@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Applied Science, Faculty of"@en, "Materials Engineering, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Melashvili, Mariam"@en ; dcterms:issued "2015-12-08T02:31:31"@en, "2015"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description "The kinetics of the oxidation of pyrite sulphide to a thiosulphate reaction were studied using a pyrite concentrate oxidized in an alkaline medium at an oxygen overpressure between 10 and 40 psig at a temperature of 80ºC. Sodium hydroxide was used to neutralize the acid produced as a result of the pyrite oxidation. The focus of this study was the calculation of thiosulphate yield as a function of sulphide sulphur concentration of the pyrite in the feed. A single rate expression combining kinetic constants of all the metastable oxyanions was derived to predict the thiosulphate yield as a function of the known pyrite sulphide sulphur concentration under the experimental conditions that were adopted. [formula omitted] It was found that at 20 psig oxygen overpressure and a temperature of 80ºC, the initial rate of sulphide oxidation and thiosulphate yield was close to 0.08 mol/h and 0.015 mol/h, respectively, at pH values greater than 12. However, a shift from linearity occurred when the pH decreased below 12. It was observed in the experiments that any decrease in pH was accompanied by an increase in solution potential, which enhanced the rate of gold leaching provided sufficient thiosulphate remained in the solution. The oxidation and reduction of gold in thiosulphate as a function of applied potential was studied using cyclic voltammetry. The responses of cyclic voltammetry of gold electrodes in thiosulphate were compared under different conditions involving catalysts such as thallium and thiourea. It was found that the gold anodic current increases in the presence of catalysts which is more likely to be related to the adsorption/desorption phenomenon and not to the depletion of the thiosulphate ion at the reaction interface. Higher anodic current increased the surface gold concentration available for reduction. The large anodic-to-cathodic peak separation that ranged from 0.25V to 0.37V allowed most of the leached gold to diffuse away from the electrode. This indicates the electrochemically-irreversible character of this system even in the presence of catalysts. The relationship between the reduction peak current and concentration of gold thiosulphate in the bulk electrolyte was estimated based on a relevant equation for electrochemically-irreversible reaction."@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/55724?expand=metadata"@en ; skos:note "KINETICS OF SULPHIDE CONVERSION TO THIOSULPHATE AND GOLD DISSOLUTION STUDY BY ELECTROCHEMICAL QUARTZ CRYSTAL MICROBALANCE by MARIAM MELASHVILI A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF Doctor of Philosophy in The Faculty of Graduate and Postdoctoral Studies (Materials Engineering) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) December 2015 © Mariam Melashvili, 2015 ii ABSTRACT The kinetics of the oxidation of pyrite sulphide to a thiosulphate reaction were studied using a pyrite concentrate oxidized in an alkaline medium at an oxygen overpressure between 10 and 40 psig at a temperature of 80oC. Sodium hydroxide was used to neutralize the acid produced as a result of the pyrite oxidation. The focus of this study was the calculation of thiosulphate yield as a function of sulphide sulphur concentration of the pyrite in the feed. A single rate expression combining kinetic constants of all the metastable oxyanions was derived to predict the thiosulphate yield as a function of the known pyrite sulphide sulphur concentration under the experimental conditions that were adopted. dCS2O3dt= 2−1k1k2[Cos−2]�PO2�∝1[OH]β2{ 1 − k4�PO2�∝4 + 2−3k4k5�PO2�∝4[OH]β5 (5 + k6�OH]β6�} It was found that at 20 psig oxygen overpressure and a temperature of 80oC, the initial rate of sulphide oxidation and thiosulphate yield was close to 0.08 mol/h and 0.015 mol/h, respectively, at pH values greater than 12. However, a shift from linearity occurred when the pH decreased below 12. It was observed in the experiments that any decrease in pH was accompanied by an increase in solution potential, which enhanced the rate of gold leaching provided sufficient thiosulphate remained in the solution. The oxidation and reduction of gold in thiosulphate as a function of applied potential was studied using cyclic voltammetry. The responses of cyclic voltammetry of gold electrodes in thiosulphate were compared under different conditions involving catalysts such as thallium and thiourea. It was found that the gold anodic current increases in the presence of catalysts which is more likely to be related to the adsorption/desorption phenomenon and not to the depletion of the thiosulphate ion at the reaction interface. Higher anodic current increased the surface gold concentration available for reduction. The large anodic-to-cathodic peak separation that ranged from 0.25V to 0.37V allowed most of the iii leached gold to diffuse away from the electrode. This indicates the electrochemically-irreversible character of this system even in the presence of catalysts. The relationship between the reduction peak current and concentration of gold thiosulphate in the bulk electrolyte was estimated based on a relevant equation for electrochemically-irreversible reaction. iv PREFACE The concept of studying an in-situ thiosulphate formation from pyrite was proposed originally by Dr. David Dreisinger. This research was supported by SGS Canada and is presented in chapters 4, 5, and 6. The theory of a stepwise increase in sulphur oxidation state was postulated by Dr. Chris Fleming but the derivation of equations, as well as data analysis, were done by the author of this thesis. The experiments described in chapter 7 were accomplished at UBC by the author and the financial support for this part was provided by Barrick Gold. The overall supervision for this thesis was provided by Dr. David Dreisinger. The following articles were published or under review for publication. Melashvili, M., Fleming, C., Dymov, I., Matthews, D., Dreisinger, D. (2015). Equation for thiosulphate yield during pyrite oxidation. Minerals Engineering Journal. (74) 105-111. Melashvili, M., Fleming, C., Dymov, I., Matthews, D., Dreisinger, D. (2015) Dissolution of gold during pyrite oxidation reaction. Minerals Engineering Journal. Pages ahead of print. Melashvili, M., Dreisinger, D. (2015) Cyclic Voltammetry Responses of Gold Electrode in Thiosulphate Electrolyte. Under Review. v TABLE OF CONTENTS ABSTRACT ....................................................................................................................... ii PREFACE ......................................................................................................................... iv TABLE OF CONTENTS ................................................................................................. v LIST OF TABLES ......................................................................................................... viii LIST OF FIGURES ......................................................................................................... ix ACKNOWLEDGEMENTS .......................................................................................... xiii 1 INTRODUCTION ...................................................................................................... 1 2 LITERATURE REVIEW .......................................................................................... 2 2.1 INTRODUCTION ....................................................................................................... 2 2.2 METALLURGY OF GOLD .......................................................................................... 2 2.2.1 Oxidative pre-treatment of ore ........................................................................ 3 2.3 THIOSULPHATE FORMATION DURING PYRITE OXIDATION ........................................ 4 2.3.1 General leaching reaction rate models ............................................................ 4 2.3.2 Pyrite oxidation rate models ........................................................................... 7 2.3.3 PH-dependent reactions between sulphur oxyanions .................................... 16 2.3.4 Pyrite catalyzed decomposition of thiosulphate ........................................... 20 2.3.5 Complexation of gold by aqueous sulphur compounds ................................ 21 2.4 GOLD THIOSULPHATE ELECTROCHEMISTRY .......................................................... 25 2.4.1 Electrochemical studies ................................................................................ 25 2.5 OBJECTIVES OF CURRENT STUDY .......................................................................... 33 3 EXPERIMENTAL METHODS .............................................................................. 34 3.1 APPARATUS .......................................................................................................... 34 3.1.1 Oxidation and leaching reactor setup ............................................................ 34 3.1.2 Electrochemical setup ................................................................................... 36 3.2 EXPERIMENTAL METHOD ...................................................................................... 39 3.2.1 Oxidation and leaching ................................................................................. 39 vi 3.2.2 Electrochemical measurements ..................................................................... 40 4 PYRITE OXIDATION REACTION ...................................................................... 44 4.1 INTRODUCTION ..................................................................................................... 44 4.2 RESULTS ............................................................................................................... 44 4.2.1 Effect of oxygen pressure ............................................................................. 44 4.2.2 Effect of sodium hydroxide .......................................................................... 49 4.2.3 Effect of residence time ................................................................................ 60 4.3 DISCUSSION .......................................................................................................... 64 5 GOLD DISSOLUTION DURING PYRITE OXIDATION ................................. 72 5.1 INTRODUCTION ..................................................................................................... 72 5.2 RESULTS ............................................................................................................... 72 5.2.1 Gold dissolution conditions during pyrite oxidation .................................... 72 5.2.2 Gold dissolution at constant alkalinity .......................................................... 78 5.3 DISCUSSIONS ........................................................................................................ 79 6 MODELLING OF PYRITE OXIDATION TO THIOSULPHATE.................... 82 6.1 INTRODUCTION ..................................................................................................... 82 6.2 PYRITE OXIDATION RATE ...................................................................................... 82 6.3 CONVERSION MODEL DEVELOPMENT .................................................................... 85 6.4 THIOSULPHATE YIELD VALIDATION ...................................................................... 89 6.5 CONCEPTUAL LEACHING MODEL DISCUSSION ....................................................... 95 7 ELECTROCHEMICAL DISSOLUTION OF GOLD .......................................... 97 7.1 INTRODUCTION ..................................................................................................... 97 7.2 CYCLIC VOLTAMMETRY IN THIOSULPHATE ........................................................... 97 7.2.1 Effect of gold thiosulphate .......................................................................... 101 7.2.2 Effect of thallium ........................................................................................ 104 7.2.3 Effect of thiourea ........................................................................................ 113 7.3 DISCUSSION ........................................................................................................ 121 8 CONCLUSIONS..................................................................................................... 126 vii 9 RECOMMENDATIONS ....................................................................................... 130 REFERENCES .............................................................................................................. 132 viii LIST OF TABLES Table 1: Gold complexes ............................................................................................. 22 Table 2: Composition of Pyrite Concentrates - Quantitative analysis ......................... 39 Table 3: Composition of Pyrite Concentrates - ICP Scan ............................................ 39 Table 4: PY1 Sulphide oxidation at various oxygen overpressures - testing conditions 45 Table 5: PY1 Sulphide oxidation at various oxygen overpressures - thiosulphate yield 46 Table 6: Thiosalts concentration during PY1 oxidation at various oxygen pressures 48 Table 7: PY1 Sulphide oxidation at various NaOH additions - testing conditions ..... 49 Table 8: PY1 Sulphide oxidation at various NaOH additions - thiosulphate yield ..... 50 Table 9: Thiosalts concentration during PY1 oxidation at various NaOH additions .. 50 Table 10: PY1 Sulphide oxidation at various pHs - testing conditions ....................... 54 Table 11: PY1 Sulphide oxidation at various pHs - thiosulphate yield ....................... 54 Table 12: Thiosalts concentration during PY1 Sulphide oxidation at various pHs ..... 58 Table 13: PY2 oxidation summary .............................................................................. 61 Table 14: PY2 oxidation at various pHs ...................................................................... 62 Table 15: Gold dissolution during PY1 oxidation ....................................................... 73 Table 16: Gold dissolution during PY2 oxidation ....................................................... 73 Table 17: Concentration of gold, silver, and copper in solution during PY2 oxidation 74 Table 18: Gold dissolution during PY2 oxidation at constant pH ............................... 79 Table 19: Sulphides reacted and intermediates formed during pyrite oxidation at oxygen overpressure of 40 psi and 80oC .................................................................................. 89 Table 20: Sulphides reacted and intermediates formed during pyrite oxidation at oxygen pressure of 20 psi and 80oC ......................................................................................... 90 Table 21: Sulphides reacted and intermediates formed during pyrite oxidation at oxygen pressure of 10 psi and 80oC ......................................................................................... 90 Table 22: Rate of sulphide conversion and estimated rate of unmeasured intermediates ...................................................................................................................................... 93 Table 23: Reaction constants and orders ..................................................................... 94 ix LIST OF FIGURES Figure 1: Pyrite stability (1M Fe, 1M S, and eliminating hematite) as a function of pH and EH (25 oC) ................................................................................................................ 8 Figure 2: Superimposed Eh-pH diagram for sulphur and pyrite 25oC (1M S, 1M Fe eliminating Hematite, sulphate, and higher polythionates) ......................................... 12 Figure 3: Sulphide sulphur conversion paths (adapted from Mishra and Osseo-Asare, 1988) ............................................................................................................................ 14 Figure 4: Sulphide sulphur conversion paths (adapted from Druscher and Borda, 2006) ...................................................................................................................................... 15 Figure 5: Stability fields for gold (10-4 M) – sulphur (1M) complexes (25oC) ............ 23 Figure 6: Stability diagram for 1M thallium (25oC) .................................................... 31 Figure 7: Alkaline leach experimental setup ............................................................... 35 Figure 8: Schematics of reactor ................................................................................... 36 Figure 9: Electrochemical cell experimental setup ...................................................... 37 Figure 10: Electrochemical cell schematics ................................................................. 37 Figure 11: Three electrode setup .................................................................................. 38 Figure 12: A reference cyclic voltammetry scan at 15 mV/s in 0.1 M H2SO4 solution and anodic dissolution voltammetry at 15 mV/s scan rate of Cu-upd layer deposited on the same electrode after 20-minute exposure at 0.3 V using 2mM CuSO4 and 0.1 M H2SO4 solution ......................................................................................................................... 41 Figure 13: Sulphide oxidation versus oxygen overpressure ........................................ 47 Figure 14: Thiosulphate formation versus oxygen overpressure ................................. 47 Figure 15: Sulphur oxyanions distribution in Exp 1 .................................................... 51 Figure 16: Sulphur oxyanions distribution in Exp 2 .................................................... 52 Figure 17: Sulphur oxyanions distribution in Exp 3 .................................................... 52 Figure 18: Rate of NaOH addition for 9.5, 10.5, and 11.5 pulp operating pHs ........... 55 Figure 19: PY1 Sulphide oxidation at various pHs ..................................................... 56 Figure 20: PY1 Thiosulphate formation at various pHs .............................................. 57 Figure 21: Sulphur oxyanions distribution in Exp 11 .................................................. 59 Figure 22: Sulphur oxyanions distribution in Exp 12 .................................................. 59 x Figure 23: Sulphur oxyanions distribution in Exp 13 .................................................. 60 Figure 24: PY2 Sulphide oxidation at various pHs ..................................................... 63 Figure 25: PY2 thiosulphate formation at various pHs ............................................... 64 Figure 26: PY1 and PY2 moles of sulphides oxidized in time ................................... 65 Figure 27: PY1 and PY2 percent of sulphides oxidized in time ................................. 66 Figure 28: Moles of sulphides oxidized at various pHs ............................................... 67 Figure 29: Moles of sulphides oxidized at various pHs ............................................... 68 Figure 30: Stability diagram by HSC for Sulphur Species eliminating higher polythionates and sulphates (1M total Sulphur, 25oC) ................................................ 69 Figure 31: PY1 and PY2 moles of thiosulphate formed in time ................................. 71 Figure 32: Gold dissolution and thiosulphate profile in Exp 4 .................................... 75 Figure 33: Gold dissolution and thiosulphate profile in Exp 7 .................................... 75 Figure 34: Gold dissolution and thiosulphate profile in Exp 5 .................................... 76 Figure 35: Gold dissolution and thiosulphate profile in Exp 6 .................................... 77 Figure 36: Gold dissolution in time, depending on pH and thiosulphate concentration 80 Figure 37: Changes in thiosulphate concentration and gold extraction in time, depending on solution pH and EH ................................................................................................. 81 Figure 38: Sulphide oxidation ...................................................................................... 85 Figure 39: Conceptual model of pyrite sulphide sulphur oxidation ............................ 85 Figure 40: Sulphide versus time and oxygen pressure ................................................. 91 Figure 41: Thiosulphate versus time and oxygen pressure .......................................... 91 Figure 42: Tetrathionate versus time and oxygen pressure ......................................... 92 Figure 43: Trithionate versus time and oxygen pressure ............................................. 92 Figure 44: Sulphite versus time and oxygen pressure ................................................. 93 Figure 45: Comparison of model calculated and measured moles of thiosulphate ..... 95 Figure 46: E-i and E-icalc voltammograms in 0.4 M MgS2O3 at 1 mV/s scan rate ...... 98 Figure 47: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 at scan rate 1 mV/s ................................................................................................................... 99 Figure 48: Mass change versus the charge calculated by integrating the area under i-E curve on figure 46 ...................................................................................................... 100 xi Figure 49: E-i voltammograms in 0.4 M MgS2O3 at 1 mV/s scan rate with 0 mg/L and 30 mg/L gold added as a gold thiosulphate .................................................................... 102 Figure 50: E-icalc voltammograms in 0.4 M MgS2O3 at 1 mV/s scan rate with 0 mg/L and 30 mg/L gold added as a gold thiosulphate ............................................................... 103 Figure 51: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 at scan rate 1 mV/s with 30 mg/L gold added as a gold thiosulphate .................................... 104 Figure 52: E-i voltammogram in 0.4 M MgS2O3 with 0 mg/L, 0.1 mg/L, and 0.5 mg/L Tl at 1 mV/s scan rate ..................................................................................................... 