%) Ba (ppm) Ca (%) Cu (ppm) Fe (%) K (%) ICP ICP ICP ICP ICP ICP 1 8.85 370 3.63 463 6.39 1.2 2/3 8.27 340 4.75 380 6.53 0.69 Pit 1/2 8.44 340 2.77 438 8.33 1.25 Dike 1/2/3 9.24 360 1.3 1245 6.17 2.01 Bulk 1 8.69 260 2.94 121 7.9 0.67 Bulk 3-2 7.62 210 1.94 772 10.55 1.07 Bulk 5 8.23 250 3.37 177 6.84 0.73 Bulk 8 8.27 330 2.41 258 9.33 0.98 Sample Mg (%) Mn(ppm) Na (%) Pb (ppm) Sr (ppm) Zn (ppm) ICP ICP ICP A A S ICP ICP I 1.9 4200 1.34 290 238 5300 2/3 2.03 4350 1.52 350 340 7100 Pit 1/2 1.72 3880 1.12 670 265 4080 Dike 1/2/3 2.34 1560 1.3 36 181 96 Bulk 1 2.27 5020 1.53 180 369 3060 Bulk 3-2 1.37 3410 0.62 120' 152 4210 Bulk 5 1.99 5140 1.25 264 310 2160 Bulk 8 1.98 5250 0.99 196 253 1535 Aluminum was quite high (7.6-9.2%), probably due to the abundant plagioclase present in the samples, as indicated by the thin section analysis. Calcium was abundant in most of the samples (2.4-4.8%) and in agreement with the thin section and X R D analyses, Sample 1, Sample 2/3, and Bulk 5 had the highest calcium concentrations. Copper was elevated in most of the samples, especially Dike 1/2/3 (1245 ppm). Iron was relatively high (6.2-10.6%), probably due to the presence of pyrite. Extremely important in terms of water quality is zinc, which had high concentrations in all of the samples (>1500 ppm) except Dike 1/2/3 (96 ppm). In general, S i 0 2 was the single, largest constituent of each sample, comprising approximately 50% of the overall sample. Also abundant were A 1 2 0 3 and F e 2 0 3 (16-18% Table 3-6 Whole Rock Analysis Sample A1 20 3 (%) XRF CaO (%) XRF Cr 20 3(%) XRF Fe 2 0 3 (%) XRF K 2 0 (%; XRF ) MgO (%) XRF MnO (%) XRF 1 17.76 5.62 0.01 10.54 1.65 3.27 0.68 2/3 16.79 7.69 0.03 11.27 0.94 3.59 0.72 Pit 1/2 17.02 4.38 0.03 13.95 1.74 3.06 0.65 Dike 1/2/3 18.43 1.97 0.01 10.03 2.75 4.22 0.25 Bulk 1 17.65 4.7 0.01 13.14 0.86 4.04 0.83 Bulk 3-2 15.85 3.17 0.01 17.95 1.62 2.48 0.59 Bulk 5 17.07 5.56 0.03 11.85 1.05 3.63 0.86 Bulk 8 16.95 3.83 0.02 15.74 1.41 3.48 0.89 Sample Na 20 (%) P2O5 (%) Si0 2 (%) Ti0 2 (%) Loss on ignition Total XRF XRF XRF XRF XRF (%) 1 1.66 0.24 48.81 0.82 7.98 99.04 2/3 2.08 0.22 47.52 0.76 7.02 98.63 Pit 1/2 1.35 0.26 48.33 0.87 7.84 99.48 Dike 1/2/3 1.43 0.29 52.29 0.92 6.54 99.13 Bulk 1 1.94 0.29 48.19 0.81 7.3 99.76 Bulk 3-2 0.78 0.22 45.03 0.85 10.69 99.24 Bulk 5 1.57 0.27 49.56 0.8 6.92 99.17 Bulk 8 1.1 0.25 46.8 0.79 8.38 99.64 and 10-18%, respectively). CaO showed the same trend as it did for calcium, with Sample 1, Sample 2/3, and Bulk 5 having the highest percentages and Dike 1/2/3 having the lowest. Titanium oxide, chromium oxide, and potassium oxide comprised only fractions of a percent of each sample. Taking into account the amount of sample lost upon ignition, greater than 99% of the total sample composition was accounted for, with the exception of Sample 2/3 (98.63%). Quality assurance/quality control checks were performed on the ICP and whole rock analyses to confirm their accuracy and they are provided in Appendix B-5. These checks were performed by plotting the results of the ICP assay against the results of the 119 whole rock analysis. The only elements that did not exhibit a good linear correlation between the ICP and whole rock analyses were chromium and titanium. The remaining elements exhibit a high linear correlation (R2>0.90) between the ICP and whole rock analyses which indicates the two methods are providing accurate and consistent measurements (Downing, 1996 - pers. comm.). 3.2 Northwest Dump 3.2.1 REANALYSIS OF PREVIOUS RESEARCH As described earlier in Section 1.8.2.2, samples from the Northwest Dump were collected and analyzed for A R D potential using the standard acid-base accounting method. The results of these analyses were to be compared to those obtained from the current A B A results presented in Section 3.4. To aid in comparing the two sets of data, some additional analysis was performed on the previous data (Appendix C). In addition to the results originally reported (AP as kg H 2 S 0 4 , NP as kg C a C 0 3 , and NNP), A P has been converted into kg C a C 0 3 so the net neutralization potential (NNP) could be recalculated and the NP:AP ratio has been calculated. A review of the earlier data and the additional analysis led to the following observations: 120 • the majority of the material analyzed had a negative N N P and/or an NP:AP ratio of less than one, indicating a potential for acid production throughout the entire dump. • the top 23-31 feet of the dump appears to have had a lower potential for generating acid than the bottom portion of the dump (see Figure 3-1). This becomes important during dump excavation because of the sequence of excavation and the disposal of the dump material into the pit. - especially true for locations N W D #2, 4, 6, and 8 (Figures 3-1, 3-2) - the upper portion (top 23-31 feet) of the dump typically had N N P values of 20 or higher, although in most cases the NP:AP ratio was less than two. • the very bottom of the dump also appears to have had a lower acid-generating potential than the middle portion of the dump (Holes N W D #6-9; Figures 3-1, 3-2) • hole N W D #6 is the only hole with a positive N N P from top to bottom, varying from 36-85 kg CaC0 3 /tonne of material, and thus it is also the only hole with an NP:AP ratio of greater than 1 from top to bottom (1.8-21.4). The significance of these observations will be discussed in Section 4. 121 122 3.2.2 Excavation During the late summer and fall of 1995, the priority for operations personnel at Island Copper was the transport of glacial till from the Beach Dump to the mill site where the till was to be stored for use later on in dump reclamation. It was only during periods of high precipitation that work on the Northwest Dump occurred, because it was during these times that handling of the glacial till became inefficient (Rowden, pers. comm. 1995). Thus excavation of the dump did not proceed in a continuous fashion from beginning to end, but rather as a series of excavations that occurred during times of heavy rain. Due to a shortage of personnel, as a result of the mine's closure in December, 1995, specific details of the Northwest Dump's excavation were not recorded (Rowden, pers. comm. 1996). Details, such as the origin of the material within the dump and which lift of the excavation the material came from, could not be located and therefore the excavation of the dump is described using anecdotal evidence and observations made during site visits. Excavation of the dump occurred by way of three main lifts (Home, pers. comm. 1995). The first lift was approximately 8-10m in depth and excavation of this lift began in early September, 1995. The second lift was smaller than the first, approximately 6-8m in depth, and excavation occurred during October and November, 1995. The final lift was approximately 8-10m in depth and excavation was completed by the middle of December, 1995. 124 3.2.2.1 Visit 1 The middle two-thirds of the first lift had already been excavated and only small portions at the north and south ends remained (Figure 2-1) when the first site visit was conducted. During excavation of this lift, material was end-dumped into the open pit starting at a point midway along the west side of the open pit and proceeding towards the north to a point due east of the northernmost extension of the dump (see Figure 3-3). The material was pushed over the pit crest using a bulldozer. The material that was deposited into the pit during this stage of the excavation filled in the benches as it progressed towards the bottom of the pit. Thus, by the end of excavation of the first lift, a talus slope had been created from the pit crest down to the fifth bench (Plate 3-1). 3.2.2.2 Visit 2 The second site visit took place from October 29 to November 1, 1995 while the second lift was being excavated. At this time, the northwest corner of the lift had already been removed and active excavation of the remainder was occurring. It was during this site visit that the bulk of material used for static and kinetic A R D prediction tests was sampled. From Plate 3-2 it is obvious that many areas in this part of the dump were oxidizing or had oxidized. Based on observations made during sampling, there was a broad band of heavily oxidized material, approximately 2-3 m in height, running along the working surface of this lift. A rough estimate is that 10-15% of the material exposed during excavation of this lift had previously oxidized. 125 1 2 7 Disposal of the material from this lift occurred at two separate locations (Figure 3-3). The first disposal location was immediately adjacent to the dump at the pit crest (Plate 3-1). Loads were end-dumped, but not pushed into the pit, thereby creating a large berm on the pit crest. No active dumping was occurring at this location during the site visit. The second disposal location, where dumping was actively occurring during the site visit, was on the north edge of the pit (Dagenais, pers. obs., 1995). Material from the dump was being used to repair a road that runs along the north edge of the pit, between the North dump and the pit crest (Rowden, pers. comm., 1995). 3.2.2.3 Visit 3 The final site visit occurred from January 29 to February 1, 1996 after dump excavation was completed. Material from the third lift was disposed of down the talus slope adjacent to the dump (Plate 3-3). At the time of the third visit it was observed that a thin layer (1-2 m) of waste rock still remained on the dump and that dumping of glacial till in this area had begun (Dagenais, pers. obs., 1996). Due to a light snow cover, oxidized areas in this remaining rock were not observed. During this site visit, groundwater monitoring wells were installed. 3.3 Preliminary Testwork The results of the preliminary testwork are provided in Appendices D - l (shake flask tests) and D-2 (batch leach). A brief summary of the relevant findings from this testwork is given below. 128 Plate 3-2: Oxidation Zone in Northwest Dump Plate 3-3: Talus Slope in Northwest Corner of Pit photo courtesy of I. Home 129 3.3.1 SHAKE FLASK TESTS Some neutralizing minerals were present in two of the samples (Pit 1/2 and Bulk 5) as the p H in the first cycle using the p H 2 solution was 6.5 and 5.2, respectively. Based on the concentration of sulphate in the leachate, the p H 2 solution removed the most sulphate from the samples, and the concentration of sulphate in the leachate reached a minimum after the third cycle. 3.3.2 BATCH LEACH O f the four samples that underwent shake flask testing, the only sample that displayed similar trends in p H for the batch leach was Bulk 5. The p H of this sample was similar in value to that obtained from the shake flask tests and it decreased in the same manner as it did in the shake flask tests. The other samples all had much lower pHs in the batch leach compared to the shake flask tests. The majority of sulphate in the batch leach samples was removed in Cycle 1. Significantly less sulphate was removed in Cycles 2 and 3. Pit 1/2 and Bulk 1 were somewhat anomalous because sulphate removal from these samples decreased linearly. The zinc leached from the samples shows a trend similar to that for sulphate; the most zinc was leached in Cycle 1 and the least in Cycles 2 and 3. The exception to this trend was Bulk 5 which leached more zinc in Cycle 2 than in Cycle 1. Whereas most of the samples leached similar amounts of zinc in Cycles 2 and 3, Dike 1/2/3 and Bulk 1 leached less zinc in Cycle 3 than in Cycle 2, producing a linear decrease in zinc leached. The copper that leached from the samples displayed a trend that was, for the most part, the same as zinc. Most of the copper was leached in Cycle 1 while the least was leached in Cycles 2 and 3. As with zinc, more copper was leached in Cycle 2 than Cycle 1 for Sample 2/3 and Bulk 5. The copper leached from Bulk 1, on the other hand, decreased linearly from Cycles 1 to 3. Calcium was also analyzed in the batch leach. It displayed a trend similar to sulphate, zinc, and copper. Most of the calcium was leached in the first cycle and Dike 1/2/3 leached significantly less calcium than the other samples. 3.4 Static Prediction Tests The static prediction tests that were performed can be divided into two main categories: determining the acid-generating potential and determining the neutralization potential. The acid-generating potential was determined using the standard acid-base accounting method and the modified acid-base accounting method. The neutralization potential was determined using both of the above methods and the carbonate C 0 2 method. 131 3.4.1 ACID-GENERATING POTENTIAL To calculate the acid-generating potential of a sample, a sulphur analysis was performed. In the case of standard acid-base accounting, the total sulphur content of the sample must be determined. For the modified method, both the total sulphur and the sulphate sulphur must be determined. As part of these determinations, a sulphur mass balance was performed to account for the sulphur content of the sample as it moved through the various stages of testing. 3.4.1.1 Sulphur Mass Balance The mass balance performed to account for sulphur in the samples can be found in Appendix E . The sulphur content (%) of the different heads (initial and calculated heads for batch leach and kinetic testing) is presented in Table 3-7. For all of the samples TABLE 3-7 SULPHUR MASS BALANCE Sample Batch Leach (% S) Initial Head Calculated Head Kinetic Testing (% S) Initial Head Calculated Head 1 4.44 4.68 4.61 5.08 2/3 3.42 2.62 2.47 2.64 Pit 1/2 3.96 3.00 2.74 2.82 Dike 1/2/3 2.05 2.04 2.05 2.22 ' B u l k l 1.75 1.52 1.23 1.25 Bulk 3-2] 5.51 3.68 3.11 3.08 Bulk 3-22 5.51 4.89 4.35 4.38 Bulk 3-23 - - 5.51 3.30 Bulk 3-24 - - 5.51 4.72 Bulk 5 1.86 1.61 1.51 1.49 Bulk 8 2.62 2.14 1.81 1.84 132 samples, except Bulk 3-23 and 3-24, the initial kinetic testing head was assumed to be the same material as the batch leach residue (see Figure 2-3), and therefore had the same sulphur content. From the results of the mass balance it was decided to use the average value obtained from the last three methods described in Section 2.5.1.1. In general, the closure between the initial head and the calculated head is better (ie. smaller) for the kinetic testing material than it was for the batch leach material. Only two of the nine samples that were batch leached had initial and calculated heads that were within 10% of each other (Sample 1-5%, Dike 1/2/3-1%). With the exception of Bulk 3-23 (40%) and 3-24 (14%), however, the initial and calculated heads for the kinetic testing material of all the samples varied by less than 10%. 3.4.1.2 Standard Acid-Base Accounting The results of the A P determination using the standard method are presented in Table 3-8. The sulphur contents used for the original and post-batch leach material are the initial heads taken from Table 3-7. The sulphur content used for the post-kinetic testing material was determined by gravimetric analysis (see Appendix A , Analytical Methods). Sample 1 and Bulk 3-2! to 3-24 consistently had the highest acid-generating potentials of all the samples, ranging from 94 - 172 kg CaC0 3 / t . Bulk 3-2 had the highestAP of the original material (172 kg CaC03/t), while Sample 1 had the highest A P of both the post-batch leach material and the post-kinetic testing material (144 and 165kg 133 Table 3-8 Acid-generating Potential (Standard Method) Original Post- Batch Leach Post - Kinetic Testing Sample %S A P %S A P %S A P 1 4.44 139 4.61 144 5.27 165 2/3 3.42 107 2.47 77 2.7 84 Pit Vi 3.96 124 2.74 86 2.84 89 Dike 1/2/3 2.05 64 2.05 64 2.15 67 Bulk 1 1.75 55 1.23 38 1.22 38 Bulk 3-2, 5.51 172 3.11 97 3.01 94 Bulk 3-22 - - 4.35 136 4.35 136 Bulk 3-23 - - - - 3.03 95 Bulk 3-24 - - - - 4.31 135 Bulk 5 1.86 58 1.51 47 1.48 46 Bulk 8 2.16 68 1.81 57 1.84 58 CaC0 3 / t ) . Bulk 1, which had the lowest sulphur content, consistently had the lowest A P of all the samples in all three stages (38-55 kg CaC0 3 / t ) . 3.4.1.3 Modified Acid-Base Accounting The results of the A P determination using the modified method are presented in Table 3-9. The A P was calculated using the sulphide content of the sample which was assumed to be the difference between the total sulphur content and the sulphate sulphur content. Barite (BaS0 4 ) was mentioned in Section 1.7 as a mineral that can cause problems when calculating the A P because it does not contribute to the acid-generating potential but is included in the overall sulphate content of the sample. For the original samples, the sulphate content (Table 3-9) is, on average, an order of magnitude higher than the barium content (Table 3-5). Thus the influence of barite on the A P of the samples is negligible. 134 TABLE 3-9 ACID-GENERATING POTENTIAL (MODIFIED METHOD) Original Post-Batch Leach Post-Kinetic Testing Sample %S % S 0 4 A P %S % so 4 A P %S % S 0 4 A P 1 4.44 0.5 123 4.61 0.067 142 5.27 0.01 164 2/3 3.42 0.19 101 2.47 0.113 74 2.7 0.01 84 Pit 1/2 3.96 0.62 104 2.74 0.247 78 2.84 0.1 86 Dike 123 2.05 0.16 59 2.05 0.149 59 2.15 0.08 65 Bulk 1 1.75 1.04 22 1.23 0.282 30 1.22 0.08 36 Bulk 3-2 j 5.51 2.05 108 3.11 0.403 85 3.01 0.26 86 Bulk 3-22 - - - 4.35 0.605 117 4.35 0.41 123 Bulk 3-23 - - - - - - 3.03 1.52 47 Bulk 3-24 - - - - - - 4.31 1.04 102 Bulk 5 1.86 0.26 50 1.51 0.2 41 1.48 0.02 46 Bulk 8 2.62 1.15 46 1.81 0.292 47 1.84 0.15 53 There are such differences in the calculated sulphide content of the samples that no trends are readily apparent from the data in Table 3-9. This table does, however, provide an indication as to the effectiveness of the batch leach at removing stored products such as sulphate. Comparing the sulphate contents of the various samples for the original and post-batch leach material it is apparent that the batch leach did remove a significant amount of sulphate from the samples. O n average, 70% of the sulphate in the original sample was removed as a result of the batch leach, although two of the samples, Dike 1/2/3 and Bulk 5, showed much less sulphate removal than the other samples (6% and 24%, respectively). Sample 1 had the highest A P of any sample in all three stages (123-164 kg CaC0 3 / t ) , while Bulk 1 had the lowest (22-36 kg CaC0 3 / t ) . Due to their high sulphate 135 levels, Bulk 3-21, Bulk 3-22, Bulk 3-23, and Bulk 3-24 had much lower APs using the modified method than they did using the standard method. 3.4.2 NEUTRALIZATION POTENTIAL Before the neutralization potential of the samples could be calculated, a paste p H analysis had to be conducted to determine if the samples had already generated acidity. Once the paste p H was measured, the N P of the samples could be determined using acid-base accounting and carbonate analysis. The results of these analyses are presented below. 3.4.2.1 Paste p H The results of the paste p H analyses are shown in Table 3-10. In general, the p H increases from the original material to the post-batch leach material and then decreases in the post-kinetic testing material. This trend holds true for all of the samples except Sample 1 and Bulk 5, both of which show an increase in paste p H from post-batch leach material to post-kinetic testing material. Looking at the original material, the only sample that appears to have previously generated acidity (paste p H < 5) is Bulk 3-2 (pH = 4.59). The samples that still contain some reactive carbonate, indicated by a paste p H that is above 7, are Sample 2/3 (pH 7.49) and Bulk 5 (pH 7.55). Sample 1, with an original paste p H of 6.36, may still Table 3-10 Paste p H Paste PH Sample Original Post Batch Post Kinetic Leach Testing 1 6.36 7.63 7.72 2/3 7.49 8.02 7.96 Pit 1/2 6.22 6.51 6.24 Dike 1/2/3 5.67 5.81 5.14 Bulk 1 6.10 6.41 6.24 Bulk 3-2! 4.59 5.81 5.51 Bulk 3-22 - 5.24 5.12 Bulk 3-23 - - 5.36 Bulk 3-24 - - 5.03 Bulk 5 7.55 7.94 8.07 Bulk 8 5.68 5.86 5.23 have some reactive carbonate present as the paste pHs of both the post-batch leach material (pH 7.63) and the post-kinetic testing material (pH 7.72) are well above 7. 3.4.2.2 Neutralization Potential Determination This section presents the results of the analyses conducted to determine the neutralization potential of the samples. Back-titration curves plotted for the standard and modified methods can be found in Appendix F, while the overall results of the NP determination are presented in Table 3-11. In general, the standard method provided the highest estimate of the neutralization potential and the carbonate method provided the lowest estimate. There were only three exceptions to this trend. The first occurred with the post-batch leach 137 CN o u oo CN 00 C N u~> <T> o od c — •— ON «jp ON c> — G CN ON <T> rf CN JO H c l<2 T 3 <L> H3 H3 o s 00 CN m i n ,—' CN II od rf rf C rf NO oo II c oo rf CN 00 CO I/O rf CN O O N C N C N rf od CN rf NO NO ON NO II • CN o u 00 m O O CN m C N ^ CN CN L T3 O C N rf r~ C N C N C N od II C N II C N rf rf '~ H c c ( N II c od NO m CN II rf c E o CO -a e CO ON C N rf 00 <N ON C N rf CN o CN Q. R C s •-o CU Q © A. A eg N "es u 3 Z o « s s s on CN O u O 2 CO T3 C CO I C/3 NO i/-> NO ,—i II 00 rf' S 00 ON oo CN rf ON n II ON CN ir> ci CN II od rf C CN C CN ON ON C ON 00 rf oo O C N 3 PQ C N i pq 3 PQ 3 PQ CN I ^! 3 03 CN NO ON rf NO 3 PQ 00 3 PQ on co u CD D CO > C > '5b -o N CO c CO <u O (U X) E 3 C 138 material of Bulk 3-22 where the standard method and modified method produced equivalent estimates of the NP. The other two exceptions both occurred with post-kinetic testing material. The modified method provided slightly higher estimates of the N P than the standard method for both Bulk 3-2] and Bulk 3-22. Also evident from Table 3-11 is that the neutralization potential determined by all three methods decreased as the sample progressed from its original state through to the end of kinetic testing. The only exception to this trend was for Sample 2/3 where the N P increased slightly from 78.9 kg C a C 0 3 / t (post-batch leach) to 81.9 kg C a C 0 3 / t (post-kinetic testing). The back titration curves provided in Appendix F indicate that the NPs determined by the standard and modified methods were converging (ie. getting closer in value to each other) as the experiment progressed. Other Observations The results from the N P determination were plotted against the calcium contents for the original material, the post-batch leach material, and the post-kinetic testing material sample to determine if there was a correlation between these two parameters. The results are presented in Appendix G - l to G-3. The post-batch leach material tended to have the lowest average correlation between N P and calcium content (R2=0.82) while the original material had the highest average correlation (R2=0.89). The standard method produced the lowest correlation for all three materials (R2=0.73, 0.59, 0.79, respectively) while the modified method produced the highest correlation (R2=0.90, 0.82, 0.85, respectively). 139 Comparison Between Previous Data and Current Data It was originally intended to provide a comparison of the data collected during the 1990 study with the data collected during the current research. It was decided that this comparison would not be feasible for two reasons: 1) there is a relatively high degree of dump heterogeneity (due to particle segregation, compaction, etc.), and 2) the exact locations of the current sampling sites were not surveyed in so their position relative to the earlier sampling sites (drillholes) is not known. Without knowing the exact position of the current sites in relation to the previous sites, any comparison conducted would be done using "best guesses" for the locations of these sites. Due to dump heterogeneity, even if the position of the current sampling site had been surveyed and was only one or two meters away (horizontally or vertically) from the previous site it was being compared to, then one could be analyzing completely different material, thereby rendering any comparison useless. 3.4.3 Summary of Static Prediction Tests The results of the various analyses used to conduct the static prediction tests are summarized in Table 3-12. The value provided for the acid-generating potential (AP) was determined using the modified method and the value provided for the neutralization potential (NP) was the average between the modified acid-base accounting method and the carbonate analysis. 140 Table 3-12: Summary of Static Prediction Tests Original Material Sample Sulphur so 4 Sulphide AP NP NNP NP:AP (%) (%) (%) (kgCaC03/() (kgCaC03/t) (kg CaC03/t) 1 4.44 0.5 3.94 123.13 37.45 -85.68 0.30 2/3 3.42 0.19 3.23 100.94 66 -34.94 0.65 Pit 1/2 3.96 0.62 3.34 104.38 21.25 -83.13 0.20 Dike 1/2/3 2.05 0.16 1.89 59.06 9.35 -49.71 0.16 Bulk 1 1.75 1.04 0.71 22.19 22.25 0.06 1.00 Bulk 3-2 5.51 2:05 3.46 108.13 6.65 -101.48 0.06 Bulk 5 1.86 0.26 1.6 50.00 42.9 -7.10 0.86 Bulk 8 2.62 1.15 1.47 45.94 11.15 -34.79 0.24 Post-Batch Leach Material Sample Sulphur so 4 Sulphide AP NP NNP NP:AP (%) (%) (%) (kg CaC03/t) (kg CaC03/t) (kg CaCO,3/t) 1 4.61 0.067 4.543 141.97 36.25 -105.72 0.26 2/3 2.47 0.113 2.357 73.66 36.6 -37.06 0.50 Pit 1/2 2.74 0.247 2.493 77.91 15.5 -62.41 0.20 Dike 1/2/3 2.05 0.149 1.901 59.41 7.65 -51.76 0.13 Bulk 1 1.23 0.282 0.948 29.63 17.25 -12.38 0.58 Bulk 3-21 3.11 0.403 2.707 84.59 6 -78.59 0.07 Bulk 3-22 4.35 0.605 3.745 117.03 4.9 -112.13 0.04 Bulk 5 1.51 0.2 1.31 40.94 37.55 -3.39 0.92 Bulk 8 1.81 0.292 1.518 47.44 9.95 -37.49 0.21 Post-Kinetic Testing Material Sample Sulphur so 4 Sulphide AP NP NNP NP:AP (%) (%) (%) (kg CaC03/t) (kgCaC03/ l ) (kg CaC03/ l ) 1 5.27 0.01 5.26 164.38 35.35 -129.03 0.22 2/3 2.7 0.01 2.69 84.06 35.2 -48.86 0.42 Pit 1/2 2.84 0.1 2.74 85.63 13.5 -72.13 0.16 Dike 1/2/3 2.15 0.08 2.07 64.69 4.35 -60.34 0.07 Bulkl 1.22 0.08 1.14 35.63 13.8 -21.83 0.39 Bulk 3-21 3.01 0.26 2.75 85.94 5.55 -80.39 0.06 Bulk 3-22 4.35 0.41 3.94 123.13 3.95 -119.18 0.03 Bulk 3-23 3.03 1.52 1.51 47.19 4.45 -42.74 0.09 Bulk 3-24 4.31 1.04 3.27 102.19 4.35 -97.84 0.04 Bulk 5 1.48 0.02 1.46 45.63 30.6 -15.03 0.67 Bulk 8 1.84 0.15 1.69 52.81 5 -47.81 0.09 141 From Table 3-12 it is obvious that the material from the Northwest Dump was acid-generating. Although some samples were less acid-generating than others, overall it appears as though the dump was a net acid-generator. Nearly all of the samples had negative net neutralization potentials (NNPs) and nearly all of the samples had NP:AP ratios that were less than 1. The only exception to these trends was the original material of Bulk 1. This sample had an N N P of +0.06 kg CaC0 3 /tonne of material and an NP:AP ratio of 1.003. With the exception of the original material of Bulk 1, Bulk 5 was the least acid-generating sample. The original material for this sample had a slightly negative N N P (-7.10 kg CaC0 3 / t ) and an NP:AP ratio slightly below 1 (0.86). The batch leached material of this sample was apparently less acid-generating than the original material because both the N N P and NP.AP ratio were higher (-3.39 kg C a C 0 3 / t and 0.92, respectively). The post - kinetic testing material was substantially more acid-generating than the other stages with an N N P of-15.03 kg C a C 0 3 / t and an NP:AP ratio of 0.67, but it was still the least acid-generating of all the samples. Since the A P and NP were calculated for both the original and post-kinetic testing material, a number of factors can be calculated for the samples. Due to potential errors in the A P calculations, only the N P will be examined. The N P depletion rate, percent N P remaining at the end of kinetic testing, and the time to deplete N P in years can all be calculated from the information in Table 3-12 and are presented in Table 3-13. These factors will be compared later on to the same factors calculated from the kinetic test data. 142 TABLE 3-13 DEPLETION RATE, % N P REMAINING AND TIME TO DEPLETE N P Sample Deple tion Rate % N P Remaining Time to Deplete NP (years) (kg CaC03/t/week) (mg CaC03/kg/week) 1 0.056 56.0 97.5 12.3 2/3 0.088 88.0 96.2 8.1 Pit 1/2 0.125 125.0 87.1 2.4 Dike 1/2/3 0.206 206.0 56.9 0.7 Bulk 1 0.216 216.0 80.0 1.5 Bulk 3-2] 0.028 28.0 92.5 4.1 Bulk 3-22 0.059 59.0 80.6 1.6 Bulk 3-23 0.138 138.0 66.9 0.9 Bulk 3-24 0.144 144.0 65.4 0.9 Bulk 5 0.434 434.0 81.5 1.7 Bulk 8 0.309 309.0 50.3 0.6 The depletion rate was calculated by subtracting the N P of the post-kinetic testing material from the NPof the post-batch leach material and dividing by the 16 weeks of kinetic testing. In order to allow comparisons with the kinetic test data, the depletion rate was then converted into mg CaC0 3/kg/week. The percent N P remaining at the end of kinetic testing was calculated by dividing the NP of the post-kinetic testing material by the N P of the post-batch leach material. The time to deplete N P was calculated by taking the total number of weeks required to reach 0% N P remaining and dividing by 52 to convert into years. The depletion rates ranged from 28 mg CaC0 3 /kg/week (Bulk 3-2^ up to 434 mg/kg/week (Bulk 5). A l l of the samples had 50% or more of their N P remaining, but there was quite a wide range (50.3% for Bulk 8 up to 97.5% for Sample 1). The time to deplete NP was relatively short (<2 years) for most of the samples, although Sample 1 143 and Sample 2/3 had much longer times than the other samples (12.3 and 8.1 years, respectively). Since similar factors were not calculated for A P , it is unknown how the time to deplete NP compares to the time required to deplete AP. 3.5 KINETIC PREDICTION TESTS The following section presents the results of the kinetic testing conducted on the Island Copper samples. General observations will be presented, followed by major parameters of interest such as pH, conductivity, alkalinity/acidity, sulphate, molar ratios, zinc, and copper. The raw data from which the following figures were generated can be found in Appendix H - l . 3.5.1 GENERAL OBSERVATIONS In terms of the actual test results, examining the data reveals the presence of two distinct groups of samples: Group 1) Sample 1, Sample 2/3, and Bulk 5, and Group 2) Bulk 3-23, 3-24, and occasionally Dike 1/2/3. In almost every case, one or the other of these groups always stands out from the rest of the samples. For the parameters of pH, alkalinity, time to deplete A P , and time to deplete metal, Group 1 is the distinct group, while for parameters such as conductivity, acidity, sulphate production, time to deplete NP, and metal production, Group 2 is the distinct group. 144 3.5.2 p H The p H of the sample leachate is presented in Figure 3-4. From Figure 3-4 it is immediately apparent that there are two different groups of samples: one that remained neutral from the start and one that remained acidic. Sample 1, Sample 2/3, and Bulk 5 are in the neutral group while the remaining samples comprise the acidic group. The average p H of the neutral group ranged from 6.95 (Week 1) to 7.7 (Week 7) and it appears that the p H of these samples was slowly increasing over time. It is unclear as to what caused the sharp increase in p H seen in Week 7 for all three of these samples. The pH of the acidic group was quite variable, ranging anywhere from 2.8-5.1 and averaging approximately 3.8. Due to the wide variation in pH, it is hard to say i f the p H of this group was increasing or decreasing over time. On an individual basis, it appears as though the p H of Pit 1/2, Bulk 1, Bulk 3-22, and Bulk 8 was increasing slightly over time, while the p H of Dike 1/2/3 was decreasing slightly. 3.5.3 Conductivity Figure 3-5 presents the conductivity measured over the course of the kinetic testing. Bulk 3-23 and 3-24 stand apart from the other samples because they consistently had the highest conductivity. This is true throughout testing, although the conductivity of Bulk 3-23 began to drop significantly after Week 8. Starting in Week 3, Dike 1/2/3 also maintained a relatively high conductivity. The remaining samples had quite low conductivities, especially from Week 4 onwards. It appears that the conductivity was 145 146 147 decreasing over time for all the samples, although one (Bulk 3-23) was decreasing at a much greater rate than any of the others. 