T H E T R E A T M E N T OF C O P P E R - G O L D O R E S B Y A M M O N I U M T H I O S U L F A T E L E A C H I N G by E L L E N M O L L E M A N M . S c , Delft University of Technology, The Netherlands, 1995 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R OF A P P L I E D S C I E N C E in T H E F A C U L T Y O F G R A D U A T E S T U D I E S Department of Metals and Materials Engineering We accept this thesis as conforming to the lequired 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 June 1998 © Ellen Molleman, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of fWflk QrvJ MinWv^K £ l g / A g p ^ The University of British Columbia Vancouver, Canada Date DE-6 (2/88) ABSTRACT The application of ammonium thiosulfate for the treatment of copper-gold ores has been investigated. Leaching studies were conducted with copper minerals, copper minerals with gold addition to solution and copper-gold samples of different copper and gold grades. The behaviour of thiosulfate, tetrathionate and sulfate in solution was studied using ion chromatography. The copper sulfide minerals chalcopyrite and enargite seem to be unreactive toward an ammonium thiosulfate leach. Covellite and chalcocite leach to a slight extent in this leaching system. The copper extractions of the sulfide minerals seem to be independent of the availability of complexing agents. The copper oxide minerals cuprite and malachite showed high copper extractions in the presence of sufficient lixiviant. Experiments showed that both gold extraction and thiosulfate stability are influenced by a combination of aeration and cupric ions in solution. High initial gold extractions were achieved in an aerated solution in the presence of cupric ions. However, these conditions simultaneously catalyzed thiosulfate degradation, resulting in gold precipitation. Therefore, it is important to establish a balance between providing sufficient air and cupric ions for fast gold dissolution, and to minimize the amount of air in the presence of cupric ions to prevent excessive thiosulfate degradation. A promising potential alternative to these conditions is a 24 hour leach without forced aeration. i i TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iii LIST OF TABLES vi LIST OF FIGURES vii LIST OF SYMBOLS xi ACKNOWLEDGMENTS xii 1 INTRODUCTION 1 2 LITERATURE REVIEW 5 2.1 Aqueous Chemistry 5 2.1.1 Gold-water system 5 2.1.2 Copper-water system 8 2.1.3 Thiosulfate 10 2.2 Overview of Thiosulfate Technology for Gold Recovery 13 2.2.1 Thiosulfate leaching for gold extraction 13 2.2.2 Gold recovery from thiosulfate solutions 19 2.3 Chemistry of the Ammonium Thiosulfate System 21 2.3.1 Thermodynamic considerations 21 2.3.2 Copper catalysis 24 2.3.3 Thiosulfate degradation reactions... 27 2.4 Ammonia technology for copper/gold leaching 33 2.4.1 Ammonia technology for gold recovery 33 2.4.2 Ammonia technology for copper recovery 35 2.5 Summary of literature 37 2.5.1 Thiosulfate technology 37 2.5.2 Ammonia Technology 40 2.5.3 Treatment of copper-gold ores by ammonium thiosulfate ; 40 3 EXPERIMENTAL METHODS 42 3.1 Materials 42 3.1.1 Copper Minerals 42 i i i 3.1.2 Copper-Gold Samples 44 3.1.3 Reagents 46 3.2 Experimental Apparatus 47 3.3 Experimental Procedures 49 3.3.1 Preliminary testwork 49 3.3.2 Leaching of copper minerals 50 3.3.3 Leaching of copper minerals with 5 ppm gold addition 51 3.3.4 Leaching of copper-gold samples 51 3.4 Experimental Design 52 3.4.1 Preliminary experiments 52 3.4.2 Leaching of copper minerals 52 3.4.3 Leaching of copper minerals with 5 ppm gold addition 53 3.4.4 Leaching of copper - gold samples 53 3.5 Sample Analysis 54 3.5.1 Copper and gold analysis 54 3.5.2 Thiosulfate, tetrathionate and sulfate analysis • 55 4 RESULTS AND DISCUSSION 57 4.1 Eh-pH diagrams 57 4.2 Preliminary experiments 64 4.2.1 Summary 69 4.3 Leaching of copper minerals 70 4.3.1 Baseline experiments 70 4.3.2 Effect of aeration 82 4.3.3 Summary 86 4.4 Leaching of copper minerals with gold addition 87 4.4.1 Baseline experiments 87 4.4.2 Effect of aeration 96 4.4.3 Effect of temperature 100 4.4.4 Effect of reagent addition 103 4.4.5 Go ld precipitation 107 4.4.6 Summary 108 iv 4.5 Leaching of copper-gold samples 110 4.5.1 Overall Lobo Composite ,111 4.5.2 Guanaco Composite 120 4.5.3 M 4 0 Pyrite Feed 127 4.5.4 M 1 0 Pyrite Concentrate 133 4.6 Sulfur balance 140 5 GENERAL SUMMARY AND CONCLUSIONS 146 6 RECOMMENDATIONS 151 7 REFERENCES 154 Appendix A: Screen analysis of copper-gold samples 160 Appendix B: Ion Chromatography 163 Appendix C: Thermodynamic data 173 Appendix D: Experimental data 174 Appendix E: Cyanidation procedures 182 Appendix F: Sulfate determination 186 v LIST OF TABLES Table 2. A Standard reduction potentials for gold in aqueous solution] 7 Table 2.B Stability constants for some gold complexes 8 Table 3 A Composition of copper mineral samples for testwork 43 Table 3.B Composition copper-gold concentrate and ore samples 44 Table 3.C P 8 0 of the copper-gold samples 46 Table 3.D Overview of experiments with copper minerals 52 Table 3.E Overview of experiments with copper minerals and gold addition 53 Table 3.F Overview of experiments with copper-gold samples 54 Table 4. A Copper extractions (%) of copper minerals at different times during baseline experiments 109 Table 4.B Results of 24 hour cyanide leach of the copper-gold samples I l l Table 4.C Gold and copper extraction (%) of Overall Lobo Composite at different times 119 Table 4.D Gold and copper extraction (%) of Guanaco Composite at different times 126 Table 4.E Gold and copper extraction (%) of M 4 0 Pyrite Feed at different times 132 Table 4.F Gold and copper extraction (%) of M 1 0 Pyrite Concentrate at different times 140 Table A . I Screen analysis of copper-gold samples 160 v i LIST OF FIGURES Figure 2.1 Eh-pH diagram for the system A u - H 2 0 at 25°C 6 Figure 2.2 Eh-pH diagram for the system C u - H 2 0 at 25°C 9 Figure 2.3 Eh-pH diagram for the system S - H 2 0 at 25°C 10 Figure 2.4 Eh-pH diagram for the metastable system S - H 2 0 at 25°C 11 Figure 2.5 Structure of the thiosulfate ion 11 Figure 2.6 Effect of copper concentration and temperature on the dissolution of gold in 0.25 M thiosulfate, 1.0 M N H 3 , 196 k P a 0 2 , stirring velocity 200 rpm 14 Figure 2.7 Gold-ammonia-thiosulfate-water system at 25°C 22 Figure 2.8 Effect of thiosulfate concentration on the rest potential of gold 22 Figure 2.9 Copper-ammonia-thiosulfate-water system at 25°C 23 Figure 2.10 Effect of copper sulfate concentration on gold leaching rate 25 Figure 2.11 Effect of the concentration ratio of ammonia to thiosulfate on gold leaching rate ....26 Figure 2.12 The model of electrochemical-catalytical mechanism of amrnoniacal thiosulfate leaching of gold 27 Figure 2.13 Oxidation state diagrams for sulfur at several p H values 28 Figure 2.14 Effect of pH/ammonia concentration on thiosulfate decomposition 32 Figure 2.15 Eh-pH diagram for the A u - N H 3 - H 2 0 system at 25°C 33 Figure 2.16 Leach extraction as a function of time on three concentrates of different mineralogy 36 Figure 3.1. Schematic diagram of the experimental set-up 48 Figure 4.1. Eh-pH diagram of the gold-thiosulfate-ammonia-water system at 25°C 59 Figure 4.2 Eh-pH diagram of the gold-thiosulfate-ammonia-water system at 25°C 60 Figure 4.3 Eh-pH diagram of the copper-thiosulfate-ammonia-water system at 25°C 62 Figure 4.4 Eh-pH diagram of the copper-thiosulfate-ammonia-water system at 25°C 63 Figure 4.5 Eh-pH diagram of the copper-ammonia-water system at 25°C 64 Figure 4.6. Covellite, 5 g/L copper, 0.1 M S 2 0 3 , 0.47 M N H 3 , p H 9.5, aeration, 35°C 67 Figure 4.7 Covellite, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 71 Figure 4.8 Chalcocite, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 72 Figure 4.9 Chalcopyrite, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 73 v i i Figure 4.10 Enargite, 1 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 74 Figure 4.11 Cuprite, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 75 Figure 4.12 Malachite, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 76 Figure 4.13 Chalcocite, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , no aeration, 35°C 83 Figure 4.14 Cuprite, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , no aeration, 35°C 84 Figure 4.15 Covellite, 5 ppm gold, 1 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 88 Figure 4.16 Chalcocite, 5 ppm gold, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 90 Figure 4.17 Chalcocite, 5 ppm gold, 1 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 92 Figure 4.18 Chalcopyrite, 5 ppm gold, 1 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 93 Figure 4.19 Enargite, 5 ppm gold, 1 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 94 Figure 4.20 Cuprite, 5 ppm gold, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 95 Figure 4.21 Malachite, 5 ppm gold, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 35°C 97 Figure 4.22 Chalcocite, 5 ppm gold, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , no aeration, 35°C 98 Figure 4.23 Cuprite, 5 ppm gold, 1 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , no aeration, 35°C 99 Figure 4.24 Chalcocite, 5 ppm gold, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 20°C 101 Figure 4.25 Cuprite, 5 ppm gold, 1 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , aeration, 20°C 102 Figure 4.26 Chalcocite, 5 ppm gold, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , 0.05 M S 0 3 2 \ aeration, 35°C 104 Figure 4.27 Chalcocite, 5 ppm gold, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , 0.05 M S0 4 2 " , aeration, 35°C 105 Figure 4.28 Chalcocite, 5 ppm gold, 5 g/L copper, 0.20 M ( N H 4 ) 2 S 2 0 3 , 0.4 M N H 3 , aeration, 35°C 106 Figure 4.29 Effect of dilution factor on detected thiosulfate concentration 108 Figure 4.30 Overall Lobo Composite, 0.20 M (NH 4 ) 2 S 2 0 3 , 1 g/L cupric addition, aeration, 35°C 113 Figure 4.31 Overall Lobo Composite, 0.20 M (NH 4 ) 2 S 2 0 3 , 1 g/L cupric addition, aeration, 35°C, reground 114 Figure 4.32 Overall Lobo Composite, 0.20 M (NH 4 ) 2 S 2 0 3 , 1 g/L cupric addition, no aeration, 35°C 115 Figure 4.33 Overall Lobo Composite, 0.20 M ( N H 4 ) 2 S 2 0 3 , no cupric addition, aeration, 35°C 116 v i i i Figure 4.34 Overall Lobo Composite, 0.20 M ( N H 4 ) 2 S 2 0 3 , no cupric addition, aeration, 35°C, duplicate 117 Figure 4.35 Guanaco Composite, 0.20 M (NH 4 ) 2 S 2 0 3 , 1 g/L cupric addition, aeration, 35°C 121 Figure 4.36 Guanaco Composite, 0.20 M (NH 4 ) 2 S 2 0 3 , 1 g/L cupric addition, no aeration, 35°C 122 Figure 4.37 Guanaco Composite, 0.20 M ( N H 4 ) 2 S 2 0 3 , no cupric addition, aeration, 35°C 123 Figure 4.38 Guanaco Composite, 0.20 M ( N H 4 ) 2 S 2 0 3 (Thiogold™ grade), 1 g/L cupric addition, aeration, 35°C 125 Figure 4.39 M 4 0 Pyrite Feed, 0.20 M (NH 4 ) 2 S 2 0 3 , 1 g/L cupric addition, aeration, 35°C 128 Figure 4.40 M 4 0 Pyrite Feed, 0.20 M (NH 4 ) 2 S 2 0 3 , 1 g/L cupric addition, no aeration, 35°C 129 Figure 4.41 M 4 0 Pyrite Feed, 0.20 M ( N H 4 ) 2 S 2 0 3 , no cupric addition, aeration, 35°C 130 Figure 4.42 M 4 0 Pyrite Feed, 0.20 M ( N H 4 ) 2 S 2 0 3 , no cupric addition, no aeration, 35°C 131 Figure 4.43 M 1 0 Pyrite Concentrate, 0.20 M (NH 4 ) 2 S 2 0 3 , 1 g/L cupric addition, aeration, 35°C 134 Figure 4.44 M 1 0 Pyrite Concentrate, 0.20 M (NH 4 ) 2 S 2 0 3 , 1 g/L cupric addition, no aeration, 35°C 135 Figure 4.45 M 1 0 Pyrite Concentrate, 0.20 M (NH 4 ) 2 S 2 0 3 , 1 g/L cupric addition, no aeration, 35°C, 24 hours 136 Figure 4.46 M 1 0 Pyrite Concentrate, 0.20 M ( N H 4 ) 2 S 2 0 3 , no cupric addition, aeration, 35°C 137 Figure 4.47 M 1 0 Pyrite Concentrate, 0.72 M ( N H 4 ) 2 S 2 0 3 , 1 g/L cupric addition, aeration, 35°C 139 Figure 4.48 Ion Chromatograms of the A S 4 A column representing the sulfate concentration at different times during the leaching of 1 g/L Chalcocite with 5 ppm gold addition 142 Figure 4.49 Ion Chromatograms of the Omnipax column showing (from left to right) thiosulfate, trithionate and tetrationate at different times during the leaching of l g / L chalcocite with 5 ppm gold addition 143 Figure A . l Particle size distribution of Overall Lobo Composite 161 Figure A . 2 Particle size distribution of Guanaco Composite 161 Figure A . 3 Particle size distribution of M 4 0 Pyrite Feed 162 Figure A .4 Particle size distribution of M 1 0 Pyrite Concentrate 162 ix Figure D . l (a) Diagram showing the separation of a mixture of components A and B by column elution chromatography, (b) The output of the signal detector at the various stages of elution shown in (a) 164 Figure D.2 Chromatogram of the A S 4 A column 165 Figure D.3 Chromatogram of the Omnipax column 166 LIST OF SYMBOLS E ° Standard potential versus a reference electrode, V F Faraday constant; 96500 C/mol AG f° Standard free energy of formation; kJ/mol A G 0 Standard free energy change; kJ/mol K s t a b Stability constant M Molarity, mol /L R Ideal gas constant; 8.314 J/mol K T Temperature; K xi ACKNOWLEDGMENTS This research project would not have been possible without the help of a great number of people. First of all , I would like to thank Dr. David Dreisinger for his guidance and support. I am very grateful to Berend Wassink, for his assistance in dealing with the ion chromatograph and for sharing his knowledge of sulfur chemistry. Special thanks to Anita Lam, for helping me with the X R D . Thanks to my fellow graduate students of the hydrometallurgy group for their assistance in various areas, and with whom I enjoyed working with. Thanks to Onno Rutten and Ben Saito for their valuable comments. A s last, I would like to thank my parents and sister for their encouragement and support. x i i 1 INTRODUCTION Cyanide leaching is the conventional method used to extract gold from ores; it is economical, biodegradable and achieves excellent recoveries from a wide range of ores [Michaelis, 1987]. However, the use of lixiviants other than cyanide for gold recovery is a subject that has received considerable attention in the past few years. There are two reasons for this increased interest: 1) Potential environmental restraints on the use of cyanide in some areas; 2) Poor leaching characteristics of some ore-bodies to cyanidation, the so-called refractory ores. Refractory ores are not amenable to cyanide treatment due to the presence of impurities (such as copper, arsenic, antimony, tellurium, and manganese), sulfide or silicate encapsulation or preg-robbing characteristics, which cause unsatisfactory gold recovery and high cyanide consumption. Refractory ore-bodies are becoming an increasingly important source of gold, because of the decreasing supply of ores which are amenable to cyanide treatment. The application of other lixiviants for gold recovery, such as thiourea, chloride, bromide, and ammonium thiosulfate has been extensively researched in the last few years. These lixiviants would be particularly attractive i f they improve the gold recovery or result in a savings in reagent cost [Michaelis, 1987]. Ideally the lixiviant is cheap or recyclable, selective, non-toxic, and compatible with down-stream recovery processes [Avraamides, 1982]. In practice, meeting all these criteria is difficult. Thiosulfate has the ability to complex gold and silver. Moreover, ammonium thiosulfate has been used for many decades as a fertilizer and consequently, from an environmental standpoint, 1 has a definite advantage over cyanide [Atluri, 1987]. Comparing reagent unit costs, ammonium thiosulfate is far cheaper than sodium cyanide (US $0.13/kg vs. U S $1.80/kg). Consequently, with similar or even slightly higher lixiviant consumption, the application of thiosulfate for gold recovery can be economical and compete directly with cyanidation. The ammonium thiosulfate system has three indispensable components: thiosulfate, ammonia and copper ions. Thiosulfate stabilizes gold in solution, while copper and ammonia accelerate the leach reaction. Research into the ammonium thiosulfate system for gold recovery has shown that the chemistry is complex and not yet fully understood. Another complication of the system is that thiosulfate is prone to degradation. Thiosulfate is a metastable sulfur species which w i l l eventually decompose in aqueous solutions. Ammonium thiosulfate has been proposed for the treatment of refractory gold ores in which carbon components are intimately associated with the gold, the carbonaceous ores. The carbonaceous matter in the ore adsorbs the solubilized gold complexes from the leach solutions back into the ore and thus reduces gold extraction. Several patents [Kerley, 1981, 1983; Perez and Galaviz, 1987; Wan et al, 1994; Marchbank et al, 1996] describe a process development that uses an ammonium thiosulfate leach solution containing copper as a catalyst for gold recovery. A soluble gold-thiosulfate complex is formed, which is not adsorbed by the carbon content of the ore. Gold can then be recovered from solution by several methods including cementation by zinc or copper and ion exchange methods. 2 Copper-gold ores form another type of refractory gold ore. Large quantities of this ore type are available for processing. Cyanide treatment of ores with a high cyanide soluble copper to gold ratio results in unfavourable economics due to high cyanide consumption resulting from the formation of copper-cyanide complexes. The objective of this work is to investigate the application of ammonium thiosulfate for the leaching of copper-gold ores. Ammonium thiosulfate is particularly attractive as an alternative to cyanide for the treatment of copper-gold ores; the presence of copper is desired in a thiosulfate leach, whereas it is detrimental in a cyanide leach. Moreover, copper leaching may occur at the same time as gold is leached, provided enough lixiviant (thiosulfate and ammonia) is available, and thus gold and copper may be recovered. A t the moment copper recovery from a cyanide leach solution has not found wide commercial application. To study the application of ammonium thiosulfate for the treatment of copper-gold ores an experimental program consisting of three different sets of experiments was performed. First, the leaching behaviour of sulfide (chalcocite, covellite, chalcopyrite and enargite) and oxide (cuprite and malachite) copper minerals in the ammonium thiosulfate system was studied. It has been reported that copper plays a catalytic role during a thiosulfate leach, however, no information is available regarding the leaching of copper minerals in an ammonium thiosulfate solution. To observe the behaviour of gold in combination with the copper minerals, gold was added to the leach solutions in a second set of experiments. The experimental program was completed by subjecting several copper-gold concentrates of different copper and gold grades to the 3 ammonium thiosulfate system. Special attention is given to the behaviour of thiosulfate during the different experimental sets, in order to gain an understanding of the leach reactions taking place. This thesis contains five chapters. The second chapter gives a review of existing literature on the ammonium thiosulfate system for gold recovery, and a brief background on ammonia technology for the recovery of copper and gold. The experimental methods are described in chapter three, followed by a discussion of the experimental results in chapter four. The summary and conclusions are presented in chapter five, followed by the recommendations in the final chapter. 4 2 LITERATURE REVIEW The chemistry of the ammoniacal thiosulfate system for gold recovery involves many interrelated chemical equilibria which are not yet fully understood. This complexity can be attributed to the presence of three essential components which define the ammoniacal thiosulfate leaching environment: ammonia, thiosulfate and copper. Many reactions may occur between these species, dissolved metal ions, atmospheric gases and ore minerals. This chapter gives a brief review of the aqueous chemistry of copper and gold, followed by a summary of the research conducted over the years in the field of ammonium thiosulfate leaching for the recovery of gold. Some background is also given regarding the recovery of gold from thiosulfate solutions. The chemical reactions involved in ammonium thiosulfate leaching and the degradation of thiosulfate are discussed next. Ammonia, itself, is an important lixiviant for copper recovery and research has been conducted on the recovery of gold with ammonia. Both topics are discussed briefly. The chapter ends with a summary of the presented literature. 2.1 Aqueous Chemistry 2.1.1 Gold-water system Gold occurs in two oxidation states, as +1 (aurous) and +3 (auric) in aqueous solutions. A s can be seen in Figure 2.1, the Eh-pH diagram of the gold-water system shows no area of stability of A u 3 + relative to water, i.e. the standard reduction potential for A u 3 7 A u is greater than 5 the upper stability limit of water (Equation (II. 1) and (II.3)) [Bard et al, 1985]. The value of the standard reduction potential for A u 7 A u is even larger (Equation (II.2)) which makes this ionic species even less stable in water (the dashed lines delineate the stability limits of water). 1.5 1.1 LU 1.5, 1 [ 1 1 c 1 1 - '— _ B _ ^ ^ —____ —— ' — — -—-R ^ ^~ ^ — — ~—• 1 1 ! 1 1 1 PLOT LRBELS Temp = 2 9 8 . 1 5 K IHu l = 1 .0E-0M 5TRBLE RREH5 q flu B H Ru 03 < 2 - > [RQ) C flu ID H 13 H20 S T A B I L I T Y L I M I T S 1 0 2 / H 2 0 2 H 2 / H 2 0 14 P H Figure 2.1 Eh-pH diagram for the system Au-H 20 at 25°C. The activity of soluble gold species is 10'4. A u 3 + + 3e" = A u A u + + e" = A u 0 2 + 4 H + + 4 e " = 2 H 2 0 E 0 = +1.52 V S H E E 0 = +1.83 V S H E E 0 = +1.23 V S H E ( H I ) (II.2) (II.3) A s predicted by the Nernst equation for A u + (Equation II.4), a decrease in the activity of free gold ions w i l l lead to a lowering of the standard reduction potential. E = 1.83 + 0.0591og[Au+ ] (II.4) 6 Complexing the dissolved gold ions with ligands such as cyanide or thiosulfate therefore results in a dramatic lowering of the reduction potentials (see Table 2.A). Table 2.A Standard reduction potentials for gold in aqueous solution, [Bard et al, 1985] Ligand E°, Au+/Au (Volts) E°, Au37Au (Volts) H 2 0 +1.83 +1.52 c r +1.154 +1.002 B f 0.960 0.854 r 0.578 0.56 N H 3 +0.563 ± 0.006 +0.325 ± 0.003 S 2 0 3 2 " ca. +0.153 ca. +0.10* C N " -0.595 ± 0.002 -0.50* *Avraamides [1982] From Table 2 .A it can be concluded that all the ligands listed form water-stable complexes with gold, i.e. the resulting standard reduction potentials are below that of water. Therefore, from a thermodynamic viewpoint there are two basic requirements for a successful leaching system for gold [Avraamides, 1982]: 1. The presence of an oxidizing agent; 2. The presence of a suitable complexing ligand. Some stability constants associated with these gold complexes are shown in Table 2.B. 7 Table 2.B Stability constants for some gold complexes [Avraamides, 1982] Au + Au 3 + Complex Complex s^tab. Au(CN) 2 " 2 x l 0 3 8 Au(CN) 4 " ca. 105 6 A u ( S 2 0 3 ) 2 3 " 5 x 10 2 8 A u l 4 " 5 x 104 7 A u i y 4 x 10 1 9 AuBr 4 " 10 3 2 A u B r 2 ' 10' 2 A u C V 10 2 6 A u C l 2 " 10 9 From this data, ligands for complexing gold may be chosen, which, when combined with a suitable oxidant, could form a successful leaching system. Then, parameters such as kinetics, selectivity, reagent stability and reagent consumption need to be determined to assess the applicability of the system. 2.1.2 Copper-water system In the absence of complexing substances, copper is a relatively noble metal. Copper can exist in oxidation states of +1 (cuprous) and +2 (cupric) in aqueous solutions. Copper forms a large number of complexes [Atluri, 1987]: • for the +1 valency complexes with CI", CN", N H 3 and S 2 0 3 2 ", which are colorless; • for the +2 valency the yellow C u C F complexes, the intense blue ammine complexes and the brown complexes with SCN". 8 1.0 0. 5 PLOT LRBELS Temp = 2 9 8 . 1 5 K ICu] = ] . 0 E - 0 4 STABLE ARER5 fl Cu B Cu2 0 C Cu 0 D Cu <2*> (RQI E H Cu 0 2 < 2 - > (RQI H20 S T A B I L I T Y L I M I T S 1 D2 /H2D 2 H 2 / H 2 0 -0. 5 h Figure 2.2 Eh-pH diagram for the system Cu-H 20 at 25°C. The activity of soluble copper species is 10"4. The Eh-pH diagram for the copper-water system is presented in Figure 2.2. The equilibrium constant for the disproportionation reaction in water 2 C u + <-> C u 2 + + Cu° (II.5) is 10 6 (at 25°C) and therefore only small amounts of C u + can exist unless it is stabilized by complexing agents [Baes and Mesmer, 1976]. The only significant hydrolysis product of C u + is C u 2 0 , which has a low solubility. Cu 2 S is very insoluble ( [Cu + ] 2 [S 2 ] = 10"49) and w i l l l ikely form in a reducing environment in the presence of sulfide [Baes and Mesmer, 1976]. The C u 2 + ion at practical concentrations begins to hydrolyze above p H 4 and precipitates as oxides or hydroxides soon thereafter, i.e. solution hydrolysis occurs to a slight extent at moderate concentrations before precipitation occurs [Atluri, 1987]. 9 Copper complexes with N H 3 are notably stable (Equation (II.6) and (II.7)) [Sillen, 1971]. C u + + 2 N H 3 = Cu(NH 3 )2 K = 1 0 1 0 4 (II.6) C u 2 + + 4 N H 3 = C u ( N H 3 ) 2 + K = 1 0 1 2 3 (II.7) Other copper(II)ammonia complexes form in addition to the ones given in Equation (II.6) and (II.7). However, these w i l l not dominate in the p H range of interest (pH 9 to 10). 