Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A study of pyrite reactivity and the chemical stability of cemented paste backfill 1998

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

Item Metadata

Download

Media
ubc_1999-0007.pdf
ubc_1999-0007.pdf [ 36.5MB ]
Metadata
JSON: 1.0081119.json
JSON-LD: 1.0081119+ld.json
RDF/XML (Pretty): 1.0081119.xml
RDF/JSON: 1.0081119+rdf.json
Turtle: 1.0081119+rdf-turtle.txt
N-Triples: 1.0081119+rdf-ntriples.txt
Citation
1.0081119.ris

Full Text

A STUDY OF PYRITE REACTIVITY AND THE CHEMICAL STABILITY OF CEMENTED PASTE BACKFILL by V A L E R I E J. B E R T R A N D B . S c , University of Ottawa, 1991 A THESIS S U B M I T T E D I N P A R T I A L F U L F U L L M E N T OF 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 OF 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 S T U D I E S (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 B R I T I S H C O L U M B I A December 1998 ©Vale r i e J. Bertrand, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, 1 agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. The University of British Columbia Vancouver, Canada Department Date - ~ / £ ~~ DE-6 (2/88) A B S T R A C T A two-fold study was carried out to 1) characterize the evolution of the reactivity of pyrite in the early cycles of kinetic test leaching, using cyclic voltamperometry, and 2) document the weathering characteristics of various paste backfill mixtures that contain pyritic tailings, when exposed to leaching environments similar to those encountered in mine settings. Pyrite leaching experiments were carried out on 6 different pyrite samples from existing mines. Cycl ic voltamperometry was performed on carbon paste electrodes (CPE) containing fine grained pyrite samples on the unleached samples and after leaching periods of 4, 10 and 20 weeks. Pyrite reactivity profiles, supported by scanning electron microscope (SEM) observations and leachate chemistry data showed that minor phases of sphalerite and galena present in the pyrite samples were the most important parameters affecting pyrite reactivity in the initial leaching cycles. Sphalerite and galena were found to effectively retard the oxidation of pyrite in the early leaching cycles. A s sphalerite and galena were leached out, an increase in the reactivity of pyrite was observed, followed by a gradual loss of reactivity from precipitate coatings. A t a fundamental level, mineral surface characterization by cyclic voltamperometry was found useful in the interpretation of kinetic test data for the prediction of acid rock drainage ( A R D ) generation. For the backfill weathering study, paste backfill samples of 4 different mines were leached in deionized water (pH 5.5) in flooded and alternating air-flooded environments i i and in a simulated A R D solution (Fe=500 mg/1, S04=1.5 g/1 and p H 2.5) for 20 weeks. S E M , solid phase chemistry, paste p H , acid-base accounting measurements and leachate chemistry were also used to document the weathering characteristics of cemented paste backfill (CPB) . This study revealed that hydrated portland cement minerals are p H sensitive and highly soluble. Short-term exposures of portland-CPB to circum-neutral water or to A R D solution promoted the dissolution of the binder material, increasing the porosity of the backfill and further infiltration of aqueous solution. Long-term exposure or flooding of C P B was found to promote the precipitation of secondary, expansive minerals such as gypsum in addition to solubilizing primary cement minerals. Detailed chemical analyses and acid-base accounting indicated that the neutralizing potential added to the material by the cement phase is short-lived and the small volumes added are insufficient to neutralize the acid generating potential of the mixture. A l l A R D solution- leached C P B samples formed an increasingly thick crust of precipitates that, with time, reduced the ability of the C P B to neutralize the A R D solution 111 TABLE OF CONTENTS ABSTRACT JI TABLE OF CONTENTS IV LIST OF TABLES IX LIST OF FIGURES XI LIST OF PLATES XVII ACKNOWLEDGEMENT XXIII 1 INTRODUCTION 1 1.1 P R O B L E M DEFINITION A N D O B J E C T I V E S 1 1.1.1 Pyrite Reactivity Study 1 1.1.2 Chemical Stability of Cement Paste Backfill 3 1.2 B A C K G R O U N D 7 1.2.1 Pyrite Reactivity Study 7 1.2.1.1 Pyrite Oxidation Controls 9 \.2.\2 Electrochemistry Applied to the Study of Pyrite Reactivity 11 1.2.2 Cemented Paste Backfill in the Mining Industry 12 1.2.2.1 Cement Chemistry, Hydration and Chemical Stability 14 1.2.2.2 CPB in the Mine Environment 17 2 MINE SITE DESCRIPTIONS 26 2.1 C O M P A R I S O N OF M I N E SITE A N D E X P E R I M E N T A L CONDITIONS 28 3 EXPERIMENTAL METHODOLOGY 29 iv 3.1 P Y R I T E E X P E R I M E N T S 29 3.1.1 Sample Preparation and Analyses 29 3.1.2 Leaching Apparatus 30 3.1.3 Chemical analyses 32 3.1.4 Mineralogical and Electrochemical Characterisation of Pyrite 33 3.2 P A S T E E X P E R I M E N T 36 3.2.1 Paste Sample Preparation 36 3.2.1.1 Paste Components 36 3.2.1.2 Paste Mixing 37 3.2.1.3 Pre-Leaching Sample Preparation 42 3.2.2 Leaching Apparatus and Leaching Cycles 43 3.2.3 Post-Leaching Sample Preparation 44 3.2.4 Chemical Analyses of the Paste Experiment 46 3.2.4.1 Paste pH and Acid-Base Accounting Measurements 48 4 RESULTS - PYRITE REACTIVITY STUDY 49 4.1 M I N E R A L O G Y , C H E M I S T R Y A N D STOICHIOMETRY OF U N L E A C H E D PYRITES 49 4.2 E L E C T R O C H E M I S T R Y OF T H E U N L E A C H E D P Y R I T E S A M P L E S 56 4.3 E V O L U T I O N OF L E A C H A T E C H E M I S T R Y 62 4.4 E V O L U T I O N OF PRECIPITATE C O A T I N G S D U R I N G L E A C H I N G 67 4.4.1 Huckleberry pyrite 67 4.4.2 Louvicourt Pyrites 68 4.4.3 Tizapa Pyrite 68 4.4.4 Zimapan Pyrite 69 4.4.5 Brunswick Pyrite 70 4.5 E V O L U T I O N OF PYRITE R E A C T I V I T Y 77 4.5.1 Huckleberry pyrites 77 4.5.2 Louvicourt Pyrites 78 V 4.5.3 Tizapa pyrite 79 4.5.4 Zimapan pyrite 79 4.5.5 Brunswick 80 4.5.6 Evolution of Pyrite Reactivity 88 5 RESULTS - CEMENTED PASTE B A C K F I L L STUDY 88 5.1 C H E M I C A L A N A L Y S E S OF TAILINGS, B I N D E R A N D P A S T E 88 5.2 A S P E C T OF C U R E D P A S T E S A M P L E S 91 5.3 W A T E R A B S O R P T I O N B Y T H E P A S T E 92 5.4 C H E M I S T R Y OF P A S T E L E A C H A T E S 95 5.4.1 Flooded and Cycled Water-Leached Environments 95 5.4.2 Fe2(S04)3 Solution-Leached Environment 108 5.4.3 Summary of Observations from Leachate Chemistry 117 5.5 SOLID P H A S E C H E M I S T R Y OF L E A C H E D P A S T E 120 5.5.1 Tizapa 120 5.5.2 Brunswick '. 122 5.5.3 Louvicourt 123 5.5.4 Francisco I. Madero 124 5.5.5 Summary of Observations from the Solid Phase Chemistry 129 5.6 B U F F E R I N G C A P A C I T Y A N D A C I D - B A S E A C C O U N T I N G 131 5.6.1 Evolution of Acid Producing and Acid Neutralizing Potentials: 131 5.7 S C A N N I N G E L E C T R O N M I C R O S C O P E ( S E M ) O B S E R V A T I O N S 137 5.7.1 Tizapa 137 5.7.2 Brunswick 141 5.7.3 Louvicourt 143 5.7.4 Francisco I. Madero 146 5.7.5 Summary of Observations from Scanning Electron Microscopy 164 5.8 C O M P R E S S I V E S T R E N G T H M E A S U R E M E N T S 167 v i 6 DISCUSSION 172 6.1 P Y R I T E R E A C T I V I T Y S T U D Y 172 6.1.1 Effect of Precipitate Coatings on the Passivation of Pyrite: Huckleberry and Louvicourt-1 Samples 172 6.1.2 Effect of Stoichiometry on Pyrite Reactivity: Louvicourt-1 andLouvicourt-2 Samples.... 175 6.1.3 Effect of Mineral Impurities on Pyrite Reactivity 176 6.1.3.1 Galvanic Protection by Sphalerite: Louvicourt-2 and Tizapa Samples 176 6.1.3.2 Effect of the Presence of Galena: Zimapan and Brunswick Samples 178 6.1.4 The Huckleberry Problem 179 6.2 C H E M I C A L STABILITY OF P A S T E M I X T U R E S 180 6.2.1 Effect of Cementing Tailings Waste 180 6.2.2 Cemented Backfill Alteration with Leaching 182 6.2.2.1 Water-Leached Environments 182 6.2.3 Ferric Sulfate or Artificial ARD Environments 189 7 CONCLUSIONS 193 7.1 P Y R I T E R E A C T I V I T Y S T U D Y 193 7.1.1 Huckleberry Pyrite 195 7.1.2 Application of this Study to Field Investigations 195 7.2 C E M E N T E D P A S T E B A C K F I L L L E A C H I N G S T U D Y 197 8 RECOMMENDATIONS 202 8.1 P Y R I T E R E A C T I V I T Y S T U D Y 202 8.2 P A S T E B A C K F I L L S T U D Y 203 REFERENCES 206 APPENDIX I - ANALYTICAL RESULTS: PYRITE LEACHATE 214 APPENDIX II - ANALYTICAL RESULTS: PASTE BACKFILL LEACHATE 219 v i i APPENDIX III - ANALYTICAL RESULTS: PASTE BACKFILL SOLID PHASE v i i i LIST OF TABLES Table 1.1: Chemical Composition of Normal Portland Cement (Canada type 10 or Mexico no. 1) 15 Table 1.2 Concrete Subjected to Sulfate Attack 20 Table 1.3: Replacement Components of Portland Cement to Improve Sulfate Resistance of Concrete 23 Table 2.1: Summary of Sample Geological Setting and Mineralogy 26 Table 2.2 Hydrogeologic Conditions at Mine Sites where Backfi l l is or w i l l be Used.. . 28 Table 3.1 Leachate Analyses Carried Out at U A S L P 32 Table 3.2 Leachate and Solids Analyses Carried Out at C I D T 33 Table 3.3 Paste Sample Preparation 40 Table 3.4 Paste Leaching Scenarios 43 Table 3.5 Solids Phase Analyses for the Paste Experiments 48 Table 4.1 Solid Phase chemistry of Pyrite Samples 51 Table 4.2 Normative Mineralogy of Pyrite Samples 52 Table 4.3 Average Stoichiometry of Pyrites 53 ix Table 4.4 Open Current Potentials of Pyrite Samples (Volts, SHE) 56 Table 4.5 Relative Reactivities of Pyrites 90 Table 5.1 Chemical Analyses of Tailings and Binders Used in Paste Mixtures 90 Table 5.2 Final Composition of Paste Samples 90 Table 5.3a, b Paste Water Content: 5 weeks and 20 weeks 93 Table 5.4 Tizapa - Description of Leached Paste Sections 125 Table 5.5 Brunswick - Description of Leached Paste Sections 125 Table 5.6 Louvicourt - Description of Leached Paste Sections 126 Table 5.7 Francisco I. Madero - Description of Leached Paste Sections 127 Table 5.8 a, b, c, d Paste p H and Acid-Base Accounting Test Results 133 Table 5.9 Unconfined Compressive Strength of Paste Samples 169 Table 6.1 Dissociation Constants for Secondary Minerals of Z n and Pb 179 Table 6.2 Solubility Products and Dissociation Constants of Secondary Minerals Derived from Hydrated Portland Cement and Other Reference Minerals 185 LIST OF FIGURES Figure 1.1 Paste Backfi l l Leaching Environments 20 Figure 2.1 Sites Location of Huckleberry, Louvicourt and Brunswick mines, Canada... 27 Figure 2.2 Site Location of Tizapa, Zimapan and Francisco I. Madero (FIM) mines, Mexico 27 Figure 3.1 Pyrite Leaching Apparatus 32 Figure 3.2 Preparation of Carbon Paste Electrode (CPE) 35 Figure 3.3 Electrochemical Cel l (front clip attached to CPE) 35 Figure 3.4: a. M i x i n g Paste Sample b. Standard Slump Test 42 Figure 3.5 Rodding Paste into Mould Figure 3.6 Curing of Paste Samples 42 Figure 3.7 Leaching Cells Figure 3.8 Lab Set up, Paste Leaching Study 44 Figures 3.9 a, b, c: Paste Sample Preparation Steps for Mineralogical Observations 45 Figure 4.1 A l l unleached pyrite samples Figure 4.2 Close up view 60 x i Figure 4.3 Galena Figure 4.4 Sphalerite 61 Figure 4.5 Pyrite Leachate p H 63 Figure 4.6 Pyrite Leachate Redox Conditions 64 Figure 4.7 Pyrite Leachate Conductivities 64 Figure 4.8 Pyrite Leachate Sulfate Concentrations 65 Figure 4.9 Pyrite Leachate Total Iron Concentrations 65 Figure 4.10 Pyrite Leachate Zinc Concentrations 65 Figure 4.11 Pyrite Leachate Lead Concentrations 66 Figure 4.12 Pyrite Leachate Copper Concentrations 66 Figure 4.13 Pyrite Leachate Arsenic Concentration 66 Figure 4.14 Huckleberry Water-Leached Pyrites Figure 4.15 Huckleberry Solution-Leached Pyrites 82 Figure 4.16 Louvicourt-1 Pyrite Figure 4.17 Louvicourt-2 Pyrite 83 Figure 4.18 Tizapa Pyrite Figure 4.19 Zimapan Pyrite 85 x i i Figure 4.20 Zoom on Initialization points, Zimapan Pyrites 86 Figure 4.21 Brunswick pyrites Figure 4.22 Zoom of Initialization Points 87 Figure 4.23 A l l Pyrites, 4-week Leached 88 Figure 4.24 A l l Pyrites, 10-week Leached Figure 4.25 A l l Pyrites, 20-week Leached 89 Figure 5.1 Superficial Oxidation of Freshly Cured Tizapa Paste Samples 91 Figure 5.2 Water Absorption in Cyclic-Leaching (water) Environment 94 Figure 5.3 Water Absorption in Flooded-Leaching (water) Environment 94 Figure 5.4 Water Absorption in F e ^ S O ^ Solution 94 Figure 5.5 Flooded Cells - Leachate p H Figure 5.6 Cycled Cells - Leachate p H 100 Figure 5.7 Flooded Cells - Conductivity Figure 5.8 Cycled Cells - Conductivity 100 Figure 5.9 Flooded Cells - Redox Potential Figure 5.10 Cycled Cells - Redox Potential 101 Figure 5.11 Flooded Cells - Dissolved Iron Figure 5.12 Cycled Cells - Dissolved Iron 102 x i i i Figure 5.13 Flooded Cells - Dissolved Zinc Figure 5.14 Cycled Cells - Dissolved Zinc 102 Figure 5.15 Flooded Cells - Dissolved Lead Figure 5.16 Cycled Cells - Dissolved Lead 103 Figure 5.17 Flooded Cells - Sulfate Concentration Figure 5.18 Cycled Cells - Sulfate Concentration 103 Figure 5.19 Flooded Cells - Calcium Concentration Figure 5.20 Cycled Cells - Calcium Concentration 104 Figure 5.21 Flooded Cells - Calcium depletion rate Figure 5.22 Cycled Cells - Calcium depletion rate 104 Figure 5.23 Flooded Cells - Magnesium Concentration Figure 5.24 Cycled Cells - Magnesium Concentration 105 Figure 5.25 Flooded Cells - Magnesium depletion rate Figure 5.26 Cycled Cells - Magnesium depletion rate 105 Figure 5.27 Flooded Cells - Potassium Concentration Figure 5.28 Cycled Cells - Potassium Concentration 106 Figure 5.29 Flooded Cells - Potassium Depletion Rate Figure 5.30 Cycled Cells - Potassium Depletion Rate 106 xiv Figure 5.31 Flooded Cells - Silicon Concentration Figure 5.32 Cycled Cells - Silicon Concentration 107 Figure 5.33 F e 2 ( S 0 4 ) 3 Cells - p H 112 Figure 5.34 F e 2 ( S 0 4)3 Cells - Conductivity 112 Figure 5.35 F e 2 ( S 0 4)3 Cells - Redox Potential 112 Figure 5.36 Ferric Hydroxide Coating on Paste Sample 113 Figure 5.37 F e 2 ( S 0 4 ) 3 Cells - Dissolved Iron 113 Figure 5.38 F e 2 ( S 0 4 ) 3 Cells - Dissolved Zinc 113 Figure 5.39 F e 2 ( S 0 4 ) 3 Cells - Dissolved Lead 114 Figure 5.40 F e 2 ( S 0 4 ) 3 Cells - Sulfate Concentration 114 Figure 5.41 F e 2 ( S 0 4 ) 3 Cells - Calcium Concentration 114 Figure 5.42 F e 2 ( S 0 4 ) 3 Cells - Calcium Depletion Rate 115 Figure 5.43 F e 2 ( S 0 4 ) 3 Cells - Magnesium Concentration 115 Figure 5.44 F e 2 ( S 0 4 ) 3 Cells - Magnesium Depletion Rate 115 Figure 5.45 F e 2 ( S 0 4 ) 3 Leach Cells - Potassium Concentration 116 Figure 5.46 F e 2 ( S 0 4 ) 3 Leach Cells - Potassium Depletion Rate 116 xv Figure 5.47 F 62(804)3 Leach Cells - Silicon Concentration 116 Figure 5.48 Tizapa Paste - U C S with Leaching Time 170 Figure 5.49 Brunswick Paste - U C S with Leaching Time 170 Figure 5.50 Louvicourt Paste - U C S with Leaching Time 170 Figure 5.51 Francisco I. Madero Paste - U C S with Leaching Time 171 xv i L I S T O F P L A T E S Plate 4.1 Zimapan Pyrite Plate 4.2 Brunswick Pyrite Plate 4.3 Louvicourt-2 Pyrite Plate 4.4 Tizapa Pyrite Plate 4.5 Huckleberry Pyrite Plate 4.6 Louvicourt-1 Pyrite HUCKLEBERRY PYRITES Plate 4.7 Water-Leached, 20 weeks Plate 4.8 Water-Leached, 20 weeks Plate 4.9 Solution-Leached, 20 weeks Plate 4.10 Solution-Leached, 20 weeks LOUVICOURT-1 PYRITES Plate 4.11 4-week leached Plate 4.12 20-week leached Plate 4.13 20-week leached LOUVICOURT-2 PYRITES Plate 4.14 Pyrite surface at 10 weeks Plate 4.15 Pyrite surface at 20 weeks Plate 4.16 Pyrite at 10 weeks Plate 4.17 20-week leached xvi i TIZAPA PYRITES Plate 4.18 Pyrite surface at 20 weeks Plate 4.19 Oxidized sphalerite, 20 weeks Plate 4.20 Precipitate cover, 20 weeks Plate 4.21 Oxidized galena, 20 weeks 72 Zimapan Pyrites Plate 4.22 Pyrite surface at 20 weeks Plate 4.23 Corrosion pits at 4 weeks Plate 4.24 Anglesite precipitates, 20 weeks 73 BRUNSWICK PYRITES Plate 4.25 Pyrite surface at 20 weeks Plate 4.26 Pyrite corrosion pits, 20 weeks Plate 4.27 Anglesite and galena, 4 weeks 74 PASTE BACKFILL PLATES Plate 5.1 Portlandite crystals Plate 5.2 Typical E D X of portlandite Plate 5.3 Secondary gypsum crystals Plate 5.4 Typical E D X of gypsum 149 Plate 5.5 Acicular ettringite Plate 5.6 Typical E D X of ettringite 150 xv i i i Plate 5.7 Amorphous tobermorite Plate 5.8 Typical E D X of amorphous tobermorite 150 TIZAPA BACKFILL Plate 5.9 Portlandite cover on sample surface Plate 5.10 Bottom of layer 1 Plate 5.11 Bottom of layer 1 Plate 5.12 Poorly developped Tobermorite, below layer 1 Plate 5.13 Core of sample Plate 5.14 Primary ettringite on incomp-letely hydrated cement grain 151 Plate 5.15 Poorly developped Tb, layer 1 Plate 5.16 Ettringite, upper layer 1 Plate 5.17 Secondary gypsum, upper layer 1 Plate 5.18 General paste aspect, layer 1 Plate 5.19 Ettringite in core of sample Plate 5.20 Good tobermorite development at core of sample ..152 Plate 5.21 Porous layer 1 Plate 5.22 Depleted tobermorite, layer 1 Plate 5.23 Area of good tobermorite, layer 1 Plate 5.24 Layer 1, iron sulfate precipitate Plate 5.25 E D X of iron sulfate precipitate Plate 5.26 Good tobermorite cover, layer 2 154 xix BRUNSWICK BACKFILL Plate 5.27 Layer 1 Plate 5 28 Layer 1 Plate 5 29 Less resistant tobermorite, layer 1 Plate 5 30 Abundant tobermorite, layer 2 154 Plate 5.31 Good tobermorite cover, layer 1 Plate 5.32 General aspect of paste, layer 1 Plate 5.33 Paste surface Plate 5.34 General aspect of paste, layer 1 Plate 5.35 Poor tobermorite cover, layer 1 Plate 5.36 Possible 2ry ettringite, layer 1 155 Plate 5.37 General aspect of paste, layer 2 Plate 5.38 Fe-sulfate precipitate, layer 2 Plate 5.39 Inside large pore, layer 2 Plate 5.40 General aspect of paste, layer 3 Plate 5.41 Layer 3 156 LOUVICOURT BACKFILL Plate 5.42 Massive tobermorite Plate 5.43 Dendritic tobermorite Plate 5.44 E D X of tobermorite developed from slag Plate 5.45 General aspect of paste mixture 157 xx Plate 5.46 Less tobermorite in upper layer 1 Plate 5.47 More abundant tobermorite in lower layer 1 Plate 5.48 Poorly developed Tb, upper layer 1 Plate 5.49 Upper layer 1 Plate 5.50 More developed Tb, lower layer 1 Plate 5.51 Secondary gypsum, lower layer 1 Plate 5.52 Tobermorite-lined fracture in paste Plate 5.53 Close-up view of tobermorite Plate 5.54 Depleted tobermorite, layer 1 Plate 5.55 Masses of tobermorite at upper layer 1 Plate 5.56 Increased Tb content, below layer 1 Plate 5.57 Fracture developed possibly along existing plane of weakness Plate 5.58 Botryoidal mass of iron sulfate precipitate Plate 5.59 E D X of iron sulfate precipitate (type of jarosite?) Plate 5.60 Possibly ettringite, layer 2 Plate 5.61 Wel l developed tobermorite, lower layer 2 FRANCISCO I. MADERO BACKFILL Plate 5.62 Layer 1, flooded sample Plate 5.63 Core of flooded sample Plate 5.64 Layer 1, hydrated cement phase xxi Plate 5.65 General aspect of paste Plate 5.66 Good development of tobermorite, layer 1 Plate 5.67 Layer 1 Plate 5.68 Monosulfoaluminate (?), below layer 1 Plate 5.69 Core of sample xxn ACKNOWLEDGEMENT This project was the first of an international collaboration and technology exchange between the University of British Columbia and the Universidad Autonoma de San Luis Potosi. Although the logistics for this project were sometimes challenging, the project was carried out with great success thanks to the efforts of many determined people. In particular I would like to sincerely thank my supervisor and project co-ordinator D r Richard Lawrence for his time and attention, Dr Marcos Monroy, Dr George^ Poling and Dr Ignacio Gonzalez for their supervision, valuable comments and discussions throughout the study and for their support and friendship. I would also like to thank Dr Marcello Veiga, and Dr Ralph Hackle for their comments and revisions. Funding of this research was provided by a grant from the National Science and Engineering Research Council of Canada ( N S E R C ) , by Noranda Technology Centre (NTC) , Centra de Investigation y Desarollo Tecnologico (CIDT) of Pefioles, the University of British Columbia, the Universidad Autonoma de San Luis Potosi, Universidad Autonoma Metropolitana-Ixtapalapa of Mexico City. In particular, I would like to thank Luc St-Arnaud and Michael L i of N T C , Louis Racine of Louvicourt mine and Ar ie Moerman of Brunswick mining division, Noranda. Thanks also goes out to Benjamin Ruiz and Carlos Lara of CIDT, and Peter Campbell formerly of Huckleberry mines. I thank Sally, Frank, Marina, Terry, Shannon, Pary, Dr. Dunbar and Dr. Scoble at the department of Min ing and Mineral Process Engineering of U B C for their support. M e gustaria agradecer a Irene Chavira y Roel Cruz-Gaona por su invaluable ayuda con los experimentos que lleve a cabo en Mexico, y tambien por su paciencia y amistad. Tambien agradezco a Ulices, Pancho, Israel, Lorena, Teresita y Eliceo por sus atenciones y, claro, su amistad. Finalement, j 'aimerais remercier du fond du coeur Stephane D'Aoust et Therese Bertrand ainsi que Kathy et Tracy pour leur soutien moral, encouragement et amitie. xx i i i 1 INTRODUCTION 1.1 P R O B L E M DEFINITION AND OBJECTIVES The oxidation of sulfide-rich mining wastes produces drainage water of poor quality, contaminated by dissolved heavy metals and which is often of low p H . This contaminated drainage is termed acid rock drainage ( A R D ) and is the most costly environmental problem facing the mining industry today. In many cases, the only solution is long term treatment once the process of oxidation is under way. A key mineral in A R D generation is pyrite ( F e S 2 ) which, although not the most reactive sulfide mineral, is by far the most common, frequently present in large quantities in mining wastes. This study has been carried out to observe the electrochemical behaviour of pyrite oxidation to determine i f such fundamental observations might be used to refine the prediction of the behaviour of pyritic mine wastes. Building on the knowledge of pyrite reactivity, experiments have been carried out to observe the chemical stability of paste backfill containing pyritic tailings, exposed to various leaching conditions encountered in mine settings. 1.1.1 Pyrite Reactivity Study A considerable amount of research has been carried out in the last two decades to improve the understanding of mineral waste oxidation and provide methods to control the generation of A R D . Fundamental studies of pyrite oxidation such as the processes involved, oxidation rates and reaction products, abound in the literature (Singer and 1 Stumm, 1970; Lowson, 1982; Moses et al., 1987; Brown and Jurinak, 1989; Nicholson et al, 1989; Moses and Herman, 1991; Nicholson, 1994; Blowes et al, 1995; Eidsa et al, 1997). The application of this fundamental knowledge to field conditions, to predict drainage water quality, remains difficult because of the heterogeneous nature of mineral waste piles and the multitude of chemical reactions and physical factors that can affect the generation of A R D and subsequent water quality. The mining industry has typically relied on the use of simple and often short-term laboratory weathering tests such as humidity cells to attempt to predict the behaviour o f mining wastes exposed to the environment and quality of the drainage water. A n increasing number or government agencies require that specific static and kinetic tests be carried out on mineral wastes, prior to permitting a new mining operation to determine the potential of waste to generate acid. Mineral waste characterization programs required by the government of British Columbia, Canada, are presented in the B C A M D Draft A c i d Rock Drainage Technical Guide (Steffen, Robertson and Kirsten Ltd. , 1989), with revisions by Price and others (1997). The limitation to kinetic tests is that only products are measured without detailed information on sulfide oxidation kinetics or the possible change in oxidation rate, for example, with the evolution of precipitation products. In addition, the data generated by some of these tests has been shown to vary depending on the procedures followed for a given tests (Lawrence and Wang, 1997), and to be difficult to extrapolate to predict the.future chemical behaviour of mine wastes and the quality of the leachates derived from them (Bethune et al., 1997; Otwinowski, 1997). Inaccurate predictions on A R D generation can have costly consequences such as over-design of treatment facility or worse yet, unplanned environmental restoration costs due to design 2 failures. Reduction in risks and costs associated with the management of mining wastes is the objective of the great amount of research actively being carried out to improve prediction techniques. In the first part of this thesis, cyclic voltamperometry was used to study the effects of oxidative leaching on the reactivity of pyrite. Cycl ic voltamperometry is an established investigative tool used in electrochemical studies to characterize surface and/or semiconducting properties of metals and minerals. Electrochemical techniques of various kinds have been used extensively in the mineral processing industry (review by Peters, 1984). Cycl ic voltamperometry was used in conjunction with mineralogical investigation techniques to characterize the relative reactivity of various pyrite samples, and the evolution of reactivities as leaching progressed. The study was carried out with the aim to document the influence of intrinsic and extrinsic factors on the reactivity of pyrite. The objectives of this first part of the study are: 1) Investigate the usefulness of this relatively rapid and simple technique as an acid rock drainage predictive tool, 2) Measure the initial effects of leaching on pyrite to attempt to improve the understanding and subsequent interpretation of humidity cell data in the early stages of leaching. 1.1.2 Chemical Stability of Cement Paste Backfill Due to the uncertainty associated with oxidation of mining wastes and the lack of fail- proof, effective protection against A R D , especially for sulfide-rich wastes, the mining 3 industry is increasingly pushing to minimize the surface disposal of waste to try to avoid the possible environmental problems associated with that practice. A n option that is gaining popularity in the mining industry is the use of total tailings, including sulfide-rich tailings, as backfill material within the mine. Backfi l l is commonly used in mining operations to provide underground support and improve ore recovery, and as a method to dispose of some of the waste generated. Recent technological developments allow the use of total tailings, including the fine portion, which has traditionally been avoided because o f excessive water retention and associated backfill stability problems. Cement can be added in small proportion to the mixture to increase the short and/or long term strength of the backfill (Landriault et al., 1998). This material is referred to as cemented paste backfill (CPB) . Advantages to using cemented, total tailings backfill include the decrease in volume of waste to be disposed of on the surface, thereby reducing the liability associated with the long-term care of a tailings disposal facility and the possible environmental problems associated with the oxidation of that waste in the case of reactive tailings. Another reported advantage is the added neutralization capacity provided by the cement to reactive tailings. The neutralized tailings could potentially serve a dual purpose: preventing the oxidation of the waste and the subsequent generation of A R D , as well as neutralizing existing A R D actively produced within the mine when coming in contact with the backfill material. Backfi l l stability studies generally focus on the rheological or geotechnical properties of the backfill material. Many investigations are carried out to evaluate lower cost mineral additives to replace portions of Ordinary Portland Cement (OPC) while preserving the 4 strength of the cured material and into improvements in the short and/or long term strength of the mixtures (Hopkins and Beaudry, 1989; Lord and L i u , 1998; Gay and Constantiner, 1998; Noranda Technology Centre, 1998a,b,c,d; Archibald and Chew, 1998). Very few studies have been carried out on the reactivity or chemical stability of cured paste mixtures exposed to various environmental conditions. It is well known in the concrete industry that fine, sulfidic aggregate has deleterious effects on the setting ability of hydrated cement and its long-term durability. Furthermore, high concentrations of sulfate such as those commonly present in the tailings water used to make the mixtures or present in the mine waters (or A R D ) coming in contact with the backfill are recognized in the concrete industry as aggressive solutions. Sulfates react with the hydrated cement to produce expansive minerals that cause the material to crack and lose its strength. The resulting physical weakening of the backfill concrete is highly undesirable, especially when the backfill is used to provide physical support of the underground structures. The physical breakdown of the backfill can also result in the exposure of reactive sulfide particles to the environment, thereby increasing the risk of oxidation. In the second part of this study, the chemical stability of cemented paste mixtures that contain reactive waste was studied following 20-week leaching periods in environments similar to those encountered in a mine setting. The principal objective of the paste backfill stability study was to document the changes in properties occurring within various pastes when exposed to these leaching environments. The objectives of this second part of the study were to provide insights on questions and actual problems reported by mines currently using or planning to use C P B , namely: 5 1) Decreasing backfill strength with time resulting in higher than expected dilution of ore when blasting next to backfilled areas, 2) Effectiveness of C P B to neutralize reactive tailings used in the backfill mixture and to buffer A R D produced within the mine. 3) The feasibility of using reactive tailings in above-ground applications of C P B In the first part of the study, 6 samples of pyrite from mines in Canada and Mexico (Huckleberry, Louvicourt samples 1 and 2, Brunswick and Zimapan, Tizapa respectively) were studied using cyclic voltamperometry, scanning electron microscope (SEM) , energy dispersive x-ray analysis ( E D X ) , chemical analyses of solids and of the leachate to obtain information on the reactivity of each pyrite in the early cycles of leaching. In the second part, 4 paste backfill samples were prepared using the formulation specified by each mine (Louvicourt, Brunswick, Tizapa and Francisco I. Madero) to study their weathering characteristics when subjected to various leaching environments. S E M , E D X , acid-base accounting analyses as well as of solid phase and leachate chemistry was used for this investigation. In the remaining section of this chapter, the reactivity of pyrite w i l l be discussed followed by a detailed review of the use of paste backfill in the mining industry and the potential problems associated with exposure to environments encountered in a mine setting. Chapter 3 presents the methodology and sample preparation procedures developed for both the pyrite and backfill studies. Chapter 4 presents the results, analyses of data, 6 together with brief summaries of observations. A discussion of the results is presented in Chapter 5. Chapter 6 presents the conclusions of the study together with recommendations for further investigations and which could benefit similar studies. 1.2 BACKGROUND 1.2.1 Pyr i te Reactivity Study Pyrite oxidation reactions and their mechanisms are generally well understood. The overall reactions can be summarized as follows: Initial circum-neutral oxidation o f pyrite: FeS 2 + H 2 0 + 7 / 2 0 2 -> F e 2 + + 2 S 0 4 2 ~ + 2 H + (1.1) Slow, rate limiting Fe oxidation (more rapid, bacterially mediated at p H < 3.5): 2 F e 2 + + V 2 0 2 + 2 H + -> 2 F e 3 + + H 2 0 (1.2) Ferric iron oxidation of pyrite at p H < 3.5: FeS 2 + 14 F e 3 + + 8 H 2 0 -> 15 F e 2 + + 2 S 0 4 2 " + 16 H + (1.3) Although pyrite may not be the most reactive sulfide mineral, the complete oxidation of one mole of pyrite in a bacterially mediated environment p H < 3.5 can liberate 16 protons and a considerable amount of dissolved metals. B y comparison, the oxidation of one mole of a mono sulfide mineral such as sphalerite (ZnS), galena (PbS) and covellite (CuS) w i l l liberate one mole of dissolved metal (Me 2 + ) , one mole of sulfate but no protons: 7 M e S + 2 0 2 ^ Mez+ + S 0 4 2- (1.4) Under circum-neutral p H conditions, dissolved metal w i l l combine with hydroxide to precipitate sparingly soluble metal hydroxide according to equation 5, slowly decreasing the activity of O H - in solution. Reactions (1.1) to (1.5) show that, once started, the oxidation of sulfide minerals becomes difficult to stop or slow down without costly control measures. For this reason, the key to successful management of potentially acid-generating waste is to prevent the initiation of oxidation reactions. Static and kinetic tests have been devised to provide information on the potential of mineral wastes to oxidize, the time of initiation of acid generation and the expected loading of metals and low p H water to the receiving environment. This information is obtained from the interpretation of the leachate chemistry data, from the comparison of sulfide oxidation rates and carbonate mineral depletion rates measured during leaching. Extrapolation of the calculated rates and ion loading suggests a time frame for acid generation: onset, duration and metal loading to receiving waters. In many cases, however, leachate data yields data that can be difficult to interpret and, therefore, make predictions with any degree of certainty. Such is the case for the low- grade ore waste of the new Huckleberry mine in British Columbia, one of the pyrites studied in this project. Kinetic test results for Huckleberry low-grade ore was inconclusive as to the onset of oxidation of sulfide minerals contained in the rock. Static M e / + + 2 O H > M e ( O H ) 2 (1.5) 8 tests carr ied out o n this material showed a neutral izat ion potential ratio ( N P R ) o f 0.78, character iz ing it as likely acid generating according to current B r i t i s h C o l u m b i a G u i d e l i n e s (Price et al., 1997). D u r i n g 3 years o f k inet ic ( c o l u m n leach) tests, however , p H , sulfate and c a l c i u m levels remained elevated (Lawrence , 1997). T h e h i g h sulfate and c a l c i u m concentrations were attributed to the disso lut ion o f the h i g h g y p s u m content o f the mater ia l . Consequent ly , no sulfide o x i d a t i o n rates or neutral izat ion deplet ion rates were extractable f r o m the leachate data, such that no predict ions were poss ible as to the onset o f A R D or expected metal loading . A s a result, the m i n e h a d to assume the mater ia l to be reactive and dispose o f the material into the tai l ings p o n d rather than use it as construct ion material o f the inner p o n d w a l l where it w o u l d become f looded w i t h i n 8 years. In l ight o f the cost ly disposal alternative for this material o f uncertain A R D potential , the first part o f this thesis studied the reactivity o f pyrite (the p r i n c i p a l sulf ide m i n e r a l i n the waste) i n both gypsum-saturated and gypsum-free environments, to document the e v o l u t i o n o f the react iv i ty o f pyrite as leaching progressed. T h e results were expected to help evaluate the probabi l i ty o f the waste to o x i d i z e w i t h i n the t ime frame o f 8 years exposed to air and water. F i v e other pyrites were studied a long w i t h H u c k l e b e r r y to evaluate the effects o f m i n e r a l o g i c a l characteristics o n the react ivi ty o f pyri te i n general. 1.2.1.1 Pyrite Oxidation Controls Sul f ide m i n e r a l o x i d a t i o n reactions are documented as surface control led, based o n the a v a i l a b i l i t y o f reactive sites to participate i n the exchange o f charges w i t h an o x i d i z i n g agent such as o x y g e n or ferric i r o n i n the case o f pyrite ( L o w s e n , 1982; M o s e s et al., 9 1987; Brown and Jurinak, 1989; Nicholson et al, 1988; Moses and Herman, 1991; Nicholson, 1994). Bacterial oxidation of pyrite, for example by T. ferrooxidans, is also documented to be surface controlled. The oxidation of ferrous iron and/or sulfide was found to be facilitated by the attachment of the bacteria to the mineral surface (Herrera et al, 1989; Free et al., 1993). When pyrite is exposed to air and water, the oxidation products of pyrite produce coatings that decrease the surface area of the grain available for oxidation, effectively decreasing the rate of pyrite oxidation. The occurrence of precipitate coatings on pyrite and its effect of decreasing reaction rates are documented by Nicholson and others (1990). Thick precipitate coatings of iron hydroxide, iron oxyhydroxides and jarosite on oxidized pyrite grains are also documented by Jambor (1994), Bigham (1994) and Alpers and others (1994). Oxidation being a surface controlled reaction, the rate of oxidation of a pyrite is documented to be influenced by the mineralogy, stoichiometry, crystal morphology and defects of pyrite crystals. McKibben and Barnes (1986), Kwong (1993) and K w o n g and Lawrence (1994) have indicated that the relative abundance of physical or chemical defects and the occurrence of mineralogical impurities associated with pyrite, influence the distribution of surface free energy on pyrite grains. Locations of higher energy created by defects are more likely to oxidize than grains or parts of grains having lower surface energy. These researchers conclude that oxidation rate of pyrite depends heavily i on the mineralogical characteristics of the mineral. Little is known, however, on how these parameters interact to influence the reactivity of pyrite. Kwong (1994) suggested a theoretical order of importance of mineralogical factors influencing the rate of sulfide oxidation based on laboratory weathering tests. Measurements of local redox potentials 10 gave indications of local reactivity but did not provide a bulk measurement of the reactivity of a pyrite sample, which could be compared to that of another sample. Bu lk reactivity measurements on a bulk sample of pyrite (averaging all defects) would provide information on the acid generation potential of pyrite that may be closer to field conditions. 