UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Fundamental aspects of gold leaching in thiosulfate solutions Li, Cheng 2003

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

Item Metadata

Download

Media
831-ubc_2003-0505.pdf [ 5.4MB ]
Metadata
JSON: 831-1.0078636.json
JSON-LD: 831-1.0078636-ld.json
RDF/XML (Pretty): 831-1.0078636-rdf.xml
RDF/JSON: 831-1.0078636-rdf.json
Turtle: 831-1.0078636-turtle.txt
N-Triples: 831-1.0078636-rdf-ntriples.txt
Original Record: 831-1.0078636-source.json
Full Text
831-1.0078636-fulltext.txt
Citation
831-1.0078636.ris

Full Text

F U N D A M E N T A L A S P E C T S OF G O L D L E A C H I N G I N T H I O S U L F A T E S O L U T I O N S by C H E N G LI M . E . , Kunming University of Science and Technology, P. R. China, 1988 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T F O R T H E D E G R E E O F M A S T E R OF A P P L I E D S C I E N C E in T H E F A C U L T Y OF G R A D U A T E S T U D I E S Department of Metals and Materials Engineering We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A June 2003 © Cheng L i , 2003 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Metals and Materials Engineering The University of British Columbia Vancouver, Canada Date: June 18, 2003 j ABSTRACT The kinetics and mechanism of gold leaching in the ammonia-thiosulfate-copper ( A T S -Cu) system have been studied using the rotating electrochemical quartz crystal microbalance ( R E Q C M ) . Anodic polarization, cathodic polarization and leaching experiments were included in this study. The effect of concentration of different reagents, applied potential, p H , temperature, and electrode rotating velocity on the gold leaching rate have been investigated. The effect of solution copper species on the anodic reaction of gold dissolution in thiosulfate solution was elucidated using the R E Q C M . With respect to the role of copper on the anodic processes, two possible mechanisms were proposed. It was shown that, in the absence of S2O3 2", the electrochemical reaction on the cathode is the reduction of C u ( N H 3 ) 4 2 + to C u ( N H 3 ) 2 + in the potential range of 0.2 to -0.3 V vs. S H E . Based on other experimental results, it is believed that, in the presence of S2O3 2", the electrochemical reaction on the cathode also is the same as the reaction in the absence of S2O3 2". It was found that the anodic process and leaching process are under chemical reaction control, but the cathodic process is under diffusion control. It may therefore be concluded that the leaching process is under anodic control under the majority o f conditions tested (cathodic control w i l l be important when the cupric ammine species is very dilute in solution). Experiments show that increasing the ratio of NH3/S2O32" w i l l favor the leaching rate. However, excess thiosulfate or excess ammonia may inhibit the dissolution of gold. The effect of selected additives was also studied. i i i TABLE OF CONTENTS ABSTRACT .ii TABLE OF CONTENTS iv LIST OF TABLES vi LIST OF FIGURES vii ACKNOWLEDGEMENTS . xvi NOMENCLATURE xvii 1 INTRODUTION 1 2 LITERATURE REVIEW 4 2.1 Aqueous Chemistry 4 2.2 Stability of thiosulfate solution 16 2.3 Leaching of gold in thiosulfate system 19 2.4 Research on the kinetics and mechanism 25 2.4.1 Passivation phenomena 25 2.4.2 Kinetics study of gold thiosulfate leaching 27 2.4.3 Mechanism of gold thiosulfate leaching 37 2.5 Additives studies 44 3. EXPERIMENTAL APPROACH 48 3.1 Introduction 48 3.2 Basic principle of REQCM 49 3.3 REQCM 56 3.3.1 Design 56 3.3.2 Equipment set-up 58 3.3.3 Validation of the equipment 61 3.4 Experimental 64 3.4.1 Reagents 64 3.4.2 Experimental procedure and conditions 65 4. RESULTS AND DISCUSSION 69 4.1 Anodic polarization studies 69 iv 4.1.1 Preliminary experiments 69 4.1.2 Anodic polarization studies 74 4.1.2.1 Anodic polarization in the absence of copper 74 4.1.2.2 Anodic polarization in the presence of copper 79 4.1.3 Summary 90 4.2 Cathodic polarization studies 91 4.2.1 Preliminary experiments 91 4.2.2 Cathodic polarization studies 93 4.2.3 Summary 103 4.3 Leaching studies 104 4.3.1 Preliminary tests 104 4.3.2 Leaching studies under open potential 105 4.3.2.1 Leaching with higher concentrations of reagents 105 4.3.2.2 Leaching with lower concentrations of reagents 109 4.3.3 Leaching studies under applied potential I l l 4.3.3.1 Leaching with higher concentration of reagents I l l 4.3.3.2 Leaching with lower concentration of reagents 120 4.3.4 Summary 128 4.4 Additives studies 129 4.4.1 Anodic polarization studies : 129 4.4.2 Leaching studies 135 4.4.3 Summary 140 4.5 The mechanism 141 4.5.1 Anodic process 141 4.5.2 Cathodic process 143 4.5.3 The model of electrochemical mechanism 144 5 CONCLUSIONS AND RECOMMENDATIONS 146 5.1 Conclusions 146 5.2 Recommendations 149 6 REFERENCES 150 v LIST OF TABLES Table 2.1 A summary of various thiosulfate leaching conditions (Aylmore and Muir , 2000b) 23 Table 3.1 The specifications and sources o f reagents 64 Table 4.1 Influencing factors for gold oxidation on anodic polarization( no copper) 90 Table 4.2 Influencing factors for gold oxidation on anodic polarization (with copper).. 90 Table 4.3 The influence o f variables on cathodic current response 103 Table 4.4 Calculated values o f Arrhenius activation energy values for gold leaching at different applied potentials 120 Table 4.5 Influencing factors for gold oxidation on leaching tests 128 Table 4.6 Influencing factors for gold oxidation on leaching tests 128 Table 4.7 Influencing factors for gold oxidation on leaching tests 128 Table 4.8 Influencing factors for gold oxidation on leaching tests 129 Table 4.9 Effect of additives for gold oxidation on additives tests 140 Table 4.10 Effect of additives for gold oxidation on additives tests 140 v i LIST OF FIGURES Figure 2.1 A u - N H 3 - S 2 0 3 2 " system (conditions: 5 x l O ^ M A u ; 1 M S 2 0 3 2 " ; 1 M NH3/NP1/, 0.05M Cu 2 + ) (Aylmore et al, 2001a) 5 Figure 2.2 AU-NH3-S2O3 2 " system (conditions: S x l O ^ M A u ; 0 .1M S 2 0 3 2 ~; 0 .1M N H 3 / N H 4 + , 0 .05M C u 2 + ) (Aylmore et al, 2001a) 5 Figure 2.3 Effect of thiosulfate concentration on the rest potential of gold(Li et al,1996). 6 Figure 2.4 Eh-pH diagram of the gold-thiosulfate-ammonia-water system at 25°C. The activities o f the species are 2.5 xlO" 5 M A u , 0.2 S 2 0 3 2 " and 0.4 M N H 3 . AG f ° (S 2 0 3 2 " ) = -518.8 kJ/mol (Mol leman et al, 2002) 7 Figure 2.5 C u - N H 3 - S 2 0 3 2 " system (conditions: 5 x l O " 4 M A u ; 1 M S 2 0 3 2 " ; 1 M N H 3 / N H 4 + , 0 .05M Cu)(Aylmore et al, 2001a) 8 Figure 2.6 C u - N H 3 - S 2 0 3 2 " system (conditions: S x l O ^ M A u ; 0 .1M S 2 0 3 2 " ; 0 .1M N H 3 / N H 4 + , 0 .05M Cu 2 + ) (Aylmore et al, 2001a) 9 Figure 2.7 Distribution of copper species at different E h conditions for high reagent concentration (1 M S 2 0 3 2 \ 1 M N H 3 , 0.05 M C u , pH=10.0) (Aylmore et al, 2001a) 10 Figure 2.8 Distribution of copper species at different E h conditions for low reagent concentration (0.1 M S 2 0 3 2 " , 0.1 M N H 3 , 5X10" 4 M C u , pH=10.0) (Aylmore et al, 2001a) 10 Figure 2.9 Distribution of copper species at different p H concentrations for high reagent concentration (1 M S 2 0 3 2 \ 1 M N H 3 , 0.05 M C u , Eh=0.250 V ) (Aylmore et al, 2001a) 11 Figure 2.10 Distribution of copper species at different p H concentrations for low reagent concentration (0.1 M S 2O 3 2",0.1 M N H 3 , 5x l0" 4 M C u , Eh=0.250 V ) (Aylmore et al, 2001a) 11 Figure 2.11 Distribution o f copper species at different N H 3 concentrations for high reagent concentration^ M S 2 0 3 2 " , 0.05 M C u , p H 10.0, Eh=0.250 V ) (Aylmore et al, 2001a) '. 12 V l l Figure 2.12 Distribution of copper species at different NH3 concentrations for low reagent concentration^.1 M S 2 0 3 2 \ 5X10" 4 M C u , p H 10.0, Eh=0.250 V ) (Aylmore etal , 2001a) 12 Figure 2.13 Distribution of copper species at different S2O3 " concentrations for high reagent concentration(l M N H 3 , 0.05 M C u , p H 10.0, Eh=0.250 V ) (Aylmore et al, 2001a) 13 Figure 2.14 Distribution of copper species at different S20 3 2 " concentrations for low reagent concentration^. 1 M N H 3 , 5x l0" 4 M C u , p H 10.0, Eh=0.250 V ) (Aylmore et al, 2001a) 13 Figure 2.15 Distribution of copper species at different C u 2 + concentrations for high reagent concentration 1 M S 2 0 3 2 \ 1 M N H 3 , p H 10.0, Eh=0.250 V ) (Aylmore et al, 2001a) 14 Figure 2.16 Distribution o f copper species at different C u 2 + concentrations for low reagent concentration^. 1 M S 2O 3 2",0.1 M N H 3 , p H 10.0 , Eh=0.250 V ) (Aylmore et al, 2001a) 14 Figure 2.17 Effect of copper concentration and temperature on the dissolution of gold in 0.25 M S 2 0 3 2 " , 1.0 M N H 3 , 196 kPa 0 2 , stirring velocity 200 rpm (Tozawa et al, 1981) 26 Figure 2.18 Effect o f copper sulfate concentration on gold leaching rate(Li et al, 1996) 28 Figure 2.19 Effect of the concentration ratio of ammonia to thiosulfate on gold leaching rate(Lietal , 1996) 29 Figure 2.20 Effect of ammonia on gold oxidation. Experimental conditions: 0.1 M N a 2 S 2 0 3 , 30°C (Breuer et al, 2000a) 31 Figure 2.21 Kinetic plot showing the leaching of gold at fixed potential of 238 m V S H E in a solution containing either thiosulfate, thiosulfate and ammonia, or thiosulfate, ammonia and copper (Breuer et al, 2002) 34 Figure 2.22 Linear sweep voltammograms for the reduction of oxygen in ammonia-thiosulfate solutions, and for the reduction of copper in ammonia and ammonia-thiosulfate solution (Breuer et al, 2002) 35 Figure 2.23 Linear sweep voltammogram for a gold electrode in solutions containing copper ( solid line). Also shown is the calculated partial current density for the v i i i oxidation of gold to gold thiosulfate derived from the mass change (dashed line) measured using R E Q C M . Also shown as a square symbol is the measured mixed potential and reaction rate(as current density) during leaching (Breuer et al, 2002 )36 Figure 2.24 The model of electrochemical-catalytical mechanism o f ammoniacal thiosulfate leaching of gold (Jiang et al, 1993a) 38 Figure 2.25 Mechanism of gold leaching in A T S system in the presence of copper ion (Ouyang, 2001) 42 Figure 2.26 The electrochemical model for the copper catalysis mechanism of leaching gold with ammoniacal thiosulfate.( Aylmore et al, 2001a) 43 Figure 2.27 The effect of pyridine on gold leaching in A T S - C u system 45 Figure 3.1 A diagram of the R E Q C M electrode (Jeffrey et al, 2000a) 58 Figure 3.2 Assembly of cell and R E Q C M system 59 Figure 3.3 Schematic illustration of R E Q C M 60 Figure 3.4 A kinetic plot showing the deposition of silver from a solution containing 0.01 M A g ( C N ) 2 "and 0.02 M CN" . The experimental conditions were 0.1 m A , 700 rpm and 25° C 62 Figure 3.5 Validation of the Levich equation by measuring the hexacyanoferTate(IH) reduction current density as a function of (om. The experimental conditions were 0.01 M FefCN) 6 3 " , 0.1 M NaC10 4 , -250 m V versus S H E and 25°C 63 Figure 4.1 Effect o f scan rate on the gold anodic polarization in 0.2 M (NH4)2S203 solution (pH 10, 25°C, 450 rpm) 69 Figure 4.2 Effect o f scan rate on the gold anodic polarization in 0.2 M (NH4)2S203 solution (pH 10, 25°C, 450 rpm) 70 Figure 4.3 The reproducibility tests of the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution (pH 10, 25°C, 450 rpm, lmV/s ) 71 Figure 4.4 The reproducibility tests o f the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution (pH 10, 25°C, 450 rpm, lmV/s ) . Test l :Solution prepared in presence of air. Test 2: Solution prepared under nitrogen 71 ix Figure 4.5 Effect of air in preparation of solutions on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 , 250 ppm [Cu] T solution (pH 10, 25°C, 450 rpm, lmV/s ) . Test 1: Solution prepared in presence of air. Test 2: Solution prepared under nitrogen 72 Figure 4.6 Effect of adding copper (H) in solution preparation on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 , 250 ppm [Cu] T solutions (pH 10, 25°C, 450 rpm, 1 mV/s).Test 1: Adding copper (II) after adjusting p H . Test 2: Adding copper (IT) before adjusting p H 72 Figure 4.7 Effect of adding ammonia or adding ammonia ion on the gold anodic polarization at p H 10 (25°C, 450 rpm, 1 mV/s) 74 Figure 4.8 Effect of temperature on the total current density on gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution, no copper (pH 10, 450 rpm, 1 mV/s) 76 Figure 4.9 Effect o f temperature on the current density from gold mass change on gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution, no copper (pH 10, 450 rpm, 1 mV/s) 76 Figure 4.10 Effect of rotating speed on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution, no copper (25°C, p H 10, 1 mV/s) 77 Figure 4.11 Effect of rotating speed on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution, no copper (25°C, p H 10,1 mV/s) 77 Figure 4.12 Effect o f p H value on the total current density on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution, no copper (25°C, 450 rpm, 1 mV/s) . . . . 78 Figure 4.13 Effect of p H on the current density from gold mass change on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution, no copper (25°C, 450 rpm, 1 mV/s) 78 Figure 4.14 Effect o f [CU]T concentrations on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution (25°C, p H 10,450 rpm, 1 mV/s) 80 Figure 4.15 Effect of [CU]T concentrations on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution (25°C, p H 10, 450 rpm, 1 mV/s) 80 Figure 4.16. Effect of [CU]T concentrations on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution (25°C, p H 10, 450 rpm, 1 mV/s) 81 Figure 4.17 [CU]T decrease the overpotential which is required for oxidizing the gold to gold thiosulfate in 0.2 M ( N H 4 ) 2 S 2 0 3 solution ( 25°C, pHIO, 450rpm, lOmV/s) ... 82 Figure 4.18 Anodic polarization on a platinum and gold electrode in 0.2 M QSrH4)2S203 solution ( 2 5 ° C , p H 10, 450 rpm, 10 mV/s) 82 Figure 4.19 Gold anodic polarization test in 0.2 M ( N H 4 ) 2 S 2 0 3 solution ( 25°C, p H 10, 450 rpm, 1 mV/s) 84 Figure 4.20. Reproducibility test for adding 250 ppm copper (U) immediately before starting the test in 0.2 M (NH 4 )2S20 3 solution (25°C, p H 10, 450 rpm, 1 mV/s) 84 Figure 4.21 Effects of thiosulfate, ammonia, and copper on gold anodic polarization ( 25°C, p H 10, 450 rpm, 1 mV/s) 86 Figure 4.22 Effect of (NH 4 ) 2 S 2 03 concentrations on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solutions (250 ppm copper, 25°C, p H 10, 450 rpm, 1 mV/s) . . . . 86 Figure 4.23. Effect of NH3 concentrations on the gold anodic polarization in 0.2 M N a 2 S 2 0 3 solutions (250 ppm copper, 25°C, p H 10,450 rpm, 1 mV/s) 89 Figure 4.24. Effect of S2032" concentrations on the gold anodic polarization in 0.4 M NH3 solutions (250 ppm copper, 25°C, p H 10, 450 rpm, 1 mV/s) 89 Figure 4.25. Reproducibility tests of gold cathodic polarization in 0.2 M (NH 4 ) 2 S 2 03 solution (250 ppm copper, p H 10, 25°C, 450 rpm, 1 mV/s) 91 Figure 4.26. Comparing the effects of using gold electrode or platinum electrode on the cathodic polarization (0.8 M ( N H 4 ) 2 S 2 0 3 , 2 5 0 ppm copper,, p H 10, 25°C, 450 rpm, 1 mV/s) 92 Figure 4.27 Effect of adding ammonia or adding ammonia ion on the gold cathodic polarization (pH 10, 25°C, 450 rpm, 1 mV/s) 92 Figure 4.28. Effect o f ( N H 4 ) 2 S 2 0 3 concentrations on the gold cathodic polarization (copper 250 ppm, p H 10, 25°C, 450 rpm, 1 mV/s, air purge) 94 Figure 4.29. Effect o f copper concentration on the gold cathodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 (pH 10, 25°C, 450 rpm, 1 mV/s , air purge) 94 Figure 4.30. Cathodic polarization in 0.4 M N H 3 solution (pH 10, 25°C, 450 rpm, 1 mV/s , air purge or nitrogen purge) 95 Figure 4.31 Cathodic polarization in 0.4 M N H 3 , 2 5 0 ppm copper solution (pH 10, 25°C, 900 rpm, 1 mV/s , air purge). Thiosulfate was not added 96 Figure 4.32. Comparing cathodic polarization in 0.4 M N H ^ 250 ppm copper solution between air purge and nitrogen purge. N o thiosulfate present 97 x i Figure 4.33. Effect of rotating speed on the gold cathodic polarization in 0.4 M NH3, 250 ppm copper solution (pH 10, 25°C, l m V / s , air purge) 98 Figure 4.34. The relationship between the limiting current and the square root o f rotating velocity for experiments with 0.4 M N H 3 , 250 ppm copper solution (pH 10, 25°C, l m V / s , air purge) 99 Figure 4.35 Cathodic polarization in 0.4 M N H 3 , 250 ppm copper and 0.1 M S20 3 2 " solution (compared with no S20 3 2 , air purge) 100 Figure 4.36. Effect of S20 3 2 " concentrations on the gold cathodic polarization in 0.4 M N H 3 > 250 ppm copper solution (pH 10, 25°C, 450 rpm, l m V / s , air purge) 102 Figure 4.37. Effect of concentration of N H 3 on the gold cathodic polarization in 0.2 M N a 2 S 2 0 3 , 250 ppm copper (pH 10, 25°, 1 mV/s) 102 Figure 4.38. Leaching reproducibility tests in 0.2 M ( N H 4 ) 2 S 2 0 3 , 50 ppm copper solution (pH 10, 25°C, 450 rpm, 0. 25 V vs. SHE) 105 Figure 4.39 Effect o f C u ( N H 3 ) 4 2 + o n gold leaching in A T S - C u system. (pH 10, 450 rpm, 25°C, air purge) 106 Figure 4.40. Comparing leaching tests in the presence o f S 2 0 3 2 " and absence of S 2 0 3 2 " (pH 10, 450 rpm, 25°C, air purge) 107 Figure 4.41 Effect o f concentration of S 2 0 3 2 " on gold leaching in 0.4 M N H 3 , 250 ppm copper solution (pH 10, 450 rpm, 25°C, air purge) 108 Figure 4.42 Effect of concentration o f N H 3 on gold leaching in 0.2 M N a 2 S 2 0 3 , 250 ppm copper solution (pH 10, 450 rpm, 25°C, air purge) 109 Figure 4.43 Effect of concentration of S 2 0 3 2 " on the gold leaching in 0.2 M N F f 3 , 30ppm copper solution, open potential (pHIO, 450rpm, 25°C, air purge) 110 Figure 4.44. Effect of concentration of N H 3 on gold leaching in 0.1 M N a 2 S 2 0 3 solution, 30 ppm copper solution, open potential (pH 10, 450 rpm, 25°C, air purge). I l l Figure 4.45. Effect of potential on the gold leaching in 0.2 M (NFLt ) 2 S 2 0 3 solution, no copper (pH 10, 450 rpm, 25°C, air purge) , 113 Figure 4.46 Effect of potential on the gold leaching in 0.2 M ( N H 4 ) 2 S 2 0 3 solution with 30 ppm copper (pH 10,450 rpm, 25°C, air purge) 113 x i i Figure 4.47 Effect o f potential on the gold leaching in 0.2 M (NFL,)2S203 solution with 50 ppm copper (pH 10, 450 rpm, 25°C, air purge) 114 Figure 4.48 Effect o f potential on the gold leaching in 0.2 M (NH4)2S203 solution with 250 ppm copper (pH 10, 450 rpm, 25°C, air purge) 114 Figure 4.49. Effect o f concentration o f copper on gold leaching in 0.2 M (NFLt)2S203 solution with 0.20 V vs. S H E applied potential (pH 10, 450 rpm, 25°C, air purge). 116 Figure 4.50. Effect o f concentration of copper on gold leaching in 0.2 M (NH4)2S203 solution with 0.25 V vs. S H E applied potential (pH 10, 450 rpm, 25°C, air purge). 116 Figure 4.51. Effect o f concentration of copper on gold leaching in 0.2 M(NH4)2S2C>3 solution with 0.30 V vs. S H E applied potential (pH 10, 450 rpm, 25°C, air purge). 117 Figure 4.52. Effect o f temperature on gold leaching i n 0.2 M (NH4)2S203 solution, 250 ppm copper, 0. 20 V vs. S H E (pH 10, 450 rpm, 25°C, air purge) 118 Figure 4.53. Effect o f temperature on gold leaching in 0.2 M (NH4)2S203 solution, 250 ppm copper, 0.25 V vs. S H E (pH 10, 450 rpm, 25°C, air purge) 119 Figure 4.54 Effect o f temperature on gold leaching in 0.2 M (NH4)2S203 solution, 250 ppm copper, 0.30 V vs. S H E (pH 10, 450 rpm, 25°C, air purge) 119 Figure 4.55 Arrhenius plot for different potentials. . 