104 Figure 53: E-icalc voltammogram in 0.4 M MgS2O3 with 0 mg/L, 0.1 mg/L, and 0.5 mg/L Tl at 1 mV/s scan rate ................................................................................................ 105 Figure 54: E-i and E-icalc voltammograms in 0.4 M MgS2O3 and 0.5 mg/L Tl at 1 mV/s scan rate ..................................................................................................................... 106 Figure 55: Mass change versus the charge calculated by integrating the area under i-E curve in figure 54 ....................................................................................................... 107 Figure 56: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 and 0.5 mg/L Thallium at scan rate 1 mV/s ............................................................................ 108 Figure 57: E-i voltammograms in 0.4 M MgS2O3 and 0.5 mg/L Tl at 10 mV/s scan rate .................................................................................................................................... 109 Figure 58: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 and 0.5 mg/L Tl at scan rate 10 mV/s ..................................................................................... 110 Figure 59: E-i voltammograms in 0.4 M MgS2O3 and 0.5 mg/L Tl with and without 30 mg/L Au added (1 mV/s scan rate) ............................................................................ 111 Figure 60: E-icalc voltammograms in 0.4 M MgS2O3 and 0.5 mg/L Tl with and without 30 mg/L Au added .......................................................................................................... 112 Figure 61: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 and 0.5 mg/L Tl with and without 30 mg/L Au at scan rate 1 mV/s ...................................... 113 Figure 62: E-i voltammograms in 0.4 M MgS2O3 with and without 0.05 mM thiourea added .......................................................................................................................... 114 Figure 63: E-icalc voltammograms in 0.4 M MgS2O3 with and without 0.05 mM thiourea added .......................................................................................................................... 115 xii Figure 64: E-i and E-icalc voltammograms in 0.4 M MgS2O3 and 0.05 mM thiourea at 1 mV/s scan rate ............................................................................................................ 116 Figure 65: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 and 0.05 mM thiourea at scan rate 1 mV/s ............................................................................... 117 Figure 66: Mass change versus the charge ratio calculated by integrating the area under i-E curve in figure 64 .................................................................................................... 118 Figure 67: E-i voltammograms in 0.4 M MgS2O3 and 0.05 mM thiourea with and without the 30 mg/L gold at 1 mV/s scan rate ........................................................................ 119 Figure 68: E-icalc voltammograms in 0.4 M MgS2O3 and 0.05 mM thiourea with and without Au at 1 mV/s scan rate .................................................................................. 120 Figure 69: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 and 0.05 mM thiourea with and without 30 mg/L Au at scan rate 1 mV/s .............................. 121 Figure 70: Anodic portion of E-icalc voltammogram in a semi-logarithmic coordinates 124 xiii ACKNOWLEDGEMENTS I would like to thank my supervisor, Professor David Dreisinger for his guidance, support, and trust that helped me get where I am now. His kindness and tolerance have been the greatest support during my studies. Thanks are also extended to Dr. Yeonuk Choi, Manager at Barrick Gold, for his support in developing the electrochemical part of this project. Thanks to Simon Joshi and Professor Ed Asselin for sharing knowledge of EQCM apparatus and spare gold electrodes. I thank Dr. Be Wassink for his help in dealing with the theoretical and practical aspects of thiosalt chemistry while at UBC. Special thanks to Dr. Bethan McKevitt who has been a good friend and never a lazy editor of my writings. I am deeply thankful to Dr. Chris Fleming, Senior Consultant at SGS Canada, for his insightful advice, mentoring, and support to launch the internal research project (Pyrite oxidation to thiosulphate and simultaneous dissolution of gold) at SGS Canada, which became an important part of this thesis. I must thank Inna Dymov, Gold Group Manager at SGS Canada, Andre McKen, Director Metallurgical Services at SGS Canada, and Dave Matthews, Technologist at SGS. Finally, I thank my family for supporting me during this period of my life. 1 1 INTRODUCTION The formation of thiosulphate via the alkaline oxidation of pyrite has been reported by several authors (Goldhaber, 1983; Mishra and Osseso-Asare, 1988; Druschel, 2003; Chen and Morris, 1972; Zhang 2004). Pyrite is an abundant sulphide mineral that often contains gold as extremely fine (< 1 micron) particles trapped in the host mineral. If conditions are found in which the lixiviant for gold is formed at the same time as the sulphide oxidation reaction, there might be a potential for simultaneous dissolution of the gold entrapped in the host sulphide matrix. In this thesis, a simple network of reactions is proposed as the oxidation pathway of pyrite sulphide sulphur conversion to higher valence sulphur oxyanions such as tetrathionate, trithionate, sulphite, and sulphate. A systematic study of factors affecting both the yield of thiosulphate and the extent of gold leaching was undertaken and results discussed. Furthermore, the study was undertaken to investigate the electrochemical processes involved in the use of two recently discovered catalysts in order to understand their effect on gold anodic dissolution. The behaviour of gold electrodes in solutions containing magnesium thiosulphate and in the presence of thallium and thiourea is reported, covering the active dissolution, passivity regions, and the electrodeposition region. Using electrochemical techniques, this study attempted to determine if the loss of gold in the thiosulphate leaching system is prompted by the reversible nature of the gold thiosulphate complex (reversing back to the metallic state), especially in the presence of electrocatalysts by providing much higher rates of redox activities. 2 2 LITERATURE REVIEW 2.1 INTRODUCTION The literature review revisits some of the more prominent findings related to the current study. It starts with the general information about gold, its occurrence, and some of the methods that are used to overcome difficulties in processing this metal. Next, the chemistry of pyrite oxidation is summarized by providing examples from a number of relevant publications. Furthermore, the electrochemistry of gold thiosulphate is discussed. This is followed by a more detailed discussion about complexation of gold by thiosulphate and, finally, the objectives of the current study in relevance to this review are discussed. 2.2 METALLURGY OF GOLD Gold most commonly occurs in the form of native metal, either alone or mixed with silver as an electrum or gold-silver tellurides. Disseminated gold deposits are frequently found in nature where gold is associated with host minerals such as pyrite, arsenopyrite, realgar, and stibnite, as well as carbonaceous materials and silicates. Because of the low grade of gold in these ores, the extraction method of gold is often dictated by the host mineral. If gold is occluded into an impermeable mineral matrix, it can only be exposed to the lixiviant by applying some type of mechanical or chemical process to destroy the mineral matrix. One of the methods to eliminate the effect of gangue minerals is to apply an oxidative pre-treatment using one of a variety of available methods. 3 2.2.1 OXIDATIVE PRE-TREATMENT OF ORE Oxidative pre-treatment increases the extraction of gold by oxidizing the sulphide content and possibly destroying active sites on carbonaceous matter. Depending on the composition of the mineral resource, hydrometallurgical or pyrometallurgical methods may be employed. The hydrometallurgical method involves the decomposition of a sulphide mineral by increasing the oxidizing potential of the solution. This is achieved by the addition of a suitable oxidant, for example, oxygen, chlorine, or nitric acid and, in some cases, requires elevating the temperature and pressure. The chemistry of oxidation can be described based on an EH-pH diagram and taking into account kinetic constraints learned from experience. The susceptibility of a mineral to aqueous oxidation is dependent on its chemical and electrical properties (Marsden and House, 2006). Another option is the pyrometallurgical oxidation process. This can be discussed based on a high temperature, phase-stability diagram. It is achieved by roasting the sulphide and carbonaceous constituents of an ore in the presence of air or oxygen. As a result, porous oxides (calcine) are produced in which the gold is liberated. Single or two-stage processes are employed depending on ore type to be processed. Single stage roasting utilizes only an oxidizing atmosphere. In a two-stage operation, the material is initially treated under reducing conditions to create a porous intermediate product and is followed by a roast in an oxidizing atmosphere to complete the oxidation (Marsden and House, 2006). Regardless of whether the hydrometallurgical or the pyrometallurgical method is chosen, the composition of the oxidative product is largely determined by the mineralogy of the initial material. 4 2.3 THIOSULPHATE FORMATION DURING PYRITE OXIDATION Pyrite is an abundant sulphide mineral that often contains gold as extremely fine (< 1 micron) particles trapped in the host mineral. In most instances, it is necessary to decompose the pyrite to liberate and recover the contained gold. The popular route for decomposition of sulphide minerals is through oxidation by molecular oxygen. Depending on the conditions applied, the nature of the oxidation products might differ substantially. When there is no additional reagent included, the oxidation leads to aqueous ferrous/ferric ions and sulphate. But in the presence of sodium hydroxide, the oxidation of pyrite forms goethite or amorphous iron oxide as a solid product (Ciminelli, 1987) and various aqueous sulphur oxyanions. The potential for thiosulphate formation in parallel to sulphide oxidative decomposition has been reported by several authors (Goldhaber, 1983; Mishra and Osseso-Asare, 1988; Druschel, 2003; Chen and Morris, 1972; Zhang, 2004) and is discussed in the following chapters. 2.3.1 GENERAL LEACHING REACTION RATE MODELS The rate of reaction during leaching is usually described in terms of moles of reacted solids per unit interfacial surface area or volume and time in connection with the moles of disappearing aqueous reagents per volume and time. The mathematical representation of the reaction, or the model selected for the progress of the reaction, is indicative of the rate-limiting step. There are several different models known and widely used. Each model is a rate equation which can be mathematically manipulated into a linearized form to allow for selecting the appropriate model for the particular transformation reaction of interest. The best model for a given data set would produce a straight line when plotting the linearized rate expression versus time. Models describing a progress of solid-solution transformation are based on two main controlling factors: diffusion, either through the product layer or stagnant film, and 5 reaction at the interface. There is a possibility for mixed control and other complex modelling scenarios but the basic models can be grouped into three categories. For example, if the reaction is given as: A + bB C 2- 1 When diffusion through the stagnant film controls the rate of reaction, the leaching reaction can be described as (Levenspiel, 1999): 𝐭𝛕= 𝟏 − �𝐫𝐜𝐑�𝟑 = 𝐗𝐛 2- 2 with τ = ρB R3b kf C 2- 3 where τ is the time for complete conversion, kf is the reaction constant, C is a reactant concentration driving force, R is the radius of the particle, rc is the core radius of the particle, ρB is the density of the particle, �rcR�3 is the shrinkage of the core, and Xb is the fractional conversion. If diffusion through the porous product layer developed during leaching defines the rate of leaching, the leaching reaction can be described as (Levenspiel, 1999): 𝐭𝛕= 𝟏 − 𝟑 �𝐫𝐜𝐑�𝟐 + 𝟐�𝐫𝐜𝐑�𝟑 = 𝟏 − 𝟑 (𝟏 − 𝐗𝐛)𝟐/𝟑 + 𝟐 (𝟏 − 𝐗𝐛) 2- 4 6 with 𝛕 = 𝛒𝐁 𝐑𝟐𝟔𝐛 𝐃𝐩 𝐂 2- 5 where Dp is diffusion through the product layer. When progress of the reaction is unaffected by the presence of any film or product layer, the rate is then proportional to the available surface of unreacted core through this relationship (Levenspiel, 1999): 𝐭𝛕= 𝟏 − 𝐫𝐜𝐑 = 𝟏 − (𝟏 − 𝐗𝐛)𝟏/𝟑 2- 6 with τ = ρB Rb k′ C 2- 7 where k′ is rate constant for the surface reaction. In the simplest case, the progress of a reaction could be described by a single rate expression to describe its kinetic behaviour. However, in many cases, the progress of the reaction may require evaluating multiple reaction steps and require more than one rate expression. For these cases, the primary consideration is a product distribution. Multiple reaction steps can be constructed as parallel reactions or series reactions, depending on the transformation path. For example, a two-step irreversible series reaction could be described as: 7 A B C 2- 8 𝐫𝐀 = −𝐤𝟏 𝐂𝐀 2- 9 𝐫𝐁 = 𝐤𝟏 𝐂𝐀 − 𝐤𝟐 𝐂𝐁 2- 10 𝐫𝐂 = 𝐤𝟐 𝐂𝐂 2- 11 A more complicated scheme, including irreversible parallel reactions, could be: A B C 2- 12 E F 𝐫𝐀 = −𝐤𝟏𝟑 𝐂𝐀 2- 13 𝐫𝐁 = 𝐤𝟏 𝐂𝐀 − 𝐤𝟐𝟒 𝐂𝐁 2- 14 If a reaction has a number of competing paths available, it will proceed by the one with the least resistance. Knowledge of the product distribution is often the clue for determining the path that dominates (Levenspiel, 1999). Knowledge of the thermodynamics of the system will allow for predictions of possible paths, but only a study of the reaction kinetics will be able to provide the corresponding rate expressions. 2.3.2 PYRITE OXIDATION RATE MODELS A number of papers describe various mechanisms for pyrite decomposition. All agree that the most important factors affecting decomposition of the pyrite mineral matrix are: the oxygen pressure, temperature, and pH. It is well established that an elevated temperature and higher oxygen pressure improve the kinetics of the pyrite decomposition reaction. Regarding the effect of pH, there are contradictory statements (findings) published. K1 K2 K3 K4 K1 K2 8 Earlier studies (Gray, 1955-56; Warren, 1956; Stenhouse and Armstrong, 1952; Woodcock, 1961; Ciminelli and Osseo-Asare, 1995) indicated a reduction in the reaction rate with an increase in pH. They reported a shift from a chemical reaction-controlled mechanism to diffusion through the oxide film. An oxide film may be one of the products of alkaline oxidation. Other researchers (Smith and Shumate, 1970; Majima and Peters, 1966; and Goldhaber, 1983) reported higher rates at higher pH values. By carefully analysing their work, Ciminelli and Osseo-Asare (1995) concluded that the discrepancy is caused by the varying experimental conditions in these studies, causing different responses and, therefore, making a direct comparison impossible. The Pourbaix diagram in Figure 1 shows stability fields for pyrite as a function of pH and EH. Figure 1: Pyrite stability (1M Fe, 1M S, and eliminating hematite) as a function of pH and EH by STABCAL (25oC) 9 According to this diagram, under acidic conditions, iron remains either as ferrous (Rimstidt and Vaughan, 2003) or ferric (Bailey and Peters, 1976), depending on solution potential. FeS2 + 3.5O2 + H2O = Fe2+ + 2H+ + 2SO42- 2- 15 4Fe2+ +O2 +4H+ = 4Fe3+ + 2H2O 2- 16 In addition, the ferric cation, as a strong oxidant, might contribute to the decomposition of pyrite (Rimstidt and Vaughan, 2003) according to this reaction: FeS2 + 14Fe3+ + 8H2O = 15Fe2+ + 16H+ + 2SO42- 2- 17 Increasing the oxygen overpressure stabilizes the ferric ion and is expected to provide a higher oxidation rate since the ferric may act as an additional oxidant in this system. In alkaline solutions, the pyrite decomposition reaction proceeds via the formation of iron oxide (Ciminelli and Osseo-Asare 1995) as follows: FeS2 + 15/4O2 + 4OH- = FeOOH + 2SO42- + 3/2H2O 2- 18 The insoluble iron oxides precipitate on the mineral surface and/or in the bulk aqueous solution. When the oxide material precipitates on the mineral surface, the mass transfer of reactants and products to and from the solution phase is inhibited to some extent. In this case, the physical characteristics of the oxide precipitate such as porosity, grain size, and thickness, can affect control of the rate of decomposition of the pyrite. Hence, diffusion though the product layer becomes an important factor. 10 Comprehensive studies of pyrite oxidation in sodium hydroxide solution by Ciminelli (1987) confirmed the formation of a stable colloidal suspension in solution and a thin, fine texture oxide layer uniformly covering the mineral surface. According to her studies, this oxide layer fractures and detaches after some time, favoring the transport of soluble reactants and products through the product layer and, consequently, not inhibiting the rate of pyrite decomposition. Burkin and Edwards (1963; Ciminelli, 1987) and Burkin (1969; Goldhaber, 1983) suggested that the formation of a uniform oxide coating in sodium hydroxide solutions occurs through the replacement of sulphur atoms by oxygen. According to this view, the FeO formed is not stable and the iron oxidizes to a higher valence state, resulting in an alteration within the lattice which, in turn, cracks the oxide layer after a given time. In order to explain the fact that most of the iron appears to precipitate (or reprecipitate) in the bulk solution, Ciminelli (1987) admits the possibility of a formation of a stable nucleus far from the interface or, alternatively, dissolution of the initial nuclei and further reprecipitation in the bulk solution. The findings by Ciminelli (1987) and Burkin (1963, 1969) provide support to the theory that the reaction in sodium hydroxide solutions is under chemical control and the oxide precipitate does not inhibit the pyrite oxidation reaction. Ciminelli (1987) studied the oxidation of pyrite at temperatures between 50 and 80oC, oxygen partial pressures from 0 to 1 atm, and pH from acidic up to 12.5. The reaction rate was found to increase with higher values of all three factors: temperature, pressure, and pH. The rate expression with respect to oxygen and hydroxyl ions was derived (Ciminelli 1987) and the final result is shown below: -dn/dt = S b k’’ pO20.5 [OH]0.25/(1+k’’’pO20.5) 2- 19 where dn/dt is moles of pyrite consumed per unit time, b is the number of moles of pyrite reacting with one mole of oxygen, S is the solid surface area, and k’’ and k’’’ combine the rate constants for the elementary steps and the Henry’s law constant relating 11 concentration and partial pressure of oxygen. The concentration of oxygen at the interface was related to the concentration in the bulk by a Langmuir-type equation, indicating that the rate is controlled by processes involving the adsorption and desorption of oxygen. The observed apparent reaction order with respect to the hydroxyl ion concentration was explained as a combined effect of processes involving the oxidation and hydrolysis of iron, sulphur oxidation and hydrolysis and the oxygen reduction. In Ciminelli’s (1987) study, the unreacted core model for chemical control fit the data: 1 - (1-X)1/3 =kct 2- 20 where kc is an overall rate constant and X is a conversion of pyrite calculated based on the weight of the unreacted solids. To check this value, an estimate of conversion based on precipitating the soluble sulphur with barium chloride was investigated but found to have limited success. The conversion calculated by this method was always smaller than the value obtained by weighing the solid residues. It was suggested that the less oxidized sulphur species might interfere with the analysis, causing the error between the two calculations. Clearly, the conversion of sulphide sulphur to aqueous species and their interaction chemistry is much more complex than the conversion of iron. The stability fields for less oxidized sulphur species closely situated with pyrite can be estimated when their free energy calculation (eliminating sulphate) is superimposed with pyrite on an EH-pH diagram. Based on the stability field depicted in Figure 2, the tetrathionate and the elemental sulphur species predominate under acidic conditions, while thiosulphate and sulphite would likely exist in an alkaline solution. 12 Figure 2: Superimposed Eh-pH diagram for sulphur and pyrite 25oC by STABCAL (1M S, 1M Fe eliminating Hematite, sulphate, and higher polythionates) With regards to the mechanism for sulphur conversion, the majority of papers agree with the notion of a stepwise increase of the sulphur oxidation number from sulphide to sulphate. Tributsch and Gerischer (1976) postulated that oxidation follows two parallel and competing reaction paths: the first yielding sulphate sulphur and the second, yielding elemental sulphur, as shown in the reactions below: FeS2+ 7/2O2 + H2O = FeSO4 + H2SO4 2- 21 FeS2 + 2O2 = FeSO4 + So 2- 22 Other authors provide evidence of the production of sulphur as a product of chemical decomposition of thiosulphate. They propose that the thiosulphate is an immediate intermediate product of sulphide sulphur oxidation. Descostes et al. (2004) provides the 13 following multistep conversion for the oxidation of pyrite in contact with air at 25oC and pH less than 3. FeS2 + 1.5O2 = Fe2+ + S2O32- 2- 23 S2O32- + 1.2H+ = 0.4So +0.4S4O62- + 0.6H2O 2- 24 S4O62- + 3.5O2 +3H2O =4SO42- +6H+ 2- 25 With the overall reaction: FeS2 + 2.5O2 +0.6H2O = Fe2+ + 0.4So +1.6SO42- + 1.2H+ 2- 26 Druscher and Borda (2006) challenged the assumption of direct tetrathionate production upon acidification of thiosulphate and supported the reaction promoted by Williamson and Rimstidt (1992). This pathway was based on thiosulphate and ferric reacting to form tetrathionate and ferrous ions and any excess Fe3+ further oxidizes tetrathionate to sulphate (Druschel et al., 2003) as shown in the following reactions: 2S2O32- + 2Fe3+ = 2FeS2O32+ = S4O62- + 2Fe2+ 2- 27 S4O62- + 14Fe3+ + 10 H2O = 4SO42- + 14Fe2+ + 20H+ 2- 28 Mishra and Osseo-Asare (1988) supported the formation of thiosulphate as an intermediate, which converts to zero-valent sulphur at low potentials and sulphate at high potentials according to the conversion scheme shown in Figure 3: 14 Figure 3: Sulphide sulphur conversion paths (adapted from Mishra and Osseo-Asare, 1988) However, according to Goldhaber (1983), if the oxidation is conducted under alkaline conditions, the reaction might proceed as presented below: FeS2 + 3/4O2 + ½H2O = FeOOH + 2So 2- 29 2FeS2 + 4O2 +2H2O = 2FeOOH + 2H+ + S4O62- 2- 30 FeS2 +7/4O2 +2OH- = FeOOH + 1/2H2O + S2O32- 2- 31 FeS2 +11/4O2+ 4OH- = FeOOH + 3/2H2O + 2SO32- 2- 32 According to Goldhaber (1983), the relative percentage of sulphur species formed is determined by the pH. At pH 9 and 25°C, thiosulphate predominates, followed by sulphite. As the pH decreases, the concentrations of thiosulphate and sulphite decrease, while the concentrations of tetrathionate and sulphate increase. An approximate stoichiometry for the pyrite oxidation in alkaline solution at ambient temperature (22-25oC) and pressure (1 atm) based on experimental data was given by Zhang (2004). 