1.2.1.2 Electrochemistry Applied to the Study of Pyrite Reactivity Electrochemical oxidation of pyrite combined with surface spectroscopy studies has identified kinetic processes of oxidation in various ionic solutions and p H environments. The application of cyclic voltamperometry to hydrometallurgical studies is reviewed by L i and others (1992). Buckley and others (1988), Wadsworth and others (1993), and L i and Wadsworth (1993) used cyclic voltamperometry to document the formation of sulfur, or polysulfide layers on the surface of pyrite following electrode oxidation in slightly acidic or neutral p H solutions. These coatings were found to effectively decrease the leachability or reactivity of the underlying pyrite. Doyle and M i r z a (1996) used cyclic voltamperometry to characterize pyrites from different sources to evaluate the effects of pyrite composition and electric properties (rest potential, resistivity, net concentration of donors and charge carrier concentration) on the oxidation behaviour of pyrite. They found poor correlation between chemical or electric characteristics and pyrite reactivity using their methodology. Their working electrodes consisted of polished pyrite grain, documented in later studies to be largely affected by polishing of the grain, with responses often dominated by fractures or irregularities in the surface of the grain-electrode. Lazaro and others (1995) indicated that electrochemical 11 responses obtained from carbon paste electrodes (CPE) - a mixture of pyrite, graphite and non-conducting silicon o i l - are average responses of the many grains exposed at the electrode surface, including those given by grain-surface impurities and crystal defects. The averaging effect of C P E was found to produce highly repeatable results. Cycl ic voltamperometry using C P E was used in this study to document the reactivity of six different pyrite samples and try to correlate the mineralogical characteristics of each sample with its measured reactivity. The evolution of pyrite reactivity with increasing leaching time was also studied by that method. 1.2.2 Cemented Paste Backfill in the Mining Industry Backfi l l can consist of any mining residue depending on the purpose of the backfill; from large waste rock blocks to the coarser portion of mineral processing residues (tailings) or a mixture of all rock sizes. The very fine size portion of tailings has traditionally been avoided as the water it contains drains out more slowly, resulting in backfill stability problems. Mineral processing fines are typically discarded to the tailings containment areas. Decreasing grades of mineral deposits now being exploited has resulted, however, in increasing volumes of mineral processing waste to be disposed of. Shortage of land or resources to build additional tailings impoundments is forcing the mining industry to evaluate ways to reuse the additional waste being generated. Recent technological advancements in tailings dewatering have permitted the inclusion of larger proportions of fines in backfill without the problem of excessive bleeding (drainage of excess water from the mixture) (Cincil la et al, 1997; Dahlstrom, 1997; Will iams, 1997). Backf i l l containing a high proportion of fines is a popular solution to the problem of increasing 12 mineral processing wastes. Paste backfill refers to the thick yet fluid consistency of the fresh mixture, which is specially designed for pipe flow and easy emplacement in stopes of various sizes and shapes. Contrary to plain backfill, paste f i l l can contain total tailings, which include a considerable proportion of fines. To facilitate pumping and placement of the mixture and minimize erosion of the pipe used for the transport of the paste, it is recommended that the backfill mixture contain a minimum of 15 to 20 % solids finer than 20 urn (Landriault et al, 1998). Cement can be added to paste mixtures to decrease bleeding of the backfill once in place and/or to increase the strength of the backfill upon curing (Landriault et al, 1998). One of the first uses of cemented backfill in the mining industry occurred at the B H P Mount Isa mine in Australia where, since the early 1930's, large blocks of waste rock were thrown into a vertical shaft along with hydrolysed cement to f i l l open stopes and accommodate their particular mining sequence. A n overview of the Canadian experience with the various types of backfill is given by Udd (1989). Another advantage of adding cement to paste backfill is the highly alkaline composition of cement material, which provides additional neutralizing capacity to tailings that are potentially acid generating, such as pyritic tailings (Levens and Boldt, 1994). Ideally, this waste management practice could alleviate the need to dispose of reactive waste in an engineered impoundment where, apart from the construction, maintenance and land costs, the reactive waste could oxidize and become an expensive environmental liability. 13 1.2.2.1 Cement Chemistry, Hydration and Chemical Stability Cement is one of the most commonly used materials in the construction industry and much is already known about its chemistry, its strengths and weaknesses and its interaction with a variety of aggregates (Taylor, 1997). Under normal circumstances, the inherent chemical stability, physical strength and workability of concrete allows its use in a variety of settings. Table 1.1 describes the various elements that make up Ordinary Portland Cement (OPC) used in C P B , along with the principal hydration reactions involved in the curing of cement mixtures. The cement industry uses abbreviated nomenclature for the unhydrated cement components such that the oxide in the mineral phases is referred to by one letter: C=CaO, S=Si0 2 , A = A 1 2 0 3 , F = F e 2 0 3 , S=S0 4 , H = H 2 0 . In the cement literature, the alite phase of cement is written C 3 S , referring to the composition: 3 C a O - S i 0 2 , or, in analytical chemistry, CasSiOs. This terminology is used in parts of Table 1.1 and in subsequent text to permit the correlation with the cement literature. Some agents are known to interact destructively with the components of the concrete, undermining its integrity. O f these, reactive mineral aggregates (i.e. sulfidic aggregates), high concentrations of sulfate in the mixing water or excess atmospheric C 0 2 upon drying make up the principal destabilizing agents of concrete (Kosmatka et al. 1995). These processes are briefly described as: 14 Table 1.1: Chemical Composition of Normal Portland Cement (Canada type 10 or Mexico no. 1) Principal components of T\pir:il Principal hydration pioducts dr\ ci-im-nl portion Function tricalcium silicate: C 3S C a 3 S i 0 5 (alite) Rapid hardening, early strength development -50 % 1) 2 C 3 S + 6 H 2 0 ^ Ca 3 Si 2 0 7 -3H 2 0 ( a ) + 3Ca(OH)2(b> 70% reacted in 28 days "Tobermorite gel: principal binding agent of cement b)Portlandite: no cementing properties dicalcium silicate: C 2 S Ca 2 Si0 4 (belite) Slow hardening, later stage strength development (after 1 week) -25 % 2) 2 C 2S + 4 H 2 0 -> Ca 3 Si 2 Cy3H 2 0 « + Ca(OH)2(b> 30% reacted in 28 days, 90% in 1 year Same as above tricalcium aluminate: C 3 A C a 3 A l 2 0 6 Consumes Ca(OH) 2 , produces a high heat of hydration - 1 0 % 3) C 3 A + 12 H 2 0 + Ca(OH) 2 -> Ca 4 Al 2 0 6 (OH) 2 12H 2 0 ( c ) 4) C 3 A + 10 H 2 0 + CaSCv2H 2 0 -> Ca 4 Al 2 (S0 4 )12H 2 0 ( d ) 5) C 3 A + 26H 2 0 + 3 CaS0 4-2H 20-> Ca 6Al 2(S0 4) 3(OH), 2-26H 20 ( e ) c ) Tetracalcium aluminate hydrate: some early strength development. d ) Monosulfoaluminate and e ) Ettringite: expansive minerals produced from the reaction of dissolved gypsum with C 3 A . tetracalcium aluminoferrite: C 4 A F Ca4Al 2Fe 20 1o Manufacturing purpose to reduces clinkering T°, responsible for colour effects in cement - 8 % 6) C 4 A F + 10 H 2 0 + 2 Ca(OH) 2 -> Ca 6 Al 2 Fe 2 0 1 4 12H 2 0 ( 0 7) C 4 A F + 50 H 2 0 + 6 CaS0 4 -2H 2 0 + Ca(OH) 2 -> 2 Ca«Al 2 (S0 4 ) 3 (OH) l 2 -26H 2 0 ( e ) °Calcium aluminoferrite hydrate: rapid hydration but little strength contribution. Slow hydration reaction to produce ettringite Gypsum: C S H 2 CaS0 4 -2H 2 0 Very rapid dissolution, slows the rate of C 3 A hydration to avoid flash setting ~5 % Dissolved gypsum may participate in reactions (4), (5), (6) and (7), depending on the local pore water chemistry. Too much gypsum may favour the formation of ettringite over portlandite. Fe, K , M g Present in clays used to make Portland cement. Few % May be included in any cement phase in solid solution. Notes: This table adapted from Kosmatka 1995 and Taylor 1997 ( 1 ) tricalcium silicate hydrate or "tobermorite gel": composition may vary and may include trace concentrations of Fe, Mg, K. 15 Carbonation: Carbonation refers to the excessive shrinkage of concrete upon drying caused by the penetration of atmospheric C O 2 , transforming hydroxides to carbonates. These reactions lower the alkalinity of concrete, destabilizing the curing process. High waterxement ratios, low cement content and/or short curing period enhance the potential for carbonation to occur. This phenomenon is normally restricted to shallow depth or at the surface of the concrete. Alkali-aggregate reactions: The reaction between reactive mineral aggregates used in the concrete mixture and the sodium and potassium alkalis present in the cement cause expansive secondary mineral growth. Growths of the secondary minerals create internal stresses within the concrete causing it to crack and lose its strength. Sulfate attack: The interaction of the sulfate ion present in the pore water with the hydrated compounds of the cement also results in expansive secondary mineral growth (gypsum and/or ettringite) combined with disintegration of the primary binding material (tobermorite gel). Water dissolution: Another deleterious agent of particular importance in the backfill environment is the interaction of neutral p H water with concrete, dissolving and leaching out some of the p H sensitive or water-soluble components of hydrated cement such as portlandite (Ca(OH)2) 16 or tobermorite gel (Ca3Si207"3H20). This can be particularly deleterious when contact occurs during the curing period of the freshly mixed cement (Adenot and B u i l 1992; Carde and Francois 1997). 1.2.2.2 CPB in the Mine Environment The utilization of cement in the mining industry as a binder of tailings in paste backfill is a unique application as a very small proportion of cement is normally used to bind tailings, commonly less than 10 or even 5 % of dry weight, when common concentrations of cement in concrete range from 30 to 40%. In addition, water to cement (w/c) ratios, important in the hydration and subsequent curing of cement, are also considerably higher in C P B applications compared to normal concrete. This ratio is expressed as the mass of water divided by the mass of cementing materials. Kosmatka and others (1995) explain that lower w/c ratios provide the greatest unconfined compressive strength (ucs) in normal concrete. Lamos and Clark (1989) come to the same conclusion with respect to tailings backfill. Normal concrete w/c ratios, typically around 0.5, yield 28-day ucs values ranging between 25 and 35 M P a . Typical C P B applications, such as the four mines studied in this work, have w/c ratios ranging from 5 to 10. C P B mixtures can, therefore, be expected to develop poorer ucs values upon curing than concrete containing the same amount of cement. In addition, the tailings or aggregate mixed with the cement to form paste backfill is regarded in the cement industry as undesirable aggregate because of its very fine grain size and most importantly, because of its composition, in the case of high sulfide tailings. The Canadian Standards Association (Standard A23.1) specifies the lower and upper 17 limits of grain size that should make up the fine size portion of aggregate. Standard A23.1 specifies a lower limit of 5 to 10% passing 160um and upper limit of 80 to 100% passing 2.5 mm. Furthermore, the Canadian Portland Cement Association ( C P C A ) (Kosmatka et al., 1995) indicates that the fine aggregate content should be no larger than 45% by mass or volume of the total aggregate content, above which the cement cannot efficiently coat all aggregate particles. When a large proportion of small size particles are used as aggregate, the mixture w i l l require a larger cement content to effectively coat all the particles and meet specified strength requirements. In the case of C P B , the tailings used as aggregate are often finer than 150u.m. Consequently, the cement added to the paste mixture can be expected to underperform in terms of strength development upon curing compared to a similar cement proportion used in conjunction with standard aggregate. Pyritic aggregate is normally avoided in cement mixtures because of its reactivity and the consequent production of sulfates. Sulfate is a documented aggressive agent that participates in expansive secondary mineral growths within the concrete (Shayan, 1988; DeCeukelaire, 1991; Idorn, 1992; Casanova et al., 1996). These reactions create internal stresses that lead to cracking and disintegration of the cured concrete. The details of these reactions are discussed in the later paragraphs. In summary, the constituents of paste backfill are far from ideal to provide a mixture of optimum compressive strength. The strength requirement for paste backfill is, however, commonly much lower than for a building material. In the four cases studied, the ucs requirement range between 0.5 and 3.5 M P a , depending on the purpose of the backfill . Indeed, backfill used to f i l l empty voids surrounded by rock mass requires minimal 18 strength, enough to prevent liquefaction when blasting other areas of the mine or in the event of seismic activity. Higher backfill strengths are required when the material is used to hold rock faces against which active mining is planned. In these cases, the strength of the backfill material must not be decreased by cement-altering reactions. In a mine setting, the backfill is unfortunately exposed to various conditions that can be detrimental to the chemical stability of cemented tailings backfill such as sulfate attack from sulfate rich A R D generated within the mine or present in the tailings water used to make up the backfill mixture and water dilution of the cement phase of incompletely cured backfill. Figure 1.1 shows different leaching environments to which backfill can be exposed. A l l stopes are backfilled in this figure, the red colour backfill indicates exposure to A R D generated within the host rock whereas the grey colour indicates exposure to neutral p H , infiltrating water. The cyclic-leached environment (top) represents cases where backfill is temporarily exposed to either infiltrating rain water underground but above the water table or exposed to meteoritic water in above ground applications. Flooded-leached environment (bottom left) represents backfill submerged in circum-neutral p H groundwater, and the A R D environment (bottom right) represents backfill submerged in low p H A R D water containing high concentrations of sulfate and metals. These three potential leaching environments were studied in this project. 19 Figure 1.1 Paste Backf i l l Leaching Environments 1.2.2.2.1 Sulfate attack: Sulfate is omnipresent i n most metal mines, either i n the groundwater i n contact w i t h the deposit or i n the wastewater generated by ore processing. Sulfate concentrations i n m i n e r a l process ing water are often greater than 1.5 g/1, c lass i f ied as aggressive water i n the cement industry (Table 1.2). Table 1.2 Concrete Subjected to Sulfate At tack Degree o f exposure Water soluble sulfate i n s o i l sample (%) D i s s o l v e d sulfate i n contact w i t h concrete (g/1) Port land cement type to be used V e r y severe > 2 . 0 > 10.0 50 Severe 0 . 2 - 2 . 0 1 . 5 - 1 0 . 0 50 M o d e r a t e 0 . 1 - 0 . 2 0.15 - 1.5 2 0 , 4 0 or 50 Adapted from Kosmatka et al., 1995 C e m e n t e d b a c k f i l l paste is n o r m a l l y made f r o m tai l ings p i p e d direct ly f r o m the process ing plant to w h i c h cement is added i n a secondary m i x i n g step. T h e hydrat ion 20 water of the mixture consists in large part of this sulfate-rich residue water. The only method of reducing the sulfate concentration from the tailings water would be to thoroughly wash the tailings or precipitate out the sulfate as a stable mineral phase, both uneconomical and inefficient (Noranda Technology Centre, 1998a). A second source of sulfate is acid rock drainage ( A R D ) which may not necessarily be acidic but may contain high (aggressive) concentrations of sulfate generated by the oxidation of sulfide minerals. In the first case, where sulfate is found within the tailings water, the chemical stress is generated within the paste. In the case of A R D , the chemical stress or aggressivity is external. The chemical process of sulfate attack on hydrated concrete can be summarised as follows: Free sulfate ions present in solution (from A R D or from sulfate present in the pore solution of the backfill) can combine with calcium of dissolved portlandite (Ca(OH)2) to form gypsum according to reactions 1.6: S 0 4 2 " + C a 2 + + 2 H 2 0 ^ C a S 0 4 - 2 H 2 0 (1.6) Gypsum growth in pore spaces creates pressure that, when occurring at a large scale, can induce cracking of the backfill. Crystallisation pressures can reach 70 to 2000 M P a (Ouellet et al., 1998). Another expansive reaction is the formation of ettringite (Ca6Al2(S04)3(OH)i2"26 H 2 0 ) from the reaction of aqueous sulfate and calcium, (from dissolved portlandite and free sulfate or from dissolved gypsum) and the monosulfate phase of cement (Ca4Al 2 (S0 4 ) -12H20) according to reaction 1.7: 21 C a 4 A l 2 ( S 0 4 ) - 1 2 H 2 0 + 2 S 0 4 2 " + C a 2 + + 20 H 2 0 -> C a 6 A l 2 ( S 0 4 ) 3 ( O H ) 1 2 - 2 6 H 2 0 (1.7) The higher molecular weight and larger crystal lattice of ettringite compared to monosulfate phase w i l l also induce expansive forces or crystallisation pressures within the backfill that can result in cracking and disintegration (Fu et al., 1995; Taylor, 1997). Mineral additives such as ground blast furnace slag, silica fumes or fly ash can be added or partially replace O P C to modify f i l l mixture properties, such as workability of the mixture, to increase the strength of the f i l l at a particular stage of curing or resistance to chemical attack (Mangat and Khatib, 1995; Gifford and Gillott, 1997; Taylor, 1997). Mineral additives such as those presented in Table 1.3 are used to reduce the cost of the binding agent without decreasing the strength of the f i l l . In order to improve the sulfate resistance of concrete, the amount of calcium hydroxide and calcium aluminate hydrate must be minimized. 22 Table 1.3: Replacement Components of Portland Cement to Improve Sulfate Resistance of Concrete \dditi\c I", licet Fly ash: coal combustion residue; low calcium (Tor i ie ta / . 1995) (Djuric eta/. 1996) Lower air and pore volume, reduced permeability, increased resistance to sulfate absorption into concrete. Sil ica fumes: Silicon, silicon alloy smelting ash residue; minimum 75% silicon, very low calcium and aluminium oxides ( A k o z e t a / . 1995) Lower air and pore volume, very reduced permeability, decreases gypsum and ettringite formation, increased electrical resistivity (corrosion protection) Blast furnace slag: glassy iron smelting residue; calcium silicates and aluminosilicates may be high (13-15%) A 1 2 0 3 slag or low (3-5% ) AI2O3 slag (Irassar et al. 1996) Lower air and pore volume, reduced permeability, may increase mixture strength in the end. 1.2.2.2.2 Dissolution of the Cement Phase of Backfill The dissolution of cement phases in contact with meteoritic and groundwater has been studied by the French Commission on Atomic Energy with respect to degradation of cemented containers of nuclear waste. Carde and Francois (1997) who used ammonium nitrate leaching solution previously determined to leach cements in a similar but more rapid way than water, point out that a zone of lesser strength is formed in the leached area and that the decreased strength is due to increased porosity resulting from the complete leaching of the portlandite phase of cement (Ca(OH) 2 ). The increase in porosity was calculated to be equal to the proportion of portlandite in their concrete mixtures. A 23 progressive decalcification of the tobermorite phase was also observed. Leaching of these phases were the only deterioration found to occur in concrete samples kept immersed in the leaching solution. In wet-dry cycled experiments, an increase in pore solution ion concentration was found to promote the precipitation of secondary expansive minerals such as ettringite causing internal stresses and microcracking of the concrete. Adenot and B u i l (1992) used deionized water in similar leaching experiments and observed similar alteration: thin leached zones characterized by partial or complete dissolution of portlandite but preceded by the dissolution of ettringite and monosulfoaluminate phases. Calcium to silicon ratios of the tobermorite phase were also found to decrease from the core of the specimens towards the surfaces, reflecting a decreasing calcium concentration in the pore solution of the edges of the samples. Advancement of the dissolution front was calculated to be proportional to the diffusion rate of deionized water in the concrete. The cores of their flooded samples were observed to possess a similar composition to the cores of the unleached samples. In similar experiments to those of Adenot and B u i l , Revertegat and co-workers (1992) determined that dissolution of portlandite started to occur at p H 12.5 and became more severe as the p H of the pore solution dropped. Using additional samples made of cement and fly ash, they documented the increased resistance of this binder mixture to leaching. They determined that this binder mixture uses portlandite in its hydration process, effectively decreasing the amount of portlandite available to be leached, resulting in a lower loss of porosity. 24 That research indicates that deionized water is indeed an effective leaching agent of concrete, dissolving away portlandite thereby increasing the porosity of the material, as well as decalcifying or degrading tobermorite gel, the principal binding agent of cement. The considerably higher porosity of backfill material compared to concrete suggests that leaching solution w i l l more effectively penetrate the backfill and alter the pore solution chemistry, creating disequilibrium conditions between the solution and the solid phase and accelerating the dissolution of the cement phases of the backfill. 25 2 MINE SITE DESCRIPTIONS Figure 2.1 shows the location of the Canadian mines sites and Figure 2.2, the location of Mexican mine sites studied in this project. Table 2.1 below identifies the mine sites from which samples were obtained for both the pyrite reactivity and cemented paste chemical stability studies. A summary description of the geological setting, deposit type and mode of pyrite occurrence is also presented. Table 2.1: Summary of Sample Geological Setting and Mineralogy Mine Location Deposit type Pyrite occurrence Huckleberry 1997 - present British Columbia, Canada Skarn deposit Cu, M o V e i n fill, coarse granular pyritohedrons Louvicourt 1992 - present Quebec, Canada Volcanogenic massive sulfide in felsic tuff, chert and mudstone host rock 2 . Cu , Zn , A g , A u Fine-grained massive pyrite, cubic Some chalcopyrite and sphalerite Brunswick 1950's- present N e w Brunswick, Canada Volcanogenic massive sulfide in felsic volcanic and volcaniclastic host rock Pb, Zn , Cu, A u Fine-grained massive pyrite, cubic Tizapa 1994 - present Mexico state, Mexico Stratiform volcanogenic massive sulfide in andesitic metamorphised volcaniclastic host rock Zn, Pb, Cu , A u , A g Fine-grained massive pyrite, cubic Francisco I. Madero (FIM) Project Zacatecas, Mexico Stratiform volcanogenic sulfides in metamorphised andesitic volcaniclastic host rock Pb-Zn, C u (Ag) Fine-grained massive pyrite, cubic Zimapan 1920 's(?)- present Hidalgo, Mexico Skarn Pb-Zn deposit Pb, Zn , A g , Cu , (Au) Medium to large grained cubic pyrite crystals (1) Pyrite mode of occurrence in samples used for tests (2) Tourigny et al., 1994; (3) Luff et al, 1992 26 Mine Site Locations Figure 2.1 Sites Location of Huckleberry, Louvicourt and Brunswick mines, Canada 27 2.1 COMPARISON OF MINE SITE AND EXPERIMENTAL CONDITIONS Table 2.2 Describes the site conditions encountered at the different mine sites where paste backfill w i l l be used. These conditions can be compared to the experimental conditions carried out in this study, presented in Chapter 3, Section 3.2.2, Table 3.4. Table 2.2 Hydrogeologic Conditions at Mine Sites where Backfill is or will be Used Site Mine Site Water Quality Backfill Mix Water Quality Use of Backfill (or projected use) Louvicourt p H : 8.3 S 0 4 : 620 mg/1 Fe t o t : 0.05 mg/1 Cu: 0.03 mg/1 Zn: 0.10 mg/1 p H : 7.55 S 0 4 : 700 mg/1 Fetot: 0.64 mg/1 Cu: 0.03 mg/1 Zn: 0.40 mg/1 Pyritic tailings, Underground use, (since 1992), future surficial use possible. Tizapa SCV": 2.4 mg/1 n.a. Planned underground use. Brunswick p H : 2.9 ([> S 0 4 : 6900 mg/1 Fetot: 1100 mg/1 C u : 8.4 mg/1 Pb: 2.7 mg/1 Zn: 1650 mg/1 n.a. Underground use. June 1998 startup of backfill plant. Francisco I. Madero Early development stage, no data available. Planned underground use. (1) Average water chemistry from 0-mile brook, from January 13, 1989 to July 3, 1997 (Moerman, 1997). n.a.: not available 28 3 EXPERIMENTAL METHODOLOGY A l l pyrite and paste experiments, as well as parts of the analyses, were carried out in Mexico . The laboratory equipment for both studies was constructed and previously tested at the University of British Columbia (UBC) . The laboratory set up was then dismounted and rebuilt at the Universidad Autonoma de San Luis Potosi ( U A S L P ) , in Mexico . Pyrite, tailings and slag samples were sent directly from the mines to U A S L P where they were prepared and the paste samples were mixed and cured prior to leaching. The 20-week leach cycles for both experiments, as well as most mineralogical analyses, were carried out at U A S L P . Mineralogical analyses of the leached paste mixtures were carried out at U B C . X-ray diffraction analyses were carried out at U A S L P . Cycl ic voltamperometry was carried out at the Universidad Autonoma Metropolitana - Ixtapala ( U A M I ) in Mexico City. Leachate samples from both the pyrite and the paste experiments were sent for analyses to the Centro de Investigation y Desarollo Tecnologico (CIDT) of Penoles in Monterrey, Mexico. 3.1 PYRITE EXPERIMENTS 3.1.1 Sample Preparation and Analyses Pyrite was obtained from composite hand samples chosen by the geologist of each mine to represent the most common mode of occurrence at the site. The samples containing coarse pyrite crystals (Huckleberry and Zimapan) were put in a double bag of 6 m i l polyethylene and fragmented with a steel head hammer to release pyrite crystals. 29 Individual pyrite crystals containing no visible oxidation coating or inclusions of other minerals were hand-selected for use in the tests. Hand samples of pyrite from the Louvicourt and Tizapa mines consisted of massive fine grained pyrite (75-90% pyrite content), making the segregation of individual crystals by hand practically impossible. The whole samples were therefore prepared directly without the separation of gangue or minor impurities. Samples were reduced to -6.25 mm using a jaw crusher previously cleaned by repeated passings of clear glass. N o pure silica sand was available for this purpose. Pyrite crystals and crushed particles were soaked in an acid bath of 2 N HC1 for 1 hour to remove oxidation coatings, followed by six repeated rinsing in deionized water. The pyrite samples were oven dried at 35°C for 24 hours. Dry grinding was performed in two steps, first using a glass-cleaned ring pulverizer to reduce the grain size and obtain a first batch of the required particle size and second, with an agate mortar to grind down the coarse material leftover from the ring pulverizer. The ground pyrite was dry sieved using standard 105 and 150 urn Tyler sieves. The +105 -150 um fraction was retained for the leaching tests. Sieves were washed in an ultrasound bath and completely dried between the different samples. Pyrite samples were kept in a glass dessicator under vacuum between the various preparation steps and when not in use to avoid oxidation of pyrite surfaces. 30 3.1.2 Leaching Apparatus The leaching apparatus was designed to promote the oxidation of the different pyrite samples and facilitate the measurement and comparison of the early effects of leaching on pyrite grains. Mineralogical observations, as well as chemical and electrochemical analyses, were performed on the unleached pyrite and after 4, 10 and 20 weeks of leaching. The leaching apparatus consisted of 5 cm diameter Biichner Funnels™: a polyethylene container with a flat perforated bottom and open top, attached at the bottom to a funnel (Figure 3.1). A 45 um filter paper was placed at the bottom of the container to retain the pyrite particles. 20 gm of pyrite was placed in each funnel. The leaching solution was devised to simulate rainwater: distilled water with a p H adjusted to 5.5 by the addition of C O 2 , prepared immediately before use. 15 m l of leaching solution, enough to cover the sample, was added to each sample twice weekly. The pyrite samples were left inundated for 3 hours, after which the solution was extracted by vacuum suction, re-filtered and analyzed. The pyrite samples were left exposed to ambient air between each leaching cycle. Three (3) samples of each pyrite were prepared and leached simultaneously. One (1) sample was removed at the end of each period for the various analyses. Additional samples of Huckleberry pyrite were leached with Huckleberry waste rock leach solution. The solution was generated in two leaching columns, each containing 2.5 kg of crushed low-grade ore (-3.33mm +1.68 mm). The crushed ore was previously rinsed with 5 litres of deionized water to wash out all the fines (finer than 1.68 mm). To each column was added 250 ml of deionized water with the p H adjusted to 5.5 with C O 2 . The water was retained in the columns for 1 hour after which the bottom valve was 31 opened to let the solution freely drain out. The leachate was then filtered and used to leach samples of Huckleberry pyrite prepared in the same way as the water-leached pyrite. The remaining solution was acidified and submitted for chemical analyses. Figure 3.1 Pyrite Leaching Apparatus 3.1.3 Chemical analyses Routine chemical analyses were performed on leachate solutions obtained by the pyrite experiment and by the Huckleberry waste rock columns for the parameters listed in Tables 3.1 and 3 .2. The solution pH, redox potential, conductivity and ferrous iron were analysed upon collection at U A S L P , the other leachate parameters, as well as solids analyses of the unleached pyrite samples were analyzed at the chemical laboratory of the Centro de Investigation y Desarollo Tecnologico (CIDT) of Penoles in Monterrey, 32 Mexico. The pyrite leachate solutions were acidified with concentrated HC1 to a p H < 2 and stored at 4°C between monthly shipments to Pefioles via courier. Table 3.1 Leachate Analyses Carried Out at UASLP Analysis Instrument p H , redox potential Beckman (t>320 combined p H and redox meter Conductivity Cole Parmer portable conductivity meter Model 19820-00 0.1N KC1 (Calomel) probe, all values adjusted to the standard hydrogen electrode (SHE). Ferrous iron U V Spectrophotometer Beckman D U 650, using O-phenanthroline indicator Table 3.2 Leachate and Solids Analyses Carried Out at CIDT Sample; type Analyses Methodology and Instrument Leachate Solutions (both pyrite and paste backfill experiments) Fe t o t a i , Cu , Zn , Pb, A s , A l , Sb, S, Si , Ca, M g , K Atomic absorption spectrometer ( A A ) determination (Perkin-Elmer model 5000) S 0 4 z " Gravimetric determination Solids (unleached pyrite samples) C 0 3 , S 0 4 , CaO, K 2 0 , M g O , M n O , N a 2 0 , S i 0 2 , A I 2 O 3 A g , A s , B i , C , Cd , Cu , Cr, Fe, N i , Pb, S, Se, Te, Zn . Aqua-regia digestion and A A analysis (Perkin-Elmer 5000) 3.1.4 Mineralogical and Electrochemical Characterisation of Pyrite A l l pyrite samples were characterized mineralogically and chemically prior to leaching to determine stoichiometric compositions, as well as crystal habit and form. Polished 33 sections of the hand samples containing the pyrite from each mine were prepared for observation under both optical (reflected light) microscope and scanning electron microscope (SEM) . Reflected light microscopy was carried out using a Versamet Union microscope, whereas S E M observations were carried out using a Philips X L 3 0 Scanning Electron Microscope. Quantitative chemical microanalyses were carried out on the polished sections using the energy dispersive x-ray ( E D A X ) of the S E M . Loose grain mounts of the leached pyrites were prepared after each leaching period (4, 10 and 20 weeks). The grain mounts consisted a dusting of the pyrite grains over double-sided carbon tape fixed to a metal holder. These samples were carbon-coated for S E M observations. Qualitative chemical microanalyses were carried out using E D A X to identify the precipitation products formed after each leaching period. Electrochemical analyses were carried out on unleached pyrites and after 4, 10 and 20 weeks of leaching to characterise pyrite surfaces and verify the presence of precipitates imperceptible with the S E M . A t the end of each leaching cycle, the pyrite samples were oven dried at 35°C and transported to the electrochemistry laboratory of the U A M I in air tight polyethylene containers, for voltamperometric studies. A n E G & G P A R M273 Potentiostat coupled to a P C with the M270 software were used to generate a voltage cycle of 20 mV/sec and record the voltamperometric response respectively. A 3- electrode system was used, of which a carbon paste (graphite-pyrite, 50% wt) electrode (CPE) was used as the working electrode. The carbon and pyrite were carefully blended together with laboratory grade silicon oi l in an agate mortar (Figure 3.2). The resulting paste was placed in a 0.5 ml plastic syringe into which was inserted a platinum electrode welded with silver to a copper wire end, in turn connected to the current source. The 34 counter electrode consisted of a graphite rod, the reference electrode a Hg/HgSGyK^SCM (sat), S S E (Eh=0.615V, SHE) . The electrode system was placed in a Pyrex™ glass cell containing an electrolyte solution of 0.1 M N a N 0 3 with a p H of 6.5, through which was bubbled purified nitrogen gas for a minimum of 45 minutes before the start of the experiments. Nitrogen was blown on top of the solution throughout the tests to provide an inert atmosphere within the cell. The set-up is shown on Figure 3.3. Figure 3.2 Preparation of Carbon Paste Electrode (CPE) 35 Figure 3.3 Electrochemical Cell (front clip attached to CPE) T h e zero current potential (rest potential) o f the C P E was established before conduct ing the potential sweep b y lett ing the electrode system rest for a m i n i m u m o f 5 minutes u n t i l a stable reading was obtained. S ingle-cyc le potential sweeps were init iated i n both the anodic and cathodic directions (to induce pyrite surface o x i d a t i o n and reduct ion respectively) w i t h i n the range o f - 1 . 0 to +0.7 V . T h i s range was established f r o m the " n o current response" o f a C P E that d i d not contain pyrite, p laced i n the 0.1 M N a N 0 3 electrolyte so lut ion. A n e w w o r k i n g electrode ( C P E ) surface was obtained after each sweep b y squeezing out a s m a l l amount o f paste f r o m the syringe and l e v e l l i n g the C P E t ip o n a 600 grit s i l i c a carbide sand paper. M u l t i p l e analyses were carried out o f each sweep range i n order to obtain a repeatable (representative) response usable for a comparat ive study. 36 3.2 PASTE EXPERIMENT 3.2.1 Paste Sample Preparation 3.2.1.1 Paste Components Tailings were sent by courier from the mine site to U A S L P in airtight containers with enough residual water to cover the tailings and prevent oxidation during transport. Wet tailings were oven dried at 40°C for 72 hours or until constant weight. F I M tailings were shipped dry and were used without preliminary treatment. Dry tailings were used in the paste mixture in order to standardize the mixing procedure and to have control of the water content of each mixture. City tap water was used to make up the paste samples since not enough tailings water was available to make the required volumes of samples, and some tailings' water chemistry ( F I M samples) were unavailable. The tailings were not washed prior to drying, therefore the sulfate present in the tailings water and pore water likely precipitated then redissolved upon hydration of the paste mixture. Drying the tailings prior to use is not believed to have caused major changes in the chemistry of the tailings. Ordinary Portland Cement (OPC) Type 1, the Mexican equivalent of Canadian Portland Cement No . 10, was used as binding agent as prescribed in all paste recipes submitted by the mines. The cement was analyzed chemically before forming the paste samples to make sure that it met specifications. The chemical composition of the cement is shown in Table 5.5 of Chapter 5. 37 3.2.1.2 Paste Mixing The paste backfill formulation of each mine was /followed as closely as possible in the preparation of paste samples. Since the F I M project is still in the feasibility stage, the project consultant's suggested formulation (based on general guidelines) was followed for that mine. The formulation to be used at F I M in the future may differ from that made in this study. Table 3.3 shows the proportions of tailings, water and binder required for each paste recipe (shaded) and actual proportions used to make up the paste for each mine for this investigation (un-shaded). Standard slump tests ( C S A test method A23.2-5C) were carried out to determine paste consistency using a standard slump cone. Slump test results are expressed as cm of slump of the material once the cone had been removed. A large slump corresponds to a more liquid, less consistent paste. Initially, a single tailings sample was used to determine the relative proportions of cement, water and aggregate necessary to achieve the right slump since not enough tailings samples were available to carry out repeated tests. The test indicated that only the site-specific tailings could give the required slump within the given range of component proportions. Consequently paste mixtures were made by using a fixed amount of binder and tailings from each mine to which was added the minimum amount of water suggested in each recipe. The amount of water was then slowly increased i f necessary until the mixture reached the specified slump range. In some cases, the maximum water content specified by the recipe would not yield an adequate slump and the mixture would be too stiff. In these instances, a compromise between water content and acceptable slump range was made in order to obtain an adequate mixture. 