UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Reduction leaching of chalcopyrite Hackl, Ralph Peter 1983

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

Item Metadata


UBC_1983_A6_7 H32.pdf [ 5.94MB ]
JSON: 1.0078487.json
JSON-LD: 1.0078487+ld.json
RDF/XML (Pretty): 1.0078487.xml
RDF/JSON: 1.0078487+rdf.json
Turtle: 1.0078487+rdf-turtle.txt
N-Triples: 1.0078487+rdf-ntriples.txt
Original Record: 1.0078487 +original-record.json
Full Text

Full Text

REDUCTION LEACHING OF CHALCOPYRITE by RALPH PETER HACKL B.Sc. (Chemistry), The University of British Columbia, 1977 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE (Department of Metallurgical We accept this thesis as conforming to the required standard STUDIES Engi neer i ng ) THE UNIVERSITY OF BRITISH COLUMBIA June, 1983 © Ralph Peter Hackl, 1983 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 <^£^C^?U>^<^^L^ &*^A/*\£JZA-*'**-^ The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date JM^e £2j H)$>Z DE-6 (3/81) - i i -ABSTRACT This study has been concerned with deve 1 op i ng- nove 1 reduction leach methods for chalcopyrite (CuFeS2) copper con centrates. It has been found that chalcopyrite can be essen tially completely converted to chalcocite (Cu2S) by leaching in strong copper sulphate solutions at elevated temperatures, using either pressurized hydrogen gas or metallic copper pow der as reductants. The essential leach reactions appear to obey the following overall stoichiometry: CuFeS2 + 3Cu2+ + 2H2 >2Cu2S + Fe2+ + W+ CuFeS2 + Cu2+ + 2Cu° >2Cu2S + Fe2 + . Reaction rates are increased by increasing leach temperature, decreasing concentrate and copper powder particle size, and the presence of a cuprous-stabi1izing agent. The sulphide minerals bornite (Cu^FeS/^) and pyrite (FeS2), comnonly found in copper concentrates, are also quantitatively converted by reduction leaching under these conditions. Microscopic evidence indicates that chalcocite forms as layers which crack and spall away from the reacting sulph ides, allowing complete extraction of iron to take place. The mechanism for reaction probably involves transport of cuprous ions both in aqueous solution and in the solid state (CU2S), and solid state diffusion of iron outward. Cuprous ions are formed as an intermediate species during the leach, either by reaction of cupric ions with hydrogen or with copper metal. This species then probably reacts with the chalcopyrite mineral as follows: CuFeS2 + 4Cu+ >2Cu2S + Fe2+ + Cu2+. Potential methods of incorporating reduction leaching into a hydrometal1urgical process for treating copper con centrates are discussed. - i v -TABLE OF CONTENTS Page Abstract ii Table of Contents iv List of Tables viList of Fi gures  x Acknowledgements xi Chapter 1 INTRODUCTION . 1 1.1 BACKGROUND 1 1.2 EXISTING COPPER HYDROMETALLURGICAL TECHNOLOGY * 1.3 CHLORIDE ROUTES 6 1.3.1 The U.S.B.M. Copper Process 7 1.3.2 The CLEAR Copper Process 9 1.3.3 The Cyprus Copper Process 13 1.3.4 Other Chloride Processes 7 1.4 AMMINE ROUTES 19 1.4.1 The Arbiter Copper Process 20 1.5 NITRATE ROUTES 24 1.5.1 The Nitric-Sulphuric Copper Process 6 1.6 OXIDATIVE SULPHATE ROUTES 32 1.6.1 Concentrated Sulphuric Acid 31.6.2 Dilute Sulphuric Acid & Oxygen 33 1.6.3 Dilute Sulphuric Acid & Ferric Iron. 34 1.6.4 Biological Systems 35 1.7 SUMMARY 37 2 REDUCTION LEACHING AS A METHOD OF IRON REMOVAL.. 39 2.1 BACKGROUND 3- V -2.2 THE SHERRITT-COMINCO COPPER PROCESS 45 2.2.1 Copper-Iron Separation Stages 7 2.2.2 Jarosite Precipitation 49 2.2.3 Copper Recovery Stages 50 2.2.4 Precious Metals and Sulphur Recovery 53 2.2.5 Energy Consumption2.2.6 Economics 54 2.2.7 Summary  5 2.3 PURPOSE & SCOPE OF THE PRESENT INVESTIGATION 56 3 EXPERIMENTAL 9 3.1 MATERIALS 53.1.1 Copper Concentrates 59 3.1.2 Sulphide Minerals 61 3.1.3 Reagent s 62 3.2 APPARATUS 63 3.2.1 Parr Autoclave 63.2.2 Shaking Autoclave 5 3.3 EXPERIMENTAL PROCEDURE 66 3.3.1 Make-Up of Starting Leaching So 1 ut i ons3.3.2 Parr Autoclave 67 3.3.3 Shaking Autoclave 8 3.4 ANALYTICAL METHODS 69 3.4.1 Solution Analyses 63.4.2 Solids Analyses3.5 MATERIAL BALANCES AND EXTRACTION CALCULATIONS 69 4 RESULTS AND DISCUSSION 71 4. 1 REDUCT ION LEACH I NG WI TH HYDROGEN GAS 71 4.2 REDUCTION LEACHING WITH COPPER POWDER UNDER NITROGEN 7 6 4.2.1 Effect of Residence Time at 140 °C.. 77 4.2.2 Effect of Varying the Starting Molar Cu2+/(Fe+Zn) and Cu°/(Fe+Zn) Ratios. 80 4.2.3 Effect of Copper Powder Particle Size on Iron Extraction 83 4.3 POSTULATED REACTION MEACHANI SMS : GALVANIC VERSUS CUPROUS-MEDIATED 84 - v i -4.3.1 Galvanic Mechanism 84 4.3.2 Cuprous-Med i ated Mechanism 88 4.4 REDUCTION LEACHING WITH COPPER POWDER UNDER CARBON MONOXI DE 90 4.4.1 Effect of Carbon Monoxide on Reaction Kinetics 94.4.2 Effect of Leach Temperature on Reaction Kinetics 5 4.4.3 Effect of Initial Concentrate Part icle Size on Iron Extraction 99 4.5 COMPARISON OF REACTIVITY OF SEVERAL SULPHIDE MINERALS TO REDUCTION AND NEUTRAL LEACHING 101 4.6 THE MORPHOLOGY OF REDUCTION LEACHING 103 4.7 KINETIC MODELS AND MECHANISMS 108 5 APPLICATIONS, CONCLUSIONS AND RECOMMENDATIONS. . . 112 5.1 APPLI CAT I ONS OF REDUCTI ON LEACH I NG 112 5.1.1 Incorporation of Reduction Leach into S.C. Process with no Roasting.. 112 5.1.2 Incorporation of Reduction Leach into S.C. process with Partial Roasting 117 5.2 CONCLUSIONS 120 5.3 RECOMMENDATIONS FOR FURTHER WORK 121 REFERENCES 123 APPENDIX 1 130 APPENDIX 2-vii -LIST OF TABLES Tab 1e Page 1 Chemistry of the CLEAR Copper Process 11 2 Chemistry of the Cyprus Copper Process 15 3 Chemistry of the Arbiter Copper Process.... 22 4 Relative Reactivity and Elemental Sulphur Yield of Pure Minerals Leached with Nitric Acid38 27 5 Chemistry of the Nitric-Su1phuric Copper Process 9 6 Comparison of Energy Requirements 54 7 Copper Concentrates Used 60 8 "Pure" Sulphide Minerals Used 62 9 Reduction Leaching of Phoenix Concentrate With Hydrogen Gas 73 10 Effect of Varying Starting Molar Cu°/(Fe+Zn) and Cu2+/(Fe+Zn) Ratios on Reduction Leach ing of Fox Lake Concentrate 82 11 Effect of Copper Powder Particle Size on Reduction Leaching of Fox Lake Concentrate. 85 12 Residence Time Ratios at Different Iron Extractions - From Figure 14 98 13 Effect of Initial Concentrate Particle Size on the Reduction Leaching of Bethlehem Con centrate with Copper under 1 MPa CO 100 14 Reduction and Neutral Leaching Chemistry of Several Sulphide Minerals 102 15 Results of Reduction and Neutral Leach Runs on Several Sulphide Minerals 104 16 Molar Volumes of Sulphide Phases 107 - viii -17 Chemistry of Transport Processes Depicted in Figure 16 110 18 Chemistry of Reduction Leach Process with no Roasting 114 - i x -LIST OF FIGURES Figure Page 1 Schematic Flowsheet of the U.S.B.M. Copper Process^ 8 2 Schematic Flowsheet of the CLEAR Copper Process** 10 3 Schematic Flowsheet of the Cyprus Copper Process2* 4 4 Schematic Flowsheet of the Arbiter Copper Process29 21 5 Schematic Flowsheet of the Nitric-Sulphuric Copper Process^ 28 6 Pourbaix Diagram of the Cu-Fe-S-r^O System of 25 °C (Conditions: 0.1 M Fe, S Species; 0.01 M Cu Species )59 40 7 Simplified Schematic Flowsheet of the Sher r i 11-Comi nco Copper Process^ 46 8 Reduction Leaching of Ruttan Concentrate with Fine Copper Powder at 140 °C - Effect of Residence Time 78 9 Schematic Representation of the CuFeS2-Cu° Galvanic Couple in Sulphuric Acid Solution61 86 10 Schematic Representation of a Postulated CuFeS2-Cu° Couple in Copper Sulphate So 1 ut i on 87 11 Schematic Representation of a Cuprous-Mediated Reaction between Chalcopyrite and Copper 9 12 Effect of CO on the Reduction Leaching of Bethlehem Concentrate with Copper at 120 °C 92 13 Reduction Leaching of Fox Lake Concentrate with Copper at 120 °C under 1 MPa CO 94 - X -14 Effect of Temperature on Reduction Leaching of Bethlehem Concentrate with Copper under 1 MPa CO 96 15 Optical Microscope Pictures of Conversion Products from the Neutral and Reduction Leach Experiments on Cha1 copyrite/ Bornite 106 16 Morphology of Chalcocite Formation by Reduction Leaching - Solid State and Solution Transport Processes 109 17 Schematic Flowsheet of Reduction Leach Process with no Roasting 113 18 Schematic Flowsheet of Reduction Leach Process with Partial Roasting 119 - x i -ACKNOWLEDGEMENTS I would like to extend my sincere appreciation to Professor Ernie Peters for his constant guidance and encourage ment throughout the course of this study. Thanks are also extended to all members of the Department of Metallurgical Engineering for their cooperation and assist ance with experimental procedures. The financial support of the National Sciences and Engineering Research Council of Canada, and the B.C. Science Council, is gratefully acknowledged. CHAPTER 1 INTRODUCTION 1.1 BACKGROUND Approximately 90% of the world's primary copper origin ates in low-grade deposits of sulphide minerals, the most important one being chalcopyrite (CuFeS2), followed by bornite (Cu^FeS^) and chalcocite (Cu2 S ) ^ . As the copper content in an orebody is typically only Yi to 2% copper, sulphide ores are upgraded by froth flotation methods to achieve copper concentrates containing typically 20 to 3 0% copper, with a minimum of gangue materials. Iron and sulphur are always present as major impurities. Any copper recovery process from sulphide concentrates must be concerned primarily with the quantitative separation and removal of iron and sulphur from copper. At present this is accomplished almost exclusively by pyrometa 11urgica1 methods. The sulphide concentrates or partia 11y-roasted concentrates are melted at 1150-1250 °C to produce a copper-rich Cu2S-FeS matte, which is in turn blown with air to selectively oxidize most of the iron and sulphur while converting the copper sulphides to liquid metallic copper in a crude (98.5-99.5% Cu) "blister" copper form. This crude copper must be further fire- and e1ectro-refined before it is suitable for use. - 2 -Copper pyrometa 11urgy, a very old and established technology, has the advantages of high copper recovery (98%), recovery of most minor metals, and virtually 100% recovery of precious metals by refining. Smelting and converting reactions are rapid and go to completion at the high process temperatures employed. However, it has long been recognized that the single greatest problem with copper pyrometa 11urgy, and for that matter smelting in general, is the production of large quan tities of noxious sulphur dioxide containing gas; a direct consequence of oxidizing sulphur at high temperature. A chalcopyrite concentrate produces very nearly two tonnes of SO2 gas per tonne of pyrometa11urgica11y-extracted copper. As SO2 emission controls become increasingly stringent, copper plant designers are finding that a significant portion of capital costs are associated with installing sophisticated SO2 collection systems. This is particularly the case in the United States, where during the 1970's an estimated 22 cents per kilogram copper was added to the domestic production costs of the primary copper industry through improved pollution controls2. One study concluded that while capital expenditures by most U.S. copper companies have increased sharply, a sig nificant portion (25%) has been allocated for non-productive pollution abatement-^. Sulphur dioxide emitted from a roaster is relatively concentrated (5-15%) and can be fixed as sulphuric acid, but SO2 emitted from a smelting furnace is of low con centration (0.1-4%), which for economic reasons can be neither converted to acid nor neutralized with lime. In addition, - 3 -smelters have a difficult time treating the lower-grade concentrates, that is those containing high levels of pyrite or metals such as arsenic, bismuth, antimony, lead and zinc. The continuing air pollution problems associated with copper pyrometa11urgy have provided a great impetus for de velopment of technically and economically viable hydrometal-lurgical alternatives, in which sulphur can be rejected in an inert solid form. In the past 15 years, no less than 20 major hydrometa 11urgicaI processes capable of treating copper sulphide concentrates have been developed and patented in North America; yet only two of these have ever successfully reached the commercial operating stage. These are the Anaconda Company's "Arbiter" Process, which is based on an oxygen-arrmon i a leach; and Duval Corporation's "CLEAR" ch1 oride-1 each process. The Anaconda plant, an installation capable of the production of 33,000 tonnes per year of cathode copper^, was in operation from September 1974 to December 1977, at which time it was shut down due to the depressed copper market at the time^. The Duval plant started corrmercial production at 29, 000 tonnes copper in 1978, and is still in production at a current capacity of 36,000 tonnes copper per year6. The copper industry has always viewed copper hydrometal-lurgical processes as being unable to compete technically or economically with existing pyrometa 11urgica1 practice, despite the former's appeal of being non-air polluting. For example, one expert has stated that to be conrmer c i a 1 1 y viable, a new hydrometal1urgical route must demonstrate 30% lower capital cost and at least 20% lower operating costs _ IL _ than competing py r omet a 11 ur g i ca 1 routes'', target figures which have yet to be demonstrated by copper hydr ome t a 1 1 ur g i s t s . Despite the copper industry's negative outlook on hydrometal lurgical processes, research into improving existing processes and developing new ones continues. It is the object ive of this thesis to present the results of research into a newly-developed reduction leach method for copper concen trates. Suggestions for integrating the reduction leach into one of the more promising recently developed hydrometal-lurgical processes, with the objective of improving said process to make it more attractive from a commercial standpoint, will be presented. 1.2 EXISTING COPPER HYDROMETALLURGICAL TECHNOLOGY Any hydrometa 11urgica1 process for copper concentrates should be able to treat chalcopyrite if it is to gain commercial acceptance. Unfortunately, chalcopyrite is, in general, the most refractory of the copper sulphide minerals to leaching methods. The following discussion of hydrometa 11urgica1 processes will therefore focus on chalcopyrite, but it should be noted that any process effective in treating chalcopyrite should be applicable to all the other common copper sulphide minerals. Generally speaking, the ideal copper hydrometal lurgical process would satisfy all of the following requirements: - 5 -1. The ability to achieve high overall copper recoveries (at least 98%) from widely-varying grades of copper concentrates. 2. High conversion of sulphide sulphur to a marketable elemental form. 3. Production of wire-bar grade cathode copper without a separate refining step. 4. High recovery of precious metals (silver and go 1d ) . 5. Separation of iron as a marketable product, or safe disposal as an innocuous solid. 6. Recovery or safe disposal of minor metals such as selenium, arsenic, antimony, bismuth, lead, zinc, molybdenum, nickel and cobalt. 7. Use of inexpensive lixiviants, preferably ones which can be regenerated in-situ and are non-corrosive. 8. Low energy consumption by utilizing exothermic leach reactions and chemical reductants (rather than e1ectrowinning) for copper winning. Regrettably, no existing hydrometal1urgical process can claim to meet all of the above criteria, although some have the potential to be close to ideal. Copper hydrometa 11urgica1 routes can be classified in a number of ways; one conmnon procedure is to categorize them according to the type of lixiviant system used in the - 6 -leach step. Historically, the most commonly used lixiviants in copper hydrometal1urgy are: acidic oxidative chloride media, alkaline oxidative arrmine media, acidic oxidative nitrate or nitrate-su1phate med i a, acidic oxidative sulphate media. 1.3 CHLORIDE ROUTES The majority of recent hydrometa 11urgica1 research and development has focussed on chloride leach routes. Ferric chloride and cupric chloride have proven to be the most popular of the chloride lixiviants due to their ability to rapidly decompose chalcopyrite and convert the sulphide portion to elemental sulphur. The chemistry and mechanism of chloride leaching of copper sulphides has been studied extensively and is still the subject of considerable controversy. The reader is referred to several excellent recent papers on the sub j ect 8"^. For the purpose of this study, it is desirable to review three of the more promising (from a corrmercial standpoint) chloride leach processes developed in the past decade, and briefly mention important aspects of some of the other interest ing ones. - 7 -1.3.1 The U.S.B.M. Copper Process In 1971 Haver and Wong, while working for the U.S. Bureau of Mines, became the first investigators to demonstrate that very high copper extractions could be achieved using strong ferric chloride solution as a 1ixiviant12"1^. They developed what became known as the U.S.B.M. Copper Process. A simplified schematic flowsheet of this process is reproduced in Figure 1. The leach chemistry is typified by the following react i on: CuFeS2 + 3FeCl3— >CuCl + 4FeCl2 + 2S°. (1) Haver and Wong were able to obtain 99.9% copper extraction in two hours when leaching finely ground (-325 mesh) chal copyrite at 106 °C13. Pyrite, molybdenite and gold remained unattacked, but 96% silver dissolution was achieved. The ferric chloride lixiviant is regenerated by oxidation of ferrous chloride in a separate step 12FeCl2 + 302 + xH20 >8FeCl3 + 2Fe203-xH20 (2) with excess iron disposed of as an impure ferric oxide precip itate. Copper is electrowon in a diaphragm cell, and sulphur is recovered from leach residue in an ammonium sulphide leach step. The U.S.B.M. Process succeeded in achieving a very high copper recovery. Copper is electrowon from the cuprous Ballmilled Concentrate OXIDATION IS FeCI3 EVAPORATION Gangue + Sulphur CONDENSING DRYING AND MELTING T Copper f+Ag] To Electrorefining Fe203 xH20 To Tailings Pond (NH}2S LEACHING SULPHUR PRECIPITATION Sulphur Make up (NH4)2S * To Tailings Pond FIGURE 1 Schematic Flowsheet of the U.S.B.M. Copper Process14 - 9 -state at the relatively low power consumption of about 1.5 kWh/kg Cu^. This can be compared to the power consumption for electrowinning copper from the cupric state in sulphate solutions, which is often quoted as being about 2.6 kWh/kg Cu. However, the U.S.B.M. Process has made no provision for recovering precious or minor metals, and the copper product requires further refining to achieve adequate purity. An economic evaluation concluded that the process was competitive with existing pyrometa 11urgica1 practice1^, but as yet the U.S.B.M. Process has not been used corrrner c i a 11 y . 