2.1.3 Thiosulfate The Eh-pH diagram of the sulfur-water system demonstrates that thiosulfate has no dominant region in aqueous solution (Figure 2.3). It is only when sulfate formation is omitted in the construction of the Eh-pH diagram for the sulfur-water system, that the metastable sulfur species appear (Figure 2.4). 1. 5 1.0 0. 5 £ 0 -0 . 5 -1 .0 P H 1 1 i I I - — --~. "r: -5 - ~ ~ — :; i i E 1 1 1 F PLOT LRBELS' Temp = 2 9 8 . 1 5 K IS) = 0.M 0 H S 04 < - > (RQ) B S 014 < 2 - > I f iQI C S D H2 5 (RQ) E H S < - > (HQ) F S < 2 - > IRQ) H20 S T A B I L I T Y L I M I T S 1 0 2 / H 2 D 2 H 2 / H 2 0 2 LI 6 8 10 12 m Figure 2.3 Eh-pH diagram for the system S-H20 at 25°C. The activity of sulfur species is 0.4. 10 1.5 PLOT LABELS Temp = 2 9 8 . 15 K I S ] = B.M STABLE AREAS R S B S < 2 - > IRQ) C S 0 3 < 2 - > . I f lQl • 52 03 < 2 - > (RQ) E S2 08 < 2 - > IRQ! F OB < 2 - > IRQ) C H S < - > [RQI H H2 S [AQ1 1 H S 03 < - > IRQ) H20 S T A B I L I T Y L I M I T S 1 0 2 / H 2 0 2 H 2 / H 2 0 Figure 2.4 Eh-pH diagram for the metastable system S-H20 at 25°C. The activity of sulfur species is 0.4 M (AGf° (S 20 3 2) = -518.8 kJ/mol). Thiosulfates are compounds containing the group S 2 0 3 2 " which is a structural analog of sulfate with one oxygen atom replaced by a sulfur atom. The two sulfur atoms are not equivalent (Figure 2.5). s ~ O— S— 0" o Figure 2.5 Structure of the thiosulfate ion [Kirk Othmer, 1983] The unique chemistry of the thiosulfate ion, S 2 0 3 2 " or SS0 3 2 " , is dominated by the sulfide-like sulfur atom which is responsible for the reducing properties and complexing abilities of thiosulfates [Kirk-Othmer, 1983]. 11 Chemical properties of thiosulfate include: I. Tendency to be oxidized (by 0 2 , C u etc., see section 2.3.3) II. Tendency to hydrolyse at p H < 5.5 A . to S° and H S 0 3 2 - at mildly acid p H [Skoog and West, 1976] B . to more complex mixtures in strong acid [Smith and Hitchen, 1976] III. Relative hydrolytic stability in basic solution [Pryor, 1960] IV. Thiosulfate forms complex ions with a variety of metals, e.g. gold, silver, copper (Cu + and Cu 2 + ) , iron (Fe 3 +) etc. [Burns et al, 1981]. The stability of these complexes depend on the solution conditions. V . Formation of metal sulfides, e.g. with copper, silver, mercury [Burns et al, 1981] V I . Reductive stability toward reduction to free sulfide or sulfur; the E° of Equation (IV.8) is The best known reaction of thiosulfate is its oxidation to tetrathionate (S 40 6 2") by iodine, which is widely used in analytical chemistry. The two most important salts of thiosulfate are sodium thiosulfate, N a 2 S 2 0 3 or N a 2 S 2 0 3 . 5 H 2 0 , and ammonium thiosulfate, ( N H 4 ) 2 S 2 0 3 . The latter one was used for this research. -0.643 V, SHE S 2 0 2 " + 2 0 F T + 2e" -> HS" + S0 3 ~ + OFT (IV.8) 12 2.2 Overview of Thiosulfate Technology for Gold Recovery 2.2.1 Thiosulfate leaching for gold extraction The use of thiosulfate for the recovery of precious metals was first proposed in the 19 t h century and is known as the "Patera Process". It was applied in 1858 by V o n Patera for silver recovery. The process utilizes the solubility of silver chloride in a solution of sodium thiosulfate. The ore was treated by a chloridizing roast followed by a sodium thiosulfate leach. Sodium sulfide was used to precipitate silver sulfide [Liddell, 1945]. Berezowsky and Gormely [1978] revived the interest in the thiosulfate leaching system for precious metals recovery by developing an atmospheric ammoniacal thiosulfate leach system to recover gold and silver from ammoniacal oxidative pressure leach residues of sulfidic copper concentrates. The residues were comprised predominantly of hydrated ferric oxides and unleached sulfides. Extractions of 88-95% for gold and 83-98% for silver were achieved after 2 to 4 hours leaching with 0.4 to 0.8 M S 2 0 3 2 ' at 40 to 60°C. The parameters affecting the gold solubilization were thiosulfate, ammonia and cupric ion concentrations, temperature, and residence time. To minimize oxidation of the thiosulfate leach conditions initial testwork was performed under a nitrogen atmosphere. In some experiments, a decrease in gold extraction with time was observed. Berezowsky and Gormely [1978] attributed this phenomenon to thiosulfate degradation; thiosulfate decomposition in the presence of copper causes the precipitation of copper sulfide which coats the gold particles and thus inhibits gold leaching and recovery. Therefore, an optimal leach time exists; the leach time should not be needlessly prolonged. 13 A direct thiosulfate leaching test was performed on a chalcopyrite concentrate with 3 to 5 g/L C u 2 + addition and mi ld air sparging. A gold extraction of 97% was obtained after 1 hour. Analysis o f the leach liquors for thiosulfate indicated that 90-95% of the thiosulfate was accounted for and could be recycled [Berezowsky and Sefton, 1979]. The dissolution o f metallic gold in ammoniacal thiosulfate solution containing copper ions was investigated by Tozawa et al [1981]. The experiments were conducted in an autoclave under various conditions. The following results were reported (Figure 2.6): 1. In the absence of C u ( N H 3 ) 4 2 + the kinetics of dissolution o f gold are very slow. 2. A t leaching temperatures between 65 to 100°C the dissolution of gold is inhibited by the formation of a copper sulfide coating. 3. Vigorous stirring and high temperature (>140°C) decrease the gold dissolution due to excessive oxidation of thiosulfate ions. ' • < • ' ' ' ' i i ' ' ' ' i i i_ 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 T e m p e r a t u r e ( * C ) Figure 2.6 Effect of copper concentration and temperature on the dissolution of gold in 0.25 M thiosulfate, 1.0 M NH3,196 kPa0 2, stirring velocity 200 rpm, [Tozawa et al, 1981] 14 The same authors also investigated the effect of oxygen partial pressure on the dissolution of gold in thiosulfate medium. Maximum gold dissolution occurred at 98 kPa oxygen partial pressure. A t higher oxygen pressures the dissolution of gold decreased because of thiosulfate oxidation. Kerley [1981,1983] patented a process for the recovery of precious metals from refractory ores, particularly those containing manganese and/or copper, by lixiviation using an ammonium thiosulfate leach solution. The patent claims to improve upon the thiosulfate leaching of the patent of Berezowsky and Gormely [1978] by providing better control of the stability of the thiosulfate ion (see section 2.3.3). The leach solution contains 1.2 to 1.35 M S 2 0 3 2 ", 1 to 4 g/L C u 2 + , sufficient ammonia to maintain a p H of 7.5 or higher, and a minimum of 0.05% sulfite ions to control the stability of the solutions during leaching. Kerley [1981] claims that sulfite ions inhibit the decomposition of thiosulfate according to Equation (II.9) and thus prevent precipitation of metal sulfides. 4 S 0 2 ' + 2S 2 " + 6 H + <-> 3S202f + 3 H 2 0 (II.9) This reaction is however unlikely to occur at p H 7.5, since the sulfide ion is not stable in this p H region (see Figure 2.4 and section 2.3.3). The precious metals can be recovered by conventional methods, such as cementation or electrolysis. The process patented by Kerley was carried out in a plant in Mexico, but the scale-up from laboratory to plant-size failed. Perez and Galaviz [1987] describe the modifications required to 15 make plant operation feasible. The most important process adjustment was the p H value which should be maintained at a minimum level of p H 9.5, instead of 7.5 as suggested by Kerley [1981]. This higher p H inhibits the action of substantial amounts of metallic iron and ferric salts that are present in the lixiviating solution as a result of grinding the ore in a ball m i l l prior to lixiviating. According to Perez and Galaviz [1987] the ferric ion accelerates the oxidation of thiosulfate to tetrathionate (Equation 11.10) which has no lixiviating action on gold or silver. 2S20l~ +2Fe 3 + 2Fe 2 + + S 4 0 2 " (11.10) The ferrous ion displaces silver and gold from solution according to: FeO + A g 2 S 2 0 3 -> A g 2 S + F e S 0 4 (11.11) A t high p H (9.5 and higher) ammonia hydrolysis produces hydroxide ions which react with the ferrous ions according to Equation (11.12) and (11.13), thus preventing iron from displacing silver and gold from solution. F e 2 + + 2 N H 3 + 2 H 2 0 = Fe(OH) 2 + 2 N H ; (11.12) 4Fe(OH) 2 + 0 2 + 2 H 2 0 -> 4Fe(OH) 3(s) (11.13) Zipperian and Raghavan [1988] identified the parameters of importance in the dissolution of gold and silver values from a rhyolite ore with a high manganese content using ammoniacal thiosulfate solutions containing copper. The effect of thiosulfate, ammonia concentration, temperature, and copper sulfate addition was researched. Optimum conditions were established at 2 M S 2 0 3 2 \ 4.1 M N H 3 , 6 g/L C u 2 + , 50°C, and 2 hours leaching in the absence of oxygen. 16 In the absence of cupric ions only 14% gold was solubilized. Initial rates of gold extraction were enhanced by increasing cupric ion concentration, but the ultimate extraction was not influenced by the cupric ion concentration in the range investigated (up to 6 g/L Cu). Further, it was concluded that maintaining optimal p H and E h conditions (pH 10 and 200 mV) are necessary to prevent precipitation of copper as Cu 2 S. Ha l f of the thiosulfate in the lixiviant solution at p H 9.5-10 was reported to be consumed during the dissolution process. The consumption of the reagents and the recycling of the lixiviant over a long period of time was researched by Gong and H u [1990]. It was reported that the consumption of thiosulfate depends on the amount of air used, the temperature, agitation and configuration of the reactor. I f these variables are optimized the lixiviant solution can be recycled and the losses are minimal. However, details regarding agitation and reactor configuration are not described. Over 95% gold extraction was achieved under the conditions of 0.8 to 1 M S 2 0 3 2 ", 1.8 to 2.2 M N H 3 , 0.1 M N a 2 S 0 3 , 1 g/L C u 2 + , 40°C, 1.5 hours and an oxygen atmosphere. The p H was adjusted to 10. Hemmati et al [1989] studied the application of ammoniacal thiosulfate leaching for gold recovery on various types of ores. It was found that the highest gold extractions were achieved at 0.7 M S 2 0 3 2 " , 3 M N H 3 , p H of 10.5, 35°C (over the range of 25 to 85°C) and an oxygen partial pressure of 103 kPa. Gold extraction of 73% was accomplished under these conditions. N o impact was found on either gold extraction or thiosulfate consumption by variation of the oxygen pressure in the range of 0 to 206 kPa. Hemmati et al [1989] reported that the efficiency of thiosulfate leaching depends upon ore type. For treating carbonaceous ores, thiosulfate was found to be chemically superior and economically advantageous over cyanide. 17 Langhans et al [1992] focused on maximizing gold extraction while minimizing thiosulfate consumption at low reagent concentrations and ambient temperatures and a p H between 9 and 11. The range of the thiosulfate concentration tested was 0.05 to 0.2 M . The research was conducted on low-grade oxidized gold ores with application to heap, dump, or in-situ leaching techniques and found to be competitive with conventional cyanidation. After leaching for 48 hours, 83% gold extraction was achieved with 0.2 M S 2 0 3 2 \ 0.09 M N F L O H , 0.00625 M N a 2 S 0 3 , 63.5 ppm C u 2 + with 0.4 kg S 2 0 3 2 " consumed per tonne of ore. Langhans et al [1992] concluded that these results compare favorably with 86% gold extraction after 24 hours with 0.21 kg C N " consumed per tonne of ore using standard cyanidation methods. The effectiveness o f low thiosulfate concentrations was confirmed by Cao et al [1992], who studied the effects of the concentrations of thiosulfate, copper and ammonia on the extraction of gold and silver from a sulfide concentrate. Leaching was performed with a 0.2 M thiosulfate solution at 2 hours retention time which yielded 95% gold extraction. The sparge air was treated with ammonia in order to keep the ammonia concentration constant. Cao et al [1992] also researched the influence of sulfate addition to the system, in order to reduce the thiosulfate consumption, but found no effect (see section 2.3.3). Wan et al [1994] patented a process using a thiosulfate lixiviant for the treatment of carbonaceous gold ores on a heap leach without applied pre-treatment. The p H of the solution should be in the range of 9 to 10 with a thiosulfate concentration of 0.1-0.2 M . 18 Newmont Gold Company evaluated the application of ammonium thiosulfate on a demonstration heap leach of 327,000 metric tonnes of low-grade carbonaceous sulfidic ore that was pretreated by bio-oxidation. The average gold recovery for thiosulfate heap leach at a particle size of minus 1.9 cm was found to be approximately 55 per cent. Ammonium thiosulfate consumption was about 5 kg/tonne for low sulfide carbonaceous ores (without bio-oxidation) and 12-15 kg/tonne for bio-oxidised ores. Typical leach solutions used contained 0.1 M S 2 0 3 2 \ 0.1 M N H 3 and 30 ppm C u 2 + [Wan, 1997]. Construction started for a commercial size heap leach operation, but was halted due to the current low gold price. The application of a thiosulfate salt lixiviant to recover gold from an oxidative pressure leach slurry is described in a patent granted to Marchbank et al [1996]. A n ore slurry of refractory sulfidic and refractory carbonaceous ore is subjected to pressure oxidation in an autoclave under neutral or alkaline conditions followed by leaching with a thiosulfate salt in stirred tank reactors. Typical leaching conditions are 0.025 to 0.1 M ( N H 4 ) 2 S 2 0 3 , 50 to 100 ppm C u 2 + , 40 to 55°C and a minimum sulfite concentration of 0.001 M , while maintaining a p H between 7 and 8.7. 2.2.2 Gold recovery from thiosulfate solutions Process options for the recovery of gold from thiosulfate solutions are: • Cementation • Carbon adsorption • Direct electrowinning • Reductive precipitation • Ion exchange • Solvent extraction 19 Most researchers report the application of cementation for the recovery of gold from thiosulfate solutions [Berezowsky and Sefton, 1979; Kerley, 1983; Perez and Galaviz, 1987; Wan et al, 1994; Marchbank et al, 1996]. Suggested precipitants are zinc, copper, aluminum, iron or soluble sulfides. The use of copper as precipitant is preferred since it enables the precipitation of gold without also causing precipitation of copper from the lixiviant solution. Gallagher [1987] found that activated carbon has a very low affinity for the gold(I)thiosulfate complex. M c K e e and Lulham [1991] reported that gold was effectively recovered by the addition of a near stoichiometric amount of cyanide to the leach solution, enabling the recovery of the gold cyanide complex by activated carbon or an anion exchange resin. The direct electrowinning of a gold thiosulfate leach solution onto a steel cathode at 40°C was investigated by Abbruzzese et al [1995]. The kinetics of electrowinning were reported to be fast and the precious metal recovery was quantitative (99%). N o qualitative information was given. The current efficiency was 4% due to the concurrent parasitic reactions at the electrodes (oxygen evolution at the anode and reduction of water and dissolved oxygen at the cathode). Awadalla and Ritcey [1991] reported the use of sodium borohydride for the reduction of gold and silver in acidic solutions of thiosulfate. Complications of this process option are: low p H (around 6) and Cu 2 + ions decrease the efficiency of borohydride. At lur i [1987] studied the use of three anion exchange resins (Amberlite IRA-400, Amberlite IRA-68 and Amberlite IRA-94) for the selective recovery of gold and silver from simulated 20 thiosulfate leach liquors containing copper, gold and silver. A l l the three resins investigated were not selective to silver and gold over copper. Another process option mentioned is solvent extraction [Marchbank et al, 1996]. However, no literature was found which covered the selective extraction of gold from a thiosulfate solution. This brief review of recovery methods indicates that the recovery of gold from thiosulfate solutions is not easily accomplished, and offers opportunities for more research. 2.3 Chemistry of the Ammonium Thiosulfate System This section reviews some fundamental aspects of the ammonium thiosulfate system: thermodynamics, copper catalysis and the degradation pathways of thiosulfate. 2.3.1 Thermodynamic considerations Thermodynamically, the ammonium thiosulfate system is not stable. Both ammonia and thiosulfate w i l l be lost through escape or decomposition. Therefore, true equilibrium diagrams cannot be constructed for this system. B y omitting the more stable species such as sulfate ions from the system, the features of the metastable species can be examined (see Figure 2.4). The species considered are: N H 4 + , N H 3 , S 2 0 3 2 ", S°, S x 0 6 2 " (x>2), S 0 3 2 \ H S 0 3 2 \ The stability of the thiosulfate ion w i l l be discussed in section 2.3.3. The Eh-pH diagram for the gold-ammonia-thiosulfate-water system at 25°C is presented in Figure 2.7. 21 -1.0 h 0 2 4 6 8 10 12 14 PH Figure 2.7 Gold-ammonia-thiosulfate-water system at 25°C, Activity of species: 0.1 M thiosulfate, 0.1 M ammonia, 5x10^ M gold [Li et al, 1995] According to Figure 2.7, under alkaline conditions (pH>9), Au(NH 3 ) 2 + is expected to be the most stable gold species. However, rest potential measurements performed by L i et al [1996] suggest that the predominant gold species is Au(S 20 3) 2 3" since the gold rest potentials varied with thiosulfate concentrations and not with ammonia concentrations (in the range of 0 to 0.5 M NH 3 ) (Figure 2.8). -0.10 .0.30 1 1 1 1 ' L 1 1 -1.6 -1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 log[S 20 3 2] (M) Figure 2.8 Effect of thiosulfate concentration on the rest potential of gold [Li et al, 1996] 22 The half-cell reaction for gold thiosulfate is: A u ( S 2 0 3 ) 3 _ + e" -> A u + 2 S 2 0 2 _ E°= 0.153 V S H E (11.14) It can be noted from Figure 2.7 that gold remains in the metallic state i f potentials are too low throughout the pH range of 6-12. Thus thermodynamics predicts that potentials exceeding 0.05 V S H E are required for gold leaching. Figure 2.9 shows that copper(I)thiosulfate is the predominant species in solution at most common potentials and appears to be in equilibrium with copper(II)ammine (see section 2.3.2). Thus the excess copper(I)thiosulfate ions will be converted back to copper(II)ammine ions to maintain the equilibrium as dictated by potential [Li et al, 1996]. At low potentials and high pH, the decomposition of thiosulfate may lead to the precipitation of copper sulfides. 1.8 1.0 as | ao •0.5 -1.0 -1.5 O 2 4 6 8 10 12 14 PH Figure 2.9 Copper-ammonia-thiosulfate-water system at 25°C, 0.1 M thiosulfate, 0.1 M ammonia, SxlO"4 M copper [Li, 1995] 23 2.3.2 Copper catalysis The rate of gold leaching in an ammoniacal thiosulfate solution is enhanced greatly by a catalytic copper reaction [Berezowsky and Sefton, 1979, Zipperian et al, 1988, Tozawa et al, 1981]. However, conflicting opinions exist on the exact role of the copper in the leach chemistry. Agreement does exist concerning the dissolution reaction of gold in thiosulfate solutions in the presence of oxygen (Equation 11.15). 2 A u + 4 S 2 0 3 _ + y 0 2 + H 2 O o > 2 A u ( S 2 0 3 ) ' - + 2 0 H " (11.15) The above reaction can be modified to incorporate the influence of the catalytic action of the cupric ion during gold dissolution [Li et al, 1995; Marchbank et al, 1996]]: A u + 5S2Q2j + C u ( N H 3 ) 2 + -> A u ( S 2 Q 3 ) 3 ~ + C u ( S 2 Q 3 ) 3 " + 4 N H 3 (11.16) _ . Copper oxidation by oxygen or another appropriate oxidant Copper minerals in some ores can act as a source of cupric ion or copper sulfate may be added to the leach solutions. The appropriate copper half cell reaction is: C u ( N H 3 ) 2 + + 3 S 2 0 2 " +e" -> Cu(S 2 0 3 )*~ + 4 N H 3 E°=0.225 Vol t (11.17) L i et al [1996] states that oxygen in the absence of a catalyst may be an oxidant for gold leaching, but with much slower kinetics. It was found that in the absence of copper, the gold leaching rate in ammonium thiosulfate becomes negligible (Figure 2.10). 24 0.00 0.01 0.02 0.03 0.04 0.05 [ C u s c g (M) Figure 2.10 Effect of copper sulfate concentration on gold leaching rate [Li et al, 1996] At low copper levels, an increase in the copper concentration results in a dramatic rise in the gold dissolution rate. However, too high copper concentrations significantly inhibit gold leaching. L i et al [1996] explained this by a deficiency of lixiviant for complexing gold, because the copper combines with most of the ammonia and thiosulfate in solution. Furthermore, thiosulfate degradation is accelerated by high copper concentrations (see section 2.3.3). Either way, loss of reactants results in slower leaching kinetics. It is evident from Equation (11.17) that to enable the regeneration of the cupric ion, it is necessary to keep the concentration ratio of ammonia to thiosulfate in a certain range. Increasing the concentration of only one of the ligands will have a limited positive effect on the gold leaching process. However, this may have a negative effect, as an excess of either ammonia or thiosulfate may make the reaction depicted in Equation (11.17) less reversible [Li et al, 1996]. This effect is illustrated in Figure 2.11. •25 T | 1 1 1 I I I I I | 1 1 1 I I I l l | 1 1 1 I I I I I L J i i i i i 1111 i 1 . > . . • . . 4 0.1 1 10 100 [NH^H]/[(NHJ2S203] Figure 2.11 Effect of the concentration ratio of ammonia to thiosulfate on gold leaching rate [Li, 1996] Jiang et al [1993] proposed a model for gold dissolution in ammoniacal thiosulfate solutions on the basis of electrochemical investigations (Figure 2.12). According to this study, ammonia preferentially complexes gold, and the Au(NH 3 ) + , which forms at the anodic surface then reacts with S 20 3 2" and is converted to the more stable Au(S 20 3) 2 3". The Cu(NH 3 ) 4 2 + gains an electron at the cathodic surface and is reduced to Cu(NH 3 ) 2 + which is then oxidized by oxygen into Cu(NH 3 ) 4 2 + after entering the bulk solution. In this model, a prominent role is ascribed to oxygen in the system, and the effect of the ratio of ammonia to thiosulfate does not seem to play a role on the regeneration of the cupric ion. This is in contradiction to the observations of L i et al [1996]. The two models display the complexity of the ammonia thiosulfate leaching system and the need for more fundamental research. 0.10 •a "a 0.08 h g; 0.06 2 S> 0.04 • •-4 Xi o cts o o o I I I I I 11 0.02 0.00* 0.01 26 Gold Surface Solution A n o d i c A r e a Au = Au*Ve A u V 2 N H 3 = A u ( N H 3 > 2 C u ( N H 3 ) 4 * + e = C u ( N H 3 ) 2 Ca thod ic A r e a Au e • C u ( N H 3 ) | t 0 2 - * O H ~ N H 3 A u ( N H 3 ) | + 2* C u ( N H 3 ) 2 Figure 2.12 The model of electrochemical-catalytical mechanism of ammoniacal thiosulfate leaching of gold [Jiang et al, 1993] 2.3.3 Thiosulfate degradation reactions The variables affecting the stability of thiosulfate solutions are [Skoog and West, 1976] • pH; • The presence of microorganisms and redox catalysts; • The concentration of the solution; • The presence of oxygen; • Exposure to sunlight. Experiments indicate that in the absence of strong oxidants or catalysts, the stability of thiosulfate solutions is at a maximum in the pH range between 9 and 10 [Skoog and West, 1976]. Bacterial activity appears to be at a minimum at this pH, which explains the maximum stability in this pH region since bacterial activity is reported to be the most prominent cause of instability. 27 Thiosulfate decomposition is catalyzed by copper(II), and iron(III) ions as well as by the reaction products of the decomposition. Skoog and West [1976] report that the decomposition rate is greater in more dilute solutions. In the case of oxidative decomposition of thiosulfate, sulfur may exist in many different oxy-sulfur species in the +2 or higher oxidation states. These include S 50 6 2" (+2), S 20 3 2" (+2), S 40 6 2" (+5/2), S 2 0 4 2 - (+3), S 3 0 6 2 - (+10/3 ), S0 3 2" (+4), S 20 6 2" (+5), S0 4 2" (+6), S 20 8 2" (+7), S0 5 2 " (+8). Thermodynamically, sulfate is the most stable sulfur species under the preferred leaching conditions (alkaline pH, oxidative potential). This is visible in Figure 2.13, the oxidation state diagram for sulfur at several pH values. By drawing a line between HS" and S0 4 2 " at pH 10, it is visible that most sulfur species are unstable (all species lie above the line). The nearest species to the line are S° and S 20 3 2" which indicates that relative stability for these species in the pH 10 range. MEAN OXIDATION STATE Of SULPHUR Figure 2.13 Oxidation state diagrams for sulfur at several pH values [Peters]. 