120 Figure 4.56. Reproducibility tests for gold leaching in 0.1 M (NH4)2S203, 30 ppm [Cu]x solution (pH 10, 25°C,450 rpm, 0.25 V vs. S H E , air purge) 121 Figure 4.57. Effect o f concentration of (NH4)S203 on gold leaching in ( N H O ^ C h solution, 30 ppm copper, 0.25 V vs. S H E (pH 10,450 rpm, 25°C, air purge) 124 Figure 4.58. Effect o f concentration of [Cu]j on gold leaching in 0.1 M (NH4)2S2C>3 solution, 0.25 V vs. S H E (pH 10, 450 rpm, 25°C, air purge) 125 Figure 4.59. Effect o f concentration of NH3/S2O3 ratio on gold leaching in (NH4)2S203 solution, 30 ppm [Cu] T , 0.25V vs. S H E (pH 10, 450 rpm, 25°C, air purge) 125 Figure 4.60. Effect of p H value on gold leaching in 0.1 M (NH4)2S203 solution, 30 ppm [Cu] T , 0.25 V vs. S H E (pH 10, 450 rpm, 25°C, air purge) 126 x i i i Figure 4.61. Effect o f temperature on gold leaching in 0.1 M (NH4)2S203 solution, 30 ppm [Cu] T , 0.25 V vs. S H E (pH 10, 450 rpm, 25°C, air purge) 126 Figure 4.62 Arrhenius plot for rate of gold leaching using a lower concentration of reagents, 0.25 V vs. S H E 127 Figure 4.63. Effect of rotating speeds on gold leaching in 0.1 M (NH4)2S203 solution, 30 ppm [Cu] T , 0.25 V vs. S H E (pH 10, 450 rpm, 25°C, air purge) 127 Figure 4.64. Effect of 0.2 M A g + on the gold anodic polarization in 0.2 M (NH 4 ) 2 S 2 03 solution, no copper (pH 10, 450 rpm, 25°C, 1 mV/s) 131 Figure 4.65. Effect of 500 ppm N a C l on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution, no copper, (pH 10, 450 rpm, 25°C, 10 mV/s) 131 Figure 4.66. Effect of 1% N a C l on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution, no [Cu] T , (pH 10, 450 rpm, 25°C, 10 mV/s) 132 Figure 4.67 Effect of concentration of E D T A on the gold anodic polarization in 0.1 M (NTl4)2S203 solution, 250 ppm copper (pHIO, 450rpm, 25°C, lmV/s ) 133 Figure 4.68. Effect o f E D T A on gold anodic polarization in a solution of 0.2 M ( N H 4 ) 2 S 2 0 3 , 250 ppm C u 2 + (pH 10, 450 rpm, 25°C, 1 mV/s) 133 Figure 4.69. Gold anodic polarization in a solution of 0.1 M Na 2S 203, 100 ppm copper, 0.01 M E D T A (pH 10, 450 rpm, 25°C, 1 mV/s) 134 Figure 4.70 Reproducibility tests of anodic polarization of gold in 0.1 M (NH 4 ) 2 S 2 03 solution, 250 ppm copper, 0.005M E D T A (pH 10, 450 rpm, 25°C, 1 mV/s) 134 Figure 4.71. Effect o f E D T A i n leaching test i n 0.1 M Na2S2C>3,100 ppm copper (pH 10, 450 rpm, 25°C) 135 Figure 4.72 Effect of E D T A in leaching test in 0 .2M ( N H 4 ) 2 S 2 0 3 , 250 ppm copper , 0.20 V vs. SHE(pH10, 450rpm, 25°C,air purge) 136 Figure 4.73. Effect o f E D T A in test in 0.2 M ( N H 4 ) 2 S 2 0 3 , 250 ppm copper, 0.25 V vs. S H E (pH 10, 450 rpm, 25°C, air purge) 137 Figure 4.74. Effect of additives on gold leaching in 0.2 M (NH 4 ) 2 S 2 03, no copper (except where added), 0.25 V vs. S H E (pH 10, 450 rpm, 25°C, air purge), 137 Figure 4.75 Effect of concentration of Ag+ on gold leaching in 0.2 M ( N H 4 ) 2 S 2 0 3 , no copper, 0.25 V vs. S H E (pH 10, 450 rpm, 25°C, air purge) 138 xiv Figure 4.76. Effect of additives on gold leaching in 0.1 M (NH4) 2 S 2 03, 30 ppm copper, 0.25 V vs. S H E (pH 10, 450 rpm, 25°C, air purge) 139 Figure 4.77 The model A of electrochemical mechanism of gold leaching in A T S - C u system 144 Figure 4.78 The model B o f electrochemical mechanism o f gold leaching in A T S - C u system 145 xv ACKNOWLEDGEMENTS I. would like to express my sincere appreciation to Dr. David Dreisinger for his supervision, and reviewing and editing this thesis. I am very grateful to Dr. Paul West-Sells, my co-supervisor, for his constructive discussions, and reviewing and editing this thesis. I would also like to acknowledge Dr. Des Tromans, Dr. Akram Alfantazi and Dr. David Dixon for providing constructive ideas. Dr. Jianming Lu ' s kind help, especially in debugging R E Q C M system, is very much appreciated. Thanks to M s . Anita Lam and Dr. Berend Wassink for their contribution to my study. Thanks are extended to my fellow graduate students and the staff o f the Hydrometallurgy Group with whom I have enjoyed working. Thanks to the sponsoring companies Anglogold, Barrick Gold, Newcrest, Newmont, Placer Dome, and Teck Cominco for providing financial and intellectual support for this work. Finally, I would like to thank my wife, my parents, my brother and sister for their encouragement and support. x v i NOMENCLATURE A area (cm 2) D diffusion coefficient (cm 2 s"1) E potential of the electrode (V) E° standard potential (V) F Faraday constant, 96487 A s mol" 1 i current density (A cm"2) J flux ( m o l m" 2 s"1) M molarity (mol dm"3) m mass(g) n number of electrons transferred p H negative logarithm to base of the activity o f hydrogen ion ppm part per mil l ion R gas constant ( 8.314 J K" 1 mol" 1) r leaching rate ( m o l m" 2 s"1) Re Reynolds number S C E standard calomel electrode S H E standard hydrogen electrode T temperature (°C) t time(s) v kinematic viscosity ( c m 2 s"1) co rotating velocity (revolutions per minutes ) x v i i 1 INTRODUTION Cyanide has been used as a lixiviant for extracting gold for over one hundred years. Conventional cyanidation is an economical, biodegradable process and it achieves excellent recoveries from a wide range of ores. However, since cyanide is a very toxic chemical, and because cyanide solution does not effectively leach carbonaceous or complex ores, researchers have been interested in finding non-toxic chemicals and environmentally safe substitutes. One area of current research is gold leaching in thiosulfate solution. Thiosulfate has the ability to complex gold. Thiosulfate leaching can be considered a non-toxic process, the gold dissolution rates can be faster than cyanidation and, because the gold thiosulfate complex does not adsorb on carbonaceous materials and the interference o f foreign cations decreases in thiosulfate leaching, high gold recoveries can be obtained from the thiosulfate leaching of complex and carbonaceous ores. In addition, comparing reagent unit costs, ammonium thiosulfate is far cheaper than sodium cyanide(US$0.13/kg vs. US$1.807kg). Consequently, with similar or even slightly higher lixiviant consumption, the application o f thiosulfate for gold recovery can be economical and compete directly with cyanidation (Molleman et al, 2002). However, it was found that in the absence o f ammonia, gold dissolution in thiosulfate solution stops due to gold surface passivation; in the absence o f the redox couple C u 2 + / C u + as a catalyst, the leaching rate is very slow. Therefore, ammonia and the 1 C u / C u couple are needed in the leaching system. In addition to the oxidative decomposition reactions of thiosulfate, many degradation species make the thiosulfate leaching system (ATS-Cu) very complicated. In recent years many researchers focused on the thiosulfate leaching o f gold, especially on complex ores. However, both fundamental kinetic and electrochemical studies are still limited, and consequently, the reaction kinetics and mechanisms are not fully understood. So far this process has not been widely employed in the gold industry. Clearly, an understanding of all o f the major variables and how they affect the leaching process is necessary before achieving success on a commercial scale. The rotating electrochemical quartz crystal microbanlance ( R E Q C M ) is a powerful technique capable of detecting very small mass changes at the electrode surface that accompany electrochemical processes (Jeffery et al 2000a). Use o f the R E Q C M for studying gold leaching in A T S - C u system can minimize the interference of side reactions on the gold leaching kinetics. The objective of this work was to investigate the effect of different parameters (concentrations of different reagents, potential, p H , temperature, rotating velocity, etc.) on gold leaching in the A T S - C u system using the R E Q C M , and to try to gain an understanding of the reaction mechanism. Also , the effect of selected additives on the dissolution of gold in the thiosulfate system was studied. A n anodic polarization study, 2 cathodic polarization study and leaching test study were included in this work. In each case, the R E Q C M was the key apparatus for completing the study. This thesis consists o f five chapters. Following this introduction, chapter two gives a review of existing literature on gold leaching in the A T S - C u system, especially on the kinetics and mechanistic studies. The experimental methods are introduced and described in chapter three, followed by the results and discussion in chapter four. The conclusions and recommendations are presented in chapter five. 3 2 LITERATURE REVIEW 2.1 Aqueous Chemistry T h e chemistry o f the a m m o n i a thiosulfate - copper ( A T S - C u ) system is c o m p l i c a t e d due to the simultaneous process o f c o m p l e x i n g l igands such as a m m o n i a and thiosulfate, the Cu(Il)-Cu(r) redox couple and the p o s s i b i l i t y o f oxidat ive d e c o m p o s i t i o n reactions o f thiosulfate i n v o l v i n g the format ion o f tetrathionate and other addit ional sulfur compounds . Gold-thiosulfate-ammonia-water system A n Eh-pH diagram can be used to show the predominant species under different potentials and p H condit ions. A y l m o r e et a l (2001a) constructed the E n - p h d iagram for gold-thiosul fate-ammonia - water system at h i g h reagent concentrations (Figure 2.1) and at l o w reagent concentrations (Figure 2.2). Those figures show that under certain condit ions g o l d c o u l d be present i n so lut ion as A u ( N H 3 ) 2 + rather than Au(S203)23". T h e gold(I) thiosulfate c o m p l e x is the most stable species i n the leaching system up to p H 8.5. A b o v e this p H , w h e n N r L ; + converts to N H 3 , the predominant g o l d c o m p o u n d is g o l d (I) d i a m m i n e c o m p l e x . 4 Eh (Volts) 2.0 1.5 1.0 03 0.0 -O.S •1.0 •1.5 •2.0 1 1 1 r 1 , 1 i i Au(NH3)4t3(aq) Au(Sa03)2-3(aq) ! {Au(NH3),+(aq)} -Au 1 1 1 1 1 1 1 _. i 10 12 14 pH Figure 2. 1 Au-NH3-S2032" system (conditions: SxlO^M Au; 1M S2032"; 1M NH 3/NH 4 +, 0.05M Cu2+)(Aylmore et al, 2001a) Eh (Volte) 2.0 r 1 1 i . , , , , ^ 1.5 Au(NH,)4tS(aq) A u 0 » 1.0 0.5 Au(S20,)2^(aq) j {Au(NH,y(aq» 0.0 • •0.5 - --1.0 Au •IS -2.0 1 i ' ' 1 i i ' ' 4 6 8 10 12 14 Figure 2. 2 Au-NH3-S2032" system (conditions: SxlO^M Au; 0.1M S2032"; 0.1M NH 3/NH 4 +, 0.05M Cu2+) (Aylmore et al, 2001a) 5 Other researchers also constructed Eh-pH diagrams similar to Figures 2.1, 2.2 (L i et al. 1996, Molleman et al, 2002). However, as L i et al (1996) remarked, it is generally accepted that the gold (I)-thiosulfate species is the more stable species at p H 10 and this was confirmed by rest potential measurements, since the gold rest potential varied with thiosulfate concentration instead of ammonia concentration (Figure 2.3) -o.io W U -0.15 t/3 vi > .15 | -0.20 o a. 3 1 -0.25 -0.30 -1.6 -1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 LogtWKM) Figure 2. 3 Effect of thiosulfate concentration on the rest potential of gold(Li et al,1996). Aylmore et al (2001a) explained that the variation between thermodynamics and electrochemical studies may be attributed to the high activation energy for A u ( N H 3 ) 2 + formation (Meng and Han, 1993). Furthermore, i f a lower stability constant for Au(NH 3 )2 + (10 1 3 otherwise 10 2 6) is used to calculate the diagram, a much higher 6 concentration o f ammonia is required to stabilize Au(NH3)2 + and Au(S203)23" exists over the whole p H range under conditions examined in Figures 2.1 and 2.2. Molleman et al (2002) constructed the Eh-pH diagram for gold-thiosulfate-ammonia -water system. The gold concentration was 2.5xl0" 5 M (5ppm), the thiosulfate concentration was 0.2 M and the ammonia concentration was 0.4 M . Two different values of free energy of formation of the thiosulfate species were used in Molleman's work. When a value of -532.2 kJ/mol was used, the Eh-pH diagram is similar to Alymore's work. However, when a value of -518.8 kJ/mol was used, a totally different Eh-pH diagram was developed as shown in Figure 2.4. 1.5 PLOT LfiBELS T . - » • 2 9 8 . I S K I flu I . 2 . S E - 0 S IS203I • 0 . 2 INH31 - a.14 STH8LE flHEflS B Bu S M Ro 03 <2• > (HOI C H2 flu 03 <•> (POl 0 Hu (0 H 13 E Bu [52 D3 >2 <3-> Ifffll H20 STABILITY LIMITS 1 Q2/H20 2 H2/M20 Figure 2. 4 Eh-pH diagram of the gold-thiosulfate-ammonia-water system at 25°C. The activities of the species are 2.5 xlO"5 M Au, 0.2 S2032" and 0.4 M NH 3. AGf0(S2O32") = -518.8 kJ/mol (Molleman et al, 2002) 7 It can be seen that the gold(I)-thiosulfate complex is stable in the whole p H range shown in Figure 2.3. This diagram is consistent with the results of rest potential measurements. Copper-thiosulfate-ammonia-water system Aylmore et al (2001a) constructed an Eh-pH diagram for the copper-thiosulfate-ammonia-water system in high concentration and low concentration thiosulfate solutions as shown in Figure 2.5 and Figure 2.6. It can be seen that decreasing the ammonia thiosulfate and copper concentrations significantly narrows the region of stability of Cu(NH3)4 2 + and Cu(S203)35" and expands the stability region of CuO, CU2O and CU2S. It is clear that high p H values should be avoided because copper w i l l precipitate from solution as copper oxides. Eh (Volts) Cu — 1 1 1 *<aq) CuO —1 1 1 Cu(NH3)4+2(aq) 1 1 1 CuO -Cu(S 20 3 )3" —ZZiutir^^ Cu2G : Cu 2S -1 1 L _ _ Cu _ J • ' J 1 1 4 « 8 10 12 14 pH Figure 2. 5 Cu-NH3-S2032" system (conditions: 5x10"4 M Au; 1 M S2032"; 1 M NH 3/NH 4 +, 0.05M Cu)(Aylmore et al, 2001a). 8 Eh (VolU) 2.0 r Cu(Nk)4*(aq) Cu*a(ab) CuO 1.0 r CuO -1.0 Cu •2.0 4 6 8 10 12 14 p H Figure 2. 6 Cu-NH3-S2032" system (conditions: 5X10" 4 M Au; 0.1M S2032"; 0.1M NH 3/NH 4 +, 0.05M Cu2+)(Aylmore et al, 2001a). Speciation diagrams can characterise the distribution of various C u - N H 3 - S 2 0 3 2 " species co-existing in solution. Aylmore et al (2001) presented the speciation diagrams for varying concentrations of Cu(IT), S 2 0 3 2 " and N H 3 , and E h and p H as shown in Figure 2.7-2.16. These speciation diagrams are useful for understanding and optimizing this system. Figure 2.7 and Figure 2.8 show that increasing the potential (e.g., with 0 2 ) but keeping all reagent concentrations constant results in a decrease in C u ( S 2 0 3 ) 3 5 " and an increase in C u ( N H 3 )4 2 + in the solution. A t low reagents concentrations in solution, only a small E h range is available for maintaining the copper (IT) ammonia complex in solution. 9 ) 5.00E-02 0 0.1 0.2 0.3 0.4 0.5 0.6 Eh(V) Figure 2. 7 Distribution of copper species at different Eh conditions for high reagent concentration (1 M S2032", 1 M NH3, 0.05 M Cu , pH=10.0) (Aylmore et al, 2001a) 5.00E-O4 4.00E-O4 S e o I 3.00E-O4 <5 2.00E-O4 1.00E-04 0.O0E+O0 Figure 2. 8 Distribution of copper species at different Eh conditions for low reagent concentration (0.1 M S2032-, 0.1 M NH 3, 5x10^  M Cu , pH=10.0) (Aylmore et al, 2001a). 1 0 Figure 2. 9 Distribution of copper species at different pH concentrations for high reagent concentration (1 M S2032~, 1 M NH 3, 0.05 M Cu , Eh=0.250 V) (Aylmore et al, 2001a). Figure 2. 10 Distribution of copper species at different pH concentrations for low reagent concentration (0.1 M S2O32~,0.1 M NH 3, 5x10"* M Cu , Eh=0.250 V) (Aylmore et al, 2001a). 11 Figure 2.11 Distribution of copper species at different NH3 concentrations for high reagent concentration^ M S2032", 0.05 M Cu , pH 10.0, Eh=0.250 V) (Aylmore et al, 2001a). 6.0OE-O4 Total NH, cone (M) Figure 2. 12 Distribution of copper species at different NH 3 concentrations for low reagent concentration(0.1 M S2032", 5X10" 4 M Cu , pH 10.0, Eh=0.250 V) (Aylmore et al, 2001a) 12 Figure 2. 13 Distribution of copper species at different S2O3 2 " concentrations for high reagent concentration^ M NH3, 0.05 M Cu , pH 10.0, Eh=0.250 V) (Aylmore et al, 2001a). Figure 2.14 Distribution of copper species at different S2O3 2 " concentrations for low reagent concentration .^ 1 M NH3, 5x10"" M Cu , pH 10.0, Eh=0.250 V) (Aylmore et al, 2001a). 13 0.4 0.6 Cu" Cone (M) as Figure 2.15 Distribution of copper species at different Cu 2 + concentrations for high reagent concentration^ M S2032~,1 M NH 3, pH 10.0, Eb=0.250 V) (Aylmore et al, 2001a) Tanorita X )( )( X )( X )< X )( )( )( )< )<i )(. )( >E C^l(82OA,• I I I I I I I I I I I I I I I CufNHdT 3 0.002 0.0O4 0.006 0.008 0.01 Cu'*Conc(M) Figure 2.16 Distribution of copper species at different Cu concentrations for low reagent concentration(0.1 M S2O32",0.1 M NH3, pH 10.0 , Eh=0.250 V) (Aylmore et al, 2001a). 14 From Figures 2.9 and 2.10, it can be seen that to leach gold under low reagent conditions only a narrow p H region around 9.5-10 exists where the copper ammonia complex is stable without the precipitation of copper (II) oxide, tenorite. A t high thiosulfate and ammonia concentrations a broader p H range is available. A broader p H range would also be available at very low concentrations of copper. Figures 2.11, 2.12, 2.13 and 2.14 show that increasing total ammonia concentration but keeping the other reagent concentrations, p H and E h constant, increases the stability region o f the Cu(NH3 )4 2 + complex, whereas increasing the thiosulfate concentration increases the stability region of the Cu(S203)35" complex. It would appear that a higher ammonia to thiosulfate concentration would be required to achieve a higher C u ( N H 3 ) 4 2 + concentration in solution over Cu(S203)35". However, a high C u ( N H 3 ) 4 2 + concentration in solution may also result in higher losses of thiosulfate as Cu(NH3)4 2 + can oxidize thiosulfate to tetrathionate. Figures 2.15 and 2.16 show that, under the conditions used, the copper concentration has a much more pronounced effect on the stability regions o f copper species where Cu(S203)35", becomes more stable over the C u ( N H 3 )4 2 + complex. In addition, precipitation of tenorite occurs with increased copper concentration in solution. After thermodynamic analysis and the construction of Eh-pH and speciation diagrams, Aylmore et al (2001a) suggested that the optimum E h and p H lies between 0.25 and 0.30V and from 9 to 10, respectively. A t low reagent concentrations, there is a smaller 15 window of E h and p H in which copper (I) and copper(U) species exists. Excess NH3 over S2O32" favours the formation o f copper (II) amine species. 2.2 Stability of thiosulfate solution Thiosulfate ions are metastable and tend to undergo chemical decomposition in aqueous solutions. The degradation of thiosulfate not only results in loss of the lixiviant, but also can lead to the formation of sulphides which in turn may passivate gold and limit the leaching rate. How to maximize control of the stability of the leaching solution is a key factor for the success of thiosulfate leaching. When the p H of a thiosulfate solution is 5 or less, the following reaction occurs at an appreciable rate (Skoog and West, 1975): S 2 0 3 2 ' + H + o H S 2 0 3 " HSO3" + S(s) Equat ion 2.1 The rate of this reaction depends on the p H , i f in strong acidic solution, elemental sulfur forms within a few seconds. This reaction explains why gold thiosulfate leaching proceeds only in alkaline solution. In systems that contain bacteria, experiments indicate that the stability of thiosulfate solution is at a maximum in the p H range between 9 and 10, because bacterial activity appears to be at a minimum at this p H range, and solutions that are free of bacteria have limited thiosulfate degradation (Skoog and West, 1975). 16 It was reported that decomposition of thiosulfate solution is catalyzed by copper (II) ions as well as by the decomposition products themselves, and the decomposition rate is greater in more dilute solutions (Skoog and West, 1975). Wan et al (1997) also reported that dilute solutions of thiosulfate (0.01M or less) decompose more rapidly than concentrated solutions (0.1M or higher). Thiosulfate may also be oxidized to form tetrathionate under milder conditions in the presence of cupric ion (Wasserlauf et al 1982): 2S 2 0 3 2 +2Cu 2 + -» S4062"+2Cu+ (fast) Equation 2.2 2Cu++l/2Q2+2H+-> 2Cu2++H2Q (measurable) Equation 2.3 2S2032+2H+ +l/202-> S 40 6 2+H 20 Equation 2.4 A i r oxidation of thiosulfate under normal pressures and temperatures is a very slow process. Rola et al (1982) reported that a solution o f thiosulfate and polythionates, at p H 7, was aerated for 4 months under sterile laboratory conditions at Noranda research centre with less than 10% change observed in the thio salt concentration. Pryor (1960) showed that S 2 0 3 2 " is relatively stable in basic solution, especially in the absence o f catalysts. The base hydrolysis was found to be very slow in basic solution, even at 250°C. Kerley et al (1981, 1983) suggested the addition of sulfite ions to thiosulfate solutions to stabilize thiosulfate ions. Gong et al (1990) proposed this reaction as: 17 4S032"+2S2 + 3HzO o 3S203 2 +60H" Equation 2.5 Gong and H u (1993) described the substitution of sulfate for sulfite because SO42" ion inhibits the oxidation and decomposition of S2032" ion: S O 4 2 +S2 +H20->S2032 +20H Equation 2.6 However, this reaction is in question, because the SO42" species is very stable and therefore unlikely to react with sulfide ion. Copper(IT) and thiosulfate react homogeneously in the presence o f ammonia by the following equation: 2Cu(NH3)42++8S2032^2Cu(S203)35+8NH3 + S4062- Equation 2.7 Breuer et al (2000b) studied the oxidation of thiosulphate by copper(II). It was found that copper(II) can strongly catalyze the oxidation of thiosulfate, about 50% of the copper(IT) reacted with thiosulfate after 3.8 hrs. It can be concluded that degradation o f thiosulfate is very complex and that many reactions between original and generated species in solution are possible. The exact degradation pathways are not clear. 18 2.3 Leaching of gold in thiosulfate system Laboratory studies Work has been conducted on many types of raw materials: Oxide ores (Langhans et al., 1992; Ji et al, 1991,Cai 1997 ), sulfide concentrates (Zhang et al.,1987; Cao et al,1992), manganese ores (Zippering, 1998), carbonaceous ores (Hammeti et al, 1989; Wan et al, 1994), dissolution of pure gold (Tozawa et al, 1981). Most o f these experiments were performed with finely ground materials. Gold extraction and consumption of reagents are key factors to focus on in order to compete with cyanide leaching. In recent years, low thiosulfate concentration and low temperature have been widely used in tests for minimizing reagent losses. The effects of reagent concentration, such as thiosulfate, copper (If), sulfite and ammonia on the gold and silver extraction from gold concentrate, were studied by Zhang and L i (1987). It was shown that maintaining proper [S 2 0 3 2 "] / [Cu 2 + ] and [S 203 2"]/[S0 3 2"] is very critical. The gold leaching kinetics were fitted with the shrinking core model; the activation energy was calculated to be less than 5 kcal/mol. Hemmati et al (1989) investigated the thiosulfate leaching of gold from carbonaceous ore that contained 2.5% organic carbon. The optimum conditions for gold extraction were found to be 35 °C and 103 K P a oxygen over pressure, p H 10.5, 3 M N H 3 with a lixiviant containing 0.71 M (NH 4)2S 20 3 , 0.15 M C u S 0 4 and 0.1 M ( N H 4 ) 2 S 0 3 . Extraction of gold 19 was 73% in 4 hours compared with only 10% observed using cyanide over 24 hours. Hemmati et al concluded that the efficiency of thiosulfate leaching depends upon ore type. For treating carbonaceous ores, thiosulfate was found to be chemically superior and economically advantageous over cyanide. Langhans et al (1992) investigated copper-catalyzed thiosulfate leaching for extracting gold from low-grade oxidized gold ores. 83% gold extraction was achieved with 0.4 kg/t ore S 2 0 3 2 ~ consumption. These results may be competitive with conventional cyanidation, which showed 85% gold extraction and 0.21 kg C N " consumed/t ore. Cai (1997) studied the effect of different parameters on gold extraction using ammonium thiosulfate from oxide ores. It was found that only 9% of the gold was extracted in the absence of ammonia. Gold extraction increased dramatically with an increase o f ammonia from 0-2 mol /L ( N H 3 / S 2 0 3 2 + from 0 to 7), beyond which gold extraction was not increased significantly. When the ammonia concentration was over 4 mol/L, a lower extraction was obtained. Cai (1997) suggested that addition o f ammonia may lead to dissolve some sulfide precipitates, such as CuS, C u 2 S and FeCuS 2 . These sulfides are easily formed due to the nature of the ores and the decomposition of thiosulfate. The formation of some precipitates such as ( N H 4 ) 5 C u ( S 2 0 3 ) 3 , which could cover the ore's surface and hinder the attachment of thiosulfate to the ore, was suggested to explain the detrimental effect of ammonia over 4 mol/L. Cai (1997) also found that the introduction of ammonium sulfate does increase the gold extraction rate. It was also found gold extraction was increased when increasing p H from 7.5 to 10.5, then levelled off. 20 Zippering and Raghavan (1998) identified the parameters of importance in the dissolution of gold and silver values from a rhyolite ore with a high manganese content using ammonium thiosulfate solution containing copper. The effects o f thiosulfate, ammonia concentration, temperature, and copper sulfate addition were studied. Optimum conditions were established at 2 M S 2 0 3 2 " , 4.1 M N H 3 , 6 g/L C u 2 + , 50 °C, and 2 hours leaching in the absence o f oxygen. In the absence o f cupric ions only 14% of the gold was extracted. Increasing cupric ion concentration enhanced the initial rate of gold extraction, but the ultimate extraction was not influenced by the cupric ion concentration in the range o f studies (up to 6 g/l Cu). Furthermore, it was concluded that maintaining optimal p H and E h conditions (pH 10 and 200 m V ) were necessary to prevent precipitation of copper as CU2S. Ha l f o f the thiosulfate in the lixiviant solution at p H 9.5-10 was consumed during the dissolution process. Nararro et al (2002) studied gold leaching from a flotation concentrate using ammonium thiosulfate as leaching agent. The gold content i n the concentrate was 95 g/t, whereas the main mineralogical species were chalcopyrite, pyrite, pyrrhotite, tennantite and sphalerite. The highest level of gold dissolution (94%) was obtained at 0 .05M Cu(U), 0.3 M S2O32", p H 10 and 10% pulp density after 15 hours, whereas cyanidation gave the same yield but required about 46 hours of reaction. It was reported that gold is extracted without addition o f Cu(H) ions. The reason for this behaviour is the presence of copper (chalcopyrite) in the concentrate which partly dissolves as C u ( N H 3 )4 . It was found that the recovery o f gold is enhanced by increase 21 of the thiosulfate concentration (up to 0.3M) but that a negligible effect on gold dissolution is obtained at higher thiosulfate concentration. Navarro et al (2002) observed that gold dissolution is enhanced using p H values of at least 9.5. In fact, the recovery o f gold did not exceed 10% at p H of 9.0. Comparison of gold dissolution using thiosulfate or cyanide leaching was carried out. Reasonable conditions were established as, cyanidation: 8.7 kg CNVt, air bubbling, 40% pulp density and p H 11.0 (adjusted with Ca(OH) 2 ) ; thiosulfate: 117kg S 2 0 3 2 7 t , 4.7kg Cu/t, pHIO (adjusted with N H 4 O H solution) and 40% pulp density. Therefore, Nararro et al (2002) came to the conclusion that under reasonable conditions, thiosulfate leaching compares very favourably with cyanidation. However, reagent consumption data (especially thiosulfate) and evaluation (scaling-up) o f the ammoniacal thiosulfate leaching process are still under consideration. Aylmore and M u i r (2000b) have provided a summary review on thiosulfate leaching of ores (Table 2.1) 22 Table 2. 