2FeS2 +11O2 +32NaOH + 2H2O = 2Fe(OH)3 + 8Na2S2O3+ 8Na2SO3 2- 33 15 Four different conversion paths of sulphide sulphur conversion are provided by Druscher and Borda (2006), as shown in Figure 4. Figure 4: Sulphide sulphur conversion paths (adapted from Druscher and Borda, 2006) McKibben and Barnes (1986) and Druschel and Borda (2006) suggested that the presence of ferric ions in an acidic environment causes the complete oxidation of sulphide sulphur to sulphate. Since ferric ions do not persist in alkaline solutions, it can be assumed that the transformation mechanism of sulphide to oxyanions would be based purely on the sulphur chemistry. 16 2.3.3 PH-DEPENDENT REACTIONS BETWEEN SULPHUR OXYANIONS Elemental Sulphur The formation of So is favored in acidic solution but, as alkalinity increases, sulphur hydrolyzes to form thiosulphate and hydrosulphide (Kleinjan et al., 2005): 4So + 4OH- = S2O32- + 2HS- + H2O 2- 34 This hydrolysis of elemental sulphur is reported to occur quickly at temperatures close to 100oC. The colloidal particles of biologically-produced sulphur have been observed to undergo hydrolysis at 55oC and at pH 10 (Buisman et. al., 1990) with a higher reaction rate at higher pH. Hydrosulphide According to Chen and Morris (1972), the oxidation of sulphide may either produce or consume hydrogen ions, depending on the conditions. Thus: 2HS- + O2 + 2H+ = 2H2O + 2So 2- 35 2HS- + 2O2 = H2O + S2O32- 2- 36 2HS- + 4O2 = 2SO42- + 2H+ 2- 37 These authors have demonstrated that thiosulphate is the principal product at pH >8.5, regardless of the sulphide to oxygen ratio. Avrahami and Golding (1968; Lefers et al., 1978) reported that at pH > 11, thiosulphate is extremely stable and oxidation to sulphate is very slow. 17 Thiosulphate Under weakly-acidic conditions, thiosulphate is readily oxidized to tetrathionate by oxygen (Lyons and Nickless, 1968): 2S2O32- + 2H+ + ½ O2= S4O62- + H2O 2- 38 Under strongly acidic conditions, thiosulphate may disproportionate to elemental sulphur and sulphite (Davis, 1958; Goldhaber, 1983). The same decomposition reaction at low acidity but elevated temperatures of 70-130oC was reported by Smith and Hitchen (1976) but at higher acidity, these authors reported tetrathionate instead of sulphite. The tetrathionate so formed does not decompose appreciably under these conditions unless the temperature exceeds 130oC. At this temperature, thiosulphate decomposition forms sulphate and elemental sulphur (Smith and Hitchen, 1976). S2O32- + H+ = HSO3- + So 2- 39 5S2O32- + 6H+ = 2So + 2S4O62- + 3H2 2- 40 2H+ + 3S2O32- = 4So + 2SO42- + H2O 2- 41 In alkaline solutions in the temperature range of 75-87oC, the final product of the oxidation of thiosulphate by oxygen is sulphate (Rolla and Chakrabarti, 1982). S2O32- + 2O2 + H2O = 2SO42- + 2H+ 2- 42 In this study, the rate of oxidation with respect to thiosulphate was found to be the first order. The rate equation for the oxidation of thiosulphate by molecular oxygen was given as: -d[S2O32-] /dt = k[S2O32-][OH-]1.1[pO2]1.66 2- 43 18 Tetrathionate In contrast to thiosulphate, tetrathionate is stable in acid but hydrolyzes in alkaline solution and is converted predominantly to thiosulphate (Lyons and Nickless, 1968; Kurtenacker et al., 1936; Smith and Hitchen, 1976). 4S4O62- + 6OH- = 5S2O32- + 2S3O62- + 3H2O 2- 44 2S4O62- + 6OH- = 3S2O32- + 2SO32- + 3H2O 2- 45 Fava and Bresadola (1955; Rolla and Chakabarti, 1982) confirmed the first reaction (2-44) in a dilute alkaline solution and the second reaction (2-45) in a hot concentrated alkaline solution. Rolla and Chakabarti (1982) demonstrated that the yield of trithionate and thiosulphate from tetrathionate and hydroxide after 2 h of reaction time at pH 12 and 25oC corresponded to the accepted stoichiometry of the first (2-44) decomposition reaction. The rate of transformation of tetrathionate to thiosulphate and trithionate at pH >10 was found to be the first order with respect to both tetrathionate and hydroxide. -d[S4O62-] /dt = k[S4O62-][OH-] 2- 46 With the average value of k = 0.17 + 0.03 L mol-1 s-1. Various studies reported that an increase in pH and temperature greatly accelerated the tetrathionate transformation reaction. According to Zhang and Dreisinger (2002), the reaction at pH 12 became too fast to follow, even at the lowest temperature investigated (22°C). The plots of tetrathionate concentration change against time presented a straight line passing through the origin and no diversion of the linearity was observed even at the end of the reaction. The rate equation for the decomposition of tetrathionate in alkaline solutions at 22oC was basically the same as reported by Rolla and Chakabarti (1982) but with a different rate constant k= 1.38 x 103 L mol-1 s-1. The difference was explained by the absence of dissolved oxygen in the studies by Zhang and Dreisinger (2002). 19 Trithionate Decomposition of trithionate in the pH range of 5.6 and 12 at 50oC forms thiosulphate and sulphate (Kurtenacker et al., 1936; Ahern, 2006). However, in boiling alkaline (pH >=13) solution, thiosulphate and sulphite are the major species (Smith and Hitchen, 1976). S3O62- + 2OH- = S2O32- + SO42- + H2O 2- 47 2S3O62- + 6OH- = S2O32- + 4SO32- + 3H2O 2- 48 The hydrolysis of trithionate at pH 10 and 80oC was studied by Rolla and Chakrabarti (1982). The results agreed closely with the stoichiometry of the hydrolysis reaction leading to sulphate. A fundamental study of trithionate decomposition under various conditions was studied by Ahern (2005). Typical profiles of trithionate degradation and corresponding thiosulphate formation were linear over the 3 h of reaction time tested. The reaction order was identified using standard graphical methods and was found to be the first order with respect to the trithionate concentration. At higher hydroxide concentrations (>0.1 M, i.e. pH values of 12 and higher), degradation of trithionate was very fast and impossible to measure. The ratio of thiosulphate formed to the amount of trithionate reacted corresponded to reaction-forming sulphite, which is in agreement with literature observations of Rolla and Chakrabarti (1982) of the reaction of trithionate at high pH for the reaction at 50oC and pH 13.4-13.7. The degradation of trithionate at 40oC for hydroxide concentration of < 0.01M was modelled by Ahern (2005) using the following rate equation: -d[S3O62-]/dt = (ko +k1[OH]-)[S3O62-] 2- 49 20 where ko = 0.012h-1 and k1 = 0.74M-1h-1 Sulphite According to Smith and Hitchen (1976), sulphite oxidizes to sulphate according to the reaction below: SO32- + ½ O2 = SO42- 2- 50 A buildup of sulphite and thiosulphate at alkaline pH can be postulated due to the fact that these ions are intermediates in the suphur-oxidation pathway, and complete oxidation to sulphate is, to some extent, prevented in the high pH range. 2.3.4 PYRITE CATALYZED DECOMPOSITION OF THIOSULPHATE In the presence of pyrite and dissolved oxygen, aqueous thiosulphate decomposes. The experimental results by Xu and Schoonen (1995) clearly indicate that the decomposition rate of thiosulphate is accelerated in the presence of pyrite. The overall reaction was expressed as: 2S2O32- + 0.5O2 + 2H+ = S4O62- + H2O 2- 51 The oxidation of thiosulphate to tetrathionate by dissolved oxygen on the pyrite surface has a first-order dependence on the surface concentration of pyrite and a transitional-order dependence on the thiosulphate concentration (zero order at high thiosulphate concentration and low pyrite surface concentration and first order at low thiosulphate concentration and high pyrite surface concentration). This rate data fits a Langmuir-21 Hinshelwood rate law which is consistent with a heterogeneous surface-controlled reaction mechanism according to Xu and Schoonen (1995). R = ka Ks {FeS2} [S2O32−]1+Ks[S2O32−] 2- 52 where ka = 2.3 x 10-8 to 2.84 x 10-8 (M m-2 sec-1) is the apparent rate constant and Ks = 2.24 x 104 is the equilibrium constant of the adsorption equilibrium. 2.3.5 COMPLEXATION OF GOLD BY AQUEOUS SULPHUR COMPOUNDS In sulphur containing solutions, gold may form the following complexes: [Au(H2S)]+, [Au(H2S)HS], [Au(HS)], [AuS]-, [Au(HS)2]-, [Au(H2S)2S]2-, [Au(SO3)2]3-, [Au(S2O3)2]3-, and Au2S (Gudkov et al., 2010). Solubilization of gold by sulphide occurs in the presence of polysulphide according to the following reaction (Stacey, 2005): 2Au + S22- + 2S2- = 2AuS- + 2S2- 2- 53 In this reaction, polysulphide is an oxidant for gold. By hydrolysis, the sulphide transforms into hydrosulphide which is a complexant for gold. Gudkov et al. (2010) claims that the oxidation of gold by the oxygen from air in the presence of HS- is thermodynamically possible. However, the actual mechanism most likely includes the transformation of hydrosulphide into a polysulphide first, which interacts with the gold. In this reaction, polysulphide is presented both as an oxidant and a complexant for gold. S2- + H2O = HS- + OH- 2- 54 2Au + 2HS- + 1/2O2 = 2[AuS]- + H2O 2- 55 2Au + S22- = 2[AuS]- 2- 56 22 In an alkaline sulphite medium, the gold sparingly dissolves and can be represented by the following reaction (Gudkov et al., 2010) 2Au + 4SO32- + 1/2O2 + H2O = 2[Au(SO3)2]3- + 2OH- 2- 57 The stability constants, reversible potentials, and operating pH range for some of these complexes are given in Table 1 (Aylmore, 2014). Table 1: Gold complexes According to these data, the reversible potentials for these three gold complexes are far apart. Thiosulphate forms a stronger complex with gold under mild oxidative potentials compared to gold sulphite. The complexation of gold with sulphide appears under reducing conditions. The stability diagram for gold sulphide and thiosulphate are presented in Figure 5. Gold complexes log K Eo, Au+1, +3/Au (V vs. SHE) pH Au(S2O3)23- 28.7 0.17 8 to 10Au(HS)2- 29.9 -0.25 <9Au(SO3)23- 15.4 0.77 >423 Figure 5: Stability fields for gold (10-4 M) –sulphur (1M) complexes by STABCAL (25oC) According to this EH-pH diagram, gold thiosulphate is stable in a wide range of pH and potential. With respect to gold dissolution in thiosulphate, the reaction always requires an oxidant which, in most cases, is oxygen, and may be presented as follows: 4Au + 8S2O32- + O2 + 2H2O = 4Au(S2O3)23- + 4OH- 2- 58 This reaction has been shown to have slow kinetics. The reaction rate was found to be controlled jointly by diffusion of the dissolved oxygen and the chemical reaction between gold and thiosulphate/dissolved oxygen. As is usually the case, fast leaching rates are observed with higher strength of lixiviant. However, at high concentrations of thiosulphate, degradation products are formed from a reaction with oxygen and accumulate in the solution. These degradation products are predominantly polythionates, 24 which negatively impact the recovery of gold by ion-exchange resin technology (Fleming, 2003). For example, the formation of tetrathionate can be presented as: 4S2O32- + O2 + 2H2O = 2S4O62- + 4OH- 2- 59 The leaching rate of gold can be accelerated by increasing the temperature in the range of 25-60oC (Breuer and Jeffrey, 2000; Cao et al., 1992; Tozawa et al., 1981; Flett et al., 1983; Zippelian et al., 1988; Zhang, 2004). However, this increase in temperature also increases the rate of thiosulphate decomposition via oxidation. At temperatures higher than 80oC, thiosulphate decomposes rapidly, according to Klets et al. (1987), and even at ambient temperature, it is impossible to avoid some oxidation of thiosulphate under the conditions that are suitable for gold leaching. Thiosulphate starts to oxidize at about 0.08V vs. SHE, which is lower than the potential required for gold oxidation (0.15V vs. SHE). When the potential increases to 0.4 V (vs. SHE) and higher, the thiosulphate rapidly oxidizes to various sulphur oxyanions. Therefore, careful control of the redox potential is necessary to minimize the loss of thiosulphate. According to Webster (1984), the greatest stability of thiosulphate occurs at pH >7 and EH -0.4 to 0.1V. Many experimental works confirm that the gold dissolution rate is enhanced in the presence of copper and ammonia and largely controlled by diffusion rate (Quyang, 2001; Li, 2003). During leaching, copper ions in solution transfer back and forth from cupric ammine to cuprous thiosulphate, depending on the redox potential of the solution. The addition of ammonia was found to stabilize copper as a cupric ammine species. The role of copper (II) ions in the oxidation of gold is shown in the following reaction: Au + 5S2O32- + Cu(NH3)42+ = Au(S2O3)23- + 4NH3 + Cu(S2O3)35- 2- 60 In addition, the oxidation reaction promoted by copper is (Aylmore and Muir, 2001): 25 2Cu(NH3)42+ + 8S2O32- = 2Cu(S2O3)35- + S4O62- + 8NH3 2- 61 Therefore, any oxidant used for improving gold leaching kinetics would increase thiosulphate consumption. It is obvious that the thiosulphate leaching system is complex and requires a high degree of control of variables such as pH, potential via the copper (II)/copper (I) ratio, ammonia/ ammonium equilibrium, and oxygen and thiosulphate concentrations. 2.4 GOLD THIOSULPHATE ELECTROCHEMISTRY Dissolution of gold is an electrochemical process. Therefore, the fundamentals of the gold dissolution reaction in thiosulphate are often studied by applying electrochemical methods. 2.4.1 ELECTROCHEMICAL STUDIES The leaching of gold in a thiosulphate solution under normal atmospheric conditions without any catalysts has been found to progress very slowly (Zelinsky and Novgorodtseva, 2013). Various studies have been undertaken in order to determine the cause for such slow kinetics. One hypothesis concerns the possible difficulties of the positioning of two ligands in such a way that would satisfy a linear bridging with metal during leaching. Extensive work has also been done in determining whether the decomposition of thiosulphate products affects the gold leaching process or the reaction itself is inherently slow. This section explores key data from the literature to examine these hypotheses. 26 A typical voltammogram for gold in thiosulphate includes two anodic processes and is presented as a current calculated based on gold loss or gain on the electrode and a current representing the oxidation or reduction of thiosulphate during potential sweeping. Many researchers have been using the EQCM (Electrochemical Quartz Crystal Microbalance) or the REQCM (Rotating Electrochemical Quartz Crystal Microbalance) in order to differentiate these currents and identify the one representing the gold dissolution alone. These techniques allow calculating the current corresponding to gold dissolution based on the gold electrode mass change. The dissolution of gold occurs mainly within the potential range from OCP (open circuit potential) to ~ 0.5 V vs. SHE (Zelinsky and Novgorotseva, 2013), which is similar to the potential range reported by other researchers. The voltammogram taken in a 0.1 M Na2S2O3 at 10 mV/s scan rate indicated that the gold anodic current (calculated based on EQCM data) was very low with the peak current of 15 µA cm-2 at about 0.3 V vs. SHE (Zelinsky and Novgorotseva, 2013). The peak current is due to mass transport usually observed when using a stationary electrode. But when a rotating electrode is used there is often no peak in the scan. Instead, the current reaches its’ limiting value because of the controlled mass transfer process which provides a diffusion layer of constant thickness. Hence, the dependence of the gold anodic current on potential with rotating electrode is expected to have a different shape from that under stationary electrode conditions and the values of the gold dissolution rate will be higher than the results of stationary measurements due to the enhanced mass transfer. Using the REQCM, the gold anodic current of ~ 25 µA cm-2 was observed in the solution of 0.2 M sodium thiosulphate at a 300 rpm rotation rate and a 1 mV/s scan rate in the potential range of 100-400 mV (Chandra and Jeffrey, 2004). The higher current with the rotating gold disk ~ 60 µA cm-2 was seen at 0.25 V vs. SHE at 200 rpm and 5mV/s scan 27 rate in solution containing 1 M Na2S2O3 with 0.1 M NaOH background electrolyte by Zhang and Nicol (2003). The lowest current of ~ 1.4 µA cm-2 and 4 µA cm-2 was published by Baron et al. (2011) using a stationary electrode at potentials close to OCP in 0.1 M Na2S2O3 with 1 x 10-4 M NaOH and in 0.1 M Na2S2O3 with 20 mg/L Cu, respectively. It is apparent that the gold anodic current is low, ranging between 4 and 60 µA cm-2, depending on experimental conditions and methods applied. A higher gold anodic current can be achieved with higher thiosulphate concentration (Zhang and Nicol, 2003; Chandra and Jeffrey, 2004; Breuer and Jeffrey, 2002). The concentration of thiosulphate has an effect on the gold thiosulphate reduction current as well. Gold thiosulphate reduction from a solution containing 50 mg/L gold and 50 mM Ammonium thiosulphate, rotation rate of 300 rpm, 30oC, and scan rate of 1 mV/s was studied by Choo and Jeffrey (2003). The increase in the reduction current was noted at potentials more negative than -300 mV. At -600 mV, the current density was 4.5 A m-2 and increased to 13.5 A m-2 at -700 mV. The actual current corresponding to the gold cathodic reduction calculated based on the REQCM data reached a diffusion-limiting value of (-0.96) A m-2 by (-700) mV. In studying the effect of thiosulphate anions on the reduction of gold thiosulphate, it was observed that the reduction of gold thiosulphate does not occur as readily in solutions containing free thiosulphate because of high overpotential that was required to obtain an appreciable gold deposition rate (Choo and Jeffrey 2003). Sullivan and Kohl (1997) reported that the reduction voltammogram changes in accordance with the Nernst equation with increasing thiosulphate concentration. The Nernstian response shifted the reduction potential approximately 35 and 110 mV in the negative direction, as the thiosulphate concentration changed twofold and tenfold, respectively. For example, the reduction current commenced at about 0.15 V when there was no added thiosulphate but with addition of 0.1 M thiosulphate, the reduction current 28 was shifted to ~ -0.1 V. Thus, the presence of free thiosulphate shifts the gold thiosulphate reduction current towards more negative potentials from the gold thiosulphate reversible potential (0.15 V vs. SHE) and the shift is larger with higher free thiosulphate in electrolyte. Hence, the concentration of thiosulphate is an important variable for both gold dissolution and reduction processes. It increases the gold anodic current reaching at a peak at ~ 0.25-0.3 V vs. SHE and increases the overpotential for gold reduction. Several theories have been developed to explain such a slow dissolution of gold in thiosulphate. One of the theories concerns the instability of thiosulphate in the region of gold anodic activity. It suggests that thiosulphate itself decomposes, forming products that are deposited or adsorbed on the surface of gold and responsible for the slow kinetics of gold dissolution. According to Zhang and Nicol (2003), the thiosulphate decomposes, forming sulphite and aqueous sulphide, with the latter being oxidized to sulphur, which deposits on the surface of gold. The sulphur is not stable under high pH and when the increase in anodic current was detected (Zhang and Nicol, 2003) in a highly alkaline electrolyte, it was explained by partial elimination of sulphur-like films from the gold surface. The fact that the reaction rate depends on thiosulphate concentration implies that the slowest chemical step is dependent on the thiosulphate concentration. In a recent publication by Zelinsky and Novgorodtseva (2013), an original theory was proposed which seeks reasons for slow kinetics related to the coordination chemistry of gold and thiosulphate. Their argumentation was based on the fact that other metals that form multiple coordination compounds with thiosulphate such as silver and copper are characterised with facile kinetics, whereas gold thiosulphate that forms a linear coordination, is characterised with slow kinetics. The advocates of this theory reject the notion of a deposited sulphur film (product of thiosulphate oxidation), preventing gold 29 dissolution in the gold redox activity potential region. They claim that a sulphur film should have the same negative effect on silver and copper dissolution rates in thiosulphate, which is never the case. However, these metals dissolve (Choo and Jeffrey 2003) well below the thiosulphate oxidation region (< 0.08V). Therefore, there should not be any such film to affect the kinetics of these metals dissolution reactions. From thermodynamics, the standard potential of thiosulphate to tetrathionate is 0.08 V vs. SHE, which is below the gold thiosulphate reversible potential (0.15 V vs. SHE). Using cyclic voltammetry and Auger electron spectroscopy, Pedraza et al. (1988) observed that the thiosulphate ion decomposes in contact with gold at open circuit potential, leaving a film of several sulphur-containing species. However, based on his observation these species do not block the charge transfer reactions at the electrode. However, as the potential was cycled between 0 and 0.65 V, a non-conducting sulphur layer formation was evidenced, which displaced these sulphur-like species on the gold surface and caused a substantial decrease in anodic current. The changes on the electrode surface during electrochemical experiments using the SERS (Surface Enhanced Raman Scattering Spectroscopy) technique were observed by Watling (2007). The collection of SERS spectra from gold electrodes in sodium thiosulphate solution showed that at a potential of 210 mV, the bands corresponding to both thiosulphate and tetrathionate were present. At a potential of 350 mV, weak bands of thiosulphate were still evident, while those arising from the tetrathionate-like species were no longer displayed. At 410 mV, thiosulphate was still present but barely discernible and shifted to higher wavenumbers assigned to S8. Furthermore, the time-dependant spectra acquired from 0.1 mol dm-3 Na2S2O3 at potentials of 210 mV and 310 mV after a 2-minute detected S-S band was assigned to thiosulphate and tetrathionate-like species. With time, the tetrathionate-like band diminished and a band assigned to Au-S was seen to develop. At the higher potential (310 mV), the gold-sulphide bands developed more rapidly. Hence, the SERS technique confirms the occurrence of a msulphur-like species on the gold surface at the gold anodic dissolution potentials. 