1.3.2 The CLEAR Copper Process Duval Corporation's CLEAR Process (an acronym for Copper Leach Electrolysis and Regeneration), patented in 197416, has reached commercial production of 36,000 tonnes copper per year6 and as such must be regarded as the most successful of the chloride leach processes. A schematic flowsheet of the CLEAR Process is presented in Figure 2, and the pertinent chemistry is surrmarized in Table 1. The CLEAR Process differs from the U.S.B.M. Process in that two successive leach steps are employed to extract 99.5% of the copper from a chalcopyrite feed, and simultan eously dispose of iron as a goethite-jarosite mixture. As with the U.S.B.M. Process, pyrite and molybdenite remain unattacked, but a substantial portion of any silver would be solubilized. The first-stage leach, conducted at 107 °C for VA hours, is essentially a cupric chloride leach make up NaCI, — KCI Concentrate i_ FIRST-STAGE LEACH St OXIDATION AND REGENERATION -PURGE J 7> SULPHUR RECOVERY Cement Copper REDUCTION ELECTROLYSIS Si1 I o Copper S Basic Iron Salts Gangue FIGURE 2 Schematic Flowsheet of the CLEAR Copper Process6 - 11 -TABLE 1 Chemistry of the CLEAR Copper Process  First-Stage Leach (1-1) 2CuFeS2 + 3CuCl2 5>4CuCl + FeCl2 + 2S° + CuFeS2 Preliminary Reduction (1-2) CuCl2 + Cu »2CuCl Electrolys i s (1-3) 4CuCl + FeCl2, »2Cu + 2CuCl2 + FeCl2 Second-Stage Leach (Oxidation and Regeneration-Purge) (1-4) CuFeS2 + 2S° + 2CuCl2 + FeCl2 + 3/2 02 + 3H20 >3CuCl2 + 2Fe(OH)3 + 4S° (1-5) 12FeCl2 + 302 + 2H20 >4FeOOH + 8FeCl3 (1-6) 6FeSO^ + 12FeCl2 + 9/2 02 + 3KC1 + 9H20 >3KFe3(SO^)2(OH)6 + 9FeCl3 - 12 -which solubilizes about 50% of the chalcopyrite copper as cuprous chloride (equation 1-1). The pregnant liquor is directed to a preliminary reduction stage where some cement copper is added to reduce any remaining cupric chloride to the cuprous state (equation 1-2). The liquor is then ready for entry to the diaphragm-1ype e1ectrowinning cells where half of the copper is won as product at the cathode, and half is oxidized back to cupric chloride at the anode (equation 1-3). The anolyte from e1ectrowinning is recycled back to the second-stage leach, which is conducted at 140 °C under 276 kPa oxygen pressure for lYi hours. This leach serves a three-fold purpose - to dissolve the remaining chalcopyrite, to complete regeneration of the leach solution required for the first-stage leach, and to precipitate iron as a basic salt for disposal. Equation 1-4 depicts the overall chemistry of this step. A minor amount of sulphur is oxidized to sulphate which is purged from solution as potassium jarosite (equation 1-6). The remaining iron tends to precipitate as the easily filterable goethite (equation 1-5) rather than ferric hydroxide. Regarding sulphur recovery, the literature is vague and mentions only that "it appears possible to produce a marketable sulphur product from the residue by froth flotat ion"6. Unlike the U.S.B.M. Process, the CLEAR Process has apparently solved the problems of precious metals recovery and minor metals recovery/disposal to yield a relatively pure copper product, but no details are available as yet. The copper product almost certainly requires further refining - 13 -before it is suitable for use. The process is also claimed to be economically competitive with pyrometal 1 urgi cal practice, but no cost data are available to support this claim. 1.3.3 The Cyprus Copper Process The Cyprus Metallurgical Processes Corporation developed a novel process in 1972 whereby chalcopyrite was anodically decomposed in chloride solution1^. This process, named the "Cymet" Process, had the distinctive feature of being able to recover both copper and iron e 1 ectro 1 ytica1 1 y, and was in fact demonstrated in a 23 tonne per day chalcopyrite con centrate planti8>19. Unfortunately the Cymet Process suffered from very high power consumption, problems with the electrolytic cell design and the inability to produce a pure copper product. It was abandoned in 1975 and an improved process which did away with the electrolytic step, renamed the "Cyprus" Process, was developed and is now being piloted at a 100 kg/h copper rate20'21. A simplified schematic flowsheet of the Cyprus Process is presented in Figure 3 and the important reaction chemistry is surrmarized in Table 2. Copper concentrates are dissolved in a mixed ferric ch 1 oride-cupric chloride solution in a counter cur rent two-stage leach. The first leach stage serves to solubilize about half the chalcopyrite copper as cuprous chloride (equations 2-1 and 2-2) which is directed to the copper recovery stage. A minor amount of sulphur is oxidized to sulphate (equation 2-3). The second leach stage is essen-Concentrate 1 OXYDROLYSIS TAILINGS TREATMENT FT _w Jarosite.Gangue To Tailings Pond FLUID BED REACTOR \ MEL1 FURh riNG JACE Copper •HC1 To Oxydrolysis MoS2 FIGURE 3 Schematic Flowsheet of the Cyprus Copper Process - 15 -TABLE 2 Chemistry of the Cyprus Copper Process Leach 1 (2-1) CuFeS2 + 4FeCl3  (2-2) CuFeS2 + 3CuCl2  (2-3) S° + 6CuCl2 + 4H20 Oxydrolys i s (2-4) 4CuCl + 02 + 4HC1 >4CuCl2 + 2H20 (2-5) 4FeCl2 + 02 + 4HC1 >4FeCl3 + 2H20 (2-6) 3FeCl3 + 2Na2SO^ + 6H20 >NaFe3(SO^)2(OH)6 + 3NaCl + 6HC1 (2-7) FeCl3 + 6H20 >Fe(OH)3(H20)3 + 3HC1 Leach 2 (2-8) CuCl + FeCl3 >FeCl2 + CuCl2 (2-9) CuFeS2 + 4FeCl3 >CuCl2 + 5FeCl2 + 2S° Metal Reduction »CuCl2 + 5FeCl2 + 2S° -*4CuCl + FeCl2 + 2S° >6CuCl + H2SO^ + 6HC1 (2-10) CuCl + 1/2 H2 ^ •> Cu + HC1 - 16 -tially a ferric leach and dissolves the remaining chalcopyrite while regenerating lixiviant required for the first leach (equations 2-8 and 2-9). An overall copper extraction of 97.5% is obtained, with pyrite and molybdenite remaining unat tacked. Cuprous chloride from the first leach stage is recovered by flashing the hot pregnant liquor to about 40 °C to crystallize CuCl. The spent liquor from crystal lization is directed to an oxydrolysis stage where FeCl3 is regenerated for the second-stage leach (equation 2-5) and excess iron is removed as a mixture of hydrated ferric oxide and jarosite (equations 2-6 and 2-7). Copper is recovered from CuCl in a unique process in which the crystals are reduced with hydrogen in a fluid-bed reactor at 510 °C (equation 2-10). Copper forms as nodules covering sand particles, which are melted in a conventional furnace, slagged to remove impurities, poled to remove oxygen and cast into wirebars. The wire produced from these bars is apparently of a purity comparable to electrorefined copper. The Cyprus Copper Process thus exhibits a distinct advantage over the U.S.B.M. and CLEAR Processes - it avoids the energy intensive e1ectrowinning steps of the latter two and produces a pure copper product. The total energy require ment of the process is claimed to be 37.1 MJ/kg Cu. The process resorts to a copper winning step that is pyrometal-lurgical rather than hydrometal1urgical, but illustrates the desirability of seeking chemical reductive methods as alternatives to e1ectrowinning. Available literature on - 17 -the Cyprus Process mentions that methods of recovery of S, MoS2> Au, Ag and other minor metals have been developed, but no details are available as yet. 1.3.4 Other Chloride Processes A number of other chloride leach processes exist which never reached the pilot plant stage, but have interesting features nonetheless. A French process22 leaches chalcopyrite with cupric chloride in the normal manner, but the leached copper, complexed as CuCl 2~ > is then oxidized to cupric chloride while copper is solvent extracted simultaneously by a LIX reagent, and stripped from the organic by H2SO4 to yield a pure CuSO^ electrolyte. Copper is then electrowon in the normal matter. The process obviously utilizes solvent extract ion to get around the difficulties of electrowinning in chloride media and to gain a pure copper electrolyte. But it is doubtful whether the added costs of incorporating a solvent extraction step, plus the extra power consumption inherent in electro-winning from sulphate media, is justified. The University of British Columbia, in conjunction with Cominco Ltd. patented a chloride leach process^'^ in which the problems of copper winning and precious metals recovery were specifically addressed. In the U.B.C. - Cominco process, chalcopyrite is leached by a standard chloride route and product is crystallized as CuCl, which contains silver - 18 -in solid solution. A novel method of silver recovery was developed which consisted of dissolving CuCl in NH^Cl to generate the cuprous ammine complex Cu(NH3)2+, which is a strong enough reducing agent to precipitate silver as follows: Cu(NH3)2+ + 2NH3 + AgCl2_ »Ag° + Cu(NH3)^2 + +2C1-. (3) The supernatant solution is then steam stripped back to neutral pH to yield a CuNH3Cl complex, which could be thermally decom posed to CuCl and NH3 at 250 °C. The U.B.C. - Cominco Process also offered a gold recovery method. This method consisted of solubilizing gold from leach residue with ferric chloride, according to the following stoichiometry: Au + 3Fe3+ + 4C1" >AuCl/4- + 3Fe2 + . (4) A small amount of chlorine gas was used to ensure that FeCl2 was re-oxidized. Gold extractions of 91.5% were obtained at 60 °C. Other very recent chloride leach processes for which little is yet known, but which deserve mention are the Great Central Mines Process, developed by the Vancouver, B.C. firm of Bacon, Donaldson and Associates Ltd.2^, and the Dextec Process, developed by an Australian company26. The Great Central Mines Process apparently recovers iron in saleable powder form with 99.5% efficiency and boasts an innovative - 19 -e1ectrowinning cell design producing a purer product at lower cost. The Dextec Process anodically decomposes chalcopyrite in a special electrolytic cell to produce, in a single step, copper powder, elemental sulphur and precipitated iron oxides (mainly goethite). A cell voltage of as little as 0.8 volts and a power consumption of 1 kWh/kg Cu are claimed. 1.4 AVMINE ROUTES Commercial application of ammonia oxidative leaching of sulphide concentrates has best been demonstrated by Sherritt Gordon Mines Ltd. with its now classic process for treating nieke 1-copper-coba11 sulphide minerals at Fort Saskatchewan, Alberta2^. As applied to pure copper hydrometa 11urgy, the only application of importance is the Anaconda Company's Arbiter Process; developed to treat copper concentrates and used commercially from 1974-1977^> 5. The chemistry and mechanism of anrmonia oxidative leaching of copper sulphides is quite complex and has been studied by a number of investigators28'3^"3^. Leaching is possible owing to the stabilization of the cupric ion in an alkaline solution as the tetraarrmine complex, Cu(NH3)^2 + . Sulphide is oxidized ultimately to sulphate and rejected as gypsum, but goes through a number of intermediates such as thiosulphate* •*2®3^~ > trithionate, S30^2_; tetrathionate, S^Og2-; and sulphamate, NH2SO3-. Iron is oxidized and rejected as hydrated iron oxide. - 20 -1.4.1 The Arbiter Copper Process A symplified schematic flowsheet of the Arbiter Process is presented in Figure 429, with the overall reaction chemistry outlined in Table 3. Copper sulphide minerals such as chal copyrite and chalcocite are dissolved by an oxidizing arrmonia-ammonium sulphate leach (equations 3-1 and 3-2) in a series of separate closed leaching tanks at temperatures ranging from 50-80 °C. By optimizing mixing, the required oxygen partial pressure is kept low at about 35 kPa. Leach time is typically 5 hours. A second leach step would be required to obtain high copper extraction, but the Arbiter Process was originally designed to extract only about 80% of the concentrate copper value. Pyrite and molybdenite remain unattacked. The remaining copper and any precious and minor metals are recovered by flotation into a medium-grade con centrate and fed to the Anaconda smelter. Iron is rejected during the leach as hydrated iron oxide. The pregnant liquor is purified by solvent extraction using LIX 65N (equations 3-3 and 3-4) to yield a copper sulphate electrolyte which is electrowon by conventional methods. Apparently some refining is required to obtain a pure product. The raffinate contains excess ammonium sulphate which is precipitated with lime and disposed of as gypsum, and ammonia is recovered for recycle back to the leach (equation 3-5). The Arbiter Process enjoyed brief commercial success mainly because it could be used in conjunction with a smelter to treat successfully a particular concentrate. The process 1 1 PRIMARY LEACH [NHJ2SO4 SECONDARY LEACH FLOTATION t 1 Secondary Taj|jngs Concentration Lime Concentrates NH3 AMMONIA RECOVERY Gypsum to Disposal SOLVENT EXTRACTION Loaded Organic Raffinate J Stripped Organic SOLVENT STRIPPING Strong Electrolyte Spent Electrolyte ELECTROWINNING Cathode Copper FIGURE 4 Schematic Flowsheet of the Arbiter Copper Process1* - 22 -TABLE 3 Chemistry of the Arbiter Copper Process  Primary and Secondary Leaches (3-1) 2CuFeS2 + 12NH3 + 17/2 02 + (n + 2)H20 » 2Cu(NH3)4SO^ + 2(NH^)2SO^ + Fe203-nH204, (3-2) CCi2S + 6NH3 + (NH^)2SO^ + 5/2 02 > 2Cu(NH3)^SCty + H20 Solvent Extraction (3-3) Cu(NH3)/fSO^ + 2RH > CuR2 + (NH^)2SO^ + 2NH3 Solvent Stripping (3-4) CuR2 + H2SO^ >CuSO^ + 2RH Ammonia Recovery, Gypsum Formation (3-5) (Nrty^SCty + CaO —^—^CaSO^ + 2NH3 + H20 Electrowinning (3-6) CuSO^ + H20 >Cu° + H2SO^ + 1/2 02 - 23 -had a number of weaknesses inherent in atrmine leaching; namely slow leach rates and difficulties in obtaining 98% copper extraction. Also, sulphide sulphur oxidation through to sulphate is inefficient for two reasons: high oxygen consumption high lime neutralization requirements to fix sulphate as gypsum. In addition, effective recovery of precious metals, and recovery/disposa 1 of minor metals, was never proven. The process relied on an expensive solvent extraction step to transfer copper from an ammine medium to sulphate medium so as copper could be electrowon. All of the above deficiences were reflected in a high energy consumption requirement, which one study^6 estimated at 72.4 MJ/kg Cu. This figure is roughly double the energy requirements of the chloride leach processes. To their credit, the Arbiter Process developers have subsequently suggested improvements. Arbiter and Milligan-^ studied the feasibility of using sulphur dioxide to reduce the Cu(NH3)/42+ complex directly to copper metal, thereby bypassing the energy-intensive solvent extraction and electro-winning steps of the old process. The proposed reduction would be carried out in two stages: - 24 -2Cu(NH3)^SO^ + 2S02 + 4H2Q 30 °C > Cu 2 SO3 • (NHzf) 2 SO3 + 3(NH^)2SO^ (5) Cu2SQ3-(NH/t)2SQ3 , °C>2Cu° + (NH^)2SO^ +so2. (6) A purer product than that obtained by solvent extraction-el ect r owi nn i ng is claimed, however, further refining would still be necessary. 1.5 NITRATE ROUTES The action of nitric acid and mixed nitric-su1phuric acids on copper sulphide minerals has been studied by a number of investigators38-^0. Habashi39 reported the reaction between chalcopyrite and nitric acid as being highly exothermic, and represented by the following equation 3CuFeS2 + 2OHN03 >3Cu(N03)2 + 3Fe(ND3)3 + 6S° + 5N3>|v + 10H2O (7) with a portion of the sulphur being oxidized: S° + 2HN03 »H2SCty + 2NOyjc . (8) If a stoichiometric amount of acid is used, iron will pre cipitate as ferric oxide when the free acid level becomes low: - 25 -6CuFeS2 + 22HND3 »6Cu(ND3)2 .+ 3Fe203 + 12S° + lONO^ + 11H20. (9) If mixed nitric-su1phuric acid solutions are used, nitric acid plays the role of oxidant while sulphuric acid acts as a sulphate source to solubilize the cationic species. From a practical standpoint, this route is more attractive than using solely nitric acid due to the difficulty of winning copper from nitrate solutions. The dissolution of chalcopyrite in a mixed nitric-su1phuric acid solution can be represented as follows38'^0 6CuFeS2 + 22HN03 + 9H2SO^ ->6CuSO^ + 3Fe2(SO^)3 + 22ND^ + 6S° + 20H20 (10) with elemental sulphur undergoing partial oxidation as depicted by equation (10). If the free acid level falls low enough, iron will precipitate as hydronium jarosite: 3Fe3+ + 2SO/42" + 7H20 >H3OFe3(SO^)2(OH)6 + 5H+. (11) The extent of sulphur oxidation to sulphate, as opposed to oxidation to elemental sulphur depends on a number of factors. Prater et al.38 reported that increasing the concen tration of either H2SO^ or HND3 much beyond the stoichiometric requirement, or increasing leach temperature, appears to increase the extent of sulphate formation. Also, nitric - 26 -or nitric-su1phuric media attack most sulphide minerals includ ing pyrite and molybdenite. Mineral reactivity, along with elemental sulphur yield varies widely as shown in Table 4. Pyrite gives a very low S° yield, and chalcopyrite follows the general trend of being the most difficult sulphide to 1 each. 1.5.1 The Nitric-Sulphuric Copper Process A nitric-su1phuric leach process was patented by the E.I. duPont de Nemours Company in and evaluated extensively by the Kennecott Copper Corporation^2'^3. A schematic flowsheet of the process is presented in Figure 5, and the important reaction chemistry is summarized in Table 5. The process utilizes N02 gas as the chief oxidant in a two-stage counter cur rent leach conducted at 100 °C (equations 5-1 to 5-3), to obtain 98% copper extraction in a total 5 hour residence time. By-product sulphides such as pyrite and molybdenite are leached quantitatively along with chalcopyrite, and roughly 20% of the sulphide values are converted to elemental sulphur (when treating a mixed CuFeS2 - FeS2 concentrate). NZ>2 dissolves reversibly in aqueous acidic media according to the following step-wise mechanism^3: 2ND2(g) ^==^ 2ND2(aq) 2ND2(aq) ^ NO+ + ND3" (12a) (12b) - 27 -TABLE 4 Relative Reactivity and Elemental Sulphur Yield of Pure Minerals Leached with Nitric Acid38 Mineral Relat ive S° Yield, % of React i v1ty Reacted Sulphur Cu3SbS3 10 40 Cu2S 8 75 FeS 7 60 FeS2 7 3 CU9S5 6 55 CuS 6 50 Cu^FeS^ 5 60 CuFeS2 2 45 HN03 to 2nd-Stage Leach CONDENSER NO r OXIDIZER Concentrates NO.HzO srs NO.HjQ r FIRST-STAGE LEACH sV -r1\ SECOND-STAGE LEACH S\L Au and Ag Recovery V V Au Ag so GangUe Waste Product NH3 M0O3 \ t i 1 NO Mo Nitrate Recovery Removal NHa LA Purge Liquor Waste A Scrap Fe Zinc Recovery J Copper Purge Recovery i ^ CaCO Sulphate Removal Zn Product Cement Cu Iron Removal Sele Rem nium oval . Jarosite Waste Gypsum Waste Se(Cu) ro co ELECTROWINNING t Cathode Copper FIGURE 5 Schematic Flowsheet of the Nitric-Sulphuric Copper Process' - 29 -TABLE 5 Chemistry of the Nitric-Sulphuric Copper Process  Leaches (5-1) 2CuFeS2 + 5N02 + 5H2SO^ >2CuSO^ + Fe2(S04)3 + 4S° + 5H20 + 5NO^ (5-2) 2FeS2 + 3N02 + 3H2S04 >Fe2(SOff)3 + 4S° + 3H20 + 3NO^ (5-3) So + 3N02 + H20 >H2SO^ + 3NO^ Oxidizer (Reagent Regeneration) (5-4) 2NO + 02 >2N02 Nitrate Removal (5-5) N03" + 4H+ + 3Fe2 + >NO^ + 2H20 + 3Fe3 + Iron Remova1 (5-6) 3Fe3+ + NH3 + 2S0^2- + 6H20 > NH^Fe3(SO^)2(OH)6 + 5H+ Selenium Removal (5-7) H2SeQ3 + 4Cu° + 4H+ >Cu2Se+ 2Cu2+ + 3H20 - 30 -TABLE 5 Chemistry of the Nitric-Sulphuric Copper Process (contd.)  