1 A summary of various thiosulfate leaching conditions (Aylmore and Muir, 2000b). Ore type S 2 0 3 2 (M) N H 3 (M) Cu 2 + (M) t °C PH A* % B* Tozawa et al 1981 Gold plate 0.5 1 0.04 65 Kerley,1981,1983 Sulfide 2%Mn 1.2 1.1 4g/L 7-9 95 4Kg/t Block-Bolten and Tama 1985,1986 Zn-Pb sulfide flotation 0.125-0.5 0.75 21-50 7-9 90 45 lb/t Zipperian et al,1988 Rhyolite ore 7g/Kg MnO 2 2 4 0.1 35 10 90 50% Hemmati et al,1989 Carbonaceous 2.5% org C 0.71 3 0.15 50 10.5 73 15-19% Hu and Gong, 1991 0.048% Mn02,3.19%Cu 1 2 0.016 30-65 95.6 Muthy,1991 Pb-Zn sulfide 0.125-0.5 1 40 6.9-8.5 95 Cao etai, 1992 Sulfide cone. %Cu 0.2-0.3 2-4 0.047 21-70 10-10.5 95 4.8Kg/t Lanhan etal, 1992 Oxidised ore 0.02% Cu 0.2 0.09 0.001 Room Temp. 11 90 2Kg/t Wanet al,1994 Carbonaceous 1.4%C,1.0%S 0.1-0.2 0.1 60ppm Room Temp. 9.2-10 Abbruzzese et al,1995 Gold ore 2 4 0.1 25 8.5-10.5 80 Marchbank et al,1996 Pressure oxidized sulfide and carbonaceous ore 0.02-0.1 2000ppm 500ppm 55 7-8.7 70-85 Yan etal, 1996 Gold-copper ore 0.4 0.2 0.03 Room Temp. 11. 90 Wan and Brierley 1997 Carbonaceous sulfite 0.1 0.1 0.005 Room Temp. 9 50.7-65.7 Yan et al, 1998,1999 Gold copper ores ~0.36%Cu 0.5 6 0.1 Room Temp. 10 95-9.7 30Kg/t Thomas et all998 Pressure oxidized sulfide ore 0.03-0.05 -500-lOOOppm 10-lOOppm 45-55 7.5-7.7 80-85 A * : Gold extraction B * : Consumption o f thiosulfate It can be seen that it is not easy to specify an optimum condition for thiosulfate leaching. The leach conditions depend on the nature of the gold ores being tested. 23 Commercial effects Kerley et al (1981, 1983) patented a process for the recovery of precious metals from refractory ores, particularly those containing manganese and/or copper, using ammonium thiosulfate leach solution. In this patent, Kerley claimed that the sulphite ions inhibit the decomposition of thiosulfate and thus prevent the precipitation of metal sulphides. The p H recommended was at least 7.5. Perez and Galaviz (1987) modified Kerley's process to inhibit F e 3 + oxidation o f thiosulfate by adjusting the minimum level of p H to 9.5 from 7.5. Trial runs at pilot scale were carried out at L a Colorado, in the state of Sonora, Mexico. High gold recoveries were obtained, but there was no report about reagent consumption. Newmont Gold Company found that ammonium thiosulfate is effective for leaching gold from bio-oxidized "preg-robbing" ore. This process has been successfully practiced both in laboratory and pilot plant scale studies. Obtained gold recoveries were 50%-65% and ammonium thiosulfate consumption was 5.2-8.9 kg/metric tonne. Typical leach solution was p H 9.2-10.0, 0 .1M S 2 0 3 2 " , 0 .1M N H 3 and 30 ppm C u 2 + ( W a n et al, 1977). Barrick Gold Corporation patented a combined pressure oxidation, thiosulfate and resin-in-pulp process for treatment of refractory gold ores (Thomas et al,1998). In this process ore is ground to 65-85% passing 200 mesh and thickened to about 40-50% solids. Sodium carbonate is added to ensure that pressure oxidation is carried out under alkaline conditions and CI" is added to improve the kinetics and to facilitate oxidation. The ore 24 which is pressure oxidised, leaves the autoclave at about 35% solids and is directed to a leaching operation where it is contacted with ammonium thiosulfate (5 g/L) and copper sulfate (25 ppm Cu). Placer Dome Technical Services Limited have applied for a patent on a method for thiosulfate leaching of precious metal-containing materials (Ji et al, 2001). In this process, an oxygen partial pressure from about 4 to about 500 psia; A p H less than 9; A n ammonia concentration less than 0.05M, a copper concentration between 0 to 15ppm; A sulfite concentration no more than 0.01M; A n d a temperature from about 20 to about 80°C is used. A high level of precious metal extraction can be achieved - 70% and some times at least about 80% under the above conditions. 2.4 Research on the kinetics and mechanism 2.4.1 Passivation phenomena Tozawa et al (1981) demonstrated the formation of a copper sulfide coating on gold preventing it from leaching at a temperature range of 65-100 °C in an ammoniacal thiosulfate solution containing copper ions (Figure 2.17). It was suggested that cupric sulfide formed by the reaction between cupric ions and thiosulfate ions and/or the oxidation of cuprous thiosulfate complex ions led to passivation: C u 2 + + S 2 0 3 2 + H 2 0 = C u S + S O 4 2 " + 2 H + Equat ion 2.8 25 Cu(S203)23"+ l/202 + H zO = 2CuS + 2S306 2 + 20H Equation 2.9 - J — i — i — i .. i — i — i - — i — i — i — i i i i i i i_ 20 30 40 SO 60 70 80 30 100 110 120 130 140 ISO 160 170 180 T « o i p * r o t u r « (*C) Figure 2. 17 Effect of copper concentration and temperature on the dissolution of gold in 0.25 M S2032", 1.0 M NH3,196 kPa 0 2, stirring velocity 200 rpm (Tozawa et al,1981). Ter-Arakelyan et al (1984) found that the use of sodium thiosulfate as a ligand to dissolve gold was hampered by its oxidation during the leaching process with the formation of an insulating layer on the gold surface generally consisting o f colloidal sulfur. Bagdasaryan et al (1983) and Pedraza et al (1988) observed a sulfur layer as wel l as copper sulfide in the thiosulfate-copper sulfate system. It was suggested that both elemental and sulfide sulfur can be produced by the decomposition o f thiosulfate in alkaline solution (Aylmore et al, 2000b). 26 Chen et al (1996) studied gold passivation by electrochemical impedance spectra (EIS) of gold in 1 mol /L sodium thiosulfate. It was suggested that elemental sulfur may be formed to prevent thiosulfate diffusion to the gold surface and hence, to inhibit gold dissolution. The elemental sulfur coating is formed either by adsorption of elemental sulfur or by the oxidation of sulfide ions on the gold surface: Au°+S°=Au I S° Equation 2.10 or Au°+S2"=Au|s°+2e" Equation 2.11 Breuer et al (2000b) mentioned that the gold oxidation process is obviously hindered in thiosulfate solution without ammonia and copper. Also some evidence showed that the products of either the thiosulfate/copper(IT) or subsequent reactions bring about passivation of the gold surface . 2.4.2 Kinetics study of gold thiosulfate leaching Tafel current technique and leaching tests have been applied to study the kinetics o f gold leaching with thiosulfate by Jiang et al (1993b). The exchange current density ( i c o r r ) was utilized to identify the dissolution of gold. This investigation verified the catalytic effect of copper ions and ammonia, it investigated the effect of thiosulfate concentration, p H , and C u ( N H 3 ) 4 2 + on icon. It was found that the effect of C u ( N H 3 ) 4 2 + was the most notable. A s C u ( N H 3 ) 4 2 + was increased from 0.001 to 0.1 M , icon increased from 5.62 to 573.82 | j ,A/cm 2 . The activation energy for gold dissolution was reported to be 27.99 kJ/mol in 27 the absence of cupric ions and ammonia, and decreased to 15.54 kJ/mol in the thiosulfate solution containing 0.01 M C u ( N H 3 ) 4 2 + and 0.5 M N H 3 . 0.00 0.01 0.02 0.03 0.04 0.05 [CuSOJ (M) Figure 2. 18 Effect of copper sulfate concentration on gold leaching rate(Li et al,1996) L i et al (1996) found that at low copper levels, an increase in the copper concentration results in a dramatic rise in the gold leaching rate, but, too high a copper concentration significantly inhibits gold leaching (Figure 2.18). L i et al (1996) pointed out that to enable the regeneration of the cupric ion, it is necessary to keep the concentration ratio of ammonia to thiosulfate in a certain range. Increasing the concentration of only one o f the ligands w i l l have a limited positive effect on the gold leaching process. However this may have a negative effect, as an excess of either ammonia or thiosulfate may make the regeneration of the cupric ion less attractive . This effect is illustrated in Figure 2.19. A t lower ammonia/ammonium thiosulfate 28 concentrations, the gold leaching rate is reduced. There is also a good deal o f scatter in the results, possibly indicating some passivation type process slowing the leach rate. 0.10 r a • 8 | 0.08 -1 0.06 -u a hing 0.04 -o o • o 0.02 -1 0 . 0 0 * — 1 1 1 — — — * 0.01 0.1 1 10 100 [NH.OHMCra^SA] Figure 2. 19 Effect of the concentration ratio of ammonia to thiosulfate on gold leaching ratefLi et al, 1996) Ouyang (2001) investigated the effects of p H , temperature, rotating velocity and the concentrations o f different reagents on the gold leaching. The R D E was used. For the anodic process, it was found that the rotating velocity had no effect on current density. Using the Levich equation, a limiting current density was calculated that was far higher than the actual current density, so, it was suggested that the anodic process is under surface reaction control. For the cathodic process, it was found that there exists a linear relationship between the limiting current and the square root o f the rotating angular velocity, which is in agreement with the Levich equation. So, it is suggested that the cathodic process is under diffusion control or mixed control. 29 In this study, the baseline condition was: 0.2 M (NH4)2S203, 250 ppm [CU]T, p H 10, 25 °C, 450 rpm. The other results as follows (Ouyang ,2001): ft 9 1 • The magnitude of gold leaching rate is around 1x10" mol .m" . s" • The gold leaching rate increases with increasing rotating velocity • The gold leaching rate increases with increasing p H • The gold leaching rate increases with increasing ammonia concentration • The gold leaching rate increases as the thiosulfate ion concentration increases from 0.1 to 0.2 M. Further increasing leads to low gold leaching rates • A s the concentration of ammonium thiosulfate increases, the gold leaching rate increases constantly • The introduction of tetrathionate has a detrimental effect on the gold leaching rate • The gold leaching rate increases with increasing copper ion concentration • A s temperature increases from 25 to 30°C, the gold leaching rate increases a little; further increasing of temperature decreases the gold leaching rate significantly • If the same parameter has contradictory effects on the anodic and cathodic process, the gold leaching rate generally follows the pattern o f the cathodic process The gold leaching rate can be correlated with different parameters (Ouyang, 2001): roc[NH3]1-557x[S2032"]"0-63 x [ O H " f 4 4 7 x [ C u ] T a 6 4 7 x[a>] 0 6 7 5 x T " 0 3 7 2 In recent years, Breuer and Jeffrey published several papers on fundamental studies on gold leaching in A T S - C u system using the rotating electrochemical quartz crystal microbalance (REQCM) (Breuer et al, 2000a; Breuer et al, 2000b; Jeffery 2001; Jeffrey 30 et al 2001; Breuer et al 2002). R E Q C M is a very new and powerful technique in electrochemical studies. This technique w i l l also be used in this project, and w i l l be discussed in Chapter 3. The oxidation of gold in A T S - C u system was studied (Breuer et al, 2000a). Figure 2.20 shows the calculated current densities for gold oxidation at various ammonia concentrations. It seems that passivation occurs with no ammonia, and addition of ammonia can evidently result in increasing of anodic current densities. Thus, it was suggested that the action of ammonia is to alter the surface of the gold electrode, reducing the effect of the surface passivation. 20 -I 'in Q 10 J o NONE 0.2M 0.4M 0.6M / // /'• /' /.'• j / /.'' f •f / / J / / S s * *' -' _^ * ^ - ^  ^— 0 100 200 300 400 Potential ( m V v s S H E ) Figure 2. 20 Effect of ammonia on gold oxidation. Experimental conditions: 0.1 M Na2S203, 30°C (Breuer et al, 2000a). Also , it was found that the presence of copper (IT) enhances the gold oxidation process. But there is no further increase in rate when the copper (U) concentration is greater than 2 m M (Breuer et al, 2000a). 31 The effect of various parameters on the gold thiosulfate leaching kinetics and the undesirable homogeneous copper(U) reduction by thiosulfate have been investigated by Breuer et al (2000b). The leaching kinetics were measured using a rotating electrochemical quartz crystal microbalance ( R E Q C M ) . The cupric concentration in solution was monitored by measuring the limiting reducing current for copper(U) on a rotating platinum electrode at -150 m V vs. S H E . This study focused on homogeneous reduction of copper(U) by thiosulfate according to the simplified overall reaction: 2 C u ( N H 3 ) 4 2 + + 8 S 2 O 3 2 " = 2Cu(S 2 0 3 )3 5 " + 8 N H 3 + S 4 0 6 2 " Equat ion 2.12 It was found that the Cu(U) concentration decreases continuously under the following conditions: 10 m M C u S 0 4 , 0.4 M N H 3 , 0.1 M N a 2 S 2 0 3 , 30°C. This is because Cu(IT) reacts with thiosulfate rapidly . It was found that the gold leaching rate is not dependent on the initial Cu(H) concentration (The Cu(H) concentrations of the tests were 5 m M , and 10 m M ) and further electrochemical studies showed that copper(U)-thiosulfate reaction products hinder gold oxidation. It was found that as the ammonia concentration is increased (the ammonia concentrations were 0 .2M, 0 .4M, 0.6M), the gold leaching rate decreased, and the copper (U) reductive rate decreased significantly. This result is contradicted by the investigation of Ouyang (2001), which showed that at ammonia concentrations of 0.1 M - 0.8 M , the gold leaching rate w i l l increase with an increase of the ammonia concentration. 32 It was found that the gold leaching rate and the copper (U) reductive rate are increased when temperature is increased. It was found that as the thiosulfate concentration is increased (0.05 M , 0.1 M , 0.2 M ) the gold leaching rate is increased, but the rate of copper (II) reduction is increased, too. The effect o f the ammonium concentration on the rate of gold leaching and the rate of reaction between Cu(II) and thiosulfate were investigated. It was found that both are higher as the ammonium concentration increases. After compromising between fast leach kinetics and high thiosulfate consumption, Breuer et al (2000b) suggested the following conditions should be maintained: 0.4 M ammonia, 0.1 M thiosulfate, p H > 11.4 and a temperature of 30°C. The electrochemistry of gold in thiosulfate solutions containing copper and ammonia was studied using a combination of standard electrochemical techniques and R E Q C M (Breuer et al, 2002).This work was different from the previous work (Breuer et al, 2000a; Breuer et al, 2000b) due to the pure gold rather than a gold silver alloy was coated on electrode. The baseline conditions were: 0 .1M sodium thiosulfate, 0.4 M ammonia, and 0.01 M copper sulfate. Experiments were performed at 30°C, with a rotating rate of 300 rpm and a scan rate of 1 mVs" 1 . The solutions were saturated with air for cathodic studies, and deaerated using argon for anodic studies. 33 Initially, the R E Q C M was used to study the leaching of gold at a fixed potential o f 238 m V S H E (Figure 2.21). This is the mixed potential observed during leaching. In solutions containing solely thiosulfate, very little mass change is observed. When ammonia is added to the solution, the gold leaches more readily, but the reaction rate is substantially lower than the measured gold leaching rate in the copper-ammonia-thiosulfate solution. E - 1 5 Thiosulfate only Thiosulfate + ammonia Thiosulfate + ammonia *N% + copper(ll) V » N ' ' 1 ^ 1 100 200 300 t / S Figure 2. 21 Kinetic plot showing the leaching of gold at fixed potential of 238 mV SHE in a solution containing either thiosulfate, thiosulfate and ammonia, or thiosulfate, ammonia and copper (Breuer et al, 2002). In cathodic studies (Figure 2.22), oxygen reduction occurs at potentials more negative than 50 m V S H E in the absence of copper; the reduction of the copper (Uj-amine complex occurs at potentials more negative than 150 m V S H E in the absence of S2O3 2", the current density reaches a limiting plateau at -150 m V S H E , at these potentials, the 34 reduction of copper (II) is limited by the mass transfer of cupric tetrammine to the platinum surface. The most obvious change is the difference in the copper (H) reduction potential in the presence and absence of thiosulfate which indicates that copper (IT) is more readily reduced in a solution containing thiosulfate. -5 "1 -15-| -20 -25 H -30 -35-1 -40 -200 Cu(ll). NH, / / Cu(ll), NH,, SjO^' -100 100 —I— 200 300 E/mV Figure 2. 22 Linear sweep voltammograms for the reduction of oxygen in ammonia-thiosulfate solutions, and for the reduction of copper in ammonia and ammonia-thiosulfate solution (Breuer et al, 2002). In anodic studies, it is clear from Figure 2.23 that in the presence of copper (U), the gold oxidation reaction is rapid and occurs at low overpotentials. Brueuer et al (2002) identified that copper (II) is dominant in promoting the gold oxidation half reaction. Even though the solution containing copper (I) does not contain copper (U), it is believed that there w i l l be copper (H) present at the electrode interface. This is the result of the concurrent oxidation of copper (I) to copper (U), which occurs readily at potentials more positive than 100 m V in these solutions. 35 Figure 2. 23 Linear sweep voltammogram for a gold electrode in solutions containing copper ( solid line). Also shown is the calculated partial current density for the oxidation of gold to gold thiosulfate derived from the mass change (dashed line) measured using REQCM. Also shown as a square symbol is the measured mixed potential and reaction rate(as current density) during leaching (Breuer et al, 2002) It was found that copper (H) concentrations (0, 0.5, 2, 10 m M ) have a significant effect on the gold oxidation half reaction. It is recommended that a copper (H) concentration greater than 2 m M should be maintained in order to achieve appreciable gold oxidation rates. Also , it was reported that an increase in the thiosulfate concentration (0.05, 0.1, 0.2 M ) results in an increase in the current due to gold oxidation. 36 The effect o f temperature on the gold oxidation half reaction was investigated. It was shown that increasing the temperature also significantly enhances the reaction rate. A t a potential of 200 m V S H E , an activation energy o f 55 kJ .mor 1 was obtained that is quite high and indicative of the reaction being chemically controlled (Breuer et al, 2002). 2.4.3 Mechanism of gold thiosulfate leaching Many researchers studied the electrochemical-catalytic mechanism of ammoniacal thiosulfate leaching of gold (Jiang et al 1993, Zhu et al 1994, L i et al 1996, Breuer et al 2000a, Ouyang 2001, Almore et al 2001). However, there still are some conflicts in these studies, thus, no model is generally accepted. Jiang et al (1993) investigated systematically the electrochemistry of gold leaching with thiosulfate. A stationary gold electrode was used. The results showed that the current peak o f anodic dissolution of gold occurs at about 50 m V (SCE) and thiosulfate oxidation occurs at 620 m V ( S C E ) . Ammonia remarkably improves the anodic dissolution rate of gold and reduces the passivation as well as makes the current peak shift negatively. It was found that cupric ions and cupric-ammonia ions have no evident effect on the anodic process. A current step technique was used to investigate the role o f ammonia in the anodic process and the adsorption mechanism of ammonia was excluded. After this study, Jiang et al (1993a) proposed the anodic dissolution mechanism of gold as following: Au-»Au ++e" Equation 2.13 Au++2NH 3-»Au(NH 3) 2 + Equation 2.14 37 Au(NH3)2+ + 2S203 2"-> Au(S203)23" + 2NH3 Equation 2.15 The study of the cathodic process showed that the addition o f cupric ammine ions actively changes the behaviour o f the cathodic process, ammonia and cupric ions have no direct effect on the cathodic process. The reduction of C u ( N H 3 ) 4 2 + to C u ( N H 3 ) 2 + was supposed to be the real cathodic reaction. The regeneration of C u ( N H 3 ) 4 2 + was carried out by oxidation by oxygen. It was suggested that thiosulfate is not involved in the cathodic process. The mechanism of the cathodic process was put forward as follows: Cu(NH3)42+ + e" = Cu(NH3)2+ + 2NH3 Equation 2.16 Cu(NH3)2+ + 1/2 O z + H 2 0 + 2NH3 = Cu(NH3)42+ +20H' Equation 2.17 The mechanism model proposed by Jiang et al (1993a) is showed in Figure 2.24. Gold Surface Anodic Area Au = Au%e / A u V 2 N H 3 = A u C N H 3 > 2 Sy. Au Cathodic A r e a .2* Cu(NH3>4 +e=Cu(NH 3 )2 Solution N M 3 t S a 0 | - A u ( S 2 0 3 > g -Au(NH 3>2 C u ( N H 3 ) f * t 0 2 — O H " + Cu(NH 3 >2 Figure 2. 