Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Oxidative pressure leaching of chalcocite in sulphuric acid Grewal, Ishwinder Singh 1991

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

Item Metadata

Download

Media
831-UBC_1991_A7 G73.pdf [ 6.03MB ]
Metadata
JSON: 831-1.0078569.json
JSON-LD: 831-1.0078569-ld.json
RDF/XML (Pretty): 831-1.0078569-rdf.xml
RDF/JSON: 831-1.0078569-rdf.json
Turtle: 831-1.0078569-turtle.txt
N-Triples: 831-1.0078569-rdf-ntriples.txt
Original Record: 831-1.0078569-source.json
Full Text
831-1.0078569-fulltext.txt
Citation
831-1.0078569.ris

Full Text

OXIDATIVE PRESSURE LEACHING OF CHALCOCITE IN SULPHURIC ACID by Ishwinder Singh Grewal B.A.Sc, The University of British Columbia, 1989 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Metals and Materials Engineering We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October 1991 © Ishwinder Singh Grewal, 1991 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 (V\gjbn\< \ - Mc \e*.qU Bvqc^e^nWj The University of British Columbia Vancouver, Canada Date O C T - M DE-6 (2/88) ABSTRACT At INCO's Copper Cliff Copper Refinery, a copper sulphide residue containing precious metals is subjected to a pressure leach at 115°C Copper occurs predominantly as G12S and Cu : %S. The leach produces a slurry of copper sulphate solution and basic copper sulphate and precious metals solids. The basic copper sulphate is dissolved in spent acid from the electrowinning tankhouse leaving a precious metals residue for further processing. The leach periodically develops a problem referred to as a "slow cook" where leaching times are greatly extended and/or incomplete leaching of copper is encountered. Based on earlier work, chalcocite leaching was proposed to occur sequentially via the following reactions: 1. Cw,S + //2S04+102 -» CuS + CuS<94 + H20 2. CUTS + CuSOA+^02+H20 ->CuS + Cu{OH\ • CuSOA 3. Cw1S+^02+H20 -*^CuS+^(2Cu(OH)2'CuSOt) 4. CuS + 202-+CuSOt 5. CuS+H^ + ^ -^CuSO^+H^ +S° The exact reaction path is determined by the initial solution conditions (copper sulphate and sulphuric acid concentrations). Under normal batch makeup conditions, all of the CU2S is oxidized to cupric ions and sulphate via reactions 1,2 and 4. If the solution becomes depleted in copper and acid, reaction 3 could occur. Elemental sulphur can be produced via reaction 5. Experimental studies showed that the reactions were nearly sequential. Reactions 1 and 2 were found to be very fast relative to the rate of reaction 4. No slow cook conditions were observed in the laboratory under normal leaching conditions. There is evidence suggesting that -ii-the slow cooks are caused by oxygen mass transfer limitations under conditions where the slurry becomes highly viscous and pseudoplastic due to formation of finely divided basic copper sulphate. -iii-Table of Contents ABSTRACT ii Table of Tables vi Table of Figures vii Acknowledgements xi CHAPTER 1 - Introduction 1 CHAPTER 2 - Background and Literature Review 4 2.1 Overview of the INCO-CRED Process 4 2.1.1 Background of the Second Stage Leach 7 2.1.2 The Second Stage Leach 10 2.2 The UBC Screening Model 12 2.3 Scope of the Study 21 2.4 Literature Review 22 2.4.1 Copper Sulphides - Chalcocite to Covellite 22 2.4.2 Leaching of Chalcocite and Covellite 23 2.4.3 Electrochemical Studies 30 2.4.4 E h -pH Relationships and Phase Systems 33 2.4.5 Gas-Liquid Mass Transfer in Oxidative Leaching 38 2.5 Summary 41 CHAPTER 3 - Experimental Methods 43 3.1 Part A: Study of the sequential nature of the reactions 44 3.1.1 Batch make-up chemistry 44 3.1.2 Experimental procedure 45 3.1.3 Additional experiments in Part A 46 3.2 Part B: Kinetic Experiments 47 3.3 Part C: Leaching CuS in the presence of basic copper sulphate 48 -iv-CHAPTER 4 - Results and Discussion 50 4.1 Part A: Study of the sequential nature of the reactions 50 4.1.1 The behavior of copper dissolution 50 4.1.2 Iron and Arsenic 58 4.1.3 Nickel and Cobalt 61 4.1.4 Additional Observations of Part A Experiments 63 4.2 Part B: Kinetic Experiments 63 4.3 Part C: Leaching CuS in the presence of basic copper sulphate 66 4.3.1 Discussion of ORP Measurements 71 4.4 Comparison of Leaching Rates - Part A and Part C 72 4.5 Additional Observations 73 CHAPTER 5 - Conclusions and Recommendations 76 5.1 Conclusions 76 5.2 Recommendations for further work 77 REFERENCES 79 APPENDIX A - Detailed Flowsheet of the CRED Plant 82 APPENDIX B - Planned Experiments for Part B-1 83 APPENDIX C - Assay results of Part A experiments 84 APPENDIX D - Part B experimental results 90 APPENDIX E - Part C experimental results 101 APPENDIX F - Part C experimental results - Effect of iron plots 107 -v-Table of Tables Table 2.1 Typical Assay of IPC Residue 4 Table 3.1 Assay of First Stage Residue used in Part A Experiments 44 Table 3.2 List of part C experiments performed 49 Table 4.1 X-Ray Diffraction Results 54 Table 4.2 Degree of sulphur oxidation 65 Table 4.3 Time taken to leach to 10,15 and 20% oxygen consumption 66 Table 4.4 Potential and pH measurements of the leach slurry after leaching and the approximate temperatures at which they were measured 69 Table 4.5 Comaparison of oxygen flow rates 73 Table Bl. List of the planned experiments for part B 83 Table Cl . Amount of the indicated species in the leach solution at various oxygen consumption levels '. 84 Table C2. Amount of the indicated species in the releach solution at various oxygen consumption levels : 85 Table C3. Amount of the indicated species in the releach cake at various oxygen consumption levels 86 Table C4. Total amount of species (calculated as a sum of tables C1-C3) in the process 87 Table C5. Distribution of species of experiments performed with the "copper depleted cake" 88 Table C6. The pH and ORP values of the slurry after leaching to a given level of oxygen consumption 89 -vi-Table of Figures Figure 1.1 Classification of leaching methods 2 Figure 2.1 General processing route of the IPC residue at the INCO-CRED plant in Copper Cliff 6 Figure 2.2 Second Stage Leaching Circuit at INCO's - Copper Cliff Copper Refinery... 10 Figure 2.3 Approximate shape and dimensions of the Second Stage Autoclaves 11 Figure 2.4 Distribution of species as predicted by model for case (i) conditions 16 Figure 2.5 Oxygen flow rates as predicted by model for case (i) conditions 16 Figure 2.6 Distribution of species as predicted by model for case (ii) conditions 18 Figure 2.7 Oxygen flow rate as predicted by model for case (ii) conditions 18 Figure 2.8 Distribution of species as predicted by model for case (iii) conditions 19 Figure 2.9 Oxygen flow rate as predicted by model for case (iii) conditions 19 Figure 2.10 Distribution of species as predicted by model for case (iv) conditions 20 Figure 2.11 Oxygen flow rate as predicted by model for case (iv) conditions 20 Figure 2.12 Crystal structures of copper sulphide minerals relevant to this study 22 Figure 2.13 Leaching morphology for a chalcocite particle (a)0-20% copper extraction 03)20-50% copper extraction (c) 50-100% copper extraction 26 Figure 2.14 Evans diagram of applicable polarization curves during oxygen pressure leaching of chalcocite 27 Figure 2.15a-b Potential-pH diagram for the Cu-S-H20 system 34 Figure 2.16a-b Thermal precipitation diagrams for the CUSO4-H2SO4-H2O system ....... 35 Figure 2.17 Thermal precipitation diagram for the CUSO4-H2SO4-H2O system at 100°C for 3cu2+ = aS04 2- 36 Figure 2.18a-b Phase diagrams of the Cu-S system. The blaubleinder covelUte is abbreviated as "be" 37 Figure 2.19 Models for oxygen adsorption during oxidative leaching 40 Figure 3.1 Schematic of the experimental setup 43 Figure 4.1 The distribution of copper during leaching of chalcocite 52 Figure 4.2 Similar to Figure 4.1 - no copper in solids shown to magnify scale 52 Figure 4.3 Comparison of the model results to the actual behavior 53 -vii-Figure 4.4 "Reaction" temperature versus pH^ change during precipitation from 1 M CuS04 solution. The "equihbrium" boundary for 1 M solution is also shown 56 Figure 4.5 Comparison of sulphur levels in the leach residue between the model and actual results 57 Figure 4.6 Distribution of iron during leaching. 59 Figure 4.7 SEM photomicrographps - Effect of iron on agglomerate size of the precipitate 60 Figure 4.8 Distribution of arsenic during leaching 61 Figure 4.9 Distribution of nickel during leaching 62 Figure 4.10 Distribution of cobalt during leaching 62 Figure 4.11 Oxygen consumption rate at various initial copper concentrations with [Fe]=0.25 g/L. Each curve represents a different level of oxygen consumption .- 67 Figure 4.12 Oxygen consumption rate at various copper concentrations with [Fe]=0 g/L. Each curve represents a different level of oxygen consumption 68 Figure 4.13 The effect of initial copper concentration on the measured ORP at [Fe]=0 and [Fe]=0.25 g/L 70 Figure 4.14 Evans diagram of applicable polarization curves during pressure leaching of CuS (schematic) 70 Figure 4.15 A schematic highlighting areas believed to be well mixed in the second stage autoclave. The poorly mixed zones are thought to be the result of the observed pseudoplastic behavior of the slurry 75 Figure A l . Detailed flowsheet of the CRED Plant 82 Figure DI. Rate of oxygen consumption for the condition where (x2,x3,x4)=(l,l,0) and xt=0 91 Figure D2. Rate of oxygen consumption for the condition where (x2,x3,x4)=(l,0,l) and x,=0 91 Figure D3. Rate of oxygen consumption for the condition where (x2,X3/x4)=(l/-l/0) and x,=0 92 Figure D4. Rate of oxygen consumption for the condition where (x2,x3/x4)=(l/0,-l) and x,=0 92 Figure D5. Rate of oxygen consumption for the condition where (x2/X3/X4)=(0,l,-l) and x1=0 93 Figure D6. Rate of oxygen consumption for the condition where (x2,x3/x4)=(0,0,0) and x1=0 93 -viii-Figure D7. Rate of oxygen consumption for the condition where (x2/x3/X4)=(0,l/l) and x,=0 94 Figure D8. Rate of oxygen consumption for the condition where (x2/x3,X4)=(0/-l/l) and x1 =0 94 Figure D9. Rate of oxygen consumption for the condition where (x2,x3,x4)=(0,-l,-l) andx^O 95 Figure D10. Rate of oxygen consumption for the condition where (x2/x3/X4)=(0,0,0) andx,=0 95 Figure DTI. Rate of oxygen consumption for the condition where ( x ^ x ^ M - l A L O ) 96 Figure D12. Rate of oxygen consumption for the condition where (x1,x2,x3/x4)=(-l,0,-l,0) 96 Figure D13. Rate of oxygen consumption for the condition where (x1/x2/x3^4)=(-l,0J3/-l) 97 Figure D14. Rate of oxygen consumption for the condition where ( x ^ x ^ M - l A O , ! ) 97 Figure D15. Rate of oxygen consumption for the condition where (x2/x3/x4)=(-l,0,l) and x:=0 98 Figure D16. Rate of oxygen consumption for the condition where (x2,x3,x4)=(-l/0,-l) andx,=0 98 Figure D17. Rate of oxygen consumption for the condition where (x2,x3/X4)=(-l,l,0) and x1=0 99 Figure D18. Rate of oxygen consumption for the condition where (x2,x3/x4)=(-l,-l,0) andx1=0 99 Figure D19. Rate of oxygen consumption for the condition where (x:/x2,x3/X4)=(-l,-l,0,0) 100 Figure El . Rate of oxygen consumption where initial [Cu]=0 g/L and [Fe]=0.25 g/L 102 Figure E2. Rate of oxygen consumption where initial [Cu]=80 g/L and [Fe]=0.25 g/L 102 Figure.E3. Rate of oxygen consumption where initial [Cu]=10 g/L and [Fe]=0.25 g/L > 103 Figure E4. Rate of oxygen consumption where initial [Cu]=40 g/L and [Fe]=0.25 g/L 103 - i x -Figure E5. Rate of oxygen consumption where initial [Cu]=l g/L and [Fe]=0.25 g/L 104 Figure E6. Rate of oxygen consumption where initial [Cu]=0 g/L and [Fe]=0 g/L 104 Figure E7. Rate of oxygen consumption where initial [Cu]=80 g/L and [Fe]=0 g/L 105 Figure E8. Rate of oxygen consumption where initial [Cu]=10 g/L and [Fe]=0 g/L 105 Figure E9. Rate of oxygen consumption where initial [Cu]=40 g/L and [Fe]=0 g/L 106 Figure E10. Rate of oxygen consumption where initial [Cu]=l g/L and [Fe]=0 g/L 106 Figure Fl. Rate of oxygen consumption for various initial copper concentrations at the 10% oxygen consumption point 108 Figure F2. Rate of oxygen consumption for various initial copper concentrations at the 20% oxygen consumption point 108 Figure F3. Rate of oxygen consumption for various initial copper concentrations at the 30% oxygen consumption point 109 Figure F4. Rate of oxygen consumption for various initial copper concentrations at the 40% oxygen consumption point 109 Figure F5. Rate of oxygen consumption for various initial copper concentrations at the 50% oxygen consumption point 110 Figure F6. Rate of oxygen consumption for various initial copper concentrations at the 60% oxygen consumption point 110 Figure F7. Rate of oxygen consumption for various initial copper concentrations at the 70% oxygen consumption point I l l Figure F8. Rate of oxygen consumption for various initial copper concentrations at the 80% oxygen consumption point Il l -x-ACKNOWLEDGEMENTS I would like to thank Dr. David Dreisinger for his constant encouragement and support and Dr. Ernest Peters for his thought provoking discussions and ideas throughout the course of this project. I wish to thank INCO; without their financial support and faculties, this project would not have been possible. Thanks are also extended to all of the people associated with INCO who helped in various ways in the completion of the experimental work. I wish to thank my mother who has always encouraged me to press on and has shown me the value of perseverance. I also wish to thank my wife for believing in me more than I believed in myself throughout this endeavor. And a final thanks is extended to the Cy and Emerald Keyes Foundation for their financial support. -xi-CHAPTER 1 -Introduction  Copper is one of the less abundant base metals found in the earth's crust occurring at levels of approximately 7 ppm (compared to aluminum and iron at approximately 80 000 and 60 000 ppm respectively) [1]. Approximately 90% of the world's supply of copper occurs as sulphidic ores. Pyrometallurgical techniques have historically been the dominant processing method. However, pollution problems associated with sulphur dioxide emissions from pyrometallurgical operations has resulted in a considerable research effort in hydrometallurgical processing of copper and other sulphide ores. Some of the main advantages that hydrometallurgical processes offer are that: 1. Hydrometallurgy allows the processing of complex ores with multiple recoverable metals. By controlling solution conditions, it is possible to recover various metals in separate unit operations. These metals can be sold for additional revenues. 2. Hydrometallurgical operations are performed at lower temperatures and generally use less energy compared to the high temperatures often employed in pyrometallurgical operations. This is especially true for low grade ores. 3. Hydrometallurgy has often been found to be more economically viable in the treatment of low grade ores especially if the crushing and grinding steps can be minimized as in percolation leaching methods. 4. Hydrometallurgical operations produce little or no air pollution. The liquid waste generated at hydrometallurgical plants is often easier to contain and treat than effluent gases. 