38 Paste mixing and moulding was carried out according to A S T M specifications C192-90a and C470 for concrete mixtures. Key elements of these specifications are as follows: 1) M i x i n g of paste (Figure 3.4 a, b): • A plastic, non reactive container was used to mix the paste components, • The batch size of the backfill mixture was larger than the volume of the slump cone, • A l l constituents were dry and well mixed prior to water addition, • The batch was hand mixed for 20 minutes until homogeneous, following which the slump test was carried out, • The water content was adjusted i f necessary and the batch was re-mixed for 10-15 minutes following which a second slump measurement was taken, • Slump measurements were not repeated more than twice per batch, • The paste material used for the slump test was put back in the mixing container and used to f i l l the moulds. 39 n #o V* ft s- es a i- PH a> " a S « C/3 so C3 PM « be 3 u E 33 a SS o U s- a> cs I * • • " -4-> T3 <5 c s U s s. 5 "53 T3 t?; >• SI) 00 m o 'do C N 2 0 0 T3 o 00 ON o I T ) s-s o o o C N r i C N o 00 00 oo T3 C N C N in oo a ss M r i C N in oc fN 00 0 0 T3 o O C N C N C N MS O 00 a « .9 H e 00 - f r i r i T3 O c j z.• u r«". -y. ^iiiiiii 'I, ^ t-- 0 0 T3 a o 00 C N O-, T 3 o o o +-» oo « op 13 •S £ o o 1 o >> a 0 0 T3 <+H O O fl "S ^ 5 J •S t; to O oo o f- ( N m St H) - ! ^ . & 3 <U -P g <D O jS p<̂  < 0 0 3 2) M o u l d material and mould filling (Figure 3.5): • N o n reactive, non-absorptive, watertight, single use cylindrical polyethylene moulds were used having an internal diameter equal to Vi the vertical height (5 cm diameter x 10 cm height), • The filling sequence was dictated by the size and shape of the mould: • fill ing was carried out in 2 layers of 5 cm each, • each layer was given 25 strokes with a rounded end glass rod to evacuate air bubbles, • a slight tap on bottom of mould was given after rodding. 3) Curing of samples (Figure 3.6): • Samples were covered with a plastic sheet to prevent rapid drying of the surface. Wet sponges were kept under the plastic sheet as sources of humidity, • Curing was carried out in a temperature and humidity monitored environment with minimised air currents. The average curing temperature was 20°C ±2 °C and average humidity was 80% ±10%. Concrete samples, for which the above specifications were designed, possess considerably more binder or cement (commonly around 30 % binder) than paste material made by the mining industry (commonly less than 8% binder). A s a consequence, a deviation to specification CI92 had to be made where the paste samples were not de- moulded after 24 hours because of the lack of strength of some samples. A l l samples were de-moulded after a humid cure period of 14 days. 41 Figure 3.4: a. Mixing Paste Sample b. Standard Slump Test Figure 3.5 Rodding Paste into Mould Figure 3.6 Curing of Paste Samples 3.2.1.3 Pre-Leaching Sample Preparation Cured samples were wet-cut into two pieces using a thoroughly cleaned and degreased circular saw (diamond saw). The pieces consisted of a cylinder of 10 cm length for compressive strength test and a separate piece of 2 to 3 cm thick (termed a puck) polished on one side intended for mineralogical observations. The sample surfaces in contact with the mould were lightly sanded to remove the thin oxidation coating that had started to 42 form upon curing for some samples. Both pieces of each sample were oven dried at 40°C for 48 hours, weighed and measured before placement in the leaching cells. 3.2.2 Leaching Apparatus and Leaching Cycles The leach solutions and leach cycles were devised to simulate the different modes of exposure of a cemented paste backfill, as shown in Figure 1.1 of Chapter 1. Table 3.4 explains how these environments were simulated in the laboratory to recreate the various leaching scenarios. Table 3.4 Paste Leaching Scenarios Mine Setting Mode of Leaching Leaching Solution Leaching Frequency Sub-arial backfill or in underground mine above the water table Alternating dry (ambient air) and flooded cycles Simulated rain water: distilled water, p H adjusted to 5.5 with C 0 2 1 week cycle: 24 hours flooded then drained and let open to ambient air for 6 days Flooded backfill (below the water table), absence of A R D in mine water Constantly flooded cycles, 2 week water circulation period Simulated rain water: distilled water, p H adjusted to 5.5 with C 0 2 1 week cycle: Vi cell volume taken out for analyses and replaced with fresh solution Flooded backfill in contact with acidic groundwater produced in the mine Constantly flooded cycles, bi-weekly solution replacement Fe 2(S04)3 solution, 0.005 M or (1 .5g/ lS0 4 2 - ) p H 2.4 - 2.6 Vi week cycle: replacement of entire cell solution The leaching cells consisted of 8.5 cm diameter, 15 cm tall, clear acrylic cylinders equipped with an evacuation valve at the bottom and removable cap and valve at the top (Figure 3.7). Some additional cells were made at U A S L P having slightly different dimensions (5 cm diameter, 20 cm tall) for the lack of similar size material but with an 43 equal volume as the cells made at U B C . Figure 3.8 shows the laboratory set-up for the cemented paste f i l l leaching study. Three cells were prepared for each sample and leaching environment so that one sample could be removed after each designated leaching period for destructive analyses. Three (3) samples were prepared and leached simultaneously in a ferric sulfate solution, 3 in a flooded water environment and 3 in the alternating air-water environment for each mine. After every leaching period (5, 10 and 20 weeks) 1 leaching cell was dismantled for analysis of the solid phase. In each of the first five weeks, leachate solutions of the three leach cells characterizing the same environment were combined to form one large sample for analysis. Between the 5 t h and 10 t h week, leachate samples were a composite of the two remaining cells of the same environment and leachate samples between the 10 t h and 20 t h week originated from a single leaching cell. Figure 3.7 Leaching Cells Figure 3.8 Lab Set up, Paste Leaching Study 3.2.3 Post-Leaching Sample Preparation After each leaching period, the retrieved samples - both cylinder and puck - were measured, weighed and oven dried at 50°C for approximately 72 hours or to a constant weight. The difference between the wet and dry weights was noted for later analysis of 44 water absorption by the paste sample in the different leaching environments. The dry samples were placed into individual bags of 2-mil polyethylene, properly closed to exclude contact with external moisture. The cylinders were brought to the Soi l Mechanics laboratory of the Instituto Politecnico National (IPN) of Mexico City for unconfined compressive strength (ucs) analyses using a Losenhausenwerk hydraulic compressor. The leached paste pucks were very friable and fragile. Consequently, a sample preparation procedure was devised to minimize handling of the sample and maximize the subsequent quality of S E M imaging. The pucks were cut transversally with a fine tooth metal saw and sanded down using 150, 220 then 1500 grit silica sand paper to obtain a flat surface. In the case where pucks were too thin or inappropriate for mineralogical observations, slices of the cylinders were cut out and prepared in the same way. Excess dust and loose particles were lifted from the surface of the cut surface with adhesive tape. The sample was then fixed to aluminium foil with double-sided carbon tape exposing only the flattened, particle-free surface (Figure 3.9 a, b, c). A l l prepared puck surfaces were photo scanned then carbon coated for S E M observations. Some samples required 2 or 3 coatings of carbon because of their porous nature. Samples were kept in a dessicator between preparation and observation steps. 45 Figures 3.9 a, b, c: Paste Sample Preparation Steps for Mineralogical Observations T h e different o x i d a t i o n zones o f the leached samples were identi f ied b y c o l o u r and/or textural changes v i s i b l e o n the puck sections and/or cyl inders . A l l the v i s i b l y different layers, as w e l l as the core o f each sample, were careful ly scraped o f f the 20 week- leached samples (cyl inders) and col lected for analyses. The 5 and 10 week- leached samples were not used for this purpose, as the o x i d a t i o n layers were i n general too t h i n to be successful ly separated. In cases where no distinct o x i d a t i o n layers were apparent, a 1 m m t h i c k surf ic ia l layer was scraped o f f for analyses. M i n e r a l o g i c a l observations were carried out o n the prepared pucks us ing a P h i l i p s X L 3 0 scanning electron m i c r o s c o p e ( S E M ) . C h e m i c a l microanalyses were per formed u s i n g an I M I X energy dispersive x-ray ( E D X ) instrument at U B C . E l e m e n t a l spectra o f the cement phases and precipitates were taken, thereby p r o v i d i n g qualitative i n f o r m a t i o n about the general c o m p o s i t i o n o f these phases. N o quantitative in format ion c o u l d be extracted f r o m the E D X spectra as the phases o f interest were too s m a l l and scarce to p r o v i d e accurate quantitative data w i t h the analyt ical procedures used. 46 3.2.4 Chemical Analyses of the Paste Experiment Routine chemical analyses were performed on leaching solutions (water and simulated A R D or ferric sulfate solutions) and on the leachate produced by the paste backfill leaching cells. Parameters analyzed for the paste backfill study are listed in Tables 3.1 and 3.2 in section 3.1.3. Leachate p H , redox potential, conductivity and ferrous iron were analysed upon collection at U A S L P . Other leachate parameters and the hole-rock analyses of tailings, cement and slag were carried out at the analytical laboratory of the Centro de Investigation y Desarollo Tecnologico (CIDT) of Pefioles in Monterrey, Mexico . The leachate solutions were acidified with concentrated HC1 to a p H <2 and stored at 4°C between monthly shipments to Pefioles via courier. Solid phase analyses of the weathered backfill samples were carried out according to the methods listed in Table 3.5. The ferric sulfate-leached Brunswick samples were sent to Chem-Met laboratory of Vancouver for sulfur species analyses. This lab was not equipped to analyse the requested suite of metals and major ions, hence the unused portions of these samples were sent to Chemex laboratory of North Vancouver for analyses of metals and major ions listed in Table 3.5. A l l the other weathered paste samples were sent directly to Chemex laboratory for sulfur species determination as well as metals and major ions analyses (when sufficient sample mass was available for analyses). 47 Table 3.5 Solids Phase Analyses for the Paste Experiments Sample Analvses Vleihodoloi'.v .md Instrument Unleached solids (tailings, O P C and slag) FcO. C O i . S 0 4 , CaO, K 2 0 , M g O , M n O , N a 2 0 , S i 0 2 , A 1 2 0 3 A g , A s , B i , C , Cd , Cu, Cr, Fe, N i , Pb, S, Se, Te, Zn . Pefioles CIDT: Aqua-regia digestion A A analysis (Perkin-Elmer 5000) Leached paste Total Sulfur Chem Met: A c i d leach (oxidation) and gravimetric determination Chemex: Leco 420 combustion furnace with Infra Red detector Sulfate Chem Met and Chemex: Dilute (10%) HC1 leach and gravimetric determination A l , Ca, M g , Fe, Cu , Z n Chemex: H F acid digestion and ICP- A E S 1 analysis (Perkin-Elmer Optima) Pb Chemex: H F acid digestion and A A analysis (Varian 220) ICP-AES: Inductively coupled plasma atomic emission spectrocopy 3.2.4.1 Paste pH and Acid-Base Accounting Measurements Paste p H measurements were taken on dry tailings and unleached paste backfill for the purpose of comparison and to verify the presence of acidity or available buffering minerals of the samples. Acid-base accounting ( A B A ) tests were carried out at U A S L P to monitor the evolution of the paste buffering capacity as leaching proceeded. The Modified Sobek Method of A B A analysis was followed on all tailings and unleached paste samples as well as on the different oxidation layers of all 20-week leached paste samples. 48 4 R E S U L T S - P Y R I T E R E A C T I V I T Y S T U D Y 4.1 MINERALOGY, CHEMISTRY AND STOICHIOMETRY OF UNLEACHED PYRITES A l l pyrite samples studied possessed different chemical compositions, stoichiometries and mineralogical associations. Solid phase chemistry of the pyrite samples, normative mineralogy of the samples and average stoichiometry of pyrite crystals in the samples are summarized in Tables 4.1, 4.2 and 4.3 respectively. S E M micrographs showing sample mineralogy are shown in Plates 4.1 to 4.6. Mineralogical observations of unleached polished sections showed that the Brunswick and Zimapan samples possessed the greatest amount of sphalerite, galena and arsenopyrite impurities, mainly occurring as small inclusions within pyrite (Plates 4.1 and 4.2). The Brunswick pyrite sample consisted of pyrite concentrate whose particles were ground small enough to expose the separate mineral phases. Louvicourt-2 and Tizapa pyrites contained sphalerite, chalcopyrite, galena as well as some tetrahedrite and arsenopyrite impurities occurring mainly as separate phases in the interstices of pyrite grains (Plates 4.3 and 4.4). Some of these mineral impurities 1 also occurred as inclusions within pyrite crystals. Huckleberry and Louvicourt-1 samples contained the least amount of mineral impurities. Huckleberry pyrite contained no visible inclusions whereas 1 Mineral impurities refers to the presence of non-pyrite sulfide minerals in the pyrite samples. 49 Louvicourt-1 contained very few and dispersed inclusions of sphalerite, galena, chalcopyrite and arsenopyrite (Plates 4.5 and 4.6). Stoichiometric microanalyses of uncrushed pyrite crystals (polished sections) indicated that pyrites in all samples possessed some nickel (Ni), copper (Cu) and arsenic (As) in their crystal lattice. The amount of lattice impurities varied considerably within a single grain and also between the various grains of one sample, as reflected by the high standard deviations of the measurements (Table 4.3). Cross-sectional microanalyses of individual pyrite grains did not reveal any texture differences associated with chemical variations in any of the pyrites. 50 .Q (0 K- "S3 .s g 5 (0 c 5 *• Br  c > CM ISpNllit o ' > o _ l ' l l i l l Q . (0 ! N c 1 oo Q_ CO E _ N ^ f 0 0 1 o I I j CD 1 i £ o £ si 1 CD — C CD CO w CD _ C CO «= "O o © n CD — c CO w CD _ C CO «= -o O CD 91 O w "O <D — £ iS CO w CD _ C CO g" ^ 3s s •3 3 O w •o c (D E .c O 5 3 5? <o <D (0 to (0 CO " CD JZ 3 CD T3 CD JZ o CO — CD 3 £ CD CO Q . O CL co O T— V CD 3 O) c CO 0 . CD _ 3 CO TO O c CO Py  QL co" O) Py  CO O CO -s v P OO CM V T ~ I o < CD _ 3 CO TO O an  >» CL O) Q . . o CO ~s e£ CO CD CO V T Q_ 8 Q. Q. co o V P , _ CM < co" O CD 3 TO c CO TO a. co CL < CO CO o i n CL O a. co CD 3 a> c CO TO 00 I CO CD £ o CO >; a. CD O) Q. O CL < c£ co 0 3 TO C CO TO >> o 2 cu g E 51 Table 4.2 Chemical Analysis of Pyrite Samples Species unit Huckleberry Zimapan Tizapa Louv 1 Louv 2 Brunswick »iRy,.ci3nceotrate)j S wt% 47.4 47.3 50.4 40.4 47.4 58.7 S as S04 wt% 0.79 0.41 0.20 0.40 0.20 1.89 s2- wt% 46.7 46.9 50.2 40.0 47.2 56.8 Fe wt% 45.2 42.5 45.4 39.4 37.7 38.2 Zn wt% 0.01 0.70 0.53 0.004 0.98 2.60 Pb wt% 0.01 1.39 0.12 0.04 0.04 8.00 Cu wt% 0.02 0.02 0.80 0.04 0.02 0.32 Ni wt% <0.005 <0.005 <0.005 <0.005 <0.005 0.006 As wt% <0.02 1.18 0.19 0.05 0.06 0.28 Bi wt% 0.024 0.015 0.031 0.018 0.016 0.030 Si02 wt% 0.50 2.56 0.30 11.2 8.70 0.95 C03 wt% 0.14 0.04 0.01 0.43 0.10 0.74 Al203 wt% 0.09 0.17 0.11 3.47 6.50 2.12 Na20 wt% 0.04 0.07 0.07 0.07 0.07 0.01 CaO wt% 0.45 0.50 0.45 0.55 0.57 0.20 MgO wt% 0.05 0.01 0.05 0.71 0.31 0.11 MnO wt% 0.13 0.10 0.15 0.13 0.08 0.02 K20 wt% 0.02 0.02 0.02 0.02 0.02 0.01 C wt% 0.06 0.26 0.09 0.09 0.08 0.9 Ag g/Tor <2 46 80 10 8 178 Sum 94.13 96.80 98.64 96.60 102.63 113.197 Note: For all samples: Cd <0.005%, Cr <0.02%, Se<0.001 %, Te<0.002% Loss on ignition (LOI) not reported 52 c o m i d CO d to in d cn o T— o m o CO d CM 5 d <£> d o d o CM CO to CM CO CO o o CO lO d d d o _ l CD d CM 5 CO d CO d o T— d lO 00 CO iri o o d d d (0 Q_ (0 N d d T— d CO T— d CO CO CO s in o o d d d c (0 c l CO £ N in d d o o T— d CO CO LO in CD d CO o o IT) rr d d d CD ja CD u 3 C CD E LU m d m d CO o CO CO CO CO LO o o CO LO d d d 53 SEM Micrographs of Unleached Pyrites from Polished Sections ( B a c k Scatter v i e w ) Plate 4.1 Zimapan Pyrite Plate 4.2 Brunswick Pyrite Plate 4.5 Huckleberry Pyrite Plate 4.6 Louvicourt-1 Pyrite L e g e n d : P y = pyr i te ; S p = sphalerite; G a = galena; C p = chalcopyrite; A p = arsenopyrite; T e = tetrahydrite; G n = gangue 54 4.2 ELECTROCHEMISTRY OF THE UNLEACHED PYRITE SAMPLES Stable open current potentials of each pyrite sample were measured before beginning each voltamperometric test. Ocp values are presented in Table 4.4. A correlation was found to exist between ocp values and their content of mineral impurities. Unleached Huckleberry and Louvicourt-1 samples exhibited similar, relatively low ocp values, suggestive of a relatively reactive sample compared to Louvicourt-2, Tizapa and Zimapan. The latter 3 pyrite samples possessed more mineral impurities and were characterized by higher ocp values, indicative of a lower mineral reactivity. The low ocp of the Brunswick sample may be imparted by the high proportions of sphalerite and galena in that sample, which may have overwritten the ocp of the pyrite in the initial leaching cycles. The evolution of Brunswick sample reactivity with leaching time (Section 4.5.5) supports this possibility. Table 4.4 Open Current Potentials of Pyrite Samples (Volts, SSE) Sample I'nlcachcd 4 weeks 10 weeks 20 weeks Huckleberry (water-leached) -0.27 -0.22 -0.24 -0.22 Huckleberry (column leached) -0.27 -0.22 -0.22 -0.20 Louvicourt-1 -0.27 -0.21 -0.20 -0.22 Louvicourt-2 -0.19 -0.21 -0.13 -0.22 Tizapa -0.18 -0.20 -0.15 -0.18 Zimapan -0.18 -0.18 -0.16 -0.20 Brunswick -0.29 -0.26 -0.18 -0.13 55 Figures 4.1 and 4.2 compare the voltamperometric response of pyrites in the unleached samples. The arrows on Figure 4.1 indicate scan directions. For the sake of graph clarity, scan direction arrows have not been added to the other voltamograms. Each sample had a distinct electrochemical response, indicative of the reactivity differences between them. Reactivity was measured by the potential at which oxidation of the sample reached a current of 2.0 uA. A l l anodic peaks showed a catalytic behaviour, where for a given potential, a lower current was generated during the forward scan (anodic current) than in the reverse scan (cathodic current). Although the cathodic response of the leached mineral can reveal interesting information on the properties of precipitates covering mineral surfaces, this study focused on the anodic or oxidative behaviour of pyrite to document the reactivity of the pyrite surface after increasing periods of oxidative leaching. Figures 4.1 and 4.2 indicate that Huckleberry pyrite oxidized at the lowest potential and offered the least resistance as potential was increased. It was concluded to be the most reactive of all unleached samples. Following, in order of decreasing reactivity, were Louvicourt-1, Louvicourt-2 and Tizapa. The close up view of the points of initial current release (Figure 4.2) shows, for Brunswick, and slightly less pronounced for Zimapan, an initial current release prior to the typical curve of pyrite oxidation. A n electrochemical study of the oxidation of galena and sphalerite in the same voltametric cell environment (Figures 4.3 and 4.4 respectively) suggested that the lower potential at which the oxidation of the unleached sample appeared to be initiated was more likely attributable to the oxidation of galena. The oxidation of pyrite appeared to resume around 0.420 volts in 56 both the Brunswick and Zimapan curves as indicated by the sudden increase in current. The relative reactivities of the unleached Zimapan and Brunswick pyrites are not possible to establish because of the presence of galena. Once pyrite oxidation was initialized however, Figure 4.1 shows that Brunswick pyrite was the most resistant to oxidation. Humidity cell studies of fresh Brunswick tailings and mineralogical studies of old tailings carried out by Noranda Technology Centre (1998f) found Brunswick pyrite to have a relatively low reactivity and to oxidize at a lower rate than what is commonly reported for pyrite. The varying amount of mineral impurities and distinct chemistry of the pyrite samples, as well as the variable stoichiometric compositions of the pyrites within one sample did not allow for a comparison of pure FeS2 compounds. The reactivity of the pyrite in each sample could not be directly measured because of these intrinsic differences impossible to eliminate from the sample. It was possible, however, through cyclic voltamperometry, to document the effects of impurities on the reactivity of pyrite as leaching progressed. 57  u 13 M D. VI a fi 73 O CO a> «« fl o a a o la <u a. a •** > a © o IX! O o o fl W Ph U o o oo o o <=> E CO _ c o o oo o o CM o CM (vn) juaxino CU 13 j= o-CZ5 TT cu s- fl ON i n fl 13 O V S- s (yn) juejjno 4.3 EVOLUTION OF LEACHATE CHEMISTRY The chemical analyses of the leachate for all samples are presented in Appendix I (Table 1-1). The evolution of leachate composition is plotted as time-series graphs for each of the following elements: p H , conductivity, redox potential, sulfate, total iron, lead, zinc, copper and arsenic (Figures 4.5 to 4.12). The artificial rain water solution (referred to as "Disti l led Water") as well as the Huckleberry column solution used to leach a separate Huckleberry pyrite sample (referred to Hk-column) are also plotted as references. The solution-leached and water-leached Huckleberry samples are identified as Hk-1 and H k - w respectively. After a rapid decline in the first leaching cycles, p H stabilized for all pyrites (Figure 4.3). Redox values (Figure 4.4) remained in the 400-500 mvolt range (SHE) throughout the leaching cycles indicating that oxidizing conditions prevailed, as expected. Conductivity, sulfate and dissolved iron were markedly higher for both Huckleberry pyrites (water and solution-leached) compared to the other pyrites (Figures 4.5, 4.6 and 4.7) indicating that Huckleberry pyrites were being oxidized at a considerably greater rate than the other pyrites throughout the leaching experiment. Dissolved metal concentrations varied considerably from one sample to another. Leachate zinc concentrations (Figure 4.8) were directly proportional to the solid phase zinc content of the unleached sample, occurring as sphalerite impurities. Samples with the highest sphalerite content such as Louvicourt-2, Tizapa, Zimapan and especially Brunswick all leached relatively high concentrations of dissolved zinc. Tizapa and 60 Zimapan leachates showed a decline in dissolved zinc concentration after 12 cycles (6 weeks) of leaching whereas Brunswick and Louvicourt-2 zinc concentrations decreased much more slowly. This indicates that sphalerite in the Brunswick and Louvicourt-2 samples was largely exposed to the leaching solution whereas Tizapa and Zimapan sphalerite was partially locked inside pyrite grains. Dissolved lead concentrations were also proportional to the initial solid phase lead (galena) content (Figure 4.9). The Brunswick pyrite leached out highest concentrations o f lead but only in the first 12 cycles of leaching, after which aqueous lead concentrations reached steady state at approximately 10 mg/1, similar to Zimapan and Tizapa levels. S E M observations indicated that lead tended to precipitate as anglesite (PbS0 4 ) . Copper and arsenic concentrations were relatively low in all leachates (Figures 4.10 and 4.11) reflecting the low chalcopyrite and arsenopyrite content of the samples. Tizapa's aqueous copper concentration corresponds to its slightly higher chalcopyrite content observed in the polished section of the unleached sample. 10 2 0 + . .' * • * + * * + + + + + ,* + + 1 3 5 7 9 1 1 13 15 17 19 21 2 3 2 5 27 2 9 31 33 35 37 3 9 n o . o f c y c l e s Figure 4.5 Pyrite Leachate pH 61  -Q 40 E —D—Hk-I % Pb: • Hk-w (0.007%) — • — L o u 1 (0.04%) — • — L o u 2 (0.035%) • Tlz (0.12%) — • — Z i m (1.39%) — 1 — B w k (8.0%) 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 no. of cycles Figure 4.11 Pyrite Leachate Lead Concentrations 63 3 0 2 5 £T 20 S 1 5 3 O 1C ram* 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 no. of cycles Figure 4.12 Pyrite Leachate Copper Concentrations 3 0 2 5 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 no. of cycles Figure 4.13 Pyrite Leachate Arsenic Concentration 64 4.4 EVOLUTION OF PRECIPITATE COATINGS DURING LEACHING 4.4.1 Huckleberry pyrite S E M observations showed differences in precipitation products between the water and waste rock solution-leached Huckleberry samples as early as the 4 t h week of leaching. The water-leached pyrites conserved relatively sharp grain edges with very few visible coatings (Plate 4.7) or corrosion pits. The solution-leached pyrite surfaces, on the other hand, showed the presence of amorphous precipitates after 4 weeks of leaching, increasing in thickness and extent with advancing leaching cycles (Plate 4.8 and 4.9). Oxidation pits were also more extensive in the solution-leached pyrites (Plate 4.10). Although precipitates were visible under S E M , infrared spectroscopy did not succeed in identifying the secondary phases, as their amount was insufficient (less than 5%). Consequently, no data are available as to the exact composition of these precipitates. A review of the literature on surface oxidation products formed on pyrites suggests, however, that the precipitates observed under S E M may be amorphous iron oxyhydroxides similar to goethite (a-FeO(OH)) (Jambor, 1994; Bigham, 1994; Kwong , 1993). Amorphous, hydrated iron sulfates can also be present on pyrite surfaces but are less common than iron oxyhydroxides in young mine wastes according to these authors. The thin, amorphous iron precipitates observed in this study w i l l therefore loosely be referred to as iron oxyhydroxides or F e O O H . 65 4.4.2 Louvicourt Pyrites Mineralogically similar Huckleberry and Louvicourt-1 pyrite samples yielded similar precipitate coatings upon leaching. Plates 4.11, 4.12 and 4.13 show a precipitate cover similar in form and abundance to Huckleberry solution-leached pyrites. Very few corrosion pits were visible after 20 weeks of leaching. The Louvicourt-2 pyrites were considerably different from Louvicourt-1 in both the occurrence of precipitates and the aspect of oxidized pyrites. Pyrite surfaces remained relatively free of precipitates until the 20 t h week, at which point visible amorphous iron precipitates increased in abundance. After 20 weeks of leaching, however, the amount of iron precipitates covering pyrite surfaces remained much lower than the Louvicourt-1 pyrites (Plates 4.14 and 4.15). Corrosion pits were abundant after 20 weeks of leaching, contrary to Louvicourt-1 or Huckleberry pyrites (Plate 4.16). Sphalerite was observed in Louvicourt-2 sample and, where both sphalerite and pyrite were present in the same grain, sphalerite was extensively more riled and pitted than pyrite after 10 and 20 weeks of leaching (Plate 4.17). Sphalerite was being preferentially oxidized over pyrite. 4.4.3 Tizapa Pyrite Mineralogically similar Tizapa and Louvicourt-2 pyrites also showed similarities in precipitate occurrence and aspect of oxidized mineral phases. The surface of pyrite grains remained relatively free of precipitates throughout the 20 weeks of leaching and, in general, very few corrosion pits were observed (Plate 4.18). Sphalerite was also considerably more corroded than pyrite (Plate 4.19). In addition, sphalerite became more 66 extensively covered with precipitates than pyrite after 20 weeks of leaching (Plate 4.20). Galena in the unleached Tizapa sample appeared as mineral impurities within the pyrite grains and exposed to the environment. The exposed galena was readily available to oxidize or dissolve. After 10 and 20 weeks of leaching, no galena was detected but anglesite (PbSC^) was observed principally on altered galena, suggesting a direct oxidation (replacement) of galena (Plate 4.21). The precipitation of lead sulfate as anglesite explains the absence of lead in the leachate. 4.4.4 Zimapan Pyrite The Zimapan pyrites showed a similar abundance of amorphous iron precipitates to Louvicourt-1 pyrites. After 20 weeks of leaching, these precipitates covered a considerable portion of most pyrite grains (Plate 4.22). Oxidation pits were also observed as early as the 4 t h week of leaching, attesting to the highly reactive character of Zimapan pyrite (Plate 4.23). Reactivity may be accentuated, in part, by the occurrence of crystal lattice impurities of A s known to act as an electron donor when present in pyrite, thereby promoting pyrite oxidation. N o galena was observed after the 4 t h week of leaching, replaced by abundant euhedral crystals of anglesite occurring throughout the sample (Plate 4.24). The distribution of anglesite suggests that nucleation formed from the leachate solution rather than by direct replacement of galena phases, implying a previous dissolution or oxidation of galena. The abundance of anglesite appeared to remain constant with advancing leaching cycles. 67 4.4.5 Brunswick Pyrite Relatively few iron hydroxide precipitates covered the surfaces of pyrite grains after 20 weeks of leaching (Plate 4.25). Few pyrite corrosion pits were observed in the 4 t h week of leaching but their occurrence increased with advancing leaching cycles (Plate 4.26). th Precipitates of anglesite were observed in the 4 week of leaching throughout the sample along with galena (Plate 4.27). The occurrence of anglesite suggests that, similar to the Zimapan sample, it was precipitated from solution, from previously oxidized galena. Galena was present in the sample after 10 weeks, but absent after 20 weeks of leaching. Sphalerite was not observed in any of the leached samples. 68 Huckleberry Pyrites Plate 4.9 Solution-Leached, 20 weeks Plate 4.10 Solution-Leached, 20 weeks Plates 4.7 and 4.8 showing water-leached pyrites possess a smaller amount of amorphous iron hydroxide precipitates of than the solution-leached pyrites shown on Plates 4.9 and 4.10. 69 Louvicourt-1 Pyrite Plate 4.11 4-week leached Plate 4.12 20-week leached Plate 4.13 20-week leached Surface-covering iron hydroxide precipitates on Louvicourt-1 pyrites grew extensive with advancing leaching cycles. 70 Louvicourt-2 Pyrites Plate 4.14 Pyrite surface at 10 weeks Plate 4.15 Pyrite surface at 20 weeks Plate 4.16 Pyrite at 10 weeks Plate 4.17 20-week leached Little difference was observed between pyrite surface precipitates after 10 and 20 weeks of leaching (Plates 4.14 and 4.15 respectively). Plate 4.16 shows corrosion pits on pyrite grain free of visible mineral impurities. Plate 4.17 shows preferential sphalerite oxidation over pyrite. 71 Tizapa Pyrite Plate 4.18 Pyrite surface at 20 weeks Plate 4.19 Oxidized sphalerite, 20 weeks Plate 4.20 Precipitate cover, 20 weeks Plate 4.21 Oxidized galena, 20 weeks After 20 weeks of leaching, pyrite surfaces show very little oxidation and surface precipitates of iron hydroxide (Plate 4.18). Plate 4.19 shows the preferential oxidation of sphalerite over pyrite. Plate 4.20 shows more extensive cover of precipitates over sphalerite and tetrahedrite than pyrite. Plate 4.21 shows the direct oxidation of galena impurity characterized by surface precipitates of anglesite (Ang). 72 Zimapan Pyrite Plate 4.22 Pyrite surface at 20 weeks Plate 4.23 Corrosion pits at 4 weeks Plate 4.24 Anglesite precipitates, 20 weeks Plate 4.22 shows the surface of pyrite with extensive cover of amorphous iron hydroxide precipitates. Corrosion pits were abundant on pyrite grains after 4 weeks of leaching (Plate 4.23). Plate 4.24 shows anglesite (Ang) precipitates on the surface of pyrite. 73 Brunswick Pyrite Plate 4.27 Anglesite and galena, 4 weeks Relatively small amounts of amorphous iron hydroxide precipitates on pyrite surfaces after 20 weeks of leaching (Plate 4.25). Corrosion pits were not commonly observed before 20 weeks o f leaching (Plate 4.26). Anglesite (Ang) and galena (Ga) occurred together on pyrite surfaces after 4 weeks of leaching (Plate 4.27). No galena was observed in subsequent cycles. 74 4.5 EVOLUTION OF PYRITE REACTIVITY The evolution of the reactivity of pyrite with leaching time is demonstrated in cyclic voltammograms of each sample. In all graphs, each curve represents the voltamperometric response of a particular pyrite sample at a given period (unleached and after 4, 10 and 20 weeks of leaching). The navy blue curve represents the response of the unleached sample and the purple, green and red curves represent the responses of the same pyrite after 4, 10 and 20 weeks of leaching respectively. 4.5.1 Huckleberry pyrites Figure 4.14 presents the voltamperometric responses of Huckleberry pyrite leached in water. The reactivity of this pyrite decreased over the first 10 weeks, following which a slight gain in reactivity was observed. A net loss of reactivity occurred over the 20 weeks of leaching, characterized by a total positive displacement of the initialization point of 0.03 volts (measured at 2.0 uA). The solution-leached pyrite also showed a decreasing reactivity with leaching time, including a similar gain of reactivity at the 20- week mark (Figure 4.15). A net gain of 0.07 volts occurred over the 20-week leaching period with respect to the unleached pyrite. The mineralogical observations, leachate chemistry results and electrochemical characterizations suggest that the Huckleberry pyrite is considerably reactive to oxidation 75 but that a passivation 2 layer is likely formed on pyrite surfaces, effectively decreasing the reactivity of pyrite after only 4 weeks of leaching. The passivating layer derived from the waste-rock column solution appeared to be more effectively decreasing pyrite reactivity than that derived from water leaching. 4.5.2 Louvicourt Pyrites A s with sample mineralogy and evolution of precipitate coatings, the Louvicourt-1 pyrite showed electrochemical responses after each leaching period, that were very similar to Huckleberry solution-leached pyrites (Figure 4.16). The loss of reactivity was slightly lower in the Louvicourt-1 sample than in the Huckleberry solution-leached sample, with a net positive displacement of only 0.055 volts compared to 0.07 volts for Huckleberry. Contrary to Louvicourt-1, Louvicourt-2 pyrite became more reactive after 4 weeks of leaching, with a negative displacement of the point of initial oxidation of 0.05 volts (Figure 4.17). In the 10 t h and 20 t h week, however, pyrite reactivity decreased, resulting in a net positive advancement of 0.04 volts over the 20 weeks of leaching with respect to the unleached pyrite. The gain of reactivity within the first 4 weeks of leaching coincide with the preferential oxidation and loss of sphalerite over pyrite in the sample observed under S E M and supported by the leachate chemistry data. The subsequent passivation could be attributable to the formation of surficial, amorphous iron precipitates observed 2 The terms 'passivation' or 'passivated' in this study refer to a relative decrease in reactivity of the mineral measured as a positive displacement of the point of initialization of mineral oxidation, observed by cyclic voltamperometry. This term is not used to describe the mechanisms or the products responsible for the observed loss of reactivity. 76 in appreciable quantity in the 20-week leached sample. These precipitates could have effectively passivated pyrite surfaces. 4.5.3 Tizapa pyrite Tizapa pyrite reactivity evolved in a similar way to that of Louvicourt-2 pyrite, although at a smaller scale. Figure 4.18 shows an initial increase in reactivity in the 4 t h week of leaching, followed by a reactivity decrease which remained until the 20 t h week of leaching. A net positive displacement of 0.05 volts occurred after 20 weeks of leaching with respect to the unleached pyrites. The higher resistance to oxidation observed in the 10 t h week compared to the 20 t h week, suggests that the precipitate coatings formed at an early stage (10 weeks) were different than those formed after 20 weeks of leaching. 4.5.4 Zimapan pyrite The Zimapan voltammogram showed an increase in the width of the anodic peak with leaching time (Figure 4.19). With advancing leaching cycles, the reverse scan of the anodic peak released a consistently higher current than the forward scan, measured at a given potential. The forward scan showed a net advancement of the initialization point of pyrite oxidation of 0.1 volts in 20 weeks of leaching. The reverse scan showed an increasing efficiency to oxidize the pyrite surfaces after initialization. A possible explanation for this behaviour is the observed increased abundance of precipitates combined with the increased surface area created by corrosion pits as leaching progressed. The thicker precipitate coating possibly offered increasing resistance to oxidation, but once the potential became sufficiently high to break this barrier, more 77 current could be generated from the same grains because of the increased surface area created by the corrosion pits. Although pit density remained constant with advancing leaching cycles, pits have a tendency to deepen inside the crystal rather than to enlarge at the surface of the grain (Mustin, 1992). Consequently, the actual surface area of a pitted grain could be increased without a notable increase in the number or width of the pits. The effect of oxidation pits on the reactivity of pyrite was not verified in this study but merits further investigation. The presence of galena in the unleached sample was observed in Figure 4.2, Section 4.2. A close-up view of the 4, 10 and 20-week anodic scan curves in Figure 4.20 shows that galena was no longer present in the sample after the 4 t h week of leaching. S E M observations and leachate chemistry indicated that most of the available galena was oxidized in the first 4 weeks of leaching where lead was reprecipitated as anglesite throughout the sample. 