28 Smith and Hitchen [1976] described the oxidation of thiosulfate to tetrathionate in the presence of cupric ion: 2S202f + 2 C u 2 + -> S 4 0 2 " + 2 C u + (fast) (11.18) 2 C u + + \ 0 2 + 2 H + -> 2 C u 2 + + H 2 0 (measurable) (II. 19) 2 S 2 0 2 - + 2 F T + { 0 2 - > S 4 0 2 ~ + H 2 0 (11.20) or, in basic solution: 2 S 2 0 2 " + H 2 0 + y 0 2 ^ S 4 0 2 _ + O H ' (11.21) Tetrathionate may then decompose according to a variety of pathways, leading to the following stoichiometry for the overall oxidation of thiosulfate via the tetrathionate pathway: S 2 0 2 " + 2 0 F T + 2 0 2 -> 2 S 0 2 " + H 2 0 (11.22) Byerley et al [1973] cited the following reactions to describe the disproptionation of S 4 0 6 2 " to higher and lower polythionates: S 4 0 2 - + S 2 0 2 " -> S 5 0 2 " + S O 2 " (11.23) S O 2 - + S 4 0 2 - -> S 3 0 2 " + S 20 2"" 11.24) S 5 0 2 " + 3 0 H " ^ | S 2 0 2 " + | H 2 0 (11.25) 2 S 4 0 2 " + 3 0 H " -> f S 2 0 2 " + S 3 0 2 " + f H 2 0 (11.26) These reactions w i l l take place at p H > 7 [Smith and Hitchen, 1976]. The trithionate (S 30 6 2") species only decomposes under harsh conditions (pH>13 and boiling) to form thiosulfate and 29 sulfite. Marsden and House [1992] used Equation (11.26) to explain why S 2 0 3 2 " is relatively stable under basic conditions. Pryor [1960] showed that S 2 0 3 2 " is relatively stable in basic solution, i.e. in the absence of catalysts. The disproportionation of thiosulfate in water or aqueous buffers at 250-280°C follows Equation (11.27) and is irreversible. This base hydrolysis was found to be very slow in basic solution, even at 250°C, and sulfite was found not to be an intermediate in the reaction. According to several authors the addition of sulfite ions (S0 3 2") to an ammoniacal thiosulfate solution has a beneficial effect on the stability of thiosulfate [Kerley, 1981; Perez and Galaviz, 1987; Zipperian et al, 1988; Gong and Hu, 1990]. The sulfite ions react with sulfide as to regenerate thiosulfate and consequently prevent the precipitation of precious metal values with the sulfide ions. Different reactions are proposed to explain this effect: • Zipperian et al [1988]: S 2 0 2 " + O H " - » S O 2 - + HS~ (11.27) 3SO3" + 2S2~ + 3 H 2 0 2S202~ + 6 0 H 0 (11.28) • Gong and H u [1990]: 4 S 0 2 " + 2S 2 " + 3 H 2 0 o 3 S 2 0 2 " + 6 0 H " (11.29) Kerley [1981, 1983], Perez and Galaviz [1987]: 4 S 0 2 " + 2S 2 " + 6 H + <-> 3 S 2 0 2 " + 3 H 2 0 (11.30) 30 It is questionable that the reactions described in Equation (11.28) to (11.30) take place under typical thiosulfate leaching conditions (around p H 9-10, E h 150 to 250 raVSHE). Little free sulfide ions and hydrogen sulfide ions wi l l exist at these p H and E h values (Figure 2.4). Equation (11.30) describes the formation of thiosulfate under acidic conditions. This is unlikely since thiosulfate tends to hydrolyze under acidic conditions [Skoog and West, 1976]. There are a few reactions in which sulfite can participate. A n advantage of sulfite addition can be the oxidation of sulfite to sulfate (Equation 11.31). When this reaction is kinetically favored over the oxidation of thiosulfate to sulfate, sulfite can improve the stability of thiosulfate in solution. S O 2 " + j 0 2 -> S O 2 - (11.31) A further beneficial effect of sulfite addition is described by Wasserlauf and Dutrizac [1982]. Sulfite ions accelerate the decomposition of S x 0 6 2 ", x > 4, in alkaline to mildly acidic solutions. The products are trithionate, thiosulfate and protons. The reaction with tetrathionate is given in Equation (11.24). Also , elemental sulfur may react with sulfite under alkaline conditions, to form thiosulfate (Equation 11.32) [Wasserlauf and Dutrizac, 1982]: S 0 2 - + i S x - > S 2 0 2 " (11.32) Several authors described the substitution of sulfate for sulfite, in order to stabilize thiosulfate [Gong and Hu, 1994; Cao et al, 1992]. Since sulfate is quite stable toward reduction [Burns et al, 1980], no effect of sulfate with respect to thiosulfate degradation would be anticipated. 31 L i et al [1996] reported the effect of ammonia concentration on thiosulfate degradation. Increasing ammonia concentration increased the rate of thiosulfate decomposition significantly (Figure 2.14). Since this effect is more evident after several days, it is unlikely to play a significant role in stirred tank leaching, but has to be accounted for in heap leach operations. No explanation was given for this phenomena. 100 g, 80 60 8 3 <-< o c .2 40 a, e 9 ' a a 20 Y pH 7.3/0.1 M ATS/0.0 M NH4OH A pH 8.1/0.1 M ATS/0.02 M NH4OH • pH 9.2/0.1 M ATS/0.1 M NH4OH O pH 9.7/0.1 M ATS/0.2 M NH4Otf^ • pH 10.1/0.1 M ATS/0.5 M>ift4OH • pH 11.0/0.1 M NajSjO^il M NH4OH 6 9 Time (day) 12 15 Figure 2.14 Effect of pH/ammonia concentration on thiosulfate decomposition [Li et al, 1996]. It can be concluded from the above overview that the exact degradation pathways of thiosulfate are very complex. Many reactions are possible between existing and generated species in solution. The reactions quoted by the literature do not reflect this complexity, and even tend to simplify the reactions taking place. A good example of this simplification is the explanation repeatedly cited in the literature regarding the beneficial effect of sulfite addition. 32 2.4 Ammonia technology for copper/gold leaching Leaching of copper using ammonia/ammonium salts has been well established. Up to now ammonia has found no application for the extraction of gold and silver from ores due to the very slow dissolution rate of gold and silver in the absence of proper oxidants, and high temperatures and pressures. A brief overview of ammonia technology for gold and copper is given, followed by a discussion of relevant chemistry. 2.4.1 Ammonia technology for gold recovery Gold dissolves in ammonia as the gold(I)ammonia complex, [Au(NH 3) 2] +, within the range of water stability over a range of pH values (Figure 2.15). The gold(III)ammonia complex, [Au(NH 3) 4] 3 + is thermodynamically unstable in the region of water stability. 4.0 3.0 2.0 > 1.0 0 -2 0 2 4 6 8 10 12 14 16 pH Figure 2.15 Eh-pH diagram for the Au-NH 3-H zO system at 25°C. Activity of ions is 104; NH3=1.0 M [Meng and Han, 1993] 33 Au02 The dissolution of gold is thermodynamically feasible at room temperature. However, kinetic experiments showed that gold was essentially not leached. A practically acceptable dissolution rate was not observed unless the leaching temperature was greater than 120°C in the presence of an oxidant [Meng and Han, 1993]. Various oxidants such as oxygen, cupric, cobaltic (Co 3 + ) and manganic (Mn 4 + ) , alone or in combination, can be used in a solution containing ammonia and ammonium salts (e.g. ammonium chloride, ammonium carbonate or ammonium sulfate) for the leaching of gold [Han and Meng, 1992]. Oxygen in itself insufficient as an oxidant. Han and Meng [1992] reported the most desirable combination of oxidants to be: oxygen, hypochlorite and cupric ion at an appropriate ratio (the ratio was not further specified). Meng and Han [1993] found oxygen (100 kPa) and cupric ion (10 g/L Cu 2 + ) as the preferred combination. Other important variables apart from time and particle size, are temperature and ammonia concentration. Meng and Han [1993] reported that there appears to be a temperature-controlled solubility limit for the gold(I)ammonia complex: at 180°C this value is about 80 mg/L A u . Extractions of 80-96% A u from three gold ores and one concentrate in 3-4 hours were obtained by using an autoclave. Conditions were 190°C, C u 2 + 5 to 10 g/1, free N H 3 5.5 M , ( N H 4 ) 2 S 0 4 0.5 M , oxygen pressure 600-1000 kPa, particle size minus 0.147 mm, and a pulp density of 10 to 25% solids. 34 2.4.2 Ammonia technology for copper recovery Ammonia leaching of copper and nickel has been well established. Sherritt Gordon developed an ammonia pressure leach process for extracting copper, nickel, cobalt and sulfur from high grade nickel concentrates [Forward and Mackiw, 1955]. Anaconda designed the Arbiter process for the treatment of copper concentrates with a low-pressure ammonia-ammonium sulfate leach [Kuhn et al, 1974]. A suitable mixing technology was applied to overcome the necessity of high oxygen partial pressure rather than using high pressure reactors. The process requires the use of oxygen rather than air. The leach takes place at temperatures ranging from 60 to 90°C. The simplified leaching chemistries of chalcocite, covellite and chalcopyrite are shown in Equations (II. 33) to (11.35) [Kuhn et al, 1974]: j C u 2 S + f 0 2 + 4 N H 3 + { H 2 0 <-> C u ( N H 3 ) 2 + + JS0 2 . " + O H " (11.33) CuS + 2 0 2 + 4 N H 3 + H 2 0 <-> C u ( N H 3 ) f + S O 2 " + H 2 0 (11.34) CuFeS 2 + j 0 2 + 4 N H 3 + H 2 0 <-> C u ( N H 3 ) 2 + + i F e 2 0 3 + 2 S 0 2 - + 2FT (11.35) Chalcocite is probably the most amenable copper sulfide mineral to ammonia leaching. It has been demonstrated that half of the copper in chalcocite is essentially free and dissolves readily, leaving a covellite mineral matrix. A s long as there is sufficient ammonia available to complex the copper and sufficient oxygen provided, it is expected that the leaching of chalcocite and covellite reaches completion [Kuhn et al, 1974]. 35 When leaching chalcopyrite, the iron present in the mineral forms a hematite layer on the copper particle, thus becoming the rate-controlling factor in the leaching of chalcopyrite [Williams and Light, 1978]. A general formulation of enargite leaching is given in Equation (11.36) [Kuhn et al, 1974]: Cu 3 AsS 4 + 13NH 3 + 8 l 0 2 + f H 2 0 <-> 3Cu(NH 3 ) 2 + + N H 4 H 2 A s 0 4 + 4 S 0 2 - + 2 H + (11.36) Figure 2.16 shows the copper extraction as a function of leaching time for three different minerals. Kuhn et al [1974] did not explain the slow leaching characteristics of the enargite mineral, other than that the rate of enargite leaching appears to be dependent on the relative specific surface area of the mineral. Figure 2.16 Leach extraction as a function of time on three concentrates of different mineralogy. OWeed chalcocite; A Twin Buttes chalcopyrite; • Butte enargite [Kuhn et al, 1974] In summary, according to Kuhn et al [1974] the leaching of copper sulfide minerals in the arnmoma-aminonium sulfate environment depends mainly on oxygen mass transfer through the bulk solution to the surface and diffusion through an iron oxide layer, i f present. Several studies Percent 10 0 , _ 130*1 I hour 2 15' 49' 3 4 L E A C H TIME (Hour*) 5 6 36 support this theory [Halpern, 1953, Fisher and Halpern, 1956]. According to Kuhn et al [1974], the solution to this mass-transfer problem lies in the mixing system. B y increasing the stirrer speed, the chemical reaction rate is made to be the rate-controlling mechanism, instead of oxygen mass transfer, and ammonia leaching can be accomplished in a low-pressure system. However, Beckstead and Mi l l e r [1978] stated that a change in mixing system changes the morphology of the hematite deposit which causes a change in surface reaction kinetics, rather than improving the oxygen mass transfer. This was noted before by Forward and M a c k i w [1955] who found that intense agitation tends to reduce the iron oxide layer thickness. For copper oxide minerals, the following relations can be written [Fisher and Halpern, 1956]: CuO + 2 N H 3 + 2 N H ; -> C u ( N H 3 ) 2 + + H 2 0 (11.37) C u 2 0 + 2 N H 3 + 2 N H 4 + ->• 2Cu(NH 3 )£ + H 2 0 (11.38) In the presence of N H 4 + the dissolution of the copper oxide is favored. 2.5 Summary of literature 2.5.1 Thiosulfate technology From the literature review it was realized that the thiosulfate concentrations applied for the recovery of gold have decreased over the years of study. The patents granted in the early 1980's applied thiosulfate concentrations ranging from 0.4 to 2 M thiosulfate, while recent studies mention the addition of 0.1 to 0.2 M thiosulfate. The application of lower thiosulfate 37 concentrations while still achieving high gold extractions render the ammonium thiosulfate process more economical and should therefore be pursued in current and future investigations. The p H of the recently developed leaching processes is generally maintained between 9 to 10. This p H range is dictated by the ammonia/ammonium buffer point (9.25 at 25°C), since the presence of ammonia has to be ensured in order to solubilize copper as the copper(II)ammonia complex. Testing at higher temperature and lower p H values are possible because the buffer point shifts to lower values with increasing temperatures. This is probably the reason why Marchbank et al [1996] were successful in leaching at pH's of 7 to 8.7 at 50°C. The thiosulfate stability w i l l however decrease at higher temperatures and lower p H values. Therefore, a p H of 9 to 10 is generally preferred at ambient temperature because thiosulfate appears to be less prone to degradation in this region and the copper(II)ammonia complex is stable. When evaluating the optimum ammonia concentration two factors should be considered. Firstly, there has to be sufficient ammonia present in the leaching system to solubilize the copper partially as an ammonia complex. Secondly, the ammonia to thiosulfate ratio should be kept in a certain range, preferably around 1 to 2. This effect is illustrated in Figure 2.9. A too high ammonia concentration might stabilize copper as the copper(II)ammonia complex, or a too high thiosulfate concentration might result in the stabilization of copper as the copper(I)thiosulfate complex according to Equation (11.39), thus limiting the catalytic action on gold extraction: C u ( N H 3 ) 2 + + 3 S 2 0 2 " + e" -> Cu(S 2 0 3 ) 5 3 " + 4 N H 3 (11.39) Another significant difference between early studies and recent ones is the difference in cupric 38 ion addition. In the patent of Kerley [1981, 1983] a cupric addition of 1 to 4 g/L is described and Berezowsky and Gormely [1978] mention an addition of 1-10 g/L. Cupric ion concentrations of 50-100 ppm are mentioned in the patents granted more recently [Wan et al, 1994; Marchbank et al, 1996]. L o w cupric ion concentrations seem to be favourable, since high cupric ion concentrations accelerate thiosulfate degradation. The addition of sulfite ions to an ammonium thiosulfate solution in order to stabilize the thiosulfate ion is widely practiced. A s discussed in section 2.3.3., the reactions referred to in the literature seem unlikely to take place. Sulfite might, however, participate in different reactions, such as the reaction with tetrathionate to form trithionate and thiosulfate, and thus influence the leaching reactions (see section 2.3.3). The main factors effecting the thiosulfate stability in an ammoniacal thiosulfate leach solution are: cupric ions, oxygen pressure and temperature. Some authors reported a negative effect on thiosulfate consumption with high oxygen pressures, others however, did not observe any negative effect when applying higher oxygen pressures. A t high temperatures (>60°C) excessive thiosulfate consumption and the formation of a copper sulfide coating on gold particles have been observed. A t slightly higher than ambient temperatures, the kinetics of leaching are favoured, without causing excessive thiosulfate consumption. In conclusion, ammonium thiosulfate leaching for gold recovery preferably should be performed at: • L o w thiosulfate concentrations, ranging from 0.1 to 0.2 M ; • A n ammonia to thiosulfate ratio of 1 to 2; 39 • Copper concentrations ranging from 50 to 100 ppm; • A n alkaline p H , preferably 9 to 10; • Ambient temperatures to 50°C. 2.5.2 Ammonia Technology The leaching of copper sulfide minerals in ammonia solution has been extensively studied throughout the years. Less information is available concerning the behaviour of copper oxide minerals in an ammonia solution. Most studies investigated the leaching of copper minerals at higher temperatures and pressures, only in the Arbiter process ammonia leaching is accomplished in a low-pressure system with sufficient oxygen availability and high stirrer speed [Kuhn et al, 1974]. The optimum conditions for the leaching of gold in ammonia thiosulfate solution are at low temperatures and atmospheric pressure (section 2.5.1). It can be concluded that there is no data available regarding the behaviour of copper minerals under these conditions. 2.5.3 Treatment of copper-gold ores by ammonium thiosulfate Copper-gold ores are available in large quantities. However, treatment of these ores by cyanide is unfavourable for several reasons. Firstly, the use of cyanide is restricted in many areas. Secondly, cyanidation of gold ores containing cyanide soluble copper, results in high cyanide consumption. Since cyanide is an expensive reagent (US$ 1.80/kg), and the recovery of cyanide from copper cyanide complexes is not widely applied, this reagent consumption can be detrimental for the economics of a project. 40 Ammonium thiosulfate might be a potential alternative for the treatment of these ores. Copper is required as a catalyst in the leaching reaction, and thiosulfate is cheap compared to cyanide (US$0.13/kg). Furthermore, ammonium thiosulfate is used as a fertilizer, thus from an environmental standpoint it has an advantage over cyanide. Little research is available regarding the treatment of copper-gold ores by ammonium thiosulfate. The research reported by Berezowsky and Sefton [1979] concerned the treatment of ammonia pressure oxidation residues, which contained 0.54 to 2.58 % copper. Several studies were performed by Gong and H u [1990] and Cao et al [1992] on a gold sulfide concentrate containing 3.16 % C u . A l l of these studies were performed at high thiosulfate concentrations (0.4 to 1 M ) , and high ammonia concentrations (1 to 2 M ) . The catalytic effect of copper on gold extraction was recognized by these researchers, but only Berezowsky and Sefton [1979] mentioned the detrimental effect of copper on thiosulfate stability. None of the studies attempted to investigate the behaviour of copper and thiosulfate during the leach. This study is directed towards developing an economical leach process for copper-gold ores using ammonium thiosulfate. Therefore, the reagent concentrations should be as low as possible. Furthermore, an understanding should be obtained of the behaviour of copper, gold and thiosulfate in an ammonium thiosulfate leach solution. 41 3 EXPERIMENTAL METHODS This chapter first discusses the materials and reagents used for testing, followed by a description of the experimental apparatus, the experimental procedures and an overview of the performed test work. The chapter concludes with a description of the analytical methods used in this work. 3.1 Materials 3.1.1 Copper Minerals To study the behaviour of copper minerals in ammonium thiosulfate solutions, samples of natural occurring copper minerals were obtained from Ward's Natural Science Establishment, Inc. These include covellite, chalcocite, chalcopyrite, cuprite and malachite. A mineral sample of enargite was received from the Montana Bureau of Mines and Geology. The mineral samples were reduced in size by subjecting the samples to a series of cone crushers, followed by a disk grinder. A narrow particle size fraction between 230 (63 urn ) and 270 mesh (53 urn) was obtained by screening. This fraction was analysed for copper and iron content. X-ray diffraction ( X R D ) was used to determine the main mineral phases present. The results of these analyses are summarized in Table 3.A. 42 Table 3.A Composition of copper mineral samples for testwork Mineral Sample %Cu* %Fe* Mineral phases Covellite (CuS) 63.1 (66.4) 2.5(0) Covellite, Chalcocite Chalcocite (Cu 2 S) 53.9 (79.8) 8.43 (0) Chalcocite, Bornite (Cu 5 FeS 4 ) Chalcopyrite (CuFeS 2 ) 28.5 (34.6) 27.7 (30.4) Chalcopyrite Enargite (Cu 3 AsS 4 ) 23.2 (48.4) 7.0 (0) Enargite, Chalcopyrite, Tennantite ((Cu,Fe)] 2 As 4 S, 3 ) Cuprite (Cu 2 0) 38.9 (88.8) 13.9 (0) Cuprite, Goethite (FeOOH) Malachite (Cu 2 (C0 3 ) (OH) 2 ) 47.7 (57.5) 0.6 (0) Malachite in brackets represents the percentage of C u or Fe in a 100% pure mineral From this table, it can be concluded that all mineral samples contain several mineral phases other than the desired copper mineral phase. For example, X R D showed the presence of goethite (FeOOH) in the cuprite mineral sample which explains the high iron content. Upgrading these copper minerals by flotation was deemed very difficult; therefore, it was decided to conduct testing with the samples in the as-received condition. Oxidation of the copper sulfide minerals after particle size reduction was prevented by purging the sample bottles containing the copper minerals with nitrogen and sealing the lids. 43 3.1.2 Copper-Gold Samples The following copper-gold concentrates and ores were obtained for the leaching studies: • Overall Lobo Composite from Teck Corporation • Guanaco Composite from Amax Gold Inc. • M 4 0 Pyrite Feed and M 1 0 Pyrite Concentrate from Newcrest Min ing Ltd . The Overall Lobo Composite and Guanaco Composite are whole ores, whereas the M 4 0 Pyrite Feed and M 1 0 Pyrite Concentrate are flotation products. Gold and copper content and the copper mineral phases present in the samples are listed in Table 3.B. Table 3.B Composition copper-gold concentrate and ore samples Copper-gold samples Au (ppm)* Cu (ppm)* Cu mineral phases Overall Lobo Composite 2.11 1,120 Chalcopyrite, Covellite, Chalcocite Guanaco Composite 2.06 3,320 Enargite, Cuprite, Malachite M 4 0 Pyrite Feed 7.10 5,920 Chalcopyrite M 1 0 Pyrite Concentrate 45.5 16,600 Chalcopyrite Assayed by Chem Met Consultants, Vancouver, B . C . Mineralogical examination of the Overall Lobo Composite revealed that the ore was characterized by highly altered rock fragments and lesser granular vein quartz with minor finely disseminated sulfides and iron oxides/oxyhydroxides. Copper mineralization was dominated by chalcopyrite (CuFeS 2 , 76%), with minor covellite (CuS, 24%), trace chalcocite (Cu 2 S, <1%) and rare bornite (<0.2%). Trace gold / electrum was identified as small grains (<15um), and was found associated with silicates and iron oxides / oxyhydroxides [Lakefield Research Limited, 1997]. 44 The Guanaco Composite was comprised of rather strongly oxidized ore with minor remnant pyrite (FeS 2), enargite (Cu 3 AsS 4 ) , and sphalerite (ZnS). The oxidation suite includes goethite (FeOOH), with lesser amounts of psilomelane ( B a M n 2 + M n 4 + 8 0 1 6 ( 0 H ) 4 ) , cuprite (Cu 2 0) , and copper oxysalts minerals (malachite (Cu 2 (C0 3 ) (0H) 2 ) , azurite (Cu 3 (C0 3 ) 2 (OH) 2 ) , and olivenite (Cu 2 (As0 4 ) (OH)) [Honea, 1998]. N o information is available about the gold occurrence or association in this composite. The M 4 0 Pyrite Feed was characterized by 78% pyrite (FeS 2) and 10% quartz (S i0 2 ) . Some minor chalcopyrite (CuFeS 2 , 1.6%) and calcite ( C a C 0 3 , 1%) was present and the remainder was comprised of mainly clay minerals. The M 1 0 Pyrite Concentrate was found to be similar with 76% pyrite and 7% quartz as the main minerals present. Minor chalcopyrite (1.6%) and calcite (2.5%) and some dolomite (CaMg(C0 3 ) 2 , 5.0%) were identified [Oretest, 1997]. Details regarding gold association are not available for these concentrates. The Lobo composite and Guanaco composite samples were reduced in particle size by a series of cone crushers, followed by pulverization using a disk pulverizer. The product was split into smaller quantities (about 400 grams) suitable for testwork. Size reduction for the M 4 0 Pyrite Feed and M 1 0 Pyrite Concentrate was not necessary. The concentrates were also split into smaller quantities for testwork. The screen analyses of the copper-gold samples are given in Appendix A ; the 80% passing size (P 8 0) was determined for every material (Table 3.C). 45 Table 3.C P 8 0 of the copper-gold samples Copper-gold samples P8o (M-m) Overall Lobo Composite 141 Guanaco Composite 160 M 4 0 Pyrite Feed 75 M 1 0 Pyrite Concentrate 46 The P 8 0 of the Guanaco and Overall Lobo Composite is high for leaching experiments, however, both size fractions are already the result of a regrind. Further reduction of the P 8 0 was deemed difficult with the equipment available, therefore, testing was continued with this particle size distribution. 3.1.3 Reagents • Soluble gold was added to some leaching solutions. A 1,000 ppm A u reference standard solution was used. In this solution (10% hydrochloric acid) gold is present as a chloride complex (AuCl 4 "). • Unless otherwise stated, analytical grade ammonium thiosulfate in powder form was used. Tessenderlo-Kerley provided Thiogold™ grade ammonium thiosulfate solution. The solution contains between 5 0 - 60 % ammonium thiosulfate ( ( N H 4 ) 2 S 2 0 3 ) , 40 - 50 % water and 0 - 4 % ammonium sulfate ( (NH 4 ) 2 S0 4 ) . • Other reagents used during testing were cupric sulfate pentahydrate, C u S 0 4 . 5 H 2 0 , anhydrous sodium sulfite, N a 2 S 0 3 , and anhydrous sodium sulfate, N a 2 S 0 4 . Concentrated sodium hydroxide (50% solution in water) was used for p H adjustments. Ammonia solutions (1 and 3 46 M ) were prepared from concentrated reagent grade ammonium hydroxide ( N H 4 O H ) solution. Preliminary experiments used anhydrous ammonia gas (99.99% purity) to adjust the p H of the leach solution and 1 M sulfuric acid ( H 2 S 0 4 ) prepared from reagent grade sulfuric acid. A i r (19.5-23.5% 0 2 ) was supplied via a gas cylinder. 