24 The model of electrochemical-catalytical mechanism of ammoniacal thiosulfate leaching of gold (Jiang et al, 1993a) 38 Zhu et al. (1994) investigated the dissolution of gold in aqueous thiosulfate solution and the effect of the presence of ammonia in the solution the methods of voltammetry and electrochemical impedance spectroscopy (EIS). The electrochemical impedance spectra of gold in the sodium thiosulfate solution in the absence of ammonia accords with the active-passive electrochemical process on the metal surface. The authors attribute this to the formation of elemental sulfur and thus the elemental sulfur at the gold surface passivates the gold dissolution: S2O3 2 ->S° + SO32" Equation 2.18 S 20 3 2 + 6 O H S O 3 2 " + 2S2' + 3H20 Equation 2.19 The elemental sulfur may absorb on the gold surface and anodic process may occur on the gold electrode surface as: S2" + Au^Au|S° +2e' Equation 2.20 The presence of ammonia eliminates the passivation phenomena even in the absence o f copper ion. It was explained that ammonia prevents the gold electrode from passivation by sulfur coating by being preferentially adsorbed on the gold surface to dissolve gold as an amine complex, as was suggested by Jiang et al (1993a). This explanation was supported by the fact that aqueous ammonia addition was superior to ammonium sulfate addition. Normally, gold dissolution in ammoniacal thiosulfate solution in the presence of oxygen is described by the following reaction: 39 4Au + 8S2032" + 2H20 + 0 2 = 4Au(S203)23" + 40H Equation 2.21 To reflect the influence of the copper catalyst in the presence o f ammonia during gold dissolution, L i et al (1996) modified this reaction as: Au + 5S2032' + Cu(NH3)42+ Au(S203)23" + Cu(S203)35"+4NH3 Equation 2.22 Anodic reaction: The reaction for C u ( N H 3 )4 regeneration: Cu(S203)35' + 4NH3 + l/202 + H20-> Cu(NH3)42+ + 3S2032' + 20H" Equation 2.25 This mechanism is quite different from Jiang's work (Jiang et al, 1993a). It shows that ammonia does not directly take part in the anodic reaction. Also , it shows that both the ammonia and thiosulfate ligands are involved in the cathodic reaction. To generate C u ( N H 3 ) 4 2 + , it is critical to keep the concentration ratio o f ammonia to thiosulfate in a certain range according to this study. Breuer at al (2000a) studied the electrochemical aspects of gold oxidation in solution containing thiosulfate, ammonia and copper using a rotating electrochemical quartz Au + 2S2032" -> Au(S203)23" + e Equation 2.23 Cathodic reaction: Cu(NH3)42+ +3S2032_+e" -> Cu(S203)35 + 4NH3 Equation 2.24 40 crystal microbalance ( R E Q C M ) . The quartz electrode was coated with a gold/silver alloy that contained 2 wt% silver by electroplating. From the linear sweep voltammogram of gold oxidation in thiosulfate (without copper and N H 3 / N H / ) the gold oxidation process was found to be obviously hindered. It has been suggested that thiosulfate can disproportionate in alkaline solutions to form sulfide on the gold surface. 3S 20 3 2" + 6OH" -> 4 S 0 3 2 " + 2S2" + 3HzO E q u a t i o n 2.26 Breuer et al (2000a) confirmed that ammonia/ammonium can reduce the passivation, but disagreed that the oxidation of gold in ammonia thiosulfate solution occurs via A u ( N H 3 ) 2 + (Jiang et al, 1993a), because the standard reaction potential of A u ( N H 3 ) 2 + (0.572 V vs. SHE) is more positive than the potential in which passivation has been evidently reduced. It was suggested that the action o f ammonia is to alter the surface o f the gold electrode, reducing the effect of surface passivation. Breuer et al (2002) studied this process again but the quartz electrode was coated with pure gold. This study shows that the dominant cathodic reaction appears to be the reduction of copper(U) tetramine to a copper(I) thiosulfate complex, and this reaction proceeds quite rapidly. Ouyang (2001) studied the electrochemical polarization of gold using the R D E . It was found that ammonia, thiosulfate and copper species all affect the anodic process. The above species plus oxygen are involved in the cathodic process. The role o f copper species are to facilitate oxygen reduction through the Cu(NH 3)4 2 + /Cu(S 2 0 3 )3 5 " couple. 41 The gold leaching rate in ammonium thiosulfate in the presence of copper species is mainly controlled by the cathodic process, which is under diffusion or mixed kinetic control. The results of this study support the mechanism proposed by L i et al (1996) as shown in Figure 2.25. It is worth noting that the copper enhancing the gold anodic oxidation rate was found in coulometric tests. Ouyang (2001) suggested that the copper ion, like NH3, somehow activates the gold surface further to accelerate the gold dissolution rate. Anodic Zone S A 2 -Au(SAV-J1NH3MV, jCufSp,),5-Passivation film To cathode C a t h o d i c Z o n e •> CuCSjO, ) , 8 - • 4 N H , 5j +2H 2 G + 4«"-» Surface accumulation Cu<SiO^^Nfr,50H- Reactant concentration in A u - Bulk solution CU<NHJ 4*>O; ' 401*8,0,),* • O , +16NHj+ 2H,0 e , Surface depletion —> 4Cu(NHJ4'* • 12SJO,*- + 4 0 H " F r o m a n o d e Dif fusion layer Figure 2. 25 Mechanism of gold leaching in ATS system in the presence of copper ion (Ouyang, 2001) 1 42 After combining the model by Jiang et al (1993a) and L i et al(1996), Aylmore et al (2001a) suggested the electrochemical -catalytic mechanism o f gold leaching in A T S - C u system as Figure 2.26. Gold surface Anodic i r a Au=Au* • c A o * - f 2 N H , = Au(NH,) 1* Au* • 2SJO,1" = AuCSiO,)!*" A u Clthodic mrtM Cu(NH,) /* + « = Cu(NH,),* + 2NH, Cu(NH,),* • 3S,0>^ = C I K S J O J , 5 " • 2NH, Cu(NHJ)4 ,+S^), ,' Cu(NHj),* + 2NH 3 Figure 2. 26 The electrochemical model for the copper catalysis mechanism of leaching gold with ammoniacal thiosulfate.( Aylmore et al, 2001a) In Figure 2.26, It is proposed that Cu(NH .3 )4 2 + species present in solution acquired electrons on the cathodic portion of the gold surface and is directly reduced to Cu(NH3)2 + .At the same time, either ammonia or thiosulfate ions react with A u + ions on the anodic surface of gold and enter the solution to form either A u ( N H 3 ) 2 + or Au(S203)23". Depending on the concentration of S2O32", C u ( N H 3 ) 2 + converts to Cu(S203)35" ions, and likewise for A u ( N H 3 ) 2 + . Both the Cu(S203)35" species in solution are then oxidized to 43 C u ( N H 3 ) 4 with oxygen. The predominant cathodic reaction w i l l depend on the relative concentrations o f the species in solution. 2.5 Additives studies A s the consumption of the thiosulfate is a key issue which retards the commercial application of the thiosulfate leaching process, part of the current research focus was on the effect of additives in the A T S - C u system. The goal was to try to find suitable additives that could enhance the gold leaching rate or hinder the degradation of thiosulfate. Kristjansdottir et al (1996) claimed that pyridine could enhance the gold leaching in A T S - C u system (Figure 2.27). The solution was 0 .1M sodium thiosulfate, 0 .01M copper sulfate, 0 .5M ammonium hydroxide. 10"3 m L pyridine was added into 50 m L solution. Figure 2.27 shows that the sample with pyridine gives a much faster rate of gold dissolution than the sample without the activator. 44 TIME (minutes) • Thiosulfate + Pyridine —•—Thiosu l fa te Figure 2. 27 The effect of pyridine on gold leaching in ATS-Cu system L i and Kuang (1998) report that: in thiosulfate leaching of gold from both oxide and sulfide ores, adding of N a C l instead of Cu(H) significantly increased the gold extraction. It is suggested that the formation of intermediate [AuCl 2 ] " facilitates gold oxidation: [AuCl2]~ + 2 S 20 3 2 -> [Au(S203)2]3" + 2C1" Equation 2.27 However it is questionable since the standard reduction potential o f Au/[AuCl 2 ]~ is as high as 1.154 V vs. S H E (Molleman, 1998) and without a strong oxidant, gold is very unlikely to form [AuCl 2 ] " in solution. 45 It was also reported that the addition of sodium dodecyl sulphonate (SDS) significantly increased the gold extraction. This was reported to be due to the fact that SDS can decrease the solution surface tension, and increase the leaching rate. C. X i a et al (2002a, 2002b) investigated the effect of some additives on gold leaching in the A T S - C u system. The main sample used in this investigation was a copper-bearing mild-refractory low-grade gold ore. The grade was ranged from 2.80 to 3.12 grams per ton. There was 10.4% pyrite, 5.1% iron oxide and 0.5% chalcopyrite in this ore. In tests, the standard reagent composition was 0 .3M ( N H 4 ) 2 S 2 0 3 , 0 .03M C u S 0 4 and 3 M N H 3 . Solution p H was 10.2. It was found that the thiosulfate consumption could be reduced from 30kg/t to about 17kg/t by adding strong chelating agents ( E D T A , N T A ) at a proper ratio ( E D T A : C u =1:1). The same improvement could be achieved by reducing copper sulfate addition to a much lower level (0.00075M) or replacing copper sulfate with nickel sulfate at 0.001M. It also found that the gold extraction was actually improved in these three cases. The copper catalyst amount required for an acceptable gold extraction is proved to be much smaller than 0.03M in this case. It was believed that the strong chelating agents like E D T A and N T A are possible additives for optimizing the catalysis condition and minimizing the thiosulfate consumption. D . Feng et al (2002a) investigated the role o f heavy metal ions in gold dissolution in the ammoniacal thiosulfate system. It was found that the effect of heavy metalllic ions on gold dissolution strongly depended on the type of ion and concentrations, and reagent concentrations. A t high reagent concentrations, Pb could only enhance gold dissolution 46 within a concentration of 50 mg/L, and reached the largest extent at 5 mg/L. Over 50 mg/L, Pb decreased gold dissolution. At low lead concentrations, the predominant species for Pb in the leaching region could be Pb(OH) + in the Eh-pH diagram of the Pb-Au-NFL;-S2O3 2 " system. AuPb 2 could also be present in the leaching region, enhancing the gold dissolution. PbO could possibly be the predominant species at high Pb concentrations. This could explain why Pb accelerated gold dissolution at low concentrations, and retarded it at high concentrations. Zn resulted in a sharp decreased gold dissolution over 10 mg/L. Cd, Co, Cr and N i retarded gold dissolution at any ion concentration. The gold dissolution rate in the C o 3 + / C o 2 + leaching system was much lower than that in the Cu 2 + /Cu + system, while only very limited gold dissolution was observed in the C r 6 + / C r 3 + system. 47 3. E X P E R I M E N T A L A P P R O A C H 3.1 Introduction In the last twenty years, a new approach to examining electrodes and their interfaces, the electrochemical quartz-crystal microbalance ( E Q C M ) , has emerged as a powerful technique capable of detecting very small mass changes at the electrode surface that accompany electrochemical processes. This relatively simple technique only requires, in addition to conventional electrochemical equipment, an inexpensive radio-frequency oscillator, a frequency counter, and commercially available AT-cut quartz crystals. So far, the E Q C M has evolved into a routine experimental method used in numerous electrochemical laboratories. However, the E Q C M just uses a stationary electrode by which it is not possible to obtain reproducible and defined hydrodynamic conditions. Thus, in the last decade, a rotating E Q C M ( R E Q C M ) was developed to solve this problem (Ritchie et al 1994; Zheng et al 1995; Mendez-Soares et al 1998: Shirtcliffe et al 1999; P. Kern et al 2000: Jeffrey et al 2000 a). So far, no commercial product is available for the R E Q C M , Jeffrey et" al (2000) developed a R E Q C M for the study of leaching and deposition of metals. A large number of hydrometallurgical studies were carried out with this equipment. 48 In summary, the R E Q C M is a powerful technique for electrochemical studies in hydrometallurgical process. I) It can measure mass change in real time. So, it is the quickest, simplest and most accurate method to study electrochemical or leaching kinetics. II) It can distinguish the side reactions from:main electrochemical reactions. III) It is useful to study the mechanism of electrochemical reactions. IV) A R D E is used in the R E Q C M . Therefore, good hydrodynamic reproducibility is available. V) It is convenient to determine whether a reaction is chemically controlled or diffusion controlled. The electrochemical dissolution of gold in thiosulfate solution is complicated by the oxidation of thiosulfate or other thiosalts at the electrode surface. Use of the R E Q C M for the study of gold leaching in thiosulfate allows for direct measurement of gold mass change, minimizing the interference of side reactions with the gold leaching kinetics. Therefore, the R E Q C M was chosen as the main technique in this work. 3.2 Basic principle of R E Q C M Piezoelectric Effect In 1880, Jacques and Pierre Curie discovered that mechanical stress applied to the surfaces of various crystals resulted in an electrical potential across the crystal whose 4 9 magnitude was proportional to the applied stress. This behavior is referred to as the piezoelectric effect. The piezoelectric effect exists only in materials that are acentric, that is, those that crystallize into noncentrosymmetric space groups. A single crystal o f an acentric material w i l l possess a polar axis due to dipoles associated with the arrangement of atoms in the crystalline lattice. The charge generated in a quartz crystal under mechanical stress is a manifestation of a charge in the net dipole moment because of the physical displacement of the atoms and a corresponding change in the net dipole moment. This results in a net change in electrical charge on the crystal faces, the magnitude and direction of which depends on the relative orientation o f the dipoles and the crystal faces. Curie also discovered the converse piezoelectric effect, in which the application o f a potential across these crystals resulted in a corresponding mechanical strain. It is this effect that is the operational basis of the E Q C M . QCM and AT-cut quartz The E Q C M is actually the electrochemical version of the Q C M , which has long been used for frequency control and mass sensing in vacuum and air. The Q C M consists o f a thin, AT-cut quartz crystal with a very thin metal electrode "pad" on opposite sides of the crystal. The terminology " A T " simply refers to the orientation of the crystal with respect to its large faces; this particular crystal is fabricated by slicing through a quartz rod at an angle of approximately 35° with respect to the crystallographic x axis. The electrode pads 50 overlap in the center of the crystal with tabs extending from each pad to the edge of the crystal where electrical contact is made. When an electrical potential is applied across the crystal using these electrodes, the AT-cut quartz crystal experiences a mechanical strain in the shear direction to that resulting from the opposite polarity. Therefore, an alternating potential across the crystal causes vibrational motion o f the quartz crystal with the vibrational amplitude parallel to the crystal surface and the x direction. Sauerbrev Equation The vibrational motion of the quartz crystal results in a transverse acoustic wave that propagates back and forth across the thickness of the crystal between the crystal faces. Accordingly, a standing-wave condition can be established in the quartz resonator when the acoustic wavelength is equal to twice the combined thickness of the crystal and electrodes. The frequency f0 o f the acoustic wave fundamental mode is given by the equation: f0=— Equat ion 3.1 2tg Where Vtr is the transverse velocity o f sound in AT-cut quartz (3 .34xl0 4 m s"1) and tQ is the resonator thickness. The acoustic velocity is dependent on the modulus and density of the crystal. The quartz surface is at an antinode of the acoustic wave, and therefore the acoustic wave propagates across the interface between the crystal and a foreign layer on its surface. A n assumption is made here that the velocity of sound in quartz and the electrodes is identical; while not 51 rigorously true, for a small electrode thicknesses the error introduced by this approximation is negligible. In this case, a change in thickness o f the foreign layer is tantamount to a change in the thickness of the quartz crystal. Under this condition, a fractional change in the thickness results in a fractional change in the resonant frequency; appropriate substitutions yielding the well-known Sauerbrey equation. Sauerbrey demonstrated that the resonance frequency shift o f the quartz oscillator is inversely proportional to a change in deposited mass for small mass changes: (Ward, 1995) -A(MepQr5Af Am = —^ Equation 3.2 2 / o Where: A : the piezoelectrically active area (m 2) P Q : the density of quartz (kg/m 3) U Q : the shear modulus of quartz (N/m 2 ) f0: the resonance frequency of vibration(l/s) Am: the change of mass (kg) Af: the shift in the resonance frequency(l/s) This equation is the primary basis of most Q C M and E Q C M measurements wherein mass changes occurring at the electrode interface are evaluated directly from the frequency changes of the quartz resonator. 52 To achieve a good correlation between A m and A f (to obey the Sauerbrey equation), the metal coatings need to be rigidly coupled to quartz and have constant surface roughness (Jeffrey et al 2000). Faraday law The mass change can be converted into a calculated equivalent current density for gold oxidation using Faraday's Law, as shown in the following equation: nF dm AM dt E q u a t i o n 3.3 Where n is the number of electrons transferred per atom of gold oxidized. F is the Faraday constant (96484 C mol" 1), A is the surface area of the electrode m 2 , M is the atomic mass o f the metal, m is the measured mass of the electrode (kg), and t is the elapsed time of the experiment(s). Rotating disk electrode a) Rate- l imit ing-step Heterogeneous chemical and electrochemical processes proceed in three main steps: (i) transport o f reactants from the bulk solution to the solid-solution interface; (ii) chemical/electrochemical reactions at the solid surface; (iii) transport of the products away from the interface and into the bulk solution. The rate at which the overall reaction takes place is determined by the slowest of the three steps described above. This is referred to as the rate-limiting step. Thus, in order to 53 control the kinetics of an industrially important reaction, it is important to determine the nature of the rate-limiting step for the process. If step (ii) controls the overall speed of reaction, then the process is said to be chemically controlled and marked increases in reaction rate can be achieved by raising the system temperature, since chemically controlled reactions generally have high activation energies. If any o f the other steps control the reaction rate, the process is said to be diffusion controlled and increases in reaction rate can be simply achieved by increasing the agitation of the solution. b) RDE The rate at which reactants are transferred to the so l id - solution interface, often called the flux, can be critically dependent on a range of experimental variables, including the sample geometry, solution agitation and the vessel design. Thus, it is difficult to get the same results in different tests for the same experiments. However, the problem o f irreproducibility can be overcome i f the solid sample is in the form o f a disc, which can be rotated about its central axis, with one face exposed to the reactant solution. When the disc is used as an electrode, it is known as a rotating disc electrode (RDE) . The fluid flow to such a disc surface is laminar over a wide range of conditions and is very reproducible. Therefore, when experiments using a given set of experimental variables are performed with a rotating disc apparatus, the values of flux obtained from different experiments and by various experimentalists are reproducible. 54 c) Levich equation A R D E can continuously supply reactants to the electrode surface, the fluid flow to the disc surface is laminar over a wide range o f conditions. Thus, the rates of mass transfer can be controlled and good reproducibility is available. The rotating disc apparatus has the added advantage that the flux of reactants to the disc surface can be calculated from a theoretically obtained expression known as the Levich (1962) equation, Ja = 0.62Do2/ V W 2 [ o ] Equation 3.4 Where J 0 is the flux of species O (mol m ' V 1 ) , D 0 is the diffusion coefficient of species O in solution (m 2 s"1), v is the kinematic viscosity of the solution ( m V 1 ) , co is the rotating angular rates (s_ 1)and [O] is the bulk solution concentration of species O (mol m"3). The Levich equation was based on the laminar flow condition, which is described by Reynolds number Re: coR2 . „ m Re = Equation 3.5 v Where R was the radius of the electrode disk (m) When the Re exceeds a critical value (1.8~3.1 x l O 5 ) , the fluid flow regime changes qualitatively from laminar to turbulent. B y using the Levich equation, it is a simple matter to compare the calculated flux of reactants with the measured reaction rate for a 55 diffusion controlled reaction. The dependence of the rate upon co is the simplest method of determining whether a heterogeneous reaction is diffusion or chemically controlled. 3.3 REQCM 3.3.1 Design The R E Q C M is based on harmonic oscillations and the piezoelectric effect. When an alternating potential is applied to the quartz crystal, mechanical oscillations with an amplitude parallel to the surface of the crystal occur. A t a sufficiently high frequency (of the order of 10 M H z ) , resonant oscillations take place. In order to apply a potential to the quartz crystal, there are thin metal coatings, which act as electrodes, on either side. If one of these electrodes reacts with the solution in some way, the resultant change in mass, Am, causes a shift in the resonance frequency, Af, of the quartz . The circuitry A n oscillation circuit is used in the construction of the E Q C M . The A C excitation for the quartz crystal is generated by the T T L digital oscillation circuit; this oscillation is in the form o f two out-of-phase 5 V squarewaves to either side of the crystal. When one side of the crystal is at 5 V , the other side is at 0 V and vice versa. This causes the crystal to oscillate at its resonance frequency (Jeffrey et al, 2000a). 