30 The influence of sulphur-like species on gold dissolution can be eliminated by the addition of 10 mM thiourea in 0.1 M sodium thiosulphate according to Watling (2007). With thiourea, a rapid rise in the gold anodic current was observed and this effect was explained by the disappearance of the S-S bond confirmed by the SERS technique. Similar observations were published by Baron et al. (2011). In her study using the SERS technique, she detected the presence of tetrathionate and trithionate on a surface of gold immersed in thiosulphate solution. However, in the presence of thiourea, these species were no longer detected on the surface of gold. According to Baron et al. (2011), thiourea adsorbs on the gold surface with a perpendicular orientation and forms a complex with gold at the metal surface through a sulphur atom. There was no further description of a catalytic effect of thiourea given in this paper. It might be assumed that the promotion of gold thiosulphate formation only implies the elimination of polythionates passive film on the surface. A comprehensive study of thiourea adsorption on gold electrodes by Wronlowa and Green (1963) provides more interpretation that can be used to explain the influence of adsorption on the rate of electrochemical processes. The adsorption of thiourea from a sulphate-supportive electrolyte on a gold electrode was studied using a radiotracer method. This study showed that the thiourea upon adsorption on gold displaced water and locally distorted the shape of the double layer, causing nonuniform potential distribution. Perhaps the similar distortion of a double layer takes place on the gold surface in the thiosulphate solution affecting the gold dissolution rate and the displacement of polythionate ions could only be a secondary effect. The catalytic effect of thiourea on gold dissolution in thiosulphate solution has been observed in a number of studies, however, the exact mechanism by which the addition of thiourea to the thiosulphate leaching system that enhances the gold leaching is still not entirely clear and warrants further research. In recent publications, the effect of thallium (I) salts on the electrochemical properties of gold has been reported by several researchers. The increase of the gold dissolution 31 current has been noted and explained by the the presence of Tl species in a two-valence state at the electrode surface, causing a catalytic enhancement of the gold oxidation reaction. The two oxidation states (+1, +3) provide an electrochemical couple for catalysis for gold oxidation. Kabasakaloglu and Bilgic (1990) published electrochemical data that showed that in the anodic region of chemisorbed AuO formation, the oxidation of Tl(I) to Tl(III) occurs, which affects the formation of the surface gold oxide film. This reaction in the acidic media was observed at potentials more positive than 1.3 V. However, much lower potential is required in the alkaline electrolyte to oxidize the chemisorbed Tl(I) to Tl(III) oxide or hydroxide. The diagram in Figure 6 shows the upper limit of thallium (I) to thallium (III) conversion potential as a function of pH. Figure 6: Stability diagram for 1M thallium by STABCAL (25oC) 32 According to Bek and Shevtsova (2013), in nearly neutral solutions of thiosulphate, the oxidation of chemisorbed Tl(I) (β2 = 3.1) with the formation of Tl(III) (β4 = 41) complexes with thiosulphate can be presented by the following reaction: Tl(S2O3)23- + 2S2O32- -2e- = Tl(S2O3)45- 2- 62 The equilibrium potential of this reaction at equal concentrations of both complex ions (5 x 10-5 M) in 0.05 M thiosulphate is approximately 0.2 V vs. SHE. This corresponds with the potential at which the gold dissolution rate starts to increase. Thus, at the potential of E > 0.1 V, the catalytically-active adsobed Tl(III) ions associated with thiosulphate can accumulate on the gold surface by the attachment of thiosulphate anions. These authors claim that the reaction rate increases with increasing potential and weakly depends on the concentration of Tl(I) in the bulk solution. The effect of thallium additions in the concentration range between 5 x 10-6 M and 10-4 M on the anodic dissolution of gold in sodium thiosulphate with the concentration from 0.005 to 0.2 M was studied with the quartz crystal microbalance by Bek and Shevtrsova (2012). Their studies have shown that the addition of a thallium compound sharply accelerated the anodic gold dissolution. The gold anodic dissolution rate in 0.2 M thiosulphate increased from 0.02 mA cm-2 (in the absence of thallium) to 0.75 mA cm-2 after the addition of 7.5 x 10-5 M thallium. They also reported that the side reaction of thiosulphate oxidation to tetrathionate remained slow in this potential range and had a similar rate to experiments in the absence of thallium. The transfer coefficient of 0.57 was calculated for gold oxidation in thiosulphate in the presence of thallium. This value is close to the typical value for the majority of electrochemical reactions, but it is very high compared to the published value of 0.23 for gold in the thiosulphate electrolyte alone (Sullivan and Kohl, 1997). Such variations in transfer coefficient values has been observed in the electrocatalysis of anodic dissolution of gold in cyanide solutions by chemisorbed heavy metal ions (Tl+, Pb2+, Bi3+, Hg2+). Hence, the same mechanism (chemisorption) might be responsible where these metals and thallium accelerate the gold dissolution. 33 2.5 OBJECTIVES OF CURRENT STUDY There were two objectives set in this study: (1) develop a kinetic equation for sulphide conversion to thiosulphate and conditions for gold leaching with in-situ thiosulphate and (2) investigate the effect of catalysts on gold leaching in thiosulphate and the phenomenon of gold thiosulphate complex instability that is characteristic to this system using the electrochemical methods. The conversion of sulphide sulphur of pyrite in an alkaline solution forms several metastable aqueous sulphur species, as described in literature review section. Due to a complex interaction between these sulphur oxyanions, there is a lack of consistency in the proposed theories describing the sulphide oxygen reaction yielding thiosulphate. Hence, the first objective of the current study was to determine the optimal conditions for an in-situ generation of thiosulphate during pyrite oxidation and the use of this thiosulphate to simultaneously or sequentially leach the gold. This is thought to offer an interesting possibility for the process development in the gold sector. It is well established that the excess-free thiosulphate positively affects the gold leaching kinetics and stabilizes the gold as a gold thiosulphate complex. However, when thiosulphate is decomposed, it reprecipitates the gold. The exact nature of this phenomenon is not known. The second goal of this study was to determine the electrochemical reversibility of the gold thiosulphate system and to study the effect of two relatively novel catalysts on gold dissolution/reduction kinetics in thiosulphate using electrochemical methods. Such knowledge would provide the opportunity for better understanding and control of the thiosulphate leaching of gold. 34 3 EXPERIMENTAL METHODS 3.1 APPARATUS 3.1.1 OXIDATION AND LEACHING REACTOR SETUP The pyrite oxidation and gold leaching experiments were performed in a sealed 3 L glass reactor suitable for low pressure testing. The reactor was equipped with a pressure gauge, agitation, and sampling ports. The oxygen pressure was introduced through the inlet port and flashed by an off-gas line. The reactor was placed in a heating mantle connected to the temperature controller. The temperature, pressure, and agitation were controlled throughout the test. The agitation was fixed at 255 rpm, the temperature maintained at 80oC, and the oxygen over pressure was varied between 10 and 40 psig. For the controlled pH experiments, a glass burette containing sodium hydroxide and a pH probe were included in the setup. A photograph of the reactor is shown in Figure 7. The schematics are shown in Figure 8. 35 Figure 7: Alkaline leach experimental setup 36 Figure 8: Schematics of reactor 3.1.2 ELECTROCHEMICAL SETUP The Electrochemical Quartz Crystal Microbalance (EQCM) was used to study the electrochemistry of gold oxidation and reduction from thiosulphate solutions. The working electrode was a gold-coated quartz crystal. Microbalance studies were carried out using a 0.198 cm2 standard finished 9 MHz AT cut gold (sputtered on Ti ~300nm thickness) electrode. A new gold resonator (working electrode) was used for each experiment. The quartz crystal microbalance instrument QCM 922 (Princeton Applied Research) was connected to a model 273A (Princeton research) Potentiostat & Galvanostat. The experimental setup is shown in Figure 9 and schematics in Figure 10. 37 Figure 9: Electrochemical cell experimental setup Figure 10: Electrochemical cell schematics 38 A typical three electrode setup was used with an Ag/AgCl reference electrode, a graphite counter electrode, and a gold-plated quartz crystal resonator as a working electrode. During the electrochemical experiment, argon was continuously bubbling at a slow rate to prevent air ingress. The cell is shown Figure 11. Figure 11: Three electrode setup 39 3.2 EXPERIMENTAL METHOD 3.2.1 OXIDATION AND LEACHING Pyrite concentrates from two different origins were used to study alkaline oxidation of pyrite. The first concentrate (PY1) contained 6.3 g/t Au (15% CN recoverable determined in a typical carbon-in-leach experiment with 1 g/L cyanide added and maintained during the leach) and 45.7% sulphide ground to a P80 size of 14 µm. The second concentrate (PY2) contained 95 g/t Au (93% CN recoverable determined in a typical carbon-in-leach experiment with 1 g/L cyanide added and maintained during the leach) and 21.6% sulphide with a grind size P80 of 16 µm. The analyses of the concentrates are given in Table 2 and Table 3. Table 2: Composition of Pyrite Concentrates - quantitative analysis Table 3: Composition of Pyrite Concentrates ICP Scan Au Ag Fe Cu S(Total) S2- SO4 S° C(Total) C(graphitic) TOC CO3g/t g/t % % % % % % % % % %PY 1 6.3 16.2 36.8 0.61 45.7 45.7 1.0 < 0.05 < 0.01 < 0.01 < 0.05 < 0.05PY 2 95 132 19.2 0.66 22.4 21.6 1.7 - 1.12 - - 4.54SampleAl As Ba Be Bi Ca Cd Co Cr K Li Mg Mng/t g/t g/t g/t g/t g/t g/t g/t g/t g/t g/t g/t g/tPY 1 4800 0.066 77 < 0.07 55 835 < 5 50 25 264 < 5 26 69.1PY 2 13400 3500 85.4 1.87 < 20 24400 129 59 45 5130 < 20 2500 49900Mo Na Ni P Sb Se Sn Sr Ti Tl U V Yg/t g/t g/t g/t g/t g/t g/t g/t g/t g/t g/t g/t g/tPY 1 < 5 235 < 20 < 200 29 129 < 20 189 1380 < 30 < 80 29 0.9PY 2 < 20 304 141 < 80 67 < 30 < 30 42.1 633 < 40 < 30 20 6.2SampleSample40 At the start of the pyrite oxidation experiment, the required amount of pyrite concentrate was added to the sodium hydroxide solution or deionised water. Thereafter, drops of sodium hydroxide were added periodically to ensure the pH remained at target value. Samples were withdrawn periodically, filtered, and the filtrates submitted to the SGS analytical laboratory for sulphur species analysis by Dionex DX-120 chromatography using methods developed by Environmental Services Canada (Determination of Thiosulphate, Trithionate, Tetrathionate and Thiocyanate in Aqueous Samples by Ion Chromatography,‘ME-CA-[ENV]IC-LAK-AN-004’; Determination of Anions in Aqueous, Soil, Leachate, and Sludge Samples by Ion Chromatography ‘ME-CA-[ENV]IC-LAK-AN-001’). The accuracy of thiosulphate analysis is in the range of 0.8-1.5% and for other thiosalts between 0.5-4.5%, depending on the species. The solids were submitted for sulphide analyses performed by a Leco analyzer. Metals in solids were analyzed by Fire Assay for gold and silver and X-Ray Fluorescence (XRF) for copper. The solution analysis was performed by an Atomic Absorption Spectrometer and ICPOES. The analysis of all the solids and solutions have been conducted by the SGS Canada certified laboratory in Lakefield, Ontario. 3.2.2 ELECTROCHEMICAL MEASUREMENTS The electrochemical experiments were conducted using the Quartz Crystal Microbalance. The principle of operation of the EQCM is based on recording the resonant frequency change (∆f) of the quartz crystal as a function of potential. Using the proportional relationship known as Sauerbrey’s equation, the frequency change can be converted into electrode mass change (∆m): ∆f= -Cf ∆m 3- 1 The theoretical value of the conversion factor (Cf) is usually known but depends on the medium in which crystal is operating because the density and viscosity of a solution 41 affect the frequency (Bard and Faulkner, 2000). Furthermore, the Sauerbrey’s equation is applied when the mass added or lost at the oscillator surface does not experience any shear deformation (Deakin and Buttry, 1989) which may be the case for thin, rigid (Bizzotto, 2015) layers forming on the surface. It is always recommended to compare the theoretical value with an experimentally-obtained conversion factor. In order to obtain the experimental value, the method of copper underpotential deposition (Cu –upd) was used. In this method, copper is deposited from 2 mM CuSO4 and 0.1 M H2SO4 solution by holding the potential at 0.3 V vs. SHE for 20 minutes followed by a linear anodic polarization to record the copper dissolution voltammogram (Lam et al., 2011). This linear anodic sweep for copper dissolution was performed in the potential range 0.3 to 1.2 V vs SHE at 15 mV/sec scan rate. For comparison, the same procedure was repeated but in 0.1 M sulphuric acid solution with no copper added. Both voltammograms are show in Figure 12. Figure 12: A reference cyclic voltammetry scan at 15 mV/s in 0.1 M H2SO4 solution and anodic dissolution voltammetry at a 15 mV/s scan rate of Cu-upd layer deposited on the same electrode after 20-minute exposure at 0.3 V using 2mM CuSO4 and 0.1 M H2SO4 solution -4.0E-06 -2.0E-06 0.0E+00 2.0E-06 4.0E-06 6.0E-06 8.0E-06 1.0E-05 1.2E-05 0 0.2 0.4 0.6 0.8 1 1.2 1.4 I / A E / V vs SHE eletro dissolution reference l ctro dissolution blank 42 The cumulative charge for copper electrodissolution was calculated by integrating the curve in Figure 12, and was used to determine the experimental calibration constant using the following relationship (Faraday’s law applied to Sauerbrey’s equation): Cf = 𝐌(𝐂𝐮) 𝚫𝐪𝐧 𝐅 𝚫𝐟 3- 2 In this relationship, F is Faraday’s constant (96 484.5 C equiv -1); M is the atomic weight of Cu (63.547g mol-1); and n is the number of electrons (n=2). ∆q was calculated as the area under the current-time curve (8.377E-05 C) and ∆f (30.0295 Hz) was obtained from the EQCM apparatus. Using experimentally-obtained ∆q and ∆f values, the experimentally-obtained conversion factor Cf is calculated as 1.088 Hz ng -1. The conversion factor provided by Princeton Applied Research was 1.068 Hz ng-1. The experimental value is only 1.87% different from the theoretical value, which is considered acceptable. All experiments were conducted at room temperature in near-neutral (pH 7.8) solutions prepared from analytical grade magnesium thiosulphate hexahydrate and ultrapure 18 MΏ (megaohm) cm water. Other reagents used for electrochemical studies included thallium sulphate and thiourea. These reagents were both analytical grade purchased from Alfa Aesar. The solution was gently agitated during potentiodynamic cycling by bubbling nitrogen slowly through the cell. The differences in the resonant frequency of the crystal electrode were recorded as a function of potential and the corresponding mass changes were calculated using the appropriate conversion factor. The potential was measured against the silver/silver chloride reference electrode and converted to the standard hydrogen electrode scale by adding 0.197 V to the value. Prior to each experiment, the working electrode (gold-plated crystal) was subjected to electrochemical cleaning by potential cycling in 0.1 M NaOH between 0.945 and -0.125 V at 0.1 V/s until a stable voltammogram was obtained. The last cycle was stopped at the negative potential -0.125 V in order to maintain the surface free of oxides (Pedraza et al., 1988). Each 43 voltammogram presented in this thesis has been confirmed by three overlaid voltammograms using an identical electrolyte and testing conditions. 44 4 PYRITE OXIDATION REACTION 4.1 INTRODUCTION In the treatment of sulphidic refractory gold ores or concentrates, it is generally necessary to first oxidize the sulphides and liberate the gold that is trapped as fine (< 1 micron) particles within the host sulphide matrix. When sulphides are oxidized under mild conditions, one of the oxidation products is the thiosulphate anion and, since this compound is a well known lixiviant for gold, the possibility exists of achieving both objectives of liberating and leaching gold simultaneously. The following experiments were designed to determine the conditions for pyrite sulphide conversion to thiosulphate. 4.2 RESULTS 4.2.1 EFFECT OF OXYGEN PRESSURE The oxidation of pyrite was conducted at various oxygen overpressures in a reactor without in-situ pH control. In this series of experiments, about 100 g of pyrite concentrate (PY1) containing 45.7% sulphides was used. The particle size of the concentrate was P80 14 micron. Deionised water was added to the concentrate to make a pulp of approximately 8.3% solids content. About 29 g/L of 50% NaOH was initially added to the pulp. The pulp was heated to 80oC and the target oxygen pressure was applied. About 3.2 g of 50% NaOH was added on an hourly basis. The total addition of NaOH was 40 g/L (100% basis). The pulp samples (~5 g dry basis) were taken after 1, 2, 4, and 8 h of reaction time. The samples were immediately filtered. The filtrates were collected and the pH/ORP was measured and recorded. The solution samples were submitted for immediate thiosulphate analysis by ion chromatography using the Dionex unit (at SGS Canada). The filter cakes were washed several times with deionised water to displace 45 residual solution. The washed residues were dried, weighed, and submitted for sulphide assays by a Leco analyzer. Intermediate soluble sulphur species, as well as residual sulphides, were measured and results are given in Table 4 and Table 5. Table 4: PY1 Sulphide oxidation at various oxygen overpressures - testing conditions Test Time Solids Volume Solids pH EHh % ml g mV g cum , molExp 17 0 8.5 96.2 29 0.71 1039 96.2 12.6 -86 29 0.72 985 91.5 12.8 -101 31 0.84 922 85.7 12.7 -150 34 0.98 871 80.8 12.3 -188 40 1.020 psig pO 2Exp 1 0 8.3 94.5 29 0.71 1039 94.5 12.5 -172 31 0.82 991 88.1 12.9 -149 33 0.84 937 83.0 12.7 -119 36 0.98 880 78.1 12.9 -101 40 1.010 psig pO 2Exp 18 0 8.3 94.0 29 0.71 1039 94.0 12.7 -141 31 0.82 989 89.6 12.5 -141 32 0.84 934 84.7 12.6 -142 36 0.98 878 79.3 12.8 -150 40 1.0NaOH addition40 psig pO 246 Table 5: PY1 Sulphide oxidation at various oxygen overpressures - thiosulphate yield Table 5 shows the analytical assays for sulphides in solids and thiosulphate in solution. The moles of oxidized sulphides reported are the difference between the initial and the time-dependent remaining sulphides in the solids. The ratio of caustic to sulphides is based on the cumulative caustic added and the moles of corresponding oxidized sulphide. In these three tests, the ratio of added caustic-to-oxidized sulphides remained above 1.3. The profiles of oxidized sulphides and formed thiosulphate as a function of reaction time are presented in Figure 13 and Figure 14. Test Time pH OH- / ΔS2-h mol/mol % mol mg/L molExp 17 0 45.7 1.37 0.0 0.01 12.6 6.6 42.0 1.26 1800 0.0172 12.8 3.3 39.9 1.14 4400 0.0394 12.7 2.0 35.2 0.94 11000 0.0918 12.3 1.3 23.1 0.58 19000 0.14820 psig pO 2Exp 1 0 45.7 1.35 0.0 0.01 12.5 8.2 42.5 1.25 1400 0.0132 12.9 4.0 41.7 1.15 3400 0.0304 12.7 2.5 38.5 1.00 7600 0.0648 12.9 1.7 30.3 0.74 16000 0.12610 psig pO 2Exp 18 0 45.7 1.34 0.0 0.01 12.7 13.8 43.8 1.29 3300 0.0312 12.5 5.9 43.0 1.20 2200 0.0194 12.6 3.5 41.1 1.09 4900 0.0418 12.8 2.3 36.8 0.91 12000 0.09440 psig pO 2S2- S2O3 47 Figure 13: Sulphide oxidation versus oxygen overpressure Figure 14: Thiosulphate formation versus oxygen overpressure 48 Figure 13 and Figure 14 show that the sulphide oxidized and thiosulphate formed were almost linear with time. The rates for both reactions were also increased with increasing oxygen overpressure. The thiosulphate concentrations of 12, 16, and 19 g/L were reached after 8 h of reaction at 10, 20, and 40 psig oxygen pressures, respectively. The percent of sulphides oxidized were ~32, 45, and 57% (with a margin of 5% error) after 8 h of reaction at 10, 20, and 40 psig oxygen pressures, respectively. The assays of sulphur oxyanions in these tests are presented in Table 6. Three major sulphur oxyanions (thiosulphate, sulphite, and sulphate) can be identified in Table 6. The concentration of thiosulphate was the highest among other species. The concentration of tetrathionate and trithionate remained below detection limit with the exception of the last 8 h sample in Exp 17, where the trithionate concentration was 2.6 g/L. Table 6: Thiosalts concentration during PY1 oxidation at various oxygen pressures Test Time pH EH S4O6 S3O6 