3.5.4 Alkalinity/Acidity Depending on the p H of the weekly leachate, the alkalinity or acidity of the leachate was determined. If leachate p H was 5 or higher, then alkalinity was determined for that sample, i f the p H was less than 5, acidity was determined. The decision to determine if alkalinity or acidity should be measured was based on the Island Copper Methods Manual which indicated acidity should only be measured on samples with a p H of 5 or less. 3.5.4.1 Alkalinity Due to equipment and supply problems, alkalinity was measured starting in Week 4. Only three of the eleven samples had a leachate p H that was over 5 for the duration of kinetic testing: Sample 1, Sample 2/3, and Bulk 5. Therefore, these were the only samples measured for alkalinity (Figure 3-6). A l l three samples followed a similar trend. They exhibited a marked increase in alkalinity until Week 7 or 8 which was followed by a sharp drop in Week 9 and then a slow decrease until the last week of testing. The alkalinity throughout the experiment was relatively low, ranging from 10-20 mg C a C 0 3 / L , although there were two or three weeks (Weeks 6-8) where alkalinity reached as high as 27 mg C a C 0 3 / L . 148 3.5.4.2 Acidity Acidity was not measured until Week 5 due to equipment and supply problems. The acidity measured in the eight samples is presented in Figure 3-7. In general, after Week 7, it appears as though the acidity levels are decreasing in all of the samples. As observed before with other parameters, there are two distinct group of samples. The samples with much higher acidity levels than the other samples are Dike 1/2/3 and Bulk 3-24. The remaining six samples all have relatively low acidity levels, and i f Bulk 3-2! and 3-23 are excluded, this group actually had acidity levels that were less than 50 mg C a C 0 3 / L throughout kinetic testing. 3.5.5 Sulphate The weekly sulphate production was calculated because this standardizes the samples by incorporating both the leachate volume per week and the mass of the sample. The data were used to calculate the cumulative sulphate loading, the sulphide oxidation rate, the percent sulphide remaining, and the time required to deplete the acid-generating potential. 3.5.5.1 Cumulative Loading The cumulative loading of sulphate over the course of kinetic testing is presented in Figure 3-8a. The highest cumulative levels of sulphate in all of the samples were produced by Bulk 3-23 and 3-24 and the lowest levels were produced by Sample 1, Sample 2/3, and Bulk 5. Interesting to note from Figure 3-8a is that while most of the 150 151 CO ^ CM CO •<*• CM CNJ CN CN CN T- CO CO CO CO oo <D J£ CO J£ 3 3 3 3 3 3 c\i b. Q CO m CO 00 CO CO CO < • O • + O • 1 < X • • O •••)<•( O » •> OR <3Kl ° <«- i < i l « ^ i * i -4 I < I o >i o 00 1 g -3 o CD _ > J3 3 u _g c/i <U H c 2 OS OO 0) l -3 E o o o o CO o CD o o o (B)|/Biu) 6u;peo-| a}ei)d|ns 152 samples appeared to have reached a plateau and were leveling out, Bulk 3-24 was increasing linearly at approximately 840 mg S0 4 /kg of sample every week. Two other samples, Bulk 3-2 3 and Dike 1/2/3, also increased linearly near the beginning, averaging 720 mg/kg per week and 300 mg/kg per week, respectively. Bulk 3-23 began to level out after Week 9 while Dike 1/2/3 began to level out after Week 11. The remaining samples increased only to Week 5, when they were averaging anywhere from 50-200 mg S0 4 /kg per week, before leveling out and averaging only 10-50 mg/kg per week. 3.5.5.2 Sulphide Oxidation Rate The sulphide oxidation rate (Figure 3-8b) is an indication of how quickly the sulphide is being used up. It is immediately obvious from Figure 3-8b that the oxidation rate of most of the samples is not constant throughout testing. With the exception of Bulk 3-24 and Dike 1/2/3, the samples experienced a steady decrease in the sulphide oxidation rate as testing proceeded. While Bulk 3-24 and Dike 1/2/3 maintained more-or-less the same sulphide oxidation rate after Week 2 for the duration of testing at around 800 mg/kg/week and 250 mg/kg/week respectively, the remaining samples experienced decreases in the sulphide oxidation rate of approximately 50% from Week 1 to Week 16. Following the same trend seen in the other figures, Sample 1, Sample 2/3, and Bulk 5 have the lowest sulphide oxidation rates. 153 CO „_ CM CO CM CNI CM CM CM CO CO CO CO m oo ^£ CO .t; 3 3 3 3 3 D. b m m CQ III CD CO m < • o • + • 1 •4 X • • o • o • o I • o • • o • ot < I • • -» OB < » Cl +48 • < » 0) OO £ o > O M « 0) &o <u 00 c • > 0 £ </-> 1 (L) ts c o 'is T3 ' * o 0) GO _c m <U H c oo m o o o o CM O o o o o CD o o o o CM (>|3aM/6)|/6iu) ajey uo;)ep;xo ap!qd|ns 154 3.5.5.3 Percent Sulphide Remaining Using the sulphide oxidation rate, one can calculate the amount of sulphide remaining in a sample (Figure 3-8c). From Figure 3-8c it is obvious that Bulk 3-23 and 3-24, the non-batch leached replicates, had the least amount of sulphide remaining by the end of kinetic testing (around 99.92% and 99.87%, respectively). As one would expect, Sample 1, Sample 2/3, and Bulk 5 had the most sulphide remaining of all the samples (99.99%). The rest of the samples had anywhere from 99.94 - 99.97% of their original sulphide remaining. The significant loss of sulphide in Bulk 1, relative to the other samples, is surprising as this sample did not have particularly high levels of sulphate production nor did it have a particularly high rate of sulphide oxidation. 3.5.5.4 Time to Deplete A P Using the sulphide oxidation rate and the amount of sulphide remaining in a sample, one can calculate the time needed to deplete the acid-generating potential (AP) of a sample (Figure 3-8d). As one would expect, those samples that had the lowest sulphate production levels and sulphide oxidation rates will require the longest time to achieve A P depletion. In this case they are Sample 1, Sample 2/3, and Bulk 5. Using the last three weeks of data to produce an average, these samples will require approximately 140 years to deplete their acid-generating potential. Pit 1/2, and surprisingly Bulk 3-22, will also take a relatively long time to deplete their A P (70 years and 60 years, respectively). The non-batch leach replicates (Bulk 3-23 and 3-24) will need only a short 155 CO CM CO •>J-CN CM CM CM CM CM t- CO CO CO CO m oo <u J£ j£ CO 3 3 3 3 3 3 CM Q_ Q 03 03 03 03 03 03 03 < • o • + o • 1 4 X • «x o • «x o • • o «x o « • o «x o « • o «x o • «x o • «x o « • o «x o « • o + • + • + • + • +• m D f «x o « • o • + «x o « a o • + i < I -4 0) i a E .2 o O «X O O K > • + I 4 «K O W • + I 4 • + 1 4 4 K O K I + 14 + N C '3 £ a H u C b oo E + M cn oS o> cn cn 6u;u|euj3^ apiudjns % 156 > CM CO TJ -CM CNI CM CM CM C M C L Q C Q C Q C Q C Q C Q C O C Q (sjeaA) 3iu; i uofjaidaa 157 amount of time (~8 years) to reach A P depletion. Dike 1/2/3 will need slightly more time at around 10 years and Bulk 3-2! will need about 15 years to achieve A P depletion. The remaining samples (Bulk I and Bulk 8) will need approximately 20 years before their A P is completely used up. 3.5.6 Molar Ratios Various molar ratios can be calculated which provide information on the oxidation and neutralization reactions which are occurring in the sample. The two most common ratios are the carbonate:sulphate ratio and the silicate:sulphate ratio, both of which were calculated for the kinetic test data. Only calcium and magnesium were used for the carbonate ratio because the concentrations of strontium and barium were negligible in comparison to the calcium and magnesium concentrations of these samples (Table 3-5) and therefore were not measured in the weekly leachate. 3.5.6.1 Cumulative NP Produced Plotting the cumulative N P produced (Figures 3-9a and 3-9b) can provide an indication of the rate at which various components of the N P are going into solution. The graphs of cumulative N P produced using both the carbonate ratio and the silicate ratio are almost identical. In both cases, Bulk 3-23 and 3-24 produced the most N P of any of the samples and Sample 1, Sample 2/3, and Bulk 5 produced the least. The rate of NP production from Weeks 1-10 in Bulk 3-23 and 3-2 4 was approximately 750-800 mg C a C 0 3 / k g per week. After Week 10, the rate of NP production for these samples dropped 158 S2 i- CM co CM CM CM CNI CM c M Q - O m m c a c a m m f f l m CM i n •<- T> o CM ^ O o|jey JB|0|/\| tOS:aiBuoqjBQ 159 160 to approximately 250 mg/kg per week. The rate of NP production for Sample 1, Sample 2/3, and Bulk 5 was approximately 20 mg C a C 0 3 / k g per week and remained relatively constant throughout testing. The NP production rate using both ratios was approximately 50 mg C a C 0 3 / k g per week for the remaining samples. 3.5.6.2 N P Depletion Rate The NP depletion rate (Figures 3-10a and 3-10b) is an important value because it provides an indication of how quickly the neutralizing minerals are being consumed. For the carbonate ratio, the N P depletion rate of Bulk 3-23 and 3-24 for the first two weeks was relatively the same at around 775 mg/kg/week, followed by a sharp increase up to 1050 mg/kg/week. For the silicate ratio, however, there was a continual increase from Weeks 1-3 when the NP depletion rate more than doubled from 450 mg/kg/week to over 950 mg/kg/week. From Week 3 onwards, there was a steady decrease in the NP depletion rate for both samples looking at both ratios. By Week 16, using the last five weeks of data, the NP depletion rate for Bulk 3-23 was around 350 mg/kg/week and for Bulk 3-24 was around 575 mg/kg/week. Looking at the other samples, Sample 1, Sample 2/3, and Bulk 5 had the lowest NP depletion rates of all the samples. For both ratios, these samples averaged about 90 mg/kg/week in Week 1 and slowly decreased to Week 16 when they averaged approximately 20 mg/kg/week. The remaining samples had an average N P depletion rate CM CM CM CM CM ^ c M Q - o m m c o m m m m H + + + 1 H + ())a3«/B>|/coOB0 Bui) a»ey uojjaidaa 162 o o o o o o o o o o o o o o o > c o r * - to t o ^ - c o ()jaaM/6>|/eoOB0 B u 0 uo!ja|deQ 163 from both ratios of approximately 150 mg/kg/week in Week 1 which steadily decreased to approximately 50 mg/kg/week by Week 16. Compared to the NP depletion rates calculated from the acid-base accounting data, the depletion rates calculated from the kinetic test data are much lower. The only rates that are comparable are those for Bulk 3-22. The rate calculated from the A B A data was 59 mg CaC0 3 /kg/week while from the kinetic test data it was approximately 35 mg/kg/week. Only three samples had kinetic test rates that were higher than A B A rates. Bulk 3-2 1 ; 3-23, and 3-24 all had depletion rates that were 2-4 times higher than the rates calculated from the static testing. 3.5.6.3 Percent NP Remaining The percent NP remaining (Figures 3-1 la and 3-1 lb) is directly related to the cumulative N P produced and this fact is evident when one compares Figures 3-9 and 3-11. These figures are almost identical mirror images of each other. In general, there was a consistent decrease in the amount of N P remaining in each sample. It is not surprising that the samples with the highest cumulative NP production (Bulk 3-23 and 3-2 4)were the samples with the least amount of N P remaining. Sample 1, Sample 2/3, and Bulk 5 had the most NP remaining of all the samples (approximately 98%). The remaining samples had anywhere from 80-95% of their original NP remaining by the end of kinetic testing. 165 E2 <- CM n • * C M C M C M (SI C M C M I - T - C O C O C O C O I O O O C > ± i : S : 3 3 3 3 3 3 3 r - C M C L O C O C Q C Q C Q C Q C Q C Q u E H _o u o c3 Pi a 0 bo _c '£ e CD 1 +-* C <u o v-<L> P H i -=3 (so E 166 Compared to the values calculated from the A B A data, the percent of N P remaining in the samples as calculated from the kinetic test data is comparable for some samples (eg. Sample 1, Sample 2/3), higher for others (eg. Dike 1/2/3, Bulk 1), and lower for others (eg. Bulk 3-2 b 3-23, 3-24). No trend is evident that suggests the kinetic rates are consistently higher or lower than the rates calculate from the static tests. 3.5.6.4 Time to Deplete N P As mentioned above, the time required to use up all the neutralizing minerals in a sample is important because after that time, if acid is still being generated, there will be unregulated release of acid to the receiving environment. Figures 3-12a and 3-12b show the time that will elapse before the NP is completely depleted based on the carbonate and silicate molar ratios. In general, the time required to deplete the NP in the samples is increasing. The only exception to this trend is for Bulk 3-23 and 3-24, which according to Figures 3-1 l a and 3-1 lb have no N P remaining. Sample 2/3, Pit 1/2, and Bulk 5 will require the most amount of time to deplete their N P (12, 11, and 22 years, respectively). With the exception of Bulk 3-23 and 3-24, Dike 1/2/3 and Bulk 3-21 will require the least amount of time to deplete their NP (approximately 3 years). Sample 1 will need around 10 years to deplete its N P and the remaining three samples (Bulk 1, Bulk 3-22, and Bulk 8) will need approximately 6 years to completely deplete their NP. 167 168 169 When the time required to deplete NP is compared to the time required to deplete A P (Figure 3-8d), it is obvious that the samples will use up their neutralizing minerals long before they deplete their sulphide minerals. It looks as though there is almost an order of magnitude difference in the time required to deplete A P compared to the time required to deplete NP. Compared to the time to deplete N P calculated from the static test data, the time to deplete NP as calculated from the kinetic test data is significantly longer for most of the samples. The time to deplete based on the silicate ratio was slightly lower than the carbonate ratio, but both were still much higher than the time to deplete N P as calculated from the static test data. The only samples that have comparable times between the two are Sample 1 and Sample 2/3. Based on the static test data these samples had times to deplete NP of 12.3 and 8.1 years, respectively, while using the kinetic test data, these two samples both had a time to deplete NP of approximately 10 years. The biggest difference between the static test and kinetic test times to deplete NP is for Bulk 5, which had a time to deplete NP from the static testing of only 1.7 years while from the kinetic testing it had a time to deplete N P of approximately 20 years. 3.5.7 Zinc Zinc is probably the primary metal of concern regarding water quality at Island Copper. Copper is also an important metal in terms of water quality at the mine. Therefore these metals were analyzed in more detail than the other metals measured in 170 the leachate. The cumulative loading, the leaching rate, the percent metal remaining, and the time to deplete zinc are presented below. 3.5.7.1 Cumulative Loading and Metal Leaching Rate The total amount of zinc released into solution over time is presented in Figure 3-13a. Bulk 3-23 and 3-24 put the most zinc into solution, followed by Bulk 3-21 and Dike 1/2/3. The samples that released the least amount of zinc into solution were Sample 1, Sample 2/3, and Bulk 5. As with the other parameters, the leaching rate of zinc (Figure 3-13b) can be calculated from the slope of the cumulative loading curve. For most of the samples the leaching rate was highest during the first week and it slowly decreased as the testing progressed. The highest leaching rates were during Week 1 for Bulk 3-23 and 3-24 when zinc was leached at approximately 23 mg/kg of sample per week. By the end of kinetic testing, however, the leaching rate of these samples was down to approximately 3 mg/kg per week. Sample 1, Sample 2/3, and Bulk 5 had the lowest leaching rates at approximately 0.1 mg of zinc/kg of sample per week throughout kinetic testing. The rest of the samples had leaching rates that ranged from 1.0-2.5 mg of zinc/kg of sample per week. 171 C O C M C O C M C M C M C M C M T— C O C O <o C O m 0 0 _^ CD J £ JX. C O ± 2 3 3 3 3 3 3 3 C M CL b CO CO m m CO CO CO < • O • + O • 1 4 X • - • e i -• • aor-4 i 4 i • n«a-4 i • B«a--4 I 4 I 4 I 4 I •4 I < I O >f 4 I O B - • * 4 I OSH3 » o C M (6>|/6iu) Bu;peo-| ou|7 CO _^ CM CO •* CM CM CM CM CM CO CO CO CO in CO CO T— <u .t; 3 3 •<— CL Q ffl m CQ CQ CQ CQ CQ < • O • + O • 1 •4 X • < I OdB< ^ i on i <\ o • ^ on •41 on 4 1 on •41 o • O • ! 4 oo < I O M D M u> o m m < i 14 «a-ci * CM • * m o ()|33M/6)|/Bui) ajey 6u;upe3-i 173 3.5.7.2 Percent Metal Remai ning The amount of metal that is remaining in a sample (Figure 3-13c) is important because it provides information on how much metal is available for release into the environment. From Figure 3-13c it is obvious that most of the samples retained greater than 99.98% of their original zinc. Sample 1, Sample 2/3, and Bulk 5 retained almost 100% of their original zinc as would be expected based on their loading and leaching rates. The only sample that lost significant amounts of zinc was Dike 1/2/3, which by Week 16 of kinetic testing had approximately 99.76% of its original zinc content remaining. 3.5.7.3 Time to Deplete Metal Using the metal leaching rate and the amount of metal remaining, it is possible to calculate the time it will take for the sample to be completely depleted of a metal. Figure 3-13d presents this information for the Island Copper samples. Sample 1, Sample 2/3, and Bulk 5 had relatively constant depletion times throughout testing, while most of the remaining samples required an increasing amount of time to deplete their zinc contents as testing progressed. The exception to this trend was Dike 1/2/3, which will deplete its zinc in only 2 years. Sample 1, Sample 2/3, and Bulk 5 will take approximately 1600 years, 2600 years, and 700 years, respectively to deplete their zinc content. Based on the last five weeks of data, Pit 1/2, Bulk 1, and Bulk 3-22 will take from 200 to 400 years to use up all their zinc. The remaining samples (Bulk 3-2 h 3-23, 3-24, and Bulk 8) will require approximately 30 to 40 years to totally deplete their zinc content. 174 CO , CM CO •* CN CN CM CM CM CM T— CO CO CO CO 00 ^~ CD JXL JXL CO +-* 3 3 3 3 3 3 3 T— Si a. b m m m m m CO CD < • O • + O • 1 4 X • 175 CO CM CM CM CO CM CM CM CO CO CO CO IT) «*' CO Pit 1, Dike Bulk Bulk Bulk Bulk 3 3 m m < • O • + O • 1 -4 X • I o • + o • o» n o o o o o o o CM O O (saeaA) auijj. uonaidaQ 176 3.5.8 Copper Copper is the second metal of concern regarding water quality at the Island Copper Mine. The same analysis conducted for zinc was conducted for copper. As before, the cumulative loading, the leaching rate, the percent metal remaining, and the time required to deplete the copper from the samples are provided below. 3.5.8.1 Cumulative Loading and Metal Leaching Rate One can see from the cumulative loading (Figure 3-14a) that Bulk 3-24 put significantly more copper into solution than any of the other samples. Unlike most of the other trends observed (eg. sulphate, zinc) where the cumulative loading was highest at the beginning and then decreased, the cumulative loading of copper for Bulk 3-24 was lowest at the beginning and then increased. Dike 1/2/3 also released relatively large amounts of copper. The neutral samples (Sample 1, Sample 2/3, and Bulk 5) released the least amount of copper into solution. As with zinc, the leaching rate of copper (Figure 3-14b) can be calculated from the cumulative loading curve. As expected, high leaching rates were observed in Week 1, but surprisingly, the leaching rate of some of the samples actually increased in later weeks. Mid-way through the testing, Dike 1/2/3 had a maximum leaching rate of just over 3 mg/kg/week, while Bulk 3-24 had its second highest leach rate of just under 6 mg/kg/week. The other samples had more-or-less constant leaching rates throughout or 177 C O C N C M C M C O C M •* C M C N T- C O C O C O C O C O CD C O Pit b Bul Bul Bul Bul Bul Bul Bul < • O • + O • 1 < X • • • I o I o I o I o I o I o I o • o • • I o • • I o • < I o (6>|/6UJ) Bujpeoi Jaddoo 178 CO _^ CNI CO c\i CNI CNI CNI CNI CNI CO CO CO CO IT) oo <I> J*. JX. co +-> 3 3 3 3 3 3 CN Q. b m m co CO co m CO < • O • + O • 1 4 X • - B — a t • IO • IO • o • Ol Cl O I Ol Ol • 4 & Q . O «• u 0) 1 | Pi o I • 4 o Ol • 4 • o • < • i « o •> CO CD tl-()|33M/B)|/6iu) ajey 6u!upea-| 179 steadily decreased as testing progressed. By the end of kinetic testing, Bulk 3-24 had an average leaching rate of 5 mg/kg/week, Dike 1/2/3 averaged approximately 2 mg/kg/week, Bulk 3-2] and 3-23 both averaged around 1 mg/kg/week, and the remaining samples averaged from 0.1-0.3 mg/kg/week. 3.5.8.2 Percent Metal Remaining The amount of copper remaining in a sample (Figure 3-14c) is directly related to the amount of copper that has been leached into solution. O f all the samples, only Bulk 3-24 had lost an appreciable portion of its original copper by Week 16 (0.1%). Three other samples (Dike 1/2/3, Bulk 3-2 l s and 3-23) had lost around 0.03% of their original copper. The remaining samples were divided into two groups, both of which had over 99.99% of their original copper left. One group, consisting of Bulk 1, Bulk 3-22, and Bulk 8, had approximately 99.99% of their original copper content, while the other group, consisting of Sample 1, Sample 2/3, Pit 1/2, and Bulk 5, had lost less than 0.005% of their original copper. 3.5.8.3 Time to Deplete Metal The time required to completely deplete the copper in a sample (Figure 3-14d) is important in terms of understanding how long the material could be a problem regarding water quality. As seen with the time required for the sample to become depleted of zinc (Figure 3-13d), most of the samples during kinetic testing did not change in the time C O C M C O C M C M C M C M C M T— C O C O C O C O in 0 0 C O T— CD jx: ^ ±1 -id 3 3 3 3 3 3 C N a. b C Q 03 m m 03 C Q C Q < • • • + o • 1 4 X • - W B -« X «x O • on O D O • o c « K - W « O • M K - W * O • ! M C 4 « « O D I mmm O D I • K t - a « o a i 1 ^ I 8 01 CD .i PH H w v ao o c >> +3 I 4 I 4 oo r~ Buiuiewau |eja|/\] % 181 CO ,_ CM CO CM CM CM CM CM CO CO CO CO If) 0 0 T— <u JX. J£ ^ JO 3 3 3 3 3 CM CL b m m CQ CO CO CO CO < • o • + O • 1 4 X • -m—*r -cx ex CW o< CX o < o< CM CX o < O < • O < o < • o < •X • X X • + + • B> -41 CO 4 ID -44 148 41 I 03 l<H4l I <E4l + • • I CQ4 I + • « C 0 4 + • 4 0 D - 4 • « 4 + • Ol M o o CM (sjesA) aiuii uoj)3|daQ 182 required to become depleted of copper. As one would expect, Bulk 3-24 will require the shortest amount of time to fully deplete its copper content, approximately 4 years. Dike 1/2/3, Bulk 3-2! and 3-23 will require slightly more time at around 15 years. Bulk 1 will need just over 40 years, and Bulk 5 will require approximately 70 years. The curves for Bulk 3-22 and Bulk 8 are the only ones which displayed significant changes. Based on the last five weeks of testing, these samples will need about 70 and 55 years, respectively. Sample 1 and Pit 1/2 will both need approximately 140 years to completely deplete their copper content while Sample 2/3 will require the longest amount of time to fully deplete its copper content, around 160 years. 3.5.9 Other Relationships Some additional relationships were examined, including conductivity versus sulphate, the sulphate:alkalinity or acidity ratio, the calcium:magnesium or sodium ratio, and the calcium, magnesium, sodium, and potassium:sulphate ratio. Only conductivity vs. sulphate and the sulphate:alkalinity or acidity ratio are presented below. The other relationships listed above are provided in Appendix H-2 and H-3. 3.5.9.1 Conductivity vs. Sulphate Plotting conductivity versus sulphate can serve as a quick indicator of the sulphate concentration in the leachate being analyzed. The rationale is that sulphate is usually present in relatively high concentrations in kinetic test leachate and provides a 183 large amount of the T D S in a sample. Conductivity provides a measure of the T D S of a sample and so changes in sulphate concentration lead to changes in conductivity. Figure 3-15 shows the relationship between conductivity and sulphate concentration in the leachate. There is a very strong, linear correlation (R =0.97) between conductivity and sulphate in these samples. Removing the two outliers at the top right of the figure increases the correlation even further (R2=0.99). The trend line crosses the y-axis at approximately 0.1 mS/cm reflecting the fact that sulphate is not the only substance contributing to the T D S of the sample. 3.5.9.2 Sulphate:Alkalinity or Acidity Ratio The sulphate to alkalinity molar ratio (Figure 3-16) can be used as an A R D indicator. As acid generation increases, more sulphate is produced in proportion to alkalinity and the ratio increases. Since Sample 1, Sample 2/3, and Bulk 5 were the only samples measured for alkalinity, Figure 3-16 only includes these three samples. Sample 1 was the only sample to have a ratio that was greater than 1 throughout kinetic testing. Sample 2/3 had a ratio between 1 and 1.5 for most of the testing, but occasionally exhibited drops below 1. The ratio for Bulk 5 was less than 1 for most of the testing, but towards the end it increased to slightly over 1. Starting in Week 8, the ratio for Sample 1 appeared to be steadily increasing. Sample 2/3 exhibited wide fluctuations in the ratio and it is difficult to determine if the ratio was increasing or 185 I D T— CM GQ m c o i f ) C M i f > T - i o o C O C M O oijey je|0|/\| AJJUJIB^IV^OS 186 decreasing. The ratio for Bulk 5, although steadily decreasing from Weeks 4 to 12, appeared to be increasing from Weeks 12 to 16. Figure 3-17 presents the sulphate to acidity molar ratio. This ratio can be used to determine the extent of buffering that is occurring within a sample. The ratio was greater than or equal to one for all of the samples throughout kinetic testing. The ratio appeared to be decreasing as testing progressed for Bulk 3-23, 3-24, and Pit 1/2, increasing for Bulk 3-2! and Bulk 8, and holding relatively steady for the remaining three samples (Dike 1/2/3, Bulk 1, and Bulk 3-22). 3.5.9.3 Other Ratios The calcium:sodium and calcium magnesium ratios can be used to determine i f preferential leaching of neutralizing minerals is occurring. If the ratio is equal to one than no preferential leaching is occurring. If the ratio is greater than one, than there is preferential leaching of calcium (ie. carbonates, in this case calcite). If the ratio is less than one, there is preferential leaching of sodium (ie. silicates, in this case Na-feldspars such as albite) and/or magnesium (ie. silicates such as biotite, or carbonates such as magnesite). Both of the ratios for all of the samples were well above 1 throughout kinetic testing, indicating that calcium was being preferentially leached over magnesium and sodium. The calcium: magnesium ratio appeared to be steadily decreasing as testing progressed, but it never dropped below 1. 187 188 The calcium, magnesium, sodium, and potassium:sulphate ratios provide information on the chemistry occurring within a sample and can be used to characterize oxidation and dissolution reactions. The calcium and magnesium:sulphate ratios are related to carbonate dissolution reactions, and the calcium, sodium and potassium:sulphate ratios are related to the dissolution of alkali feldspar and other silicates. Carbonate dissolution (particularly calcite) was definitely occurring in Sample 1, Sample 2/3, and Bulk 5 (Ca:S0 4>l). The remaining ratios (Mg, Na, K : S 0 4 ) were all less than 1 for all the samples throughout testing indicating that sulphide oxidation was dominating the dissolution of silicates. 3.5.10 Humidity Cel l Disassembly The disassembly of the humidity cells is described below. First the flowpath examination will be provided and then the determination of the mass of material lost during kinetic testing. 3.5.10.1 Flowpaths Examination of the humidity cell material failed to uncover any evidence of flowpaths. No deposits of secondary minerals (ie. hydroxides, sulphates) were seen, nor was there any sign of gypsum precipitation. No indications were given to suggest that flowpaths had formed in any of the samples. 189 3.5.10.2 Mass Loss The samples were weighed after being air dried and then oven dried. The results are presented in Table 3-14. From Table 3-14 it is obvious that there was very little mass lost (-1% on average) as a result of the kinetic testing. Bulk 3-23 and 3-24 had slightly higher mass loss compared to the other samples (1.2% and 1.1%, respectively). These samples had not been batch leached and still probably contained a lot of fine particles. Most of these fine particles would have been removed from the other samples as a result of the batch leach. 3.6 WATER QUALITY Various aspects of water quality were examined as part of this thesis. Groundwater quality was examined to determine i f it was contaminated. Data from Table 3-14 Sample Mass (Kinetic Testing) Sample Mass Before (g) Mass After (g) 1 900 893 2/3 971 966 Pit 1/2 919 915 Dike 1/2/3 974 970 Bulk 1 976 970 Bulk 3-2, 904 900 Bulk 3-22 867 861 Bulk 3-23 1016 1004 Bulk 3-24 1012 1001 Bulk 5 942 936 Bulk 8 896 890 190 groundwater and Francis Lake were examined to determine if excavation of the dump led to improvements in water quality. The effect of Northwest Dump material on the upper benches of the pit was examined to determine the potential water quality of the freshwater cap. 3.6.1 Groundwater Quality As a result of the equipment utilized during well installation (4' gas-powered hand auger), the wells were completed to a relatively shallow depth and therefore it is unclear as to how much of the water found within the wells was actually groundwater and how much was surface infiltration. The material that was excavated during installation was used to secure the well in place and reduce surface infiltration as much as possible. Therefore most of the water within the wells was probably intercepted groundwater, although tracer tests would probably have to be conducted to confirm this conclusion. For the presentation of the results and the discussion in Section 4, it is assumed that it was groundwater in the wells and the samples will be referred to as such. Due to the relatively dry summer experienced at Island Copper and the shallow depth of the groundwater wells, all of the wells had gone dry by July. This resulted in only nine samples being collected and therefore the analysis that could be conducted was somewhat limited. 191 3.6.1.1 p H Figure 3-18a shows the p H values obtained from the six wells for the period February - June, 1996. No p H was measured for the sample taken from Well G W W on June 19. Immediately apparent from Figure 3-18a is the difference between the p H of background groundwater (BLR, p H 8.5) and the p H of groundwater emanating from the North (UTP, p H 6.0-6.5) and the Northwest (GWW, L G W , P G W , P W W , pH 4.0-4.5) Dumps. While Well L G W had a higher pH than Well G W W (5.5 compared to 4.2), Well P G W had a lower pH than Well P W W (4.2 compared to 5.5). The pH of the groundwater taken from Well L G W appeared to be increasing slightly. The first few samples for Well L G W averaged approximately pH 5.0 while the last few samples had an average p H of around 6.0. The p H in the other wells appeared to be maintaining more-or-less the same levels. 