5. Solutions and slurries in hydrometallurgical plants are easily transported by pipeline systems as opposed to moving of molten slags and mattes between furnaces using heavy refractory ladles in pyrometallurgical processes. -1-Successful hydrometallurgical processes for copper extraction from sulphide concentrates have been proposed and designed but they have not been able to compete commercially with pyrometallurgical processes. Although hydrometallurgical processes have demonstrated many advantages as listed above, they are not a panacea for extractive metallurgy. But hydrometallurgy does have an important role to play in the treatment of specific ores such as low-grade ores, complex mineral ores and secondary materials produced from other pyrometallurgical and hydrometallurgical operations. The most common hydrometallurgical step is the leaching process which serves to free the desired constituents from the gangue material via dissolution. In general, leaching methods can be classified into percolation leaching and agitated leaching (see Figure 1.1). The method used depends upon the nature of the ore and the mineral deposit. f In-situ Leaching Percolation Leaching t Heap or Dump Leaching Leaching Method t Vat Leaching Agitated Leaching Thin Layer Leaching Slime (Pulp) Leaching Pressure Leaching t Baking Process Figure 1.1 Classification of leaching methods [2]. -2-The leaching reagents used must dissolve the ore minerals as rapidly as possible and be substantially inert towards the gangue minerals. Extensive reaction with the gangue minerals causes excessive reagent consumption and causes the solution to become fouled with impurities. The reagent must also be readily available and as inexpensive as possible. Leaching in the presence of sulphuric acid is one of the most common methods in leaching copper sulphides. Sulphides are insoluble in dilute sulphuric acids but can be solubilized if oxidizing species are present in solution. These oxidizing agents include oxygen, sulphate and chloride salts of iron and copper and aqueous chlorine (ie. hypochlorous acid and hypochlorite). Acids, present as concentrated solutions, are sometimes also powerful oxidizing agents. Increasing temperature and oxygen pressures have been found to contribute significantly to the rate of copper extraction. One process for the leaching of copper from a sulphide residue is the CRED1 Second Stage Leach at INCO's Copper Cliff Copper Refinery. The second stage leach processes secondary material, mostly CujS, generated from a preceding metathetic leach. The objective of this thesis is to investigate the poor leaching behavior that has occasionally been encountered in this process. The problem of slow copper leaching kinetics has existed for approximately 15 years occasionally becoming severe enough to warrant investigation. The research work contained in this thesis was designed following some initial mathematical modelling work done by Dreisinger and Peters [3] at U.B.C. which pointed to a possible metallurgical explanation for the poor leaching behavior. This thesis is organized in the following way. Chapter 2 contains a brief literature survey on the leaching behavior of copper sulphides. Chapter 2 also contains the investigations done in the past by INCO and the model developed at U.B.C. Chapter 3 covers the experimental methods and chapter 4 contains the results and discussions. In chapter 5, some conclusions and recommendations are offered. 1 CRED - Copper Refinery Electrowinning Department -3-CHAPTER 2 - Background and Literature Review 2.1 Overview of the INCO-CRED Process INCCs Copper Cliff Nickel Refinery extracts nickel via the INCO Pressure Carbonyl (IPC) process. This process produces a residue that contains mostly copper along with the other constituents shown in Table 2.1. Approximately 50 tons of this residue are processed daily in a hydrometallurgical plant (CRED-Copper Refinery Electrowinning Department) at the Copper Cliff Copper Refinery. The purpose of the CRED plant is to separate the various constituents of the IPC residue through a number of hydrometallurgical operations (see Figure 2.1) [4]. The process steps include pressure leaching, cementation, precipitation, thickening, filtration and electrowinning. A detailed process flowchart is provided in Appendix A. Table 2.1 Typical Assay of IPC Residue [4]. Weight Percent Oz/Ton Cu Ni Co Fe S Se Te PGM + Au Ag 55-60 6-10 4-8 4-9 13-19 0.06-0.10 0.06-0.10 20-30 25-45 The IPC residue is first treated via a metathetic pressure leach in sulphuric acid (100-200 g/L) and copper sulphate (40-90 g/L) solution at 150°C. This batch process, referred to as First Stage Leaching, is used to dissolve nickel, cobalt and iron and separate these metals from copper, selenium, tellurium and precious metals which remain in the solid phase. The overall reactions taking place in the leaching process are: MeO(s) + H2SOA = MeS04+H20 Me(s) + CuSOt = MeSOA + Cu(s) -4-MeS(s) + CuSOA = MeSOA + CuS(s) where Me =Ni,Co,Fe Approximately 95-98% of the base metals, other than copper, are leached out of the solids in this step. The copper entering the first stage leach is present predominantly in the form of CujS (chalcocite) and passes through to the second stage leach unmodified. The first stage leach slurry is filtered. The filtrate is put through a copper clean-up circuit to remove some copper which still remains in solution after the leach. The first stage leach residue is treated in a total oxidative pressure leach at 115°C. This step is referred to as Second Stage Leaching. The first stage cake obtained from the filtration step is combined with water and spent electrolyte from the plant to produce a slurry of approximately 30% solids. This slurry is charged into the second stage autoclaves. The chalcocite is batch-leached to form a slurry of neutral copper sulphate (CuS04) solution (pH of -2.5-3.0) and basic copper sulphate (CuS04-2Cu(OH)2) solids. The slurry is then mixed with spent electrolyte to dissolve the basic copper sulphate to leave a residue containing precious metals and lead sulphate. Selenium, tellurium and most of the base metals are also solubilized during the leaching process. The unleached solids are filtered out for processing at another INCO plant to recover the precious metals. The solution is treated for selenium-tellurium removal and then pumped to the tankhouse for copper recovery via electrowinning. Selenium and tellurium removal is essential because these impurities tend to co-deposit with the copper during electrowinning and contaminate the cathodes. The removal of Se and Te is achieved by heating the solution to 95 °C and passing it through a column filled with copper shot. This promotes the formation of selenide and telluride precipitates that form as fine black particulates. The solution and solids are passed through four aging towers in series in which the solids settle out. The Se and Te concentrations are reduced to less than 1 mg/1 in solution. The overflow from the aging towers is passed through polishing filters and sent to the electrowinning circuit. -5-PM Residue Spent Electrolyte Sulphuric Acid Oxygen . . Second Stage Pressure Leaching First Stage Cake IPC Residue from INCO Pressure Carbonyl Process ^ First Stage Pressure Leaching Spent Electrolyte First Stage Filtrate NaSH ' Filter Aid Copper Clean-up CuS to Second Stage Pressure Leaching Se, Te Residue Filtrate Selenium, Tellurium Removal CuShot Filter Aid Steam Sulphuric Acid Steam Oxygen Lime Filter Aid Filtrate Iron/Arsenic Removal Iron hydroxide-Gypsum Solids to Effluent Filtrate Copper Electrowinning Titanium Blanks Lead Anodes Reagents Water/Steam Soda Ash Steam Filtrate Nickel-Cobalt Recovery Copper Cathodes Spent Electrolyte Nickel/Cobalt * Carbonate Vacuum Bosh Pond Figure 2.1 General processing route of the IPC residue at the INCO-CRED plant in Copper Cliff The filtrate from the first stage leach is processed through a copper clean-up circuit to remove any copper which exists in solution in the form of copper sulphate. The removal is necessary to prevent copper losses to the effluent during the iron/arsenic removal step and to prevent copper contamination of the nickel/cobalt carbonate. The copper is removed by the addition of a 30% NaHS solution at 70 °C. Most of the copper is precipitated as CuS. The process is controlled so as to prevent the evolution of H2S gas. A thickener is used to thicken the CuS precipitate. The solids are returned to the first stage filters and the overflow solution is sent to the iron/arsenic removal circuit. -6-The objective of the iron/arsenic removal circuit is to produce environmentally stable compounds of iron, arsenic and sulphur and discard them to the effluent stream without excessive loss of nickel and cobalt. The feed solution to this circuit contains 0.02 g/L copper, 25-35 g/L nickel, 15-25 g/L cobalt, 20-35 g/L iron, 1-2 g/L arsenic, and 40-70 g/L sulphuric acid. It is processed continuously at a rate of 150 L/min. Lime slurry is added to the solution as it is passed through two autoclaves operating under oxygen pressure at 90-95 °C. The iron and arsenic precipitate out and the solids are separated by filtration to produce an iron-gypsum cake. The filtrate is sent to the nickel/cobalt recovery step. The iron-gypsum cake is partially redissolved to remove co-precipitated nickel and cobalt and then filtered. The filter cake is sent to the tailings and the filtrate is returned ahead of the iron precipitation circuit. The filtrate from the iron removal circuit is mixed with a 200 g/L solution of sodium carbonate (Na2C03) in two reaction vessels in series. The pH is controlled in the ranges 7.6-7.8 and 8.1-8.3 in the two vessels respectively. The reaction product is a precipitate of basic nickel and cobalt carbonates which is thickened to 20-25% solids. The nickel/cobalt carbonate is shipped to the cobalt refinery and the barren solution is pumped to the waste pond. 2.1.1 Background of the Second Stage Leach When the CRED plant first became operational in the early 70's, the second stage leach was designed to leach CU2S completely in the presence of excess acid (H2SO4) and under oxygen pressure at 110°C. The products of the leaching process were CuS04(aq) and elemental sulphur. The equation governing the process was reported as [5]: Cu£ + 2H£Ot + 02 -> 2CuSO, + 2H20 +S° The leach products typically analyzed 90% elemental sulphur and 10% precious metals and unreacted sulphides. The factors affecting the reaction rate were determined to be the iron content of the solution, oxygen pressure, temperature and feed particle size. The acid and copper concentrations were not found to be critical factors in the leaching rate. The iron -7-content of the solution was identified as being an important variable. With iron in solution, the reaction proceeds rapidly according to the reaction above to form elemental sulphur. However in the absence of iron, the reaction proceeds more slowly and consumes up to 4 times as much oxygen and the sulphide sulphur is oxidized to sulphate according to the reaction: CifjS + H2SOA+\o2^> 2CuS04+H20 The leaching process was operated by this method for a few years. However, in the mid-seventies, historical reports [6] show that a proposal was made to change the leaching process. It was suggested that all the sulphide sulphur should be oxidized to form sulphate as opposed to elemental sulphur by leaching in a low-acid solution. Based on laboratory work, the product from this leaching method would produce a precipitate of basic copper sulphate (CuS04-2Cu(OH)2) in a solution of copper sulphate. The slurry would have to be mixed with spent electrolyte containing sulphuric acid to dissolve the precipitate and leave a residue containing precious metals and a small amount of gangue materials. The operating conditions of the process were set at 150 psi oxygen pressure at 105'C. The batch feed to the autoclave would be a mixture of spent electrolyte containing sulphuric acid, copper and iron in solution with first stage residue to form a slurry containing -40% solids by weight. The process was eventually changed to a total oxidative leach in 1975 with some modifications being continually made to improve the process. Howver, the second stage leach residue occasionally showed high levels of copper still remaining in the solids even after extended leaching times [7]. Extensive examination of IPC residue, first stage residue and poorly leached second state leach residue was carried out to determine the nature of the poor leaching behavior. The results showed that the presence of Cu 20 in the feed to second stage, associated with high levels of oxygen in the IPC residue, was linked to the poorly leached batches. The presence of Cu 20 was proposed to cause an adhering film of basic -8-copper sulphate at the point of basic copper sulphate formation. This would block oxidizing species from reaching the copper sulphide particles and hinder further leaching. Photomicrographs were presented to support this theory. During 1986, slow leaching and high copper in the residue became a severe problem and warranted further investigation of the process [8]. A complete leach should nominally take 5 hours based on plant reports and experimental tests. However, the length of leach times almost always exceeded 5 hours with the occasional leach time taking longer than 20 hours. For example, from plant data for the month of January 1988 [9], out of 93 leaches, 33% of the leaches exceeded 8 hours. A calculation in one report [10] gives the cost of downtime as approximately $25 000 /hr for deferred revenues from precious metals. It is apparent from this calculation how even small improvements in leaching times can make a significant difference to the revenues over one year. One laboratory study [11] of the second stage leach shows some interesting results on the behavior of the process with respect to pulp density, particle size of precipitate, and agitation speed. It was found that higher solids density feeds lowered the leaching rate of copper considerably due to higher viscosity of the solution. In lab tests with normal solids densities, higher viscosities were observed on the material which leached poorly in the plant as opposed to material which leached quickly. The higher viscosities observed in this case were associated with a finer particle size of the precipitate. The agitation speed of the slurry also had a significant effect on the leaching rate. All these factors appear to indicate that oxygen dispersion in the autoclave is severely affected by changing viscosities and agitation speed. The autoclaves have no level meter, so the impeller depth varies considerably between leaches. Impeller depth was suggested to be an important parameter in the leaching rate in one study and so the slurry level was lowered (amount unknown) and was found to improve the leaching process for a short while. However, this did not cure the problem permanently. -9-2.1.2 The Second Stage Leach The equipment used in the second stage leaching process consists of a batch make-up tank/2 titanium autoclaves, a dissolving tank, 2 pressure filters and a rotary vacuum filter (Figure 2.2) [4]. There are also product holding tanks between various stages of the process. PM residue slurry storage Figure 2.2 Second Stage Leaching Circuit at INCO's - Copper Cliff Copper Refinery [4]. The autoclaves were originally designed for leaching in excess acid as described earlier but were modified in the mid-seventies when the process was changed. The major change to the autoclaves was the installation of vertical cooling coils around the inside -10-perimeter of the vessels (see figure 2.3) [12]. The agitator system consists of two 45' pitched-down, 4 bladed turbine impellers attached to a single 18 cm diameter shaft rotating at 68 rpm. The internal parts are made out of 316L stainless steel. Oxygen Sparger Figure 2.3 Approximate shape and dimensions of the Second Stage Autoclaves [12]. -11-The feed is prepared as a 30% solids slurry by mixing spent electrolyte from the electrowinning plant, water and first stage residue in the batch make-up tank. The slurry is then pumped into one of the second stage autoclaves. The autoclave is then pressurized with oxygen to 150 psi while the slurry is being agitated. The temperature is maintained at 115°C during the leach. The leaching process is assumed to be complete when there is no more oxygen consumption, no more heat being generated and there is no temperature increase when cooling is off. The slurry is then pumped to the redissolving tank where it is mixed with spent electrolyte and filter aid. The acid in the spent electrolyte dissolves the basic copper sulphate. The unleached solids are then recovered by pressure filtration and the filtrate is sent to Se/Te removal. The solids are repulped in water and refiltered to recover the precious metals residue. The final residue is shipped to Port Colborne for precious metals recovery. 