SO3 SO4h mV % mg/L mg/L mg/L mg/L mg/LExp 17 0 - - 45.7 0.0 0.0 0.0 0.0 0.01 12.6 -86 44.1 1800 <2 <20 1100 16002 12.8 -101 41.8 4400 <20 <200 2400 37004 12.7 -150 36.8 11000 <20 <200 5500 50008 12.3 -188 24.1 19000 <20 2600 4600 1600020 psig pO 2Exp 1 0 - - 45.7 0.0 0.0 0.0 0.0 0.01 12.5 -172 42.8 1400 <2 <20 950 15002 12.9 -149 41.7 3400 <20 <200 1800 31004 12.7 -119 38.5 7600 <20 <200 3900 39008 12.9 -101 32.0 16000 <10 100 7900 540010 psig pO 2Exp 18 0 - - 45.7 0.0 0.0 0.0 0.0 0.01 12.7 -141 42.4 3300 <10 <100 1600 10002 12.5 -141 43.0 2200 <10 <100 1400 25004 12.6 -142 41.1 4900 <10 <100 2500 27008 12.8 -150 36.8 12000 <10 <100 5400 390040 psig pO 2S2- S2O3 49 4.2.2 EFFECT OF SODIUM HYDROXIDE Table 7 and Table 8 collect the data for experiments run at 20 psig oxygen pressure on the PY1 sample using various sodium hydroxide addition rates. The goal was to observe the changes in the sulphide oxidation rate while minimizing the consumption of NaOH. In these tests, the oxygen pressure was identical but the amount of caustic added was different. Almost half of the NaOH amount added in Exp 1 was used in Exp 2 and NaOH was sufficient to keep the molar ratio of NaOH to sulphide above 1. As a consequence, the concentration of thiosulphate increased linearly with time. Further reduction of caustic, as in Exp 3, caused this ratio to drop below 1 and resulted in a loss of thiosulphate from 4.3 g/L to 1.4 g/L. When the pH dropped to 2.3 at 8 h, any attempt to increase pH by adding drops of NaOH was ineffective, possibly owing to the rapid consumption of NaOH by reactive sulphides. At that time, all the thiosulphate that had been generated under alkaline conditions was completely decomposed. Table 7: PY1 Sulphide oxidation at various NaOH additions - testing conditions Test Time Solids Volume Solids pH EHh % ml g mV g molExp 1 0 8.3 - 94.5 - - 29 0.731 1039 94.5 12.5 -172 31 0.772 991 88.1 12.9 -149 33 0.814 937 83.0 12.7 -119 36 0.898 880 78.1 12.9 -101 40 1.01Exp 2 0 7.9 - 89.1 - - 15 0.391 1039 89.1 12.6 -134 16 0.412 989 84.4 12.7 -141 17 0.434 934 79.5 12.5 -87 19 0.488 878 74.7 11.9 -136 22 0.55Exp 3 0 7.8 - 90.5 - - 9 0.221 1066 90.5 12.3 -101 11 0.272 1007 85.2 11.3 -39 12 0.294 951 80.1 8.2 99 14 0.348 892 74.9 2.3 - 16 0.41NaOH cumulative50 Table 8: PY1 Sulphide oxidation at various NaOH additions - thiosulphate yield Table 9: Thiosalts concentration during PY1 oxidation at various NaOH additions Test Time pH NaOH / ΔS2-h mol/mol % mol mg/L molExp 1 0 - - 45.7 1.35 0.0 0.001 12.5 8.2 42.5 1.25 1400 0.0132 12.9 4.0 41.7 1.15 3400 0.0304 12.7 2.5 38.5 1.00 7600 0.0648 12.9 1.7 30.3 0.74 16000 0.126Exp 2 0 - - 45.7 1.27 0.0 0.001 12.6 5.1 42.8 1.19 1500 0.0142 12.7 2.5 41.7 1.10 3300 0.0294 12.5 1.5 38.5 0.96 7800 0.0658 11.9 1.0 32.0 0.75 16000 0.125Exp 3 0 - - 45.7 1.29 0.0 0.001 12.3 1.8 40.4 1.14 2600 0.0252 11.3 1.2 39.4 1.05 4300 0.0394 8.2 0.9 37.1 0.93 1400 0.0128 2.3 0.9 36.8 0.86 <10 0.000S2- S2O3Test Time pH S2- S2O3 S4O6 S3O6 SO3 SO4h % mg/L mg/L mg/L mg/L mg/LExp 1 0 - 45.7 0.0 0.0 0.0 0.0 0.01 12.51 42.5 1400 <20 <20 950 15002 12.85 40.0 3400 <20 <200 1800 31004 12.72 36.3 7600 <20 <200 3900 39008 12.85 30.3 16000 <20 <200 7900 5400Exp 2 0 - 45.7 0.0 0.0 0.0 0.0 0.01 12.6 42.1 1500 <10 <100 970 12002 12.7 40.6 3300 <10 <100 1700 12004 12.5 39.0 7800 <10 220 3500 20008 11.9 31.7 16000 <200 2900 1100 7300Exp 3 0 - 45.7 0.0 0.0 0.0 0.0 0.01 12.3 40.4 2600 <10 170 1000 28002 11.3 39.4 4300 <10 1400 200 46004 8.2 37.1 1400 210 5800 <500 93008 2.3 36.8 <10 400 260 <500 1900051 Experimental data of sulphur oxyanions measured at certain time intervals are presented in Table 9 and plotted in Figure 15 through Figure 17. All data in these figures correspond to Table 9 but the graphs are expressed in mmol instead of g/L to magnify the effect of pH on species distribution over time. Figure 15: Sulphur oxyanions distribution in Exp 1 52 Figure 16: Sulphur oxyanions distribution in Exp 2 Figure 17: Sulphur oxyanions distribution in Exp 3 53 Figure 15 shows that there are three major species: thiosulphate, sulphite, and sulphate. As the pH dropped below 12 between 4 hour and 8 hour, the concentration of sulphite also decreased but the concentration of sulphate increased as shown in Figure 16. Figure 17 exhibits the trithionate wave peaking at pH 8.2 but then decreasing as the pH dropped to 2.3. In this figure, sulphate was the only species that continuously increased linearly with time and this was independent of pH. In the next series of experiments, the pH was controlled in-situ. In the agitated pulp under an oxygen pressure of 20 psig, the pH was maintained at 9.5 (Exp 11), 10.5 (Exp 12), and 11.5 (Exp 13). However, the actual pH of the solution, after separating liquid from solids, was quite different. The sulphidic nature of the solids might have imparted the acidic character to the pulp to explain the higher readings of the filtrate pH. For example, in Exp 12, the pH of the filtrate (pH 11.6-11.9) was 1 to 1.5 units higher than the operating pulp pH value of 10.5. However, the opposite was observed when operating at pH 9.5 in Exp 11, where a lower pH of 8.5-8.9 was observed in the filtrate. This might be attributed to the reactivity during filtration that seems to be rapid at this pH range. There was very little discrepancy (0.2-0.3 units) between the pulp and the filtrate pH values when operating at pH 11.5 (Exp 13). The testing conditions and results are shown in Table 10 and Table 11. 54 Table 10: PY1 Sulphide oxidation at various pHs - testing conditions Table 11: PY1 Sulphide oxidation at various pHs - thiosulphate yield Test Time Solids Volume Solids pH EHh % ml g mV g molpH 9.5 of the pulpExp 11 0 8.9 - 150 - - 0 0.001 1530 150 8.9 165 4 0.112 1476 144 8.4 204 6 0.164 1440 137 8.9 189 12 0.308 1436 132 8.5 211 21 0.51pH 10.5 of the pulpExp 12 0 8.8 - 150 - - 0 0.001 1560 150 11.6 60 9 0.212 1510 144 11.6 57 11 0.274 1478 138 11.8 16 16 0.398 1460 132 11.9 26 23 0.59pH 11.5 of the pulpExp 13 0 8.7 - 150 - - 0 0.001 1582 150 11.8 -84 12 0.292 1539 144 11.7 -72 16 0.394 1489 138 11.7 79 18 0.468 1462 132 11.7 66 25 0.62NaOH cumulativeTest Time pH NaOH / ΔS2-h mol/mol % mol mg/L molpH 9.5 of the pulpExp 11 0 - - 45.7 2.14 0.0 0.0001 8.9 0.9 43.3 2.03 340 0.0052 8.4 0.9 43.5 1.96 240 0.0034 8.9 0.9 41.8 1.79 520 0.0078 8.5 1.0 39.1 1.61 400 0.005pH 10.5 of the pulp 0.000Exp 12 0 - - 45.7 2.14 0.0 0.0001 11.6 2.2 43.6 2.04 1100 0.0152 11.6 1.2 42.6 1.92 2500 0.0344 11.8 1.1 41.8 1.80 4900 0.0658 11.9 1.0 38.2 1.58 9000 0.117pH 11.5 of the pulp 0.000Exp 13 0 - - 45.7 2.14 0.0 0.0001 11.8 3.6 44.0 2.06 1400 0.0202 11.7 1.7 42.7 1.92 2900 0.0404 11.7 1.2 40.6 1.75 5500 0.0738 11.7 1.0 36.5 1.50 9500 0.124S2- S2O355 The data in Table 11indicates that maintaining the pulp pH at 9.5 in Exp 11 resulted in a low yield of thiosulphate: 0.24-0.52 g/L. The final concentration of thiosulphate was much higher ~9-9.5 g/L in Exp 12 and Exp 13, where the pH was maintained at 10.5 (pulp) and 11.5 (pulp), respectively. These data provide evidence of a constant, very small amount of thiosulphate when the ratio of sodium hydroxide to oxidized sulphides was kept at ~0.9 in Exp 11 corresponding to pH ~9. However, there was no appreciable effect of pH detected on the sulphide oxidation rate. The rate of NaOH addition is depicted in Figure 18. It appears that the slope of these lines ranges from 0.45 to 0.59 mol/h. The extrapolation to zero time intercepts the y-axis at a higher value with a higher operating pH. Figure 18: Rate of NaOH addition for 9.5, 10.5, and 11.5 pulp operating pHs 56 Figure 19 shows the sulphide oxidation rate for the six experiments conducted on PY1 at 20 psig oxygen pressures. Figure 20 shows the thiosulphate formation rate for the same six experiments. Figure 19: PY1 Sulphide oxidation at various pHs Figure 19 shows that the pH has no significant effect on the initial rate of sulphide oxidation. The rate slightly declines after 4 h, corresponding to a pH between 9 and 11. The lowest oxidation point at pH 2.3 might not be due to a decline in rate but simply a result of thiosulphate decomposition. 57 Figure 20: PY1 Thiosulphate formation at various pHs In contrast to the sulphide oxidation, the pH plays a crucial role in stabilizing the thiosulphate. The thiosulphate concentration linearly increases only for the tests where the pH remained close to 12 (Figure 20). When pH dropped below 12, the concentration of thiosulphate rapidly declined. Maintaining pH at ~9 as in Exp 11, yielded very low but constant concentrations of thiosulphate ranging between 0.2 and 0.5 g/L throughout the 8-hour experiment. Experimental data of sulphur oxyanions measured at certain time intervals for the tests at constant pulp pH are presented in Table 12 and plotted in Figure 21 through Figure 23. 58 Table 12: Thiosalts concentration during PY1 sulphide oxidation at various pHs Test Time pH EH S4O6 S3O6 SO3 SO4h mV % mg/L mg/L mg/L mg/L mg/LpH 9.5 of the pulpExp 11 0 - - 45.7 0.0 0.0 0.0 0.0 0.01 8.9 165 43.3 340 <2 480 <50 36002 8.4 204 43.5 240 2.4 430 <50 73004 8.9 189 41.8 520 <2 1400 <50 100008 8.5 211 39.1 400 200 2700 81 16000pH 10.5 of the pulpExp 12 0 - - 45.7 0.0 0.0 0.0 0.0 0.01 11.6 60 43.6 1100 <2 97 390 21002 11.6 57 42.6 2500 <2 160 810 38004 11.8 16 41.8 4900 24 490 1200 44008 11.9 26 38.2 9000 29 840 1400 6500pH 11.5 of the pulpExp 13 0 - - 45.7 0.0 0.0 0.0 0.0 0.01 11.8 -84 44.0 1400 <2 150 490 21002 11.7 -72 42.7 2900 <2 310 770 37004 11.7 79 40.6 5500 <2 590 1400 39008 11.7 66 36.5 9500 <2 1400 1200 6500S2- S2O359 Figure 21: Sulphur oxyanions distribution in Exp 11 Figure 22: Sulphur oxyanions distribution in Exp 12 60 Figure 23: Sulphur oxyanions distribution in Exp 13 In Figure 21 the only species detected in an appreciable amount was sulphate. The profiles for all sulphur oxyanions in Figure 22 and Figure 23 are similar. There were two major species: thiosulphate and sulphate. In comparison from the tests run at pH >12.5, in these tests, the sulphite curve lies well below the curve of sulphate indicating that much less concentration of sulphite was formed at pH <12.5. 4.2.3 EFFECT OF RESIDENCE TIME Table 13 illustrates the results of four experiments conducted on the PY2 concentrate. This concentrate contains 21.6% sulphide compared to 45.7% in the PY1 concentrate. The goal of these experiments was to observe the rates of thiosulphate production as a function of pH at longer reaction times. The total concentration of NaOH added was in 61 the range of 60-129 g/L, which equates to ~1 to 2 mol of NaOH per mol of initial sulphide. Table 13: PY2 oxidation summary Under the selected experimental conditions, more than 33 g/L thiosulphate accumulated in 22-24 h (Exp 4, 8) following by a gradual further increase to 40 g/L (Exp 8) in 52 h and a subsequent decrease to 36 g/L after 72 h. Exp 8 was the only test in which excess NaOH was added (~2 mol/mol S), and this was the only test in which the thiosulphate produced was reasonably stable. In all the other tests (Exp 4, 5, and 6) a sharp decline of Test Time Solids NaOH addition pH EH S2- S2O3h % cumulative, g/L mV % mg/LExp 4 0 20.8 20 - - 21.6 0.04 41 10.5 -61 - 90008 61 12.3 -131 - 1520022 61 12.3 -124 - 3600027 61 10.4 -18 - 3100028 61 9.2 22 - 1500030 61 8.2 30 8.7 11000Exp 5 0 40.1 26 - - 21.6 0.02 52 9.5 183 - 49004 78 11.3 90 - 123008 78 12.4 -21 18.9 2100022 78 9.3 143 16.7 1000024 78 7.3 251 17.0 <10Exp 6 0 20.8 20 - - 21.6 0.02 40 12.2 33 19.3 47006 60 12.6 13 16.9 1500022 60 10.3 146 9.30 2700024 60 8.4 141 8.15 20000Exp 8 0 20.8 21 - - 21.6 0.04 43 10.8 235 17.8 76008 64 12.5 73 16.8 1390024 86 11.7 45 9.30 3330030 107 12.3 83 7.69 3540052 129 12.2 17 1.71 4100072 129 12.2 349 0.74 3600062 thiosulphate concentration was evidenced as soon as the pH dropped below 12. The disappearance of thiosulphate corresponded to an increase in the oxidation/reduction potential, as shown in Table 13. Table 14 presents the data for experiments conducted with in-situ pH control. Table 14: PY2 oxidation at various pHs The experiments where pH was controlled (pulp pH 9 in Exp 14, pulp pH 10 in Exp 15, and pulp pH 11 in Exp 16) showed some increase in thiosulphate concentration initially but it rapidly reached a plateau. It is obvious that the thiosulphate does not accumulate unless the pH is close to 12. In Exp 16, the NaOH addition was unintentionally high Test Time Solids pH EH S2- S2O3 h % cumulative, g/L mV % mg/LpH 9 of the pulpExp 14 0 20.2 0 - - 21.6 0.01 5 9.7 211 - 3002 6 9.4 223 - 2404 8 9.0 243 - 1608 10 8.8 277 20.0 17024 20 8.4 284 18.1 35028 20 8.6 281 17.9 300pH 10 of the pulpExp 15 0 20.2 0 - - 21.6 0.01 5 11.4 135 - 8902 7 11.3 127 - 12004 9 11.3 114 - 12008 14 11.4 136 20.7 120024 30 11.3 - 17.3 87028 33 10.8 175 16.5 400pH 11 of the pulpExp 16 0 20.1 0 - - 21.6 0.01 6 11.0 122 - 10002 7 11.0 146 - 13504 9 11.3 147 - 13508 18 10.7 138 18.4 150024 42 11.8 - 15.2 310028 43 11.7 96 14.9 3000NaOH addition63 towards the end of the test (24-28 h), resulting in a rapid increase in thiosulphate concentration. Figure 24 shows the effect of pH on sulphide oxidation. The points for pH of 9-11 and pH 12 lie in a single trendline initially but then diverge at longer reaction times. This could be attributed to the morphology of a possible iron oxide precipitated on the surface of the mineral affecting the mass transfer of reactants and products through the product layer. This effect seems to aggregate over time. Figure 24: PY2 Sulphide oxidation at various pHs 64 Figure 25: PY2 thiosulphate formation at various pHs Figure 25 shows the rate of change in thiosulphate concentration following the change in pH for the various tests. In all cases, the concentration of thiosulphate in solution increased initially, reaching a peak after several h, and then started to decline. The experiments run at a constant pH of 9, 10, and 11 produced a very low but stable thiosulphate concentration. 4.3 DISCUSSION Figure 26 combines both PY1 and PY2 oxidized sulphides (moles) data run under similar conditions. The initial rate of oxidation for both pyrite sulphides was close to the 0.08 mol sulphide per hour but the rate decreased in time (shifting kinetics) and with 65 increasing solids density in the reactor, possibly due to a slowing oxygen mass transfer at higher solids density. Figure 26: PY1 and PY2 moles of sulphides oxidized in time 66 Figure 27: PY1 and PY2 percent of sulphides oxidized in time It is obvious that the test with higher initial sulphides (Exp 5, 40% solids) showed a higher amount (moles) of oxidized sulphides (Figure 26). However, when the fraction oxidized (calculated as moles of oxidized over moles of initial sulphide) is plotted, as in Figure 27, the curve corresponding to the higher percent solids appears below the curve for lower percent solids. The initial rate of sulphur oxidation was close to the 6.12% per hour for most of the tests. The effect of pH on the rate of sulphide oxidation was fairly independent of pH (in the range of 9-12) over the first ~8 h of reaction, but diverged at longer reaction times with apparent passivation at lower pH (Figure 28). The reason for this apparent pH effect at longer reaction times was not established, although it does suggest changing the 67 morphology of the passive layers (deposited iron oxides) and decreasing diffusivity of the mineral surface layer at lower pH. Figure 28: Moles of sulphides oxidized at various pHs The profiles for thiosulphate formation in the experiments that were conducted under pH-controlled conditions clearly demonstrate the dramatic effect of pH on thiosulphate stability in the solution. In contrast to the sulphides oxidation, the accumulation of thiosulphate was only detected when pH >12, as depicted in Figure 29. 68 Figure 29: Moles of sulphides oxidized at various pHs Much of the behaviour of thiosulphate, including its degradation and the formation pathways of polythionates, traditionally is described based on the EH-pH diagram. The stability diagram for thiosulphate (1M total sulphur) is depicted in Figure 30. 69 Figure 30: Stability diagram by HSC for sulphur species eliminating higher polythionates and sulphates (1M total Sulphur, 25oC) This EH-pH diagram shows a narrow field of stability for thiosulphate. Note that the higher polythionates and sulphate have been eliminated from this diagram. According to this diagram, the oxidative decomposition of thiosulphate forms tetrathionate, trithionate, and then sulphite at lower pH, whereas at the high end of the pH range studied here (pH 12.8), thiosulphate likely decomposes directly to sulphite and then to sulphate. In the tests at pH >12, the concentration of thiosulphate and sulphite increased linearly with time and the amounts of trithionate and tetrathionate produced were insignificant. This is in alignment with the EH-pH diagram and is also in agreement with published literature. Goldhaber (1983) demonstrated that tetrathionate and sulphate were the major species during pyrite oxidation at low pH values (pH 6-7), whereas thiosulphate and sulphite dominated when the pH was increased to 9. Various studies on thiosulphate degradation revealed tetrathionate formation under weakly acidic conditions (Lyons and 70 Nickless, 1968) which then decomposed to thiosulphate and trithionate at pH >10, while another study showed that trithionate produced thiosulphate and sulphate by hydrolysis at pH 5.5-12 (Rolla and Chakrabarti, 1982). In the current study, the concentration of sulphite decreased at pH <12 (pH 11.9), while sulphate and trithionate increased. Some insignificant increases in tetrathionate concentration can be noted in Exp 3 at pH 2.3. In all these tests, the only species that was steadily increasing, independent of pH, was sulphate. Figure 31 shows the change in thiosulphate concentration with time for tests corresponding PY1 and PY2 samples. In this plot, a tendency towards a linear rate of thiosulphate production (~ 0.0157 mol/h) sharply changes as pH decreases below 12 and declines at longer reaction times (>24 h). The latter might be explained by the depletion of sulphides possibly shifting the reaction kinetics/mechanism to a different regime. It is important to note that thiosulphate exhibited an apparent stability at pH ~12, reaching a higher peak concentration at that pH and holding the peak concentration for a longer time. 71 Figure 31: PY1 and PY2 moles of thiosulphate formed in time 72 5 GOLD DISSOLUTION DURING PYRITE OXIDATION 5.1 INTRODUCTION A further objective of this work was to investigate the simultaneous dissolution of gold with the thiosulphate that is formed in-situ during pyrite oxidation. This section provides the conditions found to achieve reasonable gold extraction from two different pyrite concentrates. 5.2 RESULTS 5.2.1 GOLD DISSOLUTION CONDITIONS DURING PYRITE OXIDATION Simultaneous oxidation of sulphide and dissolution of metals from both concentrates was studied. Leaching experiments under conditions summarized in Table 15 and Table 16 were carried out to measure the thiosulphate yield and changes in metal concentrations with time. The results indicate that gold and silver extractions as high as 96% and 75% can be obtained by complexing with in-situ generated thiosulphate. Pairs of experiments (Exp 9 and 10 with the PY1 concentrate, and Exp 4 and 7 with the PY2 concentrate) were carried out under identical conditions to evaluate the reproducibility of the experimental technique. In general, the results were similar, but there was some variability in the data output and it seems that gold recovery could be a sensitive function of small changes in the solution chemistry. The complete sets of data are given in Table 15 through Table 17 and Figure 32 through Figure 35. 73 Table 15: Gold dissolution during PY1 oxidation Table 16: Gold dissolution during PY2 oxidation Test Time Solids NaOH pH EH S2- Solids Assayscumulative S2O3 Auh % g/L mV % g/L mg/LExp 9 0 33.3 - - 45.7 0.0 0.024 33.3 12.7 15 - 18.0 0.1728 33.3 9.2 202 15.8 20.0 0.30Exp 10 0 33.3 - - 45.7 0.0 0.024 33.3 12.2 -51 - 21.0 0.2228 33.3 8.1 79 14.2 19.0 0.266.3 0.0 6.3051.1 -63.7 -6.3 0.0 6.3040.7 -72.1 0.26Extraction SolutionAu Au% g/tTest Time Solids NaOH pH EH S2- Cumulative Au S2O3 Au Auh % g/L mV % % g/L mg/L g/tExp 4 0 20 - - 21.6 - 0.0 0.0 95.022 61 12.3 -124 - 13.0 36.0 2.35 72.627 61 10.4 -18 - 25.9 31.0 4.81 69.428 61 9.2 22 - 76.7 15.0 16.4 23.930 61 8.2 30 8.7 96.3 11.0 21.5 3.63Exp 5 0 26 - - 21.6 - 0.0 0.0 95.02 52 9.5 183 - 0.18 4.9 0.10 -4 78 11.3 90 - 0.47 12.3 0.26 -8 78 12.4 -21 18.9 0.53 21.0 0.29 10122 78 9.3 143 16.7 51.5 10.0 28.4 52.824 78 7.3 251 17.0 31.4 0.0 17.3 69.6Exp 6 0 20 - - 21.6 - 0.0 0.0 95.02 40 12.2 33 19.3 0.30 4.7 0.07 94.56 60 12.6 13 16.9 1.30 15.0 0.24 85.722 60 10.3 146 9.30 50.9 27.0 8.90 40.824 60 8.4 141 8.15 85.0 20.0 18.7 15.5Exp 7 0 20.8 20 3.2 - 21.6 - 0.0 0.0 95.020 61 12.5 39 - 17.0 31.0 3.59 79.328 61 10.7 73 - 43.8 30.0 9.13 54.330 61 9.3 102 - 64.5 28.0 13.8 35.432 61 8.3 75 - 89.1 18.0 18.6 10.5Exp 8 0 20.8 21 - - 21.6 - 0.0 0.0 95.072 129 12.2 49 0.74 18.0 36.0 3.92 77.6Extraction Solution Solids 20.840.120.874 Table 17: Concentration of gold, silver, and copper in solution during PY2 oxidation Test Time Solids NaOH pH EH S2- Cumulative S2O3 Au Ag Cuh % g/L mV % g/L mg/L mg/L mg/LExp 4 0 20 - - 21.6 0.0 0.0 0.0 0.022 61 12.3 -124 - 36.0 2.35 5.23 59.827 61 10.4 -18 - 31.0 4.81 14.5 16428 61 9.2 22 - 15.0 16.4 19.1 32930 61 8.2 30 8.7 11.0 21.5 22.8 191Exp 5 0 26 - - 21.6 0.0 0.0 0.0 0.02 52 9.5 183 - 4.9 0.10 0.08 0.434 78 11.3 90 - 12.3 0.26 0.29 1.858 78 12.4 -21 18.9 21.0 0.29 0.37 8.6422 78 9.3 143 16.7 10.0 28.4 2.45 14.224 78 7.3 251 17.0 0.0 17.3 3.04 4.43Exp 6 0 20 - - 21.6 0.0 0.0 0.0 0.02 40 12.2 33 19.3 4.7 0.07 0.07 0.776 60 12.6 13 16.9 15.0 0.24 0.10 2.4122 60 10.3 146 9.30 27.0 8.90 19.6 22824 60 8.4 141 8.15 20.0 18.7 26.7 227Exp 7 0 20.8 20 3.2 - 21.6 0.0 0.0 0.0 0.020 61 12.5 39 - 31.0 3.59 3.04 47.428 61 10.7 73 - 30.0 9.13 14.2 16130 61 9.3 102 - 28.0 13.8 19.6 35132 61 8.3 75 - 18.0 18.6 22.9 411Exp 8 0 20.8 21 - - 21.6 0.0 0.0 0.0 0.072 129 12.2 49 0.74 36.0 3.92 - -Solution 20.840.120.875 Figure 32: Gold dissolution and thiosulphate profile in Exp 4 Figure 33: Gold dissolution and thiosulphate profile in Exp 7 76 Figure 34: Gold dissolution and thiosulphate profile in Exp 5 77 Figure 35: Gold dissolution and thiosulphate profile in Exp 6 Figure 32 through Figure 35 depict similar trends for thiosulphate generation and gold and silver dissolution. The greater stability of thiosulphate at pH >12 unfortunately corresponds with a significantly reduced rate of gold dissolution. When the pH drops below 12, gold leaches quite rapidly, but thiosulphate also decomposes quite rapidly. The decrease in pH is accompanied by a large increase in solution potential, which enhances the rate of oxidation of gold and silver, as well as thiosulphate. Exp 5 was conducted at 40% solids, and this was not successful. The maximum thiosulphate concentration in solution was lower than in the tests at 20% solids, despite the presence of double the quantity of thiosulphate-generating sulphide in the feed, and the maximum gold extraction was only ~50%. The poor thiosulphate generation in this test is believed to be due to oxygen starvation and slow oxygen mass transfer at the higher solids density, combined with a faster rate of thiosulphate oxidation to 78 polythionates, sulphite, and sulphate. By the end of this test, the pH had dropped to pH ~7, all the thiosulphate had disappeared from solution, and the leached gold had started to reprecipitate from solution. Exp 8 failed for a different reason. In this case, too much caustic was added at the start of the reaction (~130 g/L). The pH remained high (>12) throughout the test and, although a lot of thiosulphate was generated (36 g/L), only 18% of the gold leached in 72 h, likely due to an insufficient solution oxidation reduction potential to oxidize the gold. 5.2.2 GOLD DISSOLUTION AT CONSTANT ALKALINITY Table 18 presents the data for tests with in-situ pH control. In these tests, despite high solution oxidation potential, very little gold was extracted. In this case, the strength of the lixiviant, thiosulphate concentration was very low for gold solubilization and stabilization. 