3.2 Experimental Apparatus A schematic diagram of the experimental set-up used for the batch leaching tests is given in Figure 3.1. The experiments were carried out in a 1500 m L cylindrical glass vessel with a removable high density polyethylene l id. The vessel was vertically baffled with a baffle width of 11 mm which is 1/10 of the tank diameter. The baffles were made of acrylic. The solutions were stirred by a stainless steel Rushton impeller (35 mm diameter, 1/3 of the vessel diameter, 6 blades) on a stainless steel shaft connected to a variable speed D C motor. A l l experiments were performed at an impeller speed of 1695 R P M . It was experimentally determined that this was the minimum impeller speed required for complete particle suspension. 47 Thermocouple Heater pH probe Sparger Rushton turbine Redox probe ' Baffles Magnetic stirrer Figure 3.1. Schematic diagram of the experimental set-up. Unless otherwise stated, experiments were performed at 35°C. The glass vessel was placed in a water bath heated by an immersion heater connected to a PID controller. A thermocouple was placed in the reactor to provide temperature control. A i r was added to the solutions via a stainless steel sparger placed directly under the Rushton impeller. A gas flowmeter controlled the flow of air into the reactor. Controlled p H experiments were performed by ammonia gas and sulfuric acid addition. Two solenoid valves, connected to the p H controller, regulated the flow of the ammonia gas from a gas cylinder and the flow of sulfuric acid from a burette. The p H was controlled in a range of ± 0.10 p H . The ammonia gas was added to the solution via the air sparger. 48 A gel-filled combination p H probe with a silver/silver chloride reference electrode was used to monitor the p H of the solutions. The redox potential was measured using a platinum combination redox electrode with a silver/silver chloride reference electrode. The E° of this reference electrode at the experimental temperature of 35°C was 189 m V . In order to report the experimentally observed E h values versus the potential of the standard hydrogen electrode (SHE), all values were corrected by adding 189 m V . 3.3 Experimental Procedures Four sets of experiments were performed: 1. Preliminary testwork 2. Leaching of copper minerals 3. Leaching of copper minerals with 5 ppm gold addition 4. Leaching of copper-gold samples The procedure for each set varied and w i l l be discussed separately. 3.3.1 Preliminary testwork The preliminary testwork involved the leaching of the copper minerals in one litre of deionized water. Depending on the copper content of the different minerals (Table 3. A ) , sufficient mineral feed was added to the solution such that upon 100% copper dissolution, the copper concentration would be 5 g/L. Due to the limited amount of the 53-63 jam material available, the preliminary testwork was conducted with the <53 jam fraction. 49 The initial slurry of copper mineral feed was heated to the desired temperature. Once the desired temperature was attained, aeration of the slurry was initiated (0.23 L/min), and the ammonium thiosulfate was added. The p H was adjusted to the desired value with either ammonia gas or 3 M ammonia solution. This time was noted as time zero. In p H uncontrolled experiments, the p H was allowed to drift. During p H controlled experiments, the p H was adjusted with either ammonia or sulfuric acid in a range of ± 0.1 p H . Slurry samples of 7 m L were withdrawn at selected intervals and centrifuged for liquid/solid separation. The resulting solutions were prepared for copper, gold, thiosulfate, tetrathionate and sulfate analyses. Experiments continued until the redox potential of the solution stabilized (typically 6 - 8 hours). A t the end of the experiment the solution was filtered, the residue was washed with deionized water, dried, and collected. 3.3.2 Leaching of copper minerals After the preliminary testwork, some adjustments were made in the experimental procedure. A s above, experiments were performed in one litre of deionized water. Sufficient mineral feed was added to the solution such that upon 100% copper dissolution, the copper concentration would be either 5 g/L or 1 g/L copper. The fraction 53-63 urn of the copper minerals was used for this set of experiments. Heating, aeration and reagent addition followed the same procedure described above. The p H of the slurry was adjusted to 10 for all experiments with concentrated sodium hydroxide and the time was set to zero. The p H was allowed to drift in all experiments. 50 Slurry samples of 7 m L were withdrawn at 30, 60, 120, 180, 240 and 360 minutes, centrifuged and analysed for copper, thiosulfate, tetrathionate and sulfate. In several experiments, a solution sample was taken after 5 minutes. However, this sample time was discontinued at a later stage in order to reduce the time necessary to perform the analytical procedure with the ion chromatograph after the experiment (section 3.5). The experiments lasted 6 hours, after which the solution was filtered. The residues and the wash water were collected. The amount of residue remaining after the experiments was not deemed sufficient for copper analysis (see section 3.5). 3.3.3 Leaching of copper minerals with 5 ppm gold addition The experimental procedure for this series of experiments is similar to that described in section 3.3.2 with the exception that 5 ppm of gold from the standard reference was added to the deionized water. Accordingly, solution samples were also analysed for gold. 3.3.4 Leaching of copper-gold samples The experiments involving the concentrates were performed at a pulp density of 30% by weight (360 grams of concentrate and 840 grams of deionized water). The pulp was heated to desired temperature, after which the reagents were added. Experiments were continued as described in section 3.3.2. One adjustment was the sample volume. Because of the higher pulp density 8 m L samples were taken. Solution samples for copper, gold, thiosulfate, tetrathionate and sulfate analysis were prepared. Wash water samples were analysed for gold content. The residues were analysed for copper and gold. 51 3.4 Experimental Design The experimental design is discussed according to the four different series of experiments mentioned in section 3.3. 3.4.1 Preliminary experiments Preliminary experiments (at controlled p H and uncontrolled pH) were conducted with the copper minerals at 0.05 and 0.1 M ammonium thiosulfate at pH's ranging from 8 - 9 . 5 . A l l experiments were performed at a temperature of 35°C. 3.4.2 Leaching of copper minerals So-called baseline experiments were first conducted, after which the effects of several parameters could be studied (Table 3.D). Baseline tests included: aeration of the system, 0.2 M ammonium thiosulfate, p H 10 (uncontrolled), a copper mineral addition which would achieve 5 g/L copper or 1 g/L copper upon 100% dissolution, and 35°C. The experiments lasted 6 hours. Two experiments were performed without aeration. Table 3.D Overview of experiments with copper minerals Copper minerals Baseline No aeration Covellite X Chalcocite X X Chalcopyrite X Enargite X Cuprite X X Malachite X 52 3.4.3 Leaching of copper minerals with 5 ppm gold addition The experiments were performed under the same conditions as described in section 3.4.2. However, several other parameters were evaluated in this test series; aeration, leach temperature, ammonia concentration, sulfite concentration and sulfate concentration (Table 3.E). Table 3.E Overview of experiments with copper minerals and gold addition Copper mineral Baseline No aeration Temperature 20°C 0.4 M NH 3 0.05 M so 3 2 0.05 M so4 2-Covellite X Chalcocite X X X X X X Chalcopyrite X Enargite X Cuprite X X x Malachite X 3.4.4 Leaching of copper - gold samples Baseline experiments of this test series included: aeration of the system, 0.2 M ammonium thiosulfate, p H 10 (uncontrolled), 30 % pulp density, and 1 g/L copper addition (as cupric (CuS0 4 . 5H 2 0) ) and 6 hours. The parameters studied in this series are (see Table 3.F): aeration of the system, leaching time, ammonia thiosulfate concentration, copper addition and ammonium thiosulfate grade. 53 Table 3.F Overview of experiments with copper-gold samples Copper-gold samples Baseline No aeration No aeration, 24 hours No cupric No aeration, no cupric Extra ATS* Thiogold™ Grade Overall Lobo Composite X X X Guanaco Composite X X X X M 4 0 Pyrite Feed X X X X M 1 0 Pyrite Concentrate X X X X X * A T S = Ammonium Thiosulfate 3.5 Sample Analysis 3.5.1 Copper and gold analysis A l l the copper and gold assays on solution and solid samples were performed by Chem Met Consultants in Vancouver. Selected solution samples were analysed for gold and copper content. The solution samples (exactly 20 mL) for copper were prepared for atomic absorption spectrophotometry ( A A S ) by treatment with hydrogen peroxide to destroy any thiosulfate present since thiosulfate interferes with A A S measurements. After this, nitric acid was added to dissolve any precipitates resulting from the hydrogen peroxide addition. The solution volume was adjusted to known volume and the copper content measured by A A S . 54 Between 3-5 m L of sodium cyanide was added to the samples for gold analysis, to keep gold in solution. The gold content was determined by evaporating 25 m L of solution to dryness in a lead foil boat, followed by fire assay and cupellation. The resulting bead was digested and the gold content of the resulting solution was determined by A A S . The residues of the copper-gold samples were analysed for copper and gold for mass balance purposes. A representative sample of the residues was obtained by coning and splitting. Copper content of the residues was determined by acid digestion and A A S analysis of the resulting solution. Gold content of the residues was determined by fire assay and cupellation. The resulting bead was digested and the gold content of the resulting solution was determined by A A S . 3.5.2 Thiosulfate, tetrathionate and sulfate analysis A l l solution samples were analysed for thiosulfate, tetrathionate and sulfate by high performance liquid chromatography ( H P L C ) methods directly following the experiment. A Dionex Series 4500i Chromatograph was used for the analysis. The ion chromatography methods and procedures are discussed in more detail in Appendix B . Solution samples produced from the experiments were diluted to a concentration suitable for detection by the ion chromatograph (< 20 ppm). The typical dilution factor was 1,000 times. The diluted samples were kept refrigerated and in the dark to reduce the rate of thiosulfate degradation. The samples were allowed to warm to room temperature just before analysis. 55 It should be noted that only the free ion concentrations are detected by ion chromatography. Sulfur species complexed with metal ions w i l l either not be retained by the column (when the complex has no charge) or have such a high affinity for the ion exchange sites on the column that the complex does not elute; for example the copper(I)tri-thiosulfate complex which has a -5 charge. Further, for most experiments the presence of sulfite and trithionate was detected by ion chromatography. However, these species could not be quantified. 56 4 RESULTS AND DISCUSSION The experimental results are presented and discussed in this chapter. A s mentioned in Chapter 3, four different sets of experiments were conducted: preliminary testwork, leaching of copper minerals, leaching of copper minerals with gold addition to solution and leaching of copper-gold samples. The experimental results are discussed accordingly. First, the construction of Eh-pH diagrams which represent the experimental conditions are discussed. 4.1 E h - p H diagrams To understand and characterize the chemistry of copper-thiosulfate-ammonia solutions containing gold, Eh-pH diagrams were constructed which reflect the experimental conditions used in this research. Eh-pH diagrams represent the thermodynamic equilibria in a system and do not give information regarding the kinetics of the reactions represented in the diagram. This has to be considered when interpreting Eh-pH diagrams. The thermodynamic data used for the various copper and gold species expected to be present in these solutions are presented in Appendix C. The diagrams were constructed with the use of the computer program C S I R O Thermochemistry [Turnbull, 1986]. A n E h range of -1.5 to 1.5 V S H E and a p H range of 6 to 14 were considered. The p H range was chosen between 6 and 14; below p H 6 complex reactions involving the hydrolyses of thiosulfate start to take place which are not pertinent to this research (see section 2.3.3). The p H values of interest are around 9-10 (see section 2.5). According to the literature, the E h values during thiosulfate leaching are generally in the range of 150 to 250 m V S H E [Atluri, 1987]. 57 The Eh-pH diagrams were constructed for 25°C, while the experiments were conducted at 35°C. Entropy data are required to calculate the free energy of a species at 35°C. This data is not available for several sulfur species (e.g. S 2 0 3 2 ", S x 0 6 2 ", x>2). The diagrams were therefore constructed at 25°C. The change of the ammonia/ammonium buffer point with temperature can be calculated and changes from p H 9.25 at 25°C to p H 8.96 at 35°C. This change in p H value is not incorporated on the Eh-pH diagrams, but has to be considered when interpreting the experimental results. Moreover, Eh-pH diagrams are theoretical thermodynamic diagrams and should therefore be constructed using ionic activities. In dilute solutions these activities can be replaced by molalities (mol/kg of solvent). Due to the complex nature of the solutions in this system, representative activity coefficients are not readily available. The experimental solutions cannot be considered to be dilute either, since at least 0.4 M N H 3 and 0.2 M S 2 0 3 2 " are present. The diagrams were constructed using molarities, which in this instance, is the most practical alternative. The Eh-pH diagram for the gold-thiosulfate-ammonia-water system appears in Figure 4.1. The gold activity is 2.5 10"5 M (5 ppm), the thiosulfate activity is 0.2 M and the ammonia activity is 0.4 M . It can be seen that the gold(I)thiosulfate complex is stable in the whole p H range shown. 58 Figure 4.1. Eh-pH diagram of the gold-thiosulfate-ammonia-water system at 25°C. The activities of the species are 2.5 x 105 Au (5 ppm), 0.2 S 20 3 2and 0.4 NH 3 . AG f 0(S 2O 3 2) = -518.8'kJ/moL Comparing Figure 4.1 with the Eh-pH diagrams given in the literature (see Figure 2.7), it can be observed that the gold(I)ammonia complex has no region of stability in Figure 4.1, while Figure 2.7 shows a region of stability for this complex. A closer examination of the literature revealed that the free energy of formation value of the thiosulfate species was different. In Figure 4.1 a value of-518.8 kJ/mol was used for the thiosulfate species [Bard et al, 1985]. The Handbook of Chemistry and Physics [Weast, 1975] lists a value of -532.2 kJ/mol for thiosulfate, resulting in the Eh-pH diagram presented in Figure 4.2, which resembles Figure 2.7. 59 PLOT LABELS Temp = 2 9 8 . 1 5 K I Ru1 = 2 . 5 E - 0 5 IS2031 = 0 . 2 INM31 = 0 . U STABLE BRERS fl flu B H Ru 03 < 2 > (BQ) C H2 flu 03 < -> (BQ) D Ru [0 H 13 E Ru [N H3 12 « -> (BO) F Ru (S2 03 12 < 3 - > (BQ) H20 STABIL ITY L I M I T S 1 02 /H2D 2 H2/H2D 6 8 10 12 14 P H Figure 4.2 Eh-pH diagram of the gold-thiosulfate-ammonia-water system at 25°C. The activities of the species are 2.5 x 105 Au (5 ppm), 0.2 S 20 3 2 and 0.4 NH 3 . AG f 0(S 2O 3 2) = -532.2 kJ/mol. The gold(I)ammonia complex now appears next to the gold(I)thiosulfate complex. This large effect of this relatively small change in free energy is explained when the following equation is considered: A u ( S 2 0 3 ) 2 ~ + 2 N H 3 = A u ( N H 3 ) 2 + 2 S 2 0 3 ~ (IV. 1) The free energy change for this reaction, calculated using the AG f° value o f -518.8 kJ/mol for thiosulfate, is + 22.5 k J which indicates that this reaction is not favoured. When the free energy change is recalculated using the AG f° value of -532.2 kJ/mol for thiosulfate, the value -4.3 k J is obtained which indicates that the formation of the gold(I)ammonia complex is now favoured. It can be concluded that a small change in the free energy of the thiosulfate species has a great 60 impact on the thermodynamic equilibria presented in the Eh-pH diagram of the gold-thiosulfate-ammonia-water system. According to Figure 4.2, the gold(I)ammonia complex is the most stable species at p H 10, which does not agree with experimental observations. A s L i et al [1996] remarked, it is generally accepted that the gold(I)thiosulfate species is the more stable species at p H 10 and L i et al [1996] confirmed this by rest potential measurements (see section 2.3.1). These rest potential measurements are easily explained by Figure 4.1. This example illustrates that Eh-pH diagrams have to be interpreted with the necessary precautions. In Figure 4.1 a decrease in thiosulfate concentration results in a slight decrease in area of stability for the gold(I)thiosulfate complex over the entire p H range considered. In Figure 4.2 a decrease in thiosulfate concentration results in a larger area of stability for the gold(I)ammonia complex. When thiosulfate disappears, thermodynamically the gold(I)ammonia complex becomes the stable species over the whole p H range. In this study the concentrations of gold in ammoniacal thiosulfate solutions are usually very low (in range of 5 to 30 ppm). Hence, a variation in gold concentration should not significantly affect the thermodynamics of the system. The copper-thiosulfate-ammonia Eh-pH diagram, calculated using the value of -518.8 kJ/mol for thiosulfate, is presented in Figure 4.3. The copper(I)tri-thiosulfate complex occupies the whole p H range and is stable in the area of the leaching conditions; p H 10 and an E h range of 150-250 m V S H E . The copper(I)di-thiosulfate complex is stable at low p H values. A slight lowering of the E h w i l l result in the precipitation of copper as CuS. 61 Figure 4.3 Eh-pH diagram of the copper-thiosulfate-ammonia-water system at 25°C. The activities of the species are 0.015 Cu (0.95 g/L)„ 0.2 S 20 3 2 and 0.4 NH 3 . AGf° (S 20 3 2) = -518.8 kJ/mol. The copper(II)ammonia complex has a small stability region around the ammonia/ammonium buffer point. A s mentioned before, this buffer point w i l l shift towards p H 8.96 at 35°C. This area lies slightly above the 0 2 / H 2 0 line, and w i l l therefore not exist at significant concentrations. This complex w i l l have a region of stability within the water lines, when the thiosulfate concentration decreases. A decrease in copper concentration increases the width of the region of the copper(II)ammonia complex. The dominant copper species at the Eh-pH conditions o f the experiments is not affected by a decrease in thiosulfate concentration. Calculating the same diagram with the thiosulfate value of-532.2 kJ/mol, results in Figure 4.4. 62 PLOT LABELS Temp = 2 9 8 . 1 5 K ICul = 0 . 0 1 5 IS2D3! = 0 . 2 INH31 = 0 . U STABLE flRERS P Cu S B Cu C Cu2 S D Cu2 0 F Cu 0 F Cu [S2 0 3 1 3 < 5 - > (RQ) C Cu [N H3|L| <2+> IRQ) H20 S T A B I L I T Y L I M I T S 1 0 2 / H 2 0 2 H2 /H20 6 8 10 12 m P H Figure 4.4 Eh-pH diagram of the copper-thiosulfate-ammonia-water system at 25°C. The activities of the species are 0.015 Cu (0.95 g/L)„ 0.2 S 20 3 2 and 0.4 NH 3 . AG f 0(S 2O 3 2) = -532.2 kJ/mol. When comparing Figure 4.4 to Figure 4.3, two observations can be made. The region o f interest is p H 10. In Figure 4.4, the area for the copper(I)thiosulfate complex is greatly reduced compared to Figure 4.3. A t E h values lower than 0 V S H E , the Cu 2 S species has an area of stability directly below the copper(I)thiosulfate species instead of CuS in Figure 4.3. Furthermore, the region of stability of the copper(II)ammonia complex increases to lower E h values when compared to Figure 4.3. Figure 4.5 shows the copper-ammonia-water system. This Eh-pH diagram represents the leaching conditions when thiosulfate is no longer present in solution. 63 Figure 4.5 Eh-pH diagram of the copper-ammonia-water system at 25°C. The activities of the species are 0.015 Cu (0.95 g/L) and 0.4 NH 3 . The copper(I)ammonia complex now appears as a stable species in the area of interest. The areas of the copper(I)ammonia and copper(II)ammonia decrease in width with higher copper concentrations and lower ammonia concentrations. A t very low ammonia concentrations the stability region o f the copper(II)ammonia complex completely disappears. 4.2 Preliminary experiments The preliminary experiments were performed to obtain an understanding of the behaviour of copper minerals in an ammonium thiosulfate solution under conditions that are typical for the leaching o f gold (see section 2.5.1). The results of these experiments were used to determine the experimental conditions for further testwork. 64 Conditions A s described in section 3.3.1 the preliminary experiments were performed with the copper minerals (section 3.1.1), particle size of minus 270 mesh (53 jam). A copper mineral addition was employed which would achieve 5 g/L copper in solution upon 100% dissolution. In the remainder of this chapter this w i l l be referred to as a 5g/L copper addition. 5 g/L copper was chosen as a representative amount for an industrial operation. This would be equivalent to 0.5% copper in an ore, with leaching at 50% pulp density. A s outlined in the literature review (section 2.5.1) it is necessary to apply low ammonium thiosulfate concentrations, ranging from 0.05 - 0.20 M , to achieve an economical process for the recovery of gold by ammonium thiosulfate leaching. Therefore, testing started with concentrations of 0.05 and 0.1 M ammonium thiosulfate. Since the thiosulfate to copper ratio in the copper(I)thiosulfate complex is 3 to 1 (the Cu(S 2 0 3 ) 2 3 " complex is not stable under these conditions), there was not enough thiosulfate present in solution to complex all the copper (maximum of 5 g/L copper (=0.079 M copper)). The same applies to the ammonia concentration; the ammonia to copper ratio in the copper(II)ammonia complex is 4 to 1 and thus not enough ammonia is present to complex all the copper i f 100% dissolution would occur. However, ammonia was used for p H adjustment, resulting in sufficient ammonia concentrations at high p H values (>9.0). The p H range tested was from 8 to 9.5. Testing was started at lower p H values, considering the shift o f the ammonia/ammonium buffer point to p H 8.96 at 35°C. First, the experiments were p H 65 controlled in a narrow range (+0.1 pH). After several experiments, however, it was realized that by allowing the p H to drift during the experiment, a better understanding of the chemical reactions occurring in the leach solution can be obtained. Accordingly, the p H was allowed to drift in all further experiments. Results A l l preliminary experiments showed that the thiosulfate disappeared quickly. Thiosulfate was not detected in the leach solutions after 2 to 3 hours. A n example is presented in Figure 4.6., the leaching of covellite with 0.10 M ammonium thiosulfate at p H 9.5. The first graph reflects the p H (primary y-axis) and E h (secondary y-axis) changes during the experiment. The negative time on this plot reflects the heating of the solution to 35°C, in the absence of reagents. The addition of the reagents to the solution was noted as time zero. The large increase of p H and sharp decrease of the E h result from the reagent addition. The bottom graph shows the results of thiosulfate and tetrathionate analyses by ion chromatograph on the primary y-axis, and the copper extraction on the secondary y-axis. Sulfate concentrations are not available for this experiment. 