E1ect r owi nn i ng; (5-8) CuSCty + H20 >Cu° + H2SO/4 + 1/2 02 Sulphate Removal (5-9) H2SO^ •+ CaC03 + H20 >CaSO^-2H20 + C02 - 31 -ND+ + H20 (H2ND2 + ) HND2 + H+ (12c) HND2 + N32(aq) ^ HND3 + ND(g) . (12d) The highly reactive N0+ ion forms as an intermediate and is most likely the reactive species during leaching. The NO gas formed during the leach is collected, re-oxidized to ND2 and recycled back to the leach. Because the nitric-su1phuric system solubilizes such a wide variety of metals, pregnant leach liquors must undergo a series of rigorous purification steps before copper can be recovered. Residual nitrate is removed by reduction to NO with ferrous iron; iron is precipitated as a jarosite (which also removes Al, As, Bi, Sb, and partially removes Se, Te, S); selenium and tellurium are cemented out with copper metal; and the purge liquor still contains residual Cd, Co, Mg, Mn, Ni and Zn. Copper is electrowon at high current density (650 A/m2) in air-agitated cells to achieve a product which is comparable in purity to e1ectrorefined copper. The nitric-su1phuric leach process has the advantage of being able to rapidly solubilize all types of copper sulphides. Unfortunately this process produces a gaseous effluent (from the nitric acid recovery plant) which must be dealt with. The efficiency of sulphide to sulphur conversion is decreased for pyritic concentrates. The process is also a high energy consumer with an estimated 81.2M3/kg Cu required, which is at least 2.5 times greater than that required for flash smelting^. High energy consuming operations are elec-- 32 -trical energy for e1ectrowinning (43% of total) and steam generation for the purification steps (25% of total). The capital cost for a 45,000 tonne Cu/year facility was estimated to be $100.5 million (1978 U.S. dollars); the operating cost was estimated to be 67.9^/kg Cu. 1.6 OXIDATIVE SULPHATE ROUTES The oxidative dissolution of copper sulphide concentrates in various sulphate media has been studied extensively and is well documented in the literature. Processes have been proposed based on a number of different sulphate systems, and can be grouped into four different categories - hot con centrated sulphuric acid, dilute sulphuric acid and oxygen, dilute sulphuric acid and ferric iron, and biological systems. 1.6.1 Concentrated Sulphuric Acid Prater et al.*** studied the chemistry of sulphating, or acid-baking copper-iron sulphides, with hot concentrated H2SO4 (93-98% pure). Using chalcopyrite as an example, the overall reaction chemistry can be described as being inter mediate between the following two reactions CuFeS2 + *H2SC>4 ^CuSO^ + FeSO^ + 2S° + 2S02 +4H20 (13) - 33 -CuFeS2 + 5H2SO^ >CuSO^ + 1/2 Fe2(SO^)3 + 2S° + 5/2 S02 + 5H20 (14) with the latter reaction, depicting oxidation of iron to the ferric state, being favoured by higher reaction temperatures (200-265 °C) and longer retention times (greater than 2 hours). Sulphation is carried out in an externally heated rotary kiln, with 99% chalcopyrite decomposition achieved in about one hour. Pyrite and molybdenite are not attacked. If the reaction temperature is kept below 230 °C and only the stoich iometric acid requirement is used, S° oxidation is kept to a minimum. The Anaconda Company piloted a process based on chal copyrite sulphation (the Anatread Process) in 1971 which treated 5 tonne/day concentrate^5. The process suffered from the usual copper-iron separation problem and in the regeneration of process sulphuric acid, but was claimed to be economically competitive with smelting at the time. 1.6.2 Dilute Sulphuric Acid and Oxygen Oxygen under pressure in weakly acidic solution can act as an effective oxidant for chalcopyrite. Vizsolyi and co-workers^ at Sherritt Gordon Mines developed an acid pres sure leach process for chalcopyrite concentrates. In this process, chalcopyrite was ground to -325 mesh and leached with 95 g/L H2SO^ under 1.4-3.5 MPa 02 pressure; at 115 °C for 2-3 hours to obtain 98% copper recovery and 85% sulphur - 34 -recovery as S°. The overall leach reaction can be approximated as follows: CuFeS2 + H2SCy + 5/4 02 + 1/2 H20 >CuSO^ + Fe(OH)3 + 2S°. (15) Pyrite and molybdenite were unattacked. Leaching with a 25-50% stoichiometric excess of con centrate over acid was recommended to ensure hydrolysis of iron. However, unleached concentrate must then be recycled back to the leach step after S° removal. Other problems with this process were - pyrite would build up in the sulphide recycle stream, the iron precipitate was gelatinous and entailed a difficult liquid-solid separation, and high copper and precious metals losses were incurred in iron slimes. 1.6.3 Dilute Sulphuric Acid and Ferric Iron The mechanism of acidic ferric sulphate leaching of chalcopyrite has been the subject of widespread interest, and as with ferric chloride leaching, is still shrouded in controversy8'9'^. The leach reaction is usually expressed as CuFeS2 + 2Fe2(SO^)3 >CuSO^ + 5FeSO^ + 2S° (16) with some sulphur oxidation to sulphate occurring. In general, the copper leach rate or extraction is not nearly as high as with ferric chloride leaching8'9. This has been partially - 35 -attributed to the observation that a tenacious, impermeable sulphur layer forms around partia 11y-1eached chalcopyrite grains during ferric sulphate leaching, which severely impedes the reaction progress. One solution to this problem was proposed by Beckstead et al.^8 who found that attrition grinding of chalcopyrite concentrate to a median particle size of 0.5 um allowed 90% copper extraction to be attained in a 3 hour leach at 93 °C. However, such grinding is energy intensive (7.9 M3/kg Cu) and is likely to result in severe liquid-solid separation problems. A process for treating copper concentrates by ferric sulphate leaching was developed and piloted in Polan However, the concentrate feed was primarily a mixture of the less refractory minerals chalcocite and bornite. Retention time in the two-stage countercurrent leach was still 9 hours at 90-95 °C. Thus, while weakly acidic ferric sulphate would be attractive from a corrmercial standpoint because it is relatively cheap, non-corrosive and easy to regenerate, it is not an aggressive enough lixiviant for chalcopyrite. 1.6.4 Biological Systems The oxidation of reduced sulphur compounds and ferrous iron by the leaching bacterium Th i obac i11 us fer rooxi dans, resulting in metals solubilization and weak acid generation, has long been recognized as a naturally occurring process that can be exploited in mining operations50*5*. This unique - 36 -bacterium derives energy for growth by the oxidation of sulphide minerals and soluble ferrous iron. The bacterium requires a source of ammonia nitrogen as well as small amounts of phosphate, calcium and magnesium, and fixes carbon from carbon dioxide in the air. It is able to withstand extremely high concentrations of such metals as copper (60 g/L) and zinc (100 g/L) without apparent ill effect, and functions best at acidic pH's (1.5-2.5) and a temperature of 35 °C. Much evidence exists which suggests that T. ferrooxidans can attack sulphide minera 1s by direct attachment and' oxidation of the sulphide moiety32-33. Using chalcopyrite and pyrite as examples, the overall reactions can be represented as 12CuFeS2 + 5102 + 22H20 >12CuSO^ + 4H3OFe3(SO^)2(OH)6 + 4H2SO^ (17) 12FeS2 + 4502 + 34H20 > 4H3OFe3 ( SO^ ) 2 (OH) g + 16H2S04 (18) with iron precipitating as hydronium jarosite in the pH range favoured by the bacteria. Although bacterial leaching has been considered most applicable, from a practical standpoint, to the treatment of low-grade ores by dump or heap leaching, at least two copper concentrate bioleach processes have been proposed36'3^. However, concentrate bioleaching suffers from a number of pr ob 1 ems : - 37 -low leach rates (300-600 mg/L-h Cu). low one-pass extractions (40-80%), making concentrate re-grinding and re-cycling necessary. sulphide values are oxidized to sulphate, resulting in large volumes of weak acid which must be neutralized. On the positive side, bioleaching is carried out at near ambient temperature (35 °C) and pressure in a chemically non-aggressive lixiviant. A recent study indicated that a bioleach process may be economically competitive with smelting if used on a small scale (25,000 tonne/year Cu or less)58. 1.7 SUMMARY From the forgoing description of copper hydrometal-lurgical processes, it is clear that the chloride routes are superior from a practical standpoint. The superiority of the chloride routes can be attributed to a combination of the excellent leach rates and extractions obtained, and to the lower overall energy requirements. However, all of the processes have a common major weakness inherent in oxidative leaching - difficulty in achieving an effective copper-iron separation. Oxidative leaching, whether it be in chloride, ammine, nitrate or sulphate media, dissolves copper and iron simultaneously with iron then precipitated from solutions - 38 -high in copper. As a result, up to 5% of the copper is lost due to co-precipitation with iron. In addition, any precious metals usually end up as dilute constituents in the iron residue, making recovery difficult or impossible. A potential solution to the copper-iron separation problem is to leach chalcopyrite under reducing conditions, in which iron can be selectively dissolved. Reduction leaching of chalcopyrite has only received interest in the past 10 years, with the majority of studies being of an academic nature. For this reason it was decided to study the practical applications of reduction leaching. CHAPTER 2 REDUCTION LEACHING AS A METHOD OF IRON REMOVAL 2.1 BACKGROUND Figure 6 portrays the Pourbaix diagram for the Cu-Fe-S-H20 system at 25 °C59. It can be seen that at near neutral or negative potentials at pH's less than about 2, ferrous iron will tend to be stabilized in solution, leaving copper in a copper-rich sulphide form such as bornite, chal-cocite, or; at low enough potentials, metallic copper. Thus, based on purely thermodynamic considerations, leaching of chalcopyrite under reducing conditions should selectively remove iron. The cathodic reaction for chalcopyrite decom position to chalcocite can be expressed as follows: chalcopyrite was first demonstrated by McGauley et al. who leached copper concentrates with copper sulphate at elevated temperatures. The essential chemical reactions appeared to be: 2CuFeS2 + 6H+ + 2e- >Cu2S + 2Fe2+ + 3H2S. (19) Historically, the selective leaching of iron from CuFeS2 + CuSO^ CuFeS/j + CuSO^ >2CuS + FeSO^ »2CuS + 2Cu2S + FeSO^. (20) (21) - 40 -FIGURE 6 Eh-pH Diagram of the Cu-Fe-S-H20 System at 25°C (Conditions: 0.1 M Fe,S Species; 0.01 MCu Species)59 - 41 -However, leaching under these near neutral conditions did not prove very effective, as iron extractions achieved were low even when using high temperatures (180-200 °C) and long residence times (4-12 h). Several investigators have studied the acid reduction leaching of chalcopyrite with metals having lower rest potent ials, such as copper, iron or lead61-65. They found that the reaction was controlled by galvanic interaction of chal copyrite with the metal. This galvanic mechanism will be discussed in greater detail in Chapter 4. Hiskey and Wadsworth6* studied the kinetics of reduction leaching with fine copper powder in the temperature range 34-90 °C. The reaction observed was: CuFeS2 + Cu + 2H+ >Cu2S + Fe2+ + H2S^ (22) They found that relatively high iron extractions, in the range 80-90%, could be obtained in as little as 1 hour when leaching at very low pulp densities (1%) and using stoich iometric excesses of acid and copper. McKay and Swinkels patented a copper process based on reaction (22)62. The process successfully treated a wide range of -325 mesh concentrates containing 7 to 41% Cu and 14-31% Fe,,'using -400 mesh copper powder. Optimum leach conditions were reported to be pH 1.0-1.5, 90-95 °C leach temperature and a residence time of 1-6 h. Iron extractions obtained ranged from 66-87% depending on the type of con centrate used. The formed chalcocite is th,en separated from - 42 -the ferrous sulphate solution by filtration and is readily dissolved by oxygen pressure leaching, as follows: Cu2S + 2H2SO^ + 02 >2CuSO^ + S° + 2H20. (23) Iron is precipitated as a pure jarosite in a separate step. The formed H2S gas is absorbed in a copper sulphate solution to form copper sulphide Cu2+ + H2S >CuS + 2H+ (24) which is directed to the oxygen pressure leach step. Sulphides such as pyrite and molybdenite are not leached, and end up in the final leach residue along with sulphur and all precious me t a 1s. The McKay/Swinke1s process has the advantage of precipitating iron in a separate step, from a copper-free solution to yield a pure jarosite which is environmentally acceptable for disposal. Also, precious metals are not tied up in dilute form in the precipitated iron, as is the case with most oxidative processes, and as such are more easily recoverable. Sulphur is recovered in the preferential elemental form. Drawbacks to the process are that it produces noxious H2S gas and doesn't achieve a complete chalcopyrite to chalcocite conversion under the process conditions employed. Shirts et al.63»6^ and Nicol65 studied the acid re duction leaching of chalcopyrite with copper, iron and lead. - 43 -Reactions observed when using iron and lead were: 2CuFeS2 + Fe + 6H+ >Cu2S + 3Fe2+ + 3H2S (25) CuFeS2 + Fe + 4H+ >Cu + 2Fe2+ + 2H2S (26) 2CuFeS2 + Pb + 6H+ >Cu2S + 2Fe2+ + Pb2+ + 3H2S (27) CuFeS2 + Pb + 4H+ > Cu + Fe2+ + Pb2+ + 2H2S. (28) Metallic copper generally gave a more rapid rate of decomp osition in the first 15 minutes, but at times greater than 1 hour metallic iron or lead gave a more complete decomposition. Sohn and Wadsworth66 have investigated the reduction leaching of chalcopyrite in copper sulphate solutions using gaseous S02 as the reductant. The overall reaction could be expressed as: CuFeS2 + 3Cu2+ + 2S02 + 4H20 >2Cu2S + 6H+ + 2HSCV + Fe2 + . (29) The leach was found to be electrochemical in nature, with the cathodic reactions involving formation of an intermediate bornite product, as follows: Cathodic Reactions: 2CuFeS2 + 3Cu2+ + 4e" v "•Cu^FeS^ + Fe2 + Cu5FeS4 + 3Cu2+ + 4e" v s 4Cu2S + Fe2 + (30) (31) - 44 -Anodic Reaction: S02 + 2HzO k N3H+ + HSO^- + 2e" (32) Chalcopyrite conversion was incomplete, with a maximum 82% iron extraction obtained only when leaching very fine (50% -2um) concentrate at 90 °C for 4 h. Biegler et al.6^-69 have investigated the electro lytic reduction of chalcopyrite in HC1 solution, and con firmed the formation of chalcocite or djurleite according to the reaction 2CuFeS2 + 6H+ + 2e" ;==^ Cu2S + 3H2S + 2Fe2+ (33) for which the calculated standard potential is -0.14 V. Current efficiency decreased drastically as chalcopyrite reduction proceeded, and was mainly attributed to hydrogen evolution on the forming chalcocite. A copper extraction process was proposed based on continuous electrolytic red uction in a slurry electrolysis cell69. Investigators in the reduction leaching of chalcopyrite have shown that the method can achieve an effective, simple copper-iron separation. However, relatively little work has been done in developing reductive leach processes as practical alternatives to the oxidative leach processes. Two companies, Sherritt Gordon Mines Ltd. and Cominco Ltd., did develop a a process in which iron was preferentially separated from - 45 -chalcopyrite by a combination of pyrometa 11urgica1 and hydro-metallurgical methods. To illustrate the practical benefits of achieving an early copper-iron separation, this process will be described in some detail next. 2.2 THE SHERRITT-GOMI NCO COPPER PROCESS The Sherritt-Cominco Copper Process (hereafter referred to as the S.C. Process) is the product of a joint research development effort by Sherritt Gordon Mines Ltd. and Cominco Ltd70. Piloted in 1975-1976 at Fort Saskatchewan, Alberta70, the process was demonstrated to have a number of innovative features which make it a potentially attractive and viable alternative to the best hydrometa 11urgica1 methods available. A simplified schematic flowsheet of the S.C. Process is presented in Figure 7. The process solves the copper-iron separation problem in a novel manner - iron is preferentially extracted from copper in a series of three unit operations, and disposed of as jarosite in a separate step. The copper-rich leach residue is then leached with weak sulphuric acid under oxygen pressure followed by purification of the pregnant leach liquor and copper e1ectrowinning. Sulphur is recovered primarily in the elemental form. The process allows for recovery of precious metals and disposal or recovery of any minor metals. The following process description is based on pilot plant results on a chalcopyrite/pyrite concentrate assaying 23.7% Cu, 27.6% Fe, 3.16% Zn and 31.5% S. - 46 -COPPER CONCENTRATES THERMAL ACTIVATION 1%h Retention Time I T PRECIOUS METALS CONCENTRATES SOLID SLURRY COPPER METAL FIGURE 7 Simplified Schematic Flowsheet of the Sherritt-Cominco Copper Process71 - 47 -2.2.1 Copper-Iron Separation Stages Iron is separated from copper in three stages consisting of thermal activation, acid leaching and activation leaching to attain an overall 90% or greater iron rejection before the copper leach step. Iron is disposed of as jarosite. (a) Thermal Activation Thermal activation is essentially a pyrometa 11urgica1 step in which pelletized copper concentrate is pretreated at 700 °C in a 12-stage multiple hearth roaster to convert chalcopyrite to bornite and any pyrite to pyrrhotite. The effective stoichiometry of the reaction can be represented as foilows: -»5CuFeSK8 + S02t (34) ->Fe7Sg + 6S02/r (35) >Cu^FeS^ + 4FeS + H2S^ (36) *7FeS + H2Sf. (37) Depending on the pyrite content of the feed, about 25% of the sulphur is removed as S02 and/or H2S and recovered as H2SO^ and S° respectively, by conventional methods. Unfortunately the conversion reactions do not go to completion at 700 °C, so that thermally pretreated concentrate Roaster Top C 5CuFeS2 + 02(g)— -S02 prod. C 7FeS2 + 602(g) Roaster BottomC 5CuFeS1>8 + H2(g) -H2S prod. L Fe7Sg + H2(g) - 48 -is not completely free of chalcopyrite. Arsenic, if present in the feed, is not volatilized substantially at 700 °C and remains in the activated solids. (b) Acid Leach The activated solids are acid leached at 85 °C with weak acid obtained from jarosite precipitation, the activation leach and the acid plant (overall 200-300 g/L H2SO^), in a two-stage countercurrent system to dissolve pyrrhotite FeS + H2SO^ »FeSO^ + H2S^ (38) and produce a liquor high in iron (65 g/L Fe++) and low in acid (30 g/L H2SO^). Overall iron extraction at this stage is about 75%. If zinc is present, it will be partially dissolved and can be recovered by precipitation as ZnS with H2S. (c) Activation Leach To obtain an overall 90% or greater iron extraction before copper is leached, the S.C. Process utilizes a so-called "activation" or "neutral" leach on the residue from the acid leach. The activation leach can be originally attributed to McGauley et al.60 As applied to the S.C. Process, the essential chemical reactions are: - 49 -Cu5FeS^ + CuSO^ >2CuS + 2Cu2S + FeSO^ (39) CuFeS2 + CuSO^ > 2CuS + FeSO^ (40) Copper sulphate lixiviant (52 g/L Cu2+, 12 g/L Fe2+, 54 g/L H2SO^) is supplied as a bleed stream from solution from the oxidation leach. An undesirable side reaction also takes pi ace 5CuS + 3CuSO^ + 4H20 >4Cu2S + 4H2SO^ (41) which requires that the CuSO^ recycle be 20-30% in stoich iometric excess. The activation leach is carried out at 156 °C for 4 hours, and increases overall iron extraction from 75% to 91% to yield a residue assaying 4% Fe and 53% Cu. Zinc will be partially solubilized, if present, and can be recovered by H2S precipitation. Pyrite will not be attacked by this leach. Solution from the activation leach (3 g/L Cu2+, 30 g/L Fe2+, 79 g/L H2SO^) is recycled back to the acid leach after zinc recovery. If the copper concentrate is rich in bornite and/or chalcocite rather than pyrite, the activation leach on its own may be able to produce an adequate copper-iron separation, without the need for thermal pretreatment and acid leaching. 