3.6.1.2 Conductivity The conductivity measured from the different wells is displayed in Figure 3-18b. As with pH, conductivity was not measured in the sample from Well G W W . Wells B L R and P W W had relatively low conductivity (200-300 mS/cm) with minor fluctuations, while the rest of the wells had conductivities that were at least three times higher and varied widely. Groundwater taken from Well G W W had a conductivity ranging from 1600-2000 mS/cm and averaging around 1800 mS/cm, while the other three wells (PGW, U T P , and L G W ) ranged in conductivity from 600-1100 mS/cm and averaged approximately 900 mS/cm. The conductivity of the groundwater taken from Well L G W 192 GWW —B— LGW A PGW -e - PWW • BLR -~A- UTP 16-Jan Figure 3-18a: Groundwater Data (pH) Figure 3-18b: Groundwater Data (Conductivity) 193 appeared to be decreasing, but due to wide fluctuations in value it cannot be said with any certainty i f this decrease was actually occurring. The rest of the samples maintained relatively steady conductivity values after March 27. 3.6.1.3 Sulphate Figure 3-18c presents the sulphate values measured for the different samples. As was seen with conductivity, sulphate levels in Wells B L R and P W W were quite low (30-100 mg/L) while the rest of the samples had sulphate levels that were significantly higher. As would be expected for Well G W W , based on the conductivity values, the sulphate levels were much higher than for any of the other wells, averaging approximately 1250 mg/L (range 1000-1550 mg/L). The other three wells (PGW, L G W , and UTP) all had similar sulphate levels, ranging from 300-700 mg/L and averaging around 500 mg/L. The sulphate levels in Wells G W W , L G W , and P G W appeared to be decreasing, but were relatively steady in the remaining wells. 3.6.1.4 Calcium Calcium concentrations measured in the different wells are shown in Figure 3-18d. As seen with both conductivity and sulphate levels, Wells B L R and P W W had the lowest concentrations of calcium while Well P G W had the highest. A l l of the wells exhibited increased calcium concentrations in comparison to the background levels of approximately 10 mg/L (Well BLR) . Well G W W had the highest calcium concentrations, ranging from 240 - 430 mg/L and averaging around 325 mg/L. Wells 194 16-Jan 5-Feb 24-Jun • - G W W - B - - L G W A - P G W G - PWW • -BLR • -UTP Figure 3-18c: Groundwater Data (Sulphate) 16-Jan 5-Feb 4-Jun 24-Jun - B - G W W - O — L G W -A—PGW - O - P W W BLR -A— UTP Figure 3-18d: Groundwater Data (Calcium) 195 P G W , U T P , and L G W exhibited calcium concentrations that were all in the same range (100-200 mg/L) with an average of approximately 150 mg/L. Due to significant fluctuations, it is difficult to determine if the calcium concentrations were increasing or decreasing in the different wells, although Wells G W W and L G W appeared to be decreasing while Well P G W appeared to be increasing. 3.6.1.5 Zinc The zinc concentrations measured from the various wells are presented in Figure 3-18e. The only well with significantly elevated zinc levels was G W W , which averaged around 25 mg/L, although the last few measurements suggest that the zinc concentration in this well was decreasing. The other wells with elevated zinc were P G W , U T P , and L G W . These wells averaged approximately 2.0 mg/L and the zinc concentration did not appear to be changing. Samples taken from Wells P W W and B L R had zinc concentrations that were at or below the detection limits. 3.6.1.6 Copper Figure 3-18f shows the copper concentration in the different wells. The only well with significantly elevated levels of copper was G W W . The copper levels in samples from this well fluctuated widely, varying from 0.1 mg/L to almost 0.7 mg/L and averaging approximately 0.35 mg/L. The other wells all had extremely low copper levels, which for the most part were less than or equal to the detection limit of 0.01 mg/L. May 15 was the only sampling period when the wells exhibited an increase in copper 196 • GWW LGW A PGW O PWW • BLR UTP 16-Jan 5-Feb 4-Jun 24-Jun Figure 3-18e: Groundwater Data (Zinc) • - G W W - • - - L G W A - P G W O -PWW • -BLR -UTP 4-Jun 24-Jun Figure 3-18f: Groundwater Data (Copper) 197 concentration. The copper levels for this date varied from 0.01 mg/L up to as high as 0.05 mg/L. 3.6.2 WATER QUALITY COMPARISON 3.6.2.1 Groundwater The groundwater data presented above for Well G W W was compared to the groundwater data collected from the excavated pit in May and June of 1995, two months prior to the start of excavation of the Northwest Dump. The data from Well G W W was used because this well was completed immediately adjacent to the excavated pit. The source of groundwater in the pit and in Well G W W should be almost identical. The two sets of data are presented in Table 3-15. The values provided are the average value for all data points within the particular data set. TABLE 3-15 COMPARISON OF GROUNDWATER DATA Source p H Conductivity Sulphate (mS/cm) (mg/L) Excavated Pit 4.52 1.57 1105 Well G W W 4.32 1.78 1277 Source Z n C u C a (mg/L) (mg/L) (mg/L) Excavated Pit 16 0.06 309 Well G W W 24 0.36 312 198 The p H of the groundwater from Well G W W is lower than that collected from the excavated pit prior to dump excavation. A l l of the other parameters are higher in Well G W W than in the excavated pit. The only parameter with similar values for both the excavated pit and Well G W W is calcium (309 mg/L compared to 312 mg/L). 3.6.2.2 Francis Lake The water quality data collected from Francis Lake over a six-month period in 1994 (January-June) are compared to the water quality data collected over the same six-month period in 1996 (January-June). Also included is the average for the last three months of the 1996 water quality analysis data (ie. April-June). The results are presented in Table 3-16 and represent the average value over the indicated time period. Table 3-16 Comparison of Francis Lake Data Time Period p H Conductivity Sulphate (mS/cm) (mg/L) Jan-June/94 6.26 .100 25.3 Jan-June/96 6.30 .093 26.3 Apr-June/96 6.40 .098 24.7 Time Period Alkalinity Calcium Magnesium (mg C a C 0 3 / L ) (mg/L) (mg/L) Jan-June/94 10.4 11.3 1.7 Jan-June/96 8.2 12.2 1.6 Apr-June/96 9.4 13.3 2 199 Looking at the January-June data for both time periods, there is no consistent trend for either data set. The 1994 data has higher conductivity, alkalinity, and magnesium, while the 1996 data has higher pH, sulphate, and calcium. For all parameters except sulphate, the April-June 1996 data has higher values than the January-June 1996 data. The p H measured in both time periods is only slightly acidic and the rest of the parameters are quite low, especially in comparison to the groundwater data. 3.6.3 Predicted Water Quality It was originally envisioned to use the batch leach test, static test, and kinetic test data along with historical and current water quality data to make predictions regarding water quality in the flooded pit, groundwater around the Northwest Dump, and Francis Lake. Unfortunately the information required to make precise quantitative predictions regarding water quality could not be obtained. Therefore, only the effect of Northwest Dump material on the water quality of the freshwater cap in the flooded pit will be examined. Most of the material excavated from the Northwest Dump was disposed of in the pit and fell to the bottom. Unfortunately, some of the material from the first lift that was excavated got caught on the upper benches of the pit as it was being dumped. This material is currently above the surface of the saltwater in the pit (Plate 3-4), and some of it will still be exposed even after the freshwater cap becomes established. 200 CI C •5 o Li. -0) o a> — — * * -= 2 ^ CU 00 o >, BO c 3 O O o Q s o [3 '•Z es .e o 5 w . « ox — cU -C • • rf 201 Mass Balance The assumptions used to perform the mass balance are presented in Table 3-17. The area of the pit at the 0' and -50'-levels was estimated using air photos and scaled contour diagrams of the open pit. This method should be accurate to within 10-20%. Once it becomes fully established, the freshwater cap will be approximately 10m deep. The horizontal extent of material on the benches was estimated from photographs and from anecdotal evidence (Home, pers. comm., 1996b). The mass of material on the benches was estimated using the density value provided in L i (1990), the fact that benches are 12m high by 7m wide, and the assumption that 6 benches would be above the surface of the freshwater cap. The composition of the material on the benches was estimated from personal observations of the material and excavation of the dump. The majority of this material is believed to be from the first lift of the dump excavation. Due to heavy winter rains in 1995/96, it is assumed that the material has been rinsed a number of times since it was first excavated. Therefore the following loading rates are based on long-term loading that would occur and do not take into account the metal loading provided by the initial washing of this material. Table 3-18 provides the metal loading that would occur from flushing of the material on the benches. The total metal loading per week was determined from the loading/leaching rates calculated from the last five weeks of kinetic data for these samples. The loading/leaching rates of zinc and copper, for both Sample 1 and Sample 2/3, were 0.05 mg/kg/week. The loading/leaching rate of zinc for Pit 1/2 was 0.2 mg/kg/week and for copper was 0.05 mg/kg/week. 202 TABLE 3-17 ASSUMPTIONS OF MASS BALANCE CALCULATION Component Value Area of pit (O'-level) 1,850,000 m 2 Area of pit (-50'-level) 1,635,000 m 2 Average area of pit between 0' and -50' levels 1,742,500 m 2 Depth of freshwater cap 10 m Volume of freshwater cap 1.742 x 107 m 3 = 1.742 x 1 0 1 0 L Horizontal extent of rock on benches 300 m Cross-sectional area of filled-in bench 40 m 2 Volume of rock/bench 12000 m 3 Total volume of rock above freshwater cap (6 benches) 72000 m 3 Density of dump material 1847.4 kg/m 3 (Li, 1990) Total mass of material above cap 1 . 3 3 x l 0 8 k g Estimated composition of material - Sample 1, Sample 2/3 - Pit 1/2 95% 5% Mass of separate components - Sample 1 - Sample 2/3 - Pit 1/2 63181080 kg 63181080 kg 6650640 kg 203 TABLE 3-18 METAL LOADING OF DUMP MATERIAL Sample Metal Loading (g/week) 1 Zn: 3159 Cu: 3159 2/3 Zn: 3159 Cu: 3159 Pit 1/2 Zn: 1330 Cu: 333 Total Zn: 7648 Cu:6651 The water quality of the freshwater cap resulting from this metal loading is presented in Table 3-19. The metal loading per litre per week was calculated using the loading rates provided in Table 3-18 and dividing by the volume of the freshwater cap provided in Table 3-17. The volume of the cap was based on its final dimensions after becoming fully established (ie. area = 1.743 ha, depth =10 m). The elapsed time before TABLE 3-19 RESULTING WATER QUALITY OF FRESHWATER CAP Parameter Value Zn 0.00044 mg/L/week C u 0.00038 mg/L/week Number of years before concentrations exceed Zn: 44 discharge permit Cu: 2.5 204 water quality guidelines are exceeded was based on the concentrations given in the Water Management Pond discharge permit issued to Island Copper (zinc=1.00 mg/L, copper=0.05 mg/L) and did not take into account additional dilution provided by the annual precipitation (~1800mm/year). Based on these calculations, the concentrations of zinc and copper in the freshwater cap of the flooded pit will exceed permit guidelines in 44 and 2.5 years, respectively. 205 4. DISCUSSION Section 4 will provide an explanation of the results presented in Section 3. Any problems that were encountered while carrying out the research that could account for anomalies seen in the results will also be described. 4.1 Sample Collection and Preparation There are no results per se to discuss for these two sections. Instead, a possible source of error that could have resulted from these activities and its effect on the results will be discussed. The biggest concern for both of these activities is sample representivity. Due to the inherent heterogeneity of waste rock dumps and the large size of the dump (1 million tonnes) compared to the size of the samples (5-10 kg), it is unlikely that the samples collected were representative of the dump as a whole. In fact, it is generally acknowledged that collecting what constitutes a representative sample from a waste rock dump is a considerable challenge (Knapp et al., 1995). The potentially low representivity of the samples means that the results obtained are very sample-specific. Therefore it is difficult to extend the results from the samples to the dump as a whole with a large degree of confidence. 206 4.2 Sample Characterization 4.2.1 Particle Size Analysis On average, 20% of the original material passed through a 3.35 mm sieve compared to 50% of the humidity cell material. This means that the material in the humidity cells was composed of much smaller particles than the original material and therefore had a much greater surface area per unit mass than the original material. This higher relative surface area means that the oxidation and leaching rates calculated for the humidity cell material are probably higher than they would be for the original material. Keep in mind that the material referred to as the original material is not representative of the actual dump material because of the bias against large (>20 cm diameter) particles in the sample collection program. Therefore, the actual dump material will probably have oxidation and leaching rates that are even lower than the original material from which the humidity cell samples were taken. In terms of surface area, the differences within treatments and between treatments were fairly small. The batch leach replicates (Bulk 3-2! and 3-22) had an average surface area of 0.72 m /kg of sample with a standard deviation of SD=2.83. The non-batch leach replicates (Bulk 3-23 and 3-24) had an average surface area of 0.80 m 2 /kg (SD=5.66). Thus the coefficient of variation within the same treatment was not very significant (4% and 7%o, respectively). The difference in surface area between treatments was slightly more substantial at just over 10%. This is expected because the non-batch leach 207 replicates had a slightly higher proportion of fines than the batch leach replicates and therefore had a higher surface area. 4.2.2 T h i n Section and X R D Analyses From the thin section and X R D analyses, it is apparent that plagioclase is the most common component of the samples. This is understandable because the rocks are mostly andesitic volcanics. Andesite contains phenocrysts composed primarily of sodic plagioclase, especially andesine. Andesine has a composition that ranges from 70% albite/30% anorthite to 50% albite/50% anorthite (Bates and Jackson, 1984) and this would account for the presence of both of these minerals in the X R D analysis. The two samples lacking anorthite according to the X R D analysis (Dike 1/2/3 and Bulk 3-2) probably have phenocrysts composed primarily of albite rather than oligoclase or andesine. The thin section and X R D analyses also agree that Sample 1, Sample 2/3, and Bulk 5 are the only samples that contain significant amounts of carbonate. The X R D analysis indicates that all of the samples contain laumontite (a zeolite), but the thin section analysis only found zeolite in trace amounts in Pit 1/2 and Bulk 1. The X R D analysis detected pyrite in all but two of the samples (Bulk 5 and Bulk 8), whereas the thin section analysis indicated pyrite was present in all of the samples, although Bulk 5 had the least amount (0.5%). These discrepancies suggest that the thin sections may not have been completely representative of the samples they came from. This is understandable because the thin section is a small area of one rock from a multi-208 kilogram sample composed of many rocks. Thin section analysis is useful, however, because it provides an estimate of the mineral's distribution in the sample, whereas X R D analysis only indicates which minerals are present, not their distribution. 4.2.3 I C P and Whole Rock Analysis From the Q A / Q C checks performed on chromium and titanium (Appendix B-5) it is apparent that there is no correlation between the ICP and whole rock analyses for these two elements. This is a common result, as titanium is found in sphene (titanite) which prevents accurate measurement by ICP analysis (Downing, 1996 - pers. comm.). The presence of sphene is corroborated by the thin section analysis which detected sphene in most of the samples. The remaining elements exhibit a high linear correlation between the ICP and whole rock analyses which indicates the two methods are providing accurate and consistent measurements and the results can be used with a large degree of confidence (Downing, 1996 - pers. comm.). 4.3 NORTHWEST D U M P Reanalysis of the earlier work conducted on the Northwest Dump indicated that most of the material that was analyzed had a negative N N P and/or an NP:AP ratio of less than one. In terms of A R D prediction, these results indicate that the dump had a high potential for acid production. 209 The upper portion of the dump (top 8-10m) had a lower potential for generating acid than the lower portion. This is significant because it is the material from the upper portion of the dump that first filled the upper benches of the pit during excavation and formed a talus slope down to the bottom of the pit. A lot of this talus material was washed into the pit during the winter of 1995/96, but quite a bit of the upper dump material remains on the upper benches. The material at the very bottom of the dump (bottom 1-2 m) also appears to have had a relatively low acid-generating potential. This was especially true for the material from the bottom of holes N W D # 6-9 and could be due to the incorporation of glacial till into the sample. Glacial till has been shown to have a positive net neutralization potential (Lister, 1994). 4.4 Preliminary Testwork 4.4.1 Shake Flask Tests The neutralization that was occurring during shake flask testing was most likely due to carbonate dissolution. X R D analysis detected calcite in Bulk 5 and thin section analysis detected significant amounts of carbonate in Bulk 5 and trace quantities in Pit 1/2. Gypsum dissolution is the most likely source of the sulphate leached into solution because X R D analysis detected gypsum in all of the samples except Dike 1/2/3. The 210 high concentrations of sulphate observed in the solutions of Dike 1/2/3 are somewhat puzzling, but could be due to the dissolution of secondary minerals formed as a result of oxidation. 4.4.2 Batch Leach The most significant problem encountered during the batch leach occurred when removing the leachate at the end of each cycle. The leachate was siphoned out of the container and when the container was tilted to remove the last remaining leachate, quite a bit of silt accompanied it. This silt probably accounts for the loss of sample mass observed during batch leaching (approximately 10% of the original sample that went into the container). It is likely that the majority of this mass was in the fine fraction of the sample. The difference in p H between the shake flask tests and the batch leach could indicate that carbonate dissolution was occurring in the former, but not in the latter. This probably reflects the difference in particle size between the shake flask and batch leach tests. The small particle size of the shake flask tests would provide more reactive surface area than the larger particles of the batch leach tests. Plotting the calcium:sulphate molar ratio (Figure 4-1) can provide information on gypsum solubility. If the ratio is equal to 1, gypsum dissolution is producing the calcium and sulphate measured in solution, if the ratio is greater than 1, carbonate dissolution 211 212 outweighs sulphide oxidation, and i f the ratio is less than 1, sulphide oxidation outweighs carbonate dissolution. From Figure 4-1 it looks as though gypsum dissolution could account for the sulphate seen in Cycle 1 for Sample 1, Sample 2/3, Bulk 1 and Bulk 5 and it appears that sulphide oxidation was occurring in all of the samples by Cycle 3. Due to the high neutralizing capacity of Sample 1, Sample 2/3, and Bulk 5, zinc precipitates may have formed in this material within the waste rock dump. These secondary minerals would act as a storage area for zinc. When the strongly acidic batch leach solution was added to the samples these precipitates would have dissolved and released zinc back into solution, thereby accounting for the high zinc concentrations measured in the leachate of these samples. The increase of zinc released into solution by Bulk 5 between Cycles 1 and 2 is unusual considering the fact that the zinc released into solution by all of the other samples decreased throughout the batch leach. This could indicate that the zinc precipitates were occluded in other precipitation products (eg. gypsum) and it took time for these other precipitates to dissolve and expose the zinc precipitates. Once the zinc precipitates were exposed they too could dissolve, thus releasing more zinc into solution in the second cycle than in the first. The differences in copper concentration in the leachate of the samples are probably due to differences in original copper content, both in primary mineral form (mainly chalcopyrite) and in secondary mineral form (eg. chalcanthite). One observation that needs to be explained is the apparent increase in copper being released into solution by Sample 2/3 and Bulk 5 from Cycles 1 to 2. A l l of the other samples released less 213 copper into solution from Cycles 1 to 2. The increase observed for Sample 2/3 and Bulk 5 could be similar to the explanation given for zinc. The copper precipitates could have been surrounded by other precipitates which got dissolved in the first cycle. Once these other precipitates were gone, the copper precipitates could dissolve, thereby releasing more copper into solution during the second cycle than the first. Based on Figure 4-1, the dissolution of gypsum is the probable source of the calcium seen in Cycle 1 for Sample 1, Sample 2/3, and Bulk 5. The calcium:sulphate ratio drops well below 1 in Cycles 2 and 3, however, so dissolution of gypsum cannot explain the calcium released in the last two cycles. Other possible sources of calcium in solution could be the dissolution of laumontite which was detected in all of the samples or anorthite which was detected in all of the samples except Bulk 3-2. The calcium from Dike 1/2/3 could have been provided by the dissolution of laumontite. 4.5 Static Prediction Tests 4.5.1 ACID-GENERATING POTENTIAL 4.5.1.1 Sulphur Mass Balance A number of problems were encountered trying to perform the mass balance. The most significant one was the apparent discrepancy in sulphur content for samples at different stages. It appeared for many samples as though the sulphur content of the post-batch leach material was too low and in the post-kinetic testing material was too high in 214 relation to the original sulphur content. It was thought that these differences may have been the result of inaccuracies in the analytical method. To determine the accuracy of the analytical method, a number of samples were reanalyzed in triplicate. From these replicate samples it was determined that the analytical method used was fairly accurate because the variance was only a few percent (2-6%). A couple of other problems led to difficulties in performing the sulphur mass balance. One problem was the failure to analyze the batch leach residue for sulphur. Sulphur was analyzed from a sub-sample of this residue which had been rinsed with distilled water to remove any remnants of the batch leach solution. This meant that the sulphur content of the batch leach residue was unknown and had to be deduced from other sources. This was accomplished by back-calculating from the results of the sulphur analyses on the kinetic test residue and the washed batch leach residue, because the sulphur content in these stages was known. Another problem was the inability to measure the sulphur content of the material that was lost during batch leaching and kinetic testing. The amount of material lost during kinetic testing was insignificant, but during the batch leach, this lost material accounted for up to 10% of the overall sample mass. To perform the mass balance, it was assumed that this lost material had a sulphur content that was the same as the original material. 215 A sensitivity analysis was performed to determine how a change in sulphur content for a particular sample at a particular stage would affect the results of the static prediction tests (acid-base accounting). Two scenarios were tested: 1) a "worst-case" scenario, where the sulphur content was as high as reasonably possible for each stage, and 2) a "best-case", where the sulphur content was as low as possible. It was found that even with the best-case scenario, the results of the static prediction tests did not change significantly. That is, in most cases the N N P and NP:AP ratio did not change enough to move the sample into a different A R D potential category. This is due to the relatively high acid-generating potential and the relatively low neutralization potential of the material that was tested. The only sample that was noticeably affected was Bulk 5 which moved from a category of high A R D potential to a category of uncertain A R D potential. The original material went from a slightly negative N N P to a slightly positive N N P and therefore the NP:AP ratio went from slightly under 1 to slightly over 1. This also happened with the post-batch leach material (ie. the kinetic test head). In both cases, however, the A R D potential was still relatively high because the N N P was less than 20 kg C a C C V t and the NP:AP ratio was very close to 1. The remaining samples increased in both the N N P and the NP.AP ratio as would be expected with a decrease in sulphur content, but not enough to move the sample into a different A R D potential category. Based on the results of the sensitivity analysis, it was decided that the original values calculated from the sulphur mass balance were reasonable and did not need to be adjusted for the acid-base accounting calculations. 216 4.5.1.2 Standard Acid-Base Accounting The results of the standard acid-base accounting do not really need to be explained because they are very straight forward. The higher the sulphur content of a sample, the higher the acid-generating potential (AP) because A P is calculated by multiplying the sulphur content (in %) by a conversion factor of 31.25. Therefore the samples with the highest sulphur contents will be the samples with the highest AP. Some researchers (Morin, 1990) have suggested that the 31.25 conversion factor for converting sulphur content into A P (kg CaC0 3 / t ) is not always appropriate. In order for the 31.25 factor to be valid, a number of assumptions must be true: l)sulphur occurs only as S 2 2 \ 2) pyrite is the only source of sulphide, 3) S22~ oxidizes completely to sulphate, 4) the only oxidants available are water and molecular oxygen, 5) all of the iron is oxidized to the ferric state, and 6) all of the iron precipitates out of solution as Fe(OH) 3 . If these assumptions are not met, Morin (1990) believes that alternative conversion factors may be more suitable. For example, i f there is no precipitation of ferric iron, Morin suggests a conversion factor of 7.81, or if chalcopyrite replaces pyrite as the sulphide mineral and no copper is precipitated then Morin suggests a factor of 15.63 would be more valid. In terms of the Island Copper samples, the static prediction tests were performed as if the above assumptions had been met. The only sample that may not have met these assumptions is Dike 1/2/3 which contained chalcopyrite. Based on the thin section analysis, however, pyrite was five times more abundant than chalcopyrite (2.5% compared to 0.5%) and therefore pyrite was probably the primary source of sulphide in this sample. 217 One other factor to keep in mind regarding the acid-generating potential of a sample is the location of the sulphide. The sulphide is less likely to react i f it is occluded within a grain than i f it is exposed on the surface of the grain. The thin section analysis conducted on the samples indicated that the sulphide (predominantly pyrite) occurs as randomly disseminated, individual grains or in some cases, as small aggregates. 4.5.1.3 Modified Acid-Base Accounting Modified A B A uses the sulphide content of a sample to calculate its AP. Since the sulphate content of most of the samples was quite low, the sulphide content was quite close to the total sulphur content. Therefore the standard and modified methods produced similar values for the A P of the post-batch leach and post-kinetic testing material. The biggest difference in A P values for the samples using the two methods was for the original material because of its relatively high sulphate content. 4.5.2 Neutralization Potential 4.5.2.1 Paste pH The paste p H analysis is a standard step when conducting an acid-base account. It is used to determine if acid generation has already occurred in the sample. Since the Northwest Dump had been weathering for so long and A R D had been detected emanating from the dump, it was expected that a lot of the samples would prove to have already generated acid and the analysis was performed only as a matter of standard operating procedure for acid-base accounting. The only sample which had previously generated 218 acid (ie. paste p H < 5) was Bulk 3-2. This was the most visibly oxidized sample of them all and would be expected to have a low paste pH. 4.5.2.2 Acid-Base Accounting Both of the acid-base accounting methods use a fizz test to determine the volume of acid that should be added to the sample. This test is very subjective and can lead to significantly different NPs depending on which fizz rating you use (Lawrence and Wang, 1996). In some cases there can be an order of magnitude difference in the NP that is determined. It is possible that for some of the samples, the assigned fizz ratings were incorrect, but it is unlikely. The conditions under which the samples react during an acid-base account can lead to the dissolution of neutralizing minerals that would not normally contribute to the immediate neutralizing capacity of a sample (eg. silicates). The dissolution of these minerals leads to an overestimation of the N P calculated for that sample. The back-titration curves plotted as part of the A B A procedure (Appendix F) can be used to determine i f these relatively non-reactive minerals dissolved during the acid-base account, leading to an overestimation of the sample's NP. The plateau seen between p H 3 and 4 in a number of the back-titration curves (eg. Sample 1, Sample 2/3, Bulk 5), especially for the standard (Sobek) method, indicates that aluminosilicates, minerals that will not contribute to the immediate neutralizing capacity of the sample, dissolved during the acid-base account. Therefore, the NP of these samples is higher than it actually 219 would be in terms of realistic (ie. immediate) neutralizing capacity. This overestimation of the N P is why the N P used in Table 3-12 was the average N P determined using the modified and carbonate methods. It was felt that the modified and carbonate methods provide a more accurate prediction of the realistic neutralizing capacity of the sample. Overall Discussion Examination of Table 3-11 reveals that the NP of the samples decreased as the samples moved through each stage of testing. The carbonate N P decreased the most for all three methods (up to 80%) suggesting that the decrease in N P was due to the dissolution of carbonates. The NP as determined by the other methods showed a slightly lower decrease (up to 50-75%) than it did for the carbonate NP. This indicates that the N P of the samples determined using the standard and modified methods was composed primarily of carbonate, but was also made up of some other neutralizing minerals (eg. silicates). Other Observations The calcium contents of the samples were plotted against their respective NPs (Appendix G) to determine i f calcium content could be used as a predictor of NP. In general, calcium content showed a good correlation with the modified and carbonate NPs (average R = 0.86 and 0.85, respectively). The correlation between calcium content and N P for the standard method, however, was not quite as high (average R 2 = 0.70). It is also apparent that the correlation when using the modified and carbonate methods was highest for the original material, but when using the standard method, the highest 220 correlation was with the post-kinetic testing material. This is understandable because in most cases the original material contains higher concentrations of carbonate (predominantly calcite) than the post-batch leach or post-kinetic testing material. The N P determined by the modified and carbonate methods is dominated by carbonates so the higher the carbonate content (ie. calcium content), the higher the NP. The N P determined by the standard method on the other hand, is influenced not only by carbonate, but by minerals such as silicates. Therefore, the overall correlation between calcium and N P should be lower using the standard method than either the modified or carbonate methods. Based on the curves in Appendix G , only the modified and carbonate NPs of the original material of additional samples could have been predicted with a fair degree of accuracy from their calcium content (R2=0.90 and 0.87, respectively). Due to the low correlation between calcium and N P using the standard method, the standard N P of additional samples could not have been predicted using the curves in Appendix G. 4.6 Kinetic Prediction Tests The single biggest concern regarding the kinetic prediction tests was the length of time for which these tests were conducted. Most kinetic tests conducted these days run for tens of weeks, and the axiom "the longer, the better" certainly holds true. Unfortunately, due to time constraints, the kinetic tests carried out as part of this research could only be conducted for 16 weeks. This is quite a short period of time relative to most current kinetic tests and it is possible that the results observed for the Island Copper 221 samples are not truly representative of the kinetic behaviour of the Northwest Dump material. 