2.2 The UBC Screening Model The studies in the past have been inconclusive in determining the cause of the long leaching times and incomplete leaching in the second stage autoclaves. A mathematical model of the CRED second stage leach was developed at UBC by Dreisinger and Peters [3] in an attempt to evaluate possible metallurgical causes of the "slow cook" conditions occurring in the process. This screening model was a first approximation to possibly highlight some of the conditions that may lead to slow cook conditions. The model was developed on a number of important assumptions based on work previously done by Peters and Mao [13] on the leaching of Cu2S under slightly acidic conditions. Their work produced the following results: i. C u ^ leaches very quickly to CuS. ii. The CuS produced by leaching Cu2S tends to fracture and become finely disseminated. iii. C u ^ conversion to CuS proceeds very quickly relative to the leaching of CuS. iv. A small amount of iron in solution promotes elemental sulphur formation during leaching. The leaches with no iron showed little or no elemental sulphur in the leach residue. -12-Since CuS leaching proceeds slowly relative to Cu2S conversion to CuS, it is possible to separate the leaching steps of Cu2S conversion to CuS and CuS leaching to CuSGv In developing the model, a number of discrete leaching reactions were proposed which would proceed one at a time depending on the prevailing solution conditions and the character of the unreacted solids. These reactions are: 1. CuxS +{x- l]H2SO<+^(x- l)02-*CuS + (x- \)CuS04 + (x- \)H20 2. CuxS +^(x - WuSOt + ^ ix - l)02 + 2(x - \)H20 -> CuS +^(x - 1)2CM(0//)J « CuS04 3. CuxS +^(x - \)02 + (x -\)H20 -> ^ (3-x)CuS +^(x- l)(2Ca(0//), • CuSOj 4. CuS + 202-+CuSOt 5. CuS+H2SOi+^02^,CuSOi+H20 + S° where Cu,S refers to the average Cu/S mole ratio in the feed, not to any particular mineral form and the value of is approximately 2. There are five reaction pathways possible under various initial batch recipes. i. Begin with copper sulphate and acid in solution. Reaction 1 proceeds until acid depletion. Reaction 2 proceeds until only CuS is left in the residue Reaction 4 proceeds to total oxidative endpoint. ii. Begin with copper sulphate in solution and no acid. Reaction 2 proceeds until only CuS is left in the residue. Reaction 4 proceeds to total oxidative endpoint. iii. Begin with copper sulphate and excess acid in solution. , Reaction 1 proceeds until only CuS is left in the residue. Reaction 5 proceeds to acid depletion. Reaction 4 proceeds to total oxidative endpoint. -13-iv. Begin with acid and low copper in solution. Reaction 1 proceeds until acid depletion. Reaction 2 proceeds until copper sulphate is depleted. Reaction 3 proceeds until only CuS is left in the residue. Reaction 4 proceeds to total oxidative endpoint. v. Begin with low copper and no acid in solution. Reaction 2 proceeds until copper sulphate is depleted. Reaction 3 proceeds until only CuS is left in the residue. Reaction 4 proceeds to total oxidative endpoint. The kinetic routine in the model was divided into a two-step process. The first step is gas-liquid mass transfer defined by the following equation, d[02] - ^ - = *,([OJ -[Oz]) where kg is the gas-liquid mass transfer coefficient for the system [OJ" is the saturated oxygen cone, in solution at the oxygen partial pressure in the autoclave [02] is the oxygen concentration in the bulk solution The next step is controlled by chemical reaction defined by the following empirical equation: = k^Cu2*] [OJ + ^[CM 2 +] [OJ [CuxS] where kt and k2 are empirical rate constants. The equation is defined on the basis of copper catalysis which is first order in cupric and dissolved oxygen concentrations and a solids leaching term which is first order in cupric, dissolved oxygen and unreacted CuJ$ concentrations. The first term recognizes the significant role of copper ion catalysis and the second term accounts for the fact that the rate will drop as the solids concentration goes to zero. -14-The kinetic equations suggest that the amount of copper in solution at any given time is very important to the rate and that lower copper concentrations will cause slower leaching rates. If the copper in solution is depleted, the rate is expected to go to zero. This is the anticipated slow leaching behavior observed in the plant known as the "slow cook" condition. In this preliminary model, the rate constants were treated as fixed values because of the unavailability of additional data. These values probably change due to changing conditions during leaching. Figures 2.4-2.11 show some sample outputs of the model generated from various initial concentrations of acid and copper in solution. It can be seen that the slow cook condition occurs only when there is a depletion of copper in solution. Figure 2.4 shows the model output from a case when there are acid and copper in solution at a level which will cause the reactions to proceed as described in case (i) above. The output can be split into 3 sections. In the first section, according to reaction 1, acid is being consumed to leach copper from Cu£ and the copper concentration is increasing in solution. When the acid is depleted, at around the 16-minute mark, the process proceeds according to reaction 2 and produces basic copper sulphate from CuxS and CuS04. The copper concentration decreases as shown. Eventually, when all the Cu^ S is transformed to CuS, the leaching proceeds according to reaction 4 until all the CuS is leached. Figure 2.5 is the corresponding oxygen consumption rate curve. This shows that the initial rate of reaction rises very quickly as the amount of copper in solution rises and the rate drops as the copper is depleted via reaction 2. The reaction rates never drop to low levels because there is always a lot of copper in solution available for catalysis. The complete leach time is approximately 5 hours and is close to the times observed in plant operations for a normal leach. Figures 2.6 and 2.7 represent case (ii) in which the there is no acid available at the beginning of the leach and the initial copper concentration is approximately 100 g/L. The -15-320 -280 -240 -^ 200 -c o •a s 160 -c -120 -8 80 -40 -0 -Concentration Profiles Predicted by Model _ \ \ \ \ C ux s V \ V \ \ \ 2Cu(OH)2.CuS04  Y / \ / / / ; Cu N W -H2S04 1 1 1 I 1 1 1 I _ _ — " " ~ r — — 40 80 120 160 200 240 280 Time (min) Figure 2.4 Distribution of species as predicted by model for case (i) conditions. 0.04 0.035 Oxygen Flowrate Predicted by Model 280 Time (min) Figure 2.5 Oxygen flow rate as predicted by model for case (i) conditions. -16-model predicts a higher amount of basic copper sulphate formation via reaction 2 until all Cu,S is converted to CuS. The kinetics are good in this case because the copper in solution never drops below 50 g/1. Figures 2.8 and 2.9 shows the effect of excess acid in solution at the beginning of leach. Because there is still acid present after all the CuxS has been converted to CuS, sulphur formation is expected via reaction 5 and no basic copper sulphate is produced. Near the end of the leach process, the solubility of copper sulphate is exceeded resulting in the precipitation of CuS04.5H20. The model was operated under conditions which would cause a copper deficiency during operation. This was done by setting a high copper to sulphur ratio in the feed solids and less acid in the feed solution. The resulting output is shown in Figures 2.10 and 2.11. The acid and copper in solution drop very quickly and the result is a severe drop in the leaching rate. It is also important to note that a significant amount of basic copper sulphate solids is produced. This high amount of solids could cause a significant drop in gas-liquid mass transfer although the model does not incorporate the effect of solids loading into the gas-liquid mass transfer rate. The total leach time is predicted to exceed 25 hours. The model predicts slow leaching conditions under copper depleted conditions but the model requires verification. Firstly, the sequential reaction chemistry proposed in the model needed to be verified through experimental work. Secondly, better kinetic relationships need to be developed because the model used an empirical relationship for the rate of reaction. The objective of this thesis project was to obtain experimental data to improve the understanding of the leaching process and eventually to develop a better model. -17-c o 03 4= C o o 280 240 200 160 120 80 Concentration Profiles Predicted by Model r i i \ I \ C"xS / ^ 2Cu(OH)2.CuS04 / \ „ / / " " ^ — " Cu 1 \, 100 200 Time (min) 300 400 Figure 2.6 Distribution of species as predicted by model for case (ii) conditions. Oxygen Flowrate Predicted by Model 200 Time (min) Figure 2.7 Oxygen flow rate as predicted by model for case (ii) conditions. -18-320 -i 280 -240 -i 200 -ritratiot 160 -ncei 120 -8 80 -40 -0 -Concentration Profiles Predicted by Model \CujS Cu CuS04.5H20 Time (min) Figure 2.8 Distribution of species as predicted by model for case (iii) conditions. 0.05 Oxygen Flowrate Predicted by Model „ 0.04 c ! o £ 0.03 -s f£ 0.02 § 0.01 0 I i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i -0 20 40 60 80 100 1 20 140 160 180 200 Time (min) Figure 2.9 Oxygen flow rate as predicted by model for case (iii) conditions. -19-c o s c 500 450 400 350 300 250 200 150 100 50 0 Concentration Profiles Predicted by Model \ i I ^ 2Cu(OH)2.CuS04 C U x S Cu 'T— f— r— T—— r—T——r—T 1 500 1000 T I I 1 1500 Time (min) Figure 2.10 Distribution of species as predicted by model for case (iv) conditions. Oxygen Flowrate Generated by Model 0.04 -n — 0.035 -.g 0.03 -| 0.025 -| 0.02 -o c 0.015 -I I" 0.01 -0.005 -0 " I I I \ 1^  I I I I I I I I I 1^  0 500 1000 1500 Time (min) Figure 2.11 Oxygen flow rate as predicted by model for case (iv) conditions. -20-2.3 Scope of the Study A review of the history of the second stage leached has shown that a number of variables have been considered in trying to improve the leaching process. Clearly, the dificulties with the process have not been satisfactorily overcome. The most recent attempt at analyzing the problem was the UBC screening model discussed earlier. The present study was designed based on the results of the UBC screening model and the earlier work done at INCO labs. The main objetives of this study are: 1. to understand the reaction chemistry and check if the reactions are indeed sequential as suggested in the UBC screening model. 2. to identify any intermediate copper sulphide products formed. 3. to investigate chemical kinetics of copper leaching for the first stage residue cake. 4. to provide a basis for further work to improve the present mathematical model or develop a new mathematical model. -21-2.4 Literature Review 2.4.1 Copper Sulphides - Chalcocite to Covellite One of the most important economic copper minerals, chalcocite (CujS), is produced by the reduction of CuS04 solutions descending from oxidation zones of copper rich deposits in the earth [14]. Covellite (CuS), similarly, is usually found as an oxidation product of chalcocite or other primary copper sulphides like chalcopyrite as a zone of secondary enriched copper deposit. The oxidation of chalcocite does not lead to the direct formation of covellite. The decomposition process produces many intermediate sulphides such as djurleite (Cu196S), digenite (Cu, 7 6., g^ S), blue-remaining covellite (Cu,,.14S) and covellite (CuS). Potter [15] has shown the existence of numerous intermediate phases and provided free energy data for these phases. These phases are C u ^ S , Cu, o^S, Cu, 7 6 5S, Cu, 4S, Cu, ,S and CuS. Figure 2.12 shows the crystal structures of chalcocite, covellite and digenite. Chalcocite (hexagonal) Digenite (cubic) Covellite Figure 2.12 Crystal structures of copper sulphides relevant to this study [14]. -22-In chalcocite and djurleite, the sulphur species are arranged in a hexagonal close packed structure [14]. The copper ions are located near the triangular faces of the tetrahedral sites as opposed to being in the center of the tetrahedral interstices. The structure of digenite is a cubic close packed arrangement of the sulphur species and the copper ions are located off-center in the tetrahedral interstices. One eighth of the tetrahedra are unoccupied [14]. Natural digenite has 1 at.% iron to maintain a stable solid solution composition but a stable iron-free "low digenite" called anilite (Cu^S) is also found. The structure of covellite is more complex as can be seen in Figure 2.12. The base structure consists of three layers of hexagonal close packed sulphur species. The copper ions occupy the centers of the equilateral triangles and the centers of tetrahedral sites in the layers. 2.4.2 Leaching of Chalcocite and Covellite The leaching of chalcocite has often been reported as a two-step process according to the following reaction sequence: Cu** -> CuS + Cu1++ 2e~ CuS ->Cu2+ + S° + 2e~ A variety of leaching processes have been investigated in the laboratory and classified according to the leaching steps and the types of reagents used. The most common oxidizing agents are: Ferric sulphate in acid Ferric chloride in acid Oxygen in sulphuric acid Oxygen in ammoniacal solutions Nitric acid -23-There are other oxidizing agents but they have not received much attention, mostly because of their cost and commercial availability. Sullivan [16] studied the leaching chemistry of chalcocite in acidified ferric sulphate solutions using bottle roll tests at temperatures below 50 °C The dissolution process was found to occur in two steps. The first step was rapid until about 50% copper dissolution and the second step, oxidation of CuS, was relatively slow. The two reactions involved in the leaching process were reported as: CujS + Fe2(SOJ3 -> CuSOA + 2FeSOA + CuS CuS + Fez(SO,\ -> Cu504 + IF eSOA + S° The study showed that the dissolution rate was independent of the strength of ferric sulphate provided sufficient reagent was present. At a constant ferric concentration, the rate was also independent of acid strength. Various particle size fractions in the range of -10-mesh to 200-mesh were also studied. Although there is considerable difference in surface area per unit weight in this range, there was almost no difference in the leaching rates (ie. time taken to dissolve a given amount of copper) observed, provided that the particles were open to solution attack. The rate of dissolution was greatly affected by temperature. For example, 73% copper dissolution required 1, 5 and 15 days at 50 °C, 35 °C, 23 °C respectively. Sullivan reported no information on intermediate copper sulphide phases. Thomas et al. [17] examined the kinetics of dissolution of synthetic chalcocite and digenite in acid ferric sulphate solutions using a rotating sintered disc technique. The study found that digenite and chalcocite dissolved at similar rates. The dissolution was reported to occur in stages where chalcocite is progressively converted into djurleite, digenite, blaubleibender covellite (also known as blue-remaining covellite) and normal covellite. The covellite is transformed to elemental sulphur according to the second of two equations above. The rate of dissolution was found to be directly proportional to ferric ion concentration between the ranges tested (0.025 M - 0.2 M). This ferric ion concentration -24-dependency was not found in the studies by Sullivan. This difference is probably due to the differences in experimental methodology. The rate was also found to depend significantly on the temperature. Mao and Peters [13] studied the leaching of chalcocite under autoclave conditions and found the same two-stage leaching behavior reported in ferric sulphate leaching. During the first stage, 50% of the copper is extracted and up to 100% of the chalcocite is converted to covellite. The first stage is also separated into two steps which involves the initial conversion of chalcocite to digenite and then subsequent conversion of digenite to covellite. A leaching model to explain the leaching kinetics was based on three parts shown in Figure 2.13. The first step is a shrinking core kinetics model, the second includes particle break-up and third is the effect of elemental sulphur morphology on kinetics. The leaching process is described as a mixed-potential electrochemical model in which the first stage kinetics are predominantly cathodically controlled. The presence of iron in solution leads to higher leaching rates and decreases sulphur oxidation during the second stage. The second stage kinetics are explained by the passivation of covellite by oxygen leading to a high mixed potential. The Evans diagram in figure 2.14 shows a schematic of the applicable polarization curves. Depassivation occurs in the presence of Fe2+ ions where the process operates at point D and leads to a higher exchange current (leaching rate) and a lower mixed potential. -25-First Step: Stage I Leaching Cathodic: 02+4H++4e -^WiO (Cu£ & Cu^ Surface) Anodic: CUTS -> CKgS + 0.2Cu2+ + OAe (Cu^S only) Volume Reduction: 7.6% Second Step: Stage I Leaching Cathodic: 02 + 4H+ + 4e Anodic: CU2S -» CulJSS + 0.2Cw2++OAe (CiijS only) Cumulative Volume Reduction: 24.4% 2H20 (CuS Surface) Stage II Leaching Catnodic: 02+4H++4e^2H20 (CuS Surface) Anodic: CuS-*Cu2++S° + 2e CuS + 4H20 -> Cu++SOl'+ 8/T + Se Cumulative Volume Reduction: 43.6% (or more depending on degree of sulphur oxidation) Figure 2.13 Leaching morphology for a chalcocite particle (a) 0-20% copper extraction Ob) 20-50% copper extraction (c) 50-100% copper extraction [13]. -26-Otfluston - l \ limited Otygen i ^ hi Fww i \ 0! — n _ A - B First SleO- Stag« I teacftatg C Secofic Sm. Slage I teaching D Slags II Leaching (Fa * • present) £ Stage 1  Leaching(Fe' • absent] Legl-Cabin* Cumrtt [lenemalk) Figure 2.14 Evans diagram of applicable polarization curves during oxygen pressure leaching of chalcocite [13]. Oxygen pressure leaching experiments in an iron containing solution done by Chmielewski and Charewicz [18] show that the partial pressure of oxygen governs the process rate. The oxygen increased the kinetics mainly by oxidizing the ferrous iron to ferric iron, the main leaching agent, and not directly by interaction with the copper mineral. King et al. [19] studied the leaching of chalcocite by acidic ferric chloride solutions and found the same two stage leaching process observed in other studies. The first stage of the reaction, to approximately 50% copper dissolution, was complete in less than 4 minutes at temperatures between 40°C and 80°C. However, the second stage of leaching was strongly affected by temperature as the kinetics were much more rapid at higher temperatures. The apparent activation energy, Ea, for the first stage and second stage was 3.43 kj/mol and 101-122 kj/mol respectively. This difference in Ea was attributed to a difference in the -27-charge transfer process between the two reactions. The first stage is probably controlled by diffusion of copper ions in the particles and the second stage is chemically controlled by the reaction of S2' ions in CuS to form sulphur. The first stage of the process proceeded more slowly with the larger particle sized material due to a larger diffusion layer for the copper to travel through in the solid. The second stage was less affected by particle size. There was no effect of acid (HCl) strength on the leaching rate. An increase in ferric concentration up to 0.25 M produced an increase in the dissolution rate of the first stage but was independent of concentration beyond this point. The addition of ferrous chloride to ferric chloride solutions showed the exact same net increase that would have been found if the same amount of ferric chloride had been used instead. Use of ferrous chloride alone (ie. no ferric chloride) produced slow leaching. Particle size fractions tested were 425-600 um, 150-300 um, and 75-106 um. These size fractions were similar to the ones Sullivan studied. There was almost no difference observed in the leaching rates, a result similar to that of Sullivan. Particle sizes of 1.18-1.70 mm and 2.36-4.76 mm produced much slower leaching rates. These were attributed to a larger diffusion distance required by the copper to travel in the solid. The most interesting observations of this study were based upon the X-ray diffraction data. The results show that a whole range of intermediate non-stoichiometric copper sulphide phases are formed as copper is leached out of the solid matrix. The basic chalcocite crystal structure does not change until the copper level is below Cu1 - g 9 1S which is outside the digenite stoichiometric range. There appears to be a similar behavior as digenite transforms to covellite. This could be caused by some local areas and particles becoming more depleted of copper and achieving compositions at which phase transformations occur before others. This would explain the mixtures of phases observed. The leaching of chalcocite and covellite was studied by Grizo et al. [20] at pH values between 0.7 and 2 in the presence of sulphuric acid and ferric sulphate. They divided the leaching process into three stages. The three stages were identified by changes in kinetics -28-from linear to non-linear and back to linear dissolution rates. This is unlike previous studies which divided the stages according to the formation of intermediate species such as digenite and covellite. They do, however, suggest that during the second stage, leaching of chalcocite, digenite and covellite is occurring in parallel. The activation energies were found to progressively increase in each stage but the rise in activation energy is higher between the second and the third stage. This increase in activation energy suggests a change in mechanism from diffusion control to chemical kinetics control. An increase in particle size decreases the rate at which copper is leached but does not affect the kinetic mechanism in the three stages. The leaching data on various particle size fractions were also found to show that the second stage was controlled by the diffusion of species through a product layer. Cheng and Lawson [21] investigated the leaching of synthetic chalcocite and covellite in oxygenated acidic sulphate-chloride solutions. The leaching was described in terms of a shrinking core model with the rate being surface chemical reaction controlled in the first and second stages. The late second stage was accompanied by pore diffusion control. Elemental sulphur formation on the surface of the particles was found to retard the dissolution rate during covellite leaching. Thomas and Ingraham [22] studied the kinetics of dissolution of synthetic covellite in aqueous acidic ferric sulphate solutions via a rotating sintered disk technique in the temperature range 25 to 80°C. They identified two rate controlling steps. The first, below 60 °C, was surface chemical reaction controlled and the second, at higher temperatures, was solution transport controlled. The respective activation energies were 92 kj/mol and 33 kj/mol. The leaching rate was directly proportional to the ferric concentration below 0.005 M but not sensitive to higher ferric sulphate concentrations. Dutrizac and MacDonald [23] also studied the dissolution of synthetic CuS and high-grade natural covellite in the temperature range 25 to 95°C in acidified ferric sulphate solutions. They found little difference in the leaching rate between natural and synthetic covellite. Other leaching observations were similar to those of Thomas and Ingraham [22]. -29-2.4.3 Electrochemical Studies A number of researchers have investigated the electrochemical dissolution of copper sulphide ores and mattes in order to develop a process for the direct electrorefining of these materials. Etienne [24] studied the electrochemical aspects of aqueous oxidation of copper sulphides using rotating disk anodes of digenite and chalcocite. The results showed that the rate of chalcocite oxidation was under the control of diffusion of the cupric ions through the solution in the pores of the covellite-sulphur product layer. Etienne also explained the cause of large overpotentials observed by many researchers that occurred at some time into the electrolysis process. She theorized that the polarizations causing the overpotentials were due to the precipitation of copper sulphate in the pores formed from the leaching of copper out of the matrix. This blocked the transfer of current to the reactive surface sites where the electrolyte contacts the solid surface thus setting up a high electrical resistance. Biegler and Swift [25] also investigated the dissolution of copper sulphide anodes and the results supported Etienne's theory for the cause of the polarization. They also used other electrolytes besides copper sulphate and found that the time at which polarization occurs is directly related to the time at which the hmit of solubility of the copper salt in the solution is reached. They also noted that the structure of the product layer is poorly understood and further investigation of the non-equilibrium products formed during dissolution would be required to understand the leaching process. The study of the mechanism of the anodic dissolution of CU2S was performed in the presence of sulphuric acid under galvanostatic and potentiostatic conditions by Winand et al. [26]. In all cases, a layer of digenite, Cu : gS, was found to form on the surface according to the following reported reaction. SCu^S -» 5CulxS+Cu2+ + 2e A concentration gradient of copper was observed through the digenite layer. This digenite layer stays at a constant thickness after the Cu1;1S layer appears on the surface. The reaction in this step was reported as -30-3C«,„5 -> 4Cu1AS+Cu2++2e If the anodic potential is low during the electrolysis process, the next reaction proceeds as follows WCuLlS -» llCu2++10.S+2r? However, if the current density is sufficiently high to achieve a sharp increase in anodic potential or the potential is kept high, the reaction path is given by the following two reactions lOCulAS -> \QCuS + Cu2+ + 2e followed by CuS -+Cu2++S+2e Furthermore, sulphate is also found to be formed to some extent at high anodic potentials according to the following reactions. CuS+AH20 -» Cu2+ + S02-+W++8e It must finally be noted that the formation of a copper sulphate precipitate on the surface was stated not to be the cause of the sharp rises in anodic potential observed in the experiments because the calculated current density for such a film on the surface was a factor of ten higher than the current densities used in the study. McKay [27] studied the anodic decomposition of copper-rich mattes using particulate electrodes. Anodic decomposition of synthetic chalcocite was defined as a three-stage process according to the following reactions: 1. Cu^ -+Cu2_xS+xCu2+ + 2xe~, 1.75 < (2-x) < 1.83 2. Cu2_xS-+Cu2_yS + {y-x)Cu2*+2(y-x)e-, 0.7<(2-y)<0.9 3. Cu^S -» (2-y)Cu2+ + S° + 2(2-y)e' -31-Stage 1 is associated with crack formation along grain boundaries and in stage 2, the grains begin to deteriorate as copper is depleted from them. Total-bed polarization occurs because of the deterioration of the electrode and a reduction in conducting reaction interfaces between the electrolyte and the solid surface due to sulphur formation in stage 3. The electrochemical dissolution of copper sulphides was investigated by Hillrichs et al. [28,29,30] through cyclic voltammetric methods. The anodic dissolution of Q .^^ S in sulphuric acid pointed to three factors controlling the current density. These are, (a) solid state diffusion through the CuS product layer, (b) pore diffusion in the product layer and (c) resistance polarization due to CuS04 precipitation in the pores formed during dissolution. The formation of a thin metastable copper oxide layer was also thought to affect the dissolution of CuS. Further studies [29] confirmed the formation of this metastable, non-stoichiometric copper oxide/hydroxide layer. MacKinnon [31] investigated the anodic dissolution of chalcocite using a fluidised-bed anode method. The intermediate formation of "blue-remaining" covellite (CuuS) was observed. The dissolution process became inhibited after about 50% copper removal and was accompanied by increased oxygen evolution on the platinum current distributor. This inhibition was inferred to be caused by sulphur formation on the surface of the reactive sites. -32-2.4.4 E h -pH Relationships and Phase Systems Potential-pH diagrams show chemical equilibrium relationships for aqueous systems. The plots are generated from hydrolysis and oxidation reduction reactions. Figures 2.15a and 2.15b are E-pH diagrams for the Cu-S-H20 system at 25 and 100 "C generated at unit activity for all species [32]. Figures 2.16a-b and 2.17 [32] are thermal precipitation diagrams to show shifts in solution-solid equihbria with respect to temperature and pH. Figures 2.18a-b [33] are phase diagrams for the Cu-S system to show the stability ranges for the various species. This phase diagram does not show some of the metastable phases that have been observed by many researchers. -33--34-Figure 2.16a-b Thermal precipitation diagrams for the CuSO4-H2S04-H20 system [32]. -35-• Figure 2.17 Thermal precipitation diagram for the CuSO^HjSCvHjO system at 100-CforaCiiJt = V . [32] . -36-J20Cv, 1 i r 1000 I N 8 0 0 -r-< H 400 r 2 200r -crystalline S 0 1 — 100 Cu melt 1105' ~~ sulfide melt 813" high digenite —j 507* covellite -120° Wt. %S 80 60 40 700h 600h 500h .400h tr cr300(-UJ covellite 507s covellite 200h \00\-157* be CuS •Cu|7S \ \ Cu \ i -435* hex. Cu. chalcocite 103.5* digenite-I orth. chalcocite Cu Wt.% S CuxS X- 1.70 -t-r-n T.80 21 1.90 2.00 2.10 Figure 2.18a-b Phase diagrams of the Cu-S system [33]. The abbreviation "be" refers to "blaubleinder covellite". -37-2.4.5 Gas-Liquid Mass Transfer in Oxidative Leaching Molecular oxygen has poor solubility in aqueous solutions, particularily under atmospheric conditions, so agitation and/or sparging systems must continually be able to transfer enough oxygen into solution to maintain high leaching rates. There are three important steps in allowing oxygen to perform its objective. The first is that oxygen must dissolve into solution (or create a surrogate oxidant in solution). The second is that the dissolved oxygen or a surrogate oxidant must reach the mineral surface by mass transport and finally, the oxidant must react on the surface. In general, dissolved oxygen itself is not very reactive on most mineral surfaces because electrochemical reduction of oxygen is slow on metal sulphides [34]. Peters [34] has suggested four methods by which oxygen is utilized during oxidative leaching. The four models are shown in Figure 2.19 with mechanisms being described as follows: 1. The simple model suggests that dissolved oxygen is transported to the particle surface where it reacts directly to oxidize the metal (Figure 2.19a). 2. Oxygen can react homogeneously with reduced species in solution to form a surrogate oxidant. An example of this is the oxidation of Fe2+ to Fe3* in solution. Since it is possible for iron to be present in higher concentrations than dissolved oxygen, it is more readily available to react on the mineral surface (Figure 2.19b). 3. A surrogate oxidant can also be formed when a reducing agent generated through mineral decomposition is oxidized at the gas/liquid interface. An example of this type of a mechanism occurs in ammoniacal leaching where Cu(NH3)2+ is known to oxidize very quickly to provide a surrogate oxidant Cu(NH3)n2+. There is little dissolved oxygen under these conditions and the surrogate is responsible for mineral oxidation as well as homogeneous oxidation of other reducing agents generated by mineral decomposition (Figure 2.19c). -38-4. Theoretically, the quickest oxygen transfer rate that can be achieved is when a highly soluble surrogate oxidant is formed in the gas phase. In Figure 2.19d, the surrogate oxidant is labelled as QO z. As will be discussed later, case 2 may be an important mechanism in the leaching of chalcocite. Dawson-Amoah [35] investigated gas-liquid mass transfer in oxygen pressure leaching systems. Variables such as impeller type, size, speed and immersion depth were investigated. The effect of baffles and volumetric power consumption rates were also studied. Dawson-Amoah found that unsymmetrical, unbaffled tanks yielded higher mass transfer rates. Impeller depth was found to be the most important variable for determining the minimum required tip speed of the impellers. Mass transfer is significantly enhanced at shallow depths and higher critical tip speeds were required at greater depths. The most efficient impeller for oxygen mass transfer was found to be the flat 6-bladed disc impeller. It was found to be at least 2.8 times as efficient as the pitched-up axial impeller. The pitched-down axial impeller was the least efficient. Although the radial impeller consumes the most energy, it has the highest mass/energy ratio (kg oxygen/kW-h) for oxygen mass transfer. It was also suggested that shallow impellers pump gas and create numerous small bubbles. The values of mass transfer coefficients were high for shallow impellers, further suggesting that spargers are not necessary to achieve high gas-liquid mass transfer. Bubbles created from spargers may simply rise to the surface with little oxygen depletion. Under gas pumping conditions, bubbles that rise to the surface and mix with the freeboard gas can be recycled back into solution. -39-Dissolved Metal Dissolved Metal Dissolved Metal Figure 2.19 Models for oxygen absorption during oxidative leaching [34]. -40-2.5 Summary The second stage leach at INCO's Copper Cliff Copper Refinery has ocassionally experienced problems in leaching copper from a