2.2.2 Jarosite Precipitation Solution from the acid leach, slightly diluted from - 50- -residue washing, is subjected to conventional jarosite pre cipitation in a four-compartment autoclave with the reaction temperature increased from 120 °C in compartment 1 to 190 °C in compartment 4. The overall reaction is: 6FeSO^ + 2NH4OH + 7H20 + 3/2 02 >2NH^Fe3(SO^)2(OH)6 + 2H2SO^. (42) An oxygen partial pressure of 0.7-1.8 MPa is used; total retention time is 45 minutes. Solution from jarosite pre cipitation (5 g/L Fe, 50 g/L H2SO^) is recycled to the acid leach. About 33% of the initial feed sulphur is lost to jarosite. 2.2.3 Copper Recovery Stages (a) Oxidation Leach Residue from the activation leach consists primarily of bornite, chalcocite and covellite, with minor amounts chalcopyrite and pyrite. To recover copper, this residue is acid leached under oxygen in a continuous two-stage counter-current operation, followed by a final batch leach under more aggressive conditions to scavenge as much copper as possible. The overall leach reactions can be summarized as foilows: - 51 -Cu5FeS^ + 6H2SO4 + 302 ^5CuSO^ + FeSO^ + 4S° + 6H20 (43) Cu2S + 2H2SO^ + 02 >2CuSO^ + S° + 2H20 (44) CuS + H2SO^ + 1/2 02 ^CuSO^ + S° + H20. (45) Acid is supplied by return electrolyte from copper electro-winning and is supplemented with concentrated acid. The oxidation leach is conducted at 100 °C for a total retention time of 7'A hours; oxygen pressure is 1.05 MPa. Acid concentrations are 150 g/L for the continuous phase and 250 g/L for the batch scavenging leach. Overall copper extraction is 98.4%, with most of the remaining iron and zinc (if present) also solubilized. The pregnant copper solution (85 g/L Cu2+, 6 g/L Fe, 20 g/L H2SO^) is directed to the purification step. (b) Solution Purification Pregnant leach liquor from the oxidation leach must undergo purification to remove trace impurities such as Se, Te, As, Sb, Bi and Pb which would co-deposit with copper during e1ectrowinning. Also it is desirable to lower the iron level as much as possible, as iron reduces current effic iency in e1ectrowinning. The purification method chosen was to subject the solution to high temperature oxydrolysis (similar to jarosite precipitation) which will remove iron as Fe20-j along with adsorbed trace impurities. - 52 -Prior to oxydrolysis, however, any solubilized selenium must be quantitatively reduced to the tetravalent state as the selenite anion (Se032-) because the hexavalent form (SeCty2-) will not precipitate during oxydrolysis. Selenium reduction is carried out in the first chamber of the four-compartment oxydrolysis autoclave. By excluding oxygen from this chamber, the selenate anion can be reduced to selenite by ferrous iron: SeO^2" + 2Fe2+ + 2H+ >Se032" + 2Fe3+ + H20. (46) Reaction conditions of 200 °C for about 5 minutes are adequate to complete the reduction. The solution then undergoes oxydrolysis at 200 °C under 0.35 MPa 02 for a further 15 minutes to precipitate iron as ferric oxide (probably hydrated): 2FeSO^ + 1/2 02 + 2H20 > 2H2SOif + Fe203. (47) Minor impurities are precipitated or adsorbed concurrent with iron precipitation. The purified solution contains typically only 1 g/L Fe and 0.4 mg/L Se-r and is ready for copper e1ectrowinning. (c) Copper Electrowinning Copper is electrowon from the purified pregnant leach liquor (85 g/L Cu2+, 30 g/L H2SO^) in air-sparged cells at - 53 -high current density (330-650 A/rrr) and elevated temperature (50 °C). Spent electrolyte containing 30 g/L Cu2+ and 130 g/L H2SO4 is recycled to the oxidation leach. Pure cathode copper suitable for sale is produced at a current efficiency of 90% or greater . 2.2.4 Precious Metals and Sulphur Recovery Residue from the oxidation leach contains elemental sulphur together with precious metals, gangue and any unreacted sulphides (including M0S2 which is not solubilized by any of the leaches). Gangue is rejected by flotation, followed by sulphur removal by solvent extraction (xylene at 100 °C). The end product is a rich precious metals concentrate con taining typically 6,600 g/t Ag and 275 g/t Au, and can be sold or further processed. Recovery of Ag and Au based on concentrates is 96%. 2.2.5 Energy Consumption Energy consumption for the S.C. Process is contrasted with flash smelting in Table 672. As can be seen, the S.C. Process follows the trend of most hydrometa11urgica1 processes in being energy intensive, and cannot compete with flash smelting if the latter has an acid market. However, if SO2 produced by a smelter must be fixed as elemental sulphur or neutralized sulphate, then the energy gap between the two processes is virtually eliminated. - 54 -TABLE 6 Comparison of Energy Requirements Fos s i1 Fuel* kcal/kg Cu Elect r i c Powe r # Total S.C. Copper Process 6320 3650 9970 Flash Smelting/Electro-refining with Acid Production 3290 1280 4570 Flash Smelting/Electro-refining with Sulphur Production f r om SO2 9300 to 11000 * Natural gas calorific value = 8900 kcal/std m3 # Electric power = 860 kcal/kWh Reasons for the higher energy requirement for the S.C. Process are - the e1ectrowinning step which accounts for virtually all of the electric power requirement, the heating of large solution volumes, and the inefficient recovery and recycle of process heat. 2.2.6 Economi cs A detailed cost estimate was prepared for the S.C. Process and compared to flash sme 11 i ng/elect roref ini ng costs'72. The estimate was based on a 68,000 tonne Cu/year facility built at a Western Canada location with costs in 1977 Canadian dollars. No credits were given for the elemental sulphur - 55 -recovered in the S.C. Process or the by-product acid produced in a flash smeIter. The analysis concluded that an S.C. plant could be built for 172 million dollars versus 195-205 million for a comparable flash sme 11er/e1ectrorefinery. Operating cost for the S.C. Process was estimated at 31.5^/kg versus 28.4^/kg for smelting with acid production, or 40.0^/kg for smelting with acid neutralization. A major operating cost expense for the S.C. Process is energy consumption, which is about 9.1f£/kg Cu as opposed to 4.2^/kg for smelt ing/el ect r or ef i n i ng with acid production. 2.2.7 Surrmar y The S.C. Copper Process retains all the key advantages boasted by the best of the other copper hydrometa 11urgica1 processes, namely 98% or greater recovery of copper, conversion of sulphur to S°, disposal of iron as an innocuous solid, flexibility in treating varying grades and types of copper concentrates, and recovery/disposal of minor metals. However, the S.C. Process can claim these further advantages: Negligible copper loss to jarosite. Pure cathode copper is easily obtained. Precious metals are recovered quantitatively into a P.M. concentrate, from which they can easily be recovered. Copper is leached under mild conditions. - 56 -These advantages can all be attributed, directly or indirectly, to the fact that iron is selectively extracted from copper. However, three unit operations are required to extract 91% of the iron from the copper, including an objectional pyrometa 11urgica1 front-end step. The process cannot be called purely "hydrometa11urgica1", and as such may not be as clean as other hydrometa 11urgica1 processes. The number of unit operations required for the process is reflected in the high operating and capital cost. Clearly, if the copper-iron separation procedure could be simplified, this would lower substantially capital and operating costs. Maschmeyer et al.72 estimated capital and operating costs could both be reduced by 15% if thermal activation and acid leaching of iron could be eliminated, which is proposed for feeds high in bornite. The other obvious improvement to the S.C. Process would be to replace the e1ectrowinning step with hydrogen reduction. This could reduce energy requirements by roughly 20-25%. 2.3 PURPOSE AND SCOPE OF THE PRESENT INVESTIGATION Previous studies on the reduction leaching of chal copyrite have indicated that this method shows promise as an effective method of separating iron from copper, prior to copper dissolution. However, studies to date have been mainly of academic interest. Little work has been done to - 57 -develop optimal leach conditions which could be incorporated into an overall process. The practical advantages of selectively extracting iron from copper concentrates in a hydrometa 11urgical process have been demonstrated by the S.C. Copper Process. However, the S.C. Process resorted to using three unit operations to remove iron, one of which was pyrometa 11urgical. Reduction leaching is seen to be a potentially viable alternative which could be incorporated into the S.C. Process. Therefore, it was decided to investigate two novel methods of reduction leaching of chalcopyrite. Referring back to Figure 6, it can be seen that there is a small area above the hydrogen-water line, at about pH O or less, in which chalcocite and ferrous iron are stable with respect to chalcopyrite. This region of stability is expanded to slightly higher pH's at higher temperatures, as shown by Baratin?3 jn n-iS 200 °C Pourbaix diagram. This suggests that hydrogen gas can act as a reducing agent for chalcopyrite. Furthermore, if the reduction leach is carried out in copper sulphate solutions, evolution of H2S gas can be avoided. The overall reaction can be postulated as follows: CuFeS2 + 3Cu2+ + 2H2 >2Cu2S + Fe2+ + 4H+. (48) In addition, metallic copper should be an alternative reductant under similar conditions: CuFeS2 + Cu2+ + 2Cu° >2Cu2S + Fe2+. (49) - 58 -This work will be concerned with studying the potential practical applications of the above two reactions. The focus will be on optimizing iron leach rates and extractions; for this reason experiments will be conducted in strong copper sulphate solutions at elevated temperatures and pressures. CHAPTER 3 EXPERIMENTAL 3.1 MATERIALS 3.1.1 Copper Concentrates Reduction leach experiments were carried out on four different copper concentrates, obtained from the Phoenix (now shut down) and Bethlehem mines located in British Columbia, and from the Fox Lake and Ruttan mines in Manitoba. The as-received concentrate samples were kept stored in air tight tins or plastic bags when not in use, to minimize surface air oxidation. Information supplied by the respective mines indicated that all concentrates contained primarily chalcopyrite. In addition, the Bethlehem concentrate contained substantial bornite whereas the Phoenix, Fox Lake and Ruttan concentrates contained pyrite. The latter two also contained minor amounts of sphalerite (ZnS). Table 7 presents results of elemental analyses, calculated approximate sulphide mineral compositions, and screen analyses (Tyler series sieves) on the concentrates. Approximate mineral compositions were calculated from Cu and Fe analyses by assuming that only the above mentioned sulphide minerals were present. For example, with the Bethlehem concentrate, let x = moles CuFeS2 present and y = moles Cu^FeS^ present. Then, on a 100 g basis: - 60 -TABLE 7 Copper Concentrates Used Phoen i x Beth 1ehem Fox Lake Ruttan Elemental Analysis % Cu 24.3 32. 7 26.4 28. 2 % Fe 32.9 20. 0 33. 1 31.7 % S 31. 0* 26. 0 35. 6 35. 9 % Zn - - 3.75 2. 02 Calculated Mineral Content** % CuFeS2 70 59 75 82 % FeS2 25 - 20 14 % Cu5FeS^ - 20 - -% ZnS - - 5 3 Particle Size Distribution (%) Mesh +100 0.2 0.3 5.0 0.2 Size +140 12.7 5.9 - -(Tyler +200 18.7 10.7 - -Series) +270 19.9 23.7 - -+ 400 13.8 23. 8 - --400 34.7 35.6 72.4 64.9 Total 100. 0 100. 0 Sulphur assay is lower than expected because extensive surface oxidation with time had occurred for this very old concentrate. Does not values . consider minor sulphide minerals and gangue - 61 -total moles Cu = x + 5y = 32.7/63.54 total moles Fe = x + y = 20.0/55.85. Solving for x and y, we get: x = 0.319 moles CuFeS2 = (0.319 m) (183.52 g/m) = 59% CuFeS2 y = 0. 039 moles Cu5FeS^ = (0. 039 m) (501.80 g/m) = 20% Cu5FeS^. Leach experiments were carried out mainly on as-received material. However, to assess the effects of particle size on leaching, some experiments were performed on material carefully screened to the following size ranges: a) -140+270 mesh (53-105 um) b) -275 mesh (< 50 um) c) -400 mesh (< 38 um) 3.1.2 Sulphide Minerals Reduction leach experiments were also performed on six different sulphide minerals to assess their relative rates of reactivity. Minerals studied were pyrite, bornite, chalcopyrite, pentlandite (NiFeSj g), sphalerite and pyrrhotite (~Fen 9S). Massive rock samples, supplied by the Ward Museum and thought to be reasonably pure, were crushed and ground to -100 mesh size, after which portions were carefully sized to -270+325 mesh for use. Table 8 presents the results of elemental analyses on the mineral samples. With the exception - 62 -TABLE 8 "Pure" Sulphide Minerals Used Mineral Or i g i n Cu % Fe % S % Other % Pyr i te Quebec 43. 6 52. 2 Borni te Montana 53.7 14. 2 28. 3 -Cha1 copyr i te Quebec 20. 1 41. 2 36. 5 -Pent 1 andi te Sudbury - 31. 9 32. 1 29.5 Ni Pyr rhot i te Kimberley - 58. 1 35. 1 -Spha1er i te New York - - 34. 2 50.5 Zn of the chalcopyrite sample, the analyses correlate reasonably well with expected values. However, the chalcopyrite sample contained an inordinately high amount of iron, which SEM and X-ray diffraction analysis indicated was due primarily to the presence of pyrite. 3.1.3 Reagent s Reagent grade copper sulphate pentahydrate and armrionium sulphate were used for all leach experiments. Sulphuric acid used was 94-98% pure and was diluted as required. Reagent grade copper powder was used for all experiments. In most cases, very fine copper screened to -400 mesh was used. To assess the effect of initial copper particle size on leaching, -100+200 mesh and 90% -100+325 mesh copper sizes were used. - 63 -Standard cylinder grade hydrogen, nitrogen and carbon monoxide gases were used throughout. 3.2 APPARATUS Leach experiments were performed in two types of pressure autoclaves; a Parr series 4500 2L capacity reactor, and a 107 mL capacity shaking autoclave. 3.2.1 Parr Autoclave The Parr Autoclave assembly was equipped with controls for stirring, temperature, gas pressure and sampling. Due to the corrosive nature of some of the reduction leach exper iments, a 2L titanium reactor, 10 cm in diameter and 25 cm in height (interior dimensions), was used. Early experiments showed that a titanium stirring assembly still tended to corrode. An assembly consisting of a zirconium stirring shaft and tantalum impellors was tried and proved to be ade quately corrosion resistant. The autoclave stirring assembly was powered by a belt-driven 1/15 hp motor which could provide variable stirring speeds of up to 1,000 rpm by changing pulley sizes. For tests in which hydrogen gas under pressure was used as the reductant, two 7 cm diameter downward thrusting impellors, one situated 1 cm below the solution level and the other situated 1 cm above the bottom, were used. A high stirring speed of 750 rpm was used to ensure adequate gas dispersion into solution. For tests in which copper - 64 -powder was the reductant, solution pulp densities were much higher and a different impellor design was necessary to ensure good solids suspension. A single 9 cm diameter upward thrusting impellor, situated Yi cm above the bottom of the reactor, was used. A stirring speed of 250 rpm proved adequate for these tests. The reactor was heated by sliding it into an insulated, stainless steel sleeve which was electrically heated and pressure air cooled for temperature control. Accurate automatic temperature control was provided by a Yellow Springs Instruments Thermistemp controller (Model 71), which was connected to a thermister probe inserted into the autoclave temperature well. This arrangement controlled leach temperatures to i 1 °C. The temperature was measured and monitored by a chrome 1-alumel thermocouple wire, one end of which was inserted into the temperature well and the other end connected to a Sargeant-Welch strip chart recorder. If desired, the leach solution could be sampled during the course of a run via a titanium valve connected to a sampling tube, the end of which was situated about 3 cm above the autoclave bottom. To prevent solids from entering the sampling tube, a graphite plug or porous teflon filter was fitted to the end. Frequently, plugging problems were encountered during sampling, for two reasons - (a) leach solutions contained high concentrations of copper, iron and ammonium salts which tended to crystallize upon cooling and clog the sampling valve, and (b) considerable sulphide mineral particle size reduction occurred during leaching, and the ultra-fine solids - 65 -(<5 um) clogged the graphite or teflon filters. This sampling problem was never completely overcome. Gas pressure was measured using a 0-6.5 MPa (0-1000 psi) gauge situated on the top of the autoclave head. Pressures were corrected for the pressure of steam at the particular leach temperature. 