79 Table 18: Gold dissolution during PY2 oxidation at constant pH 5.3 DISCUSSIONS The best leaching results were achieved in Exp 4, 6, and 7. In each case, the pyrite was leached and the sulphides oxidized at 20% solids. In these tests, sufficient caustic was added at the start of the reaction (~60 g/L) to maintain pH >12 for sufficiently long enough to build up a high concentration of thiosulphate (20-30 g/L). Gold leaching was slow during this period (typically <20% in 24 h) because of the high pH. No further caustic was added and the pH was then allowed to drift downwards over the next ~12 h, to around pH 8 (due to the consumption of the excess caustic by the acid generated in the sulphide oxidation reaction). Although thiosulphate was less stable and was disappearing over this period, a sufficient amount remained in the solution to leach gold and silver, and the rate of the leaching reactions increased significantly because of the lower pH. Unfortunately, the oxidation of sulphides at pH <12 does not produce sufficient thiosulphate to effectively leach the gold. Figure 36 compares the gold extraction data Test Time pH EH S2- S2O3 Au Extraction Au Ag Cu Auh cumulative, g/L mV % mg/L % mg/L mg/L mg/L g/tExp 14 0 0 - - 21.6 0.0 0.00 0.0 0.0 0.0 95.01 5 9.7 211 - 300 0.20 0.05 0.09 0.48 -2 6 9.4 223 - 240 1.00 0.24 0.32 0.66 -4 8 9.0 243 - 160 1.70 0.44 0.25 0.22 -8 10 8.8 277 20.0 170 0.30 0.08 0.14 0.12 -24 20 8.4 284 18.1 350 4.40 1.15 2.10 0.34 -28 20 8.6 281 17.9 300 1.40 0.38 0.58 0.19 -Exp 15 0 0 - - 21.6 0.0 0.00 0.0 0.0 0.0 95.01 5 11.4 135 - 890 0.60 0.11 0.08 0.44 76.82 7 11.3 127 - 1200 0.10 0.03 0.03 0.16 86.44 9 11.3 114 - 1200 0.30 0.06 0.04 0.20 88.38 14 11.4 136 20.7 1200 2.30 0.44 0.46 0.99 84.024 30 11.3 - 17.3 870 2.20 0.45 0.13 0.30 10028 33 10.8 175 16.5 400 0.30 0.05 0.03 0.23 100Exp 16 0 0 - - 21.6 0.0 0.00 0.0 0.0 0.0 95.01 6 11.0 122 - 1000 0.20 0.22 0.32 1.40 -2 7 11.0 146 - 1350 0.20 0.23 0.34 1.08 -4 9 11.3 147 - 1350 0.20 0.15 0.23 0.57 -8 18 10.7 138 18.4 1500 0.20 0.17 0.59 0.79 -24 42 11.8 - 15.2 3100 0.10 0.09 0.07 0.51 -28 43 11.7 96 14.9 3000 0.10 0.10 0.06 0.47 -NaOH addition80 from three experiments that gave the best gold extraction (Exp 4, 6, and 7) against the tests (Exp 14, 15, and 16) where pH was maintained at <12 value. Figure 36: Gold dissolution in time, depending on pH and thiosulphate concentration 81 Figure 37: Changes in thiosulphate concentration and gold extraction in time, depending on solution pH and EH Figure 37 exemplifies factors affecting the thiosulphate stability and the gold dissolution. It is obvious that the concentration of thiosulphate that was produced as a result of pyrite oxidation reached the peak at pH 12 but sharply decreased when pH decreased to 8.2. This drop in pH was accompanied with the increase in EH from (-124) to 80 mV and a sharp increase in gold extraction. Despite the fact that these factors was in the range (pH 8.6 and EH = 82 mV) that is optimal for gold dissolution, there was very little gold extracted in the tests due to low concentrations (<5 g/L) of thiosulphate that was generated under these conditions. 82 6 MODELLING OF PYRITE OXIDATION TO THIOSULPHATE 6.1 INTRODUCTION The oxidation of sulphide under highly alkaline conditions yielded near linear rates for sulphide conversion to thiosulphate. This is particularly interesting since both reactions rates can be easily linked and controlled simultaneously. The goal of this modelling exercise was to determine the rates of sulphide oxidation and thiosulphate production under highly alkaline conditions. 6.2 PYRITE OXIDATION RATE When the progress of the reaction is unaffected by the product layer, then the rate becomes directly proportional to the available surface of unreacted sulphides. 𝐝𝐪𝐝𝐭= −𝐀 𝐉 6- 1 where q is the moles of sulphide, A (m2) is the total surface area, and J (mol m-2 h-1) is the mole flux. Assuming the particles are spherical and of identical size with radius r (m), the total number of grains 𝚗, density ρ (mol m-3), and using known relationships, one obtains: 𝐀 = 𝟒𝛑𝐫𝟐 6- 2 𝐕 = 𝟒𝟑𝛑𝐧𝐫𝟑 6- 3 𝐫 = � 𝟑𝐕𝟒𝛑𝐧� 𝟏𝟑 6- 4 83 then the total surface area can be presented as: 𝐀 = 𝟒𝛑𝚗 � 𝟑𝐕𝟒𝛑𝚗𝛒�𝟐𝟑 = 𝟑 �𝟒𝛑𝐧𝟑�𝟏𝟑 𝐕𝟐𝟑 6- 5 Since, V= 𝐪𝛒 6- 6 𝐀 = 𝟑 �𝟒𝛑𝐧𝟑�𝟏𝟑 𝐪𝛒𝟐𝟑 6- 7 Inserting this into [6-1] 𝐝𝐪𝐝𝐭 = −𝟑 𝐉 �𝟒𝛑𝐧𝟑�𝟏𝟑 𝐪𝛒𝟐𝟑 6- 8 Converting 𝐝𝐪𝐪𝟐𝟑 = −𝟑 𝐉 �𝟒𝛑𝐧𝟑�𝟏𝟑 𝟏𝛒𝟐𝟑 𝐝𝐭 6- 9 Integrating 𝐪𝟏𝟑𝟏𝟑 = 𝟑𝐪𝐨𝟏𝟑 − 𝟑 𝐉 �𝟒𝛑𝐧𝟑 �𝟏𝟑 𝟏𝛒𝟐𝟑 𝐭 6- 10 Or 𝐪𝟏𝟑 = 𝐪𝐨𝟏𝟑 − 𝐉 �𝟒𝛑𝐧𝟑 �𝟏𝟑 𝟏𝛒𝟐𝟑 𝐭 6- 11 84 Simplifying �𝐪𝐪𝐨�𝟏𝟑 = 𝟏 − 𝐉 �𝟒𝛑𝐧𝟑𝐪𝐨�𝟏𝟑 𝟏𝛒𝟐𝟑 𝐭 6- 12 Assuming that molar flux is constant and assigning the complete conversion by 𝛕 = �𝐀𝟎 �𝟒𝛑𝐧𝟑𝐪𝐨�𝟏𝟑 𝛒−𝟐/𝟑�−𝟏 6- 13 Then 𝐪𝐪𝐨= �𝟏 − 𝐭𝛕�𝟑 6- 14 By plotting the sulphide data versus time, an excellent straight line fit was obtained when plotting conversion versus time. The times for total conversion estimated from the slopes of the lines in Figure 38 are 33 h, 42 h, and 59 h at oxygen overpressures of 40 psi, 20 psi, and 10 psi, respectively. 85 Figure 38: Sulphide oxidation 6.3 CONVERSION MODEL DEVELOPMENT A simple network of reactions is presented for the hypothetical route of pyrite sulphide conversion under oxidative conditions. The rate equations are all first order so the development of rate expression does not involve complex mathematics. The conceptual model is presented in Figure 39. Figure 39: Conceptual model of pyrite sulphide sulphur oxidation S2 2- So S2O32- S4O62- S3O62- SO32- SO42- HS- S2O32- S2O32- S2O32- K1 K4 K5 K6 K7 K2 K3 86 The reactions describing the stoichiometric relationship between these species (Goldhaber, 1983; Chen and Morris, 1972; Byerley, 1975; Rolla and Chakhabati, 1982; Zhang and Dreisinger, 2002; Ahern, 2006; Naito, 1970; Bengtsson and Bjerle, 1975; Kleinjan et al., 2005, Smith and Hitchen, 1976) are as follows: 2FeS2 + 1.5O2 + H2O 2FeOOH + 4So 6- 15 4So + 4OH- 2HS- + S2O32- + H2O 6- 16 2HS- + 2O2 H2O + S2O32- 6- 17 2S2O32- + 0.5O2 + H2O S4O62- + 2OH- 6- 18 4S4O62- + 6OH- 5S2O32- + 2S3O62- + 3H2O 6- 19 2S3O62- + 6OH- S2O32- +4SO32- + 3H2O 6- 20 SO32- + 0.5O2 SO42- 6- 21 The rate of each reaction is a function of alkalinity and oxygen overpressure. The rate of change and the yield of each intermediate species in these reactions can be described as the sum of the individual rates: fractions formed minus fraction disappeared in the following reaction. Due to multiple occurrences of thiosulphate in this sequence, the ultimate performance equation will include the fraction formed due to sulphur oxidation (6-16 and 6-17), the fraction oxidized forming tetrathionate (6-18), the fraction formed as a side product in the hydrolysis reactions of tetrathionate (6-19), and trithionate (6-20) as shown in equation (6-25). Rate equations for reactions (6-22) through (6-29) are written as first order with respect to sulphur species, but having various orders with respect to oxygen concentration and alkalinity. −𝐝𝐂[𝐒−𝟐]𝐝𝐭= 𝐤𝟏[𝐂𝐨𝐬−𝟐] 6- 22 k1 k2 k4 k5 k6 k3 k7 87 𝐝𝐂𝐬𝐨𝐝𝐭= 𝐤𝟏[𝐂𝐨𝐬−𝟐][𝐏𝐎𝟐]∝𝟏 − 𝐤𝟐[𝐂𝐬𝐨][𝐎𝐇]𝛃𝟐 6- 23 𝐝𝐂𝐇𝐒𝐝𝐭= 𝟏𝟐 𝐤𝟐[𝐂𝐬𝐨][𝐎𝐇]𝛃𝟐 − 𝐤𝟑[𝐂𝐇𝐒][𝐏𝐎𝟐]∝𝟑 6- 24 𝐝𝐂𝐒𝟐𝐎𝟑𝐝𝐭 =𝟏𝟒𝐤𝟐[𝐂𝐬𝐨][𝐎𝐇]𝛃𝟐 + 𝟏𝟐 𝐤𝟑[𝐂𝐇𝐒]�𝐏𝐎𝟐�∝𝟑 − 𝐤𝟒[𝐂𝐒𝟐𝐎𝟑]�𝐏𝐎𝟐�∝𝟒 + 𝟓𝟒 𝐤𝟓[𝐂𝐒𝟒𝐎𝟔][𝐎𝐇]𝛃𝟓 ++ 𝟏𝟐𝐤𝟔[𝐂𝐒𝟑𝐎𝟔][𝐎𝐇]𝛃𝟔 6- 25 𝐝𝐂𝐒𝟒𝐎𝟔𝐝𝐭= 𝟏𝟐𝐤𝟒[𝐂𝐒𝟐𝐎𝟑]�𝐏𝐎𝟐�∝𝟒 − 𝐤𝟓[𝐂𝐒𝟒𝐎𝟔][𝐎𝐇]𝛃𝟓 6- 26 𝐝𝐂𝐒𝟑𝐎𝟔𝐝𝐭= 𝟏𝟐𝐤𝟓[𝐂𝐒𝟒𝐎𝟔][𝐎𝐇]𝛃𝟓 − 𝐤𝟔[𝐂𝐒𝟑𝐎𝟔][𝐎𝐇]𝛃𝟔 6- 27 𝐝𝐂𝐒𝐎𝟑𝐝𝐭= 𝟐𝐤𝟔[𝐂𝐒𝟑𝐎𝟔][𝐎𝐇]𝛃𝟔 − 𝐤𝟕[𝐂𝐒𝐎𝟑]�𝐏𝐎𝟐�∝𝟕 6- 28 𝐝𝐂𝐒𝐎𝟒𝐝𝐭= 𝐤𝟕[𝐂𝐒𝐎𝟑][𝐏𝐎𝟐]∝𝟕 6- 29 Sulphate, as the most stable end-product, does not react any further. By inserting the rate of sulphate formation (6-29) into the sulphite yield equation (6-28), the rate of sulphite formation can be estimated as shown in equation (6-30). All other equations can be modified in a similar fashion as follows: 𝐝𝐂𝐒𝐎𝟒𝐝𝐭 + 𝐝𝐂𝐒𝐎𝟑𝐝𝐭 = 𝟐𝐤𝟔[𝐂𝐒𝟑𝐎𝟔][𝐎𝐇]𝛃𝟔 6- 30 𝐝𝐂𝐒𝟑𝐎𝟔𝐝𝐭+ 𝟏𝟐(𝐝𝐂𝐒𝐎𝟒𝐝𝐭+ 𝐝𝐂𝐒𝐎𝟑𝐝𝐭) = 𝟏𝟐𝐤𝟓[𝐂𝐒𝟒𝐎𝟔][𝐎𝐇]𝛃𝟓 6- 31 𝐝𝐂𝐒𝟒𝐎𝟔𝐝𝐭+ �𝟐 𝐝𝐂𝐒𝟑𝐎𝟔𝐝𝐭+ �𝐝𝐂𝐒𝐎𝟒𝐝𝐭+ 𝐝𝐂𝐒𝐎𝟑𝐝𝐭�� = 𝟏𝟐𝐤𝟒[𝐂𝐒𝟐𝐎𝟑]�𝐏𝐎𝟐�∝𝟒 6- 32 𝐝𝐂𝐒𝟐𝐎𝟑𝐝𝐭+ �𝟐 𝐝𝐂𝐒𝟒𝐎𝟔𝐝𝐭+ 𝟒 𝐝𝐂𝐒𝟑𝐎𝟔𝐝𝐭+ 𝟐 𝐝𝐂𝐒𝐎𝟑𝐝𝐭+ 𝟐 𝐝𝐂𝐒𝐎𝟒𝐝𝐭� −𝟓𝟒�𝟐𝐝𝐂𝐒𝟑𝐎𝟔𝐝𝐭+ 𝐝𝐂𝐒𝐎𝟑𝐝𝐭+ 𝐝𝐂𝐒𝐎𝟒𝐝𝐭� −𝟏𝟐�𝟏𝟐𝐝𝐂𝐒𝐎𝟒𝐝𝐭 + 𝟏𝟐𝐝𝐂𝐒𝐎𝟑𝐝𝐭� = 𝟏𝟒𝐤𝟐[𝐂𝐬𝐨][𝐎𝐇]𝛃𝟐 + 𝟏𝟐 𝐤𝟑[𝐂𝐇𝐒]�𝐏𝐎𝟐�∝𝟑 6- 33 88 When intermediates are present at very small concentrations that cannot be measured, their rate of change can be taken to be zero; for example, the transient formation of aqueous sulphide (𝐝𝐂𝐇𝐒𝐝𝐭= 𝟎) under the alkaline oxidative conditions. 𝟏𝟐 𝐤𝟐[𝐂𝐬𝐨][𝐎𝐇]𝛃𝟐 = 𝐤𝟑[𝐂𝐇𝐒]�𝐏𝐎𝟐�∝𝟑 6- 34 After inserting the concentrations of species from equation 6-34 into 6-33, the rate of the thiosulphate formation becomes: 𝐝𝐂𝐒𝟐𝐎𝟑𝐝𝐭+ 𝟐 𝐝𝐂𝐒𝟒𝐎𝟔𝐝𝐭+ 𝟑𝟐𝐝𝐂𝐒𝟑𝐎𝟔𝐝𝐭+ 𝟏𝟐𝐝𝐂𝐒𝐎𝟒𝐝𝐭+ 𝟏𝟐𝐝𝐂𝐒𝐎𝟑𝐝𝐭= 𝟏𝟒𝐤𝟐[𝐂𝐬𝐨][𝐎𝐇]𝛃𝟐 + 𝟏𝟒 𝐤𝟐𝐤𝟑 𝐤𝟑[𝐂𝐬𝐨] [𝐎𝐇]𝛃𝟐�𝐏𝐎𝟐�∝𝟑 �𝐏𝐎𝟐�∝𝟑 = = 𝟏𝟐𝐤𝟐[𝐂𝐬𝐨][𝐎𝐇]𝛃𝟐 6- 35 Furthermore, the species in the rate equations in the series can be replaced by known parameters such as the initial sulphide sulphur (pyrite), alkalinity (NaOH), and oxygen pressure (pO2), as presented below. 𝐤𝟐[𝐂𝐬𝐨][𝐎𝐇]𝛃𝟐 = 𝐤𝟏𝐤𝟐[𝐂𝐨𝐬−𝟐][𝐎𝐇]𝛃𝟐[𝐏𝐎𝟐]∝𝟏 6- 36 𝐤𝟒[𝐂𝐒𝟐𝐎𝟑][𝐏𝐎𝟐]∝𝟒 = 𝟏𝟐 𝐤𝟏𝐤𝟐𝐤𝟒[𝐂𝐨𝐬−𝟐][𝐏𝐎𝟐]∝𝟏+∝𝟒[𝐎𝐇]𝛃𝟐 6- 37 𝐤𝟓[𝐂𝐒𝟒𝐎𝟔][𝐎𝐇]𝛃𝟓 = 𝟏𝟒 𝐤𝟏𝐤𝟐𝐤𝟒𝐤𝟓[𝐂𝐨𝐬−𝟐][𝐏𝐎𝟐]∝𝟏+∝𝟒[𝐎𝐇]𝛃𝟐+𝛃𝟓 6- 38 𝐤𝟔[𝐂𝐒𝟑𝐎𝟔][𝐎𝐇]𝛃𝟔 = 𝟏𝟖 𝐤𝟏𝐤𝟐𝐤𝟒𝐤𝟓𝐤𝟔[𝐂𝐨𝐬−𝟐][𝐏𝐎𝟐]∝𝟏+∝𝟒[𝐎𝐇]𝛃𝟐+𝛃𝟓+𝛃𝟔 6- 39 The thiosulphate yield can be calculated by inserting the corresponding relationships from equations 6-36 through 6-39 into equation 6-25 and combining related terms: 89 𝐝𝐂𝐒𝟐𝐎𝟑𝐝𝐭=𝟏𝟐𝐤𝟏𝐤𝟐[𝐂𝐨𝐬−𝟐]�𝐏𝐎𝟐�∝𝟏[𝐎𝐇]𝛃𝟐 − 𝟏𝟐 𝐤𝟏𝐤𝟐𝐤𝟒[𝐂𝐨𝐬−𝟐]�𝐏𝐎𝟐�∝𝟏+∝𝟒[𝐎𝐇]𝛃𝟐 ++ 𝟓𝟏𝟔𝐤𝟏𝐤𝟐𝐤𝟒𝐤𝟓[𝐂𝐨𝐬−𝟐]�𝐏𝐎𝟐�∝𝟏+∝𝟒[𝐎𝐇]𝛃𝟐+𝛃𝟓 + 𝟏𝟏𝟔 𝐤𝟏𝐤𝟐𝐤𝟒𝐤𝟓𝐤𝟔[𝐂𝐨𝐬−𝟐]�𝐏𝐎𝟐�∝𝟏+∝𝟒[𝐎𝐇]𝛃𝟐+𝛃𝟓+𝛃𝟔 6- 40 𝐝𝐂𝐒𝟐𝐎𝟑𝐝𝐭= 𝟐−𝟏𝐤𝟏𝐤𝟐[𝐂𝐨𝐬−𝟐]�𝐏𝐎𝟐�∝𝟏[𝐎𝐇]𝛃𝟐{ 𝟏 − 𝐤𝟒�𝐏𝐎𝟐�∝𝟒 + 𝟐−𝟑𝐤𝟒𝐤𝟓�𝐏𝐎𝟐�∝𝟒[𝐎𝐇]𝛃𝟓 (𝟓 + + 𝐤𝟔�𝐎𝐇]𝛃𝟔�} 6- 41 Thus, equation 6-41 is the overall thiosulphate yield equation interrelating the rate constants of the multiple reactions that were considered. 6.4 THIOSULPHATE YIELD VALIDATION To test whether the thiosulphate yields predicted by the kinetic equation correspond to the experimentally-generated data, it was necessary to estimate the rate constants for the formation of the intermediate species. Knowing that intermediates are highly reactive, their concentrations (fraction formed that is not consumed in the following reaction) have been calculated using the relationships given in equations 6-30 through 6-33 using the time dependent assays shown in Table 6. The calculated values are presented in Table 19 through Table 21. Table 19: Sulphides reacted and intermediates formed during pyrite oxidation at oxygen overpressure of 40 psi and 80oC Time S2O3 S4O6 S3O6 SO3 SO4h mol mol(calc) mol(calc) mol(calc) mol(calc) mol(calc) mol1 0.111 0.098 0.033 0.032 0.016 0.032 0.0172 0.234 0.223 0.074 0.070 0.035 0.068 0.0384 0.432 0.444 0.148 0.113 0.057 0.111 0.0488 0.791 0.790 0.263 0.219 0.109 0.195 0.145S2- 90 Table 20: Sulphides reacted and intermediates formed during pyrite oxidation at oxygen pressure of 20 psi and 80oC Table 21: Sulphides reacted and intermediates formed during pyrite oxidation at oxygen pressure of 10 psi and 80oC As can be seen from Table 19 through Table 21, the measured sulphides and calculated sulphides are typically in close agreement. The rate of sulphide conversion and unobserved intermediates were estimated as slopes of the species concentrations given in Table 19 to Table 21 versus time and presented in Figure 40 to Figure 44 and Table 22. Time S2O3 S4O6 S3O6 SO3 SO4h mol mol(calc) mol(calc) mol(calc) mol(calc) mol(calc) mol1 0.094 0.082 0.027 0.029 0.014 0.029 0.0162 0.201 0.177 0.059 0.056 0.028 0.054 0.0324 0.351 0.321 0.107 0.086 0.043 0.084 0.0388 0.609 0.584 0.195 0.137 0.069 0.136 0.050S2- Time S2O3 S4O6 S3O6 SO3 SO4h mol mol(calc) mol(calc) mol(calc) mol(calc) mol(calc) mol1 0.056 0.142 0.047 0.033 0.016 0.032 0.0112 0.138 0.125 0.042 0.044 0.022 0.043 0.0264 0.254 0.208 0.069 0.057 0.028 0.056 0.0268 0.430 0.427 0.142 0.096 0.048 0.095 0.036S2- 91 Figure 40: Sulphide versus time and oxygen pressure Figure 41: Thiosulphate versus time and oxygen pressure y = 0.1016xR² = 0.9945y = 0.0797xR² = 0.9834y = 0.0563xR² = 0.9840.000.100.200.300.400.500.600.700.800.900 5 10Sulphide (mol)Time (h)40 psi pO220 psi pO210 psi pO2y = 0.4431x - 1.8222R² = 0.9878-1.60-1.50-1.40-1.30-1.20-1.10-1.000.9 1.1 1.3 1.5 1.7-log [dC[S2-]/dCo[S2- ]/dt] (h-1)log [pO2], psiSLinear (S)y = 0.0339xR² = 0.9942y = 0.0251xR² = 0.993y = 0.0182xR² = 0.91780.000.050.100.150.200.250.300 5 10Thiosulphate (mol)Time (h)40 psi pO220 psi pO210 psi pO2y = 0.4314x - 1.9955R² = 0.9989-1.70-1.60-1.50-1.40-1.30-1.20-1.10-1.00-0.90-0.800.5 1.0 1.5 2.0log{dC[S2O3]/dCo[S-2 ]/dt} (h-1)log [pO2] (psi)S2O3Linear (S2O3)92 Figure 42: Tetrathionate versus time and oxygen pressure Figure 43: Trithionate versus time and oxygen pressure y = 0.028xR² = 0.9921y = 0.0186xR² = 0.9346y = 0.0131xR² = 0.83580.000.050.100.150.200.250 2 4 6 8 10Tetrathionate (mol)Time (h)40 psi pO220 psi pO210 psi pO2y = 0.5479x - 1.971R² = 0.998-1.50-1.40-1.30-1.20-1.10-1.00-0.90-0.800.0 0.5 1.0 1.5 2.0log{dC[S4O6]/dCo[S-2 ]/dt} (h-1)log [pO2] (psi)S4O6Linear (S4O6)y = 0.014xR² = 0.9922y = 0.0093xR² = 0.9349y = 0.0064xR² = 0.91180.000.020.040.060.080.100.120 2 4 6 8 10Trithionate (mol)Time (h)40 psi pO220 psi pO210 psi pO2y = 0.5646x - 1.9961R² = 0.9993-1.50-1.40-1.30-1.20-1.10-1.00-0.90-0.800.0 0.5 1.0 1.5 2.0log{dC[S3O6]/dCo[S-2]/dt} (h-1)log [pO2] (psi)S3O6Linear (S3O6)93 Figure 44: Sulphite versus time and oxygen pressure Table 22: Rate of sulphide conversion and estimated rate of unmeasured intermediates All reaction constants and orders were estimated by plotting logarithms of concentration change over initial sulphide contents versus logarithm of oxygen overpressure. The resulted values are collected in Table 23. y = 0.0256xR² = 0.9795y = 0.0184xR² = 0.943y = 0.0127xR² = 0.91750.000.050.100.150.200.250 2 4 6 8 10Sulphite (mol)Time (h)40 psi pO220 psi pO210 psi pO2y = 0.4884x - 1.9063R² = 0.9979-1.45-1.40-1.35-1.30-1.25-1.20-1.15-1.10-1.05-1.000.0 0.5 1.0 1.5 2.0log{dC[SO3]/dCo[S-2 ]/dt} (h-1)log [pO2] (psi)SO3Linear (SO3)pO2, psiS2- S2O3 S4O6 S3O6 SO340 0.102 0.034 0.028 0.014 0.02620 0.080 0.025 0.019 0.009 0.01810 0.056 0.018 0.013 0.006 0.013dC/dt, mol/h94 Table 23: Reaction constants and orders Using the constants from Table 23, the thiosulphate yield in 1 M NaOH solution at 80oC and oxygen overpressures of 10, 20, and 40 psi can be estimated by the equation (6-42) given below: 𝐝𝐂𝐒𝟐𝐎𝟑𝐝𝐭= 𝟐−𝟏 ∗ 𝟎.𝟎𝟏𝟓 ∗ 𝟎.𝟔𝟕𝟏 ∗ [𝐂𝐨𝐬−𝟐]�𝐏𝐎𝟐�𝟎.𝟒𝟑𝟕{𝟏 − 𝟏.𝟎𝟓𝟖 ∗ �𝐏𝐎𝟐�𝟎.𝟏𝟎𝟖 + 𝟐−𝟑 ∗ 𝟏.𝟎𝟓𝟖 ∗𝟎.𝟗𝟒𝟒 ∗ �𝐏𝐎𝟐�𝟎.𝟏𝟎𝟖 (𝟓 + 𝟏.𝟐𝟑𝟎} 6- 42 The predicted thiosulphate yields from the equation are compared to the measured yields in Figure 45. The best fit to the measured data can be observed at 20 psi oxygen overpressure. Constants calculationlog k1 -1.822 k1 0.015 α1 0.437logk1+log k2 -1.996 k2 0.671logk1+Log k2 +log k4 -1.971 k4 1.058 α1+α3 0.545logk1+Log k2 +log k4+log k5 -1.996 k5 0.944logk1+Log k2 +log k4+log k5+log k6 -1.906 k6 1.230Constants, mol-1 h-1 Reaction order95 Figure 45: Comparison of model calculated and measured moles of thiosulphate 6.5 CONCEPTUAL LEACHING MODEL DISCUSSION There are three important reactions in the overall process: sulphide oxidation, thiosulphate decomposition, and gold dissolution. It was observed that the rate of sulphide oxidation is not significantly affected by the changes in pH. However, pH had a dramatic effect on the thiosulphate stability and any decrease in pH <12 caused a rapid decomposition of thiosulphate. The decrease in pH was accompanied by an increase in solution potential which enhanced the rate of gold leaching. Hence, in the modelling of this process, an optimal intercept of all three processes must be found. The concentration of thiosulphate in solution at any point in time is a function of the rates of both sulphide oxidation and thiosulphate decomposition. The fact that the rate of sulphide oxidation is not largely affected by the pH might be used to navigate the process 96 pH towards desirable values. For example, in modelling the sulphide oxidation at 80oC under 20 psig-40 psig oxygen pressure assuming 20% pulp density and knowing the rates of sulphide oxidation and thiosulphate accumulation, the following steps might lead to the modelling of the process:  Estimate the amount of sulphide necessary to drop the pH from 12 to 10 based on known reactions  Subtract this amount from the total sulphide and calculate the amount of caustic required in order to have a ratio of 1.2 for the remaining sulphide  Knowing the rates for both the sulphide oxidation and thiosulphate accumulation to estimate the reaction time under alkaline pH >12 conditions  Determine the rate of thiosulphate decomposition as a function of pH and the remaining sulphide based on published rates in literature (complex problem) 97 7 ELECTROCHEMICAL DISSOLUTION OF GOLD 7.1 INTRODUCTION The EQCM technique in conjunction with a cyclic voltammetry method was used to identify the characteristic current profiles corresponding to the gold oxidation and reduction reactions. Furthermore, the addition of reagents, such as thallium and thiourea, were tested as additives to the thiosulphate electrolyte. The anticipation was that they would be directly involved in the charge transfer process due to their effect on gold dissolution as reported in a number of studies and discussed in the following sections. 7.2 CYCLIC VOLTAMMETRY IN THIOSULPHATE The data produced by EQCM was useful to separate the portion of the voltammogram corresponding to the gold anodic/cathodic current from the measured (total) current. Figure 46 presents the voltammogram obtained at the stationary gold electrode in the electrolyte of pH 7.8 containing 0.4 M magnesium thiosulphate. The potential scan was initiated at open circuit potential (OCP), reversed at approximately 0.37 V, continued to a minimum value of -0.2 V, reversed again, and then stopped at the initial potential (OCP). Also shown is the calculated anodic current (icalc) based on the gold electrode mass change during potential sweep. 98 Figure 46: E-i and E-icalc voltammograms in 0.4 M MgS2O3 at 1 mV/s scan rate The data revealed that the anodic current in 0.4 M MgS2O3 electrolyte peaked at ~0.217 V showing a total current of 5.79 µg cm-2 and the gold anodic current of 4.46 µg cm-2. The latter value was calculated based on the gold electrode mass change during cyclic voltammetry, which can be seen in Figure 47. The difference between measured and calculated currents has to be accounted for by a side electrochemical reaction (thiosulphate oxidation to tetrathionate). The gold anodic-to-cathodic peak-to-peak separation was close to 0.29 V. 99 Figure 47: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 at scan rate 1 mV/s From Figure 47, it is obvious that the mass of the gold electrode was constantly decreasing during the positive-going (from OCP to 0.37 V) and the initial negative-going parts of the potential scan until reaching the vicinity of OCP. In this region, there were no changes in the mass of the gold electrode, suggesting that gold is neither dissolving nor depositing. Passing this potential towards more negative values (from OCP to -0.2 V), the re-deposition of some of the dissolved gold occurred as an increase of the mass of the electrode can be seen in Figure 47. Furthermore, Figure 47 shows the charge calculated based on the total and calculated gold anodic curves presented in Figure 46. For example, the charge values corresponding to the potential region from OCP to 0.32 V were 203 µC and 154 µC for total and gold anodic curves, respectively. If the total current was entirely representing the gold dissolution, it would be equal to 2.095 µg cm-2. The actual dissolved gold, based on the gold electrode mass decrease, was 1.593 µg cm-2 which 100 upon conversion equals the charge (154 µC) corresponding to the gold anodic curve in Figure 46. The potential region from OCP to 0.32 V was selected as an example because the plot of the gold electrode mass change versus the charge of the measured (total) anodic curve exhibited a linear relationship up to this potential, as shown in Figure 48. This linearity was no longer observed at higher potentials, which matches the sudden increase in the anodic current shown in Figure 46. This is consistent with the difference between the total and calculated currents becoming large and points to a further decrease in the contribution of the oxidation of gold to the total charge passed. Figure 48: Mass change versus the charge calculated by integrating the area under the i-E curve on Figure 46 101 The mass variation versus cumulative charge (q) that passed from 0 to 0.22 mC, which corresponds to the 0.32 V potential before current reversal, shows a slope of approximately 1.585 µg mC-1 or 0.78 mol of gold per mol of electrons for q(itotal). For comparison, the theoretical cumulative charge was plotted versus the electrode mass charge, q(icalc), giving 2.04 µg mC-1 or 1 mol gold per mol of electrons. The difference between measured and calculated currents has to be accounted for by a side electrochemical reaction (thiosulphate oxidation to tetrathionate). The charge, q (icalc), covering the region of gold active dissolution or anodic charge (OCP to 0.37 and reversed to 0.07V) was 271 µC or 2.8 µg cm-2 (Figure 47). The charge following the anodic dissolution from 0.07 V to -0.06 V, where gold deposition occurred or cathodic charge (there was a slight re-dissolution occurring from -0.06V to 0.03V which was the OCP), was 49 µC or 0.5 µg cm-2. Based on these values, the ratio of cathodic-to-anodic charge becomes 0.177, which means that ~82% of the dissolved gold diffused out to the bulk electrolyte and was not available for reduction. 