chalcocite concentrate received from the first stage leach. The poor leaching behavior is identified by the extended leaching times required for some batches. Incomplete leaching results in significant levels of copper reporting to the residue. The residue is a precious metals concentrate in which unleached copper is an undesirable impurity. The INCO-CRED process and the chemistry of the second stage leach is described in section 2.1. A mathematical model was developed by Dreisinger and Peters [3] to investigate metallurgical causes of the slow leaching behavior. The model was based on information from data and historical research reports from INCO and other studies on chalcocite and covellite leaching. The given reactions were proposed to occur in sequence with the reaction path dependent upon the initial solution conditions. The model predicted that the leaching rate was dependent upon the copper concentration and that a depletion of soluble copper would result in the leaching rate reaching a value close to zero. The proposed reactions and the predicted catalytic role of copper required verification through experiments. The role of this study was to verify the assumptions made in the development of the model and attempt to discover any other parameters which may affect the leaching process. There is unanimous agreement among the various studies that the leaching of chalcocite is a multi-step process. There is also general agreement that intermediate copper sulphide compounds are formed during the leaching process. In oxygen leaching experiments, higher oxygen pressures and higher temperatures were found to increase the leaching rate. The presence of ferric ion increases the leaching rate but if a sufficient amount is present, there is no increase in rate realized by increasing the ferric ion concentration. In oxygen pressure leaching the ferric ion is believed to serve as a surrogate oxidant and acts as an intermediary between oxygen and the copper sulphide mineral. -41-Surrogate oxidant formation can increase leaching rates considerably because it is usually available in higher concentrations than dissolved oxygen. Surrogate oxidants are generally more reactive on mineral surfaces than dissolved oxygen. In terms of physical properties of agitation systems, shallow mounted impellers are more efficient for gas-liquid mass transfer. Flat-bladed impellers are more efficient also at good gas dispersion. The pitched-down impeller was found to be the least efficient. -42-CHAPTER 3 - Experimental Methods The experimental program consisted of three parts designated as part A, part B and part C. All experimental work was performed in Sudbury at the Copper Cliff Copper Refinery. The objectives and procedures for each part are described in the following sections. The experimental apparatus consisted of a Parr 2-liter titanium autoclave fitted with an internal cooling coil and a magnetic stirring drive. The oxygen delivery to the process was monitored using electronic flow meters connected to a personal computer via a data acquisition system. The temperature and stirring speed were controlled and monitored using a Parr controller unit. A schematic of the general layout of the experimental apparatus is shown in Figure 3.1. Oxygen Data Aquisistion System and Computer Valve Electronic Mass Flowmeter Temperature and Stir Speed Controller Magnetic Stirrer O O Autoclave - C D — Valve Cooling Water Figure 3.1 Schematic of the experimental setup. -43-3.1 Part A: Study of the sequential nature of the reactions. The objective of the part A experiments was to test whether the leaching process follows the reactions proposed in the model discussed earlier. The copper was predicted to leach sequentially to form basic copper sulfate and CuS as intermediate products. The theory was tested by performing interruptive experiments. These experiments were stopped at various times into the leaching process. The products were separated and analyzed. 3.1.1 Feed preparation Synthetic electrolyte was prepared using analytical-grade reagents in distilled water with analytical grade sulphuric acid (95-98 wt%). The initial electrolyte composition was 40 g/L Cu, 5 g/L Fe and 200 g/L sulphuric acid. All the analysis of solutions and solids were done by an ICP scan at the INCO Central Lab in Copper Cliff unless otherwise indicated. Numerous samples of solids (first stage residue) were obtained from the plant. The sample that was found to contain a "typical" assay of metal values was chosen for the part A experiments. The cake was thoroughly washed using a Buchner funnel apparatus to remove all entrained electrolyte and was stored wet in sealed containers to prevent oxidation. The assay of the cake is given in table 3.1. Table 3.1 Assay of First Stage Residue used in Part A Experiments. Weight Percent (dry basis) Cu Ni Co Fe S As 80.0 0.18 0.04 0.08 18.1 0.54 -44-3.1.2 Experimental procedure A feed slurry was prepared by adding 510 g of solids (dry basis) and 500 mL of the prepared electrolyte to the lab autoclave. Distilled water was then added to the autoclave to achieve a total volume of 1.2 liters. The autoclave was then sealed and heated to 115°C while being stirred at a speed of 820 rpm. This stirring speed was chosen to prevent excessive splashing onto the autoclave walls which made it difficult to recover all of the solids. This stirring speed may not overcome gas-liquid mass transfer limitations but since the leaching rates were not studied in part A, this was not an issue2. When the desired temperature (115°C) was achieved, the oxygen feed was engaged and monitored using an electronic mass flow meter connected to a personal computer. The operating pressure was set at 150 psig. In the first experiment, the leaching was allowed to proceed until the oxygen flow rate dropped to zero. The total oxygen consumed to this point was considered the 100% oxygen consumption level. This value was verified by calculation of the oxygen demand based on the initial solids analysis. All following part A experiments were performed by shutting down the experiment at intermediate levels of oxygen consumption. This was done by terminating the oxygen flow to the autoclave, rapidly quenching the slurry to a temperature of 80 °C and bleeding excess oxygen out of the autoclave quickly. The autoclave was opened and the internal parts were washed into the autoclave to the prevent loss of slurry. The temperature, pH and ORP (oxidation-reduction potential-via a calomel electrode) were measured using an Orion pH/ORP meter. The slurry was then filtered hot and washed by pouring hot water over the solids. The leach solution (filtrate) was analyzed via an ICP scan. 2 The oxygen consumption rate was found to be sensitive to the agitation speed in the neighborhood of the stirring speed used. -45-The recovered solids were reslurried in water and sulphuric acid was added to dissolve all the basic copper sulphate formed. The acid was added slowly until the slurry pH reached a value of 0.5. The slurry was then filtered again to separate the unleached solids from the solution. This solution is later referred to as the "releach solution". The solids were dried in a vacuum desiccator at room temperature to prevent further oxidation prior to analysis. The analyses performed on the experimental products were as follows: 1. Leach solution: ICP Scan 2. Releach solution: ICP Scan 3. Unleached solids: ICP Scan, X-Ray Diffraction elemental sulphur, total sulphur sulphate sulphur and sulphide sulphur The sulphur assays on the solids were done only if enough material was available after performing other tests. 3.1.3 Additional experiments in Part A A number of additional experiments were performed to investigate the affect of other variables. These experiments are described below. 1. The effect of the iron content in the electrolyte: These experiments were performed with iron-free synthetic electrolyte. In these experiments, the solids were leached to 4, 20 and 100% oxygen consumption. All other variables were kept constant. 2. The effect of "copper depletion" in the first stage solution: This product is referred to as "copper-depleted" cake. This first stage cake is produced in the plant when copper is depleted from solution during the first stage leaching step. It is frequently associated with high levels of arsenic. Copper-depleted cake was investigated because difficulties in leaching have appeared in the plant during second stage leaching of this material. Two leaches were carried out to 20 and 100% oxygen consumption. -46-3. The effect of arsenic: High arsenic (5 g/L) was added to the synthetic electrolyte initially. The cake was leached to 20 and 100% oxygen consumption. 4. A leach with high Cu/S ratio solids (Cu/S ratio = 5.5)3 was performed. The higher copper levels were achieved by adding fine copper powder to the cake. 3.2 Part B: Kinetic Experiments The main objective of the kinetic experiments was to try to determine how fast the solids leached under a variety of solution conditions. These experiments were designed to maintain approximately constant solution conditions by using small pulp densities. A slurry volume of 1.8 Uters was chosen to minimize the oxygen inventory in the autoclave relative to the amount of oxygen required to oxidize the solids. This volume was the maximum that could be used safely. An overfilled autoclave could be dangerous because, as the solution is heated, the liquid could expand to the point of creating hydraulic pressure in the autoclave. The amount of solids used in these experiments was 15 g. The quantity of oxygen required for the total oxidation of 15 g of solids is approximately 0.22 mole and the amount of oxygen occupying the free space in the autoclave is less than 0.064 mole Cbased on less than 0.2 liter of free space) at the operating temperature and pressure. These numbers were considered to be acceptable for these experiments. A factorial design was chosen for this part of the experimental program. The following variables were chosen to be studied and were coded according to the following formulae: X, (Cu/S ratio) = semi-quantitative variable, l(high), O(average), -l(low) X 2 (Acid) = ([H2SO4]-60)/60 X 3 (Copper) = ([Cu]-20)/20 X 4 (Iron) = ([Fe]-2.5)/2.5 3 The solids being used in the experiments have a Cu/S ratio of 4.4 and a low Cu/S ratio is considered to be values below 4. -47-Three values (-1,0,1) of each variable were chosen. For example, acid concentration was varied at values of 0, 60, and 120 g/L H2S04 to give values of X 2 of -1, 0 and 1 respectively. The complete list of planned experiments is given in Appendix B. Unfortunately, no high Cu/S ratio cake was available during the experimental program. All experiments to be run at high Cu/S ratio were therefore cancelled. The stirring speed was set at 1350 rpm to ensure that oxygen mass transfer would not be a rate controlling factor. The temperature and pressure were the same as in part A. Blank experiments (no solids in the batch) were performed so that the experimental results could be adjusted to obtain net leaching rates. The leach solution was filtered and analyzed via an ICP scan. The solids were weighed and recovered for an ICP scan and x-ray diffraction analysis. If enough solids were available, a sulphur analysis was also performed. 3.3 Part C: Leaching CuS in the presence of basic copper sulphate Since the slow leaching conditions are predominant during the leaching of CuS in the presence of basic copper sulphate, it was decided that this process stage should be investigated further. A series of experiments were conducted to measure the leaching rate of CuS in the presence of added basic copper sulphate under a variety of solution conditions. The tests were performed in a similar method to that of part B. The CuS was obtained by leaching first stage cake to 50% oxygen consumption at which point all remaining solids are expected to be CuS based on part A results. Basic copper sulphate was obtained by running a leach to 100% oxygen consumption. Both solids were filtered and washed prior to use. The copper concentration was varied from 0 to 80 g/1 and the iron concentration was set at 0 or 0.25 g/1. Basic copper sulphate (50 g) was added to each batch. Two blank experiments were performed with no CuS present. The list of experiments is given below in table 3.2. -48-Table 3.2 List of part C experiments performed. Experiment [Cu] (g/L) [Fe] (g/L) CuS present 1 0 0.25 No 2 0 0.25 Yes 3 80 0.25 Yes 4 10 0.25 Yes 5 40 0.25 Yes 6 1 0.25 Yes 7 0 0 No 8 0 0 Yes 9 80 0 Yes 10 10 0 Yes 11 40 0 Yes 12 1 0 Yes -49-CHAPTER 4 - Results and Discussion 4.1 Part A: Study of the sequential nature of the reactions The results of the experiments in this section are presented as graphs of the amount of metal species reporting to either the leach and releach solutions or the solids residues. The raw data used to generate the graphs are provided in Appendix C. 4.1.1 The behavior of copper dissolution Analyzing the behavior of copper during the leaching process provides the most important clues as to how the leaching process proceeds. The model's first premise is that copper leaches out of the sulphide particles sequentially according to the reactions shown earlier. These equations are given here again and will be referred to by their respective numbers in following sections. 1. CuxS + (x- l)H2SOA + ^ (x- l)02 ->CuS + (x- l)CuS04 + (x-\)H20 2. CuxS +^(x - l)CuSO, + ^ (x - l)02 + 2(x - \)H20 -> CuS +^(x - IftCuiOH^ • CuS04 3. CuxS - \)02 + (x - \)H20 -> |(3-x)CuS +^(x - l)(2Cu(OH)2 • CuS04) 4. CuS + 202-+CuSOA 5. CuS+H2SOA+^02^CuSOA + H20 +S° The results for the copper distribution are shown in Figures 4.1 and 4.2. In Figure 4.1, the leach solution assay is the level of copper in solution as soon as the leaching process is stopped. The releach solution is representative of the amount of basic copper sulphate formed since this is found by redissolving the basic copper sulphate precipitate using sulphuric acid. The amount of the copper shown in the solids is obtained from the assay of the "unreacted" cake. Figure 4.2 is a magnification of the vertical scale achieved by not plotting the amount of copper in the solids residue. -50-By analyzing the copper in solution, it can be seen that the copper in solution rises as expected if reaction 1 is proceeding up until 5% oxygen consumption. At this point, all acid should be depleted and reaction 2 should proceed. Since reaction 2 consumes copper to form basic copper sulphate, the copper in solution should begin decreasing. This observation is consistent with the expected behavior. After reaching a minimum, the copper in solution should begin to increase again as CuS dissolution begins eventually according to reaction 4. It must also be noted that the initial rate of reaction is very fast and the time taken to reach 20% oxygen consumption is less than 30 minutes. This means that all CUxS is converted to CuS in less than half an hour. Total leaching times to 100% oxygen consumption in the lab were approximately 4.5 hours. The part A experiments were sensitive to agitation speed suggesting that the gas-liquid mass transfer is an important variable. The model was run with the same initial batch recipe as used in the experiments to see which reactions are predicted to proceed under the experimental leaching conditions. The comparison of the model to the actual results is shown in Figure 4.3. The model predicts that the reactions 1,2,3 and 4 proceed in order given the batch recipe used in the lab. This would mean that copper in solution should become depleted just beyond the 20% oxygen consumption point and reaction 3 should proceed until all of the CuxS is transformed into CuS. However, by analyzing the comparison of the model and the actual results, it appears that the copper in solution never reaches as low a level as predicted. There are two possible reasons why the amount of copper in the leach solution always appears higher in the actual results. The first explanation is that the leaching reactions proceed with a small degree of overlap. As the leaching process starts, some of the CuS formed under reaction 1 reacts with the acid according to reaction 5 to produce elemental sulphur and copper sulphate. This would also explain the small amount (1-2%) elemental sulphur observed in the residue. -51-500.0 Distribution of Copper As a Function of Oxygen Consumption 40 60 Percent Oxygen Consumption (%) Leach Sol'n Releach Sol'n Solids Figure 4.1 The distribution of copper during leaching of chalcocite. -52-After the acid is depleted (-5% oxygen consumption), reaction 2 proceeds rapidly but reaction 4 may also proceed at a relatively slow rate. Reaction 4 causes some copper dissolution and prevents complete depletion of copper in solution. Reaction 4 proceeding would also cause the minimum point to move well beyond the 20% oxygen consumption point and cause some CuxS to exist beyond the 20% point as well. By looking at Figure 4.2 and the x-ray diffraction results in Table 4.1, it can be seen that this is consistent with the observations. The x-ray diffraction results show that CuxS disappears somewhere between 25 and 30% oxygen consumption. Comparison of Model to Actual Results 500 | = % Oxygen Consumed — — Cu in Solids (Actual) A Cu in Releach Solution (Actual) — ± — Cu in Solution (Actual) Figure 4.3 Comparison of the model results to the actual behavior. -53-Table 4.1 X-Ray Diffraction Results Oxygen Consumption Relative Strength of XRD Pattern Compounds Observed 0.0% Very Strong Medium Medium Chalcocite (CujS) Copper I Sulphide (Cu,.