3.2.2 Shaking Autoclave Smaller-scale leach experiments were performed in a home-made shaking autoclave set-up. A 107 mL capacity reactor constructed of corrosion resistant zirconium and equipped with temperature well, gas inlet and outlet was used. The reactor was heated by sliding it into an electric ally heated aluminum jacket. The temperature monitoring and control system used was identical to that used for the Parr autoclave, except pressurized air cooling was not required due to sufficient natural cooling by heat loss through the large autoclave head. Temperature control was again maintained within ± 1 °C with this system. Agitation was provided by a linear horizontal shaking motion at a rate of 288-3.8 cm strokes per minute, with the autoclave tilted at a 45° angle. Gas pressure was measured and monitored by a pressure transducer (Consolidated Electrodynamics Corporation, Model 4-313, 0-6.5 MPa range) coupled via a "T" connection to the gas inlet tube. The output signal from this transducer was monitored by a Sargeant-Welch strip'chart recorder. The 66 transducer was calibrated against steam pressure in the auto clave. The shaking autoclave had no provision for solution sampling during the course of a run. 3.3 EXPERIMENTAL PROCEDURE 3.3.1 Make-Up of Starting Leaching Solutions For all studies on reduction leaching with hydrogen gas, a starting leach solution of 63.5 g/L Cu2+, made with CuS04'5H20, was used. For all studies on reduction leaching with copper powder, a starting leach solution containing 90 g/L Cu2+ (as CuSO^-5H20), 20 g/L H2SO^ and 132 g/L (NH^^SO^ was used. Deionized water was used throughout. The solutions were made slightly acidic to prevent hydrolysis and precipitation of copper as antlerite, CuSO^•2Cu(OH)2, at high temperatures. In addition, some of the concentrates and minerals consumed acid initially (probably due to the presence of minor amounts of alkaline gangue and oxides). Arrmonium sulphate was always added to buffer the solutions, which in an overall process, results in improved jarosite precipitation of the iron and hydrogen reduction of copper. More will be said about this in Chapter 5. As the maximum room temperature solubility of copper sulphate in aqueous solution is about 90 g/L Cu2+} it was necessary to warm the ammonium sulphate buffered solutions to about 50 °C to achieve complete dissolution of all the - 67 -and ammonium salts. Solutions were then made up acc-to the desired volume in volumetric flasks, and irrmed-charged to the autoclave. 3.3.2 Parr Autoclave For all runs in the Parr autoclave, the experimental procedure used was as follows: 1) The required amounts of solids and leach solution were added to the autoclave, the autoclave was sealed, placed into the heater sleeve and stirring was initiated. 2) The autoclave was then heated. During the 10 minute warm-up period required to reach 100 °C temperature, nitrogen was flushed through the autoclave to purge air. At 100 °C, nitrogen flushing was stopped, the gas inlet and bleed valves were closed, and preparations were made to intro duce the appropriate gas. 3) At the desired reaction temperature, the approp riate gas (either nitrogen, hydrogen or carbon monoxide) was introduced at the desired pressure. 4) Solution sampling, if desired, was carried out as follows - at the appropriate time, the sampling valve was opened and about 10 mL of solution was flushed through and discarded. Then, about 5 mL of sample was discharged into a 50 mL volumetric flask. The sample was diluted to 50 mL from a burette of deionized water (acidified to pH 2.0 with H2SO4 to prevent metal hydrolysis). The original sample - 68 -volume was then calculated to the nearest 0.1 mL from the burette reading by difference, and the dilution factor taken as 50 mL * sample volume. Sampling from the Parr autoclave was not done for runs in which the leach temperature was greater than 120 °C, because of errors imposed by significant water loss from the sample due to flashing at higher temperatures. 5) To terminate a run, the gas inlet valve was shut, stirring was stopped, the reactor was removed from the heating jacket and quickly cooled by immersing in a pail of cold water. Once cooled, excess gas was vented from the reactor, the autoclave head removed and the contents pressure-filtered. The residue filter cake was washed thoroughly with pH 2.0 deionized water, dried at 60 °C, weighed, rolled thoroughly to homogenize and stored for subsequent analysis. The volumes of the filtered leach solution and wash water were recorded, and the filtrates stored for analysis. 3.3.3' Shaking Autoclave The shaking autoclave proved to be ideally suited for sma 11er-sea 1e runs in which solution volumes were less than about 70 mL. The experimental procedure used was identical to that described for the Parr autoclave, except that the reactor contents were suetion-fi1tered in a Buchner funnel, washed with pH 2.0 water, and the total filtrate made up to 250 mL volume in a volumetric flask. Data on leach behaviour with varying residence times could be rapidly obtained due - 69 -to the ease and rapidity with which one run could be terminated and the next run started. 3.4 ANALYTICAL METHODS 3.4.1 Solution Analyses Solution copper, iron, zinc and nickel values were determined, after appropriate sample dilution, by atomic absorption analysis using a Perkin Elmer model 306 spectro photometer. Metal concentrations reported were back-calculated to the original starting leach solution basis. Free acid concentrations were determined by titration with IN NaOH using a Metrohm automatic titrator. 3.4.2 So 1 ids Ana 1yses Concentrates, minerals and leach residues were assayed for copper, iron, zinc and nickel by digestion of the sample in aqua regia and bromine, followed by atomic absorption analysis of the solubilized metal values. Solids were analyzed for total sulphur by digestion of the sample in aqua regia and bromine, followed by precipi tation of sulphur as barium sulphate. 3.5 MATERIAL BALANCES AND EXTRACT ION CALCULATIONS The precision and accuracy of solids and solution - 70 -analyses were checked by performing complete material balance calculations for each run. Examples of such calculations are given in Appendix 1. If material balances did not agree to within 5%, the solids and solutions were re-assayed, or the run was repeated. Metal extraction percentages were calculated in two ways - on a solution basis, that is: % extr. = total weight of dissolved metal x 100, head metal weight and, wherever possible, on a residue basis, that is: % extr. = head metal weight - residue metal weight x 100, head metal weight Extraction values quoted are averages of values calculated on a solution basis and on a residue basis, whenever both values were available. CHAPTER 4 RESULTS AND DISCUSSION 4. 1 REDUCTION LEACHING WITH HYDROGEN GAS Early experimental work was concerned with determining how effectively iron could be extracted from a pyritic copper concentrate by reduction leaching with hydrogen gas. The leach chemistry was postulated to proceed, if at all, according to the following stoichiometry: CuFeS2 + 3Cu2+ + 2H2. >2Cu2S + Fe2+ + 4H+ (50) FeS2 + 4Cu2 + + 3H2 >2Cu2S + Fe2+ + 6H+. (51) Phoenix concentrate, assumed to contain 70% CuFeS2 and 25% FeS2, was used for these tests. The stoichiometric cupric requirement for this concentrate, based on equations (50) and (51), is: 70 x 3 + 25 x 4 = 1.98 moles Cu2+ per 183.52 119.98 100 g concentrate, giving a stoichiometric molar Cu2+/Fe ratio of: 1.98 - = 3.4/1. 100 x .329 55. 85: - 72 -The runs were performed in the 2L Parr autoclave using as-received concentrate and 1L of a IM copper sulphate solution. In the hopes of promoting rapid leaching, the runs were carried out under relatively high temperature, 180 °C, and high pres sure, 2.8 MPa hydrogen. Table 9 summarizes the results obtain ed Runs #1-3 were performed under hydrogen; Run #4, the control run, had 2.8 MPa nitrogen substituted for hydrogen. In Run #1, 50 g concentrate, representing the stoichio metric requirement, was leached for 2 h. The iron extraction achieved was 99.9% with only 0.017% Fe remaining in the leach residue. Therefore, iron was extracted quantitatively from both pyrite and chalcopyrite. The final leachate contained 15.0 g/L Fe and 0.57 g/L Cu, indicating an excellent copper-iron separation was achieved. The residue assayed 74.5% Cu, which is close to the composition of pure chalcocite (79.9% Cu), as predicted by equations (50) and (51). However, no attempt was made to determine the exact stoichiometry of the product. Either digenite (Cu^gS) or djurleite (CujggS) may also have been present. Runs #2 and 3 show results obtained when a stoichiometric excess of concentrate was leached. In Run #2, 65g of concen trate (~20% excess) was leached for VA h to give an iron extraction of 91.7% with 7.0 g/L Cu left in solution. In Run #3, 80 g of concentrate (~35% excess) was leached for 2 h to give an iron extraction of 94.5% with only 0.06 g/L Cu left in solution. In both cases iron extractions were higher than theoretically possible based on equations (50) and (51). - 73 -TABLE 9 Reduction Leaching of Phoenix Concentrate With Hydro gen Gas Run # 1 2 3 4 Feed Weight (g) 50 65 80 50 Initial Leachate: Vo1ume (mL) 1, 000 1, 000 1, 000 1, 000 g/L Cu 63. 5 63.5 63. 5 63. 5 Initial Mo 1 a r Cu2+/Fe Ratio 3.4 2.6 2. 1 3.4 Leach Conditions: Temperature (°C) 180 180 180 180 Pressure (MPa) 2.8 H2 2.8 H2 2.8 H2 2.8 N2 Time (h)* 2 lfc 2 2 Final Leachate: Vo 1 ume (mL) 1, 000 1 , 000 1 , 000 1 , 000 g/L Cu 0. 57 7.0 0. 06 40. 0 g/L Fe 15.0 n.d. 24. 8 5.2 Leach Residue: Weight (g) 89. 5 98.9 111.1 58. 1 % Cu 74. 5 70.6 74. 1 54. 1 % Fe 0.017 1. 8 1 . 1 16.8 Fe Extraction (%)** 99. 9 91.7 95. 4 40.7 n.d. = not determined * does not include warm-up time (~25 min. to 180 °C) ** based on head and residue iron assays - 74 -It is likely that side-reactions were taking place during the leach which lowered the initial cupric requirement. One possible competing reaction contributing to chalcopyrite conversion is the previously mentioned "neutral" or McGauley-type leach, which requires just one mole cupric per mole chalcopyr i te: CuFeS2 + Cu2+ >2CuS .+ Fe2+. (52) Note that this leach generates covellite (CuS), not chalcocite. Another possibility is that the Phoenix concentrate had under gone significant surface air oxidation with time, rendering some of the contained copper acid-soluble CuO + 2H+ >Cu2+ + H20 (53) thereby reducing the initial cupric requirement. Run #4, representing a neutral or McGau1ey-type leach (equation 52), was identical to Run #1 except nitrogen was used rather than hydrogen. After 2 h leaching an iron extract-tion of 40.7% was obtained, considerably lower than that achieved under reducing conditions. Thus, under comparable conditions, the neutral leach is considerably less effective than the reduction leach in achieving a copper-iron separation. However, these results showed that iron can be partially extracted from chalcopyrite under neutral leach conditions, supporting the above conjecture that a neutral leach may - 75 -be a supplementary side reaction during reduction leaching. It is well known that cupric-containing solutions are rapidly reduced to copper metal by hydrogen under the above-described conditions. Therefore, it can be presumed that the reduction leach might take place in two steps: 2Cu2+ + 2H2 >2Cu° + 4H+ (54a) CuFeS2 + 2Cu° + Cu2+' >2Cu2S + Fe2+ (54b) Reaction (54a) is very rapid at 180 °c, with Cu2+ being lowered from 60-90 g/L to 3 g/L in 30 minutes7**. The overall reduction leach appears to take place at a rate comparable to hydrogen reduction of copper sulphate solutions. More will be said about the kinetics later. These results show that reduction leaching of copper concentrate with hydrogen gas is an effective means of achiev ing a good copper-iron separation, with both pyrite and chal copyrite being converted to chalcocite. However, for practical purposes the use of hydrogen gas as a reductant has two major drawbacks: a) The requirement of three moles dissolved copper per mole chalcopyrite means leach pulp densities are restricted to being unacceptably low at 10% or less. - 76 -b) The solution produced is quite acidic, containing at least 65 g/L H2SO^. Removal of iron from acidic solutions is considered to be too difficult to be practically feasible. For these reasons no further studies on reduction leaching with hydrogen were carried out. 4.2 REDUCTION LEACHING WITH COPPER POWDER UNDER NITROGEN From a practical standpoint, it was desirable to develop a reduction leach method which would work at higher pulp densities (20% or greater) and not produce acid. Reduction leaching with copper powder seemed the obvious choice, as the anticipated leach reactions for chalcopyrite and pyrite require only one mole cupric per mole mineral iron, and do not produce acid. The following series of experiments were designed to study the feasibility of using copper powder as a reducing agent. The objectives were to produce, at reasonable temperature and residence time, a residue containing less than 1% Fe and a leachate containing less than 3 g/L Cu. CuFeS2 + Cu2+ + 2Cu° FeS2 + Cu2+ + 3Cu° • ->2Cu2S + Fe2+ •>2Cu2S + Fe2 + (55) (56) - 77 -4.2.1 Effect of Residence Time at 140 °C Reduction leach experiments on the as-received Ruttan concentrate, in which the leach residence time was varied, were carried out at 140 °C in the shaking autoclave. The leach solution used in each test contained initially 90 g/L Cu2+, 20 g/L H2SO4 and 132 g/L (NH^^SO^. Fine copper powder (-400 mesh) was used as a reducing agent. Assuming the Ruttan concentrate to contain 82%CuFeS2» 14% FeS2 and 3% ZnS; and. the predominant leach reactions to be, CuFeS2 + Cu2+ + 2Cu° ; r->2Cu2S + Fe2+ (57) FeS2 + Cu2+ + 3Cu°- >2Cu2S + Fe2+ (58) ZnS + Cu2+ + Cu°, *Cu2S + Zn2+ (59) the stoichiometric molar Cu2+/(Fe+Zn) ratio is 1/1 and the stoichiometric molar Cu°/(Fe+Zn) is 2.15/1. On this basis, 11.8 g of Ruttan concentrate and 9.4 g of -400 mesh copper powder were leached under 1 MPa N2 pressure for residence times (including the warm-up period) of 12, 22, 42 and 102 minutes. Figure 8 shows the results of iron and zinc leaching, and acid consumption as a function of residence time. The initial rate of iron and zinc dissolution was very rapid, with 68% iron extraction and 20% zinc extraction obtained during the 12 minute warm-up period. After 22 minutes, iron leaching had slowed considerably, and reached 93.4% extraction - 79 -after 102 minutes. Zinc leached steadily but reached a final extraction of only 55.8% after 102 minutes. The final residue solids assayed 0.6% Fe and 0.54% Zn, and the final leach solution contained 15.4 g/L Cu2+. During the leach, the free sulphuric acid concentration decreased from 20.0 g/L to 14.1 g/L, indicating that acid consuming reactions such as CuO + 2H+ >Cu2+ + H20 (60) Fe203 + Cu° + 6H+ : >2Fe + + + 3H20 '+ Cu + + (61) may be occurring to a slight extent. Also, the final Cu2+ concentration is higher than expected based on the iron extract ion achieved. This may be due to surface air oxidation having generated copper oxides which are acid soluble, as shown by equat ion (60). The high iron extraction obtained indicates that pyrite must also be leached, as was the case when reduction leaching with hydrogen. Zinc was only partially extracted, showing that the contained sphalerite is not as reactive as chalcopyrite and pyrite to reduction leaching under these conditions. The "target" level of 1% Fe in solids was achieved, but the soluble copper level in the final leachate remained too high. - 80 -4.2.2 Effect of Varying the Starting Molar  Cu2+/(Fe+Zn) and Cu°/(Fe+Zn) Ratios It has been demonstrated that reduction leaching of a copper concentrate with fine copper powder at 140 °C can achieve a high iron extraction. However, the calculated stoichiometric conditions used, based on mineral composition estimations, were not ideal because the final leach solution still contained approximately 17% of the initial cupric content. It was desirable, if possible, to achieve a low final cupric level ("^3 g/L) while maintaining a high iron extraction, by determining the optimum relative amounts of Cu2+, Cu° and concentrate to use. To achieve this, a series of reduction leach experiments were performed on the as-received Fox Lake concentrate in which the initial molar Cu2+/(Fe+Zn) and Cu°/(Fe+Zn) ratios were stoichiometric, and varied from stoichiometry by plus or minus 25%. The Fox Lake concentrate was assumed to contain 75% CuFeS2, 20% FeS2 and 5% ZnS, and reactions (57)-(59) were assumed to be the predominant leach reactions. On this basis, the starting stoichiometric molar Cu2+/(Fe+Zn) and Cu°/(Fe+Zn) ratios for Fox Lake concentrate are 1/1 and 2.12/1, res pect i ve1y. Leach tests were performed in the shaking autoclave using fine copper powder (-400 mesh) and a starting leach solution containing 90 g/L Cu2+, 20 g/L H2SO^ and 132 g/L (NH/j^SO^. To vary the Cu2 + /(Fe + Zn) ratio, the concentrate - 81 -weight was held fixed and the solution volume was varied. Similarly, to change the Cu°/(Fe+Zn) ratio the weight of copper powder used was varied. All leach tests were performed at 140 °C under 1 MPa