66 a 12 -50 *>—• • M ^ ^^ ^^ 50 150 250 Time (minutes) 350 p H -0— E h 600 400 200 W 450 DC w o c •Si u 10 100 200 300 Time (minutes) 400 60 40 c o « 20 | a o U ^ 0 500 • Thiosulfate Tetrathionate Copper Figure 4.6. Covellite, 5 g/L copper, 0.1 M S 20 3, 0.47 M NH 3 , pH 9.5, aeration, 35°C 67 A more detailed discussion of the leaching of covellite is given in a later section (4.3.1). This graph is presented here for two reasons: • The thiosulfate concentration in solution decreases quickly. This was also observed during the leaching of the other copper minerals. • Copper extraction is low, just above 20% (of 5 g/L), which is about 1 g/L of copper. Based on these results it was decided to try to maintain thiosulfate in solution by increasing the thiosulfate concentration to 0.2 M . The p H was set to 10 for further experiments. A t this p H thiosulfate is quite stable (see section 2.3.3) and the presence of ammonia instead of ammonium in solution is ensured. A t lower p H values the leaching of the copper sulfide minerals might be prevented because of an ammonia deficiency. Moreover, the experimental procedure was changed. Exposing the copper minerals to deionized water resulted in a unique solution p H for each mineral. B y adjusting the p H with ammonia, a different ammonia concentration was obtained for each experiment. To ensure the same ammonia concentration for all experiments, concentrated sodium hydroxide was used for p H adjustment. B y adding the same amount of ammonium thiosulfate to every experiment and raising the p H to 10 by sodium hydroxide, the concentrations of thiosulfate and ammonia w i l l be similar for every experiment (small differences in concentration w i l l exist due to the fact that the amount of sodium hydroxide required to raise the p H to 10 is mineral dependent). Furthermore, some difficulties were experienced with the dilution of solution samples to concentrations within the range of the ion chromatograph (2-20 ppm). Several samples turned 68 turbid upon dilution. Ammonia was added to the solutions to redissolve solid precipitates, enabling the measurement of the samples by ion chromatograph. It was realized that this addition changed the solution chemistry. For the next series.of experiments it was recorded to which samples ammonia was added (Appendix D). The ammonia addition may affect the solution chemistry in the following way: ammonia could displace any thiosulfate complexed with copper, resulting in a higher measured free thiosulfate concentration than actually present in the leach solution. In most experiments, ammonia was only added to the last samples (240, 360 minutes) resulting in no effect since in most experiments thiosulfate was no longer present after 180 minutes. For experiments with high initial copper extraction, ammonia was added to all samples. Again, this resulted in a minimum effect, since thiosulfate degraded quickly in the leach solutions because of the high copper concentration. 4.2.1 Summary The main conclusion that can be drawn from the preliminary experiments is that thiosulfate was not detected after 2 to 3 hours under the experimental conditions applied. To extend the presence of thiosulfate in solution the thiosulfate concentration was increased to 0.2 M for further testwork. The copper extractions for the sulfide minerals were not very high and this was thought to be due to low p H values, i.e. at low p H values ammonium w i l l be the dominant species instead of ammonia. To ensure the presence of ammonia in the leach solution, the p H was increased to 10 69 in later testwork. This p H value might also positively affect the thiosulfate stability, since it is reported to be at a maximum in solutions of p H 9 to 10 (see section 2.3.3). 4.3 Leaching of copper minerals In these experiments copper minerals were leached in an ammonium thiosulfate solution (see Table 3.D). The baseline experiments included a mineral addition which upon 100% copper dissolution would result in 5 g/L copper in solution, 0.20 M ammonium thiosulfate (~ 22 g/L thiosulfate), p H 10, 35°C and aeration. The p H was allowed to drift. The behaviour of thiosulfate, tetrathionate, sulfate and copper was recorded. The effect of aeration was studied. Experimental data and observations made during the experiments, such as solution colours are presented in Appendix D . A l l % extraction calculations are based on the head assays of the copper minerals. 4.3.1 Baseline experiments The results of the baseline experiments are presented in Figures 4.7 to 4.12. The top graphs represent the p H and E h values during the leach, whereas the bottom graphs show the sulfur species and copper extractions. Instrumental difficulties prevented the determination of sulfate concentrations for some experiments. 70 X c 12 10 8 6 + ' 4 2 0 800 I 600 400 > + 200 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) - * - p H - * - E h 25 *2 Vi V "3 a Vi U a 100 200 300 Time (minutes) 50 40 Co © 30 « 20 Z a 10 tj 400 Thiosulfate • Tetrathionate -•— Copper Figure 4.7 Covellite, 5 g/L copper, 0.20 M (NH4)2S203, aeration, 35°C 71 0 100 200 300 400 Time (minutes) —•— Thiosulfate —•— Tetrathionate —•>— Copper Figure 4.8 Chalcocite, 5 g/L copper, 0.20 M (NH4)2S203, aeration, 35°C 72 a 12 10 8 6 4 2 0 600 400 s izi > 1 1 1 1 1 1 1 1 -50 0 50 100 150 200 250 300 350 400 Time (minutes) 200 W 0 •pH -•— E h 25 3 VI s tZ2 100 200 Time (minutes) 300 50 40 a 30 | « 20 « O a a 10 ° 400 -•— Thiosulfate Tetrathionate - * - Sulfate ~#— Copper Figure 4.9 Chalcopyrite, 5 g/L copper, 0.20 M (NH4)2S2Q3, aeration, 35°C 73 12 10 8 6 4 2 4-0 - * > « - a — • — • — • — • ^ ^ ^ - HMt 0 0 0 ^ H h H ( -600 400 Q s £ 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH -0— E h 25 20 S 15 "3 a u "a 10 100 200 Time (minutes) 300 50 40 Is 30 | 20 2 a a 10 cj 400 -•— Thiosulfate • Tetrathionate - A - Sulfate Copper Figure 4.10 Enargite, 1 g/L copper, 0.20 M (NH4)2S203, aeration, 35°C 74 c. 12 10 8 + 6 4 2 0 4- 400 600 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) p H -0—Eh 100 200 Time (minutes) 300 100 80 a 60 | cs S-a a o 40 20 400 Thiosulfate Tetrathionate Sulfate Copper Figure 4.11 Cuprite, 5 g/L copper, 0.20 M (NH4)2S203, aeration, 35°C 75 0 100 200 300 400 Time (minutes) -•— Thiosulfate -m- Tetrathionate Sulfate -#— Copper Figure 4.12 Malachite, 5 g/L copper, 0.20 M (NH4)2S203, aeration, 35°C 76 A n examination of the top graphs of the figures reveals that all figures, except Figure 4.10, show an initial p H increase followed by a p H decrease after a certain time period. The solution p H is influenced by several reactions which occur simultaneously: • The oxidation of thiosulfate to tetrathionate (Equation IV.2), coupled with the reduction of oxygen to hydroxide (Equation IV.3), resulting in the production of hydroxide ions: 2 S 2 0 3 _ ->S 4 0 2 , "+2e" (IV.2) \ 0 2 + H 2 0 + 2e" -> 2 0 F T (IV.3) • Concurrently, hydroxide w i l l be consumed by the hydrolysis of tetrathionate in a basic solution: 2S40l + 3 0 P T -> f S2Ol~ + S3026~ + | H 2 0 (IV.4) • Hydroxide is also consumed because ammonia is removed from the system due to evaporative losses and complexation reactions. The equilibrium depicted in Equation (IV.5) changes accordingly: N H 3 + H 2 0 N H ; + O H " (IV.5) The reactions depicted in Equations (IV.2) and (IV.3) are dominant while thiosulfate is present. When the thiosulfate is gone, the hydroxide consuming reactions become more prominent and consequently, the p H decreases. This sequence of reactions is visible in for example, Figure 4.7, the leaching of covellite. The p H increases as long as thiosulfate is present, followed by a p H decrease around the same time (about 240 minutes) that thiosulfate was no longer detected in solution (bottom graph). 77 A closer look at Figure 4.7 reveals that the E h remains constant for a long time and then suddenly increases around 240 minutes. This time coincides with the instant that the p H starts to drop and thiosulfate is no longer present. Further, the colour of the leach solution changes: after about 180 minutes the solution colour changes from colourless to light blue indicating that the copper(II)ammonia complex has formed (see Appendix D). Moreover, the copper extraction curve shows a steady increase after thiosulfate is gone, which can only be explained by copper complexing with ammonia, since there is no thiosulfate present in solution. It can be remarked that the copper extraction is low. Figure 4.5, the Eh-pH diagram of the copper-ammonia system at 25°C, can be used to interpret these results. It can be seen from the bottom graph that the copper concentration is only about 0.006 M . This means that the stability regions of the copper(I)ammonia and copper(II)ammonia complex are slightly larger than represented in Figure 4.5. The p H and E h value measured indicate that it is likely that the copper(I)ammonia complex forms under these conditions. The copper(I)ammonia complex oxidizes rapidly, in the presence of sufficient oxygen, to form the copper(II)ammonia complex, explaining the colour change of the solution and the rise in Eh . Figures 4.7 to 4.12 show that the p H and E h trends described above are applicable to all figures, except for the leaching of enargite (Figure 4.10). However, this figure too confirms the reactions postulated above. N o decrease in p H value is observed, which indicates that thiosulfate was present throughout the entire experiment. Consequently, no change in E h value should occur, which is confirmed by the E h curve in Figure 4.10. 78 Chalcocite (Figure 4.8) shows an increase in potential after about 40 minutes, thereafter stabilizing until at 180 minutes the E h increases again to a higher potential. A s discussed in the literature review, section 2.4.2, half of the copper in chalcocite is essentially free and leaches rapidly leaving a covellite matrix behind ( A G 0 o f Equation (IV.6) is -313.1 kJ/mol, using -518.8 kJ/mol for S 2 0 3 2 ) : 2Cu 2 S + 8 S 2 0 2 ~ + 4 N H ; + 0 2 -> 2CuS + 2 C u ( S 2 0 3 ) 5 3 ' + S 4 0 2 _ + 4 N H 3 + 2 H 2 0 (IV.6) Figure 4.8 shows that copper extraction decreases slightly and then rises again. It was observed that the solution colour changed from colourless to light blue after 30 minutes. This indicates that the copper(I)thiosulfate complex, the copper(I)ammonia and the copper(II)ammonia complex could be present under these conditions. According to the Eh-pH diagram in Figure 4.3 the copper(I)thiosulfate species is dominant. However, the Eh-pH diagram presented in Figure 4.4. shows that under these conditions the copper(II)ammonia complex can also be present, which is indicated by the solution colour. When most of the copper is present as the copper(I)thiosulfate complex, free thiosulfate has to be present in solution to keep the equilibrium depicted in Equation (IV.6) to the right. When the thiosulfate degrades, the copper(I)thiosulfate complex also decomposes following the equilibrium. When thiosulfate is no longer present, copper dissolves again forming complexes with ammonia, which results in an E h rise (copper(I)ammonia to copper(II)ammonia) and an increase in copper extraction. Chalcopyrite shows similar behaviour with an increase in copper extraction after the thiosulfate is no longer present (Figure 4.9). 79 The patterns are different for the leaching of the copper oxide minerals, cuprite and malachite (Figure 4.11 and 4.12). High copper extractions are achieved at the start of the experiment, followed by a decrease in extraction and then a steady increase once the E h changes. Cuprite leaches according to Equation (IV.7) ( A G 0 o f Equation (IV.7) is -117.5 kJ/mol, using -518.8 kJ/mol for S 2 0 3 2 "): C u 2 0 + 6S 2 0 2 ~ + 2 N H 4 + -> 2Cu(S 2 0 3 ) 5 3 _ + 2 N H 3 + H 2 0 (IV.7) The high concentration of sulfate after 5 minutes and its steady decline does not seem to correlate with the patterns previously observed. B y summing up the concentrations of thiosulfate, tetrathionate and sulfate, a sulfur balance can be calculated at every moment of the experiment (Appendix D). This balance w i l l be discussed in more detail in section 4.6. The resulting sulfur balance for this experiment is quite close to the 0.4 M sulfur (0.2 M S 2 0 3 2 ") which was originally added to the system. A closer examination of the graph revealed that the results can be explained by sample degradation. The samples are analyzed as soon as possible by ion chromatograph. Practically, however, the time between taking the sample and analysis by ion chromatograph can be up to 7 to 10 hours for the first few samples (taken at 5, 30, 60 minutes) and less than 7 hours for the last samples. When a lot of copper is present in solution, as is the case when leaching the copper oxide minerals, and the cupric ion is generated in the presence of air, the degradation of thiosulfate is accelerated (see section 2.3.3). Re-examining Figure 4.11, the leaching of cuprite results in a high copper extraction. The graph 80 also shows that there is a relatively high tetrathionate concentration and very high sulfate concentration. Both are the products of thiosulfate degradation. The previous graphs show similar trends, but to a lesser extent because the copper concentrations are lower. Figure 4.11 shows that after the initial high copper extraction the copper concentration declines and stabilizes after about 240 minutes. The same explanation given for the leaching of chalcocite applies to cuprite. A t 30 minutes copper is complexed with both thiosulfate and ammonia, because there is not enough thiosulfate in solution to complex copper and give a free thiosulfate concentration of 5 g/L thiosulfate (see bottom graph, Figure 4.11). This is confirmed by the blue colour of the solution, indicating the presence of the copper(II)ammonia complex. From this, it can be concluded that Figure 4.4. seems to represent the actual solution more accurately, instead of Figure 4.3. If the copper was mainly complexed with ammonia during the entire experiment, the decline in copper extraction would not have been observed, since the ammonia concentration remains fairly constant during the experiment. There wi l l only be a slight ammonia loss to the atmosphere. The same explanation can be applied to malachite (Figure 4.12). For the enargite mineral sample the conditions were different. After several experiments it was concluded that the 5 g/L addition was too high in order to observe the leaching of copper with both thiosulfate and ammonia, since there was a deficiency of thiosulfate. It was decided to lower the mineral addition such that upon 100% copper dissolution, the copper concentration would be 1 g/L. In the remainder of this chapter this w i l l be referred to as a 1 g/L copper addition. 81 From Figure 4.10 it can be observed that enargite barely dissolves in ammonium thiosulfate under the applied experimental conditions. The effect of a low copper concentration in solution on thiosulfate degradation can be clearly observed from this experiment. The copper that dissolves is probably converted to cupric in the presence of air, but the concentration is too low to significantly affect thiosulfate stability. The p H of the solution changes only slightly and only small amounts of tetrathionate and sulfate are detected. 4.3.2 Effect of aeration In these experiments no forced aeration of the solution was applied, however, the vessel was not sealed airtight. It can be assumed that air was able to enter the reactor and the leach slurry, especially since a high stirring speed was applied (see section 3.2). Comparing Figure 4.8 and Figure 4.13 clearly illustrates the effect of aeration on thiosulfate degradation. Figure 4.13 shows that the thiosulfate concentration remains high; degradation does not occur appreciably. This is confirmed by the constant p H value. Because the system is not aerated the potential of the solution is lower. The erratic nature of the potential indicates that the system has not yet stabilized. The copper extraction shows a steady increase but remains fairly low. Compared to the leaching of chalcocite in an aerated system the final copper extractions do not differ significantly. It appears that the thiosulfate concentration in solution does not affect the final copper concentration in solution. The erratic behaviour of the sulfur species is explained by sample degradation. The leaching of cuprite in a non-aerated system is shown in Figure 4.14. Thiosulfate remains in 82 X a 12 10 8 6 4 2 0 + + 600 400 = 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH - © - E h 25 100 200 Time (minutes) 300 50 40 a 30 | « 20 » at a a 10 ^ 400 Thiosulfate - » - Tetrathionate - A - Sulfate -©— Copper Figure 4.13 Chalcocite, 5 g/L copper, 0.20 M (NH4)2S203, no aeration, 35°C 83 12 10 8 a 6 4 2 4 0 tee e e i-e—e-e—e—ee ^ ^ ' fo$^-0-<^ ^ ....fo ^ -I h H h 600 400 x s 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) - e - p H - • - E h V5 3 CZ3 25 20 15 10 100 200 Time (minutes) 300 100 80 60 | s-40 20 a a o U 400 -•— Thiosulfate —•— Tetrathionate —A— Sulfate —e— Copper Figure 4.14 Cuprite, 5 g/L copper, 0.20 M (NH4)2S203, no aeration, 35°C 84 solution. The low thiosulfate concentrations in the first half of the experiment are explained by sample degradation. The later samples are more representative of the true leaching conditions, because the length of time between sampling and analyses was less for these samples. A s for copper extraction, it can be observed that the initial extraction is just as high as with aeration. The major difference is that, without aeration, the copper extraction remains high throughout the course of the experiment. This indicates that the decline in copper concentration presented in Figure 4.11 is due to the degradation of thiosulfate. The colour of the solution is light blue during the experiment, indicating copper is complexed by both thiosulfate and ammonia. The stable E h and p H of the solution reflect the minimal thiosulfate degradation. The E h shows a small increase directly after the start of the experiment, indicating the quick dissolution of cuprite. From this study, it can be concluded that both air and cupric ion concentration play an important role in thiosulfate degradation. With copper in solution as cupric, thiosulfate degradation occurs readily and the cupric ion is converted to cuprous (Equation IV.8). The cuprous ions in solution w i l l be easily re-oxidized by air to cupric ions resulting in a fast rate of thiosulfate degradation (Equation IV.9). The overall reaction is represented by Equation (IV. 10). When no aeration is applied, the cuprous ion is not as readily oxidized to the cupric ion, resulting in greater thiosulfate stability. 4 C u ( N H 3 ) 2 + + 1 6 S 2 0 3 " -> 4 C u ( S 2 0 3 ) 5 3 ' + 1 6 N H 3 + 2S 4 0 2 6 " (IV.9) 4 C u ( S 2 0 3 ) ^ + 0 2 + 4 N H 4 + + 1 2 N H 3 -> 4 C u ( N H 3 ) 2 + + 2 H 2 0 + 1 2 S 2 0 2 " (IV.8) 85 Overall reaction: 4 S 2 0 ^ + 4 N H ; + 0 2 -> 2S 4 0g" + 4 N H 3 + 2 H 2 0 (IV. 10) 4.3.3 Summary In general it can be observed that the leaching of copper sulfide minerals in an ammonium thiosulfate solution with and without aeration, results in low copper extractions, around 0.5 to 1 g/L copper. Chalcopyrite and enargite seem to be unreactive toward ammonium thiosulfate leaching. The copper oxide minerals show high initial copper extractions, followed by a decline in extraction in the case of aeration. The copper extractions remain high without aeration. The Eh-pH diagram of the copper-thiosulfate-ammonia-water system presented in Figure 4.4 seems to give a more accurate representation of the actual solution conditions, then the Eh-pH diagram given in Figure 4.3. The following effects were observed regarding thiosulfate degradation: • A combination of aeration and copper, whereof a significant amount w i l l be present as the cupric ion in the presence of air, results in fast thiosulfate degradation. • In the absence of aeration, even a substantial amount of copper in solution (~ 1 to 4 g/L) does not constitute a detrimental effect on thiosulfate stability. • In the presence of a small amount of copper (-0.1 g/L) and aeration, thiosulfate remains longer in solution (leaching of enargite (Figure 4.10)). • The continuing occurrence of thiosulfate degradation in samples has to be taken into consideration when interpreting the ion chromatograph analyses. 86 4.4 Leaching of copper minerals with gold addition In this set of experiments, the copper minerals were leached with a 5 ppm gold addition to solution (see Table 3.E). The effect of the different variables on the leach reaction are only investigated with regard to chalcocite and cuprite. For the next two sections the following comment regarding the gold extraction curves is highlighted. Three solution samples were analyzed for gold content per experiment (30, 120 and 360 minutes). In the figures lines are drawn through those three sample points. This is to indicate a trend; it is not intended to imply that the gold extraction actually follows this curve. For a few experiments more samples were analyzed, resulting in a more accurate understanding of the behaviour of gold. A l l % copper extraction calculations are based on the head assays of the copper minerals. Gold extractions (%) are based on the 5 ppm gold addition. 4.4.1 Baseline experiments Figure 4.15 shows the leaching of covellite at 1 g/L copper addition. Comparing this figure to Figure 4.7, the leaching of covellite at 5 g/L without gold, a few differences can be observed. Firstly, the p H during heating (at "negative"time) is lower. This is due to the addition of the gold to the solution (as chloride complex in 10% hydrochloric acid, see section 3.1.3). The p H of the solution was immediately raised to p H 10 with concentrated sodium hydroxide after the ammonium thiosulfate addition, to avoid any hydrolysis of the ammonium thiosulfate at the low p H (see section 2.1.3). 87 X cu 12 10 8 + 4 + a A 800 600 400 f X! 200 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) p H 25 100 200 Time (minutes) 300 4- 80 100 -o o fl fl o o 7— « s « . O ft " a. o U 40 20 400 Thiosulfate - a - Tetrathionate Sulfate Copper Gold Figure 4.15 Covellite, 5 ppm gold, 1 g/L copper, 0.20 M (NH4)2S203, aeration, 35°C 88 The p H follows the pattern described previously. Since the thiosulfate is still present at the end of the experiment, the reduction of oxygen is the dominant reaction which produces hydroxide ions resulting in a steady increase in p H . The potential of the solution remains fairly stable, and at the same value as the potential in Figure 4.7 before the thiosulfate disappeared. The total amount of copper extracted is lower in the experiment with 1 g/L addition compared to the 5 g/L addition (0.15 g/L copper vs. 0.85 g/L copper). It seems that the copper extraction is independent of thiosulfate or ammonia availability. Further, it can be noted that the gold remains in solution during the leaching of covellite. The bottom graph illustrates the effect of lower copper concentration on thiosulfate degradation. Comparing the detected thiosulfate concentration with the concentration plotted in Figure 4.7 it can be concluded that the thiosulfate degrades at a slower rate when less copper is present in solution. Thiosulfate is still detected after 6 hours with lower copper concentration. This effect has already been discussed in the leaching of enargite. The leaching of chalcocite in the presence of gold shows the same patterns for p H , Eh, thiosulfate, tetrathionate and copper compared to the leaching of chalcocite without gold addition (Figure 4.16). The high sulfate concentration and its subsequent decline can be attributed to sample degradation. The behaviour of the gold in solution is quite different from the previous discussed experiment (Figure 4.15). The line is dashed to indicate that it is unknown how the gold extraction behaves between the points. More samples were not available for analysis. 89 2 + 600 400 an 200 a 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH -0—Eh .2-a 1X1 £3 "3 25 100 200 300 Time (minutes) 100 2 &n 80 e e o o * £ « s >- "S a o u Thiosulfate -•>- Tetrathionate —A- Sulfate Copper Gold Figure 4.17 Chalcocite, 5 ppm gold, 1 g/L copper, 0.20 M (NH4)2S203, aeration, 35°C 92 a 12 10 8 6 4 800 600 s 400 f 200 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) - * - p H E h a Vi SB ~5 t /5 25 100 200 300 Time (minutes) 100 -o o Ml 80 fl s o o 60 ~ « N + 40 a * 20 a U 400 Thiosulfate • Tetrathionate —A— Sulfate -#— Copper Gold Figure 4.18 Chalcopyrite, 5 ppm gold, 1 g/L copper, 0.20 M (NH4)2S203, aeration, 35°C 93 a 12 10 8 6 4 2 0 + + 800 600 s 400 f 4 200 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH - • - E h s CZ2 25 20 .8 1 5 o a « a o U 40 20 400 • Thiosulfate Tetrathionate —A— Sulfate Copper Gold Figure 4.19 Enargite, 5 ppm gold, 1 g/L copper, 0.20 M (NH4)2S203, aeration, 35°C 94 X c 12 10 8 6 4 2 0 —• •—c-o >-0-0—0——0— + + 600 400 x SO 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH E h 25 ^ 20 J 1 5 a GB S-.3 3 ! /3 10 -0-100 -o 80 o I I 3 3 © O 60 ~ ~ * ^ « 3 (U o a « a o U 4- 40 20 100 200 Time (minutes) 300 400 -•— Thiosulfate Tetrathionate Copper Gold Figure 4.20 Cuprite, 5 ppm gold, 5 g/L copper, 0.20 M (NH4)2S203, aeration, 35°C 95 Sample degradation is also visible in the leaching of malachite (Figure 4.21). The very low initial thiosulfate concentration, its increase later, the high tetrathionate, sulfate and copper concentration indicate sample degradation. Gold is not affected and remains in solution. It has to be commented that the overall copper extraction for cuprite and malachite is lower with the 1 g/L copper addition than the 5 g/L copper addition. It can be seen for chalcocite, cuprite and malachite that there is no thiosulfate left in solution after 180 to 240 minutes, this opposed to the covellite, chalcopyrite and enargite where thiosulfate remains present for 6 hours. The experiments with the first three minerals showed gold precipitation, whereas gold remains in solution during the experiments of the latter three minerals. 4.4.2 Effect of aeration Experiments without aeration were performed for chalcocite (Figure 4.22) and cuprite (Figure 4.23). The same conditions apply as described in section 4.3.2. Chalcocite shows a stable p H and potential, which is confirmed by the thiosulfate concentration which barely changes. The resulting copper extraction is lower than previous experiments. This is probably due to the low potential, which is lower than previous experiments with chalcocite (~50 m V S H E vs. 2 5 0 m V S H E ) . A s opposed to the experiment with aeration, gold remains in solution during the whole experiment. 