%S) Cuprite (Cu20) 5.0% Strong Strong CuxS(1.96<x<1.86) Copper Sulphide (Cu,.765S) 10.9 % Very Strong Medium Digenite (Cu, 76S) Covellite (CuS) 17.9 % Very Strong Weak Covellite (CuS) Digenite (Cu, 76S) 20.0 % Very Strong Medium Medium Weak Covellite (CuS) Digenite (Cu,76S) Copper Sulphate Hydroxide Hydrate Copper Hydroxide Sulphate 25.1 % Very Strong Weak Covellite (CuS) Digenite (Cu, 76S) 30.0 % Very Strong Medium Weak Covellite (CuS) Copper Sulfate Hydroxide Hydrate Copper Sulphate Penta Hydrate 37.7 % Very Strong Possible (v. Weak) Covellite (CuS) Digenite (Cu, 76S) 47.9 % Very Strong Covellite (CuS) 75.0 % Very Strong Weak Weak Covellite (CuS) Copper Sulphate Penta Hydrate Lead Sulphate (PbS04) 100% Very Strong Anglesite (PbS04) -54-Table 4.1 X-Ray Diffraction Results cont. Oxygen Consumption Relative Strength of XRD Pattern Compounds Observed No iron in leach solution. 20.0 % Very Strong Medium Weak Covellite Copper Sulphate Hydroxide Hydrate Copper Hydroxide Sulphate 100% Strong Strong Medium Covellite (CuS) Anglesite (PbS04) Copper Sulphide (Cu18S) High arsenic in solution 20.0 % Very Strong Medium Weak Covellite (CuS) Copper Sulphate Hydroxide Hydrate Copper Hydroxide Sulphate 100% — (Not enough solids recovered for analysis.) The other reason that the copper in solution appears higher in solution is possibly because some of the basic copper sulphate may dissolve upon cooling by back reacting with H + ions according to the following reaction [32]: 3Cu2+ + 3S02' + 4H20 <r> 2Cu{OH\-CuSOA + 2HSO; + 2HJf The equihbrium of this reaction shifts to the left as it is cooled. Figure 4.4 from Kwok and Robins [32] shows that as a solution of copper sulphate is heated, the basic copper sulphate precipitates resulting in an increase in acidity. However, considering that the average pH of the slurry is 2.8, there is not enough of a change in pH to account for the discrepancy observed in the results. The amount of copper in the solids is very close to the predicted behavior during the leaching process. Figure 4.5 shows a comparison of the amount of sulphur remaining in the -55-300 u o u 200 a. 2 100 C u S 0 4 - 2Cu(OH) 2 PRECIPITATION C O M M E N C E S HERE pH 25 Figure 4.4 "Reaction" temperature versus pHjs change during precipitation from lm CuS04 solution. The "equilibrium" boundary for lm solution is also shown [32]. solids as a function of oxygen consumption. There should be no sulphur oxidation before the 20% oxygen point but the results show that there is a slight amount. This again is probably due to the fact that reaction 4 proceeds prior to the complete disappearance of Cu^. -56-Comparison of Model to Actual Results Figure 4.5 Comparison of sulphur levels in the leach residue between the model and actual results. Although the reactions are not completely sequential as predicted, they appear to be very close to being sequential. Reaction 4 proceeds prior to the predicted time. This is verified by the comparison made between the model and actual results. The x-ray diffraction results show the formation of intermediate compounds of copper sulphide during the leaching process. The leaching of CuxS (2>x>l) is very fast compared to the leaching of CuS because the leach times to 20% consumption were less than 30 minutes and -57-the remaining 4 hours of leaching were devoted mostly to CuS dissolution. The lower than expected amount of basic copper sulphate observed can only be explained as result of the reactions not occurring sequentially. The most important result was that there was no slow leaching behavior exhibited in the laboratory experiments. This result is consistent with other experiments done under normal leaching conditions previously at INCO labs. 4.1.2 Iron and Arsenic The iron is slowly leached into solution during the reaction time (Figure 4.6). However, most of the iron in solution precipitates rapidly with the basic copper sulphate to a stable level by the 25% oxygen consumption point. The level of iron in the precipitate begins to decrease slightly by the end of the leach process probably due to some iron being resolubulized into the leach solution. The role of iron in the leaching process is very important. As discussed in the literature review section, iron was found to increase the leaching rates in most studies primarily by acting as a charge carrier between the oxygen and the copper sulphide particles. Iron is also thought to have an important affect on the precipitation behavior of basic copper sulphate. The experiments with no iron in solution showed some very interesting behavior. The resulting basic copper sulphate slurry was much more viscous and the agglomerate size of the precipitate was finer (see Figure 4.7). The rate of oxygen consumption also slowed down considerably at approximately 15% oxygen consumption and was very sensitive to stirring speeds. This suggests that gas-liquid mass transfer is severely affected by the increased viscosity. Arsenic in the solids is initially leached very quickly to the 5% oxygen consumption point but then precipitates out to report to the releach solution (Figure 4.8). All the arsenic that is leached out steadily for the remaining leaching time reports to the releach solution and not to the leach solution. There appears to be an error in the assay of the releach solution at the 5% point because the mass balance at this point does not add up to the total amount of -58-Distribution of Iron As a Function of Oxygen Consumption 3.0 3 $ 2.0 ID 1J 2 Leach Sol'n Releach Sol'n Solids 0 20 40 60 80 100 Percent Oxygen Consumption (%) Figure 4.6 Distribution of iron during leaching. arsenic in the system. The releach solution assay is thought to be incorrect because the arsenic is expected to be in the leach solution before it can precipitate. The other reason that this point is thought to be in error is because the results of the experiment with no iron in solution show that arsenic first reports to the leach solution and then to the precipitate beyond the 5% oxygen point (see assay results in Appendix C). -59-(a) With Iron-xl 500 (c) Without Iron-xl 500 -60-Distribution of Arsenic As a Function of Oxygen Consumption 3.0 | — — — — 0 20 40 60 80 100 Percent Oxygen Consumption (%) Figure 4.8 Distribution of arsenic during leaching. 4.1.3 Nickel and Cobalt The leaching behavior of nickel and cobalt is very similar (Figures 4.9 and 4.10). Both metals leach very quickly and stay in the leach solution without any appreciable amount of precipitation. Nickel leaches quickly in the presence of acid and reaches a steady state by 20% oxygen consumption. Cobalt also leaches quickly in the presence of acid but continues to leach slowly until the end of the process. Neither metal is thought to affect the leaching process significantly at the levels at which they are present in the system. -61-1.0 Distribution of Nickel As a Function of Oxygen Consumption Percent Oxygen Consumption (%) Figure 4.9 Distribution of nickel during leaching. Distribution of Cobalt As a Function ot Oxygen Consumption -62-4.1.4 Additional Observations of Part A Experiments The experiments done with excess arsenic in solution or using the "copper depleted" cake yielded no unusual leaching behavior. Only the experiments done with high Cu/S cake and no iron in the electrolyte showed significant differences in leaching behavior. The experiments done with a high Cu/S ratio cake showed significant sensitivity to agitation speed and the slurries were higher in viscosity due to more basic copper sulphate being formed. The higher viscosity observed was not measured but was just based on visual observations in comparison to other experiments. The effect of no iron in the original electrolyte was discussed earlier and reported as having a significant effect on the viscosity probably as a result of the finer precipitate formation. After six hours of leaching time, there was 4 times as much unleached material as expected with copper sulphides still appearing in the x-ray diffraction results (see Table 4.1). The slow leaching in these experiments appears to be more a function of high viscosity and therefore mixing/gas-liquid mass transfer rather than chemistry. It must be noted that the "no-iron" condition is almost impossible in the plant because there is always significant levels of iron present in the spent electrolyte and entrained in the first stage cake liquor. 4.2 Part B: Kinetic Experiments The part B experiments can be divided into 2 sections; those performed with and without acid in solution. The two types of experiments must be considered distinct because they take different reaction paths in the leaching of chalcocite. For this reason, the originally suggested factorial design analysis was not carried out on the results. The experiments done with acid in the initial solution are not very representative of the leaching path in the plant because the acid is present during the whole leaching time in these experiments. The leaching rates that are of interest are the ones in which the acid is depleted early in the process. -63-The leaching rate versus percent oxygen consumption plots are provided in Appendix D for the acid and no-acid experiments. These graphs were generated by subtracting the blank experiments (no solids added) from the actual experiments to obtain net rates. The "loops" formed at the initial part of the plot are caused by the subtraction of the two runs. The experiments that begin with acid in solution proceed initially via reaction 1 where CuxS is converted to CuS. This is followed by the leaching of CuS via reaction 5. However, reaction 4 must occur to a limited degree because some of the sulphur is oxidized. The chemical analyses of the leach residues contained approximately 90% sulphur and the x-ray diffraction results show very strong elemental sulphur patterns. The degree of sulphur oxidation is shown in Table 4.2 for the various runs performed with acid. It should be noted that those experiments with no copper and/or iron in solution showed higher levels of sulphur oxidation. This is consistent with the work done by Mao and Peters [13]. They found that the presence of iron lowered the levels of sulphur oxidation. They observed 90.6% elemental sulphur in their residues, values very similar to the ones observed in these experiments. All of the sulphur was found to oxidize in the experiments done with no acid in the solution. The times taken to 10,15 and 20% oxygen consumption are given in Table 4.3. They show very clearly that the initial leaching rates are very fast where CuxS is being converted to CuS. The time to 20% oxygen consumption is usually less than 3 minutes and is a very short duration relative to the total reaction time of most of these experiment of 2.5-3 hours. This initial rapid leaching of chalcocite is consistent with the work of Mao and Peters [13] as well as other researchers. -64-Table 4.2 Degree of sulphur oxidation at various initial acid, copper and iron concentrations. Cu/S [HjSOJ [Cu] [Fe] % Sulphur Ratio (g/D (g/U (g/D Oxidized Medium 60 40 5 7.6 Medium 60 40 0 15.9 Medium 60 0 0 31.6 Medium 60 0 5 21.4 Medium 60 20 2.5 16.0 Medium 60 20 2.5 10.4 Medium 120 0 2.5 23.9 Medium 120 40 2.5 1.6 Medium 120 20 5 9.6 Medium 120 20 0 28.4 Low 60 20 0 28.5 Low 60 0 2.5 26.6 Low 60 20 5 15.3 Low 60 40 2.5 15.4 In the experiments with no acid in the initial batch recipe, the material is expected to leach via reaction 2 followed by reaction 4. The reactions are not expected to be entirely sequential as the results of part A have indicated. In these experiments, all of the sulphur was to oxidize. The times to 20% oxygen consumption are very fast in the no-acid experiments also. The total leaching times in these experiments were much shorter than part A. This is most likely due to the fact that the agitation speed was higher and the pulp density was lower contributing to better mixing and much higher gas-liquid mass transfer rates. -65-Table 4.3 Time taken to leach to 10, 15 and 20% oxygen consumption at various initial acid, copper and iron concentrations. Cu/S [HjSOJ [Cu] [Fe] Time (s) to %consumption Ratio (g /U ( g / u (g/D 10% 15% 20% Medium 0 20 5 90 142 305 Medium 0 0 2.5 100 175 334 Medium 0 40 2.5 80 105 186 Medium 0 20 0 162 332 499 Medium 60 40 5 78 92 107 Medium 60 40 0 89 110 143 Medium 60 0 0 84 110 174 Medium 60 0 5 95 109 122 Medium 60 20 2.5 115 129 146 Medium 60 20 2.5 90 108 129 Medium 120 0 2.5 168 284 306 Medium 120 40 2.5 83 98 116 Medium 120 20 5 100 116 137 Medium 120 20 0 90 113 141 Low- 0 20 2.5 230 399 559 Low 60 20 0 137 250 640 Low 60 20 2.5 89 105 125 Low 60 20 5 92 110 136 Low 60 20 2.5 132 152 180 4.3 Part C: Leaching CuS in the presence of basic copper sulphate Figures 4.11 and 4.12 show the oxygen consumption rates at regular intervals during the leaching of CuS in the presence of basic copper sulphate. The rate of oxygen consumption at low copper concentrations, between 1 and 10 g/L, appears to be slower. A minimum point probably exists somewhere in this range but it is not possible to depict without more data. This dip in rate is apparent in both the experimental conditions of iron and no-iron in solution. Beyond the minimum point, the rate of oxygen consumption (CuS leaching) generally appears -66-to be copper catalyzed and increases with increasing copper in solution. Oxygen Consumption Rate Al Various Copper Concentrations 0.0 I 1 1 1 1 : 1 1 1 ' 0.0 0 20 40 60 80 Initial Copper Concentration (g/L): [Fe]=0.25 Figure 4.11 Oxygen consumption rate at various initial copper concentrations with initial [Fe]=0.25 g/L. Each curve represents a different level of oxygen consumption. Comparing the leaching rates of experiments done with and without iron in solution, it appears that iron in solution increases the leaching rate. Appendix F contains plots of rates at various oxygen consumption points with and without iron in solution. There is a crossover of rates between the 0 and 10 g/L Cu points. Table 4.4 and Figure 4.13 show that increases in the initial copper concentration or iron concentration results in an increase in the oxidation-reduction potential (ORP). This is consistent with the Nernst equation, e.g.: -67-Figure 4.12 Oxygen Consumption Rate At Various Copper Concentrations 20 40 60 Initial Copper Concentration (g/L): [Fe]=0 10% 20% 30% —•— 40% 50% 60% 70% 80% Oxygen consumption rate at various initial copper concentrations with initial [Fe]=0 g/L. Each curve represents a different level of oxygen consumption. Cu2+ + e'^Cu+ R T [Cu+1 EL = E ° — - I n - 1 nF [Cu2+] A higher solution potential will tend to increase the leaching rate by imposing a higher exchange current on the mineral. Figure 4.14 shows a schematic of an Evans E>iagram of the possible polarization curves during CuS leaching. The reversible potential of CuS leaching as shown on Figure 4.14 at 388 K is approximately 0.21 V. The actual ferric and cupric polarization curves will be a result of the mixed potential caused by both iron and copper in solution. The iron in solution is reported to be easily oxidized from ferrous to ferric in the presence of copper as the cupric-cuprous couple is thought to catalyze the oxidation of the ferrous species [36]. -68-The increase in the leaching rate observed at higher copper concentrations and higher iron concentrations is now more