N2 pressure for 1 h (not including a 12 minute warm-up time). Table 10 summarizes the results obtained. Run #1 gives the results of the stoichiometric test, and shows that 97% iron extraction was obtained (indicating again that pyrite must also be leached) but 22.0 g/L Cu remained in solution. Iron extractions were drastically lowered, to the 80-82% range, only when the Cu°/(Fe+Zn) ratio was de creased by 25% from stoichiometry (Runs #5, #7 and #8). The runs which produced the lowest iron content in residues and the lowest solution copper values were Runs #3 and #9, where the Cu2+/(Fe+Zn) molar ratio was decreased by 25% from stoichiometry. End solution copper values, at 0.4 and 0.5 g/L, were well below the suggested target value of 3 g/L. Residue iron values, at 1.1% and 1.2%, were very close to the 1% target value. In all runs, zinc was only partially extracted (52-57%). The results of this series of experiments have shown that, for the Fox Lake concentrate, the optimum Cu2+/(Fe+Zn) starting molar ratio is about 20-25% less than calculated stoichiometry. This suggests that the mineral composition estimation was not accurate especially with respect to neglect ing the oxidized species. The Cu2+/(Fe+Zn) ratio can be decreased by decreasing the solution volume rather than the cupric concentration, meaning the reduction leach can be Run// 1 2 3 4 5 6 7 8 9 Cone. Weight (g) 10.8 10.8 10.8 10.8 10.8 10.8 10.8 10.8 10.8 Copper Weight (g) 9.4 9.4 9.4 11.8 7.1 11.8 7.1 7.1 11.8 Molar Cu°/(Fe+Zn) 2.12 2.12 2.12 2.65 1.59 2.65 1.59 1.59 2.65 Initial Leachate: Volume (mL) 50 62.5 37.5 50 50 62.5 37.5 62.5 37.5 Molar Cu++/(Fe+Zn) 1.00 1.25 0.75 1.00 1.00 1.25 0.75 1.25 0.75 Leach Residue: Weight (g) 19.7 20.0 18.8 22.6 18.5 22.8 18.5 18.4 22.8 % Cu 80.6 78.3 75.5 80.6 70.8 79.2 71.6 71.2 79.1 % Fe 0.9 0.6 1.1 0.5 3.1 0.7 3.6 3.8 1.2 % Zn 0.95 0.95 0.91 0.85 1.00 0.83 0.96 1.01 0.83 % S 19.6 19.1 19.3 16.5 22.5 16.7 20.6 22.1 18.2 Final Leachate: g/L Cu 22.0 34.7 0.4 20.1 47.7 30.1 1.8 42.2 0.5 g/L Fe 70.8 54.7 88.7 70.5 56.3 54.1 78.0 45.0 90.0 g/L Zn 4.55 3.52 6.03 4.48 4.40 3.33 5.55 3.14 5.44 Extraction (96): Fe 97 96 94 98 81 95 82 80 93 Zn 55 54 57 54 54 52 54 52 52 TABLE 10 Effect of Varying Starting Molar Cu°/(Fe+Zn) and Cu2+/(Fe+Zn)  Ratios on Reduction Leaching of Fox Lake Concentrate - 83 -run at 20-25% higher pulp density without decreasing the overall iron extraction. The optimum Cu°/(Fe+Zn) molar ratio appears to be the calculated stoichiometric value, as increasing this value had no beneficial effect whereas decreasing the value resulted in dramatically reduced iron extractions. To determine if these "optimized" conditions could be applied to other copper concentrates, 237 g of the as-received Ruttan concentrate was reduction leached in the Parr autoclave with 189 g of -400 mesh copper powder, in 800 mL (a 20% deficiency) of leach solution containing init ially 90 g/L Cu2+, 20 g/L H2SO^ and 132 g/L (NH^)2SO^. The leach was performed at 140 °C for V/i h (not including a 17 minute warm-up period) under 1 MPa N2. The dried residue solids weighed 413 g and contained 76.2% Cu, 0.85% Fe, 0.68% Zn and 18.6% S. The final leach solution combined with wash water contained, on an 800 mL basis, 83.7 g/L Fe, 4.77 g/L Zn, 0.26 g/L Cu and 9.6 g/L H2SO^ (pH = 1.5). This represents an iron extraction of about 92.2% and zinc extraction of about 55.7%. These results show that an excellent copper-iron separation in the Ruttan concentrate can be made by reduction leaching with fine copper powder and a 20% deficiency in cupric copper. 4.2.3 Effect of Copper Powder Particle Size on Iron Extraction To determine whether the copper powder particle size - 84 -affected the kinetics of reduction leaching, Fox Lake con centrate was leached as described for Table 10, Run #1, except that copper powder carefully sized to -100+200 mesh was used. Reaction conditions and results are summarized in Table 11. The results from Table 10, Run #1 are also reproduced in Table 11 for comparative purposes. The use of coarser copper powder resulted in a sharply reduced iron extraction, from 97% to 81%, after leaching at 140 °C for 1 h. Zinc extraction was also decreased slightly, from 55% to 48%. Thus, as might be expected, the use of finer particle size copper improved reaction kinetics. 4.3 POSTULATED REACTION MECHANISMS: GALVANIC VERSUS CUPROUS-MEDIATED 4.3.1 Galvanic Mechanism Hiskey and Wadsworth postulated that reduction leaching of chalcopyrite with copper powder in acid solutions, which produced chalcocite and hydrogen sulphide gas, was controlled by galvanic interaction of the chalcopyrite mineral with copper6^. Electrochemical experiments performed by Nicol later confirmed a galvanic mechanism6^. Chalcopyrite and copper can couple galvanically because of the excellent elec tronic conductivity of all phases. The principal cathodic and anodic half-cell reactions are6^: - 85 -TABLE 11 Effect of Copper Powder Particle Size on  Reduction Leaching of Fox Lake Concentrate -400 mesh Cu° - 100 + 200 mesh Cu° Concentrate (g) 10.8 10.8 Copper Powder: Wt (g) 9.4 9.4 Initial Leachate: Vo 1 ume (mL) 50 50 Cu (g/L) 90 90 H2SO^ (g/L) 20 20 (NH^)2SO^ (g/L) 132 132 Reaction Conditions; Temp (°C) : 140 140 Time (h) 1 1 Leach Residue: Weight (g) 19.7 20. 1 % Cu 80. 6 75. 1 % Fe 0.9 3.7 % Zn 0. 95 1.01 % S 19.6 20.3 Final Leachate: g/L Cu 22. 0 30. 8 g/L Fe 70.8 59.6 g/L Zn 4.55 3.75 Extraction (%) Fe 97 81 Zn 55 48 - 86 -Cathod i c: 2CuFeS2 + 6H+ + 2e" v ^Cu2S + 2Fe2+ + 3H2S (62a) Anod i c: 2Cu° + H2S ^Cu2S + 2H+ + 2e" (62b) which give the overall reaction: CuFeS2 + Cu° + 2H+ >Cu2S + Fe2+ + H2S. (63) Figure 9 depicts schematically the galvanic CuFeS2-Cu couple. Chalcocite forms as a porous layer which surrounds FIGURE 9 Schematic Representation of the CuFeS?-Cu° Galvanic Couple in Sulphuric Acid Solution - 87 -both the Cu and CuFeS2 particles. Hiskey and Wadsworth con firmed the presence of a chalcocite layer by direct observation of polished sections of residue particles under a microscope. One can postulate that reduction leaching of chalcopyrite with copper in copper sulphate solutions might also be con trolled by a galvanic mechanism. In this case, however, no chalcocite should grow on the metallic copper particles because sulphur is not transported through the electrolyte. Such a modified galvanic model is depicted schematically in Figure 10. The anodic reaction now is the dissolution of copper 2Cu°^ ^2Cu2+ + 4e" (64a) Fe porous FIGURE 10 Schematic Representation of a Postulated CuFeS?-Cu0 Couple in Copper Sulphate Solution - 88 -while the cathodic reaction to form chalcocite is: CuFeS2 + 3Cu2+ + 4e" v N2Cu2S + Fe2+. (64b) More Cu2+ is consumed by the cathodic reaction than is produced by the anodic reaction, to give the overall reaction: CuFeS2 + 2Cu° + Cu2+ > 2Cu2S + Fe2 + . (64) The above galvanic mechanism could partially account for the observed chemistry of reduction leaching in copper sulphate solution. However, under stoichiometric conditions galvanic reactions do not go to completion due to the lack of perfect mixing of solid phases. Hiskey and Wadsworth found that iron extractions from chalcopyrite rarely exceeded 80% by their galvanic reaction. In addition, they showed that the initial presence of cupric ions severely inhibited the reaction. McKay and Swinkels also observed incomplete iron extraction when reduction leaching chalcopyrite with copper in acid solution62. Thus it is unlikely that a galvanic mechanism can totally account for the observed chemistry of reduction leaching in copper sulphate solution. 4.3.2 Cuprous-Mediated Mechanism It is known that the cuprous ion is in sulphate solutions by the equilibrium: si ightly stabi1ized 89 -Cu2 + + Cu° 5= 2Cu + (65) The observed value of Kg^ in dilute solutions at room temp erature is IO-6 lyj^ kut this value increases with increasing temperature, leading to much higher cuprous ion concentrations. The measured value for Kg<j at 0.1 M Cu2 + concentration at 160 °C was found to be about 0.035'75. The cuprous ion could mediate the reaction between chalcopyrite and copper, elim inating the need for galvanic content, ie: CuFeS2 + 4Cu+ >2Cu2S + Cu2+ + Fe2+ (66) Such a cuprous mediated mechanism can be represented s c h ema tically as sh own in Figure 11. If this me c h a n i sm is FIGURE 11 Schematic Representation of a Cuprou s-Med i ated Reaction between Chalcopyrite and Copper - 90 -operating at elevated temperatures, the reaction should be improved by a weak cuprous stabilizing agent such as carbon monoxide. Carbon monoxide under pressure will act as a comp-lexing agent to stabilize the cuprous ion in sulphate solutions, if Cu° is present, via the following equilibrium76: 2Cu° + 2Cu2+ + 4CO(g) ^==& 4Cu(CO)+. (67) The cuprous-car bony 1 complex could then react with chalcopyrite: CuFeS2 + 4Cu(C0)+ ->2Cu2S + Fe2+ + Cu2+ + 4CO . (68) Under the moderately acidic conditions that the reduction leach is carried out, carbon monoxide will not act as a reducing agent77 and should be quantitatively regenerated, as copper in solution is depleted. 4.4 REDUCTION LEACHING WITH COPPER POWDER UNDER CARBON MONOXIDE 4.4.1 Effect of Carbon Monoxide  on Reaction Kinetics To determine whether reduction leaching with copper powder in copper sulphate solutions could be explained by a cuprous-mediated mechanism, Bethlehem concentrate was leached under both carbon monoxide and nitrogen. The Bethlehem - 91 -concentrate was assumed to contain 59% CuFeS2 and 20% Cu^FeS^ as the main sulphide minerals. The overall reactions were assumed to be: CuFeS2 + 2Cu° + Cu2+—•• > 2Cu2S + Fe2+ (69) Cu5FeSZt + 2Cu° + Cu2+ > 4Cu2S + Fe2 + . (70) On this basis, the stoichiometric molar Cu2+/Fe ratio is 1/1 and the Cu°/Fe ratio is 2/1. Stoichiometric amounts of concentrate (391.6 g, -275 mesh) and coarse copper powder (180 g of 90% -100+325 mesh material) were leached in the Parr autoclave at 120 °C with a leach solution containing initially 90 g/L Cu2+, 20 g/L H2SO^ and 132 g/L (NHif)2SOif. A lower leach temperature and coarser copper powder were used to deliberately slow down the reaction kinetics so that effective comparisons could be made. In one test, the leach was pressurized with 1 MPa N2 during the 18 minute warm-up period, the solution was sampled when the temperature reached 120 °C, whereupon the nitrogen was vented and replaced with 1 MPa carbon mon oxide. In another test, the reactor was kept pressurized with 1 MPa N2 throughout the leach. Total residence time for both runs was 6 h (not including the 18 minute warm-up per iod ) . Figure 12 portrays graphically the results of these runs in terms of percent Fe extracted versus residence time. Initially the iron leach rates were similar for both runs, - 92 --i —r 1 T j-2 3 4 5 6 Leach Time - hours FIGURE 12 Effect of CO on the Reduction Leaching of Bethlehem Concentrate with Copper at 120°C - 93 -but after 45 minutes reduction leaching under nitrogen slowed down dramatically to give 64% iron extraction after 6.3 h. By contrast, reduction leaching under carbon monoxide continued at a reasonable rate to give a final iron extraction of 99%. These results strongly suggest that the kinetics of reduction leaching can be explained by a cuprous-mediated mechanism in which the rate of cuprous formation can be rate controlling. In another leach test, 218.4 g of Fox Lake concentrate was leached with 187.6 g of copper powder, again at 120 °C under 1 MPa CO. Leach conditions were identical to the above, with the exception that as-received concentrate (72.4% -400 mesh) and fine copper powder (-400 mesh) were used. Figure 13 portrays graphically the results obtained, in terms of % Fe and % Zn extracted versus residence time. The Fox Lake concentrate was initially even more reactive than the Beth lehem concentrate, both before and after CO was added at the 18 minute mark. This can probably be attributed to the increased rate of formation of cuprous ions, due to the use of copper powder of much finer particle size. It is also possible that mi neralogical characteristics of the Fox Lake concentrate make it more reactive to reduction leaching under these conditions, although studies to date had not indicated different concentrates to vary substantially in reactivity under similar leach conditions. Iron and zinc extractions reached 93% and 54%, respectively after 6.3 h leaching. - 94 -FIGURE 13 Reduction Leaching of Fox Lake Concentrate with Copper at 1206C under 1MPa C6 - 95 -4.4.2 Effect of Leach Temperature  on Reaction Kinetics To evaluate the temperature sensitivity of reduction leaching under carbon monoxide, a series of experiments were performed on Bethlehem concentrate in which the leach temp erature was varied between 100 °C and 140 °C. Tests were carried out in the shaking autoclave for residence times varying from 0.75 h to 9 h. In each test, 21.5 g of con centrate, carefully sized to -140+270 mesh particle size, was leached with 9 g of coarse copper powder (90% -100+325 mesh), using 50 mL of a leach solution containing initially 90 g/L Cu2+, 20 g/L H2SO^ and 132 g/L (NH^SO^. These con ditions represented an approximate 10% excess of concentrate over calculated stoichiometry. After a leach warm-up period which varied from 3-12 minutes depending upon the reaction temperature to be maintained, the reactor was pressurized with 1 MPa CO gas. Figure 14 portrays graphically the results in terms of both iron extractions and iron remaining in leach residues. It can be seen that the reaction kinetics are very temperature sensitive. An approximate estimation of the experimental activation energy can be made as follows. It can be assumed that the rate law is uniform at different temperatures for equal percent conversions. Therefore an approximate activation energy can be extracted from the Figure 14 plot by assuming that - 96 -FIGURE 14 Effect of Temperature on Reduction Leaching of Bethlehem Concentrate with Copper under 1MPa CO - 97 -k 120 TTioo k100 Vl20 where kj20 and k100 are tne reaction rates at 120 °C and 100 °C respectively; and X\2Q and "^100 are the times required, at 120 °C and 100 °C, to reach the same iron extraction value. Table 12 gives thetTioO anc' fl 2 0 values *or different percent iron extractions, from 35% to 65% , as determined from Figure 14. It can be seen that in this range,f100/Ti20 is virtually constant and has an average value of 4.14. The Arrhenius equation k = A exp (-Ea/RT) (71) k i on can be re-written in terms of , that is: k100 k120 A120 exP (-Ea/RT120) (72) k100 A100 exP (-Ea/RT100) Assuming A to be essentially constant with temperature, and taking the natural log of both sides, we get: In fk120N 'k100/ Ea Ea _ . (73) RT100 RT120 This equation can be expressed in terms of Ea as : I - 98 -TABLE 12 Residence Time Ratios at Different  Iron Extractions - From Figure 14 Leach Residence Time (h) Tioo k 1 20 Fe extr. at 100 °C at 120 °C ' = (%) (fioo) (fi2 0> fl20 k100 35 2.25 0. 55 4. 09 40 2.70 0. 65 4. 15 45 3. 20 0.75 4. 27 50 3. 80 0.90 4. 22 55 4.55 1.10 4. 14 60 5. 40 1.35 4. 00 65 6.60 1. 60 4. 13 Avg. = 4. 14 - 99 Ea ln 120 \k100 R Tl20 T100 Tl20 " T100 (74) Substituting numbers into this expression Ea = ln (4.14) (8.314 3K"1 mol"1) (393°K) (373°K) (1000 3/k3) (393°K-373°K) (75) = 87 kJ/mol. This value is considerably higher than the activation energy for the galvanic conversion of chalcopyrite as determined by Hiskey and Wadsworth6*, which was 48 k3/mol. 4.4.3 Effect of Initial Concentrate Particle Size on Iron Extraction To determine the effect of initial concentrate particle size on the kinetics of reduction leaching under CO, two experiments were performed on Bethlehem concentrate carefully sized to -140+270 and -400 mesh. The experiments were performed on the shaking autoclave with all other reaction conditions held constant. Reaction conditions and results are summarized in Table 13. After leaching for 12 h at 100 °C, the -140+270 mesh material had reached 73% Fe extraction, whereas the -400 mesh material had reached 84% Fe extraction. As might be expected, iron extractions were improved by leaching con-- 100 -TABLE 13 Effect of Initial Concentrate Particle Size on the Reduction Leaching of Bethlehem Concentrate wi th Copper under 1 MPa CO -140+270 mesh size -400 mesh size Concentrate (g) 21. 5 21. 5 Copper Powder: Wt (g) Par t i c1e S i ze 9.0 90% -100+325 9.0 90% -100+325 Initial Leachate: Vo 1 ume (mL) Cu (g/L) H2SO^ (g/L) (NH^)2SO^ (g/L) 50 90 20 132 50 90 20 132 Reaction Conditions: Temp (°C) Time (h) 100 12 100 12 Leach Residue: Weight (g) % Fe Particle Size n. d. 3.8 81% -270 27. 1 2.8 Final Leachate: g/L Fe 62. 8 73. 0 Fe Extraction (%) 73 84 n.d. = not determined - 101 -centrate of finer particle size. Leach residue from the original -140+270 mesh material underwent a screen analysis which showed, surprisingly, that 81% was now -270 mesh. This indicates that significant particle size reduction had occurred during leaching. This phenomenon will be discussed in greater detail later. 4.5 COMPARISON OF REACTIVITY OF SEVERAL SULPHIDE MINERALS TO REDUCTION AND NEUTRAL LEACHING As was mentioned in the S.C. copper process description, neutral or "McGau1ey-type" leaching of activated copper con centrates is used to increase the overall iron extraction from 75% to about 91%. To evaluate the feasibilty of using reduction leaching as an alternative to neutral leaching, a series of experiments was performed to determine the relative reactivities of several sulphide minerals commonly found in copper concentrates to reduction leaching and neutral leaching. The minerals used in this study were pyrite, bornite, chalcopyrite, pentlandite (^NiFeS^g), sphalerite (ZnS) and pyrrhotite (~FeS). The expected chemistry for reduction and neutral leaching of these minerals is compared in the equations equations shown in Table 14. Ground mineral samples were carefully screened to a narrow size range of -270+325 mesh for use in leach tests. All tests were performed in the shaking autoclave using 50 mL of a leach solution containing 90 g/L Cu2+} Mi neral Reduction Leaching Neutral Leaching FeS2 CujjFeS^ CuFeS2 NiFeSi.g ZnS FeS FeS2 + CuSO^ + 3Cu° > 2Cu2S + FeSO^ Cu5FeS^ + CuSO^ + 2Cu° »4Cu2S + FeSO/j CuFeS2 + Q1SO4 + 2Cu° »2Cu2S + FeSCty NiFeSi.8 + 2CuSO^ + 1.6 Cu° > 1.8 Cu2S + FeSCty + MSO4 ZnS + CX1SO4 + Cu° » Cu2S + ZnSO^ FeS + CuSO^ + Cu° » Cu2S + FeSO^ FeS2 + CuSCv >CuS + S + FeSCty Cu5FeSif + CuSCty. >2CuS + 2Cu2S + FeSCty CuFeS2 + CuSCty >>2CuS + FeSCty NiFeSi.g + CuSCty >1.6 CuS + 0.2 Cu2S + FeSCty + N1SO4 ZnS + CuSO^ > CuS + ZnSO^ FeS + CuSCty >CuS + FeSCty. TABLE 14 Reduction and Neutral Leaching Chemi Several Sulphide Minerals stry of - 103 -20 g/L H2SO4 and 132 g/L (NH^)2SO^. For the reduction leach runs, stoichiometric amounts of minerals and coarse powder (90% -100+325 mesh), calculated on the basis of the equations depicted in Table 14, were used. For the neutral leach runs, stoichiometric amounts of minerals were also used. All runs were performed at 140 °C under 1 MPa nitrogen for 1 h. The results of these runs are summarized in Table 15. Pyrite, bornite and chalcopyrite are about equally reactive to reduction leaching, with Fe extractions in the range 68-74% obtained. By contrast, all three minerals are very much less reactive to neutral leaching. Pyrite was essentially unattacked with only 3% Fe extraction obtained, whereas chal copyrite and bornite were only slightly reactive (8% and 19% Fe extractions, respectively). For the other three minerals, pentlandite, sphalerite and pyrrhotite, the differences between reduction and neutral leaching are less dramatic, although reduction leaching did extract more iron or zinc than did neutral leaching. This implies that the actual leach stoichiometry as suggested in Table 14 is not correct, with iron being leached somewhat preferentially to nickel. 