4.6.1 General Observations 4.6.1.1 Procedural Problems Minor differences in air flow between samples caused by differences in relative moisture, permeability, and porosity resulted in some cells drying out faster than others. This resulted in preferential flow through the drier cell (path of least resistance), which in turn reduced air flow through the other cells. The drier cell also retained more of the rinse solution during the leach, resulting in a lower leachate volume for that cycle and potentially affecting the concentrations of the different constituents in that week's leachate. To remedy this problem, the airflow through the cells was constantly adjusted using the "bubbler" to try and maintain equal drying in all of the cells. According to Pool and Balderrama (1994), fluctuations in air flow can have a significant effect on humidity cell results. They found that the weight loss of a cell was dependent on air flow into the cell and that small changes in air flow could lead to significant differences in the volume of leachate that was obtained from the cells during rinsing. If not enough rinse solution is added to a cell during the leach cycle, secondary mineral precipitation can occur (Morin et al., 1995a). Precipitation occurs because the solubility limits of some constituents are exceeded in the leachate due to insufficient volume. Pool and Balderrama (1994) found that in order to obtain a consistent leachate volume, the air flow and volume of water added during the leach had to be constant. The results of the 222 kinetic testing conducted for this thesis indicate that maintaining a constant air flow did not necessarily produce a consistent leachate volume because of the preferential drying discussed above. 4.6.2 p H The presence of the two different groups of samples regarding leachate pH is fairly easy to understand when one looks at the results of the sample characterization analyses. The thin section and X R D analyses indicated that Sample 1, Sample 2/3, and Bulk 5 (neutral throughout kinetic testing) were the only samples that contained significant quantities of carbonate (calcite). The only other samples that contained detectable carbonate (trace quantities based on the thin section analysis) were Pit 1/2 and Bulk 1, both of which had higher leachate p H than the remaining samples. The difference in pH between the batch leach replicates and the non-batch leach replicates could be due to the presence of weathering products in the non-batch leach replicates. Unlike the batch leach replicates, which had most of their weathering products removed, the non-batch leach replicates still contained all of their original weathering products which when rinsed would release acidity into the leachate causing the p H to drop. The presence of weathering products in the non-batch leach samples is indicated for the first week of testing when the p H of these two samples is significantly lower than in subsequent weeks. The first leach of these samples removed stored weathering products, releasing acidity, resulting in a low pH. The increase in p H for 223 these samples in Weeks 2 and 3 suggests that most of the stored acidity was removed in Week 1. The variation in results one could expect for any particular sample, as indicated by the pH data of the replicate samples, is only 4-6%. This variation is comparable to the variation seen in the p H data of the batch leach tests. By the third cycle of the batch leach tests, the difference between the p H of Bulk 3-21 and 3-22 was only 4%. 4.6.3 Conductivity Conductivity provides a measure of the ionic strength of a solution and reflects the reactivity of the sample (Coastech Research Inc., 1991). A n increase in conductivity indicates that more ions are being released into solution. Therefore, conductivity is expected to increase as kinetic testing progresses because more and more ions (eg. sulphate, metals, etc.) are being released into the leachate. As sulphide oxidizes it releases sulphate into solution and sulphate has a major effect on conductivity. The decrease in conductivity seen in the samples suggests that the 16 weeks of kinetic testing was not a long enough period of time in which to observe the onset of active sulphide oxidation. Looking at conductivity, the batch leach appears to have had a significant impact on the samples. The batch leach replicates had a much lower conductivity (0.45 mS/cm) compared to the non-batch leach replicates (1.8 mS/cm). After Week 4, the batch leach 224 replicates had an average conductivity of approximately 0.45 mS/cm (standard deviation = 0.14 mS/cm). From Weeks 4-8, the non-batch leach replicates had an average conductivity of around 2.2 mS/cm (SD=0.14 mS/cm) and from Week 9 onwards they averaged 1.8 mS/cm (SDK). 57 mS/cm). From Week 9 onwards the standard deviation between the non-batch leach replicates was high because the conductivity of Bulk 3-23 was steadily decreasing. In terms of variation of results, the batch leach replicates indicate that the results could be expected to vary by up to 25%. 4.6.4 Alkalinity/Acidity 4.6.4.1 Alkalinity Alkalinity provides a measure of the acid-consuming capacity of a solution and reflects the composition of a sample in regards to neutralizing minerals. Although there was a lot of variation in the alkalinity of the three neutral samples (Sample 1, Sample 2/3, and Bulk 5), overall the alkalinity appeared to be decreasing from Week 10 onwards. This decrease in alkalinity suggests the amount of neutralizing components in the samples was decreasing and could provide evidence that these samples were going to become acid-generating. 4.6.4.2 Acidity Acidity provides a measure of several aqueous species such as F e 2 + , F e 3 + , Fe(OH) 2 + , A l 3 + , and HSO 4 " (SRK, 1989) and can therefore act as an indicator of acid generation. The high acidity levels seen in Dike 1/2/3 and Bulk 3-24 indicate that 225 significant acid generation was occurring in these samples. Significant acid generation was also indicated for Bulk 3-2] and Bulk 3-23. There are obvious differences within treatments and between treatments in terms of acidity produced. Over the last three weeks of kinetic testing, the batch leach replicates averaged approximately 28 mg C a C 0 3 / L (standard deviation = 8 mg C a C 0 3 / L ) while the non-batch leach replicates averaged around 150 mg C a C 0 3 / L (SD=140 mg C a C 0 3 / L ) . The corresponding coefficients of variation for these two sets of replicates are 28% and 93%, respectively. The wide variation in the non-batch leach replicates is due to the fact that Bulk 3-24 produced four to five times as much acidity as Bulk 3-23 indicating that these two samples were significantly different in composition. Therefore the variation in the results should be assessed based on the batch leach replicates only. The variation in results thus obtained is 25-30% for any particular sample, again assuming that the batch leach replicates are identical in nature. Regardless of the variation in the non-batch leach replicates, it is obvious that they released significantly more acidity than the batch leach replicates. Therefore it can be concluded that the batch leach was effective at removing stored weathering products from the samples. 4.6.5 Sulphate 4.6.5.1 Cumulative Loading Ferguson and Morin (1991) described three types of curves observed when cumulative sulphate was plotted versus time: linear, concave, and convex. Although the 226 curves described by Ferguson and Morin were from samples that remained alkaline throughout kinetic testing, it is believed that the explanation provided as to how these curves developed is still valid even for samples that were acidic from the start of kinetic testing. Based on Figure 3-8a, all of the samples exhibit the convex behaviour except Bulk 3-24 which displays the linear behaviour. Ferguson and Morin provide four explanations for the development of the convex curve: 1) dissolution of stored oxidation products, 2) development of an oxidized rind around grains, 3) development of secondary mineral coatings around grains, and 4) variation in reaction rates with respect to sulphide morphology and crystal size. The linear-type curve develops if the reactions were controlled kinetically (ie. the rate of oxidation, not the change in reactants, controls the reaction). The concave curve was not observed and will not be discussed. The batch leach removed a lot of the stored reaction products from the samples before kinetic testing began so the first explanation provided for the convex-type curves probably only applies to Bulk 3-23 although it should apply to Bulk 3-24 as well. The second and third explanations are probably the most reasonable for the remaining samples that exhibited the convex-type cumulative sulphate curve. The difference in cumulative sulphate between the non-batch leach replicates suggests that maybe the fourth explanation provided above would be applicable to Bulk 3-23. Differences in sulphide morphology and crystal size could account for the differences seen between these two samples. 227 The significant difference in sulphate loading between the batch leach and non-batch leach replicates is attributable to the dissolution of stored oxidation products in the non-batch leach replicates. The difference in stored sulphate is obvious from the first week when the non-batch leach replicates produced approximately six times as much sulphate as the batch leach replicates (1360 mg S0 4 /kg of sample compared to 220 mg/kg). This trend continued throughout kinetic testing indicating that dissolution of stored oxidation products was probably still occurring in the non-batch leach replicates. 4.6.5.2 Sulphide Oxidation Rate The production of sulphate reflects sulphide oxidation if the following conditions are met (SRK, 1989): 1) all oxidized sulphur is released to the water, 2) all sulphur in the water is fully oxidized to sulphate, and 3) precipitation of gypsum or another sulphate mineral does not limit the aqueous sulphate concentration. If the above assumptions are true, then an increase in the concentration of sulphate from one analysis to another indicates that the rate of acid generation is increasing. The most important of the three conditions listed above is that the aqueous concentration of sulphate is not limited by the precipitation of gypsum or some other sulphate mineral. To determine i f gypsum precipitation was occurring in the samples, the sulphatexalcium molar ratio was plotted against the sulphate concentration (Figure 4-2). The saturation curve in Figure 4-2 was based on the work of Ferguson and Morin (1991) and was generated from the equation "Ratio = 3.43xl0" 6(SO 4) 2". The assumptions of this equation are: 1) complexation of calcium and sulphate was minimal, 2) ionic strength was negligible, and 3) p H was 228 |3 CM (O * CM CM CM CN CO CM CU cu "O. CM E £ - ^ = = OTCOQ.QCQCQCQCQCQCQCQ s e n s^; | i i iS Q . Q . £ M ^ - ^ c O C O C O C O i r > c o < • O • + O • I < X 229 essentially neutral. These assumptions require that an error factor be incorporated into the sulphatexalciurri ratios. Ferguson and Morin suggest a factor of 2, but i f the p H drops to 2 this could increase to an order of magnitude because of the formation of bisulphate (HS0 4~). None of the samples had pHs that were below 3, so the appropriate safety factor is probably between 2 and 5. The samples that can best be analyzed with the above figure are Sample 1, Sample 2/3, and Bulk 5 because these three samples remained neutral throughout kinetic testing. Dike 1/2/3 had the lowest p H (~3) so the figure must be used carefully when trying to interpret the behaviour of this sample. Based on Figure 4-2, Bulk 3-23 and 3-24 were the only samples that were oversaturated with respect to gypsum. If this figure is accurate, then the assumption above that gypsum precipitation was not limiting the aqueous concentration of sulphate is not true for these samples and therefore the rate of sulphate production cannot be used to determine the rate of sulphide oxidation. The rest of the samples are well within the undersaturated region so this assumption is valid and therefore sulphate production reflects sulphide oxidation. With this in mind, Bulk 3-23 and 3-24 will not be discussed any further. For the remaining samples, the sulphide oxidation rate appeared to decrease throughout kinetic testing. For Dike 1/2/3 and Bulk 3-2 1 ; this decrease became most apparent after Week 10. This decrease could be due to the formation of an oxidized rind around the mineral grains (ie. "shrinking core"; Nicholson et al., 1990; Robertson, 1994; Scharer et al., 1995) or the precipitation of secondary minerals onto the grain surface (Ferguson and Morin, 1991; Alpers et al., 1994). Both of these processes would result in a decrease in the rate of reactant (eg. oxygen) transport from the outer surface to the reactive inner surface of the grain, thereby reducing the sulphide oxidation rate. 230 4.6.5.3 Percent Sulphide Remaining and Time to Deplete Sulphide Throughout the kinetic testing, a calculation was made regarding the time that would be needed for the material to become depleted of sulphide, NP, or a particular metal. This is a theoretical calculation based on leaching rates and the original sulphide, NP, or metal content of the material, and it is acknowledged that there is a very low likelihood that a rock would ever become completely depleted of a metal. With this in mind, the discussions on the time to deplete a sample of sulphide, NP, or a particular metal are based on the results obtained from the kinetic testing and not on the actual behaviour of a rock in nature. The results of the percent sulphide remaining and the time to deplete sulphide do not really need to be discussed as they were generated based on previous parameters. The apparently significant loss of sulphide from Bulk 3-23 and 3-2 4 is misleading as it was shown above that gypsum precipitation is probably limiting the sulphate concentration and therefore affecting the sulphide oxidation rate of these samples. Thus the time to deplete the sulphide in these samples cannot be considered accurate either. For the remaining samples, a large percentage of sulphide remaining translates into a relatively long time for depletion and vice-versa. Looking at the replicate samples, there are significant differences within treatments and between treatments. In terms of time to deplete the sulphide, using the last three weeks of kinetic test data, the batch leach replicates will require on average 40 years (standard deviation = 21 years) while the non-batch leach replicates will only 231 require 5 years (SD=1.4 years). The large standard deviation in the batch leach replicates (coefficient of variation = 53%) indicates significant differences in the characteristics of these samples. Even with the large standard deviation in the batch leach replicates, there is a significant difference in the time to deplete sulphide between the batch leach replicates and the non-batch leach replicates. Using the replicate samples to indicate random variation in the results, one could expect variation of anywhere from 25-55%, but considering the fact that the replicate samples may not be true replicates of each other, the variation is probably much lower. 4.6.6 M o l a r Ratios Molar ratios can provide significant insight to chemical reactions occurring within a sample. The two most common molar ratios calculated, with regards to neutralization reactions, are the carbonate molar ratio (Ca+Mg+Ba+Sr:S04) and the silicate molar ratio (Ca+VzNa+ViKiSC^). In addition to the information these ratios provide on the chemical reactions taking place, these ratios can also be used to calculate various parameters such as cumulative NP produced, the NP depletion rate, percent N P remaining, and the time to deplete NP. Calcium far outweighs the other components of these ratios which is why the two sets of figures look so similar. The interpretation of a particular ratio is significantly affected by the p H of the leachate from which it was calculated (Morin and Hurt, 1994b). At neutral pH, the carbonate ratio would be between 1 and 2 i f carbonate dissolution was occurring in 232 response to acid generation, while the silicate ratio would be approximately 1 if silicate dissolution was occurring in response to acid generation (Morin and Hutt, 1994b). If acid was being produced in greater amounts than alkalinity or at a greater rate, then both of these ratios would be lower. Looking at the weekly ratios from the kinetic testing one can conclude that Sample 1, Sample 2/3, and Bulk 5 were providing sufficient neutralization in response to acid generation to keep the leachate neutral. Based on the thin section and X R D analyses that were conducted, calcite was the most likely source of neutralization for these samples. Pit 1/2, Bulk 1, and Bulk 3-2 were acidic throughout kinetic testing, although trace amounts of carbonate were present in these samples (thin section analysis). This could account for the carbonate ratio of 0.5-1.0 observed in these samples. The low ratios (<0.5) for Dike 1/2/3 indicate that acid generation was overwhelming neutralization in this sample. The widely fluctuating ratios of Bulk 8 (0.5-1.5) are hard to explain. They could indicate that K - or Na-feldspars (eg. albite) were providing some neutralization. If one looks at individual cations (eg. C a from calcite or Ca-feldspar such as anorthite, Na from Na-feldspar such as albite, or K from K-feldspar) then the ratios expected differ slightly. At acidic p H (~4), the ratio of C a (calcite), C a (Ca-feldspar), Na, and K to sulphate would be 1:1, 1:4, 1:2, and 1:2, respectively. At neutral p H (~7), however, these ratios change to 2:1, 1:1, 2:1, and 2:1, respectively (Morin and Hutt, 1994b). The weekly ratios of the individual cations to sulphate are provided in Appendix H-3. Looking at the calcium:sulphate ratio, only a few conclusions can be made. The neutral samples (Sample 1, Sample 2/3, and Bulk 5) were releasing calcium from both 233 calcite (ratio>2) and from anorthite (ratio=l). Dike 1/2/3, with a ratio of around 1:4, appeared to be releasing calcium from anorthite. Based on the X R D analysis, however, this sample did not contain anorthite, so the source of this calcium is unclear. Calcium could have been coming from laumontite, which upon dissolution would also produce a 1:4 calcium to sulphate ratio. Looking at the magnesium:sulphate ratio, the ratio for all the samples every week was less than 0.60, indicating that acid generation was overwhelming the dissolution of either chlorite (thin section analysis) or clinochlore ( X R D analysis). The sodiunr.sulphate ratio was less than 1:2 throughout testing with the exception of Bulk 5 in Week 12. This means that acid generation was dominating the dissolution of, most likely, albite. The potassium:sulphate ratio was less than 0.07 throughout kinetic testing, so the effect of K-feldspar dissolution on neutralization was negligible. Bulk 8 contained sericite (thin section analysis) and muscovite (XRD) which could be sources of the elevated potassium observed in Week 10. The values calculated for the NP depletion rate, the percent N P remaining, and the time to deplete N P vary significantly between those calculated from the static test data and those calculated from the kinetic test data. The static test values were calculated based on the arithmetic difference in N P between the post-batch leach material and the post-kinetic testing material. The kinetic test values, on the other hand, were calculated from the concentration in the leachate of Ca, M g , Na, K. No studies could be discovered in the literature that indicate i f one of these two methods is more accurate at calculating the different values than the other. Although these two methods have fundamental differences, one would expect them to produce comparable values. 234 Instead one finds that for most of the samples, the NP depletion rate was significantly higher using the static test data than the kinetic data. This in turn means that the percent N P remaining and time to deplete N P were much higher using the kinetic test data. When trying to provide explanations for these differences, the first factor that comes to mind is the size difference between the particles for the two different tests. The static tests use finely ground powders while the kinetic tests uses relatively coarse particles. One would expect that the NP determined using the finely ground powder of the static tests would be unrealistically high compared to the kinetic tests because the small particle size would have more reactive surface area than the coarse particles of the kinetic tests. Another possible explanation for the differences between the two methods is that the kinetic method uses actual concentrations of elements in the weekly leachate. In some cases, the concentration of these elements can be affected by processes other than the dissolution of a neutralizing mineral in response to sulphide oxidation. The formation and precipitation of secondary minerals (eg. gypsum) can significantly change the concentration of an element in the leachate. Thus both the static and kinetic methods of determining N P depletion have inherent sources of potential error. Since no studies were found that indicated which was the more accurate method, and because most of the papers reviewed over the course of this thesis used the kinetic method, it is the kinetic method values that will be used when discussing water quality in Section 4.7. 235 4.6.6.1 Cumulative NP Produced The occasional decrease in cumulative N P produced from some samples suggests that carbonate and/or silicate cations (eg. Ca, M g , Na, K ) are being removed from solution. Since calcium is the dominant cation in both of these ratios, the most likely cause of this decrease in NP is the precipitation of gypsum. Figure 4-2, however, indicated that the only samples that were oversaturated with respect to gypsum (meaning gypsum would precipitate) were Bulk 3-23 and 3-24. The high levels of N P production support the conclusion that gypsum is controlling the concentrations of calcium and sulphate for these samples. The formation and precipitation of secondary minerals such as epsomite ( M g S 0 4 - 7 H 2 0 ) or jarosite (KJre 3 a i I ) (S0 4 ) 2 (OH) 6 (Alpers et al., 1994) could account for the removal of some of the other cations from the other samples, but due to their low concentrations it is unlikely. Even if they did form, the contribution of calcium to NP for most of these samples is more significant than magnesium, sodium, or potassium put together so these secondary minerals would probably not affect the N P that much anyway. The occasional sharp increase in NP production seen in some of the samples (eg. Week 9, Bulk 1; Week 13, Bulk 3-22) suggests that mineralogy may be playing a role. As these samples weathered the mineral grains were slowly breaking down. The sharp increase in N P production could indicate when more neutralizing mineral became exposed on a grain. This increased exposure would result in more N P being released in response to acid generation. 236 Aside from the occasional fluctuation, there was relatively little variance within treatments, but significant variance between treatments. The batch leach replicates averaged approximately 20 mg C a C 0 3 / k g per week for both carbonate and silicate ratios (standard deviation = 2.8 mg/kg per week), while the non-batch leach replicates averaged approximately 775 mg C a C 0 3 / k g per week (SD=14 mg/kg for Weeks 1-8, 42 mg/kg for Weeks 9-16). The variation in results indicated by the replicate samples, is between 5% and 15%. This is a much lower range than was seen previously using the sulphate data. 4.6.6.2 N P Depletion Rate The NP depletion rate is a reflection of the N P that has been produced. Therefore an increase in NP production is seen as an increase in the N P depletion rate and vice-versa. As discussed for the previous parameters, the behaviour of Bulk 3-23 and 3-24 can be explained by the dissolution of gypsum. The decreasing rate of N P depletion for these samples probably indicates that the gypsum concentration is decreasing or that the reactive surface area is decreasing, meaning less gypsum will dissolve. For the remaining samples, the decrease in the depletion rate indicates that the NP content was decreasing. As the samples oxidized, less N P was available to react (either through decreased concentrations or through decreased reactive surface area as a result of the formation of an oxidized rind or coating by secondary minerals) and the depletion rate got lower and lower. The NP depletion rate for Dike 1/2/3 was consistently higher using the silicate ratio than the carbonate ratio. This was also seen occasionally for Bulk 1. 237 The higher NP depletion rate using the silicate ratio suggests that silicates (eg. albite, anorthite) were more important neutralizing minerals for these samples than carbonates. 4.6.6.3 Percent NP Remaining According to Figures 3-1 la and 3-1 lb, Bulk 3-23 and 3-24 had no N P remaining from Week 10 onwards. This provides further support to the fact that gypsum was controlling the concentrations of calcium and sulphate in these two samples. The high concentrations of calcium provided by gypsum dissolution made it look as though more N P was being consumed each week than was actually occurring. Thus by Week 10, based on the weekly concentrations of calcium and the original NP, the equation for % NP Remaining (Section 2) calculated that these samples had used up all their NP. The percent NP remaining in the other samples corresponds to the N P depletion rate and needs no further explanation. The sharp decrease in NP remaining for Bulk 3-22 in Week 13 was caused by a sharp increase in NP production. The fact that in Week 14 this sample appeared to have created significant NP suggests that the increase in NP production may have been caused by the dissolution of secondary minerals rather than an increase in the release of NP from the sample. 4.6.6.4 Time to Deplete NP The time to deplete NP from Bulk 3-23 and 3-24 can be discounted based on the explanations given above (ie. gypsum is controlling the aqueous concentration of calcium). For the other samples, the time to deplete N P corresponds to the production of 238 NP from these samples. The most acid-generating samples, other than Bulk 3-23 and 3-24 (ie. Dike 1/2/3, Bulk 3-2 h Bulk 3-22) had the shortest time to N P depletion while the most acid-consuming samples (Sample 1, Sample 2/3, Bulk 5) required the longest time to deplete NP. When the time required to deplete NP is compared to the time required to deplete A P (Figure 3-8d), it is obvious that the samples will use up their neutralizing minerals long before they deplete their sulphide minerals. It looks as though there is approximately an order of magnitude difference in the time required to deplete A P compared to the time required to deplete NP. This means that these samples, even the acid-consuming ones, will eventually generate acid. 4.6.7 Zinc 4.6.7.1 Cumulative Loading and Metal Leaching Rate Aside from the three neutral samples (Sample 1, Sample 2/3, and Bulk 5), the cumulative zinc loading of the other samples resembles the convex behaviour described by Ferguson and Morin (1991) for cumulative sulphate which was discussed above. The convex shape of the curve can probably be attributed to the dissolution of stored weathering products containing zinc (eg. gunningite; Alpers et al., 1994). As kinetic testing progressed, the dissolution of these stored products decreased as their concentration decreased, thereby releasing less zinc into solution. This explanation is most probable for the two samples that did not go through the batch leach. The 239 behaviour of the remaining samples might better be explained based on the formation of an oxidized rind around individual grains (shrinking core) or the coating of grains by the precipitation of secondary minerals, both of which would reduce the transport of reactants from the outer surface to the reactive inner surface. The non-batch leach replicates had approximately eight times the loading of the batch leach replicates in the first week. When one looks at the leaching rate (Figure 3-13b), however, the rates for these four samples (with the exception of Bulk 3-22) are closer than the cumulative loading would suggest. The 5-week average rates of zinc leaching for Bulk 3-2 u 3-22, 3-23, and 3-24, at the end of testing were 2.0 mg Zn/kg of sample per week, 0.5 mg/kg per week, 2.5 mg/kg per week, and 3.0 mg/kg per week, respectively. The differences between the batch leach replicates once again suggest that these replicate samples were actually quite different from each other. The other samples could be divided into two groups. Dike 1/2/3 and Bulk 8 had average leaching rates of approximately 1 mg/kg/week, while the rest of the samples had very low leaching rates, approximately 0.3 mg/kg/week. The overall leaching rate for all of the samples tested compares well to the leaching rate provided by Morin et al. (1995b). Using an international kinetic database, 185 tests provided an average zinc leaching rate of 3.06 mg/kg/week (minimum = 0.0007 mg/kg/week, maximum = 109 mg/kg/week). 240 4.6.7.2 Percent Metal Remaining The dramatic loss of zinc from Dike 1/2/3 relative to the other samples (99.76% of its original zinc content remaining at the end of testing) can be explained just by looking at original zinc concentrations. Dike 1/2/3 had the lowest original concentration of zinc (96 ppm) of all the samples, but it had leaching rates comparable to the other samples. Therefore, a greater proportion of the original zinc content was being leached from Dike 1/2/3 each week than from the other samples. This resulted in a significant loss of zinc from Dike 1/2/3 by the end of kinetic testing. The other samples have amounts of zinc remaining that correspond to their leaching rates and their original concentrations of zinc, so they do not need to be discussed any further. 