96 12 10 8 6 4 2 0 s—Ad (Jim) 1000 800 600 1 400 200 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH - • - E h > w o a "3 25 20 15 100 200 300 Time (minutes) 100 -a 80 60 40 20 o a c .2 .2 « "5 a « a. o U 400 -•— Thiosulfate •Tetrathionate —A—Sulfate —©—Copper Gold Figure 4.21 Malachite, 5 ppm gold, 5 g/L copper, 0.20 M (NH4)2S203, aeration, 35°C 97 a 12 10 8 6 4 2 #0#—#0 ^0-0#$0-0~0>^0-O-1 1 1 1 1— —0 "0^^ 800 600 400 CO > 200 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) p H - • - E h CO '3 cu a CO U i 25 20 15 10 —I 1 100 200 Time (minutes) 300 100 -a 80 400 o e c o o 60 i * 40 si i 3 o a 0 8 20 a U • Thiosulfate Tetrathionate • Copper Gold Figure 4.22 Chalcocite, 5 ppm gold, 5 g/L copper, 0.20 M (NH4)2S203, no aeration, 35°C 98 o. 12 10 8 + 6 4 2 4-0 -0 0 0 600 400 Q s 200 W 0 1 1 1 1 1 1 1 1 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH 50 "3 a u a "3 25 20 15 10 5 0 » — • — • --Hh 100 200 Time (minutes) 300 100 2 o OX) 80 and g s 60 tractic w tractic 40 Copper exi )lubi 20 Copper exi so 400 Thiosulfate -a—Tetrathionate -A—Sulfate -#— Copper Gold Figure 4.23 Cuprite, 5 ppm gold, 1 g/L copper, 0.20 M (NH4)2S203, no aeration, 35°C 99 The sulfur analysis of Figure 4.23 shows high tetrathionate and sulfate concentrations and a thiosulfate concentration which increases after 180 minutes. This trend indicates sample degradation of the first 5 samples. Copper extraction is high (1 g/L copper addition), however, the copper extraction of cuprite at 5 g/L with aeration is higher. It was observed that copper and gold extraction declined with time when aeration was applied. Figure 4.23 shows that copper and gold remain in solution without aeration. The above results indicate that the retention of gold in solution is ensured by the presence of thiosulfate. 4.4.3 Effect of temperature Two experiments were performed at 20°C to investigate the effect of temperature on thiosulfate stability. Comparing the figures representing the leaching of chalcocite and cuprite at 35°C (Figures 4.17 and 4.20 ) with those obtained at 20°C (Figures 4.24. and 4.25), it can be observed that the thiosulfate degradation is slightly delayed in the case of chalcocite. The effect is barely noticeable for the leaching of cuprite. The ion chromatograph analysis for cuprite suggest the occurrence of sample degradation for the initial samples indicated by the high tetrathionate and sulfate concentrations. The recorded p H and potential show that the patterns for both minerals are similar to those obtained during leaching at 35°C, only slightly delayed. The behaviour of gold and the copper extraction follow the same pattern at 20 and 35° for chalcocite. For cuprite, the results cannot be compared since the mineral feed was not the same (5 g/L copper vs. 1 g/L copper). 100 x 12 10 8 6 4 2 0 + 600 400 -= 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH - ^ E h 100 -o 80 o 6D O O 60 ~ ~ | | 40 i | S © a » 20 §" U 100 200 Time (minutes) 300 400 • Thiosulfate —•— Tetrathionate Sulfate Copper Gold Figure 4.24 Chalcocite, 5 ppm gold, 5 g/L copper, 0.20 M (NH4)2S203, aeration, 20°C 101 12 10 8 + 6 4 2 0 + 600 400 s s 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) p H - © - E h CO CU w 4- 200 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH -0—Eh .a *o «u c in U "3 C/2 25 20 15 10 5 4 100 200 300 Time (minutes) 100 -o 80 o li a a © o 60 ~ ~40 20 2 -a fl s cu © a « a o u 400 -•— Thiosulfate Tetrathionate -A—Sulfate -# -Coppe r Gold Figure 4.26 Chalcocite, 5 ppm gold, 5 g/L copper, 0.20 M (NH4)2S203, 0.05 M S0 3 2 , aeration, 35°C 104 12 10 8 6 4 2 0 800 600 fad £ 400 | 200 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) p H -0—Eh -J w o .3 CZ3 25 20 4-15 10 100 200 300 Time (minutes) 100 -a 80 40 20 400 © ss a o o 60 « 2 -a * 3 » "s 4> O a « a o U • Thiosulfate -m- Tetrathionate Sulfate Copper Gold Figure 4.27 Chalcocite, 5 ppm gold, 5 g/L copper, 0.20 M (NH4)2S203, 0.05 M SO aeration, 35°C 4 1 105 a 12 10 8 6 4 600 400 w X > E 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) p H E h "5 0 2 25 20 .a 1 5 ' a 10 5 0 100 200 300 Time (minutes) 100 -o 80 © c c o o 60 ~ ~ O CS » -s >- ~x o © a * a o u 4- 40 20 400 • Thiosulfate Tetrathionate Sulfate Copper Gold Figure 4.28 Chalcocite, 5 ppm gold, 5 g/L copper, 0.20 M (NH4)2S203, 0.4 M NH 3 , aeration, 35°C 106 4.4.5 Gold precipitation A t the instant that thiosulfate is no longer present in solution, the following reaction is driven to the left and gold precipitates, unless it is complexed by something else in solution: 2 A u + 4S 203~ + y 0 2 + H 2 0 <-> A u ( S 2 0 3 ) 2 " + 2 0 H " ( I V . l l ) Thermodynamically, ammonia should be able to complex the gold, but that does not seem to occur. The discussion in section 4.1 already indicated that it seems that Figure 4.1 represents more accurately the actual solution conditions instead of Figure 4.2. However, it can be noticed that in several cases the gold extraction does not completely go to zero, although there is no thiosulfate present in solution. Several explanations can be given for this phenomena: • Ammonia complexes gold to a certain extent. The literature reports 10 to 20% gold extraction in the presence of ammonia and cupric ions [Zipperian et al, 1988; Cao et al, 1992]. • Gold precipitates as colloidal gold, and is therefore detected in solution. • There is still a small amount of thiosulfate left in solution not detected by ion chromatographic analysis, providing enough background for gold to remain in solution to a certain extent The presence of a background concentration of thiosulfate is confirmed by ion chromatograph measurements (Figure 4.29). Typical dilution factor for all experiments was 1000 times. This dilution factor ensured that the samples could be measured within the calibration range (2-20 ppm, see Appendix B) . A disadvantage was that concentrations below 1-2 ppm were not detected. This implies that, with a 1000 times dilution factor, concentrations below 1-2 g/L 107 thiosulfate could still be present in solution. Figure 4.29 shows the measurement of a sample at 1000, 500 and 50 times dilution. At 1000 and 500 times no thiosulfate is detected, whereas a small thiosulfate concentration is detected at 50 times dilution. This indictates that about 0.1 g/L of thiosulfate is in solution, providing enough background for gold to remain in solution. More research has to be performed to determine the nature of the gold precipitates. lOOOx dilution 0 0.5 1 1.5 2 2.5 3 Minutes 500x dilution A U Mi 0.5 1 1.5 2 2.5 3 Minutes 50x dilution 3.5 4 0 0J 1 1.5 2 2.5 3 3.5 4 Minutes Figure 4.29 Effect of dilution factor on detected thiosulfate concentration (retention time thiosulfate is 1.9 minutes) 4.4.6 Summary It can be concluded that in the presence of chalcocite, cuprite and malachite gold precipitates out of solution, while gold remains in solution in the presence of covellite, chalcopyrite and enargite. This seems to correlate with the fast degradation of thiosulfate observed during the leaching of 108 the first three minerals. High copper concentrations were achieved during the leaching of those three minerals, thus increasing the presence of cupric ion in solution, which enhances the degradation of thiosulfate. The copper extractions achieved during the baseline experiments of this series and the series discussed in section 4.3 are summarized in Table 4 .A. It is evident from Table 4 .A that the copper extractions did not improve by lowering the mineral feed addition and thus increasing lixiviant availability. Table 4.A Copper extractions (%) of copper minerals at different times during baseline experiments. Mineral % Copper leached at time (in minutes) Feed addition Leaching of copper minerals Feed addition Leaching of copper minerals + gold addition 30 120 360 30 120 360 Covellite 5 g/L C u 5.8 6.7 16.5 1 g/L C u 8.8 14.3 15.1 Chalcocite 5 g/L C u 26.0 17.8 24.0 5 g/L C u 29.5 26.3 18.3 Chalcocite - - - - 1 g/L C u 19.7 35.7 46.4 Chalcopyrite 5 g/L C u 2.7 5.1 8.9 1 g/L C u 3.78 9.9 14.5 Enargite 1 g/L C u 3.8 6.3 9.2 1 g/L C u 3.19 5.7 9.2 Cuprite 5 g/L Cu , 79.1 36.2 22.8 5 g/L Cu , 80.9 32.2 24.4 Malachite 5 g/L C u 64.8 62.5 44.5 1 g/L C u 87.0 86.1 83.5 N o beneficial effect o f lower temperature, sulfite and sulfate addition on thiosulfate stability was observed. Lower copper extractions with 1 g/L copper addition were observed compared to the copper extractions achieved with 5 g/L copper addition. This cannot be attributed to an ammonia deficiency since an extra addition of ammonia did not improve copper extractions. It may be 109 that the copper minerals are not provided with sufficient oxidizing agents and/or the leach temperature is too low, resulting in poor copper extractions. Regarding the use of the Eh-pH diagrams given in Figure 4.1 to Figure 4.4 to explain the experimental results presented in section 4.3 and 4.4. the following can be concluded: • For the gold-thiosulfate-ammonia-water system, Figure 4.1 (using the free energy value of -518.8 kJ/mol for thiosulfate) seems to give a more accurate representation of the actual leaching conditions • For the copper-thiosulfate-ammonia-water system, Figure 4.4 (using the free energy value of -532.2 kJ/mol for thiosulfate) seems to explain the experimental results better. More fundamental research is necessary to investigate this phenomenon. 4.5 Leaching of copper-gold samples The experimental results of the copper-gold samples are discussed per copper-gold sample, to directly observe the influence of the different variables (Table 3.F). In this series baseline experiments were performed and the effects of aeration and cupric addition were studied. The gold extraction calculations are based on calculated head, the copper extraction calculations on solids head assay. The cyanide consumption and copper and gold extraction of the copper-gold materials were determined by a 24 hour cyanide leach at a constant 1 g/L cyanide level. The experimental conditions are described in Appendix E . The results are summarized in Table 4.B. 110 Table 4.B Results of 24 hour cyanide leach of the copper-gold samples Copper-gold samples Gold extraction (%) Copper extraction (%) Cyanide consumption (kg/t ore) Overall Lobo Composite 81.4 41.4 4.38 Guanaco Composite 84.3 70.4 6.77 M 4 0 Pyrite Feed 94.7 34.9 6.40 M 1 0 Pyrite Concentrate 47.5 28.3 9.91 M 1 0 Pyrite Concentrate* 81.2 50.3 n.a. *Data obtained from Newcrest Mining Ltd. [Oretest, 1997] Cyanide leaching of all copper-gold samples resulted in high cyanide consumption, and consequently a high reagent cost must be anticipated when treating these samples with cyanide. It can be seen that a substantial amount of the copper in all samples is cyanide soluble. Two test numbers are listed for the M 1 0 Pyrite Concentrate. The first was obtained during the leach described in Appendix E , the second was provided by Newcrest Min ing Ltd ; a 24 hour cyanidation at an initial cyanide concentration of 1.25 g/L cyanide, maintained at 0.75 g/L cyanide. The first numberis believed to be low because of the following reaction: C u ( C N ) 2 " -> Cu(CN)" + CIST (IV. 12) which indicated the presence of free cyanide, while in reality it was complexed with copper. 4.5.1 Overall Lobo Composite Figure 4.30 gives the results of the leaching of the Overall Lobo Composite with a 1 g/L cupric ion addition in an aerated system. Similar patterns as described in the previous sections are 111 observed for the p H and Eh. The p H initially increases, and starts decreasing around 180 minutes which coincides with an increase in Eh. The initial Eh , during heating of the solution without any reagents present, shows a sharp decrease. Apparently, a component is present in the ore which strongly consumes oxygen. The line representing the thiosulfate concentration is dashed, the 180 minutes sample was not analyzed for sulfur species. The line implies that the thiosulfate concentration is about 5 g/L after 180 minutes and zero around 240 minutes. However, the p H and E h patterns indicate that most thiosulfate was gone after 180 minutes. The erratic pattern of tetrathionate is attributed to sample degradation. Gold extractions are stable around 40%. Initially copper is high which can be explained by the l g / L cupric addition. The decline in copper extraction with time can be attributed to the disappearance of thiosulfate, a pattern that has been observed previously. The experiment was conducted with coarse material (P 8 0 = 253 um) and the aeration of the system did not work properly (the sparger plugged). The experiment was duplicated with reground material (P 8 0 =141 um) and the results are presented in Figure 4.31. Most patterns are comparable to those of Figure 4.30. The thiosulfate concentration goes to zero after 180 minutes and the tetrathionate, sulfate and copper values are resembling those of Figure 4.30. Two differences can be observed. The E h value shows a steady increase during the second half of the experiment, and the gold extraction is lower during the second experiment. 112 12 10 8 a 6 4 2 0 600 + 400 200 a 0 ffi U c« -200 -400 -50 0 50 100 150 200 250 300 350 400 Time (minutes) -«— p H E h cu CU a CO t /5 25 20 15 10 • - . 80 60 £ c 40 g CU "8 20 2 100 200 Time (minutes) 300 400 • - - Thiosulfate -®— Copper -a— Tetrathionate — Gold - A — Sulfate Figure 4.30 Overall Lobo Composite, 0.20 M (NH4)2S203,1 g/L cupric addition, aeration, 35°C 113 X a 12 10 8 6 4 2 600 400 X s 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH E h C u s i t /5 25 20 15 4 100 200 Time (minutes) 300 - H -80 60 £ Is © 40 2 es 20 400 -*>— Thiosulfate •Tetrathionate -A—Sulfate - e - C o p p e r Gold Figure 4.31 Overall Lobo Composite, 0.20 M (NH4)2S203,1 g/L cupric addition, aeration, 35°C, reground. 114 12 10 8 6 4 2 + 0 - • 0 » # • — 0-600 400 Q s > a, 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) Vi a u £ "a 0 3 25 20 15 10 + 5 0 0 w=^k ^ A = - a ^ = ^ 100 200 Time (minutes) 300 80 60 £ o 40 2 OS 20 2 400 • Thiosulfate -m- Tetrathionate - A — Sulfate Copper Gold Figure 4.32 Overall Lobo Composite, 0.20 M (NH4)2S203,1 g/L cupric addition, no aeration, 35°C 115 12 10 8 a 6 c. 4 2 0 e • H h H h 600 400 Ld X 200 ^ -200 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH -0—Eh s 1X1 °3 a a "3 25 20 15 10 + 5 0 ® fjh ill"* 100 200 Time (minutes) 300 80 60 40 20 es 400 -•— Thiosulfate —•— Tetrathionate —A— Sulfate —•— Copper Gold Figure 4.33 Overall Lobo Composite, 0.20 M (NH 4) 2S 20 3, no cupric addition, aeration, 35°C 116 12 10 8 S 6 4 2 0 > • — a — m m i a 0-—000 0 0Q0 0 0-0-0-0 0, ...^ 600 400 K > = 200 W 0 1 1 1 1 1 1 1 1 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH - © - E h 25 ^ 2 0 -J J 15 "3 CU a CO .3 s E -200 -400 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH - • - E h 25 20 4 6X1 °3 « 2 10 5 0 s (Z3 100 200 Time (minutes) 300 80 60 £ o 40 5 8 s. 20 2 400 Thiosulfate • Tetrathionate —A— Sulfate Copper Gold Figure 4.35 Guanaco Composite, 0.20 M (NH4)2S203,1 g/L cupric addition, aeration, 35°C 121 Figure 4.36 Guanaco Composite, 0.20 M (NH4)2S203,1 g/L cupric addition, no aeration, 35°C 122 0 0 - 0 0 — 0 0 600 400 U J X 200 | -200 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH - 0 — E h 25 20 4-15 cu w CU a : i o 3 « 5 100 200 Time (minutes) 300 80 60 £ la © 40 2 CU 20 £ 400 • Thiosulfate Tetrathionate - A — Sulfate Copper Gold Figure 4.37 Guanaco Composite, 0.20 M (NH4)2S20 3 , no cupric addition, aeration, 35°C 123 Aeration and no cupric ion addition resulted in thiosulfate degradation, as opposed to the result of the leaching of the Overall Lobo Composite (Figure 4.37). The copper concentration is higher in this solution explaining the occurrence of thiosulfate degradation. However, compared to Figure 4.35, leaching of the composite in the presence of air and with cupric ion addition, the thiosulfate degradation is slightly delayed. This implies that the addition of cupric ions results in faster thiosulfate degradation. Copper extraction remains fairly constant and gold extraction decreases at the end of the experiment. When comparing the three figures it can be concluded that the same gold extractions are achieved every experiment, however, at different times. When aeration is applied, initial gold extractions of around 40% are achieved, which is the same as the final gold extraction of the experiment without aeration. However, the experiment without aeration implies that higher gold extractions can be achieved when the experiment is allowed to last longer, since thiosulfate is still present in solution. One experiment was performed with Thiogold™ grade ammonium thiosulfate (see section 3.1.3). The results are presented in Figure 4.38. The high initial gold extraction is due to an analytical error. The conclusion that can be drawn from this experiment is that the purity of the thiosulfate solution has no effect on the gold extraction. 124 a 12 10 8 6 i 4 2 0 N •*—*v 600 400 u. B 200 f -200 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH • - E h 25 ^ 20 !& 15 u a t 10 5 0 tZ5 - 4 - " ^ 100 200 Time (minutes) 300 120 100 80 60 40 20 0 Is o 400 — • — Thiosulfate — • — Tetrathionate —A— Sulfate — • — Copper Gold Figure 4.38 Guanaco Composite, 0.20 M (NH4)2S203 (Thiogold™ grade), 1 g/L cupric addition, aeration, 35°C 125 Summary Similar gold extractions are achieved under all conditions, however, at different times. Initial gold extractions are higher in an aerated system, while final extractions are higher in a non-aerated system. Copper extractions are low when thiosulfate is not present during the entire experiment. In the presence of thiosulfate higher copper extractions are achieved. L ike the Overall Lobo Composite there is no obvious beneficial effect observed on gold extraction with cupric ion addition. Table 4.D lists the gold and copper extractions of the experiments. Table 4.D Gold and copper extraction (%) of Guanaco Composite at different times. Experimental Conditions %Gold extracted at time (minutes): %Copper extracted at time (minutes): 30 120 360 Total* 30 120 360 Aeration 33.6 33.0 b.d. 0.0 47.0 34.9 32.5 N o aeration b.d. b.d. 39.6 44.0 9.0 28.7 59.2 Aeration, no cupric 51.6 50.6 24.6 30.7 30.4 31.5 24.7 Aeration, Thiogold™ 112.3 36.8 35.7 36.7 46.5 37.3 36.9 *Total represents total gold extraction, including wash water b.d. = below detection limit Compared to the cyanidation results (84.3% A u , 70.4 % C u at 24 hours cyanidation) it can be seen that the extractions achieved with ammonium thiosulfate are low. For this composite, the experiment with no forced aeration should be extended to a 24 hours leach, enabling a more accurate comparison with the cyanidation test results. 126 4.5.3 M40 Pyrite Feed A s can be seen in Figures 4.39 to 4.42, the experimental results of the M 4 0 Pyrite Feed generally follow the same patterns as observed previously. One additional experiment was performed in which no cupric was added to the solution and no aeration was applied (Figure 4.42). Both the experiments shown in Figure 4.41 and 4.42 were conducted at 0.24 M ammonium thiosulfate. Comparison of the experiments with aeration and cupric addition (Figure 4.39) and aeration without cupric addition (Figure 4.41) shows that the cupric addition accelerates the thiosulfate degradation. This is visible from the pH, E h and thiosulfate curves. The experiment without cupric addition shows a steady increase in copper extraction. Comparison of the experiments presented in Figure 4.40 and 4.42, both without aeration and one with cupric addition and one without, show that similar low gold extractions are achieved. The cupric addition does not seem to influence the leaching of gold. High tetrathionate concentrations were detected (Figure 4.40) and it is difficult to predict whether this occurred in the leach solution or was due to sample degradation. The low copper extraction in Figure 4.42 demonstrates that the copper in the M 4 0 Pyrite Feed, which is mainly chalcopyrite, is unreactive toward an ammonium thiosulfate leach. This corresponds to the results of the leaching of chalcopyrite (Figure 4.18). 127 Time (minutes) ••— Thiosulfate - •— Tetrathionate — A - Sulfate —@— Copper Gold Figure 4.39 M40 Pyrite Feed, 0.20 M (NH4)2S203,1 g/L cupric addition, aeration, 35°C 128 a 12 10 8 6 4 600 400 Q x CO 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) •pH - • - E h VI '3 s, 200 W 0 -50 0 50 100 150 200 250 300 350 400 Time (minutes) p H - © - E h 100 200 Time (minutes) 300 60 40 ^ _© *-+-» u 05 U 20 "oS 13. This indicates that i f trithionate formed, it is very likely that it remains for a while in solution. Two comments have to be made: 1. The determination of the response factor in the study quoted, was performed using different equipment. This study was only used to show that the concentration of trithionate might be higher than would be expected from the size of the peak area. 2. The trithionate peak areas were determined with the tetrathionate calibration curve. The trithionate peak areas were all below 1 mg/L, which is below the minimum detection limit, resulting in inaccurate area measurements. 144 Summary The sulfur balances for the leach experiments did not account for all the sulfur added to the solutions. Further research revealed that it is very likely that most of the sulfur ends up as the trithionate species in solution. 145 5 GENERAL SUMMARY AND CONCLUSIONS This study investigated the application of ammonium thiosulfate for the treatment of copper-gold ores. Leaching studies were conducted with copper minerals, copper minerals with gold addition and copper-gold samples of different copper and gold grade. On the basis of the experimental results, the following conclusions were drawn: Leaching of copper minerals in ammonium thiosulfate solution General observations • The copper sulfide minerals covellite and chalcocite do not leach appreciably in an ammonium thiosulfate solution. Chalcopyrite and enargite seem to be unreactive toward an ammonium thiosulfate leach. Copper extraction seems to be independent of the availability of complexing agents; leaching of less material in the presence of the same amount of lixiviant did not result in higher copper extractions. • The copper oxide minerals cuprite and malachite, showed fast dissolution resulting in high initial copper extractions (around 80%). The final copper extractions were dependent on lixiviant availability; copper extractions decreased with declining thiosulfate concentrations, high copper extractions were achieved when thiosulfate remained in solution. The exact role of ammonia in copper solubilization is not clear. It can be further remarked that 80% copper extraction was achieved independent of the amount of material subjected to the leach; both mineral additions of 5 g/L copper and 1 g/L copper resulted in 80% copper extraction. 146 • The amount of copper solubilized was in all cases, except enargite and chalcopyrite (both at a mineral feed of 1 g/L copper), enough to catalyze the degradation of thiosulfate in solution. After about 180 to 240 minutes thiosulfate was no longer present in solution. Because enargite and chalcopyrite barely leaches in an ammonia thiosulfate solution, little copper was solubilized which resulted in increased thiosulfate stability; the thiosulfate was still present after 360 minutes. It can therefore be concluded that the leaching of the copper minerals in the presence of air resulted in the solubilization of copper as cupric, which catalyzed thiosulfate degradation. Thiosulfate degradation reduces overall copper extraction. The following observations were made when soluble gold was added to the leach solution in the presence of copper minerals: • Copper concentrations of 0.3 g/L or higher result in enhanced rates of thiosulfate degradation. The consequent reduction of thiosulfate concentration causes gold to precipitate. • In the presence of low copper concentrations (around 0.1 g/L Cu) thiosulfate degradation is reduced and gold remains in solution. Further study should be undertaken to determine exactly at which copper concentration a negative effect on thiosulfate stability is observed. The above shows that gold solubilization is dependent on the presence of thiosulfate. Ammonia is known as a complexing agent for gold, but does not appear to play a role under the experimental conditions investigated. 