understandable. The increase in the cupric/cuprous, ferric/ferrous ratios or a net increase of the cupric or ferric concentrations will increase the mixed solution potential and as a result increase the anodic current (leaching rate). The dip in the solution potential, and consequently the leaching rate, observed at approximately 10 g/L copper concentration was not explainable. This point is not likely to be in error because it occurs in both the experiments (ie. [Fe]=0 and [Fe]=0.25 g/L). Table 4.4 Potential and pH measurements of the leach slurry after leaching and the approximate temperatures at which they were measured. Initial Initial pH ORP (mV) Temperature CO [Cu] (g/L) [Fe] (g/L) 0 0.25 2.84 350 80.1 1 0.25 2.79 353 77.9 10 0.25 2.53 319 81.5 40 0.25 2.51 491 84.6 80 0.25 2.43 499 83.0 0 0 2.78 314 85.9 1 0 2.68 360 82.9 10 0 2.50 321 84.0 40 0 2.68 445 86.9 80 0 2.43 449 85.1 -69-> E 550.0 500.0 -•g 450.0 -o cu g 400.0 -o 350.0 300.0 Solution Potential at Various Copper Concentrations 10 20 30 40 50 60 Initial Copper Concentration (g/1) 80 Figure 4.13 The effect of initial copper concentration on the measured ORP at [Fe]=0 and [Fe]=0.25 g/L. Evans Diagram Schematic of Applicable Polarizations Curves During CuS Leaching 1.0 0.8 0.6 o > 0.4 0.2 0.0 -0.2 Increasing ferric or cupric ion concentration CuS+ 4HjO-»»- Cu + SO*+ 8H++ 8e_ Log I - Galvanic Current (Schematic) Figure 4.14 Evans diagram of applicable polarization curves during pressure leaching of CuS (schematic). -70-4.3.1 Discussion of ORP Measurements Since there is a small amount of solids present in the slurry when the ORP is measured, the value obtained is really a "mixed potential" because it is affected by both the solids present and the prevailing solution conditions. However, since the solids concentration is very low in the part C experiments, the measured ORP value is much closer to the solution potential. A simple calculation shows that the solid potential changes very little as a result of the change in copper concentration. The equilibrium reaction is: Cu2+ + SOl~ + 8/Y+ + &e <=> CuS + 4H20 This potential will vary as a function of [Cu2+]: ie RT 1 ' E = E°- 2.303—log aCu** a™*- a»+ And assuming: asol- = aCuit „ „„ 2.303(8.314) (298.15), 2 2.303(8.314) (298.15) „ E = E + 8^ 965001 l 0 g a — (96500) p H E = E° + 0.0148 \ogaCult - 0.059\6pH Therefore, the "solid potential" will vary by -15 mV per order of magnitude change in copper concentration. The cathodic reaction is: 02 + 4ht+4e -*2H20 However, it is also possible for other reactions to catalyze the reaction: Fe3++ e -> Fe2+ Cu2+ + e -> Cu -71-The observed changes in the ORP measurements are large (Table 4.4). However, according to the calculation above, the "solid potential" can only change by -15 mV per order of change in the Cu 2 + concentration. It is therefore more likely that the observed change in the ORP measurement is a result of the Cu 2 + /Cu + and/or Fe3 +/Fe2 + couples. An increase in potential of these couples increases the rate of electrochemical dissolution and hence increases the rate of leaching. This observation is important since it indicates that the levels of copper and iron in solution are important as catalysts and. may be serving as surrogate oxidants. 4.4 Comparison of Leaching Rates - Part A and Part C A comparison of leaching rates between part A and part C experiments was performed to determine an empirical relationship between them (see Table 4.5). The main difference between these experimental section was the pulp density and agitation speeds. The comparison of rates is done at 60 and 70% oxygen consumption for the following reasons: 1. The copper concentrations are similar in these experiments. 2. The predominant material (solids) remaining is CuS. 3. The flow rates are extremely stable and are not affected by initial transients. 4. There is no manual oxygen flow control, as was practiced in the early stages of the part A experiments4. If the ratio of rates is taken between these two parts, it can be seen that the rate of leaching is 4.5 and 4.2 times as high in the 60% and 70% runs respectively in the part C experiments. This lends further credence to the fact that the rates must be dependent upon agitation and/or pulp density both of which significantly affect gas/liquid mass transfer. 4 The oxygen flow rates in the part A experiments had to be limited by a flow valve in the early stages to keep the flow within the range of the flow meter. Beyond approximately 40% oxygen consumption, the valve was opened totally to allow the system to receive oxygen as needed. -72-Table 4.5 Comparison of oxygen flow rates Weight of Solids Remaining (g) [Cu] (g/D [Fe] (g /» Oxygen How Rate (cc/s) Ratio of Flow rate / Wt. Remaining At 60% Part A 159 92 <0.1 23.5 0.15 PartC 6 86 <0.1 4.1 0.68 At 70% Part A 115 112 <0.1 19.5 0.17 PartC 4.5 87.6 <0.1 3.26 0.72 4.5 Additional Observations An experiment was performed to produce basic copper sulphate for part C experiments. The solids loading was increased to 600 g and 40 g of fine copper powder was added. The oxygen consumption rate dropped significantly at the basic copper sulphate point and displayed behavior that could be classified as "slow cook". The rate was very sensitive to agitation speed. The products of this leach were not subsequently used in the part C experiments and a new experiment was performed in its place. This experiment did, however, prove very dramatically the significant effect of slurry density (or as a result, viscosity) on the rate of leaching. Some temperature measurements [37] were made on the plant autoclave walls during leaching. Since the autoclave is thin walled, the surface temperature was expected to be quite close to the slurry temperature in the autoclave. Since the autoclave is thought to be well mixed, the surface temperature was also expected to be uniform at various points on the surface of the autoclave. However, the temperature measurements on the surface of the autoclave were quite variable and well below the 115°C set point. The surface temperature ranged from 85 to 105°C at various points. This suggests that there may be a problem with mixing in the autoclave and if that is the case, then the gas-liquid mass transfer can probably considered to be poor also. -73-Viscosity measurements [38] were made on the basic copper sulphate slurry produced in the plant using a Brookfield viscometer. The results suggested that the viscosity between batches varied considerably and that the slurry displayed a "pseudoplastic" behavior. Pseudoplastic fluids display a decrease in viscosity with an increasing velocity gradient. Commonly encountered examples, as given in most fluid dynamics textbooks [39], of pseudoplastic fluids are slurries, muds, polymer solutions and blood. This would mean that slurry near the impeller area would be quite fluid while the slurry remote from the areas affected by the impeller would be significantly more viscous. In some leaches there is a significant amount of copper found to be remaining in the residue even when it appears that the leaching is complete (ie. no temperature rise or oxygen consumption detected). This could also be a function of poor mixing. If there are any "dead spaces" in the autoclave, then there would be unleached solids trapped in these areas that would report to the residue. This is very possible given the pseudoplastic behavior of the slurry. In fact, the area behind the cooling coils may be even be stagnant or experiencing very little mixing once high concentrations of basic copper sulphate are formed in the autoclave. Figure 4.15 is a hypothetical diagram to show how the well mixed and poorly mixed zones may appear in the autoclave. Further investigations must study the effects of viscosity on fluid flow to provide a better picture of the mixing in the autoclave. -74-Figure 4.15 A schematic highlighting areas believed to be well mixed in the second stage autoclave. The poorly mixed zones are thought to the result of the observed pseudoplastic behavior of the slurry. -75-CHAPTER 5 - Conclusions and Recommendations 5.1 Conclusions The leaching of chalcocite can be divided into a number of nearly sequential reactions. The reaction path is determined by the prevailing solution conditions which are in effect controlled by the initial batch recipe. The predominant reactions, as proposed in the screening model, were verified as being accurate in defining the leaching process. These reactions are: 1. CuJ + (x- WtSOt + ^ ix -\)02 -+CuS + (x- \)CuS04 + (x - \)H20 2. CuxS +^(x - l)CuSOt + ^ (x - l)02 + 2(x - \)H20 -> CuS +^(x - l)2Cw((9//)2 • CuSOA 3. CuxS + | ( J C - 1)02 + (x - l)H20 -> ^ (3-x)CuS +^(x - l)(2C«(0//)2-CuSOJ 4. CuS + 202^CuSOi 5. CuS + /72504+102 -»CuSOA + H20+S° The reactions were found not to be entirely sequential. It is believed that reactions 4 and 5 proceed slowly before all of the C u ^ is completely converted to CuS. However, reactions 1 and 2 were found to be very fast and the degree to which reactions 4 and 5 can occur in this short time is small and so the leaching process can be modelled as being sequential. The screening model based the leaching kinetics on two steps; gas-liquid mass transfer of oxygen and copper catalysis. Slow leaching kinetics that were expected under copper depletion in solution were not realized. Copper in solution, although shown to increase the leaching rate, does not inhibit the leaching process to the point of creating the "slow cook" conditions. Also, the copper in solution is never depleted because of reactions 4 and 5 proceeding before other reactions have stopped. The "slow cook" condition observed in the plant cannot be a result of -76-loss of copper catalysis. The chalcocite leaches progressively to form various intermediate copper sulphide compounds such as djurleite (Cu, %S), digenite (Cu1765S Cu176S), and covellite (CuS). Various intermediate compounds observed by other researchers were not observed in this study. Deficiency of iron in solution increases the viscosity of the slurry by promoting precipitation of a finer precipitate. The result is slower leaching. Incomplete leaching also results with a lack of iron in solution. The iron is most likely the dominant leaching agent because it is available in higher concentrations around the particles than dissolved oxygen. The lack of iron has a two-fold effect on the leaching process by promoting the formation of a higher viscosity slurry and reducing the concentration of the surrogate oxidant. The distribution of arsenic, nickel and cobalt during leaching was reported and was found to have no discernible effect on the leaching process. High solids loading in the process was also found to increase the viscosity of the slurry and cause the leaching process to slow down. Significantly higher leaching rates are realized as a result of leaching at lower pulp densities. Observations made on the surface temperature on the of the plant autoclave appear to suggest that there is also poor mixing in the autoclave. The pitched-down impeller design is not the optimum based on oxygen mass transfer studies done at U.B.C. [35]. The cause of the slow and incomplete leaching observed in the plant is most likely caused by poor gas-liquid mass transfer and/or poor mixing. 5.2 Recommendations for further work This study has shown that the major rate controlling parameter is not based on the chemical kinetics of the leach but rather the poor gas-liquid mass transfer in the plant autoclaves. Further research on the investigation of this leaching process must address the following issues: 1. The effect of iron on the precipitate particle size and viscosity of the slurry. -77-2. The effect of viscosity on the limits of dissolved oxygen concentration and gas-liquid mass transfer. 3. The effect of impeller design and impeller depth on gas-liquid mass transfer. 4. The effect of autoclave shape and baffling created by the cooling pipes. -78-REFERENCES 1. West E.G., Copppr and Its Alloys. Ellis Horwood Ltd., Chichester, England, p. 38,1982. 2. Gupta, CK. and Mukherjee, T.K., Hydrometallurgy in Extraction Processes - Volume I. CRC Press, p.42,1990. 3. Dreisinger D. and Peters E., "The Mathematical Modelling of the INCO CRED Second Stage Leach", Report prepared for INCO, April 1989. 4. P.M. Tyroler, T.S. Sanmiya and E.W. Hodkin, "Hydrometallurgical Processing of INCO's Pressure Carbonyl Residue", Paper pres. at 117th Ann. AIME Meeting, Phoenix, Ariz., Jan 25-29,1988. 5. D.A. Huggins, "General Operation of the Copper Refinery Electrowinning Department", INCO Process Tech. Report #5907, Febr. 19,1973. 6. W.E. Jones (A. Hall, E. Krause and N. Nissen), INCO Hydrometallurgy Section Monthly Report for April 1975, June 11,1975 7. B.J. Brandt (E. Krause), INCO Hydrometallurgy Monthly Report for July/August 1978, Sept. 6,1978. 8. E. Krause, "Analyses of CRED First-Stage Residues from periods of Slow Second-Stage Leaching', Memorandum to V.A. Ettel, May 7,1986. 9. CCCR Proc. Tech Monthly Report for January, 1988. 10. P.M. Tyroler, "Chemical Species in IPC Residue", Memorandum to S. Stupavsky, Feb. 2, 1988. 11. G.J. Borbely (C.Y. Kairovicius), "CRED 2nd Stage Leach Tests", Memorandum to B.R. Bowerman, April 25,1988. 12. K. Bech, "#3 2nd Stage Autoclave, E/W", Memorandum to all concerned, Feb. 13,1989. 13. Mao M.H. and Peters E., "Acid Pressure Leaching of Chalcocite", Hydrometallurgy Reasearch, Development and Plant Practice eds. K. Osseo-Asare and J.D. Miller, TMS-AIME, 1983, pp 243-260. 14. Shuey R.T., Semiconducting Ore Minerals, Elsevier, 1975. 15. Potter R.W., "An Electrochemical Investigation of the System Copper-Sulphur", Econ. Geol, 72,1524-1542, (1977). 16. J. Sullivan, U.S. Bur. Mines Tech. Paper 473, (1930). -79-17. Thomas G., Ingraham T.R. and Macdonald, R.J.C., "Kinetics of dissolution of synthetic digenite and chalcocite in aqueous acidic ferric sulphate solutions", Can. Met. Q., 6, 281-292,(1967). 18. Chmielewski T. and Charewicz W.A., " Pressure Leaching of a Sulphide Copper Concentrate with Simultaneous Regeneration of the Leaching Agent", Hydrometallurgy, 13,63-72, (1984). 19. King J.A., Burkin A.R. and Ferreira R.C.H., "Leaching of Chalcocite by Acidic Ferric Chloride Solutions", from Leaching and Reduction in Hydrometallurgy 36-45, A.R. Burkin ed., Inst. Mining Metal., 1975. 20. Grizo A., Pacovic N., Poposka F. and Koneska Z., "Leaching of Low-Grade Chalcocite-Covellite Ore Containing Iron in Sulphuric Acid: The Influence of pH and Particle size on the Kinetics of Copper Leaching", Hydrometallurgy, 8,5-16, (1982). 21. Cheng C.Y., Lawson F., "The Leaching of Synthetic Chalcocite and Covellite in Oxygenated Acidic Sulphate-Chloride Solutions", Non-ferrous Smelting Sym.,,1989 22. Thomas G., Ingraham T.R., "Kinetics of Dissolution of Synthetic Covellite in Aqueous Acidic Ferric Sulphate Solutions", Can. Met. Q. 6, #2,153-165, (1967). 23. Dutrizac J.E., MacDonald R.J.C, "The Kinetics of Dissolution of Covellite in Acidified Ferric Sulphate Solutions", Can. Met. Q. 13, #3,423-433, (1974). 24. Etienne A., "Electrochemical Aspects of the Aqueous Oxidation of Copper Sulphides", PhD Thesis, University of British Columbia, 1970. 25. Biegler T. and Swift D.A., "Dissolution Kinetics of Copper Sulphide Anodes", Hydrometallurgy, 2,335-359, (1976/77). 26. Brennet P., Jafferali S., Vanseveren J. and Winand R.,"Study of the Mechanism of Anodic Dissolution of Cu2S", Metal. Trans., 5,127-134, (1974). 27. McKay D., "The Anodic Decompositon of Copper-Rich Mattes Using Particulate Electrodes", Ph.D. Thesis, The University of British Columbia, 1990. 28. Hillrichs E. and Bertram R., "Anodic Dissolution of Copper Sulphides in Sulphuric Acid Solution I. The Anodic Decompositon of Cu2.xS", Hydrometallurgy, 11,181-193, (1983). 29. Hillrichs E. and Bertram R., "Anodic Dissolution of Copper Sulphides in Sulphuric Acid Solution R. The Anodic Decompositon of CuS", Hydrometallurgy, 11,195-206, (1982). -80-30. Hillrichs E., Greulich H., Bertram R., "Investigations of the Electrochemical Dissolution of Copper Sulfide Ores in Sulfuric Acid Solutions", Hydrometallurgy Reasearch, Development and Plant Practice, eds. K. Osseo-Asare and J.D. Miller, TMS-AIME, 1983, 277-287. 31. MacKinnon, D.J., "Fluidised-Bed Anodic Dissolution of Chalcocite", Hydrometallurgy, 1, 241-257, (1976). 32. Kwok O.J. and Robins R.G., "Thermal Precipitaion in Aqueous Solutions", International Symposium on Hydrometallurgy, eds. D.J.I. Evans and R.S. Shoemaker, 1033-1080,1973. 