4.6.7.3 Time to Deplete Metal As with the percent metal remaining, the time required for a sample to become fully depleted of a metal is dependent on the metal leaching rate. Low rates correspond to long times to metal depletion (eg. Sample 1, Sample 2/3, Bulk 5), while high rates correspond to short times to depletion (eg. Dike 1/2/3). The time required for these samples to become depleted of zinc is significantly higher (by an order-of-magnitude in some cases) than the times required to deplete the samples of their A P and NP. The time required to deplete the samples of zinc would be expected to decrease, however, because the N P will be depleted from these samples before the AP. Once the NP is depleted, one would expect that the leaching rate of zinc 241 from these samples (especially Sample 1, Sample 2/3, and Bulk 5) would increase, because the acid generated from sulphide oxidation will not be neutralized. The low p H of the drainage would mobilize greater concentrations of metals (ie. increase the leaching rate) which would lead to a decrease in the time required to deplete the samples of zinc. The amount of zinc that would be lost from the time NP is gone to the time A P has been depleted is unknown, but one would expect the leaching rate would be several times higher during this period than it was while acid was still being neutralized. Once the A P became depleted, however, it is assumed that acid generation would more-or-less stop, meaning the leaching of zinc from the sample (if there was any zinc left) would be dependent on the p H of infiltrating precipitation. 4.6.8 Copper 4.6.8.1 Cumulative Loading and Metal Leaching Rate While most of the samples display cumulative copper loading curves that are convex in shape (ie. sulphate, zinc), Bulk 3-24 exhibits a concave curve. Based on the work of Ferguson and Morin (1991), the concave shape of the Bulk 3-24 curve could indicate that mineral grains were breaking down, exposing more surface area, or that the p H of the water film around the grains dropped, increasing the leaching rate. Figure 3-4 indicates that the p H of the leachate from this solution dropped from Week 3 onwards, supporting the latter explanation. Bulk 3-23 had a relatively linear curve which implies that the leaching of copper from this sample was kinetically controlled (ie. controlled by the leaching rate, not by a change in reactants). The leaching rate for this sample (Figure 242 3-14b) was more-or-less the same from Week 2 onwards supporting the idea that kinetic control dominated the leaching of this sample. By the end of kinetic testing, the leaching rate (Figure 3-14b) for all of the samples had decreased, either from the start of testing or from Week 8 onwards. This decrease, as was seen for sulphate and zinc, suggests the formation of an oxidized rind around mineral grains or the coating of grains by precipitation of secondary minerals. The relatively low concentrations of iron and high concentrations of copper (Appendix H - l ) in the leachate from Bulk 3-2 4 and Dike 1/2/3 could allow the formation of secondary minerals. The low concentrations of copper in the leachate of the other samples suggest that the shrinking core theory may be a more reasonable explanation of the decrease in the leaching rate for these samples. The overall leaching rate of copper for all of the samples tested compares to the leaching rate provided by Morin et al. (1995b). Using an international kinetic database, Morin et al. (1995b) reported that 185 tests provided an average copper leaching rate of 0.387 mg/kg/week (minimum = 0.0003 mg/kg/week, maximum = 9.354 mg/kg/week). There were significant differences within treatments and between treatments for copper. By the end of kinetic testing, the batch leach replicates had an average loading/leaching of approximately 0.5 mg/kg per week (standard deviation = 0.4 mg/kg per week) while the non-batch leach replicates had an average loading/leaching of around 3.5 mg/kg per week (SD=2.3 mg/kg per week). The large variance in copper loading/leaching between replicates of the same treatment (coefficient of variation = 243 80% and 60% for the batch leach and non-batch leach replicates, respectively), makes it difficult to conclude that one treatment released significantly more copper into solution than the other treatment, although it appears that the non-batch leach replicates did release more copper than the batch leach replicates. Once again, the wide variation between replicates of the same treatment suggest that rather than representing variation in the results obtained, this variation represents fundamental differences in sample composition between the replicates themselves. 4.6.8.2 Percent Metal Remaining and Time to Deplete Metal The percent copper remaining in these samples (Figure 3-14c) is easily explained by looking at the leaching rate of the particular sample. Bulk 3-24 has the least amount of copper remaining because it had the highest leaching rate of all the samples. The same reasoning holds true for the other samples. The time for the samples to become completely devoid of copper is much less than that needed for zinc. In fact, the time to copper depletion is comparable to the time needed to deplete the A P from the samples. So, whereas the samples are not likely to become completely devoid of zinc, the samples are likely to release all of their copper. The time required for the copper to be depleted from the samples will decrease significantly once the NP is gone because acid will not be neutralized and the leaching rate will likely increase. Thus the times provided in Figure 3-14d are probably higher 244 than will actually be experienced. This is especially true for Sample I, Sample 2/3, and Bulk 5 which produced neutral leachate. 4.6.9 OTHER PARAMETERS 4.6.9.1 Conductivity vs. Sulphate Conductivity was plotted versus the sulphate concentration of the leachate (Figure 3-15) to determine if conductivity could be used as a predictor of sulphate concentration. Based on the results obtained from this graph, there is a very strong correlation (R2=0.97) between conductivity and sulphate concentration. Removing the two outliers (extreme data points) strengthened this correlation even further (R2=0.99). Therefore, it would have been possible to use conductivity to determine the sulphate concentration in the leachate i f kinetic testing had gone on longer. 4.6.9.2 Sulphate:Alkalinity or Acidity Ratio Sulphate:Alkalinity Ratio Alkalinity is often compared with sulphate in the form of a molar ratio to interpret kinetic test data and this ratio can provide information on the future acid-generating behaviour of the sample. The sulphate:alkalinity molar ratio was provided in Figure 3-16. A n increase in this ratio means that acidity is being produced in greater amounts or at a greater rate than alkalinity and indicates the onset of acid generation. Based on Figure 3-16, Sample 1 and Bulk 5 would be expected to go acid in the near 245 future because the ratio steadily increased from Week 12 onwards. The wide fluctuations in the ratio for Sample 2/3 prevent the prediction of future acid-generating behaviour. Sulphate:Acidity Ratio As with alkalinity, acidity can be compared with sulphate in the form of a molar ratio (Figure 3-17). If one assumes that the sulphate produced reflects the total acidity produced, this ratio indicates the amount of acidity not internally buffered by the sample. Thus a ratio of 1 indicates that there is no internal buffering and a ratio above 1 indicates that some internal buffering is taking place. Using this ratio, all of the samples that were measured for acidity experienced some internal buffering with the exception of Dike 1/2/3. In addition, the ratio appeared to be decreasing for all of the samples indicating that the internal buffering capacity of these samples was gradually disappearing. 4.6.10 Humidity Cell Disassembly The apparent lack of flowpaths indicates that the filter cloth placed above the sample in each humidity cell performed its function and dispersed the weekly rinse solution over the entire sample. If the rinse solution had been allowed to drip directly onto the surface of the sample, flowpaths may have developed in the centre of the sample. Flowpaths would have allowed the rinse solution to short-circuit the majority of the sample in the humidity cell, thereby affecting the reaction rates calculated earlier. If the rinse solution short-circuited most of the sample then the concentrations of the different constituents in the leachate would be unrealistically low because not all of the 246 reaction products would have been removed. This in turn would result in inaccurately low reaction rates being calculated for that particular sample. 4.7 Water Quality 4.7.1 Groundwater Quality It appears as though there may have been some contamination in the lab when the groundwater samples collected on May 15 were analyzed for copper. The copper concentration in almost all of the samples shows a marked increase for this sampling period compared to the ones before and after (ie. April 4 and June 19). The fact that the other parameters did not exhibit similar changes (some samples increased, others decreased) for the same sampling period suggests that the contamination was limited to copper. It is unlikely that an event such as heavy rain would produce increased concentrations of copper in all of the samples and decreased concentrations for some of the other parameters in some of the samples, but this effect has been observed (Kwong, 1991), so it could be happening here as well. Compared to the background well (BLR), the groundwater data from the monitoring wells indicates that all of the wells intercept contaminated groundwater. Therefore, the answer to the first question posed in the objectives section (Section 1.1.2), is that acid rock drainage from the Northwest Dump did in fact contaminate local groundwater. Since most natural water bodies are discharge areas for groundwater (Fetter, 1994) it is likely that the contaminated groundwater flowing from the Northwest 247 Dump discharges into Francis Lake. This discharge of contaminated groundwater is probably a leading cause of increased zinc levels observed in Francis Lake. The groundwater quality data collected from the monitoring wells suggests that there could be a spatial improvement in water quality rather than a temporal improvement. That is, the distance from the dump has a greater effect on water quality than the time that has elapsed since the dump was excavated. This spatial effect is indicated by Well G W W (adjacent to the Northwest Dump) which had lower water quality (ie. low pH, high sulphate, high metals) than Well L G W (adjacent to Francis Lake, directly between the lake and GWW). This improvement in groundwater quality as one moves further from the dump has been observed in other studies of waste rock piles as well (Herbert, 1994). A number of mechanisms could explain the spatial improvement, including: adsorption, dispersion, and dilution (Stollenwerk, 1994). Adsorption onto soil particles, especially those containing clays, iron oxides, and possibly zeolites, would result in reduced concentrations of the different constituents. Unfortunately, this adsorption could also provide a long-term source of metals to the groundwater. Dispersion in the water table would also mitigate the concentrations of various parameters. Probably the most important mechanism for reducing the concentration of the different parameters is dilution. The farther one moves from the contamination point source, the more chances there are for uncontaminated groundwater to mix with the contaminated plume, thereby reducing the concentration of the various constituents. It was initially believed that Well 248 P G W intercepted the same groundwater as Well P W W because a low hill (ie. a groundwater divide) running from the west side of the dump towards Francis Lake isolates these wells from the surrounding drainage basin. If this were true one would expect to see an improvement in water quality between P W W and P G W much like the one seen between G W W and L G W . The fact that P G W had lower water quality than P W W suggests that this well intercepts groundwater from a different source than P W W or is being contaminated by other groundwater coming from the Northwest Dump. It is possible that adsorption and dispersion are producing the better water quality in P W W compared to P G W , or it may simply be that the water intercepted by P W W is less contaminated than the water intercepted by PGW. The lower concentrations seen in L G W as compared to G W W are most likely due to dilution from less contaminated groundwater coming from the Twin Lakes system. As groundwater samples were not taken in the area of Twin Lakes, it is unknown if this groundwater is more or less contaminated than that around the Northwest Dump. With regards to temporal improvement, none of the wells show a consistent change in the various parameters (ie. increases in pH, decreases in sulphate or metals) from January to June. The occasional spikes seen in the concentrations of various constituents may indicate the effect of rain events. Heavy rains would have flushed out the reaction products from the material remaining on the dump and from the soil beneath the dump sending a pulse of contaminants into the groundwater which would be indicated by the observed spikes. With the heavy winter rains common to the area, one would expect an initial pulse of increased concentrations, followed by a steady decrease 249 in concentration as a result of dilution. If sufficient rain fell during the winter, the reaction products within the remaining dump material and the contaminated water underneath the dump should have been flushed out, thereby producing lower concentrations of the constituents in the spring and summer months (March-June). The fact that this was not observed suggests that there are mechanisms, other than infiltration and dilution, which are controlling the concentrations of the different constituents in the groundwater. It may be, as seen elsewhere (Herbert, 1994; Stollenwerk, 1994), that the precipitation and adsorption of secondary minerals, such as goethite, gypsum, or various oxides, and their subsequent dissolution is the dominant factor with regards to concentrations of the individual constituents. Gibson and Pantelis (1988) reported that it takes at least four years before groundwater in the vicinity of a overburden rock dump will respond to measures taken to reduce the leaching of metals from the dump. Thus, the fact that no improvements in water quality were seen immediately after excavation occurred is not particularly surprising. To better understand the physical and chemical mechanisms controlling the concentrations of the different parameters measured from the monitoring wells, hydrogeochemical models, such as M T N T E Q A 2 or P H R E E Q E could be employed. P H R E E Q E allows the mixing of solutions to predict the effects of dilution on the concentrations of different parameters, while M I N T E Q A 2 can predict what happens to these solutions as they are transported. Other methods that could be used to better define the hydrogeological system around the dump include slug tests, to determine hydraulic conductivity, and tracer tests, to determine dilution effects. 250 4.7.2 Water Qual i ty Comparison 4.7.2.1 Groundwater The two sets of data indicate that the groundwater in the immediate vicinity of the Northwest Dump was contaminated both prior to and after the dump's excavation. The p H of the water taken from Well G W W after excavation was lower than that taken from the excavated pit prior to excavation, and the levels of conductivity, sulphate, zinc, copper, and calcium were higher in Well G W W than in the excavated pit, so excavation of the dump has not yet led to any noticeable improvements in the quality of local groundwater. This conclusion must be qualified, however, because the source of the water in the excavated pit was not determined. Groundwater from the dump is probably a major component of the water in the excavated pit, but surface runoff and precipitation are also likely contributors. The dilution provided by the surface runoff and precipitation would have increased the p H and lowered the levels of the other parameters. Thus, the quality of the groundwater prior to dump excavation may actually have been worse than that measured in Well G W W after excavation, but was not apparent due to the masking effect of the surface runoff and precipitation. It is speculated that removal of approximately 1 million tonnes of acid-generating material from this dump would have to result in an improvement in the quality of the groundwater flowing from this area. Unfortunately, when this improvement will be 251 observed and how much the quality will improve is impossible to determine from the current data. It must be pointed out, however, that a thin layer (~lm) of rock still remains in the area of the dump and could potentially be a long-term source of metals and acidity to the local groundwater. 4.7.2.2 Francis Lake Due to inconsistencies in data collection and reporting, large amounts of data collected for Francis Lake could not be used to determine if excavation of the dump led to improved water quality in the lake. The pre-excavation data did not contain values for many of the dissolved metals (eg. zinc, copper, aluminum) that were measured and reported in the post-excavation data. As well, a lack of data for 1995 prevented the most recent pre-excavation data from being compared to the most recent post-excavation data. The only data possessed from 1995 were the results of a lake survey. Unfortunately, the results of the most recent lake survey conducted in August, 1996 were reported in orders-of-magnitude lower detection limits than the 1995 results, preventing the two sets of data from being compared. Therefore, only those parameters that were reported in both the pre-excavation and post-excavation data were compared. Some of the parameters in Table 3-16 indicate less contamination since excavation (pH, Conductivity, magnesium) while others indicate more (sulphate, alkalinity, calcium). The higher p H and lower levels of conductivity and magnesium in 1996 suggest that water quality in the lake has improved since 1994, but the higher levels 252 of sulphate and calcium and lower levels of alkalinity suggest that water quality has actually gotten worse. The last three months of lake data were examined independently of the rest to determine if the most recent lake data indicated improvement in water quality. Again, some parameters showed improvement (pH, sulphate, alkalinity) while others did not (conductivity, calcium, magnesium) so excavation of the dump has not yet led to improvement of water quality in Francis Lake. As with the groundwater, the removal of a large amount of acid-generating material from the dump should eventually lead to an improvement in water quality, but when this improvement will occur and how much the quality will improve is impossible to calculate with the given data. Additional data, reported for the same parameters and using the same detection limits as the pre-excavation data, is needed to properly assess the effect of dump excavation on water quality in Francis Lake. Based on the quality of post-excavation groundwater, it is unlikely that water quality in Francis Lake will improve in the near future. 4.7.3 Predicted Water Quality The values calculated for water quality in the flooded pit are a worst-case scenario for a number of reasons: 1) the concentrations in the kinetic leachate were below the detection limits for Z n and C u in Sample 1 and Sample 2/3 and for C u in Pit 1/2, 2) the loading/leaching rates used were based on kinetic tests, 3) it is assumed that the rates will not decrease as reactants are used up or as reactive surface area decreases, 253 4) it is assumed that all of the material currently on the benches will remain there and that all of it will react. The second point raised above is important because the material used in the kinetic tests got rinsed more often and more regularly than would be the case for the material on the benches. The leaching rates used were also taken from kinetic tests, meaning the rates are based on material composed of much smaller particles (ie. greater surface area) than the bench material. For these reasons, the time calculated before the concentrations of zinc and copper exceed discharge permit guidelines is probably shorter than it otherwise would be. Another reason the time calculated is probably shorter than it will actually be is because wind mixing will be significant in the cap and therefore the annual precipitation will provide additional dilution. Wind mixing will thoroughly combine the annual precipitation with the water in the cap, in effect adding that extra volume to the volume of the cap. This dilution will occur continuously, even though once the crest of the pit is reached, the excess volume will discharge towards the Beach Dump. If one assumes that there are 1800 mm of uncontaminated precipitation each year, then zinc concentrations will exceed permit guidelines in 52 years and copper concentrations in 3 years. If the loading/leaching rates of the kinetic test samples are reduced by half to better represent the overall rates actually occurring in the bench material, and the volume of the bench material is reduced by a quarter to account for erosion of this material into 254 the flooded pit, the time before metal concentrations in the freshwater cap exceed the discharge permit increases to 128 years for zinc and 7.4 years for copper. Unfortunately, the time to deplete NP in this material, based on kinetic data, is on the order of 15 years. After the NP has been depleted, the loading/leaching rates will probably increase since the acid generated from sulphide oxidation (which will still be occurring) will not be neutralized. Therefore, the actual time for metal concentrations in the freshwater cap to reach discharge permit guidelines is probably somewhere in between the two times given. In order to be conservative, the worst-case scenario should be looked upon as the more likely of the two and plans made accordingly. 255 5. CONCLUSIONS AND RECOMMENDATIONS 5.1 Conclusions The batch leach is an effective method for removing stored reaction products from samples of weathered waste rock. Based on sulphate concentrations in the leachate, the batch leach removed approximately 70% of the stored products from the samples. Comparing the batch leach samples to the non-batch leach samples, the removal of these stored products is believed to have significantly reduced the time required for the humidity cell material to reach equilibrium, although the short duration of the kinetic tests precludes further speculation. The batch leach also indicates that significant amounts of sulphate and metals would have been released from the Northwest Dump material upon initial submergence in the flooded pit and/or during the first heavy rain. Static prediction tests indicated that almost all of the Northwest Dump material had a high potential for producing acid rock drainage. The net neutralization potential for nearly all of the samples was negative and the NP:AP ratio was equal to or less than one for all of the samples. It is felt that the standard (Sobek) method of acid-base accounting tends to overestimate the neutralization potential of a sample, leading to erroneous conclusions regarding a sample's acid-consuming ability. The modified and carbonate methods are felt to provide more realistic estimates of a sample's neutralization potential and a combination of the two probably provides the most reasonable estimation of the actual NP available in a sample. 256 Due to the short duration of the kinetic tests, one must be careful when drawing conclusions from the results obtained. The results of the kinetic prediction tests indicated that the majority of the Northwest Dump material would produce moderately acidic leachate and release elevated concentrations of sulphate and metals (zinc, copper, calcium) over the long-term, although some of the dump material did produce neutral leachate throughout kinetic testing and released correspondingly low levels of sulphate and metals. Alkalinity levels and various ratios (sulphate:alkalinity, calcium, magnesium:sulphate) indicate that the material from some of samples that were neutral during testing will likely start producing acidic leachate in the near future. Neutralizing minerals in these samples will be depleted long before the sulphide is completely oxidized, so the rates calculated for metal leaching will probably increase with time as unneutralized acid makes contact with the dump material. It is also felt that "bubblers" are an effective substitute for controlling air flow through a humidity cell when individual air supplies for each cell are unavailable. The dike separating Twin Lakes from the open pit is highly acid-generating and is capable of releasing significant amounts of zinc and copper. Ac id rock drainage from the dike, characterized by low p H and elevated levels of zinc and copper, is probably the major source of contamination to Twin Lakes. Water quality analysis indicated that groundwater in the immediate vicinity of the Northwest Dump is contaminated as it displays depressed p H and elevated levels of sulphate and metals compared to background levels taken from the north side of Francis 257 Lake. Excavation of the dump has not yet led to any noticeable improvements in the quality of either the local groundwater or Francis Lake. Predictions made regarding water quality of the freshwater cap in the flooded pit indicate that the dump material on the upper benches could cause zinc concentrations to exceed discharge permit guidelines within 45 years and copper concentrations to exceed discharge permit guidelines within 3 years. These time frames are probably worst-case scenarios, but due to uncertainties in the mass of material on the benches of the pit and the long-term leaching rate of this material, it is advised that these time frames be used for any decisions made regarding the material on the upper benches of the pit. 5.2 Recommendations To better understand the groundwater regime in the area of the Northwest Dump and Francis Lake, and to determine i f the quality of groundwater or Francis Lake has improved, it is recommended that additional monitoring wells be installed between the dump and the lake and between Twin Lakes and Francis Lake. These wells should be installed deep enough so that they can be sampled even in the dry summer months to allow more complete characterization of the groundwater system. Geochemical models such as P H R E E Q E or M I N T E Q A 2 could also be used to better characterize the groundwater regime in terms of sources of contamination and dilution. To prevent further contamination of Francis Lake and the Stephens Creek watershed, it is recommended that Twin Lakes be drained into the pit. The drainage 258 ditch to accomplish this task has already been completed and all that is necessary is government approval. The dike separating Twin Lakes from the pit has released acid rock drainage low in p H and high in dissolved metals, especially zinc and copper, into Twin Lakes. Twin Lakes drains into Francis Lake and, in addition to the contaminated groundwater from the Northwest Dump, is probably a significant source of contamination to Francis Lake. Based on the predictions made regarding the effect of the Northwest Dump material on the water quality of the freshwater cap in the flooded pit, it is recommended that the erosion of this material into the pit be artificially enhanced. Due to the difficulty in accessing the upper benches of the pit and their current state of disrepair, using equipment such as a dozer to move this material is inadvisable. The use of high-pressure hoses along the crest of the pit might be effective at washing some of this material further down the pit wall so that it becomes submerged. Submarine disposal of acid-generating material is recognized as an effective means of controlling acid rock drainage (Rawson Academy, 1992; Morin, 1993; Aube et al., 1995) and should more-or-less halt sulphide oxidation in the submerged Northwest Dump material. 259 REFERENCES Alpers, C . N . , D .W. Blowes, D .K. Nordstrom, and J.L. Jambor, 1994. "Secondary Minerals and Ac id Mine-water Chemistry". In Shortcourse Handbook on Environmental Geochemistry of Sulfide Mine-Wastes. J.L. Jambor and D.W. Blowes (ed.'s). Mineralogical Association of Canada; Nepean, O N . Chapter 9, pp. 247-270. Aube, B . C , L . C . St-Arnaud, S.C. Payan, and E . K . Yanful, 1995. "Laboratory Evaluation of the Effectiveness of Water Covers for Preventing Acid Generation from Pyritic Rock". In Conference Proceedings of Sudbury '95: Mining and the Environment. T.P. Hynes and M . C . Blanchette (ed.'s). May 28-June 1; Sudbury, O N . Volume II, pp. 495-504. BHP, 1996. "Island Copper Mine Closure Plan Addendum - Appendices A to E " . BHP Minerals Canada Ltd., Island Copper Mine. 340 p. B H P , 1994. "Island Copper Mine Closure Plan". B H P Minerals Canada Ltd., Island Copper Mine. 300 p. B H P , 1986. "Methods Manual". Environmental Department, Island Copper Mine. pp. 9.1-9.6, 24.1-24.3,29.1,29.2. BHP-Utah Mines Ltd. and Rescan Environmental Services, 1988. "South Wall Pushback: Description of Project and Related Environmental Issues". Report prepared for the Vancouver Island Regional Reclamation Advisory Committee. 90 p. Blowes, D .W. and C.J. Ptacek, 1994. "Acid-neutralization Mechanisms in Inactive Mine Tailings". In Shortcourse Handbook on Environmental Geochemistry of Sulfide Mine-Wastes. J.L. Jambor and D .W. Blowes (ed.'s). Mineralogical Association of Canada; Nepean, O N . Chapter 10, pp. 271-292. Broughton, L . M . and P . M . Healey, 1992. "Mine site decommissioning: Technology for assessment and control of acid rock drainage". In Environmental Issues and Management of Waste in Energy and Mineral Production; R .K. Singhal, A . K . Mehrotra, K. Eytas, and J. Collins (ed's). Proceedings of the 2 n d International Conference on Environmental Issues and Management of Waste in Energy and Mineral Production; Calgary, Alberta, Canada. September 1-4. Volume 2, pp. 789-797. 260 Broughton, L . M . and A . M . Robertson, 1991. "Modelling of Leachate Quality from Acid Generating Waste Rock Dumps". In Proceedings of the 2 n d International Conference on the Abatement of Acidic Drainage. September 16-18; Montreal, Quebec. Tome 3, pp. 341-361. Cargill, D . G . , 1975. "Geology of the Island Copper Mine". Ph.D. Thesis, University of British Columbia. 250 p. Coastech Research Inc., 1991. "Acid Rock Drainage Prediction Manual". Mine Environment Neutral Drainage ( M E N D ) Project 1.16. lb. 120 p. Coulometrics Incorporated, 1984. "Coulometrics Incorporated Instruction Manual for Models 5010 and 5030". Coulometrics Incorporated; New York, New York. 200 p. Dagenais, P., 1996. "Personal observation of former location of Northwest Dump during site visit from January 29 - February 1, 1996". Dagenais, P., 1995. "Personal observation of dumping occurring on the north edge of the open pit during a site visit from October 29 - November 1, 1995". Downing, B . W , 1996. "Personal communication regarding analysis of acid rock drainage data - July, 1996". E P A , 1994. "Technical Document: Acid Mine Drainage Prediction - Draft". U.S. Environmental Protection Agency; Washington, D C . 