56 REQCM electrodes The R E Q C M electrodes were constructed using 10 M H z A T cut quartz crystals. Each 12 mm crystal was mounted onto a hollow P V C cylinder (11.95 mm inner diameter) containing a 12.5 mm X 0.5 mm recess, as shown in figure 3.1. The quartz crystal electrodes were prepared by sputtering a layer of either platinum or gold onto the electrode. The platinum or gold was sputtered with a Balzers Union sputter coating unit (model 020). The flag o f the crystal was covered with a thin layer o f insulating varnish to leave a disc of metal. To study the chemical or electrochemical behaviour of the material of interest, the material must be electrically connected to the surface o f the electrode. This can be accomplished by a number of techniques, such as sputtering, evaporating, electroplating, adsorption or self-assembly (Jeffrey et al, 2000a). The diagram of R E Q C M electrode is shown in Figure 3.1. 57 Holder (b) Figure 3.1 A diagram of the REQCM electrode (Jeffrey et al, 2000a) (a) An end view of the electrode (b) An enlarged view of the end of the electrode (c) The cross-section of the electrode 3.3.2 Equipment set-up The frequency o f the crystal oscillation was measured with an FC-7150 frequency counter, which was interfaced with an I B M compatible P C through the serial port. The electrochemical experiments were performed using an E G & G P A R 273A potentiostat and data acquisition was completed with software which was custom written in Q-Basic. This software allowed the real-time measurement and analysis of potential, current and frequency responses. 58 Mass changes were measured simultaneously with current measurement using the R E Q C M . A platinum wire was used as the counter electrode. A l l potentials were measured relative to the saturated calomel electrode, but are reported relative to the S H E . The set-up of equipment is shown in Figure 3.2 and Figure 3.3. • • ^ J Drive pulley +5V >y Support frame E Q C M Circuit Reference electrode Electrolyte bridge Figure 3. 2 Assembly of cell and REQCM system 59 Frequency counter R E Q C M Computer-Prmtef Cell Work electrode Reference electrode Counter electrode H i 1 RE fCE,fl f E G & G PARlyfodel273'; I* PbtehtiQstet/Gafyanostat Figure 3. 3 Schematic il lustration of R E Q C M 3.3.3 Validation of the equipment To make sure the reliability and accuracy of the R E Q C M , the validation of the Sauerbrey equation and Levich equation by the method recommended by Jeffery et al (2000 a) was accomplished. Validation of the Saubrev equation It was also necessary to establish that the circuitry of the R E Q C M was operating properly and to determine whether the frequency response conformed to the Sauerbrey equation. This was accomplished by measuring the rate of electrodeposition of silver from a solution containing silver cyanide, a process which has been shown to occur at close to 100% current efficiency (Zheng et al 1995). Shown in figure 3.4 is the change in mass (calculated using the Sauerbrey equation) versus the time response for the electroplating of silver at 0.1 m A from a solution containing 0.01 M silver cyanide. It is clear that, as the silver is deposited, the mass of the electrode increases linearly with time. The rate of plating with silver can be simply estimated from the slope of the data, which corresponds to 5.17 X I 0 " mol m" s" . Using the data in figure 3.4, the current efficiency is calculated to be 99.0% (it was 97% in Jeffery's work (Jeffery et al 2000a)). It is thus clear that the Sauerbrey equation gives a good correlation between changes in mass and the frequency response of the quartz crystal during the electrodeposition of silver. 61 18 16 y=0.1115x-0.4687. F^ = 0.9988,^*" 14H 2 « 1 2 i y cn c JS 8 a -1 4H .2 0 20 40 60 80 100 120 140 160 Time(s) Figure 3. 4 A kinetic plot showing the deposition of silver from a solution containing 0.01 M AgfCN)2 "and 0.02 M CN". The experimental conditions were 0.1 mA, 700 rpm and 25° C. Validation of the Levich equation In this study, hexacyanoferrate(IU) was used (with 0.1 M sodium perchlorate as background electrolyte), because its reduction was found to be diffusion controlled at sufficiently negative potentials. Shown in figure 3.5 is a plot of i against com at a fixed potential of -250 m V . From this graph, it is clear that the experimental data obey a linear relationship, which is consistent with the mass transfer to the surface o f the R E Q C M conforming to the Levich equation. This point is further demonstrated by calculating the diffusion coefficient of hexacyanoferrate(III) from the slope of the data. According to Lide (1995), v for water at 25°C has a value of 0.89 X 10"6 m 2 s"1. The calculated value of the diffusion coefficient is thus 0.732 X 10"9 m 2 s"1, which is in good agreement with 62 the reported value of 0.746 X 10 " 9 m 2 s"1 for hexacyanoferrate(m) in 0 . 1 M N a C l O 4 (Fogg and Gerrard 1991) and 0.74 X 10 " 9 m 2 s"1 in Jeffrey's work (Jeffrey et al, 2000). It can therefore be concluded that the mass transfer to the surface of the R E Q C M electrode does conform to the Levich equation Figure 3.5 Validation of the Levich equation by measuring the hexacyanoferrate(III) reduction current density as a function of com. The experimental conditions were 0.01 M FefCN) 6 3", 0.1 M NaC104, -0.25 V vs. SHE and 25°C 63 3.4 Experimental 3.4.1 Reagents A wide variety of reagents were used in this study. Table 3.1 below gives a listing of specifications and sources for most chemicals. Table 3.1 The specifications and sources of reagents Reagents Specification Category # Sources (NH 4 )2S 2 03 >99% 33672-6 Aldr ich Chem. Com., Inc C u S 0 4 . 5 H 2 0 Class I B C-489-500 Fisher Scientific N a 2 S 2 0 3 >99 S-1648 Sigma Chem. Co. N a O H 5 N solution SS256B-500 Fisher Scientific H 2 S 0 4 I N solution SA212B-1 Fisher Scientific Buffer solution p H 4,7,10 - Fisher Scientific N H 4 O H 5 N Solution, 28-30% A-669-225 Fisher Scientific A i r - - Praxair Tech. Inc Nitrogen - - Praxair Tech. Inc. Anthraquinone -2-sulphonic acid(AQ) 97% A9,000-4 Aldr ich Chem. Com., Inc C o ( N H 3 ) 6 C l 3 99% 48.152-1 Aldr ich Chem. Com., Inc A g N 0 3 0.1025N 31,943-0 Aldr ich Chem. Com., Inc P b N 0 3 Crystals A . C . S M C I B E D T A 99.88% Sigma Chem. Co. N T A 99% N840-7 Aldr ich Chem. Com., Inc 64 3.4.2 Experimental procedure and conditions Gold coating procedure For all experiments, gold was electroplated onto the platinum electrode at 25Am" from a solution containing 8 g/L K A u ( C N ) 2 , 25 g/L K C N , 25 g/L K 2 H P 0 4 and 25 g/L K 2 C 0 3 prior to each experiment. To ensure the quality of the coating, there were 2 steps in the coating procedure. A t first, a 5mA (250A/m 2 ) current was used for coating gold from 0 to -100 pg. Finally, a 0.5 m A current was used from 100 pg to -300 pg. Rotating speed is controlled at 700 rpm. After each test was completed, the remnant gold on electrode was dissolved in 0 .1M N a C N solution. A fresh gold coating was used for each test. The solution preparation procedure The procedure was to add the appropriate amount of water to the vessel, which was on the top of a magnetic stirrer, and start the stirrer, then add a certain amount of ammonium thiosulfate (or Sodium thiosulfate) into the vessel. If necessary, a certain amount of ammonia was added into solution. A certain amount of cupric sulfate was added into solution i f it was required. Also , a certain amount of additives were added in additive 65 studies. Adjust p H to desired value using N a O H solution, then transfer solution into volumetric flask and add water i f necessary. Ful ly mix the solution. Only fresh solution was used for each test. Anodic polarization procedure About 150 m L fresh solution was put into the cell, then the circulator was started to keep a constant temperature. Solution was purged with nitrogen gas for 15 minutes, then the electrode was put into solution, and rotation was started. After the purge time was up, the purge was stopped, but the purge gas was kept in the cell above the solution. After making sure that no bubbles were on the electrode surface and in the Luggin capillary, the experiment was started. Baseline conditions: a) N o copper 0.2 M (NH 4)2S 203, 25°C, p H 10 ,450 rpm, nitrogen purge and scan rate o f 1 mV/s b) Wi th copper 0.2 M (NHi)2S203, 250 ppm copper, 25°C, p H 10, 450rpm, nitrogen purge and scan rate of l m V / s Cathodic polarization procedure The procedure is same as anodic polarization, but air was used instead o f nitrogen as 66 purge gas. Baseline conditions: 0.2 M (NH 4)2S 203, 250 ppm copper, 25 °C, pH 10, 450 rpm, air purge and scan rate of lmV/s Leaching studies procedure The procedure is same as before, and air was used for the purging gas. Baseline conditions: a) Open potential with higher reagents 0.2 M(NH 4)2S 203, 250 ppm copper, 25°C, pH 10, 450 rpm and air purge b) Open potential with lower reagents 0.1 M(NH 4 ) 2 S 2 0 3 , 30 ppm copper, 25°C, pH 10, 450 rpm and air purge c) Applied potential with higher reagents 0.2M(NH 4 ) 2 S 2 O 3 , 250ppm copper, 25°C, pHIO, 450rpm, potential 0.25 V vs. SHE and air purge d) Applied potential with lower reagents 0.2M(NH 4 ) 2 S 2 O 3 , 250ppm copper, 25°C, pHIO, 450rpm, potential 0.25 V vs. SHE and air purge 67 Additives studies procedure The procedure is same as before, and air was used for the purge gas. Baseline conditions: a) N o copper 0.2M(NH 4 )2S 2 O3, 25°C, pHIO, 450rpm, 0.25 V vs. S H E and air purge b) Wi th copper 0 . 1 M ( N H 4 ) 2 S 2 O 3 , 30ppm copper, 25°C, pHIO, 450rpm, 0.25V vs. S H E and air purge 68 4. RESULTS AND DISCUSSION 4.1 Anodic polarization studies 4.1.1 Preliminary experiments Preliminary experiments were performed to ensure that the experimental results were reliable. Scan rate The effect of the scan rate on the gold anodic polarization was studied (Figures 4.1,4.2). It was found that the current density curve is similar at the scan range of 0.5, 1 and 2 mV/s. 10 i E 8H « •o ** c 9 o 75 o 1mV/s 50 100 150 200 250 300 350 400 Potential vs SHE.mV Figure 4. 1 Effect of scan rate on the gold anodic polarization in 0.2 M (NH 4 ) 2 S 2 03 solution (pH 10, 25°C, 450 rpm) 69 Figure 4. 2 Effect of scan rate on the gold anodic polarization in 0.2 M (NH 4) 2S203 solution (pH 10, 25°C, 450 rpm) Reproducibility It was found that the reproducibility is highly dependent on the sputtered metal substrate placed on the electrode and the quality of the electrodeposited gold coating. However, good reproducibility could be achieved under the experimental conditions (Figures 4.3, 4.4). To ensure the reliability of experimental data, the reproducibility was checked frequently. Effect of air in the preparation of solution Thiosulfate is known to degrade quickly under oxidizing conditions (especially when catalysts are present). About one half hour was needed for preparing the thiosulfate solutions. Thus it was necessary to investigate the effect of air during preparation of solutions. 70 Figure 4. 3 The reproducibility tests of the gold anodic polarization in 0.2 M (NH4)2S2C>3 solution (pH 10, 25°C, 450 rpm, lmV/s). 5 0 100 200 300 400 Potential vs SHE mV Figure 4. 4 The reproducibility tests of the gold anodic polarization in 0.2 M (NH 4)2S 203 solution (pH 10, 25°C, 450 rpm, lmV/s). Test IrSolution prepared in presence of air. Test 2: Solution prepared under nitrogen. 7 1 30 25 20 -\ E '55 o 10 T3 £ 5^ 3 o Total current density Current density from mass change Test 1 100 200 300 400 500 600 0 -5 -10 Potential vs SHE.mV Figure 4. 5 Effect of air in preparation of solutions on the gold anodic polarization in 0.2 M (NH4)2S2C>3 , 250 ppm [Cu]T solution (pH 10, 25°C, 450 rpm, lmV/s). Test 1: Solution prepared in presence of air. Test 2: Solution prepared under nitrogen. 35 -, 30 -25 -Aim 20 -'35 15 -c <B •o 10 -c 0) L_ 3 5 -O 0 -0 -5 --10 -Total current density Current density from mass change Test 1 Test 2 100 200 300 400 500 600 Potential vs SHE.mV Figure 4. 6 Effect of adding copper (II) in solution preparation on the gold anodic polarization in 0.2 M (NH 4)2S 203 , 250 ppm [Cu]T solutions (pH 10, 25°C, 450 rpm, 1 mV/s).Test 1: Adding copper (II) after adjusting pH. Test 2: Adding copper (II) before adjusting pH. 72 In Figure 4.5, test 1 was carried out in a solution prepared in the presence of air, test 2 was carried out in a solution prepared under nitrogen purging. Figure 4.5 shows that the anodic polarizations are similar in both cases. Therefore, it can be concluded that a short exposure to air has no effect on the preparation of solutions. Effect of adding copper (TO in the preparation of solutions A s the NH 3:NFf4 + ratio is dependent on the p H value, and copper (U) forms complexes just with NH3, it is necessary to investigate the effect of adding copper (H) on polarization tests (Figure 4.6). Test 1 is conducted in solution that is made by adding copper (U) after adjusting p H . Test 2 is conducted in solution that is made by adding copper (H) before adjusting p H . Figure 4.6 shows that anodic polarizations are similar in both cases. Therefore, it can be concluded that adding copper (U) has no effect on preparation o f solutions. Comparing the effect of adding NEU or adding N H / at same pH The effect of adding ammonia or adding ammonia ion at p H 10 were investigated (Figure 4.7). Figure 4.7 shows that both ammonia and ammonium have basically equal effects on the oxidation of gold when the p H value is the same. This result is not surprising as it is well know that an equilibrium, dependent on p H , exists between ammonia and ammonium. Therefore, no matter what form o f ammonia is added, the effect w i l l only depend on the p H value. 73 18 i -4 J Potential vs SHE.mV Figure 4. 7 Effect of adding ammonia or adding ammonia ion on the gold anodic polarization at pH 10 (25°C, 450 rpm, 1 mV/s) 4.1.2 Anodic polarization studies 4.1.2.1 Anodic polarization in the absence of copper Effect of temperature Figure 4.8 shows the effect of temperature on the total current density on the gold anodic polarization in 0.2 M (NH4)2S203 solution. Figure 4.9 shows the effect of temperature on the current density calculated from the rate of gold mass change on the gold anodic polarization in 0.2 M (NH4)2S203 solution. 74 Figures 4.8 and 4.9 show that temperature significantly affects both the total current density and current density from the rate of mass change, increasing temperature w i l l result in increasing both current densities. This result is consistent with Ouyang (2001) who measured the total current density using R D E under similar conditions. Effect of rotating speed Figure 4.10 and 4.11 show the effect of rotating speed on the gold anodic polarization in 0.2 M (NH 4)2S 203 solution. Figure 4.10 and 4.11 show that rotating speed has almost no effect on both the total current density and the current density calculated from the rate o f mass change. This result is consistent with Ouyang (2001) who measured the total current density using a R D E under similar conditions. This result once again confirms that the anodic process is not under mass transfer control. Effect of pH value Figures 4.12 and 4.13 show the effect o f p H on the gold anodic polarization in 0.2 M (NH 4)2S 203 solution. Figures 4.12 and 4.13 show that lower p H (<9) or higher p H (>11) w i l l hinder the rate of anodic oxidation of gold, the best p H value appears to be around 10. This result does not agree with Ouyang (2001) who measured the total current density using a R D E . Ouyang found that the current densities at p H 8 and 9 are significantly higher than those at p H 10 and 11. 75 4 5 - i 4 0 -CN E 35 -30 -in c o 2 5 -u "E £ 2 0 -3 U 15 -75 o 1 - 10 -5 -0 -r 4 5 ° C 1 0 0 2 0 0 3 0 0 4 0 0 Poten t i a l v s S H E m V 5 0 0 6 0 0 Figure 4. 8 Effect of temperature on the total current density on gold anodic polarization in 0.2 M(NH 4)2S 203 solution, no copper (pH 10,450 rpm, 1 mV/s). 100 200 300 400 Potent ia l v s S H E m V 500 600 Figure 4. 9 Effect of temperature on the current density from gold mass change on gold anodic polarization in 0.2 M (NH 4) 2S 20 3 solution, no copper (pH 10, 450 rpm, 1 mV/s). 76 25 20 I 15 '5! g 10 -| 4-* c fc c 3 v o Total current density Current density from mass change 200rpm 100 200 300 400 Potential vs S H E mV 500 600 Figure 4. 10 Effect of rotating speed on the gold anodic polarization in 0.2 M (NH4)2S2C>3 solution, no copper (25°C, pH 10,1 mV/s). 25 20 ~ E 15 c I 0 -10 Total currnet density Current density from mass change 1600rpm 100 200 300 400 500 600 Potential vs S H E mV Figure 4. 11 Effect of rotating speed on the gold anodic polarization in 0.2 M (NH 4)2S 20 3 solution, no copper (25°C, pH 10,1 mV/s). 77 2 5 n 0 1 0 0 2 0 0 3 0 0 4 0 0 5 0 0 6 0 0 Potential vs SHE mV Figure 4. 12 Effect of pH value on the total current density on the gold anodic polarization in 0.2 M (NH4)2S203 solution, no copper (25°C, 450 rpm, 1 mV/s) 2 5 i 0 1 0 0 2 0 0 3 0 0 4 0 0 5 0 0 6 0 0 Potential vs SHE mV Figure 4. 13 Effect of pH on the current density from gold mass change on the gold anodic polarization in 0.2 M (NH 4 ) 2 S 2 03 solution, no copper (25°C, 450 rpm, 1 mV/s). 78 4.1.2.2 Anodic polarization in the presence of copper Effect of copper a) Effect of concentration of copper Figure 4.14 shows that the anodic current density of gold dissolution increases with increasing [ C U ] T concentration from 0 ppm to 50 ppm. Figure 4.15 shows that the anodic current density of gold increases with increasing [ C U ] T concentration from 50 ppm to 250 ppm when the potential is less than 0.28V vs. SHE, but does not change after 0.28V vs. SHE. Figure 4.16 shows that the anodic current density of gold dissolution increases slightly with increasing [ C U ] T concentration from 200 ppm to 500ppm. In summary, these experiments show that copper can improve the oxidation of gold. However, it is worth noting that copper (U) can consume thiosulfate at the same time. There will therefore be a balance between fast leach kinetics and high thiosulfate consumption under certain conditions. 79 30 ! 25 120 >» '</> §15 T3 c 9> 10 H Total current density Current density from mass change 100 200 300 400 Potential vs SHE.mV 500 600 Figure 4. 14 Effect of [Cu]T concentrations on the gold anodic polarization in 0.2 M ( N H 4 ) 2 S 2 0 3 solution (25°C, pH 10,450 rpm, 1 mV/s). 30 i 120 Total current density Current density from mass change 100 200 300 400 Potential vs SHE.mV 500 600 Figure 4. 15 Effect of [Cu]T concentrations on the gold anodic polarization in 0.2 M (NH4)2S203 solution (25°C, pH 10,450 rpm, 1 mV/s). 80 600 Potential vs SHE.mV Figure 4. 16. Effect of [Cu]T concentrations on the gold anodic polarization in 0.2 M (NH 4)2S 20 3 solution (25°C, pH 10,450 rpm, 1 mV/s). b) Scanning at wider potential range Linear potential sweep voltammetry experiments over a wider range have been conducted (Figure 4.17). It was found that the gold oxidation peak was around 0.55-0.7 V vs. S H E without copper but dropped to around 0.25 V vs. S H E with 250 ppm added copper. This result indicates that the overpotential, which is required to oxidize gold to gold thiosulfate, can be decreased with added copper. To support this result, anodic polarization experiments in the presence of copper on a platinum electrode (instead of gold electrode) were conducted (Figure 4.18). It was found that there was no mass change on the platinum electrode. This result confirms that only gold oxidation contributes to the current peak calculated from the mass change, which was attributed to the mass change measured in Figure 4.17. j 81 Figure 4. 17 Effect of [Cu]T on the gold anodic polarization in 0.2 M (NH4)2S203 solution (25°C, pHIO, 450rpm, lOmV/s) Potential vs SHE.mV Figure 4. 18 Anodic polarization on a platinum and gold electrode in 0.2 M (NH4)2S203 solution ( 25°C, pH 10,450 rpm, 10 mV/s). 82 c) Adding copper(II) immediately before starting the test To understand the role o f Cu(NH3)4 2 +, the following test was designed and conducted (results shown in Figure 4.19). Test A : 250 ppm copper (IT) added to solution 20-30 minutes before testing. Test B : 250 ppm copper (IT) added immediately before starting the test. It was observed in the experimental procedure that: Test A : The solution colour in the cell was pale blue with an open circuit potential of about 0.04 V vs. S H E . Test B : The solution colour in the cell was deep blue with an open circuit potential o f about 0.28 V vs. S H E . From the observation o f different colors and different open potentials o f solution, it is obvious that the concentration of Cu(NH3)4 2 + in test B is higher than test A . So, a reduction current appeared in test B (Figure 4.19), this reaction probably is: Cu(NH3)42" + e" = Cu(NH3)2+ + 2NH3 Equation 4.1 However, from Figure 4.19, it is interesting to find that the current densities from mass change in both tests are basically the same. This result is not initially expected since a higher concentration of copper tetrammine generally leads to a higher gold oxidation current density. A reasonable explanation is that the 250ppm copper (IT) concentration is too high under these experimental conditions, and excess copper (IT) has no catalyzing role for gold oxidation but only oxidizes the thiosulfate. A repeat test was carried out and reported in Figure 4.20 which shows very good reproducibility. 83 30 i 25 -20 -CN A/m 15 ->. + J "3 10 -c o •a 5 -1 &. 3 0 -o 0 -5 --10 --15 -Test A Total current density Current density from mass change 600 Potential vs SHE.mV Figure 4. 19 Gold anodic polarization test in 0.2 M ( N H 4 ) 2 S 2 0 3 solution ( 25°C, pH 10, 450 rpm, 1 mV/s). Test A: 250 ppm copper (II) added to solution 20-30 minutes before testing. Test B: 250 ppm copper (II) added immediately before starting the test. 30 25 H 20 CN t 10 c a 2 5 c £ = 0^= o -5 -10 -15 Total current density Current density from mass change Test 1 Test 2 300 400 500 600 Potential vs SHE.mV Figure 4. 20. Reproducibility test for adding 250 ppm copper (II) immediately before starting the test in 0.2 M (NH4)2S203 solution (25°C, pH 10,450 rpm, 1 mV/s). 84 d) M o d i f i e d electrode To understand the role of copper and ammonia on gold leaching in thiosulfate solutions, a test using a modified gold electrode was designed and conducted. The modified gold electrode was prepared by immersing the gold electrode into a 0.4 M ammonia hydroxide solution containing 500 ppm copper (II). After 20 minutes, the electrode was removed and rinsed. This modified electrode was then immersed into air-saturated or nitrogen-purged 0.2 M sodium thiosulfate solution, free o f copper and ammonia. Disappointingly, it was found that there was no difference between using the modified and unmodified electrodes. Effect of S?