33. Roseboom E.H. Jr., "An Investigation of the System Cu-S and Some Natural Copper Sulphides Between 25 C and 700 C", Econ. Geol., 61,641-672,1966. 34. Peters E.,"Oxygen Utilization in Hydrometallurgy: Fundamental and Practical Issues", Proc. of the International Symposium on the Impact of Oxygen on the Productivity of Non-Ferrous Metallurgical Processes", eds. Kachaniwsky G. and Newman C , Pergamon Press, New York, 151-164,1987. 35. Dawson-Amoah J., "Gas-Liquid Mass Transfer Rates by Gas Pumping Agitators in Oxygen Pressure Leaching Systems", M. A.SC Thesis, The University of British Columbia, 1991. 36. Dreisinger D.B. and Peters E., "The Oxidation of Ferrous Sulphate by Molecular Oxygen under Zinc Pressure-Leach Conditions", Hydrometallurgy, 22,101-119,1989. 37. Shelegey J. and Grewal I., unpublished data, August 1990. 38. Shelegey J. and Grewal I., unpublished data, September 1990. 39. De Nevers N., Huid Mechanics, Addison Wesley Publishing Company, Reading, Masachusetts, pp 9-10,1970. -81-APPENDIX A - Detailed Flowsheet of the CRED Plant. -82-APPENDIX B - Planned Experiments for Part B-1. The following is a list of the experiments that were originally planned for the Part B experimental program. All of the following experiments were performed except for the ones in which the value of X,=l (high Cu/S ratio) because this material was not available from the plant at the time the experiments were being performed. Table Bl. List of the planned experiments for part B. Experiment x2 x3 *4 (Cu/S) (Acid) (Copper) (Iron) 1 -1 -1 0 0 2 1 -1 0 0 3 -1 1 0 0 4 1 1 0 0 5 0 0 -1 -1 6 0 0 1 -1 7 0 0 -1 1 8 0 0 1 1 9 0 0 0 0 10 -1 0 0 -1 11 1 0 0 -1 12 -1 0 0 1 13 1 0 0 1 14 0 -1 -1 0 15 0 1 -1 0 16 0 -1 1 0 17 0 1 1 0 18 0 0 0 0 19 -1 0 -1 0 20 1 0 -1 0 21 -1 0 1 0 22 1 0 1 0 23 0 -1 0 -1 24 0 1 0 -1 25 0 -1 0 1 26 0 1 0 1 27 0 0 0 0 -83-APPENDIX C - Assay results of Part A experiments. The assay results from part A experiments are shown in the following tables. The percent oxygen is the fraction of oxygen used compared to the total amount required to leach the sample. The assays reported as <0.0> represent a "very low" value; it was below the detection limit of the analytical method. Table Cl . Amount of the indicated species in the leach solution at various oxygen consumption levels. Percent Oxygen Amount of species in the leach solution (g). (%) Cu Ni Co Fe As S 0.0 18.5 0.00 0.00 2.65 0.00 38.8 5.0 79.1 0.54 0.08 2.29 0.80 43.8 10.9 61.4 0.63 0.10 0.57 0.00 31.6 17.9 39.6 0.66 0.11 0.28 0.00 20.6 20.0 39.0 0.62 0.11 0.15 <0.0> 20.3 25.1 36.3 0.65 0.13 0.13 0.01 18.9 30.0 43.6 0.58 0.12 0.10 <0.0> 22.6 37.7 61.3 0.67 0.15 0.17 0.03 31.5 47.9 80.6 0.66 0.16 0.13 0.06 38.7 75.0 145.0 0.61 0.16 0.07 0.11 73.0 100.0 194.3 0.76 0.20 0.06 0.14 104.3 Experiments with no iron in leach solution 4.0 74.6 0.49 0.09 0.08 0.85 19.4 20.0 38.0 0.61 0.11 <0.0> 0.03 19.6 100.0 180.7 0.63 0.16 <0.0> 0.12 92.6 Experiments with high arsenic in leach solution 20.0 48.5 0.64 0.14 0.13 0.01 25.2 100.0 197.6 0.29 0.17 <0.0> 0.15 103.3 -84-Table C2. Amount of the indicated species in the releach solution at various oxygen consumption levels. Percent Oxygen Amount of species in the releach solution (g). (%) Cu Ni Co Fe As 5.0 2.87 <0.0> <0.0> 0.09 0.80 10.9 73.3 0.009 0.002 1.98 0.96 17.9 162.9 0.035 0.008 2.36 1.18 20.0 161.9 0.029 0.004 2.28 1.20 25.1 201.8 0.040 0.013 2.54 1.44 30.0 204.1 <0.0> <0.0> 2.31 1.48 37.7 218.1 0.041 0.010 2.44 1.65 47.9 226.1 0.064 0.012 2.54 1.88 75.0 224.6 <0.0> 0.010 2.37 2.10 100.0 235.7 0.065 0.019 2.22 2.40 Experiments with no iron in leach solution 4.0 0.26 <0.0> <0.0> <0.0> 0.05 20.0 176.3 <0.0> 0.009 0.11 1.23 100.0 239.0 <0.0> <0.0> <0.0> 2.53 Experiments with high arsenic in leach solution 20.0 152.5 <0.0> 0.008 2.55 3.82 100.0 249.5 <0.0> <0.0> 2.19 5.18 -85-Table C3. Amount of the indicated species in the releach cake at various oxygen consumption levels. Percent Oxygen Amount of species in the leach cake (g). (%) Cu Ni Co Fe As S 0.0 408.1 0.94 0.20 0.40 2.76 92.3 5.0 341.3 0.31 0.11 0.24 1.85 91.2 10.9 292.0 0.28 0.09 0.40 1.81 91.9 17.9 230.5 0.26 0.08 0.26 1.61 88.3 20.0 221.9 0.23 0.08 0.21 1.55 87.3 25.1 193.6 0.21 0.06 0.21 1.37 83.3 30.0 175.5 0.27 0.06 0.16 1.30 76.2 37.7 155.2 0.23 0.04 0.29 1.11 69.6 47.9 125.8 0.20 0.03 0.13 0.87 58.2 75.0 52.9 0.21 0.03 0.10 0.49 24.8 100.0 0.13 0.18 0.02 0.34 0.27 0.65 Experiments with no iron in leach solution 4.0 355.3 0.44 0.12 0.29 1.94 93.8 20.0 220.8 0.28 0.08 0.16 1.59 87.4 100.0 13.5 0.16 0.02 0.07 0.19 5.33 Experiments with high arsenic in leach solution 20.0 226.6 0.27 0.09 0.34 1.65 86.5 100.0 0.03 0.18 0.02 0.19 0.27 0.56 -86-Table C4. provides a sum of all the species in the three phases as check to see if the assays added upto the original amounts added into the process (ie. a mass balance). The totals in each column should be approximately equal to the value given at zero oxygen conssumption. Table C4. Total amount of species (calculated as a sum of tables C1-C3) in the process. Percent Oxygen Amount of species in the process (g). (%) Cu Ni Co Fe As 0.0 426.6 0.94 0.20 3.05 2.76 5.0 423.2 0.85 0.19 2.62 3.45 10.9 426.6 0.93 0.19 2.95 2.77 17.9 433.0 0.95 0.20 2.89 2.79 20.0 422.8 0.88 0.19 2.64 2.75 25.1 431.6 0.90 0.20 2.88 2.81 30.0 423.2 0.85 0.17 2.57 2.78 37.7 434.6 0.94 0.20 2.90 2.78 47.9 432.4 0.93 0.20 2.81 2.80 75.0 422.5 0.82 0.19 2.54 2.70 100.0 430.1 1.01 0.24 2.63 2.81 Experiments with no iron in leach solution 4.0 430.2 0.93 0.20 0.36 2.84 20.0 435.1 0.89 0.20 0.27 2.85 100.0 433.2 0.79 0.18 0.07 2.83 Experiments with high arsenic in leach solution 20.0 427.6 0.92 0.23 3.03 5.48 100.0 447.1 0.47 0.19 2.39 5.59 -87-Table C5. Distribution of species on experiments performed with the "copper depleted cake". Percent Oxygen Amount of species in the process (g). (%) Cu Ni Co Fe As Leach Solution 0.0 20.0 0.00 0.00 2.50 0.0 20.0 61.8 1.63 0.39 0.46 0.01 100.0 234.5 2.59 2.53 0.02 0.005 Releach Solution 20.0 127.0 0.056 0.01 2.92 2.70 100.0 191.3 0.092 0.03 3.08 3.85 Leach Cake 0.0 392.7 2.70 2.63 0.28 4.06 20.0 227.6 1.28 2.28 0.21 1.47 100.0 0.0s 0.0 0.0 0.0 0.0 Totals 0.0 412.7 2.698 2.63 2.78 4.06 20.0 416.3 2.961 2.68 3.59 4.18 100.0 425.8 2.679 2.56 3.10 3.85 5 The amount of residue in this experiment was very small and the quantity of species listed in the table could be approximated as zero at 100% oxygen consumption. -88-Table C6. The pH and ORP values of the slurry after leaching to a given level of oxygen consumption. Percent Oxygen ORP (%) pH (mV) 5.0 1.69 366 10.9 2.80 65 17.9 2.93 295 20.0 2.68 332 25.1 2.94 285 30.0 2.76 285 37.7 2.76 300 47.9 2.85 321 75.0 2.97 323 100.0 2.20 182 No iron 4.0 1.41 290 20.0 2.62 390 100.0 2.41 182 High arsenic 20.0 2.57 336 100.0 1.99 458 -89-APPENDIX D - Part B experimental results The following pages contain the plots of the part B experimental results. They are presented as plots rather than numerical data because the data files are too large to print numerically and would occupy too many pages. Blank experiments, done with no solids present in solution, were performed for each case. These blank experiments were subtracted from the actual experiments to obtain a net leaching rate. The plots on the following pages are the net leaching rate plots. The plots have negative values of oxygen consumption. This is the result of the subtraction and has no interpretable meaning. The initial peaks (erratic behaviour) is caused by the initial rapid kinetics and the filling of the autoclaves with oxygen. The necessary data was extracted from the corresponding numerical files where the process was stable. In the plots labelled as (x2,X3,x4), the value of xl is zero. The other plots are labelled as (x1,X2,x3,x4). The concentration values for the variables is provided in the experimental procedures sections. The plots are arranged as follows: Figures DI. to DI4: Experiments in which acid was present in the solution at the beginning of the experiment. Figures D15. to D20. Experiments in which there was no acid present in the solution at the beginning of the experiment. -90-Rate of Oxygen Consumption Run Type: (1,1,0) 40 60 Percent Oxygen Consumed Figure DI. Rate of oxygen consumption for the condition where (x2/x3/x4)=(l/l/0) and x:=0. R a t e of O x y g e n C o n s u m p t i o n Run Type: (1.0,1) a. Percent Oxygen Consumed Figure D2. Rate of oxygen consumption for the condition where (x2,x3/X4)=(l,0,l) and xt=0. -91-R a t e of O x y g e n C o n s u m p t i o n Run Type: (1,-1,0) Percent Oxygen Consumed Figure D3. Rate of oxygen consumption for the condition where (x2/X3,x4)=(1,-1,0) and \y=0. Rate of Oxygen Consumption Run Type (1,0,-1) 20 1 , -5 --10 ' 1 1 1 I I | | -10 0 10 20 30 40 50 60 70 Percent Oxygen Consumed Figure D4. Rate of oxygen consumption for the condition where (x2,x3,x4)=(l,0,-l) and x,=0. -92-Figure D5. 0 20 40 60 80 100 Percent Oxygen Consumed Rate of oxygen consumption for the condition where (x2/x3/X4)=(0,1,-1) and xt=0. R a t e of O x y g e n C o n s u m p t i o n Run Type: (0,0,0) -20 0 20 40 60 80 100 Percent Oxygen Consumed Figure D6. Rate of oxygen consumption for the condition where (x2/x3/x4)=(0,0,0) and xt=0. -93-Rate of Oxygen Consumption Run Type: (0,1,1) -" 1 • i i - i — -20 0 20 40 60 80 100 Percent Oxygen Consumed Figure D7. Rate of oxygen consumption for the condition where (x2,x3,xi)=(0,\A) and x^O. Rate of Oxygen Consumption Run Type: (0,-1,1) 30 i 1 Figure D8. Percent Oxygen Consumed Rate of oxygen consumption for the condition where (x2,x3/x4)=(0/-l/l) and x:=0. -94-Rate of Oxygen Consumption Run Type: (0,-1,-1) Figure D9. Percent Oxygen Consumed Rate of oxygen consumption for the condition where (x2,x3,x4)=(0/-l/-l) and x^O. Oxygen Consumption Rate Run Type: (0,0,0) #1 Percent Oxygen Consumed Figure D10. Rate of oxygen consumption for the condition where (x2,x3,xi)=(0,0,0) and x^O. -95-O x y g e n C o n s u m p t i o n R a t e Run Type: (-1,0,1,0) ; / - V -20 0 20 40 60 tO 100 Percent Oxygen Consumed Figure Dl l . Rate of oxygen consumption for the condition where (XUX2/X2,XA)=(-\,0,1,Q). R a t e of O x y g e n C o n s u m p t i o n Run Type: (-1,0,-1,0) -20 0 20 40 60 80 100 Percent Oxygen Consumed Figure D12. Rate of oxygen consumption for the condition where (x1,x2/x3/X4)=(-l/0/-l,0). -96-O x y g e n C o n s u m p t i o n R a t e Run Type: (-1,0,0,-1) Perceni Oxygen Consumed Figure D13. Rate of oxygen consumption for the condition where (x1/x2/x3/x4)=(-l,0,0,-l). R a t e of O x y g e n C o n s u m p t i o n Run Type: (-1,0,0,1) Percent Oxygen Consumed Figure D14. Rate of oxygen consumption for the condition where (x1/x2,X3/x4)=(-l/0/0,l). -97-Rate of Oxygen Consumption Run Type: (-1,0,1) -1 1 1 ' 1 J — 1 1 J 1 1 -10 0 10 20 30 40 50 60 70 80 90 100 110 Percent Oxygen Consumed Figure D15. Rate of oxygen consumption for the condition where (x2,x3,X4)=(-i/0,l) and x^O. Rate of Oxygen Consumption Run Type: (-1,0,-1) 15 E o -5 -\ \ -, ==_ i i i i i i i i i i •10 0 10 20 30 40 50 60 70 80 90 100 110 Percent Oxygen Consumed Figure D16. Rate of oxygen consumption for the condition where (x^x^M-lA-D and x^O. -98-R a t e of O x y g e n C o n s u m p t i o n Run Type: (-1.1.0) Perceni Oxygen Consumed Figure D17. Rate of oxygen consumption for the condition where (x2/x3/X4)=(-l,l/0) and x,=0. Rate of Oxygen Consumption Run Type: (-1,-1,0) 40 60 Percent Oxygen Consumed Figure D18. Rate of oxygen consumption for the condition where (x2rx3,X4)=(-1/-i/0) and x^O. -99-O x y g e n C o n s u m p t i o n R a t e Run Type: (-1,-1,0,0) 30 I • 0 10 20 30 <0 50 60 70 80 90 100 110 Percent Oxygen Consumption Figure D19. Rate of oxygen consumption for the condition where (xjpc^x^M-l/-! AO). -100-APPENDIX E - Part C experimental results The plots on the following pages are generated from the part C experiments in which CuS is leached in the presence of basic copper sulphate. The levels of copper and iron in solution at the beginning of the leach is given for each plot. Blank experiments, done with no solids present in solution, were done for each case. These blank experiments were subtracted from the actual experiments to obtain a net leaching rate. The plots on the following pages are the net leaching rate plots. The plots have negative values of oxygen consumption. This is the result of the subtraction and has no interpretable meaning. The initial peaks (erratic behaviour) is caused by the initial rapid kinetics and the fining of the autoclaves with oxygen. The necessary data was extracted from the corresponding numerical files where the process was stable. -101-R a t e of O x y g e n C o n s u m p t i o n Modified B: Exp. #2 30 - -Percent Oxygen COnsumed Figure El. Rate of oxygen consumption where initial [Cu]=0 g/1 and [Fe]=0.25 g/1. R a t e of O x y g e n C o n s u m p t i o n Modified B: Exp #3 Percent Oxygen Consumed Figure E2. Rate of oxygen consumption where initial [Cu]=80 g/1 and [Fe]=0.25 g/1. -102-R a t e of O x y g e n C o n s u m p t i o n Modified B: Exp.#4 Percent Oxygen Consumed Figure E3. Rate of oxygen consumption where initial [Cu]=lO g/1 and [Fe]=0.25 g/1. R a t e of O x y g e n C o n s u m p t i o n Modified B: Exp. #5 40 I at. Percent Oxygen Consumed Figure E4. Rate of oxygen consumption where initial [Cu]=40 g/1 and [Fe]=0.25 g/1. -103-R a t e of O x y g e n C o n s u m p t i o n Modified B: Exp. #6 Percent Oxygen Consumed Figure E5. Rate of oxygen consumption where initial [Cu]=l g/1 and [Fe]=0.25 g/1. R a t e of O x y g e n C o n s u m p t i o n Modified B: Exp. #8 10 —•• .in --104-R a t e of O x y g e n C o n s u m p t i o n Modified B: Exp. #9 Percent Oxygen Consumed Figure E7. Rate of oxygen consumption where initial [Cu]=80 g/1 and [Fe]=0 g/1. R a t e of O x y g e n C o n s u m p t i o n Modified B: Exp. #10 10 | -5 I 1 1 1 1 1 0 20 40 60 80 100 Percent Oxygen Consumed Figure E8. Rate of oxygen consumption where initial [Cu]=10 g/1 and [Fe]=0 g/1. -105-R a t e of O x y g e n C o n s u m p t i o n Modified B: Exp. #11 Percent Oxygen Consumed Figure E9. Rate of oxygen consumption where initial [Cu]=40 g/1 and [Fe]=0 g/1. R a t e of O x y g e n C o n s u m p t i o n Modified B: Exp. #12 -10 0 10 20 30 40 50 60 70 90 100 110 Percent Oxygen Consumed Figure E10. Rate of oxygen consumption where initial [Cu]=l g/1 and [Fe]=0 g/1. -106-APPENDIX F - Part C experimental results - Effect of iron plots The following pages contain plots of the leaching rates at various oxygen consumption points as a comparison between the experiments done at [Fe]=0 g/L and [Fe]=0.25 g/L. -107-O x y g e n C o n s u m p t i o n R a t e At 10% Oxygen Consumption 05 Copper Concentration (gpl) _,_[Fe]=0.25 _»_[Fe]=0 Figure Fl. Rate of oxygen consumption for various initial copper concentrations at the 10% oxygen consumption point. O x y g e n C o n s u m p t i o n R a t e At 20% Oxygen Consumption 0* 30 40 50 Copper Concentration (gpl): . [Fe)=0.25 .[Fe]=0 Figure F2. Rate of oxygen consumption for various initial copper concentrations at the 20% oxygen consumption point. -108-O x y g e n C o n s u m p t i o n R a t e At 30% Oxygen Consumption Copper Concentration (gpl) . [Fe]=0.25 .[Fe]=0 Figure F3. Rate of oxygen consumption for various initial copper concentrations at the 30% oxygen consumption point. Figure F4. O x y g e n C o n s u m p t i o n R a t e At 40% Oxygen Consumption Copper Concentration (gpl) . . [Fe]=0.25 .[Fe]=0 Rate of oxygen consumption for various initial copper concentrations at the 40% oxygen consumption point. -109-Oxygen Consumption Rate At 50% Oxygen Consumption Copper Concentration (gpl) _^[Fe]=0.25 _ » _ [ F e ] = 0 Figure F5. Rate of oxygen consumption for various initial copper concentrations at the 50% oxygen consumption point. Oxygen Consumption Rate At 60% Oxygen Consumption Copper Concentration (gpl) _ » _ [ F e ] = 0 . 2 5 _ » _ [ F e ] = 0 Figure F6. Rate of oxygen consumption for various initial copper concentrations at the 60% oxygen consumption point. -110-Oxygen Consumption Rate Al 70% Oxygen Consumption Copper Concentration (gpl) . [Fe]=0.25 .[Fe]=0 Figure F7. Rate of oxygen consumption for various initial copper concentrations at the 70% oxygen consumption point. Oxygen Consumption Rate At 80% Oxygen Consumption Copper Concentration (gpl) _^[Fe]=0.25 _o_[Fe]=0 Figure F8. Rate of oxygen consumption for various initial copper concentrations at the 80% oxygen consumption point. - I l l -

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0078569/manifest

Comment

Related Items