49 p. Ferguson, K . D . and K . Morin, 1991. "The Prediction of Ac id Rock Drainage - Lessons from the Database". In Proceedings of the 2 n d International Conference on the Abatement of Acidic Drainage. September 16-18, Montreal, Quebec. Tome 3, pp. 83-106. Fetter, C .W. , 1994. Applied Hydrogeology. Macmillan College Publishing Company, Inc.; New York, New York. 691 p. Fleming, J.A. , 1983. "Island Copper". In Proceedings of the G A C / M A C / C G U Joint Annual Meeting; Victoria, B C . Field Trip 9, Guidebook to Mineral Deposits of Vancouver Island, pp. 21-35. Gibson, D . K . and G. Pantelis, 1988. "Forecasting the Effect of Mine Site Rehabilitation Works on Local Ground Water Quality". In Conference Proceedings of Mine Drainage and Surface Mine Reclamation. April 19-21, Pittsburgh, Pennsylvania. Volume I: 248-252. 261 Green, R . N and K. Klinka, 1994. "A Field Guide to Site Identification and Interpretation for the Vancouver Forest Region". B C Ministry of Forests Research Program, Research Branch. Victoria, B C . In B H P , 1996 (see above). Harries, J.R. and A. I .M. Ritchie, 1983. "The Microenvironment Within Waste Rock Dumps Undergoing Pyrite Oxidation". In Progress in Biohydrometallurgy G.Rossi and A . E . Torma (ed.'s). Associazione Mineraria Sarda; Inglesias, Italy, pp. 377-392. Harries, J.R. and A. I .M. Ritchie, 1981. "The Use of Temperature Profiles to Estimate the Pyritic Oxidation Rate in a Waste Rock Dump from an Open Cut Mine". Water, Air, and Soil Pollution; 15:405-423. Helsel, D.R. , 1987. "Advantages of nonparametric procedures for analysis of water quality data". Hydrological Science Journal, 32(2) 6:179-190. Herbert, R.B. Jr., 1994. "Metal Transport in Groundwater Contaminated by Acid Mine Drainage". Nordic Hydrology, 25:193-212. Home, L A . , 1996a. "Personal communication with environmental manager in form of a facsimile message - June 5, 1996". Home, L A . , 1996b. "Personal communication with environmental manager - October 28, 1996". Home, L A . , 1995. "Personal communication with environmental manager during site visit from October 29 - November 1, 1995". Jambor, J .L. , 1994. "Mineralogy of Sulfide-rich Tailings and Their Oxidation Products". In Shortcourse Handbook on Environmental Geochemistry of Sulfide Mine-Wastes. J.L. Jambor and D .W. Blowes (ed.'s). Mineralogical Association of Canada; Nepean, O N . Chapter 5, pp. 59-102. Jayawardena, A . W . and F. Lai , 1989. "Time series analysis of water quality data in Pearl River, China". Journal of Env. Eng.; Volume 115, N o . l , February 1989: 590-607. Kleinmann, R.L.P. , D . A . Crerar, and R.P. Pacelli, 1981. "Biogeochemistry of Acid Mine Drainage and a Method to Control Ac id Formation". Mining Engineering, March, pp. 300-305. Knapp, R . A . , C M . Pettit, and J.C. Martin, 1995. "Waste Rock Sampling Techniques for Acid Rock Drainage (ARD) Assessment". In Conference Proceedings of Sudbury '95: Mining and the Environment. T.P. Hynes and M . C . Blanchette (ed.'s). May 28-June 1; Sudbury, O N . Volume II, pp. 587-596. 262 Koyanagi, V . M . and A. Panteleyev, 1992. "Natural Acid-Drainage in the Mount Mclntosh/Pemberton Hills Area, Northern Vancouver Island (92L/12)". Geological Fieldwork, Paper 1993-1. pp. 445-450. Kwong, Y . T . J . , 1994. "Prediction and Prevention of Ac id Rock Drainage from a Geological and Mineralogical Perspective". Mine Environment Neutral Drainage ( M E N D ) Report 1.32.1. 47 p. Kwong, Y . T . J . , 1991. "Acid Generation in Waste Rock as Exemplified by the Mount Washington Minesite, British Columbia". In Proceedings of the 2 n d International Conference on the Abatement of Acidic Drainage. September 16-18; Montreal, Quebec. Tome l , p p . 175-190. Lapakko, K . , 1994. "Evaluation of Neutralization Potential Determinations for Metal Mine Waste and a Proposed Alternative". In Proceedings of the International Land Reclamation and Mine Drainage Conference and the 3 r d International Conference on the Abatement of Acidic Drainage. April 24-29; Pittsburgh, PA. Volume 1: Mine Drainage, pp. 129-137. Lawrence, R .W. and Y . Wang, 1996. "Determination of Neutralization Potential for A R D Prediction". Report prepared for Environment Canada and Hudson Bay Mining & Smelting Company. 32pp. Lawrence, R.W. and A . M . Robertson, 1994. "Acid Rock Drainage: Understanding the Problems - Finding Solutions". Shortcourse; Canadian Institute of Mining, Metallurgy, and Petroleum (CIM) District 6 Annual General Meeting. October 11-14; Vancouver, B C . Lawrence, R.W. , 1990. "Laboratory Procedures for the Prediction of Long Term Weathering Characteristics of Mining Wastes". In "Acid Mine Drainage: Designing for Closure" J.W. Gadsby, J.A. Malick, and S.J. Day (ed.'s). Papers presented at the G A C / M A C Joint Annual Meeting. May 16-18; Vancouver, B C . pp. 131-140. Leitch, C . H . B . , 1988. "Brief summary of mafic-potassic style alteration as applied to the Island Copper porphyry deposit, British Columbia". Unpublished report, 5 p. In Lister (1994) - see reference below. L i , M . , 1990. "Island Copper Mine, Northwest Dump Study - Stage I Report". Unpublished report submitted to Island Copper Mine. 96 p. Lister, D:, 1994. "An Assessment of Ac id Rock Drainage Potential of Waste Rock and Implications for Long Term Weathering of the North Dump at Island Copper Mine, Port Hardy, B C " . M.A.Sc . Thesis, University of British Columbia. 217 p. 263 Lohani, B .N . and M . M . Wang, 1987. "Water quality data analysis in Chung Kang River". Journal of Env. Eng.; Volume 113, No. 1, February 1987: 186-195. Morin, K . , N . Hurt, and K . D . Ferguson, 1995a. "Measured Rates of Sulphide Oxidation and Ac id Neutralization in Kinetic Tests: Statistical Lessons from the Database". In Conference Proceedings of Sudbury '95: Mining and the Environment. T.P. Hynes and M . C . Blanchette (ed.'s). May 28-June 1; Sudbury, O N . Volume II, pp. 525-536. Morin, K . , N . Hurt, and K . D . Ferguson, 1995b. "Measured Rates of Copper and Zinc Leaching in the International Kinetic Database". In Proceedings of the 19 t h Annual British Columbia Mine Reclamation Symposium. June 19-23, Dawson Creek, B C . pp. 255-263. Morin, K. , N . Hutt, and I. A . Home, 1995c. "Prediction of Future Water Chemistry from Island Copper Mine's On-Land Dumps". In Proceedings of the 19 t h Annual British Columbia Mine Reclamation Symposium. June 19-23, Dawson Creek, B C . pp. 224-233. Morin, K. and N . Hutt, 1994a. "An Empirical Technique for Predicting the Chemistry of Water Seeping from Mine-Rock Piles". In Proceedings of the International Land Reclamation and Mine Drainage Conference and the 3 r d International Conference on the Abatement of Acidic Drainage. April 24-29; Pittsburgh, PA. Volume 1: Mine Drainage, pp. 12-19. Morin, K. and N . Hutt, 1994b. "Observed Preferential Depletion of Neutralization Potential over Sulfide Minerals in Kinetic Tests: Site-Specific Criteria for Safe N P / A P Ratios". In Proceedings of the International Land Reclamation and Mine Drainage Conference and the 3 r d International Conference on the Abatement of Acidic Drainage. April 24-29; Pittsburgh, PA. Volume 1: Mine Drainage, pp. 148-156. Morin, K . , 1993. "Rates of Sulphide Oxidation in Submerged Environments: Implications for Subaqueous Disposal". In Proceedings of the 17 t h Annual British Columbia Mine Reclamation Symposium. May 4-7; Port Hardy, B C . pp. 235-247. Morin, K . , E . Gerencher, C . E . Jones, and D . E . Konasewich, 1991. "Critical Literature Review of Acid Rock Drainage from Waste Rock". Mine Environment Neutral Drainage ( M E N D ) Report 1.11.1. 175 p. Morin, K , 1990. "Problems and Proposed Solutions in Predicting Acid Drainage with Acid-Base Accounting". In "Acid Mine Drainage: Designing for Closure" J.W. Gadsby, J.A. Malick, and S.J. Day (ed.'s). Papers presented at the G A C / M A C Joint Annual Meeting. May 16-18; Vancouver, B C . pp. 93-107. 264 Nicholson, R . V . , 1994. "Iron-sulfide Oxidation Mechanisms: Laboratory Studies". In Shortcourse Handbook on Environmental Geochemistry of Sulfide Mine-Wastes. J.L. Jambor and D.W. Blowes (ed.'s). Mineralogical Association of Canada; Nepean, O N . Chapter 6, pp. 163-183. Nicholson, R . V . , R .W. Gillham, and E.J . Reardon, 1990. "Pyrite oxidation in carbonate-buffered solution: 2. Rate control by oxide coatings". Geochim. Cosmochim. Acta, Vol . 54:395-402. Perello, J .A. , J .A. Fleming, K.P. O'Kane, P.D. Burt, G.A. Clarke, M . D . Himes, and A . T . Reeves, 1995. "Porphyry copper-gold-molybdenum deposits in the Island Copper Cluster, northern Vancouver Island, British Columbia". In Special Volume 46: Porphyry Deposits of the Northwest Cordillera of North America. T . G . Schroeder (ed.). Canadian Institute of Mining, Metallurgy, and Petroleum; Montreal, Quebec, pp. 214-238. Pool, D . L . and R . M . Balderrama, 1994. "Evaluation of Humidity Cell Parameters: Their Effect on Precision and Repeatability". In Proceedings of the International Land Reclamation and Mine Drainage Conference and the 3 r d International Conference on the Abatement of Acidic Drainage. April 24-29; Pittsburgh, PA. Volume 2: Mine Drainage, pp. 326-333. Price, W . A . and J.C. Errington, 1995. "Acid Rock Drainage Guidelines for Minesites in British Columbia". British Columbia Ministry of Energy, Mines and Petroleum Resources, Mine Review and Permitting Branch, 29 p. Rawson Academy of Aquatic Science, 1992. " A Critical Review of M E N D Studies Conducted to 1991 on Subaqueous Disposal of Tailings". Mine Environment Neutral Drainage ( M E N D ) Project 2.11.1 d. Ritchie, A . J . M . , 1994a. "The Waste Rock Environment". In Shortcourse Handbook on Environmental Geochemistry of Sulfide Mine-Wastes. J.L. Jambor and D .W. Blowes (ed.'s). Mineralogical Association of Canada; Nepean, O N . Chapter 5, pp. 133-161. Ritchie, A . J . M . , 1994b. "Sulfide Oxidation Mechanisms: Controls and Rates of Oxygen Transport". In Shortcourse Handbook on Environmental Geochemistry of Sulfide Mine-Wastes. J.L. Jambor and D .W. Blowes (ed.'s). Mineralogical Association of Canada; Nepean, O N . Chapter 8, pp. 201-245. Robertson, W.D. , 1994. "Iron-sulfide Oxidation Mechanisms: Laboratory Studies". In Shortcourse Handbook on Environmental Geochemistry of Sulfide Mine-Wastes. J.L. Jambor and D.W. Blowes (ed.'s). Mineralogical Association of Canada; Nepean, O N . Chapter 6, pp. 163-183. 265 Rowden, A . , 1996. "Personal communication with senior mine engineer on May 14, 1996 regarding excavation of the Northwest Dump". Rowden, A . , 1995. "Personal communication with senior mine engineer during site visit from October 29 - November 1, 1995 regarding disposal of Northwest Dump material". Scharer, J . M . , E . C . M . Kwong, R .V . Nicholson, C M . Pettit, and D .W. Chambers, 1995. "Factors Affecting Acid Rock Drainage Production: Kinetics of Sulphide Oxidation". In Conference Proceedings of Sudbury '95: Mining and the Environment. T.P. Hynes and M . C . Blanchette (ed.'s). May 28-June 1; Sudbury, O N . Volume II, pp. 451-458. Sherlock, E .J . , 1995. "Evaluation of Static and Kinetic Prediction Test Data and Comparison with Field Monitoring Data". M.A.Sc . Thesis, University of British Columbia. 196 p. Sherlock, E . J . , R .W. Lawrence, and R. Poulin, 1995. "On the neutralization of acid rock drainage by carbonate and silicate minerals". Environmental Geology 25:43-54. Smith, L . , D . L . Lopez, R. Beckie, K. Morin, R. Dawson, and W. Price, 1995. "Hydrogeology of Waste Rock Dumps". Contract No.:23440-4-1317/01-SQ. Final Report to Department of Natural Resources Canada. 130 p. Sobek, A . A . , W . A . Schuller, J.R. Freeman, and R . M . Smith, 1978. "Field and Laboratory Methods Applicable to Overburdens and Minesoils". E P A 600/2-78-054, U.S. Environmental Protection Agency, Washington, D C . pp. 45-55. SRK, 1992. "Mine Rock Guidelines: Design and Control of Drainage Water Quality". Project No. 93301, prepared for Saskatchewan Environment and Public Safety, Mines Pollution Control Branch. 156 p. SRK, 1989. "Draft Ac id Rock Drainage Technical Guide, Volume 1". British Columbia Acid Mine Drainage Task Force Report 5.10. BiTech Publishing, Richmond, B C . 261 p. Standard Methods, 1992. Standard Methods for the Examination of Water and Wastewater".' A . E . Greenberg, L.S. Clesceri, and A . D . Eaton (ed.'s). American Public Health Association; Washington, D C . pp. 4-131 to 4-135. Stollenwerk, K . G . , 1994. "Geochemical interaction between constituents in acidic groundwater and alluvium in an aquifer near Globe, Arizona". Applied Geochemistry, Vol . 9, pp. 353-369. 266 Stumm, W. , and J.J. Morgan, 1981. Aquatic Chemistry - A n Introduction Emphasizing Chemical Equilibria in Natural Waters, 2 n d Edition. Wiley-Interscience; New York, New York. 583 p.. Vos, R.J. and T.J . O'Hearn, 1993. "The use of zeolite to treat A R D from Britannia minesite". In Proceedings of the 17 t h Annual British Columbia Mine Reclamation Symposium. May 4-7; Port Hardy, B C . pp. 223-232. White IE, W.W. and T . H . Jeffers, 1994. "Chemical Predictive Modeling of Ac id Mine Drainage from Metallic Sulfide-Bearing Waste Rock". In Environmental Geochemistry of Sulfide Oxidation, American Chemical Society, pp. 608-630. White III, W . W . , E . M . Trujillo, and C. Lin, 1994. "Chemical Predictive Modeling of Acid Mine Drainage from Waste Rock: Model Development and Comparison of Modeled Output to Experimental Data". In Proceedings of the International Land Reclamation and Mine Drainage Conference and the 3 r d International Conference on the Abatement of Acidic Drainage. April 24-29; Pittsburgh, PA. Volume 1: Mine Drainage, pp. 157-166. White III, W . W . and S.S. Sorini, 1993. "Standard Test Method for Accelerated Weathering of Solid Waste Using a "Modified" Humidity Cell". Unpublished draft method prepared for A S T M . Whiting, D . L . , 1985. "Surface and Groundwater Pollution Potential". In Design of Non-Impounding Mine Waste Dumps. M . K . McCarter (ed.). American Institute of Mining, Metallurgical, and Petroleum Engineers, Inc. Chapter 9, pp. 91-98. Young, M . J . and E.S. Rugg, 1971. "Island Copper Deposit". Western Miner, February: 31-40. 267 Appendix A "Analytical Methods" 268 fill A Barnant Digital p H meter/controller (Model No. 501-3400) and an A T I Orion low maintenance p H triode were used to measure p H on a weekly basis. Samples being analyzed for acidity had p H measured prior to the addition of hydrogen peroxide. Samples being analyzed for alkalinity had p H measured prior to the addition of sodium hydroxide. The p H meter was calibrated using standard solutions of p H 7.00 and p H 4.00 prior to sample measurement. The p H meter was checked for drift after the samples being analyzed for acidity were measured, after the samples being analyzed for alkalinity were measured, and after the acidity titrations had been completed. Conductivity Conductivity was measured on a weekly basis using a Metrohm 660 Series 01 Conductometer. The conductometer was calibrated using a standard conductivity solution (11.67 mS/cm) before the start of each sample run and was checked for drift after the samples had been measured. Measurements were recorded in mS/cm. Sulphur and Sulphate Samples of the original material, the bucket leach residue, and the post kinetic testing material were analyzed for sulphur and sulphate content. Analysis was conducted in the Centre for Coal and Mineral Processing (CMP) at the University of British Columbia using gravimetric analysis. In gravimetric analysis for sulphur, 2 gm of sample (-100 mesh) are digested with nitric acid and potassium chlorate to dryness. The sample is then dissolved in 5% hydrochloric acid and filtered, and final precipitation occurs as barium sulphate. In gravimetric analysis for sulphate, 3 gm of sample are boiled for 5 minutes in 50 ml of a 5% hydrochloric acid solution. The resulting soluble sulphate is then filtered and precipitated as barium sulphate. Sulphur and sulphate contents are determined by weight using a 4-decimal analytical balance (Metter A C 100 model). The weight of the precipitate is adjusted by a factor corresponding to the species being determined (weight x 32/233 for sulphur, weight x 96/233 for sulphate) and the concentration is reported as a percentage of the original sample. Sulphate was analyzed in the weekly humidity cell leachate using two different methods; gravimetric analysis and turbidity. For the first six weeks of testing, sulphate was determined by gravimetric analysis following the above method, but substituting 200 ml of the leachate solution for the 3 gm of pulp sample. 269 Starting on the seventh week of testing, sulphate concentration was determined using a Perkin-Elmer Lambda 3 U V / V I S Spectrophotometer and following the turbidimetric method described in Standard Methods for the Examination of Water and Wastewater (1992). Sulphate measurements were taken within two days after collection of the cell leachate using the following procedure: • a calibration curve was prepared each session. The curve was prepared by analyzing solutions of known sulphate concentration (0 mg/L, 10 mg/L, 20 mg/L, 30 mg/L, 40 mg/L) and plotting the resulting absorbances on a graph of absorbance vs. concentration. A second-order polynomial was used to fit the data points and the equation of this curve was used to calculate the sulphate concentration of the leachate sample. • 100 ml of sample was used (or portion made up to 100 ml by adding distilled water) to which 20 ml of Buffer Solution A (an acetic acid buffer) was added • barium chloride was added, using a 1 ml measuring spoon which held 1.6 g of barium chloride crystals, and the sample stirred at constant speed for exactly one minute • immediately after stirring stopped, the solution was poured into a 1 cm spectrophotometer cell and absorbance was recorded exactly five minutes after stirring was ended. • a 20 mg/L standard solution was measured in the middle of the sample set and 10 mg/L and 30 mg/L standard solutions were measured at the end of the sample set. Any significant changes between the standards and the calibration curve resulted in the standards being reanalyzed and i f significant differences were still present, a new calibration curve was plotted and the entire sample set was reanalyzed. • samples were diluted to provide a sulphate concentration of under 40 mg/L. Standard Methods suggested this procedure because above 40 mg/L sulphate, the barium sulphate suspension loses stability and accuracy is affected. • background turbidity was measured by adding the buffer solution to the sample and measuring absorbance. This value was then subtracted from the absorbance measured after the barium chloride was added and the resulting value was compared to the calibration curve. • triplicates of at least two of the eleven samples were run in each sample set. Significant differences between replicates of the same sample resulted in the sample and its replicates being reanalyzed. Total Alkalinity Only those samples with a p H of 5 or higher were analyzed for alkalinity. Below p H 5, very little alkalinity will be present in the sample. The total alkalinity of the sample leachate was calculated on a weekly basis. The procedure used was that described in the Island Copper Mine (ICM) Environmental Methods Manual (BHP, 1986). The following steps describe the basic procedure: 270 100 ml of sample sample was placed on a stirring apparatus and titrated with HC1 to p H 4.5 volume of H Q used was recorded total alkalinity was calculated using the following equation: Total alkalinity (mg C a C 0 3 / L ) = 50.000 x N x V V s Where : N = normality of HC1 used (0.0197N) V = volume (ml) of HC1 used V s = volume of sample (100 ml) It is assumed that alkalinity is due to bicarbonate ion and that carbonate and hydroxyl concentrations are nil. Only those samples with a p H of less than 5 were analyzed for acidity. Above p H 5, very little acidity will be present in the sample. Sample leachate acidity was measured on a weekly basis using the procedure outlined in the I C M Environmental Methods Manual (BHP, 1986): • 50 ml of sample was used • 5 drops of hydrogen peroxide were added and the sample boiled for 5 minutes. This step removes interference caused by iron and aluminum by precipitating them out of solution prior to titration. • sample is cooled to room temperature, placed on a stirring apparatus, and titrated with N a O H to p H 8.3 • volume of N a O H used is recorded • acidity is calculated using the following equation: Leachate was analyzed for metals on a weekly basis using atomic absorption spectroscopy (AAS). Analysis was performed in the C M P using a Perkin-Elmer 2100 Atomic Absorption Spectrophotometer. Calcium, magnesium, sodium, potassium, Acidity Acidity (mg C a C 0 3 / L ) = 50,000 x N x V V s Where : N = normality of N a O H used (0.0184N, 0.0165N) V = volume (ml) of N a O H used V s = volume of sample (50 ml) Metals 271 copper, and zinc were analyzed every week. For the first four weeks, additional metals analyzed for included iron, lead, aluminum, cobalt, manganese, molybdenum and cadmium. After four weeks, the following metals were analyzed every four weeks : iron, aluminum, manganese, cadmium, and lead. The following A A procedure was used to perform the analyses: Ca . M g . Na. K , Fe Atomic absorption analysis was performed with a matrix correction. A pipette was used to transfer a suitable aliquot from the received solution into a 200 ml volumetric flask. To this solution was added 10 ml of lanthanide chloride solution (175g/L) and 5 ml of hydrochloric acid. The sample was bulked up to the designated mark and the sample was read against standards with the same lanthanum chloride concentration. C u . Zn , Pb. M n . A l Atomic absorption analysis was performed without a matrix correction. The sample was read directly as received or after an appropriate aliquot was taken. If an aliquot was taken, the sample was read against previously prepared pure standards. 272 Appendix B B - l : P a r t i c l e S ize A n a l y s i s B - 2 : T h i n Sec t ion A n a l y s i s B - 3 : X R D A n a l y s i s B - 4 : I C P A n a l y s i s B - 5 : Q u a l i t y A s s u r a n c e / Q u a l i t y C o n t r o l f o r I C P a n d W h o l e R o c k A n a l y s e s 273 Appendix B-l: Particle Size Analysis (Humidity Cell Material) Sample: 1 Mass (g): 900 Wt. % Retained Cumulative Sieve (mm) Mass (g) on Seive W t % Retained W t % Passing 9.51 0 0 0.00 100.00 6.68 41.6 4.76 4.76 95.24 4.699 190.7 21.83 26.60 73.40 3.35 177.5 20.32 46.92 53.08 2.38 99.6 11.40 58.32 41.68 1.651 79 9.05 67.37 32.63 1.19 63.8 7.30 74.67 25.33 0.841 42.6 4.88 79.55 20.45 0.6 39.5 4.52 84.07 15.93 0.42 27.4 3.14 87.21 12.79 0.297 26.5 3.03 90.25 9.75 0.21 20.3 2.32 92.57 7.43 0.147 17.3 1.98 94.55 5.45 0.104 12.4 1.42 95.97 4.03 0.075 13 1.49 97.46 2.54 0 22.2 2.54 100.00 0.00 Total: 873.4 100.00 Particle Size Distribution 274 Appendix B-l: Particle Size Analysis (Humidity Cell Material) Sample: 2/3 Mass (g): 970 Wt. % Retained Cumulative Sieve (mm) Mass (g) on Seive W t % Retained Wt % Passing 9.45 0 0 0.00 100.00 6.68 42.9 4.51 4.51 95.49 4.699 233 24.50 29.01 70.99 3.35 199.7 21.00 50.02 49.98 2.38 118.9 12.50 62.52 37.48 1.651 89.4 9.40 71.92 28.08 1.19 62.5 6.57 78.49 21.51 0.841 42.8 4.50 83.00 17.00 0.6 34.1 3.59 86.58 13.42 0.42 26.2 2.76 89.34 10.66 0.297 22.8 2.40 91.73 8.27 0.21 16.9 1.78 93.51 6.49 0.147 15.9 1.67 95.18 4.82 0.104 9.8 1.03 96.21 3.79 0.075 12.4 1.30 97.52 2.48 0 23.6 2.48 100.00 0.00 Total: 950.9 100.00 Particle Size Distribution 275 Appendix B-l: Particle Size Analysis (Humidity Cell Material) Sample: Pit 1/2 Mass (g): 920 Sieve (mm) Mass (g) Wt. % Retained on Seive Cumulative Wt % Retained Wt % Passing 9.45 0 0 0.00 100.00 6.68 62.3 6.91 6.91 93.09 4.699 252.6 28.01 34.92 65.08 3.35 184.6 20.47 55.39 44.61 2.38 99 10.98 66.37 33.63 1.651 72.6 8.05 74.42 25.58 1.19 55.8 6.19 80.61 19.39 0.841 34.6 3.84 84.44 15.56 0.6 31.3 3.47 87.91 12.09 0.42 20.6 2.28 90.20 9.80 0.297 20.9 2.32 92.51 7.49 0.21 15.5 1.72 94.23 5.77 0.147 13.4 1.49 95.72 4.28 0.104 9.5 1.05 96.77 3.23 0.075 9.9 1.10 97.87 2.13 0 19.2 2.13 100.00 0.00 Total: 901.8 100.00 276 Appendix B-l: Particle Size Analysis (Humidity Cell Material) Sample: Dike 1/2/3 Mass (g): 975 Wt. % Retained Cumulative Sieve (mm) Mass (g) on Seive Wt % Retained Wt % Passing 9.45 0 0 0.00 100.00 6.68 37.1 3.86 3.86 96.14 4.699 254.2 26.45 30.32 69.68 3.35 206 21.44 51.75 48.25 2.38 129.5 13.48 65.23 34.77 1.651 98.4 10.24 75.47 24.53 1.19 66.2 6.89 82.36 17.64 0.841 44 4.58 86.94 13.06 0.6 32.1 3.34 90.28 9.72 0.42 23 2.39 92.67 7.33 0.297 18.8 1.96 94.63 5.37 0.21 12.7 1.32 95.95 4.05 0.147 11.7 1.22 97.17 2.83 0.104 6.1 0.63 97.80 2.20 0.075 7.3 0.76 98.56 1.44 0 13.8 1.44 100.00 0.00 Total: 960.9 100.00 Particle Size Distribution 277 Appendix B-l: Particle Size Analysis (Humidity Cell Material) Sample: Bulk 1 Mass (g): 975 Wt. % Retained Cumulative Sieve (mm) Mass (g) on Seive Wt % Retained Wt % Passing 9.45 0 0 0.00 100.00 6.68 61,3 6.40 6.40 93.60 4.699 268.4 28.04 34.45 65.55 3.35 221.6 23.15 57.60 42.40 2.38 108.7 11.36 68.96 31.04 1.651 81.6 8.53 77.48 22.52 1.19 59 6.16 83.65 16.35 0.841 36.2 3.78 87.43 12.57 0.6 29.7 3.10 90.53 9.47 0.42 20.5 2.14 92.68 7.32 0.297 17.7 1.85 94.53 5.47 0.21 12.7 1.33 95.85 4.15 0.147 10.1 1.06 96.91 3.09 0.104 6.9 0.72 97.63 2.37 0.075 6.9 0.72 98.35 1.65 0 15.8 1.65 100.00 0.00 Total: 957.1 100.00 Particle Size Distribution Wt % Passing • Wt. % Retained 278 Appendix B-l: Particle Size Analysis (Humidity Cell Material) Sample: Bu lk 3-21 Mass (g): 905 Wt. % Retained Cumulative Sieve (mm) Mass (g) on Seive Wt % Retained W t % Passing 9.45 0 0 0.00 100.00 6.68 49 5.51 5.51 94.49 4.699 283.1 31.82 37.33 62.67 3.35 198.8 22.34 59.67 40.33 2.38 97.4 10.95 70.62 29.38 1.651 65.1 7.32 77.94 22.06 1.19 45.2 5.08 83.02 16.98 0.841 30.3 3.41 86.42 13.58 0.6 23.7 2.66 89.09 10.91 0.42 18.7 2.10 91.19 8.81 0.297 17.3 1.94 93.13 6.87 0.21 13 1.46 94.59 5.41 0.147 13 1.46 96.05 3.95 0.104 7.6 0.85 96.91 3.09 0.075 9.2 1.03 97.94 2.06 0 18.3 2.06 100.00 0.00 Total: 889.7 100.00 Particle Size Distribution Wt % Passing — • — Wt. % Retained 279 Appendix B-l: Particle Size Analysis (Humidity Cell Material) Sample: Bulk 3-22 Mass (g): 865 Sieve (mm) Mass (g ) Wt . % Retained on Seive Cumulative W t % Retained Wt % Passing 9:45 0 0 0.00 100.00 6.68 51.9 6.11 6.11 93.89 4.699 237.8 27.99 34.10 65.90 3.35 199.2 23.45 57.54 42.46 2.38 89.5 10.53 68.08 31.92 1.651 64.2 7.56 75.64 24.36 1.19 47.2 5.56 81.19 18.81 0.841 30.9 3.64 84.83 15.17 0.6 27 3.18 88.01 11.99 0.42 18.2 2.14 90.15 9.85 0.297 19.7 2.32 92.47 7.53 0.21 14.2 1.67 94.14 5.86 0.147 12.6 1.48 95.62 4.38 0.104 8.7 1.02 96.65 3.35 0.075 9.8 1.15 97.80 2.20 0 18.7 2.20 100.00 0.00 Total: 849.6 100.00 Particle Size Distribution Wt % Passing — • — Wt. % Retained 280 Appendix B-l: Particle Size Analysis (Humidity Cell Material) Sample: Bulk 3-23 Mass (g ) : 1010 Sieve (mm) Mass (g) Wt. % Retained on Seive Cumulative Wt % Retained Wt % Passing 9.45 0 0 0.00 100.00 6.68 52.4 5.20 5.20 94.80 4.699 305.7 30.33 35.53 64.47 3.35 203.5 20.19 55.73 44.27 2.38 132.4 13.14 68.86 31.14 1.651 75.9 7.53 76.39 23.61 1.19 57.3 5.69 82.08 17.92 0.841 34.1 3.38 85.46 14.54 0.6 29.2 2.90 88.36 11.64 0.42 23.3 2.31 90.67 9.33 0.297 17.1 1.70 92.37 7.63 0.21 14.2 1.41 93.78 6.22 0.147 13.9 1.38 95.16 4.84 0.104 12.9 1.28 96.44 3.56 0.075 13.8 1.37 97.81 2.19 0 22.1 2.19 100.00 0.00 Total: 1007.8 100.00 Particle Size Distribution 281 Appendix B-l: Particle Size Analysis (Humidity Cell Material) Sample: Bu lk 3-24 Mass (g) : 1010 Sieve (mm) Mass (g) Wt . % Retained on Seive Cumulative Wt % Retained Wt % Passing 9.45 0 0 0.00 100.00 6.68 51.3 5.10 5.10 94.90 4.699 305.2 30.34 35.44 64.56 3.35 203.1 20.19 55.64 44.36 2.38 115.7 11.50 67.14 32.86 1.651 75.2 7.48 74.62 25.38 1.19 54.9 5.46 80.08 19.92 0.841 33.5 3.33 83.41 16.59 0.6 30.1 2.99 86.40 13.60 0.42 25.6 2.55 88.94 11.06 0.297 20.7 2.06 91.00 9.00 0.21 17.2 1.71 92.71 7.29 0.147 17.1 1.70 94.41 5.59 0.104 15.2 1.51 95.92 4.08 0.075 16.6 1.65 97.57 2.43 0 24.4 2.43 100.00 0.00 Total: 1005.8 100.00 282 Appendix B-l: Particle Size Analysis (Humidity Cell Material) Sample: Bu lk 5 Mass (g): 940 Wt. % Retained Cumulative Sieve (mm) Mass (g) on Seive Wt % Retained Wt % Passing 9.45 0 0 0.00 100.00 6.68 53.6 5.78 5.78 94.22 4.699 223.6 24.12 29.90 70.10 3.35 187.7 20.25 50.15 49.85 2.38 115.5 12.46 62.61 37.39 1.651 83.2 8.98 71.59 28.41 1.19 61.1 6.59 78.18 21.82 0.841 41.2 4.44 82.62 17.38 0.6 33.4 3.60 86.22 13.78 0.42 25.1 2.71 88.93 11.07 0.297 22.8 2.46 91.39 8.61 0.21 16.2 1.75 93.14 6.86 0.147 16.5 1.78 94.92 5.08 0.104 9.8 1.06 95.98 4.02 0.075 13.4 1.45 97.42 2.58 0 23.9 2.58 100.00 0.00 Total: 927 100.00 Particle Size Distribution Opening Size (mm) Wt % Passing — • — Wt. % Retained 283 Appendix B^l: Particle Size Analysis (Humidity Cell Material) Sample: Bu lk 8 Mass (g): 895 Wt. % Retained Cumulative Sieve (mm) Mass (g) on Seive Wt % Retained Wt % Passing 9.45 0 0 0.00 100.00 6.68 56.4 6.43 6.43 93.57 4.699 237.7 27.09 33.51 66.49 3.35 187.2 21.33 54.84 45.16 2.38 89.5 10.20 65.04 34.96 1.651 73.5 8.38 73.42 26.58 1.19 56.6 6.45 79.87 20.13 0.841 36 4.10 83.97 16.03 0.6 31.2 3.56 87.52 12.48 0.42 22.5 2.56 90.09 9.91 0.297 20.6 2.35 92.43 7.57 0.21 15.5 1.77 94.20 5.80 0.147 13.3 1.52 95.72 4.28 0.104 8.9 1.01 96.73 3.27 0.075 9.7 1.11 97.84 2.16 0 19 2.16 100.00 0.00 Total: 877.6 100.00 Particle Size Distribution Wt % Passing — • — Wt. % Retained 284 Report f o r : 8080 GLOVER ROAD, LANGLEY, B.C. V3A 4P9 PHONE (604)888-1323 • FAX (604)888-3642 Paul Dagenais, Department of Mining, University of B r i t i s h Columbia, 517 - 6350 Stores Rd., VANCOUVER, B.C. V6T 1Z4 Job 960389 August 1, 1 996 SAMPLES: 8 polished thin sections (numbered 1 through 8) with corresponding sample rejects, were submitted for petrographic examination, with special reference to AMD c h a r a c t e r i s t i c s . SUMMARY: A l l the rocks of t h i s suite are of similar general character - being andesitic volcanics composed e s s e n t i a l l y of plagioclase and c h l o r i t i z e d mafics. Epidote and sphene/leucoxene are consistent accessories. Samples 1, 2 and 3 are recognizably fragmental, and may be l i t h i c t u f f s or breccia flows. The remaining samples are homogenous, being abundantly to sparsely porphyritic rocks of extrusive aspect. A l l the samples have minor contents of sulfides - estimated as between 0.5 - 2.5%. These consist of randomly disseminated, i n d i v i d u a l grains of pyrite, 0.01 - 1.0 mm i n siz e . The rocks are only occasionally fractured, and the pyrite t y p i c a l l y shows no str u c t u r a l control. Some of the samples also contain traces of chalcopyrite, as tiny flecks i n the rock matrix, independent of the py r i t e . This mineral i s most abundant (0.5%) i n Sample 4. The sulfide s appear fresh, without recognizable limonite rimming (though the poor quality of polish i n these s l i d e s hampers detailed observation of grain boundaries). Carbonate i s absent, or present only i n traces, i n 5 of the 8 samples. The remaining three (#s 1, 3 and 6) have estimated carbonate contents of 3%, 10% and 17% respectively. The carbonate occurs as an al t e r a t i o n of plagioclase and/or mafic phenocrysts, and as diffuse replacements of the groundmass. Carbonate veining i s very rarely seen. On an individual basis, Samples 2, 4, 5, 7 and 8 are e f f e c t i v e l y devoid of natural buffering capacity, and would be net acid generators under conditions of active oxidation - although the absolute levels of sulfi d e s are at the bottom of the range generally considered SAMPLE 1: Estimated mode F e l s i t i c p l a g i o c l a s e C h l o r i t e ) A l t e r e d mafic g l a s s ) 25 62 E p idote Sphene) 3 5 Leucoxene) Carbonate P y r i t e C h a l c o p y r i t e 3 2 t r a c e T h i s rock i s a v o l c a n i c of a n d e s i t e - b a s a l t composition. An i l l - d e f i n e d fragmental t e x t u r e i s d i s c e r n a b l e i n t h i n s e c t i o n , and the rock may be a form of t u f f or a f l o w - b r e c c i a . I t c o n s i s t s of a t u r b i d groundmass of f e l s i t i c p l a g i o c l a s e , c r y p t o c r y s t a l l i n e a l t e r e d mafic m a t e r i a l ( c h l o r i t e / e p i d o t e / s p h e n e ) and f l e c k s of leucoxene. The rock co n t a i n s s m a l l p l a g i o c l a s e phenocrysts, 0.2 - 1.0 mm i n s i z e , which show v a r i e d degrees of perv a s i v e a l t e r a t i o n to epidote and carbonate (ranging from d i f f u s e f l e c k i n g to complete pseudomorphing by f i n e - g r a i n e d carbonate. Ther are a l s o some t o t a l l y a l t e r e d ( c h l o r i t i z e d ) mafic phenocrysts. O c c a s i o n a l amygdules are f i l l e d by mafic secondary m i n e r a l s . The p r i n c i p a l mode of occurrence of carbonate i n the rock i s as the a l t e r a t i o n of p l a g i o c l a s e phenocrysts, but the s e c t i o n e d area a l s o i n c l u d e s a s i n g l e h a i r l i n e v e i n l e t of carbonate, 0.1 - 0.2 mm i n t h i c k n e s s . S u l f i d e s c o n s i s t of randomly disseminated, i n d i v i d u a l subhedra of p y r i t e , 0.1 - 1.0 mm i n s i z e , and r a r e t r a c e s of c h a l c o p y r i t e as t i n y specks (10 - 50 microns i n s i z e ) i n the s i l i c a t e matrix, independent of p y r i t e . The bulk of the p y r i t e i s concentrated i n a r e s t r i c t e d patch about 1.0 x 0.7 mm i n s i z e , apparently r e p r e s e n t i a p a r t i c u l a r fragment. The l a t t e r a l s o e x h i b i t s the s t r o n g e s t carbonate a l t e r a t i o n i n the s l i d e , such t h a t the p y r i t e g r a i n s o f t e n occur i n c l o s e s p a t i a l a s s o c i a t i o n with carbonate pseudomorphs. 