(K [CUIT, FNEM The effects of thiosulfate, ammonia, and copper on gold polarization have been studied. A s can be seen from Figure 4.21, almost no current density of gold oxidation appeared in 0.2 M thiosulfate solution, and the current density of gold oxidation was very low in 0.2 M thiosulfate and 250 ppm copper solution. The gold oxidation also was reasonably slow in 0.2 M thiosulfate and 0.4 M NH3 solution, especially at lower potentials (less than 0.4 V vs. SHE) . A reasonable gold oxidation current density only appeared in 0.2 M thiosulfate, 0.4 M N H 3 and 250 ppm copper solution. This result indicates that the rate of anodic gold dissolution increases to the maximum when each of thiosulfate, ammonia and copper are present. 85 E g , 6 u CO (0 A (0 H E E p (A C V •o c 0 N a 2 S 2 0 3 0 .2M [Cu] T 2 5 0 p p m N H 3 0.4M N a 2 S 2 0 3 0.2M [Cu] T 250ppm 100 200 300 400 500 600 Potential vs SHE.mV Figure 4. 21 Effects of thiosulfate, ammonia, and copper on gold anodic polarization ( 25°C, pH 10,450 rpm, 1 mV/s). 100 200 300 Potential vs SHE.mV 400 500 Figure 4. 22 Effect of (NH 4)2S 20 3 concentrations on the gold anodic polarization in 0.2 M (NH4)2S203 solutions ( 250 ppm copper, 25°C, pH 10, 450 rpm, 1 mV/s). 86 Effect of concentration of ( N H ^ S i O j A s shown in Figure 4.22, the current density of gold oxidation increases significantly with increasing concentration of (NH4)2S203 .This result can be explained by the equation 4.2, the regeneration of C u ( N H 3 ) 4 2 + , Which is believed to be the anodic electrochemical reaction: Cu(S203)35" + 4NH3 = Cu(NH3)42+ + 3S2032_ + e" Equation 4.2 The anodic regeneration of cupric ammine from cuprous thiosulfate w i l l be favoured at higher ammonia to thiosulfate ratios. It is interesting to find that passivation occurs in high (NFLO2S2O3 concentration (0.4 M , 0.8 M ) solutions at high potentials. This phenomenon may be due to S species (as degradation products from thiosulfate) forming deposits on the gold electrode and inhibiting further oxidation of gold. Effect of concentration of N H i A s shown in Figure 4.23, N H 3 has a positive effect on gold oxidation. Increasing the concentration of NH3 w i l l increase the gold oxidation rate. Many researchers believe that the role o f NH3 on the anode process is to alter the surface of the gold electrode, reducing the effect of the surface passivation. This result is consistent with this understanding. Furthermore, this result can be understand by equation 4.2,from which it can be seen that increasing the concentration of ammonia w i l l favour 87 the anodic reaction of cuprous to cupric oxidation when the concentration of S2O3 2" is fixed. It is worth noting that the current density in 1.6 M NH3 is similar, even slightly lower than that measured at 0.8 M N H 3 in the potential range of 0-0.3 V vs. S H E . This result may be due to an excess of NH3 blocking the gold surface, preventing thiosulfate access. Thus, it can be concluded that the concentration of NH3 should be less than 1.6M under experimental conditions. Effect of concentration of S i O ^ ~ Figure 4.24 shows that the gold oxidation rate was similar between 0.1 and 0.2 M S2O3 2" and also between 0.4 and 0.8 M . Also , it was found that the gold oxidation current at 0.1 and 0.2 M S2O32" was higher than the current at 0.4 or 0.8 M when the potential was lower than 0.35 V vs. S H E . This result indicates that too much S2O3 2" w i l l hinder gold oxidation when the concentration of N H 3 is fixed. Again, this result can be interpreted by the equation 4.2. Similar phenomenon was also reported by Ouyang (2001) who found that the gold leaching rates were exactly the same at 0.1 and 0.2 M sodium thiosulfate concentration, and then fell steeply from 0.2 to 0.4 M , further increasing the sodium thiosulfate concentration has no significant influence on gold leaching. 88 600 Potential vs SHE .mV Figure 4. 23. Effect o f NH 3 concentrations on the gold anodic polarization in 0.2 M Na2S203 solutions (250 ppm copper, 25°C, pH 10,450 rpm, 1 mV/s). 600 Potential vs SHE.mV Figure 4. 24. Effect of S2032" concentrations on the gold anodic polarization in 0.4 M NH 3 solutions (250 ppm copper, 25°C, pH 10,450 rpm, 1 mV/s). 89 4.1.3 Summary In the absence of copper Table 4. 1 Influencing factors for gold oxidation with anodic polarization( no copper) Baseline conditions: 0 .2M (NHO2S2O3, no copper (pHIO, 450rpm, 25°C,nitrogen purge) Factors Gold oxidation Comments p H value Positive and negative Optimum value is around 10 Temperature Positive Rotating speed N o effect Probably under chemical control In the presence of copper Table 4. 2 Influencing factors for gold oxidation on anodic polarization (with copper). Baseline conditions: 0.2 M (NH 4 )2S 2 03 , 250 ppm copper (pH 10,450 rpm, 25°C,nitrogen purge). Factors Eff ect Comments Gold oxidation Consumption of thiosulfate Copper (adding as C u 2 + ) Positive Negative Optimum concentration exits N H 3 Positive 1.6M probably blocks the gold surface S 2 0 3 ' - Negative and positive Negative:0~0.35V vs. S H E Positive: over 0.35V vs. S H E (NH4 )2S 2 0 3 Positive 90 4.2 Cathodic polarization studies 4.2.1 Preliminary experiments Reproducibility Figure 4.25 shows that cathodic polarization experiments have good reproducibility. 3 i Figure 4. 25. Reproducibility tests of gold cathodic polarization in 0.2 M (NH 4) 2S20 3 solution (250 ppm copper, pH 10,25°C, 450 rpm, 1 mV/s). Effect of gold electrode and platinum electrode Figure 4.26 shows that the limiting reduction current densities on gold electrode and platinum electrode are same. There was no mass change on platinum over the tested potential range, but there was gold oxidation on the gold electrode at a higher potential (>0 V vs. SHE). 300 -2 Potential vs SHE.mV 91 12 10 8 Total current density Current density from mass change Gold electrode 300 Potential vs SHE.mV Figure 4. 26. Comparing the effects of using gold electrode or platinum electrode on the cathodic polarization (0.8 M (NH 4 )2S 2 03 ,250 ppm copper,, pH 10, 25°C, 450 rpm, 1 mV/s). E •o C O 1.5 1 H 0.5 -0.5 A NH 3 0.4M N a 2 S 2 0 3 0.2M Cu 250ppm 400 300 Potential vs SHE.mV Figure 4. 27 Effect of adding ammonia or adding ammonia ion on the gold cathodic polarization (pH 10,25°C, 450 rpm, 1 mV/s). 9 2 Comparing the effect of adding N H i or adding N H / at the same pH Figure 4.27 also shows that both ammonia and ammonium have similar effects on the gold cathodic polarization when the p H is the same. This result is consistent with the results o f the anodic polarization studies (Figure 4.7). 4.2.2 Cathodic polarization studies Effect of ( N H ^ S i O i concentration From Figure 4.28, it can be seen that the concentration of (NH4)2S203 has a positive effect on gold polarization, limiting current density increases with increasing the concentration o f (NH4)2S203 under the experimental conditions. This result can be explained by the impact of increasing ammonium thiosulfate on the the concentration of Cu(NH3) 4 + in solution. The cathodic limiting current density is for the reduction of cupric ammine. A s the limiting current density increases in Figure 4.28 with increasing ammonium thiosulfate, this indicates that cupric ammine is increasing. Effect of copper(II) concentration Not surprisingly, as shown in Figure 4.29, the cathodic current limiting density increases with increasing copper concentration. This is due to more copper addition giving more 2"i" C u ( N H 3 ) 4 species in solution. From the equation 4.3, it can be seen that an increase in Cu(NH3)42+ w i l l enhance the cathodic current density. C u ( N H 3 ) 4 2 + + e' = C u ( N H 3 ) 2 + + 2 N H 3 Equat ion 4.3 93 12 10 8H E '35 c a> •o c a> k. 3 o Total current density Current density from mass change 0.8M (NH 4 )2S 2 03 0.2M (NH 4 ) 2 S 2 0 3 300 Potential v s S H E . m V Figure 4. 28. Effect of (NH^SjOs concentrations on the gold cathodic polarization (copper 250 ppm, pH 10,25°C, 450 rpm, 1 mV/s, air purge). 300 Potent ia l v s S H E . m V Figure 4. 29. Effect of copper concentration on the gold cathodic polarization in 0.2 M (NH 4)2S 203 (pH 10, 25°C, 450 rpm, 1 mV/s, air purge). 94 Cathodic polarization in 0.4M NH^solution 300 Potential v s S H E . m V Figure 4. 30. Cathodic polarization in 0.4 M NH 3 solution (pH 10, 25°C, 450 rpm, 1 mV/s, air purge or nitrogen purge). A s shown in Figure 4.30, reduction occurred below a potential of -0 .05 V vs. S H E with N 2 purge, but occurred below -0.01 V vs. S H E with air purge, and the current density with air purge was much higher than with N2 purge. Both currents did not reach the limiting current in the experimental scan range. The currents may have resulted from the reduction of oxygen in solution on the gold electrode surface. 95 Cathodic polarization in 0.4 M NHj_250 ppm copper solution 10 o Mass change F r o m a - b: « -600 -500 -406 -300 -200 -100 0 1 0 0 / 2 0 0 ° 300 400 5 -10 - | / Current density from mass change (A § -20 13 C § -30 3 o -40 -50 J Total current density C u ( N H 3 ) 4 ^ C u ( N H 3 ) 2 < | F rom b - c: C u ( N H 3 ) 4 2 ^ — • C u or C u ( N H 3 ) 2 + — • C u 250 200 150 _ D) 100 "jf ro E 50 2 o 0 0 -50 -100 Potential v s S H E m V Figure 4. 31 Cathodic polarization in 0.4 M NH 3 ,250 ppm copper solution (pH 10, 25°C, 900 rpm, 1 mV/s, air purge). Thiosulfate was not added. A s Figure 4.31 shows, a reduction current appeared after the potential was lower than 0.22 V vs. S H E , and a high limiting current density (22 A / m 2 ) was observed in scan range of -0.02 to -0.3 V vs. S H E . It was very interesting to find that the mass increased dramatically after the potential was lower than -0.3 V vs. S H E (there was no mass change above -0.3 V vs. S H E . This result indicated that something was deposited on the electrode. Based on this system, it is obvious that elemental copper deposited on the electrode. So, it is reasonable that at potential lower than - 0.3 V vs. SHE(b-»c) the cathodic reaction should be: 9 6 Cu(NH3)2+ + e" = Cu + 2NH3 Or Cu(NH3)42++ e" = Cu + 4NH3 E° =-0.1 IV E°= -0.05 V Equation 4.4 Equation 4.5 Due to the lack of mass change in the potential region of 0.22 to -0.3 V vs. S H E (a-»b) , the cathodic reaction that contributed a high limiting current density was probably: Cu(NH3)42+ + e = Cu(NH3)2+ + 2NH3 E°=0.1 V Equation 4.6 Comparing cathodic polarization in 0.4M NH^250ppm copper solution between air purge and nitrogen purge -30 J Potential v s S H E , m V Figure 4. 32. Comparing cathodic polarization in 0.4 M NH3, 250 ppm copper solution between air purge and nitrogen purge. No thiosulfate present. 97 A s Figure 4.32 shows, when using nitrogen for purging, a limiting current still existed in the scan range of 0 to - 0.29 V vs. S H E , but the value was 13 A / m 2 , which was lower than the current density with an air purge (16 A / m ). This result indicates that oxygen contributes to the cathodic process either by reoxidation o f cuprous ammine to cupric ammine or by direct reduction of oxygen on the gold electrode surface. Also , after the potential was lower than -0.29 V vs. S H E , it was observed that copper was produced and precipitated onto the electrode. Effect of rotating speed on the gold cathodic polarization in 0.4M, 250ppm copper solution 350 Potential vs S H E mV Figure 4. 33. Effect of rotating speed on the gold cathodic polarization in 0.4 M NH3, 250 ppm copper solution (pH 10,25°C, lmV/s, air purge) 98 Figure 4.33 shows that the rotating speed significantly affect the cathodic reaction (equation 4.3). The limiting current density was very low (less than l A / m 2 ) with no rotation, but it was quite high (22A/m 2 ) with a rotating speed of 900 rpm. This result suggests that the reaction (equation 4.2) is probably under mass transfer - diffusion control under these experimental conditions. A s shown in Figure 4.34, there exists a linear relationship between the limiting current density and the square root of rotating velocity. This is in agreement with the Levich equation, which confirms that the cathodic reaction is diffusion controlled. 0 10 20 30 40 50 60 70 80 Figure 4. 34. The relationship between the limiting current and the square root of rotating velocity for experiments with 0.4 M NH3, 250 ppm copper solution (pH 10, 25°C, lmV/s, air purge). 99 Comparing the cathodic polarization in 0.4 M NH^250 ppm copper with and without 0.1 M S^ CW" solution 5 i -400 b-300 -200 -100 0 100 / 200 300 400 E c •o c -15 © O -20 -25 -30 J S2O3^0.1M a y From a - b: C u ( N H 3 ) 4 2 ^ Cu(NH 3) 2 + D S,032" 0 M From b - c: Cu(NH 3) 4 2-i-+ Cu C / or Cu(NH3)2 +—». Cu So, in presence of S 2 0 3 ,the cathodic reaction probably still is: Cu(NH 3 ) 4 2 + + e"= Cu(NH3)2 ++ 2NH 3 Potential vs SHE .mV Figure 4. 35 Cathodic polarization in 0.4 M NH3,250 ppm copper and 0.1 M S2O3 2 " solution (compared with no S2O3 2 , air purge) Figure 4.35 shows that the limiting current density of 16A/m 2 with no thiosulfate present dramatically dropped to 1 A/m2. However, it is worth noting that when S2O3 2" was present, the potential at which reduction occurred was almost the same as the potential ( starting at around 0.2 V vs. SHE) in the absence of S2O3 2 . Based on these results, it is fair to conclude that about 1/16 th o f the copper in solution with thiosulfate present occurs as cupric ammine. The balance of the copper has been reduced to a cuprous species (either ammine or more likely thiosulfate complex) by reaction with thiosulfate. 100 Thus, in the presence o f S2O3 2 the cathodic reaction probably is the same as that occurring in the absence of S2O3 2 , (equation 4.2) in Figure 4.32. Cu(NH3)42++ e" = Cu(NH3)2+ + 2NH3 E°=0.1V Equation 4.7 Therefore, equation 4.7 is the real cathodic reaction in A T S - C u leaching system. Effect S?Oi 2" concentration on the gold cathodic polarization in 0.4M NH^250ppm copper solution Figure 4.36 shows that S2O3 2" has a significantly negative effect on the limiting current of cathodic polarization. Increasing the concentration of S2O3 2" w i l l decrease the limiting current under experimental conditions. This is in agreement with the result o f L i et al (1996) and Ouyang (2001). The reason for this behaviour is that excess thiosulfate increases the reduction of cupric ammine (equation 4.8). The lower the concentration of cupric ammine, the lower the limiting current density. Reaction 4.8 is undesirable as S 2 03 2 " is consumed by reduction o f C u ( N H 3 ) 4 2 + . So, this explains why the cathodic current decreases with the increasing of S2O3 2 " when the concentration of NH3 is fixed. 2Cu(NH3)42+ + 8 S 2 O 3 2 " = 2Cu(S203)35" + 8 N H 3 + S4Ofi: 2- Equation 4. 8 101 1-2 J S 2O 3 2-0.1M Potential vs S H E . m V Figure 4. 36. Effect of S2032" concentrations on the gold cathodic polarization in 0.4 M NH3, 250 ppm copper solution (pH 10, 25°C, 450 rpm, lmV/s, air purge). E -500 -4 w c 01 "D c S? •_ 3 o -6 A -8 0.1 M N H , 0.4 M N H 3 I0 "250~ ^0i5 *~0 100 / / 2 0 0 300 0.8 M N H 3 1.6 M N H 3 Potent ia l v s S H E . m V Figure 4. 37. Effect of concentration of NH 3 on the gold cathodic polarization in 0.2 M Na2S203, 250 ppm copper (pH 10, 25°C, 1 mV/s). 102 Effect of the concentration of ammonia Figure 4.37 shows that NH3 has a significantly positive effect on the limiting current of cathodic polarization. Increasing the concentration of NH3 w i l l dramatically increase the limiting current density. Ouyang (2001) also observed that the cathodic current density increased constantly with increasing ammonia concentration. Ouyang pointed out that the increase of ammonia concentration helps to stabilize C u (U) in the solution (higher concentration) and therefore leads to a higher cathodic current. 4.2.3 Summary a) Influencing factors on cathodic polarization Table 4. 3 The influence of variables on cathodic current response. Baseline conditions: 0.2 M (NH4)2S203, 250 ppm copper (pH 10, 450 rpm, 25°C, air purge). Factors Effect on cathodic current Copper Positive N H 3 Positive S2O3 2" Negative (NH4)2S203 Positive Rotating speed Positive Oxygen Positive b) It can be concluded that the effective cathodic reaction is not the direct reduction of oxygen in the A T S - C u system, but oxygen contributes indirectly to the cathodic reaction by regenerating the cupric ammine species. 103 c) It was proved that the electrochemical reaction on the cathode in the potential range 0.2 to -0.3 V vs. S H E for the solution in the absence o f S2O3 "is: Cu(NH3)42+ + e" = Cu(NH3)2+ + 2NH3 Eo=0.1 V Equation 4.9 d) Based on the fact that the potential at which the cathodic current appeared in the presence of S2032"was close (around 0.2 V vs. S H E ) to that in the absence of S2O32", it is reasonable to believe that the cathodic reaction in the A T S - C u system still is equation 4.9. 4.3 Leaching studies 4.3.1 Preliminary tests Reproducibility tests To assure the reliability o f leaching rate data from the R E Q C M , many experiments testing reproducibility were conducted. It was found that the reproducibility is highly dependent on the platinum surface quality of the R E Q C M electrode. However, very good reproducibility was achieved under the experimental conditions (Figure 4.40). This result indicates that the leaching tests using the R E Q C M are repeatable. 104 200 i 180 -160 -Test 1 3 140 -«r 120 -| 100 -2 80 -O 60 -40 -20 -0 200 400 600 800 1000 1200 1400 1600 1800 T ime(Second) Figure 4. 38. Leaching reproducibility tests in 0.2 M (NH4)2S203, 50 ppm copper solution (pH 10, 25°C, 450 rpm, 0. 25 V vs. SHE). 4.3.2 Leaching studies under open potential 4.3.2.1 Leaching with higher concentrations of reagents The baseline conditions for this part o f the study were NH3/NFJ_4+ = 0.4M, [S2O32"]=0.2 Mand [Cu] T = 250 ppm. To understand the role o f Cu(NH3)4 , C u and NH3, a series o f leaching tests were designed and conducted (Figure 4.39). Test 1: A leaching test was run in 0 .2M Na 2S20 3 solution which was purged by air for 15 minutes. After 280 seconds, 50 ppm Cu(NH3)42+ was added. It was found that almost no mass change before or after the C u ( N H 3 ) 4 2 + was 0 Effect ofCu(NH 2 )4-105 added. Test 2: This test was the same as test 1 but using 0.2 M (NEL^SaOs instead o f Na2S2C>3. It shows that initially, the gold dissolved at a very low rate, but, when Cu(NH3)4 was added, the mass decreased dramatically. Test 3: This test was the same as test 2 except copper was added as C U S O 4 instead of Cu(NH3)4 2 +. Also , it can be seen that the mass decreased dramatically after the C u 2 + was added, indicating that gold was leaching. 210 -1 Time(Second) Figure 4. 39 Effect of Cu(NH3)42+on gold leaching in ATS-Cu system. (pH 10, 450 rpm, 25°C, air purge). Based on these tests, it is clear that the Cu(NH3)42+ species catalyzes gold leaching in A T S system. A s the cupric ammine species requires ammonia in solution to be stable, these tests also confirmed that NH4 + /NH3 is necessary in this system. 106 Leaching tests in presence of S?Oi 2" and in absence of S?CM Figure 4.40 shows that when there is no thiosulfate present, gold leaching does not occur to any extent, this means that Cu-NH.3 system can not leach gold under these experimental conditions. However, Han and Meng (1993) carried out a detailed investigation on the dissolution kinetics of gold in ammoniacal media between 100°C and 200°C in an autoclave. The study revealed that the kinetics of gold dissolution in ammoniacal solution under certain conditions of temperature, pressure and oxidant addition is favorable. 107 Effect of concentration of S ? O i -A s shown in Figure 4.41, the gold leaching rate increased with the increasing of concentration of S2O3 " from 0.1 M , 0 .2M, 0 .4M while the concentration of NH3 was fixed at 0 .4M. However, too high a concentration of S2O32" (0.8M) w i l l hinder gold leaching. This result is in contrast with the anodic polarization studies (Figure 4.24), which indicated that the anodic current densities were higher in 0 .1M and 0 .2M but lower in 0.4 and 0 .8M S 2 0 3 2 \ 212 n Time(Second) Figure 4. 41 Effect of concentration of S2032" on gold leaching in 0.4 M NH 3, 250 ppm copper solution (pH 10, 450 rpm, 25°C, air purge). Effect of concentration of NH^ 108 A s shown in Figure 4.42, the gold leaching rate increased with the increasing o f concentration o f ammonia from 0.2 M , 0 .4M, 0 .8M while the concentration of S2O3 " is fixed at 0 .2M. However, too high ammonia (1.6 M ) hindered the gold leaching. This result is consistent with the anodic polarization studies (Figure 4.23), which suggested that the 1.6M NH3 may block the gold surface from other species. 190 n 0 200 400 600 800 1000 1200 1400 Time(Second) Figure 4. 42 Effect of concentration of NH 3 on gold leaching in 0.2 M Na2S203, 250 ppm copper solution (pH 10,450 rpm, 25°C, air purge). 4.3.2.2 Leaching with lower concentrations of reagents The baseline conditions for studying the behaviour of this system with lower concentration o f reagents were: [NH 3 ]+[NH 4 + ] = 0.2 M , [S 2 0 3 2 "] = 0.1 M and [Cu] T = 30 ppm. 109 Effect of concentration of S?Q ~^ Figure 4.43 shows that the leaching rate decreased with the increasing of concentration o f 2 2 S2O3 at lower concentration. Again, the possible reason is due to excess S2O3 " consuming the C u 2 + , which decreases the concentration of C u ( N H 3 ) 4 2 + . 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 43 Effect of concentration of S2032" on the gold leaching in 0.2 M NH 3 , 30ppm copper solution, open potential (pHIO, 450rpm, 25°C, air purge) Effect of concentration of N f t t Figure 4.44 shows the effect of the concentration o f NH3 at lower concentrations of S2O3 2" and [Cu] T . It was found that the gold leaching rate increased as the concentration of NH3 increased from 0.2 M to 0.8 M .However, when the concentration of NH3 was 1.6 M , the gold leaching rate was less than that at 0.8 M NH3. This result indicated that NH3 has a positive effect on gold leaching but too much NH3 w i l l retard the leaching rate. 110 Also , this result is consistent with the polarization studies (Figure 4.23) and the results for leaching with a higher concentration of reagents (Figure 4.42). 292 -291 -290 -I , , , , , , , , , 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 44. Effect of concentration of NH 3 on gold leaching in 0.1 M Na2S203 solution, 30 ppm copper solution, open potential (pH 10,450 rpm, 25°C, air purge). 4.3.3 Leaching studies under applied potential The in-pulp oxidation-reduction potentials o f 0.16 to 0.30 V vs. S H E are generally found in conventional leaching experiments with air or oxygen addition. Therefore this part o f the study w i l l focus on the leaching in the presence of an applied potential. The baseline potential was 0.25 V vs. S H E . 