4.5.5 Brunswick The forward scan curves of the Brunswick pyrite voltammograms showed an apparent increase in reactivity with leaching time up to the 10 t h week of leaching, followed by relative decrease in reactivity in week 20 (Figure 4.21). The presence of galena in the unleached, 4-week and 10-week samples, however, concealed the exact point (or voltage) at which pyrite started to oxidize. It was therefore impossible to calculate the amount of curve displacement. The apparent gain in reactivity observed until week 10 was similar to the Louvicourt-2 response and also most likely attributable to the presence of 78 sphalerite. Although sphalerite was not visible under S E M in the leached samples, solid phase zinc concentration (corresponding to sphalerite) was the highest in the Brunswick sample. Leachate chemistry indicated that large concentrations of zinc were continually being leached form the Brunswick sample whereas iron was not. L ike in the Zimapan and Louvicourt-2 pyrite samples, sphalerite was most likely providing galvanic protection to pyrite at least in the early cycles. With increasing leaching time, the presence of iron hydroxides and anglesite precipitates on pyrite surfaces may have reduced the availability of sphalerite to oxidize in lieu of pyrite. A close-up view of Figure 4.22 indicated that the electrochemical signature of galena was still discernible after 10 weeks of leaching, although very diminished, and was absent form the 20-week voltamogram. Galena along with anglesite was observed under S E M in the 4-week sample, galena was not seen in the 10 and 20-week samples. The reverse scan of the Brunswick voltammogram showed a pattern similar to the Zimapan pyrites where the anodic peak widened with advancing leaching cycles. Oxidation pits were more abundant after 20 weeks of leaching in the Brunswick sample which, again, could possibly explain the increasing generation of current on the reverse scan. 79 (vn) juajjnQ     4.5.6 Evolution of Pyrite Reactivity Figures 4.23 to 4.25 show the voltametric responses of the group of pyrites after 4, 10 and 20 weeks of leaching respectively. The differences in reactivities between pyrites were significantly decreased after 4 weeks of leaching. The opposing processes of increased passivation for the initially reactive pyrites (Huckleberry, Louvicourt-1) and loss of galvanic protection for the initially less reactive pyrites (Louvicourt-2 and Tizapa and possibly Zimapan and Brunswick) appear to have effectively homogenized the electrochemical responses o f the various pyrites. Individual reactivity differences appeared again after 10 weeks of leaching, developing further after 20 weeks as the characteristics of the pyrite surfaces were likely dominated by the presence o f precipitation products and possibly oxidation pits. H u c k - w a t e r H u c k - l e a c h L o u v i c o u r t 1 L o u v i c o u r t 2 T i z a p a Z i m a p a n B r u n s w i c k 200 250 300 350 400 450 Potential (mV) Figure 4.23 All Pyrites, 4-week Leached 85  Table 4.5 shows the relative reactivity of each pyrite at different leaching periods measured at the point of 2.0 u A current release on the voltammograms (on close-up views not presented). Samples exhibiting similar reactivities appear in the same box. Although the reactivity of Huckleberry pyrites continually decreased with advancing leaching cycles, they remained the most reactive of all pyrites after 20 weeks of leaching. Similarly, Tizapa and Louvicourt-2 remained the least reactive. The comparison with Brunswick pyrite reactivity is difficult to establish because of the galena peak effectively hiding the initialization point of pyrite oxidation in the unleached sample and after 4 and 10 weeks of leaching. This study suggested that the reactivity of the pyrite samples studied depend greatly on the occurrence of mineral impurities and, after some period of leaching, on the composition of the precipitate layer. The latter could be dictated by the various mineral phases being oxidized simultaneously with pyrite. Table 4.5 Relative Reactivities of Pyrites t > o CU S-l ' in cd <L> t-i o C unleached Huckleberry Louvicourt-1 Louvicourt-2 Tizapa (Zimapan?) (Brunswick?) 4 weeks 10 weeks 20 weeks Huck. Water, Louvicourt-1, Huck. Column Louvicourt-2, Zimapan, Tizapa, (Brunswick?) Huck. Water Huck. Column Louvicourt-1 Zimapan Tizapa Louvicourt-2 (Brunswick?) Huck. Water Huck. Column, Louvicourt-1 Zimapan Tizapa, Louvicourt-2, Brunswick 87 5 R E S U L T S - C E M E N T E D P A S T E B A C K F I L L S T U D Y 5.1 CHEMICAL ANALYSES OF TAILINGS, BINDER AND PASTE Chemical analyses of the tailings, paste and slag are presented in Table 5.1. The paste mixture compositions were calculated from the chemical data of Table 5.1 and are presented in Table 5.2. Analyses indicate that Tizapa tailings possess the highest concentrations of sulfide, iron, copper and zinc of all other tailings. They have the second highest concentration of lead and arsenic after Brunswick. Brunswick tailings have a composition similar to that of Tizapa, with high concentrations of sulfide, iron, zinc, lead and arsenic. The maximum pyrite content is estimated at - 7 0 % for both Brunswick and Tizapa tailings, assuming all the iron occurs as pyrite. This estimate is slightly higher than values reported by Brunswick Min ing Divis ion Inc. (40 to 65% pyrite) (Noranda Technology Centre, 1998e). The discrepancy probably lies in the 0-5% pyrrhotite content and the iron of the sphalerite not being taken into account in this summary calculation. The Tizapa estimate falls within the values reported by Penoles (Ybarra, 1998). Tizapa tailings also have the lowest concentration of calcium whereas Brunswick has the highest. F I M and Louvicourt tailings have a similar composition, containing approximately one-third the concentration of sulfide and half the iron than Brunswick or Tizapa, giving a maximum pyrite content of 35% for each tailings. This estimate lies within the Noranda Inc. calculations of 16 to 47% pyrite for Louvicourt. F I M and Louvicourt tailings also have low concentrations of zinc, lead copper and arsenic compared to the other tailings. A l l tailings have low concentrations of 88 carbonates, a parameter associated with limestone (CaCCh) or dolomite (CaMg(C03)2), suggesting that all tailings have a low acid neutralization potential. The chemical analyses presented in Table 5.2 indicate that the principal component of O P C is calcium corresponding to the content of tri- and di-calcium silicates (Ca 3 Si05 and Ca2SiC>3 respectively) and tri-calcium aluminate ( C a s A L ^ ) , the principal binding agents of O P C . The granulated blast furnace slag used in the Louvicourt formulation has a relatively low iron content and is therefore favourable for use against sulfate rich waters (Kosmatka et al., 1995). The slag is also rich in calcium and silicon, carbonates and magnesium. 89 Table 5.1 Chemical Analyses of Tailings and Binders Used in Backfi I Mixtures Cement Tiz Stotal % 1.3 1.1 9.2 10.5 37.4 34.3 S asSO, % 0.2 0.2 1.0 0.2 0.4 3.3 s2- % 1.1 0.9 8.2 10.3 37.0 31.0 Fe % 1.2 0.3 16.1 16.7 34.4 30.5 Zn % 0.007 0.002 0.11 0.66 2.10 1.08 Pb % 0.003 0.002 0.03 0.07 0.38 1.08 Cu % 0.057 0.022 0.001 0.001 0.180 0.100 Ni % <0.001 O.001 <0.001 <0.001 <0.005 0.007 As % <0.005 O.005 0.02 0.005 0.31 0.37 Cd % <0.001 O.001 0.007 <0.001 O.001 0.13 S i 0 2 % 18.9 31.7 42.5 34.6 11.2 21.7 co 3 % 1.7 7.7 0.6 2.3 2.3 1.6 A l 2 0 3 % 4.5 6.9 9.6 5.7 2.7 1.1 Na 2 0 % 0.63 0.09 0.44 0.08 0.07 0.14 CaO % 69.0 44.2 2.8 8.8 0.7 14.3 MgO % 1.4 12.7 6.7 2.0 0.7 4.2 MnO % 0.02 0.74 0.20 1.44 0.09 0.12 K 2 0 % 1.10 0.42 0.02 0.06 0.05 0.11 C % 0.6 0.3 1.3 2.4 0.4 0.2 Sum: 100.3 106.2 89.5 85.3 92.9 110.97 Water Element 41.8 S 0 4 0.47 Fe <0.1 Zn <0.1 Pb <0.05 Cu <0.1 Ni < 1.0 As — Cd 28.2 Si — c o 3 <0.3 Al 64.8 Na 31.4 Ca 0.77 Mg <0.1 Mn 8.7 K — C For all solid samples: Bi <0.001%, Cr <0.02%, Se<0.001%, Te<0.002% L.O.I, not reported : not analysed Table 5.2 Final Composition of Paste Samples Element Louv Tiz Stotal % 8.8 10.2 35.1 32.6 S 0 4 % 1.0 0.2 0.4 3.2 Sulfide % 7.9 10.0 34.7 29.5 Fe % 15.3 16.2 32.3 29.0 Zn % 0.11 0.64 1.97 1.03 Pb % 0.03 0.07 0.36 1.03 Cu % 0.002 0.003 0.17 0.10 Ni % <0.001 <0.001 <0.005 0.007 As % 0.02 0.005 0.29 0.35 Cd % <0.001 0.005 O.001 0.12 S i 0 2 % 41.9 34.1 11.7 21.6 co 3 % 0.8 2.3 2.3 1.6 A l 2 0 3 % 9.4 5.6 2.8 1.3 Na 2 0 % 0.43 0.10 0.10 0.16 CaO % 4.9 10.6 5.0 17.0 MgO % 6.8 2.0 0.7 4.0 MnO % 0.22 1.40 0.09 0.12 K 2 0 % 0.04 0.09 0.12 0.16 C % 1.3 2.3 0.4 0.2 Sum: 90.2 85.7 93.3 110.44 90 5.2 ASPECT OF C U R E D PASTE SAMPLES T h e F I M paste samples were very friable and had little cohes ion after 14 days o f h u m i d c u r i n g . S o m e F I M samples were s l ight ly damaged u p o n d e m o u l d i n g and part ia l ly dissociated or c r u m b l e d w h e n first being exposed to water or the Fe2(S04)3 so lut ion. L o u v i c o u r t samples were also relat ively soft and friable after 14 days o f h u m i d c u r i n g a l though more resistant than F I M samples. N o disso lut ion or fracturing was observed u p o n i m m e r s i o n o f the samples. T h e cured paste samples o f B r u n s w i c k and T i z a p a were m u c h harder and appeared to have l o w e r porosi ty that the other two mixtures. T i z a p a paste was the densest and hardest o f a l l mixtures . D u r i n g the cur ing per iod, T i z a p a samples f o r m e d a t h i n o x i d i z e d coat ing o n the paste surface inside the m o u l d (Figure 5.1). T h e coating penetrated less than 1 m m and was sanded d o w n to expose fresh surfaces to the leaching solutions. Figure 5.1 Superficial Oxidation of Freshly Cured Tizapa Paste Samples 91 5.3 WATER ABSORPTION BY THE PASTE Measurements of water absorption in the puck and cylinder samples after 5 and 20 weeks of leaching are presented in Tables 5.3 a and b respectively. With the exception of the Louvicourt puck sample, the percentage of water absorbed by the pucks and cylinder samples remained relatively unchanged in both the flooded and cycled water environments (Figures 5.2 and 5.3). Tizapa paste had an average water content of 15.7 % after each leaching period, similarly Brunswick paste had a water content around 17.8 %, F I M 22.8% and Louvicourt 23.2 % water. The paste with the lowest water content (Tizapa) corresponding with the mixture having the highest proportion of cement. Following this trend, Brunswick, F I M then Louvicourt had decreasing amounts of cement and increasing water contents. Equal water contents for samples of similar composition but very different sizes and shapes (such as pucks and cylinders of one particular mixture) suggest that the samples were completely saturated with water and with the ferric sulfate solution. Saturation of the paste is likely achieved within 24 hours after immersion since the cycle-leached samples had similar water contents to the flooded-leached pastes. Water absorption in the ferric sulfate environment seemed to indicate an increased water content with time for all samples (Figure 5.4). Given that the samples became saturated with water shortly after immersion, the apparent increase in water content is instead a loss of solid mass with leaching time due to dissolution of the sample. 92 di ffe re nc e in  w at er  a b or pt io n ^8 e> O 3 s CN CN e> O c> o >8 e> CN CO CN or- CO T— e> CO % H 20  av er ag e CO m' CM CO CM CO CM CO O) in 00 csi CM <p CO CM CO T oo CM CO in CM 00 o> O CM X 15 .5  15 .1  23 .4  23 .0  23 .0  23 .2  17 .8  17 .8  16 .2  15 .6  22 .6  22 .9  23 .8  23 .3  18 .7  17 .4  19 .3  16 .3  29 .4  25 .5  26 .6  23 .9  20 .4  19 .2  •a -g; 2 ~ 5 ~ 52 .7  4 4 .5  50 2. 8 42 6. 9 12 .7  9 .8  38 2. 0 29 4. 2 40 .0  3 0 .8  37 3. 2 28 6. 7 46 .8  3 8 .5  46 7. 2 3 8 4 .1  52 .9  4 4 .3  50 2. 0 42 3. 6 33 .1  2 5 .6  41 1. 0 3 1 6 .8  56 .3  4 2 .9  38 8. 1 29 7. 6 54 .0  4 3 .9  48 9. 2 40 4. 2 41 .5  3 3 .5  49 6. 2 4 1 5 .5  13 .6  9 .6  39 7. 4 29 5. 9 32 .6  2 4 .0  39 6. 2 3 0 1 .5  48 .5  3 8 .6  47 0. 4 38 0. 3 a TI ZA P A  #  3  P U C K  C YL IN D ER  FIM  #  1 0 P U C K  C YL IN D ER  LO U V  # 5 P U C K  C YL IN D ER  B R W K  # 6 P U C K  C YL IN D ER  TI ZA P A  #  4  P U C K  C YL IN D ER  FIM  # 7 P U C K  C YL IN D ER  LO U V  #  3  P U C K  C YL IN D ER  B R W K  #  5  P U C K  C YL IN D ER  TI ZA P A  # 2 P U C K  C YL IN D ER  FI M  # 3 P U C K  C YL IN D ER  LO U V  #  8  P U C K  C YL IN D ER  B R W K  # 7 P U C K  C YL IN D ER  pepoou - jeiBM uoijnios e(K )S )29d %  H 2 0  av er ag e CO u i .— (O CM CM 22 .6 0 00 iri O) CM CM q CO CM iri CO CM CM a oo O CM X a? 15 .9  14 .6  22 .5  22 .8  6. 1 22 .6  18 .2  17 .2  15 .8  15 .7  22 .3  23 .4  23 .2  22 .8  17 .6  17 .5  16 .4  15 .4  25 .8  22 .9  25 .0  23 .4  19 .5  17 .9  W  w et  W  d ry  (g ) (g ) 26 .7  2 2 .5  50 1. 5 4 2 8 .1  15 .5  1 2 .0  40 9. 3 3 1 5 .8  32 .9  3 0 .9  40 0. 2 30 9. 7 28 .7  2 3 .5  48 0. 4 39 7. 6 42 .3  3 5 .6  49 7. 4 4 1 9 .5  31 .0  2 4 .1  36 5. 8 28 0. 2 52 .4  4 0 .2  40 2. 4 3 1 0 .7  40 .5  3 3 .3  46 4. 6 38 3. 5 43 .8  3 6 .6  51 1. 6 43 2. 9 21 .5  1 6 .0  41 3. 5 3 1 8 .9  19 .2  1 4 .4  40 9. 0 3 1 3 .4  38 .4  3 0 .9  47 5. 5 39 0. 6 Q TI ZA P A  # 6 P U C K  C YL IN D ER  FIM  # 4 P U C K  C YL IN D ER  L O U V # 9  P U C K  C YL IN D ER  B R W K  #  1  P U C K  C YL IN D ER  TI ZA P A  # 5 P U C K  C YL IN D ER  FIM  # 1 1 P U C K  C YL IN D ER  L O U V # 2  P U C K  C YL IN D ER  B R W K  #  1 0 P U C K  C YL IN D ER  TI ZA P A  #  7  P U C K  C YL IN D ER  FIM  # 8 P U C K  C YL IN D ER  L O U V # 4  P U C K  C YL IN D ER  B R W K  # 8 P U C K  C YL IN D ER  » pepAo - j a j B M pepooy. - j e j B M uounps e ( frOS )39d 93 ii ni • AK * 5 10 15 weeks leached 20 Tizapa FiM Louv Bwk Figure 5.2 Water Absorption in Cyclic-Leaching (water) Environment o u—— 3K— JK 5 10 15 weeks leached 20 -Tizapa -F iM -Louv -Bwk Figure 5.3 Water Absorption in Flooded-Leaching (water) Environment weeks leached Figure 5.4 Water Absorption in Fe2(S04)3 Solution 94 5.4 CHEMISTRY OF PASTE LEACHATES Information on the effect of the different chemical environments on paste was obtained by comparing the chemistry of leachates of the different leaching environments and the solid phase chemistry of the unleached samples. Results indicated that leachate chemistry was controlled more by the characteristics of the leaching environments and the cement content of the paste mixtures than by the mineralogy of the different tailings. 5.4.1 Flooded and Cycled Water-Leached Environments The chemistry of the flooded-environment and cycled-environment leachates are presented in Appendix II (Tables I I - l and II-2 respectively). Figures 5.5 to 5.32 present graphs of element concentrations with advancing leaching cycles for each water-leached environment. The solid phase concentration of the element or ion appears in the legend box of each graph. Depletion rates were calculated for the elements showing the greatest change with time (calcium, potassium and magnesium) in order to verify the effect of the different leachate volumes extracted from each leaching environment. Flooded environment p H values remained relatively high (pH 8-9) throughout the 20 weeks of leaching (Figure 5.5), whereas the cycled environment p H reached steady state at p H 6-7, closer to the p H of the distilled water (Figure 5.6). Leachate conductivities were much higher in the early cycles of the flooded environment and reached steady state at a slightly higher level that the leachate of the cycled environment (Figures 5.7 and 5.8). Conductivity measurements are indicative of a greater dissolution of the sample in the 95 flooded environment compared to the cycled environment. Redox potential measurements are on average higher in the cycled environments than in the flooded environments (Figures 5.9 and 5.10). The higher redox levels in the cycled environment leachate are due to a higher dissolved oxygen concentration resulting from the exposure of air between each cycle. Trace concentrations of dissolved iron are present in F I M and Louvicourt leachates of both the flooded and cycled environments in the early cycles of leaching (Figures 5.11 and 5.12). The aqueous iron concentrations appear to be unrelated to the iron content of the paste mixtures. The presence of aqueous iron either is from dissolution the tetracalcium aluminoferrite phase of the cement or from the dissolution of pre-existing soluble iron salts originally contained in F I M and Louvicourt tailings. Zinc concentrations in the flooded environment indicate that zinc is not mobile with the exception of Brunswick in the first 3 cycles of leaching where concentrations of zinc (0.2-0.3 mg/1), slightly above detection limit (0.1 mg/1), are leached out of the paste (Figure 5.13). Higher zinc mobility occurred in the cycled environment throughout the leaching cycles where zinc was mobilized from all pastes (Figure 5.14). Cycled environment zinc concentrations were close to detection limit for Louvicourt and F I M samples, both of which contain the lowest solid phase zinc concentration. Brunswick and Tizapa samples showed higher concentrations of zinc throughout the leaching cycles compared with the flooded environment and a general increase in zinc concentration with leaching time. Leachate zinc concentrations appear to be proportional to solid phase concentrations of zinc and possibly inversely proportional to the cement content of the 96 samples. Unlike zinc, lead was not very mobile in either water-leached environment. In the flooded environment, Louvicourt, F I M and Brunswick leachate lead concentrations remained close to or below detection limit throughout the 20 weeks (Figure 5.15). N o lead was leached out of the Tizapa samples. L o w quantities of lead were leached out from all samples in the early cycles of leaching and out of Louvicourt and Tizapa in the last cycles (Figure 5.16). A direct correlation cannot be made between solid phase lead concentrations and leachate concentrations. Sulfate concentrations followed a decreasing trend for all samples in both the flooded and cycled water environments (Figures 5.17 and 5.18). p H values were too high to allow for significant sulfide oxidation hence the sulfate in the leachate is most likely derived from the initial sulfate content of the tailings, most probably sulfate salts precipitated from the tailings water upon drying of the tailings. Indeed, sulfate concentrations in the leachate were proportional to the solid phase concentrations of sulfate. Calcium concentrations also showed a decreasing trend in both the flooded and cycled environments (Figures 5.19 and 5.20). Initial flooded leachate concentrations of calcium were between 682 mg/1 (FIM) and 922 mg/1 (Brunswick), falling to 185 mg/1 and 590 mg/1 respectively after 20 weeks of leaching. Louvicourt and Tizapa concentrations fall between those of F I M and Brunswick. The difference in the time series concentration of calcium between both environments resulted from the different leachate sampling protocols. The calcium depletion curves are in fact similar for both environments (Figures 5.21 and 5.22). The flooded environment showed a slightly slower 97 depletion in the early cycles but quickly reached the values of the cycled environment in the later cycles. Aqueous concentrations of magnesium, potassium and silicon in both flooded and cycled environments are proportional to the solid phase concentration of each sample for the respective element (Figures 5.23 to 5.32). In the early cycles of both water-leached environments, Louvicourt leachates had considerably higher concentrations of magnesium than the other samples (Figures 5.23 and 5.24). Magnesium concentrations reached steady state shortly thereafter. Louvicourt concentrations remained slightly higher than the other samples. A correlation can be made between aqueous magnesium concentrations and the slag component of the Louvicourt paste, the latter having the highest solid phase concentration of magnesium of all paste ingredients. It is most probable that the magnesium in the leachate originated from the dissolution of the incompletely cured slag component of the Louvicourt sample. In fact, the granulated blast furnace slag used by Louvicourt requires up to 21 days of curing before participating in cementicious reactions (Dallaire, 1997). The steady state conditions reached after the first cycles of leaching were probably the consequence of a more complete curing of the binder, reducing the availability of magnesium. Depletion profiles of magnesium were similar for all samples in both water-leached environments, with the exception of Tizapa (Figures 5.25 and 5.26). After 20 weeks of leaching, the depletion of magnesium from the Tizapa sample in the cycled environment was double that of the flooded environment. The continued depletion of magnesium from Tizapa may be related to the dissolution of a magnesium mineral (i.e. dolomite: CaMg(C03)2) , 98 unaffected by the curing of the cement phase. In the case of potassium, the most important source of that element in the paste is the cement phase (refer to Table 5.2). The highest leachate concentrations of potassium coincide with the Tizapa pastes, characterized by the highest proportion of cement (Figure 5.27 and 5.28). Potassium depletion rates are similar for both water-leached environments for all samples, with the exception of Louvicourt, which had a higher K depletion rate in the flooded environment (Figures 5.29 and 5.30). Sil icon concentrations remained low in both water-leached environments (Figures 5.31 and 5.32). The general stability of silicate minerals in the p H and redox conditions of the leaching cells suggests that silicon is dissolving from the more soluble cement phases such as the tobermorite gel (Ca3Si2CV3H2C>), the principal binding agent of cement. Continued decomposition of this phase would have serious consequences on the stability of the binder. 99 o o o o o o C N O t O r O O O O O O * o CM r-H d (S n) AiiAipnpuoo   (I/6UJ) qd (1/6) *os # i ? s> ? <P o o o o o o o o o o o O CO <D t CN (l/Biu) BO ( | /6lU) BQ 8 § 8 S 5 uoi}B|dap BO % uouaidap eo % o o o o o o o CO tO CN 0> ( 0 CO o o o o o o o CO l O CN O CD CO (|/6iu) 6|A| m co csi 1 - o uojjeidap 6|/\| %  (|/6uj) is 5.4.2 Fe2(S04)3 Solution-Leached Environment The chemical results of the ferric sulfate leachate solutions are presented in Appendix II (Table II-3). Time series concentration graphs, as well as calculated depletion rates for calcium, potassium and magnesium, are shown in Figures 5.33 to 5.50. The acidic p H (2.4 - 2.6) of the ferric sulfate solution was almost instantly neutralized when coming in contact with the paste sample (Figure 5.33). In the first leaching cycles, ferric sulfate solutions were buffered to a p H of 6-7 for all samples. After 40 cycles of leaching (20 weeks), Louvicourt, Brunswick and Tizapa solutions were buffered to an average p H of 3.5 whereas F I M leach solution was buffered to a slightly higher average p H of 5. Conductivities of the leachate solutions show a considerable drop in the first four cycles of leaching, stabilizing at a lower level (1500 uS) than that of the ferric sulfate solution (2300 uS) (Figure 5.34). This indicates that the ions are precipitating from the solution upon contact with the samples. The ferric sulfate solution added to the samples bi-weekly has an average redox potential of 620 milivolts S H E . Initial leachate redox values (after 3 to 4 day contact periods with the samples) are around 130 m V , rising to near ferric sulfate solution values in the final cycles (Figure 5.35). Louvicourt, Brunswick and Tizapa leachates had final redox values around 480 m V whereas F I M final redox conditions were around 330 m V . The evolution of the redox values in all leachates again indicates that the ferric sulfate solution is decreasingly altered or buffered by the sample as leaching progresses. In the initial stages of leaching, dissolved iron concentrations in the leachate were much 108 lower than those of the ferric sulfate leaching solution. Iron was precipitated out of solution onto the surfaces of the paste samples as a dark orange ferric hydroxide precipitate (Figure 5.36). A s leaching progressed, leachate iron concentrations of all samples slowly increased, approaching the values of the ferric sulfate solution (Figure 5.37). This indicates that the ferric hydroxide coating was forming much more slowly. Unlike the water-leached environments, a considerable amount of zinc was continuously leached out of the Brunswick paste sample, averaging a concentration of 15 mg/1 throughout the leaching cycles (Figure 5.38). Zinc was also leached out of the Tizapa sample (average of 4 mg/1), increasing in the last five cycles to about 9 mg/1. Louvicourt and F I M samples released low concentrations of zinc relative to the other samples (below 3 mg/1) but higher than in the water leached environments. The samples with the highest solid phase zinc concentrations (Tizapa and Brunswick) yielded the highest leachate concentration of zinc. The mobility of lead was slightly higher in the ferric sulfate environment than in the water environments (Figure 5.39). Tizapa and Brunswick leachates had the highest concentration of aqueous lead throughout the leaching cycles, both slowly increasing to an average of 1.6 and 1.2 mg/1 respectively in the last 10 cycles. Lead concentrations in the Louvicourt leachate were close to detection limit throughout the leaching cycles whereas F I M showed trace concentrations of lead only in the final cycles. A s with zinc, lead concentration in the leachate appears to be proportional to the solid phase concentration. In the first cycles, sulfate concentrations in the leachate were greater than the leaching 109 solution sulfate. After the third leaching cycle, leachate concentrations were essentially the same as the ferric sulfate solution. These concentrations were maintained throughout the remaining leaching cycles. This suggests that the sulfate ion of the leaching solution is not participating in cement-altering reactions within the pastes. Calcium was leached from all samples throughout the test. Relatively high concentrations were leached in the early cycles, decreasing to steady state around the 10 t h leaching cycle (Figure 5.41). The variation of calcium concentrations between the different samples is much greater in the ferric sulfate environment than in water-leached environments. Louvicourt had the lowest aqueous calcium concentration as wel l as the lowest solid phase concentration of calcium. F I M and Brunswick had generally the highest leachate concentrations of calcium as well as the highest solid phase concentration. Leachate calcium concentrations are proportional to the solid phase concentration prior to leaching. Ferric sulfate-leached depletion rates of calcium (Figure 5.42) were slightly higher than in the water environments with the exception of F I M for which the depletion rate was double that of the water environments. A s with the water-leached environments, aqueous concentrations of magnesium, potassium and silicon are also related to the solid phase concentration of the element. Aqueous magnesium concentrations were highest in the Louvicourt leachate, which also possessed the highest solid phase concentration (Figure 5.43). In the initial cycles, leachate M g concentrations were similar to those of the flooded environment. A s leaching progressed, however, the M g concentrations remained high in the ferric sulfate solution. Tizapa, Brunswick and F I M leachate concentrations of M g were also slightly 110 higher in the ferric sulfate environment than in the flooded water environment. Depletion rates of M g were, in fact, much higher in the ferric sulfate environment (Figure 5.44) compared to the water-leached environment. Potassium concentrations of each leachate followed a trend similar to that of the flooded environment (Figure 5.45). Leachate concentrations of K in the first cycles were similar for both environments. However, the decrease in concentrations occurred more suddenly and earlier in the ferric sulfate environment, falling to near detection limit around the 5 t h leaching cycle. Depletion rates of K are similar in all three environments, with the exception of Louvicourt for which the depletion rate is lower in the ferric sulfate than in flooded water (Figure 5.46). Leachate concentrations of silicon were similar to those of the flooded environments in the initial cycles, the highest concentration being that of the Louvicourt leachate (Figure 5.47). Contrary to the water-leach environments, silicon concentrations in the ferric sulfate environment showed a continued increase for all samples. Given the low p H conditions of the leachate, especially in the final cycles, the dissolved silicon in this environment may be leached from silicate minerals as well as the tobermorite phase of the binder. I l l Figure 5.33 Fe2(S04)3 Cells - pH 4000 1 3 3000 • , R — | • | I 1 4 7 10 13 16 19 22 25 28 31 34 37 40 n o . of c y c l e s - * Tiz "*• Louv ""• Brw - • FIM •- - -Solution Figure 5.34 Fe2(S04)3 Cells - Conductivity Figure 5.36 Ferric Hydroxide Coating on Paste Sample 1400 1 4 7 10 13 16 19 22 25 28 31 34 37 40 no . of cyc les - * Tiz (32.3% Fe) - • Louv (15.4%) - • Bwk (29.0%) - * FIM (16.2%) - - "Solution Figure 5.37 Fe2(S04>3 Cells - Dissolved Iron 30 25 20 15 10 5 0 V I I I I I I I I I I 1 1 ' 1 1 4 7 10 13 16 19 22 25 28 31 34 37 40 no. of cyc les Figure 5.38 Fe2(S04)3 Cells - Dissolved Zinc -Tiz (2.0%Zn) -Lou (0.1%) -Bwk ( 1 . 0 % ) -FIM (0.6%) 113 10 1 4 7 10 13 16 19 22 25 28 31 34 37 40 no. of cycles Figure 5.39 Fe2(S04)3 Cells - Dissolved Lead •Tiz (0 .36%Pb) -Lou (0.03%) •Bwk ( 1.0% ) "FIM (0.07%) (0 .4%SO4) Louv (1.0%) •Bwk (3.2%) •FIM (0.2%) - - -Solution Figure 5.40 Fe2(S04)3 Cells - Sulfate Concentration - * — Tiz (5.0% CaO) - • — L o u v (4.9%) - • — B w k (17.0% ) ••—FIM (10.6%) n o . of c y c l e s Figure 5.41 Fe2(S04)3 Cells - Calcium Concentration 114 50 no. of cyc les Figure 5.42 Fe2(S04)3 Cells - Calcium Depletion Rate 180 1 4 7 10 13 16 19 22 25 28 31 34 37 40 n o . of c y c l e s Figure 5.43 Fe2(SC>4)3 Cells - Magnesium Concentration 10 1 4 7 10 13 16 19 22 25 28 31 34 37 40 no. of cycles Figure 5.44 Fe2(S04)3 Cells - Magnesium Depletion Rate 115 180 150 s - 120 "Si E 90 * 60 30 0 l t t l W M . H I JIJMMiJUlJWJ.—••••••••• 1 4 7 10 13 16 19 22 25 28 31 34 37 40 n o . of c y c l e s Figure 5.45 Fe2(S04)3 Leach Cells - Potassium Concentration -Tiz (0 .12%K2O) -Louv (0.04%) •Bwk (0.16%) -FIM (0.09%) Figure 5.46 Fe2(S04)3 Leach Cells - Potassium Depletion Rate Tiz ( 1 1 . 7 % S i 0 2 ) - • — L o u v (41.9%) - • — B w k (21.6%) - • — FIM (34.1%) Figure 5.47 Fe2(S04)3 Leach Cells - Silicon Concentration 116 5.4.3 Summary of Observations from Leachate Chemistry The chemistry of the leachate together with the initial concentrations of each element in the paste provide the following information on processes occurring in the first 20 weeks of leaching with water and artificial A R D solution: Water-leached environments: • The paste samples provided a high immediate buffering capacity to the leaching water. The buffer was most likely coming from the dissolution of the portlandite phase of the cement (Ca(OH)2). Higher p H in the flooded environment suggests that portlandite dissolution was more intensive in the flooded environment although calcium depletion rates were similar for both environments. The longer contact time between the leaching solution and the solid sample may have allowed for the precipitation of a secondary calcium phase within the samples, thereby explaining the similarity between calcium depletion rates. • Conductivity values were indicative of the extent of sample dissolution. Higher conductivities occurred at steady-state conditions in the flooded environments for all samples. This indicates that more intensive dissolution occurred in the flooded environment throughout the leaching cycles because of longer contact times between the samples and the leaching solution. • The combination of high sulfate concentrations and low concentrations of iron and zinc suggests that sulfate was not the product of sulfide oxidation but rather was leached from a soluble sulfate phase present in the tailings (prior to cement 117 addition). Considering the high concentrations of sulfate of the tailings water (Table 2.2, Chapter 2), the sulfate in the samples likely originated from a sulfate salt precipitated when the tailings were dried, and redissolved when coming in contact with the leaching solutions. • Comparison of major ion chemistry (Ca, K , M g , and Si) in the leachate and in the solid phase of the samples suggests that the binder was actively dissolving. Calcium was most likely leached from the highly soluble portlandite phase of the cement (Ca(OH)2) (dissociation constant K e q = 6.3E22, Appelo and Postma, 1994; Pakhurst, 1995) whereas silica and potassium were likely dissolved from the tobermorite gel phase. Magnesium, in relative abundance only in the Louvicourt leachate, was l ikely dissolved from the slag portion of the binder. The decreasing loading of major ions suggests the following: 1) a depletion of the dissolving phases in the outer layers o f the paste sample and/or 2) a continued curing of the sample, binding the major ions into more stable, less soluble hydrated cement phases. Ferric sulfate-leached environment: • Leachate p H , redox potentials and aqueous iron concentrations evolved towards the values of the ferric sulfate leaching solution with advancing leaching cycles. This suggests that the ferric sulfate solution was decreasingly modified or buffered by the paste with which it came in contact. • The leaching solution sulfate did not appear to be consumed in secondary, expansive mineral growth reactions in the samples over the 40 cycles of leaching, as the sulfate 118 concentration of the leaching solution and the leachates remained unchanged throughout the leaching period. The relatively high aqueous zinc concentrations in the Brunswick sample suggests either 1) active oxidation of sphalerite that occurred in the outer layers of the paste and/or 2) dissolution of a soluble zinc phase (sulfate, oxide or hydroxide) present in the tailings. The pyrite reactivity study indicated that sphalerite present in the Brunswick pyrite samples was readily oxidized, loading the leachate with dissolved zinc to a much greater extent than the other samples. The pyrite study supports the possibility of active sphalerite oxidation. Pyrite or pyrrhotite oxidation was not verifiable from the leachate data as the leaching solution iron and sulfate concentrations masked any evidence of oxidation of these minerals. The presence of lead, albeit in low concentrations, suggest that some dissolution of galena occurred in the Tizapa and Brunswick pastes, both of which contained the largest concentration of lead in the paste prior to leaching. Major ion chemistry (Ca, K , M g , and Si) compared with the solid phase concentrations of the samples suggests that, as in the water-leached environment, binder phases were actively dissolving. Calcium was most likely leached from the portlandite phase (Ca(OH)2), with the O H - ions consumed to neutralize the acidic ferric sulfate solution. Higher depletion rates of calcium, potassium and magnesium occurred in the ferric sulfate environment (compared to the water) indicating that the A R D solution was more aggressive than water. 119 5.5 SOLID PHASE CHEMISTRY OF LEACHED PASTE Tables 5.4 to 5.7 contain scanned images of the leached paste samples showing concentric altered layers. The sample sections were cut through either pucks (rectangular) or cylinder samples (pie-shaped). The above mentioned tables also contain comparative chemical analyses of each altered layer of the samples. The detailed analytical reports are presented in Appendix III. The chemistry of each layer is presented as the difference in concentration (in percentage) with respect to the core of the particular sample in the flooded environment. This method of analysis was devised to show only the statistically significant differences in concentrations between the altered layers of a particular backfill mixture, taking into consideration the 5% margin of analytical error reported by the laboratory. The comparative studies are based on the premise that the core of the flooded samples have undergone the least amount of alteration, a hypothesis verified by various leaching studies of cemented mixtures (DeCeukelaire, 1991; Revertegat et al, 1992; Kosmatka et al, 1995; Casanova et al, 1996). 5.5.1 Tizapa Water-leached environments Table 5.4 shows a net loss of sulfate and calcium in the outer layers of both water- leached environments compared to the core of the flooded sample. The loss of calcium is greater in the 4 mm outer layer of the cycled sample than in the 0.75 mm outer layer of 120 the flooded layer. Calcium is most likely depleted from the highly soluble portlandite phase (Ca(OH)2) o f cement and possibly also from some de-calcification of the tobermorite phase (Ca3Si20s-3H20), as reported in other concrete leaching experiments (Revertegat et al, 1992; Carde and Francois, 1997). The less depleted concentration of calcium in the flooded paste may indicate the precipitation of a secondary calcium phase in the outer layer. Sulfide levels are similar in all layers suggesting that sulfide oxidation is not occurring in this layer. The sulfate leached out of the sample therefore came from the dissolution of a soluble sulfate phase already present in the paste, a similar observation extracted from the leachate geochemistry data. Ferric sulfate-leached environment Considerable differences in solid phase chemistry were observed between the first and second alteration layers of the sample exposed to the ferric sulfate solution. The outermost layer (layer 1) showed an enriched sulfate concentration but depleted concentrations of total sulfur and sulfide-sulfur, calcium, magnesium, zinc and lead compared to the underlying layer 2 and to core concentrations of the flooded sample. These results indicate that oxidation of sphalerite and dissolution of galena was probably occurring in this area. N o information is available to verify the oxidation of the iron sulfides although leachate geochemistry data suggests that it is minimal. Layer 2 was depleted in sulfate and further depleted in calcium and magnesium compared to the layer 1 and the core of the flooded sample. Solid phase concentrations in the 3rd layer had almost all reverted to the values of the core sample in the flooded environment, 121 with the exception of calcium and magnesium which still remained slightly depleted. A progressive depletion of cement occurred from the centre of the sample towards the edges. The greater loss of cement in the outer layer exposed to the ferric sulfate leach solution allowed sulfide mineral oxidation to take place. 