7.2.1 EFFECT OF GOLD THIOSULPHATE Figure 49 compares voltammograms for two electrolytes each containing 0.4 M MgS2O3 and 30 mg/L of gold added as a gold thiosulphate salt (Na3Au(S2O3)2) . 102 Figure 49: E-i voltammograms in 0.4 M MgS2O3 at 1 mV/s scan rate with 0 mg/L and 30 mg/L gold added as a gold thiosulphate The curve in Figure 49 shows that the addition of gold increased the reduction current and modestly suppressed the anodic oxidation peak. The E-icalc plot in Figure 50 also exhibits the same trend. The increase in the reduction current with the addition of gold at -0.07 V confirms that this peak is related to the gold reduction reaction. Figure 51 plots the gold electrode mass change and the charge corresponding to gold anodic current (icalc). 103 Figure 50: E-icalc voltammograms in 0.4 M MgS2O3 at 1 mV/s scan rate with 0 mg/L and 30 mg/L gold added as a gold thiosulphate 104 Figure 51: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 at scan rate 1 mV/s with 30 mg/L gold added as a gold thiosulphate 7.2.2 EFFECT OF THALLIUM When thallium was added into the thiosulphate electrolyte of the same strength, the anodic and cathodic current peaks became better defined but both peaks were shifted to more anodic potentials than those corresponding to voltammograms with thiosulphate alone. Figure 52 and Figure 53 show the effect of the Tl+ ion on the voltammograms at the gold electrode thiosulphate electrolyte of pH 7.5. Figure 52: E-i voltammogram in 0.4 M MgS2O3 with 0 mg/L, 0.1 mg/L, and 0.5 mg/L Tl at 1 mV/s scan rate 105 Figure 53: E-icalc voltammogram in 0.4 M MgS2O3 with 0 mg/L, 0.1 mg/L, and 0.5 mg/L Tl at 1 mV/s scan rate From Figure 52, it is obvious that the current densities increased from 5.7 uA cm-2 to 17.7 uA cm-2 and then to 53.4 uA cm-2 after increasing the dosage of thallium from 0 mg/L to 0.1 mg/L and 0.5 mg/L Tl in the 0.4 M MgS2O3 solution, respectively. A similar trend can be noted in Figure 53 showing the gold anodic current of close to 4.3 uA cm-2 to 15.1 uA cm-2 and then to 41.8 uA cm-2 after increasing the dosage of thallium from 0 mg/L to 0.1 mg/L and 0.5 mg/L Tl in the 0.4 M MgS2O3 solution, respectively. This effect is in accordance with other research findings and confirms the catalytic effect of thallium on the gold anodic dissolution in thiosulphate from OCP up to 0.3 V. Figure 54 compares total and calculated current in the thiosulphate electrolyte with and without thallium present. The gold anodic-to-cathodic peak-to-peak separation in the presence of thallium was ~0.26 V. 106 Figure 54: E-i and E-icalc voltammograms in 0.4 M MgS2O3 and 0.5 mg/L Tl at 1 mV/s scan rate Figure 55 plots the gold mass change as a function of the charge. The slope of mass-to-charge ratio is1.623 µg/mC or 0.80 mol of gold per mol of electrons for q( itotal), which is slightly higher than the value estimated in the thiosulphate electrolyte alone (0.78 mol of gold per mol of electrons), as shown in Figure 48. Hence, in the presence of thallium, the proportion of a charge that was consumed by something other than the gold oxidation reaction to the actual gold dissolution slightly decreased compared to the thiosulphate electrolyte where thallium was absent. 107 Figure 55: Mass change versus the charge calculated by integrating the area under i-E curve in Figure 54 More data are presented in Figure 56 to emphasize the catalytic effect of thallium. The charge corresponding to the potential region from OCP to 0.32 V in the presence of 0.5 mg/L thallium were 1413 µC and 1116 µC for the total and the gold anodic curves, respectively. If the total current was entirely associated with the gold dissolution, it would be equal to 14.57 µg cm-2. The actual dissolved gold based on gold electrode mass loss was 11.52 µg cm-2. The total anodic charge covering the region of the gold active dissolution (OCP to 0.35 and back to 0.018 V) was 1730 µC or 17.85 µg cm-2. The total cathodic charge following the anodic dissolution from 0.018 V to 0.02 V (there was a slight re-dissolution occurring from 0.02 V to 0.06 V or OCP) was 207 µC or 2.138 µg cm-2. The ratio of cathodic-to-anodic charges becomes 0.12, which means that almost 88% of the gold diffused away to the bulk electrolyte. 108 Figure 56: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 and 0.5 mg/L Thallium at scan rate 1 mV/s However, when the scan rate was increased from 1 mV/s to 10 mV/s, a dramatic increase in the gold anodic current, up to 353 µA cm-2, was noted, as shown in Figure 57. This is not surprising since with the increasing scan rate, the flux of electroactive species to the electrode surface will increase. 109 Figure 57: E-i voltammograms in 0.4 M MgS2O3 and 0.5 mg/L Tl at 10 mV/s scan rate 110 Figure 58: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 and 0.5 mg/L Tl at scan rate 10 mV/s From Figure 58, it can be noted that not only the anodic current increased with increasing the scan rate from 1 mV/s to 10 mV/s, but the height of the cathodic current increased as well. At a high scan rate, the lines corresponding to the gold electrode mass decrease/increase are much steeper compared to the slow scan rate and, as a consequence, the regions become better defined where reactions are most likely thermodynamically and kinetically unfavorable. The anodic charge for a 10 mV/s scan rate was 1159 µC or 11.95 µg cm-2 and the cathodic charge was 602 µC or 6.206 µg cm-2. The ratio of the cathodic-to-anodic charge was increased to 0.52 which indicates that half of the dissolved gold was still at the surface when the scan rate was reversed from anodic-to-cathodic polarization. This is to be expected with the increasing scan rate. Figure 59 exhibits the effect of gold added as a gold thiosulphate into the electrolyte containing 0.4 M magnesium thiosulphate and 0.5 mg/L thallium. Figure 60 exhibits the 111 same experiment but presented as a gold anodic current versus potential. As observed in these figures, the anodic current has been significantly reduced and both anodic peaks shifted towards more anodic potential with the addition of gold thiosulphate, which is similar to what was observed in the thallium-free thiosulphate electrolyte. However, in this case, the reduction current responded with an insignificant increase. Figure 59: E-i voltammograms in 0.4 M MgS2O3 and 0.5 mg/L Tl without and with 30 mg/L Au added (1 mV/s scan rate) 112 Figure 60: E-icalc voltammograms in 0.4 M MgS2O3 and 0.5 mg/L Tl without and with 30 mg/L Au added Figure 61 compared the gold electrode mass change in 0.4 M magnesium thiosulphate and 0.5 mg/L thallium at 1 mV/s scan rate with and without the added gold thiosulphate. It is apparent that the gold dissolution was hindered in the thiosulphate electrolyte where gold thiosulphate was initially present. 113 Figure 61: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 and 0.5 mg/L Tl with and without 30 mg/L Au at scan rate 1 mV/s 7.2.3 EFFECT OF THIOUREA The following voltammograms were recorded in 0.4 M thiosulphate of pH 7.5 with and without the presence of 0.05 mM Thiourea. It appears that thiourea not only accelerated the anodic curves but also changed the E-i profiles for both total and calculated currents. Note that in these experiments, the peak in the anodic current is not evident before the switching potential of 0.35 V in the E-i voltammogram presented in Figure 62. But the anodic current peak passing 0.3 V can be seen in E-icalc voltammogram in Figure 63. The anodic-to-cathodic peak-to-peak separation was ~0.37 V. 114 Figure 62: E-i voltammograms in 0.4 M MgS2O3 without and with 0.05 mM thiourea added 115 Figure 63: E-icalc voltammograms in 0.4 M MgS2O3 without and with 0.05 mM thiourea added The calculated and measured current versus potential plots for experiments with a thiourea-thiosulphate electrolyte are presented in Figure 64. Figure 65 plots gold electrode mass change as well as the cumulative charge versus applied potential. -5051015202530-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4i calc/ µA cm-2E / V vs. SHEicalc, 0.4M S2O3, 0.05mM Thioureai, 0.4M S2O3icalc, 0.4M S2O3, 0.05 mM Thioureaicalc, 0.4M S2O3, 0.00 mM Thiourea116 Figure 64: E-i and E-icalc voltammograms in 0.4 M MgS2O3 and 0.05 mM thiourea at 1 mV/s scan rate 117 Figure 65: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 and 0.05 mM thiourea at scan rate 1 mV/s The total anodic charge covering the region of gold active dissolution (OCP to 0.38 and back to 0.127 V) was 1062 µC or 10.95 µg cm2-. The total cathodic charge following the anodic dissolution from 0.127 V to 0.04 V (there was a slight re-dissolution occurring from 0.04 V to 0.08 V or OCP) was 174.9 µC or 1.803 µg cm-2. The ratio of cathodic-to-anodic charge is 0.16, which means that almost 84% of gold diffused out to the bulk electrolyte. The gold electrode mass change as a function of charge estimated based on the total and calculated anodic curves is shown in Figure 66. The slope of mass-to-charge ratio is 1.515 µg mC-1 or 0.74 mol of gold per mol of electrons for q(itotal) which means that the charge that was consumed by something other than the gold oxidation reaction increased 118 slightly compared to the thiourea-free thiosulphate electrolyte (0.78 mol of gold per mol of electrons). Figure 66: Mass change versus the charge ratio calculated by integrating the area under i-E curve in figure 64 Figure 67 depicts the measured or total currents and Figure 68 depicts the calculated currents for an electrolyte that contained thiosulphate, thiourea, and gold thiosulphate. Similar to the thallium-thiosulphate system, the addition of gold thiosulphate into thiourea-thiosulphate electrolyte had no effect on the gold reduction rate during cathodic polarization. Furthermore, a high concentration of gold thiosulphate in the bulk electrolyte showed an insignificant change on the anodic section of the cyclic 119 voltammetry in the presence of thiourea. This insignificant change is better depicted in Figure 69, the gold electrode mass change during the potential scan. Figure 67: E-i voltammograms in 0.4 M MgS2O3 and 0.05 mM thiourea with and without the 30 mg/L gold at 1 mV/s scan rate 120 Figure 68: E-icalc voltammograms in 0.4 M MgS2O3 and 0.05 mM thiourea with and without Au at 1 mV/s scan rate 121 Figure 69: Mass-potential and charge-potential curves at gold in 0.4 M MgS2O3 and 0.05 mM thiourea with and without 30 mg/L Au at scan rate 1 mV/s 7.3 DISCUSSION It is apparent that when a gold electrode was anodized in thiosulphate solution at potentials between OCP and 0.35 V, two reactions took place: the gold dissolution and thiosulphate oxidation to tetrathionate. These reactions (Zhang and Nicol, 2003; Sullivan and Kohl, 1997; Chandra and Jeffrey, 2004; Breuer and Jeffrey, 2002; Woods et al., 2006) can be described as follows: Au + 2S2O32- Au(S2O3)23- + e- Eo =0.153V 7- 1 2S2O32- S4O62- + 2e- Eo =0.08V 7- 2 122 Based on reversible potentials for these reactions, it is obvious that thermodynamically the thiosulphate oxidation to tetrathionate commences at a lower potential than gold dissolution. The data revealed that the anodic current in the 0.4 M MgS2O3 electrolyte of pH 7.8 peaked at ~0.217 V potential, showing a current as low as 5.79 µA cm-2 for both reactions and even lower 4.46 µA cm-2 for the gold anodic oxidation reaction alone. The latter value was calculated based on the gold electrode mass change during cyclic voltammetry. According to Woods et al. (2006), the reaction of thiosulphate oxidation to tetrathionate involves the following steps: S2O32- S2O32-ads 7- 3 S2O32-ads S2O3-ads + e- 7- 4 2S2O3-ads S4O62-ads 7- 5 S4O62-ads S4O62- 7- 6 The adsorbed tetrathionate inhibits the gold anodic dissolution at lower potentials according to (Watling, 2007; Barron et al., 2011; Woods et al., 2006), but the effect seems to be more evident at higher potentials (>0.35 V) due to a decline in the rate of gold anodic dissolution. Based on SERS (Surface Enhanced Raman Scattering Spectroscopy) studies (Watling, 2007; Barron et al., 2011), the tetrathionate is no longer present at potentials higher than 0.35 V but fast oxidation of thiosulphate forms other sulphur-like species such as bulk S8 via an Au-S8 intermediate which, according to these authors, eventually completely blocks any redox reactions at the gold surface. Hence, it might be contemplated that in the low potential region corresponding with this study, where tetrathionate adsorbs on the surface of gold, the gold dissolution occurs by penetration of the reactive species or the transferring of electrons through the film at a reduced rate compared to the free surface. 123 In systems where electron transfer is facilitated due to electrocatalysis, the anodic peak current can be comparable with the limiting current with respect to the ligand, as suggested by Bek and Shuraeva (2008). Applying the Fick’s 1st law the limiting diffusion current (Bek and Shevtsova, 2012) can be calculated as: iLim = FDn C2δ 7- 7 With 0.4 M or 0.4 x 10-3 mol cm-3 thiosulphate concentration (C), n- number of electrons, diffusion coefficient (D) of 1.08 x 10-5 cm2 s-1 (Sullivan and Kohl, 1997), considering that 2 atoms of thiosulphate complexes one atom of gold and assuming thickness of the diffusion layer (δ) equals to 10-2 cm, the limiting current becomes 20 mA cm-2 which is considerably larger than the experimentally-observed values which points to the fact that the charge transfer limitation still persists due to incomplete surface coverage by the activator. When the scan rate was increased from 1 mV/s to 10 mV/s, a dramatic increase in the gold anodic current from 40 to 353 µA cm-2 was noted (Figure 57). This is not surprising since, with the increasing of the scan rate, the flux of the electrochemically-active species to the surface will increase. But the diffusion limiting current, calculated above, is still far from the experimental value even at the higher scan rate, which possibly requires a search for optimal concentration of thallium in the hopes that the conditions will be found when the anodic current approaches the theoretical value. The voltammogram taken at the gold electrode in the thiourea-thiosulphate electrolyte also exhibited accelerated anodic current density but showed a distinctively different profile. After initial passivation, the anodic current peaked at ~0.36 V. The anodic-to-cathodic peak-to-peak separation was the largest 0.372 V in comparison with thallium and thiosulphate alone. Thus, it seems definite that both thallium and thiourea indeed accelerate the anodic dissolution of gold but they produce different E-i and E-icalc profiles at a gold electrode. 124 Although, when plotted in a semi-logarithmic scale (Figure 70), they both show similar slopes, which suggests a similar rate-determining step (or limiting kinetic step), perhaps through the surface activation via adsorption. It is also evident that under conditions adapted in this study, the thallium exhibited a stronger catalytic effect at lower anodic potentials compared to thiourea. Figure 70: Anodic portion of E-icalc voltammogram in semi-logarithmic coordinates Some of the dissolved gold re-deposited on electrode on reversing the direction of potential scan towards cathodic potentials. Due to large anodic-to-cathodic peak separation, only ~12-18% of the dissolved gold re-deposited on the electrode. In a catalyst-free thiosulphate electrolyte, the surface concentration of gold available for reduction was very small producing a very low reduction current. The gold electrodeposition current was seen to increase after the addition of 30 mg/L gold 125 thiosulphate reaching the diffusion-limited peak current density at ~(-0.1) V SHE. The equation for peak current �Ip� value for electrochemically-irreversible systems according to Bard and Faulkner (2000) and Sullivan and Kohl (1997) is: Ip = 2.99 X 105 n (α na)0.5 A C D0.5 υ0.5 7- 8 The calculated peak current using diffusion coefficient value (D) of 1.3 x 10-5 cm2 s-1 published by Guerra and Dreisinger (1999) for gold thiosulphate ions, 0.23 (α) transfer coefficient value published by Sullivan and Kohl (1999), the calculated current for one electron transfer (n, na) processes containing 30 mg/L or 30 x 10-3 mg cm-3 Au(S2O3)3- (C) at 1 x 10-3 V s-1 scan rate (υ) equals to 4.97 x 10-7 A or 2.51 µA cm-2 for electrode with a surface area of 0.198 cm-2 (A). The experimentally-observed gold cathodic current (icalc) was 2.35 µA cm-2 and is close to the theoretical value (2.51 µA cm-2) calculated by the equation which confirms that the cathodic current that appeared at close to (-0.1) V potential represents gold reduction. The addition of gold thiosulphate had no effect on the cathodic current when a large amount of surface gold was available for reduction, as evidenced in the presence of electrocatalysts. It seems obvious that the accumulation of thiosulphate during the gold deposition reaction, which is relatively stable in this potential region, becomes a limiting factor for gold re-deposition. The goal of the present study was to contribute to the understanding of the gold redox activity by coupling cyclic voltammetry with EQCM. The focus was to define the gold redox state as a function of applied potential. 126 8 CONCLUSIONS In-situ thiosulphate leaching of gold Although there are some discussions about the formation of thiosulphate during alkaline oxidation of pyrite in literature, no information is available describing the kinetics of the pyrite sulphide to a thiosulphate reaction or a predictive equation of thiosulphate yield. Further, there is limited information available regarding the conditions necessary for effective dissolution of gold with in–situ-generated thiosulphate. Several experiments were conducted in this study to investigate the leaching of gold with thiosulphate that was generated by the simultaneous oxidation of pyrite. It was found that conditions do indeed exist (20 psi oxygen pressure, 80OC temperature) where the sulphide oxidation and gold dissolution reactions occur simultaneously, but the conditions must be carefully optimized to achieve the desired gold recovery. The oxidation of the sulphide produces thiosulphate and also generates acid. In order to stabilize the thiosulphate in solution and maximize the rate of sulphide oxidation, it is important to neutralize this acid and maintain a pH of >12. At pH <12, thiosulphate oxidizes to polythionates, sulphite and, sulphate, and the rate of thiosulphate decomposition increases as the pH decreases. Therefore, a measure of compromise must be reached in optimizing the solution chemistry for the two desired reactions, since the optimum pH for the gold dissolution reaction is between pH 8 and 10. Gold dissolution is very slow at pH >10 while thiosulphate decomposition is rapid at pH <8. Only a limited amount of experimentation was conducted on the first pyrite concentrate (PY1) but, under the best conditions, 65% of the sulphide was oxidized and 72% of the gold was leached with the in-situ-generated thiosulphate over a period of 28 h. More tests were conducted and better results achieved with the second pyrite concentrate (PY2). In 127 this case, the best result achieved was 60% sulphide oxidation in 30 h, which was accompanied by 96% gold dissolution and 74% silver dissolution, respectively. An equation predicting thiosulphate yield as a product of rate constants and an oxygen pressure-dependent term has been derived. The derivation was based on a conceptual model for the progress of a series of reactions in which sulphide sulphur is transformed to various soluble oxyanions under the oxidative conditions. The transformation mechanism was constructed on the least electron transfer principle and was assumed that the reactions occur in series. Such a model may be simplistic and imprecise in terms of the individual kinetic constants employed, yet the thiosulphate yield predictive power of the model was reasonable and could be useful for further studies in this field. Electrochemistry A significant amount of literature is devoted to thiosulphate leaching of gold but the majority of them discuss the gold anodic dissolution and provide certain additives (copper, ammonia, thiourea, thallium, etc.) to improve the gold dissolution kinetics. However, there is a lack of fundamental knowledge of gold redox activity that explains the observed gold losses during leaching in thiosulphate. The characterization of gold anodic and cathodic behaviour in various compositions of thiosulphate electrolyte by using cyclic voltammetry in conjunction with EQCM was the novel approach taken in this study. The cyclic voltammetry provided a total current value and the EQCM rate of mass change provided insight into the amount of gold dissolved/ deposited during the anodic and cathodic sweeps. The conversion of mass change to current allowed the development of icalc versus E plots to compare with the total current. In this way, the true behaviour of gold could be discriminated. 