287 SAMPLE 2: Estimated mode Cryptocrystalline matrix Plagioclase Epidote Carbonate 70 20 7.5 trace trace 2 Zeolite Pyrite Chalcopyrite trace This sample i s another altered andesitic rock. Its fragmental character i s c l e a r l y recognizable i n thin section, which includes l i t h i c c l a s t s or xenoliths of prominently porphyritic andesite up to several mm i n s i z e . The p r i n c i p a l component i s a turbid sub-opaque matrix ( o r i g i n a l l y glassy?), probably composed dominantly of plagioclase and c r y p t o c r y s t a l l i n e secondary mafic material. This i s host to small phenocrysts of turbid plagioclase and, i n some fragments, to abundant rounded bodies of fine-grained epidote which look l i k e amygdules. Carbonate i s present only i n traces, as a couple of segments i n h a i r l i n e microfractures. Pyrite occurs as sparse, random disseminations of individual subhedra, 30 - 300 microns i n s i z e . A single 100 micron grain of chalcopyrite was seen i n an epidote amygdule. 288 SAMPLE 3: Estimated mode P l a g i o c l a s e A l t e r e d mafics Leucoxene Ep i d o t e Carbonate P y r i t e 50 33 5 1 1 0 1 T h i s rock i s composed of fragments, up to about 1 cm i n s i z e , of v a r i o u s p o r p h y r i t i c and/or amygdaloidal mafic v o l c a n i c s . T u r b i d p l a g i o c l a s e and brown/green secondary mafics are the dominant components, together with widespread d i f f u s e sub-opaque m a t e r i a l (leucoxene?). The rock may i n c l u d e a s u b s t a n t i a l p r o p o r t i o n of a l t e r e d g l a s s . Carbonate i s r e l a t i v e l y abundant. I t occurs i n v a r i e d modes i n d i f f e r e n t fragments: mainly as d i f f u s e permeations of the groundmass, but sometimes as r e c o g n i z a b l e pseudomorphs of o r i g i n a l mafic phenocrysts; as p a r t i a l p e r v a s i v e a l t e r a t i o n of p l a g i o c l a s e phenocrysts; or as f i l l i n g s of amygdules. S t r u c t u r a l l y c o n t r o l l e d carbonate (as v e i n l e t s ) i s not seen. P y r i t e occurs as sparse d i s s e m i n a t i o n s of i n d i v i d u a l euhedra, or small aggregated clumps t h e r e o f , 10 - 150 microns i n s i z e . A l o c a l i z e d area of 2 x 2 mm a t one extreme corner of the t h i n s e c t i o n ( p o s s i b l y r e p r e s e n t i n g a s p e c i f i c fragment) c o n t a i n s a much higher c o n c e n t r a t i o n of p y r i t e , as c u b i c euhedra to 300 microns i n s i z e . 289 SAMPLE 4: Estimated mode P l a g i o c l a s e 84 1 0 2 C h l o r i t e Leucoxene Ep i d o t e t r a c e 1 2.5 0.5 F e - T i oxides P y r i t e C h a l c o p y r i t e T h i s sample i s a r e l a t i v e l y f r e s h , prominently p o r p h y r i t i c a n d e s i t e , c o n t a i n i n g abundant, euhedral-subhedral p l a g i o c l a s e phenocrysts, 0.2 - 2.0 mm i n s i z e . These are e s s e n t i a l l y u n a l t e r e d , and are s e t i n a f e l s i t i c to microgranular groundmass composed mainly of p l a g i o c l a s e (with intergrown c h l o r i t e and leucoxene). The rock a l s o c o n t a i n s a few t o t a l l y a l t e r e d ( c h l o r i t i z e d ) mafic phenocrysts. P y r i t e occurs as disseminated i n d i v i d u a l anhedra, 10 - 300 microns i n s i z e , mainly a s s o c i a t e d with c h l o r i t i z e d m a f i c s . A l i t t l e of the p y r i t e i n t h i s sample a l s o occurs i n f r a c t u r e - c o n t r o l l e d mode, as a couple of s t r i n g s of semi-coalescent g r a i n s c o n s t i t u t i n g h a i r l i n e v e i n l e t s 0.1 mm i n t h i c k n e s s . C h a l c o p y r i t e i s s i g n i f i c a n t l y more abundant i n t h i s sample than i n previous ones. I t occurs as c l u s t e r s of i r r e g u l a r g r a i n s , 10 - 200 microns i n s i z e , i n the rock matrix, independent of the p y r i t e . The c h a l c o p y r i t e i s sometimes mantled by e p i d o t e . T h i s sample i s devoid of carbonate. 290 SAMPLE 5 : Estimated mode P l a g i o c l a s e 75 20 1 C h l o r i t e E p i d o t e Sphene) 3 Leucoxene) Carbonate t r a c e t r a c e 1 Z e o l i t e P y r i t e T h i s sample i s another a n d e s i t i c v o l c a n i c . I t d i f f e r s t e x t u r a l l y from previous samples i n havng a s u b - t r a c h y t i c groundmass composed of s l e n d e r , s u b - p a r a l l e l p l a g i o c l a s e l a t h s , up to 200 microns i n l e n g t h , with i n t e r s t i t i a l c h l o r i t e and f l e c k s of sub-opaque sphene/leucoxene. Sparse p l a g i o c l a s e phenocrysts are e s s e n t i a l l y f r e s h , but f o r minor e p i d o t i z a t i o n . L i k e the p revious sample (#4), t h i s rock i s a p p a r e n t l y homogenous, without evidence of fragmental c h a r a c t e r . The s e c t i o n e d p o r t i o n i s cut by a few h a i r l i n e f r a c t u r e s f i l l e d by z e o l i t e . Z e o l i t e s a l s o occur as r a r e pockets with minor intergrown carbonate (the only occurrence of the l a t t e r c o n s t i t u e n t seen i n the s e c t i o n ) . P y r i t e i s l i k e w i s e sparse. I t occurs as randomly s c a t t e r e d , small subhedra, 30 - 150 microns i n s i z e , and as m i c r o g r a n u l a r aggregates of a s i m i l a r s i z e . C h a l c o p y r i t e was not seen. 291 SAMPLE 6: Estimated mode P l a g i o c l a s e C h l o r i t e Quartz Carbonate 65 1 5 1 17 2 Sphene P y r i t e 0.5 t r a c e C h a l c o p y r i t e T h i s i s a prominently p o r p h y r i t i c a n d e s i t e c o n t a i n i n g abundant, s h a r p l y - d e f i n e d , f r e s h , well-twinned euhedral phenocrysts of p l a g i o c l a s e , 0.2 - 2.0 mm i n s i z e . M a f i c phenocrysts are of two k i n d s . One - as euhedral pseudomorphs up to 3.0 mm i n s i z e - i s made up of intergrowths of carbonate, c h l o r i t e and qua r t z . The other c o n s i s t s of c l u s t e r s of t i n y p r i s m a t i c pseudomorphs composed e n t i r e l y of c h l o r i t e . Carbonate i s notably abundant i n t h i s rock. In a d d i t i o n t o i t s occurrence i n a l t e r e d mafics, i t occurs e x t e n s i v e l y as r a t h e r evenly d i s t r i b u t e d , s m a l l granules and d i f f u s e impregnations throughout the ( f e l s i t i c p l a g i o c l a s e / c h l o r i t e / s p h e n e ) groundmass. S u l f i d e s are very minor, c o n s i s t i n g of a few i n d i v i d u a l subhedra and microgranular clumps of p y r i t e , 20 - 200 microns i n s i z e . There are a l s o one or two t i n y specks of c h a l c o p y r i t e . 292 SAMPLE 7: Estimated mode Plagioclase 55 5 25 1 0 Sericite Chlorite Epidote Carbonate Sphene) Rutile) Pyrite trace 3.5 1 .5 This is a fine-grained, sparsely porphyritic andesite. It is of similar general composition to the other rocks of the suite, being composed essentially of plagioclase and chlori te . Scattered small plagioclase phenocrysts, 0.2 - 0.5 mm in size, are l ight ly dusted with ser ic i te , and more or less strongly altered to microgranular epidote. They are set in a meshwork-textured groundmass of turbid, lath- l ike plagioclase, of grain size 0.05 - 0.2 mm (showing similar alteration to the phenocrysts), with i n t e r s t i t i a l chlorite and flecks of sphene/rutile/leucoxene. The rock is cut by occasional thin fractures delineated by chlorite and epidote. Carbonate is essentially absent - i t s presence in the thin section being confined to a single small segregation (altered mafic phenocryst?). Pyrite is minor, occurring as sparse, randomly scattered, subhedral individuals, 0.1 - 0.5 mm in size. 293 SAMPLE 8: Estimated mode P l a g i o c l a s e 60 6 25 3 2 S e r i c i t e C h l o r i t e E p i d o t e R u t i l e Sphene) 2 Leucoxene) Quartz P y r i t e t r a c e 2 T h i s i s an abundantly p o r p h y r i t i c a n d e s i t e . Randomly o r i e n t e d , stumpy, anhedral-subhedral phenocrysts of p l a g i o c l a s e , 0.2 - 1.0 mm i n s i z e , are t u r b i d , and show a l i g h t d u s t i n g of p e r v a s i v e s e r i c i t i z a t i o n and i n c i p i e n t e p i d o t i z a t i o n . Mafic phenocrysts, to 1.5 mm i n s i z e , are t o t a l l y pseudomorphed by c h l o r i t e , sometimes with a s s o c i a t e d g r a n u l a r e p i d o t e . The groundmass i s a f e l s i t i c aggregate of t u r b i d p l a g i o c l a s e and a l t e r e d mafic m a t e r i a l . Disseminated equant g r a i n s of r u t i l e are a notable accessory. The sample c o n t a i n s amygdules, 0.1 - 0.3 mm i n s i z e , i n f i l l e d by c h l o r i t e or, l e s s commonly, quartz/chalcedony. S u l f i d e s c o n s i s t of sparse, but r e l a t i v e l y l a r g e , subhedra of randomly disseminated p y r i t e , as g r a i n s 0.1 - 1.0 mm i n s i z e . T h i s rock i s homogenous and u n f r a c t u r e d . Carbonate i s absent. 294 0 <r (S c <c 0 J u 1 n; 0 - 0 c, <r ja <r -1 r_;i i—i CD "} Ul <H LD c : 0 - _3 — _J I 3 x o n • < o <r — — iii o oo o U T3 C CC Q <I 1) iU CTi ^ cn-i- ' U — C3 i n Gi ~ -Hi O O f ' T ** t3 LD T-<LD C LCD -0 al- O *> <r* - - <H t • • X J 3 0 — 2-- * IP w J w _] J G » P O - f C ft C I - u— ;TI J I T I o— w - « -C <E m - <r - --+-• - N • CO CO •+-' 0 +>o .-- < Q —«• . — i 3 N vD (« (C t> ~" " 3 0 3WS I'1' CO €> O Cu I N O I'M PI —• - M N O N ' l c <r wo o co •-• II nl (S3 «S - O w CO O fcu O Q •—• x :*: i—i O Q GI T - t rs <J3 CD Ul •>'. M G > \ D C D * - " N _f -s r- «_| yi co : I I I i i i -> Gi CT) CO Lfi >n Lfi . C^CO V 0 X en 0 0 ' 0 •H (= rd 0 c, c. ID c« B O ' >-• 1 c c 0 0 3 X 0 ~ — - o ^ <r ~ o its 0 O J i o<r U "3 c -x Q i en a;. — cn T cn o t, — S Lfi Q T 83 D O T "TLO — ufJO " -<H xi O "" • O 3 3 ~ • 5_| w |_ w +-<T C W C 3 — £, ~ Jl Jl « -1 Q —• W " W CU - c <r 3i -•<r - ~cu - 0.1 CO -t-> 0 O 1 — 0 ~ U — — (N <I IN <fj "3 * C X HSCO (DO' PL, -OCMB. CO T - N O N O - <t tnO U CO CO ZJ CZT- 1 '15 ID 03 OwCOOCuO -a—>* x —• Q O rH TH <H .jl Gi (SJ CO S Q '.H CO OM ~t* ft) <H Li"! *T —' " Q r - C CD Q M i l l fSI CTi CO LO 'XI <3* r* W CO CO 296 o <I cn C3 '-T cn Ul 0 0 ' 0 0 cn J S _x LO c (S «5 IN 0 S U s • CO t3 CO 83 0 0 i <r :l 0 - - cn cn <c C3C3 cn CO * 0 • <u">Ln i - • ifl r i—( '15 U • o — fr1 —•> J J o 0 .—. L i " ' - ' O'^-II to o <x 0'-' IN C.T3 iJJCTKE 0) +j s cn HI • £. ~ <? CD -H ' ! (D-t-1 LlTT-^  o -+-' AC 'LD ^ f—1 - CD C 0<-* • 0 Q " O S ^ o PL i I n - - w <I o - « J - • 3 T <r IS +> ^ _2 c r-t —' O w *^ - OJEfM £ N IN - ^r-pz: 3 co : .— SO 3KJ0 CL 0 0. 0 -H <I C ~- J) CO o <r.iNes -.xi Q CO CD O O+•'• 0) Cu - o <r IN tu J IN I C -<r - — CO IN DV-J itf iicors - cu <E <c U, N - i T w Q -CC 0»i<E CO CO X W E * «J n3 IDT CO J - O O O Cu I •• i x CJ Q i-i Q S rH UD U") IN Q CO C3 00 S 00 IN 01 <-H <T LO c <r co T r- co Q EJ <ri "H QC3C3 • i i i i - - 1 1 l i i i CO LO iJD >S) 01 Q 1 T t oi 297 0 0 ' 0 n 0 I <E CTi CP CD Q Ul Ul r-f J •J > ' c is T3 0 (S CN c C3 11 * <HH ts r"i CO u CO on —• - '•X*s— «•') CO - 1 H ID s U, •• 0 • 12 X t- 0 - 0 o 0 <I 0 cn -: c -I-' CT>Q IB — ffi Ul B O " -— ID - 0 c-oo<r >—1 <ri 0 CTi Q -• X Q CU I US J - Q T J 2 ( D - L I I <r PC £ -C O c 3 - P • -— <T oa i-i 0- 0 — X T* 7) c <r w o -C5<n ->-H j) ^ ti — ID Q • O CO CO •«-> CU L- CM O — •tSIMUl U <E J N - ~ J) E- • O -CO I)CU <r T CM Ii, Q r j o - IN PC 01 r mOl w - IC ITJX H CO O U - fc-X) I O Q « Q (S (S H CM T< H IN CD . coo a IN Ul T CO T- CO O | CS CS (S CD 1 i 1 1 1 - -  1    . U)iNCnpl o ro H IN 299 0 c cn IS in *H Q c - - cn XI CD w ID C ITS o --p, J p, :z a) <X III o 0 <x cn s T LO u Hi W X. i 0-0 i c £ 0 0 ' 9 Z 6 m p, 0) 0 c. tJ<C <C P. cn --• OS c <t o ' L 0 P. C ' "3 «$«H Ji X o - -**J -itl CZ (tl 0 o — +> <r —• CTi ' \ f — - D W Q LOO <D+> O T : ' uyi * i CO — «5 -" Xi —< —) t-i x ~ i cn - J O <I X 'XI - -IS P H " XI - (S CO 0 1 2 H O P . O -: EOCO Jr-: •NT - PL M r W " — " LTi • a - ' i f w « .'* <r • - o ICS * — " X € 0 CO O w o CO CO +• ' (•'•.! w — H Q ID X" <£l X r-< Pi, PH ITI * O PL — rn - « I « cniTJiN -CO z: r. - x m — • •  <I - IX ' M l r l Q ^ r - Q i ro ©cc - i IN v-i -xi- r- "T i s "sT o-~j Q H CD IS IS i I I I I I i£l CTi — -T XI XI m N IN co 300 11 (D 0 <r cr Ul cn oo —1 u ts mm CM " CO 0 o -i. i <r xi - -t"u CTi CJ» — J is cs - n T S : -• Ul LD H • - I D « OB 0 X H -C O 0 ~ - - w O ti <E O 0 ucn I 0 I' T3 - uQiJ +> © —•—"TO • - ' d-" — U 1 T ' ii +* 1) O U ' Lfi +• U T J ' - 0 L C D - ' H ; J £ 0 X " " — 3 I O J -<E <C -O 3 i—I •J 4) O X CS -m CN X O O Ji £ - I +> T CM » 3 - T C — M i <x - O — CO - CU - c <r - ID ji M'M<T ~'~--*J<S< to H — T - -- OCD COO O O • T O - -"• — CM a - CM'Xi co <sx - CO — ^ - O IN its CM co euro O — CN fc — CD T -<r— -<roo m«- <r WIN o co X 'T3 *C i— iTi H5 ITJ - U O « O O U Ul vO <r*C3 \D <D 0 0 ' 0 'Li' CD Q CS CO \T T r - T Ul (S H S H Q E l I I I I I <H ij' T Ul 'Xi T : > : T 301 <I cn & T ifi • .•••">. r H Q <I -- cn ~i is X T u X 0 0 ' 0 a TS ^ •' - , - - t i ; o CO !,<r - i o CO >—< m en cn <i »—* o o s cn L T T Gl r-l D LO IX T CO -L0 * ID TH <H ' C3 C, C < H 0 ITS - " " " —< - J -J - -X, U X X J E- o - x o G ,TJ <X — U CD i Ji -PCS) 0 ID CO - - - « T C - P CQ 0+> Lf) (0 — •+-' C " T 3 O O O'H L 0 >—i - .0 £ 0 0 Q 3 -; « Cu • o 1 nS J ~ 3 4 3 IDE <C r-t O O — T cc CM E . C C N - T X 3 - P O CO — T « UXI CQ d " OH 0 ] - ~ JI u CO o - - CM CJ X Xi Q •' CO - O O C O CD Cu .. -— - T CM O 5 'JD \ Z ~"~* nz FN •* <I CO CM —• E- • CC1 «$CM 'CO <I <E O —' T CM - ' Q • <E o ~ CM PC t} ~ co <r x PQ Zl * IT3 iTST CO - O O C J I O Q * Q Gi Q v—1 E Lfl 'XI Xi CTi Gi J CS) CO CM f CO T LO •' T CO Q T CO Q 5 i GI t-t G) ) 1 1 1 1 - - ! I 1 1 1 H If) '-.0 CO CJ <T T T 302 PH u £ PH w u aj P-H PH a P< PH — »—i ) H o s PH O o Is p -o o- - PH TO I—I U B „ PH OH 3 v j CQ PH CQ P -OH OS CQ ^ PH U IH PH 60 < CN ON *n '—f C— t— co 00 vo CN o vo o r*» ON © ^H ' CN © ' © o ON CO CO co >n co vo vd oo *o ^ 2 ^ ^ H-H ON m "3" CO 00 CO O 00 £ ~ CO VO 00 CO CO 00 >n CN CN ^ H O CN CO CN VO VO O N O O N O N C N t ^ - O O O — H - C N ^ H C N C N ^ H T - H C M >0 >n CO co so vo CN ^ . uo oo oo co r-H CN ° ^ CO vq i n r - ON CO CO CN CO CN (N VO CN CN <N VO CN CN V V V V m i n m i n wo i n IT) © V O V o V 0> o V o V O V o V o o o o o o o o COCOCOCOCNCNCNCO <40 i> "3- ON CN CO oo CN CN VO VO CN CN od od oo' ON 00 od oo CN CN CN CN ^ ^ 00 ^ H P - PH c 1-1 N PH PH PH ( J P - PH > 0 s -PH 3 U 00 PH PH p- w P . PH 3 o « l-H S1 PH S< PH 3 u ° PH o o o o o o m o o 00 vo VO t—H VO CO CO t—1 o ON o CN i n IT) CO CN ^ H o o 2 o o 2 o ° _ _ H v 00 - H y <N V c o i n r ^ ^ a N © - ^ ^ O N r - o o c N o o o o c ^ o o VO CO ON i-i CO o ° © © © © o ° 0 0 O 1 4 0 - - H 0 N C N O C 0 C O - ^ - V O O O V O m ^ i n C N C O C N ' - i C O ' - H C O C N o o o ON i n r--CN co vo vo CO ~ O O Tj- VO 00 CN VO ON O o o o o ON - H o CO o CN o o OO o o r ~ - o o r - ~ c N c o r - ( - k ' 3 -- s l - C N C N ^ C O C N m O N CO WO i . »n VO CN ON , - ; T-H' —; ' - 1 —" © ^ © i n i n i ^ i n ^ t ^ - o O T f o o o o o o o o o <n oo vo CN 1—I <n CN CO 00 IT) o CN CO < — m CO i n i n ON CO CN •<*• ON 00 O l > CO CN CO ON ON CN CN CN " P -B 00 CO CO (U ^ CN CN 5 g PH & PQ 5 / 3 oo Q C N i CO 3 CQ i n oo CQ CQ 0) * P . s CO ^ co D N N N *PH •~~ , — ' ' S S PH 3 CQ CN • C O 3 CQ i n oo CQ CQ Appendix B-5: QA/QC for ICP and Whole Rock Analyses 304 Appendix B-5: QA/QC for ICP and Whole Rock Analyses Potassium Magnesium 3 R 2 = 0.9892 / o -4 -I 2 O CN 1 -V o £ 2 1 -. . . . . . y 0 - n 0 1 2 3 K ( % ) u -c 1 1 1 2 3 Mg(%) 305 Appendix B-5: QA/QC for ICP and Whole Rock Analyses 306 Appendix C "Previous Acid-Base Accounting cn o CJ £0 CJ cn o u CJ cn o 0- CJ < C3 CJ oo o f/3 C N M | es ^ ii ,<L> O N i n O N O C N O N — O N " ' * N O O N O O m o o c > r ~ C N N O l ~ ~ © o o o © o © o o © C N NO CN r- oo C N — -^ t" m C N i—i NO as as o T f — C N — — t n s o * C N oo in' C N in in cn T! -as — _ — oo m - 2 ~ o< CN C I CN ^ ) - C N © c i o o m t ^ o o c N ^ - ' o o c N t ^ O N t ^ ^ - o o ' ^ r v o o so X ON o cn C--' 0 0 o o T l - N O N O N o m N o m T j - i n ' < t o \ c v t - - r — O O C ^ - U I N O C N v> so od as C N m' so r- ON ON N O m m v o m ' < 3 - ' ' 3 - c > m c i >—i ON •—< CN cn cn r- m cn oo t— m NO NO r~ o cn oo o X X o ON —I >—i CN so cn m m m oo r-o ON oo ~ . ^ ^ NO m C N m C N C N — C N C N 0£ CN -H "1 ON ^ . I~- ON (X i—i i n CN H i CN c> o o m o o r-' so m m i—; mi so cn — i~- n od c i CN cn cn i n <—1 c i c > c i oo NO CN CN 0 0 CN '—' NO m c i C N o >—• ON N O cn v cn NO ^ CN Tj" © c i c i cn so ON T J - m £ i —' ON O I T c i 0 \ m n t t • - H C N o o T t - m N O N O ^ c ~ - r ^ c N N o m m N o o N c i r - i t ^ m O N O N C N O O C N O —< — CN >-i CN -5t O O O *— i n <—i C N o o 9? t • O N m o m O N ^ m cn C N NO NO o o ^ ON cn C N - 5 ) ' C N ' ^ - N O r 7 r - ' — O NO 0 0 CN c t cn so . oo . . . . ,—, ON m o ON S N O r ^ t ^ o o 2 O N ' ~ ^ ON o  cn ON in cn r~ cn O N m ON OO CN O — CI f --3- f- O N cn C N C N C N i-. cn ~ O N . . oo ,„• 3- r-T cn vs Tf; CN NO NO cn CN . -3" © cn NO od ci © ' 1 2 T T cn m c i » — O N t ^ i n c i ^ H O N •—> C N o o ^ m N o r ^ r ^ -VO VO 00 T f m V O — •-i CN Cxi 00 CN CN co o cs f -VO ON 00 CN -vf ON VO >n r - ON « n ON co •<3- vo oo r o •—• t~~ CO ON VO CN «/">' ON vo 00 ON —• 00 I s I— 00 VN 00 oo t-~ r-, r- o 00 CO 00 * c n N 2 S ON ~ t~- CO ^ m N m co ON 1 —' CN r o r o O N O v o o o c N o o o r o r ^ t ^ m >—> m O N m r o - x t m o o r ^ O O O ~ O O O <—1 CO i n vo CO IT) oo 00 ON ON ON 00 ON N t 9 ^ o m 00 Ifi (N CO CO CO m 00 m — i n m -xf vo t t *o • * O N 00 1- • * t 00 .CO O o O m n ^) h (s O co CN CO co 0 1 m v o c N m c N - H t ^ o o O N O oo n f - " o n i f l ^ h o — r ^ i n c N m O ' — o o o o C f - H 0 O ^ N \ £ ) i f l \ O vo oo "3- m o U-) 1/-) vo 1^ --H co vo r o r o V CN ° in ' ° N r - t — O N m r o r - <— vo ON m t-- r o • — ' C N C N c O r O r O T f - > n i n V O O O r O m O C O C O O O O m O N ^ O i ^ n N t * O i ' i » i o o —< r o .—i CN O ' —' O co vo ON CO r -o VO VO ON i n CO ON CO CO vd CN vo O t * ON CN O ON CO O i co t 00 « CO 00 i n 00 00 00 i n v o v o u - . T i - T - ' a - ' - H TJ-ON ON O CO •<«f ON r o — ^ oo O 00 . ON VO o ON CN CN CN CN C N vo ON r o CO CO 2 m g ON' •xt CN ~ CN CN C N vo ^ i n r o >—i O N ' — CN r o r o f » m r o vo O •vf <n vo vo r-~ C N P ^ O N r - - o O ' « f r r o % l - r o r o ( ^ • - H r ^ c N T t r o r o m c N — O CN O O •—i O — O CO o ON CO oo ^ j o ON' r o 00 » h N M O N 00 \ o co V~> _ ' t—i ,-N' 00 r - m i n ^ v o c o c N o o CN CO (Nj CN >n r o oo t-~ — •xt " A v o i n r H v o i n ^ m >n O N oo r o vd O N l"- CN r o r o CN ON CN vo m — i o VO -sf rt-i-t co co oo ON vo oo i n oo M "t o t— CN r-~ VO vo , _ r i n r o ON 1 •—i CN r o r o >n r o vo 1--m vo ^o vo C3 OS Q OB 3 #© "> u PH OB OB >> e CU as u s a a 309 Appendix D D-l: Shake Flask Data D-2: Batch Leach Data Appendix D-l: Shake Flask Data Run Sample 1 2 3 p H Pit 1/2-2 6.52 4.09 3.79 Pit 1/2-4 7.38 7.66 7.13 Pit 1/2-d 7.47 7.98 6.78 Dike 1/2/3-2 3.58 3.2 3.06 Dike 1/2/3-4 6.88 7.31 7.03 Dike 1/2/3-d 6.91 7.48 7.36 Bulk 3-2 3.34 3.23 3.06 Bulk 3-4 5.06 5.46 4.91 Bulk 3-d 5.58 5.87 5.49 Bulk 5-2 5.22 3.44 3.28 Bulk 5-4 6.68 7.58 7.73 Bulk 5-d 6.82 7.6 7.79 Sulphate Concentration (mg/L) Pit 1/2-2 283 193 203 Pit 1/2-4 182 32 23 Pit 1/2-d 180 25 28 Dike 1/2/3-2 213 203 193 Dike 1/2/3-4 122 17 14 Dike 1/2/3-d 130 20 8.5 Bulk 3-2 1043 213 193 Bulk 3-4 672 128 32 Bulk 3-d 690 145 37 Bulk 5-2 393 183 173 Bulk 5-4 342 38 24 Bulk 5-d 320 55 19 •o U 1$ lu l l Cu si a E |5S Q . ro X X . o O O —' O m r o O N O o m O N m v o c N o o o X o o o o —' n M N N N •-' ro CN ro O O O o o o o o • x f v o o o m m r ^ O N - ^ r - ^ o o o o o © ©' <s o oo m " r o r o — r o r-~ ^ • x r v o c N O N r - ^ o o - x i - r ^ CN r o r o >/-> m ON ON n t t » 2 o d 2 • < d - C N C N n n C N O O O — U l r - - r o O N £ ; o o ^ r o o N m w M ' ^ m m n c o c N x f ^ - v o v o r ~ - H o cxj o oci m 1 — 1 1 — 1 co d ; C N - 0 0 ^ . - i CN O O . - H O N C N m c N ' — ' — vo 0 0 0 0 0 . 0 0 0 0 CO •— O •NT C N C N O O CN O CN CN <5> —< o o o o o r o ON 0 0 C N ON , , oo vo r- m O N ~ o o o vo vq 0 0 o >/-> vo VO o ro ro </N CN ON ro >/-> CO CN ro CN CN £5 N N N H N N tfl 06 < N * - v "5 £ £ "5 "3 PH ."2 CO "a Vs CQ CO Q ca ca •vf ON r— ©' © o o CN c o •— — r o O —' © r o c N v o c N c o v o r o r o r ^ o © o X o © o o © O O O O c s < S O O O v o o N r ^ v o v o t ^ c N T r i n o o o o o o o o o r-~ xi- xt r o c o c o r-~ ON •xi- ON r-~ **r C N r o r o ^ C N c N C N C N r o r ~ i — i « / i r - " — i r o v o o o >—| CNi i—i roi vd roi roi ro • r r c N r - ~ 0 . o o o N v o o o o o o r ^ v o ^ J r o T r v o r ^ v o ^ i l - m w o o o ^ P I T f r r o ' - , r o c N - H r M t N j c N r - ' o I ^ O N C N r o ^ — O N i n G N r O V O C N - ^ r O m O O O O O O ci o o ^ S O N ^ ^ C N - m ^ o O N t v o o o c o x t O N N t r ^ o o o o o ' X o o ' o o o o o o o o o o • ^ • o o o v o r ^ o o c N r o o o w i ' x i - v o x t r - ^ v o o o t n v o m 0 0 VO _ m vo 3 oo VD m VO r o . CO CN i n VO CO ON CN xf V O — ; O N • - i CN —; o o ~ — ' ™ © ro xt ,_ ro VO ON ro ON VO m ON VO vo VO CO CN CN co P ro ro ro CN CN CN CN CN ro CN CO CN CN £5 N N i 55 -H N-H ^ CN S H aj ' a E £ co o »H CN CN CN i/^ CO fO 3 S ca co 00 =1 =1 "a "a PQ CQ O O O - ^ O ^ c i O O c o c N > / N r o v O r o r o r ^ O © O O © O © O CO •—i >—; i -| (Nl CN] N , <S O O o o o o m r ^ v o o o r ^ r o r o m — ' O O O O O O O O O N M rn ^ N rn ON .q. _ CN CN CO CN r o m r o ^ T C N i — •—i - H ' l - ^ n N N CN O N O N ^ „ V O — . i n m ^ o ON CN ON -xt r o ON O ro ro 0 0 0 0 CN ON ON ro CN CN VO m VO ON CN 0 0 CO CN — i m ro o c> O O ro ON ON CN CO O CN VO CO m O o o o o O o o o CN no VO VO NO <n VO xt <n ro ON ON vo o ON CN o r~ — oo — CO CO 0 0 Tj" CN CN CN CN £5 — <N rxi fs — ^ ^ iio oo c o ^ i ^ ^ c o r n ^ ^ SH •£ CQ s a CQ CQ Q CQ ca OS Q w -S3 PH Q •3 e eu a < 312 Appendix E "Sulphur Mass Balance" Appendix E : Sulphur Mass Balance Sample 1 Product Mass(g) Assay Units (%S) ( g o f S ) Initial Head 1218 4.44 54.08 B L Residue 1034 4.45 46.06 Leach Solutions 5.13 Lost Material 184 4.44 8.17 A c i d Added 3.99 Calculated Head 1218 4.55 55.37 Kinetic Head 1034 4.45 46.06 Washed B L R 129.4 3.75 4.85 Rinse Solutions 0.02 Kinetic Residue 893 5.27 47.06 Weekly Leachate 0.13 Calculated Head 1034 5.04 52.06 Sample 2/3 Product Mass(g) Assay Units (%S) ( g o f S ) Initial Head 1247 3.55 44.27 B L Residue 1109 2.47 27.44 Leach Solutions 4.35 Lost Material 138 3.55 4.90 A c i d Added 3.89 Calculated Head 1247 2.63 32.80 Kinetic Head 1109 2.47 27.44 Washed B L R 136.8 2.28 3.12 Rinse Solutions 0.01 Kinetic Residue 966 2.7 26.08 Weekly Leachate 0.08 Calculated Head 1109 2.64 29.29 Pit 1/2 Product Mass(g) Assay (%S) Units ( g o f S ) Initial Head 1210 3.98 48.16 B L Residue 1061 2.74 29.05 Leach Solutions 6.16 Lost Material 149 3.98 5.93 A c i d Added 4.82 Calculated Head 1210 3.00 36.32 Kinetic Head 1061 2.74 29.05 Washed B L R 136 2.72 3.70 Rinse Solutions 0.01 Kinetic Residue 915 2.84 25.99 Weekly Leachate 0.25 Calculated Head 1061 2.82 29.94 314 Appendix E: Sulphur Mass Balance Dike 1/2/3 Product Mass(g) Assay Units (%S) ( g o f S ) Initial Head 1205 2.05 24.70 B L Residue 1123 2.34 26.27 Leach Solutions 4.64 Lost Material 82 2.05 1.68 A c i d Added 4.74 Calculated Head 1205 2.31 27.85 Kinetic Head 1123 2.34 26.27 Washed B L R 146.9 2.43 3.57 Rinse Solutions 0.01 Kinetic Residue 970 2.15 20.86 Weekly Leachate 1.30 Calculated Head 1123 2.29 25.74 Bulkl Product Mass(g) Assay (%S) Units feofS) Initial Head 1231 1.75 21.54 B L Residue 1124 1.23 13.77 Leach Solutions 7.81 Lost Material 107 1.75 1.87 A c i d Added 4.74 Calculated Head 1231 1.52 18.71 Kinetic Head 1124 1.23 13.77 Washed B L R 146.6 1.19 1.74 Rinse Solutions 0.02 Kinetic Residue 970 1.22 11.83 Weekly Leachate 0.46 Calculated Head 1124 1.25 14.06 Bulk 3-21 Product Mass(g) Assay Units (%S) ( g o f S ) Initial Head 1205 5.51 66.40 B L Residue 1031 2.87 29.59 Leach Solutions 7.35 Lost Material 174 5.51 9.59 A c i d Added 4.70 Calculated Head 1205 3.47 41.83 Kinetic Head 1031 2.87 29.59 Washed B L R 122 3.23 3.94 Rinse Solutions 0.02 Kinetic Residue 900 2.52 22.68 Weekly Leachate 0.67 Calculated Head 1031 2.65 27.31 315 Appendix £: Sulphur Mass Balance Bulk 3-22 Product Mass(g) Assay Units (%S) ( g o f S ) Initial Head 1204 5.51 66.34 B L Residue 1020 4.35 44.37 Leach Solutions 9.15 Lost Material 184 5.51 10.14 A c i d Added 4.78 Calculated Head 1204 4.89 58.88 Kinetic Head 1020 4.35 44.37 Washed B L R 147.9 4.53 6.70 Rinse Solutions 0.04 Kinetic Residue 861 4.35 37.45 Weekly Leachate 0.43 Calculated Head 1020 4.38 44.63 Product Bulk 3-23 Mass(g) Assay (%S) Units ( g o f S ) Kinetic Head 1012 5.51 55.76 Washed B L R Rinse Solutions Kinetic Residue 1004 2.98 29.92 Weekly Leachate 2.98 Calculated Head 1012 3.25 32.90 Product Bulk 3-24 Mass(g) Assay (%S) Units ( g o f S ) Kinetic Head 1010 5.51 55.65 Washed B L R Rinse Solutions Kinetic Residue 1001 4.31 43.14 Weekly Leachate 4.50 Calculated Head 1010 4.72 47.64 Appendix £: Sulphur Mass Balance Bulk 5 Product Mass(g) Assay (%S) Units ( g o f S ) Initial Head 1272 1.86 23.66 B L Residue 1096 1.51 16.51 Leach Solutions 5.27 Lost Material 176 1.86 3.27 A c i d Added 4.51 Calculated Head 1272 1.61 20.53 Kinetic Head 1096 1.51 16.51 Washed B L R 151.6 1.56 2.36 Rinse Solutions 0.01 Kinetic Residue 936 1.48 13.85 Weekly Leachate 0.07 Calculated Head 1096 1.49 16.30 Bulk 8 Product Mass(g) Assay (%S) Units ( g o f S ) Initial Head 1210 2.66 32.19 B L Residue 1053 1.81 19.08 Leach Solutions 7.23 Lost Material 157 2.66 4.18 A c i d Added 4.53 Calculated Head 1210 2.15 25.96 Kinetic Head 1053 1.81 19.08 Washed B L R 151.9 1.8 2.73 Rinse Solutions 0.03 Kinetic Residue 890 1.84 16.38 Weekly Leachate 0.28 Calculated Head 1053 1.84 19.42 Appendix F "Back Titration Curves (Standard and Modified Methods)" 319 320 324 Appendix G G-l: % Calcium versus Neutralization Potential (Original Material) G-2: Post-Batch Leach Material G-3: Post-Kinetic Test Material 325 Appendix G-l: Calcium vs. NP (Original Material) Carbonate Method Average 326 Appendix G-2: Calcium vs. NP (Post-Batch Leach Material) 327 Appendix G-3: Calcium vs. NP (Post-Kinetic Material) 328 Appendix H H-l: Kinetic Test Data H-2: Calcium: Magnesium and Sodium Ratios H-3: Calcium, Magnesium, Sodium, Potassium'.Sulphate Ratios < 3 E •is is £ IX a 1-2 «) J 0) o I —1 > l~ o C CO a> 0 . • ° " 1 6 ? f s 1 1 «5 , to o as ca O) g-11 = o o> ca O <u 0) .5? * . ® >N > CO p* "D 1 a> 87 CO "a co Jl E 5 ® 3 0) CO CO CN 00 CO CO CO CD CD CO 00 CO CN W UO N lfi CO 0 0 0 0 0 0 0 00 CO T - CO M- -sr CO CO lfi ci ci C D ( £ ) ' ^ C O ' ^ C O C O C O C O < D f 0 m cn cn co CM LO co co •<-oo oo oo co cn T M- M- ^ cn od oo co in o co LO r— co co oo CO T CD co cn co CO CO M" T co co uo cn o o LO 00 LO CO o c5 ° M-ci CO CM CM CJ) LO cn cn LO cn LO -sr T - T ~ O CN CO cn cn i n c O C M N O O N N l f i S N i n c N c d L o ' - ^ c b a i c d c d c d o c s i C N N 0 ) O ( D C N . U ) O O n S C D C D C D C D C O C D C O L O L O C D C D LO LO C D t N C O m ( D C N i n r - N ^ C O ^ C O ^ O C O C O T - S co co CO M" O N S 2 LO - oo co T - ^ r h - X c o c o c o c o C N T - T - ^ T - T - T - T -CN 00 co CN o TT 00 00 , -cn LO oo LO d 00 00 M-cn co I-- LO oo iri o o V ^ ^ ' co LO i— co cn (O CO CO N S CM ^ (D CO CO T - LO co cn LO CM CO <M CM O T - ^ T -i— i— i— i— h-LO CO CD CO CM 00 cn N -C N n ^ l O l O N O O O ) cn CO CO s _£Z CO (0 4-* o z CM M- CM M- LO cn N -N If) ro N J - CO CO N CM t- LO 00 CD M" 00 d ci ci ci ci d ci T - CN CM CM CM CO LO 00 T- CO d d LO CO CD O •<- LO h-i Ni -St co - LO LO oo co co CO CO CO CO CO CM CJ) N CO LO cn co oo CD T - CJ) LO M" CO T~ 00 CN OO LO O CJ) CJ) lfi t co cn •q- oo LO a> co oo CO N - x - oo co h-' d co CM oo cn cn o cn M" M" T J " I O T ] -oo O ) S f IO d d co CN cn cn LO N CN oci CJ) 00 T-oo CN £2 C O C D o ^ ^ L O C D ^ O O o o o o g ^ r - ^ - o c n L O 0 0 ^ 1 - h ~ C D ^ . A , C M C M ' ^ - A 3 - - CM « j ^j. ^H. r-~ CO CO o o o o o oo M- : i ^ ^ ^ cn cn • ^ r c o c o c N c n c n c o c o c o c o c d ^ d c o c o c o L o c r i c x i L O l O n s ^ N C O i f i m i f l C O S CO CO C O C O C O C D C O C D C D LO LO CO CO co M-CD CN CO CJ) d CM CN t-co CO CD CN LO T - h~ o co co r - N LO 00 CD CN co co co co co 00 T - T -lO (D ^ o cn co OO N N CO l f i cn co -3- co OO CO CD ° ^ T ~ S 0) m (D CO if) CO K (D ^ CO CO N N l f i CN CO O) O) O) N N . CO N N -r— I N N CM W N M l ) CD O 00 CD CD O H O (O M O N ID (D CO CO N N CN M" CO 00 ^ CO If) CO CD ^ L O C O ' d L O ^ C N ^ C D C N C O O j C M C M O ' - s ' - s N- N- N-cn i - C N C O ^ f m c D S 0 0 O ) ^ CO CN . J ca o o >N I l-o CD c l-o D_ Q . CO es CQ 0 u 1 3 w •3 s a* a. a < E Q. CO cu 330 •a u . $ CM < Jo E o — 3i |g O l C CO 0) CD c . ° — " J o ? f S I?? (o "9 Is* (fl u (0 (0 CO g-1= "5 j CD CD O a> — <» I O) •« J'S >, 5 (0 -o •o £2, 1 cu I >, CD co 5 E § CD I CD O CD 13 i o c N c o o ^ ^ ! ^ £ j S - x r • < - , - - 1 c o c n m m i s - ^ l > ! . c o cn CD N C O C N ^ C O C O C O C N O) CN s ^ s o) i n q d i ^ ^ c o ^ t c o ' c o ^ r i*- en m cn T - T -CO N T m cn cn cn I S -•xt 00 Is-m oo cri •xt o oo m cn oo I S - oo m <xt oo I s-i f l CD (D -•"3- •*• ^ CN cn -vt m co cn •xt -xt -xt m CD I S -•xt m i n oo cn o q m i n c N ^ . N . c N i i ~ - - ' x t CN CD m o o o • I S -m co cn cn cn CM •xt ^ £ 5 00 CO CO cn m cri co o co oo CD i r i co 2 co CN S 2 o oo CD •xt cn oo 5 £ CD CO d i ^ oo oo cn cn cn m cn °> r - °> CM co " I S -•xt cn oo o i n m co i n 05 oo CD m x-T o cn CD M CO N ^ i n oo cn oo ^ c d ^ ^ S C N i N r ^ C O T - o i i - T - i o m ^ T - T - o c o o m C D C O C O C O C O C O C D C D C D C O m m -xt CD CO CO "xt CD CM CO co I S-' I S -CD CM m O CD 00 I s-I S -CM x— T -in i oo co CM CD CD CD CO 00 m CO CD CD CO ^  CM CO !5 ~ - £ cn ix-I S -oo OO CO CO CN CM N N o co _ oo oo cn co m co CN m I— I S -00 I S - •xt "xt I s -•xt -xt CN i n CT) I S - CO I S - 'xt CO I s - I s- CD I s - I S -oo •xt m cn cn d o CD CM i r i co co I S - co cn co C D I S - OO CO I S -cn T -CN I S - CM _ - -c o m c o c M c o - x t c o m CM CN CD _ 00 S N C D N N C O m N C O C D C D m co CM oo co cri 0O S CM - - x t CM •xt CO CD CO i n co co T - l f l C O O i m ^ C M ^ C D C M C O C O C M C M O r - t x - T - t x - T - i - C n s N N N I s - I s - I s - r -( M c o x r m c