147 Effect of aeration A s stated above, an aerated solution in combination with as little as 0.3 g/L copper in solution causes fast thiosulfate degradation, resulting in gold precipitation. Thiosulfate degradation is less in the absence of aeration, even in the presence of high copper concentrations. Cuprous ions are easily regenerated into cupric in the presence of air, resulting in fast thiosulfate degradation. The oxidation of cuprous to cupric is a lot slower in the absence of air, resulting in greater thiosulfate stability, and thus gold remains in solution. Effect of temperature A leach temperature of 20°C instead of 35°C did not result in a measurable effect on thiosulfate degradation, copper extraction and gold solubilization. Effect of reagent addition In this study, the addition of sulfite and sulfate was not observed to affect thiosulfate stability, as suggested in the literature. A n increase in ammonia concentration did not result in higher copper extractions, and the thiosulfate stability was negatively affected by the increase in ammonia concentration. Leaching of copper-gold samples in ammonium thiosulfate solution General observations These experiments confirmed the above conclusion: when the thiosulfate concentration declines, gold extraction decreases. Thiosulfate degradation depends on aeration and cupric ion 148 concentration. Gold extractions ranged from low to high depending on the experimental conditions. The copper-gold samples tested did not show appreciable co-extraction of copper during an ammonium thiosulfate leach. Compared to the cyanidation test results, the ammonium thiosulfate treatment of the Overall Lobo Composite and M 1 0 Pyrite Concentrate showed that similar gold extractions were achieved at lower reagent cost. Effect of aeration High initial gold extractions are achieved in a well aerated system. Without aeration, but not in an airtight system, gold dissolves a lot slower and low gold extractions are achieved after 6 hours of leaching. A n extended residence time of 24 hours for the M l 0 Pyrite Concentrate resulted in gold extractions of 83%. It can be concluded that a balance has to be found between providing enough oxidant for fast gold dissolution, and minimizing the amount of oxidant in the presence of cupric ions to prevent excessive thiosulfate degradation. The alternative is slow gold dissolution and low thiosulfate consumption in a non-aerated system with extended residence time. Effect of cupric ion addition A high cupric ion concentration in combination with aeration results in fast thiosulfate degradation. The leaching of the M10 Pyrite Concentrate and M 4 0 Pyrite Feed showed higher initial gold extraction with the addition of cupric ions in combination with aeration. This indicates a catalytic effect of cupric ions on the dissolution of gold in an ammonium thiosulfate solution, which is described in the literature (see section 2.3.3). In the absence of aeration, no 149 beneficial effect was observed from the addition of cupric ions. For the two composites (Overall Lobo and Guanaco), higher gold extractions were observed without the addition of cupric ions. This effect was not further investigated. Analytical • Part of the sulfur could not be accounted for in the leach solutions. Initial research indicates that a large portion of the sulfur initially present does not end up as sulfate, the thermodynamically predicted final degradation product of thiosulfate, but as trithionate (s3o62-). • It is very difficult to prevent early thiosulfate degradation in the samples, resulting in an inaccurate portrayal of the actual leaching conditions. Eh-pH diagrams • Different A G 0 values are used for the thiosulfate species (-518.8 kJ/mol and -532.2 kJ/mol), which result in a totally different representation of the thermodynamic equilibria in the ammonium thiosulfate system. 150 6 RECOMMENDATIONS This work investigated the application of ammonium thiosulfate for the treatment of copper-gold ores. From the previous chapter it can be concluded that the application of ammonium thiosulfate for the treatment of copper-gold ores offers definite opportunities; however, more research should be conducted in the following areas: Process development • High gold extractions were achieved with a 24 hour leach in a non-aerated system. Additional experiments should be conducted to optimize the gold extraction and at the same time minimize thiosulfate consumption. Factors to be studied to optimize the leaching are thiosulfate concentration and the effect of the solution p H . • High initial gold extractions were achieved in an aerated system. It might be valuable to investigate the possibility of separating solution and solids at that instant without causing thiosulfate degradation and thus gold precipitation. Further, the amount of aeration should be studied, to optimize gold extraction and minimize the thiosulfate degradation. • L o w copper extractions were achieved with the experimental conditions specified. More research should be performed to determine why low copper extractions were achieved. The copper species in solution should be more closely examined using U V spectrophotometry. This may reveal more information regarding the mechanism of copper complexation in solution; whether mainly copper thiosulfate or copper ammonia complexes form and which copper ammonia complexes form: the copper(I)ammonia or the copper(II)ammonia complex. • I f trithionate is the predominant degradation product of thiosulfate in this system, it should be researched how trithionate can be prevented from building up in the system. Furthermore, 151 trithionate may have a significant effect on adsorption processes for gold and copper recovery and these effects must be quantified. Fundamental Research • Fundamental research should be performed to provide more insight in the leaching chemistry. Because the chemistry of the ammonia thiosulfate system is not yet fully understood and has proven to be very complex, it is difficult to optimize and control the leaching conditions, especially on an industrial scale. • Efforts should be undertaken to obtain correct values for the free energies of the species present in ammonium thiosulfate solutions. This w i l l enable a more accurate representation of the thermodynamic equilibria. • More research is required to determine the mechanism of gold precipitation when thiosulfate is no longer present in solution. This is coupled with the question of whether or not ammonia plays a role in gold solubilization in this system. Analytical • Difficulties were encountered in preventing thiosulfate degradation from occurring in the solution samples for ion chromatography analyses. Sample treatment, with for example ion exchange resins, might offer a solution to this problem. • Moreover, it was only possible to detect the free species and not the complexed species by ion chromatography during this research. B y developing methods for sample treatment, it might be possible to quantify both the free and complexed species. 152 Not all sulfur species could be quantified by ion chromatography, e.g. sulfite and trithionate. Since it appears that trithionate plays an important role in the solution, methods should be developed for quantifying these sulfur species by ion chromatography. Furthermore, the response factor for trithionate has to be determined for the ion chromatograph set up used in this research. U V spectrophotometry should be applied to qualitatively determine which copper ammonia complexes are present in solution. 153 7 REFERENCES A B B R U Z Z E S E , C , Fornari, P., Massidda, R., Vegl io , Ubaldini, S., Thiosulfate leaching for gold hydrometallurgy, Hydrometallurgy, 39 (1995), pp. 265-276 A T L U R I , V . P . , Recovery of gold and silver from ammoniacal thiosulfate solutions containing copper by resin ion exchange method, Masters of Science thesis, University of Arizona, 1987 A V R A A M I D E S , J., Prospects for alternative leaching systems for gold " a review", in Carbon-in-pulp technology for the extraction of gold, Australas. Inst. M i n . Metall . Melbourne, Murdoch, Australia, July 1982, pp. 369-391 A W A D A L L A , F.T., Ritcey, G . M . , Recovery of gold from thiourea, thiocyanate, or thiosulfate solutions by reduction-precipitation with a stabilized form of sodium borohydride, Separation Science and Technology, 26 (9) 1991 pp. 1207- 1228 B A E S , C.F . , Mesmer, R . E . , The hydrolysis of cations, John Wiley and Sons, N e w York, 1976, pp. 267-285 B A R D , A . J . , Parsons, R., Jordan, J., Standard potentials in aqueous solution, International Union of Pure and Applied Chemistry, 1985 B E C K S T E A D , L . W . , Mil ler , J.D., Ammonia oxidation leaching of chalcopyrite, in: Chapman, T.W. , Tavlarides, L . L . , Hubred, G . L . , Wellek, R . M . (Eds.), Fundamental aspects of hydrometallurgical processes, A I C h E Symposium Series, No . 173, V o l . 74, pp. 28-40 B E R E Z O W S K Y , R . M . , Gormely, L .S . , Recovery of precious metals from metal sulfides, U S Patent 4,070,182, January 24, 1978 B E R E Z O W S K Y , R . M . G . S . , Sefton, V . B . , Recovery of gold and silver from oxidation leach residues by ammoniacal thiosulfate leaching, 108 t h A I M E Annual, Meeting, N e w Orleans, February 18-22, 1979 B U R N S , D.T. , Townshend, A . , Carter, A . H . , Inorganic reaction chemistry, vol.2., Reactions of the elements and their compounds, Part. B : Osmium to Zirconium, El l i s Horwood Limited, 1981, pp.421-423 154 B Y E R L E Y , J.J., Fouda, S.A., Rempel, G . L . , Kinetics and mechanism of the oxidation of thiosulfate ions by copper-II ions in aqueous ammonia solution. J. Chem. Soc. Dalton Trans. (1973): 889-893 C A O , C , Hu , J., Gong, Q., Leaching of gold by low concentration thiosulfate solution, Transactions of NonFerrous Society of China, v.2 (4), 21-25 Nov. 1992 D U B Y , P., The thermodynamic properties of aqueous inorganic copper systems, International Copper Research Association, Inc., U S A , 1977 F I S H E R , J.I., Halpern, J., Corrosion of copper-gold alloys by oxygen-containing solutions of ammonia and ammonium salts, Trans. Electrochem. Soc. 103, 282-286(1956) F O R W A R D , F . A . , Mackiw, V . N . , Chemistry of the Ammonia Pressure Process for Leaching N i , C u , and Co from Sherritt Gordon Sulfide Concentrates, Journal of Metals, 7, (1955), pp. 457-463 G A L L A G H E R , N .P . , Interaction of gold cyanide, thiocyanate, thiosulfate, and thiourea complexes with carbon matrices, Masters of Science Thesis, University of Nevada, Reno, 1987 G O N G , Q., Hu , J., Treatment of copper-containing gold sulfide concentrate using thiosulfate solution, Journal of engineering chemistry and metallurgy, 1990 (11)2, 145-151 H E M M A T I , M . , Hendrix, J .L. , Nelson, J .H. , Milosavljevic, E . B . , Study of the thiosulfate leaching of gold from carbonaceous ore and the quantitative determination of thiosulfate in the leached solution, Extraction Metallurgy '89, (Institution of Min ing and Metallurgy: London, 1989), pp. 665-678 H A L P E R N , J., Kinetics of the dissolution of copper in aqueous ammonia, J. Electrochem. Soc. 100, 421 (1953) H A N , K . , Meng, X . , Ammonia extraction of gold and silver from ores and other materials, U S Patent 5,114,687, M a y 19, 1992 H O N E A , R. M . , Polished section examination of P2509 Guanaco Composite, February 6, 1998 155 J E F F R E Y , G . H . , Bassett, J., Mendham, J., Derrney, R . C . , VogeFs textbook of quantitative chemical analysis, 5 t h Edition, Longman Scientific and Technical, Great Britain, 1989, p.490 J I A N G , T., Chen, J., X u , S., Electrochemistry and mechanism of leaching gold with ammoniacal thiosulfate, X V I I I Int. Mineral processing congres, 1993, Australia, V . 5., The australasian institute of mining & metallurgy, pp. 1141-1146 K E R L E Y , B . J . , Recovery of precious metals from difficult ores, U S Patent 4,269,622, M a y 26, 1981 K E R L E Y , B . J . , Recovery of precious metals from difficult ores, U S Patent 4,369,061, Jan. 18, 1983 K I R K O T H M E R , Encyclopedia of Chemical Technology, 3 r d Edition, John Wiley and Sons, New York, V o l . 22, 1983, pp. 974-988 K U H N , M . C . , Arbiter, N . , K l ing , H . , Anaconda's Arbiter process for copper, C I M Bulletin, (1) pp. 62-71, February, 1974 L A K E F I E L D R E S E A R C H L I M I T E D , method no. 9.1.9, Sulfate determination by ion chromatography, 1990 L A K E F I E L D R E S E A R C H L I M I T E D , method no. 9.1.14, Thiosulfate and tetrathionate determination by ion chromatography L A K E F I E L D R E S E A R C H L I M I T E D , Mineralogical Examination of Overall Lobo Composite, M a y 5, 1997 L A N G H A N S JR., J.W., L e i , K . P . V . L e i and Carnahan, T .G . , Copper-catalyzed thiosulfate leaching of low-grade gold ores. Hydrometallurgy 29 (1992) 191-203 L I , J., Mi l le r , J.D., Wan, R . Y . , Le Vier, M . , The ammoniacal thiosulfate system for precious metal recovery, in: Proceedings of the 19 t h International Mineral Processing Congress, San Francisco, 1995, Chapter 7 156 L I , J., Mi l le r , J.D., Wan, R . Y . , Important solution chemistry factors that influence the copper-catalyzed ammonium thiosulfate leaching of gold, presented at the 125 t h S M E Annual meeting Phoenix, Arizona, March 11-14, 1996. L I D D E L L , D . M . , Handbook of non-ferrous metallurgy, vol.2; Recovery of metals, N e w York, M c G r a w - H i l l , 1945, p. 1052 M A R C H B A N K , A . R . , Thomas, K . G . , Dreisinger, D . , Fleming, C. , Go ld recovery from refractory carbonaceous ores by pressure oxidation and thiosulfate leaching, U S Patent # 5,536,297, July 16, 1996 M A R S D E N , J. House, I., The chemistry of gold extraction, El l i s Horwood, N e w York, 1992 M C K E E , D . , Lulham, J.P., Separation process, Interational Publication # W O 91/11539, August 8, 1991 M E N G , X . , Han, K . N . , The dissolution behaviour of gold in ammoniacal solutions, in: Hydrometallurgy, Fundamentals, Technology and innovation, S M E , 1993, pp. 205-221 M I C H A E L I S , H . V O N . , Gold processing update, The prospects for alternative leach reagents, can precious metals producers get along without cyanide? Engineering & Min ing Journal , June 1987, pp. 42-47 O R E T E S T , Metallurgical testwork on Telfer M40 and M l 0 Flotation samples, August 18, 1997 P E R E Z , A . E . , Galaviz, H . D . , Method for recovery of precious metals from difficult ores with copper-ammonium thiosulfate., U S Patent # 4,654,078, March 31, 1987. P E T E R S , E . , Oxidation state diagrams, Short course notes P R Y O R , W . A . , The kinetics of the disproportionation o& sodium thiosulfate to sodium sulfide and sulfate, Analytical Chemistry, The American Chemical Society, 82, 1960, pp.4794-4797 S I L L E N , L . G . , Stability constants of metal-ion complexes, Supplement no . l . , Special publication No.25, The Chemical Society, Burlington House, London 157 S K O O G , D . A . , West, D . M . , Fundamentals of analytical chemistry, 3 edition, Holt, Reinhart and Winston, N Y , 1976, pp.363-366 S K O O G , D A . , Leary, J. J., Principals of instrumental analysis, 4 t h edition, 1992, pp.579-667 S M I T H , C .W. , Hitchen, A . , Aqueous solution chemistry of polythionates and thiosulfate: a review of formation and degradation pathways, C A N M E T , 1976 S T E U D E L , R., Holdt, G . , Ion-pair chromatographic separation of polythionates S n 0 6 2 " with up to thirteen sulfur atoms, Journal of Chromatography, 361, 1986, pp. 379 - 384 T O Z A W A , K . , Inui, Y . , Umetsu, Y . , Dissolution of gold in ammoniacal thiosulfate solution, T M S paper selection A81-25, 1981 T U R N B U L L , A . G . , Wadsley, M . W . , The C S I R O Thermochemistry system (version V ) , Institute of Energy and Earth Resources, Divis ion of Mineral Chemistry, Australia, 1986 W A N , R . . Y . , LaVier , K .Marc , Clayton, Richard, B . , ( Newmont Gold Co.) Hydrometallurgical process for the recovery of precious metal values from precious metal ores with thiosulfate lixiviant. United States Patent # 5,354,359, O c t . l l , 1994 W A N , R . Y . , Importance of solution chemistry for thiosulfate leaching of gold, unpublished, 1997. W A N G , X . , Thermodynamic Equilibrium calculations on A u / A g lixiviant systems relevant to gold extraction from complex ores, in: Electrochemistry in Mineral and Metal Processing III, St.Louis, Missouri , 17- 22 May 1992, The electrochemical society, Inc. ( U S A ) pp. 452-477, 1992 W A S S E R L A U F , M . , Dutrizac, J.E., The chemistry, generation and treatment of thiosalts in mil l ing effluents - a non-critical summary of Canmet investigations 1976-1982, Canmet report 82-4E, March 1982 W E A S T , R .C . , C R C Handbook of Chemistry and Physics, C R C Press, Ohio, 1975 158 W I L L I A M S , R . D . , Light, S.D., Copper concentrate dissolution chemistry and kinetics in an ammonia-oxygen environment in: Chapman, T.W. , Tavlarides, L . L . , Hubred, G . L . , Wellek, R . M . (Eds.), Fundamental aspects of hydrometallurgical processes, A I C h E Symposium Series, 1978, No.173, Vol .74, pp.21-27 W O O L F , A . A . , Anhydrous sodium thiosulfate as a primary iodometric standard, Analytical Chemistry, The American Chemical Society, 1982, 54, pp. 2134-2136 Z I P P E R I A N , D. , Raghavan, S., Gold and silver extraction by ammoniacal thiosulfate leaching from.a rhyolite ore. Hydrometallurgy, 19 (1988) 361-375 159 Appendix A: Screen analysis of copper-gold samples The particle size distributions of the Overall Lobo Composite and Guanaco Composite were determined by taking a 500 gram representative sample of the pulverized ore, and subjecting it to screening. Results are presented in Table A . I . Table A.I Screen analysis of copper-gold samples Size (um) Lobo Composite Guanaco Composite M40 Pyrite Feed M10 Pyrite Concentrate Mass(g) Mass % Mass(g) Mass % Mass(g) Mass % Mass(g) Mass % 300 0.38 0.08 0.83 0.17 27.87 1.41 - -212 3.16 0.64 19.04 3.85 54.09 2.73 8.30 0.42 150 73.38 14.85 99.42 20.10 74.86 3.78 17.20 0.87 106 142.02 28.74 81.27 16.43 94.37 4.76 35.90 1.81 75 61.62 12.47 75.83 15.33 150.94 7.61 87.50 4.41 45 113.6 22.99 86.2 17.43 330.52 16.67 256.00 12.90 -45 100.02 20.24 132.02 26.69 1250.13 63.05 1578.85 79.59 Totals 494.18 100.0 494.61 100.0 1982.78 100.0 1983.75 100.0 P80(um) 141 160 75 46 The particle size distributions of the M 4 0 Pyrite Feed and M 1 0 Pyrite Concentrate were obtained from Newcrest Min ing Ltd (Table A.I) . The P 8 0 o f the four materials was determined by plotting the cumulative weight percentage passing against the screen sizes (see Figure A . 1 to A.4) 160 100 . 3^ _ 90 80 70 3? § 60 '!> i? 50 Q-S 40 E O 30 . 20 10 . 0 ^ Size, m i c r o n s ^ 10 P 8 0 1000 Figure A.2 Particle size distribution of Guanaco Composite 161 mn _ 90 80 70 ? fin 'assir Um ^ 40 E O 30 20 10 0 10 P 8 0 1 ( )0 _. Size, m i c r o n s 1000 Figure A.3 Particle size distribution of M40 Pyrite Feed 100 90 80 70 if 60 'to in S. 50 ^ 40 i O 30 20 10 0 * — 10 8 0 100 Size, microns 1000 Figure A.4 Particle size distribution of M10 Pyrite Concentrate 162 Appendix B: Ion Chromatography Chromatography is an important method that permits the separation and analysis of ionic species. In this research ion exchange chromatography was used. A small amount of l iquid sample is injected into a moving stream of liquid (termed the mobile phase), that passes through an immiscible stationary phase (resins with ion exchange sites), which is fixed in a column or on a solid surface. Separation is based on ions partitioning into the ion exchange phase to varying degrees. This process is illustrated in Figure D . l . A t time t0, the sample, contained in the mobile phase, is introduced at the head of the column. The components of the sample distribute themselves between the two phases. Introduction of additional mobile phase (the eluent) forces the mobile phase containing a part of the sample down the column, where further partition between the mobile phase and fresh portions of the stationary phase occurs (time t,). Continuous flow of the mobile phase carries analyte molecules down the column in a continuous series of transfers between the mobile and the stationary phases. Because movement of sample components can only occur in the mobile phase, the average rate at which a species migrates depends upon the fraction of time it spends in that phase. For components that are strongly retained by the stationary phase (compound B in Figure D . l ) this fraction is small, whereas the fraction is large for components that are weakly retained by the stationary phase (component A ) . Ideally, the components are separated into bands (time t2), which can be detected at the end of the column (time t} and /,) [Skoog, D . A . , 1992]. 163 Ca) Sample m Packed column 2 1 Mobile phase J H I 4 JfllS 7 T • 41 '2 Time -Detector Figure D.l (a) Diagram showing the separation of a mixture of components A and B by column elution chromatography, (b) The output of the signal detector at the various stages of elution shown in (a) [Skoog, D.A., 1992] If the signal of the detector is plotted as a function of time, a series of peaks is obtained, as shown in the lower part of Figure D . l . This plot is called a chromatogram, and is a tool for both qualitative and quantitative analysis. The positions of peaks on the time axis may serve to identify the components of the sample; the areas under the peaks provide a quantitative measure of the amount of each component. This is achieved by calibrating using standards that contain known components of known concentrations. 164 Ion chromatography in this work Two methods were used in this work to measure the different sulfur species in solution. The IonPac A S 4 A - S C analytical column and IonPac A G 4 A - S C guard column combined with a conductivity detector and anion micro membrane suppressor were used for sulfate analyses. A n example of a chromatogram is given in Figure D.2. Figure D.2 Chromatogram of the A S 4 A column: sulfate (retention time of 4.7 minutes) A n OmniPac P A X - 1 0 0 analytical and Omni Pac- P A X - 1 0 0 guard column with a U V detector was used for thiosulfate and tetrathionate determination. A n example chromatogram is shown in Figure D.3. Detailed procedures for both columns are described on the next pages. 165 0.14 0.12 0.1 Figure D.3 Chromatogram of the Omnipax column: thiosulfate (retention time of 2 minutes) and tetrathionate (retention time of 5.1 minutes) 166 Determination of thiosulfate and tetrathionate (S 20 3 and S406) by Ion Chromatography [Lakefield Research Ltd.] Objective To determine thiosulfate and tetrathionate in solutions by ion chromatography Range: 1.0 mg/L to 20 mg/L (maximum detection limit), i f there are no interferences and no dilution factor. Concentrations higher than 20 mg/L should be diluted to below maximum detection limit. Interferences: High thiocyanate interferes with the thiosulfate peak Sample Requirements: The sample is not preserved, and should be analysed the same day as the experiment took place. Apparatus: 1. Liquid chromatograph with U V detector 2. OmniPac Pax-100 analytical column 3. OmniPac Pax-100 guard column 4. 1 m L syringe for sample injection 5. 0.45 um syringe filter 167 Reagents: 1. Sodium Perchlorate eluent (0.038 M ) with 10 % methanol: dissolve 10.675 g N a C 1 0 4 . H 2 0 ( H P L C grade) in DI water, add 200 m L methanol, and dilute to 2 L with DI water using a 2 L beaker. Filter the solution through a 0.45um filter. 2. Calibration standards (prepare fresh every day) Thiosulfate Prepare 1000 mg/L S 2 0 3 standard by dissolving 1.4117 g of Na 2 S 2 0 3 (anhydrous sodium thiosulfate) in DI water and dilute to 1000 m L in a volumetric flask Tetrathionate Prepare 1000 mg/L stock solution by dissolving 0.1368 g N a 2 S 4 0 6 . 2 H 2 0 ( s o d i u m tetrathionate dihydrate) in water and dilute to 100 m L in a volumetric flask Prepare from these 1000 mg/L stock solutions calibration standards containing 2, 10, 15 and 20 ppm thiosulfate and tetrathionate Procedure: 1. Set up the ion chromatograph according to the instructions in the ion chromatograph manual 2. Guard column -OmniPac P A X - 100, analytical column - OmniPac P A X - 1 0 0 3. Eluant: 0.038 M NaC10 4 in 10% methanol 4. F low rate: 1.1 mL/min 5. sample loop volume - 50 u L 6. Detector: U V , measure at 215 nm 7. Inject one of the calibration standards and run the chromatogram of thiosulfate and 168 tetrathionate to determine the retention times. Standardize the instrument on these conditions 8. Calibrate the machine with the 2, 10, 15 and 20 ppm calibration standards and check the calibration plot 9. If the calibration is successful, determine the chromatograms of the samples (use syringe filters for injection) Ion Retention Time* S 2 0 3 1.8 - 2.2 minutes S 4 0 6 4 - 6 minutes *Note: difference in retention time depends on condition of the column Quality Control: The following samples must be analyzed after an analytical run i . Replicate calibration standard: choose the calibration standard close to prevalent concentration i i . Replicate sample: select random a sample and perform a second analytical run Note: the anhydrous sodium thiosulfate can be prepared from hydrated sodium thiosulfate ( N a 2 S 2 0 3 . 5 H 2 0 ) via a procedure described by Wool f [1982]. It can also be purchased in a granular form. 169 Determination of sulfate by Ion Chromatography [Lakefield Research Ltd., 1990] Objective: To determine sulfate in solutions by ion chromatography Range: 0. 5 mg/L to 20 mg/L (maximum detection limit), i f there are no interferences and no dilution factor. Concentrations higher than 20 mg/L should be diluted to below maximum detection limit. Interferences: There can be overestimation of sulfate values with low p H samples Sample Requirements: The sample is not preserved, and should be analysed the same day as the experiment took place. Apparatus: 1. Liquid chromatograph with conductivity detector and anion micro membrane suppressor ( A M M S ) 2. IonPac A S 4 A - S C analytical column 3. IonPac A G 4 A - S C guard column 4. 1 m L syringe for sample injection 5. 