5.5.2 Brunswick Water-leached environments Table 5.5 shows that, like the Tizapa water-leached samples, the outer layers o f both water-leached Brunswick samples underwent a net loss of sulfate and calcium, with a more severe depletion occurring in the cycled environment. The dissolved calcium and sulfate likely originated from cement phases (portlandite and tobermorite) and pre- existing sulfate phases respectively, as described for the Tizapa samples. In addition, similar sulfide concentrations and sulfur to metals ratios in all water-leached layers suggest that no sulfide oxidation occurred in any water-leached sample. Ferric sulfate-leached environment Sulfate concentrations were significantly depleted only in the 2 n d layer of the ferric sulfate-leached paste. Precipitation of a secondary sulfate phase from the leachate solution most likely replaced the lost sulfate due to dissolution of the solid phase, explaining the unchanged sulfate concentrations in the outermost layer of the paste. The depleted sulfate concentrations, unchanged sulfide concentrations and stable sulfur to metal ratios suggest that, like in the Tizapa case, no oxidation occurred in layers 2 and 3 122 of the ferric sulfate-leached paste. Sulfide mineral oxidation was also unlikely in layer 1 as indicated by the similar sulfide concentrations between layer 1 of the ferric sulfate environment and the core of the flooded paste. Also similar to Tizapa, calcium was progressively less depleted in underlying layers 2 and 3, reflecting the decreasing dissolution of portlandite (and de-calcification of tobermorite) with depth. 5.5.3 Louvicourt Water leached environments N o alteration layers were evident upon observation of the water-leached samples (Table 5.6). A n outer layer of 1 mm was, therefore, arbitrarily chosen and analysed for verification. The thin layers extracted did not provide enough sample to allow for complete elemental analyses. Sulfur species analyses showed depleted sulfate concentrations in the outer layers of both environments with respect to the core of the flooded sample, resulting from the dissolution of pre-existing sulfate phases. The absence of significant changes in sulfide concentrations and in total sulfur to sulfide ratios suggests that no oxidation of sulfide minerals occurred in these layers. Ferric sulfate-leached environment Two distinct layers of alteration were observed on this sample. However, only the second layer provided enough material to carry out elemental analyses. Similar to the Tizapa ferric sulfate-leached sample, layer 2 of the Louvicourt sample showed depleted concentrations of sulfate, calcium, magnesium, zinc, copper and lead. Some oxidation of 123 sphalerite along with dissolution of galena and possibly some soluble copper phases may have occurred in this layer. The similar sulfide concentration with respect to core values does not, however, support the occurrence of significant sulfide mineral oxidation. 5.5.4 Francisco I. Madero Water leached environments Table 5.7 shows a net loss of sulfate in layer 1 of both water-leached environments, as well as a net gain in copper in layer 1 of the flooded sample. The elevated copper concentration suggests a mobilisation of copper from an internal area to the edges of the sample where it may have precipitated as a more stable phase. The migration of elements may have been facilitated by the greater porosity and lower binder proportion of the F I M backfill mixture compared to the others. Ferric sulfate-leached environment Layer 1 was characterized by depleted concentrations of sulfate, calcium and magnesium and a considerably enriched concentration of iron. A s with the other ferric sulfate- leached samples, the depletion of sulfate and major ions may be explained by the dissolution of pre-existing sulfate phases and cement phases in that layer, namely portlandite and perhaps tobermorite. The enriched iron may be explained by a deeper penetration of surficial ferric hydroxide precipitates observed on all samples in this environment because of lower binder content of the F I M paste mixture. 124 u cu c o - Q i IT, C5 H "8 i l > III •XT I 50 c w T3 o e W UL. c Q o OB 05 cu X ™ CN CN 1 CO rVj <*> J t? — i C O 5 5J I O ™ CD C/5 HH U tit CN r-4 e T3 C M O to CD Vi H \- 03 03 • J C N 00 13 SB C c5 o •a I 5-. 03 "O SO P CD I T3 i i D a a o o O 00 2 o 03 03 O o 5. ss o o £ £ •T3 <D o -9 as O k> O -2 J CO p -a a £ ^ -rt o J O Js O 3 £ I c<~> CM CN Si cn Si co o © i ©" 'I o c 03 C oo £ CL, CL CO o S3 oo -o UJ oo _>> 03 C 03 o oo J-CD 5? 03 .c CN A 93 9J00 UI0JJ 90U3J9JJIp 0UO0 % a CN "8 o 1 w c > I &o CN B 1) i W U i<5 c u S c a > a pq C c E o Vi Vi 5 03 U 00 tN o U CN Vi O IS BO H 1 - u. <u <L> >» 03 — 0 0 o? r- O Jo © 03 50) g 5 o i a1 o g X) O 03 & i Oil c 03 O >- >> 0 3 o 03 5) 2 o o • at O o — J 3 O .SP o .= CL o +3 g .SP o S B. o CL o o •fl 2 .SP o JS CL en 2 .SP © x tx . Vi 9 5 3s £ ft in H fi. i C N gj 6, Bl oo CN E 0 CL 1 « c °5 CL i i 1 oo oo cc d ad i 81 O O O O o E CL a vo t"* o to OJ CL E 03 35 l-o ii vi >» "3 c 03 in S3 03 c 9J03 UIOJJ 30U3J9.LT.ip 0UO0 o/Q in | C N A u. o vo CN SB S © w u t» u 09 SS CU TB cw -= CD 05 S3 -3 c e n Q © o -J IT; 93 T3 - O o c CD > c O fe 3 CD 2 > , c w J D o , U m £ s 2 > c UJ -o u T3 © I fe fe X c . © -3 93 on fN ecj «3 os u co 93 — . 3 O 00 — 93 - i £ £ £ tr> 00 Cfl u .£ E l H C3 00 00 93 £ J 1 1 oo I a 12 P IN 93 >> 03 —J 03 00 i 93 00 c S3 — O M 3 93 60 e 03 03 o A3 03 ~CD 03 I o c U o 00 -o 1) oo •— 3 3 2 I a* - T3 C o a 03 l l fe oo O 00 oo" =5 2 c Q . C 00 oo" C — o CL. o oo oo" 3 O a CD E 3 X 93 fe 00 c 93 <u E £ 3 03 00 00 £ O N g, a o O GO to re 00 £, il 00 Cu £ oo c 93 3 00 3 cN £ fN r i gj Si fi, a £, a m m £ C L C L oo fN © C L £ 03 00 3 93 O £ 3 oo C 2 - C fe 9J00 uiojg 30U3J3JJTp UOUBJJU33UO0 o/0 oo = o u 93 o 93 a- o -= <J 03 o = o o - os o cv "« c 98 S - fe r~ IT' ii s ed H CD c OJ oo oo l : x C N a GO CD TJ g g o CD S? fe £ £ 0 0 © £ £ CD £ J - 9 03 oo 00 CD o IE fe — 93 >. 03 fe at 00 c 03- 03 i • CD 00 -c 03 & i CD 00 '5 03 I o 93 O U o 00 13 03 £r L - • — CU oo T 3 O £ OJ oo o 1- o oo o oo 00 2 a 1 CD T 3 O £ CD I 93 -a o £ o 00 $ 00 O a o oo 03 I # CU oo TJ O l a CD CD fe C O CU £ £ 3 03 oo "2 C ON £ o fi, o CO i r j £, C N IT, £ , a £ a r4 fi, i o C N CN oc r-i £ o. C L IT) - C 0- 3JO0 UI0i} 90U3J3JJip UOIJBJJU30UO0 % 5.5.5 Summary of Observations from the Solid Phase Chemistry Water-leached pastes: • N o signif icant o x i d a t i o n o f sulf ide minerals occurred i n any layer o f a l l water- leached samples. Sulfate contained i n the paste was readi ly leached out o f the b a c k f i l l f r o m a pre-exist ing sulfate phase present i n the mixture . In a l l but the F I M samples, sulfate deplet ion was more severe i n the c y c l e d environment than the f looded environment. • C a l c i u m was readi ly leached out f r o m the outermost layers o f a l l samples analysed, i n b o t h water-leached environments. D e p l e t i o n was more severe i n the c y c l e d environment for both T i z a p a and B r u n s w i c k samples. N o data were avai lable for L o u v i c o u r t . T h e less depleted amount o f c a l c i u m i n the f looded paste m a y indicate the prec ipi tat ion o f a secondary c a l c i u m phase i n the outer layer i n that environment . C a l c i u m was l i k e l y lost f r o m the disso lut ion o f the cement phases o f port landite (Ca(OH)2) and p o s s i b l y f r o m decalc i f icat ion o f tobermorite ( C a 3 S i 2 0 s - 3 H 2 0 ) . T h e latter was also reported by Revertegat et al, (1992) and Carde and F r a n c o i s (1997) i n s i m i l a r experiments. • T h e m a x i m u m thickness o f the alteration layers i n both water-leached environments was larger for paste mixtures containing the highest percentage o f pyr i te and largest p r o p o r t i o n o f binder. Indeed, the m a x i m u m thickness o f the alteration layer for L o u v i c o u r t and F I M pastes was 1 m m i n 20 weeks, compared to 6 m m for T i z a p a a n d B r u n s w i c k mixtures . 129 Ferric sulfate-leached pastes: • Similar or increased sulfate levels on the edges of the samples suggest that one or both of the following were occurring in the samples: 1) penetration and precipitation of sulfate from the solution as the solution came in contact with the pore solution of the paste and/or 2) active oxidation of sulfide minerals (sphalerite and possibly some iron sulfides) producing sulfate and releasing metal ions. Both cases involve the advancement of a ferric sulfate solution front into the paste. • Greater net losses of calcium and magnesium occurred in the ferric sulfate environment compared to the water-leached environments, indicating a more intensive dissolution of portlandite and perhaps tobermorite (through decalcification). Portlandite, a hydroxide mineral, is considerably more soluble in an acidic environment than in the a near-neutral water-leached environment, the hydroxyl ion being consumed to buffer the acidic solution. • The alteration layers of samples leached in ferric sulfate solution penetrated deeper than in the water environments. In addition, similar to the water-leached environments, mixtures with higher pyrite and high binder content had deeper penetrating alteration: 8 mm and 10 mm in 20 weeks for Tizapa and Brunswick respectively, compared to 2.75 mm and 1.8 mm for Louvicourt and for F I M pastes. • In general, as a result of the more intensive dissolution of the binder phases, the readily oxidized sulfide minerals such as sphalerite present in the outermost layers were most likely oxidized by the acidic (unbuffered) ferric sulfate solution. The 130 oxidation of sulfide minerals did not, however, occur in all samples in that environment. The net loss of lead in the altered layers of some samples indicated that galena was probably dissolved or oxidized from those areas. 5.6 BUFFERING CAPACITY AND ACID-BASE ACCOUNTING Paste p H measurements and A B A results are presented in Tables 5.8 a to d. Paste p H can give qualitative information on the immediate buffering capacity or acid production of soil or rocks. In this study, alkaline paste p H suggested the presence of a readily available source of neutralizing minerals and little or no acidity. The paste p H o f unleached Louvicourt (8.4), Brunswick (10.3) and Tizapa (11.1) samples were high. These values corresponded to a binder content of 4.5% for Louvicourt (0.9% cement and 3.6 % slag) and to cement contents of 5% for Brunswick and 6.2% for Tizapa respectively. For comparison, the paste p H values of these tailings ranged from 5.0 to 5.8. F I M paste sample had a paste p H of 8.9, a smaller increase from the tailings p H of 7.3. These values reflect the smaller proportion of cement (3%) of the F I M paste mixture. Paste p H values corresponded well to the cement content of the paste mixtures. 5.6.1 Evolution of Acid Producing and Acid Neutralizing Potentials: A s a preliminary step to acid-base accounting ( A B A ) , fizz ratings ranged from no fizz to moderate fizz for both the tailings and unleached paste mixtures of Tizapa, Brunswick and Louvicourt. F I M tailings and unleached paste produced a strong fizz. Ratios of neutralizing potential (NP) to acid production potential (AP) or N P R of all tailings, 131 unleached pastes and of each altered layer of the 20 week-leached pastes also figure in Tables 5.8 a to d. A l l tailings possessed N P R values below 0.5. The addition of cement provided some neutralizing capacity to all the mixtures. However, N P R values all remained below 1, hence a strong potential for acid generation remained. In all cases, the proportion of cement was not sufficient to change the classification of the backfill to a potentially non acid-generating residue. In all cases, as leaching progressed, A P values remained statistically similar (within the 5 % margin of analytical error) while N P was being depleted. N P depletion was attributed to the dissolution of portlandite, agreeing with the leachate and solid phase chemistry data. In addition, N P depletion was more severe in the cycle-leached environment than in the flooded environment. Resulting N P R values were lower for the outer layers of samples in the cycle-leached environments than in the flooded environments. Ferric sulfate or artificial A R D solution leaching depleted N P values to a greater extent than water-environments. This agrees with the solid phase chemistry, indicating that the dissolution of portlandite was considerably more significant in the ferric sulfate environment. In fact, negative N P values in the outer layers of the ferric sulfate environment in all cases analysed indicated not only complete depletion of buffering minerals but also the initiation of acidic conditions in those layers of the samples. 132 CL < CL CL < CL n -' o < 3 CO at > CD o d o d CO o d o CD o ci 8 m o 00 o CO o 0 0 o •o co IO o oo o o CN O o oo CM o CO CM 0> 3 CO o o to o> o CO o o en o CO CD 01 •*-> W I- c o u u < a> CO CO m o < •a c CO X a CO (0 CL < CL (0 CO IO CO (/} / o co s CO co o N- d co CO s 0) CO m co co cd co d oo CO if) CO CO CM m CO CO i CO CO 0 0 CO d CO CO o CO d CM CO N- CO ai CN CO N" co 0 0 ai CO CN CM o N" CO CO CM CO CO CO m CO 0 0 CM CO LO CO CL z o o CJ -I CM to to n to CO 00 CO CM CM IO CM I o z u I ai a E re CO I O n Z o X a* l-il > Q) CD O CN If) O CM CO T— o N" d cb o CM o o CM o T - d CM d co T— rsi o N" CM CO d N- d CM I O CN CO | CD CM O CM O N" O O CM O d CM d CO o N- d o N" N" if) O O I CM O d o d co CO CO o N- O o O T— d CM d CO o CO o o CM O d a> o CN o o CM O d O) o o CD If) CO 8 o N- O o CM O d O) o H i l i n 0 C O c CI) •«-« CO , co o E CD c o CO c O CD c o c c o 'o E 3 - r C L 0> a E (0 CM LO CO 11= 0) a. O CO « 1 ° CM CD >, CO a) to 1 d); CO ICN l» CD >. CO I w CD CO I CM CD >, CO I CO 1— CD >. CO CD 133 OL < Si a, < Q_ O •"3 < O a o u> T - o o O O o © © d d d d CM o o f- in m CO -<*• T— CO T— O o O o o o o o o o o o o o d d d d d d d d d d d d d d d CO T - to O) CO •<* CO 00 a* CO CN o CO o a> o co CO a> o> 0> cn CO o> 00 CO CO P i n CO o CD o CO ai a i 00 O) CO iri d CO CN CN CN CN CM CN CN CO CN CN CO in CO CN CO •̂r •gr T— h- CO CO CN in T— CO CO CO CO CO CO d d i — d d t— d T _ CO CD CM CO m CO o CO CD T _ CN d a i d ai ai d ai CO CO CO CM CO CM CM CN CO CN 00 T - co CM o co o> T" o O) CO u> ™̂ T— o> CO CO (O CO * t CO T -1 1 CO CO CO CO CO CO CD CD CO O o CO CN cn CM h- in !•» CD CM CD ai d p 00 CD d T - x— CN CN CN CM CN •3- CM CM 1 t— CN CN r o CO O O O O O m m o o m LO CN o in O o O CO CO CO CO r~~ r-- —̂ CM CN •<}• m m x— CO LO o O o o o O CO o CN O O o o o o o O o o o o o O o a> O O o o o CN CN CN CN CN CM CN CM CN d CN , ~ CN CN CN CN CM CO CO CO o O o o O O o O o O O O o O CD CO CD X— T— T— T— T— T— d d d d o d d d d d d o d d d d d O o O CN a> CM a> cn a> cn a> cn en o o o o o o o o o o o o d d d d d d d d d d d d d d d d d B CD CD CD 2 2 2 2 2 CD CD CD CD CD CD CD CD CD c c a> T3 TJ T J n TJ T J o o TJ O o O o O O c c o £ E E E E E E at ui d T - sz * \ CM CN CN CO w W \J u. i _ i _ i _ k- u-t5 CO CD CD CD CD CD CD CD CD CO 0) "O >> >, >. ^ - >> >< >s CD CO CO C0 CO O CO CO CO 5 po< _ l _ l CD w_ c le  _ l _ i W OJ _ i _ l _ l o o CD CM UL o o LL 5 3 CO CD O) c c 3 o o o < a> (0 (0 OQ TJ o < •D C (0 X a (0 (0 Q. O i z D or CD 00 If) a> Si (0 CM o (0 s 6 o S I o is z cj i Q. E n co I O to Z u X a* E -a E o — > ? 3 CO 11 c (a (0 0. I CL a E (0 C0 < CL Z CL < CL z CL < CO o CO o o mm mm CO LO d CD ci 00 CN co CO o o O) CO N- ci CM d CO ci cn IO CD CM LO CO CD CO CO CO CO CN O N" CM CM CM CM CM CM 1̂ - 06 CO 10 CO CM CD T— ci GO CO CM CD 00 O CO d co CO 0) Q_ E CO </> c CD o m c CL Z o cj o to CO o CO CO CO o CO o CO LO I o 2 o X 0) a E CO X O n Z o X at l i 1 o - x: S>5>| o — E o c i i to CO CO CL I Q. CM 1^ O CO o CM CO d N- d CD c O c o 10 CD CO o CM O O CM O t— d CM d o | CO o CD O O CM CO d N- x— d CO CO o N- CM CO d N- d co CM o N- o o CM o d CM d x: CO CO cd CO CO CM O N" o o CM O T— d CM d sz CD 10 CO CN O N" O O CM O d CN d CD c o c CO CO CO o x— o CO o d CD c o c 0) a E CO CO CO £ Tn co a. sz o CO a) o CM CO CO CO LL Q) CO I 35 CL < CL Z co at a > © © IO •o co IO < CL Z CN N" d N" d d r- N" d LO LO CO LO CN LO 3? - CN CN CO CO co CN CO LO CN CO o CO l i l l co d o CN d LO d o d o CN d CN d CO CO o CN CO CO o d CN CO CL Z o o io O <3) z E d O > !| CO en CM CO CN CO | ? 3 I n 8 ^ to > CO > x o 1 * Z 75 - x LO CN N" O CO o o CN O •t— d CN T— d LO CO N" O 00 O O CN O d CN T— d LO I CO CO o 00 o o CN o d CN d r-co o CO o o CN o T— d CN 1— d CO CO o CO o o CN o d CO CO o CO o o CN o d iq CN co o q CO o CO o o CN O T— d co o N E Li. (Q CO c O CD c o CD c O CD c o CD < CL j co cd CO CO JsilSlii 'mm. m CO 00 Q. '13 co ©I ' « I© CN 1), 0} CO CO o CO 136 5.7 SCANNING ELECTRON MICROSCOPE (SEM) OBSERVATIONS The sections shown in Tables 5.4 to 5.7 discussed earlier, were observed under S E M to identify the mineralogical changes that occurred as leaching progressed. The layer identification system and thicknesses described in these tables was used to guide the S E M observations. Energy dispersive x-ray analyses ( E D X ) were carried out on many different mineral surfaces. The fragility of the surfaces did not allow the sections to be polished flat since they would easily crumble. A l l E D X analyses reported in this section are therefore qualitative and used in conjunction with S E M micrographs to identify the general composition of mineral phases. Plates 5.1 to 5.8 show the typical morphology and qualitative composition ( E D X scans) of tobermorite, portlandite, gypsum and ettringite encountered in this study. 5.7.1 Tizapa The Tizapa paste contained the largest proportion of cement (6.2%) making possible the observation of many distinct phases of hydrated cement. Tizapa samples displayed the greatest differences in cement mineralogy between the altered layers. i) Flooded Water Environment Layer 1: 0 - 0 . 8 mm A n extensive layer of densely packed euhedral portlandite crystals covered the surface of the sample (Plate 5.9). Portlandite was also abundant down to a depth of 0.8 to 1 mm, 137 along with a well hydrated but thin layer of tobermorite covering the tailings grains (Plates 5.10 and 5.11). The local abundance of euhedral portlandite in the outer layer suggests that it was most likely secondary in nature. Below layer 1: 0 . 8 -1 .5 mm A transition zone of poorly developed tobermorite was encountered at this level, where tailings grains were not so well cemented compared to the overlying and underlying layers (Plate 5.12). Little portlandite was encountered in this layer. Core: < 1.5 mm The core of the sample was characterised by well-developed and abundant tobermorite gel covering the tailings particles (Plate 5.13). Acicular ettringite was also visible covering partially hydrated tricalcium aluminate (C3A or Ca3Ali06) phases (Plate 5.14). Ettringite is normally one of the first minerals to crystalize upon hydration of the cement. Its purpose is to slow down the otherwise very rapid hydration of C3A which causes "flash setting" of the concrete (Taylor, 1997; Kosmatka et al, 1995). The occurrence of ettringite on C3A indicates that it was a primary mineral rather than a product of secondary, expansive mineral growth reactions within the sample. In addition, the occurrence of ettringite in this area indicated an abundance of C a 2 + and SO42" ions in solution, as ettringite is formed according to the reaction (Vernet, 1994): 32 H 2 0 + C a 3 A l 2 0 6 + 3 C a 2 + + 3 S 0 4 2 " -> C a ^ ^ M O H ^ ^ H z O (5.1) (C3A) (or dissolved gypsum) (ettringite) In low sulfate cement mixtures, the gypsum phase present in the dry cement is eventually 138 exhausted (in the first hours of hydration) and the ettringite proceeds to react with the excess C3A to produce monosulfoaluminate (Ca4Ai206(S04)45H20; a wavy, leaf-like morphology - not observed in this sample) according to: 2 C a 3 A l 2 0 6 + Ca6Al2(S04)3(OH)i 2 -26H 2 0 -> 3 Ca4Al 2 0 6 (S04) -15H 2 0 + 17 H 2 0 (5.2) ( C 3 A ) (ettringite) (monosulfoaluminate) Monosulfoaluminate forms away from the C3A surface, leaving the mineral surface free to continue to hydrate, consuming excess portlandite and providing some additional strength to the mixture (see reaction (3) in Table 1.1, Chapter 1). In this case, however, sulfate continues to be present in the pore water in equilibrium with the primary ettringite, preventing further hydration of C3A and consumption of portlandite. Both factors impede the normal development of strength. ii) Cycled Water Environment Layer 1: 0 - 4 mm The outer layer of the cycle-leached sample was characterized by poorly developed tobermorite gel in the exterior portion of the layer (Plate 5.15), only partially covering tailings grains. Ettringite needles and abundant gypsum were also visible in the upper portion of this layer (Plate 5.16 and 5.17). In the deeper portion (>2mm), the cement phases were well developed and distributed. Plate 5.18 suggests that the cement phase was not very strong as many tailings grains were removed during the sample preparation process, leaving empty grain-shaped cavities. 139 Core: > 4 mm A s with the flooded core, this area of the cycle-leached paste was characterised by abundant ettringite covering phases of tobermorite and unhydrated C3A (Plate 5.19). Tobermorite cover was generally extensive and well developed (Plate 5.20). iii) Ferric Sulfate Environment Layer 1: 0 - 2 mm This layer of the ferric sulfate-leached sample was generally porous with sparse hydrated cement phases (Plate 5.21). Most pyrite grains (as well as other tailings grains) showed only traces of cement cover (Plate 5.22). Some isolated areas did have well-developed tobermorite (Plate 5.23). Oolitic masses of iron sulfate were exclusive to this layer (Plates 5.24 and 5.25). This phase could either have been precipitated directly from the ferric sulfate solution in contact with the pore water of the sample or could have precipitated following sulfide mineral oxidation in that area. Layer 2: 2 - 6 mm The hydrated cement cover was more uniform and better developed in this layer, offering better protective cover to all grains such as pyrite (Plate 5.26). Under S E M , the presence of amorphous hydrated cement cover masked the presence of amorphous iron phases such as hydrated iron oxide or iron oxyhydroxide. The rust colour of this layer suggests, however, the presence of these phases. 140 Layer 3: 6 - 8 mm This was an apparent transition zone between layer 2 and the core of the sample, where the mineralogy was similar to that of the core of the flooded sample. 5.7.2 Brunswick i) Flooded Water-Leached Environment Layer 1: 0 - 3 mm Similar to the Tizapa sample, portlandite was also present in abundance in this layer, along with fairly well developed tobermorite covering the tailings grains (Plates 5.27 and 5.28). The cement appeared to be less resistant than in deeper layers as empty grain cavities were observed (Plate 5.29). The cavities were, however, less numerous than in the ferric sulfate-leached sample. Layer2: 3 - 6 m m This layer possessed well-developed and abundant tobermorite covering the tailings particles (Plate 5.30). N o portlandite was visible. A n acicular phase, most likely ettringite, was also observed from this depth down to the core of the sample. 141 ii) Cycled Water-Leached Environment Layer 1: 0 - 3 mm Tobermorite appeared to be well developed, covering the tailings grains fairly well (Plate 5.31). The paste was porous but did not show any empty grain cavities (Plate 5.32). N o gypsum or acicular ettringite were observed at any depth in this sample. Layer 2: 3 - 6 mm N o obvious mineralogical differences were observed under S E M between this layer and Layer 1. iii) Ferric Sulfate-Leached Environment Layer 1: 0 - 1 mm Macroscopically, this layer showed an outer rim of iron oxyhydroxide precipitated from the leaching solution (refer to Figure 5.36). Under S E M , a 0.5 mm crust was visible with vertical fractures extending 1.5 to 2 mm down (Plate 5.34). Many grain cavities were visible suggesting that the strength of the hydrated cement was relatively poor (Plates 5.34 and 5.35). Acicular minerals were also present in this area but were too small to obtain a reliable E D X identification (Plate 5.36). The morphology of the phase suggested ettringite. 142 Layer 2: 1 - 3 mm The hydrated cement phases appeared to be more resistant at this depth, showing a smoother sample surface with less grain cavities (Plate 5.37). Masses of oolitic iron sulfate phases were abundant in this layer, possibly a type of jarosite (Plate 5.38). Similar to the Tizapa sample, this iron sulfate phase either was precipitated directly from the ferric sulfate solution in contact with the pore water of the paste or was a product of sulfide mineral oxidation in the area. Euhedral gypsum crystals were encountered inside sample pores (Plate 5.39). N o ettringite needles were identified in this layer. Layer 3: 3 - 1 0 mm Macroscopically, a gradient in colour was visible from 3mm to 10 mm and a very pale line marked the 10mm depth. This fine layer was not noticeable under S E M and layer 3 appeared as a transition zone between layer 2 and the core of the sample. In the upper and middle portion of layer 3, cement binder strength appeared to be greater (Plate 5.40), the topographic lows were related to the porosity of the paste rather than to empty grain cavities. Tailings grains were also better covered by cement (Plate 5.41). N o ettringite needles or gypsum was positively identified in this layer. 5.7.3 Louvicourt The high slag proportion of the Louvicourt paste gave the binder a different aspect from the O P C binder used in the other mixtures. The hydrated slag was identified by its morphology and occurrence, appearing as both masses of subangular crystals (Plate 5.42) 143 and delicate, semi-dendritic crystals covering the tailings grains (Plate 5.43). Both morphologies are typical of well-hydrated slag (Taylor, 1997). The composition of the tobermorite developed from slag is slightly different from that from Portland cement, generally having a lower concentration of Ca but higher A l and M g . Plate 5.44 presents a typical E D X spectrograph of slag tobermorite (from the flooded paste). Plate 5.45 shows the general aspect of the paste. i) Flooded Water-Leached Environment Layer 1: 0 - 1 mm Contrary to Brunswick and Tizapa pastes, no portlandite was observed in any area of the flooded paste sample. Tobermorite was well developed, covering the tailings grain to a greater extent the previous two samples in the same area. The tobermorite content was slightly lower in some areas of the outer layer (Plate 5.46) compared to deeper areas where tobermorite was more abundant and evenly distributed (Plate 5.47). N o ettringite, monosulfoaluminate or gypsum was identified in any area of the flooded paste. ii) Cycled Water-Leached Environment Layer 1: 0 - 1 mm A greater depletion of tobermorite was observed down to 500um from the surface (Plate 5.48) compared to the flooded environment. Binder cover of tailings grains in this area was still greater than Tizapa and Brunswick at that depth (Plate 5.49). Below 500 um, tobermorite abundance and distribution was similar to that of the flooded paste (Plate 144 5.50). Secondary gypsum growth occurred in the deeper portion of the paste (Plate 5.51) which could have been a source of internal stress within the sample. N o gypsum was found closer to the surface. N o portlandite, ettringite or monosulfo-aluminate was identified in any area of the cycled sample. A vertical fracture traversing the height of puck was observed under S E M (not visible to the naked eye) (Plate 5.52). Dendritic and platy crystal forms of Tobermorite were observed lining the fracture walls (Plate 5.53) indicating that the fracture was formed during leaching rather than during later-stage sample preparation. Secondary fracturing reveals the occurrence of internal stress after only 20 weeks of leaching in water. iii) Ferric Sulfate-Leached Environment Layer 1: 0 - 0.75 mm Tobermorite in the outermost layer was most depleted in the ferric sulfate environment (Plate 5.54), however, not so much as in samples containing O P C binder in a similar environment. Its morphology was slightly altered compared to deeper areas, the masses were more rounded and amorphous (Plate 5.55), possibly resulting from an altered chemistry due to leaching of some elements, as described by Taylor (1997). Tobermorite content and increased again at 1 mm (Plate 5.56), regaining its more typical morphology. A partly separated crust was present on the surface of the puck, visible under S E M (Plate 5.57). N o obvious secondary minerals were observed on the walls of the fracture 145 suggesting that fracturing may have occurred during sample preparation. The constant thickness of the fracture indicates, however, that it developed along an existing plane of weakness. Layer2: 1 - 3 mm Like in Tizapa and Brunswick, some iron sulfate precipitates were observed in this layer. The morphology of these precipitates differed from the other samples. They occurred as larger botryoidal masses (20 um - Plate 5.58) rather than the loosely bound oolitic masses occurring in Brunswick and Tizapa. E D X analysis suggests that these precipitates may be a form of jarosite (Plate 5.59) or some other hydrated iron sulfate phase. N o portlandite, gypsum or monosulfoaluminate phases were identified in this sample at any depth. Long, rectangular-shaped crystals (Plate 5.60) were analysed but the E D X spectrogram was not sufficiently precise to provide a definite identification. The morphology was similar to that of ettringite. Tobermorite gel was well developed (Plate 5.61) although slightly less abundant than deeper towards the core of the sample. Below 3mm depth, tobermorite was abundant, well developed and wel l distributed, similar to that encountered in the core of the flooded sample. N o gypsum or monosulfoaluminate were observed although the rectangular ettringite-shaped crystals were still present. 5.7.4 Francisco I. Madero F I M samples contained the lowest proportion of cement (3%) of all paste mixtures. 146 Macroscopically, neither the flooded nor the cycled water-leached environments displayed visible alteration layers after 20 weeks of leaching. S E M observations revealed few differences between superficial layers and the cores of the samples. i) Flooded Water-Leached Environment Layer 1: 0 - 1 mm The outer layer of the flooded paste showed a good distribution of well-developed tobermorite (Plate 5.62), similar to Brunswick tobermorite distribution although F I M contains 60% of the cement content. N o portlandite or ettringite were identified in this layer or deeper. Core: > 1 mm A gradual decline in tobermorite content and degree of development was visible between -750 urn and 1mm, below which the tobermorite appeared as fibrous and scattered, incompletely covering the tailings grains (Plate 5.63). This morphology is suggested in the literature to be typical of decalcified tobermorite (Taylor, 1997). These phases were, unfortunately, too scattered and thin to provide reliable E D X analyses to measure the calcium to silicon ratio. ii) Cycled Water-Leached Environment Layer 1: 0 - 1 mm Tobermorite was moderately well developed but less abundant than in the flooded 147 sample. The cement phases in the outer layer appeared as masses of small individual crystals (<lum size) gathered in bunches rather than intergrown (Plate 5.64). This morphology was encountered throughout the cycled sample and was distinct from the morphology encountered in any other mixture, where it appeared as amorphous masses. These cement phases could not be accurately identified, as the crystals were too small and not sufficiently abundant to obtain reliable E D X analyses. N o secondary minerals such as portlandite, ettringite or gypsum were observed at any depth in this sample. iii) Ferric Sulfate-Leached Environment Layer 1: 0 - 1 mm Plates 5.65 and 5.66 show the porous but homogeneous aspect of the outer edge of the sample. The small proportion of cement in the mixture appeared evenly distributed, showing better tailings grain cover than Tizapa in a similar environment (Plate 5.67). Wavy, leaf-like minerals, most likely monosulfoaluminate, were discernible deeper, at 1.8 mm (Plates 5.68). Acicular ettringite could not be positively identified in this area or anywhere in the sample. The presence of monosulfoaluminate over ettringite suggests that relatively low concentrations of aqueous C a 2 + and SO42" ions were present in the pore solution in the area - see reaction (5.2). This may be a consequence of the low sulfate levels present in the F I M tailings, the lowest of all the samples. Cement cover of tailings (pyrite) grains appeared to be slightly improved at 4 mm depth compared to the surface (Plate 5.69). 148   Tizapa - Flooded-Leached Paste Samples Plate 5.9 Portlandite cover on sample surface Plate 5.10 Bottom of layer 1 Plate 5.11 Bottom of layer 1 Plate 5.12 Poorly developped Tb, below layer 1 Plate 5.13 Core of sample Plate 5.14 Primary ettringite on incomp- letely hydrated cement grain (Tea), core of sample Pd = portlandite; Tb = tobermorite gel (amorphous); Py = pyrite; + = area of E D X analysis 151 Tizapa - Cycle-leach Paste Samples Plate 5.15 Poorly developped Tb, layer 1 Plate 5.16 Ettringite, upper layer 1 Plate 5.17 Secondary gypsum, upper layer 1 Plate 5.18 General paste aspect, layer 1 Plate 5.19 Ettringite in core of sample Plate 5.20 Good tobermorite development at core of sample Et = ettringite; Tea = tricalcium aluminate (unhydrated cement grain); G y = gypsum 152 Tizapa - F e 2 ( S 0 4 ) 3 Solution-leached Paste Samples Plate 5.21 Porous layer 1 Plate 5.22 Depleted tobermorite, layer 1 Plate 5.23 Area of good tobermorite, layer 1 Plate 5.24 Layer 1, iron sulfate precipitate 0 J lFe A l | ' J F I 1 e 1 I T — ~T 1 ' f 1 0 2 .0 4.0 6.0 B.O 10.C Plate 5.25 EDX of iron sulfate precipitate 153 Plate 5.26 Good tobermorite cover, layer 2 Brunswick - Flooded-leach Paste Samples Plate 5.29 Less resistant tobermorite, layer 1 Plate 5.30 Abundant tobermorite, layer 2 154 Brunswick - Cycle-leach Paste Samples Plate 5.31 Good tobermorite cover, layer 1 Plate 5.32 General aspect of paste, layer 1 Brunswick - F e 2 ( S 0 4 ) 3 Solution-leach Paste Samples Plate 5.33 Paste surface Plate 5.34 General aspect of paste, layer 1 Plate 5.35 Poor tobermorite cover, layer 1 Plate 5.36 Possible 2ry ettringite, layer 1 155 Brunswick - Fe 2(S04)3 Solution-leached Paste Samples Plate 5.37 General aspect of paste, layer 2 Plate 5.38 Fe-sulfate precipitate, layer 2 Plate 5.39 Inside large pore, layer 2 Plate 5.40 General aspect of paste, layer Plate 5.41 Layer 3 156 Louvicourt - Aspects of Hydrated Slag-OPC binder Plate 5.42 Massive tobermorite Plate 5.43 Dendritic tobermorite hi t II 1 s c* A r "~ —— i— — • i- • • i i i 0.0 2.0 4.0 6.0 8.0 10.0 Plate 5.44 EDX of tobermorite developed from slag Plate 5.45 General aspect of paste mixture 157 Louvicourt - Flooded-leached Paste Samples Plate 5.46 Less tobermorite in upper layer 1 Plate 5.47 More abundant tobermorite in lower layer 1 158 Louvicourt - Cycle-leached Paste Samples Plate 5.48 Poorly developed Tb, upper layer 1 Plate 5.49 Upper layer 1 Plate 5.50 More developed Tb, lower layer 1 Plate 5.51 Secondary gypsum, lower layer 1 Plate 5.52 Tobermorite-Iined fracture in paste Plate 5.53 Close-up view of tobermorite 159 Louvicourt - Fe^SCVb Solution-leached Paste Samples Plate 5.54 Depleted tobermorite, layer 1 Plate 5.55 Masses of tobermorite at upper layer 1 Plate 5.56 Increased Tb content, below layer 1 Plate 5.57 Fracture developed possibly along existing plane of weakness 160 Louvicourt - F e 2 ( S 0 4 ) 3 Solution-leached Paste Samples Plate 5.58 botryoidal mass of iron sulfate precipitate Ca * I i e Fe L - A >C—X-l—IT^ 1 1 | 0.0 2.0 1 — •• 1 ' ' 1 4.0 6.0 8.0 10.0 Plate 5.59 EDX of iron sulfate precipitate (type of jarosite?) Plate 5.60 Possibly ettringite, layer 2 Plate 5.61 Well developed tobermorite, lower layer 2 161 Francisco I. Madero - Flooded-leached Paste Samples Plate 5.62 Layer 1, flooded sample Plate 5.63 Core of flooded sample Francisco I. Madero - Cycle-leached Paste Samples Plate 5.64 Layer 1, hydrated cement phase (may be Et, Pd, Tb and/or Gy) 162 Francisco I. Madero - F e 2 ( S 0 4 ) 3 Solution-leached Paste Samples Plate 5.67 Layer 1 Plate 5.68 Monosulfoaluminate (?), below layer 1 Plate 5.69 Core of sample M s a = monosulfoaluminate 163 5.7.5 Summary of Observations from Scanning Electron Microscopy Distinct secondary precipitates were observed between mixtures containing only O P C binder and the Louvicourt mixture containing 80% slag in its binder. In general, the tobermorite developed from the slag-cement was observed to be more resistant to leaching in all environments. Flooded Water Environment: • In O P C binder mixtures, Euhedral portlandite crystals were present in relative abundance in the outer edges of flooded water-leached samples exclusively. The abundance of portlandite appeared to be proportional to the cement content of the sample. The longer contact time between the water and the sample in the flooded environment most likely allowed for the pore water to become supersaturated with calcium hydroxide and re-precipitate this phase. These observations agree with the solid phase and leachate chemistry data. • Tobermorite, the main binding agent of cement, was present in smaller amounts in the outer edges of the samples compared to the cores. The amount of depletion was much less in the slag-cement Louvicourt mixture. In all cases and all depths, the tobermorite that was present was evenly distributed and well developed. • Ettringite was easily discernible in the core of samples that contained the largest amount of O P C binder (Tizapa and Brunswick). This suggests that dissolved C a 2 + and SO42" were abundant in the pore solution in the core of these samples but were 164 leached out on the edges. The concentration gradient would drive a gradual and continued depletion of these ions from deeper areas within the paste. • N o secondary, expansive minerals such as ettringite, gypsum or monosulfoaluminate were observed in the slag-cement binder mixture. This attests to the increased resistance to water-leaching and sulfate attack of this binder mixture. Cycled water environment: • N o portlandite was found in the outer edges of any cycle-leached sample. Supersaturation of portlandite was most likely inhibited in the cycled environment as O H - ions were consumed to neutralize the leaching solution, and C a 2 + was dissolved and flushed out of the system before equilibrium could be reached. • N o portlandite was observed in the slag-binder mixture as expected, as slag generally generates little portlandite upon hydration, another feature of sulfate resistance (Taylor, 1997). • Tobermorite was depleted in the outer edges of the cycled water-leached pastes relative to the sample cores, the extent of which appears to be greater than in the flooded environments for both the O P C and slag-cement binder mixtures. • Presence or greater abundance of grain-shaped voids in the outer edges of the O P C binder pastes suggests a decreased cement binding strength in these areas relative to the core. The loss in binding strength may be caused by the lower amount of cement and/or by a degradation (i.e. decalcification) of the cement phase in this area relative 165 to the core. • Although tobermorite was depleted in the outer layer of the slag-cement mixture, no apparent loss of binder strength or empty grain cavities were observed. • Secondary, euhedral gypsum crystals were observed in the outer layers of the Tizapa sample along with acicular ettringite. Secondary gypsum was also observed in the Louvicourt paste below 4 mm depth. These minerals indicate that a source of internal stress created by the pressure of crystal growth is present within the backfill. Ferric sulfate environment: • Tobermorite was depleted in the outer edges of the ferric sulfate-leached pastes, to a similar extent than in the cycled environments, for mixtures of both binder types. The level of depletion was much lower in the Louvicourt sample than in the others. • Oolitic or botryoidal. masses of iron sulfate precipitates (possibly jarosite) were encountered in the 2 n d layer of alteration for all paste mixtures. This location may indicate the penetration front of the ferric sulfate leach solution at the interface with the high p H pore solution of the samples. • Phases of ettringite, monosulfoaluminate and/or gypsum were present at various 2"T" 2 depths in the ferric sulfate-leached samples, indicating that dissolved C a and SO4 " ions were present in sufficient quantity in the pore solution. These ions were therefore not completely depleted from the pore solution in any area of the ferric sulfate-leached samples. 