f l K o o o ) ° ^ co CM CO m CO CN co $2 •xt CO CM oo cn CM V— I S - T- O T - O CN _ o d d CO oo - x t -K_ CM cn r - i n L : m CM o - x t u < m c n v . o c o O O 0 CN o o •xt "xt O T -1s- N - x t T - CO oo co •<- cn CO CO CO CO I S - CO I S - T- CN CN m I S - i n oo i n cn -xt oo I S -r- CO CM N CO ob co i n I s - oo CM co m CD oo T3-•xt -xt •xt -xt cn •xt oo m cn oo CM "xt •xt ob cb m T - I s-•xt "xt "xt co m I s -•xt •xt I S -i r i I S -cn co co 0 0 ^ cn cn s a « « 8 § cn o o cn cn ° > T I S -I s - O I s -cn CD CM 00 cn •xt CO cb •xt CM in cn CO CO cn O o O oo oo o •xt •<- CM d IS-; CO CM ^ cn ^ ^ cn co ^ co co cb cri co o cn cn oo d cn oo •xt cb cn cn d i n CN o co oo ^ m -xt CM m cn oo CO CD cb cb o o o •<-CD "xt •xt i n i n oo cn oo N C O C N C M N K C O i - C O C O ( D O t D f - C O O T - C O m C O C O i n c D c o c o c D C D c o m i n c D i n m cb co co CN co m co CM i n m * m m o CD CO CO "xt •xt CO CM I"-•xt I S -o •xt m oo co T - CM N— N— CD CO CD CO Is-CN oo I S - CO cb ob oo I s -oo 0 1 o o T- I s -CO CM cn CM cn cn -xt ob cb cri oo •xt i n co i n co co oo oo m I S - r— co -xt CM m CM oo - I S - CD cn C N C D U 5 c b d ^ r O T : c o i r i s c M ^ t s ^ m m o ' f ^ s o N K C D N S C O C D K C O C D N 00 CO 00 o I s -d co cri co co co co i — I s - CD m co cri cn o _ CM co oo co x t se oo COCN "xt CO 00 00 m CO CD ^ i n c b d i n . ? C N ^ c b c N o b C X ) C M C M O T - | X _ T - | X -rx- |x- |x- IX- cn I S - fx- I S -c M c O ' x t m c o r x - o o c n o co •xt CD CO <D I o . i EI < IS E i s E m J 0) o I-I > i - 8 3 o l 0) 15 -g (0 ' 1*1 CO CO « o J (0 CO [ I r r CD i 5 I i= -5 1 0 CO IO a) .2> ? |a> > co =» -o •o JO a> a) >> — co •a co 5 03 1 ® ii co "^ r 00 CN o S N N CD X CN CN CO M-CD 00 CN ^ N <? M" CD 00 CD T- T-00 CO CD CD M" T -CD CO CO CN O "3" CD CD CD N r--Lo" S 9 o CN CN O CO CN CD LO CN C i O C N ^ I - I - T N - C D C D - S J - C N C O T -• ^ ^ C O C D C O L O C O C D C O C O C D N ^ - ^ c N - ^ r c o ' c o c o C O N co IO O) t CO N T- CO CN LO LO CD O T - O S U) N T -sr T oo LO a) co CN CO o CO CN CD CD T - h - CD LO M- M- •<*• ^ CO I— CD <•> £ § co o g CD S£ I " CO t ^ (N N (D CM ° « § CD CD CD CO N CO CO LO CD CN CN CD co CD CN O CD co CN o 00 l~-CD CO CD CD CD O ^ CD CD ^ ^ 00 t-N LO h- T -CD CD oo CN O CD CO CD CO 2 N o o o O CD  CD CO CO CN CD CN CO O CO CD c\i LO CD CO ^ CD CO ^ ^ O ) CO N- CO CN CD N CO CD C N I v - - < - C D ' s - L O C O C D - r t -C O O C O ^ t O C O N m L O C D C D CD CO CO IO 00 00 CN -r-' o CD CD LO - r -LO cb co CO M" T -CO CN CO _ -sr CO CD c i CD CN LO -I- N-O CO CO T - N LO •>- 00 CO CN CO CO CD CO 00 CD CN CD CD CD CD CO C\i LO CO CO CO N -i— r~-CD 00 co i-» • • • CN 00 LO CD CD LO CO — ^ ^ CD CO CO CO E— -t— LO CD T - 00 CO •<- 1- N-CO LO £ CO d O C O cd o' CO C O ^ m C O ^ i n m C O C M O ) ' -S S ^ N N ( 0 " N C O ( O N 0 0 CO CO CO CO T- ^ — • S T L O C D C D T - C D N - ' I - L O d N N-' co CD LO ^ ( D ' - C M C O C O T - C O S C O ( O N N f f l i n N N I O l D 00 C N M - C O O O C O ^ L O C O C D L o ' c o ' d L o ' ? c \ i ^ c o c N c r i C M C M O i - s r ^ i - i - O ) r— i— N N h~ h~NN-C N C O T L O C O h - C O C D CD LO < CO LO N CD <0 00 CO & 2 LO CN CO CO CD CO T— T— LO LO CO t 1- co 00 X — CD o CO CO CD 00 CN d c\i CN •<-co CM CM CO CM CN 00 LO f— O CO LO CN o o o o o CD CO CN CD I*-N o O CN O CD C D N - C D C D M - t - C N C D C D h - C O T - C M C M C O t C O N C O C M ^ O m o i ' T c o N ^ ^ c o m r o W C M ( D l O C M i - ^ N O S CD O LO 3^" T 3^" K CO O LO CO CD M- M- -^ r CN I"-- LO M" M-CO CN co r*~ •<d- LO M- -<r _ . LO T . _ P r - r s i C * t ; m ' - n c o r ; ^- °> ^ CO • - n co ri CM O ° co CD LO CN o LO CD co o °? CN ^ CO CN CD S G oo o CD CD _ CN CO •>- LO co i — CD LO •<-CD CD CN CD CN LO CO CD CD CD ^ CO co co i - N C D l D C S C M ' - I O C M •<-' ^ ^1 N.' CN CD CO CO CD y 0 ) T - N m C M O C M ( O C M S ( D L O C D C D C D C D C D L O C D L O LO LO CD CO co T - co CN co CD co CM CN CD CN LO o co oo LO T - 00 CO CM CD CD CD CO 00 CO O ) N CO N- CD C D L O O O C N c o L O C N ^ O C D N O l D C D f f l C O ^ c o N ^ T— I IS- N - N CD CO N S C B T - C O C O C O C M C O C D C O LO CD CO CD ^ CD CM •<- CO CD f ^ s t ' - M - O t M O l D ' -S N ( O N M D ( D S C 0 ( O N C O - ^ L O C N C D C N C O C O T -LO O CN CD LO 00 CN 00 7n r O T r r o c M c o r o c D i - c o S N N I D M O f f l l O N N " CD oo CD CN -5T CD CO ^— LO CO CD LO T— CO CM CM O |— I— N I— h- h~ CO LO CO CD C N C D CN CO IS- CD N N N f O f i n C D S C O O ) ^ CO CO Q. < o >^  o CO E C L < - r i CD C3 a U •3 S s a >> cu •3 S «u a a. < c o in i_ •3 O I CD o I'D 0 Q-L-L_ 0 —^' c _>> O-Q_ CO co in o 332 I? o . < col < E o u 1^ 091 ,O E O 0) •—• I I ca — O -i > 1=11 cn >> | •§ jo ? «< £ • D CO (0 o co co CO g-1= o ] 0) CO .2> ? © >x > CO = fr CO i m >x CO ~U co CO Q. » E § 0 3 I <D O l i 00 CN CO oo Is- -xt CN CN I s -"xt •xt O cn m in ™ <» § 5 Si oo CO "xt m co m oo co cn CM T-C N C N C N h - T - i n - r - _ . _ . c O N -I D O T - C \ I O C O T - S S ! C O O ) CN CN o o o o o o o o i - CO -xt CD T -N T - CO CO If) I s - T t CM CO m cb co I s-cb CO cb x— CO I s - o Is-' "xt •xt oo I s-v - cn oo m cn q co o' in cb cri CD m m CD CD \ t t ^ N t •xt N" CD in •xt "xt co in cn co cn in in IS-' cri in m •xt •xt in m •xt -xt C O c D ^ C O ^ ^ ^ ^ C N c r i ^ t r - ^ o ^ g c r i (D S N O ^ (D CM O) ° O) ^ ^ O) O) cn T- cn cn •xt cri cd cn oo •xt d oo oo Is-' in co cb cb XI U) (\| M w 1^ m K N c o c n c o g g c o c g S n t o O) O l O) cn oo o o I s- CD cri o cn cn CM CO CO ^ ^ cn o cn cn cn CD CM cn CD cn i - o o o •xt £^ oo CT) oo CD cri . in co Jo CT) CD S m oo 2 cn oo c D C D t - ' x t C N c o m c D - x t r - i n i ^ c o c b c b i ^ i r i d c b o b x — o b c n c n c o o - x t o c N - < - " > - c n c o i n m C D C D C D C O C D C D C D i n c D m T -m -xt cb cb co "xt CD CM CO cn d •xt co CM •<-oo d •xt CD CN m T -O CD 00 T -m 00 CD CD CD CD CO I s-Is-CM oo 00 CO CM CD ob CTi cb in co IS- oo I s- CD I s- I s-CN O CD CD OO CO CO I s- I s- I s - CD CD , -•xt CD 00 •xt r-J CO x— "xt CO K K co m co in m m o -xt co CM •xt CN co co co in I s- I s - CO O I s - CO . O CD O l I s- I s - CO CO I s - I s -I s- ^ k oo co cp T— I s -CM T -Is-; Is-; CO •>-I s - r— l s - 0 O r s - - x t C O 0 O C D 0 O C T ) CN m m CO CM CN CD Is- Is-T- OO O 1- T -•xt CO CO oo co m r— Is- co co CM "xt CO 00 ^- in cb cri iri CO m co CD CN T— CD CN 00 " J i x . r s - | x . r x , ' |x- I X _ I ^ I ^ C M C O x t m c O N C O O ) O T" o CM I s - ^ CL < 00 oo i - CO •xt r-; CM CN CO •xt m cri T co cn •xt m co CM CD OO 00 CO N— •>- CO co cn § g cri ix: oo T - T - T - ^ C O i n C N C M C O o c M CN CN "Xt T - CT) in CM m in Is-I s -o o o o o o o o C O N C M C D i n C M i n N C O O C O i n m c O C O C D C D C D N C O C O O ) [ ^ ^ • x t c N ' - x t c o ' c o ' c b c b h - ' c b m CT) CO o CN Is-•xt •xt OO I s -m o •xt cb cb CN I s - 00 00 •xt -xt •xt •xt oo m cn co I s - o ~ cri cb cn co _. . . o 00 OO 00 •xt 'xt I s - CO CO I s -•xt -xt CD O) CM 2 CD N N O) co -xt in -xt cn o cn o o 58 cn cn cn oo -xt •xt cri cb O x - o in cri ^ ^ cn co cn cn r-•xt •xt cn cri u-j cb LT> rxi in rt- m u ' r -cn cn o CT) o cn co cb I s -Is- CM cn cn CO "xt CM r-; CTi cri CM ro o cn cn ~ cn ^ CM i r i CM O CD "xt co in cn oo CM cb I s-o « cb CD CO °> 5 c M i n m i n o o o o o o h - o o c o cn x t x t m cn N O CM CO CM m T -•xt •xt cn co co in -xt •xt o in C D D D C D C D C O C O C D C D C D C D m in cb cb d co -xt T -T - x - CM CD CM CO m CD CM m rt\ m i—> CD 00 CO I s -I s -x— OO CD CN CO CD CO 00 o co ^ g £ § U Is- Is-CD CM CO CO O) CO I s -00 CO CO CO 00 m CD oo oo T - CO T - 'Xt 00 Is- Is-l s - 0 O i n m C T ) l s - C O C T ) C N C M C M i r i - x t o b c b d c b c b - ^ c b C T i c b i n i n r - m m c o c M O C M o o c M S N C O N N C D C D C O C O C D K CO CD CO 00 CM t-oo ^ c \ i ^ ^ d c Q N - x - ^ r c o c M c o c o c M m J n t O ' - N l s - l s - C O I s - l s - C O , - l s - 0 O C O CM CM CO •xt m T - ' CM CM 'xt CD 00 CO in CO CD ^ i n ' c b c r i i n ^ c M ^ c b c N ' o b ( O C M N O i - [ x ^ s fX- |N- [X- |X-CT> I s - I s- I s -C N c o - x t m c D N - o o c n I s -CM oo ^ CL < es o u s S W 1c cu cu .2 •5 s cu C . Q. < to CD O IZ 333 |T3 J o '. |2 E| < 3 E l-o " Jo E lo o — co — | 3 £ I - I O - | IS ? 1 ? l o 7 ^ co co ^ _ (/> u CO CO LI © O) g-I s 1 = o 1 0 CO o © — CD J'S >* 5 (0 " O •O co ' © (0 -a «0 i fl) >* I S -o CO O) Q. 3 E 5 0 0 IO •x 0 I s- CD CN LO LO LO CO CO 00 •sr oo T - CD CD CN T CN CD LO LO O f— o CO CD CD LO !£ CD N - T - T - L O C O C D C O C O rocotnsoiDN O O i - O N I i ) t N f d d d T - d o o T -00 LO LO o d d CD CN CN LO CD T - 00 CN I s- O I s- I s- I s-N S ^ C O ' ^ t CO LO CD CO co co co LO CD -^ r oo h -00 ^ CO CN CD CD CO N CO I s- LO ^ CD I s-X CD CO CD 00 LO CD C O C O LO C N CO I s- CO CD o CD CN I s-CD CD CD CO O T CD LO O CD LO m . <9 to ^ O J N ^ O C O (D ^ S I O XT CD CD ^ II ^ CD CN Is- I s- CD CD O CD 1033 CO CO CD 1035 1025 CD C N C O LO LO LO CN CN LO fv-i m r%i CD o TJ- m CO o CD _ ^ CD O CN LO T -CD CD CD , _ CO O CD —^ CN CN CO I s-O O C N O O C D C N CD CO ^ ^ CD I s-CO T -O CD CD • < r O ) r O N N i r ) ' 5 f ^ c D C O C O C D L O CD O C O C O C O N M C D ' - ( D C M ^ ( O C 0 C O ^ L O C O C D C O C O C D C D L O C O LO LO CD CO co T - C O C N C O m C D C N L O rr\ v*- rri LU r—\ r r\ cr\ CD -sr d CN CN T -10 CO r -LO - i ^ ob cd CD CD CD CO I s-CN CO CN T -cci Is* I s-I— I s-CN CD I s- C D O O C O T - C O o i m c o c o c n c n c N S C D ^ C D O O C N O C O — CD I s- CD CO OO 00 I s-I s-I s-C N C D T - L O C N C D C O C N O O O O d c d c N L O - i - c o ^ r c o - i - o ^ C N L O C D - s t , s t l s - ' < - L O ' < - | s - I ^ N N C O N N C D C D N C O C D ^ CD O I s-00 LO CO CO L O T co co 8 CO I s-s 8 C N C O C N C O d Is— CD I s- T — CD T— CD CD O 00 LO O CD I s-N ^ (D ID CO IO CO CD ^ L O c o d L O ? c \ i ^ c o c N o b C O N C M O T - S I - K ' - I - O ) ^ I s- I s- I s- I s- I s- I s-T - ( N C O ^ i r ) ( D N C O O ) ^ CD CO TO l _ 0 o C L D (/) 1_ CO 0 c S Ri CN CN CD CN LO 00 CD CD CD 00 CD •>- LO CO LO CN LO I s-co CD LO CD OO CO d c o c o J S c D C N i T - r o f i i n T - O T - ^ C N L O C N I - ^ O C N C D C N o d CN O d d N d ^ CO C N O O C O C D ^ r - ' - C D ^ J - C N C O ^ r ^ l - ' ' -C O ^ O i - N N C O O O C O C O C O S N u i n ' ^ r i c d c d c d N ^ l f i O) T CO N 00 lfi CD CO O O L O C N C O O o C D C O C N C O - ^ -c o ' c N i ^ ^ c d t i r i c r i N c d c d • ^ • i n c D m c D ' ^ c D c M i o ^ a ) •^ r ^ T -^ r -sr ^ -3-CD ^ C O c 9 C D C O . C , : ! T - ' < - , ~ C ^ , ; , : C O d S L o ^ g L o ' c b g g g o c i a ) ° a ) ° ° c o a i ° r " c T ) LO CD o> T — C N C O S 2 2 co CO ^ & CN O) g§ g CN 00 ^ I s-LO O CD CD I s- LO CD CD CO CO CN LO co K O CO CD CN Is-CD CD CD CD CD CO ~ CN CO LO O CO LO v-' CO CO CO CD 00 ^ r ( n C N c o ^ r o o T - c o T - _ c N L O ' c D s D d o ' o d o d c D o O T - " C O S ) N O ) ^ r - 0 0 C O C N i r ) C O L O C O L O C D C D C O L O C D CD LO LO CD CO CO T— C D C N C O CD d CN •^ r co C D C N L O O C D 00 LO 00 C D CO CD CO CO I s-I s-c\i CO " o C D ^ C D C D L O T - I s- -CD CD od ^ I s- CN O CD CD CO 00 I s- CD CD CD CD CN T - -^ r " I s- I s-L O C O L O O O T - C D C O C N L O - q -O O K O l D N i n O D i f i Q l f l f f i CO CN CD I s- I s- CD CN CO CD CO O) CD O I s- ^ - LO I s- I s- CD I s-00 I s- CO LO CD CN I s-CO CD CO CO CO LO T - I s-CD CO CO CD " " CO CO CO CD Lfi I s- I s-I s-oo C D co C N C D 00 T - LO CO CD LO 00 C N C N O T -^ r-- I s- I s- I s-C O L O CO CD CN ^ CD CN C O T ~ 1^. 1- T ~ CD I s- I s- I s- I s-C M C O ^ - l f i l O N C O C B 2 ^ CO cn 0 iin I 3 Jo E o J CD O I-I > x: o 1— CO Ol >. " -g co 12 (0 >, D fr i l l l^1 co co co o (0 CO O) g-01 cz CD = ° 1 0) CO cn > | > CO T3 52. 0) fr TD I >x CO •o co J"S |5 col J a> o co ro m T «P § M- I s- • m cn CD CO g <D n m nfl w Ix I s - o CD CO cri I s -i r i co i n od o o i n c o c o i n c o c o c o f r i C N i n o o o o o r x - - x r c o - x r c D r y c o c o 0 0 0 0 0 0 0 - ^ ^ 0 0 C l N t m 0 1 C 0 C D i n C O m N m r s - ^ ^ c o ' - x f c b c b c b c o r s - ' ^ ' in cn cn r-•xt x-co I s-•xr <xr •xt CO s CM CO S cri od cb m i n co •xT -xt -xt •xf CO co if) cn co O | X T " O) CO O) O O) w [x. CD CD m CO • •xt ' ' -xt •xt 'xt 00 cb CD m oo cb cb •xt "xt o o O) ¥ O x=- x=- ° ) •xt fc m r; i f ) oi S CM CD O SS CD o •«-co ob •xT CD CD i n CO CD •xT O h ^ CD K fc o ^ § O o S£ x- O) CD -xt I s-K •xt CD I s -o o h - CO cb CO x-CD O) i n cb cn •xt d CM CO CM ~ •xt Is-' cb Lo S CD CD CN CO S oo o <JJ 0~> O) o~> 00 CO 00 N " T — X — o in o ^ CD CD LO CM CM 00 •xt I s- co co I s-x — m c n o - x r c N c o o c n i n c o O T - c D O - x t x - ' x t c o c N c n ' x r C O C D C O C D C D C O C D C D C O i n c O i n T -i n -xt cb cb CO -xt CO CN CO "xf CO x— x- CM x-00 d •xt CO CM O " m T - r--CD 00 i - I s-i r i x—• cb cb csi CO CO CD CO 00 00 CD N; •<-I s- I s-I s - I s-- C D x - C O N - C N i n C N K d d d c o C O C D N S 0 ) 1 0 l f ) N f O O C M T - \ T C O f s - l s - C O C O 0 0 C 0 l s - l s -CO CD I s- CO CO CD CD m d Is- cri -xt •xt CO CD Is- I s- CO CM i n Is- I s- CD •xt CN OO x -I s - CD CD I s- I s- ;±: _ O) i n co 00 CM CO T - if) O) O) I s - i n CM co co o CM CO CO O T- N " _ _ _ l s - l s - C D I s - l s - C O i n i s - l s - C O C D I s - cn -xr co cn CD LO CO CD CD 00 C N - x t C D O O C O ^ i n c O C O i n c r i c x i L r i ^ c N ^ c b c M o b C N C M O x — |x_x— |x-x— x-CD i— i— I s- I s- Is- I s-T - C M C O \ f i f ) ( 0 S C O O ) oo CO x- CO 00 I s-™ £ i n CM CM I s - CD CO po ro CD I s-s ^ -I s-CD cb CN CO CO CM cn x— i n i n 00 cn o o" IS- "Xt T -O I s - CD x— CD x— o d d oo i n i n CM CM 00 -xT m I s-O CM d d C D x - T - C D 0 O C n 0 0 C O 0 0 x -C M ^ f O T - 0 ) C O O ) N C M x f i ^ i ^ i n c b - x r c o ' c b c o ' c b h -m O) m I s-d •xt oo r -CM CD CO cb cri cb i n i n CD o i n•xt -xt -xt x^ -•xt oo m o) co m oo m co a) ob o -xt . .I s- m co •xt -xt -xt •xt CO •xt 'xt •xt "xT Is-CO cn O) co x- CD •xf i r i •xf x— •xt •xf <xf o O) o o T — T — T — CO I s-CD x— CD cn CM O -xt (xi CD x— O CM 9 ^ 2 cn =° £ co O CO x*-^ S cn cn cn co oo co Is-; cb CO CD cn oo •xt co cri cri N - ^ I s- £ 2 o 0 5 5 - «>. X * CO N T -o oo cn cn O) oo •xT I s-CM CO co CM 00 OO CO CD CD CD 00 00 CM & i r i ob g g co oo CD OO x— CM CM T - S CD CD if) xt CO I s-I x x t x t c x i c o V w c d c M O O ) c o O T h - o o c o x — r o x - c o o o c N i n i n c o m c o c o c o c D C D i n c D i n x - C O C N C O r n C O C N i n * m m w n i n i n -xr cb cb CO -xt x— x— CM x— •xt CO CM r-; • t I s-_ . i n o co oo Is-I s-00 CD CM CO CD CD CO 00 i n •xT 00 I s-I s- h- •xt co co cn I s-ob cb 1^ i ^ cri ob cb cn i n m I s- CM o CM CD I s- Is- CO CO OO OO O) CO CD x-x- CO Is- I s-o o m - x T x - x - c o c D - x r o o c o o o ' c o c b - x r i ^ x - ^ i r i d i r i c b d COCMCOCDx— i n o c o o ) K K C D C D N C D C D N S CO CD N " I s-00 Cxi I s-O) CM CD CO CO OO N x t S C M C O C M l O C O C D c o o o o ) c o i s - ' x r i s - - x f o o C D C D C D C D i n S N C D C D CM 'xt CO CO 00 ^ i r i cb O ) i r i x— CM 00 CN CM O x— |x- x— |x- |x- |x- |x- fx-m co co ^ cb CM ob h- x- x— cn fx. |X_ r-x - C M C O x T i f J C D S C O O ) ^ i n CN ^ CO 3 o x: s £ , r o I "a x 0) c c co 1^ o CD o 'c i— CD > O 0 o c o o CD to c CO CD < D)| < « E 1 C CO . ° E o — J 0 •—• J a> o I-J > I - "I 0 ® O 1 •§ |5 co io "9 l< £ T J CO (0 o ! to co IE 5 cn g-5 g 1= o 1 0 CO lO © , — CD |o» 5 1 0 >. T3 C O 3 it) TD <D >. CO "O C O oi a 0 E g 0 0 LO CN CN co o LO IS-' LO CD LO LO 00 CD CD CO cri LO •t— CN •ST LO -51- CD CN — cn co co oo o cri CO ^ S T - M T- cn CN T- 1 0 T- o T- co 0 0 0 0 „, CN CN LO CO S i - N T N S ^ ™ « CN ° S O O T - o ^ CN CD 00 ^ - cn i f N C O M D C M f O ' -T ^ O C O C D C O C N L O C N s s ^ c o m c o c o n c o s ' f LO cn oo IS-CJ) CO CD CN CO I • LO' CD co O CD LO CD co co IS-oo LO cn N CO Ci T -CD CD CD •«r CO •vt' LO LO co cn oo co O oo IS-CN cri cn CN od IS-LO t-cn cn °° 9 LO CM cn o g 2 2 CO LO CD CD co If) m CO O) CO o CN cn IS-cn oi CD CN en cn cn cn oo IS- cn LO IS-o o 5 -CJ) CD 00 LO T -C D C N C N o cn T -CJ) CD CD — IS- C O L O CD CD — C D L O Q IS-O ) 00 CD co 2 o S LO CD CO CD CD 00 L O 0 0 C D f S - C N ' ^ - ' < - C D T _ C N S f f l C O S N ^ f D N O o S O O O N O O C D O C O C N C O C O C O L O C D C D L O C O C D C D C O C D LO - r -LO •sr CO CO co CD CN CO cn f co CD CN LO 1- IS-o co oo i - IS-LO OO CD CN CN •<- T - CD CD CD CO CO CN CD CO "ST LO O) LO CO C O r n C O C N L O C N C D C O - ^ S CM O I s -I s -O) If) L O CD IS- IS-co co IS-CD CD 00 CO I s -L O IS-O) CN CD T- CO CD CO CO IS-IS- IS- CD T - cn I s- CO CN CD ^ CO if) N d T- IS-O O t O C D O l C O i -S N C O C D N N C D N CO IS-If) CO 00 CO CO i - 00 I s-LO ^" LO CD CO CO CO CD CN 00 CD CD CD CO LO CO CD LO CD CO CO CD C N ^ C D O O C O ^ L O C O C D ^ L O C O ' C D L O ' ^ C N ^ c b C N O O C O C M C M O ' r - s , " N , " , r C I ) ^ r ^ r - ^ N N I s - I s- i — I s -C N C O ' S f L O C D N O O C D co ® D CD 5 ^ ^ oi f2 LO 00 00 & ™. <? CN LO CN _•. I N U J ~ CN C O C N C D L O C O C N CO LO CD CN 00 C O O O C D ^ r h - C O - ^ - C D 0 0 ) O N ( 0 ( D O ) N T - O T - N T - C N I - O d o d d o o d - r -M OJ rr. m ^ O -<-^ d o oo CO r - . ^ r c o c o c o L o ^ r r N l c o ^ r M - K T - C O S C O N 1 > T - C M CO N S ^ C O ^ C O C O C O IS- •^ r cn cn O CN CD oo IS-CN LO CD CN CD LO . LO CO CD 00 LO CD CO O T -CD IS- 1-CD LO LO CO LO ^ -NT CD o> o £ o T - cn oo IS-cri IS-LO •<-O) CD CO LO cri LO oo o O T -CN CD o ™ 2 CD CO CD CD CD CD 92 co CD CD ~ IS-C S) CN 00 D  CO CO cri LO •sr CD o o CD 00 I s- o CD CD od cn CD o cn cn \ r O CD IS- T -C O C D C O o CD CD cn CD CD oo r^: CO CD CD cri r - CO o o IS-00 oo C O I S - C D C N L O - > r C D 0 O L O C N C D o d v f N U ) C N L d o ) ^ c d c d c d O O T - N C D C N C D C N C D C N O C O L O C D C D L O C D L O C D C O C D C D L O T - C D C N C O m C D C N L O T - h -LO CD cd co -^ r CD -vT CO O CN I s -T- M- N CM T- T -O CD CO 2 LO CO CD CN Z CD CO CD CO CO m I s-I s-CO IS- IS-O) CD CD Ln CD CD IS-00 CO oo LO LO COggcD^ oo LO IS- IS-CD CN CO d CO IS- IS-C O C O T - ^ C O C O N N st CO CO CO If) ^ CD CD .. CD O CN 00 CD O O C O C D C O C O O O O I— N (D ifl S N CO I— CN 00 LO CN CO CD CO CD CD IS-O CM T - CO N CN CO IS- CD 00 CN IS- CD CO LO CN m in oo CD CD LO IS- IS- CD CD CM t (O (O CO U) CO CD I I i r i c o d i r i ^ c N ^ d c N o p fflCM(MOi-sfs IS- IS- IS- IS-T - CD I s- I s- I s-C M C O v r i O C D N C O C J ) ^ oo 0 c D 93 OS Q U is 9J •5 s a a. < w 0 3 3 6 o . < CW = El < I S O o l I a) O I —i > i — 8 I S * CO >. i f co co 1 to o J co co IE £ O CO g-1= o j CD CO — 0) I CO 5 J'S >» 5 co "O •a co , £ >, 5 cT co -a co • a> >x co "D CO CO Q. » E § CD I <1> O 1 CT> 00 O -xt •xt -xt CO oo r--^ CO LO CO CN g co v f ° O c b ^ o . ^ « CN ^ LO LO LO CO co LO CN cci OO i - LO LO CD N-00 N- h - x - CO LO O O O I s- T - CO o o o o c i o 00 LO o o CO CN O C O N O i m c O T - C D T - l O C O O ) C N C N O O O r - ~ C O C O C O x - x - T -i s ^ i s ^ ' x t c o ' - x t c b c o ' c o o b i s - ' , x t LO CD o \ r oo s LO CN <J> •xt •xt "xf CD LO •xf , _ "xt ^ 00 9 LO co 2 co -xt ^ -xt -xt 00 LO CD CO x— CD CD CT) LO 00 "xf "xt ( P N ( 0 . ( P C O C O « ^ ro o _^ o N - r N -2 cn 2 oo 8 § ° § ° 2 cn cn 2 ^ 00 CD °°. CO § LO 2 CD T- oo cn ^ N \ f T- CD •xt CD CD CD Ix 00 <P. CM CD d ro fN g) CT) 00 CO CD $N CO o S O ^ CD CD CO O cn S £ £ « ° S & § m * oo cn cn oo oo ro ro ro <£> oo CD LO cb o o o CD o oo CO co oo CD 00 c o c D L O t x - c o r o r o o o o o o o r - -O n r r - N O C l M D r T -o o r o o o o c N O c o o ' x t - T - c o L O L O C D C O C D C O C D C O C O C O C D LO LO cb co •>-CDCNC0 m CDCMLO m m n-* r-icDCO CO N" N" CO CM Is-' § M s ^ T - CM T - N— I s -I s -T- CO CD CM CO CD CO 00 LO I s-I s-00 T - CM xt LO LO O CD CT) LO I s- CD I s-CD co co ro T -•xt CM CT) l b I s-r-- co ro CM i -CO CO I s - OO I s-ro LO I s -CM 2 CO OO CN CN oo T— CM cn co ro CO |x- o co o -xT CD T— I s - CD CO I s-LO CO CM O -xf CN T - cn o CO CO I s-T- 00 co CO 00 CN £ ft! CO CM LO CO o ro co ° oo CD I s- CD I s- CD ^ <°. ^ O S s I s- T - |X-LO I s - T— co LO LO oo CD CD CN -xt CD 00 co T - i n co ro LO T - CN 00 CM CM O •<- rx_ i -^ |x, r- I s - I s - I s -LO CO cb CM CD _ . oo rx 1- T - ro i x |x_ |x-( M c o x t m c D s o o o ) S"«r-LO LO 0 •""" c 3 (0 CD O Z tn CO m CO _ ^ ^ L O L O L O C M - 2 S ^ N ^ ^ CM CM x r C N C O O O C O N x f l O O C D C O T f ^ O C M C O T - o o i S - T - - x r c M o o d d d d d d d d CM r - I s -LO LO o •>-d d o o c M r o c D - x r o o c N o o o o r x - C N C M C O I s - O O O L O I s - l s - ' < - C M C M N N x f c o ' ^ C O ' n c < ) C O S x t LO ro -xt oo Is-co ro LO co •xt CN cb 00 CM CO* CO CM I s - LO -xt •xt 'xt 'xt -xf -xt CO •xt •xt •xt oo LO ro co oo LO I s- o ro co h-" eb cb 00 LO CD 00 •xt "xt "xt "xt LO •xt CO CD "xt o ro ro "xt CN CD CT) CO CO o o LO 58 CD ro CN 00 ro cb i^ CO o O T-co CM LO ro CM CO CD T-' CM I s-ro ro ro Is-* ro 92 co cn ro ro •xt ro 2 •xt I s -I s -o ro ro co Is-* CM oo d CM LO ro oo ro co ro CD oo ro ro ro ro oo oo •xt I s -I s- Q g o CM Is-•xt -xt d cb •xt oo ro oo ro • x t i o - < - o o r o r o c M C D C N i s -o ' m L O C O C O ' o d d N C N o o c n o c o c o o o r o i ^ c - x t rois-oorocMoo'xt-<-cocMco L O L O C D L O C D L O C O C D C D C D C O LO T - CO CN CO LO -xt cb cb CO •xt •xt CO CM 1^ CO CM LO O CD 00 I s-I s -LO t- CO CO CN •<- CM T- T -•xth-^LCDCDCDCOOO 00 CM LO •xt cb LO' CD ro |x-o, ^ CO CM CO •xt LO \P I s - CD I s - I s -!(— CD CD I s -co ro •xt CM ob ^ i - LO I s - I s -CM LO •xt d cb ob o •xt cn I s- |x_ oo I s - CO CD I s -I s - •xt CM Is- I s -cb cb ob LO LO co ^ - LO ro o o ^ CO I s- I s - I s- I s -CO CD CD *xt CN 00 LO CO CM T-O OO CO o o I s- CD CD Is- I s-oo «N w •<- |x- T -cb « ? : CN CM 00 LO I s- CD LO CO CO CM 'xt CD CO CO T - LO co ro LO T - CN 00 CM CN O T - r -W |x- |x- I s - I s- I s -LO CO CO cb CM ob ro I s - r -T-cNco'xtLocoi s-coro o T -CM CM CO CD •^ c 3 (0 CD O IZ i4 —' c i CN CN _ CN CO —' c n c n CN c n C-- f - CN •-1 -1 CN CN CN ON CN 0> o V O o V o o o i n <n o V o o V o V O © O O O ' 1 ' 1 © V o o V d V d V d d d d d d d CN CN c n O ~1 CN CN c n CN CN m d d o O c i O o c i d o o d O CS d d d d d d c i d CN CN c i CN c n o o V o V o V o V O V o O o 0> o V o V o V o V o V o o o O o © V © o O •<t rj- oo r - r - CN CN 00 d d d d d i—i d o d c i c i d © - t M m t CN CN C l CN C l C l OS CN , _ „ c > < n c i t - - — - ^ S ^ c o O O ' ^ ^ C N C ^ S C N O C N c3 00 O s d C l ,_; CN C l Tt u-1 fvl <N . r n ^ o o o ^ » c j m - H <N • • OS SO m • • c n c n so CN CN ^ ^ - C l t c i i _ ^ - C N C i ; _ J _ ' — c i 00 ^ C N t ^ ^ . v O c i V) d rt c s n • 2 ^ ^> ~ c n ^ - ^ - C l ' - ^ ^ C l C N C l O O ^ - ' C l ^5 "es „ ^ ^ ^ C O - n ^ ^ C -• - • ' ' ~ <N © Q ^ ^ -*»» es o o c i m r— <~> ^ ° ° ^ O SO OO H H (fl N t \£> ^ Q \ © © © CN ^ ^ d \ N S O as oo r - -vt r=l NO C l IT) M » O l ^ OO ^ CN CN C l d d d N rt n C l C l m i n p - c i CN m C i ON CN M0 " rt d i s c i c N M o ^ 2 2 2 t ~ - t 2 — , r ' —< ^~! CN U l 00 CN O o m <~~ d ^ d d « m t s r H ' - ^ -eu es cu CU ii CL, e N i < i cS d d c i ^~ ^ i d o o V V . i—i — CN CN . CN ° d d d d ° d — H . I s N \ D H r f ^ . - x f d ^ N n n ^ « ^ ^ - - • ^ - ^ ^ H - ^ C I I — ' r-* I-H o i i N o d r i d c i d . M N - . CN . — — ° ° o o © ^ ^ o ° o © ~ . ~ ^ CN f - CN ^ w . ~ CN ^ ^ ^ C N - C N ^ ^ ^ ^ ^ C ~^ ~^ ~», ~ , -rr CN| c i cii •rf; r~_ M d c i o o ^ d d d d c i ' - i -3 c cu Q. a V V c i oo i n o CN • m c i_! o V 3 £ . . . m . - N T so x t • v, r i ex o o o • o : : rsi c i C D : . o o . t~- •—• oo r~ ? ? ? " 7 - CN' <N ^ 9 . c i c i i n CN • c i ? ? ? ? «^ ; - oi * « o o - c - i i n l —' ^ c i O o u ' - i s ^ r ' " ^ r 2 : N ' LZ OO M „ O f - v O O O O o S S o O " " i - i i c i n n n ^ L " ' - ' i> (U CN . oo cd Cl CN CN 5^ D 3 s CQ 5 CN CN CN CN m oo 3 3 3 3 eg pa pq oa ca co ^ ^ ^ C N C N C N C N ^ o o C l ^ ^ H ^ ^ 1 C l^ C l^ C l <y ^ Q ca ca CQ oa C*") —1 r-i in "d-c i ^ ^ 4 ^ c i c n c n c n S ^ C Q 3 3 3 3 C Q C Q Q CQ CQ PQ CQ c T CN 338 T3 U o o V V N M N m O © O © O V - - . ^ c M x i r o ^ r o O N 0 0 — 0 0 ea ca o' Q> — © ' © ' © ' © ' ©' © x , x , CN x , r o CM Q Q 6 O 6 © r-- x , vo x^ x , CN — -xt x , ON vo x , o i Q © c i c i O © O o o o o u S3 ro ^ co ro © ' © " > / - ; < - < CM co m — 5-J o V i ^ t ON f x QO CN CN CN CN ON 00 0 0 ^ vo CM r o NO ^ ON — VO S d d 6 " «' S »' 5j N d rl — — CM fx ' CM CN CM - ^ . — I I — 4 f O V O T l - l - ~ © o > > < N O O VO CM ON CO CN o . CN 00 VO — - -H « 3 - N - oo' o o — m .2 "as — - n r o — < C N c o . . c N r - ~ "J O O ' - N C N ' m S S d M ^ ] CO N b oo t m oo w-i Ji. y m m CN — r o O N O N O O v o J j g — ^ 4 . li-> f x T j - N O ON i n O N 2 0 0 vo vo r -ON CN O W l CN r o m O N r x O v o r - N O ° ° P ] o o o o W N - n O V V O O O N ^ g - i M S3 . -H . CN o VO _ ° © 9 © «n ro' • fc O CM ON CM O _ CU .£2 &4 N N N M - I CN1 C N C N 1*0 x i x i x i o o i d d o © x i <x> V V . ro »40 ro ro vo o O V V — v — . . oo «/•> r o >/-> ><o --• " J ? ^ o ' m d ' * d K o \ ^ w -5' e o. c 3 E ^ 1 m t r_N Q, © © © O ^ •sf CN . C O CN f-' ° d .CO . N" O O O x- O _J V V V ^ V 0 1 a N . r . <N "H v °' . ^ . ON . w-i CM r o ^ ? . — © © O O _• _j CN O _• CN -O O V O r - . i r N ^ 2 2 v o O 0 0 0 0 § § 0 0 o o o o g 2 g 2 o c N ^ o o m M ° J ^ ^ o o i j - o o N r -£ CB t / 3 r o CN CM — r o fc fc CM *2 <D 3 PH P3 Q CN CN CM CN i i i i co r o r o r o ^ . Y co oa pa m u~> oo 3 3 oa pa r o CM CN -H co fc; fc ^ .*s u 3 CN CH ^ pa r o r o r o ^ ^ 3 3 3 oa pa ca i/-> 00 3 3 pa pa CM CM CN CM w> oo r o fc fc^co r o r o r o y y cM.ti < u 3 ^ ^ = £ = £ 3 3 P ^ ^ C B 3 3 5 spapa p pa oa oa pa 4> cd a . < 339 -a O ' • ' • <—CN '—<—| >— I s VO ^  ^ ^ 0 0 0 0 0 0 0 ' ° d ~~, CN ON >—; -xt; CN c-; — v q ci d o o o d d d d X , — 00 — CO — ON liO — ci ci o o o o o o o o — x , x , m — oo m x . r o d o c i d o d d d d d o vo * 0 P - r o cx t c n r n c n vo M m (N r - oo fc CO ^ d ~ , • ^ ~ , ^ " <*> O N S - N » ^ 00 ^ ^ 2 d o — t o N o\ c s co o r o ^ 2 2: - co - -— CN i«o vo t ON i n °°. CN vo d d d o o M ' o d i c i ^ ^ d m . CN CN • CN 00 00 d f^ •xt d o •xt •xf • * ' no CN « s = < cu vo O r f O rO CN "3" ON ON vo ) £ ^ J o v n i n vo O m CN ON CO 00 00 VO ON CN CN CN CN 00 t— t"- CO ° "0" ON co oo x: O O i n r o M t ^ O O v ' 2 § C N r x CU •xt 1 — 1 ^ —i — CN vo t - m CN — i—i r o , _ CN r -o d o — CN o — oo •xt d d CU cu PH N o o V V CN "=t — •xf — — CO fc — • CN — co — m — r o c N c o r o — CN d d d m d d d d — d d o o o o o o o o v v v v v v v v o o V V oo m d -xf O V r o oo oo ov •xf d cK fc . . v o r ~ v o ' x t o o r - - , x t -° ° d co o '^f d vd W •3 s cu a, a. < V V 3 £ . . . vo . vo vo r -r j O, O O O . O Cl. V V V V CO O CN co o o — V V O o o o V V V V •xt d VO — —' d — V CN d ' ' . ' . ON ' • *0 ^t ON ^ ' ' '—1 •xt O C/3 o o W> CN ^  £ o o o o o o — ON _,. " CN Tt — fc o o vo t - O ^, • -xt CN fc O •xt O O O "3" CN VO o o o o o ON r - — _•. fx — CN — ~ o o 0 r-- — o fc oo CN " r- r- r» O ON CO O CN CN ^ S ' - : c N C h 2 0 2 ~ 6 cd C/3 CD CD CN CN - CN CN <N CN ^ ^ r o f c f c ^ r o r o r o r o y y CN « <u3^^^:^33 £ ^ c a 3 3 3 3 c a a a d ' ' CN ex < ( N C N - ^ ^ ^ ^ ^ 0 0 c o f c f c ^ r o r o r o r o _ ^ _ y S ^ m 3 3 3 3 o q c Q Q 03 03 03 03 fc' CN r o oo - c C N C N - ^ ^ ^ ^ i n O O r o fc fc _Y r o r o r o r o y y S ^ f f l 5 3 3 3 0 a C Q O CQ CQ 03 CQ 340 -a . . . . . <N . m • CN O O O O O ~ O —: O _ ; V V V V V ° V 0 0 V ° i4 d d d d d d <=> co d d d ,—i ,—i s o r o t~ •—i CN d d d d d d d d d d d o o i o o 6 d d d d d o O SO „ •-; CN r o OS 2 r o CN I—1 r o r o <—i m OS CN cs O s O s c N - oo m m so | Z d d >—'^ 2 '—" d r o xf <—< i—< o s o s o s . *— oo Os x t r o so © © d 22 ^ >-< © •xT •xt ~ ~ oo oo TH ( x j t ^ - x l - x j - o o r o m d © . . - * ' — < © .-H' d x t ^ . - . co oo . - ' C N - x t - ' x ! - - x t - > n r ^ - - < 0 0 . --HOO CN os 2 d d d i d ^ oo n » d t-i o d d so <N - i _ ^ ^' 2 s ° -H I N C I C A I O H H ^ Q N N * © © ' © -xf d od r o ^ co o •xf 93 CO SO CN -xT © SO s o © C J _ „ rt i © * * N i n so os os CN m 00 00 CO x f CO -xi-i n '—1 CN C O CN <—i —' — CO CN -xT CN m CN — r o t> oo O oo CU " ' 83 _ ^ - os 2 <U CU 4) c . . s o t ^ s o - x r o o t ^ x i -N ° ° d co d x f d sd r-' ° . r o so m r o r— oo CN Os V V O r o O -xf O m so C N S O — . - ' r o C N C N O s C N r O , , ^ d d d d d d d •-< d d i w „ ^ „ , _ _ _^ ;g o o o o o d d d d d d S V V V V V V V V V V V cU a a o o V V Tt- CN (O ^ x r ^ l . O CN O r o O m y 3 £ . . . <3\ . v~i Ct\ ^ . " - i n CX O O O O -xt! r< ' ^ © r-^ ' m cn s o o o o o _ • _j o o o O O O r o O CN • • o O V V V V O C N ^ y O S ^ r x n ? S m <-> 00 «xj rxl ro ~ " 00 £ O 5 00 — O C M C N C O ^ ^ c N ' ^ S t - ' - C N 0O M O l J X „ S O C N U S C N C N - ^ S C N ' ^ O O ^ ' - ^ r o CN s o IT so £ Os ^ r o c N ^ r o ^ ^ 2 2 r o ^ 2 ;  i n r-~ CN 00 ^ r o CN CN n « M CN * ; a> 3 P* ril PQ Q r o r o r o r o ^ ^ ^ M 3 3 3 CQ CQ P3 3 PQ m oo 3 3 CQ oa r o CN CN n - i ^ CN .ti i i 3 s a a s CN CN CN CN i i i i r o r o r o r o _y ^ , ^ m oo 3 3 P H ^ C Q 3 3 3 3 C Q 0 a Q pa pa oa ca r o CN CN CN CN i CN i CN i m 00 J * * r o r o r o r o 3 3 3 CQ 3 3 3 3 ca CQ ca oa pa CQ CO 2 >> ca 341 -a CJ —* >-H CN s o — c o - — r ^ T f T — m ^ . r t ^ v o - ^ N N i / i ^ — Nf rtr-rt^-THn — i n ^ f T H r i © o' © o' o © © ' © ' © ' © ' © o © o o o o © © © o o o o © o d d o o o © © o U CS 6 - * d d oi —i oi 00 — CN Ti- f- ^ <"i ^ CN 2 d © © ^ © 2 ^ 2 ^ d c ; M C f t i N * O 0 O O ' < t 0 O n h O O © © <—i ^ d © © ci T ' d © — c N x t r - - i > - — * O s 0 0 r N ' - H m d o d od ©' so id od <o o CN O O O s O O . O O C N S O O s O s ©' d d H © © CN ci — H - H C l O N v O C l C l C l O ' ^ d . o d d id d id od ci o CN CN IS e < cu 93 cj , „ ^ c so - £ £ . so i n c i *-H c i s o m 5- N \ n m ^ O C N C ^ ^ ^ ^ - — — C l i— C l — ^ . •—' 93 4* -CU CM c . . S O x t S O r~ 0 0 so — 0 0 N ° ° d © -vf ©' so r-' © •—' . i n c i oo i n oo m . Os ° ° © CN d • © ci so ° ©' . . C I S O C N — t— r ^ s o - C N ° ° © C N © ci © ci vd ° W • MM) •a B <u a. a r j a. © d d •* © CN © P _: Q. V V V v < N O r n - ^ V ° © © © _ • © V V V ^ V c i oo • CN © © Os \j . CN . m c i t~ m . i ? ? ? ? © <*' V o O ou c, CN m Si iS C/5 g 1 CN OS 1 . — c i c N - * C J . r - - 2 0 O s d r ^ v o 0 0 0 0 0 0 ^ 3 — P S ! ^ O s — m c N ^ c N § s o [ ^ ; 2 ! ^ • ~ ~ ^ - , ^ o. CO C/J C l CN CN . - i c i i i iS CN * J <U 3 OH r i CQ P — (N m CN CN CN CN i i i i C l C l C l C l =_• s-1 3 3 3 PQ PQ J d 3 03 m oo M J d 3 3 03 PQ PQ c i CN ?5 — i c ^5 J ^ M CN . t i <o 3 CM -ri PQ p CN — (M m CN CN CN I I I I C l C l C C l ^ 3 3 3 J d 3 PQ 03 PQ PQ m oo J d J d 3 3 03 PQ C l CN CN C l C l C l C l ^ CN t i OJ 3 ^ ^ ^ CM J d CO 3 3 3 03 PQ PQ 03 3 3 CQ CQ J*i O e 3 2 g 3 CD C 3 342 T 3 CJ r o v . i 0 0 0 0 0 0 0 . ~ ; , - ; 0 , _ : V V V V V V V ° ° V ° d d d d d d d d d d d C J o CJ c O S CN r o m V ON NO' V~i 00 r o 00 CO r o CN CS 00 C_J <—i 2 00 H^" ^ t— r~- r o i—i •-< O CN r o > - H C S r o m c ^ r o r - - N O . —* vi d d d MD d o< u-i t~ d CN >> It s «U 93 5 ^ 2 2 § S S ; S ^ C A S — CN . ^ . CN NO -N- ^ ^ O oo O • . n vi o CU 93 CU CU CU D — ' U ^ T f i - i . — • < n C N C N c N ' - ' ' — ' o d d CN d d o d CN d d CU o o o o o o o o o o o ^ v v v v v v v v v v v C • . oo m a oo - H c . — N ° ° d CN d r o d r o vS ° d ffl •3 s cu a. a, 3 G - c o - O N r o r o c N ° & V V v v ° ' °° V 1*1 - - £ C/3 g n ' » i i ' i ' H - ' i : , o S E; Norow-i U a ) n - H r n 5 « 2 ; O N ° ^ - N O _0J "5. CN .ti CJ CN i CN i CN i CN i Vi 00 r o r o r o r o j * i 3 3 3 03 3 3 3 3 CQ 03 ca CQ CQ CQ CD CN CN — c 3 343 f j ' - i - n n n o i o i n £ 2 ± i ^ 3 3 3 3 3 3 3 ^ ( N j Q - D O l C D C Q D a m C Q C Q oijey JB|O|/\| B|/\|:BO 344 to CM 4 ~ 4) r - CM CO <• CM CM CM CM co co ro ro in oo - • - c M a Q m m m m m m m L I 1 I I I I L I I ^ T r T t T T l T T T o o CM o 00 oijey jB|OiA| eiyj.eo 345 346 e - - C M o • * CM CM CM CM CM f j i - i - n < o o n i o o o ^ c M Q - O o a m m m r r i c D m 1 1 1 (- o n CM - - . P o o o o oiiey JB|oi/\| frOS-6|/\l 347 348 T - (1) ro co co co ID oo m ^ ^ • » - ( N Q - Q C Q C Q Q Q D Q D Q O Q C Q 349 

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.831.1-0081146/manifest

Comment

Related Items