4.6 THE MORPHOLOGY OF REDUCTION LEACHING Residue particles from both the reduction and neutral leach experiments on bornite and chalcopyrite were mounted FeS2 Cu^FeS^ CuFeS2 NiFeSi .8 ZnS FeS Red. Neut. Red. Neut. Red. Neut. Red. NeUt. Red. Neut Red. Neut. Mineral Weight (g) 9.07 9.07 27.93 27.93 9.60 9.60 6.20 6.20 9.17 9.17 6. 81 6.81 Copper Weight (g) 13.50 - 9.00 — 9.00 — 3.60 — 4.50 — 4. 95 — Residue: Wt. (g) 22.00 8.96 36.58 29.03 19.21 9.45 9.73 6.26 13.65 9.25 11. 99 6.98 % Cu 74.2 1.6 72.7 56.5 69.1 23.7 41.9 8.5 39.2 4.2 56 .8 5.4 % Fe 4.3 43.4 2.6 11.3 7.1 38.3 15.4 28.0 - - 22 .7 57.8 % Other _ _ _ - - - 19.5 Ni 30.2 Ni 29.4 Zn 44.2 Zn - -% S 18.5 53.6 21.4 26.5 18.7 36.1 . 20.8 32.2 23.7 34.4 21 .2 33.5 Final Leach Solution: g/L Cu 28.2 85.4 25.9 65.6 29.5 82.0 74.9 82.3 64.8 82.3 55 .0 82.4 g/L Fe 56.5 3.5 54.0 17.4 55.8 6.1 8.9 3.4 - - 27 .0 8.3 g/L Other - - - - - - 2.4 Ni 3.3 Ni 12.4 Zn 5.3 Zn — _ Fe Extr. (96) 74 3 72 19 68 8 23 10 33 11 "Other" Extr. (%) - - - - - - 7 M 9 Ni 13 Zn 9 Zn - -TABLE 15 Results of Reduction and Neutral Leach Runs on Several Sulphide Minerals - 105 -on epoxy, polished and viewed under a Zeiss optical micro scope. Figure 15 shows colour pictures depicting cross-sectional views of the conversion products from (a) the neutral leach and (b) the reduction leach. The different mineral phases are quite clear, with particles of pyrite (white), chalcopyrite (yellow), bornite (brown) and cha1cocite/cove11ite (blue/grey) being identifiable. Only bornite particles show significant conversion by neutral leaching, with chalcopyrite and pyrite remaining relatively unattacked. By contrast, both chalcocopyrite and pyrite show significant conversion to chalcocite by reduction leaching. All original bornite particles in Figure 15(b) have been completely converted to cha1coci te. The morphology of chalcocite formation by reduction leaching is quite clearly evident as being an expansion of the product Cu2S away from the original mineral due to rapid volume increases. The volume, per mole of sulphur, for the different mineral phases is given in Table 16. It is clear from this table that every mineral must expand if its sulphur content is quantitatively converted to chalcocite. Mineral expansion is particularly dramatic in the case of pyrite con version, as its molar volume incrases from 11.95 to 27.92 cm3 /mo 1 . The conversion products were studied extensively under the electron microprobe for visual and analytical evidence which would support a galvanic mechanism. However, copper particles were never observed to be coupled to any mineral phases, lending further support to a cuprous-mediated mechanism. -10 6-K—50 um—* K—50 um—a (a) (b) Neutral Leach Products Reduction Leach Products Yellow - chalcopyrite Wh i te - pyrite Brown - born i te Blue/grey - cha1cocite/cove 11ite (400x Magnification) FIGURE 15 Optical Microscope Pictures of Conversion Products from the Neutral and Reduction Leach Experiments on Cha1copyri te/Bornite - 107 -TABLE 16 Molar Volumes of Sulphide Phases Phase Nomi na 1 Formu 1 a Formu la Wt. , g/mol . Dens ity*, g/cm3 Vol./Mole S, cm3 /mol Pyr i te FeS2 119.98 5. 02 11.95 Bo r n i t e Cu -jFeS^ 501.80 5. 07 24. 74 Cha1 copyri te CuFeS2 183.52 4.2 21.85 Pent 1 and i te NiFeS1<8 172.27 4.8 19.94 Spha1er i te ZnS 97.44 4.0 24. 36 Pyr rhot i te Fe0.9S 82. 33 4.6 17.90 Cha1coci te Cu2S 159.14 5.7 27.92 Cove 11i te CuS 95. 60 4.7 20. 34 * Taken from: CRC Handbook of Chemistry and Physics, 54th ed. (1973-1974), pp. B192-B197. - 108 -4.7 KI NET IC MODELS AND MECHANI SMS Postulated solid state and solution transport processes for cuprous-mediated reduction leaching of chalcopyrite are identified in Figure 16, and summarized in Table 17. The mineral decomposition process takes place by (a) rearrangement of sulphur atoms to form the Cu2S basic lattice, (b) diffusion of Cu+ inward through this new lattice and (c) diffusion of Fe2+ outward through the Cu2S phase. The diffusion rate of Cu+ in Cu2S is relatively fast, and has been measured by Etienne78 to be 3.8 x 10~10 cm2/sec at 75 °C, with a corres ponding activation energy of about 49 kJ/mol. This is somewhat lower than the 87 kJ/mol activation energy estimated for the reduction leach in this study; therefore diffusion of Cu+ would not seem to be rate limiting. In general, Cu2S is known to have a small Fe solid solubility73; therefore Fe2+ diffusion should be much slower than that of Cu+. However, as depicted in Figure 16, diffusion paths for Fe2+ are probably shorter than for Cu+ because of spall ing of the product layer. This would explain why essentially complete removal of iron can be be achieved. The solid state process that forms Cu2S and eliminates Fe2 + is driven by the reaction with aqueous cuprous ions, that is: CuFeS2 + 4Cu+(aq) >2Cu2S + Cu2+(aq) + Fe2+(aq). (76) The Cu+ ions are formed by reaction of Cu2+ with Cu° or H2 - 109 -FIGURE 16 Morphology of Chalcocite Formation by Reduction Leaching - Solid State and Solution Transport Processes - 110 -TABLE 17. Chemistry of Transport Processes Depicted in Figure 16 Overall (17-D CuFeS2 + 4Cu + 2+ 2+ ->2Cu2S + Fe + Cu CuFeS2 - Cu2S Interface Zone (17-2) 2S CuFeS. -> 2S Cu2S (17-3) Cu CuFeS-. -*Cu Cu2S (17-4) Fe CuFeS. -*Fe Cu2S Cu2S - Aqueous Interface Zone (17-5) (17-6) (17-7) Fe Cu2S 3Cu+(aq) Cu+(aq) -»Fe2+(aq) + 2e" (Cu2S) -»3Cu - 3e (Cu2S) ->Cu2+(aq) + e" (Cu2S) Aqueous Solution or Cu Interface (17-8) Cu2+(aq) + h H2 ^Cu+(aq) + H+ Hydrogen as re-ductant. (17^9) Cu2+(aq) + Cu° + -»2Cu (aq) Metallic copper as reductant - Ill -(equations 16-8 and 16-9). In view of the evidence that the metallic copper particle size and the presence of a Cu+ stabilizing agent affect the kinetics, it is clear that under these circumstances the solid state processes mentioned above are not rate-determining, even though they must take place. Because all leach experiments had starting conditions close to stoichiometry, reaction kinetics were mixed in that the solid state diffusion processes were dependent on, or "coupled" to, formation of Cu+. reduction leaching was beyond the scope of this work. However, there is a temptation to apply parabolic kinetics to the reduction leach (see Appendix 2). This model is suitable for the neutral leach where there is no evidence of product spalling. However, the reduction leach kinetics are probably closer to being linear, with a rate constant dependent on the Cu+ content of the solution. It was not possible to resolve the Cu+ dependence on the leaching rate because Cu+ could not be directly measured. If it is first order, the reaction rate can probably be written where ut is the linear unreacted particle dimension. The integrated form of this equation would depend on (a) the particle size distribution and (b) the dependence of Cu+ on time. Measurement of the latter function was not easily accessible within the range of this project. Rigorous evaluation of a kinetic rate equation for dt (77) CHAPTER 5 APPLICATIONS, CONCLUSIONS AND RECOMME NDATI ONS 5.1 APPLI CAT I ONS OF REDUCTI ON LEACH I NG 5.1.1 Incorporation oi Reduction Leach into  S.C. Process with no Roasting Reduction leaching, as a single unit operation, could in principle be incorporated into the S.C. Process as an alternative to the thermal activation-acid 1 eaching-neutra 1 leaching steps. One such hypothetical flowsheet for treating sulphide concentrates in which roasting is totally eliminated is shown in Figure 17, and the reaction chemistry (for chal copyrite) is summarized in Table 18. In this hypothetical process, copper concentrate is initially reduction leached with a solution containing 80 g/L Cu2+, 20 g/L H2SO^ and 132 g/L (NH^)2SO^, for 1J4 h at 140 °C, to obtain 95% or greater iron extraction. For chal copyrite concentrates, the copper reagent burden is large-about two tonnes copper powder and one tonne copper as copper sulphate leach solution is required for each tonne of copper in the concentrate (equation 18-1). Any pyrite present will also be leached, adding to the reagent requirements. However, pyrite decomposition at this stage is preferred if it is associated with any gold and silver, as is frequently the case, because precious metals locked into pyrite are usually difficult to recover . - 113 -FIGURE 17 Schematic Flowsheet of Reduction Leach Process with no Roasting Powdered Scrap &_ Cement Copper New Sulphuric Acid Copper Concentrates Reduction Leach Recycle Sulphides FeS04 Solution _Weak Acid Oxidizing B Leach . Recycle Pregnant Solution 1^ Recycle Copper Powder Residues Pregnant Solution Purification Purified Solution L*_ Return. Acid Recycle Copper ' Powder Hydrogen D Reduction L / S Split 02 NH3 Oxydrolysis E H L Flotation Flotation Concentrate Sulphide Residues to P.M. Recovery Jarosite Residue Silicious Tails • 1 Sulphur Recovery Recycle Sulphides Split to Oxidizing Leach B Elemental Sulphur Melting and Casting Wire Bars to Market - 114 -TABLE 18 Chemistry of Reduction Leach Process with no Roasting A. Reduction Leach (18-1) CuFeS2 + 2Cu° + CuSOzj > 2Cu2S + FeSO^ B. Oxidation leach (18-2) 2Cu2S + 4H2SO^ + 202 ^ 4CuSO^ + 2S° + 4H20 C. Pur i f i cat i on (18-3) H2Se03 + 4Cu° + 2H2SO^ > Cu2Se + 2CuSO^ + 3H20 (18-4) H2SeO/f + 5Cu° + 3H2SO^ >Cu2Se + 3CuSOif '+ 4H20 D. Hydrogen Reduction (18-5) 3CuSO/f + 3H2 >3Cu<> + 3H2SO^ E. Oxyd rolysis (18-6) FeSO^ + 1/3 NH3 + 1/4 02 + 3/2 H20 > 1/3 NH4Fe3(SO^)2(OH)6 + 1/3 H2SO^ - 115 -The copper reagent burden would be much lower for bornite concentrates, because the copper requirement for bornite is only 0.4 tonnes metal and 0.2 tonnes copper as copper sulphate, for each tonne of copper contained as bornite (refer back to equat ion 70). Solids from the reduction leach contain less than 1% Fe and are subjected to acid oxygen pressure leaching under conditions similar to the S.C. Process oxidizing leach (equation 18-2). Acid feed to the oxidizing leach comes from recycled acid from the hydrogen reduction and oxydrolysis steps, and new acid (added as 96-100% H2SO4) which is required to replace the sulphate lost by jarosite precipitation. The leach solution will contain 0.8-1 M (NH^)2SO^ which is recycled from hydrogen reduction and oxydrolysis. The pregnant solution from the oxidizing leach is split, with one quarter of the dissolved copper recycled to the reduction leach and the remaining three quarters directed to a purification step. The solution may contain impurities such as Se, Te, As, Sb and Bi. Of these, Se and Te will be particularly troublesome as they will precipitate quantitatively with copper during hydrogen reduction, and so must be removed. The suggested purification method is to precipitate these elements with copper powder destined for recycling to the reduction leach,N just as proposed by Kennecott Copper for their nitric sulphuric leach process^2 (equations 18-3 and 18-4). The purified pregnant leach solution is subjected to hydrogen reduction (equation 18-5) to bring the copper - 116 -level in solution from 80 g/L down to about 20 g/L. The buffering effect of 1 M (NH^)2SO^ wi 11 ensure that the reduction proceeds rapidly at 160-180 °C7^. Of the precipitated copper, two thirds is recycled back to the reduction leach after being used for purification, and one third is melted and cast into wire bars for sale. Hydrogen reduction rather than e1ectrowinning offers substantially reduced energy costs. Solution from the reduction leach is low in copper (less than 3 g/L Cu) and high in iron (70-90 g/L Fe), and undergoes oxydrolysis to precipitate iron as a clean ammonium jarosite (equation 18-6), suitable for disposal. This step is essentially the same as the iron removal step in the S.C. Process, except iron precipitation should be more rapid and complete due to the buffering effect of ammonium sulphate. Jarosite precipitation will remove about one third of the original sulphide sulphur in the concentrate as sulphate. Residue from the oxidizing leach will contain all precious metals, elemental sulphur, gangue and some unleached sulphide minerals. This residue can be treated as in the S.C. Process, by (a) flotation to reject gangue; (b) elemental sulphur removal by stream stripping or dissolution in a suitable organic solvent; and (c) splitting into a precious metals con centrate for shipment, and a recycle sulphide concentrate to be returned to the oxidizing leach for enhanced overall copper extraction. This proposed modification to the S.C. Process has the advantage of selectively extracting iron from copper in one unit operation and a residence time of Vh h, versus - 117 -the three unit operations and a total residence time of over 10 h required for the S.C. Process. The process is entirely hydrometa 11urgica1, provides for high recovery of copper and elemental sulphur, and rejects iron as a relatively pure jarosite compound. Gold and silver are recovered from the process as a precious metals concentrate which can be sold to a refinery or subjected to a special hydrometal 1 urgical t reatment. However, this process suffers from a large circulating load of copper. The consequence of this circulating load is that for chalcopyrite concentrates, 4 tonnes of copper are leached, and 3 tonnes of copper are hydrogen reduced, for every tonne of copper recovered for sale. As a result the leaching plant is 4 times as large and the hydrogen re duction plant 3 times as large as if there were no circulating loads of copper sulphate or copper metal. 5.1.2 Incorporation of Reduction Leach into  S.C. Process with Partial Roasting One method of reducing the circulating load of copper in a reduction leach process is to eliminate part of the feed sulphur by roasting to produce copper ferrite CuFeS2 + 13/4 02 >CuFe02>5 + 2S02 (78) followed by reduction leaching of a sulphide/calcine mixture, using varying mixtures of hydrogen and copper metal as reduct-- 118 -ants. Experimental work conducted in our laboratories has proven that iron can be extracted from s u 1 ph i de/ca 1 c i ne mixtures by reduction leaching, although not as easily as from pure sulphides79. A flowsheet for such a process is shown in Figure 18. The reduction leach has the following s t o i ch i omet r y : CuFeS2 +°<.CuFe02<5 + ^CuSO^ + VCu° + <SH2SO^ + JOH2 >2Cu2S + (ot+DFeSO^ + 2.5o<H20. (79) Here oM-/J+y= 3; ^+£=e/.+ l; $+f> = 2.5«K. The limiting condition for maximum calcine utilization is for the case where = 3, ($ = 0, y= 0, S = 4 and f> = 3.5. In this case, 75% of the chalcopyrite is roasted, which results in a zero circulating load of copper in the process. An intermediate stoichiometry occurs when c< = 1 (50% of chalcopyrite roasted). In this case ft = 0 to 2, v= 2 to 0, 8= 2 to 0 and f> = 0.5 to 2.5. The circulating load of copper is now about 1 tonne copper per tonne copper in the feed. Thus, by partially eliminating sulphur in a roaster, the amount of copper required to convert the remaining sulphur to chalcocite by reduction leaching is substantially reduced. As a consequence, the sizes of the subsequent oxidizing leach and hydrogen reduction steps are reduced, resulting in improved process economics. The major drawback of this partial cal cination option is that the process is less acceptable environ-me n t a 11y. - 119 -FIGURE 18 Schematic Flowsheet of Reduction Leach Process with Partial Roasting Powdered Scrap & Cement Copper New Acid • H2 J_ Copper Concentrates I Calcines Split Reduction Leach New Acid Acid Recycle Sulphides Recycle Pregnant Solution Split Purification L, Recycle _ Copper Powder Return_ ' Acid I^Copper Powder Split Roaster (fluid bed) * FeSOd „ Air 1 Solution °2 » Acid Leach B (Fe Scavenger) NH3 _L_ Oxydrolysis F I S j-Weak Acid Oxidizing C Leach Residues Flotation Sulphur Recovery Hydrogen E Reduction Sulphides Recycle to Oxidizing Leach Split Melting & Casting Sulphide Residues to P.M. Recovery S02-containing gases to Acid Plant. Jarosite Residues Silicious Tails Elemental Sulphur Wire Bars To Market - 120 -5.2 CONCLUSIONS This study has been concerned with developing novel reduction leach methods for chalcopyrite concentrates. It has been found that chalcopyrite can be essentially completely converted to chalcocite by leaching in strong copper sulphate solutions at elevated temperatures, using hydrogen gas or copper metals as reductants. The reactions are sensitive to temperature, concentrate and copper particle size, and to the presence of a cuprous-stabi1izing agent. The essential leach reactions appear to obey the following overall stoich-i ome try: The sulphide minerals bornite and pyrite, commonly found in copper concentrates, are also quantitatively converted to chalcocite by reduction leaching under these conditions. Microscopic evidence indicates that chalcocite forms as layers which crack and spall away from the reacting sul phides, allowing complete extraction of iron to take place. The mechanism for reaction probably involves transport of cuprous ions both in aqueous solution and in the solid state (Cu2S), ar>d solid state diffusion of ferrous ions outward. Cuprous ions are formed as an intermediate species during the leach reactions, either by reaction of cupric ions with hydrogen or with copper metal: CuFeS2 + 3Cu2+ + 2H2 -CuFeS2 + Cu2+ + 2Cu° >2Cu2S + Fe2+ + 4H+ >2Cu2S + Fe2+. (80) (81) - 121 -2Cu2+ + H2 >2Cu+ + 2H+ (82) Cu2+ + Cu° >2Cu+. (83) This species then reacts with the sulphide mineral as follows, using chalcopyrite as an example: CuFeS2 + 4Cu+ >2Cu2S + Fe2+ + Cu2+ (84) Reduction leaching is potentially an effective method of separating iron from copper in a hydrometal 1 ur gi cal process. Various practical applications of reduction leaching have been proposed. In particular, reduction leaching as a single unit operation offers an attractive alternative to the thermal activation-acid 1eaching-neutral leaching steps in the S.C. Copper Process. 