166 5.8 COMPRESSIVE STRENGTH MEASUREMENTS Unconfined compressive strength (ucs) measurements were carried out on unleached paste samples and on the 5, 10 and 20-week leached samples of all 3 environments. The results shown on Figures 5.48 through 5.51 should be interpreted in a qualitative manner since only one sample was tested per environment, per leaching period. Ucs requirements by each respective mine were achieved for all paste mixtures and surpassed for the Tizapa mixture (Table 5.9). Brunswick, Louvicourt and F I M ucs appeared to remain constant as leaching progresses, regardless of the leaching environment. A small decrease in ucs was observed in the Tizapa paste in all environments as leaching progressed. This loss may not be statistically significant. Alteration of the cement phases upon leaching was not only expected, from the literature review, but was also observed under S E M . The layers where alteration occurred were probably too thin after 20 weeks of leaching to influence the compressive strength of the entire sample. Unconfined compressive strength of any material is a function of the surface area of the sample subjected to compression according to the formula: (5.3) 167 where F is the force applied to the sample by the compressor, in kg, (multiplied by gravity) and A is the area of the sample in contact with the piston. Postulating that the loss of cementing material seen under S E M resulted in a loss of strength in the oxidized layers, then the unaltered core of the samples should possess higher strength than the unleached samples. Although no measurements were made to verify this, it is possible that the cores of the leached samples gained strength over the leaching period because o f the longer curing time of the samples in an aqueous environment. This would suggest that uniaxial compressive strength may not constitute an appropriate measuring tool to characterize the strength of weathered backfill samples. 168 o co c 3 CO 3 O O "> 3 O o w OJ T 3 o >f T J a> X 3 o o CO a. CO N —I o ml T J CD o >. o I T J a> o o u_ d d CO co •o d m co • o CNJ c cu E cu cr cu a: co o d co T J 0 JZ o ro c CO 00 CO o d d •̂ r CM co o d d •sr t - CN o d d CN o o o co CD d oo d d CD CO N~ CN csi CO CD CD 5 LO T - 169 Flooded Cycled Fe2(S04)3 soln 0 5 10 20 Weeks of leaching Figure 5.48 Tizapa Paste - UCS with Leaching Time -Flooded -Cycled -Fe2(S04)3 solnl 5 10 Weeks of leaching 20 Figure 5.49 Brunswick Paste - UCS with Leaching Time -Flooded -Cycled -Fe2(S04)3 soln 0 20 5 10 Weeks of leaching Figure 5.50 Louvicourt Paste - UCS with Leaching Time 170 2 Weeks of leaching Figure 5.51 Francisco I. Madero Paste - UCS with Leaching Time 171 6 DISCUSSION 6.1 PYRITE REACTIVITY STUDY The mineral surface characterisation technique of cyclic voltamperometry provided a measure of the combined effects of the different mineral characteristics on the reactivity of pyrite, such as chemistry, crystal morphology, stoichiometry and presence o f other sulfide phases associated with pyrite (termed mineral impurities). This method of analysis established that, initially, Huckleberry pyrite was the most reactive of all samples studied, most likely due to its relatively low concentration of separate sulfide phases. Conversely, Louvicourt-2 and Tizapa pyrites were the least reactive because of the considerable amount of galena and sphalerite texturally associated with pyrite in the samples. This study suggested that the presence of these mineral phases in the pyritic samples effectively decreased the reactivity of pyrite to a greater extent than either stoichiometric proportions of Fe:S, chemical impurities in the crystal lattice, crystallinity or morphology of pyrite crystals could have increased the reactivity of pyrite. The following sections discuss the effects of the various mineral characteristics observed in this study, on the reactivity of pyrites. 6.1.1 Effect of Precipitate Coatings on the Passivation of Pyrite: Huckleberry and Louvicourt-1 Samples Higher dissolved iron and sulfate concentrations in the leachate supported the voltamperometric data suggesting that the unleached Huckleberry pyrite was the most 172 reactive of all samples analyzed. The relatively high reactivity of Huckleberry pyrites may, in part, be caused by the greater amount of fine size particles in the sample. Grain size is not believed to be the only cause of Huckleberry reactivity since the finer grain size Brunswick sample possessed the lowest reactivity after 20 weeks of leaching. A s leaching progressed, however, both water and waste rock solution-leached Huckleberry pyrites were rapidly oxidized and became less reactive. The amount of iron oxyhydroxide precipitates on pyrite surfaces that were visible under S E M formed a very discontinuous layer even after 20 weeks of leaching. These precipitates alone may not account for the decreasing reactivity exhibited in the voltamperometric study. The visible precipitates may be however, local accumulations of a more extensive F e O O H precipitates and/or polysulfide layers not visible under S E M . Mineral coatings of F e O O H on oxidized 1 pyrite were observed by many researchers. Nicholson and others (1990) measured a continuous a-Fe203 (maghemite) precipitate layer on pyrite of an average of 0.6 urn thickness after 400 days of leaching. In that study, the chemistry of the precipitate layer was believed to be a dehydrated form of y- F e O O H (lepidocrocite) transformed by the high vacuum of the auger electron spectroscopy (AES) instrument, according to: 2 FeO(OH) -> F e 2 0 3 + H 2 0 (6.1) The presence of such thin oxidation rims on pyrite were also documented by Alpers and others (1994) and Jambor (1994). Many other researchers also documented the formation of iron oxide, iron oxyhydroxide, and/or sulfur, sulfoxianion or polysulfide layers on the 173 pyrite surface following electrode oxidation, all of which effectively decreased the leachability (or reactivity) of the underlying pyrite (Peters, 1984; Buckley et al., 1988; Ahmed, 1991; Zhu et al., 1992; Ahmed and Giziewicz, 1992). In addition, Moses and Herman (1991) proposed that accumulation of Fe on pyrite surfaces in early stages o f leaching effectively decreased the rate of pyrite oxidation and that F e O O H deposits on the pyrite surfaces in slowly-stirred, long-term leaching experiments were the cause of decrease pyrite oxidation rates. Although no specific information is available on the composition of the oxidized layers observed by S E M on pyrite grains in many samples, literature indicates that thin F e O O H , sulfur or polysulfide layers could be responsible for the increasingly efficient passivation of Huckleberry pyrites or any of the other pyrites studied. Louvicourt-1 and -2 samples came from the same mineral deposit but exhibited large differences in their chemical and stoichiometric composition as well as in their oxidation behaviour. The unleached Louvicourt-1 pyrite sample showed a similar voltamperometric response to the Huckleberry sample and, likewise, a similar evolution of reactivity with increasing leaching-time. The more abundant, amorphous iron precipitates covering the surfaces of Louvicourt-1 pyrites did not appear however, to result in a greater loss of reactivity. It is probable that sulfur or polysulfide layers could have been more instrumental in passivating the pyrite surfaces than the apparently discontinuous iron oxyhydroxide precipitate coatings. 174 6.1.2 Effect of Stoichiometry on Pyrite Reactivity: Louvicourt-1 and Louvicourt-2 Samples Pyrite stoichiometry was not found a determinant factor in the reactivity of pyrite compared to the presence of sphalerite. The relative depletion of iron in the Louvicourt-1 pyrite structure consisted of an acceptor defect that should have theoretically protected the pyrite from oxidation (Shuey, 1975; Kwong, 1993) compared to Louvicourt-2 pyrite having a S:Fe ratio of exactly 2.0. Similarly, although arsenic contents were similar for both Louvicourt samples, S E M observations showed that Louvicourt-1 arsenic was mainly contained in separate arsenopyrite phases whereas Louvicourt-2 arsenic principally occurred as lattice impurities within pyrite. The presence o f arsenic impurities in the crystal lattice of pyrite creates a donor defect, which should theoretically promote the oxidation of Louvicourt-2 pyrite (Shuey, 1975; Kwong, 1993)., Voltamperometric analyses of the unleached Louvicourt pyrites indicated, however, that Louvicourt-2 pyrite was the less reactive one. The presence of arsenic may have influenced the way in which the Louvicourt-2 pyrite sample oxidized, as a greater density of oxidation pits were visible on Louvicourt-2 pyrites compared to Louvicourt-1 after 10 and 20 weeks of leaching. In summary, for all the pyrites studied, the effects of iron to sulfur ratios and the presence of arsenic impurities in the pyrite crystal lattice were overcome by an opposing, more determinant factor on the reactivity of pyrite. This factor is believed to be the presence of sphalerite and galena impurities in the pyrite samples, which offered galvanic protection to pyrite. 175 6.1.3 Effect of Mineral Impurities on Pyrite Reactivity 6.1.3.1 Galvanic Protection by Sphalerite: Louvicourt-2 and Tizapa Samples Tizapa and Louvicourt-2 pyrite samples possessed similar mineralogies and showed a similar evolution of pyrite reactivity. Both were characterized by an initial gain in reactivity after 4 weeks of leaching, followed by a decreasing reactivity. This study suggests that during the leaching period, sphalerite grains in contact with pyrite acted as an anode and were preferentially oxidized over pyrite. The cathodic reaction could involve the reduction of oxygen, as described by Murr and Metha (1983): Anodic reaction: sphalerite oxidation ZnS -> Z n 2 + + S° + 2e- (6.2) Cathodic reaction: reduction of water lA 0 2 + 2 H + + 2 e - ^ H 2 0 (6.3) Kwong and Lawrence (1994) also proposed that the formation of galvanic couples affected the reactivity of pyrite to a greater extent than mineralogical defects, stoichiometric imbalances, grain size, crystallographic orientation or bacterial activity. Results of both this and the Kwong and Lawrence studies are supported by the research of Metha and Murr (1983) who measured the contribution of galvanic interaction on the reactivity of pyrite in an acidic solution of 1 M H2SO4. They proposed that galvanic interactions were responsible for the cathodic protection of a sulfide mineral with higher rest potential (pyrite) in contact with a mineral of lower rest potential (sphalerite, 176 chalcopyrite) which acts as the anode and oxidized preferentially. The gain of reactivity in both Louvicourt-2 and Tizapa samples coincided with higher concentrations of zinc in both sample leachates, following reaction 6.2. S E M observations showed that in both samples, when sphalerite and pyrite occurred on the same grain, sphalerite phases were extensively corroded compared to pyrite phases. Pyrite generally possessed smooth surfaces and sharp grain edges in the 4 t h and 10 t h weeks for Tizapa and Louvicourt-2 respectively. The subsequent passivation of pyrite coincided with the decline in leachate zinc concentrations, occurring around the 5 t h week (10 t h cycle) for Tizapa and the 10 t h week (20 t h cycle) for Louvicourt-2. Indeed, in later leaching cycles S E M observations showed sphalerite grains becoming precipitate-covered and corroded, suggesting that sphalerite could also have been passivated and were no longer available to oxidize and offer galvanic protection to pyrite. Sphalerite passivation may be caused by a covering of the surface by elemental sulfur shown in reaction 6.2. Increases in the intensity of pyrite pitting were observed in the 10 t h week for Tizapa and 20 t h week for Louvicourt-2, suggesting a more intensive oxidation of pyrite following the decreased oxidation of sphalerite. The increasing supply of partly oxidized sulfur and dissolved iron to the leachate could then have promoted the formation of iron oxyhydroxide and/or sulfur- polysulfide coatings on all grains, effectively decreasing pyrite reactivity. A decrease in the reactivity of pyrite was indeed observed electrochemically in the 10 t h and 20 t h week of leaching in the Tizapa and Louvicourt-2 samples respectively. The absence of sphalerite in the Louvicourt-1 sample probably enabled 177 that pyrite surface to oxidize from the first leaching cycle onward, allowing for an earlier production of a passivating layer. Voltamperometric studies indicated that Louvicourt-1 pyrite was more reactive prior to leaching but was passivated at an early stage. 6.1.3.2 Effect of the Presence of Galena: Zimapan and Brunswick Samples The presence o f galena was detected on the Zimapan and Brunswick voltamograms by the occurrence of a characteristic peak around 0 m V . In that way, a concentration of galena as low as 1.5 % was electrochemically detected in the Zimapan pyrite sample. A more intense peak was observed in the Brunswick sample that contained a higher galena content (9%). The amplitude of peaks attributed to galena appeared to be proportional to the galena content of the sample. The fact that galena is highly susceptible to oxidation is well documented in the literature (Metha and Murr , 1983; Nicholson, 1994; Alpers et al, 1994; Jambor, 1994; Blowes et al., 1995). In this study, galena was also found to oxidize rapidly: only traces of galena were present in the Brunswick sample after 10 weeks of leaching. Galena was replaced with anglesite (PbSC^) either by direct replacement (i.e. in Tizapa where anglesite is found covering skeletal galena) or by precipitation from solution (i.e. in Brunswick and Zimapan where anglesite is dispersed throughout the sample). The low solubility of anglesite (Table 6.1) explains why dissolved lead readily precipitated as lead sulfate in the sample effectively scavenging the dissolved lead from the leachate solutions. Unlike lead, dissolved zinc does not tend to precipitate as a secondary zinc hydroxide or sulfate but tends to remain in the leachate solution. In the case of Brunswick, it is possible that 178 the leachate solution was oversaturated with respect to anglesite due to the large amount of galena being oxidized. Table 6.1 Dissociation Constants for Secondary Minerals of Zn and Pb Zinc Dissociation Secondary Minerals : Constant ! dog K.y I .cad Secondary Minerals Dissociation Constant ( IogK) ' Z n S 0 4 3.0 P b S 0 4 -7.8 Z n 2 ( O H ) 2 S 0 4 7.5 P b 3 0 2 S 0 4 10.4 Z n 4 ( O H ) 6 S 0 4 28.4 P b 4 0 3 S 0 4 22.1 Z n ( O H ) 2 11.5 Pb(OH) 2 8.2 '(Parkhust, 1995: PhreeqC database updated 1996) 6.1.4 The Huckleberry Problem The potentially acid generating Huckleberry pyrite was found the most reactive of all pyrites tested. This classification may be due in part by the larger fine grain proportion of the sample, increasing the surface area of the sample and possibly increasing the oxidation rate of pyrite (McKibben and Barnes, 1986; Kwong 1993). This is not necessarily the principal cause of the Huckleberry's high reactivity, however, since the Brunswick sample, which had a higher proportion of fines, exhibited the lowest reactivity (following the disappearance of galena in the sample). A n important factor remains that the reactivity of Huckleberry pyrite declined from the 4 t h week onward. Although Huckleberry remained the most reactive pyrite sample throughout the 20-week leaching period, the waste rock solution-leached pyrite showed the greatest loss of reactivity (0.07 volts) after Zimapan (0.1 volts). The reactive nature of these pyrite samples was progressively overcome by the precipitate coatings, which are believed to have caused the reactivity decrease. 179 The slight gain in reactivity observed in the 20 week of leaching in the solution-leached Huckleberry pyrites as well as in Tizapa and Louvicourt-2 samples, however, suggested that surface properties of the pyrite (or its precipitate cover) continually evolved during leaching, affecting the reactivity of pyrite. Evolution of surface properties may include changes in precipitate chemistry or morphology, bacteriological alteration of grain surfaces, or many other factors. The electrochemically observed changes in reactivity could be paralleled with the constantly evolving chemistry of mineral waste solids and leachates, making predictions difficult to establish with an acceptable degree of certainty. 6.2 CHEMICAL STABILITY OF PASTE MIXTURES 6.2.1 Effect of Cementing Tailings Waste Binder (Portland cement and slag-cement mixtures) does provide some buffering capacity to the tailings. The small proportion of binder added to the reactive tailings studied (less than 6% in this study) is, however, far from sufficient to neutralise all the potential acidity of the tailings studied which had N P R values below 0.1. N P R values of all the backfill mixtures studied remained well below 1 after cement addition, remaining potentially acid generating like the tailings that make up the backfill. Paste p H measurements suggest that both binders (OPC and slag-OPC) contain readily dissolved buffering minerals that provide efficient, early-stage buffering capacity to the backfill. A B A analyses as a function of leaching time indicated that binder N P is readily consumed in the early stages of contact with water or A R D solution. Depletion is observed to be more severe in A R D solution for all cases studied. A s a result, even i f 180 cementing material was added initially in sufficient quantity to raise the N P R of the backfill to environmentally safe levels (i.e. N P R > 3), the N P may be depleted in this environment at a very early stage, leaving the tailings unprotected. In most cemented backfilling operations, the backfill is exposed to the environment almost immediately after placement. This study suggests that exposure of the backfill to an aqeous leaching environment prior to being properly cured w i l l promote the loss of binder material, which, in the cases studied, accounted for the majority of backfill buffering capacity. This could potentially be accentuated where slag-cement binders are used, as slag is typically slower to react (hydrate and cure.) than O P C . The loss of material was further accelerated in A R D solution. Fully cured backfill could, however, show a different response to leaching. Conversly, i f the backfill is not exposed to any leaching solution, for example in a dry area of the mine, the cement content should be conserved until the mixture is properly cured. In the case of Tizapa paste backfill which contained the highest proportion of cement and the highest proportion of sulfide minerals in the tailings, surficial oxidation of the samples began to occur within the 14-day curing period. This suggests that the sulfide portion of the waste was not sufficiently covered or protected by binder to prevent its oxidation in air. Paste containing such high proportions of reactive sulfides would likely require complete submergence in water at an early stage to prevent oxidation of the sulfide portion of the tailings. Such a material may not be suitable for storage above ground, or underground above the water table, where it could be exposed to air and humidity for extended periods of time. 181 6.2.2 Cemented Backfill Alteration with Leaching In 20 weeks of leaching, differences in alteration were more influenced by the leaching environment and binder content (OPC and slag-cement) of the paste mixtures than by tailings composition. Binder content was found to be the most important factor affecting leachate water quality. Calcium, potassium, silicon and magnesium loadings in the leachates were found to be directly proportional to the solid phase concentration of these elements in the paste mixture, in turn, proportional to the binder content of the mixture. 6.2.2.1 Water-Leached Environments The leachate and solid phase geochemistry along with A B A analyses, indicated that calcium concentrations in the leachate came principally from the dissolution of the portlandite phase (Ca(OH) 2 ) of cement, where O H - ions were consumed to buffer the p H of the leaching solution. This agrees with previous investigations on the effect of water leaching of cemented mixtures (Adenot and B u i l , 1992; Revertegat, 1992; Benzaazoua, 1997). E D X microanalyses under S E M were not sufficiently precise (due to uneven sample surfaces) to measure a quantifiable decalcification of the tobermorite, as expected from the literature review. The modified morphology of the exposed tobermorite observed in altered regions, however, was similar to the morphological descriptions of Ca-depleted tobermorite by Adenot and B u i l (1992), Carde and Francois (1997) and Taylor (1997). They have found that in situations of severe leaching, tobermorite generated from both O P C and slag-OPC binder lost its volume and took the shape of dehydrated gel-like (silica-rich) filaments, loosing binding strength in the process. 182 The Tizapa and Brunswick mixtures containing the highest percentage of pyrite and largest proportion of binder showed deeper penetrating alteration layers compared to the Louvicourt and F I M mixtures. The tobermorite developed from the slag binder in the Louvicourt paste was considerably less depleted than O P C tobermorite in all environments and showed generally fewer expansive minerals after 20 weeks of leaching. This indicates that the slag-cement binder of Louvicourt offered increased resistance to water leaching as well as to sulfate attack compared to OPC-binder backfills. Flooded vs Cyclic Water-Leached Environments Portlandite supersaturation was achieved in the flooded environment for both the Tizapa and Brunswick samples, which contained the highest proportion of binder. Masses of euhedral portlandite lined the outer edge of the samples and were observed to 1 m m and 3 mm depths respectively after 20 weeks of leaching. N o portlandite was observed in the Louvicourt or F I M pastes. Penetration of the leachate solution and the resulting concentration gradients of calcium and hydroxyl ions between the sample pore solution and the leachate drove the dissolution of portlandite and the diffusion of these ions out of the sample. The chemical conditions at the sample edge may have induced the precipitation of secondary minerals at that location. The same can be suspected to have occurred for all soluble backfill components. Supersaturation conditions must occur in the leachate solution before the secondary mineral may nucleate and grow. Longer contact times between the solution and the paste, such as in the flooded environment simulated in this study, facilitate the 183 achievement of supersaturated conditions and precipitation of secondary minerals such as gypsum or portlandite. S E M observations showed, however, that no expansive ettringite or gypsum was present after 20 weeks of leaching in the flooded environment in either Tizapa or Brunswick samples. Instead, secondary gypsum and ettringite were encountered in the outer edges of the cycled environment samples. It is clear then that in the case of cemented tailings backfill , the occurrence of secondary minerals is a solubility-controlled process. Table 6.2 presents the ranges of solubility constants (Kps), associated solubilities in water (in mg/1) and equilibrium constants (Keq) of the secondary minerals encountered in concrete mixture. Lime, calcite and salt (halite) are also shown for comparison. Mineral solubility can be evaluated using the solubility product constant (Ksp) in instances where dissolution of the mineral in pure water is not affected by water p H . Given any dissolving mineral ' A B ' : x A B <-» y A + zB (6.2) Ksp is evaluated from the activities of species (activity of a solid =1) according to: K s p = [ A ] y x [ B ] z (6.3) 184 Table 6.2 Solubility Products and Dissociation Constants of Secondary Minerals Derived from Hydrated Portland Cement and Other Reference Minerals Mineral and Dissociation Reaction Solubilit\ Product (Ksp or molcs/1 at Solubility (in water at 25"C) * Dissociation ( onslant (Kcqa t25 l 'C ) Portlandite: Ca (OH) 2 + H + -> C a 2 + + 2 H 2 0 5.5 x 10"° ( 2 ) 3 . 2 x 10"5 depends on p H f o r pH<10.7 " ' " " r ' 6 ' 6 . 3 x l 0 2 2 " Ettringite: C a 6 A l 2 ( S 0 4 ) 3 ( O H ) i 2 - 2 6 H 2 0 -> 6 C a 2 + + 2 A l 3 + + 3 S 0 4 2 " + 12 0 H " + 2 6 H 2 0 P ) 1 . 2 3 x 1 0 ^ (4) ! 0 - l 12 depends on pHfo r p H < l l not available Gypsum: C a S 0 4 - 2 H 2 0 -> C a 2 + + S 0 4 2 " + 2 H 2 0 w 9 x 10"° ( 8 ) 2 . 5 x l 0 " 5 w 520 mg/1 ( 7 ) 2100 mg/1 l 0 3 . 2 x 10"3 Calcite: C a C 0 3 -> C a 2 + + C 0 3 2 " (1> 2 . 8 x 1 0 * ( 8 ) 4 . 5 x l ( T 9 (i> 3 mg/1 ( 8 ) 4 mg/1 ( 5 ) 3 . 1 x l 0 " 9 Lime: CaO + H 2 0 -> C a 2 + + 2 O H - w 6 . 3 x l 0 3 2 Halite (salt): N a C l -> N a + + C l " ^ 3 . 4 x 1 0 ' K " 360 000 mg/1 w 3 . 4 x 10 1 * Solubility calculated from Ksp values unless stated otherwise; (1) Brown and Lemay, 1977; (2)Duchesne and Reardon, 1995; (3) Myneni et. al., 1998; (4) Warren and Reardon, 1994; (5) Appelo and Postma, 1994; (6) Pakhurst, 1995 (PhreeqC database updated 1996); (7) Freeze and Cherry, 1979 (Ksp values calculated from solubilities); (8) Fetter, 1993 For minerals affected by p H , such as some oxides and hydroxides, mineral solubility is better represented by the equilibrium constant (Keq). Keq is determined from the law of mass action, which states that the rate of chemical reaction is proportional to the activity of the participating substances (including water): Given any two substances: w A + x B <-> y C + zD (6.4) 185 K e q is evaluated from the activity of species (activity of solids and H 2 O =1) according to: K e q = r C P x r D f (6.5) [ A f x [B] x In minerals such as calcite, the relationship between Ksp and K e q is explained by: CaCO-3 (in water) <-> C a 2 + + C 0 3 2 " (6.6) Keq = Ksp = [Ca 2 + ] x [C0 3 2 "] = 3.2 x 10"9 (6.7) However for portlandite, Ksp and K e q differ because of the dependence of portlandite on the p H of water. The simple dissolution of portlandite follows: Ca(OH) 2 C a 2 + + 2 O H ' (6.8) A n d therefore: Ksp = [Ca 2 + ] x [OH'] s 1.9 x 10"5 (6.9) Whereas the solubility of portlandite is expressed as: Ca(OH) 2 +2H'<-> C a 2 + + 2 H 2 0 (6.10) i Keq = r C a 2 + l = 6.3 x 10 2 2 (6.11) [H + ] Equation 6.11 shows that solution p H is inversly propotional to portlandite solubility, explaining why portlandite dissolves more readily in a low p H A R D solution. Portlandite has the same solubility range as gypsum but the above showed that 186 portlandite is much less stable in neutral or acidic p H conditions than gypsum or ettringite. In these conditions, portlandite w i l l readily dissolve and saturate the pore water with calcium. In the cases of Tizapa and Brunswick backfill (highest proportions of cement), secondary portlandite was found in the outer sample edges of the flooded environment only, whereas secondary gypsum was found in the outer edges of the samples in the cycled environment only. One possible explanation for these occurrences is that the pore solution of both water-leached environments becomes supersaturated with portlandite within 24 hours, hence precipitating this phase in the outer edges of the samples. With increasing leaching time, the secondary portlandite redissolves. in response to the declining concentration of calcium in the pore solution as other more stable mineral phases continue to dissolve and supply other elements (i.e. A l ) necessary for ettringite or gypsum to become supersaturated, nucleate and precipitate out of solution. This study suggests that in a flooded environment, where hydraulic conductivity of the saturated material is low and the backfill mixtures contain relatively high proportions of Portland cement, secondary expansive minerals may form. Consequently, in addition to the dissolution and loss of binder components and eventually loss of binder strength, expansive minerals w i l l create internal stresses accelerating the disintegration of the backfill. In cyclic leaching environments where the backfill is exposed to a leaching agent for short periods, the leachate contact time may not be sufficient to precipitate secondary, expansive ettringite and gypsum. Instead, the leachate acts as an ion sink, driving 187 dissolution and depletion of the binder components, especially the very unstable portlandite, and flushing them out before supersaturation and precipitation of expansive mineral phases can occur. Solid phase geochemistry showed that calcium and sulfate were more effectively depleted on the edges of the samples in the cycled environments compared to flooded environment. Cycl ic leaching resulted in an accelerated loss of binder material and, in the case of OPC-binder, a loss of tobermorite integrity in the periphery of the samples as indicated by the more abundant grain cavities and greater depletion of major ions in the outer layers of the cycled samples. The extent of tobermorite depletion was lower in the slag-cement binder mixture of Louvicourt pastes. S E M observations and solid phase geochemistry of the altered layers of all samples, indicate that backfill leaching is a surface phenomenon characterized by alteration layers widening inwards with increasing leaching time. The data obtained in this relatively short term leaching study does not, however, provide information on the long-term rate o f advancement of the leaching front, whether the leaching front would continue to progress or eventually cease its advancement. In all water-leached cases, the augmented porosity resulting from binder dissolution,, combined with the precipitation of expansive secondary minerals suggests that advancement w i l l l ikely continue to proceed as porosity is augmented with time and the leachate can more easily penetrate the backfill. Effect of low binder content The absence of portlandite, gypsum or ettringite in the F I M and, to some extent, in the Louvicourt samples, is likely related to both the small proportion of hydrated Portland cement available to be dissolved and to the generally higher porosity of these samples. 188 High sample porosity w i l l facilitate contact between solute and solvent and the diffusion of dissolved elements out of the sample making supersaturation more difficult to achieve. In addition, i f the initial concentration of portlandite to be dissolved is small, pore solution supersaturation may not be attained. This is especially true in the case of slag- mixed binders where the hydration reaction and formation of tobermorite consumes portlandite (Taylor, 1997; Kosmatka et al., 1995). In this case, supersaturation of high calcium phases (such as ettringite) is more difficult to achieve without a readily available source of dissolved calcium. Gypsum supersaturation can occur, however, as was observed in the deeper areas of the cyclic-leached pastes. This, in time, w i l l create internal pressure within the backfill. Wide-spread secondary gypsum growth would be responsible for internal breakdown and loss of strength of the backfill. 6.2.3 Ferric Sulfate or Artificial ARD Environments Pore water in concrete has a stable p H ~12, the stability p H of portlandite. The artificial A R D solution p H of 2.4-2.5 used in this experiment is considerably more conducive to portlandite dissolution than water. Decreased pore solution p H accelerates the dissolution of portlandite whose hydroxide ions are consumed to buffer the solution p H . This was demonstrated in these experiments by the leachate geochemistry data, where p H was buffered on contact with the pastes in the early cycles of leaching. The A B A tests also attested to the high, early dissolution of portlandite whereby N P depletion was considerably more severe and deeper penetrating in the ARD-leached pastes than in the water-leached pastes for both O P C and slag-OPC binder mixtures. Alteration layers of the ferric sulfate-leached pastes were almost double in thickness to those of the water- 189 leached pastes. Indeed, ARD-leached pastes had a measurable lost of mass in 20 weeks of leaching, contrary to the water-leached pastes. S E M observations of the altered layers of O P C binder backfills showed important losses of tobermorite volume as well as more pronounced changes in tobermorite morphology (likely from the decalcification of tobermorite) and apparent strength losses (increased grain cavities viewed with S E M ) in the altered layers compared to the water-leached samples. Tobermorite loss was slightly lower in the slag-OPC binder (Louvicourt) pastes, but still higher than in water-leached environments. The limited ucs tests carried out on all samples did not show, however, any strength loss with time. This is probably due to the relative thin r im of alteration compared to the diameter of the entire sample. A l l ARD-leached pastes samples were covered by surficial iron hydroxide precipitates. They also possessed similar amorphous, oolitic or botryoidal iron sulfate precipitates, possibly jarosite, in the second alteration layer (2 n d from edge). It remains unclear, however, whether these minerals were precipitated directly from the ferric sulfate solution or precipitated as a product of the oxidation of iron sulfide in these areas. Both cases imply the penetration of an acidic ferric sulfate solution front into the paste. The sulfide minerals would have been oxidized in the acidic environment whereas metal sulfates and hydroxides would have been precipitating on the pore solution side of the front where p H and ionic strength could allow the precipitation of these minerals. In any case, it is possible that continued growth of both or either of these precipitates (iron hydroxide and/or iron sulfate) eventually passivates the surface of samples. The evolution of leachate p H , redox potentials and aqueous iron concentrations 190 towards the values of the leaching solution support the suggestion that the samples were passivating as leaching progressed. It is unclear whether the depressed reactivity of the paste was a consequence of precipitate coatings from the ARD-solut ion (ferric hydroxide precipitates visible on all ARD-leached pastes), or i f it was due to an exhaustion of soluble buffering minerals in the area actively leached. In the latter case, loading rates of paste constituents to the leachate would be proportional to the infiltration rate of the solution into the backfill. Regardless of the processes responsible for the apparent passivation of the paste, the implication is that paste backfill exposed to A R D w i l l eventually lose its ability to buffer the drainage, following which the acidic, metal-loaded mine drainage coming in contact with the backfill would be free to flow unaltered around the backfill to the surrounding groundwater or surface water. Another aspect of the eventual passivation of the paste is that the aggressive sulfate solution would eventually be prevented from penetrating the backfill and degrade the hydrated cement phases of the mixture. In any case, the unchanging sulfate levels in the leachates suggest that the sulfate in the leaching solution may not actively participate in cement altering reactions in the first 20 weeks of leaching. A l l data suggest that in 20 weeks of leaching, external sulfate plays a minimal role in cement-altering reactions when tailings already contain high concentrations of sulfate. The presence of relatively high aqueous zinc concentrations in the Brunswick sample, combined with the results of the pyrite reactivity study, support the possibility of active sulfide mineral oxidation in the outer layers of the paste. Absence of dissolved iron in the leachate indicate one of two possibilities: 1) no oxidation of iron sulfide minerals 191 (pyrite, pyrrhotite), or more likely 2) oxidation or iron sulfides followed by precipitation of iron sulfate within the backfill. The pyrite reactivity study suggests that pyrite may get passivated at an early stage. Verification of the latter would require further investigation. 