4.3.3.1 Leaching with higher concentration of reagents The baseline concentrations for this study were: [NH 3 ]+[NH 4 + ] = 0.4 M , [ S 2 0 3 2 ] = 0.2 M and [ C U ] T = 250 ppm. Ill Effect of potential with different concentrations of copper a) No copper Figure 4.45 shows that the leaching rate increases with the increasing of potential in the absence of added copper. It also shows that almost no gold leaching when the potentials are 0.15, 0.20, 0.25 V vs. S H E . This result is consistent with the anodic polarization studies (Figure 4.14). b) 30 ppm copper Figure 4.46 also shows that the leaching rate increased with increasing potential in the presence of 30 ppm copper. c) 50ppm copper Figure 4.47 shows that the leaching rate increased with increasing potential in the presence of 50 ppm copper. This result is consistent with the anodic polarization studies (Figures 4.13 and 4.14) d) 250ppm copper Figure 4.48 also shows that the leaching rate increased with increasing potential in the presence of 250 ppm copper. The leaching rates are similar when the potentials are 0.20, 0.25 and 0.30 V vs. S H E . This point is consistent with the result in the anodic polarization studies (Figure 4.15), in which a plateau appears when the potential is over 0.20 V vs. S H E . From the Figures 4.45, 4.46, 4.47, 4.48, it is clear that, in general, a higher potential can have higher leaching rate under experimental conditions. 112 150mVSHE 200mVSHE 160 -140 -120 -100 -I 1 1 1 1 1 1 , 1 ! 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 45. Effect of potential on the gold leaching in 0.2 M (NH4)2S203 solution, no copper (pH 10,450 rpm, 25°C, air purge) 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 46 Effect of potential on the gold leaching in 0.2 M (NH 4 ) 2 S 2 03 solution with 30 ppm copper (pH 10,450 rpm, 25°C, air purge). 113 140 -120 -100 ] 1 , 1 1 ; 1 , , , , 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 47 Effect of potential on the gold leaching in 0.2 M (NH 4) 2S20 3 solution with 50 ppm copper (pH 10,450 rpm, 25°C, air purge). 120 -100 \ , , , , , , , , , 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 48 Effect of potential on the gold leaching in 0.2 M (NH4)2S203 solution with 250 ppm copper (pH 10,450 rpm, 25°C, air purge). 114 Effect of the concentration of copper at different potentials a) 0.20 V vs. S H E A s shown in Figure 4.49, the leach rate increased with the increasing concentration of copper. It is worth noting that the concentration of copper has a significant effect on the gold leaching rate when the potential was fixed at 0.20 V vs. S H E . b) 0.25 V vs. S H E A s shown in Figure 4.50, the leach rate also increased with increasing concentration o f copper. However, the effect of the concentration of copper on the leaching rate is not as great as that at 0.20 V vs. S H E (Figure 4.51). c) 0.30 V vs. S H E A s shown in Figure 4.51, the effect of the concentration o f copper is quite different between 0.20 and 0.25 V vs. S H E (Figures 4.49, 4.50). The leaching rate was highest at 30 or 50 ppm copper. The leaching rate did not increase when the concentration of copper was increased to 250 ppm. This result indicates that 30-50 ppm copper may be sufficient for effective gold leaching in 0.2 M thiosulfate solution i f the potential is 0.30 V vs. S H E . 0.30 V vs. S H E is a higher potential than is normally achieved in thiosulfate leaching of gold with air or oxygen oxidation. Figures 4.49, 4.50 and 4.51 show that the potential has a greater effect at lower [Cu] T than that at higher [ C U ] T . 115 Oppm copper 140 -120 -100 -I 1 1 1 1 1 1 1 1 1 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 49. Effect of concentration of copper on gold leaching in 0.2 M (NH4)2S203 solution with 0.20 V vs. SHE applied potential (pH 10,450 rpm, 25°C, air purge). Oppm copper 140 -120 -100 -I , , , , 1 , , , , 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 50. Effect of concentration of copper on gold leaching in 0.2 M (NH4)2S203 solution with 0.25 V vs. SHE applied potential (pH 10,450 rpm, 25°C, air purge). 116 120 -100 J , , , , , , , , , 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 51. Effect of concentration of copper on gold leaching in 0.2 M(NH4)2S203 solution with 0.30 V vs. SHE applied potential (pH 10, 450 rpm, 25°C, air purge). Effect of temperature Chemically controlled reactions usually have high activation energy, and thus changes in temperature dramatically influence the reaction rate. So an easy way o f enhancing the kinetics of a chemically controlled reaction is to increase the temperature. Figures 4.52, 4.53, 4.54 show the effect of temperature on the rate of gold leaching at different applied potentials (0.20, 0.25, 0.30 V vs. SHE) . It can be see that temperature has a significant positive effect on the rate of gold leaching at all potentials. This result is consistent with the anodic polarization studies (Figures 4.8 and 4.9). According to the Arrhenius equation, the relationship between activation energy and leaching rate can be described as follows: 117 E 1 _ lnr = x — + C R T Equation 4.10 Where r: leaching rate (mol/m .s) E : activation energy (J/mol) R: gas constant (8.314 J/K.mol) T: temperature (K) C: constant Arrhenius plots for different potentials are shown in Figure 4.55. The apparent activation energies are calculated from the slope o f the fitted straight line and are summarized in Table 4.4. A l l o f the activation energies are higher than 40 kJ/mol. This result indicates that the reaction is chemically controlled over the entire potential range. 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 52. Effect of temperature on gold leaching in 0.2 M (NH4)2S203 solution, 250 ppm copper, 0. 20 V vs. SHE (pH 10,450 rpm, 25°C, air purge). 118 o 100 200 400 600 Time(Second) 800 1000 Figure 4. 53. Effect of temperature on gold leaching in 0.2 M (NH4)2S2C>3 solution, 250 ppm copper, 0.25 V vs. SHE (pH 10,450 rpm, 25°C, air purge). Figure 4. 54 Effect of temperature on gold leaching in 0.2 M (NH4)2S203 solution, 250 ppm copper, 0.30 V vs. SHE (pH 10,450 rpm, 25°C, air purge). 1 1 9 Figure 4. 55 Arrhenius plot for different potentials Table 4. 4 Calculated values of Arrhenius activation energy values for gold leaching at different applied potentials. Potential (V vs. SHE) Activation energy(kJ/moi) 0.20 51.14 0.25 76.07 0.30 68.47 4.3.3.2 Leaching with lower concentration of reagents For the leaching of gold using lower concentrations of reagents, a baseline condition of [NH3]+[NH4+] = 0.2 M, [S2032"] = 0.1 M and [Cu2+] = 30 ppm was chosen. 120 Reproducibility tests Figure 4.56 shows good reproducibility for gold leaching tests with lower concentrations of reagents. 260 -I . . , , , , 0 200 400 600 800 1000 1200 T i m e ( S e c o n d ) Figure 4. 56. Reproducibility tests for gold leaching in 0.1 M (NH4)2S203, 30 ppm [Cu]T solution (pH 10, 25°C,450 rpm, 0.25 V vs. SHE, air purge). Effect of concentration of (NIL^SiOi Figure 4.57 shows that the leaching rate increases with the increase of concentration of (NFLO2S2O3. This result is consistent with anodic and cathodic polarization studies (Figures 4.22 and 4.28). Both of these figures indicate that (NH4)2S203 has a positive effect on anodic current density and cathodic current density. Effect of concentration of [CU IT Figure 4.58 shows that the gold leaching rate increases with the increase of concentration of [Cu]j. It can be found that the leaching rate with 50 ppm [CU]T is similar to that with 121 30 ppm[Cu] T . Also , this result is consistent with the anodic and cathodic polarization studies. A l l o f the individual studies indicate that increasing [CU]T has a positive effect on anodic current density and cathodic current density (Figures 4 . 1 4 , 4 . 1 5 , 4 . 1 6 and 4 . 2 9 ) . Effect of NH,(totan/SzQ3 ratio N H 3 ( t o t a i y S 2 0 3 = 0 : 0 M (NH 4)2S 20 3 , 0 M N H 3 , 0 . 0 5 M N a 2 S 2 0 3 NfL(total ) /S ? Ch = 1 : 0 . 0 5 M (NH 4)2S 20 3 , 0 M N H 3 , 0 . 0 5 M N a 2 S 2 0 3 N H 3 ( t o t a r j / S 2 0 3 = 2 : 0.1 M ( N H 4 ) 2 S 2 0 3 , 0 M N H 3 , 0 M N a 2 S 2 0 3 N H 3 ( t o t a l l / S 2 0 3 = 4 : 0.1 M ( N H 4 ) 2 S 2 0 3 , 0 .2 M N H 3 , 0 M N a 2 S 2 0 3 Figure 4 . 5 9 shows that the N H 3 / S 2 0 3 ratio has a significant positive effect on gold leaching in thiosulfate solutions, thus highlighting the need for a sufficient high N H 3 / S 2 0 3 ratio in the A T S - C u leaching system. Again, according to the equation 4 . 1 1 , 4 . 1 2 : Anodic electrochemical reaction: C u ( S 2 0 3 ) 3 5 - + 4 N H 3 = C u ( N H 3 ) 4 2 + + 3 S 2 0 3 2 + e" Equat ion 4.11 Regeneration o f C u ( N H 3 ) 4 2 + on the cathode: C u ( S 2 0 3 ) 3 5 ' + Vi 0 2 + H 2 0 + 4 N H 3 = C u ( N H 3 ) 4 2 + + 3 S 2 0 3 2 +20H" Equat ion 4.12 It can be seen that the N H 3 / S 2 0 3 ratio is a critical factor for regeneration of C u ( N H 3 ) 4 2 + . Both the anodic and cathodic current w i l l increase with increasing N H 3 / S 2 0 3 ratio, thus a higher N H 3 / S 2 0 3 ratio can enhance the leaching rate. L i et al ( 1 9 9 6 ) also suggests that it 1 2 2 is vital to keep the molar concentration ratio of ammonia to thiosulfate in a certain range in order to regenerate the cupric species. The present work supports this suggestion. Effect of pH value Figure 4.60 shows that the leaching rate increases with the increasing of p H from 7 to 10, but it decreases when the p H value is more than 10. In general, gold leaching w i l l benefit from increasing p H (to the limit of p H -10). One reason is that thiosulfate is more stable under alkaline conditions, the second reason is that increasing p H means having more NH3, which can stabilize the cupric ammine species in solution. However, too high p H w i l l retard the leaching rate. This result is consistent with anodic polarization studies (Figures 4.11 and 4.12). Aylmore et al (2001) pointed out that a high p H value should be avoided, because copper can precipitate from solution as copper oxides. Effect of temperature From Figure 4.61, it is clear that the leaching rate increases dramatically with increasing temperature. The Arrhenius plot is shown in Figure 4.62. The apparent activation energy was calculated from the slope of the curve as 70.1kJ/mol, which suggests that the reaction is under chemical control. This result is consistent with leaching studies in higher concentrations of reagents (Figures 4.55 and 4.56) and anodic polarization studies (Figures 4.8 and 4.9). 123 Effect of rotating speed Figure 4.63 shows that the rotating speed has no effect on the gold leaching rate under experimental conditions, which confirms that the reaction is under chemical control. This result is consistent with the anodic process (Figure 4.10, 4.11). However, it is inconsistent with the cathodic process (Figure 4.33, 4.34), which is under diffusion control. It may therefore be concluded that the leaching process is under anodic control under the conditions tested. 180 -I , 1 i • , , , , , 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 57. Effect of concentration of (NH4)S203 on gold leaching in (NH4)2S203 solution, 30 ppm copper, 0.25 V vs. SHE (pH 10,450 rpm, 25°C, air purge). 124 o 220 200 180 250ppm 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 58. Effect of concentration of [Cu]T on gold leaching in 0.1 M (NH4)2S203 solution, 0.25 V vs. SHE (pH 10,450 rpm, 25°C, air purge). NH3/S203 0 300 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 59. Effect of concentration of NH3/S2O3 ratio on gold leaching in (NH4)2S203 solution, 30 ppm [Cu]T, 0.25V vs. SHE (pH 10, 450 rpm, 25°C, air purge). 125 Time(Second) Figure 4. 60. Effect of pH value on gold leaching in 0.1 M (NH4)2S203 solution, 30 ppm [Cu]T, 0.25 V vs. SHE (pH 10,450 rpm, 25°C, air purge). 0 -I 1 1 1 1 1 1 1 1 1 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 61. Effect of temperature on gold leaching in 0.1M (NH4)2S2C>3 solution, 30 ppm [Cu]T, 0.25 V vs. SHE (pH 10,450 rpm, 25°C, air purge). 126 Figure 4. 62 Arrhenius plot for rate of gold leaching using a lower concentration of reagents, 0.25 V vs. SHE. 0 2 0 0 4 0 0 600 8 0 0 1000 1200 1400 1600 1800 Time(Second) Figure 4. 63. Effect of rotating speeds on gold leaching in 0.1 M (NH4)2S203 solution, 30 ppm [Cu]T, 0.25 V vs. SHE (pH 10,450 rpm, 25°C, air purge). 127 4.3.4 Summary a) Open potential with higher reagent concentrations Table 4. 5 Influencing factors on gold oxidation during leaching tests Factor Effect of gold oxidation Comments N H 3 Positive or negative Positive: from 0.2, 0.4, 0.8 M Negative at 1.6 M S 20 3 2- Positive or negative Positive: from 0.1, 0.2, 0.4 M Negative at 0.8 M Baseline conditions: Open potential, 0.2 M (NFL;)2S203, 250 ppm copper (pH 10, 450 rpm, 25°C, air purge) b) Open potential with lower reagent concentrations Table 4. 6 Influencing factors on gold oxidation during leaching tests Factor Effect of gold oxidation Comments N H 3 Positive or negative Positive: from 0.2-0.8 M Negative at 1.6 M S2O3 2 " Negative 0.1,0.2, 0.4 M Baseline conditions: Open potential, 0.1 M (NH4)2S203, 30 ppm copper (pH 10, 450 rpm, 25°C, air purge). c) Applied potential with higher reagent concentrations Table 4. 7 Influencing factors on gold oxidation during leaching tests Factor Effect of gold oxidation Comments Copper Positive Effect is greater at lower potentials Temperature Positive Chemical control Potential Positive Effect is greater at lower copper concentrations 128 Baseline conditions: 0.25 V vs. S H E , 0.2 M (NH4)2S203, 250 ppm copper (pH 10, 450 rpm, 25°C, air purge) d) Applied potential with lower reagent concentrations Table 4. 8 Influencing factors on gold oxidation during leaching tests Factor Effect of gold oxidation Comments Copper concentration Positive Effect o f 30 ppm and 50 ppm is similar (NH4)2S203 Positive N H 3 / S 2 0 3 2 " is larger than 4/3 N H 3 / S 2 0 3 2 " ratio Positive Critical value is 4/3 p H valve Positive or negative Optimum value is around 10 Temperature Positive Chemical control Rotating speed N o effect Chemical control Baseline: 0.25 V vs. S H E , 0.1 M ( N H 4 ) 2 S 2 0 3 , 30 ppm copper, (pH 10, 450 rpm, 25°C, air purge) 4.4 Additives studies A number of additives were tested for their impact on the rate o f gold leaching. 4.4.1 Anodic polarization studies Effect of Ag + Figure 4.64 shows that A g + addition to solution can improve the rate of gold leaching. But, the reason is uncertain. 129 Effect of NaCl a) 500pppmNaCl From Figure 4.65, it was found that N a C l had no effect on gold anodic polarization in the absence of copper. b) l%NaCl Figure 4.66 shows that the gold oxidation current decreased significantly when the concentration of N a C l was increased to 1%. Also , it is very interesting to find that the oxidation of S2O3 " was almost stopped. c) Bench tests observation Two kinds of solution were prepared and kept static for a period of time to compare the changes in each of the solutions. Solution A : 0.2 M ( N H ^ S a O s , p H 10, 500 ppm copper, no N a C l . Solution B : 0.2 M (NH4)2S203, p H 10, 500 ppm copper, 12.5 g/1 N a C l (1.25%). After about 120 hours, there was a lot of black precipitation (some green) in solution A but no colour change for solution B . After 3 weeks, it was found that there was more black precipitation in solution A , but still almost no colour change in solution B . This observation and Figure 4.66 indicates that C l " ion can reduce the degradation o f thiosulfate. 130 Figure 4. 64. Effect of 0.2 M Ag+ on the gold anodic polarization in 0.2 M (NH4)2S203 solution, no copper (pH 10,450 rpm, 25°C, 1 mV/s). E 3 60 50 40 £ 30 w c V T3 •£ 20 o 10 0 -10 Total current density Current density from mass change 800 Potential vs SHE,mV Figure 4. 65. Effect of 500 ppm NaCl on the gold anodic polarization in 0.2 M (NH4)2S203 solution, no copper, (pH 10,450 rpm, 25°C, 10 mV/s). 131 Figure 4. 66. Effect of 1% NaCl on the gold anodic polarization in 0.2 M (NH4)2S203 solution, no [Cu]T, (pH 10,450 rpm, 25°C, 10 mV/s). Effect of EDTA From Figure 4.67, it can be seen that E D T A does not improve gold leaching. In fact, E D T A hindered gold leaching under standard experimental conditions, especially at high E D T A concentrations (0.01 M ) . Compared with no E D T A , Figure 4.68 also shows that E D T A hinders gold oxidation. However, it is interesting to find that E D T A may be able to hinder the oxidation o f thiosulfate at same time. 132 J 600 Potential vs SHE.mV Figure 4. 67 Effect of concentration of EDTA on the gold anodic polarization in 0.1M (NH4)2S203 solution, 250 ppm copper (pHIO, 450rpm, 25°C, lmV/s) 30 25 ^ 20 CM E S 10 •D 2 5 3 L) 0 -5 -10 J Total current density Current density from mass change (NH^SjOs 0.2M [Cu] T 250ppm ( N H 4 ) 2 S 2 0 3 0.2M [Cu] T 250ppm EDTA 0.01 M 100 200 300 400 Potential vs SHE.mV 500 600 Figure 4. 68. Effect of EDTA on gold anodic polarization in a solution of 0.2 M (NH4)2S203, 250 ppm Cu 2 + (pH 10,450 rpm, 25°C, 1 mV/s). 133 2 E 5 1.5 1 0.5 (A o 0 •o o -0.5 H 5 -1.5 -2 J Total Current densiti Current density from gold mass change 100 200 300 400 500 Potential vs SHE mV Figure 4. 69. Gold anodic polarization in a solution of 0.1 M Na2S2C>3, 100 ppm copper, 0.01 M EDTA (pH 10, 450 rpm, 25°C, 1 mV/s). 20 E c 0) u -5 J Total current density Current density from mass change 100 200 300 400 500 Potential vs SHE.mV 600 Figure 4. 70 Reproducibility tests of anodic polarization of gold in 0.1 M (NH4)2S203 solution, 250 ppm copper, 0.005M EDTA (pH 10, 450 rpm, 25°C, 1 mV/s). 134 The volatile loss of ammonia is a serious design issue for commercial application of ammonium thiosulfate leaching o f gold. It is desirable to identify an alternate catalyst system that works well without volatile losses. Unfortunately, Figure 4.69 indicated that Na2S203, C u 2 + , E D T A system does not leach gold, which means that E D T A can not be substituted for NHs/NH/ to leach gold in the thiosulfate leaching system. Figure 4.70 shows the reproducibility in presence of E D T A 4.4.2 Leaching studies Various leaching tests were also performed in the presence of additives. Effect of EDTA 207 206 _ 205 cn I 204 ra | 203 o O 202 201 -I 200 Add 0.01 mM EDTA 100 200 300 400 Time(Second) 500 600 Figure 4. 71. Effect of EDTA in leaching test in 0.1 M Na2S203, 100 ppm copper (pH 10, 450 rpm, 25°C). 1 3 5 In Figure 4.71, a leaching test was run in 0.1 M Na2S2O3,100 ppm copper solution that was purged with air for 15 minutes. After 280 seconds, 0 .01M E D T A was added. It was found that there was almost no gold mass change before or after E D T A was added. This result indicates that E D T A is not an effective alternate catalyst to ammonia. Effect of EDTA with a high concentration of copper From Figures 4.72, 4.73, it can be seen that when the concentration o f copper is 250 ppm, the effect o f E D T A was quite different from 30 ppm copper. It is interesting to find that E D T A hinders gold leaching only in the early stage (for about 1200 seconds). After this point, E D T A no longer hinders gold leaching. 300 -, 250 -200 -"5f in ra E 150 -T3 O O 100 -50 -0 -No EDTA 0.001 M EDTA 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 72 Effect of EDTA in leaching test in 0.2M (NH4)2S203, 250 ppm copper , 0.20 V vs. SHE (pHIO, 450 rpm, 25°C, air purge) 136 I 150 -\ | 100 50 No EDTA 0.001 M EDTA 0 200 400 600 800 1000 1200 1400 1600 1800 Time(Second) Figure 4. 73. Effect of EDTA in test in 0.2 M (NH4)2S203, 250 ppm copper, 0.25 V vs. SHE (pH 10,450 rpm, 25°C, air purge) Effect of additives in the absence of copper No additives Time(Second) Figure 4. 74. Effect of additives on gold leaching in 0.2 M (NH4)2S203, no copper (except where added), 0.25 V vs. SHE (pH 10,450 rpm, 25°C, air purge), 137 Figure 4. 75 Effect of concentration of Ag+ on gold leaching in 0.2 M (NH 4 ) 2 S 2 03, no copper, 0.25 V vs. SHE (pH 10,450 rpm, 25°C, air purge). Figure 4.74 shows that Co(NH3)6 and anthraquinone(AQ) have almost no effect on gold leaching. A g + has positive effect on gold leaching. It is clear that copper (U) has the strongest effect among of all o f these additives. Figure 4.75 shows that the effect o f 2 m M and 10 m M A g + is similar. 138 Effect of additives in the presence of copper Figure 4. 76. Effect of additives on gold leaching in 0.1 M (NH 4 ) 2 S 2 03, 30 ppm copper, 0.25 V vs. SHE (pH 10,450 rpm, 25°C, air purge) Figure 4.76 shows that most additives hinder gold leaching under experimental conditions. It seems that anthaquinone and pyridine probably have a small positive effect on the rate o f the gold leaching process. 139 4.4.3 Summary Table 4 . 9 Effect of additives for gold oxidation on additives tests Species Gold leaching Degradation of thiosulfate Comments N a C l N o effect Positive at 1 % no effect at 500 ppm A g N 0 3 Positive No effect Same effect at 2 m M and 10 m M E D T A Negative Positive A Q N o effect C o J + N o effect Baseline: 0.25 V vs. S H E , 0.2 M (NH4)2S203, no copper, (pH 10, 450 rpm, 25°C, air purge) Table 4 . 1 0 Effect of additives for gold oxidation on additives tests Species Effect of gold leaching P b ( N 0 3 ) 2 Negative N T A Negative C o ( N H 3 ) 6 C l 3 Negative A g N 0 3 Negative N a 2 S 0 4 Negative N a C l Negative H g C l Negative A Q N o effect Pyridine Probably positive Baseline: 0.25 V vs. S H E , 0.1 M ( N H 4 ) 2 S 2 0 3 , 30 ppm copper, (pH 10, 450 rpm, 25°C, air purge). 140 4.5 The mechanism It is generally agreed that gold leaching in thiosulfate solution follows an electrochemical mechanism and this system is very complicated. In prior studies, several different mechanism models were proposed (Jiang et al, 1993, Zhu, 1993; L i , et al, 1996; Ouyang 2001; Muir , 2002). But, some controversy exists among these studies, and no mechanism model was generally accepted. 4.5.1 Anodic process Because of the limitation of the experimental techniques used, no model noticed the effect of copper on the anodic process in prior studies, even though Ouyang (2001) observed that copper enhanced the gold anodic oxidation rate in coulometric tests but he simply supposed that the role of copper was to decrease the passivation on the gold surface, like NH3. A s the R E Q C M was used in this study, the effect of copper on the anodic process has been distinguished and studied in detail. It was found that copper can reduce the overpotential which is needed to oxidize gold in thiosulfate solution. Therefore, the role o f copper can not be simply explained by only decreasing passivation, whereas the copper should be involved in the anodic reaction and contributes a significant effect on the anodic process. However, the exact role of copper on the anodic process is not clear. 