128 The results show that the gold anodic dissolution peak in thiosulphate appears close to 0.217 V at 1 mV/s. This anodic peak current shifts towards more anodic values in the presence of catalysts. In the thallium-thiosulphate electrolyte, the anodic peak appears at 0.226 V at 1 mV/s and 0.310 at 10 mV/s. In the thiourea-thiosulphate system, the anodic current peaked at 0.361 V. The anodic-to-cathodic peak-to-peak separation was in the range of 0.26 – 0.37 V. Owing to this large peak-to-peak separation, the cathodic-to-anodic charge ratio was between 0.12-0.17. This charge ratio increased to 0.52 with increasing the potential scan rate from 1 to 10 mV/s. Using electrochemical techniques, this study attempted to determine if the loss of gold in the thiosulphate leaching system is prompted by the reversible nature of the gold thiosulphate complex (reversing back to metallic state), especially in the presence of electrocatalysts, by providing much higher rates of redox activities. It has been proven that the anodic current is significantly affected in the presence of electrocatalysts such as thallium and thiourea, however, the anodic and cathodic peak-to-peak separation remained large enough for allowing the dissolved gold to diffuse into the bulk electrolyte. This means that the diffusion rate is always larger than the charge transfer rate, even in the presence of electrocatalysts under conditions adopted in this study. It is also contemplated that released thiosulphate during gold thiosulphate deposition (surface gold thiosulphate produced during anodic polarization) might have increased the overpotential for the gold reduction current in this potential range. This view is based on an observation that the addition of gold thiosulphate in the bulk electrolyte did not show any apparent effect on the gold reduction current. The effect of the gold thiosulphate addition was seen in the thiosulphate electrolyte without electrocatalysts present where surface concentration of gold, produced during anodic polarization and owing to the diffusion processes, was very small. In this case, the known bulk concentration of gold thiosulphate and corresponding constants allowed the calculation of a theoretical peak current that closely predicted the experimentally-obtained gold reduction peak. 129 One of the problems in the thiosulphate leaching of gold is the high consumption of thiosulphate and the generation of its degradation products. The simultaneous measurement of mass variation and current as a function of potential confirmed the oxidation of thiosulphate occurring concurrent to gold oxidation. Although it can be argued that the side reaction is not necessarily the oxidation of thiosulphate in the presence of electrocatalysts, it can be deduced based on this study (gold mass change versus total charge passed) that the side reaction proceeds at the slowest rate in the thallium-thiosulphate electrolyte and at the highest rate in the thiourea-thiosulphate electrolyte compared to thiosulphate alone. It was also evidenced that the thallium exhibited a stronger catalytic effect at lower anodic potential compared to thiourea under conditions adapted in this study. 130 9 RECOMMENDATIONS Further work on the research for the conditions of the gold thiosulphate reversibility could be carried out using the REQCM. The advantage of this method is that steady state is attained quickly and a number of equations are available to determine the rate constants precisely in order to develop a predictive electrochemical model for the gold thiosulphate system. These rate constants will show how the current is affected by the different experimental variables. The influence of additives can also be quantified. It might be possible to find the combination of factors that brings the calculated kinetic limited current to its minimum value (i.e. to move the system to mass transfer control and away from electrochemical kinetic control). Slow dissolution of gold with in-situ-generated thiosulphate is due to a low oxidative potential of alkaline solution of pH >12. The addition of thallium or thiourea might improve the gold leaching kinetics based on electrochemical studies. However, it is important to study the effect of these additives on sulphide oxidation in parallel to gold dissolution. In addition, the concentration of thiosulphate in solution at any point in time is a function of the rates of both sulphide oxidation and thiosulphate decomposition, and the optimum conditions will occur when the rate of sulphide oxidation is enhanced and the rate of thiosulphate decomposition is retarded. The search for possible catalysts for sulphide oxidation that will have the least affect on thiosulphate decomposition could be beneficial for any future work in this area. Further research should also be undertaken to determine the rates of these three important reactions (sulphide oxidation, thiosulphate yield, and gold dissolution) at various experimental conditions and including other sulphide minerals such as arsenopyrite and energite. 131 Thiosulphate is an attractive alternative to cyanide for the leaching of gold. The classic thiosulphate leaching process involves use of the copper/ammonia catalyzed leach chemistry. As pyrite produces thiosulphate during alkaline oxidation with a predictive equation for thiosulphate yield that has been developed, this could be an ideal opportunity for a development of a novel leaching system of gold. The next step is to complete the process flowsheet by including the necessary recovery and possible recycling stages and developing an integrated process cost model. 132 REFERENCES Adams, M.D., Johns, M.W., & Dew, D.W. (1999). Recovery of gold from ores and environmental aspects. Gold: Progress in Chemistry, Biochemistry, and Technology. Ahern, N. Dreisinger, D. Van Weert, G. (2006). Kinetics of trithionate degradation.Canadian Metallurgical Quarterly, Vol. 45, No. 2 pp 135-144. Ahern, N. (2005). Thiosulphate degradation during gold leaching in ammoniacal thiosulfate solutions: A focus on Trithionate. Ph.D. Thesis. The University of British Columbia Avrahami, M. Golding, R.M. (1968) J.CHem, Soc. A 647 Aylmore M.G., Muir D.M. (2001) ‘Thiosulfate leaching of gold – a review’, Minerals Engineering, 14, 135-174. Aylmore M.G., Muir D.M. (2001) ‘Thermodynamic analysis of gold leaching by ammoniacal thiosulfate using Eh/pH speciation diagrams’, Minerals and Metallurgical Processing, 18 (4), 221-227 Aylmore, M.G. (2014) Short Course notes. Hydrometallurgy 2014. 7th International Symposium. Burke, L.D., Nugent, P.F. (1997) ‘The electrochemistry of gold: the redox behaviour of the metal in aqueous media’, Gold Bulletin, 30(2) pp43-53. 133 Breuer, P.L., Jeffrey M.I. (2000) ‘Thiosulphate leaching kinetics of gold in the presence of copper and ammonia’, Minerals Engineering, 13 (10-11), 1071-1081. Breuer, P.L., Jeffrey M.I., Tan E.H.K., Bott A.W. (2002) ‘The electrochemical quartz crystal microbalance as a sensor in the gold thiosulfate leaching process’, Journal of Applied Electrochemistry, 32, 1167-1174. Baron, I.Y., Mirza J., Frydrychewicz, A., Lipkowski, J., Choi, Y.(2011) ‘ ‘Electrochemical studies of the gold-electrolyte interface under thiosulphate based leaching conditions’, COM 2011 Basolo, F. and Pearson, R.G. (1967). Mechanism of inorganic reactions: A study of metal complexes in solution. J.Wiley&Sons. Bengtsson, S. Bjerles, I. (1975). Catalytic oxidation of sulphite in diluted aqueous solutions. Chemical Engineering Science Vol. 30, Issue 11. p p.1429-1435 Breuer, P.L. and Jeffrey, M.(2000). Thiosulphate Leaching Kinetics of Gold in the presence of copper and ammonia. Miner. Eng., 13 (10-11): pp. 1071-1081. Bard, A.J., Faulkner, L.R. (2000). Electrochemical methods: Fundamentals and Applications. John Wiley and Sons Ltd. Bailey, L.K., Peters, E. (1976) Decomposition of pyrite in acids by pressure leaching and anodization: the case for an electrochemical mechanism. Canadian Metallurgical Quarterly, Vol. 14, No. 4. 134 Burkin, A.R. and Edwards, A.M., (1963). The formation of insoluble iron oxide coatings during the alkali pressure leaching of pyrite. Proc. 6th Int. Congr.Mineral Processing (Cannes), pp. 61-67 Burkin, A.R., (1969). Solid-state transformations during leaching: Mineral Sci. And Eng., V. 1, pp 4-14 Byerley, J.J., Fouda. S.A., Rempel, G.L., (1973). Kinetics and mechanism of the oxidation of thiosulfate ions by copper (II) ions in aqueous ammonia solutions. J.Chemical Society, Dalton Transactions, 889-893. Byerley, John J., Fouda, Safaa A. and Rempel, Garry L., (1975). Activation of copper(ii) ammine complexes by molecular oxygen for the oxidation of thiosulphate ions. J.C.S. Dalton. pp 1329-1338. Bek, R.Y., Shuraeva, L.I. (2008) ‘Electrocatalysis by adatoms at the gold and silver dissolution in cyanide solution’, Russian Journal of Electrochemistry, Vol. 44, No 1, pp 113-122. Bek, R.Y., Shevtsova, O.N. (2013) ‘The effect of time on electrode exposure to the solution on gold dissolution rate in thallium-containing thiosulphate electrolytes’, Russian Journal of Electrochemistry, Vol. 49, No 3, pp 203-208. Bek R.Y., Shevtsova, O.N. (2012) ‘ The effect of thallium ions on the gold dissolution rate in thiosulphate electrolytes’, Russian Journal of Electrochemistry, Vol. 48, No 11, pp 1046-1051. 135 Breuer, P.L., Jeffrey, M.I. (2002) ‘An electrochemical study of gold leaching in thiosulphate solutions containing copper and ammonia’, Hyrdometallurgy 65 pp 145-157. Buisman, C., Ijspeert, P., Janssen, A., Lettinge, G. (1990) Kinetics of chemical and biological sulphide oxidation in aqueous solutions. Wat. Res. Vol. 24, No 55, pp 667-671. Bizzotto, D. (2015) Personal communications. UBC, Vancouver Cao, C. , Hu, J. And Gong, Q. (1992). Leaching gold by low concentration thiosulphate solution. In Randol Gold Forum 92, Vancouver. (Randol Intl. Ltd.: Golden, Co, USA) pp. 193-198. Caldeira, C.L., Ciminelli, V.S.T., Dias, A. Osseo-Asare, K. (2003). Pyrite oxidation in alkaline solutions: nature of the product layer. Int. J. Miner. Process. 72. 272-286 Chandra, I., Jeffrey, M.I. (2004) ‘An electrochemical study of the effect of additives and electrolyte on the dissolution of gold in thiosulfate solutions’, Hydrometallurgy, 73, 305-312. Choo, W.L., Jeffrey, M.I. (2003) ‘A rotating EQCM study of the effect of electrolytes, additives and metal impurities on the electrodeposition of gold(I) thiosulphate’, Electrochemical Society Proceedings Volume 2003-18, pp 292-303. Chen, K.Y., and Morris, J. C., (1970). Oxidation of aqueous sulfide by O2: 1-General characteristics and catalytic influences: Internat. Conf. on Water Pollution Research , 5th, San Francisco, Calif., July-August 1970, Paper III-32. 136 Chen, K.Y., and Morris, J. C., (1972). Kinetics of oxidation of aqueous sulfide by O2: Environmental Sci. And Technology, v.6, p. 529-537. Ciminelli, V.S.T. (1987). Oxidation of pyrite in alkaline solutions and heterogenous equilibria of sulfur- and arsenic-containing minerals in cyanide solutions. Ph.D. Thesis. Pennsylvania State Univ., Pa. Ciminelli, V.S.T. and Osseo-Asare, K. (1995) Kinetics of pyrite oxidation in sodium hydroxide solutions. Metallurgical and Materials Transactions, Vol. 26B, pp 677-685. Deakin, M.R., Buttry, D.A. (1989) Electrochemical Applications of the Quartz Crystal Microbalance. Analytical Chemistry, Vol 61, No.20, pp 1147-1154. Druschel, G.K., Hamers, R. J., and Banfield, J.F.,(2003). Kinetics and mechanism of polythionate oxidation to sufate at low pH by O2 and Fe3+. Geochimica et Cosmochimica Acta, Vol. 67, No. 23, pp 4457-4469. Druschel, G., Borda, M. (2006) Comment on “Pyrite dissolution in acidic media” by M.Descostes, P.Vitorge, and C. Beaucaire. Geochmical et Cosmochimica Acta 70. Pp 5246-5250. Descostes, M., Vitorge, P., Beaucaire, C. (2004). Pyrite dissolution in acidic media. Geochimica et Cosmochimica Acta, Vol. 68, No. 22, pp 4559-4569. Davis, R.E. (1958) Displacement reactions at the sulfur atom. I. An interpretation of the decomposition of acidified thiosulfate: Am. Chem. Soc.Jour., V. 80, pp 3565-3569. 137 Fava, F., Divo, D., (1953). Studies of the decomposition of polythionates in alkaline medium and in neutral medium by the aid of radioactive sulfur. Chem. Abstr. 47, 11914d. Fava, F., Divo, D., (1954). Decomposition of solutions of polythionates by means of radioactive sulfur. Chem. Abstr. 48, 4349e. Fave, A. Bresadola, S. (1955).“Kinetics of the catalytic rearrangement of tetrathionate”. J. America Chem. Soc., Vol. 77, 5792-5794. Fleming, C A; McMullen, J; Thomas, K G; Wells, J A (2003). Recent advances in the development of an alternative to the cyanidation Process: Thiosulfate leaching and resin in pulp. Minerals and Metallurgical Processing. February 1, 2003. Fleming, C A. (2014). Personal Comunication. SGS Minerals. Lakefield. Feng, D., van Deventer, J.S.J. (2005). Thiosulphate decomposition in the presence of sulphides. Treatment of Gold ores. Proceedings of the 44th Ammual Conference of Metallurgists of CIM. Calgary, Alberta, Canada, pp 129-143. Feng, D., van Deventer, J.S.J. (2007) The effect of sulphur species on thiosulphate leaching of gold. Minerals Engineering 20, pp. 273–281. Flett, D.S., Derry, R. And Wilson, J.C. (1983).Chemical Study of Thiosulfate Leaching of Silver Sulpfide. Trans, Inst. Min. metal., Sect. C. 92 (Dec), pp. 216-223. Forward, F.A., Peters, E., Majima, H., (1964). Paper presented at AIME Annual Meeting, Feb 1964, New York. 138 Gluud, W. Ber. Dtsch. CHem. Ges. B (1921), 54B, 2425 (1921), Goehring, M., Heibing, W., Appel, I., (1949). Spontaneous decomposition of polythionate solutions. Chem. Abstr. 43, 2886e. Goldhaber, M.B., (1983). Experimental study of metastable sulfur oxyanion formation during pyrite oxidation at pH 6-9 and 30oC. American Journal of Science., Vol 283, March, pp 193-217 Glasstone, Hickling. (1954) Journal of the Chemical Society, 1932, p2345, p2800 and 1933, p 829, cited in Partington JR, ‘General and Inorganic Chemistry for University Students’, 2nd edition, Macmillan and Co Ltd., London, p726. Gray, P.M.J. (1955-56) Tran.Inst. Min. Metall., Vol. 65, pp 55-65. Gudkov, A.S. Zhuchkov, I.A., Mineev, G.G. (2010). Thermodynamics of interaction between sulfide-thiosulfate solutions and noble metals. Russian Journal of Non-Ferrous Metals, Vol. 51, No. 3, pp 233-236. Guerra, E., Dreisinger, D.B., (1999). A study of the factors affecting copper cementation of gold from ammoniacal thiosulphate solution. Hydrometallurgy 51: pp 155-172 Jiang, T.,Chen, J. And Xu, S. (1993), Electrochemistry and mechanism of leaching gold with ammoniacal thiosulphate. In XVIII International Mineral Processing Congress, Publ. Series No. 3/93, Vol. 5. (Australias. Inst. Min. Metall.: Parkville, Vic), pp 1141-1146. 139 Jiang, T., Xu, S., Chen, J. And Wu, Z.(1992). A theoretical study on gold extraction by thiosulfate leaching: chemical and thermodynamical principles of gold leaching. Huangjin, 13(2), pp 31-34. Kleinjan, Wilfred E., de Keizer, Arie, Janssen, Albert J.H. (2005).Kinetics of the chemical oxidation of polysulfide anions in aqueous solution. Water Research 39 pp 4093–4100 Kabasakaloglu, M., Bilgic, S. (1990) ‘The effect of thallium (I) salts on the electrochemical properties of gold ‘Journal of Islamic Academy of Sciences 3:4, pp 287-291 Klets, V.E. Emelianov, Y.E., Rashkovski, G.B., (1987). Stability of calcium thiosulphate under hydrothermal conditions. Izvestia Vysshikh Uchebnykh Zavedeni. Tstvetnaia Metallurgia. Severo-Kavkazskii gorno-metallurgicheski Ordzhonikidze. Vol 5, pp 53-57. Kiknadze, N.P. (1991). Fundamental chemistry for processing of complex copper and manganese containing concentrates. Ph.D. Thesis. Academy of science, Institute of inorganic chemisty and electrochemisty. Tbilisi, Georgia Kurtenacker, A., Mutschin, A., Stastny, F., (1936). Spontaneous decomposition of polythionate solutions. Chem. Abstr. 30, 980. Kirk-Othmer, (2006) Kirk-Othmer Encyclopedia of Chemical Technology. Wiley. Fifth eddition. Volume 23. Lyons, D., and Nickless, G., (1968). The lower oxy-acids of sulfur, in G. Nickless, ed., Inorganic sulfur chemistry: Amsterdam, Elsevier, pp 509-533. 140 Lefers, J.B., Koetsier, W.T., Van Swaaij, P.M. (1978) The oxidation of sulphide in aqueous solutions. The Chemical Engineering Journal, 15, pp 111-120. Li, Cheng (2003) Fundamental aspect of gold leaching in thiosulfate solution. MSc Thesis. The University of British Columbia. Levenspiel, O.(1999) Chemical reaction engineering. John Wiley & Sons. Third addition. Lam, V.W.S., Kannangara, D.C.W., Alfantazi, A., Gyenge, E.L. (2011) Electrochemical Quarts Crystal Microbalance Study of Borohydride Electro-Oxidation on Pt: The Effect of Borohydride Concentration and Thiourea Adsorption. J. Phys. C, 115, pp 2727-2737 Marsden, J.O., & House, C.I., 2006. The chemistry of gold extraction. London, Ellis Horwood. Mishra, K.K.,Osseso-Asare, K. (1988). Aspects of the interfacial electrochemistry. J. Electrochem. Soc.: Electrochemcial Science and Technology. Oct 1988. Vol 135, No 10. Pp 2502- 2509. Muir D.M., Aylmore M.G. (2002) ‘Thiosulphate as an alternative to cyanide for gold processing – issues and impediments’, Green Processing Conference, Aus IMM, Cairns, May 2002, draft copy, no page numbers available. Majima, H., Peters, E. (1966). TMS-AIME, Vo. 236, pp 193-217. 141 McKibben, M.A., Barnes, H.L. (1986) Geochim.Cosmochim. Acta, Vol. 50, pp 1509-1520. Naito K., Yoshida M., Sheih M.C., Okabe T. (1970a) ‘The chemical behaviour of low valence sulfur compounds. III. Production of ammonium sulfamate by the oxidation of ammonium thiosulfate’, Bulletin of the Chemical Society of Japan, 43, 1365-1372. Naito K., Hayata H., Mochizuki M. (1975) ‘The reactions of polythionates. Kinetics of the cleavage of trithionate ion in aqueous solutions’, Inorganic and Nuclear Chemistry, 37, 1453-1457. Naito. K., Shieh, Mu-Chang, Okabe, T., (1970). The chemical behavior of low valence sulfur compounds. V. Decompositon and oxidation of tetrathionate in aqueous ammonia solution. Bulletin of the Chemical Society of Japan. Vol 43, pp 1372-1376. Nicol, M.J. and O’Melley, G. (2002) Recovering gold from thiosulfate leach pulps via Ion Exchange. JOM October pp 44-46 Rimstidt, J.D., and Vaughan, D.J. (2003). Pyrite oxidation: A state-of-the-art assessment of the reaction mechanism. Geochimica et Cosmochimica Acta, Vol. 67, No. 5, pp 873-880. Rolla, E. and Chakrabarti, C.L., (1982). Kinetics of decompositions of tetrathionate, trithionate, and thiosulphate in alkaline Media. Environ. Sci. Technol. 16, pp 852-857 Puddephatt, R.J. (1978). The Chemistry of Gold. (Amsterdam: Elsevier Scientific Publishing Company). 142 Puddephatt, R.J. (1987). Comprehensive Coordination Chemistry. Pergamon. Oxford. Pedraza A.M. Villegas I., Freund P.L. (1988) ‘ Electro-oxidation of thiosulphate ion on gold study by means of cyclic voltammetry and auger electron spectroscopy’, J. Electroanal. Chem., 250 pp 443-449. Quyang, M. (2001) The study of the kinetics of gold leaching in the ammonium thiosulfate system in the presence of copper. MSc Thesis. The University of British Columbia. Smith, E.E., Shumate, K.S. (1970). No. 14010 FPS02/70, U.S. Department of Interior, Washington, DC. Smith, C.W., Hitchen, A. (1976) ‘Aqueous solution chemistry of polythionates and thiosulphate: a review of formation and degradation pathways’, CANMET Mineral Sciences Laboratories Report MRP/MSL 76-223 (LS). Sullivan, A.M., Kohl, P.A. (1997) ‘Electrochemical Study of the gold thiosulphate reduction’, J. Electrochem. Soc., Vol. 144, No.5 pp1686-1690 Senanauake, G. (2004). Gold leaching in non-cyanide lixiviant system: critical issues on fundamentals and applications. Miner. Eng. 17, 785-801 Stenhouse, J.F., Armstrong, W.M. (1952). Can. Min. Met. Bull., pp 49-53 143 Stacey, D., (2005) The development of the alkaline sulphide gold leaching process. Master’s Thesis. Montana Tech of the University of Montana. Tozawa, K., Inui, Y. and Umetsu, Y. (1981). Dissolution of gold in ammoniacal thiosulphate solution. Paper presented at 110th Annual Meeting of AIME, Chicage (TSM-AIME: Warrendale, PA), Paper selection A-81-25: 12pp. Tributsch, H., Gerischer, H. (1976). The oxidation and heating of metal sulphides as an electrochemical corrosion phenomenon. Journal of Applied Chemistry and Biotechnology. Vol. 26, Issue 1, pp 747-761 Williamson, M.A. and Rimstidt, J.D. (1992) Correlation between structure and thermodynamic properties of aqueous sulfur species. Geochimica et Cosmochimica Acta Vol. 56, pp 3867-3880 Webster, J.G. (1984). Thiosulphate Complexing of Gold and Silver during the oxidation of a sulphide-bearing carbonate lode system, upper ridges mina, P.N.G.. The Aus.I.M.M. Perth and Kalgoorlie Branches, Regional Conference on “Gold-Mining, Metallurgy and Geology” October 1984. Wroblowa H., Green M. (1963). Asorption of organic compounds at the metal/solution interface: thiourea on gold electrodes. ELectrochimica Acta, Vol. 8, pp 679-692. Warren, J.H. (1956) .Aus.J.Appl. Sci., Vol. 7, pp 346-358. Watling, K. M. (2007) Spectroelectrochemical Studies of Surface Species in the Gold/Thiosulfate System. Ph.D. thesis. Griffith University, Australia 144 Woodcock, J.T.. The Aust. IMM, (1961), No. 198, pp. 47-84. Woods, R., Hope, G.A., Watling, K.M., Jeffrey, M.I. 2006. A spectroelectrochemical study of surface species formed in the gold/thiosulphate system. Journal of The Electrochemical Society. 153 (7) pp 105-113 Zipperian, D, Raghavan, S, and Wilson, J.P. (1988). Gold and Silver Extraction by Ammoniacal Thiosulphate Leaching from a Rhyolite Ore. Hydrometallurgy, 19(3), pp.361-375. Zhang, H. Dreisinger, D.B. (2002) The kinetics for the decomposition of tetrathionate in alkaline solutions. Hydrometallurgy 66, 59–65 Zhang, S. (2004) Oxidation of refractory gold concentrates and simultaneous dissolution of gold in aerated alkaline solutions. Ph.D. Thesis. Murdoch University, Western Australia. Zelinsky A.G., Novgorodtseva O.N. (2013) EQCM study of the dissolution of gold in thiosulphate solutions’ , Hydrometallurgy, 138 pp 79-83 Zhang S., Nicol M.J. (2003) ‘Electrochemical study of the dissolution of gold in thiosulphate solutions Part 1: Alkaline solutions’, Journal of Applied Electrochemistry 33: pp 767-775. 145 Xu, Y. and Schoonen M.A.A. (1995). The Stability of Thiosulphate in the presence of Pyrite in Low-temperature Aqueous Solutions. Geochimica et Cosmochimica Acta, Vol.. 59, No 22, pp. 4605-4622. Yen, W.T., Guo, H. And Deschenes, G. (1999). Development in percolation leaching with ammonium thiosulphate for gold extraction of a mild refractory ore. In EPD Congress 1999, Proc. TMS Annu. Meet., San Diego. (Ed.) Mishra, B. (TMS: Warrendale, PA), pp. 441-455. "@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2016-02"@en ; edm:isShownAt "10.14288/1.0220849"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Materials Engineering"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "Attribution-NonCommercial-NoDerivs 2.5 Canada"@* ; ns0:rightsURI "http://creativecommons.org/licenses/by-nc-nd/2.5/ca/"@* ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Kinetics of sulphide conversion to thiosulphate and gold dissolution study by electrochemical quartz crystal microbalance"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/55724"@en .