192 7 C O N C L U S I O N S The following conclusions are made based on the results of this study: 7.1 PYRITE REACTIVITY STUDY • Results of the electrochemical study supported by chemical, mineralogical and leachate chemistry data, suggested that the presence of non-pyrite sulfide mineral impurities, in contact with pyrite was the most important parameter affecting pyrite reactivity in initial leaching cycles. In more advanced stages of leaching, electrochemical analyses together with S E M observations suggested that precipitate coatings, either visible F e O O H coatings or angstrom thick sulfur or polysulfide layers not visible under S E M , played a dominant role in the reactivity of pyrites, effectively decreasing pyrite reactivity as leaching progressed. • The presence o f sphalerite was believed to form a galvanic couple with pyrite until sphalerite itself became unreactive because of precipitate coatings on its surface. Once the availability of sphalerite became limited, pyrite became free to oxidize and develop a precipitate coating, which proceeded to decrease its reactivity. • The extent of galvanic protection offered by sphalerite was proportional to the sphalerite content of the sample. Lower concentrations of sphalerite in Louvicourt-2 is believed to have offered either less efficient or shorter time of galvanic protection compared to the Brunswick sample. Higher leachate concentrations of iron and 193 increased pitting of the Louvicourt-2 pyrite suggested that this sample oxidized to a greater extent than the Brunswick pyrite. Leachate concentrations of zinc were proportional to the sphalerite content of unleached pyrite samples. The high solubility of zinc sulfate and zinc hydroxide explains why no zinc precipitates were found in any of the leached samples. The presence of galena in pyrite samples was signalled by a current release at around 0 mvolts, lower than the potential at which pyrite oxidized in that environment. The intensity of the voltametric peak appeared to be proportional to the concentration of galena in the sample. Contrary to zinc, leachate lead concentrations remained low throughout the leaching period by precipitating out as anglesite (PbSC»4). It is possible that the precipitation of secondary anglesite could have indirectly provided pyrite grains some resistance to oxidation by reducing the surface area of pyrite available to oxidize. After the 20-week leaching period, all pyrites had decreased reactivities with respect to its unleached reactivity regardless of the presence of mineral impurities. In many cases, however, a slight regain of reactivity observed in the 20 t h week of leaching may indicate that surface precipitates were modified, no longer providing increasing levels of passivation. This study does not provide information as to the resistance o f the passivating layers to changing conditions in the leachate. This factor commands caution in the long-term extrapolation of the effectiveness of passivation layers. 194 7.1.1 Huckleberry Pyrite The potentially acid generating Huckleberry pyrite was found to be the most reactive of all samples studied. However, the reactivity of Huckleberry pyrites began to decrease within 4 weeks of leaching, progressively overcoming its reactive nature. Reactivity loss was slightly greater in the waste rock (low-grade ore)-leached sample, possibly because of the greater amount of ions in solution available to for precipitate coatings. The potentially reactive pyrite present in the low grade ore for which were carried out 3-year kinetic tests may also have been reactive but surfaces of the pyrites probably passivated from early cycles onward by the ionically charged leachate solution to which it was exposed. Post leaching mineralogical analyses should have been carried out to characterize the surfaces of these pyrites. 7.1.2 Application of this Study to Field Investigations A s pyrite is never the only sulfide mineral in a deposit or in mineral wastes, this study has shown that, at the scale of the study, occurrence of mineral impurities was the most determinant factor of the reactivity of pyrite, above crystal morphology or stoichiometric imbalances. Optical and electronic microscopy in conjunction with point source chemical analyses permitted to distinguish between solid solutions and separate mineral phases. The mode of occurrence of impurities present with pyrite dictates the ability of that mineral phase to oxidize and provide galvanic protection. If a sphalerite grain has a very small contact area with a pyrite grain or is separated by a micron size phase of a different mineral, galvanic protection could be diminished or ineffective. 195 Some pyrite grains that were not in contact with mineral impurities, for example in the Tizapa and Zimapan samples, oxidized to a similar extent than pyrite grains in the Huckleberry or Louvicourt-1 samples. This indicates how the intrinsic mineralogical heterogeneity of a given geologic deposits or mineral waste pile make predicting A R D generation tentative at best. Although detailed stoichiometry and mineralogy can give indications as to the expected reactivity of a mineral, these characteristics may not be the dominant factors affecting the water quality and/or overall reactivity of a waste rock pile or tailings pond. Megascopic features such as pile or pond construction, hydrology or other site conditions also influence the possible generation of A R D from a given mineral waste as observed by many researchers (Kwong, 1991; Ritchie, 1994; Schafer et al., 1994; Woyshner and St-Arnaud, 1994; Otwinowski, 1997). It is therefore imperative that representative mineralogical samples and megascopic characteristics, such as (but not limited to) multiple grain size fractions that could include sulfide mineral pockets, be included in the make up of rock samples used in kinetic tests in order to make more accurate A R D predictions. A t a fundamental level of understanding the reactivity of pyrite, however, the electrochemical characterisation technique of cyclic voltamperometry on carbon paste electrodes used in this study was an effective tool to demonstrate how the various mineralogical characteristics work together to influence the overall reactivity of pyrite. Cycl ic voltamperometry was found to effectively measure the ability of pyrite to oxidize under the conditions of the kinetic tests carried out in this study. This type of analyses can be revealing in situations where kinetic test results are inconclusive, such as in the 196 case of Huckleberry. The high degree of expertise involved in interpreting the results of an electrochemical study, however, can be an obstacle to the widespread use of this technique as an A R D predictive tool. 7.2 CEMENTED PASTE BACKFILL LEACHING STUDY The small proportion of binder added to reactive tailings in this study was insufficient to rise N P R ratios above 1 or change the status of potentially acid generating tailings to non-acid generating backfill. The addition of cement to reactive tailings does not necessarity constitute an effective prevention against generation of A R D from reactive tailings. Some backfill mixtures that contain a high concentration of reactive pyritic tailings used as the principal aggregate (such as Tizapa backfill) may oxidize in air. This type of backfill would likely require complete submergence in water at an early stage to prevent oxidation of the tailings. Such a material would not be suitable for storage above ground or underground above the water table, where it could be exposed to air and humidity for extended periods of time. • The hydrated portland cement phases of backfill readily dissolve in water. The extent and rapidity of depletion is generally more pronounced in a low p H solution (i.e. A R D ) . The small addition of neutralisation potential provided by the cement is depleted at an early stage in the external layers of the samples, leaving the tailings in these areas exposed to the environment. In 20 weeks of leaching, differences in alteration were in general more pronounced between the different leaching 197 environments than between the different tailings mineralogy making up the backfill mixtures. Binder content (cement or cement/slag) of the paste mixtures was found to be the most important factor affecting the release of major ions to the leachate in all environments. Backf i l l leaching is a surface phenomenon characterized by alteration layers widening towards the core of the sample with increasing leaching time. The same would be suspected to occur in underground mine workings filled with cemented backfill. The backfill would start to deteriorate at the contact with the host rock, with the depth of penetration of the alteration zone increasing with time. The current short-term leaching study does not provide, however, information on the long-term rate of advancement of the alteration zones, whether it would remain constant or eventually cease its advancement. It can be postulated from the experimental data, however, that in water-leached environments, the augmented porosity and precipitation of expansive minerals w i l l encourage leachate penetration into the backfill. Hence in these environments, the rate of advancement of the alteration zone would probably not slow down. This study indicated that in ARD-leached environments, the backfill material become passivated with time. The passivation mechanism remains unclear, however, either due to an increasing thickness of ARD-solut ion precipitates onto the backfill or due to the formation of self-sealing iron precipitates resulting from sulfide oxidation in the outer layers of the backfill. In either case, the consequence is a decreased capacity of the backfill to neutralize incoming A R D solution generated within the host rock. In 198 time, the acidic and metal-loaded mine drainage could flow unaltered around the backfill to the surrounding groundwater or surface water. This study suggests that cemented tailings may not constitute, in the moderate to long-term, an effective buffering material to neutralize A R D generated within the mine or untreated A R D effluent. Some oxidation of sphalerite and galena occurred in the A R D and cyclic leaching environment but not in the flooded water-leached environment. Oxidation of pyrite and pyrrhotite may have occurred in the A R D or cyclic environments but would have reprecipitated within the backfill matrix. The reactivity of iron sulfides within backfill would require specific investigation. The apparent loss in strength that was observed under S E M for some A R D and water- leached pastes was not manifested in the ucs tests of the leached samples. These results may not be statistically significant as only one sample per category was tested. Conversly, they could indicate that the thin alteration layer is not sufficiently significant to alter the strength of the entire sample measured by the apparatus. The study methodology was not designed to measure the strength of different areas within the sample. Knowledge of the peripheral strength would be very useful (i.e. periphery of backfilled stopes) as this area is exposed to blasting when mining next to the backfill. Lower peripheral strength of the backfill may cause higher than expected dilution of the ore. Ouellet et al. (1998) have pointed out that this was a particular problem at the Louvicourt mine with a previous backfill formulation. 199 This study showed a definite alteration of the backfill when subjected to water or A R D leaching solution, resulting in some apparent losses in strength of the backfill in the altered layers. Thicknesses of the alteration zones, however, ranged from less than 1mm to 10mm in 20 weeks, with the thicker alteration zones associated with the sulfide-rich tailings in ARD-leach solutions. It is necessary to place these results into the context of a mine setting where field rates of alteration can be higher or lower, depending on the local groundwater regime and annual precipitation on the site. The possible impacts of leaching paste backfill depend on the purpose for which the backfill is used. For example, a highly pyritic cemented backfill may not suit the purpose of a buffering plug meant to neutralize pre-existing acidic drainage within the mine. The same backfill could, however, effectively serve the purpose of ground support i f placed in an isolated stope and promptly submerged in a n o n - A R D environment or i f placed in a dry area. This study highlights the importance of carrying out leaching studies on cemented paste backfill mixtures in order to evaluate its stability under the conditions to which it w i l l be exposed. 200 8 R E C O M M E N D A T I O N S 8.1 PYRITE REACTIVITY STUDY Electrochemical studies of minerals should be considered where static test results are inconclusive as to the potential acid generation of a mineral waste. These studies could be realized during kinetic tests, for one or a number of sulfide minerals of concern to a particular mine site. The test set-up could consist, for example, of retrievable pockets of the sulfide mineral of concern placed into the column of rock sample being kinetically tested or placed as an appendage, similar to the Huckleberry solution-leached set-up of this study. The evolution of sulfide mineral reactivity could, in this way, be verified periodically along with the resistance of passivation coatings. The following recommendations should be followed to facilitate interpretation and enhance the understanding of sulfide mineral leaching tests: 1. In order to facilitate the observation of precipitate products and the evolution of grain surfaces, the pyrite samples should be washed in an ultrasound bath to remove the small mineral particles electrostatically stuck to pyrite surfaces. This would not only yield particle-free pyrite surface but also provide a uniform grain size sample from which could be calculated oxidation rates per surface area of pyrite. 2. Further analyses of the precipitate products should be carried out to define the composition of the precipitates and test the resistance of the passivation layer. This 201 would involve: • Detailed cathodic electrochemical analyses of the leached grains to verify the relative resistance of the grain coatings with advancing leaching cycles, increasing the cathodic current sweep to -1.5 volts. • Detailed quantitative mineral analyses of the precipitation products or grain surfaces. This would require making resin-mounted polished sections of leached samples and analyzing the grain edges by using either point source chemical analyses ( E D X , W D X or E A S ) or image analyses under S E M . 3. Additional information should be obtained on the effect of oxidation pits on the reactivity of pyrite. Cycl ic voltamperometry could be performed on unleached and pitted grains of the same sample, previously acid-washed to remove the precipitate coatings. 8.2 PASTE BACKFILL STUDY The following recommendations could facilitate future backfill leaching studies and help to resolve issues pending from this study. 1. Longer-term leaching studies would be useful for the following: • To obtain information on the progression of the leaching rate of the backfill with advancing leaching times and determine i f the rates w i l l increase, decrease or w i l l remain the same as leaching progresses. 202 To obtain thicker leaching zones, which would facilitate separation of the different layers and to provide a greater amount of sample to carry out multiple analyses of solid phase geochemistry, acid-base accounting or any other analyses. Bigger sample dimensions would also help in this respect, keeping sample diameter = Vi height specification for the ucs test. To provide the opportunity to observe under microscope and document the oxidation of sulfide minerals. Larger sample sizes would be necessary in longer-term tests to keep a relatively unleached core for comparison purposes and for the determination of the rate of penetration of the alteration zone, making sure that the alteration zone w i l l not fully penetrate the sample. Control samples should be prepared and stored for the duration of the tests in a lime solution at p H 12. Both the leached samples and the control samples could then be chemically analyzed and physically tested to compare the effects of leaching. The following analyses should be carried out as they provide very useful information as to the geochemical changes occurring within the pastes: evolution of A B A parameters in the altered layers compared to the control sample and to the core of the flooded sample; elemental analyses of the leachate, the solid phase of the altered layers, of the control sample and of the core of each leached sample. Analytical parameters should include the following: Ca, A l , M g , K , S i , C 0 3 2 \ Fe, Zn , C u , Pb, trace elements (depending on the mineralogy of the tailings, i.e. A s , Hg , Sb, Cd) as 203 well as p H , conductivity and redox potentials of the leachate solutions and the sulfide 2 2 speciation (S totai, SO4 S ") o f the solid phase samples. 2. Preparation of the samples for analyses: • A polished section of the altered paste could be prepared by using ultra low viscosity resin that would be deeply penetrating. Such samples would allow quantitative E D X analyses necessary to more easily identify the various hydrated binder phases and possibly the presence of oxidation rims on sulfide minerals. These samples should be prepared in addition to the flat unpolished sections that permit the observation of 3- dimentional crystal morphologies usefull for mineral identification. 3. Carry out leaching analyses of 3-month cured samples in parallel with 14-day cured samples (or earlier i f necessary) to verify the effects of advanced curing on the leachability of the hydrated cement in the backfill. 4. Carry out multiple (minimum 3) leaching cells of the same samples in order to perform multiple ucs analyses, for statistical significance. This could be organized as one large cell containing 3 samples, to ensure that all are subjected to similar conditions. 5. Carry out instrumented compression tests to measure differences in strengths of the different alteration zones in the leached backfill samples. 6. With respect to excessive dilution of ore when blasting next to backfilled areas: measure the effect of blasting vibrations on partially cured and fully cured backfill to 204 verify its effect on the development of strength of the backfill. Because of the likelihood of backfill dissolution once in place in a wet environment, it is recommended that C P B be subjected to similar scrutiny and attention as other mining wastes such as tailings and waste rock, especially in environmentally sentitive areas (i.e. above ground) or when reactive tailings are used in the backfill. 205 REFERENCES Adenot F. and B u i l M . , 1992: Modelling of the Corrosion of the Cement Paste by Deionized Water, Cement and Concrete Research, vol.22, nos. 2-3: 489-496. Ahmed S . M . and Giziewicz E . , 1992: Electrochemical Studies of Iron Sulfides in Relation to their Atmospheric Oxidation and Prevention of Acid Drainage; Part-II, in: Proceedings of the International Symposium on Electrochemistry in Mineral and Metal Processing: 372-389. Ahmed S . M . , 1991: Electrochemical and Surface Chemical Methods for the Prevention of the Atmospheric Oxidation of Sulphide Tailings, in: Second International Conference on the Abatement of Acid ic Drainage, Montreal, Quebec, September 1991, Conference Proceedings: 305-319. Alpers C . N . , et al., 1994: Secondary Minerals and Acid Mine Water Chemistry, in: Short Course Handbook on Environmental Geochemistry of Sulfide Mine-Wastes, Jambor and Blowes Editors: 245-270. Archibald J. and Chew J., 1998: Brunswick Paste Fill Optimization Study, Monthly Report #1, unpublished report to Noranda Mining and Exploration Inc., Brunswick Min ing Division, March 1998, 11 pages. A S T M Standard C 192-90a, 1990: Standard Practice for Making and Curing Concrete Test Specimens in the Laboratory, Annual Book of A S T M Standards, vol . 04.02.: 119-121. A S T M Standard C 470-87, 1987: Standard Specification for Molds for Forming Concrete Test Cylinders Vertically, Annual Book of A S T M Standards, vol . 04.02.: 246-249. A k o z F. , et al., 1995: Effects of Sodium Sulfate Concentration on the Sulfate Resistance of Mortars with and without Silica Fume, Cement and Concrete Research, vol . 25, no. 6: 1360-1368. Appelo C . A . J , and Postma D. , 1994: Geochemistry, Groundwater and Pollution, Balkema publishers, 536 pages. Benzaazoua M . , 1997: Etude chimique, mineralogique et microstructurale du remblai a 4.5% liant de la mine Louvicourt, Confidential Report by the Unite de Recherche et de Service en Technologie Minerale ( U R S T M ) , August 1997, 11 pages. Bethune K . J . et al., 1997: Acid Mine Drainage: Comparison of Laboratory Testing to Mine Site Conditions, in: Fourth International Conference on A c i d Rock Drainage, Vancouver, B . C . , M a y 1997, vol.1: 305-318. 206 Bigham J . M . , 1994: Mineralogy of Ochre Deposits Formed by Sulfide Oxidation, in: Short Course Handbook on Environmental Geochemistry of Sulfide Mine-Wastes, Jambor and Biowes Editors: 103-132. Blowes D . W . eta/., 1995: Microbiological, Chemical and Mineralogical Charaterization of the Kidd Creek Mine Tailings Impoundment, Timmins Area, Ontario, Geomicrobiology Journal, vol . 13: 13-31. Brown A . D . and Jurinak J.J., 1989: Mechanism of Pyrite Oxidation in Aqueous Mixtures, Journal of Environmental Quality, vol . 18: 545-550. Brown T . L . and Lemay H . E . , 1977: Chemistry the Central Science, 3 r d edition, Prentice- Hal l , 893 pages. Buckley A . N . , et al., 1988: Studies of the Surface Oxidation of Pyrite and Pyrrhotite Using X-Ray Photoelectron Spectroscopy and Linear Potential Sweep Voltammetry, in: Proceedings of the International Symposium on Electrochemistry in Mineral and Metal Processing: 234-246. Carde C. and Francois R., 1997: Effect of the Leaching of Calcium Hydroxide from Cement Paste on Mechanical and Physical Properties, Cement and Concrete Research, vol . 27, no. 4: 539-550. Casanova I. et al., 1996: Aggregate Expansivity Due to Sulfide Oxidation - 1. Reaction System and Rate Model, Cement and Concrete Research, vol . 26, no. 7: 993-998. Cinci l la W A . et al., 1997: Application of Paste Technology to Surface Disposal of Mineral Wastes, in "Tailings and Mine Waste '97": 343-356. Dahlstrom D . A . , 1997: Disposal of Fine Tailings in the Underground Mine, in "Tailings and Mine Waste '97": 327-334. Dallaire B . , Personal communication, November 13, 1997. DeCeukelaire L . , 1991: Concrete Surface Deterioration Due to the Expansion by the Formation ofJarosite, Cement and Concrete Research, vol . 21, no. 4: 563-574. Djuric M . et al., 1996: Sulfate corrosion of Portland Cement - Pure and Blended with 30% of Fly Ash, Cement and Concrete Research, vol . 26, no. 9: 1295-1300. Domvile S. J., et al., 1994: Weathering Behaviour of Mine Tailings and Waste Rock: A Surface Investigation, in: Third International Conference on the Abatement of Ac id ic Drainange, Pittsburgh, Pensylvania, Apr i l 1994: 167-176. Doyle F . M . and Mi rza A . H . , 1996: Electrochemical Oxidation of Pyrite Samples with Known Composition and Electricle Properties in: Proceedings of the 4 t h International Symposium on Electrochemistry in Mineral and Metal Processing, vol.96-6, Woods, Doyle and Richardson Editors: 203-241. 207 Duchesne J., and Reardon E.J. , 1995: Measurement and Prediction of Portlandite Solubility in Alkali Solutions, Cement and Concrete Research, vol . 25, no.5: 1043- 1053. Eidsa, G . et al., 1997: Pyrite Oxidation Measured by Oxygen Consumption and Carbon Dioxide Production in Laboratory Respirometer Tests, in: Fourth International Conference on A c i d Rock Drainage, Vancouver, B . C . , M a y 1997, vol.1: 319-331. Fetter W . C . , 1993: Contaminant Hydrogeology, Macmillan, 458 pages. Free M . L . et al., 1993: A Comparison of Leaching by Attached and Nonattached Bacteria of a Pyrite/Arsenopyrite gold-ore concentrate, Biohydrometallurgical Technologies, vol . 1, Proceedings of the 1993 International Biohydrometallurgy Symposium, Torma, Wey and Lakshmanan Editors: 459 - 468. Freeze A . , Cherry J., 1979: Groundwater, Prentice-Hall, 604 pages. Fu, Y . et al., 1995: A Kinetic Study of Delayed Ettringite Formation in Hydrated Portland Cement Paste, Cement and Concrete Research, vol . 25 no. 1: 63-70. Gay F. and Constantiner D. , 1998: Additives for Improving Paste Backfill Mixes, in Tailings and Mine Waste '98 Conference Proceedings, Fort Collins, Colorado, January 1998: 159-166. Gifford P . M . and Gillott, J.E., 1997: Behaviour of mortar and concrete made with activated blast furnace slag cement, Canadian Journal of C i v i l Engineering, vol . 24: 237-249. Goldhaber M . B . , 1983: Experimental Study of Metastable Sulfur Oxianion Formation During Pyrite Oxidation at pH 6-9 and 30°C, American Journal of Science, vol . 283: 193-217. Hen-era M . H . et al, 1989: The Kinetics of Sulfur Oxidation by Thiobacil lus ferrooxidans, in: Biohydrometallurgy 1989, International Symposium Proceedings, Salley, Cready and Wichlacz Editors; 415-421. Hopkins P . H . and Beaudry M . S . , 1989: An Evaluation of Binder Alternatives for Hydraulic Mill Tailings, in Innovations in Mining Backfi l l Technology, Hassani et al. Editors: 447-452. Idorn G . M . , 1992: Expansive Mechanisms in Concrete, Cement and Concrete Research, vol . 22, no.6: 1039-1046. Irassar E .F . , et al., 1996: Sulfate Attack on Concrete with Mineral Admixtures, Cement and Concrete Research, vol . 26, n o . l : 113-123. 208 Jambor J .L. , 1994: Mineralogy of Sulfide-rich Tailings and Their Oxidation Products, in: Short Course Handbook on Environmental Geochemistry of Sulfide Mine-Wastes, Jambor and Blowes Editors: 59-102. Kosrhatka S.H. , et al., 1995: Design and Control of Concrete Mixtures, Canadian Portland Cement Association, 6 t h edition, 221 pages. Kwong Y . T . J . , and Lawrence J.R., 1994: Mineralogical Controls of Sulfide Oxidation, National Hydrology Research Institute (NHRI) Contribution no. 94010, 87 pages. Kwong Y . T . J . , 1993: Prediction and Prevention of Acid Rock Drainage from a Geological and Mineralogical Perspective, M E N D Report no. 1.32.1, 47 pages. Kwong Y . T . J . , 1991: Acid Generation in Waste Rock as Exemplified by the Mount Wahington Minesite, British Columbia, Canada, Conference Proceedings of the 2 n d International Conference on the Abatement of Acid ic Drainage, Montreal, Quebec, September 1991: 175-190. Lamos A . W . and Clark I .H., 1989: The Influence of Material Composition and Sample Geometry on the Strength of Cemented Backfill, in Innovations in Min ing Backfi l l Technology, Hassani Scoble and Y u Editors: 89-94. Landriault D . et ah, 1998: Paste Technology for Underground Backfill and Surface Tailings Disposal Applications, Short Course Notes prepared by Golder Paste Technology Ltd. , January 25, 1998. Lawrence R . L . and Wang Y . , 1997: Determination of Neutralization Potential in the Prediction of Acid Rock Drainage, in: Fourth International Conference on A c i d Rock Drainage, Vancouver, B . C . , M a y 1997, vol.1: 449-464. Lawrence R . L . , 1997: Kinetic Tests on Huckleberry Low-Grade Ore, Waste Rock and Tailings, Summary Progress Report, unpublished consultation report prepared for Huckleberry Mines Ltd. , by Lawrence Consulting Ltd, A p r i l 1997, 7 pages. Lazaro et al., 1995: The Use of Carbon Paste Electrodes with A 1 on-Conducting Binder for the Study of Minerals: Chalcopyrite, Hydrometallurgy, vol . 38: 277-287. Levens, R . L . and Boldt, C . M . K , 1994: Mine Waste Sandfill, Min ing Environmental Management, vol . 2, no. 4: 16-20. L i J. et al., 1992: Application of Mixed Potential Theory in Hydrometallurgy, Hydrometallurgy, vol.29: 47-90. Lord E .R .F . and L i u Y . , 1998: Depositional and Geotechnical Characteristics of Paste Produced from Oil Sands Tailings, in "Tailings and Mine Waste '98" Conference Proceedings, Fort Collins Colorado, January 1998: 147-157 209 Lowson R.T. , 1982: Aqueous Oxidation of Pyrite by Molecular Oxygen, Chemical Reviews, vol . 82, no. 5: 461-497. L u f f W . M . et al., 1992: Evidence for a Feeder Pipe and Associated Alteration at the Brunswick No. 12 Massive-Sulftde Deposit, Exploration and Min ing Geology, vol.1, no.2: 167-185. Lundgren D . G . and Silver M . , 1980: Ore Leaching by Bacteria, Annual Reviews in Microbiology, vol 34: 263-283. Mangat P.S. and Khatib, J . M . , 1995: Influence offly ash, silica fume, and slag on sulfate resistance of concrete, A C I materials Journal, vol . 92, no.5: 542-552. McKibben M . A . and Barnes H . L . , 1986: Oxidation of Pyrite in Low Temperature Acidic Solutions: Rate Laws and SurfaceTextures, Geochimica et Cosmochimica Acta, vo l .50 : 1509-1520. Metha A . P . and Murr L . E . , 1983: Fundamental Studies of the Contribution of Galvanic Interaction to Acid-Bacterial Leaching of Mixed Metal Sulfides, Hydrometallurgy, vol.9: 235-256. Moerman A . , 1997: files sent to Valerie Bertrand via electronic mail, October 1997. Moses C O . et al., 1987: Aqueous Pyrite Oxidation by Dissolve Oxygen and by Ferric Iron, Geochimica et Cosmochimica Acta, vol . 51: 1561-1571. Moses C O . and Herman J.S., 1991: Pyrite Oxidation at CircumneutralpH, Geochimica et Cosmochimica Acta, vol.55: 471-482. Must in C , 1992: Approchephysico-chimique et modelisation de Voxydation bacterienne de la pyrite par Thiobacillus ferrooxidans: Role determinant de la phase miner ale. Doctoral Thesis, Universite de Nancy, Nancy, France: 222 pages. Myneni S .C .B. et al., 1998: Ettringite Solubility and Geochemistry of the Ca(OH)2- Al2(S04)s-H20 System at 1 atm Pressure and 298K, Chemical Geology, vol . 148: 1-19. Nicholson R . V . , 1994: Iron-sulfide Oxidation Mechanisms: Laboratory Studies, in: Short Course Handbook on Environmental Geochemistry of Sulfide Mine-Wastes, Jambor and Blowes Editors: 163-183. Nicholson R . V . et al., 1990: Pyrite Oxidation in Carbonate-buffered Solution: 2. Rate Control by Oxide Coatings, Geochimica et Cosmochimica Acta, vol . 54: 395-402. Nicholson R . V . et al., 1988: Pyrite Oxidation in Carbonate-bufferred Solution: 1. Experimental Kinetics, Geochimica et Cosmochimica Acta, vol . 52: 1077-1085. 210 Noranda Technology Centre, 1998a: BMD Paste Backfill Chemical and Physical Stability Project - Progress report #1, unpublished internal report prepared for Brunswick Min ing Division, Noranda Inc., January 1998, 6 pages. Noranda Technology Centre, 1998b: BMD Paste Backfill Chemical and Physical Stability Project - Progress report #2, unpublished internal report prepared for Brunswick Min ing Division, Noranda Inc., February 1998, 37 pages. Noranda Technology Centre, 1998c: BMD Paste Backfill Chemical and Physical Stability Project - Progress report #3, unpublished internal report prepared for Brunswick Min ing Division, Noranda Inc., March 1998, 16 pages. Noranda Technology Centre, 1998d: BMD Paste Backfill Chemical and Physical Stability Project - Progress report #4, unpublished internal report prepared for Brunswick Min ing Division, Noranda Inc., Apr i l 1998, 36 pages. Noranda Technology Centre, 1998e: Benchmarking of BMD Tailings Against Louvicourt Tailings, unpublished internal report prepared for Brunswick Min ing Division, Noranda Inc., A p r i l 1998, 19 pages. Otwinowski M . , 1997: Meso-Scale Characterization of the Geochemical and Physical Processes Responsible for Acid rock Drainage, in: 4 t h International Conference on A c i d Rock Drainage, Vancouver, B . C . , M a y 1997, vol.2: 569-585. Ouellet J. et al., 1998: Mechanical, Mineralogical and Chemical Characterization of a Paste Backfill, in: Tailings and Mine Waste '98 Conference Proceedings, Fort Collins Colorado, January 1998: 139-146. Pakhurst D . L . , 1995: P H R E E Q C computer program for geochemical calculations, U . S . Geological Survey, Water-Resources Inverstigations Report no. 954227, 134 pages. Peters E . , 1984: Electrochemical Mechanisms for Decomposing Sulfide Minerals, in: Proceedings of the International Symposium on Electrochemistry in Mineral and Metal Processing, Richardson, Srinivasan and Woods Editors: 343-361. Price W A . et al., 1997: Guidelines for the Prediction of Acid Rock Drainage and Metal Leaching for Mines in British Columbia: Part II. Recommended Procedures for Static and Kinetic Testing, in: 4 t h International Conference on A c i d Rock Drainage, Vancouver, B . C . , M a y 1997, vol.1: 15-30. Revertegat E . et al., 1992: Effect ofpHon the Durability of Cement Pastes, Cement and Concrete Research, vol . 22, no. 2-3: 259-272. Ritchie A . I . M . , 1994: Sulfide Oxidation Mechanisms: Controls and Rates of Oxygen Transport, in: Short Course Handbook on Environmental Geochemistry of Sulfide Mine-Wastes, Jambor and Blowes Editors: 201-245. 211 Schafer W . M . et al., 1994: Monitoring Gaseous and Liquid Flux in Sulfide Waste Rock, in: 3 r d International Conference on the Abatement of Acid ic Drainange, Pittsburgh, Pensylvania, A p r i l 1994: 410-418. Singer P .C . and Stumm W., 1970: Acidic Mine Drainage: The Rate-Determining Step, Science, vol . 167: 1121-1123. Shayan A . , 1988: Deterioration of a Concrete Surface Due to the Oxidation of Pyrite Contained in Pyritic Aggregates, Cement and Concrete Research, vol . 18, no. 5: 723-730. Steffen, Robertson and Kirsten Ltd. , 1989: Draft Acid Rock Drainage Technical Guide, vol . 1, B C A M D Task Force Report no.5.10. Taylor H . F . W . , 1997: Cement chemistry, Thomas Telford editors, 459 pages. Tori i K et ah, 1995: Sulfate Resistance of High Fly Ash Content Concrete, Cement and Concrete Research, vol . 25, no. 4: 759-768. Udd J.E., 1989: Backfill Research in Canadian Mines, in "Innovations in Min ing Backfi l l Technology", Hassani, Scoble and Y u Editors: 3-13. Vernet C , 1994: Introduction to Portland Cement Chemistry, in "Applications of N M R Spectroscopy to Cement Science", Colombet and Grimmer Editors: 29-53. Wadsworth M . E . et ah, 1993: Electrochemistry of Pyrite, in: Hydrometallurgy; Fundamentals, Technology and Innovations, Proceedings of the 4 t h International Symposium on Hydrometallurgy, Hiskey and Warren Editors, Salt Lake City, Utah, August 1993: (Chapter 5) 85-99. Warren C.J . and Reardon E.J.,1994: The Solubility of Ettringite at 25 deg. C, Cement and Concrete Research, vol . 24, no. 8: 1515-1524. Wil l iams D.J . , 1997: Effectiveness of Co-disposing Coal Washery Wastes, in: Tailings and Mine Waste '98 Conference Proceedings, Fort Collins Colorado, January 1998: 335-341. Woyshner M . R . and St-Arnaud L . , 1994: Hydrogeological Evaluation and Water Balance of a Thickened Tailings Deposit Near Timmins, ON, Canada, in: 3 r d International Conference on the Abatement of Acid ic Drainange, Pittsburgh, Pensylvania, A p r i l 1994: 198-207. Ybarra, R.R. , 1998: Prediccion de la Generacion de Drenajes Acidos en una Presa de dales con Altos Contenidos de Pirita. M.Sc . Thesis, Faculty of Chemical Sciences, Universidad Autonoma de San Luis Potosi (Mexico). 212 Zhu X . , et al., 1992: Transpassive Oxidation of Pyrite, in: Proceedings of the International Symposium on Electrochemistry in Mineral and Metal Processing, Woods y Richardson Editors: 391-409. i 213 APPENDIX I - ANALYTICAL RESULTS: PYRITE LEACHATE 214 Table 1-1 (cont'd) Pyrite - Leachate Chemistry Cycle Pyrite Leach Solution Pb (mg/l) Hk-I Hk-w Lou1 Lou2 Tiz Zim Brw Hk-I Zn (mg/l) Hk-w Lou1 Lou2 Tiz Zim Brw Hk-I As (mg/l) Hk-w Lou1 Lou2 Tiz Zim Brw 1 0.5 0.4 6.1 4.9 6.8 19 61 0.3 1.6 2.8 22 18 28 240 <1 <1 <1 <1 <1 <1 <1 2 1.4 1.4 5.1 4.3 7.4 5.9 73 2.3 0.4 8.3 47 42 94 366 <1 <1 <1 <1 <1 <1 <1 3 1.0 1.1 5.0 3.8 5.8 5.5 87 0.5 0.2 7.7 53 48 75 320 <1 <1 <1 <1 <1 <1 <1 4 0.7 92 0.1 0.1 2.5 33 8.4 51 174 <1 <1 <1 <1 <1 <1 <1 5 92 0.1 0.2 4.1 39 9.2 68 177 <1 <1 <1 <1 <1 <1 <1 6 71 0.3 0.1 3.2 40 17 59 180 <1 <1 <1 <1 <1 <1 <1 7 46 0.1 0.6 2.0 31 10 43 188 <1 <1 <1 <1 <1 <1 <1 8 49 0.1 0.4 3.0 42 12 52 225 <1 <1 <1 <1 <1 <1 <1 9 0.2 3.0 2.2 5.4 8.0 37 0.2 < 0. 2.1 36 9.0 36 273 <1 <1 <1 <1 1.1 3.4 <1 10 0.4 0.4 3.3 1.3 5.2 6.0 27 0.2 0.1 2.0 33 6.5 32 261 <1 <1 <1 <1 <1 2.7 <1 11 0.3 <0.1 4.3 2.3 7.2 17 21 0.1 <0. 0.8 18 2.7 15 295 <1 <1 <1 <1 1.0 4.4 <1 12 0.1 0.2 4.2 1.7 6.6 13 0.1 <0. 0.4 9 1.6 7 222 <1 <1 <1 <1 1.0 3.2 <1 13 0.5 0.3 4.1 2.0 6.8 10 0.3 1.2 0.7 14 2.2 10 239 <1 <1 <1 <1 < 1 3.7 <1 14 0.3 0.3 2.3 1.4 7.9 11 9.5 <0.1 0.9 1.7 28 3.4 11 183 <1 <1 <1 <1 < 1 4.1 <1 15 0.5 0.8 3.3 1.8 6.5 14 9.9 <0.1 0.5 1.2 15 2.7 8.8 176 <1 <1 <1 <1 1.4 4.6 <1 16 0.5 0.5 3.4 1.8 6.1 13 6.5 0.2 0.1 1.0 22 3.2 9.5 154 <1 <1 <1 <1 1.4 4.1 <1 17 0.4 0.5 3.7 1.0 6.0 13 7.0 0.3 < 0. 0.8 16 1.9 6.1 149 <1 <1 <1 <1 1.5 4.2 <1 18 0.9 0.2 2.8 0.8 6.3 12 8.4 0.2 < 0. 1.2 19 2.1 7.3 145 <1 <1 <1 <1 1.4 3.4 <1 19 1.6 1.7 2.4 1.5 5.9 13 9.4 0.1 0.1 1.1 22 2.4 135 <1 <1 <1 <1 1.3 3.6 <1 20 0.3 3.1 2.3 0.5 6.0 10 8.3 0.2 0.1 0.9 16 2.1 6.6 133 <1 <1 <1 <1 1.6 3.4 <1 21 0.6 0.6 2.6 0.6 7.6 11 7.4 0.3 0.1 1.1 18 2.4 6.4 131 <1 <1 <1 <1 1.6 3.4 <1 22 0.2 2.7 2.6 0.5 6.5 14 5.1 0.1 0.3 0.5 8.4 1.5 151 <1 <1 <1 <1 1.6 4.1 <1 23 0.2 0.5 2.6 0.7 7.8 5.4 <0.1 0.1 0.8 9.5 2.2 146 <1 <1 <1 <1 1.7 <1 24 0.1 0.3 2.5 0.6 6.6 13 7.3 <0.1 <0. 