141 Based on the effect o f copper, the following two possible mechanisms on the anode are proposed: Mechanism A It was supposed that the regeneration of Cu(NH3)4 2 + was the electrochemical reaction on anode. On the anode: Step 1: Regeneration o f Cu(NH3)4 2 + Cu(S203)35" + 4NH3 = Cu(NH3)42+ + 3S2032" + e" Equation 4.13 Then, C u ( N H 3 ) 4 2 + oxidizes A u to A u ( S 2 0 3 ) 2 3 " Au + Cu(NH3)42+ + 2S2032" = Au(S203)23" + Cu(NH3)2+ + 2NH3 Equation 4.14 In the bulk solution: Step 2: Cu(NH3) 2 + enters solution then reacts with S 2 03 3 " to form the more stable C u ( S 2 0 3 ) 3 5 -Cu(NH3)2+ + 2NH3 + 3S2033" = Cu(S203)35" + 4NH3 Equation 4.15 Thus, the total anodic reaction is: Au + 2S2032" = Au(S2C>3)23" + e" Equation 4.16 Mechanism B It is believed that the dissolution of gold on the anode is composed of two parts: one part is contributed by electrochemical reaction (equation 4.17), which produces anodic current, the other part is contributed by chemical reaction (equation 4.18) and no currents produced. Equation 4.17 and 4.18 occur on the anode independently. 142 On the anode Electrochemical reaction: Au + 2S2032" = Au(S203)23" + e" Equation 4.17 Chemical reaction: Au + Cu(NH3)42+ + 2S2032- = Au(S203)23" + Cu(NH3)2+ + 2NH3 Equation 4.18 4.5.2 Cathodic process From the cathodic studies (4.2), the mechanism of the cathodic process is proposed : On the cathode Step 1 Cu(NH3)42+ + e" = Cu(NH3)2+ + 2NH3 Equation 4.19 In the bulk solution Step 2 C u ( N H 3 ) 2 + enters the solution then reacts with S 20 3 3" to form the more stable C u ( S 2 0 3 ) 3 5 -Cu(NH3)4+ + 2NH3 + 3S203 2 = Cu(S203)35- + 4NH3 Equation 4.20 Step 3 Regeneration of C u ( N H 3 ) 4 2 + in the presence of oxygen Cu(S203)35' + y2 02+H20+4NH3 = Cu(NH3)42+ + 3S203 2 +20H Equation 4.21 Thus, the total cathodic reaction is: V2 02+ H zO +e" = 20H" Equation 4.22 143 4.5.3 The model of electrochemical mechanism Mode l A According to the mechanism A on anodic process, the model A of the electrochemical mechanism is proposed in Figure 4.77. GOLD SURFACE Anodic area A I Cu(S203)3 ! ,'+4NH3=Cu(NH3)42++3S2032 +e" / v Cu(NH3)2*+ 2NH 3 /j / ^ 3S203 : Au + Cu(NH3)42++2S2032"= Au^Osfe3"* Cu(NH3)2T+2NH3 Cathodic area Cu(NH 3) 4 2 ++e" =Cu(NH3)2 ++2NH3 SOLUTION Cu(S203)3 5'+4NH3 AutS.Oj),3-,Cu(NH 3) 4 2 + + 3S2032" + 20H" | H 2 0 A) t 0.5O; CulS h^+ANHj 3S 2 0 3 2 Cu(NH 3) 2 ++2NH 3 Total reaction Au + 2S203 2' +1/202 + H 2 0 = AutSzOjh3'* 20H" Figure 4. 77 The model A of electrochemical mechanism of gold leaching in ATS-Cu system 144 Model B According to the mechanism B on anodic process, the model B o f electrochemical mechanism is proposed in Figure 4.78. G O L D S U R F A C E Anodic area SOLUTION Au +2S20j2"= Au{S201)23'+ e (Electrochemical dissolve) Au + CutNHjJ^+aSzOj2^ Au(S203)23-+ Cu(NH3)2*+2NH3 (Chemical dissolve) e Cathodic area Cu(NH3)42 * + e" = Cu(NH3)2++ 2NH3 ^ Cu(NH3)42+ i / \ S*../h.lLI \ + J . O M U ' Au^O,)^ Cu(NH3)2+ 2NH3 ^CutNH^2** SSzOj 2* 20H" 0.5O 2^ H 20 CufSzO^+aNHj Cu(NH3)2++2NH3 t 3S2032-Total reaction 1) Au + 2S2032" + 1/202+H20 = Au(S203)2 3 + 20H" (Electrochemical dissolve) 2) Au + C u f N H ^ ^ S A 2 - = AutSiOj)^ Cu(NH3)2++2NH3 (Chemical dissolve) Figure 4. 78 The model B of electrochemical mechanism of gold leaching in ATS-Cu system 145 5 CONCLUSIONS AND RECOMMENDATIONS 5.1 Conclusions This study investigated the kinetics and mechanism of gold leaching in the A T S - C u system using the R E Q C M . The effect of some additives was also studied. Anodic polarization, cathodic polarization and leaching experiments were carried out in this study. On the basis o f the experimental results, the following conclusions were drawn. Both electrochemical studies and leaching studies show that all o f thiosulfate, ammonia, and copper are necessary in the A T S - C u system for gold leaching. In the absence of any of these species, gold leaching can not proceed at acceptable rates under the experimental conditions studied. Because the R E Q C M was used in the anodic polarization study, the effect of copper on the anodic process was distinguished. It seems that copper is involved in the anodic reaction under electrochemical experimental conditions. Based on the role o f copper with respect to the anode, two possible anodic mechanisms were proposed. In mechanism A , it is supposed that copper directly catalyzes the electrochemical reaction (equation 4.13). In mechanism B , it is believed that copper enhances the gold oxidation on anode just by chemical reaction (equation 4.18). Leaching studies show that it was C u ( N H 3 ) 4 2 + that catalyses gold leaching. 146 Using the R E Q C M in the cathodic polarization study, it was proved that, in the absence of S 2 O 3 2 " , the electrochemical reaction on the cathode is the reduction of Cu(NH3)4 2 + to C u ( N H 3 ) 2 + in the potential range of 0.20 to -0.30 V vs. S H E (equation 4.19). Based on the fact that the potential value at which the cathodic current appeared in the presence o f S2032" was close to that in the absence of S2O3 2". It is believed that the electrochemical reaction on cathode also is the reduction of Cu(NH3)4 2 + to Cu(NH3)2 + (equation 4.19). Anodic polarization studies show that the temperature has a significant effect but rotating speed has no impact on gold oxidation. Also , leaching tests show similar results. The activation energy values suggest that both anodic process and leaching process are under chemically control. The rotating speed dramatically affects the cathodic current density. The good relationship between cathodic current density and square root of angular rotation speed indicates that the cathodic process is under diffusion control. Experiments show that the ratio of NH3/S2O3 2" is an important parameter for gold leaching in the A T S - C u system. From the mechanism studies, it seems that the critical value is around 4/3, keeping the ratio larger than this critical value is needed to achieve a reasonable leaching rate. Copper(U) can catalyze gold leaching but also consumes thiosulfate. So, an optimum copper concentration exists in a certain condition. Ammonia usually can enhance the gold leaching. However, too much ammonia w i l l have a negative effect on the leaching rate, the reason probably is that excess ammonia blocks the gold surface from other species , 147 which need to reach the gold surface and become involved in the gold dissolution reactions. The suitable concentration o f thiosulfate is dependent on many factors: the concentration o f copper, the ratio value of NH3/S2O3 2 ", etc. Like ammonia, excess thiosulfate also may passivate the gold surface. Both anodic polarization studies and leaching studies indicate that the suitable p H value for gold dissolution probably is around 10. One reason is that thiosulfate is stable under this alkaline condition; Another reason is that lower p H w i l l decrease the concentration of ammonia, which is needed to solubilize the copper as the copper (IT) ammonia complex. However, too high p H w i l l lead to precipitation of copper as copper oxide. Leaching tests at different applied potentials show that the impact of copper concentration on leaching rate is greater at lower potentials than at higher potentials. These tests also indicate that the impact of potential on leaching rate is greater at lower copper concentrations than that at higher copper concentrations. Additive studies show that, in the absence of copper , A g + enhances gold leaching, although this effect is much less than that of copper(II). It was found that 1% N a C l can inhibit oxidation o f thiosulfate, but it hinders gold leaching at same time. E D T A hinders the oxidation of gold. However, it probably also can hinder the oxidation o f thiosulfate. Finally, the studies show that EDTA-S2032"-Cu system is not effective for gold leaching. 148 5.2 Recommendations In future studies, the passivation/surface layer formation should be investigated using A C Impedance Spectroscopy method and other surface characteristic method such as X P S . Because the high consumption of reagents is cited as a limitation for commercial application, it is necessary to continue to seek an alternative catalyst to replace ammonia/copper or to continue looking for an additive which can improve gold leaching or hinder the degradation of thiosulfate. It w i l l be interesting to further study the role o f A g + on the thiosulfate leaching system. 149 6 REFERENCES (1) Abbruzzese,C.;Fornari,P.;Massidda,R.;Veglio,F.;Ubaldini,S.,Thiosulfate leaching for gold hydrometallurgy, Hydro-metallurgy 39(1995) 265-276 (2) Adanuvor,P.K.; White, R. E . , Oxygen reduction on silver in 6.5 M caustic soda solution, J. Electrochem. Soc. Oct. 1988, 2509-2517 (3) Ar ima, H ; Fujita, T; Yen , W. , Thermodynamic evaluation on gold oxidation and reduction mechanisms in ammonium thiosulfate solution, 2002 SME Annual Meeting, Feb. 25-27, Phoenix, Arizona (4) Aylmore, M . G . ; Muir , D . M , Thermodynamic analysis of gold leaching by ammoniacal thiosulfate using Eh/pH and speciation diagrams, Minerals and Metallurgical Processing 18(4) (2001a) 221-227. (5) Aylmore ,M.G. ; M u i r , D . M . , Thiosulfate leaching of gold-a review, Minerals Engineering , V o l . l 4 No.2, ppl35-174, (2001b) (6) Aylmore, M . G . Treatment of a refractory gold-copper sulfide concentrate by copper ammoniacal thiosulfate leaching, Minerals Engineering, Vol .14, No.6, pp.615-637,2001 (7) Berezowsky, R . M . G . S . ; Sefton, V . B . , Recovery of gold and silver from oxidation leach residues by ammoniacal thiosulfate leaching, Presentation at the 108th AIME Annual Meeting, New Orleans, Louisiana, Feb. 18-22 1979 (8) Block-Bolten, A . ; Torma, A . E . , Thiosulfate leaching of gold from sulfide wastes, Metallwissenschaft Technik 150 (9) Breuer, P. L . ; Jeffrey, M . I., An electrochemical study of gold oxidation in solutions containing thiosulfate, ammonia and copper, Submitted as Conference Proceedings Fifth International Symposium on Electrochemistry in Mineral and Metal processing, The Electrochemical Society, 2000a (10) Breuer, P. L . ; Jeffrey, M . I., Thiosulfate leaching kinetics of gold in the presence of copper and ammonia, Mineral Engineering, V o l . 13.No. 10-11, pp. 1071-1081, 2000b (11) Breuer, P. L . ; Jeffrey, M . I., An electrochemical study of gold leaching in thiosulfate solutions containing copper and ammonia, Hydrometalllurgy 65(2002) 145-157 (12) Breuer, P. L . ; Jeffrey, M . I., Copper catalysed oxidation of thiosulfate by oxygen in gold leach solutions, Minerals Engineering, 16(2003)21-30 (13 )Byerley, J. J.; Fouda, S. A.; Rempel, G. L . ,Kinetics and mechanism of the oxidation of thiosulfate ions by copper(II) ions in aqueous ammonia solution, J .Chem .Soc .Dalton Trans,(\973) 889-893 (14) Byerley, J. J.; Fouda, S. A.; Rempel, G. L . , Activation of copper(II) ammine complexes by molecular oxygen for the oxidation of thiosulfate ions, J .Chem .Soc. Dalton Trans, (1975) 1329-1338 (15) Cao, changlin; Hu , Jiexue; Gong, Qian, Leaching gold by low concentration thiosulfate solution, Transaction of NonFerrous Society of China, Vol .2 No.4, 21-25 Nov. 1992 (16) Chen, Jiayong; DengTong; Zhu Guocai and Zhao Jin, Leaching and recovery of gold in thiosulfate based system-a research summary at ICM, Transaction of the Indian Institute of Metals (India). Met V o l . 49. No.6, December 1996, pp841-849 151 (17) Dasgupta, R.; Guan, Y. C.; Han, K. N, The dissolution behaviour of gold in ammoniacal solution at 75°C, Presentation at the SME Annual Meeting Phoenix, Arizona-March 11-14,1996 (18) Dreisinger, D., The leaching of gold in thiosulfate solution, June 2001, unpublished (19) Feng, D.; Van Deventer, J.S.J. Preg-robbing phenomena in the thiosulfate leaching of gold ores, Minerals Engineering, Vol.11, pp. 1387-1402, 2001 (20) Feng, D.; Van Deventer, J.S.J. The role of heavy metal ions in gold dissolution in the ammoniacal thiosulfate system Hydrometallurgy 64 (2002a) 231-246 (21) Feng, D.; Van Deventer, J.S.J. Leaching behaviour of sulphides in ammoniacal thiosulfate systems, Hydrometallurgy 63 (2002b) 189-200 (22) Feng, D.; Van Deventer, J.S.J. Effect of sulfides on gold dissolution in ammoniacal thiosulfate medium, Metallurgical and Materials Transactions B Volume 34B, February 2003 5-13 (23) Ficeriova, J.; Balaz, P.; Boldizarova, E . ; Jelen, S., Thiosulfate leaching of gold from a mechanically activated CuPbZn concentrate Hydrometallurgy 67(2002)37-43 (24) Gelves, G. A.; Arias, V. A.; Pedraza, J. E . , Recovering of refractory gold using ammonium thiosulfate solutions, Clean Technology for the Mining Industry, University of Concepcion, concepcion-Chile 1996 pp 471-487 (25) Gong, Q.; Hu, J., Treatment of copper-containing gold sulfide concentrate using thiosulfate solution, Journal of Engineering Chemistry and Metallurgy, 1990 (11) 2, 145-151 152 (26) Gong, Q.; Hu , J., Cao, C , Kinetics of gold leaching from sulphide gold concentrates with thiosulfate solutions, Transactions of Nfsoc, Vol .3 No.4 Nov.1993 pp30-36 (27) Grosse, A . C ; Dicinoski , G . W. ; Shaw, M . J.; Haddad, P. R. , Leaching and recovery of gold using ammoniacal thiosulfate leach liquors(a review) Hydrometallurgy 69 (2003) 1 -21 (28) Guan, Y . C ; Han, K . N . , The electrochemical study on the dissolution behaviour of gold in ammoniacal solutions at temperature above 100°C, J. Electrochem. Soc, Vol.143, No . 6, June 1996, 1875-1880 (29) Hemmati, M . ; Hendrix, J .H. ; Milosaufjovic, E . B . , Study of the thiosulfate leaching of gold from carbonaceous ore and the quantitative determination of thiosulfate in the leached solution, Extraction metallurgy '89, Institution of Min ing and Metallurgy, London, 1989, pp665-678 (30) Jeffrey, M. I . ; Zheng, J.; Ritchie, I .M. , The development of a rotating electrochemical quartz crystal microbalance for the study of leaching and deposition of metals, Meas. Sci. Technol. 11 2000 a 560-567 (31) Jeffrey, M. I . ; Ritchie, I .M. , The leaching of gold in cyanide solutions in the presence of impurities I, The effect of lead, Journal of the electrochemical society, 147 (9)3257-3262 2000b (32) Jeffrey, M. I . ; Ritchie, I . M . , The leaching of gold in cyanide solutions in the presence of impurities II, The effect of silver, Journal of the Electrochemical Society, 147(9)3272-3276 2000c (33) Jeffrey. M . I., Kinetic aspects of gold and silver leaching in ammonia-thiosulfate solutions, Hydrometallurgy 60 (2001) 7-16 153 (34) Jeffrey. M.I., Breuer, P. L. and Choo, W.L., A kinetic study that compares the leaching of gold in the cyanide, thiosulfate, and chloride systems, Metallurgical and Materials Transactions B, Volume 32B, December 2001 979-986 (35) J i , Caixia; Y u , Qiang, Research and optimization of thiosulfate leaching technology of gold, Rare Metals, V o l . 10,1991 275-280 (36) JI, J.; Fleming, C.,A.; West-sells, P. G.; Hackl, R. P., Method for thiosulfate leaching of precious metal-containing materials, International Patent WO 01/88212 A 2 2001 (37) Jiang, T; Chen, J.; X u , S., Electrochemistry and mechanism of gold leaching with ammoniacal thiosulfate, Proceedings, 13th International Mineral Processing Congress, Sydney, May, 1993a (38) Jiang, T; Chen, J.; X u , S.,A kinetic study of gold leaching with thiosulfate, Hydrometallurgy: Fundamentals .technology and innovation, 119-126, Edited by: Hiskey, J.B., Warren, G.W., Soc. M i n . , Metall. Explo. ,Littleton, colo. 1993b (39) Jiang, T; X u , s.; Chen, J.; Yang, Y . , Anodic oxidation of thiosulfate ions in gold leaching, J. Cent. South Univ. Technol.Vol A No.2 Nov.1997 pp89-91 (in Chinese) (40) Kerley, B . J.; Jr., Santo Tomas, Recovery of precious metals from difficult ores, US patent 4.369.061 Jan. 18, 1983 (41) Kerley, B . J.; Jr., Santo Tomas, Recovery of precious metals from difficult ores, US patent 4.269.622 M a y 26,1981 (42) Kristjansdottir, S., S; Thompson, J.S., Enhancement of gold lixiviation using nitrogen and sulfur heterocyclic aromatic compounds, US Patent 5,484,470, January 16,1996 154 (43) Kudryk, V . ; Kellogg, H . H . , Mechanism and rate-controlling factors in the dissolution of gold in cyanide solutions, J. Metals , 61(1954):541-548 (44) Lam, A . E . ; Dreisinger, D .B . , The importance of the Cu(II) catalyst in the thiosulfate leaching of gold W i l l be presented to T M S Conference Vancouver August 2003 (45) Langhans Jr., J .W.; Le i , K . P. V . ; Carnahan, T .G. , Copper-catalyzed thiosulfate leaching of low-grade gold ores, Hydrometallurgy, 29(1972)191-203 (46) Levich, V . G . , Physicochemical Hydrodynamics, translated by scripta technical, inc., Prentice- H a l l , Inc., Englewood Cliffs, N . J . 1962 (47) L i , J.; Mi l le r , J.D.; Wan, R . Y . , Important solution chemistry factors that influence the copper-catalyzed ammonium thiosulfate leaching of gold, 125th SME Annual Meeting Phoenix, Arizona, March 11 -14,1996 (48) L i , J.; Mi l le r , J .D.; Wan, R . Y . ; Levier, M . , The ammoniacal thiosulfate system for precious metal recovery., in: Proceedings of the 19th International Mineral Processing Congress, San Francisco, 1995, Chapter 7 (49) L i , R.; Kuang, S., Leaching gold with thiosulfate solution containing added sodium chloride and sodium dodecyl sulfonate, Huagong Yejin,\9 (1), 77-82(Chinese),1998 (50) Loroesch, J; Knorre, H . ; Griffiths, A . , Developments in gold leaching using hydrogen peroxide, Mining Engineering Sept. 1989 963-965 (51) Meng Xinghui and Han, K . N . , The dissolution behaviour of gold in ammoniacal solutions, Proceedings of 4th International Symposium on Hydrometallurgy, pp205-221, AIME/SME(1993) 155 (52) Molleman, E . , The treatment of copper-gold ores by ammonium thiosulfate leaching, Master thesis, University of British Columbia, 1998 (53) Molleman, E . and Dreisinger, D . , The treatment of copper-gold ores by ammonium thiosulfate leaching, Hydrometallurgy 66 (2002) 1-21 (54) Navarro, P.; Vargas,A.; Villarroel, A . ; Alguacil ,F.J. , On the use of ammoniacal /ammonium thiosulfate for gold extraction from a concentrate Hydrometallurgy 65(2002) 37-42 (55) Ouyang, M.,The study of the kinetics of gold leaching in the ammonium thiosulfate system in the presence of copper, Master thesis, University of British Columbia ,2001 (56) Perez, A . E . ; Galaviz, H . D . , Method for recovery of precious metals from difficult ores with copper-ammonium thiosulfate, US patents, 4,654,078, March 31, 1987 (57) Pryor, W . A . , The kinetics of the disproportionate & sodium thiosulfate to sodium sulfide and sulfate, Analytical chemistry, The American Chemical Society, 82,1960, pp4794-4797 (58) Ritchie, I., M . ; Zheng, J.; Labrooy, S.R. and Singh, P. 1994 Extension of electrochemical windows by the use of an electrochemical quartz crystal microbalance Extended Abstracts: 186th Annual Meeting of the Electrochemical Society V o l . (2) 94-2 (Pennington, N J : The Electrochemical Society) pp1044-5 (59) Rolla , E . ; Chakrabartl, C . L . , Kinetics of Decomposition of tetrathionate, trithionate, and thiosulfate in alkaline media , Environ .Sci .Technol. 1982,16,852-857 156 (60) Shirtcliffe, N . , Deposition of clays onto a rotating, electrochemical quartz crystal microbalance, Colloids and Surfaces A: Physicochemical and Engineering Aspects 155(1999)277-285 (61) Skoog, D . A . ; West, D . M . , Fundamentals of analytical chemistry, 1963 (62) Sui, C ; Finch, J .A. ; X u , Z . , Effect of Diethylenetriamine on Xanthate interaction with Pb-contaminated pyrite, Minerals Engineering, V o l . 11 .No.7.pp639-649, 1998 (63) Sun, Xiaowei ; Guan, Y . C . ; Han, K . N . , Electrochemical behaviour of the dissolution of gold-silver alloys in cyanide solutions, Metallurgical and Materials Transaction, Volume27B, June 1996 355-361 (64) Ter-Arakelyan, K . A.;Bagdasaryan, K . A . ; Oyanyan, A . G.;Mkrtchyan,r.t.and Babayan, G . G . , On Technological expediency of sodium thiosulfate usage for gold extraction from raw material, lzv.V.U.Z.Tsvetn, Metall.,{5) pp72-76,1984 (65) Thomas, K . G . ; Fleming, C ; Marchbank, A . R . ; Dreisinger, D . , Gold recovery from refractory carbonaceous ores by pressure oxidation, thiosulfate leaching and resin-in-pulp adsorption, U . S . patent #5,785,736, July 28,1998 (66) Tozawa,K.; Inui, Y . ; Umetsu, Y . , Dissolution of gold in ammoniacal thiosulfate solution, TMSpaper selection A 81-25,1981 (67) Trindane, R . B . E . ; Monhemius, A . , J., The use of anthraquinone as a catalyst in the cyanide leaching of gold, Minerals Engineering, Vol .6 , pp 565-574,1993 (68) Tshilombo, A . F . ; Sandenbergh, R.F . , An electrochemical study of the effect of lead and sulfide ions on the dissolution rate of gold in alkaline cyanide solutions, Hydrometallurgy 60 (2001) 55-67 157 (69) Wan, R.Y.; Vie r M.L.; Miller, J.D., Research and development activities for the recovery of gold from noncyanide solution, Hydrometallurgy: Fundamentals, technology and innovation, 119-126, Edited by: Hiskey, J.B., Warren, G.W., Soc. M i n . , Metall. Explo., Littleton,colo.l993 (70) Wan, R.Y.; Brierley, J.A., Thiosulfate leaching following biooxidation pre-treatment for gold recovery from refractory carbonaceous-sulfidic ore, Mining Engineering Aug. 1997 76-80 (71) Ward , M.D., Principles and applications of the electrochemical quartz crystal microbalance, Physical Electrochemistry: Principles, Methods and Applications Edited by Rubinstein,!, (New york, Marcel Dekker) 1995,pp293-338 (72) White, H.A., The solubility of gold in thiosulfates and thiocyanates, Journal of the Chemical Metallurgical and Mining Society of South Africa, 1905,5109-111 (73) X i a , Chen; Yen , Wan-Tai; Deschenes, G., An investigation on minimizing reagent consumption in ammonium thiosulphate, Proceedings of Annual Meeting of the Canadian Mineral Processors, January 22-24, 2002(a), Ottawa, Canada, pp354-368 (74) X i a , Chen; Deschenes, G, Improvement of thiosulfate stability in gold leaching, 2002(b) SME Annual Meeting, Feb.25-27, Phoenix, Arizona (75) Zhang, Wenge; L i , Yuling, Studies on leaching gold and silver from gold concentrates and silver pyrites associated with complex metals sulphides by ammoniacal thiosulfate, Nonferrous metals Nov. 1987, V o l . 39. No4 (in Chinese) 71-76 (76) Zheng,J.; Ritchie, I.M.; La brooy, S. R. and Singh, P., Study of gold leaching in oxygenated solutions containing cyanide-copper-ammonia using a rotating quartz crystal microbalance, Hydrometallurgy 39 (1995) 272-92 158 (77)Zhu, Guocai; Fang, Zhaoheng; Chen, Jiayong, Electrochemical studies on the mechanism of gold dissolution in thiosulfate solutions, Transactions of NFsoc Mar. 1994 Vol .4 No . 1 159 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0078636/manifest

Comment

Related Items