5.3 RECOMMENDATIONS FOR FURTHER WORK Because this study has focussed on practical appli cations, the reaction mechanisms of reduction leaching have not been fully elucidated. To gain a better understanding of the mechani sms involved, future research should be directed towards examining the solid state diffusional processes ocur-ring. This could be accomplished by the following types of exper iments: a) Massive polished sulphide mineral specimens should be reduction leached, then sectioned and subjected - 122 -to a detailed microscopic analysis of structural changes and composition of the conversion product 1ayer s . A combination of X-ray diffraction, electron micro-probe and energy dispersive analysis of the con version products should provide enough information to determine whether the products are chalcocite or some other copper sulphide such as djurleite or digenite. In addition, the detection of iron gradients in product layers would confirm the presence of solid state diffusional process. Measurement of activation energies and other rate dependencies during reduction leaching should be measured under conditions of essentially stable solution composition, which can be accomplished by performing runs at low pulp densities and with large excesses of Cu2+ and Cu°, relative to mineral. Under these conditions, solid state diffusion would be rate-determining rather than formation of cuprous ions at the copper surface. REFERENCES 1. Biswas, A.K. and Davenport, W.G. Extractive Metallurgy  of Copper . Pergarrmon Press ( 1976TJ 438 pp. 2. Rovig, D.A. and Doran, R.K. Copper - a decade of change and its meaning for the future. Min. Cong. Jour. , December (1980), pp. 32-40. 3. From: The potential economic impact of U.S. regulations on the U.S. copper industry. Industry and Trade Admin istration, Department of Commerce, April 1979. 4. Anon. Engng. and Min. Jour., June (1975), pp. 104-105. 5. Mining Annual Review. Min. Jour. , June ( 1978), p. 358. 6. Schweitzer, F.W. and Livingston, R.W. Duval's CLEAR hydrometallurgical process. Chloride Electrometallurgy, Proceedings of a Symposium sponsored by the SME of AIME, held at the 111th AIME Annual Meeting, Dallas, Texas, February 15-16, 1982, 233 pp. 7. Agarwal, J.C. The future of hydrometal1urgy in the copper industry. Paper presented at the CIM Annual  Meeting, Toronto, April 23, 1980. 8. Jones, D.L. and Peters, E. The leaching of chalcopyrite with ferric sulphate and ferric chloride. Extractive Met  allurgy of Copper, Volume 2, J.C. Yannopoulos and J.C. Agarwal, Eds., AIME, N.Y., 1976, pp. 633-653. 9. Dutrizac, J.E. The dissolution of chalcopyrite in ferric sulphate and ferric chloride media. Metal 1. Trans. B, vol. 12B, June (1981), pp. 371-378. 10. Palmer, B.R., Nebo, CO., Rau, M.F. and Fuerstenau, M.C.' Rate phenomena involved in the dissolution of chalcopyrite in chloride-bearing lixiviants. Meta 11.  Trans. B, vol. 12B, September (1981), pp. 595-601. 11. Wilson, J.P. and Fisher, W.W. Cupric chloride leaching of chalcopyrite. J. Metals, February (1981), pp. 52-57. 12. Haver, F.P. and Wong, M.M. Recovering elemental sulfur from nonferrous minerals - ferric chloride leaching of chalcopyrite concentrate. U.S.B.M. Report of Invest  igations 7474, 1971. - 124 -13. Haver, F.P. and Wong, M.M. Recovery of copper, iron and sulphur from chalcopyrite concentrate using a ferric chloride leach. J. Meta 1s, February (1971), pp. 25-29. 14. Haver, F.P., Baker, R.D. and Wong, M.M. Improvements in ferric chloride leaching of chalcopyrite concentrate. U.S.B.M. Report of Investigations 8007, 1975. 15. Phillips, T.A. Economic evaluation of a process for ferric chloride leaching of chalcopyrite concentrate. U.S.B.M. Information Circular 8699, 1976. 16. Atwood, G.E. and Curtis, C.H. Hydrometal1urgical process for the production of copper. U.S. Patent 3,785,944, January 15, 1974; and U.S. Patent 3,879,272, April 22, 1975. 17. Kruesi, P.R. Process for the recovery of me t a 1s f r om sulfide ores through electrolytic dissociation of the sulfides. U.S. Patent 3,673,061, June 27, 1972. 18. Kruesi, Paul R. Cymet copper reduction process. Min.  Cong. jour., September (1974), pp. 22-23. 19. Kruesi, P.R., Allen, E.S., and Lake, J.L. Cymet process-hydrometa 11urgica1 conversion of base-metal sulphides to pure metals. CIM Bui 1., June ( 1973), p. 81-87. 20. McNamara, J.H., Ahrens, W.A. and Franek, J.B. A hydro-metallurgical process for the extraction of copper. Paper presented at the 107th AIME Meeting, February 26, 1978, Denver, Colorado. 21. McNamara, J.H., Ahrens, W.A. and Franek, J.H. A hydro-metallurgical process for the extraction of copper. CIM Bui 1., March (1980), pp. 201-204. 22. Demarthe, J.M. , Gandon, L. and Georgeaux, A. A new hydrometa 11urgica1 process for copper, Ext r act i ve  Metallurgy of Copper, Volume 2, J.C. Yannopoulos and J.C. Agarwol, Eds., AIME, N.Y., 1976, pp. 825-848. 23. Milner, E.F.G., Peters, E., Vizsolyi, A. and Swinkels, G.M. Canadian Patent 938,793, December 25, 1973; and U.S. Patent 3,798,026, March 9, 1974. 24. Peters, E., Swinkels, G.M. and Vizsolyi, A. Copper recovery from sulphide concentrates by the U.B.C. Cominco ferric chloride leach route. Process and Fundamental  Considerations of Selected Hydrometallurgical Systems, M.C. Kuhn, Ed., SME of AIME, N.Y., 1981, pp.71-81. - 125 -25. Tompkins, J. Unique new copper process is ready for unveiling. Mining Review, March/April (1981), pp. 29-31. 26. Everett, P.K. The Dextec copper process. Ext r act i on  Metallurgy'81, I .M.M. Symposium (1981), pp. 149-156. 27. Forward, F.A. and Mackiw, V.N. Chemistry of the anrmonia pressure process for leaching Ni , Cu and Co from Sherritt Gordon sulphide concentrates. J. Meta 1s, March ( 1955), pp.457-463. 28. Kuhn, M.C., Arbiter, N. and Kling, H. Anaconda's Arbiter process for copper. CIM Bu11., February (1974), pp. 62-73. 29. Kuhn, M.C. and Arbiter, N. Recovery of metals. U.S. Patent 4,022,866, May 10, 1977. 30. Evans, D.3.L. et al. Treatment of copper-zinc concentrates by pressure hydrometal1urgy. CIM Bui 1., 628 ( 1964), pp. 857-866. 31. Stanczyk, M.H. and Rampacek, C. Oxidation leaching of copper sulphides in ammoniacal pulps at elevated temperatures and pressures. U.S.B.M. Report of Invest igations 6808, 1966. 32. Tozawa, K., Umetsu, Y. and Sato, K. On chemistry of anrmonia leaching of copper concentrates. Ext r act i ve  Metallurgy of Copper, Volume 2, 3.C. Yannopoulos and 3.C. Agarwal, Eds., AIME, N.Y., 1976, pp. 706-721. 33. Beckstead, L.W. and Miller, 3.D. Ammonia oxidation leaching of chalcopyrite. Meta11. Trans. B., vol. 7B, pp. 19-29. 34. Beckstead, L.W. and Miller, J.D. Ammonia oxidation leaching of cha1 copyrite-surface deposit effects. Metal 1.  Trans. B., vol. 8B, pp. 31-38. 35. Williams, R.D. and Light, S.D. Copper concentrate dissolution chemistry and kinetics and an arrrnon i a-oxygen environment. Fundamental Aspects of Hydrometa 11urgica1  Processes., A.I.Ch.E. Symposium Series, vol. 74, N.Y., 1978, pp. 21-27. 36. Pitt, CH. and Wadsworth, M.E. An assessment of energy requirements in proven and new copper processes. Report prepared for: U.S. Department of Energy, Division of Industrial Energy Conservation, Contract No. EM-78-S-07-1743, December 31, 1980, 362 pp. - 126 -37. Arbiter, N. and Milligan, D.A. Reduction of copper anrmine solutions to metal with sulfur dioxide. Ext ract i ve  Metallurgy of Copper, Volume 2, 3.C. Yannopoulos and 3.C. Agarwal, Eds., AIME, N.Y., 1976, pp. 974-993. 38. Prater, 3.D., Queneau, P.B. and Hudson, T.J. Nitric acid route to processing copper concentrates. Trans. SME- AIME, 3une (1973), pp. 117-122. 39. Habashi, F. Action of nitric acid on chalcopyrite. Trans. SME-AIME, September (1973), pp. 224-231. 40. Bjorling, G., Faldt, I., Lindgren, E. and Toromanov, I. A nitric acid route in combination with solvent extraction for hydr ometal lugical treatment of chalcopyrite. Extractive Metallurgy of Copper, Volume 2, 3.C. Yannopoulos and 3.C. Agarwol, Eds., AIME, N.Y., 1976, pp. 725-737. 41. Brennecke, H.M. Recovery of Metal values from ore con centrates. U.S. Patent 3,888,748, 3une 10, 1975. 42. Brennecke, H.M. et al. Nitric-su1furic leach process for recovery of copper from concentrate. Min. Engng., August (1981), pp. 1259-1266. 43. Davies, D.S. et al . Nitric-sulfuric leach process improve ments. Min. Engng., August (1981), pp. 1252- 1259. 44. Prater, 3.D., Queneau, P.B. and Hudson, T.3. The sulfation of copper-iron sulphides with concentrated sulfuric acid. 3• Meta 1s, December ( 1970), pp. 23-27. 45. Subramanian, K.N. and Jennings, P.H. Review of the hydrometal lurgy of chalcopyrite concentrates. Can. Metal 1.  Quart., 11, 2 (1972), pp. 387-400. 46. Vizsolyi et al. Copper and elemental sulphur from chal copyrite by pressure leaching. 3. Meta 1s, November (1967), pp. 52-59. 47. Baur , 3.P. Gibbs, H.L. and Wadsworth, M.E. Initial-stage sulfuric acid leaching kinetics of chalcopyrite using radiochemical techniques. U.S.B.M. Report of Investi gations 7823, 1974. 48. Beckstead, L.W. et al. Acidic ferric sulphate leaching of attritor-ground chalcopyrite concentrates. Extract ive  Metallurgy of Copper, Volume 2, J.C. Yannopoulos and 3.C. Agarwal, Eds., AIME, N.Y., 1976, pp. 611-632. - 127 -49. Letowski, F. et al. A new hydrometa 11urgica1 method for the processing of copper concentrates using ferric sulphate. Hydromet. 4 (1979), pp. 169-184. 50. Malouf, E.E. and Prater, J.D. Role of bacteria in the alteration of sulfide minerals. J. Meta 1s, 13, May (1961), pp. 353-356. 51. Malouf, E.E. The role of micro-organisms in chemical mining. Mi n. Engng., Nov. (1977), pp. 43-46. 52. Duncan, D.W. , Landesman, J. and Walden, CC. Role of Th i obac i11 us fer rooxi dans in the oxidation of sulfide minerals. Can. Jour. Microbiol., 13 (1967), pp. 397-403. 53. Beck, J.V. and Brown, D.G. Direct sulfide oxidation in the solubilization of sulfide ores by Thiobaci1lus  fer rooxidans. J. Bacter i o1., 96 (4), October (1968), pp. 1433-1434. 54. Duncan, D.W. and Walden, CC Microbiological leaching in the presence of ferric iron. Developments in industrial  microbiology, vol. 13 (1972), pp. 66-74. 55. Sakaguchi, H. , Torma, A.E. and Silver, M. Microbiological oxidation of synthetic chalcocite and covellite by Xtli°ka.£iiIu.l_l.e.LL0.02Sic'a.n.s.* Appl. Environ. Microbiol., M (1), January (19677, pp. 7-10. 56. Torma, A.E. et. al. Microbiological leaching of a chal copyrite concentrate and recovery of copper by solvent extraction and e1ectrowinning. Meta 11., 33, May (1979), pp. 479-484. 57. McElroy, R.O. and Bruynesteyn, A. Continuous biological leaching of chalcopyrite concentrates: demonstration and economic analysis. Metallurgical applications of  bacterial leaching and related microbiological phenomena. Murr, L.E., Torma, A.E. and Brierley, J.A., Eds. Academic Press Publishers, New York, 1978, pp. 441-462. 58. Lawrence, R.W., Bruynesteyn, A. and Hackl, R.P. Recent developments in the bioleaching of sulfide concentrates. Paper presented at the 28th Congress of the I.U.P.A.C, Vancouver, B.C., August 16-22, 1981. 59. Peters, E. The physical chemistry of hydrometal1urgy. International Symposium on Hydrometal1urgy, Chicago, Illinois, February 25-March 1, 1973, pp. 205-228. Evans, D.J.I, and Shoemaker, R.S., Eds. AIME, New York, New York. - 128 -60. McGauley, P.3., Cove, G. and Roberts, E.S. Process for the recovery of copper from its ores and minerals. U.S. Patent 2,568,963, September 25, 1951. 61. Hiskey, 3.B. and Wadsworth, M.E. Galvanic conversion of chalcopyrite. Metal 1. Trans. B., vol. 6B ( 1975), pp. 183-190. 62. McKay, D.R. and Swinkels, G.M. Hydrometa 11urgica1 process for treating copper-iron sulphides. Canadian Patent 953,925, September 3, 1974; and U.S. Patent 3,891,522, 3une 24, 1975. 63. Shirts, M.B., Winter, 3.K., Bloom, P.A. and Potter, G.M. Aqueous reduction of chalcopyrite concentrate with meta 1s. U.S.B.M. Report of Investigations 7953, 1974. 64. Shirts, M.B. and Staker , L. Decomposition of chalcoprite. U.S. Patent 3,985,555, October 12, 1976. 65. Nicol, M.3. Mechanism of aqueous reduction of chalcopyrite by copper, iron and lead. I .M.M. Trans. C, vol. 84 (1975) , pp. C206-C209. 66. Sohn, H.3. and Wadsworth, M.E. Reduction of chalcopyrite with S02 in the presence of cupric ions. 3. Me t a 1s, November (1980), pp. 18-22. 67. Biegler, T. and Swift, D.A. The electrolytic reduction of chalcopyrite in acid solution. 3. Appl. Electrochem. 6 (1976) , pp. 229-235. 68. Biegler, T. and Constable, D.C. Upgrading and activation of chalcopyrite concentrates by slurry electrolysis. I.M.M. Trans. C., vol. 85 ( 1976), pp. C23-C29. 69 Biegler, T. and Constable, D.C. Continuous electrolytic reduction of chalcopyrite slurry. 3. Appl. Electrochem. 7 (1977) , pp. 175-179. 70. Swinkels, G.M. and Berezowsky, R.M.G.S. The Sherritt- . Cominco copper process - part I: the process. CIM Bui 1., February (1978), pp. 105-121. 71. Kawulka, P., Kirby, CR. and Bolton, G.L. The Sherritt-Cominco copper process - part II: pilot-plant operation. CIM Bui 1., February ( 1978), pp. 122- 130. 72. Maschemeyer, D.E.G., Milner, E.F.G. and Parekh, B.M. The Sher r i t t-Comi nco copper process - part III: commercial implications. CIM Bu11., February (1978), pp. 131-139. - 129 -73. Baratin, F. An investigation of the Cu-Fe-S-r^O system at 200 °C. Ph.D. Thesis, U.B.C. Department of Metallurgy, Vancouver, B.C., October, 1980. 74. Stenhouse, J.H. Reduction of aqueous cupric sulphate by hydrogen, carbon monoxide, and their mixtures. M.Sc. Thesis, U.B.C. Department of Metallurgy, Vancouver, B.C., Apr i1, 1982 75. von Hahn, E.A. and Peters, E. The role of Cu(I) in the 0 kinetics of hydrogen reduction of aqueous cupric sulfate solutions. 3. Phys. Chem. 69, February (1965), pp. 547-552. 76. Peters, E. et. al. A carbonyl-hydrometal1urgy method for refining coper. Joint meeting of the MMI3-AIME, Tokyo, 1972. 77. Byerley, J.J. and Peters, E. Kinetics and mechanisms of the reaction between carbon monoxide and copper (II) in aqueous solution. Can. Jour. Chem. 47 (1969), pp. 313-321. 78. Etienne, A. Electrochemical method to measure the copper ionic diffusivity in a copper sulphide scale. J. Electro- chem. Soc, July ( 1970), pp. 870-873. 79. Peters, E. and Hackl, R.P. Iron-copper separation by reduction leaching. Canadian Patent Application, 1981. - 130 -APPENDIX 1 EXAMPLES OF MATERIAL BALANCE CALCULATIONS FOR LEACH RUNS - 131 -APPENDIX 1 Material Balance Calculations  for Table 11 Runs (see page 85) -400 mesh Cu° -100 + 200 mesh Cu° Head Concentrate: Wt. (g) 10.8 10.8 % Cu 26.4 26. 4 % Fe 33. 1 33. 1 % Zn 3.75 3.75 % S 35.6 35.6 Initial Leachate: Vo 1 ume (mL) 50 50 Cu (g/L) 90 90 Copper Powder: Wt.(g) 9.4 9.4 % Cu 99. 9 99.9 Leach Residue: Wt. (g) 19.7 20. 1 % Cu 80. 6 75. 1 % Fe 0.9 3.7 % Zn 0. 95 1.01 % S 19.6 20.3 Final Leachate: Vo 1 ume (mL) 50 50 g/L Cu 22. 0 30. 8 g/L Fe 70. 8 59.6 g/L Zn 4. 55 3.75 - 132 -APPENDIX 1 Material Balance Calculations  for Table 11 Runs (see page 85) (contd.) -400 mesh Cu° -100 + 200 mesh Cu° Weight In (g): Cu 16.75 16.75 Fe 3. 57 3. 57 Zn 0.41 0.41 S 3. 84 3. 84 Weight Out (g): Cu 16.98 16.64 Fe 3. 72 3.72 Zn 0.41 0. 39 S 3. 86 4. 08 Ext ract i on (So 1ut i on Bas i s): Fe 99.0 83. 4 Zn 56.2 49.9 Ext ract i on (Res i due Bas i s): Fe 95. 0 79. 2 Zn 53. 8 49.9 Ext ract ion (Average): Fe 97.0 81. 3 Zn 55.0 48. 1 - 133 -APPENDIX 2 A KINETIC MODEL FOR THE REDUCTION LEACH - 134 -APPENDIX 2 A KINETIC MODEL FOR THE REDUCTION LEACH 1. Assume that the process is parabolic to a first approx imation, with a shrinking core morphology. ut ^—5—* ^ u , % o <S = thickness of original mineral that has disappeared (related to thickness of Cu2S formed). Then, -d(ut) kp dt S that is, the rate-determining step is diffusion through the solid product layer. 2. Develop an expression for ut based on constant geometry (the "spherical" shrinking core model). 135 therefore Integrat ing, Therefore, or , uQ-ut = 2 5 from geometry. -d(ut) = +d(uQ-ut) = 2d(S), 2d(S) kn dt or 2Sd(6) = kpdt 62 = kptj fi= (kpt)^. uQ-ut = 2(kpt)* = (kkpt)y2, ut = uQ - Ukpt)» 3. The percent iron extracted is proportional to the volume fraction of mineral converted to CU2S. 1 Vt = c<ut-? , o<= g7Tfor spheres of diameter ut. V 3 Vt <*ut3 V0 'O/ [u0 - <4kpt)»]3 1 - (kkpt)Y2 Fe extracted = 100 \1- — J, therefore, Vt % Fe extr. V. 100 1-Rearranging, kp uo"uo ^• % Fe ext 100 -)1 4t - 136 -4. Now, using iron extraction data from the neutral and reduction leach runs on bornite, chalcopyrite and pyrite (Table 14), we can calculate the respective parabolic rate constants kpN and kp^. These results are tabulated in the following table. Parabolic Rate Constants for Neutral and Reduction Leaching* Neutral Leach i ng Reduct i on Leach i ng Mineral Fe extr k N Fe extr. k R kpR/kpN % (um)2h- 1 % (um)2h"1 Cu ^FeS^ 19 2. 88 72 74.7 26 CuFeS2 8 0. 470 68 62. 4 133 FeS2 3 0.064 74 81. 8 1278 * Using a uQ value of 50 um and t = 1 h. 5. Therefore, assuming both leaching processes are para bolic, the rate constants for reduction leaching of bornite, chalcopyrite and pyrite are very much higher than the corres ponding neutral leach rate constants. However, for reduction leaching, a parabolic approximation is poor because the Cu2S product ruptures and spalls away due to its own expansion. 


Citation Scheme:


Usage Statistics

Country Views Downloads
United States 8 1
Iran 5 0
Canada 5 0
China 4 10
France 4 0
South Africa 3 0
Iraq 2 0
Chile 2 1
Turkey 2 1
Netherlands 2 0
Australia 2 0
Republic of Korea 2 0
United Kingdom 1 0
City Views Downloads
Unknown 23 8
Ashburn 3 0
Beijing 3 9
Vancouver 3 0
Mountain View 2 0
Tunceli 2 1
Santiago 2 1
Nieuw-Vennep 1 0
Kumla 1 0
San Bruno 1 0
Mashhad 1 0
Amsterdam 1 0
Los Angeles 1 0

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



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items