0.5 7.2 1.9 3.6 147 <1 <1 <1 <1 <1 5.7 <1 25 0.2 0.3 2.8 0.5 6.8 11 6.9 0.1 0.2 0.6 7.1 1.6 3.3 187 <1 <1 <1 <1 <1 4.6 <1 26 0.1 0.6 1.8 0.2 7.2 5.7 4.2 0.2 0.5 4 1 1.4 201 <1 <1 <1 <1 1.1 1.5 <1 2 7 0.2 0.6 1.3 0.2 5.5 6.4 4.1 0.3 0.3 0.5 5.3 1.2 1 248 <1 <1 <1 <1 <1 2.1 <1 2 8 0.4 0.4 1.8 0.5 3.9 2.0 4.4 0.4 0.2 0.5 3.8 1.3 0.7 213 <1 <1 <1 <1 <1 3 <1 2 9 0.1 1.1 1.3 0.3 6.3 5.9 3.7 0.2 0.3 0.8 4.6 1 1.5 246 <1 <1 <1 <1 1 2.8 <1 3 0 0.2 0.3 1.4 0.3 6.7 8.1 0.3 0.1 0.6 4.8 1.1 1.4 <1 <1 <1 <1 <1 3 31 0.1 0.5 1.7 0.4 7.1 7.9 2.8 0.1 0.3 0.5 4 1.2 1.1 100 <1 <1 <1 <1 <1 2.6 <1 32 0.4 1.2 5.5 0.2 0.8 0.1 <1 <1 <1 <1 <1 <1 33 0.3 0.3 6.7 0.4 0.1 4.4 1.1 <1 <1 <1 <1 <1 <1 34 0.3 0.5 0.4 1.1 5.4 1.9 6.2 0.6 0.1 0.5 0.2 1.3 1.1 <1 <1 <1 <1 <1 1.7 <1 35 0.2 0.2 0.8 0.3 7.3 7.2 7.2 0.2 0 0.5 5 1.1 1 <1 <1 <1 <1 <1 2.1 <1 36 0.3 0.3 0.6 <0.1 7.2 6.6 9.9 0.2 0 0.7 3.2 0.8 0.9 65 <1 <1 <1 <1 <1 2.3 <1 37 0.2 0.4 0.7 0.1 7.0 6.1 6.7 0.3 0.2 0.8 4.1 1.4 1.5 62 <1 <1 <1 <1 <1 2.7 <1 3 8 0.2 0.3 0.8 0.2 6.1 4.6 5.0 0.2 0.3 0.7 3.9 1.2 2 57 <1 <1 <1 <1 <1 2.5 3 9 0.6 0.5 0.5 0.2 5.6 3.3 5.6 0.6 0.2 0.4 3.3 1.2 1 27 <1 <1 <1 <1 <1 2.1 4 0 0.4 0.59 <0.1 <0.1 6.5 3.3 0.7 0.4 0.4 2.8 1 0.8 <1 <1 <1 <1 <1 1.8 215 Table 1-1 (cont'd) Pyrite Leachate Chemistry C y c l e Pyrite Leach Solution Fe total (mg/1) Hk-I Hk-w Lou1 Lou2 Tiz Zim Brw Hk-I S04 (g/l) Hk-w Lou1 Lou2 Tiz Zim Brw Hk-I Cu (mg/l) Hk-w Lou1 Lou2 Tiz Zim Brw 1 388 679 80 85 86 27 8.0 1.0 1.3 0.3 0.3 0.2 0.1 0.1 <0.1 0.2 0.1 0.4 0.2 0.2 <0.1 2 848 764 80 97 64 118 0.5 1.7 1.9 0.3 0.4 1.7 0.3 0.2 <0.1 <0.1 2.0 <0.1 16 <0.1 <0.1 3 738 818 69 108 61 97 0.3 2.4 1.9 0.4 0.3 0.2 0.3 0.2 <0.1 <0.1 1.1 <0.1 22 0.3 <0.1 4 475 427 44 66 45 86 0.2 0.9 0.1 0.2 0.1 <0.1 0.1 0.4 0.1 0.4 0.3 <0.1 5 345 448 70 75 46 68 0.3 1.1 0.2 0.3 0.1 0.1 0.0 0.9 0.1 6.6 0.4 <0.1 6 419 364 47 66 38 53 0.3 0.9 0.1 0.2 0.1 0.0 0.0 0.5 0.0 4.2 0.3 <0.1 7 340 286 39 58 35 45 0.2 0.7 0.2 0.1 0.2 0.1 0.1 0.1 0.4 0.1 4.4 0.3 <0.1 8 273 336 49 64 43 45 0.2 0.8 0.2 0.1 0.2 0.2 0.0 0.0 0.1 0.1 8.4 0.3 <0.1 9 302 280 48 57 47 40 0.2 0.5 0.1 0.2 0.2 0.2 0.1 0.1 <0.1 0.4 <0.1 3.6 0.3 <0.1 10 246 304 50 52 45 34 0.2 0.3 0.9 0.3 0.2 0.2 0.5 <0.1 <0.1 0.4 <0.1 3.4 0.3 <0.1 11 247 173 33 35 30 27 0.2 0.8 0.4 0.2 0.2 0.2 0.5 < 0.1 <0.1 0.2 <0.1 1.9 0.2 <0.1 12 146 122 17 21 19 15 0.9 0.4 0.2 0.1 0.1 0.1 < 0.1 <0.1 0.2 <0.1 1.6 0.2 0.2 13 80 156 23 26 24 19 0.4 0.4 0.3 0.2 0.2 0.2 < 0.1 0.1 0.3 <0.1 1.4 0.2 <0.1 14 128 287 41 51 33 20 0.4 0.6 0.2 0.2 0.2 0.2 < 0.1 0.1 0.4 <0.1 2.1 0.2 <0.1 15 256 203 35 40 37 21 0.5 0.5 0.3 0.4 0.2 0.1 0.1 <0.1 0.3 <0.1 1.6 0.2 0.2 16 152 181 34 36 39 17 0.1 0.5 0.5 0.1 0.1 0.0 0.1 < 0.1 0.12 0.4 <0.1 2.2 <0.1 <0.1 17 129 160 35 30 29 14 0.2 0.1 0.2 0.1 0.1 0.0 0.2 < 0.1 <0.1 0.4 <0.1 1.7 <0.1 <0.1 18 125 180 46 28 28 14 0.3 0.5 0.4 0.1 0.1 0.1 0.1 < 0.1 <0.1 0.5 <0.1 1.5 0.3 <0.1 19 139 161 42 31 31 13 0.3 0.6 0.2 0.2 0.2 0.3 0.2 < 0.1 <0.1 0.4 <0.1 1.9 0.2 <0.1 20 142 163 41 24 33 14 0.5 0.4 0.1 0.1 0.6 0.11 <0.1 0.2 <0.1 1.6 0.2 <0.1 21 112 156 47 21 33 13 0.4 <0.1 <0.1 0.3 <0.1 1.9 0.3 <0.1 22 106 123 28 14 26 12 <0.1 <0.1 <0.1 <0.1 1.0 0.2 0.1 23 75 139 46 17 28 11 0.5 0.00 <0.1 0.2 <0.1 1.4 <0.1 24 91 89 26 12 12 0.7 <0.1 0.2 0.2 0.1 0.2 0.00 0.15 0.2 0.1 1.6 <0.1 <0.1 25 60 99 28 12 9 1.1 0.3 0.1 0.4 0.14 0.18 0.2 0.2 1.4<0.1 0.2 26 73 15 8 30 2 2.1 0.3 0.3 0.0 0.1 0.1 0.00 0.13 0.2 0.1 0.8 <0.1 0.4 27 85 17 8 3 7.9 0.3 0.3 0.1 0.1 0.1 0.1 0.2 0.17 0.14 0.2 0.1 1.0<0.1 0.1 28 70 58 15 6 <0.1 16 0.3 0.1 0.1 0.5 0.17 0.1 0.1 <0.1 0.9 <0.1 0.3 29 47 67 20 8 3 31 0.1 0.3 0.1 0.5 0.14 0.11 0.1 0.1 0.8 <0.1 0.5 30 47 69 19 9 12 3 0.2 0.3 0.1 0.1 0.12 0.13 0.2 0.1 0.9 0.1 31 73 20 11 15 3 53 0.6 0.2 0.1 0.5 0.13 <0.1 0.1 0.8 0.1 <0.1 32 46 103 57 0.3 0.5 0.00 0.28 0.1 0.1 1.1 0.1 1.5 33 68 98 55 0.4 0.13 0.15 0.2 0.1 0.9 0.1 2.4 34 71 87 15 11 17 2 66 0.3 <0.1 <0.1 0.1 0.1 0.4 0.12 0.16 <0.1 <0.1 1.3<0.1 4.2 35 58 82 21 12 19 3 82 0.2 0.2 0.1 <0.1 0.1 0.1 0.5 0.10 0.12 <0.1 <0.1 1.0 <0.1 6.4 36 60 67 16 10 15 3 76 0.2 0.2 <0.1 <0.1 0.1 0.1 0.00 <0.1 <0.1 <0.1 0.7 <0.1 4.3 37 60 88 21 13 16 3 89 0.2 0.2 <0.1 0.1 0.2 0.1 0.3 0.00 0.11 0.1 <0.1 1.1 <0.1 4.5 38 67 73 22 12 15 4 127 0.2 0.2 <0.1 <0.1 0.1 0.1 0.11 <0.1 0.1 <0.1 1.0 <0.1 39 64 72 19 10 13 3 110 0.2 0.1 0.1 0.2 0.1 0.13 0.11 <0.1 <0.1 0.8 <0.1 40 52.8 61.1 16 10 11 1.9 0.2 0.2 <0.1 <0.1 0.1 0.1 <0.1 0.14 0.1 <0.1 <0.1 0.1 216 Table 1-1 (cont'd) Pyrite Leachate Chemistry Cycle Pyrite Leach Solution conductivity (uS) Hk-I Hk-w Lou Lou Tiz Zim Hk-I Hk-w Lou1 Lou2 Tiz Zim Brw Hk-I Redox potential (mV) Hk-w Lou1 Lou2 Tiz Zim Brwk 1 1954 508 548 947 3.8 4.3 4.5 4.1 4.8 5.0 208 2 1751 2900 576 676 524 581 647 3.1 2.7 3.6 3.9 3.5 4.7 5.2 194 3 3900 2860 424 573 490 487 609 2.4 2.4 3.4 3.6 3.3 3.9 5.1 245 4 2440 1909 344 444 418 519 529 2.4 2.6 3.5 3.8 3.4 4.0 5.1 204 5 1744 2490 367 439 409 524 541 2.6 2.4 3.5 3.8 3.3 3.6 4.9 no data available 276 6 2460 2360 405 450 413 515 511 2.4 2.4 3.4 3.8 3.4 3.4 5.0 219 7 2430 1697 322 391 409 476 538 2.5 2.6 3.6 3.9 3.3 3.4 5.0 289 8 1820 2400 385 470 443 478 573 2.5 2.4 3.4 3.6 3,2 3.3 4.9 155 9 2420 1917 402 430 481 466 668 2.3 2.5 3.4 3.7 3.2 3.4 5.0 256 10 1908 2160 381 361 444 369 196 2.6 2.5 3.4 4.4 3.2 3.5 5.1 196 11 1945 1296 299 277 380 304 665 2.5 2.6 3.4 3.7 3.2 3.3 5.0 220 12 1412 972 195 207 243 174 687 2.6 2.9 3.8 5.6 3.7 3.7 5.1 332 301 131 295 286 268 13 899 1111 227 256 307 227 629 3.0 2.6 3.4 3.5 3.2 3.4 5.4 334 320 309 291 322 315 228 14 1119 1391 358 396 411 260 473 2.6 2.7 3.4 3.7 3.2 3.4 5.4 329 331 310 292 324 308 139 15 1440 1088 320 328 403 261 487 2.7 2.8 3.3 3.6 3.1 3.4 5.3 332 315 305 280 315 304 143 16 1123 1378 296 701 410 236 478 2.9 2.6 3.4 3.6 3.1 3.4 5.5 317 335 311 290 327 298 140 17 1207 1190 329 281 390 247 473 2.6 2.6 3.3 3.6 3.1 3.4 5.4 331 329 311 297 343 320 173 18 1167 1374 418 275 365 242 423 2.7 2.6 3.3 3.7 3.2 3.4 5.4 333 326 317 288 314 310 237 19 1251 1301 417 292 434 243 445 2.7 2.6 3.3 3.6 3.2 3.5 5.4 334 335 317 295 329 319 231 20 1365 1230 335 245 366 220 410 2.7 2.5 3.2 3.6 3.1 3.3 5.3 340 335 308 279 327 330 251 21 1042 1379 435 244 425 225 434 2.7 2.6 3.3 3.7 3.1 3.4 5.5 333 329 315 292 325 318 149 22 1124 883 269 159 323 305 503 2.8 2.8 3.4 3.8 3.2 3.5 5.8 331 320 291 283 307 305 196 23 757 1336 416 182 567 179 485 2.9 2.7 3.3 3.7 3.2 3.5 5.2 315 323 311 284 307 299 178 24 950 950 262 173 372 147 429 2.8 2.8 3.4 3.8 3.2 3.7 5.4 323 326 303 284 318 301 25 818 1087 301 160 313 129 517 2.9 2.8 3.5 4.0 3.3 3.9 5.4 319 325 304 274 307 269 233 26 944 695 142 110 197 135 603 2.9 3.0 4.1 3.6 3.6 3.3 5.1 325 309 242 135 241 144 209 27 674 829 196 136 222 101 683 3.3 2.8 3.5 3.8 3.3 4.1 5.0 311 323 270 240 277 236 197 28 526 798 212 120 270 400 655 3.2 2.9 3.7 4.1 3.5 4.2 4.7 302 279 299 257 305 145 204 29 231 684 223 154 240 130 666 3.9 3.1 3.7 4.2 3.6 4.0 5.3 273 286 285 261 288 255 153 30 588 679 189 124 195 69 654 3.3 3.1 3.9 4.2 3.6 4.4 5.2 302 281 242 217 267 218 31 629 720 244 130 198 75 874 3.2 3 3.7 4.2 3.7 5.0 4.1 270 318 292 276 302 241 257 32 653 837 241 157 246 77 731 3.2 2.9 3.6 3.8 3.5 5.5 4.1 303 320 286 263 286 184 257 33 748 797 222 146 256 67 720 3 2.9 3.7 4.0 3.5 4.6 3.7 293 316 288 252 288 194 266 34 754 736 166 128 221 82 783 3.1 3.1 4.1 4.5 3.8 5.6 3.5 317 319 256 213 277 128 241 35 701 701 231 141 243 63 830 3.1 3 3.7 4.0 3.6 4.5 3.5 324 318 293 271 300 220 273 36 712 580 152 110 194 68 739 3.1 3.1 3.8 4.1 3.5 4.4 3.3 324 308 37 654 685 180 119 226 55 771 3.1 3.1 3.8 4.3 3.6 4.6 3.2 319 284 240 302 223 310 38 697 635 190 118 196 69 802 3.1 3.1 3.7 4.1 3.6 4.5 3.2 322 311 281 242 289 216 312 39 691 619 158 99 160 55 721 3.1 3 3.9 4.3 3.7 5.3 3.2 306 312 240 224 273 175 317 40 575 453 128 91 153 47 795 3.3 3.2 4.0 4.2 3.7 4.7 3.2 309 306 230 221 279 190 315 217 Table 1-1 Pyrite Leachate Chemistry Days Date Cycle con fuS) Column Leach soln pH vol edo Fetot (ml) (mV) (mg/l) S04 Cu (g'l) (mg/l Zn Ca Mg Na K (mg/l (mg/l (mg/l) (mg/l (mg/l) Distilled water pH cond Redox (uS) (mV) 0 07-Nov-97 1 233 0.31 0.11 <0.1 <0.1 31 3.6 26 9.5 5.4 3 10-Nov-97 2 6.7 228 0.10 0.10 <0.1 <0.1 30 1.7 24 9.0 5.6 2.1 6 14-Nov-97 3 340 5.5 245 0.42 0.12 <0.1 <0.1 26 1.5 22 8.5 5.3 1.6 10 18-Nov-97 4 221 7.8 213 0.10 0.01 <0.1 <0.1 23 1.3 20 7.4 5.5 14 21-Nov-97 5 288 6.8 212 5.6 19 17 24-Nov-97 6 277 6.2 225 0.04 0.12 <0.1 <0.1 28 1.9 20 9.2 5.6 1.8 20 27-Nov-97 7 258 6.4 205 0.01 0.08 <0.1 <0.1 28 1.8 18 7.3 5.5 4.6 23 01-Dec-97 8 238 6.2 202 0.00 0.09 <0.1 <0.1 29 1.6 15 6.5 5.6 2.6 27 04-Dec-97 9 240 6.2 203 0.23 <0.1 <0.1 26 1.5 16 7.0 5.5 10 30 08-Dec-97 10 247 6.3 217 <0.1 0.25 <0.1 1.7 39 1.8 15 7.2 5.6 16 34 12-Dec-97 11 235 6.4 206 <0.1 0.80 <0.1 <0.1 38 1.7 15 7.4 5.5 28 269 38 15-Dec-97 12 188 6.6 202 <0.1 0.25 <0.1 <0.1 40 2.0 14 7.3 5.4 21 128 41 18-Dec-97 13 163 6.6 202 90 <0.1 0.18 <0.1 <0.1 31 1.5 10 4.8 5.4 26 232 44 29-Dec-97 14 157 6.6 195 209 <0.1 0.08 <0.1 <0.1 30 1.4 12 6.0 5.5 27 248 55 05-Jan-98 15 295 5.9 205 231 <0.1 0.10 <0.1 <0.1 51 2.5 13 7.3 5.5 17 256 62 08-Jan-98 16 287 6.1 210 122 <0.1 0.14 <0.1 <0.1 56 2.7 14 8.4 5.5 8.3 178 65 12-Jan-98 17 181 6.3 194 137 <0.1 0.11 <0.1 <0.1 45 2.3 12 6.5 5.4 7.3 269 69 15-Jan-98 18 219 6.1 204 174 <0.1 0.11 <0.1 <0.1 83 7.8 14 6.6 5.6 9.2 144 72 19-Jan-98 19 169 6.4 198 161 <0.1 0.09 <0.1 <0.1 35 2.1 13 7.0 5.4 9.4 271 76 23-Jan-98 20 170 5.7 200 231 <0.1 0.25 <0.1 <0.1 36 1.7 10 5.5 5.4 10 134 80 26-Jan-98 21 151 6.2 200 159 <0.1 0.05 <0.1 <0.1 36 2.0 11 5.9 5.5 4.5 217 83 29-Jan-98 22 154 6.4 205 157 <0.1 0.10 <0.1 <0.1 70 6.6 10 5.3 5.7 8.7 183 86 02-Feb-98 23 138 6.1 197 180 <0.1 0.07 <0.1 <0.1 28 1.3 10 4.9 5.5 6.1 248 90 06-Feb-98 24 163 5.7 181 165 <0.1 0.08 <0.1 <0.1 62 6.0 11 5.1 5.3 4.6 265 94 09-Feb-98 25 163 5.7 181 165 <0.1 0.07 <0.1 <0.1 33 1.94 11 5.1 5.5 6.4 250 97 12-Feb-98 26 147 5.7 185 217 <0.1 0.06 <0.1 <0.1 41 1.86 10 6.06 5.5 6.4 250 104 16-Feb-98 27 143 6.0 184 148 <0.1 0.07 <0.1 <0.1 32 2.0 9.6 7.12 5.6 9.3 272 107 19-Feb-98 28 143 5.9 171 167 <0.1 0.07 <0.1 <0.1 27 1.73 9.6 5.81 5.6 9.5 184 111 23-Feb-98 29 130 6.4 226 193 <0.1 0.06 <0.1 <0.1 29 1.82 11 6.36 5.5 18 213 111 26-Feb-98 30 163 6.3 205 157 <0.1 0.08 <0.1 <0.1 31 1.89 11 6.8 5.6 39 232 114 02-Mar-98 31 152 6.1 191 203 <0.1 0.07 <0.1 <0.1 35 1.73 10 6.76 5.3 6.9 221 118 05-Mar-98 32 166 6.0 190 206 <0.1 0.07 <0.1 <0.1 53 2.15 10 6.89 5.3 5.8 241 121 09-Mar-98 33 160 5.9 204 123 <0.1 0.06 <0.1 <0.1 41 1.21 9.2 6.2 5.6 5.4 250 125 12-Mar-98 34 170 5.8 214 171 <0.1 0.06 <0.1 <0.1 35 1.44 8.9 5.8 5.5 8.8 247 128 16-Mar-98 35 165 6.2 192 164 <0.1 0.06 <0.1 <0.1 28 1.5 9.5 5.64 5.6 48 163 132 20-Mar-98 36 166 5.8 168 161 <0.1 0.10 <0.1 <0.1 31 1.3 10 6.5 5.4 4.1 211 136 25-Mar-98 37 168 5.8 190 160 <0.1 0.07 <0.1 <0.1 33 1.8 9.1 6.2 5.3 16 234 141 27-Mar-98 38 169 6.1 187 204 <0.1 0.05 <0.1 <0.1 36 2.1 9.3 6.53 5.2 3.4 263 143 30-Mar-98 39 165 6.2 187 198 <0.1 0.07 <0.1 <0.1 38 2.2 9 6.36 5.6 14 204 146 02-Apr-98 40 166 6.2 185 142 <0.1 0.07 <0.1 <0.1 39 2.25 8.7 6.41 5.6 14 173 Note: empty spaces = no data available 218 APPENDIX II - ANALYTICAL RESULTS: PASTE BACKFILL LEACHATE 2 1 9 Table 11-1 Paste Leachate Chemisty - Flooded Water Cells ay Date ycl Distilled water data batch cond pH redox (uS) (mV) Paste leach water conductivity (uS) Tiz Louv Brw FIM Tiz PH Louv Brw FIM Redox potential (mVSHEl Tiz LOU Brw FIM 4 16-Dec-97 1 1 28 5.5 499 3710 3670 3550 3100 11 7.4 7.7 7.8 286 400 405 388 7 18-Dec-97 2 2 17 5.6 422 3440 3450 3260 2720 11 6.6 6.7 6.7 311 371 410 402 18 30-Dec-97 3 3 33 5.6 429 3120 2940 2960 2520 11 7.8 7.2 8.3 303 422 395 414 25 06-Jan-98 4 4 6.1 5.6 509 2660 2540 2480 2230 10 8.5 8.1 8.5 280 314 333 328 32 13-Jan-98 5 5 18 5.5 469 2410 2310 1798 1920 11 8.5 8.7 8.9 257 311 295 307 40 21-Jan-98 6 6 5.8 5.58 442 1519 1668 1678 1475 11 7.8 7.5 8.4 290 359 374 252 46 27-Jan-98 7 7 6.3 5.61 488 1384 1403 1477 1225 11 8.3 8.1 8.3 261 340 347 349 54 04-Feb-98 8 8 16 5.61 466 1346 1496 1610 1185 11 8.5 9.2 9.2 276 318 320 313 61 11-Feb-98 9 9 3.3 5.4 459 1295 1398 1261 1091 11 9.4 9.2 9.4 304 400 368 376 68 18-Feb-98 10 10 5.5 5.4 555 1242 1111 1167 1031 11 9.4 9.1 9.5 280 316 321 308 75 25-Feb-98 11 11 39 5.6 443 1212 1082 1201 838 11 9.3 9.1 9.3 288 356 350 337 82 04-Mar-98 12 12 6.1 5.3 393 1256 1097 1218 811 11 9.0 8.9 9.5 257 396 302 407 89 11-Mar-98 13 13 7.8 5.3 475 1169 1005 1127 777 11 9.2 9.2 8.9 213 354 308 333 96 18-Mar-98 14 14 15 5.2 479 1101 982 1149 701 11 9.3 9.7 9.4 234 294 276 298 103 25-Mar-98 15 15 6.5 5.5 434 1080 952 1305 699 11 8.7 9.2 9.1 236 324 301 332 108 30-Mar-98 16 16 14 5.4 447 1067 937 1120 715 11 8.5 9.1 8.4 233 412 359 394 117 08-Apr-98 17 17 30.7 5.5 466 1113 1003 1221 737 9.8 8.3 8.8 8.5 260 349 326 354 124 15-Apr-98 18 18 65 5.6 452 1140 947 1228 732 10 7.9 8.4 8.5 252 319 312 329 131 22-Apr-98 19 19 47 5.6 469 1200 1026 1260 761 10 7.9 7.7 8.1 243 361 332 327 138 29-Apr-98 20 20 10.5 5.4 400 1033 890 1104 656 10 8.1 8.4 8.5 228 347 327 330 ay Date ycle F P total (mg/l) Cu (mg/l) Zn (mg/l) Pb (mg'l) Tiz Louv Brw FIM Tiz Louv Brw FIM Tiz Louv Brw FIM Tiz Lou Brw FIM 4 16-Dec-97 1 <0.1 0.74 <0.1 0.23 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.14 <0.1 0.2 0.1 0.1 <0.1 7 18-Dec-97 2 <0.1 0.20 <0.1 0.5 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.17 <0.1 <0.1 0.1 0.1 <0.1 18 30-Dec-97 3 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.13 <0.1 <0.1 0.1 0.1 <0.1 25 06-Jan-98 4 <0.1 <0.1 <0.1 0.20 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.3 <0.1 0.2 32 13-Jan-98 5 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 40 21-Jan-98 6 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 46 27-Jan-98 7 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 54 04-Feb-98 8 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 61 11-Feb-98 9 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 68 18-Feb-98 10 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.2 <0.1 75 25-Feb-98 11 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 82 04-Mar-98 12 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.2 <0.1 89 11-Mar-98 13 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.2 <0.1 96 18-Mar-98 14 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.2 <0.1 103 25-Mar-98 15 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 <0.1 <0.1 108 30-Mar-98 16 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.2 <0.1 <0.1 117 08-Apr-98 17 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 124 15-Apr-98 18 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 131 22-Apr-98 19 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 138 29-Apr-98 20 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 Note: distilled water batch 1 put into cells Dec 12th '97, cycle 1 chemical data taken on Dec.16th '97 220 Table l l - i (cont'd) Paste Leachate Chemisty - Flooded Water Cells ay Date ycle S04(mg/l) Ca (m g/i) Mg (mg/l) K (mg/l) Tiz Louv Brw FIM Tiz Louv Brw FIM Tiz Louv Brw FiM Tiz Lou Brw FIM 4 16-Dec-97 1 1.3 1.3 1.4 1.1 840 845 922 682 3.6 47 11 4.8 127 51 107 32 7 18-Dec-97 2 2.2 2.3 2.2 1.5 883 4.0 127 19 8.3 158 85 136 53 18 30-Dec-97 3 1.9 2.1 1.9 1.7 784 945 857 852 2.8 80 14 5.0 146 74 117 45 25 06-Jan-98 4 2.1 1.7 1.7 1.5 782 776 773 774 2.7 59 10 4.5 138 60 99 41 32 13-Jan-98 5 1.5 1.6 1.6 1.3 725 860 755 664 4.7 41 7.0 2.8 117 45 66 34 40 21-Jan-98 6 1.5 1.5 1.5 1.2 700 760 751 556 5.5 30 6.4 2.0 78 36 44 28 46 27-Jan-98 7 1.3 1.3 1.5 0.9 610 622 646 452 1.3 22 5.0 4.5 50 27 32 22 54 04-Feb-98 8 1.2 1.2 1.4 0.8 626 617 699 437 1.2 16 4.4 1.9 41 25 25 20 61 11-Feb-98 9 1.2 1.2 1.5 0.7 595 591 697 392 1.4 13 4.1 1.9 29 21 17 19 68 18-Feb-98 10 1.1 1.1 1.4 0.6 572 538 692 347 1.2 10 3.6 1.7 20 17 12 16 75 25-Feb-98 11 1.1 1.0 1.4 0.6 547 526 675 301 1.2 11 3.6 1.5 16 15 10 15 82 04-Mar-98 12 0.9 1.0 1.4 0.4 554 600 880 310 0.9 12 3.2 1.0 10 12 6.0 10 89 11-Mar-98 13 1.0 0.9 1.4 0.5 546 540 870 308 1.0 11 3.2 1.0 8.3 9.2 6.2 10 96 18-Mar-98 14 0.9 0.8 1.4 0.4 504 510 760 278 1.0 9.0 2.7 0.8 7.2 8.2 5.1 9.3 103 25-Mar-98 15 1.0 0.9 1.4 0.5 515 530 855 294 0.9 13 3.2 1.5 5.7 7.9 4.8 10 108 30-Mar-98 16 0.9 0.8 1.3 0.5 417 460 810 280 0.7 12 3.0 1.7 4.8 7.1 4.7 10 117 08-Apr-98 17 1.0 492 390 650 190 1.3 8.4 2.4 1.3 4.1 5.0 5.3 9.0 124 15-Apr-98 18 0.9 0.7 1.3 0.4 467 330 623 209 0.7 12 2.4 1.8 4.6 6.5 5.0 11 131 22-Apr-98 19 0.9 0.8 1.3 0.4 472 365 610 198 0.6 14 2.3 1.9 4.4 6.0 5.1 10 138 29-Apr-98 20 0.6 0.7 1.3 0.4 312 320 590 185 0.8 12 2.0 1.6 5.5 5.2 4.3 9.1 ay Date ycle Si (mg/1) Al (mg/l) As (mg/l) IIP** Tiz Louv Brw FIM Tiz Louv Brvv FIM Tiz Louv Brw FIM 4 16-Dec-97 1 2.2 9.7 1.9 2.4 < 0.30 < 0.30 < 0.30 < 0.30 <1.0 <1.0 <1.0 <1.0 7 18-Dec-97 2 3.3 9.8 2.0 3.4 < 0.30 < 0.30 < 0.30 < 0.30 <1.0 <1.0 <1.0 <1.0 18 30-Dec-97 3 2.9 9.7 2.2 3.2 < 0.30 < 0.30 < 0.30 < 0.30 <1.0 <1.0 <1.0 <1.0 25 06-Jan-98 4 2.0 9.7 1.6 2.7 < 0.30 < 0.30 < 0.30 < 0.30 <1.0 <1.0 <1.0 <1.0 32 13-Jan-98 5 2.9 10 2.2 3.2 <1.0 <1.0 <1.0 <1.0 40 21-Jan-98 6 3.0 9.8 1.9 3.0 <1.0 <1.0 <1.0 <1.0 46 27-Jan-98 7 3.0 8.6 1.5 2.5 <1.0 <1.0 <1.0 <1.0 54 04-Feb-98 8 2.2 9.9 1.5 2.0 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 61 11-Feb-98 9 2.4 10 1.2 2.6 <0.3 <0.3 < 0.3 <0.3 <1.0 <1.0 <1.0 <1.0 68 18-Feb-98 10 2.2 9.7 1.0 2.1 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 75 25-Feb-98 11 1.9 9.8 1.4 2.6 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 82 04-Mar-98 12 1.5 6.8 0.5 1.6 0.32 < 0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 89 11-Mar-98 13 1.6 6.1 0.5 1.3 0.32 < 0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 96 18-Mar-98 14 1.7 6.7 0.5 1.5 0.34 < 0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 103 25-Mar-98 15 1.7 6.6 0.8 1.6 0.33 < 0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 108 30-Mar-98 16 1.5 6.6 0.9 2.0 0.63 < 0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 117 08-Apr-98 17 0.8 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 124 15-Apr-98 18 1.6 8.9 2.0 4.1 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 131 22-Apr-98 19 1.3 9.0 2.0 4.1 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 138 29-Apr-98 20 2.5 8.6 1.7 3.6 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 Note: distilled water batch 1 put into cells Dec 12th '97, cycle 1 chemical data taken on Dec. 16th '97 221 Table 11-2 Paste Leachate Chemisty - Cycled Water Cells Distilled water data Paste leach water Redox potential Day Date Cycle bate cond PH cdox conductivity (uS) pH (mV SHE) (uS) (mV) Tiz Louv Brw FIM Tiz Louv Brw FIM Tiz Louv Brwk FIM 1 12-Dec-97 1 1 27.5 5.5 499 1880 1660 1902 1565 10 8.2 7.6 8.4 300 358 368 361 6 18-Dec-97 2 2 28.8 5.5 359 1000 1090 1330 900 9.3 7.5 7.6 7.4 367 405 417 411 12 30-Dec-97 3 3 26.8 5.6 485 1038 1054 1268 892 7.3 6.6 6.5 6.3 399 420 421 423 20 07-Jan-98 4 4 6.09 5.6 509 885 846 1060 724 8.1 7.3 6.3 6.3 356 367 371 356 28 15-Jan-98 5 5 17.7 5.5 469 1192 1104 1322 906 7.8 6.6 6.4 7.3 300 362 364 334 34 21-Jan-98 6 6 5.83 5.58 442 692 639 842 495 6.6 6.3 6.1 6.7 321 323 338 329 41 28-Jan-98 7 7 6.3 5.61 488 935 775 1025 633 6.8 6.9 6.5 6.8 385 390 383 390 48 04-Feb-98 8 8 4.5 5.52 778 609 860 491 6.4 6.8 6.5 6.7 413 409 412 412 50 11-Febr98 9 9 4.0 5.32 498 795 635 838 504 6.7 7.1 6.8 7.1 369 374 385 385 57 18-Feb-98 10 10 17.9 5.42 779 601 813 476 7.6 7.7 7.3 7.7 375 378 388 400 64 25-Feb-98 11 11 34.9 5.64 432 803 647 928 528 7.4 7.5 7.2 7.4 352 356 347 361 71 04-Mar-98 12 12 6.57 5.34 404 810 596 818 509 7.2 7.5 7.1 7.5 345 328 365 352 78 11-Mar-98 13 13 14.9 5.2 479 684 459 669 400 7.0 7.4 7.1 7.6 379 373 390 392 84 18-Mar-98 14 14 2.0 5.2 464 758 540 758 459 6.7 7.6 7.1 7.3 381 405 396 412 90 24-Mar-98 15 15 8.8 5.55 481 721 496 708 432 6.7 6.8 6.6 6.8 445 419 434 433 99 30-Mar-98 16 16 15 5.59 490 745 520 724 745 6.9 7.5 7.0 7.5 418 393 398 394 107 08-Apr-98 17 17 55 5.62 411 820 625 847 693 6.9 7.3 7.0 7.5 374 393 396 390 113 16-Apr-98 18 18 48 5.6 470 863 623 801 676 6.9 7.2 6.9 7.7 390 395 378 357 120 22-Apr-98 19 19 4.62 5.3 389 785 567 765 773 6.9 7.3 7.5 7.6 348 358 357 360 138 29-Apr-98 20 20 18 5.59 767 592 749 628 7.3 7.4 7.4 7.8 332 329 338 338 Day Date Cycle Fc total (mg/l) Cu (mg/l) Zn (mg/l) Pb (mg/l) Tiz Louv Brw FIM Tiz Louv Brw FIM Tiz Louv Brw FIM Tiz Louv Brw FIM 1 12-Dec-97 1 <0.1 0.74 <0.1 0.23 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 <0.1 0.2 0.1 0.1 <0.1 6 18-Dec-97 2 <0.1 0.5 <0.1 1.78 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.3 0.1 0.2 0.1 0.1 <0.1 12 30-Dec-97 3 <0.1 0.3 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.5 0.2 <0.1 0.1 0.1 <0.1 20 07-Jan-98 4 <0.1 0.7 <0.1 0.2 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 28 15-Jan-98 5 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 34 21-Jan-98 6 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 41 28-Jan-98 7 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.2 <0.1 <0.1 <0.1 <0.1 <0.1 48 04-Feb-98 8 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.3 0.1 <0.1 <0.1 <0.1 <0.1 50 11-Feb-98 9 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.1 <0.1 0.1 0.1 0.4 0.1 <0.1 <0.1 <0.1 <0.1 57 18-Feb-98 10 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.1 <0.1 0.1 0.1 0.2 0.1 <0.1 <0.1 <0.1 <0.1 64 25-Feb-98 11 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 <0.1 <0.1 0.1 0.1 0.6 0.1 <0.1 <0.1 <0.1 <0.1 71 04-Mar-98 12 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.12 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 78 11-Mar-98 13 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.11 <0.1 0.3 <0.1 <0.1 <0.1 <0.1 <0.1 84 18-Mar-98 14 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.14 <0.1 0.3 <0.1 <0.1 <0.1 <0.1 <0.1 90 24-Mar-98 15 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.11 <0.1 0.3 <0.1 <0.1 <0.1 <0.1 <0.1 99 30-Mar-98 16 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.3 <0.1 0.5 <0.1 <0.1 <0.1 <0.1 <0.1 107 08-Apr-98 17 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 1.2 <0.1 2.2 <0.1 <0.1 <0.1 <0.1 <0.1 113 16-Apr-98 18 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.5 <0.1 1.4 <0.1 <0.1 <0.1 <0.1 <0.1 120 22-Apr-98 19 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.14 <0.1 0.5 <0.1 0.7 0.6 <0.1 <0.1 138 29-Apr-98 20 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.3 <0.1 1.3 <0.1 0.3 0.42 <0.1 <0.1 Note: distilled water put into cells for 24 hours prior to leachate analysis 222 Table 11-2 (cont'd) Paste Leachate Chemistry - Cycled Water Cells Day Date Cycle SO, (mg/l) Ca (mg/l) Mg (mg/l) K (mg/l) Tiz Louv Brw FJM Tiz Louv Brw FIM Tiz Louv Brw FIM Tiz Louv Brw FIM 1 12-Dec-97 1 1.3 1.3 1.4 1.1 840 845 922 682 3.6 47 11 4.8 127 51 107 32 6 18-Dec-97 2 0.9 0.8 1.0 0.7 475 543 621 530 3.0 35 6.3 4.0 113 36 86 25 12 30-Dec-97 3 0.6 0.7 0.9 0.8 318 333 431 332 2.6 30 7.8 3.8 80 27 59 20 20 07-Jan-98 4 0.5 0.5 0.8 0.4 232 225 370 204 2.0 24 5.6 2.6 55 19 32 16 28 15-Jan-98 5 0.8 0.7 1.1 0.6 340 261 481 276 3.0 48 6.0 6.8 44 19 25 17 34 21-Jan-98 6 0.4 0.3 0.5 0.2 167 214 238 111 4.5 29 4.0 2.0 17 8.2 9.0 8.5 41 28-Jan-98 7 0.4 0.3 0.6 0.3 208 200 265 123 4.8 22 4.7 2.0 13 6.7 6.7 8.6 48 04-Feb-98 8 0.4 0.3 0.4 0.2 210 148 257 120 2.2 8.8 3.3 1.6 10 5.9 5.5 8.4 50 11-Feb-98 9 0.4 0.3 0.5 0.2 230 178 296 125 2.4 10 3.9 1.7 8.6 6.1 4.9 8.9 57 18-Feb-98 10 0.3 0.3 0.4 0.2 225 150 250 116 2.3 8.3 3.4 1.6 5.9 1.9 3.5 7.9 64 25-Feb-98 11 0.4 0.3 0.6 0.2 255 174 329 129 2.3 10 4.1 1.8 6.3 5.9 5.3 9.1 71 04-Mar-98 12 0.4 0.3 0.5 0.2 340 170 353 120 2.1 1.1 3.8 1.0 4.1 3.7 2.2 5.7 78 11-Mar-98 13 0.4 0.3 0.5 0.2 340 164 325 160 2.8 9.0 2.9 1.9 3.5 3.0 1.9 5.0 84 18-Mar-98 14 0.4 0.3 0.5 0.2 335 178 320 158 2.3 9.0 2.2 1.5 3.3 3.2 1.8 5.1 90 24-Mar-98 15 0.4 0.2 0.5 0.2 285 170 310 160 3.0 8.7 2.5 1.3 3.0 3.2 1.9 5.1 99 30-Mar-98 16 0.5 0.3 0.5 0.3 354 190 350 270 2.8 12 3.6 2.1 3.0 3.6 2.0 8.6 107 08-Apr-98 17 0.5 0.3 271 190 260 210 2.4 3.1 1.7 3.0 5.0 7.4 13 6.4 113 16-Apr-98 18 0.5 0.3 0.6 0.4 260 152 232 158 2.7 12 2.0 1.8 4.4 4.8 4.7 8.1 120 22-Apr-98 19 0.4 0.3 0.4 0.3 233 124 195 168 1.8 9.2 2.3 1.4 1.7 2.4 1.4 5.2 138 29-Apr-98 20 0.6 0.3 0.4 0.2 249 150 233 142 2.0 11 3.2 1.3 2.4 3.3 2.4 5.0 Day Date Cycle Si (mg/l) Al (mg/l) As (mg/l) Tiz Louv Brw FIM Tiz Louv Brw FIM Tiz Louv Brw FIM 1 12-Dec-97 1 2.2 9.7 1.9 2.4 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 6 18-Dec-97 2 2.0 7.1 2.0 2.9 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 12 30-Dec-97 3 1.5 5.0 1.8 2.8 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 20 07-Jan-98 4 <0.5 4.5 <0.5 1.4 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 28 15-Jan-98 5 <0.5 6.1 1.3 2.8 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 34 21-Jan-98 6 <0.5 4.4 <0.5 1.1 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 41 28-Jan-98 7 <0.5 4.2 <0.5 1.7 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 48 04-Feb-98 8 0.8 5.4 1.2 1.9 <0.3 0.5 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 50 11-Feb-98 9 1.0 3.7 1.4 1.7 <0.3 0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 57 18-Feb-98 10 1.4 4.0 1.4 2.0 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 64 25-Feb-98 11 2.5 5.7 2.1 4.2 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 71 04-Mar-98 12 1.4 3.8 1.0 1.6 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 78 11-Mar-98 13 1.2 3.2 0.9 1.3 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 84 18-Mar-98 14 1.4 4.0 0.9 1.6 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 90 24-Mar-98 15 1.6 4.1 1.0 1.9 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 99 30-Mar-98 16 2.0 5.1 1.4 3.0 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 107 08-Apr-98 17 2.3 5.2 1.2 1.4 <0.3 <0.3 <0.3 < 0.3 <1.0 <1.0 <1.0 <1.0 113 16-Apr-98 18 5.3 9.7 4.8 6.0 <0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 120 22-Apr-98 19 2.0 4.8 1.7 2.9 <0.3 <0.3 < 0.3 <0.3 <1.0 <1.0 <1.0 <1.0 138 29-Apr-98 20 3.6 7.3 3.0 4.2 < 0.3 <0.3 <0.3 <0.3 <1.0 <1.0 <1.0 <1.0 223 Table 11-3 Paste Leachate Chemisty - Fe 2 (S0 4 ) 3 Solution Cells Fe2(S04)3 Solution Paste leach solution Redox potential Day Date Cycle bate cond pH OR Feto SO, conductivity (uS) pH (mV SHE) no. (uS) (mV (mg/l) (g/l) Tiz Louv Brw FIM Tiz Lou Brw FIM Tiz Lou Br.v FIM 0.5 12-Dec-97 1 1 2735 2.9 587 920 1 5 6.5 6.4 6.3 6.6 6 17-Dec-97 2 2 2360 2.3 596 826 1 5 3240 1890 3020 1800 6.2 6.2 6.1 6.0 181 154 155 169 18 29-Dec -97 3 3 2610 2.6 623 754 1 4 1855 1605 1623 1432 6.7 6.2 6.3 6.0 228 239 199 252 25 05-Jan-98 4 4 1836 2.6 624 780 1 4 1472 1362 1339 1315 6.3 6.3 6.2 5.9 234 223 226 327 28 08-Jan-98 5 5 1967 2.6 625 740 1338 1311 1268 1350 5.9 5.7 5.9 6.1 278 304 243 182 32 12-Jan-98 6 6 1977 2.6 620 734 1 5 1391 1328 1363 1337 5.7 5.6 5.9 5.6 290 374 258 360 35 15-Jan-98 7 7 1929 2.4 669 689 1 6 1407 1335 1258 1350 5.2 5.3 5.5 5.1 412 409 298 449 39 19-Jan-98 8 8 1942 2.6 629 726 1 4 1318 1314 1331 1307 5.3 5.2 5.7 5.8 341 329 278 265 43 23-Jan-98 9 9 2420 2.6 623 772 1 7 1320 1298 1275 1298 5.2 5.1 5.6 5.7 302 299 239 235 46 26-Jan-98 10 10 2320 2.5 626 692 1 6 1496 1488 1418 1351 4.1 3.9 5.0 5.1 433 470 317 331 49 29-Jan-98 11 11 2370 2.6 626 723 1 3 1448 1453 1349 1232 4.0 3.6 4.4 5.2 428 508 372 306 53 02-Feb-98 12 12 1979 2.5 624 723 1 5 1424 1512 1448 1399 4.5 4.3 5.3 5.7 400 419 365 289 57 06-Feb-98 13 13 2700 2.5 623 687 1 6 1474 1487 1434 1316 4.4 4.0 4.9 5.5 405 447 381 289 60 09-Feb-98 14 14 2400 2.5 625 638 1 3 1502 1517 1552 1304 3.9 3.8 4.1 5.1 469 495 451 370 63 12-Feb-98 15 15 1911 2.7 617 679 1 .4 1447 1397 1439 1219 4.0 3.9 4.2 5.4 433 459 420 294 67 16-Feb-98 16 16 1894 2.5 630 548 1 .3 1439 1420 1390 1272 4.1 3.9 4.4 5.6 435 457 412 311 70 19-Feb-98 17 17 2460 2.5 621 665 1 .4 1494 1490 1504 1320 3.9 3.9 4.1 5.2 443 439 430 344 74 23-Feb-98 18 18 2230 2.6 630 611 1 .3 1556 1460 1503 1328 4.0 4.0 4.3 5.5 436 432 408 289 77 26-Feb-98 19 19 2410 2.6 624 609 1 .4 1426 1380 1400 1225 3.9 3.9 4.0 5.2 435 428 421 322 81 02-Mar -98 20 20 2330 2.7 612 645 1 .4 1375 1354 1232 4.0 4.1 4.5 5.6 440 431 409 302 84 05-Mar -98 21 21 2510 2.5 628 483 1 .3 1325 1320 1361 1169 3.8 3.8 3.9 5.0 463 477 450 389 88 09-Mar -98 22 22 1896 2.5 628 695 1 .5 1462 1431 1457 1233 4.0 4.1 4.4 5.6 434 419 394 287 91 12-Mar-98 23 23 2550 2.5 621 550 1 .5 1374 1353 1387 1280 3.7 3.7 3.9 5.3 466 484 442 399 95 16-Mar-98 24 24 2500 2.6 619 598 1 .5 1536 1503 1555 1290 3.8 3.8 4.0 5.1 439 452 437 345 99 20-Mar -98 25 25 2460 2.7 622 564 1 .5 1407 1355 1391 1143 3.9 3.9 4.1 5.5 433 444 419 291 103 24-Mar -98 26 26 2480 2.5 621 604 1 .5 1466 1377 1395 1199 4.1 4.1 4.4 5.8 426 427 404 255 109 30-Mar -98 27 27 2540 2.5 623 630 1 .4 1396 1364 1375 1143 3.9 3.9 4.1 5.6 430 439 405 254 112 02-Apr-98 28 28 1951 2.5 627 641 1 .4 1458 1390 1420 1188 3.7 3.6 3.8 5.3 450 457 444 304 116 06-Apr -98 29 29 2360 2.5 627 520 1 .4 1523 1479 1535 1272 3.8 3.8 4.0 5.6 450 451 398 281 118 08-Apr -98 30 30 1934 2.5 616 630 1 .5 1465 1438 1429 1216 3.5 3.5 3.6 5.1 473 488 460 312 123 13-Apr-98 31 31 2370 2.5 615 1478 1489 1461 1302 3.9 3.7 4.0 5.6 428 451 408 277 126 16-Apr-98 32 32 1901 2.6 624 721 1 .6 3.7 3.6 3.8 5.6 130 20-Apr-98 33 33 2390 2.5 621 725 1 .6 1454 1434 1418 1320 3.9 3.7 4.0 4.9 432 445 411 333 133 23-Apr-98 34 34 2580 2.5 617 745 1 .5 1405 1399 1417 1275 3.6 3.4 3.6 4.2 469 512 466 395 137 27-Apr-98 35 35 1937 2.5 615 727 1 .5 1564 1539 1596 1310 3.8 3.5 3.7 5.2 443 489 448 330 140 30-Apr-98 36 36 2530 2.5 617 910 1 .4 1584 1565 1576 1374 3.6 3.4 3.6 4.9 462 506 482 342 144 04-May-98 37 37 2570 2.5 614 550 1 .4 1565 1542 1583 1280 3.7 3.5 3.7 5.3 455 494 470 306 147 07-May-98 38 38 2410 2.6 625 470 1 .4 1608 1543 1636 1411 3.6 3.4 3.5 4.6 459 493 467 357 151 11-May-98 39 39 2430 2.5 622 840 1 .7 1615 1587 1596 1320 3.7 3.5 3.7 5.1 454 481 444 307 154 14-May-98 40 40 2470 2.5 624 725 1 .4 1493 1466 1479 1305 3.5 3.4 3.6 4.7 465 479 455 369 Notes: Empty spaces denote unavailable results O R P : Oxidation-reduction (redox) potential (SHE) 224 Table 11-3 (cont'd) Paste Leachate Chemisty - Fe 2 (S0 4 ) 3 Solution Cells Day Date Cycle Fe total (mg/l) Cu (mg/l) Zn (mg/l) Pb(mg/I) Tiz Louv SWmBMi Tiz Lou Brw F llWiil Tiz LOU Brw FIM,, 0.5 12-Dec-97 1 260 400 328 348 <0.1 0.1 <0.1 <0.1 1.3 0.6 11 1.3 0.5 0.1 <0.1 <0.1 6 17-Dec-97 2 94 280 154 169 <0.1 0.1 <0.1 <0.1 0.6 0.4 8.6 1.0 <0.1 <0.1 0.1 <0.1 18 29-Dec-97 3 73 75 74 138 <0.1 0.1 <0.1 <0.1 0.4 3.8 4.7 0.8 <0.1 <0.1 0.2 <0.1 25 05-Jan-98 4 88 216 190 230 <0.1 0.2 <0.1 <0.1 0.9 0.5 9.1 1.1 <0.1 <0.1 <0.1 <0.1 28 08-Jan-98 5 290 329 290 155 0.6 0.5 0.4 0.4 3.3 1.4 17 1.3 <0.1 0.2 <0.1 <0.1 32 12-Jan-98 6 224 290 279 249 0.6 0.5 0.3 0.4 3.1 1.2 18 1.6 <0.1 0.2 0.2 <0.1 35 15-Jan-98 7 183 201 201 134 0.6 0.7 0.3 0.4 5.2 1.8 23 2.1 0.6 0.2 0.3 <0.1 39 19-Jan-98 8 248 368 304 261 0.6 0.3 0.4 0.4 4.2 1.7 18 1.8 0.4 0.2 0.2 <0.1 43 23-Jan-98 9 278 397 323 271 0.3 0.4 0.3 0.3 3.6 1.3 18 2.0 0.3 <0.1 0.3 <0.1 46 26-Jan-98 10 340 417 366 371 0.5 0.4 0.4 0.5 4.3 1.5 19 2.1 0.8 <0.1 0.9 <0.1 49 29-Jan-98 11 388 438 388 335 0.4 0.9 0.3 0.4 4.1 1.8 18 2.1 0.9 <0.1 1.4 <0.1 53 02-Feb-98 12 346 399 357 267 0.4 0.4 0.5 0.6 4.1 1.5 17 1.7 0.8 <0.1 0.7 <0.1 57 06-Feb-98 13 333 451 357 321 <0.1 0.1 0.2 <0.1 3.2 1.4 19 1.8 0.6 <0.1 1.2 <0.1 60 09-Feb-98 14 438 443 379 339 0.2 0.1 0.1 <0.1 3.6 1.2 17 2.0 1.0 <0.1 1.6 <0.1 63 12-Feb-98 15 417 459 378 361 0.1 0.3 0.1 <0.1 3.3 1.5 15 1.7 0.9 <0.1 1.4 <0.1 67 16-Feb-98 16 406 421 336 0.1 0.3 0.1 3.8 1.6 16 0.8 0.1 1.0 70 19-Feb-98 17 424 492 359 343 0.1 0.3 0.1 <0.1 3.2 1.4 14 1.8 1.0 0.1 1.4 <0.1 74 23-Feb-98 18 343 436 345 304 <0.1 0.2 0.1 <0.1 3.2 1.5 15 1.5 0.8 <0.1 1.1 <0.1 77 26-Feb-98 19 402 479 355 356 <0.1 0.2 0.1 <0.1 3.2 1.4 13 1.8 1.1 <0.1 1.2 <0.1 81 02-Mar-98 20 329 387 290 236 0.2 0.2 0.2 0.2 3.6 1.3 14 1.7 0.9 <0.1 0.9 0.1 84 05-Mar-98 21 490 411 270 260 0.1 0.1 0.2 0.2 4.0 1.5 12 1.6 0.3 1.3 88 09-Mar-98 22 438 407 230 221 0.1 0.1 0.2 0.2 5.0 1.6 9.7 1.0 0.3 0.6 91 12-Mar-98 23 518 510 393 250 0.2 0.1 0.3 0.4 3.8 1.3 15 1.5 0.3 1.7 95 16-Mar-98 24 502 384 310 305 0.1 0.3 0.2 0.2 4.7 1.3 12 1.2 0.1 1.0 99 20-Mar-98 25 476 370 295 260 <0.1 0.1 0.2 0.2 4.7 1.3 1.0 0.2 103 24-Mar-98 26 348 420 254 166 <0.1 0.2 0.1 0.3 4.7 1.7 0.9 0.3 109 30-Mar-98 27 416 417 360 180 <0.1 0.2 0.2 0.4 4.5 1.8 1.1 0.2 112 02-Apr-98 28 448 497 350 215 <0.1 0.2 0.2 0.3 4.7 1.8 1.6 0.2 116 06-Apr-98 29 433 370 325 312 <0.1 0.1 0.2 0.2 1.6 16 2.8 1.8 <0.1 <0.1 <0.1 118 08-Apr-98 30 549 502 390 310 <0.1 0.3 0.2 0.1 1.0 11 2.9 2.3 <0.1 <0.1 <0.1 123 13-Apr-98 31 552 594 605 423 0.3 0.1 0.3 0.5 6.9 1.4 16 2.3 1.4 0.2 1.5 0.2 126 16-Apr-98 32 530 494 900 370 0.3 0.2 0.7 0.4 6.2 1.3 22 2.7 1.9 0.5 3.3 0.2 130 20-Apr-98 33 530 935 800 395 0.2 0.3 0.7 0.4 6.7 3.0 14 2.2 1.3 0.6 1.2 0.2 133 23-Apr-98 34 630 1180 640 530 0.0 0.3 0.4 0.3 5.9 2.5 14 2.3 1.9 0.7 2.1 0.2 137 27-Apr-98 35 570 605 554 580 0.2 0.2 0.4 0.2 6.6 1.8 14 2.1 1.4 0.3 1.5 0.2 140 30-Apr-98 36 700 690 498 520 0.2 0.3 0.3 0.1 7.1 1.6 11 2.3 2.1 0.3 1.4 0.2 144 04-May-98 37 720 730 610 474 0.1 0.3 0.3 <0.1 9.5 2.1 13 2.4 2.0 0.3 1.5 0.2 147 07-May-98 38 490 485 470 330 <0.1 0.1 <0.1 <0.1 7.7 1.4 13 3.0 1.4 1.0 <0.1 151 11-May-98 39 555 480 433 380 <0.1 0.1 <0.1 <0.1 12 1.7 13 2.2 1.2 <0.1 <0.1 <0.1 154 14-May-98 40 472 430 460 393 <0.1 0.1 <0.1 <0.1 8.9 1.4 12 3.1 1.3 <0.1 <0.1 <0.1 225 Table 11-3 (cont'd) Paste Leachate Chemisty - Fe 2 (S0 4 ) 3 Solution Cells Days Date Cycle SO, (mg/l) Tiz LOUV Brw F iM Ca (mg/l) Tiz Louv Brw F IM. Mg (mg/l) Tiz Louv Brw FIMj K (mg/l) 0.5 12-Dec-97 1 1.9 1.9 2.0 1.7 586 498 590 525 7.1 63 8.0 142 39.3 104 27 6 17-Dec-97 2 2.1 2.5 2.1 2.0 756 691 742 736 8.0 108 10 177 55.3 122 36 18 29-Dec -97 3 1.8 2.0 1.5 1.6 835 773 770 631 8.0 115 8.3 120 32 51 28 25 05-Jan-98 4 1.6 1.7 1.5 1.5 761 581 800 546 8.0 61 6.4 46 18.2 20 20 28 08-Jan-98 5 1.6 1.6 1.6 1.6 562 390 477 583 8.8 45 8.9 7.5 11 6.3 5.8 14 32 12-Jan-98 6 1.8 1.6 1.6 1.5 611 355 561 440 9.7 48 11 6.3 6.3 4.3 4.4 9.2 35 15-Jan-98 7 1.6 1.5 1.5 1.4 543 381 557 433 14 72 16 8.5 4.4 4.0 3.6 8.0 39 19-Jan-98 8 1.6 1.6 1.5 1.4 494 401 484 411 9.3 57 9.7 7.0 2.6 3.7 2.1 6.2 43 23-Jan-98 9 1.7 1.5 1.6 1.7 559 350 522 522 9.2 47 9.6 8.0 2.4 3.0 2.1 7.0 46 26-Jan-98 10 1.5 1.4 1.5 1.6 446 241 435 416 8.3 35 8.3 5.5 1.4 1.9 1.5 4.8 49 29-Jan-98 11 1.6 1.4 1.7 1.4 419 211 443 394 8.7 39 9.8 6.4 1.5 2.3 1.8 4.3 53 02-Feb-98 12 1.7 1.7 1.8 1.5 472 275 461 466 9.7 50 11 7.6 1.3 2.4 1.5 3.9 57 06-Feb-98 13 1.5 1.5 1.5 1.4 420 277 529 429 9.0 42 11 6.5 1.4 2.4 1.9 3.7 60 09-Feb-98 14 1.4 1.3 1.4 1.4 353 209 444 385 9.4 34 9.7 6.4 1.0 1.9 1.4 3.8 63 12-Feb-98 15 1.4 1.5 1.3 1.5 344 244 427 395 9.9 43 9.9 6.4 1.0 2.6 1.5 3.7 67 16-Feb-98 16 1.6 1.3 1.4 393 271 442 12 51 11 1.0 1.9 1.4 70 19-Feb-98 17 1.4 1.3 1.2 1.4 307 237 381 367 9.3 45 8.6 6.2 0.9 2.1 1.5 3.6 74 23-Feb-98 18 1.2 1.4 1.3 1.3 330 266 443 372 11 56 11 6.7 0.9 2.1 1.5 3.0 77 26-Feb-98 19 1.3 1.3 1.2 1.5 305 229 337 372 10 49 8.4 6.4 1.0 2.2 1.8 3.4 81 02-Mar-98 20 1.5 1.6 1.6 1.3 356 220 407 343 12 45 9.0 5.9 84 05-Mar-98 21 1.4 1.5 1.2 285 180 280 10 37 7.0 1.2 2.1 2.6 3.0 88 09-Mar-98 22 1.6 1.6 1.2 377 222 294 15 69 8.6 1.2 2.4 3.1 3.0 91 12-Mar-98 23 1.5 1.8 1.4 244 113 340 8.2 25 9.0 1.0 1.8 2.4 1.0 95 16-Mar-98 24 1.5 1.3 1.4 328 213 320 13 48 8.8 1.0 2.1 2.2 2.0 99 20-Mar -98 25 1.6 1.2 1.4 357 218 14 56 0.9 1.8 1.0 2.0 103 24-Mar -98 26 1.7 1.8 1.6 447 320 16 79 0.9 2.5 1.0 2.0 109 30-Mar-98 27 1.6 1.8 379 254 13 62 0.8 2.0 1.0 2.0 112 02-Apr-98 28 1.6 1.6 297 180 11 42 0.8 1.7 1.0 2.0 116 06-Apr-98 29 1.7 1.4 1.8 1.6 364 169 420 410 58 14 9.0 0.8 2.0 1.4 2.9 118 08-Apr-98 30 1.5 1.8 1.4 1.7 225 107 230 285 32 7.3 7.0 2.5 1.7 2.4 3.8 123 13-Apr-98 31 1.8 1.7 2.0 1.8 355 180 440 460 13 63 11 7.8 1.6 2.2 2.4 3.5 126 16-Apr-98 32 1.6 1.3 3.0 1.6 279 134 560 420 10 40 15 9.6 1.7 2.2 3.8 3.2 130 20-Apr-98 33 1.5 2.8 2.7 1.6 320 270 480 380 12 93 15 8.6 1.2 2.7 2.8 2.9 133 23-Apr-98 34 1.5 2.7 1.7 1.5 236 207 290 280 7.9 70 7.9 6.1 1.0 2.1 1.5 2.0 137 27-Apr-98 35 1.6 1.6 1.7 1.7 284 152 327 333 11 50 9.5 7.0 0.8 1.2 1.3 2.1 140 30-Apr-98 36 1.8 1.7 1.5 1.5 280 133 255 294 10 49 6.6 5.8 0.5 1.2 1.0 1.6 144 04-May-98 37 2.0 1.9 1.5 1.6 340 154 320 348 14 64 8.6 6.7 0.6 1.1 0.9 1.7 147 07-May-98 38 1.6 1.3 1.6 1.4 314 140 321 443 8.0 45 11 9.0 0.5 1.6 1.2 2.0 151 11-May-98 39 1.7 1.5 1.7 1.6 466 183 355 400 17 65 13 9.0 0.6 2.0 0.9 1.6 154 14-May-98 40 1.6 1.4 1.6 1.5 296 132 290 350 10 43 11 9.0 0.5 1.7 0.9 1.8 226 Table 11-3 (cont'd) Paste Leachate Chemisty - Fe 2 (S0 4 ) 3 Solution Cells Days Date C y c l e S i (mg/l) A l (mg'l) As (mg/l) hz 1 ouv Brw FIM Tiz Louv brw FIM •Tiz-, Louv Brw FIM 0.5 12-Dec-97 1 7.0 12 6.2 <0.3 <0.3 <0.3 <0.3 <1 <1 <1 <1 6 17-Dec-97 2 6.6 14 6.9 <0.3 <0.3 <0.3 <0.3 <1 <1 <1 <1 18 29-Dec-97 3 5.4 14 6.2 <0.3 <0.3 <0.3 <0.3 <1 <1 <1 <1 25 05-Jan-98 4 7.0 12 5.9 <0.3 <0.3 <0.3 <0.3 <1 <1 <1 <1 28 08-Jan-98 5 9.6 17 9.8 9.4 <0.5 <0.5 <0.5 2.7 <1 <1 <1 <1 32 12-Jan-98 6 9.7 17 8.9 8.0 <0.5 <0.5 <0.5 0.7 <1 <1 <1 <1 35 15-Jan-98 7 15 23 12 9.3 1.2 1.82 <0.5 <0.5 <1 <1 <1 <1 39 19-Jan-98 8 12 23 8.9 8.0 0.6 <0.5 <0.5 <0.5 <1 <1 <1 <1 43 23-Jan-98 9 12 21 10 8.6 <0.5 <0.5 <0.5 <0.5 <1 <1 <1 <1 46 26-Jan-98 10 12 19 11 7.7 0.6 1.63 <0.5 <0.5 <1 <1 <1 <1 49 29-Jan-98 11 13 20 10 8.3 0.8 <0.5 <0.5 <0.5 <1 <1 <1 <1 53 02-Feb-98 12 13 19 10 9.6 0.7 1.1 <0.5 <0.5 <1 <1 <1 <1 57 06-Feb-98 13 12 20 12 8.6 0.7 1.5 0.9 <0.3 <1 <1 <1 <1 60;;. 09-Feb-98 14 13 18 12 8.4 1.4 2.3 0.9 <0.3 <1 <1 <1 <1 6 3 1 12-Feb-98 15 13 19 11 8.1 1.0 2.0 0.8 <0.3 <1 <1 <1 <1 67 16-Feb-98 16 14 19 10 1.4 2.5 0.9 <1 <1 <1 <1 70 19-Feb-98 17 13 20 10 8.6 1.0 1.7 1.0 0.5 <1 <1 <1 <1 74 23-Feb-98 18 12 20 10 8.3 1.1 1.7 0.8 0.6 <1 <1 <1 <1 77 26-Feb-98 19 14 20 10 10 <0.3 1.6 0.8 <0.3 <1 <1 <1 <1 81 02-Mar-98 20 13 17 9.7 8.0 <0.3 <0.3 <0.3 <0.3 <1 <1 <1 <1 84 05-Mar-98 21 13 16 8.6 1.8 1.6 0.5 1.0 <1 <1 <1 <1 88 09-Mar-98 22 13 17 8.5 1.2 0.9 0.7 <0.3 <1 <1 <1 1 91 12-Mar-98 23 11 15 8.3 2.3 2.1 1.2 1.0 <1 <1 <1 1 95 16-Mar-98 24 12 12 10 1-7 1.3 0.7 1.0 <1 1.2 <1 <1 99 20-Mar-98 25 12 13 1.1 0.8 1.0 <0.3 <1 1.0 2.0 <1 103 24-Mar-98 26 12 16 0.5 0.7 <0.3 <0.3 <1 1.4 <1 <1 109 30-Mar-98 27 13 18 0.7 0.9 <0.3 <0.3 <1 1.3 2.0 <1 112 02-Apr-98 28 12 17 1.7 1.4 1.0 <0.3 <1 <1 1 <1 116 06-Apr-98 29 15 <0.3 <0.3 <0.3 <0.3 <1 <1 <1 <1 118 08-Apr-98 30 17 <0.3 <0.3 <0.3 <0.3 <1 <1 <1 <1 123 13-Apr-98 31 17 19 15 15 1.1 1.1 1.0 <0.3 <1 <1 <1 <1 126 16-Apr-98 32 19 20 29 16 1.6 1.3 2.1 <0.3 <1 <1 <1 <1 130 20-Apr-98 33 17 36 22 15 1.3 2.1 1.5 0.3 <1 <1 <1 <1 133 23-Apr-98 34 15 33 16 12 2.3 1.4 2.1 1.6 <1 <1 <1 <1 137 27-Apr-98 35 15 18 14 13 1.4 2.1 1.4 1.0 <1 <1 <1 <1 140 30-Apr-98 36 16 18 13 12 2.0 2.4 1.3 0.9 <1 <1 <1 <1 144 04-May-98 37 18 20 13 13 1.8 2.0 1.1 0.5 <1 <1 <1 <1 147 07-May-98 38 15 15 12 13 1.8 2.0 1.4 0.4 <1 <1 <1 <1 151 11-May-98 39 19 16 11 10 2.1 1.5 0.9 <0.3 <1 <1 <1 <1 154 14-May-98 40 15 14 11 11 2.0 1.7 1.3 0.7 <1 <1 <1 <1 227 APPENDIX I I I - ANALYTICAL RESULTS: PASTE BACKFILL SOLID PHASE 228 to CD o> NT CO CO LU O o CC Q- _ l < < z l i o< X Z < sis £ w COLLLUo U - O C t S 0 | - 2 ~ - C L U COLLI y j C C t o g N ZLLI t <5 n o m>> < cc CD yj cc UJ < > Z O O ^ i— co C0 CC X CD E O _ o * 3 CD sz CC 03 C 0 L " U (D OJ CM ra- ja CO CO jo o> * J2-i§ 9 o 0 6 | ° u J — m -c "Z. t- ^- c x 1 CMCCiO- C0 OJ rr Q LU O o cc CL _l < o fc —I < z < CCh CL - z O Q o Q- CC O CO n-CO 3 < z c o 2 Q L L J O 5 ° o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o • » o o o • o o o o - o o o o o o o o o o o o o • • • o o • o o i n o o o m i n o o o i n • • o o • o o o o • o o o o o o • • o • onodon OOOCNOLOOOOOOOOOOOOOOOOOOO TH r-i tH Hriri HHHHHHHdHHdririHririOOriO rHH H NHOIflrtrllflriHHHHHlflHHHOrtririHOnHlflHHHriH •OH • O • OOO O iH O *H OOOOOOO o • o • o . . . . . O O OOO O O OOOOOOO oioioioioioioioii 01 i 9! ; 01 01 01 01 oi oi oi w oi « *g • • • • • • • • • • • • • • I I I I l l l l O O r ^ H H H H H H H H H H H H H H H H H S H H H H H a a O O > U U O i ?̂ O rH H ?̂ a? u o H to OJ OJ oj i n L n i n i n L n i n L n M L f l t A L f l t A i A i n i f l i A t n L n i A i A i A U t i A u i n m ^ o n r o r i CO O) CM 0 0 < LU I- < O LL I- cc LU o < m O o I CO cc m LL O | CO cc L U > z 3 CO CO CM DC a e 3 0 • U 00 J3 >g H o a 43 o a 43 -H •o a o ii a O * CC 'p"< 43 . M U 0 Pi n o a u 01 EH a z b> 6 H z i - i o O-rr u I0S  g 0 < 143 Q •rl •0 44 cc a to < t> 9 •H 0) a. 0 'H o -a LU a cc Us CL ca to i LU S ft -1 2 U « H 0. N U M B E R  S A M P LE S  SA  N U M B E R  S A M P LE S  m m rH rH CJSUJ in in CN 00 S Q CN CN LLJO- 5 " 116 

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
Canada 16 0
United States 13 0
China 4 1
Serbia 1 0
Ireland 1 0
France 1 0
City Views Downloads
Petawawa 7 0
Mountain View 5 0
Halifax 4 0
Beijing 4 0
Vancouver 3 0
Clarks Summit 3 0
Redmond 3 0
Rouyn-Noranda 2 0
Unknown 2 0
Pelion 1 0
Belgrade 1 0
Ashburn 1 0

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

Share

Share to:

Comment

Related Items