0.45 um syringe filter 170 Reagents: 1. Sodium carbonate - sodium bicarbonate eluent with 10 % methanol: prepare 1.8 m M N a 2 C 0 3 + 1.7 m M N a H C 0 3 by weighing 0.382 g of N a 2 C 0 3 anhydrous and 0.286 g of N a H C 0 3 anhydrous into DI water. A d d 200 m L methanol, and dilute to 2 L with DI water using a 2L beaker. Filter the solution through a 0.45um filter. 2. Sulfuric acid (25 m N H 2 S 0 4 ) : 100 ml I N sulfuric acid into 2 L beaker, dilute to 2 L . 3. Calibration standards Sulfate Prepare 1000 mg/L S 0 4 standard by dissolving 1.4796 g of Na 2 S0 4 (anhydrous sodium sulfate) in DI water and dilute to 1000 m L in a volumetric flask. Prepare from this 1000 mg/L stock solution calibration standards containing 2, 10, 15 and 20 ppm sulfate Procedure: 1. Set up the ion chromatograph according to the instructions in the ion chromatograph manual 2. Guard column IonPac A G 4 A - S C , analytical column - IonPac A S 4 A - S C 3. Eluant: 1.8 m M N a 2 C 0 3 + 1.7 m M N a H C 0 3 4. F low rate: 2.0 mL/min 5. Sample loop volume - 50 u L 6. Suppressor: An ion Micro Membrane Suppressor ( A M M S ) Regenerant: 25 m N H 2 S 0 4 at flow rate 3-5 mL/min 7. Detector: conductivity 8. Inject one of the calibration standards and run the chromatogram of sulfate to determine the retention time. Standardize the instrument on these conditions 171 9. Calibrate the machine with the 2, 10, 15 and 20 ppm calibration standards and check the calibration plot 10. I f the calibration is successful, determine the chromatograms o f the samples (use syringe filters for injection). If the sample contains thiosulfate, the peak w i l l appear around 15-20 minutes on the chromatogram, depending on the condition of the column Ion Retention Time* S 0 4 4-6 minutes S 2 0 3 15-20 minutes *Note: difference in retention time depends on condition of the column Quality Control: The following samples must be analyzed after an analytical run i . Replicate calibration standard: choose the calibration standard close to prevalent concentration i i . Replicate sample: select random a sample and perform a second analytical run 172 Appendix C: Thermodynamic data Formula AG 0 , kJ/mol H 2 0 -237.18 OH" -157.29 N H 3 -26.6 N H 4 + -79.37 S 0 s2- 86.31 79.5 s32" 73.6 s42" 69 s52" 65.7 so 3 2- -486.6 so 4 2- -744.63 s 2o 3 2- -518.8 s 2o 3 2- -532.2*, *** S 20 4 2" -600.4 S A 2 - -791. S 20 6 2" -966. S 20 8 2" -1110.4 s 3o 6 2- -958. s 4o 6 2- -1022.2 s 5o 6 2- -956. H S ' 12.05 H 2 S (aq) -27.87 HS0 3 ' -527.81 H S C V -756.01 A u 0 Formula AG°,kJ/mol A u + 176. A u 3 + 440. Au0 3 3" -51.9 HAu0 3 2" -142. H 2 Au0 3 - -218. A u ( O H ) 3 (aq) -283.5 A u ( O H ) 3 (s) -317 A u ( N H 3 ) 2 + -41 1** Au(S 2 0 3 ) 2 3 " -1048.** C u 0 C u + 50.30 C u 2 + 65.70 C u 2 0 -148.10 CuO -134. Cu(OH) 2 (s) -359.50 HCu0 2" -258.90 Cu0 2 2" -183.90 C u ( N H 3 ) 2 + -63. C u ( N H 3 ) 2 + 15.60 C u ( N H 3 ) 2 2 + -30.50 C u ( N H 3 ) 3 2 + -73.20 C u ( N H 3 ) 4 2 + -111.50 Cu(S 2 0 3 ) 2 3 - -1083.66*** Cu(S 20 3) 3 5" -1623.39*** Cu 2 S -87.60 CuS -53.20 A l l data from Bard [1986], except * from C R C Handbook of Chemistry and Physics [1975] and ** from Atlur i [1987] and Wang [1992] and *** from Duby [1977] 173 Appendix D: Experimental data Mineral CuS Cu2S CuFeSj Cu3AsS4 CujO Baseline Baseline Baseline Baseline Baseline Grams added 7.92 9.28 17.56 4.31 12.85 Residue 6.06 7.78 15.87 3.97 11.29 NaOH added (g) 24.84 24.89 24.89 24.96 22.02 NaOH added (mL) 16.40 16.43 16.43 16.48 14.53 Colour samples Time (min.) = 5 colourless colourless colourless colourless l.blue 30 colourless l.blue colourless colourless blue 60 colourless blue colourless colourless blue 120 colourless blue colourless colourless blue 180 l.blue blue very l.blue colourless blue 240 blue blue l.blue colourless blue 360 blue blue blue blue Addition of NH3 to samples: only t240, 360 all except t5,30 only O40, 360 to none to all Colour filtrate blue d.blue blue v.l.blue d.blue Colour washwater colourless colourless colourless colourless colourless Ion chromatograph analyses g/L S-species g/L S-species g/L S-species g/L S-species g/L S-species Thiosulfate 5 19.83 15.48 21.55 6.72 30 15.59 10.04 19.95 22.69 .4.80 60 12.41 9.41 16.81 22.56 2.67 120 8.70 3.36 13.44 21.77 3.30 180 2.09 0.09 3.18 20.74 1.31 240 0.00 0.19 0.75 19.81 0.05 360 0.02 0.00 0.05 17.31 0.02 Tetrathionate 5 0.53 1.20 0.42 3.21 30 1.73 1.47 1.14 0.19 2.94 60 1.76 0.49 2.06 0.27 2.10 120 0.82 0.04 1.17 0.39 0.68 180 0.40 0.00 0.64 0.49 0.63 240 0.03 0.00 0.00 0.52 0.00 360 0.00 0.00 0.06 0.26 0.00 Sulfate 5 0.24 23.98 30 0.56 0.00 20.59 60 1.54 0.00 18.00 120 0.78 0.00 8.80 180 3.65 0.00 3.82 240 5.23 0.00 3.62 360 5.46 0.77 4.07 Sulfur balance at every sample time (thiosulfate, tetrathionate and sulfate) g/LS g/LS g/LS 5 12.65 13.69 30 12.25 13.09 11.30 60 11.31 13.06 8.74 120 8.62 12.67 5.22 180 3.41 12.14 2.38 240 2.17 11.63 1.24 360 1.89 10.31 1.37 Trithionate 360 5.51 - 7.28 Total Sulfur balance at t=360, incl.trithionate: 7.40 10.31 8.66 - = no trithionate present, or too small to measure Sample calculation: CuFeS2, baseline, t=5: 21.55*0.572 (% S in S 20 3) + 0.42*0.572 (% S in S4O6)-H).24*0.334 (% S in S04)=12.65 174 Mineral CuC0 3 Cu2S Cu 20 CuS+Au Cu2S+Au Baseline no aeration no aeration Baseline Baseline Grams added 10.48 9.28 12.85 1.58 9.28 Residue 5.33 7.42 6.73 1.24 7.82 NaOH added (g) 23.3 25.52 23.03 24.95 25.58 NaOH added (mL) 15.38 16.84 15.20 16.47 16.88 Colour samples Time (min.) = 5 l.blue colourless l.blue colourless colourless 30 blue colourless l.blue colourless l.blue 60 blue colourless l.blue colourless blue 120 blue colourless l.blue colourless blue 180 blue colourless l.blue colourless blue 240 blue colourless l.blue colourless blue 360 blue colourless l.blue v.l.blue blue Addition of NH3 to samples: to all to none to all none to all, except t5 Colour filtrate d.blue/purple v.l.blue l.blue l.blue v.dark blue Colour washwater colourless-l.blue colourless l.bl-green colourless l.blue Ion chromatograph analyses g/L S-species g/L S-species g/L S-species g/L S-species g/L S-species Thiosulfate 5 4.88 21.53 2.18 21.88 13.31 30 3.49 20.35 3.25 21.19 3.27 60 4.41 19.10 1.60 20.93 3.58 120 2.23 15.00 3.11 18.92 3.75 180 0.12 15.12 3.42 17.09 2.36 240 0.07 11.22 6.93 14.08 0.00 360 0.16 19.03 12.85 2.93 0.06 Tetrathionate 5 4.37 0.24 2.99 0.00 2.20 30 3.07 0.55 2.60 0.22 2.41 60 2.05 1.08 2.91 0.51 1.55 120 0.48 2.09 3.11 0.72 0.76 180 0.00 1.99 2.94 0.47 0.08 240 0.00 2.88 2.56 0.13 0.00 360 0.00 1.26 1.84 0.00 0.00 Sulfate 5 18.88 1.63 23.56 0.15 11.44 30 15.73 2.58 22.73 0.15 21.40 60 10.03 4.48 23.61 ' 0.15 16.26 120 4.41 12.77 22.26 0.15 9.03 180 2.93 13.97 22.61 0.15 5.01 240 3.15 17.59 19.27 0.15 5.40 360 3.79 6.99 17.85 1.72 6.07 Sulfur balance at every sample time (thiosulfate, tetrathionate and sulfate) g/LS g/LS g/LS g/LS g/LS 5 11.60 13.00 10.82 12.57 12.69 30 9.01 12.81 10.94 12.30 10.40 60 7.04 13.04 10.47 12.31 8.37 120 3.03 14.04 10.99 11.28 5.59 180 1.05 14.45 11.19 10.10 3.07 240 1.09 13.94 11.86 8.18 1.80 360 1.36 13.94 14.37 2.25 2.06 Trithionate 360 4.93 - - 3.73 8.34 Tot. S.balance at t=360: 6.29 13.94 14.37 5.98 10.40 - = no trithionate present, or too small to measure 175 Mineral Cu2S+Au CuFeS2+Au Cu3AsS4+Au Cu20+Au CuCQ3+Au Baseline Baseline Baseline Baseline Baseline Grams added 1.86 3.51 4.31 12.85 2.1 Residue 1.16 3.02 3.89 10.95 0.36 NaOH added (g) 25.6 24.2 30.19 22 24.22 NaOH added (mL) 16.90 15.97 19.93 14.52 15.99 Colour samples Time (min.) = 5 colourless colourless colourless blue v.l.blue 30 colourless colourless colourless blue v.l.blue 60 v. l.blue colourless colourless blue l.blue 120 l.blue colourless colourless blue blue 180 blue colourless colourless blue blue 240 blue colourless colourless blue blue 360 colourless blue blue Addition of NH3 to samples: to 240, 360 none to none all to 120, 180.240 Colour filtrate blue v.l.blue v.l.blue blue d.blue/purple Colour washwater colourless colourless colourless l.blue v.l.blue Ion chromatograph analyses g/L S-species g/L S-species g/L S-species g/L S-species g/L S-species Thiosulfate 5 22.47 0.54 0.00 30 23.68 22.19 22.17 1.95 0.17 60 16.36 22.96 21.71 2.33 2.01 120 8.04 21.10 20.76 2.69 2.77 180 1.98 19.18 19.76 0.24 0.53 240 0.00 18.09 18.49 0.07 0.00 360 0.00 12.68 15.75 0.04 0.00 Tetrathionate 5 0.00 2.87 3.11 30 2.32 0.28 0.12 2.69 3.02 60 . 1.16 0.59 0.22 1.83 2.07 120. 0.73 0.95 0.35 0.59 ' 0.66 180 0.00 1.10 0.40 . 0.00 0.00 240 0.00 1.01 0.38 0.00 0.00 360 0.00 0.54 0.11 0.00 0.00 Sulfate 5 30 0.17 0.00 0.66 11.93 60 0.38 0.27 0.65 4.71 120 1.35 0.69 3.23 180 2.39 0.03 0.73 3.59 240 5.07 0.07 0.76 4.07 360 5.70 1.04 0.91 4.65 Sulfur balance at every sample time (thiosulfate, tetrathionate and sulfate) g/LS g/LS g/LS g/LS 5 12.85 5.76 30 14.93 12.85 12.97 1.83 60 10.15 13.56 12.77 3.91 120 5.47 12.61 12.31 3.04 180 1.93 11.61 11.78 1.50 240 1.69 10.95 11.04 1.36 360 1.90 7.91 9.38 1.55 Trithionate 360 5.94 2.99 - 6.38 Tot. S.balance at t=360: 7.84 10.90 9.38 7.94 - = no trithionate present, or too small to measure 176 Mineral CujS+Au Cu20+Au CujS+Au Cu20+Au Cu2S+Au Cu2S+Au no aeration no aeration 20°C 20°C 0.4MNH3 Sulfite addn. Grams added 9.28 2.57 1.86 2.57 1.86 1.86 Residue 7.86 1.26 1.25 1.31 1.14 1.2 NaOH added (g) 25.54 23.54 21.27 19.31 33.58 18.33 NaOH added (mL) 16.86 15.54 14.04 12.75 22.17 12.10 Colour samples Time (min.) = 5 colourless colourless colourless colourless l.blue colourless 30 colourless colourless colourless l.blue l.blue colourless 60 colourless colourless colourless l.blue blue v.l.blue 120 colourless colourless v.l.blue blue blue l.blue 180 colourless colourless v.l.blue blue blue blue 240 colourless colourless l.blue blue blue , blue 360 colourless colourless blue blue Addition of NH3 to samples: none none to 360 to 120, 180.240 to none to 240, 360 Colour filtrate very l.blue l.blue d.blue d.blue/purple d.blue blue Colour washwater colourless colourless colourless colourless colourless l.blue Ion chromatograph analyses g/L S-species g/L S-species g/L S-species g/L S-species g/L S-species g/L S-species Thiosulfate 5 21.08 0.00 22.08 0.00 30 21.21 0.00 19.86 0.00 15.93 20.20 60 20.99 0.00 12.60 0.00 11.92 17.17 120 19.06 0.00 8.62 2.05 1.93 7.73 180 17.59 0.00 5.16 0.94 0.00 2.11 240 17.33 0.84 1.59 0.00 0.00 0.30 360 19.83 4.32 0.00 0.00 0.00 0.00 Tetrathionate 5 0.22 2.51 0.32 3.39 30 0.39 2.65 1.07 4.32 1.79 0.03 60 0.58 2.72 2.75 0.99 0.90 0.24 120 0.96 2.71 2.71 2.04 0.15 0.49 180 1.19 2.81 1.04 0.00 0.00 0.21 240 1.20 2.91 0.00 0.00 0.00 0.00 360 0.77 3.13 0.00 0.00 0.00 0.00 Sulfate 5 11.38 0.56 14.95 30 0.65 12.46 0.86 1.92 60 12.55 1.24 6.35 0.65 2.91 120 1.69 2.79 1.41 4.39. 180 5.73 1.69 2.38 2.87 5.67 240 5.21 1.67 3.01 3.22 8.72 360 0.92 3.28 3.13 3.68 9.48 Sulfur balance at every sample time (thiosulfate, tetrathionate and sulfate) g/LS g/LS g/LS g/LS g/LS 5 5.24 13.00 6.93 30 1.51 12.19 6.63 10.42 12.21 60 5.75 9.19 2.69 7.55 10.93 120 1.55 7.05 3.27 1.66 6.17 180 3.52 4.11 1.33 0.96 3.22 240 3.88 1.47 1.01 1.07 3.08 360 4.57 1.09 1.05 1.23 3.17 Trithionate 360 - 7.38 - 6.47 5.29 Tot. S.balance at t=360: 4.57 8.48 1.05 7.70 8.45 - = no trithionate present, or too small to measure 177 Mineral Cu2S+Au Lobo Lobo Lobo Lobo Sulfate addn. Baseline Baseline, dupl. no aeration no Cu2+addn. Grams added 1.86 360 360 360 360 Residue 1.1 342.66 336.53 339.42 335.09 NaOH added (g) 25.08 23 22.63 22.45 21.63 NaOH added (mL) 16.55 15.18 14.94 14.82 14.28 Colour samples Time (min.) 5 colourless v.l.blue v. l.blue colourless colourless 30 colourless v.l.blue v.l.blue colourless colourless 60 v.l.blue v.l.blue v.l.blue colourless colourless 120 l.blue l.blue l.blue colourless colourless 180 blue l.blue l.blue colourless colourless 240 blue blue colourless colourless 360 Addition of NH3 to samples: to 240, to none to none to none to none Colour filtrate blue blue blue v.l.blue colourless Colour washwater colourless colourless l.blue colourless colourless Ion chromatograph analyses g/L S-species g/L S-species g/L S-species g/L S-species g/L S-species Thiosulfate 5 30 21.42 13.91 13.01 4.58 21.87 60 17.65 13.30 10.62 9.22 21.44 120 12.19 8.78 3.30 7.10 20.52 180 4.99 0.00 0.00 11.58 19.52 240 0.79 0.00 0.00 10.29 18.70 360 0.00 0.00 15.17 16.86 Tetrathionate 5 30 0.54 2.03 2.26 3.36 0.51 60 0.92 0.97 0.94 3.04 0.66 120 0.38 5.40 0.17 3.16 0.78 180 0.22 0.00 0.00 2.61 0.82 240 0.00 0.00 0.00 2.68 0.81 360 0.00 0.00 1.97 0.65 Sulfate 5 30 4.80 2.40 2.83 3.88 0.94 60 4.69 2.10 2.91 3.86 0.95 120 4.93 4.38 3.71 3.45 0.96 180 6.18 6.25 6.30 3.22 0.97 240 7.15 6.55 6.63 3.17 0.98 360 9.46 7.37 2.29 0.99 Sulfur balance at every sample time (thiosulfate, tetrathionate and sulfate) C g/LS g/LS g/LS g/LS g/LS J 30 14.17 9.92 9.68 5.84 13.12 60 12.19 8.87 7.58 8.30 12.96 120 8.84 9.58 3.23 7.02 12.51 180 5.04 2.09 2.10 9.19 11.96 240 2.84 2.19 2.21 8.48 11.49 360 3.16 0.00 2.46 10.57 10.35 Trithionate 360 4.07 4.00 5.45 - 3.09 Tot. S.balance at t=360: 7.23 4.00 7.91 10.57 13.44 - = no trithionate present, or too small to measure 178 Mineral Lobo Guanaco Guanaco Guanaco Guanaco no Cu2+addn.,dupl. Baseline no aeration no Cu2+addn. Thiogold Grams added 360 360 360 360 360 Residue 338.78 343.13 339.72 338.65 341.97 NaOH added (g) 21.74 22.76 22.76 20.69 24.11 NaOH added (mL) 14.35 15.02 15.02 13.66 15.91 Colour samples Time (min.) = 5 colourless l.blue colourless colourless l.blue 30 colourless blue colourless v.l.blue blue 60 colourless blue colourless l.blue blue 120 colourless blue colourless blue blue 180 colourless blue colourless/v.l.b. blue blue 240 colourless blue v.l.b. blue blue 360 Addition of NH3 to samples: to none to all to none to 180, 240, 360 to all Colour filtrate colourless blue l.blue blue d.blue/purple Colour washwater colourless blue v.l.b. l.blue blue Ion chromatograph analyses g/L S-species g/L S-species g/L S-species g/L S-species g/L S-species Thiosulfate 5 30 22.73 12.54 22.24 15.94 11.32 60 22.03 9.37 19.86 14.64 7.82 120 21.09 3.07 14.55 6.84 0.66 180 20.09 0.00 12.96 1.50 0.00 240 19.60 0.00 13.52 0.00 0.00 360 17.52 0.00 14.36 0.00 0.00 Tetrathionate 5 30 0.45 1.78 0.34 1.88 0.97 60 0.64 0.78 0.98 1.12 0.45 120 0.67 0.12 2.03 0.43 0.00 180 0.72 0.00 2.36 0.00 0.00 240 0.76 0.00 2.40 0.00 0.00 360 0.57 0.00 2.41 0.00 0.00 Sulfate 5 30 0.88 3.47 2.54 1.46 3.93 60 0.87 3.62 2.86 1.36 4.04 120 0.94 4.02 3.43 2.17 6.35 180 0.92 6.27 3.61 3.61 7.43 240 1.00 6.70 3.52 5.21 7.75 360 1.06 7.10 3.13 5.83 8.41 Sulfur balance at every sample time (thiosulfate, tetrathionate and sulfate) c g/LS g/LS g/LS g/LS g/LS J 30 13.55 9.35 13.77 10.69 8.35 60 13.26 7.01 12.88 9.47 6.08 120 12.75 3.17 10.63 4.88 2.50 180 12.21 2.09 9.97 2.07 2.48 240 11.98 2.24 10.28 1.74 2.59 360 10.70 2.37 10.64 1.95 2.81 Trithionate 360 2.89 8.34 - 5.17 6.15 Tot. S.balance at t=360: 13.59 10.71 10.64 7.12 8.96 - = no trithionate present, or too small to measure 179 Mineral M40 M40 M40 M40 M10 Baseline no aeration no Cu2+addn. no Cu 2 + , no air Baseline Grams added 360 360 360 360 360 Residue 334.39 336.6 336.35 338.04 333.4 NaOH added (g) 21.5 19.85 25.63 24.72 20.94 NaOH added (mL) 14.19 13.10 16.92 16.32 13.82 Colour samples Time (min.) 5 v.l.blue colourless colourless colourless l.blue 30 l.blue colourless colourless colourless l.blue 60 blue colourless colourless colourless blue 120 blue colourless colourless colourless blue 180 blue colourless l.blue colourless blue 240 blue colourless blue colourless blue 360 blue colourless blue colourless blue Addition of NH3 to samples: to all, except 5 to none to 360 to none to all, exept 5 Colour filtrate d.blue/purple colourless, yello blue colourless, yellowis d.blue/purple Colour washwater l.blue colourless v.l.blue colourless blue Ion chromatograph analyses g/L S-species g/L S-species g/L S-species g/L S-species g/L S-species Thiosulfate 5 1.60 9.04 23.87 19.67 8.67 30 7.32 12.11 17.67 25.13 8.06 60 4.51 9.45 15.25 25.81 7.02 120 0.66 11.95 10.34 26.18 2.55 180 0.00 11.71 10.28 25.04 0.00 240 0.00 12.02 1.36 25.05 0.00 360 0.00 12.95 2.20 24.78 0.00 Tetrathionate 5 3.68 3.89 1.09 5.45 0.00 30 3.08 3.41 2.37 0.63 4.27 60 0.62 3.67 2.32 0.52 3.07 120 0.00 3.07 1.57 0.66 1.49 180 0.00 3.09 1.38 0.52 0.30 240 0.00 2.82 0.12 0.53 0.00 360 0.00 2.68 0.24 0.72 0.00 Sulfate 5 7.93 4.54 1.94 0.27 4.46 30 4.66 4.12 2.33 2.09 4.75 60 5.04 5.09 2.63 0.46 3.93 120 5.00 4.24 3.31 0.15 3.97 180 6.62 4.41 3.88 0.55 5.66 240 7.39 3.98 6.86 0.00 6.55 360 8.76 3.81 7.95 0.00 8.22 Sulfur balance at every sample time (thiosulfate, tetrathionate and sulfate) g/LS g/LS g/LS g/LS g/LS 5 5.67 8.91 14.92 14.46 6.45 30 7.50 10.26 12.24 15.44 8.64 60 4.62 9.21 10.93 15.21 7.08 120 2.05 10.00 7.92 15.40 3.64 180 2.21 9.94 7.97 14.81 2.06 240 2.47 9.82 3.14 14.63 2.19 360 2.93 10.21 4.05 14.59 2.74 Trithionate 360 5.17 - 5.68 - 8.10 Tot. S.balance at t=360: 8.09 10.21 9.73 14.59 10.85 - = no trithionate present, or too small to measure 180 Mineral M10 M10 M10 M10 no aeration no aeration no Cu2+addn. 0.72 M A T S Grams added 360 360 360 360 Residue 335.52 341.73 333.81 337.71 NaOH added (g) 21.58 21.48 21 38.49 NaOH added (mL) 14.24 14.18 13.86 25.41 Colour samples Time Time (min.) = 5 v.l.blue 30 v.l.blue colourless l.blue 30 v.l.blue 120 v.l.blue v.l.blue blue 60 v.l.blue 300 v.l.blue l.blue blue 120 v.l.blue 480 v.l.blue blue d.blue 180 v.l.blue 1440 l.blue blue d.blue 240 v.l.blue blue d.blue 360 v.l.blue blue Addition of NH3 to samples: to none to 1440 to 120,180,240 to none Colour filtrate 1. green l.blue d.blue/purple d.blue/purple Colour washwater v.l. green v.l.blue blue blue Ion chromatograph analyses g/L S-species g/L S-species g/L S-species g/L S-species Thiosulfate 5 7.37 0.51 30 4.10 4.90 2.09 12.87 60 10.71 5.76 2.67 17.56 120 9.06 7.11 0.14 12.55 180 10.13 9.57 0.00 8.75 240 9.42 7.97 0.00 6.04 360 12.54 0.00 1.09 Tetrathionate 5 4.10 3.18 30 3.92 3.81 2.67 5.25 60 3.42 3.29 0.93 3.14 120 3.52 2.92 0.00 3.74 180 3.12 2.46 0.00 3.24 240 3.01 1.14 0.00 2.14 360 2.48 0.00 0.24 Sulfate 5 5.95 6.92 30 7.61 6.81 3.43 6.15 60 4.65 6.40 4.28 5.33 120 5.12 5.72 5.72 5.63 180 4.55 5.03 7.07 5.24 240 8.31 5.35 8.02 5.46 360 3.05 9.72 5.76 Sulfur balance at every samp] e time (thiosulfate, tetrathionate and sulfate) g/LS g/LS g/LS g/LS 5 8.55 4.42 30 7.13 7.26 3.87 12.42 60 9.64 7.32 3.49 13.62 120 8.90 7.65 1.99 11.20 180 9.10 8.56 2.36 8.61 240 9.89 7.00 2.68 6.50 360 9.61 3.25 2.69 Trithionate 360 - 2.95 5.86 21.08 Tot. S.balance at t=360: 9.61 2.95 9.11 23.76 - = no trithionate present, or too small to measure 181 Appendix E: Cyanidation procedures Sample: Overall Lobo Composite Cyanidation Test Report Date: M a y 9/1998 Test Conditions: Slurry Weight [g]: 100.12 Ca(OH) 2 added [g]: 0.11 Target pulp density: 30 Initial N a C N concentration [g/L]: 1 H 2 0 to add [g]: 233.61 N a C N t o a d d [g]: 0.234 Mass of solids [g]: 100.12 Leach time [firs]: 24 Mass of liquids [g]: 233.613 Temperature [°C]: 19.7 Pre-lime p H : 5.47 Stirrer speed [rpm]: 512 Post-lime p H : 9.12 Leach Data Time [hr] N a C N [g/L] N a C N added [g] N a C N cumul. [g] p H Ca(OH) 2 added [g] p H 0 1.00 0.234 0.234 9.86 0.18 10.50 2 0.38 0.147 0.381 10.36 - 10.54 4 0.75 0.062 0.443 10.61 - 10.66 8 0.76 0.059 0.502 10.61 - 10.67 12 0.78 0.055 0.557 10.81 - 10.85 24 0.45 0 0.557 10.75 - -Solids Balance mass [g] A u [ppm] C u [ppm] Solids 100.12 2.11 1120 Residue 99.39 0.4 640 Solution Balance mass [g] A u [ppm] C u [ppm] Filtrate 234.38 0.64 1.75 Titr. waste 100 0.24 34 Au extraction [%] 81.4 assayed head [ppm] 2.11 calculated head [ppm] 2.14 Reagent Consumption N a C N [kg/t] 4.38 Ca(OH) 2 [kg/t] 2.90 Cu extraction [%] 41.1 assayed head [ppm] 1120 calculated head [ppm] 1080.27 Notes: 1. Sodium cyanide concentration was determined by titration of 5 n iL solution samples with silver nitrate and rhodanine indicator at selected time intervals. The solution samples were collected and analyzed for copper and gold content (Titr. Waste) 2. The extractions are based on calculated head 182 Sample: Guanaco Composite Cyanidation Test Report Date: M a y 9/1998 Test Conditions: Slurry Weight [g]: 100.04 Ca(OH) 2 added [g]: 0.03 Target pulp density: 30 Initial N a C N concentration [g/L]: 1 H 2 0 to add [g]: 233.43 N a C N t o a d d [g]: 0.233 Mass of solids [g]: 100.04 Leach time [hrs]: 24 Mass of liquids [g]: 233.427 Temperature [°C]: 20.7 Pre-lime p H : 7.93 Stirrer speed [rpm]: 514 Post-lime p H : 9.72 Leach Data Time [hr] N a C N [g/L] N a C N added [g] N a C N cumul. [g] p H Ca(OH) 2 added [g] p H 0 1.00 0.233 0.233 10.49 0.02 10.59 2 0.09 0.213 0.446 10.27 - 10.74 4 0.45 0.131 0.577 10.58 - 10.78 8 0.55 0.107 0.684 10.46 - 10.64 12 0.68 0.079 0.763 10.70 - 10.76 24 0.33 0 0.763 10.34 - -Solids Balance mass [g] A u [ppm] C u [ppm] Solids 100.04 2.06 3320 Residue 99.78 0.28 1500 Solution Balance mass [g] A u [ppm] C u [ppm] Filtrate 236.697 0.6 1460 Titr. waste 100 0.08 109 Au extraction [%] 84.3 assayed head [ppm] 1.6 calculated head [ppm] 1.78 Reagent Consumption N a C N [kg/t] 6.77 Ca(OH) 2 [kg/t] 0.50 Cu extraction [%] 70.4 assayed head [ppm] 3320 calculated head [ppm] 5061.47 Notes: 1. Sodium cyanide concentration was determined by titration of 5 m L solution samples with silver nitrate and rhodanine indicator at selected time intervals. The solution samples were collected and analyzed for copper and gold content (Titr. Waste) 2. The extractions are based on calculated head 183 Sample: M 4 0 Pyrite Feed Cyanidation Test Report Date: M a y 10/1998 Test Conditions: Slurry Weight [g]: 100.06 Ca(OH) 2 added [g]: 0.07 Target pulp density: 30 Initial N a C N concentration [g/L]: 1 H 2 0 to add [g]: 233.47 N a C N to add [g]: 0.233 Mass of solids [g]: 100.06 Leach time [hrs]: 24 Mass of liquids [g]: 233.473 Temperature [°C]: 19.7 Pre-lime p H : 6.03 Stirrer speed [rpm]: 505 Post-lime p H : 9.89 Leach Data Time [hr] N a C N [g/L] N a C N added [g] N a C N cumul. [g] p H Ca(OH) 2 added [g] p H 0 1.00 0.233 0.233 10.40 0.02 10.62 2 0.09 0.213 0.446 9.91 - 10.63 4 0.46 0.129 0.575 10.84 - 10.96 8 0.59 0.099 0.674 10.78 - 10.86 12 0.66 0.082 0.756 10.70 - 10.81 24 0.46 0 0.756 10.16 - -Solids Balance mass [g] A u [ppm] C u [ppm] Solids 100.06 7.1 5920 Residue 99.85 0.5 3680 Solution Balance mass [g] A u [ppm] C u [ppm] Filtrate 239.283 3.36 775 Titr. waste 100 0.96 117 Au extraction [%] 94.7 assayed head [ppm] 7.1 calculated head [ppm] 9.5 Reagent Consumption N a C N [kg/t] 6.40 Ca(OH) 2 [kg/t] 0.90 Cu extraction [%] 34.9 assayed head [ppm] 5920 calculated head [ppm] 5645.93 Notes: 1. Sodium cyanide concentration was determined by titration of 5 m L solution samples with silver nitrate and rhodanine indicator at selected time intervals. The solution samples were collected and analyzed for copper and gold content (Titr. Waste) 2. The extractions are based on calculated head 184 Cyanidation Test Report Sample: M 1 0 Pyrite Concentrate Date: M a y 10/1998 Test Conditions: Slurry Weight [g]: 100.06 Ca(OH) 2 added [g]: 0.13 Target pulp density: 30 Initial N a C N concentration [g/L]: 1 H 2 O t o a d d [g]: 233.47 N a C N to add [g]: 0.233 Mass of solids [g]: 100.06 Leach time [firs]: 24 Mass of liquids [g]: 233.473 Temperature [°C]: 20 Pre-lime p H : 5.76 Stirrer speed [rpm]: 494 Post-lime p H : 9.57 Leach Data Time [hr] N a C N [g/L] N a C N added [g] N a C N cumul. [g] p H Ca(OH) 2 added [g] p H 0 1.00 0.233 0.233 10.34 0.04 10.73 2 0 0.233 0.466 10.20 - 11.08 4 0.07 0.217 0.683 10.86 - 11.21 8 0.16 0.196 0.879 11.01 - 11.20 12 0.23 0.182 1.061 11.08 - 11.23 24 0.29 0 1.061 10.25 - -Solids Balance mass [g] A u [ppm] C u [ppm] Solids 100.06 45.5 16600 Residue 98.39 21.5 11400 Solution Balance mass [g] A u [ppm] C u [ppm] Filtrate 242.103 7.6 1750 Titr. waste 100 1.32 192 Au extraction [%] 47.5 assayed head [ppm] 45.5 calculated head [ppm] 41.54 Cu extraction [%] 28.3 assayed head [ppm] 16600 calculated head [ppm] 15645.27 Reagent Consumption N a C N [kg/t] 9.91 Ca(OH) 2 [kg/t] 1.70 Notes: 1. Sodium cyanide concentration was determined by titration of 5 m L solution samples with silver nitrate and rhodanine indicator at selected time intervals. The solution samples were collected and analyzed for copper and gold content (Titr. Waste) 2. The extractions are based on calculated head 185 Appendix F: Sulfate determination Determination of sulfate as barium sulfate (derived from Jeffery, [1989]) The method consists in slowly adding a solution of barium chloride to a solution of the sulfate: B a C l 2 + S O 2 " -> B a S 0 4 (s) + 2C1" (F. 1) Normally this procedure is carried out at boiling temperature, to obtain large barium sulfate crystals. For this study, the measurement was performed at room temperature. To determine i f it was possible to obtain representative sulfate numbers at room temperature the following procedure was followed: • Prepare standard solutions of 0.1 M ammonium thiosulfate and 0.1 M ammonium sulfate and 1 M barium chloride • Measure out 100 m L of the first two solutions and transfer them into an 300 m L erlenmeyer • While stirring add slowly 1 M of barium chloride • Filter, wash and dry precipitate The sulfate concentration was calculated from the resulting weight of the barium sulfate precipitate. The amount of sulfate added was 0.96 gram, the method gave 1.08 gram. The filtrates resulting from the leach experiments were subjected to the same procedure, only 50 m L of filtrate was used. 186